Effects and mechanisms of a concentrated goyasaponin fraction in 3t3 l1 cell line

First of all, insulin can stimulate the peripheral glucose uptake in muscle around 60% of insulin-stimulated whole body glucose uptake, liver around 30% and adipose tissue about 10% [12]

CHAPTER 1

LITERATURE REVIEW AND OVERALL HYPOTHESES

1 Type 2 Diabetes and Obesity

1.1 Type 2 Diabetes

Diabetes mellitus is characterized by abnormally high blood glucose levels

(hyperglycemia) and caused by an altered secretory amount of insulin from pancreas as well

as decreased effectiveness in insulin action [1, 2] According to recent classification, Type 1 and Type 2 are two main types of this chronic disease [2, 3] Type 1 diabetes, also known as juvenile-onset or insulin-dependent diabetes, is an autoimmune disease in which pancreatic islet cells are destroyed by antibodies produced by the body [2, 4] Therefore, absolute deficiency in insulin production by the pancreas is always associated with Type 1 diabetes [3] Type 2 diabetes, also known as adult-onset or non-insulin dependent diabetes, is the consequence of altered insulin secretion as well as resistance to insulin action [2, 4]

The prevalence of type 2 diabetes, which accounts for approximately 90 - 95% of all diagnosed cases of diabetes, is increasing worldwide and the global diabetic population is estimated to double from 151 million in 2000 to 300 million in 2025 [5, 6] Meanwhile, annual medical expenditure associated with type 2 diabetes imposed a significant financial burden on society In United States, type 2 diabetes accounts for more than $100 billion in healthcare costs annually [2] while in western European countries, 2-7% of total national health budgets has been spent on diabetes care [7] Even more disturbing is the growing number of children and adolescents diagnosed with type 2 diabetes which was previously present in adults only [7, 8] Additionally, as diabetes mellitus can lead to serious long-term

complications including heart disease, retinal damage, renal failure and stroke [9], it was ranked sixth as the leading cause of death in United States in 2004 [10]

In normal subjects, the plasma glucose levels are maintained with the aid of insulin [11] Insulin is a multipotent hormone secreted from β-cell in pancreas and its most

important effect is on glucose homeostasis [12] First of all, insulin can stimulate the peripheral glucose uptake in muscle (around 60% of insulin-stimulated whole body glucose uptake), liver (around 30%) and adipose tissue (about 10%) [12] At the same time, insulin also stores excess glucose as glycogen in liver and muscle [13, 14] Insulin resistance, characterized as defects in insulin signaling in insulin-responsive tissues, occurs many years prior to diabetes onset and serves as a strong predictor for the type 2 diabetes development [15, 16] Therefore, both normal insulin secretion and sensitivity to insulin action are of importance for maintenance of plasma glucose level

The developmental process of type 2 diabetes is a complex and multifactorial

consequence as shown in Figure 1 proposed by Olefsky et al in 1995 [16] Several risk

factors, including both genetic and acquired factors, make significant contributions to type 2 diabetes In addition, further new insights into mechanism in the future would provide a better understanding of type 2 diabetes pathogenesis

Figure 1 Pathology of type 2 diabetes concluded by Olefsky et al in 1995

1.2 Obesity

Obesity is another chronic disease associated with the excessive presence of body fat and it has become a leading public health problem in the world [17] Similar to type 2

diabetes, increased prevalence of obesity among both adults and children has been observed

in many countries throughout the world [18, 19] According to National Centre for Health Statistics, it is reported that 61% of adults are overweight and 26% are obese in United States

in 1999, as defined by body mass index (BMI) [20] Globally, the prevalence of obesity has increased more than 75% since 1980 [17]

Obesity is a consequence of an imbalance in energy metabolism resulting from

excessive food intake concurrent with decreased energy expenditure [17] There are several explanations for this global increase in obesity rate First of all, genetic factor makes

dominant contribution to body weight as inheritability for obesity is estimated to be 50-90% [17] Moreover, some acquired factors, including obesity-promoting changes in diet and sedentary lifestyle, also exacerbate this increased prevalence [17]

1.3 Close Connection between Type 2 Diabetes and Obesity

Obesity is closely associated with insulin resistance and is considered to be a leading risk factor for both type 2 diabetes and cardiovascular disease [21, 22] Recent

insulin resistance and impaired glucose tolerance develop in response to physiological

dysfunction in muscle, liver and pancreas where excess lipids are deposited [24-26]

1 Increased number of adipocytes (hyperplasia)

2 Adipose Tissue and 3T3-L1 Cell Line

2.1 Adipose Tissue

Mature adipocytes are the main cellular component of adipose tissue In addition, adipose tissue also contains other components, such as undifferentiated preadipocytes, immune cells (leukocytes, macrophages), nerve fibers, vascular stroma, lymph nodes and a matrix of collagen and reticular fibers [27]

