Forty heathy female C57BL/6J mice (specific pathogen free), aged 8∼12 weeks with an average weight of 20∼22 g, were purchased from Hunan SJA Laboratory Animal Co., Ltd. (Hunan, China) and fed in the Laboratory Animal Center of Guangxi Medical University. All mice were raised in same environment at 22∼26℃ with adequate lighting. The mice were provided with enough food and water, and the padding was replaced at regular intervals to keep clean and dry. The design of this research was approved by the Animal Ethics Committee of the First Affiliated Hospital of Guangxi Medical University.

hUCMSCs were purchased from Saliai Stem Cell Science and Technology Co., Ltd. (Guangzhou, China) and cultured in DMEM/F12 (gibco, USA) consisting of 10% fetal calf serum (FBS) and 1% penicillin-strep-tomycin. The cells were placed into an incubator which was maintained at 37℃ with 5% CO₂ and 95% humidity. Cells at the logarithmic phase were collected for experi-ments.

After three to six passages, well-conditioned hUCMSCs were selected for exosome extraction. Medium of the cells was replaced with DMEM/F12 without serum or penicillin/streptomycin when the cells reached 80% confluence. About 48 h later, 750 ml of culture supernatant was collected and centrifuged at 4℃, 300 g for 10 min. Supernatant obtained was then centrifuged at 2,000 g for 10 min. Supernatant obtained from the former step was centrifuged at 10,000 g for 30 min. At last, the supernatant was centrifuged at 140,000 g for 90 min. An ultra-centrifuge (optima XE-100, Beckman, USA) was used here for exosome isolation. Sediments were obtained after the final centrifugation, from which exosomes were obtained. The sediments were washed in PBS buffer and resuspended. The suspension was centrifuged at 140,000 g for 90 min and the sediments were resuspended in 100 μl of PBS buffer. Morphology of the exosomes was observed under a transmission electron microscope (H-7650, Hitachi, Tokyo, Japan). Concentration and size of the exosomes were measured by a nanoparticle tracking analyzer (Zetaview, Particle Metrux, Bavaria, Germany). The concentration of the exosomes was 1.66×10⁸/ml. The yield of exosomes was about 1.66×10⁷ particles. The protein concentration was 1.63 μg/μl, and the purity of exosomes was 1.02×10⁵ particles/μg of protein. The exosomes were preserved at −80℃ for later use.

The mice were randomized into the following four groups: the control group, DAH group, DAH+exo group and DAH＋Methylprednisolone group. Each group was comprised of ten mice. Mice in the DAH group, DAH＋exo group and DAH＋Methylprednisolone group were intraperitoneally injected with 0.5 ml of sterile filtrated pristine (Sigma-Aldrich, Saint-Quentin Fallavier, France) for DAH model establishment. The pristine was purchased from Sigma-Aldrich (Saint-Quentin-Fallavier, France) and stored at room temperature. Mice in the control group were treated with 0.5 ml of PBS. After DAH model establishment, mice in the DAH group and DAH＋exo group were respectively injected with PBS (100 μl per mouse) and 0.2 mg/ml hUCMSC-exosome suspension (100 μl per mouse) via tail vein every other day for 14 days. Mice in the DAH＋Methylprednisolone group were intraperito-neally injected with 8 mg/kg methylprednisolone every other day; the concentration of methylprednisolone was reduced to 4 mg/kg after one week. Mice in the control group were left untreated. All the mice were euthanatized 14 days after DAH induction. Alveolar macrophages, lung tissue and bronchoalveolar lavage fluid (BALF) were extracted from the mice.

The paraffin block of mouse lung was cut into 5-μm sections and dewaxed in xylene I (15 min) and xylene II (15 min). The dewaxed lung tissue was then washed in absolute alcohol for 2×5 min and hydrated in gradient alcohol (from 95%, 80% to 70%) for 5 min each time. After being washed, the tissue was stained in hematoxylin for 10 min and then washed in running water for 5 min. Then the tissue was immersed in 1% hydrochloric acid-ethanol for 30 s and subjected to wash. After that, the tissue was stained in 1% eosin for 30 s. After being washed, the tissue was dehydrated in gradient alcohol (from 95%, 95%, 100% to 100%) for 1 min each time and permeabilized in xylene for 2×3 min. The tissue was mounted using neutral balsam and observed under a microscope.

