Corosolic acid inhibits adipose tissue inflammation and ameliorates insulin resistance via AMPK activation in high-fat fed mice

Phytomedicine 23 (2016) 181–190

Contents lists available at ScienceDirect

Phytomedicine journal homepage: www.elsevier.com/locate/phymed

Corosolic acid inhibits adipose tissue inflammation and ameliorates insulin resistance via AMPK activation in high-fat fed miceR Jie Yang a, Jing Leng a, Jing-Jing Li a, Jing-fu Tang b, Yi Li a, Bao-Lin Liu a, Xiao-Dong Wen a,∗ a State Key Laboratory of Natural Medicines, Department of Chinese Medicines Analysis, China Pharmaceutical University, Nanjing 210009, People’s Republic of China b Shanghai Hua Yu Chinese Herbs Co., Ltd, People’s Republic of China

a r t i c l e

i n f o

Article history: Received 27 August 2015 Revised 7 December 2015 Accepted 15 December 2015

Keywords: Corosolic acid AMPK Inflammation Insulin resistance Adipose tissue

a b s t r a c t Background: Adipose tissue inflammation is tightly associated with the development of insulin resistance. Corosolic acid (CRA), a natural triterpenoid, is well known as “phyto-insulin” due to its insulin-like activities. However, its underlying mechanism remains unknown. Purpose: In this study, we investigated the mechanisms of CRA on improving insulin resistance both in vivo and in vitro. Methods: C57BL/6 mice were fed with normal diet, high-fat diet (HFD) or HFD with CRA, respectively. General biochemical parameters in blood and glucose intolerance in mice were assayed. Meanwhile, proinflammatory cytokines and macrophage infiltrations in adipose tissues were analyzed by real-time PCR and immunohistochemical staining. The effects of CRA on insulin signaling transduction and AMPK activity in adipose tissues were investigated by western blot. Furthermore, the effects of CRA on AMPK were confirmed on 3T3-L1 cells by using both AMPK inhibitor and AMPKα 1/2-specific siRNA Results: CRA attenuated hyperlipidemia, improved insulin sensitivity and glucose intolerance in mice. Meanwhile, it alleviated inflammation in adipose tissues, demonstrated by the suppression of IKKβ phosphorylation and down-regulation of gene expressions of proinflammatory cytokines. Histological analysis revealed that CRA attenuated macrophage infiltrations into adipose tissue. It also improved insulin signaling transduction by modification of Ser/Thr phosphorylation of IRS-1 and downstream Akt, thereby improved insulin sensitivity in HFD-fed mice. Furthermore, CRA regulated AMPK activation in a LKB1dependent manner. AMPKα knockdown in adipocytes abolished the inhibitory effects of CRA on IKKβ and IRS-1 serine phosphorylation, indicating that CRA inhibited inflammation and ameliorated insulin resistance via AMPK activation. Conclusions: CRA inhibited inflammation with improvement in adipose tissue dysfunction and ameliorated insulin resistance in an AMPK–dependent manner. © 2016 Elsevier GmbH. All rights reserved.

Introduction Over the past decade, obesity has been recognized as a low-grade and chronic inflammatory disease tightly associated with the development of insulin resistance and type-2 diabetes (Kalupahana et al. 2012). Adipose tissue actively participates in the Abbreviations: CRA, corosolic acid; HFD, high-fat diet; IR, insulin resistance; TG, triglyceride; TC, total cholesterol; FBG, fasting blood glucose; Fins, fasting insulin; Mac-CM, macrophages-derived conditioned medium; CLSs, crown-like structures; ISI, insulin sensitive index. R The name of corosolic acid in NCBI PubChem: (1S,2R,4aS,6aR,6aS,6bR,8aR,10R,11R,12aR,14bS)-10,11-dihydroxy-1,2,6a,6b,9,9,12aheptamethyl-2,3,4,5,6,6a,7,8,8a,10,11,12,13,14b-tetradecahydro-1H-picene-4acarboxylic acid). ∗ Corresponding author. Tel.: +86 25 86185045; fax: +86 25 85301528. E-mail address: [email protected] (X.-D. Wen).

http://dx.doi.org/10.1016/j.phymed.2015.12.018 0944-7113/© 2016 Elsevier GmbH. All rights reserved.

obesity-induced inflammation through recruitment of macrophages and release of pro-inflammatory cytokines which impairs insulin signaling, leading to insulin resistance and metabolic disorders (Greenberg et al. 2006). Macrophage abundance is thought to contribute to adipose tissue dysfunction in obesity, with a tendency towards overproduction of proinflammatory adipokines, and reduced production of anti-inflammatory and insulin-sensitizing adipokines such as adiponectin (Lafontan 2014). Insulin promotes glucose disposal in adipose tissue, muscle and liver through insulin receptor substrate-1(IRS-1)/PI3K/Akt signaling, whereas inflammatory molecules, such as TNF-α and IL-6, block insulin PI3K/Akt signaling by impairing IRS-1 function, leading to insulin resistance (Gual et al. 2005). It is well established that inflammation-associated deregulations of adipokine expression is a causative event for the initiation of insulin resistance.

