An in vitro assay to study multiple Candida biofilms, in parallel, has been carried out using the Calgary biofilm device (CBD). We here report: i) standardization of the CBD for Candida albicans biofilm formation, ii) kinetics of C. albicans biofilm formation, iii) biofilm formation by five Candida species, and iv) effect of dietary carbohydrates on biofilm formation. The biofilm metabolic activity on all CBD pegs was similar (p=0.6693) and C. albicans biofilm formation revealed slow growth up to 36 h and significantly higher growth up to 48 h (p<0.001). Significant differences in total biofilm metabolic activity were seen for glucose, fructose and lactose grown C. albicans compared with sucrose and maltose grown yeasts. Candida krusei developed the largest biofilm mass (p<0.05) relative to C. albicans, C. glabrata, C. dubliniensis and C. tropicalis. Scanning electron microscopy revealed that C. krusei produced a thick multilayered biofilm of pseudohyphal forms embedded within the polymer matrix, whereas C. albicans, C. dubliniensis and C. tropicalis biofilms consisted of clusters or chains of cells with sparse extracellular matrix material. We conclude that CBD is a useful, simple, low cost miniature device for parallel study of Candida biofilms and factors modulating this phenomenon.

Antimycotic perfusion through Candida biofilms was demonstrated by a modification of a simple in vitro diffusion cell bioassay system. Using this model, the perfusion of three commonly used antifungal agents, amphotericin B, fluconazole, and flucytosine, was investigated in biofilms of three different Candida species (i.e., Candida albicans, Candida parapsilosis, and Candida krusei) that were developed on microporous filters. Scanning electron microscopy revealed that C. albicans formed a contiguous biofilm with tightly packed blastospores and occasional hyphae compared with C. parapsilosis and C. krusei, which developed confluent biofilms displaying structural heterogeneity and a lesser cell density, after 48 h of incubation on nutrient agar. Minor structural changes were also perceptible on the superficial layers of the biofilm after antifungal perfusion. The transport of antifungals to the distal biofilm-substratum interface was most impeded by C. albicans biofilms in comparison to C. parapsilosis and C. krusei. Fluconazole and flucytosine demonstrated similar levels of perfusion, while amphotericin B was the least penetrant through all three biofilms, although the latter appeared to cause the most structural damage to the superficial cells of the biofilm compared with the other antifungals. These results suggest that the antifungal perfusion through biofilm mode of growth in Candida is dependent both on the antimycotic and the Candida species in question, and in clinical terms, these phenomena could contribute to the failure of Candida biofilm-associated infections. Finally, the in vitro model we have described should serve as a useful system to investigate the complex interactions that appear to operate in vivo within the biofilm-antifungal interphase.

Elucidation of bacterial and fungal interactions in multispecies biofilms will have major impacts on understanding the pathophysiology of infections. The objectives of this study were to (i) evaluate the effect of Pseudomonas aeruginosa lipopolysaccharide (LPS) on Candida albicans hyphal development and transcriptional regulation, (ii) investigate protein expression during biofilm formation, and (iii) propose likely molecular mechanisms for these interactions. The effect of LPS on C. albicans biofilms was assessed by XTT-reduction and growth curve assays, light microscopy, scanning electron microscopy (SEM), and confocal laser scanning microscopy (CLSM). Changes in candidal hypha-specific genes (HSGs) and transcription factor EFG1 expression were assessed by real-time polymerase chain reaction and two-dimensional gel electrophoresis, respectively. Proteome changes were examined by mass spectrometry. Both metabolic activities and growth rates of LPS-treated C. albicans biofilms were significantly lower (P < 0.05). There were higher proportions of budding yeasts in test biofilms compared with the controls. SEM and CLSM further confirmed these data. Significantly upregulated HSGs (at 48 h) and EFG1 (up to 48 h) were noted in the test biofilms (P < 0.05) but cAMP levels remained unaffected. Proteomic analysis showed suppression of candidal septicolysin-like protein, potential reductase-flavodoxin fragment, serine hydroxymethyltransferase, hypothetical proteins Cao19.10301(ATP7), CaO19.4716(GDH1), CaO19.11135(PGK1), CaO19.9877(HNT1) by P. aeruginosa LPS. Our data imply that bacterial LPS inhibit C. albicans biofilm formation and hyphal development. The P. aeruginosa LPS likely target glycolysis-associated mechanisms during candidal filamentation.

Although Candida albicans is the most common human yeast pathogen, other Candida species such as C. krusei are now recognized as emerging agents, especially in patients with human immunodeficiency virus (HIV) disease. C. krusei is inherently resistant to the widely used triazole antifungal fluconazole and poses therapeutic problems, especially in systemic candidiasis. In a surveillance study of leprosy patients (with arrested or burnt-out disease) in a leprosarium in northern Thailand, we found a rate of oral carriage of C. krusei (36%) significantly (P < 0.05) higher than that for a healthy control group (10%). Among the Candida-positive patients, 16 of 35 (46%) carried C. krusei, while C. albicans was the second most common isolate (12 of 35 patients; 34%). The corresponding figures for the control group were 2 of 13 (15%) and 6 of 13 (46%), respectively. Studies of the antifungal resistance of the C. krusei isolates from patients indicated that all except one of the isolates were resistant to fluconazole, two isolates were resistant to ketoconazole, and all isolates were sensitive to amphotericin B. Evaluation of their genetic profiles by randomly amplified polymorphic DNA analysis with three different primers and subsequent analysis of the gel profiles by computerized clusterderived dendrograms revealed that the C. krusei isolates from patients belonged to 10 disparate clusters, despite the origin from a single locale. These nascent findings indicate an alarmingly high prevalence of a Candida species resistant to a widely used antifungal in a part of the world where HIV disease is endemic.

The extracellular phospholipases of Candida albicans are considered to play a significant role in the pathogenesis of human infections. Therefore 30 clinical isolates of C. albicans from human immunodeficiency virus (HIV)-infected individuals were screened for phospholipase production in vitro (using an egg-yolk-agar medium). Two groups of six isolates with positive (group A) or deficient (group B) phospholipase activity were then analysed for phospholipase B1 (PLB1) gene expression both in egg-yolk-agar and yeast extract/peptone/dextrose (YPD) broth media. A total of four virulence attributes of these two groups were in turn characterized, namely their germ-tube formation, cell-surface-hydrophobicity (CSH), adhesion to buccal epithelial cells (ABEC) and haemolysin production, and these factors were subsequently correlated with PLB1 expression. In the phospholipase-producing isolates (group A) a positive correlation was demonstrated between phospholipase production and the degree of PLB1 expression in YPD medium (r ¼ 0 . 96, P , 0 . 01). No such association was observed in group A isolates for PLB1 expression in egg-yolk-agar medium. Further, PLB1 expression in egg-yolk agar was less than that in YPD medium, although a positive correlation was seen between the expression levels on regression analysis (r ¼ 0 . 86, P ¼ 0 . 026). Surprisingly, however, no significant associations were observed in either growth media between PLB1 expression and any of the four pathogenic attributes examined (P , 0 . 001). A significant correlation was seen between CSH and ABEC (r ¼ 0 . 74) in group A isolates. The phospholipase-deficient group B, however, demonstrated a significant correlation between the latter parameters (r ¼ þ0 : 50) and also between germ-tube formation and ABEC (r ¼ À0 : 59), and germ-tube formation and haemolysin production (r ¼ þ0 : 31). It appears that in oral C. albicans isolates in HIV infection there may be no significant association between the degree of PLB1 expression and other widely recognized major virulence attributes.