New olean-15-ene type gymnemic acids from Gymnema sylvestre (Retz.)R.Br. and their antihyperglycemic activity through α-glucosidase inhibition

Article abstract of DOI:10.1016/j.phytol.2019.05.005

A mixture of gymnemic acids was precipitated from the water extract of leaves of Gymnema sylvestre (Retz.)R.Br. ex Sm. (Asclepiadaceae)by acidification with 2 N H₂SO₄. The chromatographic separation of the mixture afforded five new gymnemic acids (1-5). The compounds were characterized as Δ¹⁵ oleanane glycosides on the basis of extensive spectral data analysis. The compounds (1-5)showed dose dependent inhibition of α-glucosidase, which was found to be comparable to acarbose (IC₅₀ 95 μg/ml). Maximum inhibition was achieved with compound 4 (IC₅₀ 57 μg/ml)followed by 3 (IC₅₀ 62 μg/ml), 1 (IC₅₀ 80 μg/ml), 2 (IC₅₀ 120 μg/ml)and 5 (IC₅₀ 128 μg/ml). The results revealed that the overall pattern of hydroxyl and acyl substitutions of compounds affected their inhibitory activity. In oral sucrose tolerance test, pre-treatment with crude gymnemic acid mixture and isolated compounds 1 and 4 at a dose of 10 mg/Kg b.w. significantly blunted the effect of sucrose challenge in mice. Based on these results, the antihyperglycemic effect of G. sylvestre can be, at least partly, attributed to the inhibition of α-glucosidase by its gymnemic acids. The current study provides relatively more direct evidence explaining the effectiveness of G. sylvestre against hyperglycemia.

Full text of DOI:10.1016/j.phytol.2019.05.005

Phytochemistry Letters 32 (2019) 83–89
Contents lists available at ScienceDirect
Phytochemistry Letters
journal homepage: www.elsevier.com/locate/phytol
New olean-15-ene type gymnemic acids from Gymnema sylvestre (Retz.)
R.Br. and their antihyperglycemic activity through α-glucosidase inhibition
a
b
c
a,d
a,⁎
Naila Hassan Ali Alkefai , Saima Amin , Manju Sharma , Javed Ahamad , Showkat R. Mir
b
c
d
a
Department of Pharmacognosy and Phytochemist, School of Pharmaceutical Education and Research, Jamia Hamdard, New Delhi, India
Department of Pharmaceutics, School of Pharmaceutical Education and Research, Jamia Hamdard, New Delhi, India
Department of Pharmacology, School of Pharmaceutical Education and Research, Jamia Hamdard, New Delhi, India
Department of Pharmacognosy, Faculty of Pharmacy, Tishk International University, Erbil, Iraq
A R T I C L E I N F O
A B S T R A C T
Keywords:
A mixture of gymnemic acids was precipitated from the water extract of leaves of Gymnema sylvestre (Retz.) R.Br.
Gymnema sylvestre
olean-15-ene glycoside
Gymnemic acid
ex Sm. (Asclepiadaceae) by acidification with 2 N H SO . The chromatographic separation of the mixture af-
2
4
15
forded five new gymnemic acids (1-5). The compounds were characterized as Δ oleanane glycosides on the
basis of extensive spectral data analysis. The compounds (1-5) showed dose dependent inhibition of α-gluco-
sidase, which was found to be comparable to acarbose (IC50 95 μg/ml). Maximum inhibition was achieved with
compound 4 (IC50 57 μg/ml) followed by 3 (IC50 62 μg/ml), 1 (IC50 80 μg/ml), 2 (IC50 120 μg/ml) and 5 (IC50
α-glucosidase inhibition
Antihyperglycemia
1
28 μg/ml). The results revealed that the overall pattern of hydroxyl and acyl substitutions of compounds af-
fected their inhibitory activity. In oral sucrose tolerance test, pre-treatment with crude gymnemic acid mixture
and isolated compounds 1 and 4 at a dose of 10 mg/Kg b.w. significantly blunted the effect of sucrose challenge
in mice. Based on these results, the antihyperglycemic effect of G. sylvestre can be, at least partly, attributed to
the inhibition of α-glucosidase by its gymnemic acids. The current study provides relatively more direct evidence
explaining the effectiveness of G. sylvestre against hyperglycemia.
1
. Introduction
Diabetes mellitus is a chronic metabolic disorder characterized by
flatulence, meteorism and diarrhoea that can be partly ascribed to the
non-specific inhibition of both α-glucosidase and α-amylase enzymes
by these agents. Selective α-glucosidase inhibitors are expected to bring
about stricter glycemic control while keeping the pancreatic (insulin)
pathway undisturbed (Sheliya et al., 2015). Recently several triterpenes
have aroused greater interest in tackling disturbances in carbohydrate
and lipid metabolism. They are also reported to inhibit the formation of
advanced glycation end products that are implicated in the pathogen-
esis of diabetic nephropathy, neuropathy or atherosclerosis (Nazaruk
and Borzym-Kluczyk, 2015). The curative potential of triterpenoids is
very high yet still poorly recognised.
the relative or absolute absence of insulin or its insensitivity resulting in
disturbed homeostasis (Alberti and Zimmet, 1998). Post-prandial hy-
perglycemia (PPHG) is the most important risk factor in the onset of
diabetes and development of its comorbidities (Dong et al., 2012). The
elevated blood glucose levels can be the cause of several complications
associated with diabetes, such as retinopathy, nephropathy, neuropathy
and impaired wound healing (Ban and Twigg, 2008). α-Glucosidase is
primarily a hydrolysing enzyme located primarily on the epithelial wall
of the small intestine. Its inhibition delays the metabolism of carbo-
hydrates as well as their absorption, to consequently suppress PPHG
Gymnema sylvestre (Retz.) R.Br. ex Sm. (Asclepiadaceae) is a widely
distributed medicinal herb in India. It is locally known as Gur-mar as it
suppresses the ability to taste sweetness. Sushruta - an ancient compi-
lation of Indian medicinal plants - refers to G. sylvestre as a destroyer of
Madhumeha (glycosuria). It is thus believed that it neutralizes the excess
(
Krentz and Bailey, 2005). Acarbose, miglitol and voglibose are some of
the clinically-used carbohydrate metabolism inhibitors. These agents
are associated with common side effects such as abdominal distension,
Abbreviations: COSY, correlation spectroscopy; DEPT, distortion-less enhancement by polarization transfer; DMSO, dimethyl sulfoxide; FTIR, fourier transform infra
red; HR-ES-MS, high resolution electrospray mass spectrometry; MPLC, medium pressure liquid chromatography; MW, molecular weight; NMR, nuclear magnetic
resonance; HMBC, heteronuclear multiple bond correlation; HMQC, heteronuclear multi-quantum coherence; PNPG, p-nitrophenyl-α-D-glucopyranoside; PPHG,
postprandial hyperglycaemia; TLC, thin layer chromatography; TMS, tetramethyl silane
⁎
Corresponding author at: Phyto-Pharmaceutical Research Laboratory, Department of Pharmacognosy and Phytochemistry, School of Pharmaceutical Education
and Research, Jamia Hamdard, PO Hamdard Nagar, New Delhi, 110 062, India.
