Effect of supercritical carbon dioxide fluid extract from Chrysanthemum indicum Linné on bleomycin-induced pulmonary fibrosis

Background As a prevalent type of cryptogenic fibrotic disease with high mortality, idiopathic pulmonary fibrosis (IPF) still lacks effective therapeutic drugs. The compounds extracted from buds and flowers of Chrysanthemum indicum Linné with supercritical-carbon dioxide fluid (CISCFE) has been confirmed to have antioxidant, anti-inflammatory, and lung-protective effects. This paper aimed to clarify whether CISCFE could treat IPF induced by bleomycin (BLM) and elucidate the related mechanisms. Methods Rats (Sprague-Dawley, male) were separated into the following groups: normal, model, pirfenidone (50 mg/kg), CISCFE-L, −M, and -H (240, 360, and 480 mg/kg/d, i.g., respectively, for 4 weeks). Rats were given BLM (5 mg/kg) via intratracheal installation to establish the IPF model. A549 and MRC-5 cells were stimulated by Wnt-1 to establish a cell model and then treated with CISCFE. Haematoxylin-eosin (H&E) and Masson staining were employed to observe lesions in the lung tissues. Quantitative real-time polymerase chain reaction (qRT-PCR) and Western blot (WB) were performed to observe changes in genes and proteins connected with the Wnt/β-catenin pathway. Results CISCFE inhibited the proliferation of MRC-5 cells (IC50: 2.723 ± 0.488 μg/mL) and A549 cells (IC50: 2.235 ± 0.229 μg/mL). In rats, A549 cells, and MRC-5 cells, BLM and Wnt-1 obviously induced the protein expression of α-smooth muscle actin (α-SMA), vimentin, type I collagen (collagen-I), and Nu-β-catenin. The mRNA levels of matrix metalloproteinase-3 (MMP-3) and − 9 (MMP-9), two enzymes that degrade and reshape the extracellular matrix (ECM) were also increased while those of tissue inhibitor of metalloproteinase 1 (TIMP-1) were decreased. However, CISCFE reversed the effects of BLM and Wnt-1 on the expression pattern of these proteins and genes. Conclusion These findings showed that CISCFE could inhibit IPF development by activating the Wnt/β-catenin pathway and may serve as a treatment for IPF after further investigation. Supplementary Information The online version contains supplementary material available at 10.1186/s12906-021-03409-9.

antitumour [16], antioxidant [17], antimicrobial [18], anti-inflammatory, and immunomodulatory effects [19,20]. The supercritical carbon dioxide extraction is widely used to extract chemical constituents from Chrysanthemum indicum Linné, not only because the supercritical carbon dioxide extraction is efficient and environmentally friendly, but also because it can obtain more biologically active compounds and ensure the structural integrity of volatile compounds. Specifically, the extract from buds and flowers of Chrysanthemum indicum Linné in supercritical carbon dioxide fluid (CI SCFE ) has been widely used in cosmetics, functional foods, and toiletries [21]. Moreover, our previous study showed that it could enhance the antitumour effect and reduce pulmonary damage of bleomycin (BLM) [22]. In addition, CI SCFE has been demonstrated to protect against UV-induced skin injury and lipopolysaccharide (LPS)-induced lung injury [23,24] We previously observed that CI SCFE can significantly inhibit inflammatory cytokines induced by LPS and produced by alveolar epithelial cells, thereby alleviating LPS-mediated lung injury [24]. When CI SCFE and BLM were administered together to treat tumours, we found that BLM combined with CI SCFE could remarkably attenuate collagen deposition and inflammatory damage in lung tissues. Based on the above findings, we hypothesized that CI SCFE has a lung protective effect. However, whether CI SCFE could be used to treat IPF remains unknown.
In this study, to determine whether CI SCFE could relieve or treat pulmonary fibrosis, we used a BLM-induced IPF model and Wnt-1-induced cell model to verify this hypothesis.

