Attachment of Candida albicans, an important opportunistic pathogen, to host tissues is an initial step in the development of the infection. The events occurring in the fungal and in the host cells after interaction are poorly understood. In this study we concentrated on the events occurring in the mammalian cells after the interaction with Candida, with emphasis on the cytoskeleton actin. Human cell line cells (HEp2) were exposed to C. albicans or C. albicans-secreted material (culture filtrate) (actin-rearranging Candida-secreted factor, arcsf). The HEp2 cells were examined for cellular changes using confocal laser microscopy (CLSM), transmission and scanning electron microscopy (TEM and SEM). The CLSM studies, using fluorescein isothiocyanate-labeled C. albicans and rhodamine phalloidin actin staining, revealed yeasts adhering to the HEp2 cells or internalized into the cells, with actin surrounding the fungi. Furthermore, actin rearrangement from filamentous network to actin aggregates was noticed. Interaction between the HEp2 cells and C. albicans could be demonstrated also by SEM and TEM after a 2–4-h exposure of the cells to the fungus. Yeasts and hyphae were found attaching to the surface and within the cells. CLSM studies revealed that exposure of HEp2 cells to arcsf was also followed by cellular actin rearrangement, reduced membrane ruffling and decreased cellular motility. The effect was dose- and time-dependent. All these data indicate that the interaction of Candida with HEp2 cells involves signaling events and affects the cellular actin.
Candidiasis, which is caused by Candida albicans and other Candida species, is considered one of the most frequent fungal infections. All of the tissues and organ systems are subject to invasion, and the pathology evoked is as variable as are the clinical syndromes. In its mucocutaneous forms it is a most common infection, and as a deep-seated mycosis it poses a serious problem among compromised individuals .
The fungus C. albicans is a versatile organism that thrives as a commensal of the human gastrointestinal tract and possesses the ability to invade host tissues under the appropriate environmental conditions .
The most extensively investigated putative virulence factors of Candida include the ability of the fungus to attach to host tissues, and the potential to penetrate into host tissues, which may follow the attachment step . Attachment of C. albicans to mammalian cells, as in other microbial systems, is regarded as the initial step in the infectious process, enabling the microorganism to survive on the host and colonize his tissues, resulting eventually in development of candidiasis [3,4]. The adhesins of C. albicans are diverse, reflecting the ability of the fungus to colonize and invade a variety of host cells and tissues. The host cell ligands recognized by C. albicans are also diverse, but appear to be broadly classified into at least two types: glycosides and in part the peptides of several extracellular matrix proteins [4,5]. Although adherence of C. albicans has been studied extensively by different research groups, the events subsequent to adherence are not entirely understood. It is believed that the conversion from yeast to hyphae and secretion of essential enzymes promote the invasion of the microoorganism into subepithelial or subendothelial tissues. These events indicate that a contact-induced regulation of gene expression may be occurring, and that C. albicans can respond to its host in various ways . Data indicate that following adherence specific proteins are made by the fungus, and further phosphorylation of proteins occurs, indicating possible signaling.
The data regarding the events in the mammalian cells after interaction with Candida are scarce [7–9], unlike the information in other microbial systems [10–14]. It was shown for various microbial systems that the cellular actin is involved in microbial internalization, but the exact role of actin in this process is not entirely understood. It was also shown that there is a dramatic change of actin organization upon phagocytosis, but again, it is not clear if the phagocytic process is dependent on actin polymerization.
Electron microscopic (EM) observations made in our laboratory of C. albicans organisms adhering to epithelial cells suggested that the mammalian cells do respond by physical changes noticeable by EM. In view of the reported involvement of actin in microbial–mammalian interactions, and aiming to further elucidate the events occurring in the mammalian cells following the interaction with C. albicans, we initiated a study employing confocal laser scanning microscopy (CLSM) to investigate possible effects of the fungal interaction on the mammalian cytoskeleton, particularly the actin. CLSM is a technology which by use of fluorescent staining can enable the in-depth study of microbial–mammalian cell interaction, as it facilitates the demonstration and localization of cellular components which are associated with the interaction .
Thus, the present report describes results obtained from a study using CLSM, transmission and scanning electron microscopy (TEM and SEM) to investigate the effects on cellular actin following the interaction between C. albicans or factor(s) secreted by the fungus and the human cell line HEp2.
