Lipoxin A(4) regulates bronchial epithelial cell responses to acid injury

Abstract

Aspiration of gastric acid commonly injures airway epithelium and, if severe, can lead to respiratory failure from acute respiratory distress syndrome. Recently, we identified cyclooxygenase-2 (COX-2)-derived prostaglandin E(2) (PGE(2)) and lipoxin A(4) (LXA(4)) as pivotal mediators in vivo for resolution of acid-initiated acute lung injury. To examine protective mechanisms for these mediators in the airway, we developed an in vitro model of acid injury by transiently exposing well-differentiated normal human bronchial epithelial cells to hydrochloric acid. Transmission electron microscopy revealed selective injury to superficial epithelial cells with disruption of cell attachments and cell shedding. The morphological features of injury were substantially resolved within 6 hours. Acid triggered and early marked increases in COX-2 expression and PGE(2) production, and acid-induced PGE(2) significantly increased epithelial LXA(4) receptor (ALX) expression. LXA(4) is generated in vivo during acute lung injury, and we observed that nanomolar quantities increased basal epithelial cell proliferation and potently blocked acid-triggered interleukin-6 release and neutrophil transmigration across well-differentiated normal human bronchial epithelial cells. Expression of recombinant human ALX in A549 airway epithelial cells uncovered ALX-dependent inhibition of cytokine release by LXA(4). Together, these findings indicate that injured bronchial epithelial cells up-regulate ALX in a COX-2-dependent manner to promote LXA(4)-mediated resolution of airway inflammation.

Figure 1

Transmission electron micrographs of well-differentiated NHBE cells exposed to acid. A and B: NHBE cells were differentiated in air-liquid interface culture for 21 days. The epithelium comprised ciliated (c), goblet (g), and basal (b) cells. C, D, and E: NHBE cells were transiently exposed to 0.1 N HCl (pH 1.5, 5 minutes) and incubated in fresh medium at neutral pH for 2 hours (F, G) and 6 hours (H, I). Arrows indicate examples of cell injury (C) and arrowheads indicate loss of cell attachments after acid injury (G). Magnifications are indicated.

Figure 2

Injury increases NHBE cell COX-2 expression and activity. A: Time course for COX-2 expression in well-differentiated NHBE cells after transient acid exposure (see Methods). Results show fold induction of COX-2 mRNA by quantitative polymerase chain reaction compared to cells without acid for each time point. COX-2 protein expression was also analyzed 2 hours after acid injury by immunoblot (inset). B: Time-dependent PGE₂ production after acid injury. C: NHBE cells were exposed for 1 hour to a COX-2 selective inhibitor or vehicle, before acid injury (see Methods) and incubated with fresh medium for an additional 2 hours. Results are the mean ± SEM for n = 3, *P < 0.05 (acid compared to control) and **P < 0.05 (COX-2 inhibitor compared to vehicle).

Figure 3

Effect of acid and PGE₂ exposure on ALX expression in NHBEs. A: Time course of ALX mRNA expression in well-differentiated NHBEs after transient exposure to acid. Results show fold induction of ALX mRNA expression compared to cells without acid for each time point. B: NHBEs were incubated with PGE₂ or vehicle for 2 hours, and ALX mRNA expression was determined by semiquantitative polymerase chain reaction (see Methods). Inset is a representative gel of three independent experiments; 1, control; 2, PGE₂ 0.1 ng/ml; 3, PGE₂ 10 ng/ml. C: NHBEs were incubated for 1 hour with a selective COX-2 inhibitor or vehicle and then transiently exposed to acid and incubated with fresh medium for 2 hours. Inset is a representative gel of three independent experiments; 1, control; 2, acid; 3, COX-2 inhibitor + acid. *P < 0.05 (compared to control) and **P < 0.05 (COX-2 inhibitor compared to vehicle) (n = 3).

Figure 4

Impact of LXA₄ on epithelial cell proliferation and inflammatory responses. A: LXA₄, PGE₂ (200 nmol/L), EGF (0.5 ng/ml), or vehicle were added to basal NHBE cells and cell number was determined after 24 hours (see Methods). Values are the mean ± SEM, n ≥ 3, d = 3. B: Cells were exposed (15 minutes) to LXA₄ (100 nmol/L), PGE₂ (200 nmol/L), or vehicle before acid injury, and then incubated in fresh medium for 6 hours (see Methods). IL-6 levels were measured in cell-free supernatants by enzyme-linked immunosorbent assay. Results are expressed as mean ± SEM (n = 3). C: PMNs were exposed (15 minutes, 37°C) to LXA₄, and transmigration toward the chemoattractant LTB₄ (1 μmol/L) was determined by myeloperoxidase activity (see Methods). Results are expressed as percent inhibition of LTB₄-induced PMN transmigration and represent mean ± SEM for n = 3. *P < 0.05 (compared to control), **P < 0.05 (compared to acid), and ^#P < 0.05 (compared to acid + PGE₂).

Figure 5

ALX signals LXA₄-mediated anti-inflammatory actions for airway epithelial cells. A: Expression of ALX mRNA in A549 cells transfected with rhALX cDNA. B: Determination by flow cytometry of rhALX surface expression on A549 cells transfected with pcDNA3 vector alone (left) or vector containing rhALX cDNA (right). hALX on cell surfaces is expressed as a logarithmic increase in mean index of fluorescent intensity when A549 cells were incubated with either anti-hALX (black) or control antibody (white). The x axis represents log increase in fluorescence (fluorescein isothiocyanate), and the y axis represents relative cell number (counts). C: A549 cells transfected with pcDNA3 vector alone or rhALX cDNA were exposed to LXA₄ before acid injury and incubated in fresh medium for 6 hours. IL-6 levels in cell-free supernatants were measured by enzyme-linked immunosorbent assay. Results are expressed as percentage of IL-6 release compared to acid injury alone (100%) and represent mean ± SEM for n ≥ 3. *P < 0.05 when compared with A549 cells exposed to acid. D: A549 transfected with pcDNA3 vector alone (○) or with vector containing rhALX cDNA (•) were exposed to LXA₄ before TNF-α (1 ng/ml) and incubated for 6 hours. IL-8 release was determined by enzyme-linked immunosorbent assay. Results are expressed as percent inhibition of TNF-α-induced IL-8 release and represent mean ± SEM for n = 3. *P < 0.05 compared to vehicle.