You are on page 1of 11


The metabolic response of Candida albicans to farnesol under

hyphae-inducing conditions
Ting-Li Han1, Richard D. Cannon2 & Silas G. Villas-Boas1
Centre for Microbial Innovation, School of Biological Sciences, The University of Auckland, Auckland, New Zealand; and 2Department of Oral
Sciences, The University of Otago, Dunedin, New Zealand

Correspondence: Silas G. Villas-Boas, School Abstract

of Biological Sciences, Centre for Microbial
Innovation, The University of Auckland, 3A Farnesol is a quorum-sensing molecule (QSM) produced, and sensed, by the
Symonds Street, Auckland 1010, New polymorphic fungus, Candida albicans. This cell-to-cell communication
Zealand. Tel.: +64 9 373 7599; molecule is known to suppress the hyphal formation of C. albicans at high cell
fax: +64 9 373 7416; density. Despite many studies investigating the signalling mechanisms by which
QSMs influence the morphogenesis of C. albicans, the downstream metabolic
effect of these signalling pathways in response to farnesol-mediated morpho-
Received 4 April 2012; revised 22 July 2012;
accepted 26 July 2012. Final version
genesis remains obscure. Here, we have used metabolomics to investigate the
published online 4 September 2012. metabolic response of C. albicans upon exposure to farnesol under hyphae-
inducing conditions. We have found a general up-regulation of central carbon
DOI: 10.1111/j.1567-1364.2012.00837.x metabolic pathways when hyphal formation was suppressed by farnesol
evidenced by a considerably larger number of central carbon metabolic inter-
Editor: Jens Nielsen mediates detected under this condition at an overall lower intracellular level.
By combining the metabolic profiles from farnesol-exposed cells with previous
metabolomics data for C. albicans undergoing morphogenesis, we have identi-
Candida albicans; filamentous growth;
fied several metabolic pathways that are likely to be associated with the
quorum-sensing molecules; farnesol;
metabolomics. morphogenetic process of C. albicans, as well as metabolic pathways such as
those involved in lipid metabolism that appeared to be specifically affected by
farnesol. Therefore, our results provide important new insights into the meta-
bolic role of farnesol in C. albicans metabolism.

Maidan et al., 2005; Han et al., 2011). Furthermore,


a population of C. albicans cells has the capacity to use

Candida albicans is a commensal, opportunistic, fungal chemical signals in order to co-ordinate a synchronous
pathogen that is commonly found in the normal human morphogenetic process. Candida albicans is known to pro-
flora. This fungus causes little disease in the healthy duce several quorum-sensing molecules (QSMs) that regulate
population but can result in fatal infections in immunocom- morphogenesis, of which farnesol is the best characterised
promised patients. Indeed, C. albicans has been attributed (Albuquerque & Casadevall, 2012). Farnesol has been shown
as the fourth leading cause of mortality in nosocomial to stop germ tube formation in the later stage of biofilm
bloodstream infections of severely immunocompromised development, as well as under other hyphal-inducing condi-
patients (Pfaller & Diekema, 2007; Moran et al., 2010). tions (e.g. when grown on N-acetylglucosamine, proline or
The ability to rapidly switch from budding yeasts to serum) (Hornby et al., 2001; Alem et al., 2006), but it is
hyphal growth and vice versa in response to environmental unable to block the elongation of pre-existing hyphae (Mosel
factors is regarded as an important virulence factor for et al., 2005; Navarathna et al., 2005).
C. albicans (Whiteway & Oberholzer, 2004). The morpho- After farnesol was first isolated from C. albicans cultures
logical change of C. albicans is triggered by environmental by Hornby et al. (2001), a number of studies have
signalling. The yeast-to-hyphal development can be investigated the molecular mechanisms by which QSMs
induced by N-acetylglucosamine, proline, serum and starva- influence C. albicans morphogenesis. For example, Sato
tion (Reynolds & Braude, 1956; Holmes & Shepherd, 1987; et al. (2004) showed, using RT-PCR analysis, that farnesol

FEMS Yeast Res 12 (2012) 879889 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
880 T.-L. Han et al.