Drugs
We dissolved BLM (Zhejiang Hai Zheng Pharmaceuticals, China. purity> 99%) in 0.9% normal saline. CI SCFE  (5,10,15,20,30,40,60, and 100 ng/mL) on A549 and MRC-5 cells were also studied using MTS. Cells were seeded into 96-well plates determine cell viability. Serum-free medium was added when the cells adhered. Then, the cells were stimulated with 20 ng/mL Wnt-1 for 24 h after serum starvation for 6 h. Then, the cells were treated with medium containing 400 ng/mL CI SCFE . Cells were assessed after 24 h of treatment.

Experimental animal procedures and pulmonary fibrosis model establishment
The Experimental Animal Centre of Guangzhou University of Chinese Medicine (Certificate number 44005800005378, Guangzhou, China) provided male adult Sprague-Dawley (SD) rats. The procedures were conducted under the guidance of the Animal Care and Welfare Committee of Guangzhou University of Chinese Medicine. Rats were placed in an SPF environment (temperature 24-26 °C, humidity 70-75%). Rats were provided free access to food and water. The animal experiments were conducted according to the guidelines established by the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals.
Sixty rats were randomly assigned to the following six groups (n = 10; group by random number table method): normal, model, PFD, CI SCFE -L, −M, and -H. Except for rats in the control group, all the rats were given BLM (5 mg/kg) via intratracheal installation to establish a pulmonary fibrosis model. Three days after BLM injection, intragastric administration was started. Normal saline was administered to rats in the normal and model groups. PFD (50 mg/kg) was given to rats in the PFD group, and rats in the CI SCFE -L, −M, and -H groups were treated with CI SCFE (240, 360, and 480 mg/ kg, respectively). All drugs were intragastrically administered once per day. During the experiment, none of the rats died.
After 28 days, all rats were weighed and administered 1% pentobarbital sodium (40 mg/kg) via intraperitoneal injection. Then, the rats were sacrificed by bleeding from the abdominal aorta, and the lung tissues were rapidly collected and washed with ice-cold normal saline. The lung tissues were weighed, and the left lung tissues (0.1 g) of all rats were fixed with 4% paraformaldehyde. The remaining lung tissue was stored at − 80 °C.

Relative lung weight and body weight changes
The relative lung weight was calculated as follows: lung coefficient = lung weight (g)/body weight (kg) × 100%.

Histological analysis
After fixation for 24 h, the fixed lung tissue was dehydrated with different concentrations of alcohol, and different xylenes were used for transparency. Then, lung tissues were embedded in paraffin. After the paraffin was cooled, 5 μm sections were cut to observe inflammatory infiltration and collagen deposition with haematoxylineosin (H&E) and Masson's trichrome staining, respectively. Then, the stained sections were observed under a microscope. The grades of pathologic changes that indicated lung injury were evaluated referencing others' reports [25,26]. Lung injury includes oedema, congestion, inflammatory cell infiltration, and interstitial inflammation, each of which was graded from 0 to 4. The scores of each category per individuals were added to the final score of lung injury.

HYP analysis
Lung tissues (0.1 g) were homogenized in HYP Assay Kit hydrolysate using a tissue homogenizer. After halting the hydrolysate reaction according to the manufacturer's instructions, the absorbance of the supernatant was measured at 550 nm on a spectrophotometer.

Cell transfection and treatment
MRC-5 (1 × 10 5 cells/mL) and A549 (1.25 × 10 5 cells/mL) cells were seeded into 6-well plates. Then, the cells were stimulated with Wnt-1 (20 ng/mL) when they reached 70-80% confluence. After incubation for 24 h, cells were treated with medium containing CI SCFE (400 ng/mL). RNA and protein were extracted after 24 h. The small interfering RNA (siRNA) targeting β-catenin and the negative control were provided by RiboBio (Guangzhou, China). Lipofectamine 2000 (Invitrogen) was adopted to transfect β-catenin siRNA (40 nM) or negative control siRNA into the cells according to the manufacturer's instructions. Then, the cells were treated as described above.

