Available online at www.sciencedirect.

com
ScienceDirect
ScienceDirect
Procedia Computer Science 00 (2018) 000–000
Available online at www.sciencedirect.com
Procedia Computer Science 00 (2018) 000–000 www.elsevier.com/locate/procedia
www.elsevier.com/locate/procedia
ScienceDirect
Procedia Computer Science 135 (2018) 632–642

3rd International Conference on Computer Science and Computational Intelligence 2018

3rd International Conference on Computer Science and Computational Intelligence 2018
In vivo anti-diabetic activity test of ethanol extract of the leaves of
In vivo anti-diabetic activity
Cassiatest of ethanol
Siamea Lamkextract of the leaves of
Cassia Siamea Lamk
Heruna Tantya*, Syarifah Diana Permaib, Herena Pudjihastutib
a
Heruna Tantya*, Syarifah Diana Permaib, Herena Pudjihastutib
Department of Mathematics, School of Computer Science, Bina Nusantara University, Jakarta 11480
a
b
Department
Department ofof Statistics, School
Mathematics, ofof
School Computer Science,
Computer Bina
Science, Nusantara
Bina University,
Nusantara Jakarta
University, 11480
Jakarta 11480
b
Department of Statistics, School of Computer Science, Bina Nusantara University, Jakarta 11480

Abstract
Abstract
Cassia siamea Lamk has been used as a traditional medicine in Indonesia including to treat diabetes mellitus (DM). This study
aimed to
Cassia investigate
siamea Lamk in hasvivo
beenanti-diabetic activities of
used as a traditional the ethanol
medicine extract, ethyl
in Indonesia acetate
including fraction
to treat and n-hexane
diabetes fractionThis
mellitus (DM). of Cassia
study
siameatoLamk
aimed in alloxan-induced
investigate mice. The
in vivo anti-diabetic study concluded
activities that: extract,
of the ethanol (1) the ethanol extractfraction
ethyl acetate and n-hexane fractionfraction
and n-hexane of Cassiaof siamea
Cassia
Lamk
siamea(Juar)
Lamkleaves have anti-diabetic
in alloxan-induced mice.activity in Webster
The study albino
concluded mice
that: (1)induced with extract
the ethanol alloxan,and (2)n-hexane
The extract of Cassia
fraction siameasiamea
of Cassia Lamk
leaves, the fractions
Lamk (Juar) of 500
leaves have mg ethylactivity
anti-diabetic acetateinand 500 mg
Webster n-Hexana
albino of Cassia
mice induced withsiamea
alloxan, Lamk provided
(2) The extract better performances
of Cassia siamea Lamk in
lowering
leaves, theblood glucose
fractions of levels
500 mg compared to Ethanol
ethyl acetate extracts
and 500 both 500 mg
mg n-Hexana and 1000
of Cassia mg. InLamk
siamea the form of ethyl
provided acetate
better and n-hexane
performances in
fraction
loweringatblood
a dose of 150levels
glucose mg/kgcompared
BW provided the highest
to Ethanol extractsanti-diabetic
both 500 mg activity compared
and 1000 mg. Intothetheform
other
of test
ethylgroups
acetatethat
andare able to
n-hexane
decrease
fraction atblood
a dose sugar level
of 150 by 10.25%
mg/kg and 9.98%
BW provided respectively.
the highest Its effect
anti-diabetic is equivalent
activity compared to to
glibenclamide
the other testatgroups
a dosethat
of 0.65 mg/kg
are able to
BW which
decrease can sugar
blood lowerlevel
bloodbysugar
10.25% levels
and by 9.27%.
9.98% Thus Cassia
respectively. siameais Lamk
Its effect leaf toisglibenclamide
equivalent very potentialatasa andosealternative drug
of 0.65 mg/kg
antidiabetes
BW which can mellitus,
lowerand (3) the
blood 1000
sugar mg Ethanol
levels by 9.27%. extract,
Thus150 mg Ethyl
Cassia siamea acetate
Lamkand leaf150ismg n-Hexana
very hadasno
potential andifference
alternativeeffects
drug
in lowering blood
antidiabetes glucose
mellitus, levels
and (3) compared
the 1000 to the anti-diabetic
mg Ethanol extract, 150 chemical
mg Ethyl drug glibenclamide.
acetate and 150 mg n-Hexana had no difference effects
in lowering blood glucose levels compared to the anti-diabetic chemical drug glibenclamide.
© 2018The Authors. Published by Elsevier Ltd.
© 2018
This The
is an openAuthors. Published
accessPublished
article underby Elsevier Ltd.
the CC BY-NC-ND license(https://creativecommons.org/licenses/by-nc-nd/4.0/)
© 2018The
This Authors.
is an open access article underby Elsevier
the CC Ltd.
BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
Selection
This is an and
open peer-review
access under
article responsibility
under the CC of the 3rdlicense(https://creativecommons.org/licenses/by-nc-nd/4.0/)
BY-NC-ND International Conference on Computer Science and Computational
Selection and peer-review under responsibility of the 3rd International Conference on Computer Science and Computational
Intelligence
Selection
Intelligence 2018.
and2018.
peer-review under responsibility of the 3rd International Conference on Computer Science and Computational
Intelligence 2018.
Keywords:Cassia siamea Lamk; In Vivo anti-diabetic test; Anova
Keywords:Cassia siamea Lamk; In Vivo anti-diabetic test; Anova

