(A–D) Polarized NHBE were treated with or without LT for 48 h. (A) The epithelial layer was subjected to resistance measurements (left panel) or incubated with FITC-albumin on the apical side for permeability measurements (right panel). Data are represented as mean+/−SEM of 3–5 inserts/per condition, n = 3. (B) TEM images of crosscuts obtained from untreated (-) and LT-treated lung epithelial layers. Crosscuts with ciliated cells on the apical side are depicted. Scale bar represents 10 µm. (C) Confocal images of the apical side of untreated and LT-treated lung epithelial layers using a tight junction marker (ZO1, green). Z-series was analyzed with Imaris5 (Bitplane Inc., MN). Disruptions of multicellular junctions are indicated by white arrows. The scale bar represents 50 µm. (D) LDH release of untreated and LT-treated NHBE 3D layers. Data are represented as mean+/−SEM of 3 inserts, n = 3.

(A, C) Immunofluorescence images of NHBE cells with or without LT treatment (48 h). Cells were stained for actin (A, red) or α-tubulin (C, green). Scale bar represents 20 µm. (B, D) Cells with or without LT treatment were incubated for 1 h with 1 µg/ml cytochalasin B (CytoB) (B) or 10 nM nacodazole (Nac) (D) and stained for actin (red, B) or acetylated tubulin (D, green). Scale bar represents 50 µm. Remaining actin stress fibers (B) or stable microtubules (D) were quantified in 500 cells/per condition. (E) Confocal images of NHBE cells treated with or without LT and stained with vinculin (green). Scale bar represents 50 µm. (F) Fractionation of LT-treated and untreated (-) NHBE cells into cytosol (CYT), membrane (MEM) and cytoskeletal (CSK) fractions. Immunoblotting was performed using the indicated antibodies. Rho GDI served as cytosol marker.

(A) Live cell wound healing assay was performed in untreated or LT-treated airway cells (SALE) expressing non-cleavable MKK1/2, MKK3/6 or MKK4/7 mutants or EGFP alone as control. Cells were imaged at time zero and 24 h after the scratch. Scale bar represents 50 µm. Quantification of experiment is shown in lower panel. Data (mean+/−SEM) in each experiment were obtained from 10 different wound areas, n = 3. The overall analysis of variance between all groups was highly significant: F_7,72 = 49.6, p<0.0001. (B) Chemotaxis of untreated and LT-treated airway cells expressing non-cleavable MKK1/2, MKK3/6 or MKK4/7 using Boyden chambers (migration for 21 h). Cells on the bottom of the filter were stained with DAPI (blue). Upper panel depicts untreated cells, lower panel LT-treated cells. Scale bar represents 50 µm. (C) A representative frame of bright field live cell movies of untreated and LT-treated airway cells expressing EGFP or non-cleavable MKK1/2-/EGFP after global growth factor stimulation. Scale bar represents 20 µm. Complete movies (S9–12) can be found in Supplemental data.

(A, B) NHBE cells were treated with or without LT for 48 h and stained with ZO1 (A) or E-cadherin (B). Confocal images were taken consecutively. Scale bar represents 50 µm. (C) Fractionation of LT-treated and untreated (-) NHBE cells into cytosol (CYT), membrane (MEM) and cytoskeletal (CSK) fractions. Immunoblotting was performed using anti-E-cadherin and anti-ZO1. Rho GDI served as cytosol marker.

(A) A representative frame of bright field live cell movies of untreated and LT-treated NHBE (48 h) after global growth factor stimulation (see Methods). Scale bar represents 20 µm. The complete movies (Movie S1 untreated and Movie S2 LT-treated) are in Supplemental data. (B) One representative frame of bright field live cell movies of untreated (-), LT-treated (48 h) or MKK/MAPK inhibitor-treated (48 h) NHBE cells after global growth factor stimulation (see Methods). Scale bar represents 20 µm. The corresponding movies (S3–8) can be found in the Supplemental part. (C) Chemotaxis of untreated (-), LT-treated (48 h) or MKK/MAPK inhibitor-treated (48 h) NHBE in Boyden chambers. 3INH is a combination of three inhibitors (U0126, SB203580, JNKII). Cells were stained with rhodamine phalloidin (red) and DAPI (blue). Upper panels depict cells on the top of the membrane after 1 h of plating before growth factors were added to the bottom chamber. The lower panels show cells on the bottom of the filter 21 h after addition of growth factors. Scale bar represents 50 µm. (D) Immunofluorescence imaging of wound healing assay using polarized NHBE (actin, red). Untreated (-) 0 h image displays the original scratch, whereas all other images were taken 24 h later. Cells were left untreated or treated with LT, catalytically inactive LT (E687C), 3INH combination, or inhibitors for MKK1/2 (U0126), p38 (SB203580) or JNK (JNKII) for 48 h before the scratch was performed. Scale bar represents 100 µm. (E) Quantification of experiment shown in (D). Data (mean+/−SEM) are represented as width of the wound of 4–8 different areas with 2 inserts/condition. The overall analysis of variance between all groups was highly significant: F_6,604 = 675.4, p<0.0001.

(A) Permeability measurements were performed in the ex vivo differentiated NHBE treated with or without LT (1 µg/ml LF/PA), a three inhibitor combination (3INH / U0126, SB203580, JNKII) or catalytically inactive LT (1 µg/ml LF E687C/PA) for 48 h. (B–D) Polarized NHBE layers were treated for 1 h with KGF (250 ng/ml) or dexamethasone (10 µM) before LT (1 µg/ml) was added for 48 h. Layers were analyzed for resistance (B), permeability (C) and LDH release (D). Data (A–D) are represented as mean+/−SEM of 3 inserts/condition. The overall analysis of variance in A–D between all groups was highly significant: (A) F_3,10 = 145.6, (B) F_5,19 = 22.6, (C) F_5,12 = 36.4, (D) F_5,12 = 63.3, p<0.0001.