Correspondence: Sueli Fumie Yamada-Ogatta, Departamento de Microbiologia, Centro de Ciências Biológicas, Universidade Estadual de Londrina, Rodovia Celso Garcia Cid s/n, Campus Universitário, Londrina, Paraná, Brazil, CEP 86051-990. Tel.: +55 43 3371 4297; fax: +55 43 3371 4192; e-mail: firstname.lastname@example.org
Candida tropicalis is a common species related to nosocomial candidemia and candiduria. Most Candida spp. infections are associated with biofilm formation on implanted medical devices or on host epithelial cell surfaces. Sessile cells display phenotypic traits dramatically different from those of their free-living, planktonic counterparts, such as increased resistance to antimicrobial agents and to host defenses. The characteristics of C. tropicalis biofilm formation in vitro are described. By an XTT-reduction assay, an increase in metabolic activity was observed up to 24 h of biofilm formation, and this activity showed a linear relationship with sessile cell density. Scanning electron microscopy was used to further characterize C. tropicalis biofilms. The initial adherence of yeast cells was followed by germination, microcolony formation, filamentation and maturation at 24–48 h. Mature biofilms consisted of a dense network of yeast cells and filamentous forms of C. tropicalis. Increased resistance of sessile cells against fluconazole and amphotericin B was also demonstrated. Real-time reverse transcription-PCR quantification showed that sessile cells overexpressed ERG11 (coding for lanosterol 14 α-demethylase) and MDR1 (coding for an efflux protein belonging to the major facilitator superfamily). These mechanisms may contribute to the fluconazole resistance of the C. tropicalis biofilm.
Species of the genus Candida are endogenous commensals of the gastrointestinal and urogenital tracts in healthy individuals. However, as opportunistic pathogens, they can cause diseases ranging from mucosal candidiasis to life-threatening disseminated infections, mainly in immunocompromised hosts (Soll, 2002).
Candida albicans has been regarded as the most common causative agent of fungal infection in humans. However, Candida species other than C. albicans have become a significant cause of infection, particularly in invasive candidal infections, and the population at risk includes cancer patients and transplant recipients treated with immunosuppressive drugs. In addition, healthcare-related factors, including intravascular catheter, broad-spectrum antibiotic use and surgical procedure, are risk factors for Candida invasive infections (Ruhnke, 2006).
A substantial proportion of candidal infections are associated with biofilm formation, especially on the surface of implanted medical devices (Douglas, 2003; Ramage et al., 2006). Biofilm consists of surface-attached communities of cells embedded within an exopolymeric matrix that these cells produce. Sessile cells within the communities display an altered phenotype with respect to growth rate and gene transcription compared with that of planktonic cells (Donlan & Costerton, 2002). The marked clinical significance of biofilm concerns the enhanced resistance of yeast to a variety of antimicrobial agents (Hawser & Douglas, 1995; Ramage et al., 2001a, b, c) and its ability to withstand host defenses (Vuong et al., 2004). Thus, biofilm-associated infections are difficult to treat, representing a source of reinfections (Ramage et al., 2006).
Similar to bacterial biofilms, C. albicans biofilms display a complex, three-dimensional architecture with structural heterogeneity, the presence of exopolymeric material and decreased susceptibility to antimicrobial agents (Chandra et al., 2001; Ramage et al., 2001c). In contrast to the extensive literature dealing with C. albicans biofilms (Hawser & Douglas, 1995; Chandra et al., 2001; Ramage et al., 2001b, c), little attention has been paid to C. tropicalis. It has been reported that this yeast can form extensive biofilms in vitro on the surface of polyvinyl chloride (PVC) catheter (Hawser & Douglas, 1994) and polystyrene (Shin et al., 2002; Parahitiyawa et al., 2006). Therefore, the purpose of this study was to investigate the characteristics of C. tropicalis biofilm development with respect to identifying growth phases, morphology and antifungal susceptibility. The results of these studies can contribute to one's understanding of the antifungal resistance of C. tropicalis biofilm as well as to its biology.
