Primary AM isolation and culture conditions

Human AMs or PMA (phorbol 12-myristate 13-acetate)-differentiated THP-1 cells were used for in vitro experiments. THP-1 cells were differentiated for 48 h with 10 nM PMA. For experiments with primary cells, AMs were isolated via ex vivo lavage from de-identified lungs rejected for transplant and obtained from Lifeline of Ohio Organ Procurements agency (Columbus, OH). All lungs were from subjects with no history of chronic lung disease or cancer and were non-smokers for at least 1 year. After collection, RBCs were lysed and cells were enumerated and frozen down in FBS and 10% DMSO. Prior to experiments, AMs were rapidly thawed, added to RPMI (10% FBS, 1% antibiotic/antimycotic), and centrifuged at 400 × g for 5 min. Total concentration and viability of cells were determined with trypan blue staining before seeding.

Primary polymorphonuclear cell isolation and culture conditions

Primary human PMNs were isolated from peripheral blood of healthy human donors by Ficoll density gradient separation method. Informed consent was obtained from all donors under an approved IRB protocol (IRB 2011H0059) in accordance with the Declaration of Helsinki. Briefly, 14 ml of Ficoll-Paque Premium sterile solution was added to the 35 ml blood/saline mixture slowly by a pipette, and then centrifuged at 1400 r.p.m. for 40 min at room temperature. The plasma was removed, and the PMN-rich red pellet was dissolved by adding cold 0.9% saline followed by cold 3% dextran, and incubated on ice for 30–40 min. After centrifugation, the upper PMN-rich layer was transferred to a new 50 ml conical tube, and 3× volume of cold 1× HBSS (without Ca & Mag & phenol red) was added. The PMN-rich cells were then pelleted by centrifugation at 1500 r.p.m. for 5 min at 4 °C and resuspended in 1× HBSS. Cell counts and viability were determined with trypan blue staining by automated cell counter. The purity of PMNs was assessed by differential cell count analysis following cytospin preparation. PMN cells were centrifuged again, and resuspended in phenol/FBS free RPMI supplemented with 1% penicillin/streptomycin (P/S) and seeded on collagen-coated transwell permeable membrane inserts at a cell density of about 2 × 10⁶ cells per insert. PMNs were allowed to attach for 30 min prior to the oscillatory pressure experiment. Cells without oscillatory pressure were used as a control. Cell viability after 16 h at air–liquid interface was assessed by live/dead staining using calcein and propidium iodide.

Pressure apparatus

Oscillatory pressure was applied to AMs at an air–liquid interface using a custom design apparatus²⁴. Briefly, AMs were seeded on Transwell 6-well inserts at a density of about 1 × 10⁶ cells per insert and allowed to adhere for 4 h in complete RPMI media supplemented with 10% FBS and 1% P/S at 37 °C and 5% CO₂. Media was exchanged and replaced with FBS-free RPMI supplemented with 1% P/S in the basal chamber and media was removed from the apical chamber. The well was sealed using rubber stoppers fixed to a central channel connected to the tubing. Tubes from each well were connected to a water manometer and a small animal ventilator (Harvard Apparatus, Holliston, MA, USA). Oscillatory pressure was applied for 16 h with a 0–20 cm-H₂O range at 0.2 Hz. The media were subsequently collected for analysis by ELISA, and cellular RNA was isolated for RT-qPCR.

Cell transfection

AMs were seeded on 6-well Transwell inserts as described above for transfection⁴⁴. Following adherence, AMs were transfected in Opti-MEM (with antibiotic supplement) with pre-miR-146a (Thermofisher Scientific, Waltham, MA, USA) loaded nanoparticles (200 nM miR) or lipofectamine (100 nM miR, Thermofisher Scientific) solution for 24 h. Following transfection, the media was replaced with RPMI and cells were subjected to pressure as described above.

Animal use

All animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) at Ohio State under protocols 2011A00000081-R2 and 2013A00000105-R1. The studies were carried out in compliance with all relevant ethical regulations. MiR-146a KO mice⁴⁵ (stock: 016239) were purchased from the Jackson Laboratory (Bar Harbor, ME, USA) and crossed with in-house WT C57Bl/6J mice. KO and WT litter-mate controls were used for all experiments, with groups matched for age (age range 8–12 weeks) and sex.

