molecules

Article
Bioactivity-Guided Fractionation and Identification of
Antidiabetic Compound of Syzygium polyanthum (Wight.)’s
Leaf Extract in Streptozotocin-Induced Diabetic Rat Model
Tri Widyawati 1,2, *, Nor Adlin Yusoff 3, *, Idris Bello 2 , Mohd Zaini Asmawi 2 and Mariam Ahmad 2

1 Pharmacology and Therapeutics Department, Medical Faculty, Universitas Sumatera Utara,

Medan 20155, Indonesia
2 Department of Pharmacology, School of Pharmaceutical Sciences, Universiti Sains Malaysia,
Penang 11800, Malaysia
3 Department of Toxicology, Advanced Medical and Dental Institute, Sains@Bertam, Universiti Sains Malaysia,
Penang 13200, Malaysia
* Correspondence: [email protected] or [email protected] (T.W.); [email protected] (N.A.Y.);
Tel.: +62-8126-582-932 (T.W.); +60-111-777-7592 (N.A.Y.)

Abstract: (1) Background: An earlier study on the hypoglycemic activity of S. polyanthum (Wight.)
leaf methanol extract identified squalene as the major chemical compound. The present study was
conducted to assess the hypoglycemic effect of fractions and subfractions of the methanol extract
of S. polyanthum compared to the squalene using a bioassay-guided in vivo study. (2) Methods: The
methanol extract was fractionated using the liquid–liquid fractionation method. Streptozotocin-
induced type 1 diabetic rat was used to study the hypoglycemic effect. (3) Results: The findings
showed that chloroform fraction significantly (p < 0.05) lowered blood glucose levels of diabetic
rats as compared to the control. Further fractionation of chloroform fraction yielded subfraction-1
Citation: Widyawati, T.; Yusoff, N.A.;
and -2, whereby subfraction-1 exhibited a higher blood-glucose-lowering effect. The lipid profile
Bello, I.; Asmawi, M.Z.; Ahmad, M.
Bioactivity-Guided Fractionation and
test showed that the total cholesterol level of subfraction-1 and squalene-treated groups decreased
Identification of Antidiabetic significantly (p < 0.05). An immunohistochemistry study revealed that none of the treatments
Compound of Syzygium polyanthum regenerated pancreatic β-cells. Gas chromatography–mass spectrophotometer analysis identified the
(Wight.)’s Leaf Extract in presence of squalene in the active methanol extract, chloroform fraction, and subfraction-1. In silico
Streptozotocin-Induced Diabetic Rat analysis revealed a higher affinity of squalene against protein receptors that control lipid metabolism
Model. Molecules 2022, 27, 6814. than metformin. (4) Conclusions: Data obtained from the present work suggested the crude methanol
https://doi.org/10.3390/ extract exerted the highest hypoglycemic effect compared to fraction, subfraction, and squalene,
molecules27206814
confirming synergistic effect may be responsible for the hypoglycemic activity of S. polyanthum.
Academic Editor: Jianbo Xiao
Keywords: antihyperglycemic; Indonesian bay leaves; Syzygium polyanthum (Wight.); diabetes;
Received: 1 August 2022
squalene
Accepted: 5 October 2022
Published: 12 October 2022

Publisher’s Note: MDPI stays neutral

with regard to jurisdictional claims in 1. Introduction
published maps and institutional affil-
Diabetes mellitus is a chronic metabolic disorder characterized by persistent hyper-
iations.
glycemia resulting from insufficiency in insulin action or insulin secretion, or both [1].
Persistent hyperglycemia can lead to macrovascular and microvascular complications that
affect several vital organs, including the kidneys, eyes, and heart [2]. The majority of
Copyright: © 2022 by the authors.
diabetes cases are divided into type 1 diabetes mellitus or insulin-dependent and type
Licensee MDPI, Basel, Switzerland. 2 diabetes mellitus or noninsulin-dependent [3]. The prevalence of diabetes has been
This article is an open access article steadily increasing over the decades. In 2021, approximately 537 million diabetes cases
distributed under the terms and were reported, and the disorder was the ninth leading cause of death worldwide [4]. Cur-
conditions of the Creative Commons rently, oral hypoglycemic agents and insulin have been used to control the disease and
Attribution (CC BY) license (https:// its complications. Several associated adverse effects, however, have been reported, re-
creativecommons.org/licenses/by/ flecting the need for continuous research to develop new effective therapeutic agents for
4.0/). diabetes mellitus.

Molecules 2022, 27, 6814. https://doi.org/10.3390/molecules27206814 https://www.mdpi.com/journal/molecules

Molecules 2022, 27, 6814 2 of 15

As plants are known to contain a wide range of active compounds, many studies inves-
tigate plant sources for potential antidiabetic agents. Plant-derived active compounds serve
as pure drugs, as well as lead compounds, for the development of synthetic drugs directly
or indirectly [5]. Bioassay-guided fractionation is one of the widely used techniques in
plant drug discovery. In this technique, the extraction and biological screening of the extract
take place simultaneously in order to identify the active compound [6]. By applying this
technique, several studies have successfully isolated compounds with antihyperglycemic
effects, such as xyloccensin-I from Xylocarpus granatum [7], nicotiflorin and tulipanin from
Punica granatum var. [8], and a novel decahydro-1H-xanthene from Garcinia cowa [9].
Syzygium polyanthum (Wight.) Walp. (Myrtaceae) is an herb widely used in Indonesian
and Malaysian cuisines [10]. It is also commonly used as a traditional medicine to treat dia-
betic patients in Indonesia [11]. Our previous study demonstrated the antidiabetic activity
of an S. polyanthum methanolic extract in a streptozotocin-induced diabetic rat model, in
which squalene was identified as one of the active constituents [12]. Squalene has been
reported to improve the lipid profile associated with the progression of diabetes. Gabas-
Rivera et al. [12] reported that dietary squalene increased the levels of HDL cholesterol and
paraoxonase-1 and decreased total cholesterol levels at a 1 g/kg dose in Apoe-deficient
mice. Similarly, Liu et al. [13] showed that the intake of squalene dietary supplements
effectively increased HDL levels in obese/diabetic KK-Ay mice. These findings were fur-
ther confirmed by Mirmiranpour et al. [14], who studied the effect of squalene on type
2 diabetic patients, and found that consuming squalene for 84 days caused HDL levels to
rise, subsequently lowering total cholesterol, LDL, and VLDL levels in diabetic patients.
The present study utilized a bioactivity-guided approach to further investigate the
hypoglycemic activity of S. polyanthum leaves and the squalene isolated from the plant.
A streptozotocin-induced diabetic rat model that mimicked type 1 diabetes mellitus in
humans was used to screen for possible hypoglycemic effects.

2. Results and Discussion

2.1. Gas Chromatography–Mass Spectrometry Analysis
Gas chromatography–mass spectrometry analysis was conducted to identify the pres-
ence of bioactive compound/s in the active extract (methanol extract), the fraction (chloro-
form fraction), and the subfraction (subfraction-1). Table 1 lists the detected compounds
of high library matching quality. The analysis revealed the presence of several bioactive
compounds that may contribute to the observed glucose-lowering effect; one compound
was found in methanol extract, five in chloroform fraction, and three in subfraction-1. Inter-
estingly, only squalene was detected in all samples. Squalene, a triterpene, is an isoprenoid
compound that belongs to the terpenoid family [15]. It has been implicated in several
studies as a compound that contributes to the hypoglycemic and anti-obesity activities of
plants. Wang et al. [16] have demonstrated the excellent antidiabetic of Sanbai melon seed
oil with squalene as one of the major bioactive compounds. Ravi Kumar et al. [17] reported
that squalene modulated the metabolism of fatty acids in obese diabetic mice models, thus
contributing to the glucose-lowering effect.

Table 1. Phytochemical components identified in the methanolic extract (ME), chloroform fraction
(CF), and n-hexane subfraction-1 (SF-1) using GC–MS.

RT * Compound Molecular
Samples Peak (%)
Name Formula
ME 14.96 7.60 Squalene C30 H50
10.74 6.90 Hexadecanoic acid, methyl ester C17 H34 O2
11.49 2.03 9,12-Octadecadienoic acid, methyl ester C19 H34 O2
CF 14.97 8.92 Squalene C30 H50
18.37 4.73 Vitamin E C29 H50 O2
21.66 22.57 Stigmasterol, 22,23-dihydro- C29 H50 O
14.92 4.54 Squalene C30 H50
SF-1 18.28 4.26 Vitamin E C29 H50 O2
21.50 33.37 Stigmasterol, 22,23-dihydro- C29 H50 O
* RT = retention time; Peak area = the percentage of the compound it represents.
 18.37 4.73 Vitamin E C29H50O2
21.66 22.57 Stigmasterol, 22,23-dihydro- C29H50O
14.92 4.54 Squalene C30H50
SF-1 18.28 4.26 Vitamin E C29H50O2
21.50 33.37 Stigmasterol, 22,23-dihydro- C29H50O
Molecules 2022, 27, 6814 3 of 15
* RT = retention time; Peak area = the percentage of the compound it represents.

