If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
National Children's Research Centre, Our Lady's Children's Hospital, Crumlin, Dublin 12, IrelandDepartment of Molecular Medicine RCSI-ERC Beaumont Hospital, Royal College of Surgeons in Ireland, Dublin 9, Ireland
National Children's Research Centre, Our Lady's Children's Hospital, Crumlin, Dublin 12, IrelandInstitut National de la Santé et de la Recherche Médicale, U1151, Faculté de Médecine Paris Descartes, France
National Children's Research Centre, Our Lady's Children's Hospital, Crumlin, Dublin 12, IrelandDepartment of Paediatrics, Royal College of Surgeons in Ireland, Dublin 2, Ireland
National Children's Research Centre, Our Lady's Children's Hospital, Crumlin, Dublin 12, IrelandInstitut National de la Santé et de la Recherche Médicale, U1151, Faculté de Médecine Paris Descartes, FranceDepartment of Molecular Medicine RCSI-ERC Beaumont Hospital, Royal College of Surgeons in Ireland, Dublin 9, Ireland
Cystic Fibrosis (CF) lung disease is characterised by dysregulated ion transport that promotes chronic bacterial infection and inflammation. The impact of the specialised pro-resolution mediator resolvin D1 (RvD1) on airway surface liquid (ASL) dynamics and innate defence had not yet been investigated in CF airways.
Methods
Ex vivo studies were performed on primary cultures of alveolar macrophages and bronchial epithelial cells from children with CF and in human bronchial epithelial cell lines; in vivo studies were performed in homozygous F508del-CFTR mice treated with vehicle control or RvD1 (1–100 nM).
Results
RvD1 increased the CF ASL height in human bronchial epithelium and restored the nasal trans-epithelial potential difference in CF mice by decreasing the amiloride-sensitive Na+ absorption and stimulating CFTR-independent Cl− secretion. RvD1 decreased TNFα induced IL-8 secretion and enhanced the phagocytic and bacterial killing capacity of human CF alveolar macrophages.
Conclusion
RvD1 resolves CF airway pathogenesis and has therapeutic potential in CF lung disease.
Efficient mucociliary clearance relies on adequate hydration of the airway surface liquid (ASL). This is achieved through a balance between sodium absorption, mediated by the Epithelial Sodium Channel (ENaC), and chloride secretion via CFTR and calcium-activated chloride channels. In cystic fibrosis (CF), this ion transport equilibrium is impaired, leading to a reduced ASL height that favours chronic bacterial infection and persistent inflammation [
Despite the robust inflammatory response, CF lungs fail to clear bacteria and are more susceptible to infections. Pseudomonas aeruginosa is a key CF pathogen and its early acquisition is predictive of an accelerated decline in lung function [
]. Moreover, one of the consequences of the excessive activation of the inflammation in patients is the production of terminal electron acceptors for anaerobic respiration that allow P. aeruginosa to persist and outcompete other pathogens [
]. Recent studies in young children with CF have identified neutrophil elastase, as a key risk factor for the onset and early progression of CF lung disease [
Several reports provide evidence for a correlation between chronic inflammatory disease and abnormal production or activity of the specialised pro-resolution lipid mediators (SPMs) including resolvins and lipoxins [
] and there is a significant correlation between the levels of RvD1 in plasma and sputum of CF patients with the biomarkers of inflammation (IL8 and IL1β) and lung function [
SPMs have been shown to halt neutrophil infiltration, enhance macrophage phagocytosis of apoptotic neutrophils and attenuate NFĸB activation in mouse models of lung inflammation [
]. Moreover, RvD1 promoted differentiation of alternatively activated (M2) macrophages, improved bacterial killing and the containment of a bacterial challenge in mouse models of lung infection by P. aeruginosa [
In this study, we demonstrate, that RvD1 produces restorative effects on key aspects of CF lung disease specifically; airway epithelial ion transport and surface liquid height, NFĸB-mediated inflammation and CF macrophage phagocytosis activity.
