22 Aug 2013: Han TL, Tumanov S, Cannon RD, Villas-Boas SG (2013) Correction: Metabolic Response of Candida albicans to Phenylethyl Alcohol under Hyphae-Inducing Conditions. PLOS ONE 8(8): 10.1371/annotation/d395f7a5-ac0a-41e2-abd8-fa947717f6a3. https://doi.org/10.1371/annotation/d395f7a5-ac0a-41e2-abd8-fa947717f6a3 View correction
Phenylethyl alcohol was one of the first quorum sensing molecules (QSMs) identified in C. albicans. This extracellular signalling molecule inhibits the hyphal formation of C. albicans at high cell density. Little is known, however, about the underlying mechanisms by which this QSM regulates the morphological switches of C. albicans. Therefore, we have applied metabolomics and isotope labelling experiments to investigate the metabolic changes that occur in C. albicans in response to phenylethyl alcohol under defined hyphae-inducing conditions. Our results showed a global upregulation of central carbon metabolism when hyphal development was suppressed by phenylethyl alcohol. By comparing the metabolic changes in response to phenylethyl alcohol to our previous metabolomic studies, we were able to short-list 7 metabolic pathways from central carbon metabolism that appear to be associated with C. albicans morphogenesis. Furthermore, isotope-labelling data showed that phenylethyl alcohol is indeed taken up and catabolised by yeast cells. Isotope-labelled carbon atoms were found in the majority of amino acids as well as in lactate and glyoxylate. However, isotope-labelled carbon atoms from phenylethyl alcohol accumulated mainly in the pyridine ring of NAD+/NADH and NADP−/NADPH molecules, showing that these nucleotides were the main products of phenylethyl alcohol catabolism. Interestingly, two metabolic pathways where these nucleotides play an important role, nitrogen metabolism and nicotinate/nicotinamide metabolism, were also short-listed through our previous metabolomics works as metabolic pathways likely to be closely associated with C. albicans morphogenesis.
Citation: Han T-L, Tumanov S, Cannon RD, Villas-Boas SG (2013) Metabolic Response of Candida albicans to Phenylethyl Alcohol under Hyphae-Inducing Conditions. PLoS ONE8(8): e71364. https://doi.org/10.1371/journal.pone.0071364
Editor: Jacob Guy Bundy, Imperial College London, United Kingdom
Received: October 17, 2012; Accepted: June 30, 2013; Published: August 12, 2013
Copyright: © 2013 Han et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by University of Auckland Faculty of Science Research Funding and by the University of Auckland Doctoral Scholarship to Ting-Li Han. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
In C. albicans, the switch from yeast to hyphal growth or vice versa is determined by environmental signals that trigger signal transduction pathways and change gene expression. The environmental perturbations can arise from modifications in the level of nutrients, or from molecules that are secreted by C. albicans in a cell density-dependent fashion known as quorum sensing molecules (QSMs) . There are a number of QSMs which have been identified in C. albicans, including farnesol, farnesoic acid, tyrosol, tryptophol, and phenylethyl alcohol –.
Phenylethyl alcohol was one of the first QSMs to be identified in C. albicans . This molecule inhibits both filamentous growth and germ tube formation of C. albicans once its extracellular concentration reaches a certain threshold value. On the other hand, Chen & Fink  demonstrated that phenylethyl alcohol has an opposite effect in S. cerevisiae, where it stimulates filamentous growth in response to ammonia starvation and high cell density. These authors proposed that phenylethyl alcohol is involved in nitrogen- and signalling-mediated morphogenesis in S. cerevisiae. They suggested that ammonia starvation triggers morphogenesis by regulating the production of phenylethyl alcohol which increases in concentration because the reduced amount of ammonium ion alleviates repression of ARO9, ARO10, and other genes required for aromatic alcohol production. Phenylethyl alcohol is also thought to promote S. cerevisiae morphogenesis by upregulating an essential filamentous gene, FLO11, via a PKA pathway-dependent mechanism . In comparison, little is known about the molecular mechanisms by which phenylethyl alcohol regulates the morphological switch of C. albicans. What is known is that the extracellular concentration of phenylethyl alcohol increases in the presence of phenylalanine supplementation in the growth medium, at low ammonia concentration, under alkaline pH and under anaerobic conditions . In addition, phenylethyl alcohol production is not affected by alkaline pH when the transcriptional regulators, ARO80 and RIM101 genes, are disrupted .
Therefore, to further investigate the metabolic response of C. albicans to phenylethyl alcohol, we have applied a gas chromatography-mass spectrometry (GC-MS) metabolomics approach, and isotope labelling experiments, under hyphae-inducing conditions. C. albicans is one of the most prevalent pathogenic yeast species in humans that can cause candidiasis at multiple sites, from mucosae to internal organs. Therefore, understanding the metabolic mechanisms behind one of its key virulence traits, morphogenesis, may provide insights for novel therapeutic interventions to prevent C. albicans infections.
