Phagocytosis of Aspergillus fumigatus conidia by primary nasal epithelial cells in vitro

Fungal strains and preparation of conidia

Conidia were obtained from A. fumigatus strain IP 2279.94 (Pasteur Institute), originally isolated from a patient with IA, and from P. chrysogenum strain IP 1652.86 (Pasteur Institute) isolated from the environment. Fungi were cultured on 2% malt agar slants (Bio-Rad, Marnes-la-Coquette, France). Conidia were harvested after 7 days at 30°C by rinsing the slants with phosphate-buffered saline (PBS) supplemented with 0.1% Tween 20. Conidia were obtained after filtration through a 40 μm pore-size cell strainer (Millipore, Molsheim, France) to remove mycelium and suspended in cell culture medium.

HNEC primary cultures and infection with conidia

The HNEC culture was adapted from a culture model originally developed with human tracheo-bronchial cells [30]. Nasal polyps were harvested from patients with nasal polyposis undergoing ethmoidectomy at the Intercommunal Hospital of Créteil. Preparation of HNEC cultures was performed as described previously [13, 14]. The transport medium, DMEM/F12 (Invitrogen, Cergy-Pontoise, France), consisted of a mixture of Dulbecco's modified Eagle's nutrient medium (Life Technologies, France) and Ham's-F12 nutrient medium (Life Technologies, France) supplemented with antibiotics (100 U/ml penicillin, 100 mg/ml streptomycin, 100 mg/ml gentamicin, 2.5 μg/ml amphotericin B). The polyps were immediately transported to the laboratory and stored at 4°C for 2 h. Each nasal polyp was rinsed in PBS-antibiotics with 5 nM dithiothreitol and placed overnight at 4°C in a PBS-antibiotics solution containing 0.1% pronase (Sigma, France). The sample then was rinsed in DMEM/F12 with 5% fetal calf serum (FCS). After centrifugation (1000 g, 7 min), the cell pellets were suspended in a 0.25% trypsin solution (Life Technologies, France) diluted in DMEM/F12 with 5% FCS for 3 min, centrifuged and then suspended in DMEM/F12-5% FCS. Finally, 10⁶ cells were plated in 12-mm insert wells (Transwell, Costar, Dutscher, France) with micropore membranes coated with type IV collagen (Sigma, France) and incubated at 37°C in 5% CO₂. After 24 h, the liquid medium was removed from the apical compartment to obtain an air-liquid interface. The basal compartment was filled with 1 ml of HNEC culture medium, consisting of DMEM/F12 with 2% ultroser G (Life Sciences, Cergy-Pontoise, France) with antibiotics (100 U/ml penicillin, 100 mg/ml streptomycin, 100 mg/ml gentamicin). The HNEC culture medium was changed daily and the electrophysiological properties of the HNEC were checked twice a week using a microvoltmeter (EVOM^® World Precision Instruments, Aston-Stevenage, UK). All experiments were performed on 14-day-old HNEC cultures in which cell differentiation was well established.

Each HNEC well was seeded with 10⁷ conidia. To speed up sedimentation and to avoid prolonged immersion of the cells, the wells were centrifuged at 60 g for 5 min and washed three times to remove unbound conidia. For each experiment, the number of conidia bound to the HNEC was evaluated by subtracting all unbound conidia collected in the three washings from the total number of conidia added to the cells. After removal of unbound conidia, the wells were incubated for the indicated times at 37°C in a 5% CO₂ atmosphere.

TEM and confocal studies

HNEC infected with A. fumigatus conidia were fixed for 2 h at 4°C in 0.045 M cacodylate buffer, pH 7.4, containing 2.5% glutaraldehyde, and post-fixed in buffered 1% osmium tetroxide for 90 min, stained in 2% uranyl acetate and dehydrated through graded solutions of ethanol. Cell culture membranes were removed from the inserts and embedded in Epon. Semi-thin sections (1 μm) were stained with toluidine blue and examined by light microscopy to evaluate the specimen prior to thin sectioning. Ultrathin sections (80 nm) were examined on a transmission electron microscope (Philips EM 301) at an acceleration of 80 kV and a final magnification ranging from × 2500 to × 30 000.

For the confocal microscopy study, freshly harvested conidia were labeled with fluorescein isothiocyanate (FITC) (Sigma, Saint-Quentin Fallavier, France) in 0.1 M carbonate buffer for 1 h with shaking at 37°C. The HNEC cells with FITC-labeled conidia were fixed for 1 h at 4°C in 2% paraformaldehyde, permeabilized with 2% methanol for 1 h at -20°C, and stored in PBS at 4°C. After incubation in 5% goat serum and 5% human serum in PBS to avoid non-specific binding of antibodies, HNEC were incubated for 90 min in 1:50 diluted mouse monoclonal anti-human cytokeratin specific antibody (mAb) (Immunotech, Marseille, France) to label HNEC. They were then washed and finally incubated for 90 min with 1:200 diluted rhodamine-conjugated goat anti-mouse antibody (Immunotech). After washing with PBS, the membranes with HNEC were removed from the inserts, mounted in Mowiol (Sigma), and examined with a Zeiss Laser Scanning 410 Microscope (LSM) (Göttingen, Germany). Images were obtained with LSM 3.98 Software (Zeiss) and further processed with Adobe Photoshop software.

