Loureirin B activates GLP‐1R and promotes insulin secretion in Ins‐1 cells

Loureirin B (LB) is a natural product derived from Sanguis draconis, which has hypoglycaemic effects. In order to research the possible target of LB in the treatment of diabetes, molecular docking was used to simulate the interaction between LB and potential targets, and among them, glucagon‐like peptide‐1 receptor (GLP‐1R) had the optimal results. Further, spectroscopy and surface plasmon resonance (SPR) experiments were applied to detect the interaction between LB and GLP‐1R. Ultimately, after GLP‐1R siRNA interfering the expression of GLP‐1R in Ins‐1 cell, the promoting insulin secretion of LB was weaken, which directly proved that GLP‐1R plays an important role. These results show that LB promotes insulin secretion of Ins‐1 cells through GLP‐1R. Hence, the strategy of LB as a prodrug will provide a potential approach for non‐peptide GLP‐1R agonist.

emptying, thereby increasing satiety, and suppressing appetite. 15 Therefore, the weight-loss effect of GLP-1R agonists can improve both the quality of life of patients, and the metabolic abnormalities of obese T2DM. 16 However, due to easily degrade by gastric juice and lose their activity in the body, these polypeptide drugs cannot be taken orally. 17,18 Therefore, the study of non-peptide GLP-1R agonists has become a hot topic.
Resina draconis is derived from the resin of Dracaena cochinchinensis, which has a variety of pharmacological effects, such as anti-blood stasis, anti-coagulation and hypoglycaemic. 19,20 Increasing researches are focused on how Resina draconis plays a role in the treatment of diabetes. Zhang et al reported that Resina draconis has good hypoglycaemic effect on rat hyperglycaemia model and increasing insulin secretion. 21 The current research focuses on the hypoglycaemic mechanism of Resina draconis. In Resina draconis, the content of dihydrochalcones has relatively potential with hypoglycaemic effect such as loureirin B (LB). 19 With the continuous determination of biomacromolecules and development of computer science, computer-aided drug design has been widely used in the research and development of compounds. Through molecular docking to mimic the interaction of LB with various insulin secretion-related targets, it was found that GLP-1R may be a potential target for LB.
Therefore, whether GLP-1R is the target of LB was studied. The purpose of this study is to lay a foundation for the structural optimization of LB.

| Material
The recombinant protein GLP-1R was obtained from Creative BioMart. Loureirin B (LB) was purchased from National Institutes for Food and Drug Control. The stock 0.01 mol/L LB solution was prepared with dimethyl sulphoxide (DMSO).

| Molecular modelling and docking analyses
The extracellular domain of GLP-1R was got from the PDB database (4ZGM, http://www.rcsb.org/pdb). The structure of LB was obtained from the PubChem database (https://pubch em.ncbi.nlm.nih.gov).
The simulated interaction between LB and GLP-1R was measured by Discovery Studio 4.1 (DS) (a suite of software for simulating small molecule and macromolecule systems).
The interactions between LB and GLP-1R, DPP-IV, ffar1 and GK were simulated by discovery studio 4.1 molecular docking software.
The specific operation is as follows: (τ 0 represents the average lifetime of fluorescent molecules in the absence of loureirin, taking 10-8s), K q was obtained. 23,24 (1)

| Cell culture
Rat insulinoma Ins-1 cells were purchased from American Type

| Assessment of insulin release and total insulin content
To measure insulin secretion, Ins-1 cells were first seeded onto 24- 11. The concentration of samples and quality control serum was read from the standard curve.

| Real-time PCR expression
The total RNA of Ins-1 cells was extracted by TRIzol method (TRIzol-based method for sample preparation) and measured by spectrophotometer. Complementary DNA was generated using the PrimeScript™ reverse transcript reagent Kit (Takara, Tokyo, Japan).
The relative expression of genes associated with insulin secretion was determined by quantitative RT-PCR (Bio-Rad, Hercules, USA) according to the manufacturer's instruction. The experiment was repeated 3 times. The primer sequences used for PCR were as follows: Pdx1,5′-AAATCCACCAAAGCTCACGC −3′ (forward) and 5′-AAGTTGAGCATCACTGCCAGC-3′(reverse).
1. After adding trizon to the sample, blow several times repeatedly to make the sample fully split. The protein nucleic acid complex was completely separated after being placed at room temperature for 5 minutes. (4) Add chloroform, add 0.2 ml chloroform every 1 ml trizon, cover the tube, shake violently for 15 seconds, and place at room temperature for 3 minutes.
3. After centrifugation at 4°C at 12 000 rpm for 15 minutes, the upper colourless aqueous phase (400 μL in 1 ml trizon) was removed and transferred into a new RNase free centrifuge tube.
4. Add equal volume of isopropanol into the extracted aqueous solution, mix it upside down and keep it at room temperature for 10 minutes.
5. The supernatant was centrifuged at 4°C 12 000 rpm for 10 minutes, and the supernatant was discarded.
6. 75% ethanol (prepared with water without RNase) was added to wash the precipitate. Wash the precipitate with 1 ml 75% ethanol every 1 ml trizon.
7. The supernatant was centrifuged at 4°C 12 000 rpm for 5 minutes, and the supernatant was discarded.
9. Place at room temperature for 3 minutes and dry. 30-100 μL of RNase free water was added to dissolve RNA, and the concentration of extracted RNA was determined. The obtained RNA can be stored in -80°C refrigerator for subsequent experiments.