Materials and Methods
Methanol, chloroform, sodium bicarbonate, and sodium hydroxide were obtained from MERCK (Darmstadt, Germany). The internal standard 2,3,3,3-d4-alanine, the derivatization reagent methyl chloroformate (MCF), pyridine, and D-glucose-13C6 were purchased from Sigma-Aldrich (St. Louis, USA). Anhydrous sodium sulphate and pheneylethyl alcohol were obtained from Fluka (Steinheim, Germany). All chemicals were of analytical grade.
Fungal Strain and Culture Media
C. albicans strain SC5314  was maintained on YPD agar medium containing yeast extract (6 g/L), peptone (3 g/L), glucose (10 g/L), and agar (15 g/L); at 30°C. Pre-inocula were prepared in minimum mineral medium (MM medium) at pH 5.5 containing D-glucose (10 g/L), (NH4)2SO4 (5 g/L), MgSO4⋅7H2O (0.5 g/L), KH2PO4 (3 g/L), vitamins and trace metals as previously described . Three different culture media were used for the metabolomic and isotope labelling experiments. (1) Minimum mineral medium (MM medium) at pH 7.5; (2) phenylethyl alcohol medium (PA medium), which consisted of MM medium supplemented with phenylethyl alcohol (1.5 mM); and (3) phenylethyl alcohol medium with 13C-labelled glucose (PA 13C-glucose medium), which consisted of MM medium supplemented with phenylethyl alcohol (1.5 mM) and with glucose (10 g/L) uniformly labelled with 13C instead of 12C-glucose.
C. albicans was cultured in 250 mL MM medium (pH 5.5) at 30°C using shake flasks in a rotary shaker overnight. The cells were collected by centrifugation at 2000 g (4°C) for 5 min and washed in phosphate buffered saline (8 g/L NaCl, 0.2 g/L KCl, 1.44 g/L Na2PO4, 0.24 g/L KH2PO4, at pH 7.5). The cells were resuspended in the three different growth media described above, at an initial OD600 of 0.2. The cells were incubated in a rotary shaker-incubator at 37°C for 12 hr. The morphology of C. albicans cells in each growth medium was monitored using a phase contrast microscope (DMR, Lecia).
Sampling and Quenching of Cell Metabolism
Five replicate shake-flask cultures (30 mL) for each growth medium were harvested at middle to late exponential growth phase (12 hours). Samples (2 mL) of the microbial cultures were membrane-filtered (0.2 µm) to remove C. albicans cells, and the filtrate was used for the analysis of extracellular metabolites. The remaining 28 mL of culture were rapidly filtered under vacuum (Air Cadet vacuum/pressure station, Thermo), quickly washed with cold phosphate buffered saline solution (1–2°C) and quenched in cold methanol water (1∶1v/v) at −30°C as described by Smart et al. . The whole sampling procedure took less than 30 s per sample.
Sample Preparation for Metabolite Analysis
The intracellular metabolites were extracted from the quenched cell pellets using cold methanol water and freeze-thaw cycles following the protocol described by Smart et al. . The Internal standard 2,3,3,3-d4-alanine (0.3 µmol/sample) was added to each sample before extraction. The intracellular metabolite extracts and 1 mL of spent culture medium containing extracellular metabolites were freeze-dried (BenchTop K manifold freeze dryer, VirTis) before chemical derivatization.
Chemical Derivatization of Metabolites
The freeze-dried samples were derivatized using the methyl chloroformate (MCF) protocol developed in-house and described in Smart et al. .
Gas Chromatography-Mass Spectrometry (GC-MS) Analysis
The MCF derivatives were analysed in an Agilent GC7890 system coupled to a MSD5975 mass selective detector (EI) operating at 70 eV. The column used for all analyses was a ZB-1701 GC capillary column (30 m × 250 µm id × 0.15 µm with 5 m guard column, Phenomenex). The GC-MS parameters were set according to Smart et al. . Samples were injected under pulsed splitless mode with the injector temperature at 290°C. The helium gas flow through the GC-column was set at 1.0 mL/min. The interface temperature was set to 250°C and the quadrupole temperature was 200°C.
The cell debris collected after intracellular metabolite extraction was dried using a domestic microwave (250 W for 20 min) and weighed in order to measure the total biomass content (dry weight) of each sample.
Data Mining, Data Normalization, and Data Analysis
AMDIS software (NIST, Boulder, CO, USA) was used for deconvoluting GC-MS chromatograms and identifying metabolites using our in-house MCF MS library. The identifications were based on both the MS spectrum of the derivatized metabolite and its respective chromatographic retention time. The relative abundance of identified metabolites was determined by ChemStation (Agilent) using the GC base-peak value of a selected reference ion. These values were normalized by the biomass content in each sample as well as by the abundance of internal standard (2,3,3,3-d4-alanine). A univariate analysis of variance (ANOVA) was applied to determine whether the relative abundance of each identified metabolite was significantly different between growth conditions. Our Pathway Activity Profiling (PAPi) algorithm  was used to predict and compare the relative activity of different metabolic pathways in C. albicans during the growth conditions tested based on metabolite profiling results. This programme connects to the KEGG online database (http://www.kegg.com) and uses the number of metabolites identified from each pathway and their relative abundances to predict which metabolic pathway is likely to be active in the cell. The entire data mining, data normalization and pathway activity predictions were automated in R software as described in Smart et al.  and Aggio et al. . Graphical representations of the results were generated by gplots and ggplot2 R packages , .
