Selective targeting of nuclear receptor FXR by avermectin analogues with therapeutic effects on nonalcoholic fatty liver disease

Abstract

Non-alcoholic fatty liver disease (NAFLD) has become a predictive factor of death from many diseases. Farnesoid X receptor (FXR) is an ideal target for NAFLD drug development due to its crucial roles in lipid metabolism. The aim of this work is to examine the molecular mechanisms and functional roles of FXR modulation by avermectin analogues in regulating metabolic syndromes like NAFLD. We found that among avermectin analogues studied, the analogues that can bind and activate FXR are effective in regulating metabolic parameters tested, including reducing hepatic lipid accumulation, lowering serum cholesterol and glucose levels, and improving insulin sensitivity, in a FXR dependent manner. Mechanistically, the avermectin analogues that interact with FXR exhibited features as partial agonists, with distinctive properties in modulating coregulator recruitment. Structural features critical for avermectin analogues to selectively bind to FXR were also revealed. This study indicated that in addition to antiparasitic activity, avermectin analogues are promising drug candidates to treat metabolism syndrome including NAFLD by directly targeting FXR. Additionally, the structural features that discriminate the selective binding of FXR by avermectin analogues may provide a unique safe approach to design drugs targeting FXR signaling.

Figure 1. Chemical structures of compounds used in the study.

Figure 2. Effects of avermectin analogues on liver fat.

(a) The liver morphology of mice. (b) H&E staining of liver sections (original magnification, ×200). (c) Oil Red O staining of liver sections (original magnification, ×200).

Figure 3. Effects of avermectin analogues on various mice metabolic parameters.

(a) Hepatic triglyceride. (b) Body weight. (c) Liver/body weight ratio. (d) Food intake. (e) Serum triglyceride. (f) Serum cholesterol. (g) Relative mRNA levels of genes related with lipid metabolism by real-time PCR. (h) GTT. (i) ITT. n = 6 per group, Values are the means ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001 versus vehicle.

Figure 4. The potency of avermectin analogues in regulating metabolism correlates with their FXR binding capabilities.

(a) Various coactivator motifs bind to FXR in response to 0.5 μM avermectin analogues or GW4064 by AlphaScreen assay. (b) 1 μM compounds induce FXR activity in reporter assays. (c) Dose responses of compounds in inducing FXR to recruit SRC1-2 coregulator binding motif by AlphaScreen assay. (d) Dose response of compounds in inducing the activity of FXR in reporter assays. (e) The recruitment of NCoR-2 to FXR in response to avermectin analogues by AlphaScreen assay. Values are the means ± SEM of three independent experiments.

Figure 5. Relative binding affinity of various peptide motifs to the FXR LBD in the presence of avermectin analogues as determined by peptide competition assays.

Various unlabeled peptides (20 μM) are used to compete off the binding of the biotin-labeled SRC2-3 LXXLL motif to FXR LBD in response to 1 μM ivermectin, abamectin, doramectin, or 0.5 μM GW4064, respectively. Values are the means ± SEM of three independent experiments. Sequences of peptides used in the AlphaScreen assays are as described previously.

Figure 6. The therapeutic effects of avermectin analogues to reduce lipid accumulation were dependent on FXR.

(a) Oil Red O staining of liver sections (original magnification, ×200). (b) Food intake. (c) Hepatic triglyceride. (d) Serum triglyceride. (e) Serum cholesterol. (f) Serum glucose. WT, Wild type C57BL/6J mice; KO, FXR knockout mice. n = 6 per group, Values are the means ± SEM. *p < 0.05, and ***p < 0.001 versus vehicle.

Figure 7. Protetin-ligand docking of avermectin analogues in the ligand binding pocket of FXR.

(a) A general chemical structure of ivermectin, abamectin and doramectin. (b–d) Left:The structure of FXR bound with ivermectin and the docking structures of the FXR docked with abamectin and doramectin in ribbon representation. Right: Schematic representation of interactions between FXR and avermectin analogues. Hydrogen bonds are indicated by arrows from proton donors to acceptors. The helix 3 of the FXR LBD is highlighted in green, the NCoR-2 motif is in brownish red.

Figure 8. Functional correlation of the avermectins/FXR interactions.

(a–d) Molecular determinants of the interaction between FXR with avermectins. Overlays of avermectin (green) and GW4064 (salmon red) in the FXR structure (grey). The selected residues important for ligand interaction are shown in stick representation with wild-type and mutant depicted in white and blue, respectively. The hydrogen bonds are shown with arrows. The potential hydrophobic interactions, if the corresponding mutations are made as indicated in (e), are shown in dashed lines. (e) Differential effects of mutations of key FXR residues on its transcriptional activity in response to 1 μM avermectin analogues in reporter assays. Values are the means ± SEM of three independent experiments.