2Materials and methods
2.1Candida growth conditions, fluorescein isothiocyanate (FITC) labeling and extraction of Candida-secreted material (actin-rearranging Candida-secreted factor, arcsf)
C. albicans CBS-562 was maintained as described previously . For the interaction with HEp2 cells and FITC labeling the fungus was grown in M-199 (Biological Industries) at 28°C for 18 h under constant shaking. The cells were pelleted, washed three times in phosphate-buffered saline (PBS, pH 7.2) and suspended to the desired concentration.
For FITC labeling C. albicans cells at a concentration of 109 microorganisms ml−1 were incubated for 24 h at room temperature in the dark with 0.1 mg ml−1 FITC (Sigma) in a solution containing 50 mM NaHCO3 and 100 mM NaCl (pH 9.0). The yeasts were then washed with PBS to remove free FITC and resuspended in PBS to a concentration of 108 microorganisms ml−1.
To obtain secreted material, C. albicans was grown in yeast nitrogen base (Difco) without amino acids for a period of 7 days, at 28°C under constant shaking. The culture was filtered and the filtrate was collected. The filtrate was dialyzed against distilled water for 24 h using dialysis tubes (cutoff 6000–8000) and lyophilized. The lyophilized material was designated actin-rearranging Candida-secreted factor (arcsf; see Section 3).
2.2Interaction of HEp2 cells with Candida or with arcsf
HEp2 human larynx epithelial cells were maintained in Dulbecco's modified Eagle's medium ((Biological Industries) with 2 mM L-glutamine and 10% fetal calf serum supplemented with 200 μg of streptomycin and 200 IU of penicillin ml−1.
For interaction assays the HEp2 cells were grown in 8-well tissue chamber slides (Permanox Slide, Nunc) for confocal microscopy, or in 6-well plates (Sterilin) for electron microscopy, for 24 h, or until the culture was confluent. The cells were washed with PBS and then exposed for various times to FITC-labeled or unlabeled C. albicans organisms, or to arcsf (various concentrations).
2.3F-actin staining and CLSM analyses
F-actin staining was carried out using rhodamine phalloidin (Rh-phalloidin) (Molecular Probes, Eugene, OR, USA) as previously described  with slight modifications. Following incubation with FITC-labeled C. albicans, arcsf or no treatment, the HEp2 cells were fixed, permeabilized and stained in a one-step procedure, as suggested by the manufacturer (Molecular Probes; simultaneous fixation protocol). Briefly, treated and untreated cells were incubated for 20 min with 3.7% formaldehyde containing 100 μg ml−1 lysophosphatidylcholine (Sigma, St. Louis, MO, USA) and 0.3 μM Rh-phalloidin and then washed. Slides and coverslips were mounted using Gel Mount (Biomeda, Foster City, CA, USA). Analyses were performed using a Zeiss confocal laser scanning microscope. The Zeiss (Oberkochen, Germany) LSM 410 is equipped with a 25-mW krypton–argon laser and a 10-mW helium–neon laser (488, 543). Images were stored on an optical disk drive and printed using a Codonics NP1600 printer (Codonics, Middleburg Heights, OH, USA).
The specimens (HEp2 cells and C. albicans cells after interaction) were fixed in 1% glutaraldehyde overnight at room temperature, postfixed with 2% osmium tetroxide solution for 2 h at room temperature, then treated with a graded series of ethanol and embedded in Epon 812. Ultrathin sections were stained with uranyl-acetate and examined in a Joel 100B at 80 kV.
The specimens were fixed in 1% glutaraldehyde overnight at 4°C, postfixed with 1% OsO4 solution for 2 h at room temperature and dehydrated in an alcohol series, air-dried, coated with gold (15 nm) and examined with a Joel 840 SEM at 20 kV.
2.6Ruffling and cellular movement
Ruffling was measured using CLSM phase-contrast time-lapse microscopy as previously described . Using the CLSM program we determined the cellular ruffling that occurred in 15 s by subtracting the 10-min images from the images of 10 min+15 s, for untreated and treated cells. The resulting areas of cellular ruffling are shown as red areas superimposed on 10 min+15 s untreated and treated images. Effect of arcsf on cellular movement was measured using phase-contrast images of 45-s intervals.