* Corresponding author.
E-mail address:author.
* Corresponding [email protected]
E-mail address: [email protected]
1877-0509© 2018 The Authors. Published by Elsevier Ltd.
This is an open
1877-0509© access
2018 Thearticle under
Authors. the CC by
Published BY-NC-ND license(https://creativecommons.org/licenses/by-nc-nd/4.0/)
Elsevier Ltd.
Selection
This is an and
openpeer-review under
access article responsibility
under of the 3rdlicense(https://creativecommons.org/licenses/by-nc-nd/4.0/)
the CC BY-NC-ND International Conference on Computer Science and Computational Intelligence 2018.
Selection and peer-review under responsibility of the 3rd International Conference on Computer Science and Computational Intelligence 2018.

1877-0509 © 2018 The Authors. Published by Elsevier Ltd.

This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
Selection and peer-review under responsibility of the 3rd International Conference on Computer Science and Computational
Intelligence 2018.
10.1016/j.procs.2018.08.223
 Heruna Tanty et al. / Procedia Computer Science 135 (2018) 632–642 633
2 Author name / Procedia Computer Science 00 (2018) 000–000

1. Introduction

The people having Diabetes Mellitus (DM) in Indonesia continues to increase every year. According to the Ministry
of Health of the Republic of Indonesia, by 2030, about 21.3 million people in Indonesia are estimated to have diabetes.
About 16.7% of patients with DM are under the age of 40 years, 46.5% are at the age of 40-59 years, and the rest at
age above 60 years 1.
DM is a disease caused by high levels of glucose in the blood as a result of defects in insulin secretion. The elevated
blood glucose is the risk factor that can cause various disorders in some organs such as liver, heart, kidneys, eyes,
nerves, and sexual function 2. Diabetes Mellitus can be managed by taking chemical drugs continuously for long
periods of time. However, the use of chemical drugs can produces negative side effects to other organs. Hence, the
search for alternative source of medicine from medicinal plants has continued.
Juar (Cassia siamea Lamk) is a plant from the fabaceae family. A qualitative analysis showed that the leaves of
Cassia siamea contains a secondary metabolic compound such as antrona, flavona, triterpenoide, alkaloids and
casiadinine 3, isoquinolone Siaminine A, Siaminine B and Siaminine C 4, new bischromone and chrobisiamone A 5
and 5-acetylmethyl-7-hydroxy-2-methyl chromone 6. The Cassia siamea leaves are believed able to heal various
diseases such as malaria, diarrhea, hives, anti-diabetic and anti-inflammatory 7. In vitro studies showed that the extract
of ethanol of Juar leaves in the fractions of ethyl acetate and n-hexane can inhibit the enzyme of alpha glucosidase
about 52.319%. This suggests that Cassia Siamea is very potential as an alternative drug for anti-diabetic 8. This study
was conducted to continue the anti-diabetic test that has been performed by in vitro.
The main objectives of this study including: (1) to test whether Cassia Siamea Lamk extracts have an effect on
lowering blood glucose levels, (2) to examine whether 4 types of extracts i.e. Ethanol 500 mg/kgBW, Ethanol 1000
mg/kgBW, Ethyl acetate 150 mg/kgBW and n-Hexana 150 mg/kgBW have a different effect on lowering blood
glucose levels, and (3) to find out which extract gives the best performance in lowering blood glucose levels. The tests
were performed using statistical analysis including descriptive analysis, ANOVA and the Tukey Honest Significance
Difference test.
There are 4 types of extracts that will be compared, because there are more than two types of extract that will be
compared, then the method that can be used is one way ANOVA. There are several studies of extract test in in vivo
activity using one way ANOVA method, such as Ethanol extract of Aloe vera for anti-diabetic test 9, Ethanol extract
of Tithonia diversifolia (Hemls.) A. Gray for anti-hyperglycemic test 10, Water extract of Pandan Wangi leaf to
decrease blood glucose levels 11 and Extract of Actinidia columnikta for anti-diabetic test 12.
ANOVA can only test whether there are differences between 4 types of extracts. But to investigate differences
among the various extracts cannot provide by ANOVA. One method that can be used is t-test between each of the
pairs of extracts. However it is not good approach because of repeated statistical tests on the same data. This will cause
a higher probability of making a Type I error. There are several alternatives to the multiple comparison analysis tests
that can be used among Bonferroni, Sheffèe, Tukey, Newman-Keuls and Dunnett. Each method has strengths and
limitations. Based on the comparison of different multiple comparison analysis tests by McHugh 13, this research used
Tukey Honest Significance Difference (HSD) method because it can be avoid the probability of making a Type I error.

