1
INTRODUCTION
1. The justification of research
Diabetes mellitus (DM, or diabetes for short) is a metabolic disorder
characterized by chronic hyperglycemia- the condition where blood sugar
(glucose) levels are abnormally high, as a result of insulin deficiency or
decreased activity or a combination of both. The consequences of
hyperglycemia are serious complications that can be life-threatening.
Globally, the proportion of people with diabetes in 2016 was 8.5% of the
adult population (422 million people) and the number is expected to rise to
9.9% by 2045. The prevalence of diabetes is more common in low- and
middle-income countries, where more than 80% of deaths occur. The World
Health Organization believes that DM will be one of the seven leading causes
of death by 2030.
Plant species are always a valuable source of medicinal materials, with
many of the drugs currently available on the market derived directly or
indirectly from them. There are at least 1,200 plant species used in traditional
medicine to combat diabetes, but only about 450 plants have been studied
with scientific publications. “Chè dây” plant (Ampelosis cantoniensis (H. &
A.) PL.), belonging to Grape family (Vitacae), although has been investigated
on the chemical composition and antioxidant activity, but so far there has
been no scientific research on anti-diabetic activity. In addition, “lá đắng”
(Vernonia amygdalina Del.), belonging to Aster family (Asteraceae) have had
investigated on the world against diabetes, but the studies are not
comprehensive. Therefore, the search for new diabetes medicines from
natural plants is still highly interesting because the plants can contain new and
safer compounds in the treatment of diabetes.
On that basis, to contribute to the research and development of antidiabetic drugs from plants in the Central region of Vietnam, I have conducted
the thesis: “Studies on isolation and effective treatment for type 2 diabetes

2
mellitus of bioactive compounds from some plant species collected in Central
region”.
2. The objectives of the thesis
- Screening various local plants with hypoglycemic effect on type 2
diabetic mouse model and identify plants with the best hypoglycemic effect,
thereby purifying active compounds.
- Determining their chemical structures, and investigating the mechanism
of action of their hypoglycemia.
- Combining plants that have been studied effectively in the treatment of
diabetes mellitus to form an extract mixture that is more effective in treating
diabetes, thereby studying the mechanism of hypoglycemia and determining
the acute toxicity of the extract mixture.
3. Content of the thesis
- Collecting specimens of 20 species of plants in the Central region, most
of them are considered to be effective in treating diabetes . Extracting of plant
samples with ethanol 70% (v/v) and screening the hypoglycemic effect of the
extracts on type 2 diabetic mouse model.
- Extracting “chè dây” and “lá đắng” with solvents of different polarities
and testing the hypoglycemic abilities of these extract fractions on type 2
diabetic mice.
- Isolate and determine the structure of the compounds from the fractions
of “chè dây” and “lá đắng” extract with the best hypoglycemic activity
- Investigating inhibiting effects of purified compounds on α-glucosidase
and α-amylase enzymes; testing their effects on proinflammatory cytokine
production of TNF-α, IL -6, IL-8 and IL-10, which have been associated with
the development of insulin resistance, and restoring effect on the expression
of cellular pIRS-1 and pY20 proteins, which are involved in the insulin
signaling pathway.
- Studying the hypoglycemic effect of an extract mixture, and the effect of this

mixture on blood cholesterol, triglyceride levels, liver’s glycogen content as well


3
as inhibiting activities of α-glucosidase and α-amylase enzymes in the type 2
diabetic mice. The acute toxicity of this extract mixture has also been assessed.
4. New contributions of the thesis
After a literature review of related researches in the country and the
world, the results of the thesis have made new scientific and practical
contributions as follows:
1. In this research, 5 herbal sources “chè dây” (A. cantoniensis) , papaya
seeds (C. papaya), papaya leaves (C. papaya), leaves and stems of sweet
grass (S. rebaudiana), and “lá đắng” (V. amygdalina Delile) have been
studied the hypoglycemic activity for the first time in Vietnam, particularly
“chè dây” has not been studied for hypoglycemic activity in the world nor has
it been mentioned in folklore about this use.
2. From the CEtOAc fraction of the ethanol extract of A. cantoniensis,
five purified compounds have been isolated, of which quercetin is not a new
substance but is isolated for the first time in this tree.
3. For the first time, an investigation of α-amylase and α-glucosidase
inhibitory activity of compounds isolated from A. cantoniensis has been
carried out.
4. Phloretin compound (isolated from A. cantoniensis) was studied for the
effect of reducing insulin resistance in 3T3-L1 cells caused by TNF-α as well
as restoring the expression of pIRS-1 and pY20 proteins of 3T3-L1 cells.
5. From the CBuOH fraction of the ethanol extract of V. amygdalina, a
new compound, called vernonioside VN (LDB), has been isolated and
determined the structure. This is a novel compound isolated from V.
amygdalina and also isolated from nature for the first time.
6. Vernonioside VN was investigated on the ability to inhibit the

expression of proinflammatory cytokines- TNF-α, IL-6, IL-8, and noninhibiting effect on anti-inflammatory cytokines IL-10.