Analysis of Isotope Labelling Distribution in the Detected Metabolites
Due to the commercial unavailability of isotopically labelled phenylethyl alcohol, we decided to apply an inverse isotope labelling approach whereby all C. albicans metabolites were fully labelled with 13C by culturing them in MM medium with 13C-U-labelled glucose as the sole carbon source. This way, we searched for 12C-enrichment in the metabolite profile originating from the metabolism of 12C-phenylethyl alcohol. First, we identified the 13C-labelled metabolites based on their chromatographic retention times obtained by GC-MS analysis of metabolite extracts of cells grown under the same conditions but with 12C-glucose as sole carbon source. The electron-impact fragmentation pattern of each identified MCF derivatized 13C-labelled metabolite was determined based on its corresponding 12C spectrum from our MS library. The degree of labelling was estimated based on the variation observed in the mass spectrum fragmentation pattern of fully 13C-labelled metabolites in comparison to their counterpart 12C mass spectrum.
Suppression of Hyphal Formation by Phenylethyl Alcohol
C. albicans morphogenesis was completely inhibited by 15 mM pheneylethyl alcohol (Figure 1). Phenylethyl alcohol (15 mM) reduced both the growth rate and biomass yield of C. albicans (Figure 2, Table 1) but did not kill the cells. At the sampling point (t = 12 h), microscopic examination showed that phenylethyl alcohol (15 mM) completely inhibited hyphal formation under the growth condition tested, whilst at least 95% of cells grew as hyphae when phenylethyl alcohol was not present (Figure 1).
The Images were taken by Nomarksi contrast microscopy with 800× magnification.
The Extracellular Metabolite Profile of C. albicans Under Suppression of Hyphae Formation by Phenylethyl Alcohol
We detected over 50 metabolites in the spent culture samples of C. albicans and we were able to accurately identify 26 of them using our in-house MS library, including phenylethyl alcohol (Table 2). Nineteen metabolites were detected at significantly different levels when comparing MM and phenylethyl alcohol-supplemented MM cultures (Figure 3). Of these extracellular metabolites 16 appear to have been secreted by C. albicans cells in phenylethyl alcohol supplemented cultures. These metabolites included alanine, β-alanine, benzoate, carbamate, citraconate, citramalate, fumarate, isopalmitate, lactate, nicotinate, proline, succinate, valine, 2-hydroxybutyrate, and 2-isopropylmalate. Other compounds such as glyoxylate seem to have been taken up more extensively by C. albicans cells growing in the presence of phenylethyl alcohol, because they were detected at significant lower concentration compared to the MM medium without phenylethyl alcohol. Interestingly, control cultures without phenylethyl alcohol supplementation secreted small amounts of phenylethyl alcohol to the extracellular medium and cultures with phenylethyl alcohol supplementation showed 16% reduction in the extracellular phenylethyl alcohol levels when compared to non-inoculated medium, therefore confirming that C. albicans cells took up phenylethyl alcohol during growth.
Phenylethyl alcohol treatment (P) is represented by blue triangles and control (C) hyphae-inducing conditions (MM) are represented by red circles. The concentrations of identified metabolites have been normalised by internal standard (d4-alanine) and biomass before the relative concentrations of the corresponding metabolites found in un-inoculated culture medium were subtracted. Standard deviations are indicated by the vertical line-range. The difference between secretion and consumption of extracellular metabolites is distinguished by the dashed lines (y = 0). Secretion of metabolites is indicated by positive values. Consumption of metabolite is indicated by negative values. Only the metabolites generating statistically significant ANOVA scores (p-value<0.05) are shown.
The Intracellular Metabolite Profiles of C. albicans Under Suppression of Hyphae Formation by Phenylethyl Alcohol
In order to investigate further how C. albicans cells respond to phenylethyl alcohol, we compared the profiles of intracellular metabolites from cells grown in MM medium with and without phenylethyl alcohol. We detected over 100 metabolites in the intracellular metabolite extracts and 67 of them were accurately identified across samples. Of these, the concentrations of 51 metabolites were significantly different in cells grown in the phenylethyl alcohol-containing medium (Table 2, Figure 4). Interestingly, all intracellular metabolites were detected at lower concentrations in samples from cultures supplemented with phenylethyl alcohol. These intracellular metabolites include a range of intermediates from the central carbon metabolism such as amino acids, organic acids, fatty acids, and nucleotides. In particular, the concentrations of aromatic amino acids, tyrosine and tryptophan, ornithine, β−alanine, and lysine were reduced up to 16-fold in cells growing in phenylethyl alcohol medium. Furthermore, metabolites such as alanine, β-alanine, benzoate, carbamate, citramalate, lactate, nicotinate, proline, succinate, and valine were found at reduced concentrations intracellularly, but increased concentrations extracellularly in cells exposed to phenylethyl alcohol, suggesting that the cells actively secreted those metabolites when growing in culture media supplemented with phenylethyl alcohol.
