N U TR IT ION RE S EA RCH 3 2 ( 2 0 12 ) 62 6 –6 36

Available online at www.sciencedirect.com

w w w. n r j o u r n a l . c o m

Aqueous extract of tamarind seeds selectively increases

glucose transporter-2, glucose transporter-4, and islets'
intracellular calcium levels and stimulates β-cell proliferation
resulting in improved glucose homeostasis in rats with
streptozotocin-induced diabetes mellitus☆

Sushant Shivdas Sole⁎, B.P. Srinivasan

Department of Pharmacology, Delhi Institute of Pharmaceutical Sciences and Research, University of Delhi, PushpVihar,
New Delhi 110017, India

ARTI CLE I NFO A BS TRACT

Article history: Tamarindus indica Linn. has been in use for a long time in Asian food and traditional medicine for
Received 15 January 2012 different diseases including diabetes and obesity. However, the molecular mechanisms of these
Revised 23 June 2012 effects have not been fully understood. In view of the multidimensional activity of tamarind
Accepted 26 June 2012 seeds due to their having high levels of polyphenols and flavonoids, we hypothesized that the
insulin mimetic effect of aqueous tamarind seed extract (TSE) might increase glucose uptake
Keywords: through improvement in the expression of genes of the glucose transporter (GLUT) family and
GLUT-2 sterol regulatory element-binding proteins (SREBP) 1c messenger RNA (mRNA) in the liver. Daily
GLUT-4 oral administration of TSE to streptozotocin (STZ)–induced (90 mg/kg intraperitoneally) type 2
Sterol regulatory element-binding diabetic male Wistar rats at different doses (120 and 240 mg/kg body weight) for 4 weeks showed
proteins-1c positive correlation with intracellular calcium and insulin release in isolated islets of
Islet's cytosolic calcium Langerhans. Tamarind seed extract supplementation significantly improved the GLUT-2
Glucagon protein and SREBP-1c mRNA expression in the liver and GLUT-4 protein and mRNA
Somatostatin expression in the skeletal muscles of diabetic rats. The elevated levels of serum nitric oxide
Tamarindus indica seeds (NO), glycosylated hemoglobin level (hemoglobin A1c) and tumor necrosis factor α (TNF-α)
STZ-induced model of type 2 decreased after TSE administration. Immunohistochemical findings revealed that TSE
diabetes mellitus abrogated STZ-induced apoptosis and increased β-cell neogenesis, indicating its effect on
islets and β-cell mass. In conclusion, it was found that the antidiabetic effect of TSE on STZ-
induced diabetes resulted from complex mechanisms of β-cell neogenesis, calcium handling,
GLUT-2, GLUT-4, and SREBP-1c. These findings show the scope for formulating a new herbal
drug for diabetes therapy.
© 2012 Elsevier Inc. All rights reserved.

Abbreviations: ANOVA, analysis of variance; BW, body weight; [Ca2 +]i, intracellular calcium; DM, diabetes mellitus; dUTP, deoxyuridine
5-triphosphate; ELISA, enzyme-linked immunosorbent assay; GLUT, glucose transporter; HbA1c, hemoglobin A1c; mRNA, messenger RNA;
NO, nitric oxide; PBS, phosphate-buffered saline; PCR, polymerase chain reaction; SREBP, sterol regulatory element-binding proteins; STZ,
streptozotocin; TNF-α, tumor necrosis factor α; TSE, tamarind seed extract; TUNEL, terminal deoxynucleotidyl transferase–mediated
deoxyuridine 5-triphosphate nick-end labeling.
☆
Conflict of interest: The authors declare that there is no conflict of interest.
⁎ Corresponding author. Tel.: +91 11 29554327; fax: + 91 11 29554503.
E-mail address: sush.sole@gmail.com (S.S. Sole).

