Vegetal oil-based ketogenic diet improves inflammation and fibrosis in experimental metabolic dysfunction-associated steatohepatitis

Abstract

Background and aims: Metabolic dysfunction-associated steatohepatitis (MASH) represents a growing cause of liver cirrhosis and hepatocellular carcinoma (HCC). However, effective therapy for MASH is still lacking. Despite recent studies suggest that ketosis might improve MASH evolution, the mechanisms involved have not been explored since common ketogenic diets cause severe steatohepatitis in mice. In this study, we have investigated the capacity of a new-formulated ketogenic diet (KD) containing vegetal fat in improving liver alterations associated with experimental MASH.

Methods: MASH was induced in C57BL/6 mice by feeding a cholesterol-enriched Western Diet (WD) for up to 16 weeks, followed by switching animals to KD for an additional eight weeks.

Results: We observed that KD administration greatly increased ketone body production and significantly reduced liver and body weights. Moreover, liver proteomic analysis and functional tests evidenced an improved glucose and lipid metabolism along with insulin resistance in KD-fed mice. These metabolic effects were associated with an amelioration in MASH-associated gut dysbiosis and with an improvement of hepatic steatosis, parenchymal injury and liver fibrosis. From the mechanistic point of view mice receiving KD showed a significant reduction in liver TREM2-positive monocyte-derived macrophages forming crown-like aggregates along with a lowering in the hepatic expression of pro-inflammatory/pro-fibrogenic markers such as CCL2, IL-12, CD11b, α1-procollagen, TGF-β1, osteopontin, and galectin-3. Consistently, in vitro experiments showed that β-hydroxybutyrate supplementation reduced TREM2 and galectin-3 expression by cultured Raw 264.7 macrophages.

Conclusions: Altogether, these results indicate that ketogenic diet based on vegetal fat effectively improves MASH metabolic derangements and steatohepatitis, and it might represent a potential therapeutic strategy in this disease.

Keywords: gut dysbiosis; ketone bodies; liver fibrosis; liver inflammation; metabolic dysfunction-associated steatotic liver disease; non-alcoholic fatty liver disease; non-alcoholic steatohepatitis.

Figure 1

Ketogenic diet (KD) administration does not induce hepatic injury, inflammation, or fibrosis in mice. Wildtype C57BL/6 mice were fed with standard (SD) or KD diets for 8 weeks. (A) Body and liver weights. (B) Hematoxylin/Eosin staining of liver sections (magnification 20×). (C) Circulating levels of alanine aminotransferase (ALT) and hepatic transcripts for the inflammatory markers TNF-α and CD11b as evaluated by RT-PCR. (D) Intrahepatic collagen deposition as evaluated by the transcripts of procollagen-1α and the staining of liver sections with Sirius Red (magnification 20×). Dots correspond to individual animals and the boxes include the values within the 25th and 75th percentile. The horizontal bars represent the medians, while the extremities of the vertical bars (10th–90th percentile) comprise 80% of the values.

Figure 2

Ketogenic diet administration modifies hepatic metabolic functions in mice with MASH. MASH was induced in 8-weeks old wildtype C57BL/6 mice by 16 weeks feeding with Western diet before the switching to ketogenic diet for further 8 weeks (WD16+KD8w). Reference mice received standard diet or WD for 24 weeks (SD24w or WD24w, respectively). (A) Experimental Plan. (B) Changes in body and liver weights and hepatic triglyceride content. (C) Hepatic transcripts for genes of glucose transporter 2 (GLUT2) insulin receptor substrate-1 (IRS-1) and peroxisome proliferator-activated receptor-γ coactivator 1-α (PPARγC1α). RT-PCR values are expressed as fold increase of 2^-ΔCT after normalization to the β-actin gene. Glucose tolerance test (GTT) was assessed as area under the curve (AUC). Dots correspond to individual animals and the boxes include the values within the 25th and 75th percentile. The horizontal bars represent the medians, while the extremities of the vertical bars (10th–90th percentile) comprise 80% of the values. (D) Ingenuity pathway analysis (IPA) of the changes in liver proteins involved in lipid metabolism, inflammatory response, hepatic steatosis, glucose metabolism and insulin resistance in mice switching to KD as compared to those that received WD only. (*=p<0.05).

