Author’s Accepted Manuscript Response of gut microbiota and inflammatory status to bitter melon (Momordica charantia L.) in high fat diet induced obese rats Juan Bai, Ying Zhu, Ying Dong www.elsevier.com/locate/jep
PII: DOI: Reference:
S0378-8741(16)31326-5 http://dx.doi.org/10.1016/j.jep.2016.10.043 JEP10501
To appear in: Journal of Ethnopharmacology Received date: 4 March 2016 Revised date: 13 September 2016 Accepted date: 14 October 2016 Cite this article as: Juan Bai, Ying Zhu and Ying Dong, Response of gut microbiota and inflammatory status to bitter melon (Momordica charantia L.) in high fat diet induced obese rats, Journal of Ethnopharmacology, http://dx.doi.org/10.1016/j.jep.2016.10.043 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Response of gut microbiota and inflammatory status to bitter melon (Momordica charantia L.) in high fat diet induced obese rats Juan Bai, Ying Zhu, Ying Dong* School of Food and Biological Engineering, Jiangsu University, Zhenjiang, 212013, China *Corresponding author: Tel: +86-511-88797202, fax: +86-511-88780201, E-mail: [email protected] Notes: The authors declare no competing financial interest.
1
Abstract Ethnopharmacological relevance Bitter melon (Momordica charantia L.) is rich in a variety of biologically active ingredients, and has been widely used in traditional Chinese medicine (TCM) to treat various diseases, including type 2 diabetes and obesity. Aim of the study We aimed to investigate how bitter melon powder (BMP) could affect obesity-associated inflammatory responses to ameliorate high-fat diet (HFD)-induced insulin resistance, and investigated whether its anti-inflammatory properties were effected by modulating the gut microbiota. Materials and methods Obese SD rats (Sprague–Dawley rats, rattus norregicus) were randomly divided into four groups: (a) normal control diet (NCD) and distilled water, (b) HFD and distilled water, (c) HFD and 300 mg BMP/kg body weight (bw), (d) HFD and 10 mg pioglitazone (PGT)/kg bw. Results We observed remarkable decreases in the fasting glucose, fasting insulin, HOMA-IR index, serum lipid levels, and cell sizes of epididymal adipose tissues in the BMP and PGT groups after 8 weeks. BMP could significantly improve the proinflammatory cytokine tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6), anti-inflammatory cytokine interleukin-10 (IL-10), and local endotoxin levels compared to the HFD group (p < 0.05). BMP suppressed the activation of nuclear factor-κB (NF-κB) by inhibiting inhibitor of NF-κB alpha (IκBα) degradation and phosphorylation of c-Jun 2
N-terminal kinase/ p38 mitogen-activated protein kinases (JNK/p38 MAPKs) in adipose tissue. Sequencing results illustrated that BMP treatment markedly decreased the proportion of the endotoxin-producing opportunistic pathogens and increased butyrate producers. Conclusions These results demonstrate that BMP ameliorates insulin sensitivity partly via relieving the inflammatory status in the system and in white adipose tissues of obese rats, and is associated with a proportional regulation of specific gut microbiota. Keywords: Bitter melon; Obesity; Insulin resistance; Inflammation; Gut microbiota
3
1. Introduction Obesity is a growing epidemic that has been recognized as a global problem. Studies have suggested that obesity is a state of low level inflammation induced by an increased release of proinflammatory cytokines, which influences insulin resistance and progression to type 2 diabetes (T2D) (Ota, 2013). It is well known that food in the intestine interacts with microbial flora which plays a significant role in inflammation. The gut microbiota is a well known factor regulating human health. Evidences indicate that gut microbiota plays a pivotal role in obesity, obesity-associated inflammation, and insulin resistance (Cani et al., 2008; Tremaroli and Bäckhed, 2012; Zhao, 2013). Gut microbiota participates in several host functions, including energy harvest, fat storage, immune response et al.(Shen et al., 2013; Velagapudi et al., 2010). In particular, gut microbiota breaks down undigested food and produces short-chain fatty acids (SCFAs), which have been suggested to supply energy to the intestine and protect the intestinal mucosa (Cani et al., 2008). It has been demonstrated that SCFAs produced by certain strains like Bifidobacterium, enhances the beneficial functions of host epithelial cells (Fukuda et al., 2011). Changed gut microbiota by HFD is now implicated in obesity-associated inflammation and termed “metaflammation,” orchestrated by metabolic cells and tissues in response to excess nutrients and energy (Gregor and Hotamisligil, 2011). Bacterial lipopolysaccharides (LPS), or endotoxins, produced by gram-negative bacterial cell walls, are the main inflammatory factors responsible for the onset of insulin resistance and obesity (Cani et al., 2007). Disruption of the gut microbiota by HFD may release more endotoxin and increase gut permeability, which gradually augments the serum endotoxin level, activates several stress pathways, ultimately leads to the production of inflammatory cytokines (Leonel and Alvarez-Leite, 2012; Muslin, 2008). For example, one study found an overgrowth of Enterobacter cloacae B29, an endotoxin-producing
