Pseudomonas aeruginosa Alters Critical Lung Epithelial Cell Functions through Activation of ADAM17

Abstract

Severe epithelial dysfunction is one major hallmark throughout the pathophysiological progress of bacterial pneumonia. Junctional and cellular adhesion molecules (e.g., JAMA-A, ICAM-1), cytokines (e.g., TNFα), and growth factors (e.g., TGFα), controlling proper lung barrier function and leukocyte recruitment, are proteolytically cleaved and released into the extracellular space through a disintegrin and metalloproteinase (ADAM) 17. In cell-based assays, we could show that the protein expression, maturation, and activation of ADAM17 is upregulated upon infection of lung epithelial cells with Pseudomonas aeruginosa and Exotoxin A (ExoA), without any impact of infection by Streptococcus pneumoniae. The characterization of released extracellular vesicles/exosomes and the comparison to heat-inactivated bacteria revealed that this increase occurred in a cell-associated and toxin-dependent manner. Pharmacological targeting and gene silencing of ADAM17 showed that its activation during infection with Pseudomonas aeruginosa was critical for the cleavage of junctional adhesion molecule A (JAM-A) and epithelial cell survival, both modulating barrier integrity, epithelial regeneration, leukocyte adhesion and transepithelial migration. Thus, site-specific targeting of ADAM17 or blockage of the activating toxins may constitute a novel anti-infective therapeutic option in Pseudomonas aeruginosa lung infection preventing severe epithelial and organ dysfunctions and stimulating future translational studies.

Keywords: exosomes; junctional molecules; lung infection; metalloproteinase; proteolysis; regeneration.

Figure 1

Regulation of a disintegrin and metalloproteinase (ADAM) 17 protein expression and maturation in bacterial infection. A549 (A–D) or human small airway epithelial cells (HSAEpC) (E) were grown until confluence. Cells were either left unstimulated or infected with either Pseudomonas aeruginosa (P. aeruginosa, multiplicity of infection (MOI) 5) (A,E) or Streptococcus pneumoniae (S. pneumoniae, MOI 5) (B), stimulated with Exotoxin A (ExoA, 100 ng/mL) (C) or infected with heat-inactivated (HI) P. aeruginosa (MOI 5) (D). Samples were taken after 4 h of incubation, and the protein expression and maturation of ADAM17 were analyzed through Western blot with an antibody against the C-terminal domain of ADAM17 in comparison to the protein expression of glyceraldehyde-3-phosphat dehydrogenase (GAPDH, internal loading control). Densitometry was used to evaluate band intensities, which were further normalized to the expression of the unstimulated cells. Quantitative data are shown as means + SD of three independent experiments. Asterisks indicate significance difference to the control calculated using two-tailed two samples t-test (** p < 0.01, *** p < 0.001, **** p < 0.0001, n.s. not significant).

Figure 2

Activation and exosomal release of ADAM17 by P. aeruginosa and ExoA. (A,B) A549 cells expressing plasmid-encoded TGF-alpha (TGF-a) coupled with alkaline phosphatase (AP, N-terminal) were pre-incubated for 30 min with the ADAM17 inhibitor TAPI-1 (10 µM) or DMSO (0.1%). Subsequently, cells were left untreated, challenged with P. aeruginosa (A, MOI 5) or ExoA (B, 100 ng/mL) for 4 h. The activity of alkaline phosphatase was analyzed in both cell lysate and the medium as an indicator for TGF-α cleavage and release. Quantitative data are shown as means + SD of three independent experiments. Asterisks indicate significance among treated cells calculated using two-way ANOVA and Tukey post-test (** p < 0.01, *** p < 0.001, **** p < 0.0001). (C,D) A549 cells were either left untreated or challenged with P. aeruginosa (MOI 5) followed by a differential centrifugation of the supernatant (300, 1000, 10,000, 100,000 g). The resulting pellet from each centrifugation step was lysed with SDS buffer and analyzed by Western blot (C) developing against the C-terminus of ADAM17, Flotiline-1 and CD9 (exosome markers). The pellet obtained from the 100,000 g centrifugation in step C, containing extracellular vesicles (EVs), was used to further separate the EVs according to their density. The resulted fractions were analyzed by Western blot and probed against ADAM17, Flotiline-1 and CD9 (D). Representative blots of at least three independent experiments are shown. No ADAM17 expression in EVs/exosomes could be observed.