There are essentially two types of adipose tissue in humans referred to as white

adipose tissue (WAT) and brown adipose tissue (BAT) [27, 28] BAT is found in fetuses and newborn infants and is practically absent in adults as its principal function is to burn fat to generate heat for newborns during the initial hours after birth [27, 28] Uncoupling Protein-1 (UCP-1), exclusively expressed in brown adipocytes, plays a vital role in heat production as

it uncouples electron transport from adenosine-5'-triphosphate (ATP) production allowing energy to dissipate as heat [27]

However, the predominant type of adipose tissue in humans is WAT [29] Two types

of WAT exist and are classified as visceral and subcutaneous [30] Functional differences exist between these two types and it appears that individuals with visceral fat accumulation are more likely to develop metabolic and cardiovascular diseases [30] In response to energy demands, WAT serves as a site for storing excess energy as triglyceride via lipogenesis and

regulation, lipid metabolism, insulin resistance, immunological response and vascular disease [32-34] Moreover, WAT also expresses and secretes some other cytokines and chemokines, such as tumor necrosis factor-alpha (TNF-α), interleukin-6 (IL-6) and monocyte

chemoattractant protein-1 [35]

Besides these two roles of adipose tissue as mentioned above, Cao et al in 2008 found

that palmitoleate, a free fatty acid metabolite generated in and released from adipocytes, is a circulating factor that promotes the insulin sensitivity in liver and muscle [36] This latest finding is a further step to demonstrate that lipid metabolism and glucose homeostasis are highly interconnected processes [37]

Therefore, due to its multifunctional characters, WAT has been focused by

researchers as a possible central mediator of whole body insulin resistance in recent years [38]

2.2 3T3-L1 is an established in vitro model of adipocyte biology

The molecular and cellular events during the adipogenesis process have been studied

on various cell culture models, including both preadipocyte cell lines and primary culture of adipose-derived stromal vascular precursor cells [32] 3T3-L1 is a substrain of Swiss 3T3 murine cell line derived from disaggregated 17- to 19- mouse embryos [39] It has been proved to be an established murine preadipocyte fibroblast which can be induced to

adipocyte differentiation (adipogenesis) and completely convert into oil droplet-containing mature adipocytes [22] Therefore, both differentiation process and adipocytokines secretion from mature adipocyte can be investigated under this cell line condition [22, 40]

Progress has been made in understanding adipocyte differentiation process First of all, these new findings provided the molecular and cellular basis of the adipose tissue growth of physiological and pathophysiological state Additionally, they also provided means of

developing strategies for both prevention and treatment of obesity [39] Characterized by increased adipose tissue mass, obesity is determined by both enlarged size and increased number of adipocytes [41] Therefore, several applicable anti-obesity mechanisms, including decreased preadipocyte proliferation, inhibition of adipocyte differentiation, reduced

lipogenesis, increased lipolysis and enhanced free fatty acid oxidation were proposed by

Wang et al [42]

2.3 Peroxisome Proliferator-Activated Receptor γ (PPARγ)

Peroxisome proliferator-activated receptors (PPARs) constitute a subfamily of nuclear hormone receptors which regulate storage and catabolism of dietary fats [43-45] There are three subunits of PPARs: α, δ and γ [46] Among those PPAR subunits, the expression of PPARγ exhibits the predominant specificity in adipose tissue while smaller amounts are also present in skeletal muscle, liver, pancreatic β-cells, vascular endothelial cells and

macrophages [46, 47]

The roles of PPARγ in both adipogenesis and differentiated adipocytes have been

essential and irreplaceable role in the adipocyte differentiation [49, 50] Secondly, PPARγ

also play an important role in mature differentiated adipocytes Tamori et al found PPARγ

functions at least in part by regulating relevant gene expressions to maintain the

characteristics of mature adipocytes including free fatty acid (FFA) uptake and triglyceride

accumulation [51] Similarly, Way et al also found that PPARγ activation with the potent

PPARγ ligand GW1929 stimulated the expression of genes involved in lipogenesis and fatty acid metabolism in adipose tissue in Zucker diabetic fatty rat [52]

The therapeutic usage of PPARγ agonists is mainly focused on type 2 diabetes treatment PPARγ has been identified as the receptor for thiazolidinediones (TZDs) and the antidiabetic effects of TZDs are mediated through PPARγ [4, 53, 54] Therefore, adipose tissue, where PPARγ predominantly expressed, has become a main target tissue of TZDs However, PPARγ agonists also bring in a paradox On one hand, PPARγ agonists effectively reduce plasma glucose and ameliorate insulin resistance [55] On the other hand, PPARγ agonists also promote adipocyte differentiation which potentially leads to obesity, a major risk factor for the development of type 2 diabetes One of postulated explanations to this paradox is that PPARγ activation in rodents induced increased number of small adipocytes which typically demonstrate greater insulin sensitivity, more glucose uptake and lower rates

of lipolysis when compared to large adipocytes [55-57] However, further investigations are still in need to clarify this issue