3.1Interaction of C. albicans with HEp2 cells – CLSM studies
HEp2 cells were inoculated with FITC-labeled C. albicans for 2–4 h. Following the interaction the cells were fixed, stained for F-actin with Rh-phalloidin, and analyzed by CLSM (Fig. 1). The CLSM analysis revealed C. albicans adhering to HEp2 cells or internalized within the cells. Internalization of the FITC-labeled C. albicans into HEp2 cells is demonstrated by the intercalation of the C. albicans into F-actin fibers located inside the cells (Fig. 1a).
A Z-section through the specimen, in which the scan plane is depicted by a yellow line (Fig. 1a,a′), was carried out. The image generated from this focal plane (Fig. 1a′) shows C. albicans which are intercalated with the F-actin cytoskeleton (red).
To further investigate the pattern of actin association with C. albicans in HEp2 cells, a co-localization analysis of the regions of interaction was carried out (Fig. 1b–g). A dense cap-like structure of actin arrangement is observed in the F-actin staining (Fig. 1d). The overlaid image (Fig. 1c) shows a yellow ring surrounding the fungus.
A graphic representation of the images was generated using the Zeiss co-localization procedure (Fig. 1g). This graph depicts fluorescence intensities of the F-actin staining (X coordinate, red staining) and the C. albicans staining (Y coordinate, green fluorescence) of all the image pixels. The graph demonstrates co-localized C. albicans with F-actin depicted by the points along the 45° angle. An image which contains the selected pixels in their original X–Y location shows co-localization of C. albicans and F-actin surrounding the microorganism. Taken together these results demonstrate that actin rearrangement was induced in HEp2 cells by C. albicans. Furthermore, actin rearrangement was noted also in cells not having direct contact with the fungus. This observation led us to the assumption that C. albicans produces a factor which affects the cellular actin. This working hypothesis was explored in a later stage of the study, in which HEp2 cells were exposed to the culture filtrate of C. albicans and the effect on actin was assessed.
3.2SEM and TEM studies
The effect of interaction of C. albicans with HEp2 cells on the cellular ultrastructure of the mammalian cells was also studied by means of electron microscopy, both scanning and transmission. Fig. 2 depicts illustrations of SEM and TEM observations following exposure of HEp2 cells to yeasts or hyphae of C. albicans for 2–4 h. The SEM demonstrates C. albicans yeasts and hyphae attaching to HEp2 cells. It also demonstrates that the exposure of HEp2 cells to C. albicans resulted in changes in the mammalian cells. The cells seem to extend ‘ameboid appendages’ towards the fungus. This effect is noted for exposure of cells to both the yeast form and the hyphal form (Fig. 2A,B). Furthermore, the cellular appendages seem to engulf some of the fungi.
The TEM observations corroborate the observations of the SEM, and emphasize that the process is initiated by the fungal attachment, which may be followed by complete internalization of the microorganisms (Fig. 2C,D).
3.3Effect of C. albicans culture filtrate (arcsf) on HEp2 F-actin
Since the effect on actin rearrangement, which was noticed in HEp2, was not solely restricted to cells interacting directly with C. albicans, we examined the possibility that C. albicans secretes factors that induced actin rearrangement. Filtrates from cultures in which the fungus was grown for 7 days were dialyzed and lyophilized.
To study the effects of the Candida culture filtrate material, HEp2 cells were incubated with various concentrations (5–50 mg ml−1) for various time intervals (30 min, 1, 2, 3, 6, and 24 h), stained with Rh-phalloidin and analyzed by CLSM.
Exposure of HEp2 cells to the Candida filtrate material for 1 or 2 h (there was no difference in the effect between 1- and 2-h interaction, so most experiments were done with 1-h interaction) resulted in a dramatic reorganization of actin from the short filamentous actin network ordinarily found in HEp2 cells (Fig. 3A, control) to focal aggregates (Fig. 3B–D). This effect was dose-dependent: detectable at 6.25 mg ml−1 (Fig. 3B) and observed to be most prominent at 25–50 mg ml−1 (Fig. 3C,D). A toxic effect on the cells was noted at concentrations >50 mg ml−1 (data not shown).