2.2. Effect on Blood Glucose Level

2.2. Effect on Blood Glucose Level
Active methanol extract of S. polyanthum was fractionated into chloroform, ethyl ac-
Active methanol
etate, n-butanol, extract
and water of S. polyanthum
fractions. was fractionated
The result showed into chloroform,
all the fractions reduced bloodethyl
glucose n-butanol,
acetate,levels and wateroral
after repeated fractions. The resultwith
administration showed all the fractions
a significant lowering reduced blood
effect only
glucose
seen in thelevels after repeated
chloroform (p < 0.05)oral administration
fraction-treated withas
group, a significant
compared to lowering effectcon-
the diabetic only
seen in the chloroform (p < 0.05) fraction-treated group, as compared to
trol group (Figure 1). Hence, the chloroform fraction was further fractionated. Subfrac- the diabetic control
group (Figure
tionation of 10 g1).ofHence, the chloroform
chloroform fraction was
fraction produced 63%further fractionated.and
of subfraction-1 Subfractionation
14% of sub-
fraction-2. Only subfraction-1 significantly reduced blood glucose level (p <subfraction-2.
of 10 g of chloroform fraction produced 63% of subfraction-1 and 14% of 0.05) when
Only subfraction-1
compared significantly
with the diabetic controlreduced
(Figureblood glucose
2). Overall, thelevel (p <of0.05)
results thiswhen
studycompared
pertain-
with the diabetic control (Figure 2). Overall, the results of this study
ing to the hypoglycemic activity of S. polyanthum extract were consistent with the previ- pertaining to the
hypoglycemic activity of S. polyanthum extract were consistent with the
ous studies, which demonstrated the ability of ethanolic extract of the leaves to substan- previous studies,
which
tially demonstrated
reduce the bloodthe abilitylevels
glucose of ethanolic extract of thediabetic
of alloxan-induced leaves to substantially
rats [18,19]. reduce the
blood glucose levels of alloxan-induced diabetic rats [18,19].

Effectsofofthe
Figure1.1.Effects
Figure fractionsofofS.S.polyanthum’s
thefractions polyanthum’smethanol
methanolextract
extractononthetheblood
bloodglucose
glucoselevels
levelsofof
streptozotocin-induced diabetic rats after being treated orally twice daily for
streptozotocin-induced diabetic rats after being treated orally twice daily for six days. NC, normal six days. NC, normal
control (normal saline, 10 mL/kg); DC, diabetic control (normal saline, 10
control (normal saline, 10 mL/kg); DC, diabetic control (normal saline, 10 mL/kg); M, metformin mL/kg); M, metformin
(500
(500mg/kg);
mg/kg); CF,
CF,chloroform
chloroformfraction
fraction(500
(500 mg/kg);
mg/kg);EAF,
EAF, ethyl
ethyl acetate fraction;
fraction;n-BF,
n-BF,n-butanol
n-butanolfraction
frac-
tion (500 mg/kg); WF, water fraction (500
(500 mg/kg); WF, water fraction (500 mg/kg). The mg/kg). The values are expressed
expressed as mean ± SEM (n= =
as mean ± SEM (n 6).6).
# #indicates significant differences as compared to the NC ( ###
###p < 0.001 and p
####
indicates significant differences as compared to the NC ( p < 0.001 and p < 0.01). * indicates < 0.01). * indicates
significant differences compared to the DC (*** p < 0.001, ** p < 0.01, and * p < 0.05) as analyzed using
Dunnett’s as post hoc test.

Figure 3 presents the blood glucose level of active extract, fraction, subfraction, squa-
lene, as well as controls in streptozotocin-induced diabetic rats. Compared to the nor-
mal rats, streptozotocin induction caused a 2–3-fold increase in blood glucose levels.
Blood glucose levels of methanol extract, chloroform fraction, subfraction-1, squalene,
and metformin-treated groups were significantly decreased on day 12 as compared to day
0. The descending order of the hypoglycemic activity with reduction percentage was as fol-
lows: metformin = methanol extract (56%) > chloroform fraction = subfraction-1 = squalene
(43%). Interestingly, the blood-glucose-lowering effect of methanol extract was comparable
to that observed in the metformin-treated groups (p < 0.001). Overall, crude extract exerted
a better hypoglycemic effect than fraction and subfraction. As can be seen in Figure 3, the
magnitude of the hypoglycemic effect decreased when the extract was further fractionated.
As compared with the NC, only the ME group showed an insignificant difference, which
further indicated treatment with ME successfully eliminated hyperglycemia in the diabetic
rats. In addition to that, the analysis also suggested the possible synergistic effect between
squalene and other bioactive compounds of the methanol extract, chloroform fraction, and
Molecules 2022, 27, 6814 4 of 15

Molecules 2022, 26, x FOR PEER REVIEW 4 of 16

subfraction-1. Compared to the pharmacological effects observed with individual isolated
compounds, better results are obtained with whole plant extracts due to the presence of
various bioactive
significant differencescompounds that
compared to the DCwork
(*** p in synergy
< 0.001, ** p <by targeting
0.01, and * p < either
0.05) asthe sameus-
analyzed or different
pathways
ing Dunnett’s[20].
as post hoc test.

Figure
Figure2.2.Effects of of
Effects thethe
subfractions of S. of
subfractions polyanthum‘s chloroform
S. polyanthum‘s fraction fraction
chloroform on the blood glucose
on the blood glucose
levels
levels ofofstreptozotocin-induced
streptozotocin-induced diabetic rats after
diabetic rats being
after treated
being orally twice
treated daily
orally for six
twice days.
daily for six days.
Molecules 2022, 26, x FOR PEER REVIEW
NC, normal control (normal saline, 10 mL/kg); DC, diabetic control (normal saline, 10 mL/kg);5 of M,16
NC, normal control (normal saline, 10 mL/kg); DC, diabetic control (normal saline, 10 mL/kg); M,
metformin (500 mg/kg); SF-1, subfraction-1 (250 mg/kg); SF-2, subfraction-2 (250 mg/kg). The val-
metformin
ues (500 as
are expressed mg/kg); SF-1,(nsubfraction-1
mean ± SEM = 6). # indicates(250 mg/kg);
significant SF-2, subfraction-2
differences as compared to (250
the mg/kg).
NC The
### p < 0.001 and # p < 0.05). * indicates significant differences compared to the DC (*** p < 0.001 and
(values are expressed as mean ± SEM (n = 6). # indicates significant
for a broad spectrum of compound screening, it is unable to directly analyze compoundsdifferences as compared to the
*NC
p < 0.05)
(### pas<analyzed
0.001 andusing
#p< Dunnett’s as post hocsignificant
0.05). * indicates test. differences compared to the DC (*** p < 0.001
that are polar and nonvolatile [21]. Chemical derivatization to create volatile forms of
the *compounds
and p < 0.05) as is needed using
analyzed to make them amenable
Dunnett’s as post hocfor this
test.chromatographic analysis.
Figure 3 presents the blood glucose level of active extract, fraction, subfraction,
squalene, as well as controls in streptozotocin-induced diabetic rats. Compared to the
normal rats, streptozotocin induction caused a 2–3-fold increase in blood glucose levels.
Blood glucose levels of methanol extract, chloroform fraction, subfraction-1, squalene,
and metformin-treated groups were significantly decreased on day 12 as compared to
day 0. The descending order of the hypoglycemic activity with reduction percentage was
as follows: metformin = methanol extract (56%) > chloroform fraction = subfraction-1 =
squalene (43%). Interestingly, the blood-glucose-lowering effect of methanol extract was
comparable to that observed in the metformin-treated groups (p < 0.001). Overall, crude
extract exerted a better hypoglycemic effect than fraction and subfraction. As can be seen
in Figure 3, the magnitude of the hypoglycemic effect decreased when the extract was
further fractionated. As compared with the NC, only the ME group showed an insignifi-
cant difference, which further indicated treatment with ME successfully eliminated hy-
perglycemia in the diabetic rats. In addition to that, the analysis also suggested the pos-
sible synergistic effect between squalene and other bioactive compounds of the metha-
nol extract, chloroform fraction, and subfraction-1. Compared to the pharmacological ef-
fects observed with individual isolated compounds, better results are obtained with
whole plant extracts due to the presence of various bioactive compounds that work in
synergy by targeting either the same or different pathways [20].
In reference to the gas chromatography–mass spectrometry analysis (Table 1), five
compounds were detected in chloroform fraction, three in subfraction-1, and one com-
Figure from
pound 3. Effects of S. polyanthum’s
methanol active extract,the
extract. Additionally, fraction, and subfraction
chloroform fraction on
hasthe blood glucose
a higher
Figure 3. Effects of S. polyanthum’s active extract, fraction, and subfraction on the per-
blood glucose
levels ofof
centage streptozotocin-induced
squalene (8.92%) than diabetic rats afterextract
methanol being treated
(7.60%)orally
andtwice daily for twelve
subfraction-1 days.
(4.54%).
levels of streptozotocin-induced
NC,number
normal control (normal saline, diabetic
10 mL/kg); rats
DC, after beingcontrol
diabetic treated orallysaline,
(normal twice daily for twelve days.
The of detected compounds and percentage of squalene, however, do10not mL/kg);
reflectM,
NC, normal(500
metformin control
mg/kg);(normal saline, 10extract
ME, methanolic mL/kg);g/kg);
DC, diabetic control (normal saline, 10 SF-
mL/kg); M,
the magnitude of the observed hypoglycemic(1effect CF, chloroform
of each sample.fraction (500
This further mg/kg);
implies
1, n-hexane(500
metformin fraction (250 mg/kg);
mg/kg); ME, SQ, squalene
methanolic (160(1mg/kg).
extract g/kg); The
CF, values are expressed
chloroform fraction as mean
(500 ±
mg/kg); SF-1,
the presence of other potentially bioactive compounds in the methanol extract that are
SEM (n = fraction
n-hexane 6). # indicates
(250 significantSQ,
mg/kg); differences
squalene as(160
compared
mg/kg).to the
The DC (### p <are
values 0.001). * indicates
expressed as sig- ± SEM
mean
not detected
nificant through
differences gas to
compared chromatography–mass
the NC (*** p < 0.001, ** pspectrophotometry* p < 0.05) analysis.
< 0.01, and ### as analyzedEven
using
(n = 6). #gas
though indicates significant differences as compared to is DC ( p <the
theconsidered 0.001).
gold* indicates
standard significant
Dunnett’s aschromatography–mass
post hoc test. spectrophotometry
differences compared to the NC (*** p < 0.001, ** p < 0.01, and * p < 0.05) as analyzed using Dunnett’s
as post
2.3. hocon
Effect test.
Body Weight
Figure 4 shows the effect of S. polyanthum extract, fraction, and subfraction on body
weight. Normal rats showed a significant increase in body weight after the 12-day exper-
imental period (p < 0.01). As expected, diabetic control rats exerted a significant loss in
body weight (p < 0.05) on day 12. The significant loss of body weight in untreated diabet-
ic rats may be due to the increased muscle wasting and loss of muscle mass and adipose
tissue caused by the excessive breakdown of tissue proteins and fatty acids [22,23]. Sev-
 Molecules 2022, 27, 6814 5 of 15