2.4 Adiponectin

Adiponectin is predominantly secreted from mature adipocytes to influence insulin sensitivity by improving glucose and lipid metabolism [58, 59] Adiponectin has been

regarded as a clinical marker of type 2 diabetes as high levels of adiponectin are associated with reduced risk of diabetes while reduced levels of adiponectin have been observed in type

2 diabetics and obese patients [60, 61] Moreover, adiponectin also has become one of therapeutic targets for type 2 diabetes First of all, adiponectin has been shown to suppress hepatic glucose production [62] Moreover, via activating AMP-activated protein kinase (AMPK) pathway, adiponectin increases glucose uptake and stimulates fatty acid oxidation [63, 64]

3 Antidiabetic Drugs: Thiazolidinediones (TZDs) and Metformin

Besides the use of insulin in the treatment of diabetes, oral hypoglycemic agents such

as sulphonylureas, thiazolidinediones (TZDs), α-glucosidase inhibitor, and metformin are also used to regulate plasma glucose level [65] The principal modes of action of these

antidiabetic agents are listed in Table 1 [4, 54, 66]

Class Molecular Target(s) Sites of Action Main mode of action Adverse Effects

Biguanide (Metformin) Unknown Liver, muscle

Decrease hepatic glucose output;

Increase peripheral glucose uptake

Gastrointestinal Disturbances; lactic acidosis

TZDs

(Pioglitazone,

rosiglitazone) PPARγ Fat, liver, muscle

Increase insulin sensitivity

Weight gain; Oedema;

Gastrointestinal disturbances

Insulin Insulin receptor Fat, muscle, liver

Decrease hepatic glucose output;

Increase peripheral glucose uptake;

Decrease lipolysis

Hypoglycaemia; Weight Gain

Table 1 Current therapeutic agents for type 2 diabetes

TZDs and metformin are two important classes of drugs for the treatment of Type 2 diabetes but they counter insulin resistance via different cellular mechanisms [54] Currently, two TZDs, pioglitazone and rosiglitazone remain on the US market while troglitazone was taken off because of liver toxicity [55] TZDs function as PPARγ agonists The main actions

of TZDs include increasing systematic insulin sensitivity, increasing peripheral glucose uptake, reducing plasma fatty acid concentration and enhancing adiponectin secretion [15, 54] Adiponectin influences insulin sensitivity by improving glucose and lipid metabolism and its decreased expression has been reported in models of obesity and diabetes [58, 59] Moreover, TZDs also inhibited secretion of TNF-α, IL-6 and resistin which promoted muscle insulin resistance [67] In addition, TZDs were proved to activate AMPK in rat liver and adipose tissue but it remains unclear whether this is a direct effect and/or mediated by

PPARγ via increasing plasma level of adiponectin [68] However, the safety of earlier

generation TZD has been questioned as well as concerns over common side effects of newer generation of TZD such as weight gain, edema and heart failure [69]

Metformin, on the other hand, achieve hypoglycemic effects principally via

suppressing hepatic glucose output [54] It was found that metformin might mediate its insulin-sensitizing effect by directly activating AMPK pathway in rat liver and muscle [66]

In addition, a clinical study on type 2 diabetes patients demonstrated that metformin caused a

was observed that metformin could not enhance secretion of adiponectin in both in vivo [73] and in vitro experiments [74]

4 Momordica Charantia and Goyasaponins

Although current therapeutic agents for type 2 diabetes are effective in controlling hyperglycemia, they also cause significant side effects as shown in Table 1 Some of the

drugs such as sulphonylureas and TZDs frequently lead to weight gain which may further exacerbate the hyperglycemic conditions [23] In view of these undesirable side effects as well as the increasing prevalence of type 2 diabetes, there is demands to search more

efficacious agents with fewer side effects [38] This fueled the search for alternative

therapeutic substances [75] and led to a rising interest in dietary adjuncts and herbal products that exhibit hypoglycemic properties [76]

4.1 Momordica Charantia

The use of traditional functional foods, such as plants and herbal remedies to treat disease and symptoms has a long history of use in Asia and other developing countries [77,

78] Momordica charantia is commonly known as bitter gourd or bitter melon [77] It is a

tropical climber belonging to the Cucurbitaceae family and has a unique bitter taste and is

cultivated worldwide for its edible fruit [38] Besides being consumed as a vegetable, M

charantia has been used to maintain health, prevent illnesses as well as manage chronic

diseases, most notably diabetes, in the traditional medicinal systems of many cultures

worldwide, including those of the Asian Indians, Chinese and South Americans [79] Besides

anti-diabetic properties, M charantia has also been credited with antiviral, antitumor,

antileukemic, antibacterial, anthelmintic, antimutagenic, antimycobacterial, antioxidant, antiulcer, anti-inflammatory, hypocholesterolemic, hypotriglyceridemic, hypotensive,

immunostimulant, and insecticidal properties [77] However, it has been used extensively as

a folk remedy for diabetes in Asia and is most widely studied with regards to its antidiabetic effects [78, 80, 81]