VILI protocol

To induce VILI, mice were ventilated for 4 h with a tidal volume (TV) of 12 ml kg⁻¹ to induce volutrauma and 0 cm-H₂O PEEP to induce atelectrauma using a Flexivent small animal ventilator (Scireq, Montreal, QC, Canada). Mice were anesthetized with ketamine and xylazine. Following induction of anesthesia, a tracheostomy cannula was placed, and mice were connected to the ventilator. Body temperature was maintained by using a heating pad beneath the animal. Baseline lung physiology measurements were obtained by performing a recruitment maneuver followed by forced oscillation to determine tissue elastance. Blood oxygenation was monitored using a mouse thigh pulse oximeter (Starr Lifesciences). At each hourly time-point, the lungs were recruited (two deep inflations) and lung physiology parameters were measured. Animals were volume resuscitated every 2 h ventilation with 10 μl g⁻¹ bolus saline. At the conclusion of ventilation or SB controls, mice were given a lethal overdose of anesthetic and BAL was performed by instilling 1 ml of PBS into the lungs twice and withdrawing it each time. BAL fluid was then centrifuged for 10 min at 500 × g and the supernatant was collected for subsequent analysis. Red blood cells were lysed with RBC lysis buffer, and differential stain was performed (Hema 3 Stat Pack, Thermofisher Scientific) according to the manufacturer’s instruction. The remaining cells were resuspended in TRIzol (Qiagen, Hilden, Germany) for RNA extraction. Following BAL, lungs were harvested and snap-frozen in liquid nitrogen. For whole lung gene expression, lungs were thawed, 750 μl of TRIzol was added, and homogenized using a handheld tissue homogenizer (Omni International, Kennesaw GA, USA).

Clodronate depletion

AMs were depleted using liposomal clodronate (Clodrosome, Brentwood, TN, USA). Mice were anesthetized with ketamine and xylazine and then administered 50 μl of 5 mg ml⁻¹ liposomal clodronate or unloaded liposomal controls via intratracheal instillation 48 h and 24 h prior to MV experiments.

Bone-marrow-derived macrophage adoptive transfer

BMDMs were isolated from the femur and tibia of 6–8-week-old WT or miR-146a KO mice. BMDMs were generated by differentiating bone marrow cells for 7 days in 20 ng ml⁻¹ MCSF⁴⁶. Following differentiation, BMDMs were trypsinized, counted, and resuspended at 20 × 10⁷ cells ml⁻¹. Fifty microliters of BMDM suspension (1 × 10⁶ total cells) was intratracheally instilled into mice 6–8 h prior to ventilation⁴⁷.