Red circles represent concentrations for samples from cells incubated in the absence of phenylethyl alcohol (hyphae-inducing conditions) - that were set to 0. Blue triangles indicate metabolite concentrations in cells exposed to phenylethyl alcohol relative to those in cells grown without phenylethyl alcohol. The metabolite levels relative to the hyphal samples have been plotted using a log2 scale. Negative values indicate that the metabolite concentrations were reduced in response to phenylethyl alcohol. Only the metabolites generating statistically significant ANOVA scores (p-value<0.05) are shown.
Prediction of the Metabolic State of C. albicans Under Suppression of Hyphal Formation by Phenylethyl Alcohol
Using the profile of intracellular metabolites identified in the samples of C. albicans cells growing in the presence or absence of phenylethyl alcohol, we created a comparative metabolic activity profile using PAPi software . The metabolic activities of C. albicans cells under conditions suppressing hyphal formation by phenylethyl alcohol were compared with those in cells cultured in the absence of phenylethyl alcohol (Figure 5). All of the 48 metabolic pathways that showed significant changes in metabolic activity in the cells growing in phenylethyl alcohol-supplemented medium were up-regulated. These include a range of metabolic pathways from the metabolism of amino acids, carbon, cofactors, energy, lipids, nucleotides, and secondary metabolites. In particular, ubiquinone biosynthesis, tryptophan metabolism, and D-arginine/D-ornithine metabolism exhibited a marked upregulation probably in response to phenylethyl alcohol. This is in agreement with our metabolite profile results because two of the main assumptions of the PAPi software, which is a hypothesis generating tool, is that the higher the flux through a given metabolic pathway the larger the number of detected intermediates from that pathway will be and, most importantly, the lower will be the concentration of those compounds inside the cell, because we believe that if a pathway is operating at a high flux there will be a high turn-over rate between its intermediates and a higher incorporation of those intermediates and end products into the biomass, reducing their intracellular concentration.
Red circles represent metabolic activities under hyphae-inducing conditions that were set to 0. Blue triangles indicate metabolic activities in cells treated with phenylethyl alcohol. The metabolic activities relative to the hyphae-inducing samples have been plotted using a log2 scale. Positive values indicate the metabolic pathways had their activity up-regulated in response to phenylethyl alcohol. Only the pathways generating statistically significant ANOVA scores (p-value<0.05) are shown.
12C-label Distribution Through the Metabolite Profile of C. albicans Grown in 13C-U-glucose Medium Supplemented with 12C-phenylethyl Alcohol
When cells grown in 13C-glucose were exposed to 12C-phenylethyl alcohol, a number of metabolites contained at least one carbon from phenylethyl alcohol (Table 3). These metabolites included a number of amino acids (alanine, asparagine, asparate, cysteine, glutamate, histidine, isoleucine, leucine, lysine, ornithine, phenylalanine, proline, serine, valine, 2-aminobutyrate, D-2-aminoadipiate), glyoxylate, and lactate. The majority of threonine and β-alanine molecules contained at least two 12C-atoms. However, we found that the pyridine ring from the NADP+/NADPH and NAD+/NAD molecules were almost fully labelled with 12C in 13C-cultures supplemented with 12C-phenylethyl alcohol (Figure 6). In contrast, intermediates from the TCA cycle (citrate, cis-aconitate, α-ketoglutarate, fumarate, malate, and succinate), two aromatic amino acids (tryptophan, tyrosine), GABA, N-acetylglucosamine, nicotinate, oleate, and pyroglutamate showed no significant 12C-enrichment.
Stacked column plot indicates the percentage that the number of carbons derived from phenylethyl alcohol contributes to the total ion mass in each identified metabolite (see Table 3). Only the pyridine ring of NADPH and NADH are considered for those molecules.
We have detected a global upregulation of central carbon metabolism when C. albicans hyphal formation was suppressed by high concentrations of the quorum sensing molecule, phenylethyl alcohol. This result is in accordance with our previous two metabolomic studies of C. albicans morphogenesis, in which we also observed a global upregulation of central carbon metabolic pathways when hyphal formation was suppressed by farnesol, or a general metabolic downregulation when hyphal growth was promoted by various growth media , . The latter observation was further validated by quantifying intracellular ATP levels, which confirmed a lower ATP concentration when hyphal growth was promoted by various growth media (14).
In an attempt to understand the downstream molecular mechanism of C. albicans morphogenesis, we compared the metabolic profiles of C. albicans cells when growing in the presence of phenylethyl alcohol with the profiles obtained in our previous studies: exposure to farnesol and hyphae-inducing conditions , . This permitted us to short-list 7 metabolic pathways that appear to be altered in all three studies (Figure 7). These metabolic pathways encompassed alanine, asparate and glutamate metabolism; β-alanine metabolism; cysteine and methionine metabolism; histidine metabolism; nitrogen metabolism; nicotinate and nicotinamide metabolism; and pantothenate and CoA biosynthesis. Therefore, we believe that these seven metabolic pathways are likely to be closely associated with the morphogenetic process of C. albicans.