182

J. Yang et al. / Phytomedicine 23 (2016) 181–190

Fig. 1. Chemical structure of corosolic acid.

As an energy sensor, AMPK regulates glucose and lipid metabolism, and has been considered as a potential target for the treatment of metabolic disorders (Grahame 2014). Increasing evidences have demonstrated its anti-inflammatory action implicated in attenuating insulin resistance. It has been reported that AICA riboside (AICAR), an AMPK agonist, improves insulin resistance due to the reduction of obesity-induced inflammation (Yang et al. 2012). As a potent inflammation inhibitor, salicylate could ameliorate insulin resistance by suppression of inflammation (Yuan et al. 2011), and it could regulate the activity of AMPK (Hawley et al. 2012). These findings raise the possibility that regulation of AMPK activity might be a therapeutic strategy for the management of insulin resistance in which inflammation is involved. Banaba (Lagerstroemia speciosa (L.) Pers., Lythraceae) have been used in Southeast Asia as a folk medicine for the treatment of diabetes and obesity. Its antidiabetes and antiobesity effects have been demonstrated in several animal models as well as in a number of human studies (Stohs et al. 2012). Many studies have indicated that corosolic acid (CRA, Fig. 1), one of the main components in banaba (Sivakumar et al. 2009), is responsible for its antidiabetic activity. It has been reported that CRA promotes glucose disposal after glucose load and exhibits glucose lowering action in diabetic mice (Takagi et al. 2010; Miura et al. 2006). Although these evidences demonstrated that CRA has beneficial effects on glucose homeostasis, its potential therapeutic targets and underlying mechanism remain unknown. Some studies showed that CRA inhibits inflammation and has significant effects on attenuating metabolic disorders (Chen et al. 2012; Yamaguchi et al. 2006). These observations raise an interesting question whether the beneficial effects of CRA on glucose and lipid homeostasis were associated with suppressions of obesity-induced inflammation. Since it was reported that CRA could modulate the activation of AMPK in tumor cells (Lee et al. 2010), we also wonder to know whether AMPK is involved in the anti-inflammation and anti-diabetic effects of CRA. To address these issues, in the present study we observed the effects of CRA on high-fat diet fed mice by focusing on the improvement of inflammation and regulation of insulin signaling in adipose tissue. Our work showed that CRA inhibited inflammation with improvement in adipose tissue dysfunction and ameliorated insulin resistance in an AMPK dependent manner. Materials and methods Materials CRA (≥98% in purity, Molecular weight: 472.7) was purchased from Chendu Biopurify Medical Technology Co., Ltd. (Chendu, China; 13120404). Pioglitazone (Pio, ≥98% in purity, molecular

weight: 356.44) was purchased from Feiyu Co., Ltd. (Wuxi, China; FY1308). For cell experiments, CRA was dissolved in dimethyl sulfoxide (DMSO, final concentration 0.1%, v/v). Lipopolysaccharide (E. coli serotype 055:B5, LPS) and compound-C were obtained from Sigma (St. Louis, MO, USA). Insulin was purchased from Wanbang Biochemical Pharmaceutical Company (Xuzhou, Jiangsu, China). Antibodies including anti-IRS-1, anti-p-IRS-1 (Ser307), antiAkt, anti-p-Akt, anti-Ikkβ , Anti-p-Ikkα /β (Ser176/180), anti-LKB1, anti-p-LKB1, anti-AMPKα , and anti-p-AMPKα (Thr172) were obtained from Cell Signaling Technology, Inc. (Cell Signaling Technology, Beverly, MA, USA). PY99, small interfering RNA (siRNA) duplexes specific for AMPKα 1/2, control siRNA and siRNA transfection reagent were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Anti-GAPDH, anti-F4/80 antibody (1:150) and 5aminoimidazole-4-carboxamide1-β -D-ribofuranoside (AICAR) were from Bioworld Technology (St. Paul, MN, USA), Abcam, and Beyotime Institute of Biotechnology (Shanghai, China), respectively. The commercial kits for triglyceride (TG), total cholesterol (TC), free fatty acid (FFA) and glucose were obtained from (Jiancheng, Nanjing, China). The ELISA kit for insulin was obtained from Millipore (Millipore, USA). SYBER Green PCR kits were obtained from Applied Biosystems (Foster City, CA, USA). Animals Male C57BL/6 mice at 5 weeks of age were purchased from the Animal center of Yangzhou University (Yangzhou, Jiangsu, China). Mice were maintained under controlled room temperature (20 ± 2 °C) and humidity (60–70%) with day /night cycle (12 h/12 h). All animals had free access to food and water. The animal care and study protocols were maintained in accordance with the Provisions and General Recommendation of Chinese Experimental Animals Administration Legislation. Preparation of macrophages-derived conditioned medium (Mac-CM) Peritoneal macrophages were collected from mice and activated with LPS (E. coli serotype 055:B5) as described previously (Liu et al. 2011). Cell culture, differentiation and transfection 3T3-L1 cells, obtained from the cell bank of the Chinese Academy of Sciences, were cultured and differentiated as described previously (Wang et al. 2013). Transfection of small interfering RNA (siRNA) duplexes specific for AMPKα 1/2, or control siRNA into cells were carried out using siRNA transfection reagent, according to the manufacturer’s instructions for 5–7 h at 37 °C in a CO2 incubator. After transfection, cells were cultured in medium for 18–24 h, and then replaced with fresh medium for another 24 h. The cells were treated with indicated agents in the presence or absence of MacCM for 0.5 h. Then the proteins specific for the siRNA transfection levels were assayed for by Western blot. High-fat feeding After acclimatization for 5 days, mice were randomly divided into five groups: (1) normal group (n = 24) fed with standard chow diet (10% kcal fat, 70% kcal carbohydrates, 20% kcal proteins); (2) HFD group (n = 24) fed with high-fat diet (HFD) (60% kcal fat, 20% kcal carbohydrates, 20% kcal protein) for 16 weeks; (3) HFD and CRA1 group (n = 24) fed with HFD for 8 weeks and then switched to HFD supplemented with CRA (10 mg/kg/day) for another 8 weeks; (4) HFD and CRA2 group (n = 24) fed with HFD for 8 weeks and then switched to HFD supplemented with CRA (20 mg/kg/day) for another 8 weeks; (5) HFD and pioglitazone (Pio) group (n = 24) fed with HFD for 8 weeks and then switched to HFD supplemented with Pio (10 mg/kg/day) for another 8 weeks. The experiment was designed according to our previous study

