Animals and COPD model
Female Balb/c mice (20-22 g) were obtained from the University’s main animal care center. All experiments were performed in accordance with the National guidelines for the care of laboratory animals and the study was approved by the Ethical Committee of the College of Applied Medical Sciences (agreement number: CAMS05/3334). During the different experimental procedures, animals were anesthetized by intramuscular administration of a mixture of 0.1 mL of 4 mL of ketamine (50 mg/mL), 1 mL of xylazine (2 %), and 5 mL of physiological serum. COPD model was induced by intrapulmonary instillation (1 mg.kg−1; V = 100 μl) of Lipopolysaccharide (LPS) from Escherichia Coli (Santa Cruz Biotechnology, Inc., CA, USA) using a MicroSprayer aerosolizer (Penn-Century Inc., PA, USA). At 48 h post LPS challenge, mice were intrapulmonary instilled with either physiological saline, iron labeled ex vivo polarized M1 or M2 macrophages (106 cells suspended in 100 μl phosphate buffered saline (PBS) solution).
Macrophages polarization and magnetic labeling
Bone marrow (BM) derived M1 and M2 macrophages were obtained and polarized as previously reported . Briefly, BM cells from tibiae and femora of donor mice were incubated for 7 days at 37 °C in complete IMDM medium containing L-glutamine and phenol red (Gibco, Lifetechnologies, CA, USA) and supplemented with 10 ng/ml of macrophage clone stimulating factor (R&D systems, Abingdon, UK) to obtain adherent M0 macrophages. Macrophage polarization was then induced by incubating adherent M0 cells for 20 h at 37 °C in complete IMDM medium supplemented with 1 ng.ml−1 LPS (Santa Cruz Biotechnology, Inc., CA, USA) and 10 ng.ml−1 INFγ to obtain M1-polarized cells or with 10 ng.ml−1 IL-10 and 20 ng.ml−1 IL-4 (R&D systems, Abingdon, UK) to obtain M2-polarized macrophages.
M1 or M2 macrophages were labeled with Dextran-coated SPIO nanoparticles (Micromod Partikeltechnologie GmbH, Germany) at an extracellular iron concentration of 2 mM with 1 h incubation time at 37 °C, which was chosen as the best compromise between labeling efficiency and biocompatibility to the cells. For efficient macrophages labeling, SPIO nanoparticles were functionalized by the addition of polyethylene glycol (PEG) with amine terminal (NH2) . We have previously reported that these nanoparticles showed an enhanced labeling efficiency of macrophages with a better biocompatibility .
Macrophages were incubated in serum-free RPMI medium containing L-glutamine and phenol red (Gibco, Lifetechnologies, CA, USA) to avoid proteins interactions with the uptake mechanism . The incubation step was followed by an overnight chase period in SPIO-free culture medium to allow sufficient time for iron oxide internalization. Prior to administration, iron content in the different macrophages subsets were quantified using Ferrozine-based assay and their biocompatibility was assessed using MTT assay as previously reported .
Ferrozine is an iron-chelating agent that forms a complex with iron and exhibits characteristic UV/Vis absorption. By comparing to a standard calibration curve, accurate quantification of iron uptake by macrophages can be obtained. Briefly, 2x105 labeled M1 or M2 macrophages were digested with 5 M hydrochloric acid and absorbance was measured at 351 nm using Multiskan Go Spectrophotometer (Thermo Scientific, NH, USA). Using a calibration curve of the absorbance vs. iron concentration of the SPIO nanoparticles prepared under the same experimental protocol, the iron content in the macrophages was determined and expressed as pg of iron per cell.
Cell viability was evaluated by MTT (3-[4,5-dimethylthiazol-2-yl]-2,5- diphenyltetrazolium bromide) Cell Growth Assay Kit (Merck Millipore, MA, USA) according to the manufacturer protocol. Briefly, iron labeled M1 and M2 macrophages (104 cells/well) were placed in a 96-well plate (n = 3) and absorbance was measured using Multiskan Go Microplate Spectrophotometer (Thermo Scientific, NH, USA) with a test wavelength of 570 nm and a reference wavelength of 630 nm. The relative percentage of cell viability for each condition was calculated related to unlabeled M1 and M2 macrophages.