suppresses the HST7 and CPH1 gene expression involved g L 1), KH2PO4 (3 g L 1), vitamins and trace metals as
in MAP kinase cascades. Furthermore, a stress-activated previously described (Verduyn et al., 1992).
protein kinase (Hog1p) was also found to be phosphory-
lated and accumulate in the nucleus in response to farne-
Culture conditions
sol (Smith et al., 2004). cDNA microarray analysis
showed that some genes related to morphogenesis are Candida albicans was cultured in 250 mL of MM medium
down-regulated (e.g. CRK1 and PDE2), while others are (pH 5.5) at 30 C in a rotary shaker overnight. The cells
up-regulated (e.g. TUP1) in C. albicans biofilms exposed were collected by centrifugation at 2000 g (4 C) for
to farnesol (Cao et al., 2005). Davis-Hanna et al. (2008) 5 min and washed in phosphate-buffered saline (8 g L 1
suggested that farnesol stops hypha formation by repress- NaCl, 0.2 g L 1 KCl, 1.44 g L 1 Na2PO4, 0.24 g L 1
ing the RAS1-cAMP signalling pathway. Kebaara et al. KH2PO4, at pH 7.5). The cells were resuspended in 20
(2008) demonstrated that farnesol suppressed hyphal shake-flasks containing 100 mL of MM medium at pH
growth by up-regulating the TUP1 gene (a transcriptional 7.5 (to induce filamentous growth) to an initial OD600 of
repressor) because it no longer stops germ tube formation 0.2. Five flasks were supplemented with farnesol (1 mM),
in tup1/tup1 and nrg1/nrg1 null mutants. Roman et al. and another five flasks were used as control. The cell
(2009) revealed that farnesol prevents the phosphoryla- suspensions were incubated in a rotary shaker at 37 C
tion of Cek1p, part of the MAPK signal transduction for 6 h. The morphology of C. albicans in each growth
pathway. And recently, Hall et al. (2011) showed that medium was monitored using a phase-contrast micro-
farnesol inhibited cAMP-PKA cascade via the inactivation scope (DMR, Lecia).
of adenylyl cyclase (CYR1p). Together, these studies sug-
gest that farnesol inhibits germ tube formation by
Sampling and quenching of cell metabolism
repressing both MAPK and cAMP-PKA signalling path-
ways and by stimulating the expression of hyphal sup- Five flasks for each growth condition were harvested after
pressor genes such as TUP1 and HOG1. However, the 3- and 6-h incubation. A portion (10 mL) of the cultures
metabolic effect of farnesol on the central carbon metabo- was filtered (0.2 lm pore size) to remove the C. albicans
lism of C. albicans, which is downstream to signalling cells, and the filtrate was used for the analysis of extracel-
pathways, is yet to be studied. Therefore, in this work, we lular metabolites. The remaining 90 mL of culture was
have employed a gas chromatographymass spectrometry rapidly filtered under vacuum (Air Cadet vacuum/pres-
(GC-MS)-based metabolomics approach to investigate the sure station, Thermo), quickly washed with cold saline
metabolic response of C. albicans cells to farnesol under solution (12 C) and quenched in cold methanol/water
hyphae-inducing conditions. (1 : 1 v/v) at 30 C as described by Smart et al. (2010).

Materials and methods Sample preparation for metabolite analysis

The intracellular metabolites were extracted from the
quenched cell pellets using cold methanol/water and
Methanol, chloroform, sodium bicarbonate and sodium freezethaw cycles following the protocol described by
hydroxide were obtained from MERK (Darmstadt, Smart et al. (2010). The internal standard 2,3,3,3-d4-alanine
Germany). The internal standard 2,3,3,3-d4-alanine, the (0.3 lmol) was added to each sample before extraction.
derivatisation reagent methyl chloroformate (MCF), and The intracellular metabolite extracts and 1 mL of spent
pyridine were purchased from Sigma-Aldrich (St. Louis, culture medium containing extracellular metabolites were
MO). Anhydrous sodium sulphate and farnesol were freeze-dried (BenchTop K manifold freeze dryer, VirTis)
obtained from Fluka (Steinheim, Germany). All chemicals before chemical derivatisation.
used in this study were of analytical grade.
Chemical derivatisation of metabolites
Fungal strain and culture media
The freeze-dried samples were derivatised using the MCF
Candida albicans strain SC5314 was maintained on YPD method according to the protocol described by Smart
agar medium containing yeast extract (6 g L 1), peptone et al. (2010). In summary, the freeze-dried samples were
(3 g L 1), glucose (10 g L 1) and agar (15 g L 1), at 30C. resuspended in 200 lL sodium hydroxide (1 M) and
Candida albicans cells were grown in minimum mineral transferred to a silanised glass tube, then mixed with
medium (MM medium) at pH 5.5 containing D-glucose 167 lL methanol and 34 lL pyridine. The derivatisation
(10 g L 1), (NH4)2SO4 (5 g L 1), MgSO47H2O (0.5 began by adding 20 lL MCF followed by vigorously

2012 Federation of European Microbiological Societies FEMS Yeast Res 12 (2012) 879889
Published by Blackwell Publishing Ltd. All rights reserved
The metabolic response of C. albicans to farnesol 881