E-mail address: showkatrmir@gmail.com (S.R. Mir).
https://doi.org/10.1016/j.phytol.2019.05.005
Received 18 March 2019; Received in revised form 6 May 2019; Accepted 8 May 2019
1874-3900/ © 2019 Phytochemical Society of Europe. Published by Elsevier Ltd. All rights reserved.

N.H.A. Alkefai, et al.
Phytochemistry Letters 32 (2019) 83–89
sugar present in diabetics (Nadkarni, 1986). The plant is popular in
Indian systems of traditional medicine and is listed in the Indian
Pharmaceutical Codex as a remedy for diabetes (Singh et al., 2008). The
leaf extract is reported to reduce hyperglycaemia in humans (Khare
et al., 1983) and in experimental animals (Srivastava et al., 1985;
Shanmugasundaram et al., 1983, 1990). The blood glucose-lowering
effect is postulated to be mediated through the stimulation of insulin
release (Persaud et al., 1999), regeneration of β-cells (Ahmed et al.,
2
010) or inhibition of glucose uptake (Shimizu et al., 1997). The anti-
sweet constituents of the plant are loosely referred to as gymnemic
acids. A number of triterpenoidal derivatives have been isolated from
the plant. The anti-sweet activity of the extracts and some of the sa-
ponins from G. sylvestre have also been reported (Liu et al., 1992; Sahu
et al., 1996; Yoshikawa et al., 1989a, 1989b, 1991; Ye et al., 2000,
2
001). The literature reports concerning the anti-diabetic or antisweet
effects of triterpenoids from G. sylvestre have been compiled in an ex-
tensive review (Fabio et al., 2014). It is interesting that there are only a
few reports on the gut glucosidase inhibition by individual gymnemic
acids (Daisy et al., 2009; Kimura, 2006).
As part of our ongoing research on the preparative isolation of
bioactive natural compounds from G. sylvestre (Alkefai et al., 2018), this
article describes the isolation and characterization of five new gym-
nemic acids and delineation of their antihyperglycemic effect. A mix-
ture of gymnemic acids was precipitated from the water extract of
leaves of G. sylvestre by acidification with 2 N H
2
SO The residue was
4
dissolved in minimal amount of methanol and fractionated over silica
gel (50–60 μm) column using MPLC. Prep-TLC was used for further
purification of eluents from MPLC. This resulted in the isolation of five
1
5
new gymnemic acids (1-5). Their structures were elucidated as Δ
oleanane-type triterpene glycosides by the interpretation of spectral
data. The chemical structures of compounds 1-5 are presented in Fig. 1.
The orientation of glucopyranosyl and glucuronopyranosyl residues
1
13
was determined on the basis of H and C NMR spectral data (Tables 1
and 2, see Supplemental material Fig. S1–S25). Furthermore, D-con-
figuration of the sugar residues was established by TLC comparison of
the acid hydrolysate with the reference standards.
2
. Results and discussion
Compound 1 was obtained as brown sticky mass from ethyl acetate
eluents. Its FTIR spectrum exhibited absorption bands for hydroxyl
−
1
−1
group (3392 cm ), carbonyl group (1722 cm ) and unsaturation
−
1
(
1448 cm ). Its molecular formula was derived to be C53
H
84
O
18 (MW
1
3
1
008) on the basis of C/DEPT NMR and mass spectra. Its HR-ES-MS
(
Fig. S6, see Supplemental material) displayed a fragment ion peak at
+
m/z 844.6627 for [C47
H
72
O
13
]
(calc. 844.4973)arising due to the loss
Fig. 1. Structures of compounds 1-5 isolated from Gymnema sylvestre.
of a glucose residue. A detailed study of 1D and 2D NMR spectra of 1
(
Fig. S1-S5, see Supplemental material) led to the assignment of all
proton and carbon resonances (Tables 1 and 2). A downfield doublet at
Acid hydrolysis of 1 yielded D-glucose and D-glucuronic acid that were
1
δ 5.61 (2H, J =1.2 Hz, H-15/16) in H NMR and a signal at δ 129.2 (C-
identified by TLC comparison of the hydrolysate with the respective
1
3
13
1
5/16) in C NMR spectra indicated a vinylic linkage between two
standards. The C/DEPT NMR spectra of 1 revealed the presence of
equivalent carbons that was placed at C-15. This allocation was cor-
roborated by long range correlation observed between C-15/16 and H-
three oxygenated methine carbons [δ 82.0 (C-3), 72.7 (C-21) and 69.7
(C-22) associated with its aglycone. Comparison of NMR data of 1 with
those of substituted oleananes (Liu et al., 1992; Mahato and Kundu,
1994; Sahu et al., 1996), established its aglycone as 3α, 21β,22α-tri-
hydroxy olean-15-ene. For determining the position of attachment of
2
1/22 in HMBC spectrum. It also revealed the presence of two vicinal
carbinol carbons (C-21 and C-22). A double-doublet at δ 3.21 (1H,
J = 1.6, 4.8 Hz) was assigned to H-3 carbinol proton placed in β-or-
ientation due to its lower coupling constant values.
1
13
acyl/sugar groups, H- C chemical shift correlation spectra were ex-
amined in detail. Key correlational cross peaks were exhibited between
H-1′ and C-3; H-1′′ and C-3′; H-21 and C-1a; and H-22 and C-1b in
1
The H NMR spectrum of 1 exhibited signals for two methyl bu-
tyroyl moieties [δ 2.12 (m, H-2a), 2.17 (m, H-2b), 1.68 (m, H
2
-3a), 1.46
1
3
(
m, H
2
-3b), 0.61 (br s, H
3
-4a, H
3
-4b), 1.29 (d, J =7.2 Hz, H
3
-5a) and
HMBC spectrum. In C NMR spectrum, a downfield shift of +3.3 ppm
was observed for C-3 (δ 82.0 compared to δ 78.7 for oleanolic acid) due
to its glycosylation (Kasai et al., 1979). The signals for C-21/22 carbons
also showed a downfield shifts that were attributed to their acylation
(Ishii et al., 1978). Based on the foregoing account, the structure of 1
was elucidated as 3α,21β,22α-trihydroxy-21,22-bis(2-methyl-1-ox-
1
1
1
.27 (d, J =6.8 Hz, H -5b)]. The signals for two anomeric carbons at δ
3
05.4 and 104.3; and two anomeric protons at δ 4.38 (d, J =7.6 Hz, H-
′) and 4.30 (d, J =7.6 Hz, H-1′′) suggested the presence of two sugar
units. A detailed examination of the saccharide regions of the NMR
spectra revealed a significant downfield shift experienced by the C-3′ (δ
8
9.0 from 76.2) that indicated a 3′→1′′ linkage between the sugar units.
obutoxy)-olean-15-en-23-methyl
carboxylate-3yl-3-O-β-D-
84

N.H.A. Alkefai, et al.
Phytochemistry Letters 32 (2019) 83–89
Table 1
Table 2
1
13
H NMR spectroscopic data (δ in ppm, J in Hz) for compounds 1-5.
C NMR spectroscopic data (δ in ppm) for compounds 1-5.
