Adipose-derived stem cells therapy effectively attenuates PM2.5-induced lung injury

Background The adverse health effects of fine particulate matter (PM2.5) exposure are associated with marked inflammatory responses. Adipose-derived stem cells (ADSCs) have immunosuppressive effects, and ADSC transplantation could attenuate pulmonary fibrosis in different animal disease models. However, whether ADSCs affect PM2.5-induced lung injury has not been investigated. Method C57BL/6 mice were exposed to PM2.5 every other day via intratracheal instillation for 4 weeks. After that, the mice received tail vein injections of ADSCs every 2 weeks. Results ADSC transplantation significantly attenuated systemic and pulmonary inflammation, cardiac dysfunction, fibrosis, and cell death in PM2.5-exposed mice. RNA-sequencing results and bioinformatic analysis suggested that the downregulated differentially expressed genes (DEGs) were mainly enriched in inflammatory and immune pathways. Moreover, ADSC transplantation attenuated PM2.5-induced cell apoptosis and pyroptosis in the lungs and hearts. Conclusion ADSCs protect against PM2.5-induced adverse health effects through attenuating pulmonary inflammation and cell death. Our findings suggest that ADSC transplantation may be a potential therapeutic approach for severe air pollution-associated diseases. Supplementary Information The online version contains supplementary material available at 10.1186/s13287-021-02441-3.


Introduction
Increased airborne fine particulate matter (PM 2.5 , aerodynamic diameter ≤ 2.5 μm) concentrations (higher than the World Health Organization air quality guideline level -10 μg/m 3 ) have been implicated in the development of respiratory and cardiovascular diseases, including chronic obstructive pulmonary disease (COPD) [1], asthma [2], bronchitis [3], coronary artery disease [4], and atherosclerosis and congestive heart failure [5,6]. Due to its small size, PM 2.5 is easily inhaled into the airway and largely deposited in lung alveoli. Although the mechanism underlying the cytotoxicity of PM 2.5 remains elusive, dysregulation of the inflammatory response is considered a major trigger for cell apoptosis and lung injury [7,8]. Thus, strategies with anti-inflammatory effects might be effective in attenuating PM 2.5 -induced disease [9].
It is well documented that adipose-derived mesenchymal stem cells (ADSCs) have multilineage differentiation potential and can differentiate into different cell types, including adipocytes, osteoblasts, chondrocytes, and hepatocytes [10]. ADSCs can also repair damaged organs and tissues through releasing a variety of paracrine factors and extracellular vesicles (EVs) [11]. These properties render ADSCs attractive therapeutic agents for tissue defects and degenerative diseases. Interestingly, ADSCs have the ability to differentiate into type 2 alveolar epithelial cells [12], and ADSC therapy ameliorates lung injury and fibrosis in a series of pulmonary disorders, including bronchiolitis obliterans [13], idiopathic pulmonary fibrosis [14], and emphysema [12]. Moreover, EVs derived from ADSCs could also alleviate PM 2.5 -induced lung injury in mice [15]. However, whether ADSCs exert a direct protective effect on PM 2.5 -induced lung injury remains unclear.

ADSCs isolation and characterization
White adipose tissues were harvested from male C57BL/ 6J mice at the age of 6-8 weeks. After washing two times with ice-cold phosphate buffered saline (PBS), the adipose tissue was mechanically diced, digested, washed, filtered, and centrifuged. The cell pellet was washed with DMEM/F12 one more time and then resuspended in DMEM/F12 supplemented with 10% FBS, and 1% penicillin and streptomycin. Finally, ADSCs were placed in a 10-cm dish and cultured in an incubator at 37°C with 5% CO 2 .

