All animal procedures were approved and supervised by the University of Rochester University Committee on Animal Resources (UCAR permit number 2007-127). For euthanasia, mice were anesthetized with an intraperitoneal injection of avertin (2,2,2-tribromoethanol, 250 mg/kg), followed by exsanguination and removal of the lungs for analysis, and all efforts were made to minimize suffering. Primary human lung fibroblasts were derived from culture of tissue explants following biopsy specimens from patients undergoing lung resection surgery for hamartoma, a benign lung tumor. The fibroblasts were grown tissue from anatomically normal areas of the specimen distal to the lesion. Human blood monocytes were isolated from peripheral blood collected from healthy donors. All blood and tissue donors provided informed written consent as described in study protocols approved by the University of Rochester Institutional Review Board and conforming to the Helsinki Declarations. Primary human small airway epithelial cell cultures were obtained from a commercial source.

Primary human lung fibroblast cultures were established from lung tissue explants as previously described [28] and used at early passage (P3–P9). Primary human small airway epithelial cells (SAEC) were purchased from a commercial vendor (Lonza, Allendale, NJ) and cultured in small airway epithelial cell growth medium supplemented as recommended by the supplier. Blood monocytes were cultured in RPMI 1640 as described [29].

Serum starvation was performed to reduce the effect of serum components on induction of cyclooxygenase-2 (COX-2) and other inflammatory mediators. Aqueous cigarette smoke extract (CSE) was prepared as previously described [30]. Briefly, smoke from two reference cigarettes (1R3F, University of Kentucky) was bubbled through 20 ml of serum free medium. This is arbitrarily defined as “100% CSE.” To adjust for batch-to-batch variation, the average absorbance at 320 nm for several batches of CSE was measured and determined to be 0.65 absorbance units. Thereafter, CSE preparations are diluted as needed so that 100% CSE has a standard OD₃₂₀ = 0.65. The pH of the CSE is adjusted to 7.4, and the CSE is then sterile filtered with a 0.45-µm filter (25-mm Acrodisc; Pall, Ann Arbor, MI). CSE was freshly prepared for each experiment and diluted with serum-free MEM immediately before use. Cultures were pre-treated with 10 nM, 100 nM or 1 µM RvD1 or vehicle control solution (0.1% ethanol) for 24 hours prior to the addition of 1% or 5%CSE or 1 ng/ml IL-1β (R&D Systems, Minneapolis, MN) for an additional 24 hours. As the RvD1 was supplied in ethanol, vehicle control cultures were treated with the same final concentration of ethanol (0.1%). Inflammatory mediators were determined in cell culture supernatants as described below. Cell lysates were collected and analyzed by western blotting using an antibody specific for COX-2 (Cayman Chemical, Ann Arbor, MI) and or GAPDH (Abcam, Cambridge, MA), which serves as a normalization control.

Adult female C57BL/6J mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and used at 8–12 weeks of age. Mice were exposed to dilute mainstream cigarette smoke (CS) for 1 hour, twice per day, for 3 days, as described [31]. Resolvin D1 (7S,8R,17S-trihydroxy-4Z,9E,11E,13Z,15E,19Z-docosahexaenoic acid) and 17R-RvD1 (7S,8R,17R-trihydroxy-4Z,9E,11E,13Z,15E,19Z-docosahexaenoic acid) were obtained from Cayman Chemical. In the pre-treatment exposure model, 100 ng RvD1 in 40 µl saline was given by oropharyngeal aspiration on day 0, and 1 hour prior to the first smoke exposure on days 1, 2 and 3. Control animals were treated with saline vehicle. Smoke exposures were quantified by determining total suspended particulate matter (TSP) by gravimetric sampling. The smoke concentration was set at a nominal value of 450 mg/m³ TSP by adjusting the number of cigarettes loaded onto the carousel and the flow rate of the dilution air. Control animals were exposed to filtered air in identical equipment on the same schedule. Mice were euthanized on day 4 and cells and tissues were harvested for analysis. To analyze the resolution of active inflammation, mice were exposed to CS for 2 hours per day on days 1–3. On day 3, after the final smoke exposure, and on days 4 and 5, the mice were treated with vehicle or 100 ng RvD1 by oropharyngeal aspiration as described. The mice were euthanized and analyzed on day 4 and 6.

