Anacardic acid mitigates liver fat accumulation and impaired glucose tolerance in mice fed a high‐fat and high‐sucrose diet

Abstract In this study, we evaluated the effects of anacardic acid (AA), a phenolic lipid found in cashew nuts (Anacardium occidentale), on metabolic disorders related to obesity, fatty liver disease, and diabetes using both in vitro and in vivo models. The application of AA led to a reduction in lipid accumulation in 3T3‐L1 cells without observable cytotoxicity. Results from Western blot analysis revealed that AA treatment also led to decreased expression of fatty acid synthase and peroxisome proliferator‐activated receptor gamma. In vivo studies were performed to evaluate the effects of AA treatment on fatty liver disease and diabetes. Mice fed a high‐fat and high‐sucrose diet had significantly higher body and liver weights, and higher levels of liver fat, cholesterol, fasting glucose, and homeostasis model assessment of insulin resistance (HOMA‐IR). However, 12 weeks of oral treatment with 500 μg/kg BW AA slowed down lipid accumulation rates in the liver and mitigated insulin resistance in these mice. Thus, AA may reduce lipid levels and have an antidiabetic effect.


| INTRODUC TI ON
Metabolic dysfunctions such as lipogenesis and insulin resistance are major contributors to obesity, type 2 diabetes, and cardiovascular diseases, which are major global health concerns (Kahn, Hull, & Utzschneider, 2006;Milic, Lulic, & Stimac, 2014). Nonalcoholic fatty liver disease (NAFLD) is also considered a metabolic syndrome as it shares common mechanisms with pathologies such as insulin resistance (Anstee, Targher, & Day, 2013). NAFLD is mainly caused by excessive free fatty acids and chronic inflammation in the adipose tissue (Milic et al., 2014). As fat accumulates in the liver, insulin signaling declines, leading to abnormal hepatic metabolism (Sookoian & Pirola, 2017).
Several agonists play important roles in the treatment of metabolic diseases. For example, peroxisome proliferator-activated receptors (PPARs) and fatty acid synthase (FAS) agonists have been shown to regulate lipid and glucose metabolism, as well as inflammation (Bensinger & Tontonoz, 2008;Botta et al., 2018;Gross, Pawlak, Lefebvre, & Staels, 2017;Moseti, Regassa, & Kim, 2016). However, agonists come with several side effects such as weight gain, cardiovascular risk, and hepatopathy. In contrast, medicinal plants have long been used for medical treatments and are considered a safe alternative. Thus, medicinal plants may be useful for preventing metabolic disorders.
Natural ingredients such as herbal extracts may alleviate metabolic diseases by modulating target molecules such as PPAR-γ and FAS. For example, the ethanol extracts of the peels of Citrus junos Tanaka and Prunus mume contain abundant flavonoids, which activate PPAR-γ to stimulate glucose uptake in C2C12 cells Shin et al., 2013). In addition, epigallocatechin-3-gallate (EGCG) has been shown to attenuate adipogenesis, thus downregulating PPAR-γ and FAS expression in 3T3-L1 cell lines (Wu et al., 2017).
Cashew nut extracts and AA have been shown to increase glucose uptake by upregulating plasma membrane glucose transporters in vitro (Tedong et al., 2010). However, AA effects have not been well studied, particularly in in vivo models. Therefore, in this study, we investigated whether AA affects lipid accumulation in 3T3-L1 cells and whether AA treatment mitigates fatty liver disease and insulin resistance in mice. Changes in liver fat and glucose levels, as well as lipid and insulin parameters, were measured in mice fed a high-fat and high-sucrose diet.