2. Methods
2.1 Clinical samples
Bronchoalveolar lavage fluid (BAL) and bronchial brushings were collected through the Study of Host Immunity and Early Lung Disease in CF [
]. Studies were carried out in accordance with European community guidelines and approved by the Research Ethics Committee of Our Lady's Children's Hospital Crumlin (Dublin).
2.2 Human airway epithelial cell culture
Primary cultures of bronchial epithelial cells were grown from bronchial brushings or biopsies obtained from 5 healthy donors and 6 children with CF (4 F508del-CFTR homozygous and 2 F508del-CFTR heterozygous (F508del/2789 + 5G > A and F508del/H199Y)). The CF epithelia showed similar electrophysiological profiles in untreated conditions. Human bronchial epithelial cell lines were also used; Non-CF NuLi-1 and CF (F508del homozygous) CuFi-1 [
]. Epithelial cells were cultured on permeable supports under an air-liquid interface until reaching a high trans-eithelial electrical resistance, (TEER >700 Ω/cm2) [
Lipoxin a(4) stimulates calcium-activated chloride currents and increases airway surface liquid height in normal and cystic fibrosis airway epithelia, (in eng).
Texas red (2 mg/ml, Invitrogen) was applied to the ASL of bronchial epithelial cells, 24 h prior imaging and Perfluorocarbon-72 (3 M, St. Paul, USA) was added before acquisition to prevent evaporation. The ASL images were captured with a Zeiss LSM 510 Meta microscope (40×) and analysed using Zeiss LSM Image Browser. Each biological repeat represents the mean of 27 ASL height measurements per culture insert.
2.4 Nasal potential difference measurements
Nasal potential difference measurements were performed on homozygous F508del-CFTR mice (FVB/N) and their wild-type normal homozygous littermates (WT) as previously described [
] and approved by the ethics committee of Necker Hospital (Paris, France) and conformed to European Community regulations for the use of animals in research (authorization no. P2.AE.092.09). Changes in nasal VTE obtained after amiloride 100 μM and low Cl− solution perfusion reflect the ionic current contribution of Na+ absorption via ENaC and Cl− secretion, respectively (see on line supplement).
2.5 Short-circuit current (ISC) recordings
Differentiated human bronchial epithelia were mounted in Ussing chambers and short-circuit-current SCC was measured under voltage clamp conditions and a Cl− gradient across the epithelium (see online Supplement). The SCC decreased after amiloride (100 μM) and increased after forskolin (10 μM)/IBMX (100 μM) treatment. The use of these drugs served as an indicator of SCC changes reflecting ENaC and CFTR activity, respectively.
2.6 Enriching primary alveolar macrophages
Alveolar macrophages (AM) were isolated from the BAL of 3 CF female children (<6y, F508del homozygous), re-suspended in primary AM medium (online data), plated in 96 well plates and incubated (humidified, 37.2 °C, 21% oxygen, 5% CO2) overnight. The following morning, non-adherent cells were aspirated and discarded. The adherent cells were washed twice with pre-warmed Ca2+ and Mg2+ free PBS.
2.7 Alveolar macrophage phagocytosis assay
The phagocytic capacity of Alveolar Macrophages (AM) was measured by their ability to engulf IgG & FITC labelled beads (Cayman Chemical, Ann Arbour, MI). Phagocytosis was quantified by the fluorescence intensity of engulfed FITC labelled complexes using a plate reader (Synergy MX Biotek Instruments, Winooski, VT).
2.8 Bactericidal assay against Pseudomonas aeruginosa
Alveolar Macrophages were washed twice with antibiotic-free medium and incubated with PAO1 (2 × 10^14 CFU/ml) for 3 h. AM were then washed and gentamicin (400 μg/ml) was applied for 1 h. After 2 washing AM were lysed with Triton-X 100 (0.4% v/v) for 15 min to release internalised bacteria. Quantification of the viable intracellular bacterial load was performed and expressed as colony forming units per ml.
2.9 Statistical analysis
Results are presented as mean and standard error of the mean (SEM). The non-parametric Wilcoxon-Mann-Whitney rank sum test was used when comparing two groups. The one-way analysis of variance (ANOVA) was used in the cases of multiple comparisons.
