In vitro synergistic effect of farnesol and human gingival cells against Candida albicans

YeastYeast 2006; 23: 673–687.Published online in Wiley InterScience(www.interscience.wiley.com) DOI: 10.1002/yea.1389

Research Article

In vitro synergistic effect of farnesol and humangingival cells against Candida albicansSaid Saidi, Cyril Luitaud and Mahmoud Rouabhia*Faculte de Medecine Dentaire, Groupe de Recherche en Ecologie Buccale, Universite Laval, Quebec City, Quebec, Canada G1K 7P4

*Correspondence to:Mahmoud Rouabhia, Faculte deMedecine Dentaire, Pavillon deMedecine Dentaire, Local 1728,Universite Laval, Quebec City,Quebec, Canada G1K 7P4.E-mail: [email protected]

Received: 3 February 2006Accepted: 22 May 2006

AbstractFarnesol prevents the germination of yeast cells into mycelia, a fact that may be usefulin eliminating C. albicans pathogenicity. Given the clinical potential of farnesol, itsimpact on C. albicans and host cells merited further investigation. We thus studied theeffect of farnesol on C. albicans growth and filamentation and on gingival epithelialcells and fibroblasts and the synergistic effect of both gingival cells and farnesolon C. albicans filamentation. Repeated additions of farnesol reduced the growth ofC. albicans. Farnesol was also effective at reducing C. albicans germ tube formation.While farnesol inhibited germ tube formation under the conditions tested, it was mosteffective at inhibiting C. albicans filamentation when added to the culture mediumat the same time as the serum. Farnesol also had an effect on gingival cells. In aserum-free medium, farnesol reduced fibroblast adhesion and proliferation, promotedepithelial cell differentiation and reduced proliferation up to 48 h post-treatment.These effects were not seen in the presence of serum. When C. albicans, farnesoland gingival cells were present in the same culture, significantly greater inhibitionof the yeast-to-hyphal transition was observed than germ tube inhibition in culturescontaining only C. albicans and farnesol, suggesting a synergistic effect between thegingival cells and farnesol in inhibiting the transition. Overall, the data suggest thatfarnesol is effective against C. albicans and may have an effect on host cells at certainconcentrations. Copyright 2006 John Wiley & Sons, Ltd.

Keywords: epithelial cells; Candida albicans; candidiasis; oral mucosa; farnesol;antifungal

Introduction

The oral cavity is the gateway for a wide arrayof antigenic challenges (Dongari-Bagtzoglou andFidel, 2005; Smith and Taubman, 1992) and alsohas a variety of niches that allow the estab-lishment of unique microbial communities. Manymembers of the oral microbiota, including Can-dida albicans, have a symbiotic relationship withthe host. C. albicans frequently resides in theoral cavity without causing disease (Odds, 1988),but asymptomatic carriage may place some indi-viduals at higher risk of complications fromyeast infections as they age (Fanello et al., 2006)or if they are immunosuppressed (Gottfredssonet al., 1999; Sanchez-Vargas et al., 2005). Amongpatients initially asymptomatic for C. albicans

infection, clinical thrush only develops in thosewho persistently carry C. albicans prior to devel-oping symptoms (Fetter et al., 1993).

The oral mucosal epithelium is the most impor-tant barrier to physical, microbial and chemicalagents that may cause local cell injury (Preslandand Dale, 2000; Presland and Jurevic, 2002; Sel-varatnam et al., 2001). The oral mucosal epithe-lium is involved in the pro-inflammatory processthrough the production of cytokines, either consti-tutively or following a variety of stimuli (Andrianet al., 2005; Bos et al., 1993; Rouabhia et al., 2002,2005), which suggests that they play an active rolein controlling oral infections and maintaining thesymbiotic relationship with microbial agents. Thebalance between the various areas in the mouthinfluences the stability and integrity of oral mucosal

Copyright 2006 John Wiley & Sons, Ltd.

674 S. Saidi, C. Luitaud and M. Rouabhia

tissue, which may be disrupted by various stressfactors, such as transplantation, medications, etc.(Blijlevens et al., 2005, 2004; Presland and Jure-vic, 2002), or by a decrease in the capacity ofmucosal tissue to maintain health, resulting inhost infections such as oral candidiasis. The yeastspecies most commonly isolated from the oral cav-ity is by far C. albicans (Dongari-Bagtzoglou andFidel, 2005). While a median carrier rate of 38.1%has been observed for C. albicans in community-dwelling outpatients (Odds, 1988), a rate reach-ing 78% has been reported in hospitalized elderlypatients (Wilkieson et al., 1991) and is even higherin HIV-infected individuals (Schmidt-Westhausenet al., 2004). C. albicans is the primary cause ofnosocomial fungal infections leading to candidiasis(Odds, 1988).

Candidiasis is treated with antifungal agents suchas polyene antibiotics, flucytosine and azoles (Bus-tamante, 2005; Kuriyama et al., 2005). Currentlyavailable antifungals are limited in both numberand effectiveness. This may be due to a lack ofspecific targets that make it possible to discrim-inate between eukaryotic fungal cells and mam-malian cells. In addition to their efficacy, the clini-cal usefulness of these antifungals is hampered bythe undesirable side-effects often associated withthem and by the emergence of resistance to them(Kuriyama et al., 2005; Pfaller et al., 2005). Anti-fungal drug resistance is fast becoming a majorproblem in the growing number of immunocom-promised persons and has resulted in a dramaticincrease in the incidence of opportunistic and sys-temic fungal infections. The rise in resistance andthe changes in the spectrum of Candida infectionshave generated greater interest in the developmentof new antifungal drugs.

Cell–cell signalling, particularly quorum sens-ing, has been one of the focuses of microbiologicalresearch over the past decade. It has been demon-strated that quorum-sensing products, such as far-nesol, are essential for bacterial biofilm formationand that a threshold concentration triggers biofilmformation (Cao et al., 2005; Miller and Bassler,2001; Ramage et al., 2002). Farnesol (trans, trans-3,7,11-trimethyl-2,6,10-dodecatrien-1-ol) is a ses-quiterpene/isoprenoid originally isolated fromplants, although it is now produced synthetically.It is a key intermediate in the mevalonate pathway(Grunler et al., 1994). Farnesol is produced in cellsby the enzymatic dephosphorylation of farnesyl