BALF was collected from smoke-exposed mice and analyzed as previously described [31]. The right lung was frozen in liquid nitrogen for later analysis. Total protein in BALF was determined by the bicinchoninic acid (BCA) colorimetric assay (Thermo Scientific, Rockford, IL). The right lung was homogenized in RIPA buffer as described [32], and lung proteins were analyzed for expression of COX-2 (Clone CX229, Cayman Chemical, Ann Arbor, MI), Arg-1 (H-52; Santa Cruz Biotechnologies, Santa Cruz, CA) and total actin (CP01; Calbiochem, San Diego) by Western blotting. In some experiments, the lungs were inflated and fixed with 10% neutral buffered formalin without undergoing lavage. Tissues were embedded with paraffin, sectioned (5 µm), and stained with hematoxylin and eosin (H&E).

Lung-infiltrating neutrophils were stained with rat anti-mouse Ly-6B.2 (MCA771GA, Serotec, Oxford, UK) and HRP-conjugated goat anti-rat antibodies, followed by visualization with AEC chromagen (Vector Lab, Burlingame, CA) and hematoxylin counter staining.

Human IL-6, IL-8 and MCP-1 were measured in cell culture media by commercial ELISA according to the manufacturer’s instructions (R&D Systems). Prostaglandin E₂ (PGE₂) was measured by competitive enzyme immunoassay (EIA) using commercially available reagents as described(Cayman Chemical) [33]. Mouse cytokines and chemokines were measured in BALF by ELISA (MIP-2, KC; R&D Systems) or by bead-based multiplex analysis of lung homogenates (IL-1β, IL-6, MCP-1, IFN-γ and IL-10; Millipore, Billerica, MA) using a Luminex FlexMAP3D instrument (Luminex, Austin, TX). Results are expressed as picograms of cytokine per milligram of total protein in the homogenate.

Lungs were harvested from mice exposed to smoke and treated with RvD1 as described above. One lobe was digested with collagenase and a single cell suspension was prepared. Total lung macrophages were isolated by CD11b magnetic bead separation (Miltenyi Biotech, Cambridge, MA) [34]. The cells were then labeled with FITC-conjugated anti-mouse F4/80 antibody (BM8; Invitrogen; Camarillo, CA) for 30 minutes, then fixed with 4% of paraformaldehyde and permeabilized with 0.1% Triton X-100. Ingested neutrophils were labeled by staining permeabilized cells with PE-conjugated anti-mouse Gr-1 (RB6-8C5; BD Pharmingen) antibodies for 30 minutes and analyzed by flow cytometry (FACSCanto II; BD Biosciences, San Jose, CA). Forward and side scatter was first used to exclude doublets or macrophages that are attached to but did not fully engulf neutrophils. Cells were then gated based on F4/80 positive staining. The frequency of cells that are F4/80 and GR-1 double positive was then quantified.

Alveolar macrophages were harvested from naïve mice by lavage in PBS buffer containing 0.6 mM EDTA and glucose (0.1%). Macrophages were activated with 1 ng/ml IL-1β for 30 minutes and then incubated with RvD1 (100 nM in 0.1% ethanol) or 0.1% ethanol alone for an additional 30 minutes before the addition of a 10-fold excess of FITC-conjugated latex microspheres for 1 hour. Uptake of FITC-labeled latex microspheres was measured by flow cytometry (FACSCanto II, BD Biosciences).