Quantitative real-time polymerase chain reaction (qRT-PCR)
We extracted total RNA from two kinds of cells and lung tissues with TRIzol reagent (Invitrogen) according to the manufacturer's instructions. Ultra-micro spectrophotometer (Thermo Fisher Scientific, USA) was used to measure the purity and concentration of extracted RNA. When the OD260/OD280 value was in the range of 1.8 ~ 2.1, the RNA purity is considered to meet the experimental requirements. A reverse transcription kit (Vazyme Biotech Co., Ltd., China) was applied to reverse transcribe total RNA. The cDNA of two cells and lung tissues were stored at − 20 °C. Then, qRT-PCR amplification was accomplished using a ChamQ ™ Universal SYBR qPCR Master Mix Kit (Vazyme Biotech Co., Ltd., China) under the following cycling conditions: 95 °C for 30 s followed by 40 cycles of 95 °C for 10 s, 60 °C for 30 s, and 72 °C for 60 s. The respective primers are listed in Table 1. GAPDH served as the reference gene. This study employed the 2 -∆∆Ct method to calculate gene expression based on the following formula: Fold change = 2 -∆∆Ct , ∆∆Ct = (Ct Sample -Ct GAPDH ) -(Ct Control -Ct GAPDH ).

Western blotting
Total proteins were extracted from two kinds of cells and lung tissues with radio-immunoprecipitation assay (RIPA) buffer, and the proteins from the nuclear and cytoplasmic fractions of two cells and lung tissues were obtained with a Nuclear and Cytoplasmic Protein Extraction Kit (Keygen Biotech, Jiangsu, China). The protein concentrations of all the samples were estimated using the Bicinchoninic Acid Protein Kit (Best Bio, Shanghai, China). Fifty micrograms of two cells and lung tissues proteins were resolved using SDS-PAGE on 12% or 8% gels and were transferred onto PVDF membranes (Millipore, Billerica, USA). The membranes were blocked with 5% non-fat milk for 1 h. The membranes were incubated with anti-GAPDH (1:1000), anti-H3 (1:1500), antiα-SMA (1:500), anti-vimentin (1:500), anti-β-catenin (1:500) and anti-collagen-I (1:500) antibodies overnight at 4 °C. Then, the membranes were treated with secondary antibodies (1:2000) for 2 h. Protein expression of two cells and lung tissues were measured by a chemiluminescence system (Tanon). The band densities of GAPDH and H3 were used as a reference.
Statistical analysis SPSS 23.0 software was used to analyse the results, and the data are expressed as the means ± SD. Duncan's test and one-way ANOVA were used to analyse the statistical significance among different groups. P < 0.05 was considered statistically significant.    Figure 1E, F shows that compared with the Wnt-1 alone group, the CI SCFE (80, 100, 200, 400 ng/mL) groups showed significantly reduced cell viability (P < 0.05). The data illustrated that CI SCFE can significantly suppress Wnt-1-mediated proliferation of A549 cells and MRC-5 cells. Based on these results, Wnt-1 (20 ng/mL) and CI SCFE (400 ng/mL) were utilized for subsequent experiments.

Effect of CI SCFE on relative lung weight and level of HYP in BLM-treated rats
We analysed the effect of CI SCFE on the experimental rat lung coefficient. BLM upregulated the lung coefficient (P < 0.05) ( Fig. 2A). However, after CI SCFE treatment, the lung coefficient value of rats significantly decreased.
HYP is the principal constituent of collagen. The degree of IPF can be reflected by the content of HYP. Figure 2B shows that CI SCFE -M and -H treatment can reduce the HYP level.