4
bacterium, in the gut of obese volunteers, which could also induce obesity and insulin resistance in germ-free mice (Fei and Zhao, 2013). Endotoxin-induced inflammation is critical for the development of obesity-associated systemic inflammation, involving the infiltration or expansion of many immune cells, especially in adipose tissues (Lee et al., 2011; Moreno-Navarrete et al., 2012). Elevated endotoxin levels would activate several stress-related signaling pathways such as NF-κB and mitogen-activated protein kinase (MAPK)(Cani et al., 2008), inducing the production of inflammatory cytokines (Muslin, 2008). Both pro- and anti-inflammatory adipokines link the adipose tissue to other insulin-sensitive organs. Selective inhibition of NF-κB in adipose tissues can protect against insulin resistance in nutritional and genetic animal models of obesity (Cai et al., 2005). Bitter melon (Momordica charantia L.) is a common medicinal and edible plant native to the semi-tropical regions of Asia. The BMP we used includes 38.44% dietary fiber, 34.01% carbohydrate, 12.17% protein, 11.75% water, 1.40% ash and 1.05% lipid. And we also determined the microelement, containing 0.63% phenolics, 0.32% flavonoids and 0.21% saponins (Table 1). In addition, a systemic review of different herbal products commonly applied for the treatment of type 2 diabetes confirms the potential blood glucose lowering effects of bitter melon (Yeh et al., 2003). Various experimental studies have proved its potential effects in hypoglycemic, hypolipidemic and improving insulin resistance in the obese (Chan et al., 2005; Chen et al., 2003; Fuangchan et al., 2011; Huang et al., 2008). Related mechanistic studies have found that bitter melon accelerates glucose transporter type 4 (GLUT4) abundance (Wang et al., 2011) and its translocation to the cell membrane to increase the glucose uptake (Tan et al., 2008). The lyophilized powder of bitter melon has been shown to lower the activity of fatty acid synthase and acetyl-CoA carboxylase-1 (ACC-1) (Huang et al., 2008), and upregulate the peroxisome proliferator-activated receptor α (PPARα) and
5
peroxisome proliferator-activated receptor γ (PPARγ) expression levels (Chao et al., 2011) to ameliorate adipogenesis and lipolysis. The alcohol extract of bitter melon greatly enhanced fibroblast growth factor 21 (FGF21) and AMP-activated protein kinase-silent information regulator T1 (AMPK-SIRT1) to improve energy metabolism (Yu et al., 2013). Bitter melon extracts have been reported to strongly suppress LPS-induced inflammation in RAW264.7 cells (Hsu et al., 2013), TNF-α-induced inflammation in FL83B cells (Cheng et al., 2012), Propionibacterium acnes-induced inflammation in THP-1 cells (Hsu et al., 2012), HFD-associated neuroinflammation in C57BL/6 female mice (Nerurkar et al., 2011), and organ inflammation in fructose-fed adult offspring born of fructose-fed dams (Li and CHING, 2014). Studies also have suggested that bitter melon reduces macrophage and mast cell infiltration in adipose tissues of obese mice (Bao et al., 2013) and inhibits NF-κB and JNK pathways to improve insulin resistance and diabetes (Yang et al., 2015). The safety of pioglitazone, an oral antidiabetic agent in the thiazolidinedione class, is controversial. Pioglitazone is effective at reducing hyperglycemia, hyperlipidemia, hyperinsulinemia, and glucose intolerance characterized as insulin resistant states (Ikeda et al., 1990; Lincoff et al., 2007). As pioglitazone also reacts on PPARγ, a master regulator of adipocytes, it is rational that pioglitazone increases bodyweight and subcutaneous adipose tissue weight, which is in accordance with our results in the present study (Hirose et al., 2002). The gut plays a vital role in metabolic disorders and adipose inflammation (Bertin et al., 2010; Lam et al., 2012). From this perspective, it is proposed that a dietary intervention containing bitter melon would modulate the gut microbiota while concomitantly improving intestinal barrier integrity, inflammation, and metabolic phenotypes (Xiao et al., 2014). Therefore, this study was aimed to evaluate the metabolic influence of bitter melon in rats with HFD-induced obesity and focus on 6
determining whether the resultant anti-inflammatory effects on adipose tissues in turn help modulate gut microbiota.
