Nodakenin represses obesity and its complications via the inhibition of the VLDLR signalling pathway in vivo and in vitro

Abstract Objectives Nodakenin (NK) is a coumarin glucoside that is found in the roots of Angelicae gigas. A limited number of studies have been conducted on the pharmacological activities of NK. Although NK is an important natural resource having anti‐inflammatory and antioxidant effects, no investigation has been conducted to examine the effects of NK on obesity and obesity‐induced inflammation. Materials and Methods The present study investigated the therapeutic effects of NK treatment on obesity and its complications, and its mechanism of action using differentiated 3T3‐L1 adipocytes and high‐fat diet (HFD)‐induced obese mice. Oil red O staining, western blot assay, qRT‐PCR assay, siRNA transfection, enzyme‐linked immunosorbent assay, H&E staining, immunohistochemistry, molecular docking and immunofluorescence staining were utilized. Results Treatment with NK demonstrated anti‐adipogenesis effects via the regulation of adipogenic transcription factors and genes associated with triglyceride synthesis in differentiated 3T3‐L1 adipocytes. Compared with the control group, the group administered NK showed a suppression in weight gain, dyslipidaemia and the development of fatty liver in HFD‐induced obese mice. In addition, NK administration inhibited adipogenic differentiation and obesity‐induced inflammation and oxidative stress via the suppression of the VLDLR and MEK/ERK1/2 pathways. This is the first study that has documented the interaction between NK and VLDLR structure. Conclusion These results demonstrate the potential of NK as a natural product‐based therapeutic candidate for the treatment of obesity and its complications by targeting adipogenesis and adipose tissue inflammation‐associated markers.

and obesity. [2][3][4][5] Several studies have highlighted the association between adipocytes and macrophages in obesity-mediated complications. 6 In an obese state, the expansion of the adipose tissue leads to the infiltration of adipose tissue macrophages (ATMs). Enhanced ATMs are the primary source of inflammatory responses. Stimulated ATMs induce cytokine secretion, which promotes the reciprocal release of free fatty acids from the adipocytes, forming a vicious loop. 7 Adipocyte differentiation has been demonstrated in the 3T3-L1 cell line, in which the cellular processes of adipogenesis were extensively characterized by three steps, namely, the arrest of the growth of confluent pre-adipocytes, mitotic clonal expansion (MCE) and terminal differentiation. During MCE, cell proliferation and cell cycle progression occur in sequence. At the end of the cell cycle, the cells undergo terminal differentiation. 8 The process of adipogenesis is regulated by the activation of the signalling pathways. During terminal differentiation, such as peroxisome proliferator-activated receptors (PPAR) γ and C/EBPα. 9 Meanwhile, it is also known that production of reactive oxidative stress (ROS) is markedly increased during adipocyte differentiation, indicative of the connection between ROS and adipogenesis. 10 Recent evidence suggests that the expression of the verylow-density lipoprotein receptor (VLDLR) is regulated by PPARγ. 11 VLDLR, a member of the low-density lipoprotein receptor family, is highly expressed in the adipose tissue and plays a prominent role in lipid uptake and adipogenesis. 12 There is convincing evidence that VLDLR is associated with obesity as demonstrated in humans and animal models. 13,14 In addition to adipocytes, VLDLR is also expressed in immune cells, including macrophages. Very low-density lipoprotein (VLDL)-VLDLR signalling in ATMs exacerbates adipose tissue inflammation in obesity, indicative of the role of VLDLR in adipose tissue inflammation and adipocyte-macrophage interactions. 15,16 Although several medications are designed to treat obesity, most of them act as either appetite suppressants (such as sibutramine and amfepramone) or gastrointestinal lipase inhibitors (such as orlistat).
Several agents have been subjected to clinical trials; however, only orlistat (Orli) and sibutramine were approved for long-term drug therapy. However, the administration of these drugs results in certain adverse effects, including diarrhoea, tympanites, stomachache and severe liver injury. 17 Hence, there is a need to develop substitutes based on a natural resource that could exert long-term beneficial effects and reduce the side effects.
The natural product Nodakenin (NK), a furanocoumarin glycoside initially isolated from the roots of Angelicae gigas (Umbelliferae) has been reported to possess anti-inflammatory, antioxidant and anti-hyperglycaemic effects. [18][19][20] There are a limited number of studies focusing on the biological effects of NK, except those demonstrating its anti-inflammatory effects. 21,22 Studies using macrophage cells and animal models have reported that NK exhibits anti-inflammatory effects. 23,24 Our previous study also demonstrated the mechanisms responsible for the anti-allergic inflammatory activity of NK. 25 However, the effect of NK on obesity and obesity-induced inflammation has not yet been documented. Therefore, to evaluate the therapeutic effects of the NK on obesity and its complications, we investigated the molecular mechanism(s) underlying the anti-adipogenic, anti-inflammatory and antioxidant effects of NK in 3T3-L1 cells and a high-fat diet (HFD)-induced obesity mouse model.
Next, the MTT solution (5 mg/mL) was added for 2 hours. After soaking off the supernatant, the formazan product was dissolved in DMSO, and the extent of cytotoxicity was measured at 570 nm using a BioTek™ Epoch microplate spectrophotometer (Biotek).
Experiments were performed in triplicate in a parallel manner and the values were represented as mean ± standard deviation (SD).