Experimental animals
Twenty-four 8-week-old male C57BL/6J mice were randomly divided into the control group (8 mice) and PM 2.5 -exposed group (16 mice). The mice were first treated with 10 μl PBS (control group) or 10 mg/kg PM 2.5 in 10 μl PBS (PM 2.5 -exposed group) every other day via intratracheal instillation for 4 weeks. Based on a previous report [17], the dose used here was equal to daily exposure to~1500 μg/m 3 PM 2.5 for 1 month. After PM 2.5 exposure, mice were randomly assigned to 2 groups, 1 group (8 mice) received ADSCs (2 × 10 5 cells/ mouse) via tail vein injection at the fourth and sixth weeks and the other treated with PBS. At the end of the eighth week, the mice were sacrificed by CO 2 inhalation followed by cervical dislocation. To avoid the errors arising from bias, the group information was blind to the researchers who performed biochemical assays and histological analysis. During the whole experimental period, the mice were housed in cages with corn cob bedding, and were fed commercial mouse chow and distilled water ad libitum. The mice were housed under controlled temperature (22 ± 2°C) and relative humidity (40-60%) conditions with a 12-h light/dark cycle. Mice with any abnormalities in skin, hair, bodyweight, and behavior were dropped out from the group.

Histopathological staining
After perfusion with PBS, the mouse lungs were harvested, washed, fixed with formalin, and embedded in paraffin. Lung sections (5 μm) were stained with hematoxylin and eosin (H&E) stain, a Masson's trichrome stain kit (#G1340, Solarbio Science & Technology Co. LTD, Beijing, China), dihydroethidium (DHE), or a TUNEL staining kit. The neutrophil/Ly-6B is a heavily glycosylated protein expressed on neutrophils [18], while galectin-3 (Gal-3, also called Mac-2 antigen) is a carbohydrate-binding protein expressed on the surface of inflammatory macrophages [19]. To assess neutrophil and macrophage infiltration, lung sections were also stained with Gal-3 and neutrophil antibodies, respectively. Five mice per group were used for these experiments.

ELISA measurement
Serum TNFα, IL-1β, and IL-6 levels and pulmonary 3′-NT and 4-HNE levels were determined using commercial ELISA kits according to the manufacturer's instructions. In brief, samples or standard plus antibody cocktail were added to the appropriate wells of the precoated microplate. The plate was then incubated overnight at 4°C with gentle shaking. After washing and adding the chromogen substrates for 10 min, the reactions were terminated by adding the stop solution. The content was calculated according to the standard curve and OD value.

Echocardiographic measurement
After anesthetization with 1.5% isoflurane, the mice were subjected to echocardiographic measurement using a VisualSonics high-resolution Veve 2100 system (Visual Sonics, Toronto, ON, Canada).

RNA isolation and RNA-sequencing
Total RNA was extracted from the lungs (3 mice per group) using TRIzol reagent and RNA quality was measured with an Agilent 2100 bioanalyzer (Thermo Fisher Scientific, MA, USA). The isolated RNA was further purified by DNase I treatment and rRNA removal. Library construction and RNA-sequencing were then performed on a BGISEQ500 platform (BGI-Shenzhen, China).

Read mapping and differentially expressed gene analysis
As described previously [20], the raw data were cleaned with SOAPnuke and trimmomatic software [21], and then the clean reads were mapped to the reference genome (Mus_musculus, GCF_000001635.25_GRCm38.p5) using HISAT [22] or Bowtie 2 [23] software. RESM software [24] was used to obtain the gene expression level and the differential expression of genes (DEGs) between two groups was assessed by DEGseq [25] with the following thresholds: fold change ≥ 2 and adjusted P value ≤ 0.001. The identified DEGs for each pair were mapped to terms in the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://www.genome.jp/kegg/ pathway.html). The p value was adjusted for the false discovery rate (FDR) to obtain the q-value, and a q-value ≤ 0.05 was considered to indicate significant enrichment.
Gene set enrichment analysis (GSEA) was also performed with the KOBAS 3.0 online tool 1 [26] and Java GSEA2 [27]. An FDR q-value ≤ 0.25 was considered to indicate significant enrichment.