4
7. Providing the first proof that the ethanol extract of A. cantoniensis (at
the dose of 500mg/kg) enabling the recovery of damaged pancreas and liver
in the type 2 diabetic mouse model.
8. For the first time, developing an extract mixture, which not only has a
good effect on hypoglycemia in type 2 diabetic mice (64.40% reduction
compared with 0 h) but also reduces cholesterol and triglyceride
concentrations, prevents glycogen depletion in liver tissue as well as inhibits
α-glucosidase and α-amylase.
5. The structure of the thesis
The thesis consists of 138 pages. The opening of 04 pages, conclusion and
recommendations of 04 pages, published works of 01 page, 18 pages for
references and appendices. The main content of the thesis is divided into 03
chapters: Chapter 1: Literature Review consists of 40 pages; Chapter 2:
Materials and Research Methods is 21 pages; and Chapter 3: Results and
discussion 50 pages.
CHAPTER 1. LITERATURE REVIEW
The literature review brings together the latest national and international
studies on issues related to diabetes (concept and impact), type 2 diabetes, and
herbal resources in the treatment of diabetes.
CHAPTER 2. MATERIALS AND RESEARCH METHODS
2.1 Materials and subjects of study
- Materials: 20 different plants including: “chè dây” (A. cantoniensis), “lá
vằng” (J. subtriplinerve), “lá nếp” (P. amaryllifolius), “quả chuối hột” (M.
acuminate), “bùm bụp” (P. angulate), papaya seeds (C. papaya), papaya
leaves, figs (F. racemosa), “quả ké đầu ngựa” (X. strumarium), stems and
leaves of “nở ngày đất” (G. globosa), corn silk (Z. mays), “lá đắng” (V.

amygdalina), Breadfruit’s leaves (A. altilis), “giảo cổ lam” (G. pentaphyllum),
“dây thìa canh” (G. sylvestre), lotus leaves (N. nucifera), “lược vàng”
(Callisia fragrans), cinnamon bark (C. loureirii), basil leaves (O.


5
basilicum), stems and leaves of sweet grass (S. rebaudiana). Plant samples
were collected mainly in July-September in 2016-2017, and some samples
were purchased locally.
- Research subjects:
+ Swiss white male mouse, weight from 18-22 g, provided by Suoi Dau
breeding facility - Institute of Vaccines and Biological Products in Nha Trang.
+ Raw macrophage cells 264.7 and 3T3-L1 fat cells were provided by
The National Laboratory of Enzyme Technology of the National University
of Hanoi.
2.2. Research Methods
2.2.1. Extraction method: plant sample processing; solid phase extraction
with ethanol 70% v/v; fractional extraction with solvents of increasing
polarity: n-hexane, ethylacetate, n-buthanol.
2.2.2. Preparation of type 2 diabetic mouse model: raising fat mice through a
high-fat diet (35% fat, 34% carbohydrate, 20% protein and 11% other
ingredients - HFD-high fat diet); determining biochemical indicators
(cholesterol, triglyceride); inducing type 2 diabetic mice by STZ injection at a
dose of 120 mg/kg; quantification of blood glucose, glucose tolerance test, as
well as quantification of blood insulin concentration by ELISA technique.
2.2.3. Screening for hypoglycemic effects of 20 plant samples on the type 2
diabetic mice
2.2.4. Biological slice samples of investigated mice’s liver and pancreas
2.2.5. Investigating the hypoglycemic effect of fractional extracts of “chè
dây” and “lá đắng” in the type 2 diabetic mice

2.2.6. Investigation of the ability to inhibit in-vitro enzymes α-glucosidase
and α-amylase of compounds isolated from “chè dây” and “lá đắng”
2.2.7. Evaluation of the effect on the expression of proinflammatory
cytokines and on the improvement of insulin resistance based on essays in


6
Raw 264.7 and 3T3-L1 cell lines: The process of cell activation; Toxicity
Assessment of purified compounds to the viability of Raw 264.7 and 3T3-L1
cells; Evaluation of the cytokine production by ELISA kit; Method for
assessing the ability to inhibit insulin resistance in cells .
2.2.8. Preparing an extract mixture of herbs with the ability to lower blood
sugar: combining herbal plants to increase the effectiveness of diabetes
treatment; determining glycogen content.
2.2.9. Method of isolation of compounds: thin layer chromatography; liquid
column chromatography
2.2.10. Methods of determining chemical structure
2.2.11. Data processing: The experiments were repeated with appropriate
numbers of times. Data processing using SPSS software (Version 22) with
appropriate statistical tests.
CHAPTER 3. RESULTS AND DISCUSSION
3.1. Results of sample extraction
The results of extracting 20 plant samples with ethanol 70% v/v showed
that different plant samples had different extraction rate into alcohol extract.
Leaves of “chè dây”, and “dây thìa canh”, gave the highest extract yields,
26.00% and 22.62% of total dry matter in the material, respectively. “Lược
vàng”, and papaya seeds had the lowest extract yields, 11.25% and 9.25% of
the total dry matter in the material, respectively.
3.2. Establishment of experimental type 2 diabetic mouse model
Mice were divided into two groups: the “common food” group (ND)

and the “high-fat food” group (which contains 35% fat) called the fat-fed
group (HFD). After 8 weeks of the fat-fed diet, the mice were measured their
blood fat index, then were injected with STZ to cause diabetes and after 10
days of STZ injection, their insulin concentrations were measured to confirm
type 2 diabetes by comparison with the ND group.