mixing for 30 s, and then, a further 20 lL MCF was

Data mining, data normalisation and data
added followed by vigorously mixing for 30 s. To separate
MCF derivatives from the reactive mixture, 400 lL chlo-
roform was added and vigorously mixed for 10 s followed AMDIS software was used for deconvoluting GC-MS chro-
by the addition of 400 lL sodium bicarbonate (50 mM) matograms and identifying metabolites using our in-
and mixing for an additional 10 s. The aqueous layer was house MCF mass spectra library. The identifications were
removed and dehydrated with anhydrous sodium sulphate based on both MS spectrum of the derivatised metabolite
before samples were transferred to GC-MS vials. and its respective chromatographic retention time. The
relative abundance of identified metabolites was deter-
mined by ChemStation (Agilent) using the GC base peak
Gas chromatographymass spectrometry
value of a selected reference ion. These values were nor-
(GC-MS) analysis
malised by the biomass content in each sample as well as
The MCF derivatives were analysed in an Agilent GC7890 by the abundance of internal standard (2,3,3,3-d4-alanine).
system coupled to a MSD5975 mass selective detector A complete list of all metabolites identified in all samples
(EI) operating at 70 eV. The column used for all analyses and their respective normalised relative abundances is
was a ZB-1701 GC capillary column (30 m 9 250 lm provided as Supporting Information (Table S1). A uni-
id 9 0.15 lm with 5 m guard column, Phenomenex). variate analysis of variance (ANOVA) was applied to deter-
The analysis parameters were set according to Smart et al. mine whether the relative abundance of each identified
(2010). Samples were injected under pulsed splitless mode metabolite was significantly different between growth con-
with the injector temperature at 290 C. The helium ditions. Our Pathway Activity Profiling (PAPi) algorithm
gas flow through the GC column was set at 1 mL min 1. (Aggio et al., 2010) was used to predict and compare
The interface temperature was set to 250 C, and the the relative activity of different metabolic pathways in
quadrupole temperature was 200 C. C. albicans during the growth conditions tested. This pro-
gram connects to the KEGG online database (http://www. and uses the number of metabolites identified
Biomass quantification
from each pathway and their relative abundances to pre-
The cell debris collected after intracellular metabolite dict which metabolic pathway is likely to be active in the
extraction was dried using a domestic microwave (250 W cell. The entire data mining, data normalisation and path-
for 20 min) and weighed in order to estimate the total way activity predictions were automated in R software as
biomass content (dry weight) of each sample. described in the study by Smart et al. (2010) and Aggio

Fig. 1. The morphology of Candida albicans

cells incubated in the presence or absence of
farnesol (1 mM) at 37 C after 3- or 6-h
incubation. MM is minimum mineral medium.
The images were taken by Nomarski contrast
microscopy with 10009 magnification.

FEMS Yeast Res 12 (2012) 879889 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
882 T.-L. Han et al.

et al. (2010). Graphical representations of the results were both culture media (with or without farnesol). However,
generated by ggplot2 R packages (Wickham, 2009). after 6-h incubation, cultures growing in the absence of
farnesol had significantly higher biomasses (Table 1).
The effect of farnesol on the extracellular
Suppression of hyphal formation by farnesol metabolite profile of C. albicans
To investigate the metabolomic changes associated with We detected over 50 GC-MS peaks in extracellular
hyphal suppression by farnesol, we cultivated C. albicans C. albicans samples, and 29 of them were identified using
in hyphae-inducing conditions supplemented with farne- our in-house MS library (Table 2). The concentrations of
sol for 6 h. Microscopic examination showed that farne- 20 metabolites appeared to be significantly different
sol (1 mM) completely inhibited hyphal formation under between samples from cells exposed to farnesol and
the growth condition tested, while at least 95% of cells samples from those not exposed to farnesol (Fig. 2).
underwent filamentous growth when farnesol was not Interestingly, after 3-h incubation with farnesol, the con-
present (Fig. 1 and Table 1). At the first sampling point centrations of extracellular metabolites secreted into the
(3 h), the biomass concentration was roughly equal in medium were low when compared with those for cells
growing in the absence of farnesol except for caprinate
Table 1. Biomass and morphology of Candida albicans cells cultured and EDTA (EDTA was in the medium and could be
in different growth media taken up by the cells). These metabolites consisted of vari-
Sampling time
ous TCA cycle intermediates (e.g. fumarate, succinate),
amino acids (e.g. alanine, glutamate, proline and valine),
3h 6h
fatty acids (e.g. caprinate, oleate) and quorum-sensing
Biomass Biomass molecules (e.g. phenylethyl alcohol).
Medium (mg mL 1) F (%) (mg mL 1) F (%)

MM 0.39 > 95 1.64 > 95

The effect of farnesol on the intracellular
MM + Farnesol 0.33 0 0.97 0
(1 mM)
metabolite profile of C. albicans

MM, minimum mineral medium; F, percentage of filamentous growth From a total of 60 metabolites identified in intracellular
determined by counting the number of yeast cells and filaments in samples of both farnesol-treated and untreated C. albicans
1-mm3 volume. cells, the concentrations of 34 metabolites were affected

Table 2. Candida albicans metabolites identified in all culture media

Classification of
metabolites Intra Extra Metabolites

Amino acids 19 6 Alanine*, asparagine, aspartate, cysteine, glutamine, glutamate*, glycine*,