3. Results
3.1 Resolvin D1 restores ASL height in CF bronchial epithelial cells
We investigated the effect of RvD1 on ASL height in CF airway epithelia. Under vehicle control conditions, the ASL height was low and the hydrated layer overlaying the epithelium was discontinuous in primary cultures of CF bronchial epithelial cells derived from F508del CFTR homozygous patients, (3.7 ± 0.2 μm, n = 7), (Fig. 1A ). A similar low and disrupted ASL was observed in CuFi-1 cell lines (5.7 ± 0.3 μm, n = 8), (Fig. 1B, C, D and E). After exposure of CF bronchial epithelial cells to RvD1, the ASL height was higher with a more continuous hydrated layer (Fig. 1E). Exposure to RvD1 (1 nM, 30 min) produced an increase in ASL height to 6.8 ± 0.3 μm (P < 0.001, n = 7) in CF bronchial epithelial primary cultures (Fig. 1A) and to 7.1 ± 0.15 μm (P < 0.01, n = 6), in CuFi-1 cells (Fig. 1B, C, D). As shown in the dose-response figure, treatment of CuFi-1 cells with higher concentrations of RvD1 resulted in a further increase in ASL height compared to vehicle control conditions (Fig. 1B). In CuFi-1 cells, the ASL height increased to 7.7 ± 0.2 μm (P < 0.01, n = 5) and to 8.3 ± 0.3 μm (P < 0.001, n = 5) with RvD1 10 nM and 100 nM, respectively (Fig. 1B).
Fig. 1Resolvin D1 effects on airway surface liquid (ASL) height in polarised, differentiated CF Bronchial Epithelial Cells. ASL height in μm was measured by live cell confocal fluorescence microscopy using Texas red®-dextran to stain the ASL. (A) Primary cultures of human CF Bronchial Epithelial Cells were stimulated with either vehicle control (Cont) or RvD1 (1 nM) for 30 min prior to image acquisition. Representative images of the ASL are presented with quantification of the ASL height measured in inserts prepared from bronchial epithelium derived from 2 separate children with CF (***P < 0.001, n = 7, Wilcoxon). (B) CuFi-1 epithelial monolayers were stimulated with either vehicle control (Cont), amiloride (10 μM), human neutrophil elastase (NE) alone or followed by RvD1 (1 nM–100 nM) 30 min prior to image acquisition.(*P < 0.05, **P < 0.01, ***P < 0.001, n = 6, ANOVA). (C) CuFi-1 cells were treated with either vehicle control (Cont) or with Boc2 (10 μM) for 20 min followed by RvD1 (1 nM) for 30 min (*P < 0.05, **P < 0.01, n = 6). (D) CuFi-1 cells were treated with either vehicle control (Cont) or with BAPTA-AM (10 μM) (intracellular calcium chelator) for 20 min followed by RvD1 (1 nM) for 30 min (**P < 0.01, ***P < 0.001, n = 5, ANOVA). (E) Representative images of the ASL under the various conditions are shown.
3.2 ENaC activity contributes to the increased ASL height induced by Resolvin D1
The possible role of ENaC in causing an ASL height increase in response to RvD1 was tested in CuFi-1 cells by either inhibiting or stimulating ENaC activity using amiloride or human neutrophil elastase, respectively (Fig. 1B). Apical amiloride (10 μm) exposure of the CF epithelial cells produced an ASL height increase from 5.5 ± 0.1 to 6.6 ± 0.1 μm, P < 0.001, n = 6. The ASL height increase (8.2 ± 0.2 μm, P < 0.001, n = 6) induced by RvD1 (100 nM) was larger than that produced by amiloride alone, however, when cells were exposed to both RvD1 and amiloride, no additive effects were observed (7.9 ± 0.1 μm, P > 0.5, n = 6). Conversely, ENaC stimulation by apical exposure to neutrophil elastase induced a small but significant reduction in ASL height (5.1 ± 0.1 μm, P < 0.05, n = 6). RvD1 treatment reversed the ASL height decrease induced by human neutrophil elastase by restoring the ASL height above basal values (7.1 ± 0.16 μm, P < 0.001, n = 6). This latter result, which is consistent with a greater stimulatory ASL response to RvD1 compared to amiloride alone, suggests that the ASL height increase induced by RvD1 involves both inhibition of ENaC and regulation of another ion transport pathway.