7

Figure 3.1: Establishment of experimental type 2 diabetic mouse model
Notes: * indicates p-value < 0,05 (p-value of t-test compared HFD group to
the control one at the same time the test); A. Body weights of mice between
HFD and ND groups; B. Differences in lipid concentrations in HFD and ND
mouse groups; C. Blood glucose levels as the result of HFD and single STZ
injection as compared to the control group; D. Insulin level of different mouse
groups.
The results of Figure 3.1. showed that the change in body weight of mice
after 8 weeks of feeding, the group of mice raised on a high-fat diet (HFD)
had an average weight of 60.42 ± 1.03 g, which was tripled from the baseline
(20.69 ± 0.36 g) and nearly doubled compared to the group that ate standard
food (ND) at the same time (38.95 ± 0.68 g) (panel A). The fat index
measured by cholesterol and triglyceride levels of HFD mice statistically
significantly increased when compared to the ND mouse lot (t-test had p
<0.01) (panel B). The total cholesterol index of the HFD group increased
(78.26%). Similarly, the triglyceride index of HFD increased sharply with the
amount of 160.16%.
After 10 days of injecting STZ, the blood glucose concentrations of the
HFD-fed group at different times after the injection were very high; these
differences were statistically significant when compared with ND + STZ



8
mice (the control group) at the same times (panel C). Thus, the HFD + STZ
group at 10 days had a clear appearance of diabetes when compared with the
remaining non-diabetic mice groups (ND + buffer, ND + STZ and HFD +
buffer) with a strong statistically significance (p-values <0.01).
We found that the HFD + STZ mice had a normal insulin concentration
(18.84 ± 2.11 µIU / ml) compared to the control mice ND, ND + STZ and
HFD (21.10 ± 1, 78; 27.32 ± 3.80 and 27.53 ± 4.25 µIU / ml), the difference
is not statistically significant (panel D). According to Sawant et al., mice with
blood glucose levels ≥ 300 mg / dl and insulin levels between 12.41 and 49.65
µIU / ml were considered type 2 diabetic mice.
3.3. Hypoglycemic effects of 20 plant extracts in the type 2 diabetic mice

Figure 3.2: Blood glucose levels of the type 2 diabetic mice after 21
days of treatment with the plant extracts
The results of Figure 3.2 show that, in the first screening group (A),
drinking the ethanol extracts of “chè dây, and fruits of “chuối hột”, after 21
days, led to the decreases of the blood sugar concentrations by 56.60% and
45.84%, respectively, as compared to the time of 0 hours (and one-way
ANOVA had p <0.01). For the extract of “bùm bụp” fruits, the results


9
showed a hypoglycemic effect compared to the control group, but not very
significant, as the mice after 21 days of treatment had a relatively high blood
glucose level (16.49 mmol/l) (one-way ANOVA p <0.05). The remaining
plant samples, “lá nếp” and “chè vằng”, showed no hypoglycemic effect
when compared to the control sample. In the second screening group (B),
among 5 plant extracts, only the extracts of papaya leaves, papaya seeds, and

figs showed hypoglycemic activity, in which papaya leaves and papaya seeds
had the best activities. The diabetic mice drinking the papaya leaves’ extract at
day 21 had the blood sugar decreased by 47.62%; the ones drinking the
papaya seeds’ extract had a decrease of 53.18% compared to 0 hour. The
third screening group (C) showed significantly hypoglycemic abilities of the
extracts of “dây thìa canh” and of sweet grass with one-way ANOVA had p
<0.01. The diabetic mice drinking “dây thài canh”’s extract had 61.08%
decrease of blood sugar; the mice drinking “cỏ ngọt” ’s extract had a decrease
of 54.93%. For the fourth screening group, among the 5 extracts, only ones of
“giảo cổ lam” and of “lá đắng” showed significant decreases in blood sugar in
the type 2 diabetic mice, 58.86% and 56.21%, respectively (p ≤ 0.01).
Thus, after the screening process of these 20 plant samples, we found that
there were 8 plant samples showing good results in hypoglycemia in the type
2 diabetic mouse models: “chè dây”, “chuối hột”, papaya leaves, papaya
seeds, “dây thìa canh”, sweet grass, “giảo cổ lam”, and “lá đắng”. Reviewing
domestic and international researches, we found that “chè dây, and “lá đắng”
are two of the 8 plant samples with hypoglycemic activity and had not been
studied in Vietnam. In particular, for “chè dây” there is not any research in the
world about its hypoglycemic activity. Therefore, we decided to choose those
two plants for more in-depth research on anti-diabetic activity.


10
3.4.1. Effects of the extrcts of “chè dây” and “lá đắng” on histopathological
results of pancreas and liver
3.4.1.1. Effect on pancreatic tissue structure
To test the effect of the extracts on the status of the pancreatic tissue
structure of the diabetic mice, we conducted a general evaluation of the
histopathology of the mice in experimental batches after administration of
500mg/kg of the extracts with the results are shown in Figure 3.3.


Figure 3.3: Paraffin sections of the pancreas (HE) of STZ-diabetic
mice (original magnification ×400).
Note: A. Non-diabetic control; B. Diabetic control treated with distilled
water; C. Diabetic mice treated with the extract of the leaves A. cantoniensis;
D. Diabetic mice treated with the extract of the leaves V. amygdalina
In the normal mice (A), the islet density was normal, the morphology was
also normal. The islet cells have normal morphology and size, uniformly
distributed without signs of damage. The type 2 diabetic mice drank distilled
water (B) showed the damaged pancreas with the density of pancreatic islet
decreased, pancreatic islet deformed, morphologically and decreased in size.
Comparing with diabetic rats fed with the two extracts, we observed that the


11
density and size of the islet is smaller than normal but the islet showed a
recovery, not shriveled deformation (C, D).
3.4.1.2. Effect on liver tissue structure