pyrophosphate (FPP). FPP plays an importantrole as a precursor of protein prenylation, apost-translation modification of proteins (Edwardsand Ericsson, 1999). Known farnesylated proteinsinclude Ras and Ras-related GTP-binding proteins(G proteins), which control cell growth, differen-tiation, proliferation and survival. Protein farne-syltransferase (Ftase) is involved in Ras/mitogen-activated protein kinase signalling pathways inmammalian systems (Sattler and Tamanoi, 1996;Campbell et al., 1998). Ftase inhibitors (FTIs)block a wide range of human tumours, pointingto the importance of farnesylation in the growthof cancer cells and transformation of their phe-notypes (Sattler and Tamanoi, 1996; Gibbs andOliff, 1997). These inhibitors also affect the cellcycle and induce apoptosis in transformed cell lines(Suzuki et al., 1998; Gibbs et al., 1997). A widerange of organisms, including yeast, produce Ftase,which recognizes the C-terminal CAAX motif insubstrate proteins such as Ras and the γ -subunits ofyeast heterotrimeric G-protein and transducin. Pro-tein farnesylation plays a critical role in a varietyof cellular processes (Sattler and Tamanoi, 1996;Pei et al., 1998). In recent years, interest in far-nesol has focused on its anticancer properties, andit has been shown to be an effective chemopre-ventive agent (Wargovich et al., 2000). It inhibitsthe growth of cancer cell lines (Hudes et al., 2000;Poon et al., 1996) and dietary supplementation withfarnesol has been shown to be effective in reducingpancreatic cancer growth (Burke et al., 1997).

Above a certain threshold level, farnesol preventsthe yeast-to-hyphal transition of (Ramage et al.,2002) and biofilm formation by C. albicans (Enjal-bert and Whiteway, 2005; Hornby et al., 2003;Mosel et al., 2005) and is therefore of great inter-est as a potential antifungal drug for the preventionof Candida infections. As antifungal drugs are incontact with both the host tissue and the infectiousagent, and as farnesol shows promise as an anti-fungal drug, we decided to examine its effect onC. albicans and on the oral mucosa, using gingivalfibroblast and epithelial cell cultures.

Experimental procedures

Oral epithelial cell isolation and cultures

Small samples of palatal mucosa were collectedfrom gingival graft patients following informed

Copyright 2006 John Wiley & Sons, Ltd. Yeast 2006; 23: 673–687.DOI: 10.1002/yea

Anti-fungal activity of farnesol 675

consent. The biopsies were treated with ther-molysin (500 µg/ml) to separate the epitheliumfrom the lamina propria (Rouabhia et al., 1994).Epithelial cell suspensions were obtained by treat-ing the tissue with a 0.05% trypsin/0.01 M EDTAsolution. Freshly isolated epithelial cells (9 ×103 cells/cm2) were cultured in a 3 : 1 mixtureof a Dulbecco–Vogt’s modified Eagle’s (DME)medium and Ham’s F12 (H) (Invitrogen Life Tech-nologies, Burlington, ON, Canada) supplementedwith 24.3 µg/ml adenine, 10 µg/ml human epider-mal growth factor (Chiron Corp., Emeryville, CA,USA), 0.4 µg/ml hydrocortisone (Calbiochem, LaJolla, CA, USA), 5 µg/ml bovine insulin, 5 µg/mlhuman transferrin, 2 × 10−9 M 3,3′,5′-triiodo-L-thyronine, 10−10 M cholera toxin (Schwarz/Mann,Cleveland, OH, USA), 100 U/ml penicillin, 25 µg/ml gentamicin (Schering, Pointe-Claire, QC,Canada) and 10% fetal calf serum (NCS, fetal cloneII; Hyclone, Logan, UT, USA). Following charac-terization, the oral epithelial cells were cultured andused at passage two for this study.

Oral fibroblast isolation and cultures

Following separation from the epithelium, the lam-ina propria was placed in a collagenase P solu-tion (0.125 U/ml; Boehringer-Mannheim, Laval,QC, Canada) for 45 min at 37 ◦C to extract thefibroblasts (Rouabhia et al., 1994). Isolated cells(2 × 106) were seeded in 75 cm2 flasks (Falcon,Becton-Dickinson, Cockeysville, MD, USA) andgrown in DME medium (Invitrogen Life Tech-nologies) containing 10% fetal calf serum (Invit-rogen Life Technologies), 100 IU/ml penicillin G,25 µg/ml streptomycin and 0.5 µg/ml fungizone.The fibroblasts were used once the cultures hadreached 90% confluence.

C. albicans cultures

A previously characterized clinical C. albicansisolate (Candida-associated stomatitis; Rouabhiaet al., 2002) was cultured on Sabouraud dextroseagar plates (SAB; Becton-Dickinson) at 30 ◦C. Forthe C. albicans suspensions, one colony was usedto inoculate 10 ml phytone–peptone (PP) medium(Becton-Dickinson) supplemented with 0.1% glu-cose, pH < 6. The culture was grown to the station-ary phase for 18 h at 37 ◦C in a shaking water bath.The blastoconidia were collected, washed with

PBS, counted using a haemocytometer (Rouabhiaet al., 2002), adjusted to 107/ml and used to infectthe various cultures.

Effect of farnesol on C. albicans

C. albicans growth in the presence of farnesol

C. albicans was seeded on SAB plates, which wereincubated at 30 ◦C. One colony was used to inoc-ulate 10 ml PP supplemented with 0.1% glucoseto produce suspensions. The culture was grownto the stationary phase (18 h at 30 ◦C) in a shak-ing water bath. The blastoconidia were collected,washed with PBS and counted using a haemocy-tometer and Trypan blue dye exclusion (Rouab-hia et al., 2002). A 4 M stock solution of far-nesol (Sigma-Aldrich, Oakville, ON, Canada) wasdiluted to the appropriate working concentrations(10–300 µM) in methanol, as previously reported(Cao et al., 2005; Ramage et al., 2002). The con-centration of methanol was under 1.5% (v/v) in thecultures. For the antifungal assay, various concen-trations of farnesol (10–300 µM) were added to theC. albicans cultures (106 cells, 5 ml PP medium),which were incubated for 2, 4, 8 and 24 h at 30 ◦C,pH < 6. A C. albicans culture supplemented with1.5% methanol alone was used as a negative con-trol. A C. albicans culture (no growth) without far-nesol but with 0.5 µg/ml fungizone (Amphotericin-B, Sigma-Aldrich) was used as a positive con-trol. Following each incubation period, 3 × 50 µlaliquots of each culture were diluted in 450 µl PBSand the cells were counted using a haemocytome-ter. The results are reported as the means ± SD ofsix different experiments.