2. Materials and Methods 2.1 Preparation and analysis of chemical components of bitter melon powder (BMP) Fresh bitter melons were collected from Lvjian Agricultural Station (Yangzhong City, China) and authenticated by Jiangsu Academy of Agricultural Science. Unripe bitter melons were washed thoroughly in water and the seeds were removed, and the pulp was thinly sliced. The pulp was freeze-dried in a lyophilizer (FD-8, Beijing, China) at a pressure of 30 Pa for 36 h at 5 °C. The dried materials were then first coarsely milled using a disc mill, and then a Planetary Micro Mill (Changsha mi qi instrument equipment Co., Ltd., Changsha, China), and finally strained through a sieve to separate the granulates (diameter < 100 μm). The chemical profile of BMP was determined according to previous study (Shih et al., 2009). The protein, lipid, water cotent of bitter melon were determined by the Kjeldahl method, the Soxhlet extractor method and the ambient pressure drying method, respectively. The dietary fiber cotent of bitter melon was determined by using a combination of enzymatic and gravimetric procedures according to AOAC. The ash cotent was measured by a combination of ashing and gravimetric procedures. The carbohydrate cotent of bitter melon was the sum of starch and sugar. The starch and total sugar content were determined by enzyme hydrolyzation method and DNS colorimetric method respectively. The total phenolics and flavonoids contents were determined by using the Folin–Ciocalteu method and the aluminum chloride colorimetric method, respectively. The total saponin content was determined used the method described by Xu & Dong (Xu et al., 2005).
7
2.2 Animal studies Male SD rats (Sprague–Dawley rats, rattus norregicus) weighing 200 ± 20 g were obtained from the Laboratory Animal Research Center of Jiangsu University (LARC, Zhenjiang, China) with the license number SCXK (SU) 2013–0011. The rats were caged individually in LARC at 22 ± 2°C with a relative humidity of 40–60%, and artificially illuminated to provide an approximately 12-h light:dark cycle, and provided free access to standard laboratory diet and water. The rats were randomly separated into two groups: normal control and obese groups. To establish the obese rat model, rats in the control group were fed a basic diet (NCD; 70% corn, 1.8% bean pulp, 17% fishmeal, 5% grass powder, 2% yeast powder, 1.9% vegetable oil, 0.4% essential amino acid, 0.8% vitamin, 0.7% mineral substance, 0.2% CaHPO4, 0.2% salt), and rats in the experimental group were fed a high-fat diet (HFD; containing 73% basic diet, 12% lard, 10% sucrose, and 5% yolk powder) for 8 weeks. HFD were prapared by mixing the basic diet, lard, sucrose, yolk powder with proper water thoroughly, baking in the oven for a certain time and finally cutting uniformly. The body weight, body length, and oral glucose tolerance of all the rats were measured after 8 weeks, and the rats displaying 20% higher body weight than the rats in the normal group and lower glucose tolerance were selected for use as the obese rat model for experiments. The experimental rats given normal control diet (NCD) and HFD were divided into four groups (n = 8 each): in Group 1, normal rats were given a NCD and 4 mL distilled water/kg body weight (bw); Group 2, obese rats were given a HFD and 4 mL distilled water/kg bw; Group 3, obese rats were given a HFD and 300 mg BMP/kg bw; Group 4, obese rats were given a HFD and 10 mg pioglitazone (PGT)/kg bw (Jiangsu De Yuan Pharmaceutical Co., Ltd., China). After an additional 8 weeks of treatment, fecal samples were collected from the rats immediately after defecation, maintained at 4°C, brought to the laboratory, and stored at –70°C for 8
further analysis. During the 8-week experimental treatment period, the body weights of each rat and the volumes of food/water intake were measured weekly. Then, the animals were anesthetized with chloral hydrate. Blood samples were collected via abdominal aorta puncture. The blood samples were centrifuged at 3500 rpm for 10 min to obtain serum, which was stored at −20 °C for further analysis. After blood collection, the kidney, lievr, spleen and some other tissues were excised, weighed and stored at −80 °C until further use.
2.3 Oral glucose tolerance test (OGTT) OGTT was performed before the start and after 16 weeks of feeding and gavage. All the rats were deprived of feed overnight before OGTT. Blood was taken from the tail before and at 30, 60, and 120 min after gavage with 2 g oral glucose/kg bw. The plasma glucose concentration was determined using One Touch Ultra Blood Glucose Meter (Johnson & Johnson Medical Ltd., Shanghai, China).
