microRNA-181b is increased in cystic fibrosis cells and impairs lipoxin A4 receptor-dependent mechanisms of inflammation resolution and antimicrobial defense

The involvement of microRNA (miR) in cystic fibrosis (CF) pathobiology is rapidly emerging. We previously documented that miR-181b controls the expression of the ALX/FPR2 receptor, which is recognized by the endogenous proresolution ligand, lipoxin (LX)A4. Here, we examined whether the miR-181b-ALX/FPR2 circuit was altered in CF. We examined human airways epithelial cells, normal (16HBE14o-), carrying the ΔF508 mutation (CFBE41o-) or corrected for this mutation (CFBE41o-/CEP-CFTR wt 6.2 kb), as well as monocyte-derived macrophages (MΦs) from CF patients. CFBE41o- cells exhibited higher miR-181b and reduced ALX/FPR2 levels compared to 16HBE14o- and CFBE41o-/CEP-CFTR wt 6.2 kb cells. An anti-mir-181b significantly enhanced ALX/FPR2 expression (+ 60%) as well as LXA4-induced increase in transepithelial electric resistance (+ 25%) in CFBE41o- cells. MΦs from CF patients also displayed increased miR-181b (+ 100%) and lower ALX/FPR2 levels (− 20%) compared to healthy cells. An anti-mir-181b enhanced ALX/FPR2 expression (+ 40%) and normalized receptor-dependent LXA4-induced phagocytosis of fluorescent-labeled zymosan particles as well as of Pseudomonas aeruginosa by CF-MΦs. These results provide the first evidence that miR-181b is overexpressed in CF cells, impairing some mechanisms of the ALX/FPR2-dependent pathway of inflammation resolution. Thus, targeting miR-181b may represent a strategy to enhance anti-inflammatory and anti-microbial defense mechanisms in CF.

To investigate on mechanisms underlying these changes, we analyzed ALX/FPR2 transcriptional activity. CFBE41o-cells, transfected with a plasmid expressing the ALX/FPR2 core promoter upstream the luciferase gene 28 , exhibited significantly lower luciferase activity compared to 16HBE14o-or CFBE41o-/CEP-CFTR wt 6.2 kb cells (Fig. 1D). These results indicate that the ALX/FPR2 transcriptional machinery may be altered in CF cells.
On the other hand, we recently uncovered epigenetic regulatory mechanisms of ALX/FPR2 expression, namely chromatin post-translational modifications 27 and mir-181b expression 26 . Therefore, we examined whether this miR was involved in ALX/FPR2 expression in CF cells. Figure 2 shows that CFBE41o-cells expressed higher miR-181b levels compared to 16HBE14o-and CFBE41o-/WT cells. We observed a similar increment in bronchial airway primary cells collected from 3 non-CF subjects and 3 CF patients homozygous for the ΔF508 mutation (Fig. 2B).

MΦs.
To obtain further evidence of a "cause and effect" relationship between miR-181b and ALX/FPR2, we transfected CF-MΦs with a miR-181b inhibitor. Real-time PCR analysis showed that miR-181b levels were reduced by ~ 70% (p = 0.0013) in transfected cells. In these cells, ALX/FPR2 protein expression increased by ~ 40% (p = 0.0019) ( Fig. 4A and B). Next, we determined whether miR-181b-mediated regulation of ALX/FPR2 had an impact on agonist-evoked biological responses. Because macrophage phagocytosis is the hallmark of resolution and it is strongly enhanced by pro-resolving lipid mediators, such as LX and Rv [13,31], we examined phagocytosis of fluorescent-labeled zymosan particles as well as of the Pseudomonas aeruginosa strain PAO1 to mimic bacterial clearance from inflamed tissue. We exposed MΦs to increasing concentration of LXA 4 (0.001-10 nM) and compared the phagocytic capability of healthy (HS) MΦs, CF-MΦs and CF-MΦs transfected with a miR-181b inhibitor. The zymosan phagocytic activity of unstimulated CF-MΦs was slightly, although not significantly, reduced compared to HS-MΦs (results not shown). When exposed to LXA 4 , HS-MΦs displayed a significant increment in zymosan and PAO1 uptake, which was maximal with 1 and 0.1 nM LXA 4 , respectively ( Fig. 4C and D). In contrast, CF-MΦs showed a smaller increment in phagocytic activity when incubated with LXA 4 ( Fig. 4C and D). However, when these cells were transfected with the miR-181b inhibitor, an almost full recovery of LXA 4 -induced phagocytosis was observed, particularly at lower LXA 4 concentrations ( Fig. 4C and D). Receptor dependence was assessed using the WRW4 peptide, which abrogated LXA 4 -induced phagocytosis of zymosan, both in normal and CF cells ( Supplementary Fig. 1). Under these experimental settings, we were unable to detect significant changes in the release of selected cytokines (IL-8, IL-10, IL-1β, RANTES, GM-CSF).