Metabolic pathways responding to phenylethyl alcohol (red) are derived from Fig. 5. Metabolic pathways responding to farnesol (green) and hyphae-inducing conditions (blue) are derived from previous studies , . A small Venn diagram is displayed to illustrate the unique and common metabolic pathways affected by culturing the cells in the presence of farnesol, phenylethyl alcohol or under various hyphae-inducing conditions. a indicates the metabolic pathways that respond in common to all three conditions. b indicates the metabolic pathways that differ from control cells in response to both farnesol and phenylethyl alcohol. The metabolic activities relative to their corresponding controls (black line set to 0) have been plotted using a log2 scale. A positive value indicates that a metabolic pathway was up-regulated in comparison to the control. A negative value means the metabolic pathway activity was down-regulated when compared to the control. Only the metabolic pathways with statistically significant (p-value<0.05) changes in activity are shown.
Furthermore, QSMs such as phenylethyl alcohol and farnesol at relatively high concentrations (1–15 mM) seem to influence the lipid metabolism of C. albicans independently from morphological changes (Figure 7). We have detected five metabolic pathways related to lipid metabolism that appeared to be significantly upregulated only in the presence of phenylethyl alcohol and farnesol . Phenylethyl alcohol at concentrations between 60 and 140 mM is known to exhibit antimicrobial effects against bacteria (e.g. Escherichia coli, Staphylococcus aureus, and Enterococcus faecium) ,  and fungi (e.g. C. albicans itself, Saccharomyces cerevisiae, Kluyveromyces marxianus, and Candida dubliniensis) , . One of its proposed antimicrobial mechanisms is alterations in membrane functions such as permeabilisation of the cell envelope and leakage of potassium ions , . Therefore, the upregulation of fatty acid metabolism in response to the presence of phenylethyl alcohol may support our previous hypothesis  that C. albicans changes its membrane composition in order to reduce the antimicrobial effects of QSMs at high concentrations. In addition, C. albicans is likely to benefit from secreting these toxic QSMs during inter-species competition.
On the other hand, phenylethyl alcohol appears not only to act as a signalling molecule inducing gene expressions as reported before , but it is also taken up from the extracellular medium and metabolised intracellularly as demonstrated by our results. When we provided labelled cells with unlabelled phenylethyl alcohol we observed that almost all amino acids belonging to the histidine, serine, pyruvate, and glutamate families, as well as lactate and glyoxylate, incorporated at least one carbon from phenylethyl alcohol (Figure 6 and Table 3). However, it was a surprise to find the majority of unlabelled-carbon atoms ending up in NAD+/NADH and NADP+/NADPH molecules. Therefore, we hypothesized that once taken up by C. albicans cells, high concentrations of phenylethyl alcohol induce its oxidation back to its known precursor, phenyl acetaldehyde, which is catalysed by a benzyl alcohol dehydrogenase  (Figure 8). Although this enzyme has not been described in C. albicans yet, we have found a putative benzyl alcohol dehydrogenase gene (IFD7) based on orthologous gene identification from the C. albicans genome sequence (C. albicans SC5314, orf19.629). Phenylacetaldehyde could then be converted into phenylacetate and subsequently into 2-hydroxy-phenylacetate  (Figure 8). Moreover, we speculate that 2-hydroxy-phenylacetate could be broken down into catechol and acetic acid via a cis-1,6-dihydroxycyclohexa-2,4-diene-1-acetate intermediate (Figure 8). Similar reactions have been described in bacteria, and are catalysed by oxireductases acting on CH-OH groups with incorporation of one or two atoms of oxygen , . Aerobically, acetate would be converted into acetyl-CoA via acetyladenilate, feeding into the TCA cycle and lipid biosynthesis or converted into malate. However, acetate could also be reduced to acetaldehyde via C. albicans’ aldehyde dehydrogenase (encoded by ALD99). Although we believe the reduction of acetate into acetaldehyde would be less favoured under aerobic conditions, it may explain how some threonine molecules were found to be labelled with two 12C-atoms, because this amino acid can be synthesised from glycine and acetaldehyde via threonine aldolase (GLY12) in C. albicans. But, we cannot explain why we did not detect much isotope-labelling in the TCA cycle intermediates considering that some amino acids from the glutamate family were found to contain 12C-carbons. We can only speculate that the high carbon flux throughout the central carbon metabolism pathways must have diluted the 12C-isotopes to concentrations below the detection limits of our GC-MS method, whilst some amino acids would have a lower turnover inside the cell allowing the detection of their 12C-atoms.