2.4 Blood biochemistry The serum samples were assayed for triglycerides (TG), cholesterol (CHOL), high-density lipoprotein cholesterol (HDLC), and low-density lipoprotein cholesterol (LDLC) levels with an Olympus AU2700 Clinical Chemistry Analyzer (Olympus Inc., Japan). The fasting insulin levels were determined using commercial rat insulin ELISA kits (eBioscience, Inc., San Diego, CA, USA). The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as follows: fasting insulin (mU/L) × fasting glucose (mmol/L)/22.5.
2.5 Measurement of plasma endotoxin and other systemic inflammation indices Endotoxin levels were quantified on a Kinetic Turbidimetric LAL Kit (Xiamen Limulus Experimental Reagents Factory, China) according to the manufacturer’s instructions. Briefly, the
9
venous blood sample was centrifuged at 1000 rpm for 10 min, and a 100-μL aliquot of the sample was transferred to a 900-μL sample solution. Samples were incubated at 70℃ for 10 min, and centrifuged at 3500 rpm for 10 min to obtain the supernatant. Limulus reagents were then added into 100 μL of the supernatant, and 200 μL of the reaction mixture was transferred into tubes for further analysis. A standard curve was used to calculate the blood endotoxin concentration, which was assessed by measuring the optical density of the reaction mixture (EU/mL). We measured the plasma TNF-α, IL-6, monocyte chemotactic protein-1 (MCP-1), and IL-10 concentrations using ELISA kits (eBioscience, Inc., San Diego, CA, USA), according to the manufacturer’s instructions.
2.6 Western blot analysis The tissues homogenized with the indicated reagents were lysed in RIPA buffer with protease and phosphatase inhibitors (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) and centrifuged at 10000 rpm for 10 min. The supernatant was decanted. Protein concentrations were determined by the Bradford method with Coomassie brilliant blue. Briefly, the lysate (30 μg protein/lane) was separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a polyvinylidene fluoride (PVDF) membrane. The membranes were blocked with 5% bovine serum albumin/Tris-HCl-buffered saline containing 0.05% Tween-20, and probed with antibodies against IκB, p-NF-κB (p65), JNK1, p-JNK1, p38 MAPK, p-p38 MAPK, and β-actin, and incubated with the membranes at 4°C overnight. Horseradish peroxidase (HRP)-conjugated sheep anti-rat secondary antibodies were diluted in Tris-buffered saline-Tween 20 (TBST) and incubated with the membranes for 2 h at room temperature. Bands were detected using electrochemiluminescence (ECL) reagents (Millipore) according to the manufacturer’s
10
instructions. All primary and secondary antibodies were obtained from Abcam (Abcam, Cambridge, MA). Quantity One software (BioRad) was used to quantify band densities.
2.7 Histopathological analysis of epididymal adipose tissue Epididymal adipose tissues were harvested and fixed in 10% neutral buffered formalin, embedded in paraffin, cut to approximately 4-μm-thick sections, and stained with hematoxylin and eosin. The slides were viewed on a Zeiss Axiovert 40 Microscope (Carl Zeiss, Oberkochen, Germany). Image J software (National Institutes of Health) was used to quantify the adipose cell size in different groups.
2.8 Fecal DNA extraction and sequencing Fresh stool samples from five rats of the same group were obtained immediately after defecation and transported in the shortest possible time to –80 °C for storage until analysis of gut microbiota. The total bacterial DNA was extracted from the fecal samples using QIAamp DNA Stool Mini Kit (Qiagen, Germany). The extracted DNA (30 ng) was used as the template to amplify the V4 region of the 16S rRNA gene using primers 515F (5′-GTGCCAGCMGCCGCGGTAA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′). The products from different samples were sequenced pair end on the MiSeq Illumina Sequencing Platform (San Diego, CA, USA) with the sequencing strategy PE250 (PE + 251 + 8 + 8 + 251; Miseq Reagent kit) (Fadrosh et al., 2014). All high-quality sequencing reads were clustered using USEARCH (v7.0.1090) with 97% similarity, and operational taxonomic units (OTUs) were obtained with a certain threshold. Each OTU was selected as the representative sequence and subjected to RDP Classifer software (v2.2) for taxonomical assignment. The number of sequences per sample was corrected for differences in sequencing depth between samples by rarefication, that is, the same number of reads is randomly subsampled in each sample. Second, the absolute number of sequences of each OTU in each sample 11
was converted to relative abundance to reduce the effect of differences in sequence reads. The representative sequences, together with the abundance data, were used for taxon-based analysis.