Collectively, these results demonstrate that changes in miR-181b expression impair the ALX/FPR2 proresolution signaling in CF cells.

Discussion
Exuberant lung inflammation and infection, leading to respiratory insufficiency and death are trademarks of CF. Recent evidence indicates that in addition to a marked overproduction of pro-inflammatory mediators, an impairment in endogenous anti-inflammatory proresolving mechanisms does occur in CF 4,21 .
Here, we investigated on the ALX/FPR2 receptor, a GPCR that transduces signals by the endogenous proresolving mediators LXA 4 , RvD1 and ANXA1 [12][13][14] and it is regarded as a main proresolving receptor that intercepts multiple proresolution pathways. Since changes in ALX/FPR2 levels influence the outcome of an inflammatory response 32 we evaluated ALX/FPR2 expression in two cellular models of CF, i.e. airway epithelial cells as a paradigma of the respiratory pathology and macrophages as key effectors of innate immune responses. In both models ALX/FPR2 levels were reduced compared to the normal counterpart (Figs 1 and 3). Combining this evidence with early data showing reduced LXA 4 concentration in BAL from CF patients 21 ; impaired LX biosynthesis, due to deficient 12-LO activity in CF platelets 4 ; and lower ANXA1 in CF neutrophils 25 , we can conclude that the proresolution circuit converging on the ALX/FPR2 receptor is impaired in CF. Together with data showing beneficial actions of LXA 4 in CF settings 33,34 these results encourage further research to test whether the upregulation of the ALX/FPR2 pathway can be regarded as a suitable strategy to combat inflammation in CF patients. Along these lines, ALX/FPR2 and its proresolution agonists are defective in other human diseases characterized by non resolving inflammation [17][18][19][20] , confirming the relevant immunomodulatory role of this system.
The proresolving actions of LXA 4 can be strengthened by the upregulation of ALX/FPR2 expression 27 . Therefore, acting on these mechanisms may help to develop innovative anti-inflammatory, proresolving pharmacology. We previously identified miR-181b as a main epigenetic regulator of ALX/FPR2 expression 26 . Here, we tested the hypothesis that the lower ALX/FPR2 expression in CF cells was related to changes in miR-181b level.
We present evidence of miR-181b upregulation in CF respiratory epithelial cells as well as in CF monocyte-derived macrophages (Figs 1 and 3). Although the role of this miR in inflammatory disorders remains to be conclusively defined, evidence indicates that it may exert pro-inflammatory functions, at least in selected settings. For instance, miR-181b regulates TNF-α-induced transcription of pro-inflammatory genes in liver cells 35 . It also regulates vascular inflammation mediated by NFkB 36 . More recently, a role of miR-181b in atherosclerosis and aneurysms has been proposed 37 . Our results with CF cells confirm that miR-181b levels can be elevated in pathological settings characterized by unresolved inflammation and provide clues to decipher the pathophysiological significance of this miR.