In reference to the gas chromatography–mass spectrometry analysis (Table 1), five

compounds were detected in chloroform fraction, three in subfraction-1, and one compound
from methanol extract. Additionally, the chloroform fraction has a higher percentage of
squalene (8.92%) than methanol extract (7.60%) and subfraction-1 (4.54%). The number of
detected compounds and percentage of squalene, however, do not reflect the magnitude of
the observed hypoglycemic effect of each sample. This further implies the presence of other
potentially bioactive compounds in the methanol extract that are not detected through gas
chromatography–mass spectrophotometry analysis. Even though gas chromatography–
mass spectrophotometry is considered the gold standard for a broad spectrum of compound
screening, it is unable to directly analyze compounds that are polar and nonvolatile [21].
Chemical derivatization to create volatile forms of the compounds is needed to make them
amenable for this chromatographic analysis.

2.3. Effect on Body Weight

Figure 4 shows the effect of S. polyanthum extract, fraction, and subfraction on body
weight. Normal rats showed a significant increase in body weight after the 12-day ex-
perimental period (p < 0.01). As expected, diabetic control rats exerted a significant loss
in body weight (p < 0.05) on day 12. The significant loss of body weight in untreated
diabetic rats may be due to the increased muscle wasting and loss of muscle mass and
adipose tissue caused by the excessive breakdown of tissue proteins and fatty acids [22,23].
Several studies have reported similar significant weight reductions in untreated diabetic
rats [24–26]. Diabetic rats treated with methanol extract, chloroform fraction, subfraction-1,
and squalene, as well as metformin, also showed a reduction in body weight after 12 days of
the experiment. The reduction, however, was not significant as compared to pre-treatment
levels on day 0. The results suggested that the given treatments managed to prevent further
significant weight loss in diabetic rats. The minimal weight loss of treated groups was due
Molecules 2022, 26, x FOR PEER REVIEWto the gradual decrease in blood glucose levels. Additionally, the effect of the treatments
6 of 16
on body weight might be more prominent if the diabetic rats were treated for a longer
duration. Taher et al. [27] and Azad and Sulaiman [28] reported a significant increase in
body weight of treated diabetic rats only after 14 days of treatments.
250
*
200
*
Body weight (g)

150

100

50

0
NC DC M ME CF SF-1 SQ

Pre Treatment Post Treatment

Figure 4. Effects of S. polyanthum on the body weights of streptozotocin-induced diabetic rats. All
Figure 4. Effects
treatments wereofadministered
S. polyanthumorally
on thetwice
bodydaily
weights ofdays
for 12 streptozotocin-induced
as follows: NC, normaldiabetic rats.(saline,
control All
treatments were administered orally twice daily for 12 days as follows: NC, normal control (saline,
10 mL/kg); M, metformin (500 mg/kg); DC, diabetic control (saline, 10 mL/kg); ME, methanol extract
10 mL/kg); M, metformin (500 mg/kg); DC, diabetic control (saline, 10 mL/kg); ME, methanol ex-
(1 g/kg); CF, chloroform fraction (500 mg/kg); SF-1, n-hexane fraction (250 mg/kg); SQ, squalene
tract (1 g/kg); CF, chloroform fraction (500 mg/kg); SF-1, n-hexane fraction (250 mg/kg); SQ, squa-
(160
lene mg/kg).
(160 mg/kg).TheThevalues
valuesareare
expressed
expressedas as
mean
mean± SEM
± SEM(n (n
= 6). * indicates
= 6). significant
* indicates differences
significant differ-
between pre-treatment (day 0) and post-treatment (day 12) of the same group as
ences between pre-treatment (day 0) and post-treatment (day 12) of the same group as analyzed analyzed by using
bythe paired
using t-test (*t-test
the paired p < 0.05).
(* p < 0.05).

2.4. Effect on Lipid Profile

Figure 5 presents the effects of S. polyanthum on the lipid profiles of streptozotocin-
induced diabetic rats. As compared with the diabetic control, only subfraction-1 and
squalene showed the ability to decrease total cholesterol levels significantly (p < 0.05), an
 Molecules 2022, 27, 6814 6 of 15

2.4. Effect on Lipid Profile

Figure 5 presents the effects of S. polyanthum on the lipid profiles of streptozotocin-
induced diabetic rats. As compared with the diabetic control, only subfraction-1 and
squalene showed the ability to decrease total cholesterol levels significantly (p < 0.05), an
effect which was also observed in the metformin-treated group (p < 0.05). Metformin and
squalene-treated groups also significantly reduced low-density lipoprotein levels (p < 0.001
and p < 0.05 respectively). Those findings confirmed the promising hyperlipidemic effect
of squalene, as reported by previous studies. Liu et al. [29] showed that a high squalene
Molecules 2022, 26, x FOR PEER REVIEW diet significantly reduced triglycerides and cholesterol levels in rats. Several years later, 7 of 16
Gabas-Rivera et al. [30] further suggested that squalene managed to reduce cholesterol
levels in a dose-dependent manner.

Figure 5. Effects of S. polyanthum on the lipid profiles of streptozotocin-induced diabetic rats after
Figure 5. Effects of S. polyanthum on the lipid profiles of streptozotocin-induced diabetic rats after
12-day twice-daily
12-day twice-dailyoral
oraltreatments. NC,normal
treatments. NC, normal control
control (normal
(normal saline,saline, 10 mL/kg);
10 mL/kg); DC, diabetic
DC, diabetic
control (normal saline, 10 mL/kg); M, metformin (500 mg/kg); ME, methanol
control (normal saline, 10 mL/kg); M, metformin (500 mg/kg); ME, methanol extract (1 g/kg); extract (1CF,
g/kg); CF,
chloroform
chloroform fraction (500 mg/kg); SF-1, subfraction-1 (250 mg/kg); SQ, squalene (160 mg/kg); Chol, Chol,
fraction (500 mg/kg); SF-1, subfraction-1 (250 mg/kg); SQ, squalene (160 mg/kg);
cholesterol; TG,TG,
cholesterol; triglycerides;
triglycerides; HDL, high-density
HDL, high-density lipoprotein;
lipoprotein; LDL,LDL, low-density
low-density lipoprotein.
lipoprotein. The The
values are expressed as mean ± SEM (n = 6); * p < 0.05 and *** p < 0.001 indicate significant
values are expressed as mean ± SEM (n = 6); * p < 0.05 and *** p < 0.001 indicate significant differences differ-
encescompared
compared to DC
to the the as
DC as analyzed
analyzed using Dunnett
using Dunnett as post hoc astest.
post hoc test.

