Amelioration of non-alcoholic fatty liver disease by targeting adhesion G protein-coupled receptor F1 (Adgrf1)

  1. Mengyao Wu
  2. Tak-Ho Lo
  3. Liping Li
  4. Jia Sun
  5. Chujun Deng
  6. Ka-Ying Chan
  7. Xiang Li
  8. Steve Ting-Yuan Yeh
  9. Jimmy Tsz Hang Lee
  10. Pauline Po Yee Lui
  11. Aimin Xu
  12. Chi-Ming Wong

  • Curated by eLife

    eLife logo

    eLife assessment

    These valuable findings presented by Wu et al. advance our understanding in novel cell signaling regulators of hepatic metabolism. The evidence supporting these conclusions are solid, utilizing in vivo and in vitro gain and loss of function studies. These work will be of interest to biologists working in the field of hepatic steatosis.

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Recent research has shown that the adhesion G protein-coupled receptor F1 ( Adgrf1; also known as GPR110; PGR19; KPG_012; hGPCR36 ) is an oncogene. The evidence is mainly based on high expression of Adgrf1 in numerous cancer types, and knockdown Adgrf1 can reduce the cell migration, invasion, and proliferation. Adgrf1 is, however, mostly expressed in the liver of healthy individuals. The function of Adgrf1 in liver has not been revealed. Interestingly, expression level of hepatic Adgrf1 is dramatically decreased in obese subjects. Here, the research examined whether Adgrf1 has a role in liver metabolism.

Methods:

We used recombinant adeno-associated virus-mediated gene delivery system, and antisense oligonucleotide was used to manipulate the hepatic Adgrf1 expression level in diet-induced obese mice to investigate the role of Adgrf1 in hepatic steatosis. The clinical relevance was examined using transcriptome profiling and archived biopsy specimens of liver tissues from non-alcoholic fatty liver disease (NAFLD) patients with different degree of fatty liver.

Results:

The expression of Adgrf1 in the liver was directly correlated to fat content in the livers of both obese mice and NAFLD patients. Stearoyl-coA desaturase 1 ( Scd1 ), a crucial enzyme in hepatic de novo lipogenesis, was identified as a downstream target of Adgrf1 by RNA-sequencing analysis. Treatment with the liver-specific Scd1 inhibitor MK8245 and specific shRNAs against Scd1 in primary hepatocytes improved the hepatic steatosis of Adgrf1 -overexpressing mice and lipid profile of hepatocytes, respectively.

Conclusions:

These results indicate Adgrf1 regulates hepatic lipid metabolism through controlling the expression of Scd1 . Downregulation of Adgrf1 expression can potentially serve as a protective mechanism to stop the overaccumulation of fat in the liver in obese subjects. Overall, the above findings not only reveal a new mechanism regulating the progression of NAFLD, but also proposed a novel therapeutic approach to combat NAFLD by targeting Adgrf1 .

Funding:

This work was supported by the National Natural Science Foundation of China (81870586), Area of Excellence (AoE/M-707/18), and General Research Fund (15101520) to CMW, and the National Natural Science Foundation of China (82270941, 81974117) to SJ.

Article activity feed

  1. Version published to 10.7554/elife.85131 on eLife
  2. eLife

    Author Response

    Reviewer #2 (Public Review):

    In this manuscript, the authors have proposed that the suppression of hepatic GPR110, known as a tumorigenic gene, could improve non-alcoholic fatty liver disease (NALFD). With AAV-mediated GPR110 overexpression or a GalNAc-siGPR110 experiment, they have suggested that GPR110 could increase hepatic lipids through SCD1.

    Major comments

    1. Although the authors claimed that GPR110 could enhance SCD1-mediated hepatic de novo lipogenesis, the level of GPR110 expression was decreased in obese mice (Figure 1E-F). However, it has been reported that the levels of de novo lipogenic genes, including SCD1, are upregulated in HFDfed mice (PMID: 18249166, PMID: 31676768). Thus, they should show the levels of hepatic lipids and lipogenic gene expression, including SCD-1, in liver tissues from NCD vs. HFD-fed mice, which will provide insights between GPR110 level and hepatic lipogenic activity.

    Thank you for the comment. The levels of hepatic lipids and lipogenic gene expression, including SCD-1, in liver tissues from NCD vs. HFD-fed mice are summarized in Supplementary Table 4 on page 63. Additionally, we measured the de novo lipogenic activity of primary hepatocytes with varying levels of GPR110 using stable isotopes 3H-acetate. The data are presented in Figure 5D on page 36 of the revised manuscript. These findings suggest that the HFD diet may affect hepatic lipid metabolism through changes in gene expression and lipid accumulation.

    1. In Figure 2, the authors have characterized metabolic phenotypes of hepatic GPR110 overexpression upon HFD, exhibiting significant phenotypes (including GTT, ITT, HOMA-IR, serum lipids, and hepatic lipid level). However, it is likely that these phenotypes could stem from increased body weight gain. Since they cannot explain how hepatic GPR110 overexpression could increase body weight, it is hard to conclude that the increased hepatic lipid level would be a direct consequence of GPR110 overexpression. Also, given the increased fat mass in GPR110 overexpressed mice, they should test whether GPR110 overexpression would affect adipose tissue. Along the same line, they have to carefully investigate the reason of increased body weight gain in GPR110 overexpressed mice (ex., food intake, and energy expenditure).