The pathways illustrate how phenylethyl alcohol could be metabolized in C. albicans based on isotope-labelling results. The C. albicans genes annotated as encoding particular enzymatic activities are in red text and their corresponding enzyme codes are in green boxes. The genes and enzymes are as follows: IFD7 (benzyl alcohol dehydrogenase), ADL4 (aldehyde dehydrogenase), PAH1 (oxireductase acting on the CH-OH group of donors and incorporation of one atom of oxygen), ACS2 (acetyl-CoA synthetase), ALD99 (aldehyde dehydrogenase), MLS1 (malate synthase), BNA1 (3-hydroxyanthranilic acid dioxygenase), nadC (nicotinate-nucleotide pyrophosphorylase), NMNAT (nicotinamide mononucleotide adenylyltransferase), NADSYN1 (NAD synthetase 1). Enzymes not yet described in C. albicans are in grey boxes. The putative enzymes are as follow: 1.12.12- (oxireductase acting on the CH-OH group of donors and incorporation of two atoms of oxygen), 1.3.1- (oxireductase acting on the CH-CH group of donors), 1.14.-.- (oxireductase), 1.7.1.- (nitrobenzene nitroreductase), 184.108.40.206 (hydroxylaminobenzene mutase), Amn (2-aminophenol-1,6-dioxygenase), 220.127.116.11 (aminocarboxymuconate-semialdehyde decaborxylase), 18.104.22.168 (NAD+ kinase).
Furthermore, we speculate that the benzene ring from phenylethyl alcohol could be incorporated into nucleotide molecules through its metabolic conversion into the pyridine ring of NADH and subsequently NADH is phosphorylated to form NADPH (Figure 8). Based on the current understanding of NADH biosynthesis, there are three possible metabolic routes by which a pyridine ring of NADH could be potentially synthesized: i) via de novo synthesis from aspartate, but aspartate was not found to be labelled to any great extent; ii) via salvage pathways by recycling of compounds containing nicotinamide, but phenylethyl alcohol does not contain nicotinamide; or iii) via de novo synthesis from tryptophan catabolism. There is no evidence that a benzene ring from phenylethyl alcohol could be directly converted into pyridine through biochemical reactions. One possibility would be for C. albicans to convert catechol into 2-aminophenol. Some bacteria have been shown to be able to do this through bezene nitroreductase reactions  (Figure 8). C. albicans has a dioxygenase that is capable of opening the benzene ring from 3-hydroxyanthranilic acid, a similar molecule to 2-aminophenol, through the incorporation of two atoms of dioxygen and spontaneously rearranging the molecule into a pyridine structure (Figure 8). For example, in the C. albicans tryptophan catabolism pathway, 3-hydroxyanthranilate dioxygenase (BNA1) catalyses the cleavage of the benzene ring from 3-hydroxyanthranilate into 2-amino-3-carboxymuconate semialdehyde , . This unstable compound spontaneously cyclises to form a pyridine structure and becomes quinolinate, an intermediate involved in the de novo biosynthesis of NADH from tryptophan. Therefore we speculate that the C. albicans 3-hydroxyanthranilate dioxygenase could potentially accept 2-aminophenol as substrate converting it into 2-aminomuconate semialdehyde, which could be converted into 2-amino-3-carboxymuconate semialdehyde via the aminocarboxymuconate-semialdehyde decarboxylase reverse reaction (Figure 8). Aminocarboxymuconate-semialdehyde decarboxylase (EC. 22.214.171.124) has been described in different mammals and bacteria, and plays an important role in tryptophan catabolism [http://www.brenda-enzymes.info/php/result_flat.php4?ecno=126.96.36.199]. The enzyme regulates NADH biosynthesis from amino acids, directly affecting quinolinate and picolinate formation . This enzyme has very low Km values (0.001 to 0.09), which suggest that its reaction should be reversible [http://www.brenda-enzymes.info/php/result_flat.php4?ecno=188.8.131.52]. However, we would expect the cells to have some pyridine ring unlabelled due to de novo synthesis, but they did not. Thus, the fully labelled pyridines of cofactors remain puzzling and require further investigations.
An alternative hypothesis regarding the isotope labelling results is that C. albicans assimilated 12C-carbons by fixing exogenous CO2. S. cerevisae has been shown experimentally to fix CO2 into phosphoenolpyruvate by phosphoenolpyruvate carboxykinase (Pck1p) forming oxaloacetate, an intermediate of the TCA cycle , . C. albicans often grows within the host in an environment where the CO2 level (5%) is more than 150-fold greater than in the atmosphere (0.033%) . The ability to utilise additional carbon from CO2 for biosynthetic purposes during morphogenesis, seems to be a good growth and invasion strategy, considering that an atmosphere containing 5 to 15% of CO2 also stimulates germ tube formation in C. albicans , . However, there is no direct evidence that C. albicans is capable of fixing CO2 for cellular growth, even though C. albicans does have a putative phosphoenolpyruvate carboxykinase gene (PCK1) based on orthologous gene identification from its genome sequence. Nevertheless, this could explain why a large proportion of lactate molecules were labelled with one 12C carbon.