2. One Way ANOVA

The completely randomized design is a set of independent samples from a set of several populations. Analysis of
variance (ANOVA) is one of statistical method that can be used to compare the means of more than two groups /
levels. But there are assumptions to use the analysis of variance: (1) the response variable have to normally distributed
for each group, (2) the population variances of response variable in each group are the same, (3) the observations must
be independent of one another 14. Suppose that k groups, the general form of hypothesis testing to compare population
means is
H0 : µ1 = µ2 = µ3 = ... = µk
H1 : at least one of µi ≠ µj for i ≠ j
634 Heruna Tanty et al. / Procedia Computer Science 135 (2018) 632–642
Author name / Procedia Computer Science 00 (2018) 000–000 3

Table 1. ANOVA table for a completely randomized design

Source of Variation Degree of Freedom Sum of Square Mean Square F test

Treatments k–1 SSTR MSTR F
Error nT – k SSE MSE
Total nT – 1 SST

Calculate total sum of square (SST), sum square of treatments (SSTR) and sum square of error (SSE). The x̅i is
represent the mean sample i (i = 1, 2, ..., k) and x̿ is the grand mean 15.

SSTR =  ni x i − x
k

i =1
( )2
(1)

( )
k ni
2
SSE =  ni xij − x i (2)
i =1 j =1

k
SST =  ni xij − x
i =1 j =1
ni
( )
2
(3)

The mean sum of square for treatments (MSTR) and mean sum of square for error (MSE) as follows

𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 = (4)
𝑘𝑘−1

𝑆𝑆𝑆𝑆𝑆𝑆
𝑀𝑀𝑀𝑀𝑀𝑀 = (5)
𝑛𝑛𝑇𝑇 −𝑘𝑘

The F test statistic used to test the equality of k population means. Reject H0 if the F test is larger than F distribution
with k – 1 numerator degree of freedom and nT – k denominator degree of freedom at α level of significance 16. The F
test statistic is

𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴
𝑭𝑭 = (6)
𝑴𝑴𝑴𝑴𝑴𝑴

3. Tukey’s Honest Significance Difference (HSD) Test

If the null hyphotesis in analysis of variance is rejected, then the next step is discover which ones have been shown
to be different. There are several methods that can be used, one of them is Tukey’s Honest Significance Difference
(HSD) test. This research use HSD test because this method can be avoid the probability of making a Type I error 13.
Tukey’s HSD test follow the distribution of studentized range or q statistic 17.
H0 : µi = µj for i ≠ j
H1 : µi ≠ µj for i ≠ j

MSE
HSD = q k ,df , (7)
n

Where qk,df,α is the critical value with k levels, df (degree of freedom) = nT – k and α level of significance. MSE is
mean square of error from ANOVA table and n is number of sample for each population. In order to evaluate the
 Heruna Tanty et al. / Procedia Computer Science 135 (2018) 632–642 635
4 Author name / Procedia Computer Science 00 (2018) 000–000

means difference between populations is take the absolute value of the difference between means and compare to HSD
value. Then, H0 is rejected if

|𝑋𝑋̅𝑖𝑖 − 𝑋𝑋̅𝑗𝑗 | ≥ 𝐻𝐻𝐻𝐻𝐻𝐻 (8)

4. Methodology

Samples of C.siamea leaves were collected from the village of Setu, in Cirebon district, West Java, Indonesia. The
leaves were cleaned, dried for three weeks and then made into powder. About 454.99 gram of the dried and powdered
leaves of C.siamea leaves were extracted with 4 liter of 90% ethanol for 24 hours. The extraction were conducted in
3 replications (3x4Lx24 hours). The filtrates were strained using filter paper and evaporated at 40oC using Rotary
Evaporator to get the crude ethanol extract (15.1 gram). The extract were then fractionated using the mixture of 1 part
of water and 9 part of n-hexane to obtain n-hexane fraction and water fraction. The water fraction was given ethyl
acetate to obtain an ethyl acetate fraction. The n-hexane fraction and ethyl acetate fraction were then concentrated to
5.1 gram and 2.4 gram respectively 8.
The test were using animals induced with aloxan, since aloxan price is relatively cheaper compared to other
diabetogen and it can make the animal experience DM condition in a short time. Some researchers tested in Vivo
using white mice induced with alloxan 9 11, while S.Kumar performed in Vivo tests with white mice using another
diabetogen, i.e. streptozotocin induction 18. The animal used for experiment were 8-week old Swiss Webster albino
mice with weights ranging from 20 to 30 gram. The animals were kept for 1 week in advance to adapt to the conditions
of the cage before being treated. The mice were placed in a cage that is given a husk base to absorb dirt from mice.
During the adaptation period, the mice were fed, watered and weighted daily. The mice are considered healthy when
their weight increases or remains the same or decreases no more than 10%.
The mice used for experiment were 21 animals, which were divided into 7 groups, containing 3 animals in each
group. Before treatment, the mice fasted for 18 hours (ad libitum). Blood samples were drawn from each mouse‘s tail
for measuring the first blood glucose levels (baseline). All mice then were treated intra peritoneally with 250 mg/kg
BW of alloxan, to elevate their blood glucose levels. After 48 hours being treated, blood samples were drawn from
each mice for measuring their blood glucose levels. If the blood glucose levels of mice > 200 mg/dL then the mice are
considered to have hyperglycemia. Furthermore, the alloxan-induced groups of diabetic miceorally received the
following treatment schedule for 7 days:
• Group 1 served as normal control, received 2% PGA suspension without induction
• Group 2 served as negative control, received alloxan induction in 2% PGA
• Group 3 served as positive control, received 0.65 mg/kg BW glibenclamide in 2% PGA suspension
• Group 4 served as the test group, received ethanol extract at a dose of 500 mg/kg BW in 2% PGA suspension
• Group 5 served as the test group, received ethanol extract at a dose of 1000 mg/kg BW in 2% PGA suspension
• Group 6 served as the test group, received the ethyl acetate fraction at a dose of 150 mg/kg BW in 2% PGA
suspension
• Group 7 served as the test group, received the n-hexane fraction at a dose of 150 mg/kg BW in 2% PGA
suspension