Effect of CI SCFE on IPF in BLM-treated rats
The model group showed significant inflammation, collagen deposition and severely damaged lung tissue structure (Fig. 3A, B). The CI SCFE -M and -H groups showed that the above situation was relieved after treatment compared with the model group. Figure 3C shows that BLM can obviously induce lung injury in rats. However, CI SCFE -M and CI SCFE -H alleviated the lung injury caused by BLM in rats (P < 0.05). These results indicated that CI SCFE can alleviate IPF caused by BLM to a certain extent.

Effect of CI SCFE on the expressions of α-SMA, Vimentin, collagen-I, nu-β-catenin in A549 cells, MRC-5 cells and BLM-treated rats
In this study, BLM obviously increased α-SMA, vimentin, collagen-I and Nu-β-catenin protein expression in A549 cells (Fig. 4A)  The expression of these proteins was downregulated in A549 and MRC-5 cells treated with CI SCFE . In addition, Fig. 4C shows that the expression of the above proteins in BLM-treated rat lung tissues was decreased in the CI SCFE -M and -H groups (P < 0.05). Thus, CI SCFE relieved the abnormal deposition of ECM proteins in BLM-treated rats.

Effect of CI SCFE on MMP-3, MMP-9 and TIMP-1 mRNA levels in A549 cells, MRC-5 cells and BLM-treated rats
A549 cells and MRC-5 cells treated with CI SCFE showed downregulation of MMP-3 (Fig. 5A), MMP-9 (Fig. 5B), and TIMP-1 (Fig. 5C) mRNA levels (P < 0.05), when compared with Wnt-1 group. After treatment with CI SCFE -M or CI SCFE -H, MMP-3 (Fig. 5D) and MMP-9 (Fig. 5E), and TIMP-1 (Fig. 5F) mRNA levels were decreased (P < 0.05) in BLM-treated rats, compared with model group. The data showed that IPF induced by BLM can be relieved by CI SCFE by affecting the gene expression associated with abnormal deposition of collagen in the ECM. Figure 6 shows that α-SMA, vimentin, collagen-I, and Nu-β-catenin protein expression were unchanged following β-catenin siRNA knockdown in A549 cells (Fig. 6A) and MRC-5 cells (Fig. 6B) when cells were treated with Wnt-1 (20 ng/mL) for 24 h. After treatment with CI SCFE , the expression of the above proteins was also unchanged in both cell lines when β-catenin was knocked down compared with the wnt-1 group. The data illustrated that CI SCFE adjusted the Wnt/β-catenin signalling pathway to alleviate the process of IPF.