J. Yang et al. / Phytomedicine 23 (2016) 181–190

183

Fig. 2. Effect of corosolic acid on adipocyte sizes in high-fat diet fed mice by hematoxylin-eosin staining. (Normal, normal group; HFD, HFD-fed group; HFD + CRA1, HFD supplemented with 10 mg/kg corosolic acid group; HFD + CRA2 group, HFD supplemented with 20 mg/kg corosolic acid group; HFD + Pio, HFD supplemented with 10 mg/kg pioglitazone group.) Bar = 50 μm. All results were expressed as mean ± S.E.M. (n = 8). ∗ p < 0.05 vs. the HFD; # p < 0.05 vs. the normal. Table 1 Effects of corosolic acid on general characteristics of high fat diet fed mice. Parameter

Normal

HFD

Body weight (g) TG (mg/dl) TC (mg/dl) FBG (mmol/l) FFA(umol/l)

32.40 ± 2.22 76.35 ± 11.27 91.54 ± 5.76 4.12 ± 0.89 624.02 ± 24.82

45.60 ± 1.86 124.04 ± 9.01# 255.69 ± 1.24# 7.71 ± 1.02# 1867.53 ± 38.32# #

HFD+Pio

HFD+CRA1

HFD+CRA2

44.70 ± 2.49 50.49 ± 6.33∗ 106.71 ± 12.37∗ 5.37 ± 1.32∗ 976.65 ± 26.76∗

45.00 ± 2.74 111.87 ± 13.15∗ 119.92 ± 2.99∗ 6.52 ± 1.42∗ 876.32 ± 28.42∗

41.60 ± 2.66∗ 89.02 ± 4.13∗ 99.71 ± 4.35∗ 4.41 ± 0.38∗ 854.99 ± 21.10∗

Normal, normal group; HFD, HFD-fed group; HFD+CRA1, HFD supplemented with 10 mg/kg corosolic acid group; HFD + CRA2 group, HFD supplemented with 20 mg/kg corosolic acid group; HFD+Pio, HFD supplemented with 10 mg/kg pioglitazone group. All results were expressed as mean ± S.E.M. (n = 8). ∗ p < 0.05 vs. the HFD; # p < 0.05 vs. the normal.

(Li et al. 2014) which showed CRA treatment for 8 weeks could effectively improve insulin sensitivity in HFD mice. The doses of CRA were chosen according to previous studies (Miura et al. 2004; Chen et al. 2012). At the end of experiment, blood was collected from retinal venous plexus, centrifuged and the serum was harvested. The epididymal adipose tissue was isolated and frozen in liquid nitrogen, and stored at −80 °C until use. Oral glucose tolerance After oral administration of CRA for eight weeks, the mice were fasted for 6 h and orally administrated of glucose (2 g/kg). The blood was collected from tail venous at determined times and

blood glucose was measured using test strips on an One Touch profile glucose Meter (Johnson & Johnson, New Brunswick, USA). The area under the curve (AUC) for glucose was calculated as described previously (Wang et al. 2013).