Immunolabeling with LAMP1 marker

To study the late endosome/lysosome compartment, HNEC were fixed for 1 h at 4°C in 2% paraformaldehyde and permeabilized with PBS containing 0.05% saponin as described previously [6]. HNEC were then incubated for 30 min at room temperature with a 1:100 diluted mouse anti-human LAMP1 mAb specific for human lysosomal membrane associated protein (BD Pharmingen, Franklin Lakes, NJ, USA). Purified mouse IgG1 isotype (Pharmingen) was used as a negative control. After washing, HNEC were incubated for 30 min with 1:200 diluted anti-mouse Texas-red-conjugated secondary antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, USA). After washing, cells were mounted in Mowiol, sealed on microscope slides with nail varnish, and stored at 4°C until observation. Slides were examined with a fluorescence Leica DL microscope coupled to a Coolsnap Cf monochrome camera (Photometrics, Roper Scientific, Evry, France). Images were obtained with the Metaview program (Universal Imaging Corporation, Downingstown, PA, USA) and further processed with Adobe Photoshop software.

Number of HNEC-associated conidia

HNEC cultures with unlabeled conidia were dissociated for 30 min at 37°C with 0.1% collagenase (Sigma, France) and 15 min at 37°C with 0.25% trypsin (Sigma, France). Reactions were stopped with 10% FCS. After centrifugation (5 min, 500 g), cells were resuspended, smeared and stained with May-Grünwald-Giemsa. One hundred randomly chosen HNEC were examined microscopically to calculate the number of HNEC with at least one associated conidium. We also determined the average number of conidia associated with each HNEC by counting 100 HNEC associated with at least one conidium. The experiments were repeated twice for each fungal species with two different primary cultures.

Percentage of phagocytosed conidia and viability of conidia

To quantify phagocytosis among the bound A. fumigatus conidia, we used FITC-labeled conidia and an additional staining of non-phagocytosed conidia adapted from the method of Sturtevant et al. [29]. For specific labeling of non-phagocytosed conidia, HNEC were incubated for 30 min with 1:50 diluted rabbit anti-A. fumigatus conidia antiserum. After washing, HNEC were incubated for 30 min with 1:200 diluted Texas-red-conjugated anti-rabbit antibody (Sigma, France). Therefore, phagocytosed conidia, inaccessible to the anti-conidia antibody, were only FITC-labeled, whereas non-phagocytosed conidia were labeled with both FITC and Texas-red. Samples were viewed with a Zeiss Axiophot microscope equipped with epifluorescence filters. Extracellular and intracellular conidia were counted after merging the red and green channels in eclipse. The percentage of phagocytosed conidia was calculated as: number of FITC-positive and Texas-red-negative conidia/number of FITC-positive conidia × 100.

To quantify phagocytosed conidia of both fungal species we used the fluorescent brightener Blankophor-P-fluessig^® (4,4'-Bis [(4-anilino-6-substituted-1,3,5-trazine-2-yl)] amino stilbene-2,2'-disulfonic acid), kindly provided by Bayer, Leverkusen, Germany. This brightener is a stilbene compound like Calcofluor white and Uvitex, and presents no toxicity to mammalian cells [22]. The brightener stains only external conidia. Tenμl of the dye was diluted in 1 ml of culture medium and 300 μl of this dilution were added to the wells and incubated for 10 min at 37°C to allow staining of non-internalized conidia. The wells were washed twice to recover any remaining unattached conidia and to remove the dye. The HNEC were then lysed with 300 μl distilled water, and the remaining cells were scraped from the inserts and recovered in 300 μl. On excitation with ultraviolet light below 400 nm, the brightener emits a very intense bluish yellowish white light [31]. The filter combination included a barrier filter at 420 nm. Thus, the blue conidia were counted as non-internalized whereas the white conidia were assumed to have been internalized.

To test the viability of conidia obtained after brightener staining, propidium iodide (PI) staining (Sigma) was used. The same field was examined successively using different filter combinations. For the brightener staining, the filter combination included a barrier filter at 420 nm and allowed excitation at wavelengths lower than 400 nm. The other filter used was λ excitation at 540 nm and λ emission ≥ 570 nm, for PI staining. Several fields were read to count 200 conidia. Images were obtained using an Axioskop 40 microscope (Zeiss) with a digital camera DX20H (Kappa, Donneville, France) fitted to the microscope.

Statistical analysis

Statistical analysis was performed using EPI-INFO 6.04c computer software (Centers for Disease Control and Prevention, Atlanta, GA, USA). Each value was derived from triplicate data points and was expressed as the mean ± SEM. The mean counts of A. fumigatus and P. chrysogenum conidia were compared using the Student's t test or Kruskall-Wallis test. A p value < 0.05 was considered to be statistically significant.