C. albicans cultures in the presence of serum andfarnesol

To determine the effect of farnesol on the yeast-to-hyphal transition, C. albicans (106 cells) wasgrown in 5 ml PP containing 20% FCS. Farnesolwas added at various concentrations (10, 50, 100,150 and 300 µM), either at the same time as the FCSor 30 min and 120 min later. Control C. albicanscultures with and without FCS and with 1.5%methanol alone were included. The cultures wereincubated at 30 ◦C, observed microscopically at 2,4 and 6 h, and photographed to record the yeast-to-hyphal transition. To calculate the percentage oftransition, three aliquots from each culture were



used to determine the number of blastospores andhyphae, using a haemocytometer and an opticalmicroscope, as described previously (Rouabhiaet al., 2002). The percentage of transition wasdetermined using the following formula:

Number of hyphae

Number of blastospores + hyphae× 100

The results are the means ± SD of six differentexperiments.

Effect of farnesol on gingival cell growth

Fibroblasts were seeded in six-well plates (5 ×104 cells/well) in DME supplemented with 5%FCS. Epithelial cells were seeded in six-well plates(3 × 105 cells/well) in DMEH supplemented with10% FCS. Farnesol (10–300 µM) was added whenthe cultures reached 80% confluence. Epithelialcells and fibroblasts cultured in 1.5% methanol-supplemented medium were used as negativecontrols. Cell morphology and detachment wereassessed using an optical microscope and pho-torecorded. Following the incubation period, thecells were treated with a 0.05% trypsin/0.01 EDTAsolution for 5 min at 37 ◦C to detach them fromthe culture plates. They were washed twice withculture medium, and the cell numbers and viabilitywere assessed by Trypan blue exclusion (Moste-faoui et al., 2002). Cell suspensions (100 µl) weremixed with the same volume of Trypan blue solu-tion and incubated on ice for 5 min. Cell numbersand viability were determined in triplicate for eachsuspension. Results are reported as the means ± SDof six different experiments.

Effect of gingival cells and farnesol on C. albicansgrowth

Fibroblasts (5 × 104 cells/well) and epithelial cells(3 × 105 cells/well) were seeded separately in six-well plates and cultured in FCS-supplemented.When the cultures reached 80% confluence, themedium was replaced with fresh medium with orwithout 20% FCS and containing no antibiotics orantifungal. Immediately following the change inmedium, farnesol was added at various concentra-tions (50–300 µM) in the presence or absence ofC. albicans (106 cells/well). Cells cultured in 1.5%methanol-supplemented medium (without farnesol)were used as controls. Gingival cell morphology

and detachment as well as C. albicans transitionwere assessed using an optical microscope and pho-torecorded. Culture supernatants were collected at2, 4, 8 and 24 h to determine the ratio of blas-tospores to hyphae. The fibroblasts and epithelialcells were detached from the culture plates andused to determine cell numbers and viability byTrypan blue exclusion, as described above. Theresults are reported as the means ± SD for six dif-ferent experiments.

Statistical analyses

Each experiment was performed at least four times.Representative photographs are shown. The exper-imental values are given as means ± SD. The sta-tistical significance of the differences between thecontrol and test values was evaluated using aone-way ANOVA. Subsequent comparisons wereperformed using Tukey’s method. Normality andvariance assumptions were verified using theShapiro–Wilk and Brown and Forsythe tests,respectively; p values were declared significant at0.05. The data were analysed using the SAS version8.2 statistical package (SAS Institute Inc., Cary,NC, USA).

Results

Farnesol reduced C. albicans growth

C. albicans cultures were incubated with farnesolfor 2, 4, 8 or 24 h to determine whether farnesoladversely affected growth. The control experimentsshowed that 1.5% (v/v) methanol had no effecton C. albicans viability or growth. Only highconcentrations of farnesol (100, 150 and 300 µM)significantly reduced growth (C. albicans culturedalone) over the 24 h contact period (Table 1).The number of C. albicans cells in the farnesol-containing culture remained elevated. In the clinicalsituation, following infection, a patient has toreceive antibiotics or an antifungal several timesa day for several days. This raised the questionas to whether repeated treatments with farnesol(as opposed to a single treatment) would inhibitgrowth. C. albicans cultures were therefore grownin the presence of farnesol (added at time zero) andidentical concentrations of farnesol were added at2, 4 and 8 h. As shown in Table 2, the repeatedtreatments with farnesol significantly inhibited the



Table 1. C. albicans cultures were treated with farnesol at different concentrations. Farnesol was added once at the sametime of C. albicans seeding. Growth of C. albicans was determined at different time points

Contact No AmphotericinNumber × 106 of C. abicans after treatment with farnesola

periods (h) treatment B (0.5 µg/ml) 10 µM 50 µM 100 µM 150 µM 300 µM

2 4 ± 0.4 1.23 ± 0.3 3 ± 0.3 2 ± 0.1∗ 2 ± 0.15∗ 2 ± 0.16∗ 2 ± 0.15∗4 15 ± 2 1.6 ± 0.9 11 ± 1∗ 5 ± 2∗ 5 ± 1∗ 4 ± 0.3∗ 4 ± 0.3∗8 110 ± 13 1.5 ± 0.3 85 ± 9∗ 50 ± 13∗ 46 ± 7∗ 32 ± 10∗ 27 ± 3∗24 525 ± 25 1.3 ± 0.4 462 ± 50 392 ± 60 340 ± 20∗ 261 ± 40∗ 242 ± 33∗

∗ p < 0.01 when compared to the non-treated C. albicans cultures.a Values are given as means ± SD of six separate experiments.

Table 2. C. albicans were treated with farnesol at different concentrations. Treatment was repeated every 2 h up to 8 h

ContactMultiple treatment with farnesola

periods (h)¶ No treatment 10 µM 50 µM 100 µM 150 µM 300 µM

2 4 ± 0.3 2.6 ± 0.2 2 ± 0.2∗ 1.5 ± 2∗ 1.4 ± 0.1∗ 1.5 ± 0.08∗4 15 ± 3 9 ± 2∗ 5 ± 1∗ 4 ± 2∗ 3 ± 0.6∗ 2.9 ± 0.4∗8 111 ± 18 58 ± 12∗ 25 ± 5∗ 23 ± 6∗ 13 ± 4∗ 14 ± 4∗24 525 ± 34 158 ± 22∗ 140 ± 18∗ 118 ± 15∗ 100 ± 16∗ 124 ± 32∗

¶ Farnesol was added at 0, 2, 4 and 8 h.∗ p < 0.01 when compared to the non-treated C. albicans cultures.a Values are given as means ± SD of six separate experiments.

growth of C. albicans, with the maximum numberof cells decreasing from approximately 240 × 106

after one treatment with farnesol (Table 1) toapproximately 120 × 106 after several treatments(300 µM) (Table 2). With the repeated treatments,significant growth inhibition was obtained at 8 and24 h with both high and low concentrations offarnesol.