histidine, isoleucine, leucine, lysine, phenylalanine, proline*, serine, threonine,
tryptophan, tyrosine, valine* and b-alanine*
Amino acid derivatives 7 0 Creatinine, cystathionine, N-acetylglutamate, norvaline, ornithine,
2-aminobutyrate and pyroglutamate
TCA cycle intermediates 6 4 Fumarate*, citrate*, succinate*, cis-aconitate, malate and 2-oxoglutarate*
Fatty acids 10 4 Caprinate*, caprylate, docosanoate, D-2-aminoadipate, myristate, oleate*,
pentadecanoate, 14-methylpentadecanoate*, stearate* and 3-hydroxyoctanoate
Glycolytic intermediates 2 1 Pyruvate* and phosphoenolpyruvate
Cofactors and vitamins 3 1 NADP/NADPH, nicotinate* and 4-amino-n-butyrate
Others 13 8 Benzoate*, cabamate*, citraconate*, citramalate*, eicosanoate, itaconate,
lactate*, malonate, 2-isopropylmalate*, 2-hydroxybutyrate, 10,13-dimethyl
tetradecanoate*, 3,5-bis(1,1-dimethylethyl)-4-hydroxy-benzenepropanoic
acid* and 10,12-octadecadienoate
Metabolites only found in extracellular media 0 5 Glyoxylate/glyoxalate*, phenylethyl alcohol*, 2,4-bis(1,1-dimethylethyl)-phenol*,
EDTA* and 2-methoxysuccinate*
Total number of identified metabolites 60 29

Intra, number of intracellular metabolites identified; Extra, Number of extracellular metabolites identified.
*Metabolites found extracellularly. Glyoxylate/glyoxalate, phenylethyl alcohol, 2,4-bis(1,1-dimethylethyl)-phenol, EDTA and 2-methoxysuccinate
were only found extracellularly.

2012 Federation of European Microbiological Societies FEMS Yeast Res 12 (2012) 879889
Published by Blackwell Publishing Ltd. All rights reserved
The metabolic response of C. albicans to farnesol 883

Fig. 2. Relative concentration of extracellular metabolites in the presence and absence of farnesol after 3- and 6-h incubation. Farnesol-treated
samples (suppression of hyphae) are indicated by blue triangles (), and nonfarnesol-treated samples by red round dots (). Standard deviations
are illustrated by vertical lines. The relative concentrations of identified metabolites have been normalised by internal standard (d4-alanine) and
biomass before the relative concentrations of the corresponding metabolites found in uninoculated culture medium were subtracted. The line
y = 0 distinguishes secretion of metabolites (positive values) from the consumption of metabolites from the medium (negative values). Only the
metabolites detected with statistically significant ANOVA (P-value < 0.05) are shown.

significantly by the presence of farnesol (Table 2, Fig. 3). incubation in the presence of farnesol (Fig. 3). After 3-h
Most intracellular metabolites were detected at lower lev- incubation, cysteine was detected at higher concentrations
els in samples from cultures supplemented with farnesol. in samples from cells cultured in the presence of farnesol,
Myristate (a saturated fatty acid) showed the greatest con- while at 6-h incubation, the level of cysteine was lower in
centration decrease in response to farnesol (> 3-fold). cells exposed to farnesol.
Phenylalanine, isoleucine, glutamate, aspartate, leucine,
2-isopropylmalate, dimethyl myristate and methyl isopalmi-
Effect of farnesol, and the suppression of
tate also showed significant decreases in response to
hyphal formation, on the metabolic pathways
farnesol. On the other hand, tryptophan and histidine
of C. albicans
were detected intracellularly at significant higher concen-
trations in response to farnesol (Fig. 3). Other metabo- We analysed the concentrations of identified intracellular
lites such as 2-aminobutyrate, ornithine, tyrosine, NADP/ metabolites of C. albicans cells grown in the presence or
NADPH, citramalate and cystathionine were present at absence of farnesol with PAPi software (Aggio et al.,
increased concentrations in samples harvested after 3-h 2010) to generate a comparative metabolic activity profile
incubation in the presence of farnesol, but not after 6-h (Fig. 4). The metabolic activities of C. albicans cells grow-
incubation. Cysteine, however, showed different trends in ing under the suppression of hyphae formation by farne-
samples taken at 3-h compared with those taken after 6-h sol were compared with those of cells growing without

FEMS Yeast Res 12 (2012) 879889 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
884 T.-L. Han et al.

in the presence of farnesol. On the other hand, one

pathway that was down-regulated in response to farnesol
was tryptophan metabolism, which is coincidently associ-
ated with the production of tyrosol, another quorum-
sensing molecule but one which stimulates hyphal
formation in C. albicans (Alem et al., 2006). Only the
biosynthesis of aromatic amino acids such as phenylalanine
and tyrosine appears to be down-regulated when cells were
exposed to farnesol for 3-h incubation, but they were up-
regulated after 6-h incubation in the presence of farnesol.
Interestingly, we observed that the majority of the path-
ways associated with central carbon metabolism and energy
metabolism were significantly up-regulated after 6 h of
exposure to farnesol. These pathways encompass the TCA
cycle, glycolysis/gluconeogenesis, glyoxylate/dicarboxylate
metabolism and oxidative phosphorylation (Fig. 4).

