Dynamic changes in macrophage populations and resulting alterations in Prostaglandin E2 sensitivity in mice with diet-induced MASH

Abstract

Background: The transition from metabolic dysfunction-associated steatotic liver disease (MASLD) to steatohepatitis (MASH) is characterized by a chronic low-grade inflammation, involving activation of resident macrophages (Kupffer cells; KC) and recruitment of infiltrating macrophages. Macrophages produce cytokines and, after induction of Cyclooxygenase 2 (COX-2), the key enzyme of prostanoid synthesis, prostaglandin E₂ (PGE₂). PGE₂ modulates cytokine production in an autocrine and paracrine manner, therefore playing a pivotal role in regulating inflammatory processes. Changes in the hepatic macrophage pool during MASLD progression to MASH could influence PGE₂- and cytokine-mediated signaling processes. The aim of this study was to characterize these changes in mice with diet-induced MASH and further elucidate the role of COX-2-dependently formed PGE₂ on the inflammatory response in different macrophage populations of mice with a macrophage-specific COX-2-deletion.

Methods: Male, 6-7-week-old wildtype mice were fed either a Standard or high-fat, high-cholesterol MASH-inducing diet for 4, 12 and 20 weeks. Liver macrophages were isolated and analyzed by flow cytometry. For in vitro experiments primary KC, peritoneal macrophages (PM) and Bone-marrow-derived macrophages (BMDM) were isolated from macrophage-specific COX-2-deficient and wildtype mice and treated with lipopolysaccharide (LPS) and/or PGE₂.

Results: During MASH-development, the proportion of KC (Clec4F⁺Tim4⁺) decreased, while the proportion of monocyte-derived macrophages (Clec4F^-Tim4^-) and monocyte-derived cells exhibiting a phenotype similar to KC (Clec4F⁺Tim4^-) significantly increased over time. In vitro experiments showed that exogenous PGE₂ completely abrogated the LPS-induced mRNA expression and secretion of tumor necrosis factor-alpha (TNF-α) in primary KC, PM and BMDM from wildtype mice. PM and BMDM, as in vitro models for infiltrating macrophages, were more sensitive to PGE₂ compared to KC. Deletion of COX-2 in all macrophage populations led to an impaired PGE₂-dependent feedback inhibition of TNF-α production. LPSinduced TNF-α mRNA expression was higher compared to the respective wildtype macrophage population.

Conclusion: The current study, using a murine MASH model, indicates that PGE₂ may have a protective, anti-inflammatory effect, especially by inhibiting the expression of pro-inflammatory cytokines such as TNFα in infiltrating monocyte-derived macrophages. An inhibition of endogenous PGE₂ synthesis in macrophages by pharmacological inhibition of COX-2 could potentially increase inflammation and promote the progression of MASH.

Keywords: Cyclooxygenase; Infiltrating macrophages; Inflammation; Kupffer cells; MASH; Prostaglandin E2; TNF-α.

Fig. 1

Dynamic changes in hepatic macrophage populations of mice with diet-induced MASH. (A) Representative density plots of liver macrophage populations at indicated feeding periods. After gating on CD45⁺, live and singlet cells, macrophages were identified as F4/80⁺ cells (top panel) and further analysed for expression of Tim4 and Clec4F (bottom panel). (B) Quantification of F4/80⁺ cells per milligram liver tissue. (C) Proportion of Clec4F⁺Tim4⁺ Kupffer cells (KC), Clec4F⁻Tim4⁻ monocyte-derived macrophages (MoMF), and Clec4F⁺Tim4⁻ monocyte-derived Kupffer cells (MoKC) as fractions of F4/80⁺ cells. Values are median (line), upper- and lower quartile (box) and extremes (whiskers) of n = 4–8 (STD), n = 8–11 (4 W MASH-D), n = 8–13 (12 W MASH-D), n = 9–11 (20 W MASH-D) mice. Statistics: One-way-ANOVA with Tukey´s post hoc test for multiple comparison. *p < 0.05