Previous studies on hypoglycemic activity of M charantia have mainly focused on

chemically-induced diabetic rodents, such as streptozotocin (STZ)-induced rats [82-84]

Momordica charantia extracts (methanol, ethanol and aqueous extracts, or fresh juice) were

found to depress the level of plasma glucose [84, 85], stimulate glucose uptake into skeletal muscle cells after incubation with glucose [86], improve insulin sensitivity by ameliorating insulin signaling cascade in both skeletal muscle [87] and liver [88] of high-rat-red rats, enhance glucose transporter (GLUT4) protein content of plasma membrane in muscle tissue

of rats [89], improve blood glucose tolerance [90], increase the number of pancreatic cells [82], regenerate beta cells in islets of Langerhans in pancreas [91], reduce triglyceride content, decrease LDL-cholesterol and increase HDL-cholesterol [92, 93] Clinical dietary

beta-trials using fruit juice of M charantia have also shown similar serum hypoglycemic effects,

such as decreased serum glucose concentration [94], improved glucose tolerance [94, 95], and a reduction of both fasting and postprandial serum glucose levels [92]

cucurbitane-triterpene glycosides, oleanane-triterpene saponins [97] and proteins including insulin-like polypeptide-p and napin-like protein [98, 99]

4.2 Saponins

Saponins are widely distributed in the plant kingdom and include a structurally diverse group of compounds [100] They are amphiphilic in nature, consisting of triterpenoid, steroidal or steroid alkaloidal aglycones that are substituted with a varying number of sugar

side chains [101-103] M charantia provides a rich source of triterpenoid saponins [104]

The triterpene aglycones share a similar basic structure where the 30 carbon atoms of the 6 linked isoprene units are arranged into 4- or 5- ring structures The sites of glycone attachment may be one (monodesmosides), two (bisdesmosides) or three (tridesmosides) [103] Triterpene saponins are typically bidesmosidic saponins, often with one sugar chain attached through an ether linkage at C-3 and one attached through an ester linkage at C-28[103] The unsubstituted, non-polar aglycones are classified as sapogenins

Triterpenes can be subdivided into around 20 groups depending on their molecular structures, including oleananes and cucurbitanes [104] Saponins were initially a rather neglected area of research primarily because of great difficulties in their isolation and characterization [103] With the advent of more sophisticated methods of isolation and structure elucidation through the last two decades, there has been great interest in these compounds [103] Saponins occur as major constituents in active fractions isolated from many plants used in traditional medicine, and have been shown to possess a large variety of physiobiological activity, including anti-inflammatory, hemolytic, cholesterol lowering, and

anticancer properties [101-103, 105] However, due to the structural complexity of saponins, only a few of these properties are common to all members of this diverse group [100]

4.3 Goyasaponins

Goyasaponin (in Japanese) means saponins extracted from bitter melon

Goyasaponin fraction, as elucidated by Murakami et al [97], includes abundant types of

cucurbitane-triterpene glycosides and oleanane-triterpene saponins, such as goyaglycoside group (a, b, c, d, e, f, g, h), goyasaponin group (I, II, III) and momordicoside group (A, C, F1, F2, I, K, L) The molecular formulae and structures of these compounds are shown in

Figures 3 – 5 Over 40 different cucurbitane-type and oleanane-type triterpene saponins

have been isolated from various parts of the of M charantia plant [106] including the

fruits[97, 106-108], seeds [109] and vines [110]

Recently, significant research interest has focused on isolation, identification and

purification of cucurbitane-type triterpenoids from M charantia Utilizing mass spectrogram (MS) and nuclear magnetic resonance spectroscopy (NMR), there have been a number of reported new discoveries of cucurbitane-type triterpenoids isolated from M charantia [106-

108, 110-112] Moreover, it is noteworthy that one recent report documented that n-butanol

soluble M charantia goyasaponin fraction inhibited sucrose-loading serum glucose elevation

of goyasaponins In addition, there are very few reports identifying the possible mechanisms related to the anti-diabetic action at the cellular level of adipocyte, especially on regulation of adipocyte differentiation and adiponectin secretion

Table 2 Yield percentage of each single goyasaponin compounds from fresh fruit [97]

Compound Yield % from

Fresh Fruit Compound

Yield % from Fresh Fruit Compound

Yield % from Fresh Fruit Goyaglycoside a 0.00008% Goyaglycoside h 0.00008% Goyasaponin I 0.00010%

Goyaglycoside b 0.00005% Momordicoside A 0.0013% Goyasaponin II 0.00027%

Goyaglycoside c 0.00007% Momordicoside C 0.00016% Goyasaponin III 0.00009%

Goyaglycoside d 0.00008% Momordicoside F1 0.00011%

Goyaglycoside e 0.00010% Momordicoside I 0.00006%

H R

O O OH H O O CH3

O OH H OH OH CH3

O O H OH OH

CH3 O

OH H OH OH

O OH O H O OH COOH

O OH H OH OH OH

Goyasaponin I (C65H102O31)