Non-significant
2.5. Insulin changes in lipid parameters
Level and Immunohistochemistry were seen in diabetic rats treated with
Assessment
methanol extract and chloroform fraction. Diabetes was shown to be associated with hy-
Figure 6 shows
perlipidemia theoptimal
[31], and immunohistochemically stained
glycemic control ameliorates pancreatic
lipid tissues under
profile abnormalities [32]. 40 × 10
magnification power. The brown-colored area represents insulin in the viable
However, in the author’s study, normalization of plasma glucose levels in methanol extract β-cells. In
the Islets of Langerhans
and chloroform fractionof the normal
treated rat,not
groups did clear andthe
restore large
lipidbrownish
parameters, spots werethat
implying seen. The
factors, in addition to glycemic control, are also involved. An example
β-cells were found to make up 87.92 ± 1.9% of Islets of Langerhans in the normal of this factor is rats. In
insulin. Insulin has been reported to prevent hypercholesterolemia in type 1 diabetes. In-
the streptozotocin-induced diabetic rats, the size of the brown-colored areas was signifi-
sulin, by suppressing the hepatic FoxO1 gene, lowered 12α-hydroxylated bile acids, plasma
cantly (p < 0.001)
cholesterol, and reduced
cholesteroland comprised
absorption 21.36
[33]. This ± 3.7%
means of the total
the cholesterol islets
levels of the
remain highLanger-
hansinarea (Figure 7A).
the deficiency With the breaking of the DNA strand, interruption of the glucose
of insulin.
transport, and interference with the function of glucokinase, the streptozotocin damaged
the β-cells [34]. All the treated groups showed a tendency to have larger immunostained
areas. The effect, however, was insignificant when compared with the diabetic control
group. This indicates that the treatments were ineffective in terms of regenerating and
protecting viable pancreatic β-cells. Figure 7B depicts the concentration of serum insulin
in the treated groups. As expected, the serum insulin levels in all diabetic rats were
three-fold less than those of the normal control (p < 0.001). Only subfraction-1 signifi-
Molecules 2022, 27, 6814 7 of 15

2.5. Insulin Level and Immunohistochemistry Assessment

Figure 6 shows the immunohistochemically stained pancreatic tissues under 40 × 10 mag-
nification power. The brown-colored area represents insulin in the viable β-cells. In the
Islets of Langerhans of the normal rat, clear and large brownish spots were seen. The
β-cells were found to make up 87.92 ± 1.9% of Islets of Langerhans in the normal rats. In
the streptozotocin-induced diabetic rats, the size of the brown-colored areas was signifi-
cantly (p < 0.001) reduced and comprised 21.36 ± 3.7% of the total islets of the Langerhans
area (Figure 7A). With the breaking of the DNA strand, interruption of the glucose trans-
port, and interference with the function of glucokinase, the streptozotocin damaged the
β-cells [34]. All the treated groups showed a tendency to have larger immunostained areas.
The effect, however, was insignificant when compared with the diabetic control group.
This indicates that the treatments were ineffective in terms of regenerating and protecting
viable pancreatic β-cells. Figure 7B depicts the concentration of serum insulin in the treated
groups. As expected, the serum insulin levels in all diabetic rats were three-fold less than
those of the normal control (p < 0.001). Only subfraction-1 significantly increased the level
of serum insulin as compared to the diabetic controls (p < 0.05). This finding explained
the positive effect of subfraction-1 in lowering cholesterol levels and confirmed the role
Molecules 2022, 26, x FOR PEER REVIEW 8 of 16
of insulin in normalizing cholesterol levels. Semova et al. [33], who studied the role of
insulin regulation on the cholesterol mechanism in mouse model type 1 and humans with
type 1 diabetes, have suggested that insulin modulated the level of plasma cholesterol via
Effects of the
inhibition treatment
of hepatic on the
FoXO1, parameters
which such
led to the as body of
reduction weight and lipid profile
12α-hydroxylated bilecan be
acids,
cholesterol
more profoundabsorption, and study
in a longer plasma cholesterol levels.
period.

Insulin Exocrine
secreting area area
Insulin
Insulin secreting
Islet of secreting area area
Langerhans
Islet of
Exocrine Islet of Langerhans
area Langerhans

Exocrine
area

50µm 50µm 50µm

NC DC ME

Exocrine
area

Insulin
secreting area

Islet of
Langerhans
Islet of
Langerhans Insulin
secreting
area

Exocrine
area

50µm 50µm 50µm

CF SF-1 SQ

the β-cells
Figure 6. Immunohistochemistry staining of the β-cells pancreas after being treated
treated for
for 12
12 days.
days.

Considering diabetes mellitus is a multifactorial chronic disease, further research is

needed to better understand the possible mechanism of the hypoglycemic action of this
plant by exploring the effect of S. polyanthum on the hepatic glucose production, intestinal
glucose absorption, and peripheral glucose uptake in the muscle and adipose tissues.
Additionally, the current study has limitations in terms of the duration of the treatment.
Effects of the treatment on the parameters such as body weight and lipid profile can be
more profound in a longer study period.
 Islet of
Langerhans
Islet of
Langerhans Insulin
secreting
area

Exocrine
area

50µm 50µm 50µm

Molecules 2022, 27, 6814 CF SF-1 SQ 8 of 15

Figure 6. Immunohistochemistry staining of the β-cells pancreas after being treated for 12 days.

Figure 7. (A) Percentage of the

Figure immunostained
7. (A) area of each area
Percentage of the immunostained treated group
of each treatedand (B)and
group Concentration
(B) Concentration
of serum insulin of each of serum insulin
treated of each NC,
group. treated group. NC,
normal normal(saline,
control control (saline, 10 mL/kg);
10 mL/kg); M,M,metformin
metformin (500
mg/kg); DC, diabetic control (saline, 10 mL/kg); ME, methanol extract (1 g/kg); CF, chloroform
(500 mg/kg); DC, diabetic control
fraction (saline,SF-1,
(500 mg/kg); 10 mL/kg); ME, methanol
n-hexane fraction extract
(250 mg/kg); (1 g/kg);
SQ, squalene (160CF, chloroform
mg/kg). The values
fraction (500 mg/kg); SF-1, n-hexane
are expressed fraction
as mean ± SEM(250 mg/kg);
(n = 6). SQ,
* indicates squalene
significant (160 mg/kg).
differences Thetovalues
as compared the NC. #
indicates significant differences compared to the DC (*** p < 0.001 and # p < 0.05) as analyzed using
± SEM
are expressed as meanDunnett (n = 6). * indicates significant differences as compared to the NC.
as post hoc test.
# indicates significant differences compared to the DC (*** p < 0.001 and # p < 0.05) as analyzed using
Dunnett as post hoc test.2.6. Molecular Docking Analysis
Using iGEMDOCK, a molecular docking simulation was carried out to determine
2.6. Molecular DockingtheAnalysis
effectiveness of squalene’s binding to the chosen diabetes proteins. Metformin was
applied for comparison purposes. Figure 8 shows the binding energy of all two ligands
Using iGEMDOCK, a molecular docking simulation was carried out to determine
on the 13 protein targets. Based on the analysis of binding energy, the lowest negative
the effectiveness of squalene’s
scores of totalbinding to thea chosen
energy indicate diabetes
higher affinity between proteins.
the ligandMetformin
and protein’swas
recep-
applied for comparison tors, purposes.
which furtherFigure
signifies8 that
shows the binding
the ligands energy of
could modulate all two
or inhibit theligands
respective
on the 13 protein targets. Based on the analysis of binding energy, the lowest negative
scores of total energy indicate a higher affinity between the ligand and protein’s recep-
tors,2022,
Molecules which
26, x FORfurther signifies that the ligands could modulate or inhibit the respective
PEER REVIEW 9 of 16
protein target. The binding energy of squalene to all the target proteins was found to
be lower than that of the standard drugs, metformin, except for 4Y14, 1XU7, and 1IR3.
Squalene showed theprotein highesttarget. The binding
affinity against energy
2Q5S of (squalene to all the target
−75.9 kcal/mol), proteins was
followed by found
2HWQ to be
lower than that of the standard drugs, metformin, except for 4Y14, 1XU7, and 1IR3.
(−71 kcal/mol), 1ZON (−65.56 kcal/mol), and 1FM9 (−64.28 kcal/mol). Receptors 2Q5S,
Squalene showed the highest affinity against 2Q5S (−75.9 kcal/mol), followed by 2HWQ
2HWQ and 1FM9 are(−71 thekcal/mol),
proteins1ZONthat regulate and control
(−65.56 kcal/mol), adipogenesis,
and 1FM9 energy
(−64.28 kcal/mol). balance,
Receptors 2Q5S,
and biosynthesis of lipids [35–37]. This in silico analysis is in agreement with the findings
2HWQ and 1FM9 are the proteins that regulate and control adipogenesis, energy bal-
of Mirmiranpour et ance, andon
al. [14] biosynthesis
the effect of lipids [35–37]. This
of squalene on in silicoprofiles
lipid analysis is
inintype
agreement with the
2 diabetic
findings of Mirmiranpour et al. [14] on the effect of squalene on lipid profiles in type 2
patients. Further study is needed to predict the active site and associated amino acids of
diabetic patients. Further study is needed to predict the active site and associated amino
target diabetic proteins. acids of target diabetic proteins.