By combining the fact that phenylethyl alcohol could be potentially catabolised mainly into the pyridine structure of NADH/NADPH molecules with the pathways we have previously short-listed as potentially related to the morphogenetic process, we have identified two candidate primary metabolic pathways - nitrogen metabolism and nicotinate/nicotinamide metabolism - which could play a central role in C. albicans morphogenesis. These two pathways not only significantly change their activity during various hyphal-inducing conditions as indicated by our previous metabolomics studies, but they are also responsible for the biosynthesis and replenishment of NADH and NADPH molecules in the cell. Nitrogen metabolism plays a role in maintaining the redox balance between the reduced and oxidized states of NADH and NADPH. Nicotinate and nicotinamide metabolism produces NAD+ via reutilizing compounds containing nicotinamide, and by de novo synthesis of NAD+ from tryptophan or aspartate. Subsequently, NAD+ can be phosphorylated into NADP+ by NAD+ kinase . It is important to note that NADH and NADPH have a significant influence on central carbon metabolism. NADH not only acts as an electron carrier mediating energy metabolism, it is also involved in other cellular process such as post-translational modifications that affect the activity of metabolic enzymes. NADPH is a potent reducing agent that drives biochemical reactions toward anabolic metabolism including biosynthesis of lipids, nucleic acids, and amino acids. Therefore, nitrogen metabolism and nicotinate/nicotinamide metabolism have great potential to be the main link between signalling pathways and downstream primary metabolism during C. albicans morphogenesis.
This is the first investigation of the metabolic response of C. albicans to phenylethyl alcohol. We observed a global upregulation of central carbon metabolism when filamentous growth was suppressed by phenylethyl alcohol, and we demonstrated that this QSM is not only secreted by C. albicans cells, but it is also actively taken up and catabolised intracellularly. We found strong evidence that phenylethyl alcohol is catabolised mainly into the pyridine ring of NADH and NADPH molecules which could affect the concentration of these molecules inside the cells as well as their redox state; explaining its marked effect on the central carbon metabolism of C. albicans and consequent inhibition of hyphal formation. NADH and NADPH are important electron carriers responsible for the redox balance of the cell. The disruption of the cell’s redox balance would have major effects on primary metabolism. Therefore, our studies suggest that nitrogen metabolism and nicotinate/nicotinamide metabolism play an important role in the morphogenetic process of C. albicans, and further investigations are required to better understand how these pathways are connected to the upstream signalling pathways of morphogenetic regulation in C. albicans.
We thank M. Sabherwal for helping with sample preparation, R. Aggio for data analysis assistance, and T. Liu for proof reading.
Conceived and designed the experiments: TLH SVB RDC. Performed the experiments: TLH. Analyzed the data: TLH ST. Contributed reagents/materials/analysis tools: SVB RDC. Wrote the paper: TLH SVB RDC.
- 1. Han TL, Cannon RD, Villas-Bôas SG (2011) The metabolic basis of Candida albicans morphogenesis and quorum sensing. Fungal Genetics and Biology 48: 747–763.
- 2. Hornby JM, Jensen EC, Lisec AD, Tasto JJ, Jahnke B, et al. (2001) Quorum sensing in the dimorphic fungus Candida albicans is mediated by farnesol. Applied and Environmental Microbiology 67: 2982–2992.
- 3. Chen H, Fujita M, Feng Q, Clardy J, Fink GR (2004) Tyrosol is a quorum-sensing molecule in Candida albicans. Proceedings of the National Academy of Sciences of the United States of America 101: 5048–5052.
- 4. Lingappa BT, Prasad M, Lingappa Y, Hunt DF, Biemann K (1969) Phenethyl alcohol and tryptophol: Autoantibiotics produced by the fungus Candida albicans. Science 163: 192–194.
- 5. Hornby JM, Nickerson KW (2004) Enhanced production of farnesol by Candida albicans treated with four azoles. Antimicrobial Agents and Chemotherapy 48: 2305–2307.
- 6. Chen H, Fink GR (2006) Feedback control of morphogenesis in fungi by aromatic alcohols. Genes & development 20: 1150–1161.
- 7. Ghosh S, Kebaara BW, Atkin AL, Nickerson KW (2008) Regulation of aromatic alcohol production in Candida albicans. Applied and Environmental Microbiology 74: 7211–7218.
- 8. Gillum AM, Tsay EYH, Kirsch DR (1984) Isolation of the Candida albicans gene for orotidine-5′-phosphate decarboxylase by complementation of S.cerevisiae ura3 and E. coli pyrF mutations. Molecular and General Genetics 198: 179–182.
- 9. Verduyn C, Postma E, Scheffers WA, Van Dijken JP (1992) Effect of benzoic acid on metabolic fluxes in yeasts: A continuous-culture study on the regulation of respiration and alcoholic fermentation. Yeast 8: 501–517.
- 10. Smart KF, Aggio RBM, Van HoutteJR, Villas-Bôas SG (2010) Analytical platform for metabolome analysis of microbial cells using methyl chloroformate derivatization followed by gas chromatography-mass spectrometry. Nature Protocols 5: 1709–1729.