3.3 ALX/FPR2 receptor and intracellular calcium mediate Resolvin D1 effects on ASL height
Exposure of CuFi-1 cells under basal conditions to the ALX/FPR2 receptor antagonist, Boc2, did not significantly alter ASL height (Boc2 (10 μM): 5.9 ± 0.6 μm vs control: 5.7 ± 0.3 μm, n = 6). In contrast, pre-treatment of CuFi-1 cells with Boc2 completely prevented the increase in ASL height produced by RvD1 (5.7 ± 0.5 μm in RvD1 + Boc2 compared with 7.1 ± 0.2 μm in RvD1 alone (P < 0.05, n = 6)), (Fig. 1C and E). Incubation of CuFi-1 cells with the intracellular calcium chelator BAPTA-AM (10 μM) did not significantly affect the basal ASL height (5.2 ± 0.4 μm vs 5.7 ± 0.3 μm, n = 5), (Fig. 1C). However, the RvD1 mediated increase in ASL height was abolished by co-treatment with BAPTA-AM. Under these conditions the ASL height remained low in the combined presence of RvD1 and BAPTA-AM (4.9 ± 0.2 μm, n = 5) compared to the enhanced ASL height in the presence of RVD1 alone (7.1 ± 0.2 μm, P < 0.001, n = 5), (Fig. 1D and E). These data indicate that the ASL height increase induced by RvD1 in CF airway epithelia is transduced by the ALX/FPR2 receptor and involves the mobilisation of intracellular calcium.
3.4 Resolvin D1 modulates ASL height by non-CFTR pathways
We compared the effects of RvD1 on ASL height in non-CF NuLi-1 and CF CuFi-1 bronchial epithelial cells. The ASL height recorded under basal conditions in NuLi-1 cells (7.2 ± 0.3 μm, n = 11), (Fig. 2) was significantly higher than that measured in CuFi-1 cells (5.7 ± 0.3 μm, P < 0.01, n = 10). In contrast to the CF epithelium, the ASL height in NuLi-1 epithelial monolayers did not change significantly under low or high concentrations of RvD1 (7.8 ± 0.3 μm in RvD1 (1 nM), n = 13 and 7.6 ± 0.3 μm in RvD1 (100 nM), P > 0.5, n = 10), (Fig. 2). We further examined the effect of RvD1 on ASL height dynamics following CFTR inhibition in CFTR-expressing NuLi-1 epithelia. Treatment of Nuli-1 monolayers with CFTRinh172 resulted in a significant reduction in ASL height compared to untreated monolayers (7.2 ± 0.3 μm in control vs 4.7 ± 0.3 μm in CFTRinh172, P < 0.001, n = 6). These data indicate that CFTR activity makes a large contribution to maintaining the basal ASL height in normal bronchial epithelium. In contrast, RvD1 (100 nM) treatment significantly increased the ASL height in CFTR inhibited NuLi-1 cells (6.7 ± 0.3 μm in RvD1 + CFTRinh172 compared to 4.7 ± 0.3 μm in CFTRinh172 alone, P < 0.001, n = 6), (Fig. 2). Taken together, these data suggest that RvD1 can modulate ASL height dynamics by ion transport pathways other than CFTR in CF bronchial epithelium.
Fig. 2Resolvin D1 effects on airway surface liquid height in polarised NuLi-1 bronchial epithelial cells. NuLi-1 bronchial epithelial cell preparations were stimulated with either vehicle control (Cont), RvD1 (1 nM), RvD1 (100 nM), CFTR inhibitor (CFTRinh172, 20 μM) or with RvD1 (100 nM) in combination with CFTR inhibitor 172 (20 μM) 30 min prior to image acquisition. (A) Mean ASL height in vehicle control treated preparations (***P < 0.001, ANOVA). (B) Representative images of the ASL under the various conditions are presented.