Figure 8. Binding energies of the ligands against thirteen protein receptors related to diabetes
Figure 8. Binding energies of the ligands against thirteen protein receptors related to diabetes mellitus.
mellitus. 1FM9, heterodimer of the human RXR α and peroxisome proliferator–activated receptor
1FM9, heterodimer of theγ human RXR αdomains
ligand binding and peroxisome
bound with proliferator–activated receptor
9–cis retinoic acid and GI262570 ligand binding
andγco–activator peptides;
1IR3, phosphorylated insulin receptor tyrosine kinase in complex with peptide substrate and ATP
analog; 1XU7, tetrameric 11B–HSD1; 1ZON, CD11A I–domain without bound cation; 2HWQ, pe-
roxisome proliferator–activated receptor agonists; 2Q5S, peroxisome proliferator–activated recep-
tor γ bound to partial agonist NTZDPA; 2QMJ, N–terminal subunit of human maltase–
glucoamylase in complex with acarbose; 2ZJ3, isomerase domain of human glucose:fructose–6–
phosphate amidotransferase; 3C45, human dipeptidyl peptidase IV/CD26 in complex with a fluo-
roolefin inhibitor; 3CTT, N–terminal human maltase–glucoamylase with casuarina; 3L2M, pig
Molecules 2022, 27, 6814 9 of 15

domains bound with 9–cis retinoic acid and GI262570 and co–activator peptides; 1IR3, phos-
phorylated insulin receptor tyrosine kinase in complex with peptide substrate and ATP analog;
1XU7, tetrameric 11B–HSD1; 1ZON, CD11A I–domain without bound cation; 2HWQ, peroxisome
proliferator–activated receptor agonists; 2Q5S, peroxisome proliferator–activated receptor γ bound to
partial agonist NTZDPA; 2QMJ, N–terminal subunit of human maltase–glucoamylase in complex
with acarbose; 2ZJ3, isomerase domain of human glucose:fructose–6–phosphate amidotransferase;
3C45, human dipeptidyl peptidase IV/CD26 in complex with a fluoroolefin inhibitor; 3CTT, N–
terminal human maltase–glucoamylase with casuarina; 3L2M, pig pancreatic alpha–amylase with
alpha–cyclodextrin; 4A5S, human DPP4 in complex with a novel heterocyclic DPP4 inhibitor; 4Y14,
tyrosine phosphatase 1B complexed with inhibitor.

3. Materials and Methods

3.1. Chemicals
Metformin 500 mg, a standard oral antidiabetic drug, was used as the positive control.
Streptozotocin and squalene (CID 638072) were purchased from Sigma-Aldrich Chemical
Company (St. Louis, MO, USA). All reagents and chemicals used are analytical grades.

3.2. Plant Collection and Identification

Syzygium polyanthum leaves were collected from Titi Kuning, Medan, Indonesia (Ge-
ographical coordinates: 3.522988, 98.682834) from July to October 2011. The plant was
identified by Dr. Nursahara Pasarbibu from the School of Biological Sciences, Bioteknologi
no.1 Kampus USU, University of Sumatera Utara, Medan, Indonesia (voucher specimen
number: no.13/MEDA/2012).

3.3. Preparation of Samples

The dried leaves were powdered using a milling machine. To obtain the extracts of S.
polyanthum’s, about 1.5 kg of the powder was sequentially macerated (40 ◦ C–60 ◦ C) with
4.9 L each of the respective solvents: petroleum ether, chloroform, and methanol. The
extracts were filtered using Whatman No.1 filter paper and concentrated in vacuo by using
a rotary evaporator (Labortechnik, AG CH-9230 Flawil, Switzerland) at reduced pressure.
The concentrated extracts were dried in an oven (40 ◦ C) until traces of the organic solvent
completely evaporated. An earlier study has shown that methanol extract exerted the
most potent hypoglycemic effect [12]. Hence, in this study, the active methanol extract was
selected for further bioassay-guided fractionation.
The active methanol extract (25 g) was sequentially fractionated by a liquid–liquid par-
tition with four solvents, chloroform (250 mL), ethyl acetate (250 mL), n-butanol (250 mL),
and water. This resulted in four fractions of chloroform, ethyl acetate, n-butanol, and water.
The water fraction underwent lyophilization using a freeze dryer (Labonco Corporation,
Kansas, MO, USA). The preliminary result showed that chloroform fraction significantly
lowered the blood glucose level of streptozotocin-induced diabetic rats as compared to the
control. Thus, the most active fraction, the chloroform fraction, was further fractionated
by dissolving 0.5 g in n-hexane (100 mL). Meanwhile, the n-hexane was drained and re-
plenished until no color had formed. The n-hexane fraction was filtered, concentrated by a
rotary evaporator, and freeze-dried to obtain subfraction 1. The remaining residue, which
was dissolved in chloroform, was also filtered. The chloroform portion was concentrated in
vacuo to yield subfraction 2. Figure 9 depicts the schematic extraction and fractionation
of S. polyanthum leaves. All extracts, fractions, and subfractions were kept in a freezer
(−20 ◦ C) until further use. Doses were freshly prepared using 5% Tween 80 in 0.9% normal
saline prior to oral administration.

3.4. Gas Chromatography–Mass Spectrometry (GC–MS) Analysis

For GC–MS analysis, 4 mg of each active methanol extract, chloroform fraction, and
subfraction-1 were respectively dissolved in 1 mL of chloroform. The reference com-
pound, squalene (410.72 g/mol, >98% liquid), was diluted in methanol at the concen-
 significantly lowered the blood glucose level of streptozotocin-induced diabetic rats as
compared to the control. Thus, the most active fraction, the chloroform fraction, was fur-
ther fractionated by dissolving 0.5 g in n-hexane (100 mL). Meanwhile, the n-hexane was
drained and replenished until no color had formed. The n-hexane fraction was filtered,
Molecules 2022, 27, 6814 concentrated by a rotary evaporator, and freeze-dried to obtain subfraction101.of 15 The re-
maining residue, which was dissolved in chloroform, was also filtered. The chloroform
portion was concentrated in vacuo to yield subfraction 2. Figure 9 depicts the schematic
tration ofand
extraction 1 mg/mL. The presence
fractionation of squalene in
of S. polyanthum methanol
leaves. extract, chloroform
All extracts, fractions, fraction,
and subfrac-
tions were kept in a freezer (−20 °C) until further use. Doses were freshlyofprepared
and subfraction-1 was confirmed by using ion fractionation. The percentage squalene using
was calculated based on the peak heights in methanol extract, chloroform fraction, and
5% Tween 80 in 0.9% normal saline prior to oral administration.
subfraction-1 respective spectra.

Figure 9. 9.Schematic
Figure Schematicrepresentation ofbioactive-guided
representation of bioactive-guided fractionation
fractionation of S.of S. polyanthum
polyanthum leaves.leaves.

3.5. In Vivo Antihyperglycemic Studies

3.4. Gas Chromatography–Mass Spectrometry (GC–MS) Analysis
3.5.1. Induction of Type 1 Diabetes Mellitus in Rats
For GC–MS analysis, 4 mg of each active methanol extract, chloroform fraction, and
Healthy male Sprague Dawley rats weighing between 200 and 250 g were obtained
subfraction-1 were
from the Animal respectively
Research dissolved
and Service in 1 mL Sains
Centre, Universiti of chloroform. The reference
Malaysia, Penang, Malaysia. com-
pound, squalene (410.72 g/mol, >98% liquid), was diluted in methanol at the
Before commencement of the study, animals were acclimatized in the Animal Transit Room, concentra-
tion of 1 of
School mg/mL. The presence
Pharmaceutical of squalene
Sciences, UniversitiinSains
methanol extract,
Malaysia, chloroform
for one ◦ C and
fraction,
week at 20–22
subfraction-1 was confirmed
with a 12 h light/dark cycle. by using ion fractionation. The percentage of squalene was
Freshly prepared streptozotocin
calculated based on the peak heights in 0.9% sodium chloride
in methanol was
extract, injected intraperitoneally
chloroform fraction, and sub-
at a dose of 55 mg/kg to
fraction-1 respective spectra. 16-h fasted rats. The rats were provided with 10% dextrose
drinking water for the first 24 h to avoid the incidence of fatal hypoglycemia. After 72 h,
the blood glucose levels of the fasting animals were measured, and animals with blood
glucose levels above 200 mg/dL (11 mmol/L) were selected for the study.