0271-5317/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.nutres.2012.06.015
 N U TR IT ION RE S E ARCH 3 2 ( 2 0 12 ) 62 6 –6 3 6
1. Introduction
Diabetes mellitus (DM) is a disease that results in chronic
inflammation and apoptosis in pancreatic islets in patients
with either type 1 or 2 DM and is characterized by abnormal
insulin secretion or insulin receptor or postreceptor events
affecting the metabolism in addition to damaging the liver,
kidney, and β cells of the pancreas. Chronic damage is
associated with elevated oxidative or inflammatory activities
with a continuum of tissue insults leading to more severe
diabetic complications [1]. Insulin-resistant glucose use in
peripheral tissues such as muscle and adipose tissues is a
universal feature of both insulin-dependent DM and non–
insulin-dependent DM. In this process, glucose transporter
(GLUT) and sterol regulatory element-binding proteins (SREBP)
1c along with other components play crucial roles [2]. Glucose
transporter 4, a member of the GLUT family, is mainly
expressed in skeletal muscle, heart, and adipose tissues. It
plays a critical role in insulin-stimulated glucose transport in
these tissues, with glucose uptake occurring when insulin
stimulates the translocation of GLUT-4 from the intracellular
pool to the plasma membrane [3]. Glucose transporter 2, being
the primary GLUT isoform in the liver, plays a pivotal in
glucose homeostasis by mediating bidirectional transport of
glucose [4]. Furthermore, it has been reported that SREBP-1c
expression and its nuclear abundance is low in livers of STZ-
induced diabetic rats and increases markedly with insulin
treatment. This fact presents a direct relation between insulin
and fatty acid metabolism [5]. A rise in cytoplasmic-free
calcium concentration (intracellular calcium [Ca2 +]i) owing to
influx through voltage-gated L-type Ca2 + channels in the
plasma membrane is a central component of the stimulus-
secretion coupling mechanism leading to insulin release by
pancreatic β cells [6]. This pulsatility of calcium is lost in
patients with type 2 DM and thereby affects insulin secretion.
Oxidative stress is one of the key mechanisms in the
pathogenesis of diabetes-related vascular dysfunction. Strep-
tozotocin (STZ)–induced β-cell damage involves many com-
plicated mechanisms, one of which is production of reactive
oxygen species, particularly nitric oxide (NO), either from STZ
or neighboring macrophages. Peroxynitrite and proinflama-
tory cytokines such as TNF-α have been found to cause DNA
damage and apoptosis in human and rat islets [7]. However,
the prevention from selective destruction of insulin-produc-
ing β cells of pancreatic islets can be one of the targets for
either the cure or delay of DM.
Besides drugs in modern medicine, several species of
plants have been described in the scientific and popular
literature as having hypoglycemic activity. Because of their
perceived effectiveness, minimal side effects in clinical
experience and relatively low-cost herbal drugs are pre-
scribed widely even when their biologically active com-
pounds are unknown. Tamarindus indica Linn. (commonly
known as tamarind), a tree-type plant belonging to the
Leguminosae (Caesalpiniaceae) family, occurs in the tropical
regions of the world and can be found in more than 50
countries. It is indigenous to tropical Africa but has become
naturalized in North and South America from Florida to
Brazil and is also cultivated in subtropical China, India,
627
Pakistan, Indochina, Philippines, Java, and Spain. T indica
pulp fruit is used for seasoning, as a food component, and in
juices. Its fruit is regarded as a digestive, carminative,
laxative, expectorant, and blood tonic [8]. Other parts of the
plant present antioxidant [9], antihepatotoxic [10], and anti-
inflammatory [11] activity. T indica was used as a traditional
medicine for the management of DM [12]. Furthermore,
aqueous extract of tamarind seeds was found to have potent
antidiabetic and antihyperlipidemic activities in STZ-induced
diabetic rats [13]. Tamarind seeds are regularly used in the
preparation of curry and spices for human consumption in
Southeast Asian countries including India. Therefore, the
strong rationale of tamarind seed extract (TSE) for antihy-
perglycemic potential obtained from the present study may
serve the use of tamarind as an antidiabetic alone or in
combination with other hypoglycemic agents for human
subjects of diabetes.
The antihyperglycemic potential of tamarind seeds with
respect to GLUT proteins and islets' physiology such as
cytosolic calcium oscillation and endocrine function (insulin,
glucagon, and somatostatin) has not been explored as yet. In
view of the multidimensional activity of plant drugs, we
hypothesized that tamarind seeds would have a complex