Blood glucose level were measured every day for 8 days after treatment, using amperometric method that utilizes
enzymatic glucose dehydrogenase reaction with Glucometer instrument. The relative blood glucose level can be
calculated by the following formula:

blood glucose level at time t

Relative blood glucose level= x 100% (9)
intial blood glucose level

5. Results and Discussions

The result of blood glucose measurements from each group are in the following Table.
636 Heruna Tanty et al. / Procedia Computer Science 135 (2018) 632–642
Author name / Procedia Computer Science 00 (2018) 000–000 5

Table 2. Blood glucose levels of each group during the examination of anti-diabetic activity of extracts of Cassia Siamea leaves fraction
Blood Glucose Levels at day
Group No D1
D0 D2 D3 D4 D5 D6 D7 D8
(Diabetes)

1 80 77 64 100 85 95 90 88 87

2 101 68 68 101 80 80 100 80 92

Group 1
(Normal)
3 84 80 61 84 78 83 86 98 73

88.33 75.00 64.33 95.00 81.00 86.00 92.00 88.67 84.00

1 77 325 331 328 326 320 315 309 303

2 78 327 366 355 345 339 334 327 320

Group 2
(Negative)
3 80 274 283 298 313 304 296 274 252

78.33 308.67 326.67 327.00 328.00 321.00 315.00 303.33 291.67

1 90 471 457 448 440 415 391 343 296

Group3 2 96 387 383 348 313 276 240 230 220

(Positive ) 3 100 488 476 435 395 342 290 259 229

95.33 448.67 438.67 410.33 382.67 344.33 307.00 277.33 248.33

1 101 269 418 288 262 238 180 168 157

Group 4
2 78 292 415 314 261 231 222 204 202
ethanol extract
500 mg/kg of 3 98 263 379 219 190 164 157 147 145
weight
92.33 274.67 404.00 273.67 237.67 211.00 186.33 173.00 168.00

1 98 266 242 240 232 212 197 154 126

Group 5
2 83 264 248 238 225 220 206 181 140
ethanol extract
1000 mg/kg of 3 78 273 257 232 220 185 183 166 143
weight
86.33 267.67 249.00 236.67 225.67 205.67 195.33 167.00 136.33

1 90 385 323 310 300 300 273 223 163

Group 6
2 93 381 303 302 261 256 211 191 165
ethyl acetate
fraction 150 3 87 370 357 329 261 223 219 210 204
mg/kg of weight
90.00 378.67 327.67 313.67 274.00 259.67 234.33 208.00 177.33
 Heruna Tanty et al. / Procedia Computer Science 135 (2018) 632–642 637
6 Author name / Procedia Computer Science 00 (2018) 000–000

Blood Glucose Levels at day

Group No D1
D0 D2 D3 D4 D5 D6 D7 D8
(Diabetes)

1 90 322 305 299 283 272 210 190 147

Group 7 2 96 366 363 249 226 180 156 128 118

n-hexane
fraction 150 3 101 348 340 258 228 218 207 193 163
mg/kg of weight
95.67 345.33 336.00 268.67 245.67 223.33 191.00 170.33 142.67

The blood glucose level of mice was observed before alloxan induction (D0) and after the mice had diabetes (D1)
and continued by giving the 8 days suspension. Table 2 above explained that the blood glucose levels of all treatment
groups before alloxan induction (H0) ranged from 78.33 ± 2.87 to 95.67 ± 7.78 mg/dL. This indicates that all test
animals before alloxan induction had normal blood glucose levels, ranging from 62-175 mg/dL. Furthermore, at day
1 (D1) the mice intraperitoneally were induced with alloxan at a dose of 250 mg/kgBW
After alloxan induction, blood glucose levels of mice were monitored for 3 days to see the effect of hyperglycemia.
After 3 days of induction, there would be mice that have hyperglycemia and mice that was still in normal condition.
As for mice who were still in normal condition would be re-induced with the same dose of alloxan.
The observation showed that the blood glucose levels of the mice with hyperglycaemia ranged between 267.67 ±
3.86- 438.67 ± 40.12 mg/dL. The hyperglycemia mice were then divided at random into 6 groups, consisting of 3
animals in each group.
Mice with hyperglycemia were then orally treated according to their group for 8 days. Group 3, as a positive control
group was treated with glibenclamide at a dose of 0.65 mg/kgBW. Glibenclamide is insoluble in water so it is
suspended with PGA agent. Group 2 as negative control was only induced by alloxan, it used to determine the decrease
in blood glucose levels from normal circumstances. The measurement of blood glucose level was performed on day
1 (after mouse became diabetic) until day 8 (Table 2).
The relative blood glucose level and its percentage of reduction were calculated based on the data in Table 2. The
relative blood glucose levels of each group are presented in Table 3 below