2.9 Statistical analysis Data were presented as the means ± standard deviation (SD) and analyzed using SPSS version 14.0 for Windows. One-way analysis of variance with Tukey’s range test was employed to identify differences between independent sample groups. A p value of
PII: DOI: Reference:
S0378-8741(16)31326-5 http://dx.doi.org/10.1016/j.jep.2016.10.043 JEP10501
To appear in: Journal of Ethnopharmacology Received date: 4 March 2016 Revised date: 13 September 2016 Accepted date: 14 October 2016 Cite this article as: Juan Bai, Ying Zhu and Ying Dong, Response of gut microbiota and inflammatory status to bitter melon (Momordica charantia L.) in high fat diet induced obese rats, Journal of Ethnopharmacology, http://dx.doi.org/10.1016/j.jep.2016.10.043 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Response of gut microbiota and inflammatory status to bitter melon (Momordica charantia L.) in high fat diet induced obese rats Juan Bai, Ying Zhu, Ying Dong* School of Food and Biological Engineering, Jiangsu University, Zhenjiang, 212013, China *Corresponding author: Tel: +86-511-88797202, fax: +86-511-88780201, E-mail: [email protected] Notes: The authors declare no competing financial interest.
1
Abstract Ethnopharmacological relevance Bitter melon (Momordica charantia L.) is rich in a variety of biologically active ingredients, and has been widely used in traditional Chinese medicine (TCM) to treat various diseases, including type 2 diabetes and obesity. Aim of the study We aimed to investigate how bitter melon powder (BMP) could affect obesity-associated inflammatory responses to ameliorate high-fat diet (HFD)-induced insulin resistance, and investigated whether its anti-inflammatory properties were effected by modulating the gut microbiota. Materials and methods Obese SD rats (Sprague–Dawley rats, rattus norregicus) were randomly divided into four groups: (a) normal control diet (NCD) and distilled water, (b) HFD and distilled water, (c) HFD and 300 mg BMP/kg body weight (bw), (d) HFD and 10 mg pioglitazone (PGT)/kg bw. Results We observed remarkable decreases in the fasting glucose, fasting insulin, HOMA-IR index, serum lipid levels, and cell sizes of epididymal adipose tissues in the BMP and PGT groups after 8 weeks. BMP could significantly improve the proinflammatory cytokine tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6), anti-inflammatory cytokine interleukin-10 (IL-10), and local endotoxin levels compared to the HFD group (p < 0.05). BMP suppressed the activation of nuclear factor-κB (NF-κB) by inhibiting inhibitor of NF-κB alpha (IκBα) degradation and phosphorylation of c-Jun 2
N-terminal kinase/ p38 mitogen-activated protein kinases (JNK/p38 MAPKs) in adipose tissue. Sequencing results illustrated that BMP treatment markedly decreased the proportion of the endotoxin-producing opportunistic pathogens and increased butyrate producers. Conclusions These results demonstrate that BMP ameliorates insulin sensitivity partly via relieving the inflammatory status in the system and in white adipose tissues of obese rats, and is associated with a proportional regulation of specific gut microbiota. Keywords: Bitter melon; Obesity; Insulin resistance; Inflammation; Gut microbiota
3
1. Introduction Obesity is a growing epidemic that has been recognized as a global problem. Studies have suggested that obesity is a state of low level inflammation induced by an increased release of proinflammatory cytokines, which influences insulin resistance and progression to type 2 diabetes (T2D) (Ota, 2013). It is well known that food in the intestine interacts with microbial flora which plays a significant role in inflammation. The gut microbiota is a well known factor regulating human health. Evidences indicate that gut microbiota plays a pivotal role in obesity, obesity-associated inflammation, and insulin resistance (Cani et al., 2008; Tremaroli and Bäckhed, 2012; Zhao, 2013). Gut microbiota participates in several host functions, including energy harvest, fat storage, immune response et al.(Shen et al., 2013; Velagapudi et al., 2010). In particular, gut microbiota breaks down undigested food and produces short-chain fatty acids (SCFAs), which have been suggested to supply energy to the intestine and protect the intestinal mucosa (Cani et al., 2008). It has been demonstrated that SCFAs produced by certain strains like Bifidobacterium, enhances the beneficial functions of host epithelial cells (Fukuda et al., 2011). Changed gut microbiota by HFD is now implicated in obesity-associated inflammation and termed “metaflammation,” orchestrated by metabolic cells and tissues in response to excess nutrients and energy (Gregor and Hotamisligil, 2011). Bacterial lipopolysaccharides (LPS), or endotoxins, produced by gram-negative bacterial cell walls, are the main inflammatory factors responsible for the onset of insulin resistance and obesity (Cani et al., 2007). Disruption of the gut microbiota by HFD may release more endotoxin and increase gut permeability, which gradually augments the serum endotoxin level, activates several stress pathways, ultimately leads to the production of inflammatory cytokines (Leonel and Alvarez-Leite, 2012; Muslin, 2008). For example, one study found an overgrowth of Enterobacter cloacae B29, an endotoxin-producing