Figure 3.4: Paraffin sections of the liver (HE) of STZ-diabetic mice
(original magnification ×400)
Note: A. Non-diabetic control; B. Diabetic control treated with distilled
water; C. Diabetic mice treated with the extract of the leaves A. cantoniensis;
D. Diabetic mice treated with the extract of the leaves V. amygdalina
From the results presented in Figure 3.4, it could be observed that normal
mouse liver cells (A) have a polyhedron shape, in the middle is the nucleus
surrounded by cell membranes. In the nucleus usually had 1-2 large round
nuclei, stained green. Cytoplasm stained by eosin relatively in uniform so all
liver cells were pale pink. The liver cells belonging to each of the Remak rafts
packed closely. Mice with type 2 diabetes drank physiological saline salts (B)

had liver cells showing signs of degeneration, forming clusters of various
shapes, deformed cells spreading around, some cells with nucleus and
cytoplasm were darker, the liver cell cord was severely deformed. Thus, the
injection of STZ after some time has seriously damaged liver cells. The type 2
diabetic mice treated with the extracts of “chè dây” and “lá đắng” (C, D) had


12
the shape and size of the cells and the nucleus of the cells not deformed, the
liver cells still had the nucleus in the middle, the center of the cells and the
nucleus of the cells were normal, the liver cell cord was less deformed. The
liver cells of the type 2 diabetic mice treated with the extracts were less
vulnerable than those of the control and recovered to approximately the same
structure as the normal rat liver samples.
3.4.2. Study on the extract of “chè dây” (A. cantoniensis)
3.4.2.1. Study the hypoglycemic effects of the fractions of A. cantoniensis’ extract
From the study results, from day 7, CEtOAc fraction showed the
hypoglycemic activity (one-way ANOVA p <0.05) and the hypoglycemic
activity was the best at day 21 (p <0.01). CBuOH fraction had a
hypoglycemic effect that started on day 14 and was highest on day 21. CHe
fraction had no hypoglycemic effect, and the mice took that fraction after 21
days showed no change in blood glucose concentration.
3.4.2.2. Isolate and determine the structure of the compounds from the
fractions of Ampelosis cantoniensis’ extract with the best hypoglycemic
activity
* Isolation of substances
From the CEtOAc fraction, 5 purified compounds were isolated and
denoted from CDE1 to CDE5
* Spectral data of the isolated compounds
a. Myricetin (CDE1)

CDE1 compound: the substance was isolated in the form of light yellow
needle crystals, formula: C15H10O8. ESI-MS m / z 319 [M + H] +, m / z 317
[M-H] -. 1H-NMR (CD3OD, 500 MHz, δ: ppm): 6.20 (1H; d; 2.0 Hz; H-6); 6.40
(1H; d; 2.0 Hz; H-8); 7.36 (1H; s; H-2’); 7.36 (1H; s; H-6 ’). 13C-NMR
(CD3OD, 125 MHz, δ: ppm): 148.0 (C-2); 137.3 (C-3); 177.2 (C-4); 162.4 (C-5);


13
99.2 (C-6); 165.5 (C-7); 94.3 (C-8); 158.2 (C-9); 104.5 (C-10); 123.1 (C-1 ’);
108.5 (C-2 ’); 146,7 (C-3 ’); 136,9 (C-4 ’); 146.7 (C-5 ’) and 108.5 (C-6’).
b. Dihydromyricetin (CDE2)
CDE2 compound: isolated in white amorphous powder, formula:
C15H12O8. ESI-MS m / z 342.8 [M + Na] +, m / z 318.8 [M-H] -. 1H-NMR
(acetone-d6, 500 MHz, δ: ppm): 4.88 (1H; d; 11.0 Hz; H-2); 4.52 (1H; d; 11.0
Hz; H-3); 5.91 (1H; d; 2.5 Hz; H-6); 5.95 (1H; d; 2.5 Hz; H-8); 6.59 (1H; s;
H-2’); 6.59 (1H; s; H-6 ’). 13C-NMR (acetone-d6, 125 MHz, δ: ppm): 84,6
(C-2); 72.9 (C-3); 198.0 (C-4); 163.9 (C-5); 95.9 (C-6); 168.1 (C-7); 96.9 (C8); 164.8 (C-9); 101.3 (C-10); 128.7 (C-1 ’); 107.9 (C-2 ’); 146.3 (C-3 ’);
134.3 (C-4 ’); 146,3 (C-5 ') and 107,9 (C-6').
c. Phloretin (CDE3)
CDE3 compound: the substance was isolated in the form of white
amorphous powder, formula: C15H14O. ESI-MS m / z 274.9 [M + H] +, m / z
272.8 [M-H] -. 1H-NMR (CD3OD, 500 MHz, δ: ppm): 7.07 (1H; d; 8.5 Hz;
H-2); 6.72 (1H; d; 8.5 Hz; H-3); 6.72 (1H; d; 8.5 Hz; H-5); 7.07 (1H; d; 8.5
Hz; H-6); 2.88 (2H; dd; 8.0; 7.5 Hz; H-7); 3.33 (2H; dd; 8.0; 7.5 Hz; H-8);
5.84 (1H; s; H-3 ’); 5.84 (1H; s; H-5 ’). 13C-NMR (CD3OD, 125 MHz, δ:
ppm): 134.0 (C-1); 130.3 (C-2); 116.1 (C-3); 156.4 (C-4); 116.1 (C-5); 130.3
(C-6); 31.4 (C-7); 47.2 (C-8); 206.4 (C-9); 105.3 (C-1 ’); 166.1 (C-2 ’); 95.7
(C-3 ’); 165.8 (C-4 ’); 95.7 (C-5 ’) and 166.1 (C-6’).
d. Myricitrin (CDE4)
CDE4 compound: The substance was isolated in the form of a pale yellow