Figure 3

Ketogenic diet improves MASH-associated steatosis and hepatic injury in mice. MASH was induced in 8-weeks old wildtype C57BL/6 mice by 16 weeks feeding with Western diet (WD) before the switching to ketogenic diet (KD) for further 8 weeks (WD16+KD8w). Reference mice received standard diet or WD for 24 weeks (SD24w or WD24w, respectively). (A) Hematoxylin/eosin-staining of liver sections (magnification 20×); (B) Changes in circulating levels of alanine aminotransferase (ALT) and Dipeptidyl Peptidase-4 (DPP4). (C) Hepatic gene expression of pro-inflammatory markers TNF-α, CD11b, CCL2, and IL-12p40. RT-PCR values are expressed as fold increase of 2^-ΔCT after normalization to the β-actin gene. Dots correspond to individual animals and the boxes include the values within the 25th and 75th percentile. The horizontal bars represent the medians, while the extremities of the vertical bars (10th–90th percentile) comprise 80% of the values.

Figure 4

Ketogenic diet administration reshapes the gut microbiota in mice with MASH. MASH was induced in 8-weeks old wildtype C57BL/6 mice by 16 weeks feeding with Western diet before the switching to ketogenic diet (KD) for further 8 weeks (WD16+KD8w). Reference mice received standard diet or WD for 24 weeks (SD24w or WD24w, respectively). (A) Principal component analysis (PCA) describing the changes in stool microbiota composition at different time points during mice feeding with WD (red) as compared to those remaining in SD (green) or following the switching from WD to KD (orange). (B) PCA describing cecum microbiota composition in mice receiving WD, SD or after switching from WD to KD. (C) Changes in the serum levels of propionic and acetic acids following mice switching from WD to KD. Dots correspond to individual animals and the boxes include the values within the 25th and 75th percentile. The horizontal bars represent the medians, while the extremities of the vertical bars (10th–90th percentile) comprise 80% of the values.

Figure 5

Ketogenic diet administration modulates the phenotype of hepatic macrophages and MASH-associated hepatic crown-like structures (hCLSs). MASH was induced in 8-weeks old wildtype C57BL/6 mice by 16 weeks feeding with Western diet (WD) before the switching to ketogenic diet (KD) for further 8 weeks (WD16+KD8w). Reference mice received standard diet or WD for 24 weeks (SD24w or WD24w, respectively). (A) Flow cytometry analysis of the distribution of CD11b^high/F4-80⁺ monocyte-derived macrophages (MoMFs) within the liver of mice with the different dietary regiments. (B) Phenotypic features of liver macrophages forming hCLSs as evidenced by double immunofluorescence of FFPE liver sections stained with fluorochrome-labelled antibodies against F4-80 and TREM2. (C) Changes in hepatic transcripts of CD9, TREM2, galectin-3 and osteopontin. Dots correspond to individual animals and the boxes include the values within the 25th and 75th percentile, while the horizontal bars represent the medians. The extremities of the vertical bars (10th–90th percentile) comprise 80% of the values. (D) KD effect on the prevalence of hCLSs evidenced by immunofluorescence staining with anti-F4-80 antibodies. (E) Effect of β-hydroxybutyrate (β-OHbut; 50 µM) in vitro supplementation on TREM2 and galectin-3 expression by cultured Raw 264.7 macrophages. The experimental groups are labelled as Cont (vehicle) and β-OHbut and each dot represents an experimental point.

Figure 6

Ketogenic diet administration improves MASH-associated fibrosis in mice modulating hepatic crown-like structures (hCLSs) expressing galectin-3 and osteopontin. MASH was induced in 8-weeks old wildtype C57BL/6 mice by 16 weeks feeding with Western diet (WD) before the switching to ketogenic diet (KD) for further 4 (WD16w+KD4w) or 8 weeks (WD16+KD8w). Reference mice received standard diet or WD for 24 weeks (SD24w or WD24w, respectively). (A) RT-PCR analysis of the hepatic transcripts for pro-fibrogenic markers TGF-β1 and procollagen-1α. Dots correspond to individual animals and the boxes include the values within the 25th and 75th percentile. The horizontal bars represent the medians, while the extremities of the vertical bars (10th–90th percentile) comprise 80% of the values. (B) Effect of 8 weeks feeding with KD on liver fibrosis. The morphological analysis quantification refers to individual readings on liver sections derived from 3-4 animals for each experimental group. (C) Effect of 4 weeks feeding with KD on the prevalence hCLSs expressing Galectin-3 (Gal-3) as evidenced by immunohistochemistry with anti-Gal-3 antibodies and (D) on circulating levels of osteopontin (OPN). The morphological analysis refers to individual readings on liver sections derived from 3-4 animals for each experimental group.