The effect on actin was noted already after 30 min of exposure (12.5 mg ml−1) (data not shown). Actin aggregation was most evident after 1-h exposure (Fig. 3B–D). Exposure for more than 2 h resulted in cell damage and detachment from the substrate (data not shown).
These experiments showed that the HEp2 cells demonstrate reorganization of actin from the short filamentous actin network into aggregate forms. Hence, we designated the C. albicans secreted factor arcsf.
3.4Effect of arcsf on cellular movement and ruffling
Ruffling of cell membrane and cell motility are the most prominent manifestations of actin polymerization. To study the effects of arcsf on cellular movement we compared CLSM time-lapse photography (45-s intervals) of HEp2 cells treated for 10 min with various concentrations of arcsf (6, 12, 25 mg ml−1). The untreated HEp2 cells moved extensively and spread on the substrate (Fig. 4B, a–c, see arrow) while the treated cells moved significantly less and pseudopodia formation was rarely detected (Fig. 4B, d–f). The effect was noted immediately after exposure to the treatment with 6 mg ml−1 arcsf and was reversible after washing (Fig. 4B, g–i).
To analyze the effect on membrane ruffling we compared rapid CLSM time-lapse photography (15-s intervals) of untreated and arcsf-treated HEp2 cells. CLSM images of treated and untreated cells after 10 min (Fig. 4A, b,e respectively) and 10 min+15 s were acquired (Fig. 4A, c,f respectively). Using the CLSM program we subtracted the 10-min images from the images at 10 min+15 s (Fig. 4A, c minus b and f minus e). The differences between images were calculated and then areas where changes occurred were superimposed on the corresponding 10-min time images. These areas (shown in red, Fig. 4A, a,d) represent positional changes of cellular organelles and membranes, thus indicating that movement took place. The movement and ruffling in untreated cells were extensive as shown by the significant amount of red in Fig. 4A, a). Furthermore, untreated cells showed extensive formation of motile cell surface protrusions, i.e., ruffling. No significant ruffling or intracellular movement was observed in the arcsf-treated cells, therefore the image scarcely features any red (Fig. 4A, d). The effect was reversible, as deduced from the observation that cells washed with fresh medium regained their ruffling capacity (Fig. 4A, g). These results suggest that Candida produces a factor that alters F-actin organization, reducing cell ruffling and movement.
Actin association with microbial internalization has been shown for several species . Moreover, direct association between internalized bacteria and cellular F-actin in the shape of a cap was also reported for several bacteria . Although actin rearrangement is a common feature of most bacterial internalization events, factors involved in entry are specific to each bacterium. Another step during which pathogens harness the actin cytoskeleton takes place in the cytosol, within which some bacteria  or viruses are able to move. Movement is coupled to a polarized actin polymerization process, with the formation of characteristic actin tails.
This study demonstrates that C. albicans affects the rearrangement of the cellular actin directly by interaction with the cellular actin and by secretion of a factor which alters actin organization. To the best of our knowledge, this is the first report regarding a factor secreted by C. albicans that alters actin organization. The factor also significantly reduced membrane ruffling and cellular movement.
Of interest is a recent study  which investigated the internalization of C. albicans in mouse macrophages and demonstrated that uptake of Candida required intact actin filaments and depended on activity of protein kinase C. The authors suggest that internalizing C. albicans may contribute to its survival. It was also shown in an earlier study by Filler and colleagues  that actin microfilaments were necessary for phagocytosis of C. albicans in an endothelial cell system. Other studies , with different cell systems, showed that hyperpolymerization of actin by jaspamide did not completely block phagocytosis of C. albicans, as was also the case with Proteus mirabilis. These data suggest that different penetration systems may be involved, which possibly are related to different cell systems. Thus, it is too early at this stage of our research to specify the role of actin in the interaction of Candida with mammalian cells, and further experiments toward this end are warranted.
In summary, the experimental data of our study, taken together, indicate that the interaction of mammalian cells with C. albicans is associated with signaling events that occur in the mammalian cells and affect the cellular actin. A better understanding of the interplay between the fungus and host cells might contribute to the elucidation of the pathogenesis of the infection caused by this opportunistic pathogen.