Quantitative real-time PCR and western blotting
A PrimeScript RT reagent kit (#RR036B, TaKaRa, Otsu, Japan) was used for cDNA synthesis. A quantitative realtime polymerase chain reaction (qPCR) assay was performed using the SYBR Premix Ex Taq™ II Kit (#RR820DS, TaKaRa) and the results were normalized to the level of 18 S ribosomal RNA. The primers used in the qPCR assay are listed in Table S1.
Proteins were extracted from the lungs using buffer (50 mM Tris-Cl, 150 mM NaCl, 100 μg/ml phenylmethylsulfonyl fluoride, protease and phosphatase inhibitor cocktail (#046931124001 and #4906837001, Roche, Basel, Switzerland), and 1% Triton X-100) on ice for 30 min. After centrifugation at 12000×g at 4°C for 20 min, the supernatant was used for western blot analysis. The detail information for antibodies are listed in Table S2.

Data and statistical analysis
All values are expressed as the mean ± standard deviation (SD). Normal distribution was assessed by the Kolmogorov-Smirnov normality test. The significance of differences was tested using one-way analysis of variance (ANOVA) followed by Fisher's least significant difference test or the Kruskal-Wallis nonparametric test followed by Dunn's test using GraphPad Prism 7 (GraphPad Software Inc., CA, USA). Statistical significance was defined as p < 0.05.

Characterization of ADSCs and in vitro differentiation assay
As shown in Figure S1, the isolated ADSCs were positive for CD44 (93.5%), CD73 (97%), and CD90 (95.8%) and negative for CD34 (3.49%). Microscopic observation showed that the cultured ADSCs exhibited a fibroblastlike appearance and formed a monolayer ( Figure S2A). To determine the differentiation potential, ADSCs were cultured in adipogenic or osteogenic differentiation induction medium for 14 days. Positive staining with Oil red O and alizarin red indicated that ADSCs could differentiate into adipocytes and osteocytes, respectively ( Figure S2B-C).