| Western blot analysis
The mice liver and adipose tissues as well as 3T3L-1 cells were homogenized in a commercial lysis buffer PRO-PREP (Intron) and centri- Western blotting experiments were performed in triplicate, and data were quantified and expressed as means ± SDs in relative protein level graph.

| Molecular docking
To predict the binding of NK on VLDLR, molecular docking was carried out using Autodock Vina. 26 The structure of NK was downloaded from PubChem (CID: 73191). The value of exhaustiveness in AutoDock Vina was set to 40. NK-bound structure of VLDLR was visualized and analysed by PyMol 27 and two-dimensional interactions between NK and VLDLR were estimated by LigPlot+. 28

| VLDLR transfection
The fully differentiated 3T3-L1 cells received VLDLR siRNA (Bioneer Corporation, Table 2) or pmaxGF0050™ vector using 4D-Nucleofector™ system (Lonza). After 4 hours incubation, the cells were treated with the presence and absence of NK100. The impact of VLDLR knockdown transfection was detected via qRT-PCR assay.

| Animal
Male

| Induction of obesity and drug administration
Using a blinded method, mice were randomly distributed into six

| Histological analysis
The liver and adipose tissues collected from each group were fixed with 4% formalin and embedded in paraffin. The tissues were cut into 4-mm sections. The sections were stained with haematoxylin and eosin (H&E) for histological examination. Images were acquired with a Leica microscope (Leica DFC295).

| Immunohistochemistry (IHC)
All IHC experiments were performed with the formalin-fixed, paraffin-embedded sample. Paraffin blocks were cut into 5 μm thick sections, mounted onto poly-L-lysine-coated slides, and dried. After the dried slides were de-paraffinized, antigen retrieval was performed for 20 minutes by using an automated antigen retrieval machine in the presence of ethylenediaminetetraacetic

| VLDL assay
The levels of VLDL in adipose tissue were quantified with mouse VLDL assay Kit (DoGenBio) according to the manufacturer's handbook.

| Immunofluorescence (IF) staining
After deparaffinization and rehydration, slides were incubated with anti-mouse VLDLR antibodies and anti-rabbit iNOS antibodies and visualized with FITC conjugated anti-mouse and TRITC conjugated anti-rabbit secondary antibodies, respectively. Slides were mounted and detected with a Nikon X-cite series 120 Q microscope (Nikon). The exposure parameters were kept the same for each sample.

| Statistical analysis
Experiments were performed in triplicate, and data are expressed as mean ± SD. Statistically significant values were determined using analysis of variance (ANOVA) and Dunnett's post hoc test.
P-values ˂ .05 were considered statistically significant. Statistical analysis was performed using GraphPad Prism 5.