Fabrication of miR-146a-loaded LNPs

To overexpress miR-146a both in vitro and in vivo, we fabricated miR-146a-loaded LNPs. First, empty LNPs were fabricated using the ethanol dilution method where a mixture of lipids including 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE, Avanti Polar Lipids, Alabaster, AL, USA), 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP, Avanti Polar Lipids), linoleic acid (LA, Sigma Aldrich, Natick, MA, USA), and d-α-Tocopherol polyethylene glycol 1000 succinate (TPGS, Sigma Aldrich) were diluted in ethanol in a molar ratio of 40:10:48:2 for DOPE:DOTAP:linoleic acid:TPGS, respectively. This mixture was injected into 20 mM HEPES buffer and this solution was sonicated to form empty LNPs. Pre-miR-146a or pre-miR-scramble control (ThermoFisher Scientific) were mixed with PEI (2000 MW, Sigma Aldrich) at an N:P ratio of 10 (the ratio of moles of PEI amine groups to nucleic acid phosphate groups) in HEPES buffer and incubated at room temperature for 10 min to form PEI/miRNA polyplexes. The empty LNP and polyplex solutions were then mixed at a lipid to nucleic acid final mass ratio of 10 and this mixture was sonicated for 5 min at room temperature and then incubated for 10 min at room temperature. miR-146a-loaded LNPs were then concentrated to 1 nmol miR in 50 μL by spinning at 3500 × g with 10K NMWL centrifugal filters (EMD Millipore, Burlington, MA, USA). Particle size, PDI, and zeta potential were measured via dynamic light scattering using NanoZS (Malvern). The morphology of the nanoparticles was assessed on a Thermo Scientific Glacios Cryo-TEM instrument. Briefly, Cryo-TEM samples were prepared by adding 3 μl of the miR-146a-loaded nanoparticles onto a 400 mesh copper lacey carbon grid. The grid was plunge-frozen in liquid ethane using the Vitrobot system. Images were acquired using Glacios CryoTEM. Encapsulation efficiency was determined by agarose gel electrophoresis⁴⁸. Briefly, pre-miR-146a-loaded LNPs were lysed with 0.5% sodium dodecyl sulfate (SDS) (BioRad, Hercules, CA), and loaded on to 2% agarose (Fisher BioReagents) gel containing Labsafe nucleic acid stain (GBiosciences). Pre-miR-146a-loaded LNPs without SDS were also loaded to the gel for comparison. Gel electrophoresis was run at 100 V for 45–60 min in a 1× Tris-Acetate-EDTA (TAE) running buffer (Fisher BioReagents). Images of the gel were captured under UV light using ChemiDoc Imaging System (BioRad) and the amount of pre-miR-146a with and without SDS dissolution was quantitated. For in vitro experiments, cells were transfected with miR-loaded LNPs for 24 h prior to oscillatory pressure application. For in vivo experiments, 50 μl of miR-loaded LNPs (i.e. 1 nmol miR) were intratracheally instilled into mice 24 h and 0 h before ventilation.

Immunofluorescence staining

Fifty microliters of Cy3-labeled pre-miR negative control (Ambion, AM17120) loaded LNPs were administered intratracheally into WT mice. Control mice were treated with an equivalent volume of PBS without LNPs. Mice were euthanized 4 h after nanoparticle administration and the right heart was perfused with 1× Ca²⁺/Mg²⁺-free Dulbecco’s phosphate-buffer saline (PBS) (Gibco) followed by 2% (v/v) paraformaldehyde (PFA) (Thermo Scientific) in PBS. Lungs were inflated to 20 cm H₂O and then immersed in 2% PFA for 2 h. After fixation, lung tissues were embedded in optimal cutting temperature (OCT) compound (Fisher HealthCare) and frozen in 2-methylbutane (Honeywell) in a beaker submerged in liquid nitrogen. OCT embedded lungs were cryosectioned in the coronal plane at 7 μm using a cryostat microtome (Leica CM 1510S). Cryosections were fixed on slides with 2% PFA for 15 min at room temperature and washed several times with 1× PBS. Cryosections were permeabilized with 0.1% (v/v) Triton X-100 (Sigma) in PBS for 5 min, then incubated in blocking solution, 1% bovine serum albumin (BSA) (Fisher Scientific) in PBS, for 1 h at room temperature. The sections were immunostained for EpCAM (epithelial cells) and CD68 (macrophages) using fluorescently conjugated primary antibodies at a 1:50 dilution overnight at 4 °C. Rabbit anti-EpCAM antibody-Alexa Fluor 647 (ab237385), and rat anti-CD68 antibody-Alexa Fluor 488 (ab201844) were purchased from Abcam. Sections were washed three times with 1× PBS. Cryosections and coverslips were mounted with Immu-mount (Thermo Scientific). Images were acquired with a fluorescence microscope (Olympus IX81-ZDC), and processed using Fiji software.

Nanoparticle cellular distribution analysis

Nanoparticle distribution was quantitated by co-localization analysis of fluorescence images in MATLAB based on a previously published method⁴⁹. Monochromatic images of each channel (epithelial cells-Alexa Fluor 647; macrophages-Alexa Fluor 488; nanoparticles-Cy3) were converted to 8-bit images. Pixel area of the epithelial cells and macrophages were computed by thresholding images and counting above threshold pixels. The threshold level was determined using control images without staining to remove background autofluorescence. Pixel area of red nanoparticles was measured in a similar manner. Nanoparticle distribution in epithelial cells was measured as pixel area overlapping between the purple and red channels while nanoparticle distribution in macrophages was measured as pixel area overlapping between green and red channels. The percentage of nanoparticle distribution was computed as the ratio of the epithelial or macrophage cell area with nanoparticle overlapped area to the total nanoparticle area. The percentage co-localization was calculated as the ratio of epithelial or macrophage cell area overlapping with red nanoparticle area to the total pixel area of epithelial cells or macrophages, respectively.