Magnetic resonance imaging
To noninvasively monitor the biodistribution of M1 or M2 iron-labeled macrophages subpopulations in LPS-induced pulmonary model, mice were imaged using a 4.7 T Pharmascan 47/16 Bruker magnet interfaced to ParaVision 5.1 software (Bruker Biospin GmbH, Rheinstetten, Germany). A free-breathing MR imaging protocol was optimized to allow simultaneous detection of macrophages subsets and the visualization of inflammation progression in the lung using a radial UTE sequence (TR/TE = 100/0.4 ms) with 100 x 100 μm pixel resolution according to the protocol previously reported . Pulmonary MRI was performed on mice (n = 6 per group) before LPS challenge (Control), 48 h post LPS challenge (LPS) chosen as time 0 for investigating the effect of macrophages intrapulmonary administration on the resolution of lung inflammation, and at 4 h, 24 h, 48 h, 72 h and 168 h after either iron-labeled M1 or M2 macrophages subsets administration (LPS + M1 or LPS + M2, respectively). A set of 12 consecutive axial slices with 1 mm thickness were positioned approximately at the same level for all the animals in order to cover the whole lung volume during the 1-week follow-up study.
MR images were analyzed with freeware medical image analysis software (MIPAV, National Institutes of Health, Bethesda, MD, USA). The inflamed lung volume (ILV) was quantified with a semiautomatic segmentation procedure . Briefly, a low-intensity threshold was applied to exclude the non-inflamed regions from the lung images. LPS-induced inflammatory regions were extracted with a semiautomatic tool capable of generating contours whilst moving over the different lung structures by recognizing the intensity level of each pixel and selecting regions according to the default preset parameters (MIPAV ‘LevelSet’ active-contour algorithm). The total volume of high-intensity signals was computed by multiplying the slice thickness by the sum of all the areas segmented in the 12 consecutive slices. The segmentation parameters were the same for all the analyzed images, chosen to segment regions corresponding to high-intensity signals. As the signals from the LPS-induced edema and the vessels were of comparable intensities, the volume corresponding to the vessels was assessed on control images and then subtracted from the volumes determined on post-challenge images.
Computed tomography imaging
To monitor using an alternative noninvasive imaging modality the progression of LPS-induced inflammation and assess the possibility of inflammation resolution after intrapulmonary administration of either M1 or M2 macrophages subpopulations, mice (n = 6 per group) were scanned using a dedicated small animal high resolution micro-CT scanner (SkyScan 1176, Kontich, Belgium). The following parameters were used: 50 kV, 1 mm Al filter, 385 μA source current, 600 ms exposure time. Micro-CT imaging was only performed at 24 h, 72 h and 168 h post-macrophages administration on the same mice, which were previously imaged using MRI. Mice were allowed to recover for 4 to 6 h after the MRI acquisitions. CT scans were limited to three time points during the 1-week investigation to avoid high radiation exposure to the mice (i.e. ~ 0.1Gy per scan) that were imaged at the different investigation time points. Projection images were recorded in steps of 0.5 degrees from 0 to 360 degrees. Images were acquired throughout the spontaneous respiratory cycle. Respiratory motions were recorded with a visual camera, detecting the up- and downward movement of the thorax and then translated into a pseudo-sinusoidal signal to allow retrospective respiratory gating. 3D acquisitions with a 18 μm isotropic resolution were performed to cover the entire lung for a total acquisition time of 24 min. Images were reconstructed with SkyScan NRecon software (version 18.104.22.168) using a Feldkamp cone-beam reconstruction algorithm. Reconstruction parameters were smoothing “6”, beam-hardening correction “22 %”; post-alignment and ring artifact correction were optimally set for each individual scan. Reconstructed 3D images have a total of 864 slices with isotropic 18 μm voxel size and 1024 × 1024 resolution. A water phantom was used to calibrate the image to Hounsfield units (HU). Lungs were automatically delineated in the 3D micro-CT images as previously described  and the airways were removed from the lung delineation . Inflamed lung volume (ILV) was finally measured on the segmented lungs to quantify the level of LPS-induced inflammation detected using CT and subtracted from values quantified in control mice to exclude the volume corresponding to the vessels and correlate with MRI readouts.
Lungs (n = 3 in each group and at each time point) were directly removed after the micro-CT scanning and fixed overnight in 4 % paraformaldehyde. Processing of tissues for histological analyses was performed on sets of consecutive 5 μm thick sections. F4/80 rat monoclonal antibody (1:100) as universal marker for macrophages, NOS2 rabbit polyclonal antibody (1:1000) and Arginase goat polyclonal antibody (1:100) (Santa Cruz Biotechnology, Inc., CA, USA) as marker for M1 and M2 macrophages respectively, were applied as primary antibodies. Respective mouse ABC staining systems were used as sources for secondary antibodies (Santa Cruz Biotechnology, Inc., CA, USA) and consequent detection was performed as per manufacturer’s instructions. Slides were counterstained with hematoxylin before being observed using a BX53 Olympus microscope and analyzed using CellSens Entry digital imaging software (Olympus Corporation, Tokyo, Japan).
Data were presented as the mean standard deviation. ANOVA with post-hoc analyses were performed using SPSS software v12.0 (SPSS Inc., Chicago, IL, USA) to compare ILV assessments among the different groups. A p-value < 0.05 was considered significant for all tests.