ADSC transplantation attenuates PM 2.5 -induced inflammation and pulmonary fibrosis in mice
To investigate the protective effects of ADSCs on PM 2.5induced lung injury and cardiac dysfunction, we treated the PM 2.5 -exposed mice with ADSCs twice via tail vein injection. The processes of the ADSC transplantation experiment are illustrated in Fig. 1A. At the end of the experimental period, there were significant increases in serum TNFα, IL-1β, and IL-6 levels in PM 2.5 -exposed mice, while these increases were diminished in the mice that received ADSC transplantation (Fig. 1B-D). Next, we performed qPCR to examine the effects of ADSCs treatment on the mRNA levels of genes related to inflammation and fibrosis. PM 2.5 exposure caused significant increases in pulmonary collagen I and III, transforming growth factor beta (TGF-β), IL-1β, and TNFα mRNA levels; however, these increases were significantly attenuated by ADSC transplantation (Fig. 1E). Histopathological analysis of lung sections using H&E and Masson's staining demonstrated that the collapse of alveoli, thickening of airway epithelium, and deposition of collagen in the lungs of PM 2.5 -exposed mice were attenuated by ADSC treatment (Fig. 1F, G). Immunohistochemical staining with neutrophil and Gal-3 antibodies also showed that PM 2.5 -induced infiltration of neutrophils and macrophages in the lungs was diminished by ADSC transplantation (Fig. 1F, H, I).
ADSC transplantation prevents PM 2.5 -induced pulmonary oxidative stress and cell apoptosis in mice PM 2.5 exposure significantly increased pulmonary 3′-NT and 4-HNE levels, while these increases in the levels of oxidative stress markers [28] were attenuated by ADSC transplantation (Fig. 2A, B). Moreover, DHE staining revealed that ADSC decreased superoxide levels in PM 2.5exposed lungs (Fig. 2C, D), indicating that ADSC treatment ameliorated PM 2.5 -induced pulmonary oxidative stress. TUNEL staining showed that ADSC transplantation significantly attenuated the increases in apoptotic cell number in the lungs of PM 2.5 -exposed mice (Fig.  2C, E). Consistently, the expression of the proapoptotic protein Bax and cleaved caspase-3 was increased, whereas the expression of antiapoptotic protein Bcl-2 was decreased in the lungs of PM 2.5 -exposed mice. Moreover, PM 2.5 exposure also decreased superoxide dismutase 1 (SOD1) and peroxiredoxin 4 (PRDX4) protein expression in the lungs. However, the changes in apoptosis and antioxidant related proteins were significantly attenuated by ADSC transplantation (Fig. 2F).
ADSC transplantation has profound effects on the gene expression profile in PM 2.5 -exposed lungs To investigate the molecular mechanism by which ADSC transplantation attenuated PM 2.5 -induced lung injury in mice, RNA-sequencing was performed to analysis the whole-genome expression profiling changes in these groups. We identified 1217 differential DEGs (866 upregulated and 351 downregulated) in the control group vs PM 2.5 group and 585 DEGs (423 upregulated and 162 downregulated) in the PM 2.5 vs PM 2.5 -ADSC group, and the fold changes of these DEGs were visualized in volcano plots (Fig. 3A). KEGG pathway enrichment analysis demonstrated that the upregulated DEGs in the control group vs PM 2.5 group were mainly significantly enriched in inflammatory and metabolic pathways, including the PPAR signaling pathway, cytokinecytokine receptor interaction, chemokine signaling pathway, and adipocytokine signaling pathway (Fig. 3B). Notably, the downregulated DEGs in PM 2.5 group vs PM 2.5 -ADSC group were significantly enriched in inflammation and immune pathways, including the cytokine-cytokine receptor interaction, IL-17, chemokine signaling, TNF signaling, and NOD-like receptor signaling pathways (Fig. 3B). To explore the immunosuppressive mechanism of ADSCs in PM 2.5 -exposed lungs, the expression of the DEGs was visualized in a heatmap (Fig. 3C), which demonstrated that the expression of most of the inflammation-related genes that were upregulated in PM 2.5 -exposed lungs was repressed by ADSC transplantation. The qPCR results demonstrated that ADSC treatment attenuated the upregulation of Ccl3, Ccl6, Csf3r, Gdf15, Mefv, Orm2, S100a9, Saa3, and Tnfrsf19 expression in PM 2.5 -exposed mice (Fig. 3D).

ADSC transplantation attenuates pyroptosis in PM 2.5exposed lungs
Recent studies have shown that PM 2.5 -induced pulmonary inflammation is associated with activation of NODlike receptor protein 3 (NLRP3) and caspase-1 pathway, which can trigger cell pyroptosis [29]. GSEA results showed that the DEGs in the control group vs PM 2.5 group were significantly enriched in inflammation and immune pathways, including the immune system process, adaptive/innate immune response, inflammatory response, chemotaxis, and immune response ( Figure S3). Consistently, the DEGs in the control group vs PM 2.5 group were also significantly enriched in the pyroptosis pathway (Fig. 4A). To further explore the mechanism (See figure on previous page.) Fig. 1 Adipose-derived stem cell transplantation alleviated PM 2.5 -induced inflammation and pulmonary fibrosis. C57BL/6 mice were administered PBS or 10 mg/kg PM 2.5 every other day via intratracheal instillation for 4 weeks. At the fourth and sixth weeks, some of the PM 2.5 -exposed mice were treated with adipose-derived stem cells (ADSCs) via tail vein injection (2 × 10 5 cells/mouse). A Schema of the experimental procedure. B-E After mice were sacrificed, serum TNFα (B), IL-1β (C), and IL-6 (D) levels and the mRNA levels of pulmonary inflammatory and fibrotic genes (E) were measured. F Representative lung sections were stained with hematoxylin and eosin (H&E) stain, Masson's trichrome stain, and antibodies specific for neutrophils and macrophages (galectin-3, Gal-3) (brown staining). Scale bar = 100 μm. G-I The relative collagenous fiber area (G) and the numbers of neutrophils (H) and Gal-3-positive cells (I) were quantified. N = 5; the data are presented as the mean ± SD; asterisk indicates p < 0.05; double asterisks indicate p < 0.01 underlying the activation of the pyroptosis pathway in PM 2.5 -exposed lungs, the expression profiles of genes involved in the pyroptosis pathway were visualized in the heat map (Fig. 4B), which demonstrated that the expression of most of the genes was upregulated in PM 2.5 -exposed lungs, while the upregulation of the expression of these genes was significantly attenuated by ADSC transplantation. The changes in gene expression were further validated by real-time qPCR, which confirmed that PM 2.5 -induced upregulation of pulmonary Aim2, Casp4, Gsdme, IL18rap, IL1a, Naip2, and Nlrp3 gene expression was attenuated by ADSC transplantation (Fig. 4C).
Western blotting also showed that PM 2.5 exposure resulted in significant increases in the expression of NLRP3, IL-1β, cleaved caspase-1, and gasdermin E (GSDME), while the increase in the expression of these genes was significantly attenuated by ADSC treatment (Fig. 4D).