General biochemical parameters in blood The blood levels of TG, TC, FFA and blood glucose in plasma were determined using commercial kits. The insulin concentration was measured by ELISA according to the manufacturer’s instruction. Besides, insulin sensitive index (ISI) was calculated as follow: ISI = 1/(blood glucose × blood insulin).

184

J. Yang et al. / Phytomedicine 23 (2016) 181–190

Fig. 3. Effects of corosolic acid on glucose intolerance (A and B) and insulin sensitivity (C and D) in HFD-fed mice. (Normal or N, normal group; HFD or H, HFD-fed group; HFD + CRA1 or C1, HFD supplemented with 10 mg/kg corosolic acid group; HFD + CRA2 group or C2, HFD supplemented with 20 mg/kg corosolic acid group; HFD + Pio or Pio, HFD supplemented with 10 mg/kg pioglitazone group.) All results were expressed as mean ± S.E.M. (n = 8). ∗ p < 0.05 vs. the HFD; # p < 0.05 vs. the normal.

Fig. 4. Effects of corosolic acid on IKKβ phosphorylation (A) and inflammation-related gene expression (B, C and D) in adipose tissue. (Normal, normal group; HFD, HFDfed group; HFD + CRA1, HFD supplemented with 10 mg/kg corosolic acid group; HFD + CRA2 group, HFD supplemented with 20 mg/kg corosolic acid group; HFD + Pio, HFD supplemented with 10 mg/kg pioglitazone group.) The results were expressed as mean ± S.E.M. and data were derived from at least 8 mice. ∗ p < 0.05 vs. the HFD; # p < 0.05 vs. the normal.

J. Yang et al. / Phytomedicine 23 (2016) 181–190

185

Fig. 5. Corosolic acid decreased macrophage recruitment in adipose tissue evidenced by immunohistochemical staining. (Normal, normal group; HFD, HFD-fed group; HFD + CRA1, HFD supplemented with 10 mg/kg corosolic acid group; HFD + CRA2 group, HFD supplemented with 20 mg/kg corosolic acid group; HFD + Pio, HFD supplemented with 10 mg/kg pioglitazone group.) Arrows indicate CLS. Bar = 50 μm. The results were expressed as mean ± S.E.M. and data were derived from at least 8 mice. ∗ p < 0.05 vs. the HFD; # p < 0.05 vs. the normal.

Real-time PCR analysis Total RNA was extracted from adipose tissue using Trizol reagent (Invitrogen, Carlsbad, CA, USA) and reversely transcribed into cDNAs using M-MLV Reverse Transcriptase (Toyobo, Osaka, Japan). The resulting cDNAs were amplified using SYBER Green PCR kit. Quantitative polymerase chain reaction (PCR) was performed on an ABI Step one plus real time-PCR system. The primers used are listed in Supplemental Table 1. Samples were assayed in triplicate and the relative abundance of mRNAs was calculated with the CT method with β -actin as the invariant control. Western blot Protein in adipose tissue or adipocytes was separated by SDSPAGE and transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, USA). The membranes were blocked with 5% fat-free milk and incubated with different primary antibodies. The bound antibodies were detected using horseradish

peroxidase-conjugated anti-rabbit antibodies. Antibody reactivity was detected by ECL Western Blotting Detection Systems. Histology and immunohistochemistry Adipose tissues were fixed in 4% buffered paraformaldehyde, embedded in paraffin and sectioned at 4 μm. The histological characterizations, including adipocyte sizes (hematoxylin–eosin staining) and crown-like structure (anti-F4/80 antibody, 1:150) were carried out as previously described (Dong et al. 2014). Five random fields from each section were examined and Image-Pro Plus Version 7.0 (Media Cybernetics) was used to measure adipocyte diameters. Detection of glucose uptake in 3T3-L1 cells with fluorescence microscopy Glucose uptake was determined by the method described by Chen et al. (2009).

186

J. Yang et al. / Phytomedicine 23 (2016) 181–190

Fig. 6. Effects of corosolic acid on regulated p-IRS-1 at S307 (A), p-IRS1 at tyrosine residues (B) and p-Akt (C) in adipose tissue. (Normal, normal group; HFD, HFDfed group; HFD + CRA1, HFD supplemented with 10 mg/kg corosolic acid group; HFD + CRA2 group, HFD supplemented with 20 mg/kg corosolic acid group; HFD + Pio, HFD supplemented with 10 mg/kg pioglitazone group.) The results were expressed as mean ± S.E.M. and data were derived from at least 8 mice. ∗ p < 0.05 vs. the HFD; # p < 0.05 vs. the normal.

Statistical analysis

CRA improved glucose tolerance

Results were expressed as mean ± S.E.M. Differences were analyzed by one-way ANOVA followed by Newman–Keuls test. Values of p < 0.05 were considered statistically significant.

