Phosphatidylcholines as regulators of glucose and lipid homeostasis: Promises and potential risks

Authors

  • Simon Hohenester M.D.,

    1. Tytgat Institute for Liver and Intestinal Research Department of Gastroenterology & Hepatology University of Amsterdam Amsterdam, The Netherlands
    Search for more papers by this author
  • Ulrich Beuers M.D.

    1. Tytgat Institute for Liver and Intestinal Research Department of Gastroenterology & Hepatology University of Amsterdam Amsterdam, The Netherlands
    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

Abstract

Nuclear hormone receptors regulate diverse metabolic pathways and the orphan nuclear receptor LRH-1 (also known as NR5A2) regulates bile acid biosynthesis. Structural studies have identified phospholipids as potential LRH-1 ligands, but their functional relevance is unclear. Here we show that an unusual phosphatidyl-choline species with two saturated 12 carbon fatty acid acyl side chains (dilauroyl phosphatidylcholine (DLPC)) is an LRH-1 agonist ligand in vitro. DLPC treatment induces bile acid biosynthetic enzymes in mouse liver, increases bile acid levels, and lowers hepatic triglycerides and serum glucose. DLPC treatment also decreases hepatic steatosis and improves glucose homeostasis in two mouse models of insulin resistance. Both the antidiabetic and lipotropic effects are lost in liver-specific Lrh-1 knockouts. These findings identify an LRH-1 dependent phosphatidylcholine signalling pathway that regulates bile acid metabolism and glucose homeostasis.

Lee JM, Lee YK, Manrosh JL, Busby SA, Griffin PR, Pathak MC, et al. A nuclear receptor-dependent phosphatidylcholine pathway with antidiabetic effects. Nature 2011;474:506-510. (Reprint permission.)

Abstract

Nuclear hormone receptors regulate diverse metabolic pathways and the orphan nuclear receptor LRH-1 (also known as NR5A2) regulates bile acid biosynthesis. Structural studies have identified phospholipids as potential LRH-1 ligands, but their functional relevance is unclear. Here we show that an unusual phosphatidyl-choline species with two saturated 12 carbon fatty acid acyl side chains (dilauroyl phosphatidylcholine (DLPC)) is an LRH-1 agonist ligand in vitro. DLPC treatment induces bile acid biosynthetic enzymes in mouse liver, increases bile acid levels, and lowers hepatic triglycerides and serum glucose. DLPC treatment also decreases hepatic steatosis and improves glucose homeostasis in two mouse models of insulin resistance. Both the antidiabetic and lipotropic effects are lost in liver-specific Lrh-1 knockouts. These findings identify an LRH-1 dependent phosphatidylcholine signalling pathway that regulates bile acid metabolism and glucose homeostasis.

Comment

The orphan nuclear receptor liver receptor homolog-1 (LRH-1, NR5A2) is regarded as a central regulator of bile salt biosynthesis and bile salts are increasingly recognized as modulators of glucose and lipid metabolism in mice and men. In their remarkable study, Lee et al.1 identified a ligand for LRH-1, dilauroyl phosphatidylcholine (DLPC), a C12:0/C12:0 phospholipid, which had potent effects on glucose, lipid, and bile salt homeostasis in vivo.

In a cell-free system, Lee et al. demonstrated by mass spectrometry that DLPC specifically binds to a recombinant LRH-1 ligand-binding domain. Agonism for LRH-1 could be confirmed in an elegant mammalian two-hybrid assay for DLPC and its sister-molecule diundecanoyl phosphatidylcholine (DUPC; C11:0/C11:0).

On functional level, DLPC and even more DUPC were strong activators of both human and mouse LRH-1, whereas other nuclear receptors including FXR, CAR, PXR, PPARα and PPARγ were all unaffected in cell culture. DLPC and DUPC induced transactivation of the native mouse Shp and Oct4 promoters, in line with previous studies on Lrh-1.2, 3 In the human hepatoma cell line HepG2, DLPC induced the expression of CYP8B1.

When orally applied to wildtype mice, DLPC and DUPC induced the expression of hepatic Cyp7a1, Cyp8b1, and Sr-b1 but repressed Shp, leading to a modest increase in serum bile salts and total bile salt pool. These findings were consistent with previous observations in liver-specific Lrh-1 knockouts.4 More strikingly, DUPC- and DLPC-treated mice showed significantly decreased serum glucose, serum nonesterified fatty acids (NEFAs), and hepatic triglycerides. The effects of DLPC were lost in LRH-1 floxed (Lrh-1f/f) mice after administration of adenoviral Cre (Ad-Cre) vector, deleting LRH-1. Comparative oral administration of cholate (100 mg/kg body weight twice daily) improved serum NEFAs and hepatic triglycerides to a similar degree, but did not affect serum glucose.

The surprising effects of DLPC on glucose metabolism were further investigated in a diabetic model, utilizing insulin-resistant leptin receptor deficient db/db mice. DLPC improved glucose homeostasis, as assessed by serum insulin, glucose tolerance test (GTT), and insulin tolerance test (ITT). Furthermore, DLPC improved serum cholesterol, hepatic triglycerides, and serum NEFAs, possibly due to the observed repression of the hepatic lipogenic gene Srebp-1c and downstream targets.