ELISA kits were purchased from Biolegend (San Diego, CA, USA) and applied to measure the expressions of IL-10, TGF-β, IL-6, TNF-α and IL-1β in BALF according to the manufacturer’s instruction. Firstly, the kit was rewarmed to room temperature. The control (50 μl) and BALF (50 μl) were added to the plate for vibration at room temperature for 2 h, after which the samples were subjected to the following steps: 4 washes, added with 100 μl of diluted antibody detection solution, 1 h of vibration; 4 washes, added with 100 μl of avdin-HRP solution, 30 min of vibration; 5 washes, added with 100 μl of TMB substrate solution, 15 min of vibration. A microplate reader (Biotek Synergy2) was used to measure the absorbance at 450 nm within 30 min after the reaction was terminated. Quantities of those cytokines were counted according to the standard absorbance graph.

The mouse was anesthetized using 10% pentasorbital sodium. The trachea was exposed and incised in the middle. A type 9 needle was introduced into the trachea in the direction of the heart and fixed by fine-pointed tissue forceps. The lung was injected with 1 ml of PBS via the trachea using a standard 1-ml syringe and gently massaged, after which the BALF was slowly pumped. This procedure was duplicated for five times. The BALF was centrifuged at 800 g for 10 min. The cell pellet was resuspended in DMEM/F12 (gibco, USA) consisting of 10% FBS and 1% penicillin-streptomycin. Then the cells were cultured in an incubator at 37℃ with 5% CO₂ and 95% humidity. Aliquots of the cell supernatant were preserved at −20℃.

Carboxylate-modified red fluorescent latex beads (L3030, 2 μm, Sigma-Aldrich, St. Louis, MO USA) were diluted with 10% FBS-based DMEM to a concentration of 1:100 and then incubated at 37℃ for 30 min. Macrophages were mixed with the beads at a density of 1×10⁵ cells/100 μl and incubated at 37℃ for 4 h. Then the macrophages were collected and washed twice in PBS. Fluorescence intensity of the macrophages was tested by flow cytometry.

The total RNA was extracted from macrophages using TRIZOL reagent (Life Technologies, NY, USA) and the RNA concentration and purity were measured by a microplate reader (Biotek Synergy2). A PCR amplification instrument was used to synthesize cDNA. The real-time qRT-PCR was performed using a fluorescent quantitative PCR analyzer (CFX Connect, Bio-Rad, USA). GAPDH served as the internal control of mRNA. The reaction conditions were as follows: predegeneration (10 min) at 95℃; 40 cycles of degeneration (10 s) at 95℃, annealing (20 s) at 60℃ and extension (34 s) at 72℃. The statistics were analyzed using 2^−ΔΔCt method according to the following formula: ΔΔCt=[Ct(_{target gene})-Ct(_{reference gene})]_{experimental group}-[Ct(_{target gene})-Ct(_{reference gene})]_{control group}. Primers used in the PCR were synthesized by Sangon Biotech (Shanghai, China) and the primer sequences are shown in Table 1.

Concentration of the proteins extracted from the isolated exosomes was measured using a BCA kit (Vazyme Biotech Co., Ltd., Nanjing, China). The proteins were separated by electrophoresis in 10% SDS-polyacrylamide gel and then transferred onto a polyvinylidene fluoride (PVDF) membrane (Millipore, Billerica, MA). The proteins were confined in 5% skim milk at room temperature for 1 h. After that, the membrane was added with rat anti human antibodies of ALIX (ab117600, 1：1,000), CD63 (ab59479, 1：1,000) and TSG101 (ab83, 1：1,000) (Abcam, Cambridge, MA, USA) and incubated at 4℃ overnight. After 3×10 min TBST wash, the proteins were incubated with horse radish peroxidase labeled goat anti rat IgG antibody (1：5,000, ComWin Biotech Co., Ltd., Beijing, China) at room temperature for 1 h and subjected to 3×10 min TBST wash. The protein expressions were analyzed by a chemiluminescence imaging system (Tanon Science & Tech-nology Co., Ltd., Shanghai, China).