Glucose tolerance test is an important indicator for insulin action, and therefore we examined the effect of CRA on oral glucose tolerance in mice. HFD impaired glucose tolerance in mice, demonstrated by the delayed glucose disposal. Oral administration of CRA reversed glucose intolerance in a dose-dependent manner (Fig. 3A). Consistent with the promotion of glucose disposal, the increased total glucose was also reduced by CRA, as indicated by a 15.84% and 31.23% decrease (10 and 20 mg/kg, respectively, Fig. 3B) in AUC. Moreover, HFD insult impaired insulin sensitivity, but this alteration was effectively restored by CRA, as evidenced by a 2-fold and 7-fold increase (10 and 20 mg/kg, respectively, Fig. 3C, D) in ISI compared to the HFD group.

Results Effects of CRA on body weights and blood parameters in HFD-fed mice C57BL/6 mice were fed a normal or HFD for 16 weeks. There were no differences in food intakes among all groups (data not shown). Body weight was significantly higher in the HFD group compared with the normal group (Table 1), and this increase was accompanied by a significant increase in adipocyte sizes (Fig. 2). Compared with the HFD group, oral administration of CRA (20 mg/kg) decreased body weight and adipocyte sizes by 28.4% and 36.4% respectively (Table 1, Fig. 2). However, there was no significant difference in body weight between the group treated with 10 mg/kg CRA and the HFD group. High-fat feeding led to lipid disorders in mice, as we observed the significant increase of TC, TG, FBG and FFA in HFD-fed mice. By contrast, compared with the HFD group, oral administration of CRA (10 and 20 mg/kg) effectively decreased the level of TG by 9.81% and 28.23% (Table 1), respectively. Similarly, the levels of TC were also reduced by 53.10% and 61.00% (Table 1), respectively. Meanwhile, compared with the HFD-fed group, the FBG and FFA in CRA (20 mg/kg) treated groups were significantly decreased by 42.80% and 53.08% (Table 1) respectively. These results indicated the beneficial effects of CRA on the regulation of glucose and lipid homeostasis.

CRA alleviated inflammation and regulated adipokine expressions in adipose tissues Adipose inflammation is associated with the development of insulin resistance. HFD feeding initiated inflammatory response in adipose tissue, demonstrated by enhanced IKKβ phosphorylation and the deregulation of inflammation-related gene expression. By contrast, compared to these in HFD-fed mice, oral treatment of mice with CRA (20 mg/kg) attenuated IKKβ phosphorylation (p < 0.05, Fig. 4A) and significantly reduced gene expressions of proinflammatory cytokines, as evidenced by the 76.44, 65.84 and 85.82% decrease in the gene expressions of IL-6, MCP-1, and TNF-α , respectively (Fig. 4B, C). Meanwhile, the gene expression of Fizz1, the marker gene of M2 macrophage polarization, was increased more than 10-fold by CRA (20 mg/kg) treatment (Fig. 4D). These results well demonstrated the beneficial effects of CRA on regulation of adipose function.

J. Yang et al. / Phytomedicine 23 (2016) 181–190

187

Fig. 7. Corosolic acid regulated AMPK activation in adipose tissue (A) and adipocytes (B). (Normal, normal group; HFD, HFD-fed group; HFD + CRA1, HFD supplemented with 10 mg/kg corosolic acid group; HFD+CRA2 group, HFD supplemented with 20 mg/kg corosolic acid group; HFD + Pio, HFD supplemented with 10 mg/kg pioglitazone group.) The results were expressed as mean ± S.E.M. of three independent experiments. ∗ p < 0.05 vs. the control; # p < 0.05 vs. the indicated treatment.

CRA attenuated the macrophage infiltration in adipose tissue

CRA regulated AMPK activation in adipose tissue and adipocytes

To further understand the impact of CRA on adipose tissue immune cell populations that are major contributors to adipose tissue inflammatory status, we examined the relative proportions of macrophages in adipose tissues. Immunohistochemical staining showed that high-fat feeding led to macrophage recruitment in adipose tissues, indicated by increased F4/80+ positive cells which formed crown-like structures (CLSs) in the adipose tissue of HFD-fed mice (Fig. 5). CRA (10 and 20 mg/kg) and pioglitazone (10 mg/kg) treatment significantly decreased the number of CLSs (p < 0.05, Fig. 5), suggesting less macrophage population in adipose tissue. These results showed CRA could attenuate the macrophage infiltration in obesity.

As AMPK may ameliorate insulin resistance by inhibition of inflammation, it is of interest to investigate the potential regulation of AMPK by CRA. HFD feeding suppressed the phosphorylation of AMPK in adipose tissue, whereas this alteration was effectively reversed (p < 0.05, Fig. 7A) by CRA treatment (10 and 20 mg/kg), revealing the positive regulation of AMPK activity by CRA in vivo. Meanwhile, we also observed the regulation of basal AMPK phosphorylation in adipocytes. As shown in Fig. 7B, CRA (10 μM) increased LKB-1 (p < 0.05) and AMPK phosphorylation (p < 0.05) at 0.5 h, and the enhanced phosphorylation lasted for 4 h. These results indicated that CRA enhanced AMPK in a LKB1-dependent manner.