Farnesol-inhibited germ tube formationby C. albicans

C. albicans cultures were supplemented with 20%FCS. Farnesol was added at the same time as theFCS and 30 min and 120 min later the FCS wasadded. When the farnesol was added at the sametime as the serum, germ tube formation was signif-icantly reduced as early as 4 h later (Figure 1Ab)compared to the untreated culture (Figure 1Aa).The inhibition of germ tube formation was signif-icant (p < 0.05) at high concentrations of farnesol(150 and 300 µM; Figure 1B). When added 30 minand 120 min after the serum, farnesol also had asignificant inhibitory effect on germ tube forma-tion. However, the effect was not as significant aswhen farnesol was added at the same time as the

serum. These results demonstrate that farnesol washighly effective against the yeast-to-hyphal transi-tion of C. albicans. This effect was observed evenwhen the farnesol was added after germ tube for-mation had begun.

Farnesol detached fibroblasts and reducedgrowth

Farnesol was used to treat fibroblast culturesfor 24 h to determine whether one dose of far-nesol would adversely affect gingival fibroblastgrowth. The control experiments showed that 1.5%(v/v) methanol had no effect on fibroblast via-bility or growth. As shown in Figure 2, farnesolhad no effect on cell morphology up to a con-centration of 50 µM. The fibroblasts were elon-gated with a dense nucleus and small cytoplasm(Figure 2A). However, at high concentrations, thecells began to detach and degrade. No fibrob-lasts remained attached when they were treatedwith 100 or 300 µM farnesol. This effect was con-firmed by cell counts. As seen in Figure 2B, asignificant (p < 0.001) decrease in the numberof fibroblasts was observed in cultures incubated



A

Ctrl 10 50 100 150 3000

10

20

30

40

50

60

70

80

90

100

T-0T-30T-120

Farnesol (µM)

% o

f hy

phe

form

s of

C. a

lbic

ans

B

Figure 1. Effect of farnesol on the yeast-to-hyphal transition of C. albicans. C. albicans was cultured in SAB mediumcontaining 20% FCS, either without farnesol or with farnesol at various concentrations (10, 50, 100, 150 and 300 µM) at30 ◦C for 4 h. Farnesol was added either at the same time as the serum or 30 min and 120 min later. After 4 h, the cultureswere observed under an inverted microscope and photorecorded. The numbers of yeast and hyphal forms were counted.The percentage of hyphae was obtained by dividing the number of hyphae by the total number of cells (blastospores andhyphae) in each culture. The mean relative values for six separate experiments are shown in (B). (A) Photographs takenfrom C. albicans cultured in the presence of serum with farnesol (b) and without farnesol (a) added at the same time as theserum. The levels of significance for the percentages of yeast-to-hyphal transition in the presence of farnesol compared tothose in the absence of farnesol are p < 0.05 (∗), p < 0.01 (∗∗) and p < 0.001 (∗∗∗)

with 100 and 300 µM farnesol. To specificallydetermine the minimal concentration at which far-nesol affected the fibroblasts, they were culturedin the presence of 60, 70, 80 or 90 µM far-nesol. As can be seen in Figure 2C, 50, 60 and

70 µM farnesol resulted in a limited decrease inthe number of fibroblasts after 24 h. However,at higher concentrations (80 and 90 µM), farnesolhad a significant inhibitory effect on fibroblastgrowth.



A

Ctrl 10 50 100 150 3000

10

20

30

40

50

60

70

80

BP< 0.001

Num

ber

of o

ral f

ibro

blas

ts x

106

Ctrl 50 60 70 80 90 1000

10

20

30

40

50

60

70

80

Farnesol (µM)

C

P< 0.001

P< 0.05

Figure 2. Effect of farnesol on fibroblast morphology and growth. Human gingival fibroblasts were seeded in six-well platesand cultured in DMEM supplemented with 20% FCS in a 5% CO2 humid atmosphere at 37 ◦C. When the cells reached 80%confluence, the cultures were pulsed with various concentrations of farnesol. The cultures were then observed at differenttime points and photographs were taken 24 h post-contact (A). Selected photos of cultures supplemented with variousconcentrations of farnesol, as indicated. These photos are representative of six separate experiments. Twenty-four hourslater, the fibroblasts were detached and used to assess cell growth by Trypan blue exclusion (B,C). The mean relativevalues ± SD for six separate experiments are shown. The levels of significance for fibroblast growth in the presence offarnesol compared to growth in the absence of farnesol were p < 0.05 (∗) and p < 0.01 (∗∗)



Farnesol-promoted epithelial cell differentiationand reduced growth when used over a longperiod

To investigate the effect of farnesol on epithelialcells, human gingival epithelial cells were culturedin the presence of farnesol for 24 h in a mediumcontaining 10% FCS. Cell morphology assessedusing an inverted microscope revealed that, evenat elevated concentrations (150 and 300 µM), far-nesol did not modify epithelial cell morphology.Figure 3A shows that the morphology of epithelialcells treated with 300 µM farnesol was similar tothat of the controls after 24 h. In addition, there wasno significant reduction in cell numbers of farnesol-supplemented cultures compared to control cultures(Figure 3B).

To mimic clinical situations where antibioticsor antifungal may be used for longer period oftime, we determined whether longer periods of con-tact with farnesol affect cell morphology. Gingivalepithelial cells were cultured for 48 h in the pres-ence of farnesol. As shown in Figure 3A, the mor-phology of cells in contact with 300 µM farnesol for48 h developed faint nuclei, large cytoplasms andvacuoles. Cell morphology was also affected at 100and 150 µM farnesol (data not shown). These mor-phological changes were accompanied by lowercell numbers. As shown in Figure 3B, 48 h in thepresence of 100, 150 or 300 µM farnesol signifi-cantly reduced epithelial cell growth.

Effect of farnesol and C. albicans on fibroblastmorphology and proliferation

To mimic in vivo conditions, we tested the effectof farnesol on fibroblasts in the presence of

C. albicans. When fibroblasts were cultured in theabsence of serum, both C. albicans and farnesolaffected cell adhesion and morphology (Figure 4).These effects were observed at 2, 4 and 8 h. Celldetachment was significant after 2 h with 90 µM

farnesol and after 4 h with 50 µM farnesol, com-pared to fibroblasts cultured in the presence ofC. albicans without farnesol (data not shown).After 24 h, no fibroblasts remained in the culturescontaining 50 or 90 µM farnesol (Figure 4). Whenthe fibroblasts were cultured in the presence of 20%FCS, they continued to adhere and display normalmorphology up to 24 h, even in the presence of anelevated concentration of farnesol (90 µM).