| NK inhibits droplet accumulation and adipogenic differentiation in 3T3-L1 cells
To investigate the effect of NK ( Figure 1A

| VLDLR inhibition contributes to the impact of NK on adipogenesis in 3T3-L1 cells
Very low-density lipoprotein receptor is absent in preadipocytes but is remarkably manifested during adipogenesis and actively participates in adipocyte hypertrophy. 16 Figure 2A Figure 2C).

| NK suppresses weight gain in the HFDinduced obesity mouse model
To investigate the therapeutic effects of NK on obesity, we established the HFD-induced obesity mouse model as described in the methods section. After 13 weeks, the HFD mice exhibited an increase in fat pad compared with the mice fed with a normal diet, whereas Orli20 and NK20 administration in mice alleviated the overproduction of fat pad induced by HFD ( Figure 3A). As shown in Figure 3B,C, mice in the HFD group demonstrated a significantly increased body weight and weight gain values compared to mice in the control group. In contrast with the HFD group, treatment with Orli10, Orli20, NK10 and NK20 significantly attenuated the weight gain by 34.53%, 48.80%, 37.45% and 38.72%, respectively. No significant difference was observed in food intake among all experimental groups for 13 weeks ( Figure 3D).

| NK suppresses HFD-induced dyslipidaemia in the HFD-induced obesity mouse model
Animals fed a HFD demonstrated a significant dyslipidaemic plasma profile, with high levels of AST, ALT, BUN, triglyceride and total cholesterol. As shown in Table 3, after 5 weeks of treatment with Orli20 and NK20, the AST levels decreased significantly compared to that of the HFD group. In addition, NK20 treatment notably suppressed

| NK represses the development of fatty liver in the HFD-induced obesity mouse model
Long-term consumption of HFD not only leads to weight gain but also exacerbates the development of fatty liver, leading to several metabolic complications, which may be linked to oxidative stress. 29 Consistent with serum profiles, the administration of NK also demonstrated beneficial effects on lipid metabolism in liver tissues.
As seen from Figure 3E,F, mice with obesity had a notable colour change that may represent the accumulation of a high amount of fat and increased liver weight compared to that in the control group.
In contrast with mice in the HFD group, Orli and NK treatments However, the treatment with Orli and NK reduced HFD-induced lipid production in liver tissues ( Figure 3G). In addition, Orli and NK treatments significantly suppressed the relative protein and mRNA

F I G U R E 3 Effect of NK on weight gain and hepatic fat accumulation in mice with HFD-induced obesity. A, Photograph of the mouse body and epidydimal adipose tissues in the HFD-induced obesity model. B, Body weight and (C) weight gain values were assessed every
week. D, Food intake was recorded two or three times per week. E, A representative photograph of liver tissues in each experimental group (F) Liver weight was measured in each experimental group. G, Representative image of H&E staining of liver tissues from each group. Original magnification 100×. H, The protein levels of adipogenic factors in liver tissues. The total protein lysates were prepared and evaluated for determining the levels of SREBP-1, PPARγ and C/EBPα by western blot analysis using specific antibodies. Relative ratio level normalized to β-actin was determined by densitometric analysis (Bio-rad Quantity One ® Software) (I) mRNA expression of adipogenesisrelated genes (SREBP-1, PPARγ and C/EBPα) in liver tissues were analysed by the qRT-PCR assay. The values are expressed as mean ± SD of 8 mice per group. ### P < .001 vs the Con group; *P < .05, **P < .01, ***P < .001 vs HFD group levels of SREBP-1, PPARγ and CEBPα in the liver tissues compared to that in the HFD group ( Figure 3H,I). These results indicate that the administration of NK exhibits therapeutic effects alleviating HFDinduced hepatic steatosis.

| NK inhibits enlargement of adipocytes in the HFD-induced obesity mouse model
As shown in Figure 4A Figure 4D).  Figure 5D).