Figure 3

ADAM17 induces protein permeability during P. aeruginosa infection. shRNA sequences against ADAM17 (A17KD1 or A17KD2), delivered by lentiviral particles, were used to mediate ADAM17 deprivation in A549 cells while an unspecific shRNA sequence served as scramble control (scr). (A–D) Cells were seeded on transwells until formation of a monolayer and pre-incubated with DMSO (0.1%) or TAPI-1 (10 µM). Subsequently, cells were either left untreated, challenged with P. aeruginosa (A/B, MOI 5) or stimulated with ExoA (C/D, 100 ng/mL) for 4 h in the absence of presence of 10 µM TAPI-1. Subsequently, the upper chamber’s medium was replaced by 70-kDa TRITC-dextran and FITC-albumin suspension (1 mg/mL and 0.25 mg/mL, respectively, in PBS supplemented with 0.2% BSA), and the permeability was evaluated by the diffusion of the suspension to the lower chamber. The percentages of both total and paracellular permeabilities are shown relative to an empty transwell (maximal permeability, 100%). Quantitative data are shown as mean + SD of three independent experiments. Asterisks indicate significance among treated cells calculated using one-way ANOVA and Tukey post-test (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001). In both cases, we observed a significant reduction in protein permeability either upon infection with P. aeruginosa or stimulation with ExoA. Thus, the lack of ADAM17 induction may improve barrier function.

Figure 4

Impact of ADAM17 on epithelial regeneration. shRNA sequences against ADAM17 (A17KD1 or A17KD2), delivered by lentiviral particles, were used to mediate ADAM17 deprivation in A549 cells while an unspecific shRNA sequence served as scramble control (scr). Cells were grown to confluence and preincubated with either DMSO (0.1%) or TAPI-1 (10 µM) for 30 min. (A,B) Cells were treated with mitomycin (5 µg/mL) for 2 h to avoid cell proliferation and then either challenged for 4 h with ExoA (100 ng/mL) or left untreated. Subsequently, the stimulant was removed, an automated scratch was performed and the wound closure was monitored for 24 h using a live cell imaging system. The percentage of wound closure was calculated relative to a fully closed wound. Example images are shown in (A). (C,D) Cells were challenged with P. aeruginosa for 4 h or left untreated. After 4 h, the cells were lysed and the expression of JAM-A was analyzed by Western blot using an antibody against the N-terminal domain (extracellular part). GAPDH served as a loading control. An exemplary Western blot is shown in (C). Quantitative data are shown as mean + SD (n = 7 in (B), n = 5 in (D)). Asterisks indicate significance among treated cells calculated using one-way ANOVA and Tukey post-test (A,B) or two-tailed two samples t-test for (C,D) (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

Figure 5

ADAM17 regulates THP-1 cells transepithelial migration and adhesion during P. aeruginosa infection. shRNA sequences against ADAM17 (A17KD1 or A17KD2), delivered by lentiviral particles, were used to mediate ADAM17 deprivation in A549 cells while an unspecific shRNA sequence served as scramble control (scr). (A,B) Cells were seeded on transwells until formation of a monolayer and pre-incubated with DMSO (0.1%) or TAPI-1 (10 µM). Subsequently, cells were left untreated, challenged with P. aeruginosa (A, MOI 5) or stimulated with ExoA ((B), 100 ng/mL) for 4 h in the absence or presence of 10 µM TAPI-1. The random and the chemotactic (against 3 nM CCL2) transepithelial migration were evaluated by the addition of 2 × 10⁵ THP-1 cells to the upper chamber. The number of transmigrated THP-1 cells was analyzed after 45 min by evaluation of endogenous β-glucoronidase activity in the lower well. (C,D) Cells were challenged with P. aeruginosa for 4 h or left untreated. Subsequently, 5 × 10⁵ fluorescently labeled THP-1 cells were added, centrifuged at 300× g for 3 min and washed with warm PBS before measuring the fluorescence intensity of the adhered cells. Quantitative data are shown as mean + SD of three independent experiments. Asterisks indicate significance as indicated by lines calculated using one-way ANOVA and Tukey post-test (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

Figure 6

Reduced ADAM17 activity increases epithelial cell survival. (A–F) A549 cells were grown until confluence, stained with the cell membrane permeable dye NucRed™ Live 647 for 15 min and either left untreated or challenged with P. aeruginosa (A-C, MOI 5) or stimulated with ExoA (D–F, 100 ng/mL) in the presence of the cell membrane impermeable SYTOX™ green. SYTOX™ green fluorescence was evaluated every 30 min from each cell. Quantitative data are shown as means + SD of three independent experiments (A/D, representative image; B/E, fluorescence intensity over time of one experiment; C/F, area under the curve (AUC) of SYTOX™ green fluorescence over time of three independent experiments). Asterisks indicate significance among treated cells calculated using two-way ANOVA and Tukey post-test (*** p < 0.001, n.s. not significant).

Figure 7

Model of ADAM17 activation during P. aeruginosa infection. S. pneumoniae and P. aeruginosa are recognized through TRL2 and TLR4. ADAM17 activation occurs in a cell-associated manner dependent on the secreted toxin repertoire. Amongst those, ExoA induces the cleavage of junctional adhesion molecules such as JAM-A leading to increased permeability and impairment of regeneration (A). Shedding of JAM-A and especially of other adhesion molecules (e.g., VCAM-1, ICAM-1), induced through different exoenzymes (e.g., ExoU), may lead to reduced adhesion of monocytes, limiting the transmigration (B). These effects are strongly correlated to reduced lung epithelial cell survival, potentially mediated through TNFR1 shedding (C).