needle, formula: C21H20O12. ESI-MS m / z 464.9 [M + H] +, m / z 462.9 [MH] -. 1H-NMR spectrum (CD3OD, 500 MHz, δ: ppm): 6.22 (1H; d; 2.0 Hz;
H-6); 6.38 (1H; d; 2.0 Hz; H-8); 6.97 (1H; s; H-2 ’); 6.97 (1H; s; H-6 ’); 5.34
(1H; d; 1.5 Hz; H-1 ”); 4.24 (1H; dd; 1.5; 2.0 Hz; H-2 ”); 3.81 (1H; dd; 3.5;
9.5 Hz; H-3 ”); 3.6 (1H; m; H-4 ”); 3.53 (1H; m; H-5 ”); 0.99 (3H; d; 6.0 Hz;


14
H-6 ”). 13C-NMR (CD3OD, 125 MHz, δ: ppm): 158.07 (C-2); 134.9 (C-3);
178.3 (C-4); 161.8 (C-5); 98.4 (C-6); 164,5 (C-7); 93.3 (C-8); 157.1 (C-9);
104.5 (C-10); 120.6 (C-1 ’); 108.2 (C-2 ’); 145.5 (C-3 ’); 136,5 (C-4 ’); 145.5
(C-5 ’); 108.2 (C-6 ’); 102.2 (C-1 ”); 70.5 (C-2 ”); 70.6 (C-3 ”); 72.0 (C-4 ”);
70.7 (C-5 ”) and 16.3 (C-6”).
e. Quercetin (CDE5)
CDE5 compound: The substance was isolated in the form of light yellow
powder, formula: C15H10O7. ESI-MS m / z 302.9 [M + H] +, m / z 300.8 [MH] -. 1H-NMR (CD3OD, 500 MHz, δ: ppm): 6.20 (1H; d; 2.0 Hz; H-6); 6.41
(1H; d; 2.0 Hz; H-8); 7.75 (1H; d; 2.0 Hz; H-2’); 6.91 (1H; d; 8.5 Hz; H-5 ’);
7.65 (1H; dd; 2.5; 8.5 Hz; H-6 ’). 13C-NMR (CD3OD, 125 MHz, δ: ppm):
148.1 (C-2); 137.2 (C-3); 177.3 (C-4); 162.5 (C-5); 99.3 (C-6); 165,6 (C-7);
94.4 (C-8); 158.2 (C-9); 104.5 (C-10); 124.2 (C-1 ’); 116,0 (C-2 ’); 146,2 (C3 ’); 148.8 (C-4 ’); 116,3 (C-5 ') and 121,7 (C-6').
3.4.3. Study on the extract of “lá đắng” (V. amygdalina)
3.4.3.1. Study the hypoglycemic effect of the fraction of “lá đắng” ’s extract
The results showed that CEtOAC and CBuOH fractions had clear
hypoglycemic effects after 7 days of treated mice (p <0.05). The hypoglycemic
effects of these fractions were the highest on days 14 and 21 (p <0.01).
3.4.3.2. Isolate and determine structures of compounds from the fractions of
Vernonia amygdalina’s extract with the best hypoglycemic activities
* Isolation of substances
From the CEtOAc fraction, a purified compound noted LDE was isolated.
From the CBuOH fraction, a purified compound noted LDB was isolated.
* Spectral data of the isolated compounds

a. Cynaroside (LDE)
LDE: The substance was isolated in the form of light yellow needle,
formula: C21H20O11. ESI-MS m / z 448.9 [M + H] +, m / z 446.9 [M-H] -. 1HNMR (DMSO-d6, 500 MHz, δ: ppm): 6.73 (1H; s; H-3); 6.44 (1H; d; 2.0 Hz;


