Vertical sleeve gastrectomy reduces hepatic steatosis while increasing serum bile acids in a weight-loss-independent manner

Abstract

Objective: Our objective was to investigate the role of bile acids in hepatic steatosis reduction after vertical sleeve gastrectomy (VSG).

Design and methods: High fat diet (HFD)-induced obese C57Bl/6 mice were randomized to VSG, Sham operation (Sham), Sham operation with pair feeding to VSG (Sham-PF), or nonsurgical controls (Naïve). All mice were on HFD until sacrifice. Mice were observed postsurgery and data for body weight, body composition, metabolic parameters, serum bile acid level and composition were collected. Further hepatic gene expression by mRNA-seq and RT-PCR analysis was assessed.

Results: VSG and Sham-PF mice lost equal weight postsurgery while VSG mice had the lowest hepatic triglyceride content at sacrifice. The VSG mice had elevated serum bile acid levels that positively correlated with maximal weight loss. Serum bile composition in the VSG group had increased cholic and tauroursodeoxycholic acid. These bile acid composition changes in VSG mice explained observed downregulation of hepatic lipogenic and bile acid synthetic genes.

Conclusion: VSG in obese mice results in greater hepatic steatosis reduction than seen with caloric restriction alone. VSG surgery increases serum bile acids that correlate with weight lost postsurgery and changes serum bile composition that could explain suppression of hepatic genes responsible for lipogenesis.

Figure 1. A. Illustration of completed vertical sleeve gastrectomy

B. Absolute body weight and body weight change over the 60 days post-surgery period. The difference in body weight loss between VSG and Sham mice was observed as early as 4 days post-surgery with more profound loss in the VSG group. By the end of the experiment Sham mice gained much more weight compared to VSG mice. N: Sham=9, VSG=9. [**=p<0.01, two-way ANOVA]. C. Cumulative food intake and intestinal fat absorption. Except first few days post-surgery no significant difference was observed in cumulative food intake between VSG and Sham mice. N: Sham=9, VSG=9. [Two-way ANOVA]. Intestinal fat absorption was not different between two groups. N: Sham=9, VSG=9. [t-test]. D. Body fat mass, fat ratio and lean mass 49 days post-surgery. Sham mice contained much more body fat compared to VSG mice. However, no difference was observed in lean mass between VSG and Sham groups. N: Sham=9, VSG=9. [*=p<0.05, t-test]. E. Plasma active ghrelin levels 14 days post-surgery. Sham mice had much higher plasma active ghrelin levels than mice after VSG. N: Sham=8, VSG=8. [**=p<0.01, t-test]. F. Hematoxylin & eosin staining of the liver and hepatic triglyceride content at day 60 post-surgery.
Representative 20x magnification liver sections show no pathological changes in the liver of mouse which underwent VSG, while multiple cytoplasmic lipid droplet accumulation is visible in case of Sham mouse liver. VSG mice showed significantly lower hepatic triglyceride levels compared to Sham mice. N: Sham=9, VSG=9. [***=p<0.0001, t-test].

Figure 2. A. Absolute body weight and body weight change over the 60 days post-surgery period in four groups

VSG mice lost the most body weight compared to HFD Naïve, Sham and Sham-PF mice at day 7 post-surgery. However, no significant difference in body weight gain was observed between VSG and Sham-PF groups at day 60 post-surgery. Both these groups gained much less weight compared to HFD Naïve and Sham groups by the end of the study and were significantly lighter. N: HFD Naïve=6, Sham=7, Sham-PF=6, VSG=6. [#=p<0.01 VSG vs. HFD Naïve, Sham and Sham-PF, two-way ANOVA; **=p<0.01 VSG and Sham-PF vs. HFD Naïve and Sham, two-way ANOVA]. B. Average daily food intake per week in HFD Naïve, Sham and VSG groups. Except for the first week post-surgery no significant difference was observed in food intake between HFD Naïve, Sham and VSG mice. N: HFD Naïve=6, Sham=7, VSG=6. [**=p<0.01, two-way ANOVA]. C. Fasting serum glucose and insulin levels, and HOMA-IR 56 days post-surgery. Serum glucose levels were lower in the VSG group compared to HFD Naïve and Sham groups, and serum insulin levels and HOMA-IR were the lowest in the VSG group compared to other three groups. N: HFD Naïve=6, Sham=7, Sham-PF=6, VSG=6. [*=p<0.05, **=p<0.01, ***=p<0.0001, one-way ANOVA and t-test]. D. Serum cholesterol levels at day 14 post-surgery. Fasting serum cholesterol levels were significantly lower in VSG compared to HFD Naïve and Sham-PF groups. N: HFD Naïve=3, Sham=4, Sham-PF=6, VSG=6. [*=p<0.05, **=p<0.01, one-way ANOVA and t-test]. E. Serum ALT levels at day 14 post-surgery. ALT levels were the lowest in the VSG group compared to HFD Naïve, Sham and Sham-PF groups. N: HFD Naïve=3, Sham=4, Sham-PF=6, VSG=6. [*=p<0.05, **=p<0.01, one-way ANOVA, t-test]. F. Liver histology and hepatic triglyceride content 60 days post-surgery in four groups. Representative 20x magnification Oil Red O stained frozen sections show very slight neutral lipid accumulation in the livers of VSG mice compared to the mice of other groups, pointing that liver steatosis was markedly reduced in the VSG group. Liver triglyceride levels were the lowest in the VSG group compared to HFD Naïve, Sham and Sham-PF groups. N: HFD Naïve=6, Sham=7, Sham-PF=6, VSG=6. [*=p<0.05, **=p<0.01, ***=p<0.001, one-way ANOVA].