These intriguing results were confirmed in a diet-induced obesity and insulin resistance model. Liver-specific knockout of Lrh-1 had no effect on development of obesity and diabetes when a high-fat diet was applied over 15 weeks. However, DLPC treatment substantially improved glucose homeostasis, and decreased hepatic glucose production, and serum glucose and insulin levels. Improved hepatic insulin sensitivity may have been caused by increased insulin-dependent phosphorylation of the insulin receptor IRS2. Despite unchanged food intake, total body weight, and liver weight, hepatic triglycerides and NEFAs were reduced following DLPC administration and mouse livers showed reduced steatosis. Mechanistically, DLPC markedly decreased expression of genes associated with de novo lipogenesis, especially the lipogenic transcription factor Srebp-1c and its key downstream targets Acc-2, Scd-1, and Fasn. However, no effects of DLPC were observed on hepatic expression of a number of genes controlling glucose homeostasis. It is noteworthy that both serum and hepatic bile salts nearly doubled following DLPC treatment, alongside an induction of Cyp7a1 and Cyp8b1 in the liver. All reported effects of DLPC were absent when LRH-1 was conditionally deleted in the liver.

DLPC thus proved to be of potential therapeutic benefit in both genetic and diet-induced models of insulin resistance. Encouraged by these results, Lee et al. suggest that DLPC might be a promising therapeutic agent for the treatment of metabolic disorders. Consequently, the group has initiated a clinical trial to explore the effect of DLPC in prediabetic patients.

How does DLPC improve insulin sensitivity and reduce steatosis? The authors reason that the beneficial effect of DLPC on steatosis might be a result of the markedly decreased expression of Srebp-1c and/or decreased insulin levels. They propose the following regulatory loop (Fig. 1): Lrh-1-dependent repression of Srebp-1c expression may improve steatosis, increase insulin sensitivity, and hence decrease serum insulin; and decreased insulin levels in turn may reinforce repression of Srebp-1c,5 further ameliorating steatosis. This model is supported by previous data from other groups: the repression of Srebp-1c by way of Lrh-1 is consistent with a functional antagonism of SREBP-1c transactivation by LRH-16 and Srebp-1c target genes Acc-2 and Scd-1 have been shown to modulate β-oxidation, hepatic steatosis, and insulin resistance.1

Figure 1.

Positive regulatory loop improving NAFLD and diabetes following DLPC administration in mice (modified1) DLPC-induced, Lrh-1-dependent repression of Srebp-1c expression improves hepatic steatosis, increases insulin sensitivity, and hence decreases serum insulin. Decreased insulin levels in turn reinforce repression of Srebp-1c. In parallel, DLPC-induced dramatic increase in hepatic and systemic bile salt levels may have additional beneficial effects, mediated by bile salt sensors FXR and TGR5.

Do bile salts contribute to the antidiabetic and antisteatotic effects of DLPC? Upon DLPC treatment, bile salts in serum and more strikingly in liver tissue were markedly increased. This is remarkable because hepatic bile salt levels are tightly controlled. Feeding a 1% cholate (w/w) diet only induces an increase of hepatic bile salts by approximately 50% in mice.7 The striking doubling of hepatic bile salt levels in response to DLPC thus calls for mechanistic explanations. The nuclear bile salt receptor FXR has been shown to protect against insulin resistance8 and fatty liver8, 9: antidiabetic effects were mechanistically linked to repressed Pepck8 and increased hepatic IRS-2 phosphorylation. Both mechanisms were also reported in DLPC-treated mice in the current study.1 Antisteatotic effects in the presented study might thus rely on bile salt-/FXR-mediated repression of Srebp-1.9 The membrane bile salt receptor TGR5 improves glucose homeostasis by release of GLP-110 and by increasing energy expenditure in brown adipose tissue.10, 11 Concertedly, these bile salt sensors might thus mediate antidiabetic and antisteatotic effects as a result of DLPC-/LRH-1-induced stimulation of bile salt synthesis.

To further explore its molecular mechanisms and the possible contribution of bile salt receptors to its antidiabetic and antisteatotic effects, studies of DLPC should be expanded to mouse models that lack Tgr5 or Fxr expression.

As DLPC is now being administered in a clinical trial, potential risks should be considered that might be associated with DLPC treatment in men:

  • The hydrophilic, nontoxic bile salt pool in mice differs markedly from the more hydrophobic, potentially toxic bile salt pool in humans. Human hydrophobic bile salts are potent inducers of hepatocellular apoptosis.12, 13 The DLPC-induced increase in hepatic bile salt levels could thus result in a potential risk in men. As hepatocellular steatosis increases bile salt toxicity,14 patients with steatosis and steatohepatitis might be at increased risk to develop bile salt-induced liver injury following DLPC administration. It is a limitation of the present study that the effect of DLPC on biochemical markers of liver injury was not assessed in mouse models of steatosis.

  • DLPC induced expression of Oct4 in vitro in the present study. OCT4 has been implicated in tumorigenesis15 and was reported to be predictive of poor survival in HCC.16 Therefore, potential procarcinogenic effects of DLPC should be considered in further in vivo studies.

In summary, the identification of DLPC as an antidiabetic and antisteatotic ligand of Lrh-1 in mice is highly intriguing and might prove to be a long-sought new therapeutic tool in metabolic disease in men. However, mice are not men, and careful monitoring of patients for DLPC-induced hepatic and extrahepatic side effects is warranted.

Ancillary