We have observed an overall up-regulation of metabolic
pathways from the central carbon and energy metabo-
lism when the hyphal formation of C. albicans was sup-
pressed by farnesol (Fig. 4). This result is in general
agreement with our previous study that demonstrated a
global down-regulation of central carbon metabolism
during early hyphal growth of C. albicans confirmed by
lower ATP formation in filamentous growing cells (Han
et al., 2012). There are several reports showing an
increase in mRNA and protein expression in C. albicans
Fig. 3. The relative intracellular metabolite concentrations for in response to farnesol (Cho et al., 2007; Shirtliff et al.,
Candida albicans cells grown in the presence or absence of farnesol 2009). In general, the up-regulated genes and proteins
after 3- and 6-h incubation. Metabolite concentrations are given are involved in the TCA cycle, glycolysis, gluconeogene-
relative to samples from cultures grown without farnesol, using a log2 sis, acetyl-CoA pathway, amino acid biosynthesis and
scale. Deeper red colours (positive values) indicate that metabolite nitrogen metabolism, or they are heat-shock proteins/
concentrations were increased in response to farnesol, while green
chaperones that protect the cells against environmental
shades (negative values) indicate decreased concentrations. The
colour key on the top right is superimposed with a histogram that
and oxidative stress. Although these studies suggest that
counts the relative concentrations of all the metabolites. Only the cells are likely to be more metabolically active in
metabolites for which there was a statistically significant change in response to farnesol, another study by Deveau & Hogan
concentration (P-value < 0.05) are shown. (2011) found, using Alamar Blue, that farnesol reduced
the metabolic activity of C. albicans during biofilm for-
mation. However, yeast cells, hyphae and biofilms are in
farnesol (filamentous growth). Of the 41 metabolic different metabolic states (Montserrat et al., 2009) and,
pathways that showed significant changes in metabolic indeed, proteomic profiles of cells during early hypha
activity in response to farnesol, 34 were up-regulated formation and biofilm formation appear to be signifi-
(Fig. 4). In particular, the pathways associated with nitro- cantly different (Montserrat et al., 2009). Nonetheless,
gen metabolism, acetyl-CoA biosynthesis, fatty acid filamentous cells present a much lower metabolic activity
metabolism (e.g. biosynthesis of unsaturated/saturated when compared to cells growing as yeast (Han et al.,
fatty acids, fatty acid elongation in mitochondria and 2012).
metabolism of linoleic acid), nicotinate and nicotinamide By comparing the metabolic pathways that were
metabolism and amino acid metabolism (e.g. alanine, affected by farnesol treatment with the 17 metabolic
b-alanine, aspartate, cysteine, glutamate, histidine, isoleu- pathways identified in our previous study as being closely
cine, leucine, lysine, methionine, D-alanine, valine) associated with the morphogenetic process (Han et al.,
appeared up-regulated after both 3- and 6-h incubation 2012), we were able to shortlist nine metabolic pathways

2012 Federation of European Microbiological Societies FEMS Yeast Res 12 (2012) 879889
Published by Blackwell Publishing Ltd. All rights reserved
The metabolic response of C. albicans to farnesol 885

Fig. 4. Candida albicans metabolic pathway

activities predicted from intracellular
metabolite profiling data. The metabolic
pathway activities relative to the samples from
cultures grown without farnesol are presented
using a log2 scale. Deeper red colours (positive
values) indicate pathways that were up-
regulated in response to farnesol, while green
shades (negative values) indicate a down-
regulation in metabolic activity. The colour key
on the top right is superimposed with a
histogram that counts all the pathway
activities. Only the metabolic pathways for
which there was a statistically significant
change in activity (P-value < 0.05) are shown.

that appear to be up-regulated under farnesol exposure C5-dibasic amino acids, histidine, glutamate, methionine,
(3 h and 6 h) and down-regulated under hyphae-inducing nicotinate/nicotinamide, nitrogen and acetyl-CoA biosyn-
conditions (Fig. 5). These pathways include the metabo- thesis. Indeed, Cho et al. (2007) analysed the global tran-
lism of alanine, b-alanine, D-alanine, aspartate, cysteine, scription profile of C. albicans in response to farnesol in

FEMS Yeast Res 12 (2012) 879889 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
886 T.-L. Han et al.

Fig. 5. Venn diagram to illustrate the metabolic pathways of Candida albicans that are up-regulated and down-regulated (labelled with
superscript a) when hyphal growth was suppressed by exposure to farnesol for 3 h or 6 h or under different filamentous growth conditions.
Metabolic pathways that respond to farnesol are derived from Fig. 4. Metabolic pathways that respond to hyphae-inducing conditions are
derived from Han et al. (2012). The figures in brackets indicate the number of metabolic pathways.