mechanism of action; therefore, it was imperative to
consider different approaches involved in antidiabetic
mechanisms. We specifically assumed that insulin secreta-
gogue action might be due to improvement in calcium of
islets in addition to its effect on transport proteins and fatty
acid metabolism.
Therefore, the present study was designed to investigate
the insulin mimetic impact of TSE on the molecular
mechanisms of glucose uptake on STZ-induced diabetic
rats. With these views, we have studied the effect of TSE on
GLUT-2 protein and SREBP-1c messenger RNA (mRNA)
expression in the liver and GLUT-4 protein and mRNA
expression in skeletal muscle in STZ model of type 2 DM. In
addition, we evaluated the effect of TSE treatment on β-cell
proliferation, apoptosis, and neogenesis in pancreatic islets
of STZ rats.
2. Methods and materials
2.1. Drugs and chemicals
The biochemical kits used in the experiment were obtained as
follows: serum NO (Biovision, Milpitas, CA, USA; Cat no. K262-
200), apo-BrdU-IHCTM, in situ DNA Fragmentation Assay Kit
(catalogue no. K403-50; Biovision), TNF-α (catalog no. ELR-
TNF-α-001; Raybiotech, Inc, Norcross, GA, USA), STZ (Sigma
Chemicals), metformin (Ranbaxy Ltd, Gurgaon, India), Taq
DNA polymerase (catalog-A5209, 0200; Bioline Ltd, London,
UK), HbA1c assay kit (catalog #COD 11044; Biosystem, Barce-
lona, Spain), glimepiride (manufacturer: Hetero Labs Ltd,
Hyderabad, India; provided by Panacea Biotech Ltd, Malpur,
Baddi, Solan, India), benzylpenicillin and streptomycin (Sigma
Aldrich, Munich, Germany), horseradish peroxidase–conju-
gated immunoglobin G (Jackson Immuno Research Laborato-
ries, West Grove, PA, USA), and anti-GLUT-2 and GLUT-4
antibodies (Abcam, Cambridge, UK).
628 N U TR IT ION RE S EA RCH 3 2 ( 2 0 12 ) 62 6 –6 36
2.2. Preparation of aqueous extract of seeds of T indica
Seeds of T indica were purchased from Khari Baoli, New Delhi,
in May 2011, and authentication was done by a scientist of the
National Bureau of Plant Genetic Resources, Pusa, New Delhi
(voucher no. NHCP/NBPGR/2010-52). Aqueous extract of seeds
of T indica was prepared by the method of the National
Institute of Health and Family Welfare, India [14]. Briefly, after
incubation for 2 days at 40°C, the seeds of T indica were
powdered in a grinder; 100 g powder was suspended in 500 mL
redistilled water, and the extraction was performed in Soxhlet
apparatus for 18 hours. A deep-brown aqueous extract was
obtained, which was filtered using a coarse sieve filter paper.
The filtrate was then dried under reduced pressure and finally
lyophilized. Phytochemical screening and standardization of
the extract were done before commencement of the in vivo
study. The aqueous extract yielded 2.8-g lyophilized powder
(0.28%) from 1 kg of tamarind seeds.
2.3. Animals
Male Wistar rats weighing 150 to 200 g used for the study
were obtained from the Animal House of the Delhi Institute
of Pharmaceutical Sciences and Research. Rats were housed
in colony cages (4 rats per cage) at an ambient temperature
of 25°C with 12-hour light to 12-hour dark cycle. Rats had free
access to standard food and water ad libitum. The Principles
of Laboratory Animal Care (NIH, 1985) were followed
throughout the duration of the experiment. All experimental
procedures were conducted according to the institutional
animal ethical committee (protocol no. 2/DIPSAR/IAEC/2010)
and Committee for the Purpose of Control and Supervision
on Experiments on Animals (CPCSEA) guidelines.
2.4. Experimental design/animal group
2.4.1. Induction of type 2 DM
Streptozotocin was injected at a dose level of 90 mg/kg
intraperitoneally in 0.1 mol/L freshly prepared citrate buffer
pH 4.5 to 2-day-old neonatal rats. Controls were injected with
an equivalent volume of citrate buffer. After 6 weeks of
injection, animals were evaluated for fasting blood glucose
level. A fasting glucose level of 140 mg/dL was the criterion for
selection of diabetic rats [15]. A total of 40 male rats were used,
and they were divided into 5 groups with 8 rats per group:
group 1, normal untreated rats; group 2, diabetic control rats;
group 3, diabetic rats treated with TSE (120 mg/kg); group 4,
diabetic rats treated with TSE (240 mg/kg); and group 5,
diabetic rats treated with metformin (100 mg/kg). The doses in
the present study were selected based on antidiabetic and
hypolipidemic action of tamarind seeds proposed by Maiti et
al [13]. After final administration of treatment of 4 weeks,
blood samples were collected under fasting conditions, and
the body weights (BWs) were measured.
2.5. Immunocytochemistry
Whole pancreas from the rats was removed under anesthesia
and fixed in 10% buffered formalin for 24 hours. Tissues were
dehydrated in graded series of alcohol, embedded in paraffin,
sectioned at 5-μm thickness, and used for immunostaining.