Materials and methods
Candida tropicalis isolates and growth conditions
Candida tropicalis strain 112MC was isolated from a patient with vulvovaginal candidiasis who was seen at the Departamento de Análises Clínicas, Universidade Estadual de Maringá, Paraná, Brazil. Candida tropicalis strain U9815 was isolated from the urine of a patient admitted to Hospital Universitário Regional Norte do Paraná, Londrina, Paraná, Brazil. All strains were maintained on Sabouraud dextrose (SD) agar (Himedia, India) and subcultured monthly. The yeasts were also stored in sterile-distilled water at 25 °C (McGinnis et al., 1974). Species identification was confirmed by a PCR-based method using specific primers directed against the 3′ end of 5.8S and the 5′ end of 28S rRNA gene regions (Ahmad et al., 2002). Genomic DNA was extracted following the procedures described previously (Jain et al., 2001).
Biofilm formation and growth kinetics on the surface of polystyrene microtiter plates
Biofilm production by the different C. tropicalis isolates was performed in polystyrene, flat-bottomed 96-well microtiter plates (Techno Plastic Products, Switzerland) using a procedure described previously (Shin et al., 2002). Briefly, the yeast isolates were grown at 37 °C for 24 h in SD broth, pH 6.0, and the cells were counted using a hemocytometer (Neubauer Improved Chamber). A suspension of 6 × 105 yeast cells in 20 μL RPMI 1640 pH 7.0 (Invitrogen-Gibco) was placed in each well containing 180 μL of the same medium. The plates were incubated for various time intervals (3, 6, 9, 12, 24 and 48 h) at 37 °C. After each incubation period, the medium was aspirated off and nonadherent cells were removed by washing thoroughly three times with sterile 0.15 M phosphate-buffered saline (PBS) pH 7.2. Biofilm formation was quantified using the 2,3-bis(2-methoxy-4-nitro-5-sulfo-phenyl)-2H-tetrazolium-5-carboxanilide (XTT)-reduction assay as described elsewhere (Ramage et al., 2001b). A 100-μL aliquot of XTT-menadione [0.1 mg mL−1 XTT, 1 μM menadione (Sigma Chemical Co.)] was added to each well, and the plates were incubated in the dark for 2 h at 37 °C before spectrophotometric readings at 490 nm with a microtiter plate reader (Universal Microplate Reader ELx 800, Bio-Tek Instruments). Experiments were carried out in triplicate on three different occasions. The differences in biofilm metabolic activity among the isolates were compared with a t-test using statistica 6.0 software (StatSoft Inc.). The significance level for P values is given in the figure legend.
Scanning electron microscopy (SEM)
Strips of PVC (surface area 0.5 cm2) were aseptically cut and placed in wells of 24-well tissue-culture plates (Techno Plastic Products, Switzerland). A standard inoculum of 3.0 × 106 cells, from overnight cultures of the yeast strains, was prepared in 1 mL of RPMI 1640, pH 7.0, medium and used to form a biofilm on this surface. The strips were then immersed in these cell suspensions and incubated statically at 37 °C for 48 h. After this incubation, nonadherent organisms were removed by washing gently three times with PBS pH 7.2. Data collections were carried out at the different times of biofilm development as described above. Biofilms formed on these strips were fixed with 2.5% (v/v) glutaraldehyde in 0.1 M cacodylate buffer (pH 7.2) at room temperature. Postfixation, the cells were dehydrated with a series of ethanol washes (15%, 30%, 50%, 70%, 80%, 90%, 95% and 100%), critical-point dried in CO2, coated with gold and examined with a SHIMADZU SS-550 scanning electron microscope.
Antifungal susceptibility testing
The minimum inhibitory concentrations (MIC) of fluconazole (Pfizer Central Research, United Kingdom) and amphotericin B (Sigma Chemical Co.) for planktonic cells of all isolates were determined by broth microdilution assays for yeasts according to the Clinical and Laboratory Standards Institute [CLSI (M27-A2 document, NCCLS, 2002)]. Quality control Candida parapsilosis ATCC 22019 (FIOCRUZ, Rio de Janeiro, Brazil) was included in each experiment. Two wells of each plate served as growth and sterility controls. The following breakpoint definitions for fluconazole were used according to CLSI guidelines to categorize the isolates: MIC≥64 μg mL−1, resistant; MIC of 16–32 μg mL−1, susceptible dose dependent; and MIC≤8 μg mL−1, susceptible. For amphotericin B, isolates with MIC>1 μg mL−1 were considered to be resistant (Pinto et al., 2006). To determine antifungal susceptibilities of sessile cells, biofilms were formed as described above. After 24 h of biofilm formation, the medium was aspirated off and each well was washed three times with sterile PBS. A 200-μL aliquot of RPMI 1640 medium containing serially double-diluted concentrations of antifungal (fluconazole 512.0–0.5 μg mL−1, amphotericin B 8.00–0.03 μg mL−1) was added and the plates were incubated further for 48 h at 37 °C. Controls included antifungal-free wells and biofilm-free wells. Sessile minimum inhibitory concentrations were determined at 50 % inhibition (SMIC50) and at 80% inhibition (SMIC80) compared with antifungal-free control wells using the XTT-reduction assay described above. Experiments were carried out in triplicate in three different assays.