The second parameter assessed was the pres-ence or absence of C. albicans hyphae. The yeast-to-hyphal transition of C. albicans was inhibitedin all cultures, at all times and at all farnesolconcentrations, even with 20% FCS (Figure 4),unlike cultures with no farnesol (Figure 1). Toconfirm these microscopic observations, the num-ber of fibroblasts and the percentage of hyphaewere quantified at each time point. Our results(Table 3) confirmed that farnesol had a signifi-cant impact on fibroblast morphology and attach-ment. Viable cells were found only in culturestreated with low concentrations of farnesol. More-over, even at a concentration of 10 µM, the numberof viable fibroblasts had significantly (p < 0.01)decreased by 24 h. C. albicans alone had no signif-icant effect on fibroblast morphology and numbers.When fibroblasts were cultured in the presenceof 20% FCS, farnesol had no effect on morphol-ogy or viability (Table 3). When cultured in thepresence of C. albicans without farnesol, no effecton cell numbers was observed at early and late

Table 3. Number of viable gingival fibroblasts obtained after there culture in the presence of C. albicans and farnesol atdifferent concentrations

Fibroblast numbers at different time pointsa

Time ofFarnesol (µM) without serum Farnesol (µM) with 20% serum

reaction (h) 0 10 50 70 80 90 0 10 50 70 80 90

2 52 ± 11 29 ± 12∗ 0.3∗ 0∗ 0∗ 0∗ 52 ± 14 46 ± 12 50 ± 6 50 ± 15 51 ± 22 53 ± 234 63 ± 4 30 ± 7∗ 0∗ 0∗ 0∗ 0∗ 62 ± 2 58 ± 4 63 ± 12 65 ± 7 57 ± 9 51 ± 158 30 ± 1 26 ± 9∗ 0∗ 0∗ 0∗ 0∗ 63 ± 6 56 ± 10 56 ± 13 42 ± 3 30 ± 9 50 ± 624 43 ± 10 0.7∗ 0∗ 0∗ 0∗ 0∗ 67 ± 3 58 ± 3 57 ± 3 49 ± 2 49 ± 11 64 ± 2

a Values are given as means ± SD of three separate experiments.∗ Values are statistically significant (p < 0.01) when compared to the controls (farnesol-untreated cultures).



ACtrl 300 µM

24 h

48 h

Ctrl 10 50 100 150 3000

1

2

3 24h48h

Ora

l epi

thel

ial c

ell n

umbe

r x

106

Farnesol (µM)

B

Figure 3. Effect of farnesol on human gingival epithelial cell morphology and growth. Human gingival epithelial cells wereseeded in six-well plates and cultured in DMEM supplemented with 20% FCS in a 5% CO2 humid atmosphere at 37 ◦C.When the cells reached 80% confluence, the cultures were pulsed with farnesol (10, 50, 100, 150 and 300 µM), observedat different time points and photographed (magnification 300×). At 24 and 48 h post-contact (A), the cells were detachedand used to assess cell growth by Trypan blue exclusion. The mean relative values for six separate experiments are shown(B). The levels of significance for the epithelial cell growth in the presence of farnesol compared to growth in the absenceof farnesol were p < 0.05 (∗) and p < 0.01 (∗∗)



50 µM

0 µM

90 µM

Without Serum With Serum

Figure 4. Fibroblast and C. albicans morphology following co-culture in the presence of farnesol. Gingival fibroblasts werecultured as reported above. At 80% confluence, 106 C. albicans cells were added and the cultures were pulsed with farnesolat various concentrations in the presence or absence of 20% FCS. Fibroblast and C. albicans morphology and adherencewere observed at different time points and photographed after 24 h. The photographs are representative of six separateexperiments. Arrows show C. albicans adhering to the fibroblasts or the plastic culture dish (magnification, 300×)

contact periods. We determined the percentage ofC. albicans that shifted from the blastospore tothe hyphal form under the same culture condi-tions. Table 4 shows that, in the absence of FCS,C. albicans filamentation was almost totally absentin the presence of farnesol. However, when thecultures were supplemented with 20% FCS, thefilamentation of C. albicans varied with the cul-ture conditions. Indeed, a significant decrease inC. albicans filamentation was observed as earlyas 2 h after contact with farnesol, when the per-centage of hyphae was lower with 90 µM farnesolthan with 10 µM. At 4, 8 and 24 h, the percent-age of hyphae in cultures containing farnesol was

very low compared to that observed with control(no farnesol) cultures. Overall, our findings sug-gest that serum was required to prevent the effectof farnesol on gingival fibroblasts. In the presenceof serum, the sole effect of farnesol was to preventC. albicans filamentation.

Synergistic effect of farnesol and epithelial cellson C. albicans filamentation

When gingival epithelial cells were cultured inthe presence of C. albicans and farnesol but inthe absence of serum, farnesol (150 and 300 µM)promoted cell differentiation beginning 2 h post-contact. At 4 and 8 h, farnesol (150 µM) caused



Table 4. Percentage of C. albicans transition from blastospore to hyphae form when cultured in the presence of gingivalfibroblasts and farnesol at different concentrations

Hyphae form of C. albicansa


reaction (h) 0 10 50 70 80 90 0 10 50 70 80 90

2 0.16 0.14 0 0 0 0 51 ± 20 7 ± 5∗ 6 ± 2∗ 3 ± 1∗ 3 ± 2.2∗ 2 ± 0.3∗4 0 0 0 0 0 0 51 ± 2 2 ± 0.4∗ 2 ± 1.2∗ 6 ± 2∗ 2 ± 0.1∗ 1 ± 0.9∗8 0 0 0 0 0 0 52 ± 1 4 ± 1.0∗ 1 ± 1.3∗ 1 ± 0.9∗ 0∗ 0∗24 0 0 0 0 0 0 6.7 ± 3 2 ± 0.4∗ 1 ± 0.1∗ 0.9 ± 2∗ 0∗ 0∗


Table 5. Number of viable gingival epithelial cells obtained after their culture in the presence of C. albicans and farnesol atdifferent concentrations

