Semaglutide Has Beneficial Effects on Non-Alcoholic Steatohepatitis in Ldlr-/-.Leiden Mice

Abstract

Semaglutide, a glucagon-like peptide-1 receptor agonist, is an antidiabetic medication that has recently been approved for the treatment of obesity as well. Semaglutide is postulated to be a promising candidate for the treatment of non-alcoholic steatohepatitis (NASH). Here, Ldlr-/-.Leiden mice received a fast-food diet (FFD) for 25 weeks, followed by another 12 weeks on FFD with daily subcutaneous injections of semaglutide or vehicle (control). Plasma parameters were evaluated, livers and hearts were examined, and hepatic transcriptome analysis was performed. In the liver, semaglutide significantly reduced macrovesicular steatosis (-74%, p < 0.001) and inflammation (-73%, p < 0.001) and completely abolished microvesicular steatosis (-100%, p < 0.001). Histological and biochemical assessment of hepatic fibrosis showed no significant effects of semaglutide. However, digital pathology revealed significant improvements in the degree of collagen fiber reticulation (-12%, p < 0.001). Semaglutide did not affect atherosclerosis relative to controls. Additionally, we compared the transcriptome profile of FFD-fed Ldlr-/-.Leiden mice with a human gene set that differentiates human NASH patients with severe fibrosis from those with mild fibrosis. In FFD-fed Ldlr-/-.Leiden control mice, this gene set was upregulated as well, while semaglutide predominantly reversed this gene expression. Using a translational model with advanced NASH, we demonstrated that semaglutide is a promising candidate with particular potential for the treatment of hepatic steatosis and inflammation, while for the reversal of advanced fibrosis, combinations with other NASH agents may be necessary.

Keywords: NAFLD; NASH; fibrosis; gene expression; inflammation.

Figure 1

Semaglutide improved metabolic parameters in Ldlr-/-.Leiden mice. Body weight (A), food intake (B), blood glucose levels (C) and plasma insulin levels (D) were determined at several timepoints throughout the study. Values are presented as mean ± SEM for n = 8 mice on chow diet, n = 15 on fast-food diet (FFD) control mice and n = 15 mice on FFD supplemented with semaglutide. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. FFD control.

Figure 2

Body composition and plasma levels of cholesterol, triglycerides and ALT were ameliorated in semaglutide-treated mice. Fat mass and lean mass (A), perigonadal white adipose tissue (WAT), visceral WAT, subcutaneous WAT and interscapular brown adipose tissue (BAT) (B), plasma cholesterol (C), plasma triglycerides (D) and plasma alanine transaminase (ALT) (E) were determined at several timepoints or at the study endpoint. Values are presented as mean ± SEM for n = 8 mice on chow diet, n = 15 fast-food diet (FFD) control mice and n = 15 mice on FFD supplemented with semaglutide. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. FFD control.

Figure 3

Hepatic steatosis and inflammation were ameliorated with semaglutide treatment in Ldlr-/-.Leiden mice. Representative histological photomicrographs of H&E-stained liver cross-sections (A), liver weight (B), macrovesicular steatosis (C) and microvesicular steatosis (D) as percentage of surface area, hepatic concentrations of triglycerides, cholesteryl esters and free cholesterol (E) and the number (#) of inflammatory aggregates per mm² microscopic field (F) were all determined at the study endpoint. Values are presented as mean ± SEM for n = 8 mice on chow diet, n = 15 on fast-food diet sacrificed at t = 0 weeks, after a 25-week run-in period (FFD reference), n = 15 on FFD until the study endpoint (FFD control) and n = 15 on FFD supplemented with semaglutide. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. FFD control.

Figure 4

Semaglutide improved hepatic fibrosis architecture in Ldlr-/-.Leiden mice. Representative images of Sirius-Red-stained liver cross-sections and of computerized analysis of collagen fibers of the same section (A), fibrosis as percentage of surface area (B), hepatic collagen content (C), fibrosis stage (D), parenchymal-tissue-corrected area ratio of fine and aggregated collagen fibers (E) and collagen reticulation index (F). Values are presented as mean ± SEM for n = 8 mice on chow diet, n = 15 on fast-food diet sacrificed at t = 0 weeks, after a 25-week run-in period (FFD reference), n = 15 on FFD until the study endpoint (FFD control) and n = 15 on FFD supplemented with semaglutide. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. FFD control.

Figure 5

The effects of semaglutide on parameters of atherosclerosis are minimal. Representative images of HPS-stained liver cross-sections (A), cholesterol exposure (concentration*weeks) (B), atherosclerotic lesion area per cross-section (C), number of atherosclerotic lesions per cross-section (D) and lesion severity (E) were determined at the study endpoint. Values are presented as mean ± SEM for n = 8 mice on chow diet, n = 15 on fast-food diet sacrificed at t = 0 weeks, after a 25-week run-in period (FFD reference), n = 15 on FFD until the study endpoint (FFD control) and n = 15 on FFD supplemented with semaglutide. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. FFD control.

Figure 6

Semaglutide reverses FFD-induced expression of upstream regulators in the Ldlr-/-.Leiden mouse model, which closely represents human NASH. Venn diagram (A) showing the overlap of the predicted activation state of upstream regulators (URs) in the liver based on expression changes in known target genes. The white circle indicates Ldlr-/- mice that received the FFD for a total of 37 weeks (FFD control) vs. the healthy chow group, and the blue circle indicates mice treated with semaglutide for the final 12 weeks of the study vs. FFD control mice. The top 15 most significantly changed URs (−log(p-value)) for URs affected by FFD and reversed by semaglutide (left) and for URs affected by semaglutide but not induced by FFD (right) are shown. Heatmap (B) showing expression of genes differentially regulated in human NASH patients with severe fibrosis (stage F3 or F4) vs. NASH patients with mild fibrosis (stage F0 or F1), recapitulated in Ldlr-/-.Leiden mice fed FFD for a total of 37 weeks (FFD control) relative to chow-fed mice (left column) and semaglutide-treated mice relative to FFD control mice (right column). Red color indicates upregulation, blue color indicates downregulation and asterisks (*) indicate genes that are significantly (p < 0.05) upregulated or downregulated in semaglutide-treated mice.