Cigarette Smoke Modulates Repair and Innate Immunity following Injury to Airway Epithelial Cells

Abstract

Cigarette smoking is the main risk factor associated with chronic obstructive pulmonary disease (COPD), and contributes to COPD development and progression by causing epithelial injury and inflammation. Whereas it is known that cigarette smoke (CS) may affect the innate immune function of airway epithelial cells and epithelial repair, this has so far not been explored in an integrated design using mucociliary differentiated airway epithelial cells. In this study, we examined the effect of whole CS exposure on wound repair and the innate immune activity of mucociliary differentiated primary bronchial epithelial cells, upon injury induced by disruption of epithelial barrier integrity or by mechanical wounding. Upon mechanical injury CS caused a delayed recovery in the epithelial barrier integrity and wound closure. Furthermore CS enhanced innate immune responses, as demonstrated by increased expression of the antimicrobial protein RNase 7. These differential effects on epithelial repair and innate immunity were both mediated by CS-induced oxidative stress. Overall, our findings demonstrate modulation of wound repair and innate immune responses of injured airway epithelial cells that may contribute to COPD development and progression.

Fig 1. Effects of CS on airway epithelial barrier recovery and innate immunity.

Barrier integrity in ALI-PBEC was disrupted using calcium depletion, and cells were subsequently exposed to air or CS. (A) The trans-epithelial electrical resistance (TEER) was subsequently measured at 0, 3, 6, and 24 h after exposure to assess loss and recovery of barrier integrity in air- and CS-exposed cultures. TEER values in ohm (Ω). n = 7 independent donors. (B) At 24 h, mRNA expression of RNASE7 was assessed in ALI-PBEC that were incubated with calcium-depleted medium (w/o Ca²⁺) versus control medium (ctrl), and subsequently exposed to either air or CS and further incubated in calcium containing medium. Normalized mRNA expression compared to RPL13A and ATP5B is depicted in the graph. n = 4 independent donors. (C) Secretion of IL-8 in the basal culture medium was assessed by ELISA. n = 5 independent donors. Data are shown as mean; error bars represent SEM; experiments were conducted using duplicate exposures in all donors, * p < 0.05.

Fig 2. Effect of CS on airway epithelial wound healing and innate immunity.

ALI-PBEC were mechanically injured and subsequently exposed to air (control) or whole cigarette smoke (CS). (A) Phase-contrast light microscopy images were made of air- and CS-exposed ALI-PBEC at 0, 6, 24 and 48 h after exposure. (B) Wound closure is shown in percentage in air- versus CS-exposed cells and (C) wound closure rate in percentage per hour at different time intervals was calculated. n = 8 independent donors. (D) Wound closure rates per hour in air- and CS-exposed ALI-PBEC up to 12 h after exposure were determined using live imaging. n = 7 independent donors. (E) RNASE7 mRNA expression was determined in intact or wounded ALI-PBEC exposed to air or CS, at 6 h after exposure. Values shown represent normalized mRNA expression compared to RPL13A and ATP5B. n = 7 independent donors. (F) IL-8 secretion was determined in the basal culture medium. n = 9 independent donors. Data are shown as mean; error bars represent SEM; experiments were conducted in duplicate i, * p < 0.05.

Fig 3. p63⁺ cells at the wound edge of ALI-PBEC.

(A) Immunofluorescence staining for p63 (green) and nuclei (blue (DAPI)) of mechanically injured ALI-PBEC. (B) Percentage of p63⁺ cells at the first line of cells directly at the wound edge or in intact areas, in air- versus CS-exposed cells. (C) Number of p63⁺ cells and p63^- cells at the wound edge per 400 μm length of wound edge, in air- versus CS-exposed cells. (D) Internuclear distance in μm between p63⁺ cells located directly at the leading wound edge and the first adjacent p63⁺ cell. All graphs: data are shown as mean; error bars represent SEM, experiments were conducted in duplicate, n = 3 independent donors, * p < 0.05.

Fig 4. Role of CS-induced oxidative stress in modulating airway epithelial repair and innate immunity.

Wounded ALI-PBEC were pre-incubated with NAC (10 mM) and subsequently exposed to air or CS. mRNA expression of the oxidative stress-induced genes (A) HMOX1 and (B) SCAL1 was determined 6 h after exposure. Values shown represent normalized mRNA expression compared to RPL13A and ATP5B. n = 3 independent donors, * p < 0.05. (C) Wound closure in presence or absence of NAC (10 mM) 6 h after wounding, in air- versus CS-exposed cells. Data are shown as percentage wound closure compared to t = 0 h. n = 7 independent donors. (D) mRNA expression of RNASE7 and (E) CXCL8 was assessed in wounded ALI-PBEC incubated with NAC (10 mM), at 6 h after exposure to air or CS. Data are shown as normalized mRNA expression compared to RPL13A and ATP5B. n = 3 independent donors. In all graphs data are shown as mean; error bars represent SEM; experiments were conducted in duplicate; * p < 0.05.

Fig 5. EGFR and ERK1/2 signaling in wounded ALI-PBEC.

(A) Intrinsic wound healing of ALI-PBEC was determined in the presence of the EGFR tyrosine kinase inhibitor AG1478 (1 μM) or the MEK1/2 inhibitor U0126 (25 μM) at 6 h after exposure with either air or CS. Data are shown as percentage wound closure compared to t = 0 h. (B) mRNA expression of RNASE7 was determined by qPCR. Data are shown as normalized mRNA expression compared to RPL13A and ATP5B. (C) Western blot analysis of EGFR and ERK1/2 phosphorylation of wounded ALI-PBEC exposed to air or CS in the presence of AG1478, U0126, and NAC, at 6 h after exposure. (D) Bands were quantified by densitometry for analysis of EGFR and (E) ERK1/2 phosphorylation and corrected for total-EGFR and total-ERK1/2, respectively. For all graphs data are shown as mean; error bars represent SEM; experiments were conducted in duplicate; n = 3 independent donors; * p < 0.05.

Fig 6. Proposed model.

EGFR signaling is activated by CS and injury, and this leads to MEK1/2-mediated phosphorylation of downstream ERK1/2. CS furthermore directly causes phosphorylation and activation of ERK1/2 via oxidative stress, which is independent of EGFR signaling. EGFR/ERK1/2-mediated wound repair is suppressed by CS via oxidative stress. In contrast, activation of ERK1/2 due to a combined effect of CS-induced oxidative stress and injury, results in an enhanced innate immune response. Solid lines represent the effect of CS, dashed lines the effect of injury. NAC, AG1478 and U0126 were used to inhibit oxidative stress, EGFR phosphorylation, and ERK/12 phosphorylation respectively.