Patient samples

After informed consent was obtained, excess BAL fluid was obtained from patients with suspected infection during a clinically indicated bronchoscopy at The Ohio State Wexner Medical Center under an approved IRB protocol (IRB 2016H0009) in accordance with the Declaration of Helsinki. Potential subjects were identified by the pulmonary consult and intensive care unit teams. All patients ≥18 years of age were considered eligible. BAL was completed by the subject’s treating physician and samples were sent to the clinical laboratory. Once all of the ordered tests had been performed, excess BAL fluid was put on ice and transported to our laboratory. Total and differential cell counts were performed prior to centrifugation at 300 × g. The supernatant was aliquoted and frozen at −80 °C. TRIzol was added to the cell pellets for future RNA isolation.

RNA extraction

RNA was extracted for downstream analysis by RT-qPCR using the standard phenol–chloroform RNA extraction and purification was performed with TRIzol according to the manufacturer’s protocol.

ELISA

Enzyme-linked immunosorbent assays (ELISA) were performed to assess secreted cytokine/mediator levels in media or BAL. For human cytokines, OptEIA kits (BD Biosciences, Franklin Lakes, NJ, USA) were used (IL6, IL8, IL1β, TGFβ, IL12) and the manufacturer’s protocol was followed. For mouse mediators, Duoset ELISA kits (R&D Systems, Minneapolis, MN, USA) were used (IL6, CXCL1/KC) and the manufacturer’s protocol was followed.

RT-qPCR

RT-qPCR with TaqMan primer/probes was used to determine relative expression levels of miR-146a compared to U18 or U6 (human) or sno251 (mouse) endogenous control genes (ThermoFisher, assay IDs: 478399_mir, 001204, 001093, 001236, respectively). TRAF6 relative expression was determined relative to GAPDH control (ThermoFisher, assay IDs: Hs00939742_g1 (TRAF6) Hs02786624_g1 (GAPDH) for human samples, and Mm00493836_m1 (TRAF6), Mm99999915_g1 (GADPH) for mouse). Following RNA extraction, cDNA was synthesized using a high-capacity cDNA transcription kit (Thermofisher). qPCR was then performed with Taqman Master Mix (Thermofisher) on Roche LightCycler480 and data were quantified by the ΔΔCT method.

Statistics

For laboratory-based experiments, statistical analysis was performed using Graphpad Prism 8. The distribution of all data was tested for normality via a Shapiro-Wilk test. All data are presented as mean + SEM except where noted otherwise. Experimental sample size was determined by performing a pilot experiment with six animals and the sample mean and standard deviation estimates were used to calculate the final sample size with a power of 0.8 and α = 0.05. To compare two groups with normal distribution, a Student’s t-test was used for comparisons or t-test on log-transformed data if data were log-normally distributed. A Mann–Whitney test was used to the compare ranks if the data were not normally distributed. For analysis of qPCR data, the statistical comparison was performed on log-transformed fold change data. For comparison among multiple groups, all data were normally or log-normally distributed and an ANOVA was performed on log-transformed or untransformed data as appropriate. For experiments with two-independent variables, a two-way ANOVA was performed after testing data for normality as described. If a significant effect was determined by ANOVA, a Tukey post-hoc test for individual group comparisons was performed. A p-value of <0.05 was considered statistically significant. Outliers were identified using non-linear regression via the ROUT method with a Q threshold of 1%⁵⁰. For the clinical data, association analyses between pairs of variables were conducted with Fisher’s exact tests (for categorical variables) and two-tailed t-tests or Kruskal–Wallis tests (for continuous variables as appropriate based on the normality of the data) using SAS version 9.4 (Cary, NC). Generalized linear models were used to assess for an association between variables with more than two groups for comparison.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.