N-acetylglucosamine and proline medium. They found significantly up-regulated after 3 h incubation with
that genes encoding enzymes involved in amino acid farnesol and detected an up-regulation of central carbon
biosynthesis, acetyl-CoA biosynthesis and nitrogen metab- metabolism and energy metabolism in response to farne-
olism were up-regulated. Pathways involving the metabo- sol exposure for 6 h (Fig. 4). These metabolic responses
lism of butanoate, D-glutamine, D-glutamate and pyruvate are likely to be specific for farnesol because none of
were only found to be differentially expressed when these pathways appeared to be differentially expressed
C. albicans cells were grown in the presence of farnesol under the three hyphae-inducing conditions we tested
for 3 h and under hyphae-inducing conditions (Fig. 5). previously (Han et al., 2012). Farnesol, at high concen-
However, the metabolism of arginine, proline, glycine, trations, has been reported to induce apoptosis in other
threonine, purine and biosynthesis of isoleucine, leucine, fungi, bacteria, mammalian cells and C. albicans itself
valine were found to be differentially expressed after 6 h (Semighini et al., 2006; Fairn et al., 2007; Scheper et al.,
incubation in the presence of farnesol and under hyphae- 2008; Liu et al., 2009; Shirtliff et al., 2009), but also
inducing conditions (Fig. 5). Therefore, farnesol appears protects C. albicans from oxidative stress (Westwater
to suppress germ tube formation by up-regulating amino et al., 2005). This indicates some interesting features
acid metabolism, nitrogen metabolism, CoA biosynthesis about the effects of farnesol on the metabolism of
and nicotinate/nicotinamide metabolism. C. albicans. Not only does C. albicans produce the high-
Furthermore, farnesol seems to affect the lipid metab- est level of farnesol among Candida species (Weber
olism and central carbon metabolism independently et al., 2008), but it also exhibits the highest level of
from morphogenesis. We observed that the majority of tolerance to farnesol toxicity (Weber et al., 2010). This
the pathways associated with lipid metabolism were suggests that instead of just being a quorum-sensing

2012 Federation of European Microbiological Societies FEMS Yeast Res 12 (2012) 879889
Published by Blackwell Publishing Ltd. All rights reserved
The metabolic response of C. albicans to farnesol 887

molecule, there must be other advantages for C. albicans

to expend energy for the production of farnesol. We spec- Conclusions
ulate that the up-regulation of fatty acid metabolism by To our knowledge, this is the first metabolomic study
C. albicans in response to farnesol reflects the remodelling focused on the metabolic response of C. albicans during
of its membrane composition in order to minimise the the suppression of hyphal formation by farnesol. We have
harmful effects of farnesol at high concentrations. Owing shown that C. albicans globally up-regulates its cellular
to this unique metabolic response, farnesol may benefit metabolism in response to farnesol. By combining this
C. albicans by reducing the growth of other species in the metabolite profile with previous hyphae-inducing experi-
environment and to suppress macrophage attack against ments, a set of metabolic pathways associated with mor-
C. albicans by promoting oxidative stress and apoptosis of phogenesis has been identified. Moreover, there is an
macrophages (Abe et al., 2009). On the other hand, our up-regulation of lipid metabolism and subsequent up-
discovery of the up-regulation of central carbon metabo- regulation of central carbon metabolism upon exposure to
lism and oxidative phosphorylation by C. albicans in farnesol that is independent from morphogenesis. There-
response to farnesol is novel. Rozp dowska et al. (2011) fore, this metabolomic study provides new insights not
suggested that C. albicans does not possess an efficient only in our understanding of morphogenetic processes, but
glucose repression of respiration mechanism, and Veiga equally importantly on the farnesol-mediated redistribu-
et al. (2000) demonstrated that C. albicans can under- tion of metabolic fluxes in the central carbon metabolism.
take cyanide-resistant respiration and therefore is often
considered as a crabtree-negative yeast. Yamaguchi, 1974;
reported that yeast respiratory activity declined immedi- Acknowledgements
ately after cells were resuspended in a hyphae-inducing We thank M. Sabherwal for assisting with sample prepara-
medium. Land et al. (1975) showed that electron transfer tion; R. Aggio for data analysis assistance; and T. Liu for
through oxidative phosphorylation is required to maintain proof reading. This work was supported by the University
growth in the yeast morphology. The same authors also of Auckland Faculty of Science Research Funding and by a
showed that the yeast form exhibits higher levels of University of Auckland Doctoral Scholarship awarded to
respiration than hyphae, based on the oxygen consump- Ting-Li Han. The authors declare no conflict of interest.
tion rate, which is supported by our previous findings
where we found a higher pool of ATP in yeast cells com-
pared with hyphal cells (Han et al., 2012). In contrast,
Aoki & Ito-Kuwa (1982) showed that there is no differ- Abe S, Tsunashima R, Iijima R, Yamada T, Maruyama N,
ence in the respiratory activity between yeast and hyphal Hisajima T, Abe Y, Oshima H & Yamazaki M (2009)
forms and by blocking the respiratory chain at complex II Suppression of anti-Candida activity of macrophages by a
via thenoyltrifluoroacetone significantly inhibited the quorum-sensing molecule, farnesol, through induction of
hyphal development of C. albicans (Watanabe et al., oxidative stress. Microbiol Immunol 53: 323330.
2006). It is important to note, however, that unlike other Aggio RBM, Ruggiero K & Villas-Boas SG (2010) Pathway
studies that induced specific morphological forms by activity profiling (papi): From the metabolite profile to the
manipulating media components or pH, Aoki & Ito-Kuwa metabolic pathway activity. Bioinformatics 26: 29692976.
Albuquerque P & Casadevall A (2012) Quorum sensing in
(1982) compared the respiratory activity of yeast and
fungia review. Med Mycol 50: 19.
hyphal cells by manipulating the temperature. Therefore,
Alem MAS, Oteef MDY, Flowers TH & Douglas LJ (2006)
temperature-induced morphogenesis could have a differ-
Production of tyrosol by Candida albicans biofilms and its
ent mechanism than morphogenesis induced by other fac- role in quorum sensing and biofilm development. Eukaryot
tors. Nonetheless, thenoyltrifluoroacetone has been Cell 5: 17701779.
reported as a potent inhibitor of carboxylesterase (Zhang Aoki S & Ito-Kuwa S (1982) Respiration of Candida albicans
& Fariss, 2002). This drug not only suppresses mitochon- in relation to its morphogenesis. Plant Cell Physiol 23:
drial complex II activity but it may also disturb other 721726.
esterases such as acetylcholinesterase and lipase (Zhang & Cao YY, Cao YB, Xu Z, Ying K, Li Y, Xie Y, Zhu ZY,
Fariss, 2002). In addition, it is possible that the up- Chen WS & Jiang YY (2005) cDNA microarray analysis of
regulation of central carbon metabolism and oxidative differential gene expression in Candida albicans biofilm
phosphorylation is merely to fulfil the need to increase exposed to farnesol. Antimicrob Agents Chemother 49:
energy and cofactor production for metabolic activities 584589.
such as the up-regulation of amino acid biosynthesis for Cho T, Aoyama T, Toyoda M, Nakayama H, Chibana H &
rapid biomass increase during mid-exponential growth Kaminishi H (2007) Transcriptional changes in Candida
albicans genes by both farnesol and high cell density at an