Table 3. The relative blood glucose levels of each group

The relative blood glucose levels of each group (%)
Group
H0 H1 H2 H3 H4 H5 H6 H7 H8

K Normal 100.00 86.27 73.32 108.33 92.77 98.92 104.63 101.96 95.58

K(-) 100.00 394.60 417.62 417.87 418.98 410.07 402.43 387.68 372.92

K(+) 100.00 471.49 460.91 431.76 403.31 363.54 324.81 293.23 262.35

Extract ethanol 500 100.00 303.02 444.22 303.73 262.63 233.05 207.68 192.63 187.46

Extract ethanol 1000 100.00 313.17 291.74 276.36 263.29 239.52 227.94 196.01 160.19

Fraction EtOAc 150 100.00 420.91 365.01 349.11 304.66 288.31 260.65 231.51 197.67

Fraction n-Heksan 150 100.00 361.19 351.22 282.35 258.53 235.19 200.26 178.51 149.21

Figure 1 shows the average relative glucose level profile of each group in graphical form. Table 3 and Figure 1
showed that during the test the blood glucose level of all the groups, except the normal group and the negative control
638 Heruna Tanty et al. / Procedia Computer Science 135 (2018) 632–642
Author name / Procedia Computer Science 00 (2018) 000–000 7

group, decreased gradually to near normal levels. While the normal and the negative control group still had
hyperglycemia until the end of the test (day 8). The test group (group 4 to group 7) had a significant decrease in blood
glucose levels, even lower than the positive control group treated with glibenclamide at a dose of 0.65 mg/kgBW.
Group 4 that treated with Ethanol extract dose 500 mg / kg BW was still not able to lower blood glucose levels on day
2, while the other test groups have shown the effect of lowering blood glucose. However on the 3rd day, group 4
showed a very sharp decrease of blood glucose levels.

500.00
450.00
400.00 K Normal
The relative blood glucose

350.00 K(-)
300.00
K(+)
250.00
Eks etanol 500
200.00
Eks etanol 1000
150.00
Fr EtOAc 150
100.00
50.00 Fr n-Heksan 150
-
H0 H1 H2 H3 H4 H5 H6 H7 H8

Fig. 1. The average relative glucose level profile of each test group

The percentage of decrease in blood glucose levels of each group can be seen in Figure 2 below:

100.00

80.00
K(+)
60.00
Eks etanol 500
40.00 Eks etanol 1000
Fr EtOAc 150
20.00
Fr n-Heksan 150
-
1 2 3 4 5 6 7
-20.00

Fig. 2. The percentage of decrease in blood glucose levels of the test groups

The provision of ethanol extract at a dose of 1000 mg/kgBW may decrease blood glucose level, but was not better
than at a dose of 500 mg/kgBW. Thus, it can be concluded that the increase of dose on ethanol extract was not showed
differences in effect of decreasing of blood glucose levels. The ethanol extract at a dose of 500 mg/kg BW tended to
have a faster effect on decreasing blood glucose level, shown in the third day (H3) of measurement.
 Heruna Tanty et al. / Procedia Computer Science 135 (2018) 632–642 639
8 Author name / Procedia Computer Science 00 (2018) 000–000

The groups treated with n-hexane fraction and ethyl acetate fraction at a dose of 150 mg/kgBW, both showed
almost similar decrease in blood glucose levels, although on day 2 to day 8, the n-hexane fraction showed smaller
levels. The relative glucose level of the group 5 with ethanol extract test group 1000 mg/kg BW, the group 6 and 7
with n-hexane and ethyl acetate fractions on day 2 appear smaller compared to that of the group 4 with ethanol extract
group 500 mg / kg BW, positive control group 3 and negative control group.
The formula to determine the best group with blood glucose level reduction is

relative glucose level of K(-)- relative glucose level of K (test)

P (%)= x 100% (10)
relative glucose level of K(-)- relative glucose level of K (N)

where
P = percentage of relative blood glucose level reduction
K (-) = negative control group
K (test) = test group
K (N) = normal control group
Since the blood glucose levels at day 1 were different for each treatment group, the mean of difference in the
reduction of blood glucose levels per group during the test day were calculated to conclude the highest decrease in
glucose levels (Figure 3)

12.00
10.00
8.00
6.00
4.00
2.00
-
K(+) Eks etanol Eks etanol Fr EtOAc Fr n-Heksan
500 1000 150 150
Fig. 3. Bar graph of the average of the reduction of blood glucose levels per group