Lungs were lavaged extensively (0.5 ml for 5–10 times) with PBS containing 0.1% glucose to collect lung macrophages. Collected cell suspensions were enriched for macrophage by adherence on tissue culture-treated plastic dishes for 1 hour in minimum essential medium (MEM; Gibco, Camarillo, CA) containing 10% fetal bovine serum. Non-adherent cells were discarded by gently washing twice with PBS. RNA was extracted from the enriched adherent macrophages by RNeasy kit according to manufacturer’s instructions (Qiagen; Valencia, CA). RNA (1 µg) was reverse-transcribed using the iScript cDNA synthesis Kit (BioRad; Hercules, CA) and the cDNA was subjected to quantitative real-time PCR using SYBR GreenER in a BioRad iQ5 cycler. In brief, a 20 µl mixture was used containing 10 µl iQ SYBR Green Supermix, 0.5 µl forward and reverse primer, 7.5 µl sterile water, and 2 µl of the 1∶5 diluted cDNA template. The real-time PCR was performed under the following conditions: 1 cycle at 95°C for 3 minutes, then 40 cycles at 95°C for 10 seconds, 55°C for 30 seconds, 95°C for 1 minute followed by 55°C for 1 min. The intron-spanning primers were designed by using sequence information from the NCBI database. The Ct values were normalized to the endogenous control (18sRNA). Primer sequences are listed below:

m18sRNA: 5′-GCTTGCTCGCGCTTCCTTACCT-3′

m18sRNA: 5′ TCACTGTACCGGCCGTGCGTA-3′

mArg1-F: 5′- CAGAAGAATGGAAGAGTCAG–3′

mArg1-R: 5′- CAGATATGCAGGCAGGGAGTCACC -3′

mMrc1-F: 5′- CTCTGTTCAGCTATTGGACGC -3′

mMrc1-R: 5′- CGGAATTTCTGGGATTCAGCTTC-3′

miNOS-F: 5′- GTTCTCAGCCCAACAATACAAGA -3′

miNOS-R: 5′- GTGGACGGGTCGATGTCAC -3′

mTNFα-F: 5′- GACGTGGAACTGGCAGAAGAG -3′

mTNFα -R: 5′- TTGGTGGTTTGTGAGTGTGAG -3′

mIL-10-F: 5′- GCTCTTACTGACTGGCATGAG -3′

mIL-10-R:5′- CGCAGCTCTAGGAGCATGTG -3′

Results are reported as the mean ± SEM or mean ± SD as indicated. T-test (two tailed) or one-way and two-way analysis of variance (ANOVA) with Bonferroni multiple comparisons were performed using GraphPad Prism version 5.0d for Mac (GraphPad Software, La Jolla California USA, www.graphpad.com). A P value <0.05 was considered significant.

Lung fibroblasts produce many pro-inflammatory molecules and amplify inflammatory cascades following exposure to pro-inflammatory insults including cigarette smoke and IL-1β [35]. To evaluate the ability of RvD1 to suppress inflammatory signaling in lung fibroblasts, primary human lung fibroblasts were pre-treated for 24 hours with the indicated concentrations of RvD1 then stimulated with cigarette smoke extract (CSE) or IL-1β for an additional 24 hours. IL-1β strongly induced production of pro-inflammatory mediators including IL-6, IL-8, MCP-1 and PGE₂ (Figure 1). RvD1 significantly inhibited the production of MCP-1 and PGE₂ at 100 nM, and production of IL-6, IL-8, MCP-1 and PGE₂ at 1000 nM. CSE also induced these pro-inflammatory mediators in human lung fibroblasts, although at lower levels than IL-1β. IL-6, IL-8, MCP-1 and PGE₂ were significantly attenuated by RvD1 treatment (Figure 1, panels E, F, G and H). The pro-inflammatory enzyme COX-2 is responsible for PGE₂ production and is elevated in a variety of inflammatory pulmonary diseases include COPD [36]. IL-1β and CSE both induced COX-2 expression in human lung fibroblasts. Exposure to RvD1 (1000 nM) significantly attenuated COX-2 expression (Figure 1I), suggesting a potent anti-inflammatory function for RvD1 in human lung fibroblasts.