4
bacterium, in the gut of obese volunteers, which could also induce obesity and insulin resistance in germ-free mice (Fei and Zhao, 2013). Endotoxin-induced inflammation is critical for the development of obesity-associated systemic inflammation, involving the infiltration or expansion of many immune cells, especially in adipose tissues (Lee et al., 2011; Moreno-Navarrete et al., 2012). Elevated endotoxin levels would activate several stress-related signaling pathways such as NF-κB and mitogen-activated protein kinase (MAPK)(Cani et al., 2008), inducing the production of inflammatory cytokines (Muslin, 2008). Both pro- and anti-inflammatory adipokines link the adipose tissue to other insulin-sensitive organs. Selective inhibition of NF-κB in adipose tissues can protect against insulin resistance in nutritional and genetic animal models of obesity (Cai et al., 2005). Bitter melon (Momordica charantia L.) is a common medicinal and edible plant native to the semi-tropical regions of Asia. The BMP we used includes 38.44% dietary fiber, 34.01% carbohydrate, 12.17% protein, 11.75% water, 1.40% ash and 1.05% lipid. And we also determined the microelement, containing 0.63% phenolics, 0.32% flavonoids and 0.21% saponins (Table 1). In addition, a systemic review of different herbal products commonly applied for the treatment of type 2 diabetes confirms the potential blood glucose lowering effects of bitter melon (Yeh et al., 2003). Various experimental studies have proved its potential effects in hypoglycemic, hypolipidemic and improving insulin resistance in the obese (Chan et al., 2005; Chen et al., 2003; Fuangchan et al., 2011; Huang et al., 2008). Related mechanistic studies have found that bitter melon accelerates glucose transporter type 4 (GLUT4) abundance (Wang et al., 2011) and its translocation to the cell membrane to increase the glucose uptake (Tan et al., 2008). The lyophilized powder of bitter melon has been shown to lower the activity of fatty acid synthase and acetyl-CoA carboxylase-1 (ACC-1) (Huang et al., 2008), and upregulate the peroxisome proliferator-activated receptor α (PPARα) and
5
peroxisome proliferator-activated receptor γ (PPARγ) expression levels (Chao et al., 2011) to ameliorate adipogenesis and lipolysis. The alcohol extract of bitter melon greatly enhanced fibroblast growth factor 21 (FGF21) and AMP-activated protein kinase-silent information regulator T1 (AMPK-SIRT1) to improve energy metabolism (Yu et al., 2013). Bitter melon extracts have been reported to strongly suppress LPS-induced inflammation in RAW264.7 cells (Hsu et al., 2013), TNF-α-induced inflammation in FL83B cells (Cheng et al., 2012), Propionibacterium acnes-induced inflammation in THP-1 cells (Hsu et al., 2012), HFD-associated neuroinflammation in C57BL/6 female mice (Nerurkar et al., 2011), and organ inflammation in fructose-fed adult offspring born of fructose-fed dams (Li and CHING, 2014). Studies also have suggested that bitter melon reduces macrophage and mast cell infiltration in adipose tissues of obese mice (Bao et al., 2013) and inhibits NF-κB and JNK pathways to improve insulin resistance and diabetes (Yang et al., 2015). The safety of pioglitazone, an oral antidiabetic agent in the thiazolidinedione class, is controversial. Pioglitazone is effective at reducing hyperglycemia, hyperlipidemia, hyperinsulinemia, and glucose intolerance characterized as insulin resistant states (Ikeda et al., 1990; Lincoff et al., 2007). As pioglitazone also reacts on PPARγ, a master regulator of adipocytes, it is rational that pioglitazone increases bodyweight and subcutaneous adipose tissue weight, which is in accordance with our results in the present study (Hirose et al., 2002). The gut plays a vital role in metabolic disorders and adipose inflammation (Bertin et al., 2010; Lam et al., 2012). From this perspective, it is proposed that a dietary intervention containing bitter melon would modulate the gut microbiota while concomitantly improving intestinal barrier integrity, inflammation, and metabolic phenotypes (Xiao et al., 2014). Therefore, this study was aimed to evaluate the metabolic influence of bitter melon in rats with HFD-induced obesity and focus on 6
determining whether the resultant anti-inflammatory effects on adipose tissues in turn help modulate gut microbiota.