Real-time PCR was performed to determine the relative ERG11 and MDR1 mRNA levels of C. tropicalis biofilm. The planktonic cells were obtained from the supernatant of the C. tropicalis cultures incubated with agitation (200 r.p.m.) for 24 h at 37 °C. The 24 h-sessile cells from the polystyrene plate were harvested by gentle scraping with a sterile toothpick. The total RNAs were extracted using the RNAeasy kit (Qiagen Inc.) following the manufacturer's instruction and were treated with DNAse-RQ1 (Promega, Brazil). The nucleotide sequences of ERG11 (accession number AY942646), MDR1 (accession number AF194419) and ACT1 (coding for actin, accession number AJ237918) deposited in the GenBank/EMBL databases were used for specific primer design. The primer pairs were: ERG11, 5′-ATGGCTATTGTTGATACTGC-3′ and 5′-GCATTGTAAATGAATTCGTG-3′; MDR1, 5′-CCCAGAAGTTTTCATTCCA-3′ and 5′-CCCCAAGCAACAGGATAAT-3′; and ACT1, 5′-ATGGACGGGGGTATGTTTCA-3′ and 5′-GACATAAGTAATTTCCAATGTG-3′. Total RNA (1 μg) was converted to cDNA by incubation with oligo(dT) and ImProm-II™ reverse transcription (Promega) for 2 h at 42 °C. Samples were purified by centrifugation through a Microcon YM-30 filter (Millipore, Brazil). Two-step real-time RT-PCR assays were performed in an ABI PRISM 7000 Sequence Detection System (Applied Biosystems, Brazil). Fifteen nanograms of cDNA and the recommended concentration of SYBR Green Master Mix (Applied Biosystems, Brazil) were added to a 25 μL reaction mixture. The specific primers were added at a concentration of 200 nM in all cases. PCR conditions were as follows: 2 min at 50 °C, 10 min at 95 °C, followed by 45 cycles of 95 °C for 15 s, 55 °C for 30 s and 72 °C for 30 s. Thermal dissociation confirmed that RT-PCR generated a single amplicon. A standard curve method was used, based on Ct values, to assess the expression of the genes studied. Triplicate 1 : 2 dilutions of eight known concentrations of cDNA were used to generate curves extending from 20 to 0.16 ng of cDNA. A standard curve was constructed for each of the genes studied and for the control gene (ACT1). The cDNA concentration was calculated by dividing the value obtained for the gene under investigation by the value obtained for the control gene. Differences in expression are reported, using planktonic cells as the reference population.
Biofilm formation by C. tropicalis
Biofilm formation by C. tropicalis on the surface of polystyrene wells was monitored using the XTT-reduction assay. This method is based on the ability of mitochondrial dehydrogenases in viable cells to convert the substrate into a reduced-formazan product that can be measured in a spectrophotometer. The kinetics of biofilm formation over 48 h, as revealed by this assay, are illustrated in Fig. 1. An initial rapid increase in biofilm metabolic activity was detected in the first 12 h, which indicated the increase in cell density. After 24 h, the metabolic activity remained high but reached a plateau. The growth kinetic was similar for the two C. tropicalis strains, but variations in the metabolic activity level of the biofilms were detected in the individual strains (P<0.01). The highest activity was observed in the 112MC strain, a vaginal isolate. The initial inoculum of the cells consisted mainly of yeast cells (data not shown).
SEM of C. tropicalis biofilm
Biofilm formation by C. tropicalis on PVC strips was examined by SEM (Fig. 2). In the first 3 h, distinct microcolonies of yeast and filamentous cells were visualized on the surface of the strip. Interestingly, during the biofilm growth, the majority of C. tropicalis cells were present as filamentous forms (6–24 h). Mature biofilm consisted of a dense and heterogeneous network of yeast, pseudohyphae and hyphae. In addition, intense hyphal budding was observed in this film (Fig. 3). The extrapolymeric material could not be identified because of the destructive nature of SEM. Overall, these results indicate that biofilm growth kinetic on a polystyrene surface is similar to that on catheter surfaces.