15
H-6); 6.78 (1H; d; 2.0 Hz; H-8); 7.41 (1H; d; 2.0 Hz; H-2 ’); 6.91 (1H; d; 8.5
Hz; H-5 ’); 7.45 (1H; dd; 2.0; 8.5 Hz; H-6 ’); 5.08 (1H; d; 7.5 Hz; H-1 ”); 3.26
(1H; m; H-2 ”); 3.29 (1H; m; H-3 ”); 3.18 (1H; t; 5.0 Hz; H-4 ”); 3.43 (1H;
dd; 1.5; 5.5 Hz; H-5 ”); 3.46 (1H; d; 3.0 Hz; Ha-6 ”); 3.71 (1H; d; 5.5 Hz; Hb6 ”). 13C-NMR (DMSO-d6, 125 MHz, δ: ppm): 164.4 (C-2); 103.1 (C-3);
181.8 (C-4); 161.1 (C-5); 99.5 (C-6); 162.9 (C-7); 94.7 (C-8); 156.9 (C-9);
105.3 (C-10); 121.3 (C-1 ’); 113.5 (C-2 ’); 145.7 (C-3 ’); 149.9 (C-4 ’); 115.9
(C-5 ’); 119.1 (C-6 ’); 99.9 (C-1 ”); 73.1 (C-2 ”); 76.4 (C-3 ”); 69.5 (C-4 ”);
77.1 (C-5 ”) and 60.6 (C-6”).
b. Vernonioside VN (LDB)
LDB: The substance was isolated in the form of white powder, melting
point 253-255°C, formula: C35H54O12. ESI-HR-MS m / z 667.3693 [M + H]
+ (Calcd for C35H55O12: 667.36935). 1H-NMR (DMSO-d6, 500 MHz, δ:
ppm): 1.17 (1H; m; Ha-1); 1.81 (1H; m; Hb-1); 1.44 (1H; m; Ha-2); 1.86
(1H; m; Hb-2); 4.16 (1H; br d; 7.0 Hz; H-3); 1.22 (1H; m; Ha-4); 1,78 (1H;
m; Hb-4); 1.35 (1H; m; H-5); 1.79 (1H; m; Ha-6); 1.91 (1H; m; Hb-6); 5.36
(1H; br s; H-7); 5.48 (1H; d; 7.0 Hz; H-11); 1.94 (1H; m; Ha-12); 2.09 (1H;
m; Hb-12); 2.40 (1H; m; H-14); 1.66 (1H; m; Ha-15); 1.92 (1H; m; Hb-15);
3.55 (1H; m; H-16); 1.87 (1H; m; H-17); 0.47 (3H; s; H-18); 0.82 (3H; s; H19); 2.04 (1H; m; H-20); 5.36 (1H; br s; H-21); 4.63 (1H; br d; 6.0 Hz; H-22);
4.46 (1H; d; 6.0 Hz; H-23); 1.95 (1H; m; H-25); 0.88 (3H; d; 6.5 Hz; H-26);
0.87 (3H; d; 6.5 Hz; H-27); 1.34 (3H; s; H-29); 4.23 (1H; d; 7.5 Hz; H-1 ’);
2.89 (1H; m; H-2 ’); 3.06 (1H; m; H-3 ’); 3.02 (1H; m; H-4 ’); 3.04 (1H; m;
H-5 ’); 3.40 (1H; m; Ha-6 ’); 3.63 (1H; m; Hb-6 ’). 13C-NMR (DMSO-d6,
125 MHz, δ: ppm): 33.7 (C-1); 29.2 (C-2); 74.8 (C-3); 34.1 (C-4); 38.5 (C-5);
29.4 (C-6); 120.5 (C-7); 135.3 (C-8); 143.3 (C-9); 35.5 (C-10); 118.1 (C-11);

40.6 (C-12); 42.7 (C-13); 48.0 (C-14); 34.1 (C-15); 76.3 (C-16); 54.7 (C-17);
13.9 (C-8); 19.2 (C-19); 47.2 (C-20); 98.0 (C-21); 79.3 (C-22); 90.1 (C-23);
80.3 (C-24); 31.7 (C-25); 16.8 (C-26); 18.1 (C-27); 109.1 (C-28); 23.1 (C-


16
29); 100.9 (C-1 ’); 73.5 (C-2 ’); 76.8 (C-3 ’); 70.1 (C-4 ’); 76.7 (C-5 ’) and
61.1 (C-6’).

Figure 3.5: Structure of compound Vernonioside VN
3.4.4. Inhibition of α-amylase and α-glucosidase enzymes of the isolated
compounds
Of the 7 compounds isolated from “chè dây” and “lá đắng”, only 5
compounds exhibited the inhibitory activities of α-amylase and glucosidase.
Results of IC50 values for α-amylase enzyme showed that myricitrin had
the lowest IC value (IC50 = 9.64 ± 0.68 µM), followed by cynaroside,
myricetin and quercetin with IC50 values of: 83.30 ± 3.28, 86.31 ± 4.91, and
136.58 ± 6.77 µM, respectively. Phloretin had an IC50 value of 199.11± 7.60
µM, showing that the ability of phloretin to inhibit α-amylase activity was
weakest. Similarly, the results of the ability to inhibit the enzyme αglucosidase sequentially are myricitrin (8.92 ± 3.65 µM), myricetin (9.20 ±
0.04 µM), quercetin (10.64 ± 1.62 µM), phloretin. (18.74 ± 0.07 µM) and
cynaroside (21.52 ± 2.33 µM).
3.4.5. Anti-inflammatory activities and improved insulin resistance of the
isolated compounds
3.4.5.1. The toxicity effect of the isolated compounds on Raw cells 264.7 and
3T3-L1
Myricetin, dihydromyricetin, phloretin, myricitrin, quercetin, cymaroside,
and Vernonioside VN showed that they were not likely to be toxic to Raw
cells 264.7 and 3T3-L1 at concentrations of 5-40 µg/mL.



17
3.4.5.2. The anti-inflammatory ability of the isolated compounds
The results presented in Figure 3.6 showed that cells incubated with LPS
significantly increased cytokine content (IL-6, IL-8, TNF-α and IL-10) when
compared to controlled cells (not incubated with LPS). Of the 7 compounds
isolated from “chè dây” and “lá đắng”, only phloretin (from “chè dây”) and
Vernonioside VN (from “lá đắng”) had the ability to effectively inhibit the
production of pro-inflammatory cytokines- TNF-α, IL-6, IL-8 and did not
inhibit IL-10 anti-inflammatory cytokines. The expression level of TNF-α in
cells after being treated with phloretin and Vernonioside VN ranged from
317.23 ± 5.39 pg/mL and 311.16 ± 10.94 pg/ mL at the concentration of 5
μg/mL decreased to 211.53 ± 11.63 pg/mL and 219.85 ± 7.14 pg /mL at 40
μg/mL. After treatment of phloretin and Vernonioside VN, IL-6 secretion
was reduced to 211.53 ± 11.64 pg/mL and 219.8 ± 7.14 pg/mL at 40 μg/mL.
They also inhibited IL-8 bioavailability at 40 μg/mL (166.58 ± 10.17 pg/mL
and 170.18 ± 16.84 pg/mL). In contrast, phloretin and vernonioside VN did
not impair the secretion of IL-10 anti-inflammatory cytokines. The antiinflammatory effect of phloretin had been studied in the world, while
Vernonioside VN was reported anti-inflammatory activity for the first time in
this study.