Serial sections of the rat pancreas were immunostained by
streptavidin-biotin peroxidase method using prediluted poly-
clonal antibodies. All sections were deparaffinized in xylene
bath to remove the excess wax. The slides were placed in 2
changes of absolute alcohol for 3 minutes each. The same
procedure was repeated with 90% alcohol. The slides were
placed in a blocking reagent to block the endogenous
peroxidase activity for 5 minutes, which was prediluted with
5 volumes of 100% ethanol. The slides were placed in 2
changes of 70% alcohol for 3 minutes each. The excess alcohol
around the sections was removed, and the slides were quickly
immersed in Tris buffer, pH 7.6 for 5 minutes. Two drops of
tissue conditioner were added, and the sections were incu-
bated for 5 minutes and then rinsed in a buffer solution.
Prediluted primary polyclonal antiguinea pig antibody to
insulin (1:1000; Genetex, Irvin, CA, USA) raised against
human insulin, antirabbit glucagon antibody (1:375; Novus
Biologicals, Littleton, CO, USA) raised against Procine glucagon
conjugated to bovine serum albumin, and antirabbit somato-
statin antibody (1:500; Novus Biologicals) raised against
synthetic somatostatin were added to the sections and
incubated for 1 hour. The secondary antibodies for insulin,
glucagon, and somatostatin were antirabbit polyclonal anti-
bodies. After incubation for 0.5 hour, the sections were rinsed
with Tris buffer, peroxidise solution was added, incubated for
30 minutes, and later rinsed with the buffer. 3-amino, 9-ethyl
carbazole chromogen substrate was added to the sections and
was incubated for 15 minutes and rinsed with distilled water.
The sections were counterstained with Harris hematoxylin for
45 seconds to facilitate nuclear identification [16].
2.6. DNA fragmentation assay
For detection and localization of apoptosis in the pancreas, we
used the technique of terminal deoxynucleotidyl transferase–
mediated deoxyuridine 5-triphosphate (dUTP) nick-end label-
ing (TUNEL). Briefly, sections were deparaffinized, hydrated,
and digested with proteinase K (20 μg/mL), and then,
biotinylated dUTP was added to the 3′ end of DNA fragments
by incubating sections in 0.05 mol/L Tris-HCl buffer (pH 7.6)
with 0.03 U/μL TdT and 0.04 nmol/μL biotin-11-dUTP at 37°C
for 1 hour. The sections were rinsed in phosphate-buffered
saline (PBS). Endogenous peroxidase was blocked with 0.3%
H2O2 in distilled H2O. The sections were rinsed with PBS and
covered with 2% blocking solution in 0.1 mol/L sodium
maleate to reduce background staining. The sections were
then incubated with avidin-peroxidase complexes in PBS
(1:50) for 30 minutes and rinsed with PBS (3 × 5 minutes).
Peroxidase activity was visualized with 3,3′-diaminobenzidine
until the brown product was clearly visible. The sections were
then counterstained with methyl green. The positive apopto-
tic cells were the cells with brown nucleus [17].
2.7. Isolation and culture of islets from normal rats
Animals were killed by decapitation and pancreatic islets of
Langerhans isolated as previously described [18] using diges-
tion with collagenase obtained from Clostridium histolyticum
(Sigma Aldrich). Digestion and sedimentation of islets were
 N U TR IT ION RE S E ARCH 3 2 ( 2 0 12 ) 62 6 –6 3 6
carried out in Hanks solution containing 5.5 mmol/L glucose.
Islets were then handpicked under a stereomicroscope and
transferred to petri dishes containing culture medium RPMI
1640 (Sera Laboratories International Ltd, West Sussex, UK)
supplemented with 100 U/mL benzylpenicillin, 0.1 mg/mL
streptomycin, 2 mmol/L L-glutamine, and 10% (vol/vol) heat-
inactivated fetal calf serum (Sigma Chemicals Ltd). Islets were
cultured free-floating at 37°C, with an atmosphere of 5% CO2,
95% O2. The period of culture was 48 hours.
2.8. Measurement of cytosolic Ca 2 + concentrations
Pancreatic islet cells were prepared from adult rats and
cultured overnight in RPMI 1640 culture medium supplemen-
ted with 10% (vol/vol) fetal calf serum, 100 IU/mL penicillin,
100μg/mL streptomycin, and 60 μg/mL gentamycin. The cells
were loaded with Ca2 + indicator by incubation with 2 μM fura-
2/AM for 45 minutes in the culture medium. The Ca2 +-loaded
islets cells were then washed in a medium containing 140
mmol/L NaCl, 5.9 mmol/L KCl, 1.28 mmol/L CaCl2, 1.2 mmol/L
MgCl2, 25 mmol/L 4-(2-hydroxyethyl)-1-piperazineethanesul-
fonic acid (HEPES), pH 7.4, and 1 mg/mL bovine serum
albumin. The cells (about 4 × 106) were suspended in 2.5 mL
of fresh washing medium, stirred and placed in a spectroflu-
orometer cuvette 50 B; (Perkin Elmer, Waltham, MA, USA) for
monitoring [Ca2 +]i changes by 340 or 340/380 nm excitation
ratio, and emitted fluorescence of 510 nm at 37°C. Intracellular
Ca2 + levels could not be directly calibrated, and the [Ca2 +]i