CRA modulated insulin signaling in adipose tissue CRA inhibited inflammation in adipose tissue, and it was tempting to know whether this action contributed to improving insulin signaling. We observed that enhanced serine phosphorylation of IRS-1 in HFD-fed mice, while IRS-1 tyrosine phosphorylation (detected by PY99) in response to insulin was attenuated. CRA administration (10 and 20 mg/kg) inhibited IRS-1 serine phosphorylation (p < 0.05) and then effectively restored insulin-mediated IRS-1 tyrosine phosphorylation (p < 0.05, Fig. 6A, B). As an expected result, CRA treatment (10 and 20 mg/kg) also effectively restored the loss of insulin-mediated Akt phosphorylation in adipose tissue (p < 0.05, Fig. 6C), indicative of the improvement of insulin signaling.

CRA regulated NF-κ B p65 and Akt phosphorylation in adipocytes CRA positively regulated AMPK activity, and then it was tempting to know whether this action contributed to the amelioration of insulin resistance. Exposure of adipocytes to Mac-CM led to inflammation, evidenced by increased NF-κ B p65 phosphorylation. Similar to its regulation in adipose tissue, CRA treatment (10 μM) inhibited NF-κ B p65 phosphorylation (p < 0.05, Fig. 8A). Meanwhile we also observed that CRA effectively normalized insulin-mediated Akt phosphorylation (p < 0.05) in adipocytes exposed to Mac-CM (Fig. 8B). Co-treatment with AMPK inhibitor compound C diminished the effects of CRA on NF-κ B p65 and Akt phosphorylation, suggesting the possible involvement of AMPK in the regulation.

188

J. Yang et al. / Phytomedicine 23 (2016) 181–190

Fig. 8. Effect of corosolic acid on phosphorylation of p65 (A) and Akt (B) in adipocytes as well as IKKβ (C) and IRS-1 serine (D) phosphorylation in adipocytes transfected with AMPKα 1/α 2 or control scrambled siRNAs. The results were expressed as mean ± S.E.M. of three independent experiments. ∗ p < 0.05 vs. the control; # p < 0.05 vs. the indicated treatment.

CRA improved insulin-mediated glucose uptake in adipocytes

Discussion

One of the insulin actions in adipocytes is to promote glucose uptake. Mac-CM challenge impaired insulin action indicated by 61.95% loss of insulin-mediated glucose uptake. Pretreatment of adipocytes with CRA (10 μM) improved insulin-mediated glucose uptake (p < 0.05), whereas this action was blocked by co-treatment with compound C (p < 0.05, Fig. 9), indicative of the involvement of AMPK in the regulation.

Chronic HFD feeding resulted in increasing body weight and adipocyte size with the development of glucose intolerance and hyperlipidemia. Consistent with previous studies (Takagi et al. 2010; Miura et al. 2006), our present results showed that CRA (10 and 20 mg/kg) attenuated glucose intolerance and hyperlipidemia in HFD-fed mice, well demonstrating that CRA regulated glucose homeostasis through enhancing insulin sensitivity. Interestingly, we found that CRA treatment at 20 mg/kg suppressed HFD-induced adipocyte size and body weight gain, suggesting that CRA may inhibit lipid accumulation in adipose tissue. Adipose tissue inflammation is a major contributor to the pathogenesis of obesity-associated insulin resistance. As CRA exerts anti-inflammation activity (Chen et al. 2012; Yamaguchi et al. 2006), we speculated that CRA could ameliorate HFD-induced adipose tissue inflammation. This hypothesis was confirmed in our studies. Oral administration of CRA (10 and 20 mg/kg) inhibited IKKβ /NF-κ B activation and effectively reduced pro-inflammatory adipokine expressions including TNF-α , MCP-1 and IL-6, indicating that CRA inhibited inflammation and this action was

CRA inhibited inflammation in a AMPK-dependent manner For further confirming the role of AMPK, we transfected 3T3-L1 adipocytes with AMPKα 1/2-specific siRNA. As seen in Fig. 8C, CRA attenuated IKKβ phosphorylation (p < 0.05), but this action was attenuated (p < 0.05) by AMPKα knockdown. Meanwhile we also observed that knockdown of AMPK diminished the inhibitory effect of CRA on serine phosphorylation at insulin receptor substrate-1 (IRS-1) (p < 0.05, Fig. 8D). These results demonstrated that CRA inhibited inflammation and protected IRS-1 function through AMPK activation.