Macrophages and epithelial cells are important sources of inflammatory cytokines in cigarette smoke-exposed lungs. To evaluate the effect of RvD1 on these cell types, human primary small airway epithelial cells (SAECs) and blood-derived monocytes were pre-treated with the indicated concentrations of RvD1 before the addition of IL-1β or CSE for an additional 24 hours. RvD1 significantly dampened production of IL-6 and IL-8 from SAEC treated with IL-1β (Figure 2A–B). In our hands, IL-1β did not stimulate release of PGE₂, and CSE did not stimulate production of IL-6 or IL-8 from SAEC (data not shown). Both IL-1β and CSE induced expression of IL-6 and IL-8 from human monocytes. A one-hour pre-treatment with RvD1 significantly inhibited production of IL-6 and IL-8 in a dose-dependent manner (Figure 2C–F). In these experiments the monocytes produced high background levels of PGE₂ which was not significantly increased further by CSE, therefore an effect of RvD1 on PGE₂ production could not be determined (data not shown). Interestingly, monocytes were more resistant to CSE than fibroblasts and required treatment with 5% CSE to induce a strong response, but were more sensitive to RvD1, with potent inhibition seen at 10 nM (Figure 2C–F).

Based on our in vitro results, we hypothesized that administration of RvD1 would attenuate cigarette smoke-induced acute lung inflammation in vivo. Mice received 100 ng RvD1 by inhalation 1 hour prior to cigarette smoke exposure daily for 3 days. On day 4, the mice were euthanized, bronchoalveolar lavage fluid (BALF) was collected, and inflammation was evaluated by differential cell counting. Cigarette smoke (CS) exposure induced significant increases in total BAL cells, neutrophils and lymphocytes, but not macrophages or eosinophils (Figure 3), consistent with prior experience with this model [37], [38]. The increase in total cell number is almost entirely due to recruitment of neutrophils (Figure 3C). CS-exposed mice pre-treated with RvD1 had significantly fewer neutrophils in BALF compared to vehicle-treated controls, expressed either as neutrophil number (Figure 3C) or percentage (Figure 3F) of total cells. There was no change in the absolute numbers of macrophages between CS and CS+RvD1-treated mice (Figure 3B). As noted, RvD1 had no significant impact on the numbers of BAL lymphocytes (Figure 3D) and eosinophils (Figure 3E). Interestingly, while total macrophage numbers were not changed, there was a significant increase in the percentage of macrophages in the BALF with RvD1 treatment (Figure 3F).

The location of infiltrating neutrophils in lung tissues was determined by immunohistochemical staining for the neutrophil specific marker Ly-6B.2. In RvD1-treated, air-expose lung, neutrophils were sparsely detected (Figure 4A and 4B). Acute smoke exposure causes accumulation of neutrophils around vessels (v), airways (a) (Figure 4C, arrows) and in the parenchyma (Figure 4D), with neutrophils prominent in the interstitial spaces (Figure 4D, arrows) and the alveoli (Figure 4D, solid triangles). RvD1 reduces the appearance of CS-induced neutrophils in lung tissue consistent with the results of differential cell counts in BALF that RvD1 reduces neutrophil accumulation in the lung.

Levels of pro-inflammatory cytokine and chemokine were determined in BALF and lung homogenates of smoke-exposed mice. Total protein in BALF, an indication of vascular leakage and edema, was also increased with CS exposure and significantly reduced with RvD1 treatment (Figure 5A). Acute cigarette smoke exposure induces the expression of neutrophil chemoattractants KC and MIP-2, which was significantly inhibited by RvD1 (Figure 5, panels B and C). CS exposure also resulted in increased production of pro-inflammatory cytokines including IL-1β, IL-6, and MCP-1. RvD1 markedly decreased IL-1β and IL-6 but not MCP-1 (Figure 5, panels D-F). IFN-γ content was increased with CS and decreased with CS+RvD1 treatment, however the difference was not significant (Figure 5G). Interestingly, the anti-inflammatory cytokine IL-10 was significantly increased in lungs from CS-exposed mice pre-treated with RvD1 (Figure 5H). We also evaluated expression of COX-2 protein and PGE₂ levels in lung homogenates, which showed results consistent with our in vitro findings that CS increases PGE₂ production and COX-2 expression, which is strongly inhibited by RvD1 (Figure 5I and 5J, respectively).