changes presented are relative changes in fluorescence. After
a stable baseline was reached, islets were supplemented with
0 or 100 μg/mL TSE or 0 or 50 μmol/L glimepiride in the
presence or absence of 20 mmol/L D-glucose. The time that the
cells spent in the absence of glucose, that is, the time needed
for washing, suspending, and establishment of baseline, was
10 to 15 minutes [19].
2.9. RNA isolation and real-time polymerase chain
reaction analysis
Real-time polymerase chain reaction (PCR) amplifications for
SREBP-1c (gene ID: 78968) and GLUT-4 (gene ID: 25139) were
conducted using Light-Cycler 480 SYBR Green I Master
(Roche, Basel, Switzerland) according to the manufacturer's
instructions. Reaction conditions were 10× PCR buffer (20
mmol/L Tris, pH 8.4, 50 mmol/L KCl, and 2.5 mmol/L MgCl2),
20 pM oligonucleotide, 300 μmol/L deoxyribonucleotides
(dNTP), 1:1000 SYBR Green I nucleic acid stain, 0.5 U Taq
DNA polymerase (recombinant), and 50 ng templates com-
plementary DNA in a 25-mL reaction volume. Sets of PCR
primers of GLUT-4 sense 5′-AGC GTAGGTACCAACACTTTCT-
3′ and antisense 5′-CCG CCC TTA GTT GGT CAG AAG-3′; for
SREBP-1C gene sense 5′-GGAGCCATGGATTGCACATT-3′ and
antisense 5′- AGGAAGGCTTCCAGAGAGGA-3′; and for β-actin
sense 5′-TCA CCC ACA CTG TGC CCC ATC TAC GA-3′ and
antisense 5′ CAG CGG AAC CGC TCA TTG CCA ATGG-3′ were
designed with reference to the National Centre for Biotech-
nology Information (NCBI) database of conserved coding
regions. Thermal cycling conditions were 95°C for a 3-minute
denaturation step followed by 40 PCR cycles (94°C for 30
seconds, 56°C for 30 seconds, and 72°C for 1 minute), and
629
reactions were performed in a Light-Cycler 480 (Roche)
instrument. Fluorescence was detected at the end of the
56°C segment in the PCR step. Polymerase chain reaction
products of β-actin primer gene were used as an internal
standard. All assays were carried out in triplicate. Real-time
PCR analysis and subsequent calculations were performed
on Light-Cycler 480 software (version LCS480 1.2.0.169;
Roche) [20].
2.10. Glucose transporter 2 expressions in the liver and
GLUT-4 expressions in the soleus muscle
For determination of GLUT-2 protein expressions in the liver
and GLUT-4 protein expressions in the skeletal muscle, each
sample prepared was mixed with 1% sodium dodecyl sulfate
and 50 mmol/L dithiothreitol, and the mixture was subjected
to electrophoresis with 10% polyacrylamide gel. The separated
proteins in the gel were electrotransferred to a polyvinylide-
nedifluoride membrane. After blocking with 5% skimmed milk
solution including 0.05% Poly (oxyethylene) sorbitan mono-
laurate (Tween 20) overnight at 4°C, the membrane was
reacted with anti–GLUT-2 antibody and anti–GLUT-4 antibody
for 2 hours. Subsequently, it was incubated with horseradish
peroxidase-conjugated immunoglobin G (diluted 1:2000) for 2
hours at room temperature. The blots were detected with
chemiluminescence reagents (Western blot) [21].
2.11. Influence on biochemical metabolic parameters
The fasting blood glucose and BWs were measured on days 0,
14, 21, and 28. Plasma insulin was determined by enzyme-
linked immunosorbent assay (ELISA) (Mercodia, Uppsala,
Sweden) on day 28. Four weeks after the administration of
drugs (TSE, metformin), blood was collected under ether
anesthesia from the hearts of animals by cardiac puncture
before killing, placed in EDTA vacutainer tubes and centrifuged
at 4000g at 4°C for 15 minutes and stored at −80°C until
analysis. The aliquoted plasma samples were used for the
quantification of TNF-α and insulin levels. Serum was
separated by subjecting the coagulated blood to centrifugation
at 4000g for 15 minutes and further used for NO determination.
2.11.1. TNF-α, serum NO, and HbA1c
Plasma TNF-α level was determined using commercially
available ELISA kits. Briefly, wells of ELISA plate were first
coated with a TNF-α capturing antibody followed by washing
and drying. Samples were added in wells. Once the samples
were loaded, a biotinylated antirat TNF-α antibody was
subsequently added in each well. Both the antigen and
antibody were allowed to coincubate for 3 hours followed by a
series of washes. A solution of horseradish peroxidise-strepta-
vidin complex was added to each well and incubated for 20
minutes. After this, a ready-to-use chromogen was added, and
the plates were kept in the dark for 10 to 15 minutes for color to
develop. The reaction was stopped using a stop solution, and
the absorbance was read at 450 nm (Biorad, Hercules, CA, USA)
using reference filter at 630 nm. Simultaneously, a series of
dilutions of standard TNF-α was also run to obtain a standard
graph. The concentrations of TNF-α in the samples were
determined using the standard curve obtained [22].
630 N U TR IT ION RE S EA RCH 3 2 ( 2 0 12 ) 62 6 –6 36

