MUC1 expression by human airway epithelial cells mediates Pseudomonas aeruginosa adhesion

Abstract

Human MUC1 (Muc1 in animals) is an extensively O-glycosylated membrane-tethered mucin expressed on the surface of epithelial cells and some cells of the hematopoietic system. Recently, we showed that the hamster Muc1 on Chinese hamster ovary (CHO) cells served as a binding site for Pseudomonas aeruginosa (PA) through interaction between bacterial flagellin and the Muc1 ectodomain. Because CHO cells are known to produce an atypical pattern of protein glycosylation, we determined whether or not PA interacted with MUC1 endogenously expressed on human airway epithelial cells. Knock down of MUC1 expression in bronchial (NuLi-1) or alveolar (A549) epithelial cells by RNA interference significantly reduced PA binding to the cells. Conversely, over-expression of MUC1 in HEK293 cells increased bacterial adherence. By confocal microscopy, PA and MUC1 were colocalized on the surface of NuLi-1 cells. Taken together, these results confirm our previous observations in CHO cells and suggest that MUC1 serves as a binding site for PA on the surface of airway epithelial cells, which may have important consequences in the pathogenesis of PA lung infections.

Figure 1

MUC1 siRNA decreases MUC1 expression in NuLi-1 human airway epithelial cells. NuLi-1 cells were seeded at 2.0 × 10⁶ cells/well in 12-well plates, cultured for 24 hr, and transfected with 5 nM of a control siRNA or MUC1 siRNA using 2.0 μl/well of Lipofectamine2000. (A) MUC1 mRNA levels were quantified by real time RT-PCR at 72 hr post-transfection. Each bar represents the mean ± SEM values (n = 3). The Ct value was defined as the number of PCR cycles required for the specific fluorescence signal to exceed the detection threshold value set by the software installed in the iCycler. The levels of MUC1 transcripts were normalized to GAPDH transcripts using the −2^ΔΔCt method. *, significantly decreased mRNA level (p < 0.05). (B) MUC1 protein levels in cell lysates were detected by immunoblotting with MUC1 EC antibody GP1.4 and MUC1 CT antibody CT33. Immunoblotting for beta-tubulin confirmed equal sample loadings. (C) MUC1 protein expression on the cell surface was detected by FACS analysis with GP1.4 and Alexa Fluor 488-conjugated rabbit anti-mouse IgG secondary antibody. (D) A549 cells were seeded at 1.0 × 10⁶ cells/well in 12-well plates and transfected as above except that 20 nM of siRNAs were used. At 72 hr post transfection, cell surface expression levels of MUC1 were determined with the GP1.4 MUC1 EC antibody by FACS.

Figure 2

Transfection of NuLi-1 cells with MUC1 siRNA inhibits PAK binding. NuLi-1 cells were seeded at 2.0 × 10⁶ cells/well in 24-well plates, cultured for 24 hr, and transfected with 5 nM of a control siRNA or MUC1 siRNA using 1.0 μl/well of Lipofectamine2000. At 72 hr post-transfection, the cells were fixed with 2.5% glutaraldehyde, incubated for 40 min at 37°C with the indicated input CFU of ³⁵S-labeled PAK in a total volume of 0.5 ml, the cells were washed to remove unbound bacteria, and PAK remaining in the wells determined by liquid scintillation counting. Each bar represents the means ± SEM (n = 3). The p values comparing control siRNA vs. MUC1 siRNA transfections for each input CFU of PA are indicated.

Figure 3

Transfection of NuLi-1 cells with MUC1 siRNA inhibits GFP-PAO1 binding. NuLi-1 cells were seeded at 2.0 × 10⁶ cells/well in 12-well plates, cultured for 24 hr, and transfected with 5 nM of a control siRNA or MUC1 siRNA using 2.0 μl/well of Lipofectamine2000. At 72 hr post-transfection, the fixed-cells were untreated or incubated for 40 min at 37°C with 4.0 × 10⁷ (A) or 4.0 × 10⁸ (B) CFU of GFP-PAO1 in total volume of 1.0 ml, the cells were washed to remove unbound bacteria, and GFP-PAO1 remaining attached to the cells was determined by FACS analysis. Data were obtained from a total of 1 × 10⁴ viable cells for each analysis. Results are representative of 3 experiments.

Figure 4

Transfection of A549 cells with MUC1 siRNA inhibits GFP-PAO1 binding. A549 cells were seeded at 1.0 × 10⁶ cells/well in 12-well plates, cultured for 24 hr, and transfected with 20 nM of a control siRNA or MUC1 siRNA using 2.0 μl/well of Lipofectamine2000. At 72 hr post-transfection, the fixed-cells were untreated or incubated for 60 min at 37°C with 4.0 × 10⁶ (A) or 4.0 × 10⁷ (B) CFU of GFP-PAO1 in total volume of 1.0 ml, the cells were washed to remove unbound bacteria, and GFP-PAO1 remaining attached to the cells was determined by FACS analysis. Data were obtained from a total of 1 × 10⁴ viable cells for each analysis. Results are representative of 3 experiments.

Figure 5

Over-expression of MUC1 in HEK293 cells increases GFP-PAO1 binding. (A) Confocal microscopy of pcDNA3.1-HEK293 and MUC1-HEK293 cells. (B) Equal protein aliquots of lysates of pcDNA3.1-HEK293 (lane 1) and MUC1-HEK293 (lane 2) cells were analyzed by immunoblotting with MUC1 CT antibody CT33. Immunoblotting for beta-tubulin confirmed equal sample loadings. (C) pcDNA3.1-HEK293 and MUC1-HEK293 cells were fixed with 2.5% glutaraldehyde, incubated for 1 hr at 37°C with saline or 4.0 × 10⁷ CFU of GFP-PAO1, the cells were washed to remove unbound bacteria, and GFP-PAO1 remaining attached to the cells was determined by FACS analysis. Data were obtained from a total of 1 × 10⁴ viable cells for each analysis. Results are representative of 3 experiments.

Figure 6

Colocalization of GFP-PAO1 and cell surface MUC1 on NuLi-1 cells. NuLi-1 cells were cultured in 8-well chambers, incubated with 1.0 × 10⁷ CFU of GFP-PAO1 for 30 min at 37°C, washed 5 times with PBS, fixed with 4% paraformaldehyde, and stained with GP1.4 antibody plus Alexa Fluor 555-conjugated secondary antibody. The cells were visualized by confocal microscopy as follows: (A) GP1.4 antibody (red); (B) GFP-PAO1 (green); (C) DAPI staining of nuclei (blue); (D) overlay of (A), (B) and (C); (E) overlay of (A), (B), and (C) and visualized by z stack imaging. The colocalization of GFP-PAO1 with MUC1 is indicated by yellow staining of the merged images (arrows in (D) and (E)). The dashed line in (D) shows the section used for z stack imaging in (E).