Figure 3.6: Effect of cynaroside and vernonioside V on cytokines
TNF-α, IL-6, IL-8, and Il-10 produced in Raw 264.7 by LPS-induced
inflammation


18
3.4.5.3. Effect of phloretin on reducing insulin resistance in 3T3-L1 cells
induced by TNF-α
Of the test compounds, only phloretin was able to reduce insulin

resistance in 3T3-L1 adipocytes through the ability to absorb sugar and the
results are shown in Figure 3.7.

Figure 3.7: Anti-insulin resistance activity of phloretin in TNF-α
treated 3T3-L1adipocytes
Insulin-resistant adipocytes treated with phloretin (40 µg / mL) were able
to absorb increased glucose (85.22 ± 2.84%) compared to when insulinresistant cells were not treated with phloretin (39.45 ± 2.04%) Phloretin was
first investigated for the effect of reducing insulin resistance in 3T3-L1 cells
induced by TNF-α.
3.4.5.4. Effects of phloretin on expression of p-IRS1 and Y20 in the insulin
signaling pathway
TNF-α has been implicated in the development of insulin resistance. It
breaks down the phosphorylation of several proteins involved in the insulin
signaling pathway, including IRS-1 and Y20 (phosphorylated tyrosine).

Figure 3.8: Band intensities observed via Western blotting, showing
the different expression levels of pIRS1 and pY20, in groups treated with
phloretin and RM in comparison to the normal and insulin-resistant
(IR) groups


19
From the results of Figure 3.8, both proteins (IRS-1 and Y20) were clearly
expressed (after the incubation of phloretin (200 µg/mL) and rosiglitazoneRM (120 𝜇M)) or no expression (no incubation of phloretin) in insulinresistant (IR) cells, this indicated that there was a disruption in the insulin
signaling pathway. Hypoglycemic effect of phloretin by the way of restoring
the expression of the two IRS-1 and Y20 proteins had not been previously
reported.
3.5. Preparation of a novel mixture of the extracts that have the potential
to lower blood sugar
3.5.1. Combining the extracts to increase the effectiveness in the treatment

of diabetes
In this study, we conducted a combination of the studied plants that were
effective in the treatment of diabetes to create the most effective combination
in treating diabetes. When combining the extracts together, each component
used in an equal amount, we found that the mixture of papaya leaves, papaya
seeds, “chuối hột” and “lá đắng” was not affecting the hypoglycemic activity
in the mice, whereas without “chè dây” and sweet grass, the hypoglycemic
activity decreased. This result once again confirms the important role of “chè
dây” in the treatment of diabetes. Extracts of papaya leaves, papaya seeds and
“chuối hột” were removed from the final mixture. In the end, we have
selected 5 species of plants as the ingredients, including: “chè dây”, “lá đắng”,
sweet grass, “giảo cổ lam”, and “dây thìa canh”.
3.5.2. Hypoglycemic effect of the mixture
The results of Figure 3.9 showed that after 14 and 21 days of glycemic
treatment of both groups at doses of 500 mg/kg and 1000 mg/kg (A), the
blood glucose significantly decreased compared to the control group at the
same testing time (p <0.01). After 21 days of treatment with the mixture, the
blood glucose decreased by 64.40% and 58.84%, respectively, compared
with before treatment. In addition, the results showed that the blood glucose


20
levels of the two groups after the treatment of the mixture and pioglite (20
mg/kg) did not differ significantly after 21 days of treatment.

Figure 3.9: Hypoglycemic activity of polyherbal formulation
Notes: * p < 0,05; ** p < 0,01, (p compared to the control group at the
same time the test). A. Blood glucose levels of the type 2 diabetic mice after 21
days of treatment with polyherbal formulation; B. Lipid concentrations of the
type 2 diabetic mice after 21 days of treatment with polyherbal formulation; C.

Hepatic glycogen content in type 2 diabetic mice after treatment with
polyherbal formulation; D. Inhibition of α-amylase and α-glucosidase activities
by polyherbal formulation.
In addition, the type 2 diabetic mice only drank distilled water had the
indicators of cholesterol (4.37 ± 0.23 mmol/l) and triglycerides (2.93 ± 0.13
mmol/l) were very high (B). Meanwhile, the mice treated with the mixture had
cholesterol and triglyceride levels decreased significantly compared to untreated
mice, respectively 1.87 ± 0.16 and 1.4 ± 0.11 mmol/l. In addition, the results
showed that the untreated mice’s liver had glycogen content (C) significantly
decreased when compared to normal mice (46.10% reduction), while the
mixture-treated and the pioglite-treated groups had the glycogen content
increased significantly by 75.62% and 77.27%, respectively, compared to the
untreated group of mice (p <0.01), and there were no statistically significant
difference compared to the glycogen content in normal mice.