| NK inhibits obesity-induced inflammation via the MAPK/ERK signalling pathway in the HFDinduced obesity mouse model
As shown in Figure 4B, the crown-like structure formed by infiltrating macrophages was well-developed in the HFD group compared to the phenotype in the control group, whereas the administration of NK alleviated its appearance, suggesting the inhibitory effects of NK on activated macrophages in adipose tissues. When immunostaining was conducted using anti-CD68 and anti-F4/80 antibodies, which may be oxidative stress attributable to macrophage markers; the TA B L E 3 Effect of NK on plasma biochemistry in mice with HFD-induced obesity Note: ALT, AST, BUN, triglyceride and total cholesterol levels were estimated in mice with HFD-induced obesity. The values are expressed as mean ± SD of 8 mice per group. ## P < .01.
accumulation of macrophage markers around the adipocytes was notably increased in the HFD group, whereas the administration of NK inhibited CD68 and F4/80 ( Figure 6A). Western blot data also demonstrated that the protein expression of CD68 significantly decreased in the NK10 and NK20 groups compared to that in the HFD group ( Figure 6B). Next, we examined whether MAPK/ERKdependent adipose tissue inflammation is regulated by the administration of NK. As shown in Figure 6C, NK administration significantly suppressed the obesity-induced increase in MEK/ERK phosphorylation, as well as an increase in c-Fos protein expression. These results corroborate the hypothesis stating that NK inhibits adipogenesis and obesity-induced macrophage infiltration, simultaneously.

| D ISCUSS I ON
In this study, we utilized 3T3-L1 adipocytes and HFD-induced obese  (Table 3). Consistent with these results,

F I G U R E 4
Effect of NK on adipogenesis and fat accumulation in mice with HFD-induced obesity. A, The weight of the epididymal fat pad was measured in each experimental group. B, H&E staining analysis was performed using epididymal adipose tissue sections. Original magnification 100×. C, Based on H&E staining, adipocyte size was measured. D, The protein level of SREBP-1, PPARγ and C/EBPα in adipose tissue was estimated by western blot analysis. Relative ratio level normalized to β -actin was determined by densitometric analysis (Bio-rad Quantity One ® Software). All data are shown as the average value of each experimental group and are represented as mean ± SD (n = 8). ### P < .001 vs the Con group; **P < .01, ***P < .001 vs HFD group NK notably suppressed the hypertrophy and hyperplasia of adipocytes ( Figure 4B,C). Th anti-adipogenic effects of NK were also confirmed by the significant suppression of protein expression of the adipogenic transcription factors, including SREBP-1, PPARγ and C/ EBPα ( Figure 4D). By interacting with the adipose tissue, liver tissue plays a vital role in lipid metabolism. The HFD-driven aberrant F I G U R E 5 Effect of NK on VLDL/VLDLR signalling in mice with HFD-induced obesity. A, The concentration of VLDL in adipose tissues was determined by enzyme immunoassay kits. B, The expression of VLDLR in adipose tissues was observed by IHC. C, VLDLR protein levels were determined by western blot analysis using specific antibodies. β -actin was used as an internal control. The relative ratio level was determined by densitometric analysis. Values are represented as mean ± SD (n = 8); ### P < .001 vs Con group; ***P < .001 vs HFD group; significances between treated groups were determined using ANOVA and Dunnett's post hoc test. D, The colocalization of VLDLR and iNOS in ATM from mice with HFD-induced obesity was observed. Whole-mount IF analysis demonstrated nucleus (blue), VLDLR (green) and iNOS (red) in adipose tissues from each experimental group. The right panels show merged images from the individual left and middle panels. Note that the yellow-coloured regions indicate the co-localization of the target molecules in the right panels. Scale bar = 25 μm metabolic process would damage normal hepatic function and lead to the upregulation of oxidative stress and inflammation in systemic circulation and lipid disorders. 35 As shown in Table 3, we observed altered serum levels of AST and ALT, which are well-known enzymatic parameters of liver damage, in the HFD group compared to those in the control group. However, the NK20 administration significantly reduced the levels of these enzymes. In addition, the liver tissue of mice in the HFD group turned pale. NK administration alleviated this symptom of hepatic fat accumulation. It has been observed that the activation of hepatic SREBP-1, PPARγ and C/EBPα factors is associated with the upregulation of several lipid uptake-associated proteins and the formation of lipid droplets under obese state. [36][37][38] Our results from the gene expression studies indicated that the administration of NK significantly reduced the overexpression of SREBP-1,