CHO

C

O O

O OH H O O CH3

O OH H OH OH CH3

O O H OH OH

CH3 O

OH H OH O

O OH O H O OH COOH

O OH H OH OH

OH

O

OH H OH OH

Goyasaponin II (C70H110O35)

COOH

Glc - Glc - O

H

OH OH

OR

OGlc OHC

Momordicoside R Molecular Formula

Figure 5 (Continued) Molecular formulae and structures of selected momordicosides

5 Overall Hypotheses, Objectives and Implications of This Study

5.1 Overall Hypotheses

A concentrated goyasaponin fraction (CGF) from M Charantia fruit was obtained

using an extraction and concentration method It is hypothesized that this fruit CGF shows a potential as a PPARγ agonist and plays a similar role as TZDs which can

influence preadipocyte proliferation, process of adipocyte differentiation and adiponectin secretion of differentiated adipocytes by modulating cell signaling In addition, neither

the saponin contents in M Charantia seed nor bioactivity has been investigated

thoroughly It is proposed that seed CGF shows similar effects as fruit CGF on this in

vitro model These findings may be important to elucidate the effects and relevant

mechanisms of M Charantia on anti-obesity and type 2 diabetes prevention and/or

management

5.2 Overall Objectives

M charantia is widely used as a traditional functional food for the treatment of

type 2 diabetes while adipose tissue is one of the targets for anti-diabetic drugs Based on proven hypoglycemic effect of goyasaponin fraction on rats [81], the overall objective of this study is to investigate whether a concentrated saponin fraction (CGF) extracted from

both M charantia fruit and seed show a potential as PPARγ agonist by evaluating their

effects and relevant mechanisms of in the 3T3-L1 murine cell line

(1) To optimize methods of extracting and concentrating saponin fractions from both M

Charantia fruit and seed and to obtain lyophilized CGF in powder for further in vitro

experiments

(2) To investigate the effect of CGF extracted from Momordica charantia fruit and seed

in 3T3-L1 cell model related to diabetes and obesity

(3) To elucidate related mechanisms of these observed activities

5.3 Implications of This Study

There are two implications of this study First of all, because adipose tissue is one

of the target tissues for antidiabetic agents, assessments of both fruit CGF and seed CGF effects on 3T3-L1 preadipocyte proliferation, adipocyte differentiation and adiponectin secretion can shed more light on the underlying mechanisms of the bioactive compounds

responsible for the extensively investigated antidiabetic properties of M charantia Secondly, the discovery of any enhanced bioactivity of M charantia goyasaponins from

fruit or seed will constitute a basis for determination of any potential for its development into a nutraceutical product

(2) Component analysis of CGF by HPLC-MS based on molecular weight confirmation

2 Materials and Methods

2.1 Preparation of Concentrated Goyasaponin Fraction from M charantia Fruit

M charantia fruit was purchased from a local supermarket, washed, separated

from the seeds and aerial fibers, cut into small pieces and lyophilized Dried fruit pieces were powdered and stored at -15℃ until extraction Lyophilized samples were refluxed

in methanol for 4 h, filtered (No.1 Whatman paper, Maidstone, England) and vacuum evaporated The residue was dissolved in deionized water and centrifuged for 5 minutes

at 500 x g The supernatant was then applied equally to five Amberlite XAD4 columns

(Sigma, St Louis, MO) with a bed volume of 140 cm3 each at a flow rate of 3 bed

volume / hour (BV/H) followed by 1 L water wash at a rate of 5 BV/H The sample was eluted by ethanol (500 mL) in each column at a rate of 3 BV/H, and concentrated by vacuum evaporation The residue was dissolved in water and lyophilized and is herein

referred to as the fruit concentrated goyasaponin fraction (Fruit CGF) as shown in Figure

6

Removal of ethanol and dissolved in deionized water

Aqueous extract

Ethanolic extract

Concentrated aqueous extract

Fruit CGF in powder form

Lyophilized M charantia fruit powder

Figure 6 Schematic representation of methodology used in preparation of concentrated

goyasaponin fraction from M charantia fruit

2.2 Preparation of Concentrated Goyasaponin Fraction from M charantia Seed

Sun-dried M charantia seeds were purchased from a local herbal store (Ban

Lee Huat Seed Pte Ltd., Singapore) Seeds (4kg) were grounded and stored at -15°C

Each time, ground M charantia seeds were weighed (35 g), refluxed in 500mL methanol

for 4h The mixture was filtered through filter paper twice (Whatman No.1) and vacuum evaporated Concentrated methanolic extracts from all batches of refluxed extracts were combined before liquid-liquid extraction to minimize variations between batches The concentrated methanol extract (70ml) was first partitioned between 140mL ethyl acetate (EtOAc) (Tedia, Fairfield, USA) and 70mL deionized water The EtOAc phase was discarded while the aqueous phase was further partitioned between 125mL n-butanol (Sigma, St Louis, MO) and 75mL deionized water The aqueous phase was discarded while the n-butanol phase, containing the saponins, was vacuum evaporated The residue

was dissolved in 400ml deionized water and centrifuged for 5 minutes at 500 x g The

supernatant was then applied equally to one Amberlite XAD4 column with a bed volume

of 140 cm3 at a flow rate of 3 bed volume / hour (BV/H) followed by 1 L water wash at a rate of 5 BV/H The sample was eluted by ethanol (500 mL) in each column at a rate of 3 BV/H, and concentrated by vacuum evaporation The residue was dissolved in water and lyophilized and is herein referred to as the seed concentrated goyasaponin fraction (Seed