Serum NO concentration was determined by the Griess

reagent method using available reagents and the manufac-
turer's protocol [23]. Briefly, serums were added into wells
(96-well enzymatic assay plate). A sulfanilamide solution
was added to all experimental samples, and after incubation,
N-1-naphtylethylenediamine dihydrochloride solution was
added. Then, absorbance was measured by a microreader at
520 nm wavelength. The concentration of NO samples was
determined by comparison with the nitrite standard refer-
ence curve.
The HbA1c concentration was measured by the principle of
turbidometric assay using available reagents and standards
(Biosystems, Barcelona, Spain). After preparing the hemoly-
sate using tetradecyltrimethylammonium bromide as a
detergent, the HbA1c concentration was determined by
spectrometry [24].

2.12. Statistical analyses

All values are means ± SEM. Data analysis was done with
1-way analysis of variance (ANOVA) followed by Dunnet's
multiple test where a diabetic group was considered a
positive control (Sigma Plot11, San Jose, CA, USA). Group means
were considered to significantly differ at P < .05, as determined
by Dunnet's multiple range analysis. In case of islets Ca2 + P < .05
and ⁎⁎P < .01 for difference from 0 mmol/L glucose, #P < .05, and
###
P < .001 for difference from 20 mmol/L glucose.
3. Results
3.1. Effect of TSE on BW, blood glucose, and HbA1c levels
Before the supplementation of TSE and metformin, there were
no significant differences of baseline BW of the rats (Fig. 1). The
rats treated with TSE (120 and 240 mg/kg) and metformin (100
mg/kg) showed significant increase in BW as compared with
diabetic groups after 4 weeks of study. Total HbA1c content is an
indicator of irreversible condensation of glucose with the N-
terminal residue of the β-chain of hemoglobin A. The HbA1c
concentration in blood directly denotes the mean concentration
of glucose prevailing in the previous 4 to 8 weeks, equivalent to
the lifetime of the erythrocytes. Before treatment, there were
significantly higher fasting glucose and HbA1c levels (P < .05) in
all groups when compared with the normal. The daily
administration of TSE at both dose levels (120 and 240) for the
period of 4 weeks resulted in dose-dependent reduction of
fasting glucose (milligrams per deciliter) and HbA1c (percentage)
vs the diabetic group (P < .05, Table).
3.2. Effect of TSE on plasma insulin, serum NO, and
TNF-α concentration
Insulin levels were significantly decreased in diabetic rats in
comparison with nondiabetic controls. After 28 days of TSE
supplementation to the diabetic rats, there was a significant
elevation in insulin levels in respect to diabetic control group
in a dose-dependent manner (P < .05, Table). Of the 2 doses of
TSE tested, the 240 mg/kg dose was found to be more effective
in reducing fasting blood glucose levels.
Fig. 1 – Effect of TSE on BW (grams) in diabetic rats. The
values are the means ± SEM from 8 animals in each group.
*
P < .05 vs diabetic group. Bar graph with dark shed ( )
represents BW before treatment, whereas with light shed ( )
after treatment (TSE, 120 and 240 mg/kg; metformin, 100 mg/
kg; saline, for normal rats, orally for 4 weeks). Data analysis
was done with 1-way ANOVA followed by Dunnet's multiple
test (Sigma Plot11). The diabetic group without drug treatment
was compared with the normal, TSE-treated, and metformin
(standard) groups.
Vehicle-treated diabetic rats led to significant increase in
NO and TNF-α concentrations (10.02 ± 0.50 nmol/μL, 24.50 ±
0.75 pg/mL) compared with the corresponding vehicle-treated
normal rats (1.85 ± 0.27 nmol/μL, 11.25 ± 0.59 pg/mL),
respectively. Daily administration of TSE at both dose levels
(120 and 240 mg/kg) showed dose-dependent decrease in TNF-
α concentrations when compared with corresponding diabetic
controls (P < .05, Table). In the TSE-supplemented groups, the
levels of NO were decreased by 22% and 43% at 120 and 240
mg/kg dose levels, respectively. This suggests that TSE could
be a potential NO scavenger along with its anti-inflammatory
activity that was represented by reduction in TNF-α level.
3.3. Effect of TSE on immunohistochemical changes and
cell apoptosis in the pancreas of diabetic rats
Immunohistochemical evaluation of the pancreas of diabetic
rats revealed a high frequency of degenerative changes, such
as atrophy, pyknosis, and deterioration in the islets. The
numbers of immunoreactive β cells in diabetic rats were
extremely low, and the cells stained weakly, suggesting
 N U TR IT ION RE S E ARCH 3 2 ( 2 0 12 ) 62 6 –6 3 6 631