F I G U R E 6
Effect of NK on obesity-induced inflammation in mice with HFD-induced obesity. A, The expression of CD68 and F4/80 in the adipose tissues was observed by IHC. Immunoblot results demonstrated the levels of (B) CD68 and (C) MEK/ERK/c-Fos pathway in the adipose tissues. Densitometric protein levels of (B) CD68, (C) p-MEK1/2, MEK1/2, p-ERK and c-Fos are represented as mean ± SD and plots of each protein are shown. β -actin and ERK were used as internal controls. ### P < .001 vs Con group; ***P < .001 vs HFD group; significant differences between the treatment groups were determined using ANOVA and Dunnett's post hoc test PPARγ and C/EBPα in the liver compared to their levels in the HFD group ( Figure 3H). Hence, we predict that NK may exert inhibitory effects on obesity-induced metabolic complications, such as hepatic steatosis.
In our study, the anti-adipogenic effects of NK were assumed to be involved in the inhibition of the VLDLR signalling pathway.
In addition, our results demonstrate that the administration of NK remarkably suppressed the overexpression of VLDLR protein and reduced VLDL levels in adipose tissues (Figures 2 and 5 As explained in the introduction, it is clear that VLDLR actively participates in adipocyte hypertrophy, excess lipid accumulation and overproduction of oxidative stress. 13 In contrast, previous studies have shown that VLDLR deficiency not only reduced the size of adipocytes and but also concurrently reduced the ATM content, which is indicative of the role of VLDLR in the association between adiposity and inflammatory action. 16 It has been also noted that macrophage VLDLR-mediated VLDL uptake might influence inflammatory responses, thereby potentiating adipose tissue inflammation and insulin resistance in obesity. 41 In accordance with the aforementioned studies, results from our study demonstrated a strong and consistent association between VLDLR and macrophage-mediated adipose tissue inflammation. Meanwhile, NK administration notably reduced the overexpression of VLDLR in iNOS + ATMs compared to that in the HFD group, supporting the hypothesis indicating the inhibitory role of NK in obesity and obesity-induced inflammation ( Figure 5).
CD68 was utilized to identify total macrophage infiltration in human adipose tissues where total adipose tissue areas and degree of adipocyte hypertrophy were positively associated with gene expression of CD68. 42 Our data demonstrated that macrophage factors CD68 and F4/80 were overexpressed in the HFD group, whereas these effects were abolished by administration with NK ( Figure 6A,B). Subsequently, we also demonstrated the inhibitory effects of NK on the phosphorylation of MEK1/2 and ERK1/2 as well as on the overexpression of MEK1/2 and c-Fos in obesity-induced mice ( Figure 6C). These observations are supported by previous studies that demonstrate targeting the ERK1 isoform in the regulation of adipocyte differentiation and HFDinduced obesity. 43 Other studies have observed that regulation of phosphorylation of ERK1/2 in adipose tissue is associated with macrophage infiltration of adipose tissue. This action mediates proinflammatory activation of adipocytes within the adipose tissue. 44 In addition, it has conclusively been demonstrated that MAPK signalling cascades influence the regulation of pro-inflammatory and oxidative stress pathways that are intimately connected with adipose tissue inflammation. 45 The purpose of this study was to determine the pharmacological effects of NK on obesity and its complications, including hepatic steatosis and obesity-induced inflammation. The evidence from this study suggests that NK is a more potent anti-adipogenic candidate that modulates the expression of several genes associated with lipid metabolism and adipogenesis in 3T3-L1 adipocytes. To the best of our knowledge, this is the first study that has documented the interaction between NK and VLDLR structure. NK administration also al- This study investigated that NK suppresses obesity via the inhibition of oxidative stress-mediated adipogenesis and the VLDLR signalling pathway. Inhibitory effect of NK is superior to that of Orlistat, highlighting the potential of NK as a therapeutic agent for treatment of obesity and its complications. In addition, to the best of our knowledge, this is the first study that has documented the interaction between NK and VLDLR structure.

CO N FLI C T S O F I NTE R E S T
The authors declare no conflict of interest.

I N FO R M E D CO N S E NT S TATE M E NT
Not applicable.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.