| Adipocyte differentiation and Oil Red O staining
The 3T3-L1 cells were cultured in DMEM supplemented with 10% (v/v) calf serum, antibiotics, and antimycotics. Cells grown at 100% in a 6-well plate were cultured for an additional 2 days and then transferred to a medium supplemented with 0.5 mM IBMX, 10 μg/ ml insulin, 0.5 μM dexamethasone, and heat-inactivated 10% FBS (MDI medium). Cells were cultured in the MDI medium for 2 days to induce cell differentiation. Subsequently, cells were cultured in 10 μg/ml insulin in the presence or absence of AA. This medium was replaced every 2 days. For Oil Red O staining, cells were first washed twice with phosphate-buffered saline (PBS) and then fixed in 2 ml 3.7% paraformaldehyde solution (diluted with PBS) for 5 min at room temperature. The solution was then removed, and the cells were fixed again for 1 hr, after which the paraformaldehyde solution was removed, and the cells were dried at room temperature. The dried cells were stained in Oil Red O working solution for 10 min and washed five times with distilled water. Stained cells were photographed under a microscope. Then, 100% isopropanol was added to elute the accumulated lipids in the cells, and the absorbance of the subsequent liquid was measured at 510 nm using a Micro plate reader (Molecular Devices).

| Cell viability test
Cell viability was measured using 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) solution. After the lipids were eluted from the cells, 100 μg/ml MTT solution was added to the cells and they were incubated for 3 hr, forming a blue product. The solution was then discarded, and the precipitate was solubilized using dimethyl sulfoxide (DMSO). The absorbance of the resulting solution was measured at 540 nm using a Micro plate reader (Molecular Devices).

| Western blot analysis
Proteins were extracted using a radioimmunoprecipitation assay buffer that contained protease and phosphatase inhibitors.
Protein quantification was carried out using the Bradford reagent (Biosesang). Electrophoresis was performed using sodium dodecyl sulfate-polyacrylamide gel with 30 μg of proteins, and samples were then transferred to nitrocellulose membranes. Western blot analysis and band detection were performed using an EZ-Western chemiluminescence detection kit (Dongin Biotech) according to the manufacturer's instructions.

| Animal experiments and diet-induced obesity
Animal experiments were performed at INVIVO Inc.. We obtained 4-week-old male C57BL/6 mice (Samtakobio) and housed them in a climate-controlled environment at 22°C and relative humidity of 50% under a 12-hr light/dark cycle for 1 week. The mice were then randomly divided into four groups of 10 mice each. Mice were fed (a) a normal diet (ND); (b) a high-fat and high-sucrose (HFS) diet; (c) a HFS diet with 250 μg/kg BW AA; and (d) a HFS diet with 500 μg/kg BW AA. Diets were purchased from Research Diets; we used diet D12450B (10% of total calories from fat) as the ND and diet D12079B as the HFS diet. The composition of the HFS diet is described in Table 1. The AA was dissolved in DMSO and diluted with corn oil (C8276; Sigma-Aldrich). Mice were fed these diets for 12 weeks while having free access to autoclaved tap water. Food intake was quantified once a week by measuring the amount of food left the day after feeding. At the end of the experiment, the mice were sacrificed, and their blood was sampled from the abdominal vein after ether anesthesia. Liver and epididymal fat was also extracted and weighed. One liver lobe from each mouse and the epididymal fat were fixed in 4% formalin for further analysis.
Our protocol for this animal experiment was approved (WKU17-101) by the Institutional Care and Use Committee at Wonkwang University, Iksan, Republic of Korea.

| Glucose tolerance test (GTT)
Mice were starved for more than 8 hr, and their fasting blood glucose levels were measured by inserting a blood glucose meter (Autochek DIATech) into a vein. Blood glucose levels were measured again after 30 min, and 2 g/kg BW glucose was added to the results of all mice, as recommended in the experimental protocol. Subsequently, blood glucose levels were measured through the subcutaneous vein every 30 min for 120 min. From these results, the area under the curve (AUC) was calculated, and the blood sugar change was analyzed for each treatment group.

