The Thyroid Hormone Analog GC-1 Mitigates Acute Lung Injury by Inhibiting M1 Macrophage Polarization

Abstract

Acute lung injury (ALI)/acute respiratory distress syndrome (ARDS) is a life-threatening condition with a high mortality rate of ≈40%. Thyroid hormones (THs) play crucial roles in maintaining homeostasis of the cellular microenvironment under stress. The previous studies confirmed that the clinical-stage TH analog GC-1 significantly alleviates pulmonary fibrosis by improving the function of mitochondria in epithelial cells. However, the effects of GC-1 on macrophages in lung injury and the related mechanisms remain unclear. This study evaluated the therapeutic effects of GC-1 in two murine models of lipopolysaccharide (LPS)- or hydrochloric acid (HCl)-induced ALI. Additionally, mouse alveolar macrophages (AMs) and human THP-1-derived macrophages are utilized to investigate the impact of GC-1 on macrophage polarization. GC-1 effectively reduces the inflammatory response and lung injury in ALI mice, as evidenced by neutrophil infiltration, cytokine levels, alveolar fluid clearance, and pulmonary pathology. Notably, GC-1 selectively inhibits M1 macrophage polarization, which may be achieved by impeding NF-κB signaling activation through the DNMT3b-PPARγ-NF-κB pathway in a TH receptor β1 (TRβ1)-dependent manner, consequently suppressing the polarization of macrophages toward the M1 phenotype and overproduction of inflammatory cytokines. Overall, these findings highlight the immunomodulatory property of GC-1 as an anti-inflammatory strategy for ALI/ARDS and inflammation-related diseases.

Keywords: ALI/ARDS; DNMT3b‐PPARγ‐NF‐κB pathway; GC‐1; macrophage polarization; thyroid hormone.

Figure 1

GC‐1 mitigated lung injury, neutrophil infiltration, and proinflammatory cytokine production in mice with LPS‐ or HCl‐induced ALI. A) Schematic representation of the experiment. C57BL/6 mice were challenged intratracheally with LPS (5 mg kg⁻¹ body weight) or HCl (0.1 m, pH 1.0) on Day 0 followed by daily intraperitoneal injection of GC‐1 (100 µg kg⁻¹ body weight) or saline for 3 days. B) Representative histological sections of LPS‐induced lung tissue stained with H&E (n = 3), Scale bars: 500 and 20 µm (insets). LPS‐induced lung injury was evaluated by C) alveolar fluid clearance (n = 6), and D) the concentration of white blood cells (WBC) recovered from BAL fluid (n = 4). E) Representative LPS‐induced BAL cell smear stained with Diff‐Quik (n = 4), Scale bars: 100 and 20 µm (insets). F) IL‐1β and G) TNF‐α levels in LPS‐induced BAL fluid were measured by ELISA (n = 6). H) Representative H&E staining of HCl‐induced lung histological sections (n = 3), Scale bars: 100 and 20 µm (insets). I) Representative HCl‐induced BAL cell smear stained with Diff‐Quik (n = 6), Scale bars: 100 and 20 µm (insets). The values are shown as mean ± SD. ^* p < 0.05; ^** p < 0.01; ^**** p < 0.0001.

Figure 2

GC‐1 selectively inhibited the M1 macrophage polarization. A) Representative images of IHC staining for macrophage marker F4/80, M1 macrophage marker CD86, or M2 macrophage marker CD206 in lung sections of LPS‐induced ALI model mice (n = 3), Scale bars: 100 and 20 µm (insets). B) Quantitative statistical results of IHC‐stained positive cells (n = 3). C) Representative pictures of F4/80, CD86, and CD206 IF staining in lung sections from ALI mice induced by HCl (n = 3), Scale bars: 50 µm. Quantitative statistical results of fluorescence intensity from D) F4/80⁺CD86⁺ macrophages and E) F4/80⁺CD206⁺ macrophages (n = 3). F) Representative flow cytometry plots showing the proportions of M1 (CD11b⁺CD86⁺CD206⁻) and M2 (CD11b⁺CD86⁻CD206⁺) macrophages (n = 3). THP‐1 cells were treated with LPS (100 ng mL⁻¹) alone or in combination with GC‐1 (100 nm) for 24 h after PMA induction. G) CD11b⁺CD86⁺CD206⁻ M1 macrophages and H) CD11b⁺CD86⁻CD206⁺ M2 macrophages quantitative statistical results (n = 3). I) Immunoblot analysis of lysates from THP‐1 cells (n = 3). J) Immunoblot gray values statistical quantification (n = 3). The values are shown as mean ± SD. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001; ns = not significant.

Figure 3

GC‐1 inhibited the LPS‐induced activation of the NF‐κB signaling. A) Schematic representation of the experiment. B) Western blot analysis was performed to assess the NF‐κB signaling activation in THP‐1 cells (n = 3). Prior to treatment with LPS (200 ng mL⁻¹) for 30 min, THP‐1 cells were pretreated with 100 nM GC‐1 for 12 h. C) Immunoblot gray values statistical quantification (n = 3). D) Representative IF images of THP‐1 cells (n = 3), depicting the nuclear translocation of NF‐κB subunit p65 treated with LPS (200 ng mL⁻¹) for 30 min, Scale bars: 20 µm. E) IL‐1β and F) TNF‐α levels in supernatants from AMs and THP‐1 cells treated with LPS (100 ng mL⁻¹) and GC‐1 (100 nM) for 24 h were measured by ELISA (n = 3). G) qRT‐PCR analysis of p65 in murine lung tissue homogenate induced by LPS (n = 3). H) Representative pictures of p65 IF staining in lung sections from ALI model mice (n = 3), Scale bars: 20 µm. The values are shown as mean ± SD. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001; ns = not significant.