CGF) as shown in Figure 7

Methanolic extract

Concentrated methanolic extract

Aqueous phase Ethyl acetate phase

(Discarded)

n-butanol phase

Aqueous extract

Ethanolic extract

Reflux with methanol,

Removal of methanol under vacuum

Partition between ethyl acetate and deionized water

Aqueous phase (Discarded)

Removal of n-butanol and dissolved in deionized water

Removal of ethanol and dissolved in deionized water

Ground M charantia seeds

Partition between n-butanol and deionized water

2.3 Identification of Goyasaponin Compounds in CGF from both fruit and seed by High Performance Liquid Chromatography-Mass Spectrum (HPLC-MS)

A Waters Symmetry C18 column (3.9 × 150 mm, 5 µm) was used for the

separation of fruit CGF Fruit CGF powder was dissolved in methanol and syringe filtered (0.45 µm) Solvent A consisted of 0.05 % (v/v) acetic acid in water; solvent B was 0.05 % (v/v) acetic acid in acetonitrile and the column temperature was held constant

at 30 °C The flow-rate was 1 mL/min and the elution program was as follows: Time: 0 -

25 min (95-0% A, 5-100% B), 25 - 30 min (0% A, 100% B), 30 - 35 min (0-95% A, 100 - 5% B) The detection wavelength was 208 nm and injection volume was 20 µL

A Phenomenex Synergi column (RP100A, 2.00 × 50 mm, 2 µm, CA, USA) was used for the separation of seed CGF Seed CGF powder was dissolved in methanol and syringe filtered (0.45 µm) Solvent A consisted of 0.05 % (v/v) acetic acid in water; solvent B was 0.05 % (v/v) acetic acid in acetonitrile and the column temperature was held constant at 30 °C The flow-rate was 0.5 mL/min and the elution program was as follows: Time: 0 - 35 min (95-0% A, 5-100% B), 35 - 40 min (0% A, 100% B), 40 - 45 min (0-95% A, 100 - 5% B) The detection wavelength was 208 nm and injection volume was 20 µL

Detection and molecular weight confirmation of the goyasaponin compounds were established by a LC - MS (Finnigan LCQ quadrupole ion trap) system in both positive and negative mode The MS conditions were set to an ion spray voltage of 4.5

kV with a capillary temperature of 300°Cand capillary voltage of 31 eV, 1 mL/min (Fruit CGF) or 0.5 mL/min (Seed CGF) was delivered to ESI-MS and the remaining diverted to waste The scan mass spectra were in the m/z range of 100–2000

3 Results

The yield of lyophilized CGF powder (1663.62 mg) from fresh fruit (4.915 kg) is approximately 0.03385% while the yield of lyophilized CGF powder (3336 mg) from sun-dried seed (4 kg) is approximately 0.0834%

A variety of goyasaponin compounds were identified in the both fruit CGF and

seed CGF on the basis of molecular fragment match to reported molecular weights in the

literature [97] The suspected compounds contained in the Fruit CGF are listed in Table

3 For a given retention time, up to three compounds in Fruit CGF were identified based

on identical molecular weight Compounds momordicilin, momordicoside A,

goyaglycoside g, momordicoside E were identified individually without overlapping

molecular weights Similarly, suspected compounds in Seed CGF were listed in Table 4

Table 3 Molecular weight confirmation of isolated goyasaponin compounds in fruit

4 Discussion

In the first stage of project, methods for extraction and concentration of

goyasaponin fractions from both M Charantia fruit and seed were investigated and

optimized From HPLC-MS results, a variety of goyasaponin compounds were identified

in both fruit CGF (Table 3) and seed CGF (Table 4) on the basis of molecular fragment

matched to reported molecular weights in the literature In addition, lyophilized CGF

from both fruit and seed were obtained for further in vitro experiments

Further investigation could be focused on isolation and purification of single goyasaponin compounds from both fruit and Seed First of all, a successive ethyl acetate and n-butanol liquid-liquid extraction could be used to remove impurities and

concentrating the saponin fraction in methanolic extract This two-step chromatography separation method, on a laboratory scale, would offer an equally efficient isolation as compared to multiple-steps chromatography that is commonly used Secondly, the cucurbitane-type triterpene glycosides could be separated into its individual components

by normal-phase silica gel followed by HPLC manual collection This method can be scaled-up with ease However, further studies need to be done to optimize the parameters and to reduce the cycle time of isolation When one isolated component was obtained, it could be identified using common spectroscopy methods, such as LC-MS, according to

as enzymatic, acid or basic hydrolysis to confirm the structures of further isolated goyasaponin compounds