    Thank you for the comment. Firstly, we checked the expression of GPR110 in the adipose tissues of rAAV-GPR110 mice. We did not observe any change in the mRNA expression level of GPR110 in adipose tissues including SWAT, EWAT and BAT as compared to their controls (Supplementary Figure 3A on page 50). All the Ct levels for adipose GPR110 mRNA were over 40. As suggested, we use metabolic cage system to explore whether the metabolic phenotypic differences between rAAV-GFP and rAAV-GPR110 mice were due to other factors. However, we did not observe any difference in the locomotion, distance in cage locomotion, energy expenditure, daily food intake, daily water intake and respiratory exchange ratio remained similar in these two groups as shown in Supplementary Figure 3.B-G on page 50. Therefore, they shall not be the root cause of the reason of increased body weight gain in GPR110 overexpressed mice.

    1. GPR110 enhances hepatic lipogenesis via SCD1 expression (Figures 5 and 6). To verify whether GPR110 would specifically regulates SCD1 transcript, they have to provide the expression levels of other lipogenic genes, including Srebf1, Chrebp, Acaca, and Fasn.

    Thank you for the comment. As suggested, we added the expression levels of these lipogenic genes in Figure 5B-C on page 36 of the revised manuscript. In addition, we also measured the de novo lipogenic activity using primary hepatocytes with either overexpressing or knockdown of GPR110 to confirm that GPR110 enhances hepatic lipogenesis.

    1. In Figure 6, the author should provide the molecular mechanisms how GPR110 signaling could enhance SCD-1 transcription.

    Thank you for the comment. SREBP1 is a key transcription factor that regulates the expression levels of the SCD1 gene [21]. A study published in March (at the time of revising this manuscript) showed that GPR110 plays a role in mediating the activation of SREBP1 pathways by palmitic acid. This ultimately promotes the synthesis of fats in mammary gland tissues [10]. In our RNA sequencing analysis, we also found that the expression of hepatic SREBP1 was correlated with the expression of GPR110. To further investigate this relationship, we added the mRNA levels of SREBP1 in our experiments, as shown in Figure 5B-C on page 36 of the revised manuscript. Specifically, we found that the expression level of SREBP1 was increased in the GPR110 overexpression group and decreased after using ASOs to knock down hepatic GPR110 levels. These findings suggest that GPR110 regulates hepatic lipid metabolism through the SREBP1-SCD1 pathway.

    1. Figure 9C shows the increased level of GPR110 with NAFLD severity. They should test whether the levels of hepatic GPR110 and SCD-1 might be elevated in a severe NAFLD mouse model. If it is the case, it would be better to show the beneficial effects of GPR110 suppression against NAFLD progression using a severe NAFLD (ex., NASH) mouse model.

    Thank you for the comment. To further explore the expression pattern of GPR110 in a more severe NAFLD mouse model, we injected either CCl4 or STZ to induce NAFLD severity in HFD-fed mice. We found that after treating with CCl4 or STZ, the expression levels of GPR110 and SCD1 mRNAs were significantly increased compared to the control group without treatment with CCl4 or STZ (please see Figure 9F-G). We attempted to knock down the expression of hepatic GPR110 in the CCl4 or STZtreated HFD-fed mice. However, our ASOs were only effective in knocking down high levels of GPR110 mRNA in the virus mediated GPR110 expression systems (please see Figure 5 and 6). The expression level of hepatic GPR110 mRNA in HFD-fed mice after CCl4 or STZ treatment was too low to be effectively knocked down by ASOs. However, a previous study demonstrated that Gpr110-/- mice were resistant to liver tumorigenesis induced by DEN plus CCl4 injection [22]. We believe that GPR110 suppression also can prevent the progression of NAFLD in these severe NAFLD mouse models.

    Reviewer #3 (Public Review):

    In this study, the authors examined the expression of GPR110 in a HFD-fed mouse model and validated their findings in human samples. They then performed both gain- and loss-of-function studies on the cellular and systemic metabolic effects of manipulating the levels of GPR110. They further demonstrated that SCD-1 was a downstream effector of GPR110, and the effects of GPR110 could be mediated by SCD-1. This study provides a novel target in NAFLD. Overall, the data and analyses well performed and convincing. As the GPR110-SCD1-lipid metabolic phenotype axis is a central theme of the study, I would suggest that the authors further discuss the connection between GPR110 and SCD1, especially the persistent upregulation of SCD1 at late stage of HFD-fed mice (obese mouse model) when GPR110 is very low, for example, whether another regulator plays a more relevant role at this time point.