FEMS Yeast Res 12 (2012) 879889 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
888 T.-L. Han et al.

early stage of morphogenesis in N-acetyl-D-glucosamine albicans bloodstream infections. Am J Infect Control

medium. Jpn J Med Mycol 48: 159167. 38: 7880.
Davis-Hanna A, Piispanen AE, Stateva LI & Hogan DA (2008) Mosel DD, Dumitru R, Hornby JM, Atkin AL & Nickerson
Farnesol and dodecanol effects on the Candida albicans KW (2005) Farnesol concentrations required to block germ
Ras1-cAMP signalling pathway and the regulation of tube formation in Candida albicans in the presence and
morphogenesis. Mol Microbiol 67: 4762. absence of serum. Appl Environ Microbiol 71: 49384940.
Deveau A & Hogan DA (2011) Linking quorum sensing Navarathna DHMLP, Hornby JM, Hoerrmann N, Parkhurst
regulation and biofilm formation by Candida albicans. AM, Duhamel GE & Nickerson KW (2005) Enhanced
Methods Mol Biol 692: 219233. pathogenicity of Candida albicans pre-treated with
Fairn GD, MacDonald K & McMaster CR (2007) A subinhibitory concentrations of fluconazole in a mouse
chemogenomic screen in Saccharomyces cerevisiae uncovers a model of disseminated candidiasis. J Antimicrob Chemother
primary role for the mitochondria in farnesol toxicity and 56: 11561159.
its regulation by the Pkc1 pathway. J Biol Chem 282: 4868 Pfaller MA & Diekema DJ (2007) Epidemiology of invasive
4874. candidiasis: a persistent public health problem. Clin
Hall RA, Turner KJ, Chaloupka J, Cottier F, de Sordi L, Microbiol Rev 20: 133163.
Sanglard D, Levin LR, Buck J & Muhlschlegel FA (2011) Reynolds R & Braude AI (1956) The filament inducing
The quorum-sensing molecules farnesol/homoserine lactone property of blood for Candida albicans; its nature and
and dodecanol operate via distinct modes of action in significance. Clin Res Proc 4: 40.
Candida albicans. Eukaryot Cell 10: 10341042. Rozp dowska E, Galafassi S, Johansson L, Hagman A, Piskur J
Han TL, Cannon RD & Villas-Boas SG (2011) The metabolic & Compagno C (2011) Candida albicans a pre-whole
basis of Candida albicans morphogenesis and quorum genome duplication yeast is predominantly aerobic and a
sensing. Fungal Genet Biol 48: 747763. poor ethanol producer. FEMS Yeast Res 11: 285291.
Han TL, Cannon RD & Villas-Boas SG (2012) Metabolome Sato T, Watanabe T, Mikami T & Matsumoto T (2004)
analysis during the morphological transition of Candida Farnesol, a morphogenetic autoregulatory substance in the
albicans. Metabolomics, doi: 10.1007/s11306-012-0416-6. dimorphic fungus Candida albicans, inhibits hyphae growth
Holmes AR & Shepherd MG (1987) Proline-induced germ- through suppression of a mitogen-activated protein kinase
tube formation in Candida albicans: role of proline uptake cascade. Biol Pharm Bull 27: 751752.
and nitrogen metabolism. J Gen Microbiol 133: 32193228. Scheper MA, Shirtliff ME, Meiller TF, Peters BM & Jabra-Rizk
Hornby JM, Jensen EC, Lisec AD, Tasto JJ, Jahnke B, MA (2008) Farnesol, a fungal quorum-sensing molecule
Shoemaker R, Dussault P & Nickerson KW (2001) Quorum triggers apoptosis in human oral squamous carcinoma cells.
sensing in the dimorphic fungus Candida albicans is Neoplasia 10: 954963.
mediated by farnesol. Appl Environ Microbiol 67: 29822992. Semighini CP, Hornby JM, Dumitru R, Nickerson KW &
Kebaara BW, Langford ML, Navarathna DHMLP, Dumitru R, Harris SD (2006) Farnesol-induced apoptosis in Aspergillus
Nickerson KW & Atkin AL (2008) Candida albicans Tup1 is nidulans reveals a possible mechanism for antagonistic
involved in farnesol-mediated inhibition of filamentous- interactions between fungi. Mol Microbiol 59: 753764.
growth induction. Eukaryot Cell 7: 980987. Shirtliff ME, Krom BP, Meijering RAM, Peters BM, Zhu J,
Land GA, McDonald WC, Stjernholm RL & Friedman L Scheper MA, Harris ML & Jabra-Rizk MA (2009) Farnesol-
(1975) Factors affecting filamentation in Candida albicans: induced apoptosis in Candida albicans. Antimicrob Agents
changes in respiratory activity of Candida albicans during Chemother 53: 23922401.
filamentation. Infect Immun 12: 119127. Smart KF, Aggio RBM, Van Houtte JR & Villas-Boas SG
Liu P, Deng B, Long CA & Min X (2009) Effect of farnesol on (2010) Analytical platform for metabolome analysis of
morphogenesis in the fungal pathogen Penicillium microbial cells using methyl chloroformate derivatization
expansum. Ann Microbiol 59: 3338. followed by gas chromatography-mass spectrometry. Nat
Maidan MM, Thevelein JM & Van Dijck P (2005) Carbon Protoc 5: 17091729.
source induced yeast-to-hypha transition in Candida Smith DA, Nicholls S, Morgan BA, Brown AJP & Quinn J
albicans is dependent on the presence of amino acids and (2004) A conserved stress-activated protein kinase regulates
on the G-protein-coupled receptor Gpr1. Biochem Soc Trans a core stress response in the human pathogen Candida
33: 291293. albicans. Mol Biol Cell 15: 41794190.
Montserrat M-G, Perumal P, Mekala S, Nombela C, Chaffin Veiga A, Arrabaca JD & Loureiro-Dias MC (2000) Cyanide-
WL & Gil C (2009) Proteomic analysis of cytoplasmic and resistant respiration is frequent, but confined to yeasts incapable
surface proteins from yeast cells, hyphae, and biofilms of of aerobic fermentation. FEMS Microbiol Lett 190: 9397.
Candida albicans. Proteomics 9: 22302252. Verduyn C, Postma E, Scheffers WA & Van Dijken JP (1992)
Moran C, Grussemeyer CA, Spalding JR, Benjamin DK Jr & Effect of benzoic acid on metabolic fluxes in yeasts: a
Reed SD (2010) Comparison of costs, length of stay, and continuous-culture study on the regulation of respiration
mortality associated with Candida glabrata and Candida and alcoholic fermentation. Yeast 8: 501517.