H
1
2
3
4
5
C
1
2
3
4
5
3
3.21 dd
3.20 dd
(1.6, 4.8)
4.26 dd
3.33 dd (1.6, 3.33 dd
3.75 dd
1
38.1
26.3
82.0
39.7
55.4
17.3
33.8
41.5
46.5
39.7
23.3
31.7
46.6
38.8
129.2
129.2
42.0
41.2
45.4
37.4
72.7
69.7
170.2
13.2
16.1
18.4
27.1
19.4
28.5
19.6
38.2
26.0
82.3
39.6
55.3
18.3
66.5
46.0
51.8
38.2
23.4
31.7
46.6
36.9
129.2
129.2
42.5
41.1
46.5
36.8
72.7
69.1
26.5
13.2
16.5
18.3
27.5
19.3
28.6
19.4
38.0
27.5
79.4
39.4
54.9
20.2
66.3
46.4
52.1
41.0
23.5
33.1
46.7
36.5
129.4
129.4
41.3
40.3
47.3
36.7
73.3
69.0
26.0
13.5
15.4
18.1
28.4
19.2
28.3
19.3
38.8
27.1
82.2
39.7
54.9
18.4
66.3
41.5
51.8
36.0
29.2
31.6
41.2
36.2
129.2
129.2
40.2
46.6
35.6
36.6
72.3
67.8
174.2
13.2
15.7
18.2
26.6
15.4
177.1
19.4
38.4
26.6
81.9
41.5
55.4
16.7
66.0
39.7
51.8
38.8
29.1
31.5
46.6
36.4
129.2
129.2
42.1
31.7
38.8
29.4
73.9
68.4
61.3
13.3
15.9
18.5
27.2
63.2
29.1
19.7
(
1.6, 4.8)
4.8)
(1.5, 3.0)
(2.0, 8.5)
4.49 d (7.5)
2
7
–
4.21dd (7.0) 4.26 dd
(7.0)
3
(
6.8)
4
1
1
2
2
2
5
5.61 d
5.61 br s
5.61 br s
5.73 br s
5.73 br s
3.80 d (4.5)
4.15 d (4.5)
0.90 s
5.74 br s
5.74 br s
5.73 br s
5.73 br s
3.84 d (4.5)
4.16 d (4.5)
3.43 d
5
(
1.2)
6
6
1
2
3
5.61 d
1.2)
3.70 d
4.8)
4.05 d
7
(
8
3.69 d
(4.8)
3.81 d
(5.0)
4.17 d
(4.5)
–
9
(
10
11
12
13
14
15
16
17
18
19
20
21
4.04 d
(4.8)
(
4.8)
–
0.89 s
(
10.5)
2
2
2
2
2
2
3
4
5
6
7
8
9
0
0.96 s
1.26 s
1.27 s
0.91 s
1.29 s
1.01 s
1.38 s
0.91 s
1.26 s
1.27 s
0.93 s
1.29 s
1.01 s
1.38 s
4.48 d
(7.6)
3.79
3.63
3.72
3.53
0.91 s
1.30 s
1.27s
0.91 s
1.26 s
1.28 s
0.94 s
1.28 s
–
0.90 s
1.26 s
1.29 s
0.90 s
1.30 s
1.02 s
1.39 s
0.94 s
3.29 d (9.5)
1.01 s
1.38 s
1.31 s
2
2
2
2
2
2
2
2
3
4
5
6
7
8
Glc 1′
4.38 d
4.29 d (7.0)
4.31 d
4.51 d (6.5)
(
7.6)
(8.5)
2
3
4
5
′
′
′
′
3.98
4.42
3.63
3.41
4.15-3.67
4.15-3.67
4.15-3.67
4.15-3.67
4.30-3.53
4.30-3.53
4.30-3.53
4.30-3.53
4.40-3.90
4.40-3.90
4.40-3.90
4.40-3.90
Glu 1′′
4.30 d
4.38 d
4.19 d (6.5)
4.23 d
4.41 d (7.0)
29
30
(
7.6)
(6.8)
(8.0)
2
3
4
5
6
′′
′′
′′
′′
′′
3.78
3.78
4.16-3.67
4.16-3.67
4.16-3.67
4.16-3.67
4.30-3.53
4.30-3.53
4.30-3.53
4.30-3.53
4.40-3.90
4.40-3.90
4.40-3.90
4.40-3.90
3.43 d (9.5)
3.49
3.44
Glc 1′
105.4
76.1
89.0
71.8
78.8
172.4
104.3
73.7
76.5
71.2
78.2
62.9
Mba
177.1
41.5
24.9
10.8
15.7
104.3
110.0
105.4
105.4
76.1
89.7
73.6
77.7
171.9
104.3
76.0
77.7
72.0
75.0
63.4
Mba
176.1
39.7
26.6
10.9
15.9
3.32
3.24
2
3
4
5
6
′
′
′
′
′
75.1
89.1
72.9
76.9
171.7
75.7
89.7
73.3
76.4
174.9
75.8
89.7
73.8
76.5
172.9
3.59
3.56
3.50 dd
3.44 br d
(6.8)
3.63 d (10.0) 3.43 d
(9.0)
(
4.0, 10.8)
First acyl
Mba
Mba
Mba
–
–
–
–
–
Mba
2
3
4
5
a
a
a
a
2.12 m
1.68 m
0.61 br s
1.29 d
2.54 m
1.54 m
0.60 br s
1.25 d
(7.2)
2.59 m
2.48 m
1.60 m
0.72 s
Glu 1′′
114.3
73.7
76.5
71.8
78.2
62.4
Mba
177.2
41.5
24.7
10.9
15.4
115.0
73.4
75.4
70.3
77.2
63.2
Mba
175.1
41.5
24.5
11.2
15.1
104.3
76.0
77.7
72.3
75.0
63.2
–
–
–
–
–
–
1.54 m
2
3
4
5
6
′′
′′
′′
′′
′′
0.71 br s
1.21 d (6.0)
1.19 d (7.0)
(
7.2)
Other acyl Mba
Mba
Tig
–
–
–
–
–
Tig
2
3
4
5
b
b
b
b
2.17 m
1.46 m
0.61 br s
1.27 d
2.20 m
1.43 m
0.60 br s
1.27 d
(6.8)
–
–
First acyl
6.78 m
1.85 m
2.08 s
7.01 br m
1.61 br d
2.03 s
1
2
3
4
5
a
a
a
a
a
(
6.8)
COOMe
3.68 s
–
–
–
–
Other acyl
Mba
174.2
41.9
23.3
12.1
14.8
51.6
Mba
175.7
41.9
23.4
12.2
14.4
–
Tig
–
–
–
–
–
–
–
Tig
Glc: glucuronic acid, Glu: glucose, Mba: methyl butyroyl, Tig: tigloyl.
Overlapped signals are indicated without designated multiplicity.
1
2
3
b
b
b
168.5
127.9
131.1
13.2
14.4
–
167.4
129.2
138.8
12.1
16.2
–
glucopyranosyl(1→3) O-β-D-glucuronopyranoside.
4b
Compound 2 was obtained as yellowish brown amorphous powder
from ethyl acetate-methanol (98:2 v/v) eluents. Its FTIR spectrum
5b
COOMe
-
1
displayed absorption bands for hydroxyl group (3391 cm ), carbonyl
Glc: glucuronic acid, Glu: glucose, Mba: methyl butyroyl, Tig: tigloyl.