    Thank you for the comment. As SCD1 is the rate limiting enzyme catalysing the biosynthesis of monounsaturated fatty acids, a very tight and complex regulation of SCD1 gene expression in response to various parameters including hormonal and nutrient factors is reported [23]. HFD treatment itself can induce the expression of hepatic SCD1 [21, 23, 24], and our study demonstrated that the expression of SCD1 can be further increased by overexpressing GPR110 in the liver of HFDfed mice (Fig. 9F and G on page 44) that will contribute to the acceleration and aggravation of NAFLD. The discussion of the connection between GPR110 and SCD1was presented on page 21, lines 455-464.

  3. eLife

    eLife assessment

    These valuable findings presented by Wu et al. advance our understanding in novel cell signaling regulators of hepatic metabolism. The evidence supporting these conclusions are solid, utilizing in vivo and in vitro gain and loss of function studies. These work will be of interest to biologists working in the field of hepatic steatosis.

  4. eLife

    Reviewer #1 (Public Review):

    Wu et al. sought to investigate the biological role of GPR110 in modulating hepatic lipid metabolism. The authors demonstrate a pathological role of GPR110 in promoting hepatic steatosis and generalized metabolic syndrome in a mouse model of diet-induced obesity. Furthermore, the authors identify enhanced SCD1 expression as an underlying mechanism promoting GPR110-induced metabolic dysfunction. Finally, the authors provide clinically relevant human data demonstrating a positive correlation between GPR110 expression and degree of hepatic steatosis. The strengths of this study include the rigorous design and execution of experiments, the utilization of gain and loss of function as well as pharmacological and genetic approaches, and the clinically relevant human data presented. The claims are supported by robust data. These findings have the potential to impact the field of metabolism in general, given their findings indicate targeting GPR110 can reverse diet induced obesity and metabolic syndrome. Only minor weaknesses were noted in regard to further interpretation of the data.

  5. eLife

    Reviewer #2 (Public Review):

    In this manuscript, the authors have proposed that the suppression of hepatic GPR110, known as a tumorigenic gene, could improve non-alcoholic fatty liver disease (NALFD). With AAV-mediated GPR110 overexpression or a GalNAc-siGPR110 experiment, they have suggested that GPR110 could increase hepatic lipids through SCD1.

    Major comments
    1. Although the authors claimed that GPR110 could enhance SCD1-mediated hepatic de novo lipogenesis, the level of GPR110 expression was decreased in obese mice (Figure 1E-F). However, it has been reported that the levels of de novo lipogenic genes, including SCD1, are upregulated in HFD-fed mice (PMID: 18249166, PMID: 31676768). Thus, they should show the levels of hepatic lipids and lipogenic gene expression, including SCD-1, in liver tissues from NCD vs. HFD-fed mice, which will provide insights between GPR110 level and hepatic lipogenic activity.

    2. In Figure 2, the authors have characterized metabolic phenotypes of hepatic GPR110 overexpression upon HFD, exhibiting significant phenotypes (including GTT, ITT, HOMA-IR, serum lipids, and hepatic lipid level). However, it is likely that these phenotypes could stem from increased body weight gain. Since they cannot explain how hepatic GPR110 overexpression could increase body weight, it is hard to conclude that the increased hepatic lipid level would be a direct consequence of GPR110 overexpression. Also, given the increased fat mass in GPR110 overexpressed mice, they should test whether GPR110 overexpression would affect adipose tissue. Along the same line, they have to carefully investigate the reason of increased body weight gain in GPR110 overexpressed mice (ex., food intake, and energy expenditure).

    3. GPR110 enhances hepatic lipogenesis via SCD1 expression (Figures 5 and 6). To verify whether GPR110 would specifically regulates SCD1 transcript, they have to provide the expression levels of other lipogenic genes, including Srebf1, Chrebp, Acaca, and Fasn. Also, measurement of de novo lipogenic activity using primary hepatocyte with GPR110 overexpression or knockdown would be valuable to affirm the authors' proposed model.

    4. In Figure 6, the author should provide the molecular mechanisms how GPR110 signaling could enhance SCD-1 transcription.

    5. Figure 9C shows the increased level of GPR110 with NAFLD severity. They should test whether the levels of hepatic GPR110 and SCD-1 might be elevated in a severe NAFLD mouse model. If it is the case, it would be better to show the beneficial effects of GPR110 suppression against NAFLD progression using a severe NAFLD (ex., NASH) mouse model.

  6. eLife

    Reviewer #3 (Public Review):

    In this study, the authors examined the expression of GPR110 in a HFD-fed mouse model and validated their findings in human samples. They then performed both gain- and loss-of-function studies on the cellular and systemic metabolic effects of manipulating the levels of GPR110. They further demonstrated that SCD-1 was a downstream effector of GPR110, and the effects of GPR110 could be mediated by SCD-1. This study provides a novel target in NAFLD. Overall the data and analyses well performed and convincing. As the GPR110-SCD1-lipid metabolic phenotype axis is a central theme of the study, I would suggest that the authors further discuss the connection between GPR110 and SCD1, especially the persistent upregulation of SCD1 at late stage of HFD-fed mice (obese mouse model) when GPR110 is very low, for example, whether another regulator plays a more relevant role at this time point.

  7. Version published to 10.1101/2022.12.07.22283206v1 on medRxiv