Figure 4

GC‐1 suppressed the activation of NF‐κB signaling in a PPARγ‐dependent manner. A) qRT‐PCR analysis of PPARγ mRNA expression in THP‐1 cells treated with LPS (100 ng mL⁻¹) and GC‐1 (100 nm) for 24 h (n = 3). B) Representative images of P65 and PPARγ IF staining of THP‐1 cell slides treated with LPS (100 ng mL⁻¹) and GC‐1 (100 nm) for 24 h (n = 3), Scale bars: 20 and 10 µm (insets). C) WB demonstrated phosphorylation of P65 in THP‐1 cells pretreated with GC‐1 (100 nm) along with the PPARγ inhibitor GW9662 (GW, 5 µm) for 12 h, followed by treatment with LPS (200 ng mL⁻¹) for 30 min, as well as the expression of PPARγ and IL‐β in THP‐1 cells treated with LPS (100 ng mL⁻¹), GC‐1 (100 nm) and GW9662 (5 µm) for 24 h. D) Immunoblot gray values statistical quantification (n = 4). E) WB analysis demonstrated the impact of the PPARγ agonist pioglitazone (Piog, 10 µm) on the phosphorylation of P65 and the expression levels of both PPARγ and IL‐1β in THP‐1 macrophage under similar treatment conditions mentioned above. F) Statistical quantification based on immunoblot gray values (n = 4). The levels of G) IL‐1β and H) TNF‐α in AM and THP‐1 cell supernatants were detected by ELISA (n = 3). The values are shown as mean ± SD. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001.

Figure 5

GC‐1 suppressed the upregulation of DNMT3b induced by LPS to increase the expression of PPARγ. A) Representative images of IF staining for DNMT3b and PPARγ in THP‐1 cells (n = 3), scale bars are 20 and 10 µm (insets). B) The level of protein expression of DNMT3b, PPARγ, and IL‐1β were analyzed by Western blotting in THP‐1 cells treated with the DNA methylation inhibitor 5‐azacytidine (Aza) (10 µm), LPS (100 ng mL⁻¹), and GC‐1 (100 nm) for 24 h. C) Quantification of immunoblotting grayscale values (n = 4). D) Western blotting in THP‐1 cells was performed to investigate the impact of DNMT3b knockdown on PPARγ and IL‐1β production. E) Quantification of grayscale values in immunoblotting (n = 3). F) Endogenous Co‐IP experiment in THP‐1 cells to investigate the role of GC‐1 in the interaction of PPARγ and DNMT3b. G) The luciferase assays showed knocking down DNMT3b significantly increased PPARγ transcriptional activity in THP‐1 cells (n = 3). H) Schematic illustration of the primers used in the ChIP experiment, showing the position on the PPARγ promoter. I) ChIP assays demonstrated the binding of DNMT3b to the promoter region of PPARγ (n = 3). The values are shown as mean ± SD. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001; ns = not significant.

Figure 6

GC‐1 inhibited the activation of DNMT3b induced by LPS via TRβ1. A) Representative images of IF staining for DNMT3b and TRβ1 in LPS‐induced ALI model mice (n = 3), scale bar is 20 µm. Illustrative images of IF staining for B) TRβ1 overexpression (OE) and C) TRβ1 knockdown (KD) in THP‐1 cells (n = 3), scale bars are 20 and 5 µm (insets). D) The impact of TRβ1 overexpression on DNMT3b production was investigated in THP‐1 cells using Western blotting. E) Gray values statistical quantification (n = 3). F) Western blotting in THP‐1 cells to investigate the impact of TRβ1 knockdown on DNMT3b production. G) Quantification of grayscale values in immunoblotting (n = 3). H) The luciferase assays demonstrated TRβ1 overexpression significantly attenuated DNMT3b transcriptional activity in THP‐1 cells (n = 3). I) Schematic illustration of the primers used in the ChIP, indicating their position on the DNMT3b promoter. J) ChIP assays provided evidence of TRβ1 binding to DNMT3b promoter (n = 3). The values are shown as mean ± SD. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001; ns = not significant.

Figure 7

Inhibition of DNMT3b restrained M1 macrophage polarization and lung injury in ALI model mice. A) Schematic representation of the experiment. After treatment with LPS, C57BL/6 mice were intraperitoneally injected with DNA methylation inhibitor 5‐azacytidine (Aza) (1 mg kg⁻¹ body weight), while other treatment methods are mentioned in Figure 1. B) WB analysis of the effects of Aza on the expression of DNMT3b, PPARγ, and IL‐1β in mouse lungs. C) Quantification results for immunoblotting grayscale values (n = 4). D) IF staining for F4/80 and CD86 in lung slices from mice (n = 3). Scale bars: 200 and 20 µm(insets). E) Representative histological sections of lung stained with H&E (n = 3). Scale bars: 100 and 20 µm (insets). Lung injury was evaluated by measuring F) the BAL fluid total protein content, G) lung wet/dry weight ratio, and H) BAL fluid WBC concentration (n = 5). The concentrations of I) IL‐1β and J) TNF‐α in mouse BAL fluid were determined using ELISA assays (n = 5). The values are shown as mean ± SD. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001.