J. Yang et al. / Phytomedicine 23 (2016) 181–190

189

phosphorylation in response to insulin. As a resultant improvement of insulin signaling, insulin-mediated glucose uptake was restored in CRA-treated cells. All these data suggest that CRA could overcome the HFD-induced impairment in IRS-1/PI3K/Akt insulin signaling pathway, leading to the improvement of glucose disposal in HFD-fed mice. Accumulating evidences demonstrate that AMPK ameliorates insulin resistance by inhibition of inflammation (Yang et al. 2010). Long-term high-fat feeding attenuated AMPK phosphorylation in adipose tissue, but this change was prevented by oral administration of CRA, suggesting the possible involvement of AMPK in CRA regulation. As an upstream kinase, LKB1 phosphorylates the Thr172 residue of the α -kinase subunit of AMPK, leading to AMPK activation. CRA increased LKB1 activity, indicating that it promoted AMPK activity in a LKB1-dependent fashion. AMPKα knockdown abolished the inhibitory effects of CRA on IKKβ and IRS-1 serine phosphorylation, confirming that CRA improved IRS-1 function by inhibition of inflammation via regulation of AMPK activity. Consistent with the published study which shows that CRA activated AMPK in cancer cells (Liu et al. 2011), our work further demonstrated that CRA enhanced AMPK activation through regulation of LKB1 and this action contributed to the amelioration of insulin resistance.

Fig. 9. Corosolic acid regulated glucose uptake in adipocytes. The results were expressed as mean ± S.E.M. of three independent experiments. ∗ p < 0.05 vs. the control; # p < 0.05 vs. the indicated treatment.

implicated in the improvement of adipose function. In addition, obesity-induced insulin resistance has been observed along with macrophage and immune cell infiltration (Lafontan 2014). Our study also proved HFD induced macrophages infiltration in adipose tissue which exacerbates adipose tissue dysfunction and leads to insulin resistance. CRA administration reduced macrophages infiltration, contributing to blocking the inflammatory cross-talk between macrophages and adipocytes. Insulin resistance is characterized by the specific impairment of insulin PI3K signaling, in which the malfunction of IRS-1 is a crucial event (Gual et al. 2005). Many inflammatory molecules such as TNF-α and IL-6, negatively regulate IRS-1 function by increasing its serine phosphorylation, leading to a decrease in insulin-mediated IRS-1 tyrosine phosphorylation (Gual et al. 2005; Feinstein et al. 1993). CRA inhibited inflammation in adipose tissue, and then it was tempting to know whether this action was responsible for improving insulin signaling. HFD feeding induced IRS-1 serine phosphorylation and attenuated insulin-mediated tyrosine phosphorylation in response to insulin, suggesting the impairment of IRS-1 function. As expected, CRA improved IRS-1 function and facilitated insulin PI3K signaling, indicated by enhanced Akt

Conclusions In conclusion, our study demonstrated that CRA inhibited inflammation with beneficial regulation of adipokine expression in adipose tissue, and thereby facilitated insulin IRS-1/PI3K signaling, leading to the improvement of insulin sensitivity in HFD-fed mice. Furthermore, we confirmed its regulation of AMPK activity contributed to inhibition of inflammation implicated in insulin resistance. The proposed working pathway was shown in Fig. 10. Better understanding of the anti-diabetic action of CRA and underlying mechanism would be beneficial for its possible application in the management of insulin resistance and diabetes. Conflict of interest The authors declare no conflicts of interest. Acknowledgments This work was supported by the National Natural Science Foundation of China (nos. 81102763, 81373919, 81573567 and 81373957). We also greatly appreciate the financial support from the Priority Academic Program Development of Jiangsu higher education institutions (PAPD).

Fig. 10. The proposed action pathway for CRA.