ADSCs attenuate PM 2.5 -induced cardiac dysfunction and fibrosis
To determine whether ADSCs could alleviate cardiac dysfunction in PM 2.5 -exposed mice, we assessed the left ventricular (LV) function. Echocardiographic examination showed that PM 2.5 exposure significantly decreased ejection fraction (EF) values, while this reduction was attenuated by ADSC treatment (Figure S4A-B). Furthermore, ADSCs also significantly attenuated PM 2.5induced upregulation of myocardial atrial natriuretic peptide (ANP; a marker for cardiac stress) ( Figure S4C). As revealed by Masson's staining and TUNEL staining, obvious fibrosis and cardiomyocyte apoptosis occurred C-E Lung sections from the control and PM 2.5 -exposed mice were stained with dihydroethidium (DHE) or DAPI (blue) and dye (red) from a TUNEL assay kit (scale bar = 50 μm, the arrows point to TUNEL-positive cells). The relative DHE fluorescence intensity (D) and TUNEL-positive cells (E) were quantified. F Lysates of lung tissue were subjected to western blotting to determine the expression levels of Bax, Bcl-2, cleaved caspase-3, superoxide dismutase 1 (SOD1), and peroxiredoxin 4 (PRDX4). β-Tubulin was used as a loading control. In A-E, N = 5; in F, N = 3; the data are presented as the mean ± SD; asterisk indicates p < .05; double asterisks indicate p < 0.01 in the hearts of the PM 2.5 -exposed mice, while the collagenous fibrotic area and apoptotic cell number in these hearts were significantly reduced by ADSC transplantation ( Figure S4D-F). Western blotting showed that ADSC transplantation attenuated the PM 2.5 -induced decrease in Serca2a expression and increases in IL-1β, cleaved caspase-1, and GSDME expression in the heart ( Figure S4G). Together, these results indicated that ADSCs effectively attenuates PM 2.5 -induced cardiac dysfunction in mice.