-
1
-1
group (1726 cm ) and unsaturation (1441, 1267 cm ). Its molecular
1
3
formula was derived to be C52
H
84
O
17 (MW 980) on the basis of C/
of two vicinal carbinol carbons (C-21 and C-22). Two double doublets
at δ 3.20 (J = 1.6, 4.8 Hz) and 4.16 (J = 6.8, 6.8 Hz), each integrating
for one proton, were assigned correspondingly to H-3 and H-7 carbinol
protons placed in α- and β-orientation based on their coupling constant
DEPT and mass spectra. Its HR-ES-MS (Fig. S13, see Supplemental
material) displayed a diagnostic fragment ion peak at m/z 878.3750 for
+
[
C
47
H
74
O
15
]
(calc. 878.4028)arising due to the loss of a methyl bu-
1
3
tyroyl residue. Intensive study of 1D and 2D NMR spectra of 2 (Fig. S7-
S12, see Supplemental material) led to assignment of all the protons
and carbon resonances unambiguously (Tables 1 and 2). The presence
values. The C/DEPT NMR spectra of 2 also supported the presence of
four oxygenated methine carbons [δ 82.3 (C-3), 66.5 (C-7), 72.7 (C-21)
and 69.7 (C-22)] in its aglycone. Comparison of NMR data of 2 with
those of substituted oleananes (Mahato and Kundu, 1994; Ye et al.,
2001) established 3α, 7β, 21β, 22α-tetrahydroxy olean-15-en as its
1
of downfield broad singlet at δ 5.61 (2H, H-15/16) in H NMR, and a
1
3
signal at δ 129.2 (C-15/16) in C NMR spectra supported a vinylic
linkage between two equivalent carbons that was placed at C-15. This
allocation was supported by long range correlations observed between
C-15/16 and H-21/22 in HMBC spectrum. It also indicated the presence
1
13
aglycone. A close examination of H and C NMR chemical shift of the
saccharide portion indicated the presence of the same sugar chain at C-
3 position of the aglycone as in 1. Acid hydrolysis of 2 afforded D-
85

N.H.A. Alkefai, et al.
Phytochemistry Letters 32 (2019) 83–89
glucuronic acid and D-glucose that were identified by comparison of
TLC of the hydrolysate with the respective standards. For determination
66.3(C-7), 72.3 (C-21) and 67.8 (C-22)] in its aglycone. The spectra also
revealed the presence of three carboxylic carbons [δ 174.2 (C-23),
177.1 (C-29) and 172.9 (C-6′)]. Comparison of NMR data of 4 with
oleanane derivatives (Mahato and Kundu, 1994; Ye et al., 2001) and the
other compounds reported here clearly indicated the absence of any
acyl substitution. Thus the structure of 4 was deduced to be 3α, 7β,
21β, 22α-tetrahydroxy olean-15-en-23, 29-dioic acid-3-yl-3-O-β-D-glu-
copyranosyl (1→3) O-β-D-glucuronopyranoside.
1
1
1
13
of the linkage of acyl/sugar chain, H- H and H- C chemical shift
correlation were examined in detail. Key cross peaks between H-3 and
H-1′; H-16 and H-22; and H-21 and H-22 in COSY; along with corre-
lations between H-1′ and C-3; H-1′′ and C-3′; H-21 and C-1a; H-22 and
1
3
C-1b in HMBC spectra were observed. In C NMR spectrum, a glyco-
sylation shift of +3.6 ppm was observed for C-3 (δ 82.3 from δ 78.7 for
oleanolic acid). The signals for C-21/22 also exhibited downfield shifts
due to acylation of these carbons (Ishii et al., 1978). These finding led
to determination of 2 as 3α, 7β, 21β, 22α-tetrahydroxy-21,22-bis(2-
methyl-1-oxobutoxy)-olean-15-en-3yl-3-O-β-D-glucopyranosyl (1→3)-
O-β-D-glucuronopyranoside.
Compound 5 was obtained as light green amorphous powder from
ethyl acetate-methanol (90:10 v/v) eluents. Its FTIR spectrum showed
−
1
−1
absorption bands for hydroxyl (3418 cm ), carbonyl (1723 cm ) and
−1
13
olefinic (1645, 1448 cm ) functionalities. Based on C/DEPT and
mass spectra (Fig. S23-S25, see Supplemental material), the molecular
weight of 5 was established to be 1010 corresponding to the molecular
Compound 3 was obtained as brownish amorphous powder from
ethyl acetate-methanol (98:2 v/v) eluents. Its FTIR spectrum exhibited
formula C52
H
82
O
19. Its HR-ES-MS showed a diagnostic fragment ion
-
1
−1
+
absorption bands for hydroxyl (3450 cm ), carbonyl (1723 cm ) and
peak at m/z 846.5386 for [C46
H
70
O
14
]
(calc. 846.4766) arising due to
−1
13
olefinic (1650, 1441 cm ) functionalities. Based on C/DEPT and
mass spectra (Fig. S15-S17, see Supplemental material), the molecular
the loss of a glucose residue from the molecular ion.
1
The H NMR spectrum of 5 (Fig. S22, see Supplemental material)
weight of 3 was determined to be 978, consistent with molecular for-
exhibited signals for one tigloyl moiety [δ 7.01 (brm, H-3b), 1.61 (br d,
1
mula C52
H
82
O
17 (calc. 978.5052, obsd. 978.4949). The H NMR spec-
H
3
-4b) and 2.03 (s, H -5b)] and one methyl butyroyl moiety [δ 2.48 (m,
3
trum of 3 (Fig. S14, see Supplemental material) exhibited signals for a
tigloyl moiety [δ 6.78 (m, H-3b), 1.85 (m, H -4b) and 2.08 (s, H -5b)]
and a methyl butyroyl moiety [δ 2.59 (m, H-2a), 1.54 (m, H -3a),
.71(brs, H -4a) and 1.21 (d, J =6.0 Hz, H -5a)]. The signals for two
H-2a), 1.60 (m, H
2
-3a), 0.72 (s, H
3
-4a) and 1.19 (d, J =7.0 Hz, H -5a)].
3
3
3
2
The presence of two anomeric proton signals at δ 4.51 (d, J =6.5 Hz, H-
1′) and 4.40 (d, J =7.0 Hz, H-1′′), and two anomeric carbon signals at δ
105.4 (C-1′) and 104.3 (C-1′′) supported the presence of two sugar
units. Comparison of TLC of the hydrolysate of 5 with reference stan-
dards confirmed the presence of D-glucuronic acid and D-glucose.
0
3
3
anomeric carbons at δ 110.0 (C-1′) and 115.0 (C-1′′), and two anomeric
protons at δ 4.29 (d, J = 7.0 Hz, H-1′) and 4.19 (d, J =6.5 Hz, H-1′′)
revealed the presence of two sugar units. A detailed examination of
NMR spectra particularly that of the saccharide region supported the
presence of the same sugar chain in 3 as in 2. Acid hydrolysis of 3
afforded D-glucuronic acid and D-glucose that were identified by
comparison of TLC of the hydrolysate with the respective standards.