190

J. Yang et al. / Phytomedicine 23 (2016) 181–190

Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.phymed.2015.12.018. References Chen, H., Yang, J., Zhang, Q., Chen, L.H., Wang, Q., 2012. Corosolic acid ameliorates atherosclerosis in apolipoprotein E-deficient mice by regulating the nuclear factor-κ B signaling pathway and inhibiting monocyte chemoattractant protein-1 expression. Circ. J. 76, 995–1003. Chen, Y., Li, Y., Wang, Y., Wen, Y., Sun, C., 2009. Berberine improves free-fattyacid-induced insulin resistance in L6 myotubes through inhibiting peroxisome proliferator-activated receptor gamma and fatty acid transferase expressions. Metabolism 58, 1694–1702. Dong, J., Zhang, X., Zhang, L., Bian, H.X., Xu, N., Bao, B., Liu, J., 2014. Quercetin reduces obesity-associated ATM infiltration and inflammation in mice: a mechanism including AMPKα 1/SIRT1. J. Lipid Res. 55, 363–374. Feinstein, R., Kanety, H., Papa, M.Z., Lunenfeld, B., Karasik, A., 1993. Tumor necrosis factor-alpha suppresses insulin-induced tyrosine phosphorylation of insulin receptor and its substrates. J. Biol. Chem. 268, 26055–26058. Grahame Hardie, D., 2014. AMP-activated protein kinase: a key regulator of energy balance with many roles in human disease. J. Intern. Med. 276, 543–559. Greenberg, A.S., Obin, M.S., 2006. Obesity and the role of adipose tissue in inflammation and metabolism. Am. J. Clin. Nutr. 83, 461S–465S. Gual, P., Le Marchand-Brustel, Y., Tanti, J.F., 2005. Positive and negative regulation of insulin signaling through IRS-1 phosphorylation. Biochimie 87, 99–109. Hawley, S.A., Fullerton, M.D., Ross, F.A., Schertzer, J.D., Chevtzoff, C., Walker, K.J., Peggie, M.W., Zibrova, D., Green, K.A., Mustard, K.J., Kemp, B.E., Sakamoto, K., Steinberg, G.R., Hardie, D.G., 2012. The ancient drug salicylate directly activates AMP-activated protein kinase. Science 336, 918–922. Kalupahana, N.S., Moustaid-Moussa, N., Claycombe, K.J., 2012. Immunity as a link between obesity and insulin resistance. Mol. Asp. Med. 33, 26–34. Lafontan, M., 2014. Adipose tissue and adipocyte dysregulation. Diabetes Metab. 40, 16–28. Lee, M.S., Lee, C.M., Cha, E.Y., Thuong, P.T., Bae, K., Song, I.S., Noh, S.M., Sul, J.Y., 2010. Activation of AMP- activated protein kinase on human gastric cancer cells by apoptosis induced by corosolic acid isolated from Weigela subsessilis. Phytother. Res. 24, 1857–1861.

Li, Y., Li, J.J., Wen, X.D., Pan, R., He, Y.S., Yang, J., 2014. Metabonomic analysis of the therapeutic effect of Potentilla discolor in the treatment of type-2 diabetes mellitus. Mol. Biosyst. 10, 2898–2906. Liu, K., Luo, T., Zhang, Z., Wang, T., Kou, J., Liu, B., Huang, F., 2011. Modified Si-MiaoSan extract inhibits inflammatory response and modulates insulin sensitivity in hepatocytes through an IKKβ /IRS-1/Akt-dependent pathway. J. Ethnopharmacol. 136 473–139. Miura, T., Itoh, Y., Kaneko, T., Ueda, N., Ishida, T., Fukushima, M., Matsuyama, F., Seino, Y., 2004. Corosolic acid induces GLUT4 translocation in genetically type 2 diabetic mice. Biol. Pharm. Bull. 27, 1103–1105. Miura, T., Ueda, N., Yamada, K., Fukushima, M., Ishida, T., Kaneko, T., Matsuyama, F., Seino, Y., 2006. Antidiabetic effects of corosolic acid in KK-Ay diabetic mice. Biol. Pharm. Bull. 29, 585–587. Sivakumar, G., Vail, D.R., Nair, V., Medina-Bolivar, F., Lay, J.O., 2009. Plant-based corosolic acid: future anti-diabetic drug. Biotechnol. J. 12, 704–711. Stohs, S.J., Miller, H., Kaats, G.R., 2012. A review of the efficacy and safety of banaba (Lagerstroemia speciosa L.) and corosolic acid. Phytother. Res. 26, 317–324. Takagi, S., Miura, T., Ishihara, E., Ishida, T., Chinzei, Y., 2010. Effect of corosolic acid on dietary hypercholesterolemia and hepatic steatosis in KK-Ay diabetic mice. Biomed. Res. 31, 213–218. Wang, M., Gao, X.J., Zhao, W.W., Zhao, W.J., Jiang, C.H., Huang, F., Kou, J.P., Liu, L., Liu, K., 2013. Opposite effects of genistein on the regulation of insulin-mediated glucose homeostasis in adipose tissue. Br. J. Pharmacol. 2, 328–340. Yamaguchi, Y., Yamada, K., Yoshikawa, N., Nakamura, K., Haginaka, J., Kunitomo, M., 2006. Corosolic acid prevents oxidative stress, inflammation and hypertension in SHR/ NDmcr-cp rats, a model of metabolic syndrome. Life Sci. 79, 2474–2479. Yang, Z., Kahn, B.B., Shi, H., Xue, B.Z., 2010. Macrophage alpha1 AMP-activated protein kinase (alpha1AMPK) antagonizes fatty acid-induced inflammation through SIRT1. J. Biol. Chem. 285 19051–11909. Yang, Z., Wang, X., He, Y., Qi, L., Yu, L., Xue, B., Shi, H., 2012. The full capacity of AICAR to reduce obesity-induced inflammation and insulin resistance requires myeloid SIRT1. PLoS One 11, e49935. Yuan, M., Konstantopoulos, N., Lee, J., Hansen, L., Li, Z.W., Karin, M., Shoelson, S.E., 2011. Reversal of obesity- and diet-induced insulin resistance with salicylates or targeted disruption of Ikkß. Science 293, 1673–1677.