Antifungal susceptibility pattern of planktonic and sessile cells of C. tropicalis
A clear difference in antifungal susceptibilities was seen between planktonic and sessile cells of the same C. tropicalis strain. Fluconazole and amphotericin B showed decreased activity against the biofilm of both yeast strains tested. The results, determined by XTT-reduction assay and expressed as SMIC50s and SMIC80s, are shown in Table 1. Biofilms were intrinsically resistant to fluconazole and amphotericin B, although their counterpart planktonic cells remained susceptible to the same antifungals.
Table 1. Antifungal susceptibility testing of Candida tropicalis under planktonic (MIC) and sessile (SMIC) growing conditions
Amp B MIC
Values are in μg mL−1.
Fcz, Fluconazole; Amp B, Amphotericin B.
Expression of genes ERG11 and MDR1 in the C. tropicalis biofilm
To determine the expression of the genes ERG11 and MDR1 in the biofilm, real-time PCR was performed using RNAs extracted from C. tropicalis mature biofilm and planktonic cells. The pattern of expression observed for these genes was similar between the vaginal and urinary isolates (Fig. 4). The analysis of expression revealed that both genes were overexpressed in the sessile cells compared with their planktonic counterparts. MDR1 overexpression was 3.1- and 2.7-fold for the U9815 and 112 MC isolates, respectively. For gene ERG11, a slight increase in expression of 2.2- and 1.5-fold was observed in the U-9815 and 112 MC isolates. Moreover, the expression of both genes was noted to be higher in the biofilm cells of the urinary isolate.
The formation of biofilms by Candida species has been demonstrated on a number of abiotic surfaces, including medical devices (Ramage et al., 2006). The initial phase of C. albicans biofilm formation begins with the adherence of yeast cells to the substrate surface, followed by germination and microcolony formations. The metabolic activity is intense and the extracellular material emerges during the intermediate phase. Mature biofilms consist of a dense network of yeast and hyphal elements embedded within exopolymeric material (Chandra et al., 2001; Ramage et al., 2001c). Similar structural features have been observed in biofilm formed in rat (Andes et al., 2004) and rabbit (Schinabeck et al., 2004) central venous catheter models. Additional host cells, including neutrophils, macrophages, red blood cells and platelets, have been observed within biofilm formed in vivo.
In this study, the ability of C. tropicalis to adhere and form biofilm on polystyrene and PVC surfaces under static conditions was demonstrated. Shin et al. (2002) compared the different species of Candida with regard to their ability to produce biofilms on a polystyrene surface. Using SD broth containing a final concentration of 8% glucose, these authors observed that C. tropicalis was the most frequent biofilm-producing species. In both the surfaces used in this study, the development of biofilm proceeds through three distinct phases, adhesion, proliferation and maturation, as occurs with C. albicans and the other species (Chandra et al., 2001; Ramage et al., 2001a; Kuhn et al., 2002a). In spite of the similar developmental phases, variations in the metabolic activities of the biofilm formed on the polystyrene surface were detected between the C. tropicalis strains tested. Different C. albicans isolates also showed variations in the biofilm formation on the PVC surface (Hawser & Douglas, 1994). These results probably reflect the physiological differences between the strains.
Mature biofilms of C. tropicalis U9815 and 112 MC on a PVC surface are structurally similar to those described for C. albicans on the same surface by Hawser & Douglas (1994). Clearly, an enormous quantity of hyphal elements was observed during the growth and maturation phase of the biofilm development of C. tropicalis in RPMI medium. The mixture of yeasts and filamentous forms was not seen when the microorganism was grown in liquid culture alone. In broth medium, C. tropicalis undergoes reversible morphological transitions between yeast cells and pseudohyphae, which can be induced by ethanol (Tani et al., 1979; Suzuki et al., 2006). In Sabouraud broth medium, it was observed that C. tropicalis biofilms contained only blastospore forms (data not shown). Chandra et al. (2001) showed that biofilms of C. albicans grown on polymethylmethacrylate strips in yeast nitrogen base (YNB) medium supplemented with glucose contained mainly yeast forms. On the other hand, the biofilms grown in RPMI medium, which induces hyphal formation in C. albicans (Hoyer et al., 1995), consisted mostly of filamentous forms. Using the YNB medium supplemented with glucose, Kuhn et al. (2002a) observed that most mature biofilms of C. tropicalis on the surface of serum-preconditioned silicone elastomer disks displayed only blastospore forms. Only one isolate of C. tropicalis produced a thin layer of hyphae. On the polystyrene surface (Calgary biofilm device), C. tropicalis biofilm consisted of large coaggregated microcolonies of blastospores with a thick extracellular polymeric layer (Parahitiyawa et al., 2006). Almost all microorganisms display structural heterogeneity of biofilm architecture (Wimpenny et al., 2000) and this characteristic appears to be common in biofilms formed by C. tropicalis.