Significant downfield shift were observed for C-3 of glycone to δ 79.4
and for C-3′ of glucuronic acid to δ 89.7 due to glycosylation. The
signals for C-21 and C-22 also exhibited downfield shifts to δ 73.3 and δ
1
3
Further, C NMR spectrum showed significant downfield shift experi-
enced by C-3 to δ 81.9 and by C-3′ to δ 89.7 supporting the glycosy-
lation at δ C-3 of aglycone and C-3′ of glucuronic acid. The signals for C-
21 and C-22 also displayed downfield shift to δ 73.9 and 68.4, re-
spectively due to acylation at these positions (Ishii et al., 1978).
1
3
The C/DEPT NMR spectra of 5 also displayed signals for three
hydroxy methyl carbons at δ 61.3 (C-23), 63.2 (C-28) and 63.4 (C-6′′);
and four carbinol carbons at δ 81.9 (C-3), 66.4 (C-7), 73.9 (C-21) and
68.4 (C-22). Comparison of spectra data of 5 with that of 1 and other
hydroxylated gymnemic acids (Yoshikawa et al., 1989a, 1989b), sup-
ported 5 to be a hexahydroxy oleanane derivative. A doublet of doub-
lets at δ 3.75 (J = 2.0, 8.5 Hz) was assigned to H-3 carbinol proton that
was placed in α-orientation based on high coupling constant value. On
the basis of the foregoing account, the structure of 5 was elucidated as
3β, 7β, 21β, 22α, 23,28-hexahydroxy-21-(2-methyl-1-oxobutoxy)-22-
[(2-methyl-1-oxobutenyl)oxy]-olean-15-en-3-yl-3-O-β-D-glucopyr-
anosyl(1→3) O-β-D-glucuronopyranoside. It is noteworthy that the
6
9.0, respectively that can be attributed to acylation f these carbons
13
(
Ishii et al., 1978). C NMR spectrum of 3 also supported the presence
of four oxygenated carbons [δ 79.4 (C-3), 66.3 (C-7), 73.3 (C-21) and
6
9.0 (C-22)] in its aglycone. Comparison of NMR data of compound 3
with that of compound 2 revealed that 3 has one tigloyl and one methyl
butyroyl unit compared to two methyl butyroyl units in 2. Thus com-
pound 3 was identified as 3α, 7β, 21β, 22α-tetrahydroxy-21-(2-methyl-
1
-oxobutoxy)-22-[(2-methyl-1-oxobutenyl)oxy] olean-15-en-3yl-3-O-β-
D-glucopyranosyl(1→3) O-β-D-glucuronopyranoside.
1
5
Compound 4 was obtained as white amorphous powder from ethyl
acetate-methanol (90:10 v/v) eluents. Its FTIR spectrum showed ab-
compounds 1-5 represent Δ oleanane-type gymnemic acids (all acy-
12
lated except 4) compared to the arylated Δ oleanane series of gym-
nemic acids reported earlier by our group (Alkefai et al., 2018).
Compounds 1-5 were screened for α-glucosidase inhibitory activity
at concentrations ranging from 31.25–500 μg/ml. All the compounds
showed dose-dependent inhibition of α-glucosidase as shown in Fig. 2.
The maximum inhibition was achieved with compound 4 (IC50 57 μg/
ml) followed by 3 (IC50 62 μg/ml), 1 (IC50 80 μg/ml), 2 (IC50 120 μg/
ml) and 5 (IC50 128 μg/ml). IC50 values for acarbose and crude gym-
nemic acid (CGA) were found to be 95 μg/ml and 170 μg/ml, respec-
tively. It is noteworthy that the compounds 3 and 4 were about three
fold more potent than the CGA (Table S1, see Supplemental material).
Compounds 3 could not be screened further for the lack of sufficient
quantity. Compounds 1 and 4 were screened for their effect on blood
glucose in oral sucrose tolerance test in mice at a dose of 10 mg/kg b.w.
after oral administration.
-
1
-1
sorption bands for hydroxyl (3399 cm ), carbonyl (1724 cm ), olefinic
-
1
(
1643, 1455 cm ) functionalities. Its HR-ES-MS (Fig. S21, see
Supplemental material) displayed
a
molecular ion peak at m/z
8
8
72.5179 that accounted to the molecular formula C42 19 (calc.
64
H O
72.4997). The formula indicated the presence of eleven double bond
equivalents, five of which were attributed to pentacyclic framework,
three to carboxylic groups, two to sugar units and the remaining one to
1
a vinylic linkage. A two-proton broad signal δ 5.74 in H NMR spectrum
1
3
and a single signal in the olefinic region of the C NMR spectrum at δ
1
29.2 indicated the presence of the only vinylic linkage between two
equivalent carbons that was consequently placed at C-15. The presence
of two anomeric proton signals at δ 4.31 (d, J =8.5 Hz, H-1′), and 4.23
(
d, J =8.0 Hz (H-1′′), and two anomeric carbon signals at δ 105.4 (C-1′)
and 104.3 (C-1′′) further supported presence of two sugar units (Fig.
S18 and S19, see Supplemental material). TLC comparison of the hy-
drolysate of 4 revealed the sugars to be D-glucuronic acid and D-glu-
Oral administration of sucrose (4 g/kg, b.w.) produced a significant
increase (p < 0.01) in blood glucose level (BGL) of sucrose-challenged
control animals, resulting in carbohydrate induced hyperglycemia. It is
evident from Fig. 3 that a pre-treatment compounds 1 and 4 (each10
mg/kg, b.w., p.o.) blunted the effect of sucrose challenge as signified by
less than 65 mg/dL increase in peak blood glucose level after 30 min of
sucrose overload compared to about 95 mg/dL increment in sucrose
1
3
cose. Further, C NMR spectrum showed significant downfield shift
experienced by C-3 to δ 82.7 and by C-3′ to δ 89.7, indicating the
1
3
glycosylation at δ C-3 of aglycone and C-3′ of glucuronic acid. C/
DEPT NMR spectra (Fig. S19, S20, see Supplemental material) of 4
indicated the presence of four oxygenated carbons [δ 82.2 (C-3),
86

N.H.A. Alkefai, et al.
Phytochemistry Letters 32 (2019) 83–89
effectiveness of gymnemic acids against hyperglycemia.
. Materials and methods
.1. General experimental procedures
3
3
Melting points were determined using one end open capillary tubes
on Melting Point M-560 apparatus (Perfit, India) and are uncorrected.
UV spectra were measured on UV-160 spectrometer (Shimadzu, Japan).
Infra-red spectra were recorded on IR Affinity1 Spectrophotometer
1
13
(
Shimadzu, Japan) using KBr pellets. H NMR (400 or 500 MHz),
C
NMR (100 or 125 MHz) and 2D NMR spectra were recorded on DRX
NMR instruments (Bruker, Switzerland). Deuterated methanol was used
as solvent and TMS an internal standard. ES-Mass spectral data were
recorded on Synapt mass spectrometer (Waters, UK) equipped with
direct inlet probe system.
Fig. 2. α-Glucosidase inhibitory activity of compounds 1-5 and crude gym-
nemic acid mixture from Gymnema sylvestre vs acarbose.