2012 Federation of European Microbiological Societies FEMS Yeast Res 12 (2012) 879889
Published by Blackwell Publishing Ltd. All rights reserved
The metabolic response of C. albicans to farnesol 889

Watanabe T, Ogasawara A, Mikami T & Matsumoto T (2006) macromolecules and respiratory activity. J Gen Appl
Hyphal formation of Candida albicans is controlled by Microbiol 20: 8799.
electron transfer system. Biochem Biophys Res Commun 348: Zhang JG & Fariss MW (2002) Thenoyltrifluoroacetone, a
206211. potent inhibitor of carboxylesterase activity. Biochem
Weber K, Sohr R, Schulz B, Fleischhacker M & Ruhnke M Pharmacol 63: 751754.
(2008) Secretion of E, E-farnesol and biofilm formation in
eight different Candida species. Antimicrob Agents Chemother Supporting Information
52: 18591861.
Weber K, Schulz B & Ruhnke M (2010) The quorum-sensing
Additional Supporting Information may be found in the
molecule E, E-farnesol Its variable secretion and its impact
online version of this article:
on the growth and metabolism of Candida species. Yeast 27:
727739. Table S1. The relative normalized abundances of identi-
Westwater C, Balish E & Schofield DA (2005) Candida fied metabolites over all culture conditions.
albicans-conditioned medium protects yeast cells from
oxidative stress: a possible link between quorum sensing and Please note: Wiley-Blackwell is not responsible for the
oxidative stress resistance. Eukaryot Cell 4: 1654016610. content or functionality of any supporting materials sup-
Whiteway M & Oberholzer U (2004) Candida morphogenesis and plied by the authors. Any queries (other than missing
host-pathogen interactions. Curr Opin Microbiol 7: 350357. material) should be directed to the corresponding author
Wickham H (2009) ggplot2: Elegant Graphics for Data Analysis. for the article.
Springer, New York.
Yamaguchi H (1974) Dimorphism in Candida albicans. I.
Morphology dependent changes in cellular content of

FEMS Yeast Res 12 (2012) 879889 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved