Ginseng saponin metabolite 20(S)-protopanaxadiol relieves pulmonary fibrosis by multiple-targets signaling pathways

Abstract

Background: Panax ginseng Meyer is a representative Chinese herbal medicine with antioxidant and anti-inflammatory activity. 20(S)-Protopanaxadiol (PPD) has been isolated from ginseng and shown to have promising pharmacological activities. However, effects of PDD on pulmonary fibrosis (PF) have not been reported. We hypothesize that PDD may reverse inflammation-induced PF and be a novel therapeutic strategy.

Methods: Adult male C57BL/6 mice were used to establish a model of PF induced by bleomycin (BLM). The pulmonary index was measured, and histological and immunohistochemical examinations were made. Cell cultures of mouse alveolar epithelial cells were analyzed with Western blotting, co-immunoprecipitation, immunofluorescence, immunohistochemistry, siRNA transfection, cellular thermal shift assay and qRT-PCR.

Results: The survival rate of PPD-treated mice was higher than that of untreated BLM-challenged mice. Expression of fibrotic hallmarks, including α-SMA, TGF-β1 and collagen I, was reduced by PPD treatment, indicating attenuation of PF. Mice exposed to BLM had higher STING levels in lung tissue, and this was reduced by phosphorylated AMPK after activation by PPD. The role of phosphorylated AMPK in suppressing STING was confirmed in TGF-β1-incubated cells. Both in vivo and in vitro analyses indicated that PPD treatment attenuated BLM-induced PF by modulating the AMPK/STING signaling pathway.

Conclusion: PPD ameliorated BLM-induced PF by multi-target regulation. The current study may help develop new therapeutic strategies for preventing PF.

Keywords: 20(S)-protopanaxadiol; Adenosine 5'monophosphate-activated protein kinase; Pulmonary fibrosis; Stimulator of interferon genes.

Graphical abstract

Fig. 1

Effects of PPD on BLM-induced pulmonary fibrosis in mice. (A) Chemical structure of PPD. (B) Schematic of the experimental effect of PPD on BLM-induced PF in mice. Mice were sacrificed, and lung tissue was taken on day 21 after intratracheal administration of 3.5 mg/kg BLM. Intragastric PPD (10 or 40 mg/kg) or NDN (40 mg/kg, positive drug) was given for 14 consecutive days after 7 days of intratracheal BLM administration. (C) Mouse weights during the experimental period. (D) Percent survival. (E) Pulmonary index. (F) Representative pathological lung sections (H&E and Masson's staining), scale bar = 50 μm. (G&H) Lung inflammation and fibrosis scores. Means ± S.D. are presented (n = 9). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. PPD: protopanaxadiol. NDN: nintedanib.

Fig. 2

The effect of PPD on TGF-β1 and AMPK signaling pathways in lung tissue of mice with BLM-induced pulmonary fibrosis. (A&B) TGF-β1 protein and mRNA levels. (C) p-Smad2 protein expression. (D-G) Col Ⅰ and α-SMA protein and mRNA levels. (H–K) E-cadherin, Vimentin, p-AMPK and SIRT1 protein expression. Protein was detected in lung tissues by Western blotting and mRNA levels by qRT-PCR, and data are shown as mean ± S.D. (n = 4). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Fig. 3

Regulation of STING by PPD during bleomycin-induced lung fibrogenesis in mice. (A) Lung STING expression by Western blotting. (B) Scheme showing regulation of STING levels by PPD from BLM-induced mice. Mice were sacrificed for lung tissue sampling on days 14 and 21 after intratracheal administration with 3.5 mg/kg BLM. Intragastric PPD (40 mg/kg) or NDN (40 mg/kg, positive drug) was given for 14 consecutive days after 7 days of intratracheal BLM administration. (C-D) Representative micrographs of H&E, Masson's staining and STING immunohistochemistry in lung pathological sections at days 14 and 21 after intratracheal BLM administration, scale bar = 50 μm. (E&F) Lung inflammation and fibrosis scores at days 14 and 21 after intratracheal BLM administration. (G) Lung immunohistochemical scores; (H) STING mRNA in lung tissue at days 14 and 21 after intratracheal BLM administration by qRT-PCR. (I&J) AMPK and p-AMPK expression in lung at days 14 and 21 after intratracheal BLM administration by Western blotting. Data are expressed as mean ± S.D. (n = 4). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Fig. 4

The role of AMPK during PPD regulation of TGF-β1-treated MLE-12 cells. MLE-12 cells were treated with 10 ng/ml TGF-β1 or protopanaxadiol for 48 h. (A) PPD cytotoxicity (0-10 μM) towards MLE-12 was determined using an MTT assay. (B&C) E-cadherin and Vimentin proteins were detected by Western blotting. (D&E) CETSA-Wes investigation of AMPK-PPD interactions. (F) Molecular docking images of protopanaxadiol with AMPK. (G&H) p-AMPK and STING protein detected by immunofluorescence. Data are presented as mean ± S.D. (n = 4). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Fig. 5

AMPK activation and STING involvement in the antifibrotic effects of PPD in TGF-β1-treated MLE-12 cells. (A-D) MLE-12 cells were transfected with AMPK-siRNA, and expression of E-cadherin and Vimentin protein was detected by Western blotting. E-cadherin and Vimentin mRNA were detected by qRT-PCR. (E&F) MLE-12 cells were transfected with siRNA-AMPK, and STING mRNA and protein levels were determined. (G) p-AMPK binding to STING in MLE-12 cells exposed to TGF-β1. (H-L) MLE-12 were transfected with siRNA-STING and E-cadherin, Vimentin and p-AMPK protein detected by Western blotting. E-cadherin and Vimentin mRNA were detected by qRT-PCR. Data are presented as mean ± S.D. (n = 4). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Fig. 6

Key roles for AMPK activation and STING inhibition in the anti-PF effect of PPD. (A) Experimental scheme showing co-treatment of BLM-induced PF mice with 40 mg/kg PPD and 5 mg/kg Compound C (AMPK inhibitor) or 150 mg/kg CMA (STING inhibitor). Metformin (250 mg/kg) and C-176 (9 mg/kg) were used as positive drugs. (B) Weight changes during the experimental period. (C) Percent survival. (D) Pulmonary index. (E) Representative lung pathological sections (H&E and Masson's staining), scale bar = 50 μm. (F&G) Lung inflammation and fibrosis scores. (H-J) E-cadherin, Vimentin, and STING mRNA were detected by qRT-PCR. (K&L) E-cadherin and Vimentin proteins were detected by Western blotting (n = 4). Data are expressed as mean ± S.D. (n = 9). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.