3.5 Resolvin D1 restores nasal potential difference in CF and non-CF mice
We investigated the impact of RvD1 in regulating ion transport in CF airway in vivo, by testing the effects of this SPM on the nasal transepithelial electrical potential difference (VTE) in mice homozygous for F508del-CFTR (FVB/N) and in their wild-type CFTR litter-mates. In agreement with previously published studies, F508del-CF mice displayed a raised (more negative) baseline nasal transepithelial potential (VTE = −14.2 ± 1.1 mV, n = 10) compared to WT mice (−6.1 ± 1.3 mV, n = 4), (Fig. 3A and B ). The baseline VTE was significantly depolarised (shifted to less negative potentials) with RvD1 (10 nM) treatment in the CF mice (−9.0 ± 2.1 mV, P < 0.005, n = 10), whereas it was unaltered in WT mice (−5.7 ± 1.4 mV, P > 0.5, n = 4), (Fig. 3A and B). The relative contribution of Na+ absorption and Cl− secretion to the VTE response to RvD1 was tested using amiloride (100 μm) and low Cl− solutions, respectively. RvD1 reduced the amiloride-sensitive VTE in CF mice (∆VTE 10.1 ± 1.0 mV with amiloride alone compared to a ∆VTE of 6.3 ± 1.1 mV with combined amiloride and RvD1 treatment, P < 0.05, n = 10). In contrast, RVD1 treatment did not affect the ∆VTE changes induced by amiloride in WT mice (∆VTE 3.5 ± 0.36 mV with amiloride alone compared to ∆VTE 3.1 ± 1.1 mV with amiloride and RvD1, P > 0.5, n = 4), (Fig. 3A and C). As already reported, perfusion of the nasal cavity with low Cl− solutions induced a small VTE change in CF mice (−1.1 ± 0.4 mV) while producing a significantly larger VTE change in WT mice (−3.9 ± 0.6 mV). This latter response was not significantly affected by RvD1 treatment of WT mice (−3.2 ± 0.5, P > 0.1, n = 4). In contrast, when RvD1 was added to the nasal infusion fluid of CF mice, the VTE response to low Cl− solution was significantly increased (−3.1 ± 0.8 mV, P < 0.05, n = 10) to levels obtained in WT mice (without RvD1) (Fig. 3A and D). These data are consistent with an inhibitory effect of RvD1 on ENaC and a stimulation of a non-CFTR dependent Cl− secretion in the CF airway.
Fig. 3Effect of RvD1 on nasal potential in WT and F508del-CFTR mice. (A) Representative nasal potential (VTE) recordings obtained in F508del-CFTR mice. Amiloride (10 μM) and low Cl− solution were perfused at the time indicated on the graph. (B) Baseline VTE measured during nasal perfusion in a normal Cl− solution with or without RvD1 (10 nM), in F508del-CFTR (n = 10) and WT mice (n = 4). (C) Effect of RvD1 on the amiloride-sensitive nasal potential difference (∆VTE amiloride) in F508del-CFTR and WT mice. (D) Effect of RvD1 on nasal Cl− secretion measured by the nasal potential difference change generated by removal of Cl− in the perfusion solution (∆VTE low Cl−) in F508del-CFTR and WT mice. Each mice treated with RvD1 had its own untreated control. (*P < 0.05/**P < 0.01, Wilcoxon).
