Glycyrrhetinic Acid Improves Insulin-Response Pathway by Regulating the Balance between the Ras/MAPK and PI3K/Akt Pathways

Glycyrrhetinic acid (GA), a bioactive component in the human diet, has been reported to improve hyperglycemia, dyslipidemia, insulin resistance and obesity in rats with metabolic syndrome. However, GA-specific target proteins and the mechanisms involved in the downstream signaling and cross-talk to improve insulin sensitivity have not been fully elucidated. In this study, the potential targets of GA were identified by chemical proteomics strategies using serial GA probes for target fishing and cell molecular imaging. Intracellular enzyme activity evaluation and insulin resistance models were used for validating the function of the target proteins on the downstream insulin signaling pathways. Collectively, our data demonstrate that GA improved the insulin-responsive pathway and glucose consumption levels via multiple diabetogenic factors that activated the insulin signaling pathway in HepG2 cells. GA improved Glucose transporter 4(GLUT4) expression by targeting the Ras protein to regulate the mitogen-activated protein kinase (MAPK) pathway. GA exhibited a strong inhibitory effect on IRS1ser307 phosphorylation in cells treated with the Protein kinase C (PKC) activator Phorbol 12-myristate 13-acetate (PMA.) Consistently, IRS1ser307 phosphorylation was also inhibited by GA in Free fatty acid (FFA)-treated HepG2 cells. GA also inhibited the PMA-induced phosphorylation of IκB kinase α/β (IKKα/β), c-Jun N-terminal kinase (JNK) and p38 proteins (P38), suggesting that IKKα/β, JNK and P38 activation is dependent on PKC activity.


Target prediction of GA
To screen the potential target proteins of GA, the top 20 protein targets (determined based on fit value using PharmMapper were software) were analyzed using bioinformatics tools. Next, the interacting proteins were analyzed using String 10.0 (http://www.string-db.org/). We used AutoDock 4.2 software to conduct molecular docking studies to evaluate the interaction targets and GA. At last, we performed capture assays to verification of target proteins.
Reverse docking of GA was performed using the PharmMapper server (http://59.78.96.61/pharmmapper/). We prepared a 3D structure of GA in the sdf format using the ChemBio3D Ultra 13.0 software (PerkinElmer Inc., U.S.A.), and submitted the sdf file to the PharmMapper server with the choice of "human protein targets only" and the maximum generated conformations set to 300. In our study, we employed the 2010 version. The first 20 candidate targets of GA are listed in (Supplementary Table S1). Next, the interacting proteins were analyzed using String 10.0 (http://www.string-db.org/). Three targets in the insulin-related signaling pathways, namely, HRAS, PRKCA (PKCα) and MAP2K1 (MEK1). In addition, two targets in the steroid hormone biosynthesis signaling pathways, HSD11B1 and HSD17B1 The three-dimensional structures of the MAP2K1, PKCα , HRas, HSD11B1 and HSD17B1 proteins (PDB ID code 4U81, 3IW4, 4EFL, 3CZR and 3HB4) were obtained from the Protein Data Bank (http://www.rcsb.org/pdb). The structures of the MAP2K1, PKCα, HRas, HSD11B1 and HSD17B1 proteins were constructed and minimized using the SYBYL software (Chemical Computing Group, Inc.); then, AutoDock version 4.2 (Olson Laboratory, La Jolla, CA) was applied to perform a docking study using a hybrid Lamarckian genetic algorithm (LGA). The number of LGA runs was set to 30. The step size parameters of quaternion and torsion were 30. The binding energies of the GA target proteins HRAS, PKCα, MEK1, HSD11B1 and HSD17B1 were -7.47, -8.77, -9.93, -10.31 and -10.12 kcal/mol kcal/mol, kcal/mol, respectively.

General Chemical Reagents and Methods
All reagents purchased for synthesis. Thin-layer chromatography (TLC) was performed on silica gel GF254 plates with detection using shortwave UV light (λ=254 nm) and staining with 10% phosphomolybdic acid in EtOH, followed by heating on a hotplate. Flash chromatography was performed with silica gel (100-200 mesh) with EtOAc/petroleum ether or CH2Cl2/MeOH as eluent. 1 H and 13 C NMR spectra were recorded on a Bruker AV 400 spectrometer at 400 MHz ( 1 H NMR) and 100 MHz ( 13 C NMR), using CDCl3 as solvents. Coupling constants are displayed in Hertz.

Azide modified-MMs Synthesis (Compound 4)
5 mL NH2-MMs (25 mg, 5 mg/mL) were suspended in 2 mL borate buffer and Sulfo-SADP (compound 3) (0.5 mg, 11 µmol) were added in sequence, the mixture was shocked at room temperature for 12 h. After that, the azide modified-MMs was enriched through magnetic separation and washed with water for three times. The gathered azide modified-MMs (Compound 4) was used directly for next steps.

Probe 1 (Compound 5)
CuBr (10 mg, 0.7 mmol) was dissolved in degassed methanol (3 mL) under argon atmosphere and DIPEA (35 µL, 0.18 mmol) was added. The resulting yellowish suspension was degassed for 30 min under a stream of argon and with exclusion of light. Alkynyl-modified GA probe compound 2 (5 mg, 10 µmol) was dissolved in degassed methanol (0.5 mL) and treated with 1 mL of the freshly prepared suspension of CuBr-DIPEA and azide modified-MMs (25 mg, 5 mg/mL). The reaction mixture was shocked at room temperaturewith exclusion of light for 24 h. Then the GA-modified functionalized MMs were separated with magnet and washed three times each with methanol and water. The gathered GA-modified functionalized MMs (Compound 5) was used directly for next steps.

Compound 6
To a solution of probe1, probe1 (25 mg, 5 mg/mL) in 1 mL methanol, DTT (100 mM) were added. The reaction mixture was shocked at room temperature for 1h. Then the was removed the MMs with magnet and get compound 6 Probe 2 characterization section in ( Figure S5) Figure S5. Synthetic route for click chemistry product probe2 (compound 8). Reagents and conditions: (a) CuSO4, sodium ascorbic acid, DMSO, Alkynyl-GA.