CHAPTER 3

EFFECTS OF FRUIT CGF ON 3T3-L1 CELL LINE

1 Hypothesis

CGF from M Charantia fruit influences preadipocyte proliferation; affects

adipocyte differentiation and influences secretion of adiponectin in the 3T3-L1 murine cell model

2 Objectives

First of all, effect of fruit CGF on preadipocytes viability is determined by MTT

assay in order to demonstrate the bioactivity of CGF The mechanisms of this altered preadipocyte proliferation after CGF treatment would be investigated by further LDH assay, apoptosis assay and cell cycle analysis Secondly, effect of fruit CGF on adipocyte differentiation could be displayed via both Oil-Red-O staining and triglyceride content test on adipocytes The mechanism of the fruit CGF treatment in the whole adipogenesis would be demonstrated by PPARγ expression of adipocytes via western blot assay More importantly, the effect of fruit CGF on adiponectin secretion from differentiated

3T3-L1 preadipocytes were purchased from American Type Culture Collection (Manassas, VA) Preadipocytes were incubated in Dulbecco's modified Eagle medium (DMEM) (Sigma, Germany) supplemented with 10% fetal bovine serum (FBS) (Biowest, Miami, Florida), and 1% antibiotic-antimycotic (100 U/mL penicillin, 100 µg/mL streptomycin sulfate, 0.25 µg /mL amphotericin B) (Gibco, Invitrogen,Canada) at 37℃

in a humidified 5% CO2 incubator (Sanyo, Tokyo, Japan) Cells were maintained at a cell concentration not exceeding 6 x 104 cells/mL and were subcultured by total medium replacement using 0.25% (w/v) trypsin - 0.53 mM EDTA (Gibco, Canada) every 2 - 3 days The viable cell number was assessed in quadruplicate by trypan blue exclusion dye using a hemocytometer

Preadipocytes were seeded in 24-well plates at a density of 5 x 104 cells/well for

48 h Adipogenesis (Day 0) was initiated with medium supplemented with 0.5mM Isobutyl-3-methylxanthine (IBMX) (Sigma) and 1 μM dexamethasone (DEX) (Sigma) The culture medium was replaced 48 h later (Day 2) with DMEM containing 10 µg/mL insulin (Sigma) for 48 h replaced thereafter with DMEM every 48 h On Day 8,

1-adipogenesis was complete and adipocytes had acquired intracellular lipid droplets [113]

3.2 Cell Viability (MTT Assay)

Preadipocyte cell viability was assessed by the MTT

(3-(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide) viability assay 3T3-L1 preadipocytes were seeded in 96-well plates (2500 cells/well) and allowed to attach overnight After 24

h, culture medium with CGF (200-400 µg/mL) was added to each well and the cells were incubated for 24, 48 and 72 h and untreated cells acted as control After each period,

medium and test compounds were replaced with 10 µL MTT solution (5 mg/mL MTT in PBS) and 90 µL DMEM to each well After 4 h, 100 µL solubilization solution (10% SDS in 0.01M HCl) was added to each well and incubated overnight as described by Popovich and Kitts [114] The plates were measured by a spectrophotometric micro-plate reader (Thermo Multiskan Spectrum, Helsinki, Finland) at 570 nm with a reference

wavelength of 650 nm Absorbance (A) of each treated group was equal to A570nm – A650nm The preadipocyte viability percentage was expressed by A CGF / A untreated

control x 100% and the 50% Inhibitory Concentration (IC50) values were calculated by linear regression

3.3 Cell Toxicity Test by LDH Assay

LDH was measured as previously described [114] with minor modifications Pyruvate and NADH was purchased from Sigma Tris, EDTA buffer was purchased from National University of Singapore Medical Institute (NUMI) Preadipocytes were seeded

in 24-well plates at a density of 5 x 104 cells/well and allowed to attach overnight CGF extract in culture medium was added at a concentration of 308 µg/mL, equivalent to the

IC50 value determined from a 72-hour MTT assay, described above Untreated

preadipocytes acted as control groups and all cells were incubated at 37℃ in a 5% CO2

velocity was measured in the first three minutes by a spectrophotometric micro-plate reader at 340 nm at 37℃

3.4 Apoptosis Assay and Cell Cycle Analysis

Preadipocytes were seeded in 24-well plates at the density of 5 x 104 cells/well After 24 h, CGF powder was added to each well at a concentration of 200, 300, and 400 µg/mL with the untreated cells acting as a control After treatment for 24, 48 and 72 h, adherent and non-adherent cells were harvested by trypsinization and centrifugation at

400g for 10 minutes The pellet was washed twice in PBS and centrifuged to remove the