| Western blot
Protein extracted from Ins-1 cells after treatment was quantitated using Bradford's micro estimation assay. 20 μg samples were separated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE). Polyvinylidene fluoride (PVDF) membrane was utilized to electro-transfer proteins, and then blocked with 5% nonfat dry milk. And the membranes were incubated in different primary antibodies at 4°C overnight, including IRS-2, pAKT(Ser473), FOXO1, H2A and GAPDH (CST). Then, the membrane was incubated with appropriate secondary antibody tagged with the horseradish peroxidase-conjugated anti-rabbit IgG secondary antibodies for 1 h at room temperature. Blots were detected and imaged by chemiluminescence using luminol as a substrate and ChemiDoc Imaging System (Bio-Rad), respectively. The experiment was repeated 3 times.

| SPR analysis
GE Biacore T200 instrument and GE S SA chips are used to com- SPR experiments. 26 The experiment was repeated 3 times.

| siRNA transfection
Ins-1 cells were transferred to 6-well plates and cultured at 37°C in a humidified atmosphere with 5% CO 2 . SiRNA was transfected with Lipofectamine™ 2000 Transfection Reagent according to the manufacturer's instructions.

| Statistical analysis
Statistical significance was determined using the one-way analysis of variance (ANOVA). All analyses were performed using SPSS Statistics ver. 19.0 (SPSS Inc). P values of less than .05 indicated statistical significance. Results are presented as mean ± SD.

| Simulation of the interaction between LB and GLP-1R by molecular docking
After confirming the insulin secretion-promoting function of LB, relative targets associated with the effect were evaluated. GLP-1R, FFAR1, GK and DPP-IV are general targets for anti-diabetic drugs.
Discovery studio 4.1 was used to simulate the binding ability between LB and targets. CDOCKER as the docking method showed the highest accuracy, which indicated the best ability to reproduce the natural binding mode. The attribute of '-CDOCKER_ENERGY' was used as a metric to measure the binding affinity. From Table 1 and

| Types of Interaction between LB and GLP-1R
Based on the mentioned results, the interaction between LB and GLP-1R was further explored by three-dimensional(3D) fluorescence spectroscopy. The 3D fluorescence spectrum of GLP-1R and LB reactions is shown in Figure 4A (Table 2). As can be seen from  the fluorescence quenching of GLP-1R caused by LB ( Figure 4E) was plotted. The binding constant Ka between LB and GLP-1R was calculated (Table 2).

| Effect of LB on the conformation of GLP-1R
The above experimental results showed that LB may affect tyros-

| The role of GLP-1R in LB promoting insulin secretion
Since   Since good insulin secretion-promoting effect of LB, new compounds derived from LB might be potential drug against T2DM. PDX1 is not only related to insulin secretion, but also to the proliferation of islet cells. Cell activity experiments have confirmed that LB can promote the activity of Ins1 cells. Therefore, we believe that LB may also have an effect on the proliferation of islet cells. This is the main direction of our follow-up work. Optimizing molecular structure, improving water solubility, and investigating the mechanism of action of drugs will be the focus of further research in the future.
Although the interaction between LB and GLP-1R has been supported using fluorescence spectrometry and CD, these experiments could not provide the information of interactional specificity between LB and the potential prodrugs. Therefore, the docking models presented in this study belong to predictive modelling, which can be applied to any type of unknown event, regardless of when it occurred. [28][29][30] However, in order to obtain a more accurate docking model, interactional specificity between molecules needs to be obtained. NMR spectroscopy is a powerful tool to study specificity of biomolecule-biomolecule [31][32][33][34] or biomolecule-ligand interactions and dynamics. [35][36][37] In future studies, the most accurate docking modelling based on the measured NMR data will be incorporated into our study.

| CON CLUS ION
The main purpose of this paper is to find the target of LB. Molecular simulation results show that GLP-1R is the potential target of LB.
SPR detection confirms the interaction between LB and GLP-1R.
Fluorescence spectra and circular dichroism prove that LB has an effect on the conformation of GLP-1R. In cell experiments, by interfering GPL-1R, GLP-1R was proved as the target of LB. Our findings may play a role in the study of anti-diabetic drugs.