| Hematoxylin and eosin staining
Liver tissues were excised and fixed in 10% formalin solution. Tissues were then embedded in paraffin, cut into sections of 5-μm thickness, and stained with Hematoxylin and eosin (H&E). The prepared tissues were examined using an optical microscope.

| Measurement of biochemical parameters
Sampled blood was coagulated at room temperature for 30 min and then centrifuged at 3,000 rpm for 10 min to separate the blood serum. The serum was analyzed by Green Cross Medical Foundation (GC Labs) for total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C) content. Serum insulin content was analyzed using an ALPCO ELISA kit.

| Statistical analyses
All results from the animal experiments are presented as the mean ± standard error (SE). Results were compared using one-way analysis of variance (ANOVA), followed by Duncan's test using SPSS version 12.0 (SPSS Inc.). Differences were considered statistically significant at p < .05.

| Effect of AA on adipocyte differentiation
We used the adipocyte differentiation model to evaluate the effect of AA treatment on lipid accumulation. From the examination of Oil

| Effects of AA on body and liver weights, food intake, and liver and epididymal fat in mice fed an HFS diet
We investigated whether AA treatment can lower lipid levels and mitigate obesity-related diseases in vivo using mice fed a HFS diet. Mice fed the HFS diet had significantly higher body and liver weights, and higher levels of liver and epididymal fat, than ND mice.
AA treatment led to slightly reduced liver weights and significantly reduced levels of liver fat in HFS-treated mice. However, AA had no effects on the body weights and epididymal fat levels of HFStreated mice (Figure 2a-f). Similar to results observed in the cell culture experiment, AA treatment reduced the liver weights and fat accumulation rates in HFS-treated mice. We did not observe any differences in food intake among treatment groups (Figure 2b).

| Effects of AA on the biochemical parameters of mice fed a HFS diet
Serum TG, TC, LDL-C, and HDL-C levels were significantly higher in mice fed a HFS diet than in ND mice. AA treatment led to significant reductions in LDL-C levels and slight reductions in TC levels in HFStreated mice. However, AA treatment did not significantly affect TG and HDL-C levels in HFS-treated mice (Figure 3a-d). Our results TA B L E 1 Composition of the high-fat and high-sucrose (HFS) diet suggest that AA can reduce fat accumulation in the liver and LDL-C levels induced by a HFS diet.

| Effects of AA on fasting glucose levels and impaired glucose tolerance
We conducted GTT on mice that had fasted overnight. Fasting blood glucose levels were significantly higher in mice fed a HFS diet than in ND mice (Figure 4a). AA treatment slightly reduced fasting glucose levels in HFS-treated mice (Figure 4a). In the GTT, no significant differences were observed among the treatment groups (Figure 4b), but the AUC was slightly lower for HFS-treated mice exposed to AA than for mice fed the HFS diet alone (Figure 4c). In addition, serum insulin levels were significantly higher in HFS-treated mice than in ND mice ( Figure 4d), but AA treatment led to decreases in insulin levels in the former group. Finally, we conducted the homeostatic model assessment of insulin resistance (HOMA-IR) using the results obtained in these experiments. HOMA-IR levels were significantly higher in HFS-treated mice than in ND mice (Figure 4e), but AA treatment led to a decrease in insulin resistance in the former group. These results indicate that AA can mitigate insulin resistance induced by HFS diet.

| D ISCUSS I ON
We first evaluated AA effects on lipid accumulation in vitro using 3T3-L1 cells and then confirmed our results in vivo by examining