PBS The pellet was vortexed vigorously while 1 mL of ice-cold 70% ethanol was added slowly and allowed to fix overnight at 4 ℃ Ethanol was removed by centrifugation at

300g for 10 minutes and 1 mL PBS containing Propidium iodide (PI) (50 µg/mL, Sigma)

and RNase A (100 µg /mL, AppliChem, Germany) was added and incubated for 1 hour Cell cycle was analyzed using a FACS Vantage flow cytometer (Becton-Dickinson, Mountain View, CA) The percentage of each cell cycle phase after acquiring 10,000 cells was analyzed using WinMDI 2.8 software (Joseph Trotter, Scripps, CA)

3.5 Oil-Red-O Staining

Preadipocytes were seeded on quadriPERM 4-well plates (Greiner Bio-One, Germany) and adipocyte differentiation was induced as described above Three CGF concentrations (100, 200 and 300 µg/mL) were tested and untreated cells acted as

controls On Day 8, 10% formalin in PBS was added to replace the cultured medium and allowed to incubate for 5 minutes at room temperature, followed by fresh formalin for 1

hour Each well was washed by 60% isopropanol and allowed to dry, incubated with 60% Oil Red O solution for 10 minutes and washed (5 times) with deionized water Cell morphology was assessed using a CX 31 biological microscope equipped with a C-5060 digital camera (Olympus, Nagano, Japan)

3.6 Cell Lysate Preparation and Protein Concentration Determination

Mammalian Cell Lysis / Extraction Reagent (CelLytic-M) and adipocyte lysate was prepared according to the manufacturer’s instructions (Sigma) Briefly, cells were washed with PBS and 150 µL cell lysis reagent was added to each well and allowed to mix on a rotary stage for 15 minutes The cell lysate was collected and centrifuged at

15000g for 15 minutes to pellet the cellular debris The supernatant was transferred to a

chilled test tube and stored at -80℃ The protein concentration of the cell lysate was determined with Bio-Rad Protein Assay (Bio-Rad, Hercules, CA) according to the

manufacturer’s protocol and were measured by a spectrophotometric micro-plate reader

at 595nm

3.7 Triglyceride Content Test

Preadipocytes were seeded in 24-well plates and adipocyte differentiation was

described above The results were expressed as triglyceride (mg) / protein (mg) in each group

3.8 Expression of PPARγ by Western Blot

Rabbit polyclonal primary antibody of PPARγ, adiponectin, Beta Actin and goat polyclonal secondary antibody to rabbit IgG - H&L horse radish peroxidase (HRP) were purchased from Abcam (Cambridge, UK) The same cellular protein (9 µg per lane) as used in triglyceride content test was separated by sodium dodecyl sulfate -

polyacrylamide gel electrophoresis (SDS-PAGE) using a 12% polyacrylamide separating gel (Mini-PROTEAN Tetra Cell, Bio-Rad, Hercules, CA) The protein was transferred to

a nitrocellulose membrane (ClearPAGE, C.B.S Scientific, Del Mar, CA) by semi-dry transfer (C.B.S Scientific) The membranes were blocked for 1 hour in 5% blotting grade blocker (Bio-Rad) in PBST (0.05% Tween 20 in PBS, pH 7.4) Membranes were incubated with diluted primary antibody (PPAR -γ (1:300), Beta Actin (1:2000)) at room temperature overnight and with the secondary antibody (1:2000) in PBST for 1 hour Between each step, membranes were washed 3 times for 5 minutes each The

antibody/protein complexes were visualized by enhanced chemiluminescence

(SuperSignal West Femto Maximum Sensitivity Substrate, Pierce, Rockford, IL)

Fluorchem FC2 Imaging System (Alpha Innotech, San Leandro, CA) was used to detect

the chemiluminescent signals of target protein bands The expression of proteins was quantified densitometrically using the software GelPro 32 (Media Cybernetics, Bethesda, Md) and expressed relative to reference bands of Beta Actin

3.9 Secretion of Adiponectin from Differentiated Adipocytes by Western Blot

Rabbit polyclonal primary antibody of adiponectin was purchased from Abcam

On Day 8, differentiated adipocytes were washed with serum-free medium twice Four concentrations of CGF (50, 100, 200 and 300 µg/mL) in medium were replaced and incubated for 12h and 24h respectively Untreated cells act as negative control while 10

µM Troglitazone (0.02% dimethyl sulfoxide (DMSO) in medium, v/v) (sigma) acts as positive control Medium of 10 µL was taken for detection of released adiponectin by SDS-PAGE and Western Blot methods as mentioned above with diluted primary

antibody of Adiponectin at 1:2000 The number of viable adipocytes was measured by Trypan Blue exclusion Adiponectin expression was quantified using the software GelPro

32 and normalized to viable cell number

3.10 Statistical Analysis

One-way ANOVA (SPSS 12.0) was used to detect the significant differences (p <

0.05) using the Duncan Post hoc multiple comparisons of observed means Each

experiment was performed in triplicate and the results were expressed as mean ± standard deviation