Discussion
IPF is a pulmonary interstitial disease with a high mortality rate. Currently, both the morbidity and mortality of IPF has increased year after year. Even the mortality rate is higher than that of most tumours. However, owing to the poorly understood potential pathogenesis of IPF, there is still a lack of drugs that can completely cure IPF.
Our previous study illustrated that CI SCFE can improve the antitumour effect of BLM while simultaneously attenuating its toxicity effects and CI SCFE could inhibit acute lung injury induced by LPS [22,24,27]. This current study clarified that CI SCFE can inhibit IPF development by affecting the balance of the Wnt/β-catenin pathway. IPF is caused by a variety of factors. The idea that abnormal ECM deposition is a cardinal feature of pulmonary fibrosis has been generally accepted. The ECM, which comprises collagens, proteoglycans, elastin, and other molecules, is involved in proper lung function. Myofibroblasts have been confirmed to contribute to ECM accumulation in vivo and in vitro [28]. Studies have found that both epithelial cells and fibroblasts in the lungs can transform into myofibroblasts through EMT, which is the process by which epithelial cells transform into mesenchymal cells. According to molecular and functional characterization, EMT is divided into three categories: type I EMT related to embryogenesis, type II EMT linked to fibrosis and wound healing and type III EMT involved in cancer metastasis [29]. It is generally accepted that type II EMT has a substantial effect on the development of IPF [30,31]. In addition, studies have shown that the occurrence of EMT is accompanied by an increase in the expression of a mesenchymal phenotype marker (vimentin) and abnormal deposition of ECM proteins, such as α-SMA and collagen-I [32,33]. MRC-5 cells, a cell line that retains the biochemical characteristics of lung fibroblasts, are commonly utilized as lung fibroblast cells in pulmonary fibrosis research [34,35]. Although the A549 cell line is a human-derived lung adenocarcinoma cell line, it is widely used to establish an EMT model by stimulating the cells with Wnt-1 or other cytokines [36,37] because its morphology and basic cell functions are similar to those of human alveolar type II epithelial cells. Therefore, to investigate whether CI SCFE could affect the transformation of epithelial cells and fibroblasts to myofibroblasts in vitro, we used A549 cells and MRC-5 cells to establish a cell model. MRC-5 cells were treated with Wnt-1 to establish a cell transdifferentiation model, and A549 cells were stimulated by Wnt-1 to establish an EMT model. Our experimental results indicated that CI SCFE could significantly decrease vimentin, α-SMA and collagen-I expression in A549 cells, MRC-5 cells, and lung tissues of BLM-induced IPF rats. The data illustrated that BLM-induced IPF could be inhibited by CI SCFE via suppression of EMT and abnormal ECM deposition.
MMPs and tissue inhibitors of metalloproteinases (TIMPs) have been reported to participate in the formation and degradation of the ECM. MMPs are multidomain enzymes that have outstanding roles during the cleavage of constituents of the ECM. Twenty-four MMP genes have been identified in humans, and 8 MMPs have been implicated in IPF development. Not only did in vitro studies indicate that there was high expression of MMP-3 in the alveolar epithelial cells and lungs of IPF patients, but in vivo studies have also confirmed this phenomenon. Researchers observed high expression of MMP-9 in the lung tissues of IPF patients. Additionally, TIMPs are endogenous inhibitors that control the catalytic activity of MMPs. Usually, under normal physiological conditions, the ratio of MMPs to TIMPs is approximately 1:1, but under pathological conditions, this balance may be destroyed, leading to excessive deposition of ECM components [38]. Therefore, an imbalance of MMPs/TIMPs is associated with IPF. Studies have shown that not only MMP-3 levels but also MMP-9 levels were increased in IPF lungs [39]. TIMP-1 levels were also upregulated in IPF lungs but to a lesser extent than those of MMP-3 [40]. Y. Wang et al. [41] found that artesunate could inhibit the occurrence of IPF by decreasing MMP-3, MMP-9 and TIMP-1 levels. Hai-Tao Zhang et al. [42] confirmed that the balance of MMP-9/TIMP-1 expression in IPF patients was disrupted, and this ratio was obviously higher than that in normal patients. After glucocorticoid intervention, the MMP-9/TIMP-1 ratio in IPF patients was restored closer to homeostatic levels, and the MMP-9 and TIMP-1 levels were reduced. The experimental data indicated that CI SCFE could significantly decrease MMP-3 and MMP-9 mRNA levels in the lung tissues of BLM-induced IPF rats and in A549 and MRC-5 cells. At the same time, the level of TIMP-1 gene expression could also be downregulated by CI SCFE . The MMP/TIMP ratio was also decreased after CI SCFE treatment. These data illustrated that CI SCFE could inhibit the progression of BLM-induced IPF by regulating the imbalance of MMPs/TIMPs, which facilitates the elimination of abnormal ECM deposition.
