GAMG alleviates liver fibrosis through inducing ferroptosis in inflammatory macrophages via the IRF1/SLC7A11 signaling pathway

Abstract

The activation of inflammatory macrophages plays a pivotal role in the development of liver fibrosis (LF). Ferroptosis contributes to the clearance of inflammatory macrophages and the release of profibrotic factors. Glycyrrhetic Acid 3-O-Mono-β-d-glucuronide (GAMG) is a natural compound, the potential role of which on LF remains uncertain. In this study, GAMG treatment significantly reduced hepatocyte steatosis, fibroplasia, inflammatory cell infiltration, and collagen fiber deposition in LF mice. In addition, GAMG remarkably decreased the content of collagen protein and improved liver function indicators. Single-cell RNA sequencing revealed that GAMG significantly affected the changes of macrophage subsets in LF, and Funrich analysis identified IRF1 as a key transcription factor regulating the macrophage genome. IRF1 was significantly increased while ferroptosis related SLC7A11 was significantly down-regulated in GAMG treated inflammatory macrophages. Mass spectrometry metabolomics analysis showed that GAMG significantly affected metabolites associated with LF. In vivo and in vitro experiments further verified that GAMG induced ferroptosis of inflammatory macrophages through the IRF1/SLC7A11 axis, and ultimately alleviated LF. Therefore, GAMG induces ferroptosis of inflammatory macrophages by activating the IRF1/SLC7A11 axis, which provides a new strategy for the treatment of LF.

Keywords: Ferroptosis; Glycyrrhetic acid 3-O-Mono-β-d-glucuronide; IRF1; Inflammatory macrophages; Liver fibrosis; SLC7A11.

Graphical abstract

Fig. 1

Effect of GAMG on CCL4-induced liver fibrosis. (A) Chemical structure formula of GL, GAMG and GA. (B) Schematic treatment of mouse model. (C) Mouse liver sections stained with hematoxylin-eosin and masson (×200). (D) Liver hydroxyproline content. (E) Spleen weight. (F) Liver weight. The data were expressed as mean ± SD (n = 5/group). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 2

Relative content changes of 27 potential markers in liver fibrosis group and GAMG administration group. (A) OPLS-DA score chart. (B) OPLS-DA verification diagram, R²Y = 0.989, Q² = 0.752. (C) Heat maps of differential metabolites. (D) KEGG functional enrichment analysis results of differential metabolites in MetaboAnalyst database. (E) SMPDB functional enrichment analysis results of differential metabolites in MetaboAnalyst database. (F–I) The relative content changes of 27 potential markers in control, LF and GAMG administration groups. The data were expressed as mean ± SD (n = 3/group). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 3

Cluster analysis of RNA-seq data. (A) Flow chart of sample processing and sequencing of RNA-seq. (B) The proportion of different cell subpopulations in each sample. (C) The 8 cell clusters grouped by UMAP cluster analysis. (D) Distribution of individual cell clusters in different samples (control, LF and GAMG administration groups). (E) UMAP expression map of specific marker genes of each cell cluster. (F) The first five differential genes of each cell cluster. (G) Cell communication circles of control, LF and GAMG administration groups, with color shades representing interaction intensity.

Fig. 4

Effects of GAMG on the function of inflammatory macrophages. (A) The content proportion of different macrophage subsets in each sample. (B) The percentage of each macrophage subpopulation. (C) The cluster heat map showed the difference of gene expression in different macrophage subpopulations. (D) Analysis of molecular functions, biological processes and biological pathways of differential genes.

Fig. 5

Effect of GAMG on the release of inflammatory cytokines in inflammatory neutrophils and pseudo-time analysis. (A) The content proportion of different cell subsets in each sample. (B) The proportion of neutrophil subpopulations expressed as a percentage. (C) The expression difference of genes in different neutrophil subpopulations demonstrated by clustering heat maps. (D–F) The trajectory distribution of each neutrophil population in different samples over time. (G) The distribution of neutrophil percentage in the two-dimensional state space defined by Monocle 2.

Fig. 6

GAMG promotes ferroptosis in inflammatory macrophages. (A) Mitochondria observed by transmission electron microscopy (TEM). (B) The concentration of iron ions in GAMG-treated macrophages (GAMG concentrations were 10, 20 and 40 μM, respectively). (C) The cell viability of macrophages treated with Erastin (ferroptosis inducer), GAMG, and GAMG + Fer-1 (ferroptosis inhibitor). (D) The cell viability of macrophages treated with Fer-1, Z-VAD (apoptosis inhibitor) and Nec-1 (necrosis inhibitor). (E) The cell proliferation of macrophages detected by EdU experiment. (F) The GSH content in macrophages. (G) The MDA content in macrophages. (H) The ROS content in macrophages. The data were expressed as mean ± SD (n = 3/group). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 7

GAMG promotes ferroptosis and inactivation of inflammatory macrophages by inhibiting SLC7A11 transcription via regulating IRF1. (A) The protein expressions of ACSL4, GPX4 and SLC7A11 in inflammatory macrophages after GAMG treatment. (B) The differential genes for key transcription factors in the single-cell transcriptome identified by using Funrich software. (C) The expression of IRF1 in inflammatory macrophages after GAMG treatment. (D–E) The mRNA expression levels of IRF1 and SLC7A11 in vivo after GAMG treatment. (F) The expression of IRF1 and SLC7A11 in macrophages transfected with sh-IRF1. (G) The involvement of IRF1 in regulating transcription of SLC7A11 predicted by using the hTFtarget database. (H) Schematic diagram of potential IRF1 binding sites in the SLC7A11 promoter. (I) Enrichment of IRF1 on the promoter of SLC7A11 identified by ChIP analysis. (J) The relative luciferase activity of SLC7A11 promoter of wild type (WT) and mutant type (MUT) (control or IRF1 plasmid was co-transfected). The data were expressed as mean ± SD (n = 3/group). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 8

GAMG improves CCL4-induced liver fibrosis through the IRF1/SLC7A11 axis. (A) Mouse liver sections stained with hematoxylin-eosin and masson (×200). (B) Liver hydroxyproline content. (C) Spleen index. (D) Liver index. (E–F) The degree of liver fibrosis (oil red staining) and the immunofluorescence staining of Collagen I and α-SMA. (G) Mitochondrial damage in mouse liver tissues observed by transmission electron microscopy (TEM). (H) Levels of GSH, MDA, and ROS in mouse liver tissues. The data were expressed as mean ± SD (n = 5/group). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 9

GAMG improves BDL-induced liver fibrosis via the IRF1/SLC7A11 axis. (A) Mouse liver sections stained with hematoxylin-eosin and masson (×200). Liver hydroxyproline content. (C) Spleen index. (D) Liver index. (E–F) The degree of liver fibrosis (oil red staining) and the immunofluorescence staining of Collagen I and α-SMA. (G) Mitochondrial damage in mouse liver tissues observed by transmission electron microscopy (TEM). (H) Levels of GSH, MDA, and ROS in mouse liver tissues. The data were expressed as mean ± SD (n = 5/group). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.