3.2. Chemicals
Acarbose was obtained as gift sample from Medley Pharmaceuticals
Ltd, India. α-Glucosidase (EC 3.2.1.20), p-nitrophenyl-α-D-glucopyr-
anoside (PNPG) and methanol-d (NMR grade) were purchased from
4
Merck (Mumbai, India). Hexane, ethyl acetate and methanol of analy-
tical grade were procured from S.D. fine chemicals (Mumbai, India).
Silica gel Si60 (50–60 μm) for column chromatography, prep-TLC silica
gel Si60 plates and pre-coated HPTLC silica gel Si60 F254 plates from
Merck (Germany) were used for chromatographic separations.
3.3. Plant material
The leaves of G. sylvestre were procured from the local crude drug
market, Delhi (M/s Universal Biotic, Delhi, India) and identified by Dr.
H. B. Singh, Chief Scientist (RHMD), NISCAIR, New Delhi, India. A
voucher specimen (NISCAIR/RHMAD/Consult/-2011-12/1856/02)
was deposited in RHMD, NISCAIR, New Delhi, India.
Fig. 3. Antihyperglycemic effect of crude gymnemic acid mixture and com-
pounds 1 and 4 from Gymnema sylvestre vs acarbose in sucrose challenged mice.
challenged animals at the same time level. Acarbose restricted the in-
crement in peak BGL to less than 50 mg/dL. Pre-treatment with both the
isolated compounds exhibited significant reduction in peak BGL com-
pared to the sucrose-challenged group (p < 0.01). Pre-treatment with
CGA and acarbose (10 mg/kg, b.w.) significantly (p < 0.01) lowered
peak BGL compared to sucrose-challenged animals, respectively (Table
S2, see Supplemental material). From the above discussion it is clear
that the antihyperglycemic effect of 1 and 4 was comparable to that of
acarbose.
3
.4. Preparation of crude gymnemic acids mixture
Leaves of G. sylvestre were cleaned, washed under water and dried in
an oven at 45 °C. Dried leaves were pulverized to a coarse powder using
a grinder. About 5 kg of powder was defatted with hexane and then the
defatted leaves were air dried and extracted with water (10 l) under air-
reflux for 6 h with occasional stirring. This was followed by filtration.
The filtrate was adjusted to pH 3.0 by adding 2 N H
2
SO and cen-
4
trifuged at 5000 rpm for 10 min to yield a crude gymnemic acids mix-
The results revealed that the overall pattern of hydroxyl and acyl
substitution in compounds affected their α-glucosidase inhibitory ac-
tivity. As is clear from Fig. 1, the structure of compound 4 has three OH
groups that can form hydrogen bonds at the catalytic site of α-gluco-
sidase enzyme. Even the two COOH groups in 4 are amenable to hy-
drogen bonding. These groups in 4 provide points of stronger interac-
tion with α-glucosidase compared to other compounds that either have
one or no OH group. On the other hand, a comparable inhibitionof α-
glucosidase by acylated ymnemic acids 1 and 3 point towards the role
of hydrophobic interaction as well. Thus, it would be safe to infer that
inhibition involved both the hydrophobic and hydrophilic interactions
of the compounds at the catalytic site. The results of the inhibitory
studies corroborated the earlier studies involving triterpenoids and
their synthetic analogues (Ali et al., 2002; Gowri et al., 2007; Hou et al.,
ture (60 g, 1.2% w/w) (Kurihara, 1969).
3.5. Isolation of compounds
The mixture of gymnemic acids was subjected to normal-phase
MPLC. The isolation was carried out on a Sepacore Easy Purification
System (Buchi Labortechnik, Flawil, Switzerland) consisting of two C-
6
05 pump modules, a C-615 control unit and a C-640 UV detector.
Fractionation was carried out with a 70 × 460 mm plastic-glass column
Buchi, Switzerland) packed with silica gel Si60 (50–60 μm). Sample
(
solution was prepared in methanol. The solution was homogenized,
filtered and loaded on to the column through the injector loop.
The eluents from the column were detected by an on-line UV de-
tector set at 225, 245, 290 and 325 nm for simultaneous detection.
Eluents were manually collected based on the changes in absorbance
and aliquots were analysed by TLC on silica gel 60 F254 pre-coated
plates. Chromatographically identical eluents were combined and
concentrated to yield five compounds. Compound 1 was obtained as a
brownish sticky mass from ethyl acetate fractions. Compounds 2 and 3
were obtained as amorphous powders from 2% methanol in ethyl
acetate fractions. Elution with 10% methanol in ethyl acetate afforded a
2
009; Kuang et al., 2011; Uddin et al., 2012). A comparison of the
findings here and that of our earlier report highlights the fact that the
non-acylated gymnemic acid (4) is a more effective α-glucosidase in-
hibitor than the acylated (1-3 and 5) and the arylated gymnemic acids
(
Alkefai et al., 2018). G. sylvestre has been used as a traditional remedy
for diabetes in many countries. The current study lends credence to this
ethno-medical practice. Moreover, our study highlights the
87

N.H.A. Alkefai, et al.
Phytochemistry Letters 32 (2019) 83–89
mixture of compounds 4 and 5 that were purified using prep-TLC
to a control which had 60 μl of buffer solution in place of the extract.
For blank incubation (to allow for absorbance produced by the ex-
tracts), the enzyme solution was replaced with a buffer solution and
absorbance recorded. The α-glucosidase inhibitory activity was ex-
pressed as a percentage inhibition and was calculated as follows:
(
BuOH-EtOAc-H O, 20:0.5:0.5) as white and light green powders, re-
2
spectively. The purity of compounds was ascertained by HPTLC per-
formed on pre-coated silica gel Si60 F254 plates.
3.6. Characterization of compounds
A control (A test A background)
%
Inhibition =
× 100
A control
3
.6.1. Compound 1
Brown sticky mass (0.25 g, 0.05% yield); mp 48-51 °C; UV (MeOH)
where Acontrol, Atest, Abackground are defined as the absorbance of 100%
enzyme activity, test sample with the enzyme and test sample without
the enzyme, respectively.
λ
max 213, 282 nm; IR (KBr) νmax 3392, 2941, 1722, 1640, 1512, 1448,
-1 1 13
1
383, 1031 cm ; H and C NMR data, see Tables 1 and 2; +ve ES-MS
+
m/z (rel. int.): 1008 [M]
C
53
H
84
O
18 (N.O.), 844.6627 (calc. for
The concentration of inhibitors required for inhibiting 50% of en-
zyme activity under assay conditions was presented as IC50 value.
C
47
H O13, 844.4973) (8), 353 (60), 182 (65)
72
3
.6.2. Compound 2
3.8. Antihyperglycemic activity
Yellowish brown amorphous powder (0.2 g and 0.04% yield); mp
76–178 °C; UV (MeOH) λmax 210, 282 nm; IR (KBr) νmax: 3391, 2929,
1
1
3.8.1. Animals
−
1
1
13
726, 1644, 1514, 1441, 1267, 1031 cm
;
H and C NMR data, see
Wistar albino mice (30–50 g) were obtained from Central Animal
Facility, Jamia Hamdard, New Delhi and maintained under controlled
condition of illumination (12 h light/12 h darkness) and temperature
Tables 1 and 2; -ve ES-MS m/z (rel. int.): 980 [M]- C52
H O17 (N.O.),
84
8
78.3750 (calc. for C47
H O15, 878.4028) (20), 399 (100)
74
2
0–25 °C. They were housed under ideal laboratory conditions, main-
3.6.3. Compound 3
tained on standard pellet diet (Lipton rat feed Ltd., Pune, India) and
water ad-libitum throughout the experimental period. Animals were
acclimatized to the conditions before the start of experiments. The ex-
perimental study was approved by the Institutional Animal Ethics
Committee (IAEC) of Jamia Hamdard, New Delhi, India (JH/CAHF/
173/CPCSEA/28th January/2000, approval no. 2017/986). All the test
compounds and the standard drugs were administered orally.