Even though we know that EMT and the ECM are closely related to IPF development, we still have a poor understanding of the clear mechanism by which EMT and the ECM influences IPF. Recently, compelling evidence has revealed that Wnt/β-catenin reactivation is connected with EMT and the ECM [43,44]. The The results are presented as the means ± SD. (n = 6). aa , P < 0.01 vs the control group; b , P < 0.05 and bb , P < 0.01 vs the Wnt-1 group. ## , P < 0.01 vs the normal group; * , P < 0.05 and ** , P < 0.01 vs the model group Wnt/β-catenin pathway participates in adult stem cell maintenance, embryonic development and homeostasis [45,46]. However, its abnormal activation leads to the progression of carcinomas of the liver, colon, lung and other organs [47][48][49]. Recently, some studies have shown that its abnormal activation is also involved in fibrotic diseases, such as IPF and renal fibrosis [13,50,51]. Therefore, Wnt/β-catenin signalling is considered a promising new target in the treatment of fibrotic disease [52]. In addition, the Wnt/β-catenin pathway was shown to be involved in the process of ECM deposition and EMT induction. Therefore, we speculated that CI SCFE could inhibit the progression of BLM-induced IPF by downregulating the balance of the Wnt/β-catenin pathway. Studies have confirmed that β-catenin is a critical element of the Wnt/β-catenin pathway [53]. When external factors stimulate the tissue, β-catenin translocates to and accumulates in the nucleus, thereby activating the entire Wnt/β-catenin pathway. Van et al. [54] confirmed that β-catenin accumulated in the cell nucleus of lung tissues from IPF patients. Liang [55] also confirmed that during myofibroblast proliferation, β-catenin levels would increase in the nuclei of lung tissue cells. Our results showed that in the cell nuclei of lung tissues, β-catenin expression can be improved by BLM. However, CI SCFE could inhibit this phenomenon. Based on the above results, we hypothesize that CI SCFE affects IPF by regulating the balance of Wnt/β-catenin pathway activity. Fig. 6 Effect of CI SCFE on the Wnt/β-catenin signalling pathway in A549 and MRC-5 cells following siRNA-mediated β-catenin knockdown. A Effect of CI SCFE on the Wnt/β-catenin signalling pathway in A549 cells. B Effect of CI SCFE on the Wnt/β-catenin signalling pathway in MRC-5 cells. The results are presented as the means ± SD. (n = 3). To fit into the manuscript properly, the gel was reasonably trimmed. a , P < 0.05 vs control group To confirm the hypothesis that CI SCFE could inhibit the progression of BLM-induced IPF by downregulating Wnt/β-catenin pathway activation, we used MRC-5 cells and A549 cells to investigate whether CI SCFE could affect BLM-induced IPF after knockdown of the β-catenin gene. The β-catenin gene is a key gene in the Wnt/βcatenin pathway which will only be activated when β-catenin transferred into the cell nucleus. Therefore, we knockdown single gene β-catenin to block the activation of Wnt/β-catenin pathway. Human β-catenin siRNA was transfected into MRC-5 and A549 cells. The study results illustrated that CI SCFE could suppress vimentin, α-SMA, β-catenin and collagen-I expression in Wnt-1-treated MRC-5 and A549 cells. By contrast, in MRC-5 and A549 cells transfected with β-catenin siRNA, CI SCFE did not affect vimentin, α-SMA, β-catenin or collagen-I expression levels. The results revealed that CI SCFE could inhibit the progression of BLM-induced IPF by influencing the balance of the Wnt/β-catenin pathway.
Our study still has some limitations. The pathogenesis of IPF is quite complicated. In this study, we investigated whether CI SCFE inhibited IPF by downregulating Wnt/βcatenin pathway activity. However, we have not studied whether the attenuating effect of CI SCFE is related to other signalling pathways. In addition, we have not studied the specific components in CI SCFE that elicited these changes. Therefore, further research regarding these issues is necessary, and we will continue to study the issues related to the attenuating effect of CI SCFE on IPF.

Conclusions
CI SCFE could alleviate IPF induced by BLM. The studies demonstrated that CI SCFE could alleviate EMT by adjusting the balance of the Wnt/β-catenin pathway and ultimately attenuate BLM-induced IPF. These results illustrated that after further research, CI SCFE could become a potential drug for IPF.