Epithelial cell numbers at different time pointsa


reaction (h) 0 10 50 100 150 300 0 10 50 100 150 300

2 54 ± 11 51 ± 19∗ 47 ± 14∗ 29 ± 15∗ 10 ± 4∗ 7 ± 4∗ 60 ± 14 61 ± 12 54 ± 11 67 ± 20 51 ± 13 60 ± 74 52 ± 4 49 ± 7∗ 40 ± 13∗ 10 ± 6∗ 0∗ 0∗ 62 ± 2 58 ± 16 57 ± 12 55 ± 7 57 ± 9 51 ± 158 53 ± 1 36 ± 9∗ 31 ± 10∗ 4 ± 1∗ 0∗ 0∗ 63 ± 6 56 ± 10 56 ± 13 58 ± 3 59 ± 9 57 ± 624 44 ± 10 20 ± 7∗ 13 ± 7∗ 0∗ 0∗ 0∗ 70 ± 3 67 ± 3 63 ± 3 65 ± 2 64 ± 11 64 ± 2


significant cell detachment (data not shown). At24 h, both 150 and 300 µM farnesol affected thecell morphology and detachment of epithelial cells(Figure 5). The effect of farnesol on the epithelialcells was inhibited by 20% FCS. After 24 h, theepithelial cells were small and cuboidal in shapewith a small nucleus and cytoplasm. On the otherhand, C. albicans in farnesol-treated cell culturesshowed no evidence of filamentation, even with20% FCS (Figure 6) compared to the cultures con-taining no farnesol (Figure 1). To confirm thesemicroscopic observations, epithelial cell numbersand the percentage of C. albicans hyphal formswere quantified at each time point. C. albicansalone had no effect on epithelial cell morphologyand viability (Table 5). When epithelial cells werecultured in the presence of C. albicans and a highconcentration of farnesol, but in the absence ofserum, the number of epithelial cells was signifi-cantly (p < 0.01) reduced at 4 h (Table 5). Supple-menting the culture medium with 20% FCS inhib-ited the effect of farnesol on the epithelial cells

but not the effect on C. albicans filamentation. Asearly as 2 h post-treatment, the percentage of pseu-dohyphae and hyphae decreased significantly, withthe percentage decrease tracking the increase in theconcentration of farnesol (Table 6). FCS thus elim-inated the inhibitory effect of farnesol on epithelialcells but did not reduce the inhibitory effect onC. albicans filamentation.

Discussion

C. albicans is responsible for a variety of infec-tions in humans, ranging from superficial muco-cutaneous candidiasis to life-threatening dissem-inated infections (Fradin and Hube, 2003; Gowet al., 2002; Reznik, 2005). The limited numberof available antifungal drugs and the emergenceof resistance to them point to the need for thedevelopment of new and more effective antifun-gal drugs. Farnesol is known to act in vitro as anaturally occurring quorum-sensing molecule thatreduces hyphae formation by C. albicans. In the



0 µM

Without Serum With Serum

150 µM

300 µM

Figure 5. Epithelial cell and C. albicans morphology following co-culture in the presence of farnesol. Gingival epithelial cellswere cultured as reported above. At 80% confluence, 106 C. albicans cells were added and the cultures were pulsed withfarnesol at various concentrations in the presence or absence of 20% FCS. Epithelial cell and C. albicans morphology andadherence were observed at different time points and photographed after 24 h. The photographs are representative of sixseparate experiments. Arrows show C. albicans adhering to the epithelial cells (magnification, 300×)

present study, we demonstrated that farnesol waseffective in reducing C. albicans growth as earlyas 2–4 h post-contact. The efficacy of farnesolagainst C. albicans growth was greater when it wasadded repeatedly to the C. albicans culture. Theseresults are in agreement with those reported else-where. Farnesol has also been shown to be effec-tive against C. albicans biofilm formation (Ramageet al., 2002).

Farnesol has been reported to prevent thisfilamentation (Calderone and Fonzi, 2001; Leigh

et al., 2001). In the present study, we were ableto demonstrate that low concentrations (10 µM)of farnesol inhibited the yeast-to-hyphal transitionof C. albicans. With high concentrations of far-nesol (100–300 µM), we observed fewer than 30%hyphal forms, compared to 90% with the controls.These results suggest that, even at low concentra-tions, farnesol inhibits the yeast-to-hyphal transi-tion of C. albicans, although it does not block theelongation of existing germ tubes. These resultssupport the findings of Mosel et al. (2005), Kruppa



Table 6. Percentage of C. albicans transition from blastospore to hyphae forms when cultured in the presence of gingivalepithelial cells and farnesol at different concentrations

Hyphae form of C. albicansa


reaction (h) 0 10 50 100 150 300 0 10 50 100 150 300

2 2.85 ± 3 0.71 ± 0.1 0 0 0 0 44 ± 1 26 ± 10∗ 10 ± 5∗ 7 ± 3∗ 5 ± 3.2∗ 2 ± 0.3∗4 5 ± 2 0 0 0 0 0 54 ± 10 11 ± 0.4∗ 4 ± 1.2∗ 3 ± 2∗ 2 ± 0.1∗ 08 1 ± 0.09 0 0 0 0 0 35 ± 2 2 ± 1.0∗ 2 ± 1.3∗ 4 ± 0.9∗ 0∗ 0∗24 0 0 0 0 0 0 30 ± 7 1 ± 0.4∗ 1 ± 0.1∗ 0.6 ± 0.1∗ 0∗ 0∗


et al. (2004) and Staab et al. (1999), who reportedthat farnesol decreases C. albicans filamentation.This suggests that farnesol-sensing by C. albicansmay be mediated by signal transduction proteins(Chkip, Hwp1) that inhibit the yeast-to-hyphal tran-sition of yeast cells. In addition, Hornby et al.(2003) reported that blocking sterol biosynthesis inC. albicans increases intracellular and extracellularfarnesol levels, which in turn reduces C. albicansfilamentation. Our study and others suggest thatfarnesol does not kill C. albicans or disturb thephysiological balance between the host and the nor-mal microflora, but rather affects the metabolismof C. albicans, which in turn controls the yeast-to-hyphal transition.

Oral candidiasis is associated with gingival tissuewhere epithelial cells and fibroblasts are the maincells of the oral mucosa (Dongari-Bagtzoglou andFidel, 2005; Fidel, 2002; Rouabhia et al., 2002).The effect of farnesol on gingival cells was there-fore investigated. Our results show that at ele-vated concentrations, farnesol had a more signif-icant effect on fibroblast adhesion and growth thanon epithelial cells. Moreover, this effect appeared tobe time-dependent, as the longer the contact periodbetween gingival cells and farnesol, the greaterthe damage to the cells. Farnesol has previouslybeen reported to enhance CHO-K1 cell apopto-sis (Wright et al., 2001), promote tobacco BY-2cell death (Hemmerlin, 2005), inhibit hepatocytenodule formation (Horn et al., 2005) and suppressthe growth of murine B16 melanoma (McAnallyet al., 2003). The effect of farnesol on normal gin-gival cells was documented for the first time inthe present study. These findings address a keyissue, viz. the effect of farnesol on host cells.

However, prior to being used as a means to con-trol C. albicans growth and filamentation, furtherexperiments are needed to significantly reduce orprevent side-effects on host cells. All together,these data support the possible use of farnesol as anantifungal agent. This may overcome the antifun-gal resistance the medical community is currentlydealing with (Kuriyama et al., 2005; Pfaller et al.,2005).