3.6 Resolvin D1 restored amiloride-sensitive current without affecting CFTR dependant Cl− transport in CF bronchial primary cultures
The effect of RvD1 on transepithelial ion transport was investigated using short-circuit analyses of bronchial primary cultures on permeable filters, from 4 patients with F508del CFTR mutation and from 5 healthy donors. Exposure to amiloride (100 μM, apical), followed by exposure to a cocktail of forskolin (10 μM) and 3-Isobutyl-1-methylxanthine (Fsk/IBMX, 10 μM/100 μM, apical and basolateral sides application) and CFTRinh172, in order to determine ENaC, and CFTR respective contributions to the transepithelial Isc. The amiloride-sensitive SCC was significantly higher in CF bronchial epithelia compared to healthy controls (Fig. 4B ). Treatment with RvD1 (10 nM, 100 nM) or LXA4 (10 nM, 100 nM), significantly decreased the amiloride-sensitive transepithelial short-circuit current in CF epithelia (Fig. 4A and B). As expected, the effect of Fsk/IBMX exposure was significantly lower in CF epithelia compared to healthy controls (Fig. 4C). Neither RvD1 nor LXA4 affected the CFTR-dependant Cl− secretion (induced by Fsk/IBMX or inhibited by CFTR INH172) in bronchial primary cultures from patients with F508del CFTR mutations (Fig. 4A, C and D).
Fig. 4RvD1 (10–100 nM) and LXA4 (10–100 nM) effects on transepithelial ion transport in primary cultures of human bronchial epithelium grown on permeable filters under air-liquid interface. (A) Representative short-circuit current (Isc) recordings obtained in primary cultures of human CF bronchial epithelium (F508del homozygous). Amiloride (10 μM) and a cocktail of forskolin (10 μM) and IBMX (100 μM) (Fsk/IBMX) were perfused at the time indicated on the graph. (B) Mean amiloride-sensitive Isc measured in primary culture of healthy control bronchial epithelium (white bar) and in primary culture of human CF bronchial epithelium (black bars) upon exposure to EtOH (control solvent) or to RvD1 (10–100 nM), or to LXA4 (10–100 nM). (C) Mean Isc variation induced by Fsk/IBMX measured in primary culture of healthy control bronchial epithelium (white bar) and in primary culture of human CF bronchial epithelium (black bars) with or without RvD1 (10–100 nM), or LXA4 (10–100 nM). (D) Mean Isc variation induced by CFTR inhibitor (INH172) measured in primary cultures of healthy control bronchial epithelium (white bar) and in primary cultures of human CF bronchial epithelium (black bars) with or without RvD1 (10–100 nM), or LXA4 (10–100 nM).
Mean and SEM values for (μA/cm2) were calculated from median values per patient. Comparison by non-parametric Mann Whitney test. At least 2 filters were analysed per patient and condition. The P-values for statistically significant differences are indicated as * < 0.05.
3.7 Resolvin D1 enhances the phagocytic capacity of CF primary alveolar macrophages
The phagocytic activity of CF (F508del) alveolar macrophages was measured after treatment with either vehicle control or RvD1 (100 nM) and incubation with fluorescently latex beads. A greater proportion of alveolar macrophages treated with RvD1 was observed to have engulfed labelled beads. In addition, the fluorescence intensity measured from RvD1 treated macrophage samples was significantly higher compared to vehicle control samples (7490 ± 950, in RvD1 compared to 4420 ± 1020, in untreated cells, P < 0.01, n = 9), (Fig. 5 A–E ). In order to quantify the bacterial killing capacity of RvD1 treatment, CF alveolar macrophages were pre-treated with either vehicle control or RvD1 (100 nM) and exposed to P. aeruginosa lab strain PAO1 (2 × 1014 CFU/ml) for 3 h. The intracellular viable bacterial load of PAO1 was significantly decreased in preparations treated with RvD1 compared to vehicle control conditions (1.7 ± 0.6 × 106 CFU/ml, in RvD1 and 6.0 ± 1.4 × 106 CFU/ml in vehicle control, P < 0.05, n = 4), (Fig. 5F).
Fig. 5Resolvin D1 effect on the phagocytic capacity and bacterial killing of primary alveolar macrophages from children with CF (CFAM). CFAM were isolated from BAL and pre-treated for 30 min with either vehicle control (A&B) or RvD1 100 nM (C&D). CFAM were exposed to Latex Beads-rabbit IgG-FITC complex for 1 h. Light microscopy (LM) images were captured concurrently with fluorescence images (B&D) from phagocytosed FITC-labelled beads (Panels A&C are LM/fluorescence composites). Representative images are presented (scale bar = 100 μm). (E) Fluorescence intensity of phagocytosed FITC-labelled beads were quantified using a fluorescence plate reader (n = 9). (F) CFAM were treated with either vehicle control (Cont) or RvD1 (100 nM) for 3 h and then exposed to PAO1 bacteria. After 30 min, the non-engulfed bacteria were removed and CFAM were treated with gentamicin to kill residual extracellular and membrane-bound bacteria. CFAM were then lysed and the bacterial load within the lysate was quantified (n = 9 from 3 patient samples for phagocytosis and n = 4 preparations from 2 patients for bacterial killing **P < 0.01, *P < 0.05, Wilcoxon).