F I G U R E 2
Effects of AA on body weight, food intake, and liver and epididymal fat in mice fed a high-fat and high-sucrose (HFS) diet. (a) Mice fed a HFS diet had significantly higher body weight than mice fed a normal diet (ND). However, the consumption of 500 μg/kg BW AA slightly reduced the body weights of HFS-treated mice. (b) Food intake (g) was significantly lower in HFS-treated mice than in ND mice. However, no significant differences were observed between mice on a HFS diet and mice that consumed both a HFS diet and AA. (c) Upon visual observation, HFS-treated mice accumulated significantly more liver fat than ND mice; however, this additional fat was reduced when AA was consumed. (d) Examination of tissues stained with hematoxylin and eosin indicated that the HFS diet led to an increase in the accumulation of liver fat, which was subsequently reduced when AA was consumed. (e) Liver weights were slightly reduced in HFS-treated mice that also consumed AA. (f) Epididymal fat weights were significantly higher in HFS-treated mice. AA consumption did not affect epididymal fat content (the different lowercase letters indicate significant differences at 0.05 level; p < .05, n = 7, data are presented as the mean ± standard error (SE)) AA has been shown to prevent metabolic disorders in in vitro studies. For example, cashew seed extracts and AA induced adenosine monophosphate-activated protein kinase and stimulated glucose uptake by upregulating plasma membrane glucose transporters in C2C12 cells (Tedong et al., 2010). However, AA did not induce the production of protein kinase B and insulin receptor phosphorylation (Tedong et al., 2010). In breast cancer cells, AA inhibited cell proliferation by modulating mitochondrial bioenergetics (Radde, Alizadeh-Rad, Price, Schultz, & Klinge, 2016). Similar anticancer effects have also been observed in prostate cancer cells (Park et al., 2018;Tan et al., 2017). Therefore, AA, a natural compound, may be able to prevent metabolic diseases such as diabetes and cancer. In this study, treatment with AA improved symptoms of obesity-linked metabolic syndrome, including adipocyte differentiation and elevated levels of lipids, glucose, and insulin (Figures 1-4), similar to results from previous studies.
It is important to control lipid levels to manage metabolic diseases, as high lipid levels in the body cause chronic diseases.
Because lipids are susceptible to oxidation, high levels of free fatty acids in the plasma and large amounts of fat stored in adipose tissues can induce oxidative stress. Inflammation from this stress can then lead to obesity (Manna & Jain, 2015). Adipose tissue dysfunction caused by hypertrophied adipocytes stimulates the secretion of proinflammatory adipokines such as leptin, resistin, visfatin, and tumor necrosis factor (TNF)-α. In turn, the overproduction of adipokines promotes obesity-associated metabolic diseases, including diabetes and cardiovascular disease (Katsareli & Dedoussis, 2014).
Adipokines are associated with inflammation that is linked to metabolic diseases, although the specific mechanisms remain unknown.
Nevertheless, PPAR-γ and FAS have been identified as biomarkers for evaluating adipocyte differentiation and are associated with adipokine productions. When levels of PPAR-γ, a lipid-activated transcription factor, increase in adipose tissue, fatty acids bind to receptors and are converted to triglycerides (Evans, Barish, & Wang, 2004;Kersten, Desvergne, & Wahli, 2000). This process activates a target gene that induces the conversion of preadipocytes to mature adipocytes (Evans et al., 2004;Kersten et al., 2000). FAS is a key lipogenic enzyme that catalyzes the synthesis of long-chain fatty acids, that is, converting acetyl-CoA and malonyl-CoA to palmitate, which are then stored in adipose tissue (Griffin & Sul, 2004;Wakil, Stoops, & Joshi, 1983). Results from Western blot analysis in this study revealed that AA inhibited lipid accumulation and decreased PPAR-γ and FAS expression levels, which were initially increased during MDI-induced adipocyte differentiation (Figure 1d). Therefore, AA may inhibit fat accumulation by suppressing the production of PPAR-γ and FAS-mediated cytokines. The antioxidant properties of AA may also help prevent adipocyte-induced oxidative stress and adipokine production.