Figure 3 showed that group 6 that was treated with ethyl acetate test fraction at a dose of 150 mg/kgBW, provided
the highest reduction of blood glucose (i.e. 10.25%), while group 7, that was treated with n-hexane fraction at a dose
of 150 mg/kgBW gave 9.98% reduction. Moreover, it showed that group 6 and 7 gave a decrease in the blood glucose
almost equivalent compared to the positive control group (group 3) that was treated with glibenclamide at a dose of
0.65 mg/kgBW. While group 5, that was treated with ethanol extract at a dose of 1000 mg/kgBW gave a larger
decrease rate (i.e. 7.18%) than group 4 that treated with ethanol extract at a dose of 500 mg/kgBW (i.e. 5.31%). It
seems that the increasing dose of Ethanol extract will rise the anti-diabetic effect.
The results of in vivo test showed that the extract of Cassia siamea.Lamk leaves, in the form of ethyl acetate and
n-hexane fraction at a dose of 150 mg/kgBW, were able to decrease blood sugar level by 10.25% and 9.98%
respectively. It means that the anti-diabetic activity of C. siamea Lamk leaves extract is equivalent and even higher
than the glibenclamide anti-diabetic drug at a dose of 0.65 mg/kgBW (i.e. 9.27%). Therefore it suggests that the extract
of C. siamea.Lamk leaves can be used as an alternative medicine for people with diabetes mellitus (DM).
Furthermore, a statistical analysis was performed to test whether there was a difference in blood glucose levels
before and after being treated with Cassia Siamea Lamk extract. If there is a difference then Cassia Siamea Lamk
extract treatment is proven to lower blood glucose levels, instead it means that Cassia Siamea Lamk extract cannot be
used to lower blood glucose levels. The following is a comparison test of blood glucose levels before and after
treatment of Cassia Siamea Lamk extracts.
640 Heruna Tanty et al. / Procedia Computer Science 135 (2018) 632–642
Author name / Procedia Computer Science 00 (2018) 000–000 9

H0 : µ D = 0
H1 : µ D ≠ 0
The test showed p-value = 2.385 x 10-7. Because the p-value is less than α = 0.05, it would be led to reject H 0.
Thus, there was sufficient evidence to conclude that the mean of blood glucose levels differed between before and
after being treated with the extracts or fractions of Cassia Siamea Lamk for 8 days. Based on these results, it can be
concluded that the extracts or fractions of Cassia Siamea Lamk can be used as an alternative medicine to lower blood
glucose levels.
Furthermore, analysis was conducted to find out what extracts/fractions had an effect on lowering blood glucose
levels. This study used 4 types of extracts/fractions i.e. Ethanol 500 mg (group 4), Ethanol 1000 mg (group 5), Ethyl
acetate 150 mg (group 6) and n-Hexana 150 mg (group 7). The one-way ANOVA was used to test whether there are
differences in effect of the 4 extracts/fractions in lowering blood glucose levels. The data that be used in this research
was fulfill ANOVA assumptions. Based on Levene’s test for homogeneity of variance showed that p-value equal to
0.7311. It means variance of blood glucose levels in each group are same. This experiment used 21 animals, which
were divided into 7 groups, containing 3 animals in each group, it means observations was independent of one another.
Here are the results of ANOVA test

Table 4. ANOVA table

Source df Sum Square Mean Square F p-value
Treatment 3 21582 7194 10.3 0.00403
Residuals 8 5589 699
Total 11 27171

H0 : µ 4 = µ 5 = µ 6 = µ 7
H1 : at least one of µi ≠ µj for i ≠ j
Using the F test statistic on Table 4, showed p-value equal to 0.00403, which is less than α = 0.05. Therefore, the
H0 is rejected which showed that there is at least a pair of extracts/fractions have a different mean of blood glucose
level. To determine the best extract in lowering blood glucose levels, it is necessary to conduct further test using
Tukey’s Honest Significance Difference test.

Table 5. Tukey’s Honest Significance Difference test for extracts of Cassia Siamea
diff lower upper p-value
Group 5 – Group 4 24.66667 -44.44201 93.77534 0.6755397
Group 6 – Group 4 94.66667 25.55799 163.77534 0.0100015
Group 7 – Group 4 96 26.89132 165.10867 0.0092342
Group 6 – Group 5 70 0.89132 139.10867 0.0471790
Group 7 – Group 5 71.33333 2.22466 140.44201 0.0432603
Group 7 – Group 6 1.33333 -67.77534 70.44201 0.9999078