We report the novel finding that RvD1 ameliorates key components of CF lung pathogenesis such as abnormalities of ion transport and ASL height, altered macrophage mediated killing of P. aeruginosa and NFĸB -driven inflammation.
The effect of RvD1 to increase ASL height in human CF bronchial epithelia is consistent with its actions observed in vivo on nasal potential difference in F508del-CFTR mice. Similarly, the effect of the related SPM, LXA4 to increase ASL height in human CF bronchial epithelia, that we previously reported [
Lipoxin a(4) stimulates calcium-activated chloride currents and increases airway surface liquid height in normal and cystic fibrosis airway epithelia, (in eng).
] is consistent with its actions observed in vivo on nasal potential difference in F508del-CFTR mice (on line data, fig. S1). The in vitro and in vivo effects of RvD1 on ASL height in CF airway epithelia were shown to be mediated via stimulation of non-CFTR Cl− secretion and inhibition of ENaC Na+ absorption. This non-CFTR Cl− secretory pathway activated by RvD1 in CF epithelia appears to be dependent on intracellular calcium mobilisation given the inhibitory effect of intracellular calcium chelation with BAPTA-AM on the RvD1-mediated ASL height increase in CF bronchial epithelial cells. This conclusion is supported by previous reports showing that ALX/FPR2 stimulation by LXA4, induced intracellular calcium mobilisation and calcium-activated chloride currents in bronchial epithelial cells [
]. The differential effects of RvD1 in CF and non-CF cells could be explained by the amplified calcium mobilisation which is related to expanded ER calcium stores in CF compared to non-CF airway epithelial cells [
]. Several pathways for calcium-activated Cl− secretion, including members of the CLCA family or TMEM16A could be candidates for the RvD1 induced Cl− secretory response [
We have shown previously that inhibition of ENaC current increases ASL height in CuFi-1 bronchial epithelial cells and this response can be further augmented by treatment with LXA4 [
]. Here we show that RvD1 can also inhibit ENaC activity and reverse the stimulatory effect of neutrophil elastase on ENaC to increase ASL height. Amiloride and RvD1 did not produce additive effects on ASL height, suggesting that RvD1 and amiloride are at least partially sharing a common pathway. In addition, RvD1 as well as LXA4 decreased the amiloride-sensitive nasal VTE in CF mice and the transepithelial SCC in human CF primary cultures which is consistent with the inhibitory effect of RvD1 and LXA4 on ENaC. The BAPTA-AM inhibition of the RvD1 induced increase in ASL height is consistent with known effects of elevated calcium to inhibit ENaC channel activity [
The complete inhibitory effect of Boc2 used as an ALX/FPR2 receptor antagonist, supports RvD1 action on ASL dynamics via the FPR2/ALX receptor. RvD1 is reported also to signal through DRV1, formally known as GPR32 orphan receptor [
] and we have also found that LXA4 causes a mobilisation of ALX/FPR2 from the cytosol to the apical membranes of primary human CF bronchial epithelium (online data, Fig. S3). In resolving exudates, LXA4 biosynthesis precedes RvD1 [
], and the effect of LXA4 on ALX/FPR2 expression and trafficking might be important in sensitising the tissue to receive and act upon the RvD1 signal.
Taken together, the in vitro and in vivo studies provide strong evidence for a role of RvD1 in inhibiting amiloride-sensitive Na+ absorption and stimulating CFTR-independent Cl− secretion resulting in an increase CF ASL height, which would be expected to improve airway mucociliary clearance.