Brownish amorphous powder (0.26 g, 0.05% yield); mp 144–147 °C;
UV (MeOH) λmax 209, 278 nm; IR (KBr) νmax: 3450, 2861, 1723, 1549,
−
1
1
13
1
2
9
441, 1383, 1267, 1031 cm ; H and C NMR data, see Tables 1 and
+
; +ve ES-MS m/z (rel. int.): 978.4949 [M] (calc. for C52
H
O
82 17
78.5052) (3), 845 (16), 831 (25), 685 (45)
3.6.4. Compound 4
White amorphous powder (0.3 g, 0.06% yield); mp 111–113 °C; UV
3.8.2. Oral sucrose tolerance test
(
MeOH) λmax 212, 283, 318 nm; IR (KBr) νmax 3399, 2946, 2850, 1724,
The animals were randomly divided into six groups, each consisting
of six mice (n = 6). The animals were fasted overnight before the start
of the experiment. Group I served as normal control which received
1 ml/kg b.w. vehicle (0.5% CMC in distilled water). Group II served as
sucrose challenged control that received sucrose (4 g/kg b.w.). Group
III received acarbose as a standard drug (10 mg/kg b.w.). Groups IV, V
and VI were administered crude gymnemic acid mixture, compound 1
and 4 (10 mg/kg, p.o., each), respectively. This was followed by sucrose
challenge (4 g/kg, b.w.) 20 min after the treatment in Groups III-VI.
Blood was withdrawn from the tail vein at 0, 30, 60, 90 and 120 min
after the carbohydrate challenge and blood glucose level determined
using a one-touch glucometer (myLifePura, Switzerland).
−
1
1
13
1
1
8
643, 1455, 1383, 1269, 1032 cm ; H and C NMR data, see Tables
+
and 2; +ve ES-MS m/z (rel. int.): 872.5179 [M] (calc. for C42
H
O
64 19
72.4997) (5), 869 (20), 831 (50), 789 (100)
3.6.5. Compound 5
Light green amorphous powder (0.5 g, 0.1% yield); mp 131–135 °C;
UV (MeOH) λmax 216, 286, 319 nm; IR (KBr) νmax 3418, 2947, 1723,
−
1
1
13
1
2
645, 1448, 1270, 1043 cm ; H and C NMR data, see Tables 1 and
+
; +ve ES-MS m/z (rel. int.): 1010 [M]
C
52
82
H O19 (N.O.), 846.5386
(
calc. for C46
H O14, 846.4766) (5), 685 (25), 663 (100)
70
3.6.6. Acid hydrolysis of compounds
The isolated compounds (each 15–20 mg) were refluxed with 15%
3.8.3. Statistical analysis
HCl in 60% ethanol (5 ml) for 6 h. Each reaction mixture was diluted
with water and neutralized with NaOH, and extracted with ethyl
acetate. The organic was concentrated under reduced pressure to afford
aglycone. The neutral hydrolysate revealed the presence of glucuronic
All data were presented as mean ± standard error of mean (SEM).
Statistical analysis was performed by one-way analysis of variance
®
(ANOVA) followed by Dunnett’s test (GraphPad InStats version 3.06,
USA). P-values less than 0.05 were considered to be significant.
Conflict of interest
acid and/or glucose by HPTLC (CHCl
3
-CH
3
OH-H O, 6.5:3.5:1) when
2
compared with authentic samples [Rf 0.42 (D-glucose) and 0.36 (D-
glucuronic acid)].
The authors report no conflict of interest.
Acknowledgments
3.7. In-vitro α-glucosidase inhibition
The enzyme inhibition assays were carried out with compounds 1–5
and crude gymnemic acids mixture and were compared with acarbose
at doses ranging from 31.25–500 μg/ml as per the method given by
Dong et al. (2012), with slight modifications. Briefly, a volume of 60 μl
of sample solutions in DMSO of test samples or acarbose and 50 μl of
The authors are grateful for the financial support from All India
Council for Technical Education, New Delhi under Research Promotion
Scheme (8-71/RIFD/RPS/POLICY-1/2016-17) to the corresponding
author and from University Grants Commission, Delhi under Special
Assistance Program grant (DRS II vide F. 3-32/2009) to Department of
Pharmacognosy and Phytochemistry, School of Pharmacetical
Education and Research, Jamia Hamdard, New Delhi. The suggestions
of Professor Mohammad Ali are sincerely acknowledged. We also wish
to thank Advanced Instrumentation Research Facility, Jawaharlal
Nehru University, New Delhi, India for recording NMR and mass
spectra.
0
.1 M phosphate buffer (pH 6.8) containing α-glucosidase solution (0.2
U/ml) was incubated in 96 well plate at 37 °C for 20 min. After pre-
incubation, 50 μl of 5 mM p-nitrophenyl-α-D-glucopyranoside (PNPG)
solution in 0.1 M phosphate buffer (pH 6.8) was added to each well and
incubated again at 37 °C for another 20 min. Then the reaction was
stopped by adding 160 μl of 0.2 M Na
2
CO into each well, and the ab-
3
sorbance (A) recorded at 405 nm by a micro-plate reader and compared
88

N.H.A. Alkefai, et al.
Phytochemistry Letters 32 (2019) 83–89
Appendix A. Supplementary data
mellitus. Drugs 65, 385–411.
Kuang, H.X., Li, H.W., Wang, Q.H., Yang, B.Y., Wang, Z.B., Xia, Y.G., 2011. Triterpenoids
from the roots of Sanguisorba tenuifolia var. alba. Molecules 16, 4642–4651.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.phytol.2019.05.005.
Kurihara, Y., 1969. Antisweest activity of gymnemic acids A
537–543.
1
its derivatives. Life Sci. 8,
Liu, H.M., Kiuchi, F., Tsuda, Y., 1992. Isolation and structure elucidation of gymnemic
acids, antisweet principles of Gymnema sylvestre. Chem. Pharm. Bull. 40, 1366–1375.
References
13
Mahato, S.B., Kundu, A.P., 1994. C NMR spectra of pentacyclic triterpenoids - a com-
pilation and salient features. Phytochemistry 37, 1517–1575.
Ahmed, A.B.A., Rao, A.S., Rao, M.V., 2010. In vitro callus and in vivo extract of Gymnema
sylvestre stimulate β-cells regeneration and anti-diabetic activity in Wistar rats.
Phytomedicine 17, 1033–1039.
Nadkarni, K.M., 1986. Gymnema Sylvestre: Indian Materia Medica with Ayurvedic, Unani
and Home Remedies, third ed. Popular Prakashan, Bombay, India, I, pp. 596–599.