F I G U R E 3
Effects of AA on levels of serum triglycerides (TGs), total cholesterol (TC), and low-density lipoprotein (LDL-C) and highdensity lipoprotein (HDL-C) cholesterol in mice fed a HFS diet. (a) HFS-treated mice had significantly higher TG levels than ND mice. Consumption of 500 μg/kg BW AA slightly reduced TG levels in HFS-treated mice. (b) TC levels were significantly higher in HFS-treated mice than in ND mice, but decreased slightly in the former group when AA was consumed. Levels of (c) LDL-C and (d) HDL-C were significantly higher in HFS-treated mice than in ND mice, but LDL-C significantly decreased in the group when 500 μg/kg BW of AA was consumed (the different lowercase letters indicate significant differences at 0.05 level; p < .05, n = 7, data are presented as the mean ± SE) Lipid metabolic pathways, such as lipogenesis, are closely linked with insulin resistance and NAFLD. Lipid accumulation in the liver leads to impaired insulin signaling and insulin resistance, which are key symptoms of NAFLD progression (Marchesini et al., 1999;Samuel & Shulman, 2012). In our in vivo model, mice fed a HFS diet had higher levels of liver fat accumulation, fat tissue weight, TG, TC, LDL-C, fasting glucose, serum insulin, and HOMA-IR (Figures 2-4).
However, AA treatment significantly decreased levels of liver fat, LDL-C, serum insulin, and HOMA-IR in these mice. and nonhistone proteins, and the activity of histone acetyltransferases (HATs) and histone deacetylases (Vahid, Zand, Nosrat-Mirshekarlou, Najafi, & Hekmatdoost, 2015). In cell culture studies, EGCG and curcumin also suppressed the activation of NF-κB, which is associated with inflammation and the inhibition of p300/ CBP expression (Choi et al., 2009;Liu, Chen, Cui, & Zhou, 2005;Shishodia, Amin, Lai, & Aggarwal, 2005). EGCG is reported to inhibit the differentiation of preadipocytes as well (Lin, Della-Fera, & Baile, 2005). In general, the dietary consumption of bioactive compounds may affect the epigenome, activating metabolic pathways that prevent obesity and diabetes caused by lipogenesis, inflammation, and glucose metabolism.
As with EGCG and curcumin, AA can function as an epigenetic regulator. In adipose cells, AA reduced the expression of TG, SREBP-1c, and ACC, and decreased rates of histone H3 acetylation (M. K. Kim, Kim, Kim, Lee, & Chung, 2018). AA also decreased TG, SREBP-1c, and ACC expression in p300-knockdown cells, indicating that p300 inhibits HAT activity and that AA inhibits lipogenesis (M. K. Kim et al., 2018). Similarly, we observed that AA inhibited HAT activity in vitro (unpublished data).
The antioxidant function of AA is important for preventing metabolic disorders. AA and its derivatives contain salicylic acid, a type of phenolic acid that has therapeutic properties such as anticancer, anti-inflammatory, and antiobesity properties. The inhibition of NF-κB kinase and HAT activities by AA has been demonstrated in several studies (30) (Hemshekhar, Sebastin Santhosh, Kemparaju, & Girish, 2012). Therefore, the strong antioxidant properties of AA may inhibit the markers and cytokines associated with metabolic diseases, thus mitigating fatty liver disease and insulin resistance. AA may also act as a HAT inhibitor to control the expression of lipid-related genes via epigenetic regulation.

| CON CLUS ION
AA can help alleviate symptoms arising from fatty liver and insulin resistance by inhibiting lipid accumulation in cells, as well as lowering liver fat content and enhancing insulin activity in mice. Therefore, our data suggest that the AA may help to protect against the development of fatty liver and diabetes.

ACK N OWLED G M ENTS
This study was supported by the Main Research Program (E0150301-05) of the Korea Food Research Institute.

CO N FLI C T S O F I NTE R E S T
The authors declare no potential conflicts of interest in this study.

E TH I C A L S TATEM ENTS
All animal experiments were approved by the Institutional Care and Use Committee at Wonkwang University (Iksan, Republic of Korea).