In the specific case of CF, a direct correlation between the genetic defect and miR-181b upregulation appears to occur. We in fact observed that miR-181b expression is under the control of CFTR. This was established in MΦs by experiments with CFTRinh-172, a selective CFTR inhibitor 38 (Fig. 3), and in respiratory epithelium by using cells carrying the ΔF508 mutation and cells where the genetic defect was corrected by the overexpression of wild type CFTR (Fig. 3). Together, these data support the emerging concept of CFTR as epigenetic modulator 39 . Along these lines, it has been recently reported that CFTR regulates miR-125b expression with a bicarbonate-dependent mechanism 40 . Whether ion fluxes are also involved in CFTR-dependent miR-181b regulation remains to be determined.
On the other hand, our present results indicate that, at least in the CF model, the regulatory mechanisms of ALX/FPR2 expression may be cell specific. In fact, ALX/FPR2 mRNA was reduced in CF airway cells (Fig. 1), but not in CF-MΦs (data not shown), suggesting that transcriptional regulatory events are altered in CF respiratory epithelium. We previously documented the impact of epigenetic modifications of the ALX/FPR2 promoter and H3 histones on ALX/FPR2 expression 27,28 . Although the relevance of these changes remains to be determined in CF cells, our results clearly demonstrate that miR-181b controls ALX/FPR2 protein expression in both CF respiratory cells and MΦs (Figs 2 and 4). Moreover, in both cell types, by acting on miR-181b it was possible to enhance the functional responses of the pro-resolution agonist LXA 4 . In CF airway epithelial cells, we observed a significant increase in TEER (Fig. 2), an indirect readout of the monolayer integrity and of the strength of intercellular junctions, which are altered in CF cells and promote local inflammation 41 . In MΦs, a miR-181b inhibitor normalized LXA 4 -triggered phagocytic activity (Fig. 4). This was observed with both zymosan particles and Pseudomonas aeruginosa, which chronically colonizes the airways of the majority of CF patients and it is very difficult to eradicate. These results confirm that by blocking miR-181b it is possible to enhance LXA 4 bioactions useful to control bacterial colonization in CF. On the other hand, under the present experimental settings, we were unable to detect significant changes in the release of selected cytokines by airway cells, in the presence or not of miR-181b inhibitor. This is likely to be related to the predominant antagonist action of LXA 4 on the release of inflammatory cytokines, which can be uncovered by pre-exposing cells to pro-inflammatory stimuli. Along these lines, the downstream signaling leading to the enhanced LXA 4 -induced phagocytic activity, observed when miR-181 was inhibited, requires further investigation. Whether elements of the autophagy cascade are involved, as recently reported in murine and human macrophages, remains to be determined 31 . This aspect is particularly relevant in CF, since autophay activators can correct misfolded ∆F508 CFTR and promote clearance of Pseudomonas aeruginosa by ∆F508 CFTR macrophages [42][43][44] .
Regardless of these limitations, our present data indicate that the LXA 4 -ALX/FPR2 signaling may represent a novel target to stimulate the anti-inflammatory, anti-microbial defense in CF.
In conclusion, here we uncovered the upregulation of miR-181b in CF cells, which contributes to impair the endogenous anti-inflammatory, anti-microbial defense pathway centered on the ALX/FPR2 receptor. Together, these results expand our knowledge of mechanisms of CF inflammation and point to regulatory mechanisms of ALX/FPR2 expression as to potential targets for the development of innovative pharmacology for CF.
Monocyte isolation and macrophage differentiation. Monocytes were isolated from peripheral blood (15 ml collected in sodium citrate-containing tubes) of healthy subjects [mean age 34 ± 4.6 (SD), 44% female] and age-and sex-matched CF patients [mean age 27 ± 5.4 (SD), 47% female] referring to the Regional Reference Center for Cystic Fibrosis, Atri (TE), Italy. Patients recruited for this study were free of pulmonary exacerbations at the time of recruitment and had not received i.v. antibiotics, nor steroids and non-steroidal anti-inflammatory drugs in the 2 weeks preceding blood withdrawal.
The study was approved by the Ethics Committee of the ASL Teramo, and carried out in accordance with the Declaration of Helsinki, as revised in 2004 and following the guidelines for observational studies published by AIFA (20.03.2008 GU n. 76 of 31.03.2008). Written informed consent was obtained by all human participants.