H0 : µ i = µ j
H1 : µ i ≠ µ j
When comparing group 4 that was treated with ethanol extract at a dose of 500 mg/kgBW and group 5 that was
treated with ethanol extract at a dose of 1000 mg/kgBW, Table 5 showed that the p-value is 0.6755397. Since the p-
value is larger than α = 0.05, then H0 cannot be rejected, which means that there is no difference in blood glucose
levels between group 4 and group 5. Thus, it can be concluded that a dose of 500 mg and 1000 mg of Ethanol extract
gives the same performance in lowering blood glucose levels, meaning that additional doses of ethanol extract have
no effect on lowering the blood glucose level. Do the same thing for other groups, comparing group 4 that received
ethanol extract at a dose of 500 mg/kg BW and group 6 that received Ethyl acetate fraction at a dose of 150 mg/kg
BW, showed that the blood glucose levels of groups 4 and group 6 are significantly different. It concluded that 150
mg/kgBW Ethyl acetate fraction gave a better effect in lowering blood glucose level compare to 500 mg/kg BW
 Heruna Tanty et al. / Procedia Computer Science 135 (2018) 632–642 641
10 Author name / Procedia Computer Science 00 (2018) 000–000

Ethanol extract. The result of p-value of comparing group 4 that received ethanol extract at a dose of 500 mg/kg BW
and group 7 received n-Hexana fraction at a dose of 150 mg/kg BW is smaller than α=0.05. it can be concluded that
the 150 mg/kgBW n-Hexana fraction provided a better effect in lowering blood glucose level compare to 500 mg/kg
BW Ethanol extract.
On the other hand, the blood glucose levels of group 5 and group 6 are significantly different, or it means that the
1000 mg/kg BW Ethanol extract and 150 mg/kg BW Ethyl acetate fraction provide a significantly different
performance in lowering blood glucose levels. Since the difference of the value of the blood glucose level between
group 7 and 4 is positive 70, it can be concluded that the Ethyl acetate fraction at a dose of 150 mg/kgBW provided a
better effect in lowering blood glucose level compare to 1000 mg/kg BW Ethanol extract. The same result of
comparing group 5 and group 7. The n-Hexana fraction at a dose of 150 mg/kg BW provided a better effect in lowering
blood glucose level compare to Ethanol extract at a dose of 1000 mg/kg BW. But, in comparing group 6 that received
Ethyl acetate at a dose of 150 mg/kg BW and group 7 that received n-Hexana at a dose of 150 mg/kg BW, the Tukey
test gave p-value = 0.9999 which is larger than α=0.05. Hence, it showed that there is no sufficient evidence to reject
H0 and it suggested that the blood glucose levels of groups 6 and group 7 are not significantly different, or it means
that Ethyl acetate fraction at a dose of 150 mg/kg BW and n-Hexana fraction at a dose of 150 mg/kg BW provide the
same performance in lowering blood glucose levels.
Based on the results of the test, it showed that the 150 mg Ethyl Acetate fraction (group 6) and 150 mg n-Hexana
fraction (group 7) are better in lowering blood glucose levels compared to 500 mg ethanol extract (group 4) and 1000
mg ethanol extract (group 5). However, Ethyl Acetate 150 mg (group 6) and n-Hexana 150 mg (group 7) extracts
have no difference effect in lowering blood glucose levels. Therefore it can be concluded that Ethyl acetate 150 mg
and n-Hexana 150 mg are two types of fractions which gives the best result in lowering blood glucose levels.
The analysis was continued to examine the differences in blood glucose levels between groups receiving Cassia
Siamea Lamk extracts/fractions and K (+) group receiving anti-diabetic drugs i.e. glibenclamide at a dose of 0.65
mg/kgBW. The following are the results of the test using Tukey Honest Significance Difference test.

Table 6. Tukey Honest Significance Difference test between positive control and extracts of Cassia Siamea
diff lower upper p-value
Group 4 – Group 3 93.66667 5.425831 181.90750 0.0365162
Group 5 – Group 3 69 -19.240836 157.24084 0.1494212
Group 6 – Group 3 -1 -89.240836 87.24084 0.9999995
Group 7 – Group 3 -2.33333 -90.574169 85.90750 0.9999845

In comparing group 4 that received Ethanol 500 mg/kgBW and the positive control (group 3) that received
glibenclamide at a dose of 0.65 mg/kg BW, the Tukey test gave p-value = 0.03651 which is smaller than α = 0.05.
Hence, H0 was rejected and it means that Ethanol 500 mg/kgBW provided different effect in lowering blood glucose
levels compared to the anti-diabetic chemical drug glibenclamide. The Tukey tests in comparing group 5 to 3, and
group 7 to 3 gave p-values that larger than α=0.05. Hence, H0 was not rejected and it means that the Ethanol extract
1000 mg/kg BW and n-hexane fraction 150 mg provided no difference effect in lowering blood glucose levels
compared to the anti-diabetic chemical drug glibenclamide.