3.5.2. Bioassay-Guided Antihyperglycemic Activity of S. polyanthum

Hypoglycemic activity of S. polyanthum’s fractions and subfractions were assessed
using streptozotocin-induced diabetic rats. Diabetic rats were randomly divided into
groups of six (n = 6). Normal saline (10 mL/kg BW) was given and acted as a negative
control group, and metformin at the dose of 500 mg/kg BW was used as the positive control.
To screen the hypoglycemic activity of fractions, diabetic rats were treated with 500 mg/kg
of chloroform, ethyl acetate, n-butanol, and water fractions, respectively. All treatments
were administered orally, twice daily, for 6 days. Blood glucose levels were measured
before and after the treatment. The same step was repeated to evaluate the hypoglycemic
activity of subfractions, subfraction-1, and subfraction-2. These subfractions, at the dose
of 250 mg/kg BW, were given orally, and blood glucose levels before and after treatment
were measured.
Finally, once the active fraction and subfraction have been determined, the hypo-
glycemic activity of the active extract, fraction, and subfraction, as well as squalene, was
conducted again at the same time to avoid time period bias. The procedure is summarized
in Figure 10. The treated groups were as follows: Four groups of diabetic rats (n = 6) were
respectively treated with methanol extract (1 g/kg B.W.), chloroform fraction (500 mg/kg
B.W.), subfraction-1 (250 mg/kg B.W.), and squalene (160 mg/kg B.W.). The fifth and sixth
groups received metformin (500 mg/kg B.W.) and normal saline (10 mL/kg B.W.) and
Molecules 2022, 27, 6814 11 of 15

served as respective positive control and diabetic control. Six normoglycemic rats were
treated with saline (10 mL/kg) and served as the normal control. The minimum effective
Molecules 2022, 26, x FOR PEER REVIEW 12 of 16
dose of squalene was determined previously, as shown in Supplementary Figure S1. All
treatments were administered orally twice daily for 12 days. After a 12-day treatment, the
rats were euthanized using carbogen (95% CO2 and 5% O2 ), and a blood sample (3 mL) was
Figure 10. In vivo antihyperglycemic test.
obtained via cardiac puncture and centrifuged at 3000 rpm for 10 min to collect the serum.
The serum was stored at −20 ◦ C until biochemical parameters analysis. The pancreas was
harvested for immunohistochemistry analysis.

Figure 10. In vivo antihyperglycemic test.

3.5.3. Measurement of Fasting Blood Glucose
3.5.3. Blood
Measurement of Fasting
glucose level Blood Glucose
was measured at two interval points, day 0 (before treatment)
and Blood
day 12glucose
(after repeated
level wastreatment).
measured at Thetworats were fasted
interval points,overnight before
day 0 (before blood glu-
treatment) and
cose measurement. Approximately fifty microliters of blood was obtained
day 12 (after repeated treatment). The rats were fasted overnight before blood glucose from the tail
vein of each rat,
measurement. and blood glucose
Approximately level was determined
fifty microliters of blood was using Accu-Check
obtained from theAdvantage
tail vein of
Clinical
each rat, Glucose Meter
and blood (Roche
glucose Diagnostics
level Co., Indianapolis,
was determined IN, USA).Advantage Clinical
using Accu-Check
Glucose Meter (Roche Diagnostics Co., Indianapolis, IN, USA).
3.5.4. Measurement of Serum Insulin Level, Lipid Profile, and Body Weight
3.5.4. The
Measurement of Serum
concentration Insulin
of insulin Level,
in the serumLipid
wasProfile, and Body
determined Weight using an ul-
in triplicates
The concentration
tra-sensitive rat insulin of insulin
ELISA kit in the serum
(Crystal Chemwas Inc,determined in triplicates
Elk Grove Village, Illinois,using
USA).an
ultra-sensitive rat insulin
Total cholesterol, ELISAand
triglycerides, kit (Crystal Chem Inc,
HDL cholesterol Elk determined
were Grove Village, IL, an
using USA). To-
auto-
tal cholesterol,
mated Siemenstriglycerides, and HDL cholesterol
ADVIA 2400 Chemistry were determined
Analyzer (Erlangen, Germany).using ancholesterol
Total automated
Siemens ADVIA 2400
and triglycerides wereChemistry Analyzer
measured based (Erlangen,
on the Germany).
enzyme-coupled Total cholesterol
reaction, whereas HDL and
triglycerides were
cholesterol was measured based
determined using on the enzyme-coupled
a direct HDL-cholesterolreaction, whereas
assay. LDL HDL choles-
cholesterol was
terol was determined
calculated using a direct
using the Friedewald HDL-cholesterol
equation. The body assay. LDL
weights of cholesterol
the rats werewasmeasured
calculated
using the Friedewald equation. The body weights of the rats were measured
on day 0 and day 12 by using an electronic balance (Navigator , Ohaus Corporation,
TM on day 0 and day
12 by using an electronic balance (Navigator TM , Ohaus Corporation, Nanikon, Switzerland).
Nanikon, Switzerland).

3.5.5.
3.5.5. Immunohistochemistry
Immunohistochemistry Study Study of of Pancreas
Pancreas
AA histological
histological assessment
assessment of of the
the pancreas
pancreaswaswasconducted
conductedaccording
accordingtotothetheimmuno-
immuno-
histochemical method described by Yusoff et al. [38]. The pancreatic
histochemical method described by Yusoff et al. [38]. The pancreatic tissues were fixed tissues were fixed
in 10% buffered formalin, processed using a tissue processor, and embedded
in 10% buffered formalin, processed using a tissue processor, and embedded in paraffin. in paraffin.
The
Theparaffin-embedded
paraffin-embedded tissues
tissues were
were sectioned
sectioned into
into 55 µm
μm slices
slices and
and mounted
mounted on on aapoly-L-
poly-
lysine-coated
L-lysine-coated microscope slide. Immunohistochemically staining was performedusing
microscope slide. Immunohistochemically staining was performed using a
guinea-pig
a guinea-pig polyclonal
polyclonalantibody
antibody of rat
of insulin. TheThe
rat insulin. sections werewere
sections treated with with
treated 3% hydrogen
3% hy-
peroxide in methanol to quench endogenous peroxidases, followed
drogen peroxide in methanol to quench endogenous peroxidases, followed by washing by washing in buffer.
Then, the sections were incubated in a diluted normal serum for 20
in buffer. Then, the sections were incubated in a diluted normal serum for 20 min. The min. The primary
antibody, guinea-pig
primary antibody, polyclonal
guinea-pig insulin antibody,
polyclonal was applied
insulin antibody, for 30 min,
was applied followed
for 30 min, fol-by
buffer washing for 5 min. The sections were incubated for 30 min with the
lowed by buffer washing for 5 min. The sections were incubated for 30 min with the bio- biotinylated sec-
ondary
tinylatedantibody, followed
secondary by washing.
antibody, followed Following
by washing. a further 30-min
Following incubation
a further 30-minperiod in a
incuba-
Vectastain ABC kit, washing was performed for 5 min before adding diaminobenzidine
tion period in a Vectastain ABC kit, washing was performed for 5 min before adding di- for
3–5 min. The sections
aminobenzidine were
for 3–5 min.slightly counterstained
The sections with Harris
were slightly Hematoxylin,
counterstained dehydrated,
with Harris He-
matoxylin, dehydrated, cleared, and mounted. The pancreatic islets were examined un-
der a light microscope (Leica® DMi1, Leica Microsystems, Wetzlar, Hesse, Germany). An
image analyzer (Leica® microsystem Qwin plus) was used to analyze the digital image
and calculate the percentage (%) of the insulin-containing area of β cells.
Molecules 2022, 27, 6814 12 of 15

cleared, and mounted. The pancreatic islets were examined under a light microscope
(Leica® DMi1, Leica Microsystems, Wetzlar, Hesse, Germany). An image analyzer (Leica®
microsystem Qwin plus) was used to analyze the digital image and calculate the percentage
(%) of the insulin-containing area of β cells.

