Early gestational mesenchymal stem cell secretome attenuates experimental bronchopulmonary dysplasia in part via exosome-associated factor TSG-6

Background Mesenchymal stem cells (MSCs) are promising tools for the treatment of human lung disease and other pathologies relevant to newborn medicine. Recent studies have established MSC exosomes (EXO), as one of the main therapeutic vectors of MSCs in mouse models of multifactorial chronic lung disease of preterm infants, bronchopulmonary dysplasia (BPD). However, the mechanisms underlying MSC-EXO therapeutic action are not completely understood. Using a neonatal mouse model of human BPD, we evaluated the therapeutic efficiency of early gestational age (GA) human umbilical cord (hUC)-derived MSC EXO fraction and its exosomal factor, tumor necrosis factor alpha-stimulated gene-6 (TSG-6). Methods Conditioned media (CM) and EXO fractions were isolated from 25 and 30 weeks GA hUC-MSC cultures grown in serum-free media (SFM) for 24 h. Newborn mice were exposed to hyperoxia (> 95% oxygen) and were given intraperitoneal injections of MSC-CM or MSC-CM EXO fractions at postnatal (PN) day 2 and PN4. They were then returned to room air until PN14 (in a mouse model of severe BPD). The treatment regime was followed with (rh)TSG-6, TSG-6-neutralizing antibody (NAb), TSG-6 (si)RNA-transfected MSC-CM EXO and their appropriate controls. Echocardiography was done at PN14 followed by harvesting of lung, heart and brain for assessment of pathology parameters. Results Systemic administration of CM or EXO in the neonatal BPD mouse model resulted in robust improvement in lung, cardiac and brain pathology. Hyperoxia-exposed BPD mice exhibited pulmonary inflammation accompanied by alveolar-capillary leakage, increased chord length, and alveolar simplification, which was ameliorated by MSC CM/EXO treatment. Pulmonary hypertension and right ventricular hypertrophy was also corrected. Cell death in brain was decreased and the hypomyelination reversed. Importantly, we detected TSG-6, an immunomodulatory glycoprotein, in EXO. Administration of TSG-6 attenuated BPD and its associated pathologies, in lung, heart and brain. Knockdown of TSG-6 by NAb or by siRNA in EXO abrogated the therapeutic effects of EXO, suggesting TSG-6 as an important therapeutic molecule. Conclusions Preterm hUC-derived MSC-CM EXO alleviates hyperoxia-induced BPD and its associated pathologies, in part, via exosomal factor TSG-6. The work indicates early systemic intervention with TSG-6 as a robust option for cell-free therapy, particularly for treating BPD. Electronic supplementary material The online version of this article (10.1186/s13287-018-0903-4) contains supplementary material, which is available to authorized users.


Bronchoalveolar lavage (BAL)
Mouse pups were euthanized by pentobarbital overdose (250 mg/kg intraperitoneal) at PN14. The trachea was exposed by blunt dissection, and a 24GA 0.56in caliber tube (BD Insyte-N TM Autogaurd shielded IV catheter) was inserted into the airway and secured. Three washes of 0.2 ml PBS were instilled and gently aspirated and pooled to obtain BAL fluid (BALF). A small volume of BALF was used to calculate total cell counts using a hemocytometer. BALF samples were then cytocentrifuged at 1500 g for 15 min at 4º C to recover cells and the supernatant were collected and stored at -80º C to estimate total BALF protein concentration using BCA kit (Pierce, Rockford, IL), as per the manufacturer's recommendations. Differential cell counts were obtained after staining with HEMA stain as previously described (4). Absolute neutrophil count was determined by counting a minimum of 300 cells in each sample and multiplying it with its total cell number.
Percentage macrophage was calculated by examining the HEMA stained cells under microscope at 200x magnification. Five to six non-overlapping fields were examined to count a minimum of 300 cells. Percentage macrophages were expressed as (total number of macrophages/total number of cells) x100.