Our study of the effect of farnesol on C. albicansor gingival cells alone failed to mimic the in vivoconditions in which the yeast is in contact withthe host tissue. To simulate these in vivo condi-tions, gingival cells were cultured in the pres-ence of C. albicans and various concentrations offarnesol. In the absence of serum, farnesol dis-rupted fibroblast adhesion, even at a low concen-tration. It also had a significant impact on epithe-lial cell morphology. While serum completely pre-vented the cell disorganization initiated by far-nesol, it had no effect on the ability of farnesolto reduce C. albicans filamentation. Moreover, thedecrease in the degree of yeast-to-hyphal transitionwas more pronounced when gingival cells werepresent than when C. albicans was cultured alone.Indeed, at defined concentrations (e.g. 10 µM far-nesol), close to 70% of the yeast-to-hyphal tran-sition occurred when only farnesol was added tothe C. albicans cultures. However, these valuesdropped to approximately 20% in the presenceof fibroblasts and to nearly 10% with epithelialcells. These results therefore confirm the synergis-tic effect of farnesol and gingival cells in inhibitingthe yeast-to-hyphal transition. The effect of epithe-lial cells on C. albicans filamentation has beenreported by others (Barousse et al., 2001; Rouabhia



et al., 2002, 2005). However, our study is the firstto show that fibroblasts are involved in control-ling the yeast-to-hyphal transition of C. albicans.We are also the first to report a synergistic effectbetween gingival cells and farnesol. We are cur-rently investigating the cellular and molecularmechanisms used by normal gingival cells in asso-ciation with farnesol to control C. albicans growthand filamentation.

In conclusion, our study confirmed the efficacyof farnesol in inhibiting C. albicans growth andfilamentation. A damaging effect of farnesol ongingival cells was also observed, suggesting thatit be used under better-controlled conditions (con-centration, contact period). This is the first timethat a synergistic effect between gingival cells andfarnesol against C. albicans filamentation has beendemonstrated. Additional studies are required todetermine the potential of farnesol for treating fun-gal infections such as candidiasis.

Acknowledgements

The authors would like to thank the clinicians of theFaculty of Dental Medicine of the Universite Laval fortheir help in collecting the tissue samples. This study wassupported by operating grants from the Natural Sciencesand Engineering Research Council of Canada (NSERC),the Canadian Institutes of Health Research (CHIR) and theFRSQ Reseau de Recherche en Sante Buccodentaire.

References

Andrian E, Grenier D, Rouabhia M. 2005. Porphyromonasgingivalis lipopolysaccharide induces shedding of syndecan-1 expressed by gingival epithelial cells. J Cell Physiol 204:178–183.

Barousse MM, Steele C, Dunlap K, et al. 2001. Growth inhibitionof Candida albicans by human vaginal epithelial cells. J InfectDis 184: 1489–1493.

Blijlevens NMA, Donnelly JP, DePauw BE. 2005. Inflammatoryresponse to mucosal barrier injury after myeloablative therapyin allogeneic stem cell transplant recipients. Bone MarrowTransplant 36: 703–707.

Blijlevens NMA, van’t Land B, Donnelly JP, M’Rabet L, dePauw BE. 2004. Measuring mucosal damage induced bycytotoxic therapy. Support Care Cancer 12: 227–233.

Bos JD, Kapsenberg ML. 1993. The skin immune system: progressin cutaneous biology. Immunol Today 14: 75–78.

Burke YD, Stark MJ, Roach SL, Sen SE, Crowell PL. 1997.Inhibition of pancreatic cancer growth by the dietary isoprenoidsfarnesol and geraniol. Lipids 32: 151–156.

Bustamante CI. 2005. Treatment of Candida infection: a viewfrom the trenches! Curr Opin Infect Dis 18: 490–495.

Calderone RA, Fonzi WA. 2001. Virulence factors of Candidaalbicans . Trends Microbiol 9: 327–335.

Campbell SL, Khosravi-Far R, Rossman KL, Clark GJ, Der CJ.1998. Increasing complexity of Ras signalling. Oncogene 17:1395–413.

Cao YY, Cao YB, Xu Z, et al. 2005. cDNA microarray analysisof differential gene expression in Candida albicans biofilmexposed to farnesol. Antimicrob Agents Chemother 49: 584–589.

Del Villar K, Dorin D, Sattler I, et al. 1996. C-terminal motifsfound in Ras-superfamily G-proteins: CAAX and C-sevenmotifs. Biochem Soc Trans 24: 709–713.

Dongari-Bagtzoglou A, Fidel PL Jr. 2005. The host cytokineresponses and protective immunity in oropharyngeal candidiasis.J Dent Res 84: 966–977.

Edwards PA, Ericsson J. 1999. Sterols and isoprenoids: signallingmolecules derived from the cholesterol biosynthetic pathway.Ann Rev Biochem 68: 157–185.

Enjalbert B, Whiteway M. 2005. Release from quorum-sensingmolecules triggers hyphal formation during Candida albicansresumption of growth. Eukaryot Cell 4: 1203–1210.

Fetter A, Partisani M, Koenig H, Kremer M, Lang JM. 1993.Asymptomatic oral Candida albicans carriage in HIV infection:frequency and predisposing factors. J Oral Pathol Med 22:57–59.

Fidel PL Jr. 2002. Distinct protective host defenses against oraland vaginal candidiasis. Med Mycol 40: 359–375.

Fradin C, Hube B. 2003. Tissue infection and site-specific geneexpression in Candida albicans . Adv Appl Microbiol 53:271–290.

Gibbs JB, Graham SL, Hartman GD, et al. 1997. Farnesyltrans-ferase inhibitors versus Ras inhibitors. Curr Opin Chem Biol 1:197–203.

Gottfredsson M, Cox GM, Indridason OS, et al. 1999. Associationof plasma levels of human immunodeficiency virus type 1RNA and oropharyngeal Candida colonization. J Infect Dis 180:534–537.

Gow NA, Brown A, Odds FC. 2002. Fungal morphogenesis andhost invasion. Curr Opin Microbiol 5: 366–371.

Grunler J, Ericsson J, Dallner G. 1994. Branch-point in thebiosynthesis of cholesterol, dolichol, ubiquinone and prenylateproteins. Biochim Biophys Acta 1212: 259–277.

Hemmerlin A, Reents R, Mutterer J, et al. 2005. Monitoringfarnesol-induced toxicity in tobacco BY-2 cells with afluorescent analog. Arch Biochem Biophys (in press).

Horn TL, Long L, Cwik MJ, et al. 2005. Modulation of hepaticand renal drug metabolizing enzyme activities in rats bysubchronic administration of farnesol. Chem Biol Interact 152:79–99.

Hornby JM, Jensen EC, Lisec AD, et al. 2001. Quorum sensing inthe dimorphic fungus Candida albicans is mediated by farnesol.Appl Environ Microbiol 67: 2982–2992.