Nazaruk, J., Borzym-Kluczyk, M., 2015. The role of triterpenes in the management of
diabetes mellitus and its complications. Phytochem. Rev. 14, 675–690.
Alberti, K.G.M.M., Zimmet, P.Z., 1998. Definition, diagnosis and classification of diabetes
mellitus and its complications. Part 1: diagnosis and classification of diabetes mel-
litus. Provisional report of a WHO Consultation. Diabet. Med. 15, 539–553.
Ali, M.S., Jahangir, M., Hussan, S.S., Choudhary, M.I., 2002. Inhibition of α-glucosidase
by oleanolic acid and its synthetic derivatives. Phytochemistry 60, 295–299.
Alkefai, N.H., Ahamad, J., Amin, S., Sharma, M., Mir, S.R., 2018. Arylayed gymnemic
acids from Gymnema sylvestre R.Br. as potential α-glucosidase inhibitors. Phtyochem.
Lett. 25, 196–202.
Persaud, S.J., Al-Majed, H., Raman, A., Jones, P.M., 1999. Gymnema sylvestre stimulates
insulin release in vitro by increased membrane permeability. J. Endocrinol. 163,
2
07–212.
Sahu, N.P., Mahato, S.B., Sarkar, S.K., Poddar, G., 1996. Triterpenoid saponins from
Gymnema sylvestre. Phytochemistry 41, 1181–1185.
Shanmugasundaram, K.R., Panneerselvam, C., Samudram, P., Shanmugasundaram,
E.R.B., 1983. Enzyme changes and glucose utilization in diabetic rabbits: the effects
of Gymnema sylvestre, R. Br. J. Ethnopharmacol. 7, 205–234.
Ban, C.R., Twigg, S.M., 2008. Fibrosis in diabetes complications: pathogenic mechanisms
and circulating and urinary markers. Vasc. Health Risk Manag. 4, 575–596.
Daisy, P., Eliza, J., Mohamed Farook, K.A.M., 2009. A novel dihydroxy gymnemic tria-
cetate isolated from Gymnema sylvestre possessing normoglycemic and hypolipidemic
activity on STZ-induced diabetic rats. J. Ethnopharmacol. 126, 339–344.
Dong, H.-Q., Li, M., Zhu, F., Liu, F.-L., Huang, J.-B., 2012. Inhibitory potential of trilo-
batin from Lithocarpus polystachyus Rehd against α-glucosidase and α-amylase linked
to type 2 diabetes. Food Chem. 130, 261–266.
Shanmugasundaram, E.R.B., Gopinath, K.L., Shanmugasundaram, K.R., Rajendran, V.M.,
1
990. Possible regeneration of islets of Langerhans in streptozotocin-diabetic rats
given Gymnema sylvestre leaf extracts. J. Ethnopharmacol. 30, 265–279.
Sheliya, M.A., Rayhana, B., Ali, A., Pillai, K.K., Aeri, V., Sharma, M., Mir, S.R., 2015.
Inhibition of α-glucosidase by new prenylated flavonoids from Euphorbia hirta L.
herb. J. Ethnopharmacol. 176, 1–8.
Shimizu, K., Iino, A., Nakajima, J., Tanaka, K., Nakajyo, S., Urakawa, N., Atsuchi, M.,
Wada, T., Yamashita, C., 1997. Suppression of glucose absorption by some fractions
extracted from Gymnema sylvestre leaves. J. Vet. Med. Sci. 59, 245–251.
Singh, V.K., Umar, S.A., Ansari, S.A., Iqbal, I., 2008. Gymnema sylvestre for diabetics. J.
Herbs Spices Med. Plants 14, 88–106.
Fabio, G.D., Romanucci, V., Marco, A.D., Zarrelli, A., 2014. Triterpenoids from Gymnema
sylvestre and their pharmacological activities. Molecules 19, 10956–10981.
Gowri, P.M., Tiwari, A.K., Ali, A.Z., Rao, J.M., 2007. Inhibition of α-glucosidase and
amylase by bartogenic acid isolated from Barringtonia racemosa Roxb. seeds.
Phytother. Res. 21, 796–799.
Srivastava, Y., Nigam, S.K., Bhatt, H.V., Verma, Y., Prem, A.S., 1985. Hypoglycemic and
life-prolonging properties of Gymnema sylvestre in diabetic rats. Israel J. Med. Sci. 21,
Hou, W., Li, Y., Zhang, Q., Wei, X., Peng, A., Chen, L., Wei, Y., 2009. Triterpene acids
isolated from Lagerstroemia speciosa leaves as α-glucosidase inhibitors. Phytother.
Res. 23, 614–618.
5
40–542.
Uddin, G., Rauf, A., Al-Othman, A.M., Collina, S., Arfan, M., Ali, G., Khan, I., 2012.
Ishii, H., Tori, K., Tozyo, T., Yoshimura, Y., 1978. Structures of polygalacin-D and -D
2
,
Pistagremic acid, a glucosidase inhibitor from Pistacia integerrima. Fitoterapia 83,
1
648–1652.
and their monoacetates, saponins isolated from Platycodon grandiflorum A. DC., de-
termined by carbon-13 nuclear magnetic resonance spectroscopy. Chem. Pharm. Bull.
Ye, W.C., Zhang, Q.W., Lin, X.M., Che, C.T., Zhao, S.X., 2000. Oleanane saponins from
Gymnema sylvestre. Phytochemistry 35, 893–899.
2
6, 674–677.
1
3
Ye, W.C., Liu, X., Zhang, Q.W., Che, C.T., Zhao, S.X., 2001. Antisweet saponins from
Gymnema sylvestre. J. Nat. Prod. 64, 232–235.
Kasai, R., Okihara, M., Asakawa, J., Mizutani, K., Tanaka, O., 1979. C-NMR study of α-
and β-anomeric pairs of D-mannopyranosides and L-rhamnopyranosides.
Tetrahedron 35, 1427–1432.
Yoshikawa, K., Amimoto, K., Arihara, S., Matsuura, K., 1989a. Gymnemic acid V, VI and
VII from Gur-Ma, the leaves of Gymnema sylvestre R. Br. Chem. Pharm. Bull. 36,
Khare, A.K., Tandon, R.N., Tewari, J.P., 1983. Hypoglycemic activity of an indigenous
drug (Gymnema sylvestre, Gurmar) in normal and diabetic persons. Indian J. Physiol.
Pharmacol. 27, 257–258.
8
52–854.
Yoshikawa, K., Amimoto, K., Arihara, S., Matsuura, K., 1989b. Structure studies of new
antisweet constituents from Gymnema sylvestre. Tetrahedron Lett. 30, 1103–1106.
Yoshikawa, K., Arihara, S., Matsuura, K., 1991. A new type of antisweet principles oc-
curring in Gymnema sylvestre. Tetrahedron Lett. 32, 789–792.
Kimura, I., 2006. Medical benefits of using natural compounds and their derivatives
having multiple pharmacological actions. Yakugaku Zasshi 126, 133–143.
Krentz, A.J., Bailey, C.J., 2005. Oral antidiabetic agents: current role in type 2 diabetes
89