Primary human CF airway epithelial cells, isolated from different patients harboring the ∆F508 CFTR genotype subjected to lung transplantation, were obtained and provided by Dr. LJV Galietta (Telethon Institute of Genetics and Medicine, Pozzuoli, NA) as a public service of the Italian Cystic Fibrosis Research Foundation. Cells were grown in serum free growth medium (LHC9:RPMI 1640 1:1) supplemented with growth factors 45 . For differentiation, epithelia cells were cultured at high density (5 × 10 5 cells/cm 2 ) in air liquid interface condition on Transwell filters (0.4 µm pore Ø) for 8-10 days in differentiation medium (Ham's F12, 2% FBS) 46 . Epithelium formation was confirmed by measuring TEER. miR-181b analysis. miR-181b levels were determined as previously reported 26 . In brief, we used a silica-based spin column system (Norgen, Thorold, ON, Canada) for extraction of miRNA-enriched fractions. Samples were reverse-transcribed with the miScript II RT kit (Qiagen). Real-time PCR analyses were carried out with 1.5 ng of cDNA using specific primers (miScript Primer Assays) and a SYBR Green master mix (also from Qiagen) with a 7900HT Fast Thermal cycler (Invitrogen). miR-181b relative abundance was quantitated using the 2 −ΔΔCt method 47 . The exogenous cel-miR-39 or endogenous RNU6 and SNORD95 were used to normalize input cDNA. ALX/FPR2 expression. ALX/FPR2 mRNA was evaluated as in Pierdomenico et al. 26 . Total RNA was extracted using a silica-based spin column system (Norgen, Thorold, ON, Canada), quantified using a NanoDrop spectrophotometer (Thermo Scientific, Waltham, MA), and reverse-transcribed with the M-MLV Reverse Transcriptase (Sigma). Real-time PCR determinations were carried out using 500 ng of cDNA using the following primers: 5′-GGCCAAGACTTCCGAGAGAG-3′ (forward); 5′-CCGTGTCATTAGTTGGGGCT-3′ (reverse) and
ALX/FPR2 protein was assessed by flow cytometry. To this end, cells (1 × 10 6 /sample) were incubated with 0.5 µg of the Human FPRL1/FPR2 APC-conjugated Antibody (R&D Systems, Minneapolis, MN, USA). Analyses were carried out using a FACS Canto flow cytometer equipped with the Diva software (BD Bioscience). Total ALX/FPR2 was measured in cells permeabilized with Permeabilizing Solution 2 (BD Bioscience). Membrane ALX/FPR2 was assessed using non permeabilized cells.
Trans epithelial electrical resistance (TEER). CFBE41o-cells, transfected with 10 nM miR-181b inhibitor or non-targeting negative control single-strain RNA vector using INTERFERin (Polyplus Transfection TM) were seeded (6 × 10 5 cells/well) in Transwell permeable supports (6.5 mm insert, 24 well plate) (Corning, NY, USA) 24 h post transfection. The following day, cells were washed twice with DPBS and treated with LXA 4, LXA 4 + WRW4 or vehicle (0.01% EtOH) in DPBS. After 4 h, TEER was measured by a EVOM Voltmeter (World Precision Instruments, Sarasota, FL, USA). ALX/FPR2 promoter activity. CFBE41o-and 16HBE14o-cells were seeded in 6-well plates. After 24 h, cells were washed twice with PBS and supplemented with DMEM medium containing 1% FBS. Cells were then transfected with 4 μg of the pGL4 basic plasmid or pGL4 basic + the 346 bp sequence of the ALX/FPR2 core promoter as reported 28 . Luciferase activity was measured 48 h post-transfection using the luciferase assay kit (Promega) according to the manufacturer's protocol. Values were normalized for protein concentration as reported 28 . Statistical analysis. Results are reported as mean ± SEM. Statistical significance was evaluated by the Student's T-test. P values < 0.05 were taken as statistically significant.