Table – Effect of TSE on TNF-α, serum NO, and HbA1c level

TNF-α NO HbA1c (%) Fasting blood glucose (mg/dL)
(pg/mL) (nmol/μL)
Before treatment After treatment Before treatment After treatment

Normal 11.25 ± 0.59 ⁎ 1.85 ± 0.27 ⁎ 3.45 ± 1.85 ⁎ 3.95 ± 0.17 ⁎ 87.62 ± 2.94 ⁎ 89.25 ± 11.31 ⁎
Diabetic 24.50 ± 0.75 10.02 ± 0.50 7.89 ± 0.87 8.58 ± 0.51 173.50 ± 6.59 174.87 ± 9.74
TSE, 120 mg/kg 17.87 ± 0.61 ⁎ 7.74 ± 0.23 ⁎ 8.57 ± 0.89 5.74 ± 0.13 ⁎ 167.37 ± 4.13 ⁎ 140.12 ± 9.94 ⁎
TSE, 240 mg/kg 14.37 ± 0.73 ⁎ 5.64 ± 0.13 ⁎ 8.14 ± 0.58 5.34 ± 0.12 ⁎ 166.87 ± 4.80 ⁎ 130.00 ± 8.70 ⁎
Metformin 12.75 ± 0.36 ⁎ 5.46 ± 0.18 ⁎ 7.94 ± 1.02 4.85 ± 0.06 ⁎ 171.25 ± 12.03 ⁎ 113.50 ± 7.10 ⁎

The values are the means ± SEM. Data analysis was done with 1-way ANOVA followed by Dunnet's multiple test (Sigma Plot11). The diabetic
group without drug treatment was compared with the normal, TSE-treated (120 and 240 mg/kg), and metformin (standard) groups. The route of
drug (saline in case of the normal group) administration for all the groups was oral. Duration of treatment was 4 weeks after diabetes induction.
⁎ P < .05 vs diabetic group (n = 8).

Fig. 2 – Effect of TSE on immunohistopathologic changes and apoptosis (deoxynucleotidyl-transferase nick-end labeling
[TUNEL] assay) in the pancreata of diabetic rats; antiglucagon, antisomatostatin, and anti-insulin antibody immunostaining
from (A, D, and G) normal rats, (B, E, and H) vehicle-treated diabetic rats, and (C, F, and I) TSE-treated (TSE, 120 and 240 mg/kg;
saline, for normal rats, orally for 4 weeks) diabetic rats, respectively. In the TSE-treated pancreas, antirabbit glucagon antibody
(1:375) and antirabbit somatostatin antibody (1:500) raised against synthetic somatostatin yielded a staining pattern where
glucagon and somatostatin positive cells were decreased compared with diabetic rat. In contrast, prediluted primary polyclonal
antiguinea pig antibody to insulin (1:1,000) raised against human insulin revealed that β cells were abundant and intensely
stained as that in control (G) in TSE-treated rats. I, K, and L denotes effect of TSE on apoptosis (deoxynucleotidyl-transferase
nick-end labeling [TUNEL] assay) changes in the pancreata of normal (I), diabetic (K), and TSE-treated (L) rats. Morphological
features of apoptosis, including pyknotic nuclei, were readily detectable in pancreatic sections from diabetic rats that were
reduced to considerable extent in TSE-treated rats (the arrows denote TUNEL-positive cells). Original magnification ×400.
632 N U TR IT ION RE S EA RCH 3 2 ( 2 0 12 ) 62 6 –6 36

dysfunction of the β cells because of a decrease in their
number or failure to synthesize and secrete adequate
amounts of immunoreactive insulin or both. In contrast,
degenerative changes occurred at a low frequency in TSE-
treated groups, with only a slight decrease in the number of
insulin-positive granules and no marked islet atrophy,
degeneration, or necrosis. In the TSE-treated pancreas, anti-
rabbit glucagon antibody (1:375) and antirabbit somatostatin
antibody (1:500) yielded a staining pattern where glucagon and
somatostatin positive cells were decreased compared with
diabetic rats. The STZ-induced diabetic rats showed several
somatostatin positive cells (Fig. 2), suggesting the role of a
feedback mechanism of somatostatin on glucagon.
The TUNEL assay revealed that TSE supplementation in
diabetic rats reduces pancreatic cell apoptosis. In contrast,
morphological features of apoptosis, including pyknotic
nuclei, were readily detectable in pancreatic sections from
diabetic rats (Fig. 2).
3.4. Effects of TSE on muscle GLUT-4 and liver SREBP-1c
mRNA concentrations of diabetic rats
In view of the fact that acetyl coenzyme A carboxylase and
cytosolic 3-hydroxyl-3-methylglutaryl coenzyme A synthase,
the key enzymes of fatty acid synthesis and cholesterol
synthesis, respectively, are regulated by SREBP-1c during the
transcriptional step [25], we investigated the influence of TSE
on SREBP-1c mRNA. As shown in Fig. 3, the amplification
efficiency (E) for genes was determined by linear regression
analysis of the fluorescent data from the exponential phase of
PCR [26]. Quantitative reverse transcription–PCR results
showed increase in muscle GLUT-4 mRNA and liver SREBP-1c