Figure 3. A. Fasting serum total bile acid levels pre-surgery, at day 14, 28 and 42 post-surgery

Fasting serum bile acid levels were the highest in VSG group at day 14 and 28 post-surgery compared to all other groups, while there was no difference in bile acid levels between the groups before surgery and at day 42 post-surgery. N: HFD Naïve=6, Sham=7, Sham-PF=6, VSG=6. [*=p<0.05, ***=p<0.001, one-way ANOVA]. B. Serum total bilirubin levels. There was no difference in serum total bilirubin levels between four groups at day 60 post-surgery. N: HFD Naïve=6, Sham=7, Sham-PF=6, VSG=6. C. Fasting serum total bile acid levels at day 28 post-surgery. Fasting serum bile acid levels were the highest in the VSG group compared to all other groups at 28 days post-surgery. N: HFD Naïve=6, Sham=7, Sham-PF=6, VSG=6. [**=p<0.01, ***=p<0.0001, one-way ANOVA]. D. Postprandial serum total bile acid levels at day 60 post-surgery. Postprandial serum bile acid levels were higher in VSG mice compared to Sham and Sham-PF mice at time of sacrifice. N: HFD Naïve=6, Sham=7, Sham-PF=6, VSG=6. [*=p<0.05, one-way ANOVA, t-test]. E. Serum bile acid levels correlation with maximal body weight loss after VSG. A strong correlation was seen between fasting total serum bile acid levels at day 14, postprandial total serum bile acid levels at day 60 after VSG with maximal weight loss post-surgery. N=6. [R²=0.83, p=0.021, and R²=0.86, p=0.008, respectively. Pearson correlation analysis].

Figure 4. A. Serum bile acid composition analysis

Bile acid composition analysis showed that the majority of bile acids in all groups were taurine conjugated. Proportion of taurine conjugated acids was higher in Sham-PF and VSG groups, while Naïve and Sham groups had higher proportion of muricholic acid. Percentage of cholic acid was the highest in VSG mice. The following bile acids are also indicated: muricholic acid (MCA), cholic acid (CA), deoxycholate (DCA), chenodeoxycholate (CDCA), ursodeoxycholate (UDCA) and glycine conjugated deoxycholate (GCDCA). N: HFD Naïve=6, Sham=7, Sham-PF=6, VSG=6. B. Cholic and tauroursodeoxycholic acid levels at day 60 post-surgery. Postprandial serum cholic acid levels were higher in VSG compared to Sham-PF mice, while levels of tauroursodeoxycholic acid were significantly elevated in the VSG compared to all other groups. N: HFD Naïve=6, Sham=7, Sham-PF=6, VSG=6. [*=p<0.05, **=p<0.01, ***=p<0.001, t-test and one-way ANOVA].

Figure 5. A. Next Gen mRNA-sequencing gene expression heat map of Sham-PF and VSG mice at day 60 post-surgery

Multiple genes responsible for bile acid synthesis was downregulated in the livers of VSG mice (blue rows) compared to Sham-PF mice at 60 days post-surgery. N: Sham-PF=5, VSG=5. B. Hepatic bile acid synthesis gene expression at day 60 post-surgery. mRNA levels of the genes coding for the bile acid production (Cyp7a1 and Cyp8b1) were measured by RT-PCR and expressed in relative expression units. These bile acid production rate limiting genes were downregulated in the VSG group compared to HFD Naïve, Sham and Sham-PF groups. HFD Naïve=4, Sham=5, Sham-PF=5, VSG=5. [*=p<0.05, **=p<0.01, ***=p<0.0001, one-way ANOVA, t-test]. C. Hepatocyte bile acid uptake gene expression at day 60 post-surgery. mRNA levels of the genes coding for the bile acid import into the hepatocyte (Ntcp and Oatp4) were measured by RT-PCR and expressed in relative expression units. These hepatocyte bile acid uptake genes were downregulated in the VSG group compared to HFD Naïve, Sham and Sham-PF groups. HFD Naïve=4, Sham=5, Sham-PF=5, VSG=5. [*=p<0.05, one-way ANOVA, t-test]. D. Hepatic lipid metabolism gene expression at day 60 post-surgery. mRNA levels of the genes coding for the lipid metabolism (Fasn and Cpt1a) were measured by RT-PCR and expressed in relative expression units. Fasn mRNA was decreased in VSG mice compared to Sham-PF mice. Cpt1a mRNA was the lowest in the VSG group compared to other groups. N: HFD Naïve=4, Sham=5, Sham-PF=5, VSG=5. [*=p<0.05, t-test].

Figure 6. Scheme of bile acid driven signaling pathway in liver and ileum

Bile acids are reabsorbed in the terminal ileum, affect transcription factor FXR in the enterocyte and as a result FGF15 secretion which through the blood flow is able to suppress bile synthesis and lipid metabolism in the liver. On the other hand, reabsorbed bile acids are transported to the liver by the portal vein blood flow, where their hepatocyte uptake is regulated by the import genes (Ntcp and Oatps). Bile acids’ interaction with FXR effects SHP transcription, which in turn inhibits bile and lipid synthesis in the liver. In mice after VSG serum bile acid levels were elevated, expression of bile acid production genes was decreased as a consequence of the bile acid synthesis negative feedback regulation, and lipid metabolism genes were downregulated in the liver which resulted in the reduction of hepatic steatosis.