3.6. Molecular Docking Analysis

Selection of protein targets was conducted by referring to Rao et al. [39]; 3D struc-
tures of 13 proteins that are vitally important in diabetes mellitus were retrieved from
the Research Collaboratory for Structural Bioinformatics (RCSB) Protein Data Bank (https:
//www.rcsb.org/structure/2Q5S), accessed on 20 September 2022 and saved in PDB format.
Proteins with PDB ID: 1FM9 (Heterodimer of the human retinoic acid receptor α and per-
oxisome proliferator-activated receptor γ ligand binding domains bound with 9-cis retinoic
acid and GI262570 and co-activator peptides), 1IR3 (Phosphorylated insulin receptor ty-
rosine kinase in complex with peptide substrate and ATP analog), 1XU7 (Tetrameric 11B-
HSD1), 1ZON (CD11A I-domain without bound cation), 2HWQ (Peroxisome proliferator-
activated receptor agonists), 2Q5S (Peroxisome proliferator-activated receptor γ bound to
partial agonist NTZDPA), 2QMJ (N-terminal subunit of human maltase-glucoamylase in
complex with acarbose), 2ZJ3 (Isomerase domain of human glucose:fructose-6-phosphate
amidotransferase), 3C45 (Human dipeptidyl peptidase IV/CD26 in complex with a flu-
oroolefin inhibitor), 3CTT (N-terminal human maltase-glucoamylase with casuarine),
3L2M (Pig pancreatic alpha-amylase with alpha-cyclodextrin), 4A5S (Human DPP4 in
complex with a novel heterocyclic DPP4 inhibitor), and 4Y14 (Tyrosine phosphatase 1B
complexed with inhibitor) were employed in focused molecular docking investigations
on diabetes receptor proteins. Preparation of ligand was conducted applying canonical
SMILES format of the compounds. Canonical SMILES of ligand squalene was retrieved
from the PubChem compound database (https://pubchem.ncbi.nlm.nih.gov/, accessed
on 20 September 2022) with PubChem CID: 638072, and the three-dimensional structure
of the molecule was simulated using the online server NovoPro Bioscience Inc., Shanghai,
China (https://www.novoprolabs.com/tools/smiles2pdb, accessed on 20 September 2022)
and was saved in PDB format. Comparative analysis of the ligand was conducted against
metformin drug obtained from PubChem database metformin (PubChem CID:4091). The
2D structures of squalene and metformin were generated using Novoprolabs (Figure 8).
Molecular docking between the ligand and the receptors was carried out using the Generic
Evolutionary Method for molecular DOCKing (iGEMDOCKv4.2). iGEMDOCK employs
generic evolutionary algorithms (flexible docking method). Docking each ligand to the
13 target proteins was carried out at the standard docking option of 70 generations, 200 pop-
ulation size, and 2 solutions. Bond energies of squalene were compared to those of the
standard drug metformin.

3.7. Statistical Analysis

The data were expressed as mean ± standard error of the mean (SEM). Statistical
significance was determined by Graphpad prism version 7. One-way ANOVA was used,
followed by Dunnet’s post hoc test. Pre-treatment and post-treatment comparisons were
performed using the paired t-test. Differences were considered significant, with the p-value
being less than 0.05.

4. Conclusions
In conclusion, the present work successfully establishes the hypoglycemic activity of
S. polyanthum leaf extract, demonstrating evidence that S. polyanthum is the most effective
antidiabetic plant when administered as a whole extract rather than fractions. This suggests
that the hypoglycemic effect is a product of the synergy of several compounds in the
plant rather than a few or a single compound therein. In addition to that, the observed
hypoglycemic effect has no association with enhanced insulin secretion by β-cells. These
Molecules 2022, 27, 6814 13 of 15

warrant future studies to explore the possible mechanism of the hypoglycemic effect of S.
polyanthum and squalene at the extrapancreatic level.

Supplementary Materials: The following supporting information can be downloaded at: https://
www.mdpi.com/article/10.3390/molecules27206814/s1, Figure S1: Effects of 20 mg/kg, 40 mg/kg,
80 mg/kg, and 160 mg/kg of squalene, administered twice daily, on the blood glucose levels of
streptozotocin-induced diabetic rats.
Author Contributions: Conceptualization, M.Z.A. and M.A.; Formal analysis, T.W. and N.A.Y.;
Funding acquisition, M.Z.A.; Investigation, T.W. and N.A.Y.; Supervision, M.Z.A.; Writing—original
draft, T.W.; Writing—review and editing, I.B. All authors have read and agreed to the published
version of the manuscript.
Funding: This research was funded by the Ministry of Higher Education, Malaysia, Fundamental
Research Grant Scheme, FRGS (203/PFARMASI/6711304).
Institutional Review Board Statement: The study was conducted according to the guidelines of
the Declaration of Helsinki and approved by the Institutional Animal Care and Use Committee of
Universiti Sains Malaysia (Approval number: USM/Animal Ethics Approval/2012/(76)(366)).
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Acknowledgments: The authors express their appreciation to Roseli Hasan from Pharmacology
Laboratory, School of Pharmaceutical Sciences, USM, for the technical and laboratory assistance.
Conflicts of Interest: The authors declare no conflict of interest.
Sample Availability: Samples of the compounds are available from the authors.

Abbreviations
The following abbreviations are used in this manuscript:
n-BF n-butanol fraction
CF Chloroform extract
Chol Cholesterol
DC Diabetic control
EAF Ethyl acetate fraction
GCMS Gas chromatography–mass spectrometry
HDL High-density lipoprotein
LDL Low-density lipoprotein
M Metformin
ME Methanol extract
NC Normal control
RT Retention time
SEM Standard error mean
SF-1 Subfraction-1
SF-2 Subfraction-2
SQ Squalene
TG Triglycerides
WF Water fraction

References
1. Dandekar, P.; Ramkumar, S.; RaviKumar, A. Chapter 12—Structure–Activity Relationships of Pancreatic α-Amylase and α-
Glucosidase as Antidiabetic Targets. In Studies in Natural Products Chemistry; Atta-ur-Rahman, Ed.; Elsevier: Amsterdam, The
Netherlands, 2021; pp. 381–410. ISBN 9780323910989.
2. Mouri, M.; Badireddy, M. Hyperglycemia. Available online: https://www.ncbi.nlm.nih.gov/books/NBK430900/ (accessed on
1 June 2022).
3. Kaul, K.; Tarr, J.M.; Ahmad, S.I.; Kohner, E.M.; Chibber, R. Introduction to Diabetes Mellitus. Adv. Exp. Med. Biol. 2012, 771, 1–11.
[CrossRef] [PubMed]
Molecules 2022, 27, 6814 14 of 15

