Sleeve-gastrectomy results in improved metabolism and a massive stress response of the liver proteome in a mouse model of metabolic dysfunction-associated steatohepatitis

Abstract

Background: Bariatric surgery has been shown to improve the histopathological findings in patients with obesity and metabolic dysfunction-associated steatohepatitis, but there are also reports about non-responders or progressive disease after bariatric interventions. Therefore, it is of utmost importance to understand the pathophysiological processes in the liver after bariatric surgery.

Materials and methods: In the present study, 4 weeks old male C57/Bl6 mice were fed a Western Diet to induce metabolic dysfunction-associated steatohepatitis and sleeve-gastrectomy (SG), or sham operation in the pair-fed and ad libitum control group were performed. Mice were observed for two or eight weeks after surgery and metabolic assessment was performed throughout the experiment. Histopathology, flow cytometry and proteomic analyses were conducted to evaluate hepatic inflammation, liver metabolism and affected signaling pathways.

Results: Weight loss was higher, and metabolism significantly improved after SG. Two weeks after SG major inflammatory and regulatory disturbances in the liver were observed. The proportion of hepatic CD3⁺NK1.1⁺ cells were decreased, and proteins involved in apoptosis like Fas, Casp1 and Casp9 or in the acute phase response were upregulated in SG mice. These disturbances decreased in the long-term and we observed an increase of many proteins involved in lipid metabolism eight weeks following SG.

Conclusions: The rapid weight loss and decrease of hepatic fat after SG lead to a proinflammatory response in the liver in the early phase after surgery, which changes to a more moderate immune response in the long-term. We suggest a preoperative risk stratification and postoperative surveillance depending on the histopathological findings.

Keywords: Bariatric surgery; Metabolic dysfunction-associated steatohepatitis; Metabolic dysfunction-associated steatotic liver disease; Nonalcoholic fatty liver diseases; Nonalcoholic steatohepatitis; Sleeve-gastrectomy.

Fig. 1

Study design and surgical procedure. (a) Mice were fed sucrose-enriched water (10 %) and a western diet with 40 kcal% fat, 20 kcal% fructose and 2 kcal% cholesterol (WD, D09100301, Research Diets, Inc., New Brunswick, NJ, USA) ad libitum at four weeks of age. After 12 weeks of diet mice were allocated to the experimental groups (control, Pf and SG) and underwent sham operation (control and Pf) or sleeve-gastrectomy. Food and weight were measured weekly. IP-GTTs and blood samples were collected at baseline, before surgery as well as 2 and 8 weeks after surgery. Two or eight weeks after surgery mice were sacrificed for organ explantation. (b) Representative pictures of the SG procedure. After midline laparotomy and mobilizing the stomach a non-traumatic surgical (vascular) clamp was placed on the resection margin. Then the stomach including the entire forestomach was resected and the remaining section of the stomach closed with a running suture (Created with BioRender.com). (c) Experimental groups.

Fig. 2

Body weight and clinical status. Monitoring of body weight throughout the (a) 2-week and (b) 8-week study. (c) %TWL before termination of the experiment was significantly higher in SG mice in the 2- and 8-week experiment. For animal welfare assessment a clinical scoring system was used. (d) shows the clinical score and degree of strain in the first postoperative week and (f) before termination of the experiment. ∗p < 0.05, ∗∗p < 0.01 and ∗∗∗p < 0.001.

Fig. 3

Metabolic assessments. (a, b, c, d) Area under the glucose time curve (AUC) of intraperitoneal glucose tolerance test (IP-GTT) for the 2- and 8-week experiments; (e, f, g, h) IP-GTT curves at baseline, preoperative, 2 weeks postoperative and 8 weeks postoperative of the 8-week study. (a–b) and (e–f) show no differences between the groups at the baseline and the preoperative timepoint. (c) Glucose tolerance improved after SG 2 weeks after surgery, which is depicted as a decrease in IP-GTT AUC compared to sham operated animals. (d) Regarding the long-term experiment, 8 weeks after surgery the IP-GTT AUC is still decreased after SG compared to control and Pf, but without being statistically significant. (i) cholesterol levels before termination of the experiment (2 and 8 weeks after surgery). Cholesterol levels are decreased in the SG groups 2- and 8-weeks following surgery. (j) Triglyceride levels do not differ between groups at both time points. ∗p < 0:05, ∗∗p < 0:01 and ∗∗∗p < 0:001.

Fig. 4

Liver injury. (a) NAFLD Activity Score (NAS) displayed the severity of MASH. (b–d) shows the different components (steatosis, inflammation, ballooning) of the NAS. (e) Fibrosis score according to Kleiner et al. (f) Immunohistological results of CD45⁺ and (g) F4/80⁺ cells, (h–j) Flow cytometry was used to further identify the immune cell types. (h) The proportion of CD11b⁺F4/80⁺ macrophages were significantly higher in SG mice compared to control and Pf mice 2 weeks after surgery. (i) T-cells were and (j) CD3⁺NK1.1⁺ cells were significantly decreased in SG mice compared to control and Pf mice 2 weeks after surgery. These effects were no longer detectable in the long-term study. (k) AST and (h) ALT levels in serum before termination of the studies. (m) Liver weight measured at explantation. Data represent mean ± SD; ∗p < 0:05, ∗∗p < 0:01 and ∗∗∗p < 0:001.

Fig. 5

Proteomic analysis of the liver. (a–c) Volcano plots of proteins upregulated in (a) sleeve compared to control in the 2 weeks experiment, in (b) sleeve compared to control in the 8 weeks experiment and in (c) 2 weeks sleeves compared to 8 weeks sleeves. (d) Predicted regulation of transcription factor activities in the comparison of 2 weeks sleeve vs control using the decoupleR package (see supplementary text file for details). (e) PPI network of proteins upregulated in sleeve livers compared to control livers obtained from the 2-week experiment. Proteins involved in the stress/inflammation (including elevated neutrophil involvement, such as neutrophil infiltration)-associated pathways (using the (red color; GO0006950). Proteins associated with increased mortality or altered cellular aging patterns (blue color; MP 0010768 “Mortality/aging”). (a–b) minimum 2-fold regulation, p-value max. 0.05; (c) 1.5-fold level changes. PPI networks were generated using string-db.org. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 6

Proteomic analysis of the liver [2]. PPI network of proteins upregulated in sleeve livers of the 2-week experiment compared to sleeve livers of the 8-week experiment. Proteins involved in the stress/inflammation (including elevated neutrophil involvement, such as neutrophil infiltration)-associated pathways (using the (red color; GO0006950). Proteins associated with increased mortality or altered cellular aging patterns (blue color; MP 0010768 “Mortality/aging”). PPI networks were generated using string-db.org. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

figs1

Figure S1 Proteomic analysis of the liver (a) Principal component analysis plot indicates clustering of the individual biological replicates from liver tissue samples obtained from mice that underwent sleeve-gastrectomy. The two time points are separated as well. (b) PPI network of proteins upregulated in control livers compared to sleeve livers of the 2-week experiment. Proteins involved in lipid biochemistry appear significant (red color). Minimum 2-fold regulation, p-value max. 0.05

figs2

Figure S2. Proteomic analysis of the liver (2) PPI network of proteins upregulated in sleeve livers compared to control livers obtained from the 8-week experiment. Proteins involved in stress/inflammation (red) as well as altered mortality/aging associated processes appear significant. Minimum 2-fold regulation, p-value max. 0.05