21
The mixture showed an ability to inhibit: 84.69±0.51% and 80.09± 2.1% of
the activity of enzyme α-glucosidase and α-amylase at concentration of 100
μg/ml with IC50 were 19.31±1.39 μg/ml and 51.26±3.30 μg/ml, respectively.
The mixture showed stronger inhibition of α-glucosidase and α-amylase
enzyme than each individual extract previously studied.
3.6. Study the acute toxicity of the mixture
At the highest possible oral dose for mice as high as 38.4 mg/kg, there was
no manifestation of acute toxicity of the mixture on tested mice. For medicinal
materials, a LD50 dose is more than 10 times the treatment dose that is
considered to have a good therapeutic safety range, such as the case here of the
mixture. The LD50 dose of the mixture was not determined exactly, because at
very high doses of 38.4 mg/kg, still no animal was dead. This indicated that the
mixture was very safe in acute toxicity test in mice.



22
CONCLUSIONS AND RECOMMENDATIONS
CONCLUSIONS
Based on the results of this project, we can arrive at the following
conclusions:
1. Results of screening of Vietnamese herbal plants for hypoglycemic effect:
8/20 plant samples have been shown to have hypoglycemic effects on a
mouse model of type 2 diabetes mellitus. These plants are “chè dây”, fruits of
“chuối hột”, papaya seeds, papaya leaves, leaves of “dây thìa canh”, leaves and
stems of sweet grass, “giảo cổ lam” and “lá đắng”. “Chè dây”, papaya seeds,
papaya leaves, leaves and stems of sweetgrass are studied for the first time of
hypoglycemic activity in Vietnam; especially, “chè dây” is not known in the
world for hypoglycemic activity before this study.
2. Research results for leaves of “chè dây”:
For the first time, studying the effect of “chè dây” extract (500 mg/kg) on
the recovery of damaged pancreas and liver in the type 2 diabetic mice is carried
out. The two best extract’s fractions for the hypoglycemic effect have been
identified, which are CEtOAc and CbuOH fractions. I have isolated and
identified 05 compounds in the CEtOAc fraction of the “chè dây” extract,
which are myricetin (CDE1), dihydromyricetin (CDE2), phloretin (CDE3),
myricitrin (CDE4), and quercetin (CDE5). In which, quercetin, though not a
new substance, is first isolated from “chè dây”. This study is also the first report
of α-amylase and α-glucosidase inhibitory activity of purified compounds from
“chè dây”. The IC50 values for α-amylase enzyme of myricitrin, myricetin,
quercetin, and phloretin were 9.64±0.68, 86.31±4.91, 136.58±6.77, and
199,11± 7.60 μM, respectively. The IC50 value for α-glucosidase enzyme of
myricitrin is: 8.92± 3.65 µM, of myricetin is 9.20± 0.04 µM, of quercetin is
10.64± 1.62 µM, of phloretin is 18.74± 0.07 µM. These compounds- myricitrin,

myricetin, quercetin, dihydromyricetin, and phloretin do not show any toxicity
on cell line Raw 264.7 and 3T3-L1 at concentrations from 5 to 40 µg/mL.


23
Phloretin has the ability to effectively inhibit the generation of proinflammatory
cytokines TNF-α, IL-6, IL-8, and does not inhibit IL-10 anti-inflammatory
cytokines. This compound, phloretin, was first investigated for the effect of
reducing insulin resistance in 3T3-L1 cells induced by TNF-α and restoring the
expression of pIRS-1 and pY20 proteins of cells.
3. Research results for “lá đắng”:
The ethanol extraction of “lá đắng” at the concentration of 500 mg/kg not
only lower the blood glucose concentration but also restore the pancreas and
liver tissues of the type 2 diabetic mice. The two extract fractions that have the
best effectivity for hypoglycemia in the type 2 diabetic mice are CEtOAc and
CbuOH fractions. I have isolated and identified a compound in the CEtOAc
fraction of “lá đắng”, which is cynaroside (LDE). From the CBuOH fraction, I
have isolated and identified a compound named Vernonioside VN (LDB). This
compound- Vernonioside VN (LDBB) is a novel compound that was isolated
from “lá đắng” for the first time and is also a new compound isolated from
nature. The inhibitory activities of α-amylase and α-glucosidase enzymes of the
isolated cynaroside have been measured and IC50 values were 83.30± 3.28 and
21.52±2.33 µM, respectively. The cynaroside and Vernonioside VN
compounds show no toxicity on Raw cell lines 264.7 and 3T3-L1 at
concentrations of 5-40 µg/mL. Vernonioside VN has the ability to effectively
inhibit the generation of proinflammatory cytokines TNF-α, IL-6, IL-8, and
does not inhibit IL-10 anti-inflammatory cytokines.
4. Results of a new polyherbal mixture (consisting of “chè dây”, “lá đắng”,
sweet grass, “giảo cổ lam”, and “dây thìa canh”) in the treatment of diabetes:
This mixture not only has a good effect on hypoglycemia in the type 2 diabetic

mice (induce a decrease of 64.40% in the blood glucose) but also is able to
reduce cholesterol and triglyceride levels as well as prevent glycogen depletion
in liver tissue. The mixture shows no acute toxicity at a dose of up to 38.4
mg/kg.


24
RECOMMENDATIONS FOR FURTHER RESEARCHES
1. Continue the research on mechanisms of the hypoglycemic effects of the
mixture and isolated active compounds.
2. Research on preparation of the mixture as well as the determination of
microbiological standard, heavy metal content, and other chemical contents in
the mixture.
3. Study of semi-chronic toxicity of the mixture and the isolated compounds
as well as clinical trials through different stages on type 2 diabetic patients in
order to register new drugs.