Determination of particle size distributions and gas concentrations in the exposure chamber. To measure the particle size distributions in suspension and gas concentrations, we used Dust TrakⅡ aerosol detector (TSI, Shoreview, USA) and smoke Test340 portable gas analyzer (Testo, Lenzkrch, Germany) to evaluate the quality control parameters of the exposure system. The value of PM1, PM2.5, and PM10 were 27.77 ± 8.66 mg/m3, 28.07 ± 8.84 mg/m3, and 28.23 ± 8.86mg/m3 in the BMF exposure room, respectively (Supplementary Table 2). The CO concentration was maintained at a low level of 55.16 ± 13.77 ppm, and NO and SO2 weren’t detected.
BMF induced Lung morphological changes and AMs infiltration. In order to investigate whether air pollution matter exposure causes emphysema in our exposed rat model, we did hematoxylin and eosin (H༆E) staining to examine lung morphometric character. Alveolar enlargement was calculated as the mean linear intercept(MLI), and the bronchial wall thickness was quantified by wall thickness = (total bronchial area-lumen area)/total bronchial area. Our data showed that BMF exposure induced emphysematous changes and airway remodeling (Fig. 1a-d). Long-term BMF exposure damaged the lung parenchyma and airway wall, which led to alveolar enlargement and distal airway remodeling. Histological analysis demonstrated that the airway wall thickness increased (p༜0.01), and the mean linear intercept decreased dramatically (p༜0.01) at 6 months, whereas, there was no change at 1 month compared to controls (p = 0.366 and 0.557). Total BALF cells in BMF exposure groups were increased compared with the control groups after 4 days, 1 month, and 6 months BMF exposure (Fig. 1e, p = 0.013, 0.001, and 0.003, respectively). AMs were labeled with pan macrophage surface marker CD68, and defined as CD68 + subpopulation with the purity displayed as a percentage of parent population gated on FSC-A/SSC-A. The numbers of infiltrated AMs were more than CON groups after 4 days, 1 month, and 6 months BMF exposure (Fig. 1f, p = 0.01, 0.04, and 0.003, respectively), and reached a peak at 6 months of BMF exposure.
BMF exposure induced BALF cytokine expression. In order to investigate how BMF exposure influences pulmonary inflammation, which may affect M1/M2 phenotype, 27 cytokines/chemokines multiplex tests were performed (Fig. 2). We measured G-CSF, Eotaxin, GM-CSF, IL-1α, Leptin, MIP-1α, IL4, IL1β, IL2, IL6, EGF, IL13, IL10, IL12p70, IFNγ, IL5, IL17A, IL18, MCP-1, IP10, GRO/KC/CINC-1, VEGF, Fractalkine, LIX, MIP-2, TNFα, and RANTES protein levels by using Rat Cytokine/Chemokine Magnetic Bead Panel. It showed that IL1α, IL1β, IL12p70, LIX, EGF as well as VEGF increased significantly after 4 days of BMF exposure (p = 0.018, 0.008, 0.043,0.001, 0.001, and 0.007, respectively). IL1β, TNFα, and LIX were higher than the control groups at 1 month (p = 0.038, 0.031, and 0.021, respectively). Compared to the control groups, there was no cytokine change after 6 months of BMF exposure except for VEGF(p = 0.014). The result indicated that high levels of inflammatory cytokines were induced in the early stage of BMF exposure. Interestingly, cytokine analysis also showed that IL4 in BALF increased significantly compared to the control group after 4 days of BMF exposure(p = 0.002), and restored to near controls during the subsequent time(p = 0.124 and 0.118 ).
Phenotypic characterization of AMs polarization induced by BMF exposure. In order to investigate gene expression of AMs when exposed to BMF, we also used quantitative PCR to determine the mRNA expression for a few key genes (Fig. 3). The result showed that iNOS and IL1β significantly ascended at 4 days of BMF exposure (Fig. 3a, p = 0.005 and 0.001), and descended to near normal levels during the subsequent time. TNFα moderately elevated in 1 month BMF exposure (Fig. 3b, p = 0.028), and declined in 6 months BMF exposure (Fig. 3c, p = 0.374), consistent with BALF cytokines expression. Whereas, TLR2 and TLR4 had no change during the whole exposure course, consistent with previous studies30,31. The level of EGF mRNA was upregulated in AMs of rats exposed to 4 days BMF(Fig. 3a, p༜0.01), which was consistent with the level of EGF protein expression in BALF.
To further investigate the effect of BMF exposure on the dynamic phenotype change of AMs in rats, in addition to the gene expression, we assessed the CD206 (M2 marker) and CD86 expression (M1 marker) in AMs (Fig. 3). The result showed that CD206 MFI decreased at 4 days of BMF exposure (Fig. 3d,e, p༜0.01), and increased to near controls following 1 month exposure(p = 0.207), and was significantly higher than the control group following 6 months exposure(p = 0.035). Conversely, CD86 MFI had no change in AMs during the whole exposure of biomass fuel smoke (Fig. 3f,g, p = 0.730, 0.831, and 0.995, respectively). The result indicated that BMF exposure reduced the anti-inflammatory marker expression in AMs at the beginning, and the anti-inflammatory marker expression was increasing with the accumulation of exposure time.