6. CONCLUSION

The blood glucose levels after treated with the extracts or fractions of Cassia siamea Lamk were lower than before
it received the extract or fractions. The ethanol extract and n-hexane fraction of Cassia siamea Lamk (Juar) leaves
have anti-diabetic activity in Webster albino mice induced with alloxan. It means Cassia Siamea Lamk can be used as
an alternative medicine to lower blood glucose levels. Based on the result of comparing 4 types of extracts/fractions
i.e. Ethanol 500 mg (group 4), Ethanol 1000 mg (group 5), Ethyl acetate 150 mg (group 6) and n-Hexana 150 mg
(group 7), it showed that Ethyl acetate 150 mg and n-Hexana 150 mg are two types of fractions which gives the best
result in lowering blood glucose levels.
642 Heruna Tanty et al. / Procedia Computer Science 135 (2018) 632–642
Author name / Procedia Computer Science 00 (2018) 000–000 11

The extract of Cassia siamea Lamk leaves, in the form of ethyl acetate and n-hexane fraction at a dose of 150
mg/kg BW provided the highest anti-diabetic activity compared to the other test groups that are able to decrease blood
sugar level by 10.25% and 9.98% respectively. Its effect is equivalent to glibenclamide at a dose of 0.65 mg / kg BW
which can lower blood sugar levels by 9.27%. The Ethanol extract 1000 mg/kg BW and n-hexane fraction 150 mg
provided no difference effect in lowering blood glucose levels compared to the anti-diabetic chemical drug
glibenclamide. Thus C.siamea Lamk leaf is very potential as an alternative drug antidiabetes mellitus.

References

1. Wisnu W. Cegah Komplikasi Diabetes Sedini Mungkin. Jakarta: RS Pondok Indah; 2017.
2. Riskesdes K. Departemen Kesehatan. [Online].; 2013 [cited 2018 January 11. Available from:
http://www.depkes.go.id/resources/download/pusdatin/infodatin/infodatin-diabetes.pdf.
3. Heyne K. Tumbuhan Berguna Indonesia Jilid 2 Jakarta: Yayasan Sarana Wana Jaya; 1987.
4. Samia S, Sayyad E, Ross SA, Sayed HM. New Isoquinone Alkaloids from Leaves of Cassia Siamea. Journal of
Natural Product. 1984; 47(4).
5. Oshimia S, Tomizawaa Y, Hirasawaa Y, Hondaa T, Ekasarib W, Yanti AW, et al. Chrobisiamone A, a new
bischromone from Cassia siamea and a biomimetic transformation of 5-acetonyl-7-hydroxy-2-methylchromone
into cassiarin A. Bioorganic & Medicinal Chemistry Letters. 2008; 18(13).
6. Tanty H. Isolasi Chromone dalam daun Juar (Cassia siamea.Lamk). In Seminar Nasioal Kimia UNJANI; 2014;
Bandung.
7. Kamagaté M. Ethnobotany, phytochemistry pharmacology and toxicology profiles of Cassia siamea Lam. The
Journal of Phytopharmacology. 2014; 3(1).
8. Tanty H. Antidiabetic Activity Test for Leaves of Cassia Siamea.Lamk. Matter International Journal of Science
and Technology. 2018; 3(3).
9. Hutabarat ER, Bangsawan PI, Raharjo W. Uji Efek Hypoglikemik Ekstrak Etanol Daun Lidah Buaya ( Aloe
Vara.L) Terhadap Kadar Glukos Darah Tikus Putih Jantan Galur Wistar Yang Diinduksi Aloksan. Laporan
Skripsi. Pontianak: Universitas Tanjungpura, Program Studi Pendidikan Dokter, Fakultas Kedokteran; 2014.
10. Kulsum U. Uji Efek Antihiperglikemia Ekstrak Etanol 95% Daun Kembang Bulan (Tithonia diversifolia
(Hemsl.) A. Gray) Terhadap Tikus Sprague-Dawley Jantan dengan Metode Induksi Aloksan Secara In Vivo.
Laporan Skripsi. Jakarta: Universitas Islam Negeri Syarif Hidayatullah, Program Studi Farmasi, Fakultas
Kedokteran dan Ilmu Kesehatan; 2016.
11. Prameswari OM, Widjanarko SB. Uji Efek Ekstrak Air Daun Pandan Wangi Terhadap Penurunan Kadar
Glukosa Darah dan Histopatologi Tikus Diabetes Mellitus. Jurnal Pangan dan Agroindustri. 2014; 2(2).
12. Yuan X, Hu Y, Liu Y, Sun H, Zhang Z, Cheng D. In Vitro and In Vivo Anti-Diabetic Activity of Extract From
Actinidia kolomikta. International Journal of Biology. 2014; 6(3).
13. McHugh ML. Multiple Comparison Analysis Testing in ANOVA. Biochemia Medica. 2011; 21(3).
14. Anderson DR, Sweeney DJ, Williams TA. Statistics for Business and Economics. 10th ed. Mason: Thomson
Higher Education; 2008.
15. Hayter A. Probabiity and Statistics for Engineers and Scientists Boston: Brooks Cole Cengage Learning; 2012.
16. Ostertagova E, Ostertag O. Methodology and Application of One-way ANOVA. American Journal of
Mechanical Engineering. 2013; 1(7).
17. Abdi H, Williams LJ. Tukey’s Honestly Significant Difference (HSD) Test”, in Neil Salkind (eds),
Encyclopedia of Research Design Sage: Thousand Oaks CA; 2010.
18. Kumar S, Kumar V, Prakash O. Antidiabetic and Antilipemic Effect of Cassia siamea Leaves Extract in
Streptozotocin Induced Rats. Asian Pacific Journal of Tropical Medicine. 2010; 3(11).