Hornby JM, Kebaara BW, Nickerson KW. 2003. Farnesol biosyn-thesis in Candida albicans: cellular response to sterol inhibi-tion by zaragozic acid B. Antimicrob Agents Chemother 47:2366–2369.

Hudes GR, Szarka CE, Adams A, et al. 2000. Phase I pharma-cokinetic trial of perillyl alcohol (NSC 641066) in patients withrefractory solid malignancies. Clin Cancer Res 6: 3071–3080.

Kruppa M, Krom BP, Chauhan N, et al. 2004. The two-component signal transduction protein Chk1p regulates quorumsensing in Candida albicans . Eukaryot Cell 3: 1062–1065.



Kuriyama T, Williams DW, Bagg J, et al. 2005. In vitro suscepti-bility of oral Candida to seven antifungal agents. Oral MicrobiolImmunol 20: 349–353.

Leigh JE, Barousse M, Swoboda RK, et al. 2001. Candida-specific systemic cell-mediated immune reactivities in humanimmunodeficiency virus-positive persons with mucosal candidi-asis. J Infect Dis 183: 277–285.

McAnally JA, Jung M, Mo H. 2003. Farnesyl-O-acetylhydro-quinone and geranyl-O-acetylhydroquinone suppress the pro-liferation of murine B16 melanoma cells, human prostate andcolon adenocarcinoma cells, human lung carcinoma cells, andhuman leukemia cells. Cancer Lett 202: 181–192.

Miller MB, Bassler BL. 2001. Quorum sensing in bacteria. AnnuRev Microbiol 55: 165–199.

Mosel DD, Dumitru R, Hornby JM, Atkin AL, Nickerson KW.2005. Farnesol concentrations required to block germ tubeformation in Candida albicans in the presence and absence ofserum. Appl Environ Microbiol 71: 4938–4940.

Mostefaoui Y, Bart C, Frenette M, Rouabhia M. 2004. Candidaalbicans and Streptococcus salivarius modulate IL-6, IL-8, andTNF-α expression and secretion by engineered human oralmucosa cells. Cell Microbiol 6: 1085–1096.

Mostefaoui Y, Claveau I, Ross G, Rouabhia M. 2002. Tissuestructure, and IL-1β, IL-8 and TNF-α secretions after contactby engineered human oral mucosa with dentifrices. J ClinPeriodontol 29: 1035–1041.

Odds FC. 1988. Candida and Candidosis: A Review andBibliography. Bailliere Tindall: London.

Pei ZM, Ghassemian M, Kwak CM, McCourt P, Schroeder JI.1998. Role of farnesyltransferase in ABA regulation of guardcell anion channels and plant water loss. Science 282: 287–290.

Pfaller MA, Boyken L, Hollis RJ, et al. 2005. In vitro activitiesof anidulafungin against more than 2500 clinical isolates ofCandida spp., including 315 isolates resistant to fluconazole.J Clin Microbiol 43: 5425–5427.

Poon GK, Vigushin D, Griggs LJ, et al. 1996. Identification andcharacterization of limonene metabolites in patients withadvanced cancer by liquid chromatography/mass spectrometry.Drug Metab Dispos 24: 565–571.

Presland RB, Dale BA. 2000. Epithelial structural proteins of theskin and oral cavity: function in health and disease. Crit RevOral Biol Med 11: 383–408.

Presland RB, Jurevic RJ. 2002. Making sense of the epithelialbarrier: what molecular biology and genetics tell us about thefunctions of oral mucosal and epidermal tissues. J Dent Educ66: 564–574.

Ramage G, Saville SP, Wickes BL, Lopez-Ribot JL. 2002.Inhibition of Candida albicans biofilm formation by farnesol,a quorum-sensing molecule. Appl Environ Microbiol 68:5459–5463.

Reznik DA. 2005. Oral manifestations of HIV disease. Top HIVMed 13: 143–148.

Rouabhia M, JobinN, Doucet R Jr, Bergeron J, AugerA. 1994. CD36(+)-dendritic epidermal cells: a putativeactor in the cutaneous immune system. Cell Transplant 3:529–536.

Rouabhia M, Ross G, Page N, Chakir J. 2002. Interleukin-18and gamma interferon production by oral epithelial cells inresponse to exposure to Candida albicans or lipopolysaccharidestimulation. Infect Immun 70: 7073–7080.

Rouabhia M, Schaller M, Corbucci C, et al. 2005. Virulenceof the fungal pathogen Candida albicans requires the fiveisoforms of protein mannosyltransferases. Infect Immun 73:4571–4580.

Schmidt-Westhausen AM, Bendick C, Reichart PA, SamaranayakeLP. 2004. Oral candidosis and associated Candida species inHIV-infected Cambodians exposed to antimycotics. Mycoses 47:435–441.

Selvaratnam L, Cruchley AT, Navsaria H, et al. 2001. Permeabil-ity barrier properties of oral keratinocyte cultures: a model ofintact human oral mucosa. Oral Dis 7: 252–258.

Shibuya K, Coulson WF, Wollman JS, et al. 2001. Histopathologyof cryptococcosis and other fungal infections in patients withacquired immunodeficiency syndrome. Int J Infect Dis 5:78–85.

Smith DJ, Taubman MA. 1992. Ontogeny of immunity tooral microbiota in humans. Crit Rev Oral Biol Med 3:109–133.

Staab JF, Bradway SD, Fidel PL, Sundstrom P. 1999. Adhesiveand mammalian transglutaminase substrate properties ofCandida albicans Hwp1. Science 283: 1535–1538.

Suzuki N, Urano J, Tamanoi F. 1998. Farnesyltransferaseinhibitors induce cytochrome c release and caspase 3 activationpreferentially in transformed cells. Proc Natl Acad Sci USA 95:15356–15361.

Vazquez JA. 2000. Therapeutic options for the managementof oropharyngeal and esophageal candidiasis in HIV/AIDSpatients. HIV Clin Trials 1: 47–59.

Wargovich MJ, Jimenez A, McKee K, et al. 2000. Efficacyof potential chemopreventive agents on rat colon aber-rant crypt formation and progression. Carcinogenesis 21:1149–1155.

Wilkieson C, Samaranayake LP, MacFarlane TW, Lamey PJ,MacKenzie D. 1991. Oral candidosis in the elderly in long termhospital care. J Oral pathol Med 20: 13–16.

Wright MM, Henneberry AL, Lagace TA, Ridgway ND, McMas-ter CR. 2001. Uncoupling farnesol-induced apoptosis from itsinhibition of phosphatidylcholine synthesis. J Biol Chem 276:25254–25261.


Documents

In vitro synergistic effect of farnesol and human gingival cells against Candida albicans