BMF exposure triggered signaling pathways of macrophage polarization and activation. To study which signaling pathways involved in AMs polarization and activation under the BMF exposure, especially those involved in M2 polarization to attenuate the inflammatory response and promote tissue remodeling13,32, such as Stat6, Stat3, PPARγ, and TGFβ1. We used quantitative PCR, western blot, and immunofluorescence to determine the mRNA and protein level of Stat6, Stat3, PPARγ, and TGFβ1 in BMF exposed rats. It showed that Stat6 mRNA expression in AMs increased significantly after 4 days of BMF exposure (Fig.S1a, p < 0.01), and descended to near controls during the subsequent time (Fig.S1b,c, p = 0.149 and 0.661). The level of p-Stat6 increased after 4 days of BMF exposure (Fig. 4a,b, p < 0.01). Stat3 mRNA expression in AMs had no change compared to control groups after 4 days, 1 month, and 6 months of BMF exposure (Fig.S1a,b,c, p = 0.112, 0.209 and 0.832). In contrast, the level of p-Stat3 level elevated after 4 days of BMF exposure (Fig. 4a,b, p = 0.003), and then declined to near control group after 1 month and 6 months exposure(Fig. 4a,b, p = 0.898 and 0.484). PPARγ mRNA expression in AMs elevated at 4 days of BMF exposure (Fig.S1a, p < 0.01), and declined to a normal level during the subsequent course (Fig.S1b,c, p = 0.66 and 0.543). There was no change of PPARγ protein in lung tissue between controls and exposure groups after 1 month and 6 months of BMF exposure (Fig. 4a,b, p = 0.934 and 0.572). But, PPARγ protein significantly increased at 4 days of BMF exposure (Fig. 4a,b, p = 0.005), consistent with PPARγ mRNA expression in AMs.
On the other hand, TGFβ1 and p-Smad3 protein levels in lung tissue didn’t increase until 6 months of BMF exposure (Fig. 5a-c, p = 0.017 and 0.017). Additionally, AMs expressing p-STAT6, PPARγ, TGFβ1, and p-Smad3 were double examined by Immunofluorescence staining (Fig. 4c-d, 5d-e). It showed PPARγ, p-STAT6, TGFβ1, and p-Smad3 protein were induced in CD68 positive cells (AMs) after 4 days and 6 months of BMF exposure, consistent with PPARγ, p-STAT6, TGFβ1, and p-Smad3 protein level in lung tissue.
PPARγ primed BMDMs exposed to PM into alternative macrophages. To study whether PPARγ reversed M1 phenotype induced by biomass ambient particulate matter, and drove the macrophages into M2 phenotype, we cultured bone marrow derived macrophages(BMDMs), and stimulated them with PM extracted in our lab, and used PPARγ agonist and PPARγ KO lentivirus as the intervention in vitro. 30µg/ml PM was selected as an intervention concentration(Fig.S2a-c). The effect of PPARγ on inflammatory factors was determined via quantitative PCR, western blot, and immunofluorescence. 30µg/ml PM induced BMDMs to secrete iNOS, IL1β, TNFα, and TLR2(Fig. 6a, p < 0.01 for all genes), and BMDMs showed a pro-inflammatory phenotype. Transcriptional factor PPARγ inhibited pro-inflammatory genes (Fig. 6a, p < 0.01 for all genes). 30µg/ml PM triggered phosphorylation of IKBα from 6 hours to 12 hours, and triggered p-P65 began to rise after 6 hours (Fig.S3a,b). PPARγ overexpression significantly inhibited phosphorylation of P65 and IKBα (Fig. 6b, p < 0.01 for all comparisons). PPARγ overexpression also repressed upregulation of IL1β protein level induced by PM (Fig. 6b). PPARγ KO lentivirus increased the level of p-P65 and p-IKBα (Fig. 6c, p < 0.01 for all comparisons). Immunofluorescence staining showed that PPARγ KO lentivirus promoted p-P65 nucleus translocation, while PPARγ overexpression inhibited p-P65 nucleus translocation induced by PM(Fig. 6d).
Our data also showed that PPARγ in BMDMs was upregulated by stimulation with IL4, and promoted the expression of M2 markers. IL4 stimulated PPARγ and p-STAT6 expression (Fig. 7a,b, p < 0.01). PPARγ overexpression increased the level of p-STAT6 (Fig. 7a,b, p < 0.01), while PPARγ KO lentivirus attenuated phosphorylation of STAT6 (p < 0.01). In addition, PPARγ overexpression upregulated CD206 expression (Fig. 7c, p = 0.026). PPARγ KO lentivirus downregulated CD206 expression (Fig. 7c, p = 0.041).
TGFβ1 promoted CD206 expression in BMDMs. 30µg/ml PM decreased CD206 expression (Fig. 8, p < 0.01), TGFβ1 attenuated the downregulation of CD206 induced by PM(Fig. 8, p < 0.01), but had no effect on BMDMs without stimulation of PM(p = 0.522).