- 11. Aggio RBM, Ruggiero K, Villas-Bôas SG (2010) Pathway activity profiling (papi): From the metabolite profile to the metabolic pathway activity. Bioinformatics 26: 2969–2976.
- 12. Wickham H (2009) ggplot2: elegant graphics for data analysis: Springer New York.
- 13. Warnes GR, Bolker B, Lodewijk Bonebakker, Gentleman R, et al.. (2010) gplots: Various R programming tools for plotting data. CRAN website. Available: http://CRAN.R-project.org/package=gplots. Accessed 2013 Jul 15.
- 14. Han TL, Cannon RD, Villas-Bôas SG (2012) Metabolome analysis during the morphological transition of Candida albicans. Metabolomics: DOI 10.1007/s11306-11012-10416-11306.
- 15. Han TL, Cannon RD, Villas-Bôas SG (2012) The metabolic response of Candida albicans to farnesol under hyphae-inducing conditions. FEMS Yeast Research: DOI: 10.1111/j.1567-1364.2012.00837.x.
- 16. Corre J, Lucchini JJ, Mercier GM, Cremieux A (1990) Antibacterial activity of phenethyl alcohol and resulting membrane alterations. Research in Microbiology 141: 483–497.
- 17. Lucchini JJ, Corre J, Cremieux A (1990) Antibacterial activity of phenolic compounds and aromatic alcohols. Research in Microbiology 141: 499–510.
- 18. Fabre CE, Blanc PJ, Goma G (1998) Production of 2-phenylethyl alcohol by Kluyveromyces marxianus. Biotechnology Progress 14: 270–274.
- 19. Martins M, Henriques M, Azeredo J, Rocha SM, Coimbra MA, et al. (2007) Morphogenesis control in Candida albicans and Candida dubliniensis through signaling molecules produced by planktonic and biofilm cells. Eukaryotic Cell 6: 2429–2436.
- 20. Suhara K, Takemori S, Katagiri M (1969) The purification and properties of benzylalcohol dehydrogenase from Pseudomonas SP. Archives of Biochemistry and Biophysics 130: 422–429.
- 21. Neidle E, Hartnett C, Ornston LN, Bairoch A, Rekik M, et al. (1992) Cis-diol dehydrogenases encoded by the TOL pWW0 plasmid xylL gene and the Acinetobacter calcoaceticus chromosomal benD gene are members of the short-chain alcohol dehydrogenase superfamily. European Journal of Biochemistry 204: 113–120.
- 22. Harayama S, Rekik M, Bairoch A, Neidle EL, Ornston LN (1991) Potential DNA slippage structures acquired during evolutionary divergence of Acinetobacter calcoaceticus chromosomal benABC and Pseudomonas putida TOL pWW0 plasmid xylXYZ, genes encoding benzoate dioxygenases. Journal of Bacteriology 173: 7540–7548.
- 23. Park HS, Kim HS (2000) Identification and characterization of the nitrobenzene catabolic plasmids pNB1 and pNB2 in Pseudomonas putida HS12. Journal of Bacteriology 182: 573–580.
- 24. Panozzo C, Nawara M, Suski C, Kucharczyka R, Skoneczny M, et al. (2002) Aerobic and anaerobic NAD+ metabolism in Saccharomyces cerevisiae. FEBS Letters 517: 97–102.
- 25. Jones T, Federspiel NA, Chibana H, Dungan J, Kalman S, et al. (2004) The diploid genome sequence of Candida albicans. Proceedings of the National Academy of Sciences of the United States of America 101: 7329–7334.
- 26. Pucci L, Perozzi S, Cimadamore F, Orsomando G, Raffaelli N (2007) Tissue expression and biochemical characterization of human 2-amino 3-carboxymuconate 6-semialdehyde decarboxylase, a key enzyme in tryptophan catabolism. FEBS Journal 274: 827–840.
- 27. Oura E, Haarasilta S, Londesborough J (1980) Carbon dioxide fixation by baker’s yeast in a variety of growth conditions. Journal of General Microbiology 118: 51–58.
- 28. Stoppani AO, Conches L, De Favelukes SL, Sacerdote FL (1958) Assimilation of carbon dioxide by yeasts. The Biochemical journal 70: 438–455.
- 29. Klengel T, Liang W-J, Chaloupka J, Ruoff C, Schröpel K, et al. (2005) Fungal adenylyl cyclase integrates CO2 sensing with cAMP signaling and virulence. Current Biology 15: 2021–2026.
- 30. Mock RC, Pollack JH, Hashimoto T (1990) Carbon dioxide induces endotrophic germ tube formation in Candida albicans. Canadian journal of microbiology 36: 249–253.
- 31. Bahn YS, Mühlschlegel FA (2006) CO2 sensing in fungi and beyond. Current Opinion in Microbiology 9: 572–578.
- 32. Lehninger AL (2005) Lehninger principles of biochemistry. New York: W.H Freeman.