Fig. 2

Modulation of TNF-α mRNA and protein expression by exogenous PGE₂. Primary Kupffer cells (KC), peritoneal macrophages (PM) and bone marrow-derived macrophages (BMDM) were stimulated for 24 h with LPS and/or PGE₂. (A) Relative mRNA expression of TNF-α was determined by RT-qPCR with Hprt as the reference gene. (B) TNF-α protein levels were quantified by ELISA in cell culture supernatants. Values are median (line), upper- and lower quartile (box) and extremes (whiskers) of (A) n = 33 (KC), n = 26 (PM) or n = 29 (BMDM) or (B) n = 27 (KC), n = 26 (PM) or n = 26 (BMDM) independent experiments. Statistics: Two-way-ANOVA with Tukey´s post hoc test for multiple comparison. *p < 0.05

Fig. 3

PGE₂ inhibits TNF-α mRNA and protein expression in a dose-dependent manner. Primary Kupffer cells (KC), peritoneal macrophages (PM) and bone marrow-derived macrophages (BMDM) were stimulated for 24 h with LPS and indicated concentrations of PGE₂. (A) Relative mRNA expression of TNF-α was determined by RT-qPCR with Hprt as the reference gene. (B) TNF-α protein levels were determined by ELISA in cell culture supernatants. Values are mean ± SEM of n = 5 independent experiments. Statistics: Student´s t-test for unpaired samples. * vs. w/o PGE₂ with p < 0.05

Fig. 4

Modulation of OSM mRNA expression by exogenous PGE₂. Primary Kupffer cells (KC), peritoneal macrophages (PM) and bone marrow-derived macrophages (BMDM) were stimulated for 24 h with LPS and/or PGE₂. Relative mRNA expression of OSM was determined by RT-qPCR with Hprt as the reference gene. Values are median (line), upper- and lower quartile (box) and extremes (whiskers) of n = 28 (KC), n = 26 (PM) or n = 30 (BMDM) independent experiments. Statistics: Two-way-ANOVA with Tukey´s post hoc test for multiple comparison. *p < 0.05

Fig. 5

Modulation of LPS-mediated expression of cyclooxygenase 2 (COX-2) and secretion of prostaglandin E₂ (PGE₂) in macrophages from wildtype controls (WT) and COX-2-deficient mice (KO). Primary Kupffer cells (KC), peritoneal macrophages (PM) and bone marrow-derived macrophages (BMDM) were stimulated for 24 h with LPS. (A) Relative mRNA expression of COX-2 (gene name Ptges2) was determined by RT-qPCR with Hprt as the reference gene. (B) Protein lysates were analyzed by immunoblotting for COX-2 protein expression, with β-Actin serving as the loading control. All original blots are provided in Supplementary figure S5-7. (C) PGE₂ level were determined by ELISA in cell culture supernatants. Values are median (line), upper- and lower quartile (box) and extremes (whiskers) of n = 26–32 (WT) or n = 5–10 (KO) independent experiments. Statistics: Two-way-ANOVA with Tukey´s post hoc test for multiple comparison. *p < 0.05. n.d.: not detectable

Fig. 6

Modulation of cytokine mRNA and protein expression in macrophages from wildtype controls (WT) and COX-2-deficient mice (KO). Primary Kupffer cells (KC), peritoneal macrophages (PM) and bone marrow-derived macrophages (BMDM) were stimulated for 24 h with LPS. Relative mRNA expression of TNF-α (A) and OSM (C) were determined by RT-qPCR with Hprt as the reference gene. (B) TNF-α protein levels were determined by ELISA in cell culture supernatants. Values are median (line), upper- and lower quartile (box) and extremes (whiskers) of n = 26–32 (WT) or n = 5–10 (KO) independent experiments. Statistics: Two-way-ANOVA with Tukey´s post hoc test for multiple comparison. *p < 0.05