4. IDF. IDF Diabetes Atlas, 10th Edition. Available online: https://diabetesatlas.org/atlas/tenth-edition/ (accessed on 14 Septem-
ber 2022).
5. Veeresham, C. Natural Products Derived from Plants as a Source of Drugs. J. Adv. Pharm. Technol. Res. 2012, 3, 200. [CrossRef]
6. Najmi, A.; Javed, S.A.; Al Bratty, M.; Alhazmi, H.A. Modern Approaches in the Discovery and Development of Plant-Based
Natural Products and Their Analogues as Potential Therapeutic Agents. Molecules 2022, 27, 349. [CrossRef]
7. Das, S.K.; Samantaray, D.; Sahoo, S.K.; Pradhan, S.K.; Samanta, L.; Thatoi, H. Bioactivity Guided Isolation of Antidiabetic and
Antioxidant Compound from Xylocarpus Granatum, J. Koenig Bark. 3 Biotech 2019, 9, 198–206. [CrossRef] [PubMed]
8. El Deeb, K.S.; Eid, H.H.; Ali, Z.Y.; Shams, M.M.; Elfiky, A.M. Bioassay-Guided Fractionation and Identification of Antidiabetic
Compounds from the Rind of Punica granatum Var. nana. Nat. Prod. Res. 2019, 35, 2103–2106. [CrossRef] [PubMed]
9. Phukhatmuen, P.; Raksat, A.; Laphookhieo, S.; Charoensup, R.; Duangyod, T.; Maneerat, W. Bioassay-Guided Isolation and
Identification of Antidiabetic Compounds from Garcinia Cowa Leaf Extract. Heliyon 2020, 6, e03625. [CrossRef]
10. Milow, P.; Ramli, M.R.; Chooi, O.H. Preliminary Survey on Plants in Home Gardens in Pahang, Malaysia. J. Biodivers. 2017, 1,
19–25. [CrossRef]
11. Widyawati, T.; Purnawan, W.W.; Atangwho, I.J.; Yusoff, N.A.; Ahmad, M.; Asmawi, M.Z. Antidiabetic Activity of Syzygium
polyanthum (Wight) Leaf Extract, the Most Commonly Used Herb among Diabetic Patients in Medan, North Sumatera, Indonesia.
Int. J. Pharm. Sci. Res. 2015, 6, 1698–1704.
12. Widyawati, T.; Yusoff, N.A.; Asmawi, M.Z.; Ahmad, M. Antihyperglycemic Effect of Methanol Extract of Syzygium polyanthum
(Wight.) Leaf in Streptozotocin-Induced Diabetic Rats. Nutrients 2015, 7, 7764–7780. [CrossRef] [PubMed]
13. Liu, S.; Hosokawa, M.; Miyashita, K.; Liu, S.; Hosokawa, M.; Miyashita, K. Dietary Effect of Squalene on Lipid Metabolism of
Obese/Diabetes KK-Ay Mice and Wild-Type C57BL/6J Mice. Food Nutr. Sci. 2018, 9, 1498–1513. [CrossRef]
14. Mirmiranpour, H.; Ashoori, M.R.; Mikaeili, A.S.; Pezeshki, S.; Serani, A.; Vassar, R.; Martirosyan, D. The Effect of Squalene
on Lipid Profile and Some Oxidative Biomarkers in Patients with Type 2 Diabetes Mellitus. Funct. Food Sci. 2022, 2, 144–156.
[CrossRef]
15. Song, X.; Xiao, H.; Luo, S.; Wang, X.; Wang, W.; Lin, S. Biosynthesis of Squalene-Type Triterpenoids in Saccharomyces Cerevisiae
by Expression of CYP505D13 from Ganoderma Lucidum. Bioresour. Bioprocess. 2019, 6, 19. [CrossRef]
16. Wang, F.; Li, H.; Zhao, H.; Zhang, Y.; Qiu, P.; Li, J.; Wang, S. Antidiabetic Activity and Chemical Composition of Sanbai Melon
Seed Oil. Evid.-Based Complement. Altern. Med. 2018, 2018, 5434156. [CrossRef]
17. Ravi Kumar, S.; Yamauchi, I.; Narayan, B.; Katsuki, A.; Hosokawa, M.; Miyashita, K. Squalene Modulates Fatty Acid Metabolism:
Enhanced EPA/DHA in Obese/Diabetic Mice (KK-Ay) Model. Eur. J. Lipid Sci. Technol. 2016, 118, 1935–1941. [CrossRef]
18. Sutrisna, E.; Trisharyanti, I.; Munawaroh, R. Antioxidant and Antidiabetic Activity of 70% Ethanolic Extract of Syzygium
polyanthum (Wight) Leaf from Indonesia. Int. J. Res. Ayurveda Pharm. 2016, 7, 214–216. [CrossRef]
19. Hikmah, N.; Yuliet, Y.; Khaerati, K. Effect of Administration of Bay Leaf Extract (Syzygium polyanthum Wight.) on Glibenclamide
in Reducing Blood Glucose Levels of Mice (Mus Musculus) Induced by Alloxan. J. Farm. Galen. (Galen. J. Pharm.) 2016, 2, 24–30.
[CrossRef]
20. Wijaya, S.H.; Fadillah, C.; Kusuma, W.A. Prediction of Synergistic Effect between Multiple Compounds Related to Diabetes
Mellitus. IOP Conf. Ser. Earth Environ. Sci. 2019, 299, 012038. [CrossRef]
21. Lynch, K.L. Toxicology: Liquid Chromatography Mass Spectrometry. In Mass Spectrometry for the Clinical Laboratory; Nair, H.,
Clarke, W., Eds.; Academic Press: Cambridge, MA, USA, 2017; pp. 109–130.
22. Aldekhail, N.M.N.P. Pharmacotherapy and Weight Management: Efficacy and Clinical Effectiveness in Patients with Obesity and Type
2 Diabetes; University of Glasgow: Glasgow, Scotland, UK, 2018.
23. Raj, R.; Bhatti, J.S.; Badada, S.K.; Ramteke, P.W. Genetic Basis of Dyslipidemia in Disease Precipitation of Coronary Artery Disease
(CAD) Associated Type 2 Diabetes Mellitus (T2DM). Diabetes. Metab. Res. Rev. 2015, 31, 663–671. [CrossRef]
24. Holmes, A.; Coppey, L.J.; Davidson, E.P.; Yorek, M.A. Rat Models of Diet-Induced Obesity and High Fat/Low Dose Streptozotocin
Type 2 Diabetes: Effect of Reversal of High Fat Diet Compared to Treatment with Enalapril or Menhaden Oil on Glucose
Utilization and Neuropathic Endpoints. J. Diabetes Res. 2015, 2015, 307285. [CrossRef]
25. Pournaghi, P.; Sadrkhanlou, R.-A.; Hasanzadeh, S.; Foroughi, A. An Investigation on Body Weights, Blood Glucose Levels and
Pituitary-Gonadal Axis Hormones in Diabetic and Metformin-Treated Diabetic Female Rats. Vet. Res. Forum 2012, 3, 79–84.
26. Hossein Dashti-Rahmatabadi, M.; Morshedi, A. Chronic Consumption of Kombucha and Black Tea Prevents Weight Loss in
Diabetic Rats. Iran. J. Diabetes Obes. 2010, 2, 23–26.
27. Taher, M.; Tg Zakaria, T.M.F.S.; Susanti, D.; Zakaria, Z.A. Hypoglycaemic Activity of Ethanolic Extract of Garcinia Mangostana
Linn. in Normoglycaemic and Streptozotocin-Induced Diabetic Rats. BMC Complement. Altern. Med. 2016, 16, 135. [CrossRef]
28. Azad, A.K.; Azizi, W.M.; Sulaiman, W. Antidiabetic Effects of P. Macrocarpa Ethanolic Fruit Extract in Streptozotocin-Induced
Diabetic Rats. Futur. J. Pharm. Sci. 2020, 6, 57. [CrossRef]
29. Liu, Y.; Xu, X.; Bi, D.; Wang, X.; ZHang, X.; Dai, H.; Chen, S.; Zhang, W. Influence of Squalene Feeding on Plasma Leptin,
Testosterone & Blood Pressure in Rats—PubMed. Indian J. Med. Res. 2009, 129, 150–153.
30. Gabás-Rivera, C.; Barranquero, C.; Martínez-Beamonte, R.; Navarro, M.A.; Surra, J.C.; Osada, J. Dietary Squalene Increases
High Density Lipoprotein-Cholesterol and Paraoxonase 1 and Decreases Oxidative Stress in Mice. PLoS ONE 2014, 9, e104224.
[CrossRef]
Molecules 2022, 27, 6814 15 of 15

31. Kowluru, R.A.; Mishra, M.; Kowluru, A.; Kumar, B. Hyperlipidemia and the Development of Diabetic Retinopathy: Comparison
between Type 1 and Type 2 Animal Models. Metabolism 2016, 65, 1570. [CrossRef]
32. Vergès, B. Dyslipidemia in Type 1 Diabetes: A Masked Danger. Trends Endocrinol. Metab. 2020, 31, 422–434. [CrossRef]
33. Semova, I.; Levenson, A.E.; Krawczyk, J.; Bullock, K.; Gearing, M.E.; Ling, A.V.; Williams, K.A.; Miao, J.; Adamson, S.S.; Shin,
D.J.; et al. Insulin Prevents Hypercholesterolemia by Suppressing 12-Hydroxylated Bile Acids. Circulation 2022, 145, 969–982.
[CrossRef]
34. Khandelwal, P.; Khanna, S. Diabetic Peripheral Neuropathy: An Insight into the Pathophysiology, Diagnosis, and Therapeutics.
In Wound Healing, Tissue Repair, and Regeneration in Diabetes; Academic Press: Cambridge, MA, USA, 2020; pp. 49–77. [CrossRef]
35. Bharti, S.K.; Kumar, A.; Sharma, N.K.; Prakash, O.; Jaiswal, S.K.; Krishnan, S.; Gupta, A.K.; Kumar, A. Tocopherol from Seeds
of Cucurbita Pepo against Diabetes: Validation by In Vivo Experiments Supported by Computational Docking. J. Formos. Med.
Assoc. 2013, 112, 676–690. [CrossRef]
36. Guasch, L.; Sala, E.; Mulero, M.; Valls, C.; Salvadó, M.J.; Pujadas, G.; Garcia-Vallvé, S. Identification of PPARgamma Partial
Agonists of Natural Origin (II): In Silico Prediction in Natural Extracts with Known Antidiabetic Activity. PLoS ONE 2013,
8, e55889. [CrossRef]
37. Kumar, R.; Ramachandran, U.; Khanna, S.; Bharatam, P.V.; Raichur, S.; Chakrabarti, R. Synthesis, In Vitro and In Silico Evaluation
of l-Tyrosine Containing PPARalpha/Gamma Dual Agonists. Bioorg. Med. Chem. 2007, 15, 1547–1555. [CrossRef]
38. Yusoff, N.A.; Lim, V.; Al-Hindi, B.; Razak, K.N.A.; Widyawati, T.; Anggraini, D.R.; Ahmad, M.; Asmawi, M.Z. Nypa Fruticans
Wurmb. Vinegar’s Aqueous Extract Stimulates Insulin Secretion and Exerts Hepatoprotective Effect on STZ-Induced Diabetic
Rats. Nutrients 2017, 9, 925. [CrossRef]
39. Rao, M.M.V.; Hariprasad, T.P.N. In Silico Analysis of a Potential Antidiabetic Phytochemical Erythrin against Therapeutic Targets
of Diabetes. Silico Pharmacol. 2021, 9, 5. [CrossRef]