Single cell suspension of primary mouse alveolar macrophages (1×10⁶ cells/100 μl) was fixed in 4% paraformaldehyde for 15 min. After that, the macrophage suspension was incubated with rabbit anti CD11b-FITC antibody (M1/70, 10 μl, BD Bioscience, USA), monoclonal F4/80-PerCP-Cyanine5.5 antibody (BM8, 10 μl), monoclonal CD86-PE antibody (B7-2, 10 μl) and polyclonal CD206-APC antibody (PA5-46879, 10 μl) (eBioscience, USA) at 4℃ in the dark for 10 min. The proportions of F4/80^＋CD11b^＋CD86^＋CD206⁻ cells (M1 macrophages) and F4/80^＋CD11b^＋CD86⁻CD206^＋ cells (M2 macrophages) were measured by the FACS Calibur^TM system (BD, USA).

All records were statistically analyzed on SPSS 17.0 and presented in mean±SD (standard deviation). Group differences were analyzed using One-way analysis of variance (ANOVA). The LSD method was applied for homogeneity test of variances and the Games-Howell method was for heterogeneity test of variances. p<0.05 was considered statistically significant.

Particles isolated from hUCMSCs were in round or elliptical shape with an intact capsule (Fig. 1A). These particles had positive expression of exosome makers Alix, CD63 and TSG101 (Fig. 1B). The particle diameter ranged from 40∼120 nm with an average value of 100 nm; the average and maximum diameters were consistent with those of exosomes (Fig. 1C). From the above, these isolated particles were hUCMSC-exosomes.

H&E staining showed that there were slight hemorrhage and neutrophilic vasculitis in the mouse lung after pristine treatment (Fig. 2A) Pro-inflammatory cytokines IL-6 and TNF-α were increased and anti-inflammatory cytokines IL-10 and TGF-β were decreased in the BALF of DAH mice compared to the control group (Fig. 2B∼E, p<0.01). The above results indicate successful establishment of DAH models.

The hemorrhage and small-vessel vasculitis in mouse lung were attenuated 14 days after hUCMSC-exosome injection or methylprednisolone treatment (Fig. 3A). ELISA showed that IL-6 and TNF-α were decreased while IL-10 and TGF-β were induced in the DAH group compared to the control group (p<0.01); the anti-inflammatory effect of methylprednisolone was slightly stronger than that of hUCMSC-exosomes, but the difference was insignificant (Fig. 3B∼E). hUCMSC-exosomes can lighten the pathological symptoms of DAH mice.

Alveolar macrophages were isolated from the mice. Expression of macrophage markers and phenotype of macrophage were detected to investigate the mechanism underlying the effect of hUCMSC-exosomes on DAH. Macrophages in the DAH group highly expressed iNOS, IL-6, TNF-α and IL-1β compared to the control group (Fig. 4A∼D, p<0.01), exhibiting M1 macrophage phenotype. After hUCMSC-exosome or methylprednisolone treatment, iNOS, IL-6, TNF-α and IL-1β were decreased in macrophages (Fig. 4A∼D, p<0.01) while Arg1, IL-10, TGF-β and chi3l3 were increased (Fig. 4E∼H, p<0.01), presenting M2 macrophage phenotype.

Flow cytometry was applied to test the phenotypic transformation of primary mouse alveolar macrophages (Fig. 4I∼K). F4/80^＋CD11b^＋CD86^＋CD206⁻ cells (M1 macrophages) were increased in the DAH group compared to the control group (31.0±2.0% vs. 0.52±0.34%, p<0.01). After hUCMSC-exosome treatment, F4/80^＋CD11b^＋CD86^＋CD206⁻ cells were reduced (2.81±0.67% vs. 31.0±2.0%) while F4/80^＋CD11b^＋CD86⁻CD206^＋ cells (M2 macrophages) were increased (21.0±3.0% vs. 5.67±1.53%, p<0.01), suggesting that M2 polarization was associated with the regulatory effect of hUCMSC-exosomes in DAH.

The macrophages were incubated with fluorescent beads to evaluate their phagocytosis capacity. The phagocytosis was impeded by DAH compared to the control group while enhanced in the DAH＋exo group and DAH＋Methylprednisolone group compared to the DAH group (Fig. 4L). Both hUCMSC-exosomes and methylpredni-solone can improve the phagocytosis of macrophages in DAH. Taken together, hUCMSC-exosomes facilitated the transformation of macrophages from M1 to M2 phenotype and further enhanced the macrophage phagocytosis.