2. Materials and Methods 2.1 Preparation and analysis of chemical components of bitter melon powder (BMP) Fresh bitter melons were collected from Lvjian Agricultural Station (Yangzhong City, China) and authenticated by Jiangsu Academy of Agricultural Science. Unripe bitter melons were washed thoroughly in water and the seeds were removed, and the pulp was thinly sliced. The pulp was freeze-dried in a lyophilizer (FD-8, Beijing, China) at a pressure of 30 Pa for 36 h at 5 °C. The dried materials were then first coarsely milled using a disc mill, and then a Planetary Micro Mill (Changsha mi qi instrument equipment Co., Ltd., Changsha, China), and finally strained through a sieve to separate the granulates (diameter < 100 μm). The chemical profile of BMP was determined according to previous study (Shih et al., 2009). The protein, lipid, water cotent of bitter melon were determined by the Kjeldahl method, the Soxhlet extractor method and the ambient pressure drying method, respectively. The dietary fiber cotent of bitter melon was determined by using a combination of enzymatic and gravimetric procedures according to AOAC. The ash cotent was measured by a combination of ashing and gravimetric procedures. The carbohydrate cotent of bitter melon was the sum of starch and sugar. The starch and total sugar content were determined by enzyme hydrolyzation method and DNS colorimetric method respectively. The total phenolics and flavonoids contents were determined by using the Folin–Ciocalteu method and the aluminum chloride colorimetric method, respectively. The total saponin content was determined used the method described by Xu & Dong (Xu et al., 2005).
7
2.2 Animal studies Male SD rats (Sprague–Dawley rats, rattus norregicus) weighing 200 ± 20 g were obtained from the Laboratory Animal Research Center of Jiangsu University (LARC, Zhenjiang, China) with the license number SCXK (SU) 2013–0011. The rats were caged individually in LARC at 22 ± 2°C with a relative humidity of 40–60%, and artificially illuminated to provide an approximately 12-h light:dark cycle, and provided free access to standard laboratory diet and water. The rats were randomly separated into two groups: normal control and obese groups. To establish the obese rat model, rats in the control group were fed a basic diet (NCD; 70% corn, 1.8% bean pulp, 17% fishmeal, 5% grass powder, 2% yeast powder, 1.9% vegetable oil, 0.4% essential amino acid, 0.8% vitamin, 0.7% mineral substance, 0.2% CaHPO4, 0.2% salt), and rats in the experimental group were fed a high-fat diet (HFD; containing 73% basic diet, 12% lard, 10% sucrose, and 5% yolk powder) for 8 weeks. HFD were prapared by mixing the basic diet, lard, sucrose, yolk powder with proper water thoroughly, baking in the oven for a certain time and finally cutting uniformly. The body weight, body length, and oral glucose tolerance of all the rats were measured after 8 weeks, and the rats displaying 20% higher body weight than the rats in the normal group and lower glucose tolerance were selected for use as the obese rat model for experiments. The experimental rats given normal control diet (NCD) and HFD were divided into four groups (n = 8 each): in Group 1, normal rats were given a NCD and 4 mL distilled water/kg body weight (bw); Group 2, obese rats were given a HFD and 4 mL distilled water/kg bw; Group 3, obese rats were given a HFD and 300 mg BMP/kg bw; Group 4, obese rats were given a HFD and 10 mg pioglitazone (PGT)/kg bw (Jiangsu De Yuan Pharmaceutical Co., Ltd., China). After an additional 8 weeks of treatment, fecal samples were collected from the rats immediately after defecation, maintained at 4°C, brought to the laboratory, and stored at –70°C for 8
further analysis. During the 8-week experimental treatment period, the body weights of each rat and the volumes of food/water intake were measured weekly. Then, the animals were anesthetized with chloral hydrate. Blood samples were collected via abdominal aorta puncture. The blood samples were centrifuged at 3500 rpm for 10 min to obtain serum, which was stored at −20 °C for further analysis. After blood collection, the kidney, lievr, spleen and some other tissues were excised, weighed and stored at −80 °C until further use.
2.3 Oral glucose tolerance test (OGTT) OGTT was performed before the start and after 16 weeks of feeding and gavage. All the rats were deprived of feed overnight before OGTT. Blood was taken from the tail before and at 30, 60, and 120 min after gavage with 2 g oral glucose/kg bw. The plasma glucose concentration was determined using One Touch Ultra Blood Glucose Meter (Johnson & Johnson Medical Ltd., Shanghai, China).