Western Blot analysis
The tissue was homogenized in RIPA buffer and protease inhibitor cocktail (Santa Cruz; USA) using tissue disrupter homogenizer. After centrifugation, the supernatant was lysed and subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Samples equivalent to 30 µg of protein content was loaded and size separated by 4% -20% gradient SDS-PAGE (BioRad). The protein on the acrylamide gel were transferred to a polyvinylidene difluoride 6 (PVDF) membrane (Millipore, Bedford, MA, USA) or nitrocellulose membrane at constant voltage 90 V in a transfer buffer containing 25 mM Tris and 192 mM glycine, pH 8.4. The PVDF membrane was blocked in PBS with 5% (w/v) skimmed milk overnight followed by incubation with the appropriate primary antibodies TSG-6 (R&D systems, 1:800) overnight at 4 o C. The membrane was washed thoroughly with PBS with 0.1% (v/v) Tween, five washes for 5 minutes each. The membrane was then incubated with horseradish peroxidase conjugated secondary antibody (anti-mouse IgG2B, Santa Cruz Inc.) and β-actin HRP-conjugate (Cell Signaling Technology, 1:3000) or secondary antibodies labeled with IRDye near-infrared fluorescent dyes (LI-COR Biosciences, 1:10,000). for 1h at room temperature. After PBS-Tween wash, the membranes were visualized by a chemiluminescent ECL detection kit from Perkin Elmer Inc. or by Odyssey Imaging system (LI-COR). For CD81 and Alix-1 staining, the blots were incubated with primary antibody mouse CD81 (Santa Cruz Inc. 1:200) or mouse Alix-1 (Santa Cruz Inc; 1:200) followed by incubation with anti-mouse LI-COR fluorescent secondary antibody. After PBS wash, the blots were scanned at appropriate wavelength using Odyssey Imaging system.

Dot Blot Assay
CM, EXO and their respective control fractions were spotted on the methanol-treated PVDF membranes and allowed to air dry at least for 2 h. The blots were rinsed briefly in PBS and were incubated overnight at 4 o C in 5% (w/v) skimmed milk in PBS to block the residual binding sites in the membrane. The blots were washed three times in PBS-Tween (0.1% v/v Tween 20) for 10 min and were then incubated with mouse primary antibodies diluted in 3% (w/v) skimmed milk in PBS (anti-CD63, Santa Cruz: 1:200 dilution; anti-TGN48, Santa Cruz: 1:200) for 2h at 37 o C. The blots were washed in PBS-Tween (0.1%) followed by appropriate secondary antibody incubation 7 at 37 o C for 1 h 15 min. After washing with PBS-Tween, the blots were developed using Chemiluminescent ECL detection kit. A clearly defined black spot at the site where the antigen was spotted was considered a positive result. A trace reaction or absence of any reaction was considered a negative result.

Terminal Deoxynucleotidyl Transferase dUTP Nick End Labeling (TUNEL) assay
The immunofluorescent TUNEL staining with an in situ cell death detection kit (Fluorescein Version 17, Roche) was done to measure the extent of apoptosis in the lung and brain. Paraffin sections slides were de-paraffinized, rehydrated followed by antigen retrieval by heating it in microwave in citrate buffer pH 6.0. Sections were then washed with PBS and incubated in blocking solution (10% BSA with 0.3% Triton-X 100 in 1x PBS) for 30 min at 25 ̊ C. The slides were washed with 1x PBS for five times and then incubated with TUNEL reaction mixture (1 part of enzyme solution and 9 parts of label solution) for 1h at 37 o C in dark. Slides were washed with PBS, mounted with Vectashield mounting solution with DAPI (Vector Laboratories) and visualized with a fluorescent microscope using an exciting wavelength of 460-490 nm. Eight to ten non-overlapping fields under 400x magnification was examined to count TUNEL positive cells. Percentage TUNEL positive cells were expressed as (total number of green positive cells/total number of DAPI positive cells) x100.

Immunofluorescence and Immunohistochemistry
Immunofluorescence and immunohistochemistry was performed on de-paraffinized 5 µm thick lung section embedded tissue sections. Antigen retrieval was performed for all slides with 10 mM citrate buffer pH 6.0 (Sigma). Primary antibodies incubations were performed overnight at 4 o C.

Enzyme-Linked Immunosorbant Assay (ELISA)
Protein was isolated from the lungs of RA, BPD and MSC-CM/EXO/TSG-6 treated BPD mice.
Total protein content was measured by BCA kit (BioRad). Concentration of mouse IL-6 in the lung homogenates was analyzed using ELISA kit (DuoSet IL-6, R&D Systems) as per manufacturer's instructions 9 Supplementary Information Table S1. Comparison of different parameters in the exosome dose.