Discussion
To the best of our knowledge, the present study provides the first direct evidence that ADSC transplantation is effective in attenuating PM 2.5 -induced lung injury and cardiac dysfunction. The mechanism underlying the protective effect of ADSCs was associated with the suppression of the inflammatory response and attenuation of cell death. The finding that ADSC therapy decreased serum TNFα, IL-6, and IL-1β levels and pulmonary TNFα and IL-1β mRNA levels in PM 2.5 -exposed mice suggested that ADSCs protect against PM 2.5 -induced lung injury, at least partially, by suppressing the inflammatory response. The anti-inflammatory effects of ADSCs have also been observed in different pulmonary disease models, including models of bleomycin-induced interstitial pneumonia [30], lipopolysaccharide-induced pulmonary microvascular barrier damage [31], and ventilator-induced lung injury in rats [32]. Based on previous reports [33,34], the molecular mechanism underlying the anti-inflammatory actions of ADSCs includes but is not limited to, secretion of immunosuppressive cytokines/EVs [35,36], inhibition of microphage activation/infiltration [30,37], a decrease in inflammatory cytokine production, and regulation of T-reg cells [38,39]. Consistently, we demonstrated here that ADSCs had inhibitory effects on inflammatory cell infiltration and inflammatory pathway activation. PM 2.5 exposure can induce apoptosis in different cell lines and apoptosis induction mediated by PM 2.5 contributes to its adverse health effects [16,40]. Here, we found that the lung injury and cardiac dysfunction in PM 2.5 -exposed mice were associated with increases in the number of TUNEL-positive cells and the expression of cleaved caspase-3. It is notably that caspase 3 can also be enrolled in non-apoptotic processes [41]. The proapoptotic role of PM 2.5 in the lungs was further confirmed by the decrease in Bcl-2/Bax ratio, which is an important inducer for mitochondrial cytochrome c release and caspase activation [42].
There is evidence that PM 2.5 triggers pyroptosis by activating the NLRP3/caspase-1 signaling pathway in cells and lung tissues [29,43]. In agreement with previous studies, we also observed that the mRNA expression of Nlrp3, Aim2, Naip2, and I1a, as well as protein expression of NLRP3, IL-1β, and cleaved caspase-1, were upregulated in PM 2.5 -exposed lungs. Pyroptosis can also be regulated by a caspase-1 independent pathway, in which caspase-4, caspase-5, and caspase-11 are apical activators [44]. Moreover, GSDME can induce a switch from caspase-3-mediated apoptosis induced by chemotherapy drugs to pyroptosis when it is highly expressed [45]. In the present study, we found that PM 2.5 exposure increased the mRNA levels of Casp4 and Gsdme and the protein expression of cleaved caspase-3 and GSDME in the lungs, indicating that PM 2.5 induced pyroptosis via multiple mechanisms.
The finding that ADSC transplantation attenuated the increases in the number of TUNEL-positive cells, mRNA levels of pyroptosis-related genes and apoptosis-and pyroptosis-related proteins in the lungs and heart suggested that the protective effects of ADSCs against PM 2.5 -induced lung injury and cardiac dysfunction were associated with inhibition of cell apoptosis and pyroptosis. The antiapoptotic effect of ADSCs has also been observed in silicosis-exposed mice [46], heterotopic tracheal transplantation models [47], and radiation treated rats [48]. However, the antipyroptotic effect of ADSCs has not yet been reported. Considering that pyroptosis plays an important role in the pathogenesis of many respiratory and cardiovascular diseases, it is possible that ADSCs may have more potential clinical applications than expected.
It has been demonstrated that inflammatory chemokines provide axes for effective recruitment of therapeutic adult stem cells to home and engraft in the appropriate injured target tissues [49]. In the present study, we observed that the expression of chemokines, including Ccl3/6, Cxcl1/2/3, and Cxcr1/2, was upregulated in PM 2.5 -exposed lungs, which may result in a suitable inflammatory microenvironment for the recruitment of ADSCs. As previous studies have proven that ADSCs have the capacity to home and engraft in injured lungs [12,13,47], we did not label the ADSCs or track their homing and engraftment during the experimental process. Anyway, this is the limitation of our study. Another limitation of our study is we did not analysis the inflammatory cell infiltration with fluorescence activated cell sorting, which can reveal more information on inflammatory processes.

Conclusion
In summary, our study indicates that ADSCs exhibit anti-inflammatory, anti-fibrotic, and anti-cell death effects in PM 2.5 -exposed lungs and that the underlying mechanism is associated with their immunosuppressive effects. Our results suggest that ADSC transplantation may be a potential therapeutic approach for air pollution-associated diseases.