The extrapolymeric material could not be identified in this study because of the destructive nature of SEM. Recently, Al-Fattani & Douglas (2006) reported that C. tropicalis biofilm synthesized large amounts of extrapolymeric material even when grown statically, and such a matrix is composed of hexosamine (the major component), carbohydrate, protein, phosphorus and uronic acid.
It has been suggested that morphogenesis is triggered when the organism contacts a surface (Chandra et al., 2001; Ramage et al., 2001c; Douglas, 2003; Kumamoto & Vinces, 2005), and hyphae may have an important role in the structural integrity and multilayered architecture of mature biofilm (Baillie & Douglas, 1999; Ramage et al., 2002b). Indeed, the surface contact results in various cellular behaviors, including biofilm formation and invasion. These phenotypic changes involve differential gene expression, which will ultimately cause the organism to respond according to the environmental conditions. In this context, Davies & Geesey (1995) reported that the production of alginate, an exopolymeric material of Pseudomonas aeruginosa biofilm, occurs in the early phase of the process. In fact, cells attached to a glass surface for at least 15 min exhibited up-regulation of algC, the main gene in the alginate biosynthesis pathway.
The molecular mechanisms involved in the triggering and regulation of the biofilm formation process in Candida are still unknown. Recently, several reports described the global transcriptional pattern of C. albicans biofilm development under different in vitro conditions. Differential expression of several genes can be detected in the early (30 min to 6 h) biofilm developmental phase. In general, significant differential expression in sessile cells was observed in genes involved in protein synthesis, amino acid, nucleotide, lipid and carbohydrate metabolism, transcription and control of cellular organization (Garcia-Sanchez et al., 2004; Cao et al., 2005; Murillo et al., 2005; Yeater et al., 2007).
Overexpression of MDR1 and CDR1 [ATP-binding cassette pump (Barchiesi et al., 2000)] and ERG11 (Vandeputte et al., 2005) has been associated with the resistance of planktonic cells of C. tropicalis to fluconazole, but the molecular mechanisms of resistance during biofilm growth are unclear. The results of this study showed the increased expression of MDR1 and ERG11 in biofilm of C. tropicalis. The expression of gene MDR1 was greater than that of ERG11, suggesting an active role of MDR1 in the resistance of biofilm of this yeast, at least in the first 24 h of its formation. Currently available lines of evidence indicate that the mechanisms of biofilm resistance are complex and multifactorial in C. albicans. Ramage et al. (2002a) showed by Northern blot analysis that mRNA levels for both CDR1 and CDR2 were higher in sessile C. albicans cells at 24 and 48 h of biofilm formation. MDR1 was transiently overexpressed in 24-h biofilms. However, mutants carrying single- and double-deletion mutations in these efflux pump genes retained the resistant phenotype during biofilm growth. Mukherjee et al. (2003) showed that biofilms formed by double and triple mutants for CDR1, CDR2 and MDR1 were more susceptible to fluconazole in the early phase (6 h) than the wild-type strain in the same phase. In the intermediate (12 h) and mature (48 h) biofilms, parent and mutant strains became resistant to the drug, indicating the lack of involvement of efflux pumps in resistance during these phases. These authors also showed that C. albicans biofilms have an altered sterol composition during intermediate and mature phases compared with the early phase. They suggested that these changes may contribute to the resistance phenotype at late stages of biofilm formation.
It was shown that the development and architecture of C. tropicalis biofilm are similar to those described for other Candida species. In addition, an elevated resistance of the sessile cells to fluconazole and amphotericin B was demonstrated. The sessile cells overexpressed ERG11 and MDR1, and these mechanisms may contribute to the drug resistance of C. tropicalis biofilm.
This work was supported by grants from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Pro-Reitoria de Pesquisa e Pós Graduação (PROPPG) of Universidade Estadual de Londrina (UEL). This work was part of the M.Sc. dissertation of F.C. Bizerra. The authors thank Dr A. Leyva for English editing of the manuscript.