We also provide evidence for a role of RvD1 in supressing TNFα induced IL8 secretion by preserving the integrity of IĸB. RvD1 has been previously shown to supress NFĸB activation in a mouse model of LPS induced lung inflammation and to inhibit IL8 production by bronchial epithelial cells stimulated with IL4 [
"AT-RvD1 modulates CCL-2 and CXCL-8 production and NF-κB, STAT-6, SOCS1, and SOCS3 expression on bronchial epithelial cells stimulated with IL-4," (in eng).
]. Therefore, by increasing ASL height and inhibiting NfkB dependent IL8 production, RvD1 enhances two distinct functions of airway epithelium in innate immunity which are both altered in CF.
The macrophage defects in CF patients include impaired clearance of apoptotic cells [
]. We found that RvD1 improves the phagocytic and bacterial killing capacity of CF alveolar macrophages from CF donors. These findings are consistent with the effects of RvD1 to improve bacterial killing and the containment of a bacterial challenge in mouse models of lung infection by Pseudomans aeruginosa [
]. Besides enhancement of the bacterial killing capacity of CF macrophages, there are several additional mechanisms by which RvD1 may increase antimicrobial capacity of the ASL, including restoration of a more alkaline mucosal fluid, stimulation of host defence via antimicrobial peptides and a more effective mucociliary clearance. All of these potential mechanisms can be tested experimentally by ASL pH, bacterial killing capacity of secreted defensins, mucus clearance and ciliary beat frequency measurements.
Our study provides cellular insights into RvD1 functions in the CF airway. The pro-resolving effects of RvD1 on ASL height, ion transport and immune function are coherent with the recent findings of a significant correlation between the levels of RvD1 in the plasma and sputum of patients with CF, the biomarkers of inflammation (IL8 and IL1β) and the lung function (FEV1) [
]. In addition, RvD1 biosynthesis involves two sequential lipoxygenation steps both catalysed by 15 lipoxygenase which we have shown to be defective in the CF airway bronchoalveolar lavages [
]. Therefore, CF pathology might be at least in part considered as a consequence of an imbalance between pro-inflammatory and pro-resolving mediator generation, where RvD1 appears to play an important role.
In conclusion, we report that RvD1 has multiple roles in reversing CF airway epithelial dysfunction by synergistically correcting abnormalities in airway epithelial ion transport and airway surface liquid dynamics; airway epithelial cell IL8 production; and bacterial killing capacities of CF alveolar macrophages. RvD1 thus displays high therapeutic potential in CF lung disease.
Acknowledgements
This work was supported by grants from the National Children's Research Centre (Dublin) the Health Research Board of Ireland, the French National Institute of Health (INSERM) and the French CF Foundation (VLM) and EU BMBS COST Action BM1003 Microbial cell surface determinants of virulence as targets for new therapeutics in Cystic Fibrosis.
Lipoxin a(4) stimulates calcium-activated chloride currents and increases airway surface liquid height in normal and cystic fibrosis airway epithelia, (in eng).
"AT-RvD1 modulates CCL-2 and CXCL-8 production and NF-κB, STAT-6, SOCS1, and SOCS3 expression on bronchial epithelial cells stimulated with IL-4," (in eng).
☆Critical revision of the article: FR, BJH, PMN, and VU
☆Final approval of the version to be published: FR, GH, AH, AS, AM, CFT, MH, ISG, BJH, PMN, VU
☆Funding: This work was supported by the Institut National de la Santé et de la Recherche Médicale (INSERM, France), Vaincre la Mucoviscidose (VLM, France grant RF20150501364/1/1/50), The National Children's Research Center (NCRC, Ireland, TRAPCF project), EU BMBS COST Action (BM1003), The Health Research Board of Ireland (HRB, HRA/POR/2011/21 - 1403).
30974-8&id=gr1.jpg){kind=link}
30974-8&id=gr2.jpg){kind=link}
30974-8&id=gr3.jpg){kind=link}
30974-8&id=gr4.jpg){kind=link}
30974-8&id=gr5.jpg){kind=link}