Resolvins are reported to be pro-resolving as well as anti-inflammatory [39]. To evaluate whether resolvins could accelerate the resolution of CS-induced inflammation, we exposed mice to CS for 3 days, and initiated resolvin treatment after the final smoke exposure. For these experiments, we used the 17R epimer of RvD1 (17R-RvD1) as it is less susceptible to in vivo inactivation [4]. RvD1 and 17R-RvD1 bind the same receptor and have similar activities in vitro [21]. Mice were euthanized on day 4 (after a single 17R-RvD1 treatment on day 3) or day 6 (after receiving 17R-RvD1 on days 3, 4 and 5). A single dose of 17R-RvD1 accelerated the resolution of acute CS-induced inflammation, with significant reductions in total BAL cells, macrophages and neutrophils on day 4 (Figure 6, panels A, B and C). Although the number of inflammatory cells was significantly reduced by 17R-RvD1, the percentages of neutrophils and macrophages did not significantly change, as the number of macrophages, neutrophils, and total cell numbers were reduced together (Figure 6E, F). Inflammation on day 6 was also reduced with 17R-RvD1, although less dramatically than on day 4. 17R-RvD1 also significantly reduced production of the pro-inflammatory cytokine IL-6 (Figure 6F), neutrophil chemokines KC and MIP-2 (Figure 6G and H, respectively) and PGE₂ (Figure 6I).

We hypothesized that one mechanism by which RvD1 promotes resolution of CS-induced inflammation is by increasing the phagocytosis of neutrophils by lung macrophages. To investigate this mechanism, lung macrophages were harvested from CS-exposed mice, fixed and permeabilized, then co-stained for the macrophage marker, F4/80, and the neutrophil marker Gr-1. Treatment with RvD1 increased the percentage of F4/80 and Gr-1 double positive cells, suggesting that clearance of apoptotic neutrophils is enhanced in the presence of RvD1 (Figure 7A). We also investigated whether RvD1 increased macrophage phagocytosis ex vivo. Alveolar macrophages were isolated from naïve mice and stimulated with IL-1β (1 ng/ml) for 30 minutes, followed by RvD1 or vehicle for and additional 1 hour. The phagocytic ability of the macrophages was assessed using FITC-labeled latex beads and analyzed by flow cytometry. RvD1 enhanced overall bead uptake by 59% (Figure 7, panel B and C). In addition to increasing the percentage of macrophages that had phagocytosed at least one bead, RvD1 also increased the number of beads phagocytosed per cell, with a 220% increase in the percentage of cells ingesting 3 and a 660% increase in the percentage of cells ingesting 4 or more beads (Figure 7C).

Given the increased IL-10 production and increased phagocytic activity seen with RvD1 treatment, we hypothesized that RvD1 drives macrophage polarization to an M2 phenotype. RvD1 increased protein levels of arginase I (Arg1), a marker of M2 macrophages, in total lung homogenates (Figure 8A). We also extracted RNA from lung macrophages from mice exposed to air or CS, with or without RvD1 treatment, and determined mRNA levels by quantitative RT-PCR. In macrophages from air-exposed mice, RvD1 induced the M2 genes Arg1, mannose receptor 1 (Mrc1) and IL-10 demonstrating that the number of M2 macrophages increased after RvD1 treatment in the absence of inflammatory stimuli (Figure 8B–D). In CS-exposed mice, RvD1 significantly increased mRNA expression of Arg1, Mrc1 and IL-10 from alveolar macrophages compared to vehicle-treated CS-exposed mice (Figure 8, panels B-D). RvD1 significantly reduced mRNA expression of iNOS and TNF-α (Figure 8, panels E and F), two mediators highly expressed in classically-activated (M1) macrophages. Taken together, these results clearly demonstrate that RvD1 significantly augments the polarization of alveolar macrophages to the M2 phenotype.