2.4 Blood biochemistry The serum samples were assayed for triglycerides (TG), cholesterol (CHOL), high-density lipoprotein cholesterol (HDLC), and low-density lipoprotein cholesterol (LDLC) levels with an Olympus AU2700 Clinical Chemistry Analyzer (Olympus Inc., Japan). The fasting insulin levels were determined using commercial rat insulin ELISA kits (eBioscience, Inc., San Diego, CA, USA). The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as follows: fasting insulin (mU/L) × fasting glucose (mmol/L)/22.5.
2.5 Measurement of plasma endotoxin and other systemic inflammation indices Endotoxin levels were quantified on a Kinetic Turbidimetric LAL Kit (Xiamen Limulus Experimental Reagents Factory, China) according to the manufacturer’s instructions. Briefly, the
9
venous blood sample was centrifuged at 1000 rpm for 10 min, and a 100-μL aliquot of the sample was transferred to a 900-μL sample solution. Samples were incubated at 70℃ for 10 min, and centrifuged at 3500 rpm for 10 min to obtain the supernatant. Limulus reagents were then added into 100 μL of the supernatant, and 200 μL of the reaction mixture was transferred into tubes for further analysis. A standard curve was used to calculate the blood endotoxin concentration, which was assessed by measuring the optical density of the reaction mixture (EU/mL). We measured the plasma TNF-α, IL-6, monocyte chemotactic protein-1 (MCP-1), and IL-10 concentrations using ELISA kits (eBioscience, Inc., San Diego, CA, USA), according to the manufacturer’s instructions.
2.6 Western blot analysis The tissues homogenized with the indicated reagents were lysed in RIPA buffer with protease and phosphatase inhibitors (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) and centrifuged at 10000 rpm for 10 min. The supernatant was decanted. Protein concentrations were determined by the Bradford method with Coomassie brilliant blue. Briefly, the lysate (30 μg protein/lane) was separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a polyvinylidene fluoride (PVDF) membrane. The membranes were blocked with 5% bovine serum albumin/Tris-HCl-buffered saline containing 0.05% Tween-20, and probed with antibodies against IκB, p-NF-κB (p65), JNK1, p-JNK1, p38 MAPK, p-p38 MAPK, and β-actin, and incubated with the membranes at 4°C overnight. Horseradish peroxidase (HRP)-conjugated sheep anti-rat secondary antibodies were diluted in Tris-buffered saline-Tween 20 (TBST) and incubated with the membranes for 2 h at room temperature. Bands were detected using electrochemiluminescence (ECL) reagents (Millipore) according to the manufacturer’s
10
instructions. All primary and secondary antibodies were obtained from Abcam (Abcam, Cambridge, MA). Quantity One software (BioRad) was used to quantify band densities.
2.7 Histopathological analysis of epididymal adipose tissue Epididymal adipose tissues were harvested and fixed in 10% neutral buffered formalin, embedded in paraffin, cut to approximately 4-μm-thick sections, and stained with hematoxylin and eosin. The slides were viewed on a Zeiss Axiovert 40 Microscope (Carl Zeiss, Oberkochen, Germany). Image J software (National Institutes of Health) was used to quantify the adipose cell size in different groups.
2.8 Fecal DNA extraction and sequencing Fresh stool samples from five rats of the same group were obtained immediately after defecation and transported in the shortest possible time to –80 °C for storage until analysis of gut microbiota. The total bacterial DNA was extracted from the fecal samples using QIAamp DNA Stool Mini Kit (Qiagen, Germany). The extracted DNA (30 ng) was used as the template to amplify the V4 region of the 16S rRNA gene using primers 515F (5′-GTGCCAGCMGCCGCGGTAA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′). The products from different samples were sequenced pair end on the MiSeq Illumina Sequencing Platform (San Diego, CA, USA) with the sequencing strategy PE250 (PE + 251 + 8 + 8 + 251; Miseq Reagent kit) (Fadrosh et al., 2014). All high-quality sequencing reads were clustered using USEARCH (v7.0.1090) with 97% similarity, and operational taxonomic units (OTUs) were obtained with a certain threshold. Each OTU was selected as the representative sequence and subjected to RDP Classifer software (v2.2) for taxonomical assignment. The number of sequences per sample was corrected for differences in sequencing depth between samples by rarefication, that is, the same number of reads is randomly subsampled in each sample. Second, the absolute number of sequences of each OTU in each sample 11
was converted to relative abundance to reduce the effect of differences in sequence reads. The representative sequences, together with the abundance data, were used for taxon-based analysis.
2.9 Statistical analysis Data were presented as the means ± standard deviation (SD) and analyzed using SPSS version 14.0 for Windows. One-way analysis of variance with Tukey’s range test was employed to identify differences between independent sample groups. A p value of