Fig. 3 – Effects of TSE on muscle GLUT-4 mRNA (A), liver SREBP-1c mRNA (B), GLUT-4 (C) expression in the muscles, and GLUT-2
(D) expression in the liver of diabetic rats (TSE, 120 and 240 mg/kg; saline, for normal rats, orally for 4 weeks). The values are the
means ± SEM from 8 animals in each group. Amplification efficiency (E) for gene was determined by linear regression analysis
of the fluorescent data from the exponential phase of PCR [26]. In the liver and muscle membrane fractions of diabetic rats, the
translocation of GLUT4 (C) and GLUT2 (D) was very much reduced when compared with that of band density of healthy controls.
 N U TR IT ION RE S E ARCH 3 2 ( 2 0 12 ) 62 6 –6 3 6 633

mRNA concentrations after 4 weeks treatment of TSE as
compared with corresponding vehicle-treated diabetic rats.
3.5. Effect of TSE on liver GLUT-2 and muscle
GLUT-4 expression
Because there was an increase in islets cytosolic Ca2 +as well as
plasma insulin levels in TSE-treated diabetic rats and because
of the physiologic importance of insulin-dependent GLUT-2
and GLUT-4 translocation to the cell membrane, attempts have
been made to see the effect of TSE treatment on skeletal muscle
tissue membrane GLUT-4 and liver GLUT-2 levels. In the liver
and muscle membrane fractions of diabetic rats, the translo-
cation of GLUT-2 and GLUT-4 was very much reduced when
compared with the band density of healthy controls. This is
quite rational because the deficiency of insulin in the diabetic
state would decrease the translocation of GLUT-2 and GLUT-4
from the vesicles to cell membranes. Treatment with TSE
resulted in the reversal of membrane GLUT-2 and GLUT-4
levels. The modulation of GLUT-2 and GLUT-4 protein could
thus be one of the mechanisms of antidiabetic properties of
TSE. The GLUT-2 and GLUT-4 levels of normal control, diabetic,
and TSE-treated diabetic rats are shown in Fig. 3.
3.6. Cytosolic Ca 2 + concentrations
Stimulation of isolated islets with glucose produced the
expected increase of intracellular Ca2 +, especially at the
lowest glucose concentration (Fig. 4). It has been shown before
that Ca2 + oscillations of individual β cells within the mouse
and rats islet are synchronized, so that oscillations are
observed also at the level of the whole islet [27]. In the
absence of glucose, both TSE (100 μg/mL) and glimepiride (50
μmol/L) significantly increased the islet cytosolic Ca2 + (TSE,
22.22%; ⁎P < .05; glimepiride, 40%; ⁎⁎P < .01; n = 8) contents. The
addition of 20 mmol/L D-glucose alone increased the islet
cytosolic Ca2 + contents. In the presence of 20 mmol/L
D-glucose, both TSE and glimepiride significantly increased
islet cytosolic Ca2 + contents (TSE, 42%; #P < .05; glimepiride,
60%; ###P < .001; n = 8) when compared with islets treated with
D-glucose alone (Fig. 4).
4. Discussion
The neonatal-STZ Wistar model is a well-characterized model
of type 2 diabetes. The STZ rats develop persistent diabetes
rapidly after 6 weeks of age and shows diabetes-like symptoms
such as lack of insulin release in response to glucose, glucose
intolerance, and depletion of pancreatic insulin store [15]. In
the current study, TSE was tested after chronic dosing (once a
day) in a preclinical model of STZ-induced type 2 diabetes for
period of 4 weeks. The BW of diabetic rats was found to be less
during the course of the development, which may be because
of accelerated lipolysis, whereas weight gain was significantly
observed in rats treated for 4 weeks with TSE and metformin.

Fig. 4 – Effect of TSE on the pattern of cytosolic calcium concentration in islets of pancreas. The values are the means ± SEM from
8 animals in each group. Data analysis was done with 1-way ANOVA followed by Dunnet's multiple test. Islets maintained in
culture for 2 to 4 days were loaded with fura 2. The cells (about 4 × 106) were suspended in 2.5 mL of fresh washing medium and
placed in a spectrofluorometer cuvette (Perkin Elmer 50 B) for monitoring [Ca2 +]i changes by 340 or 340/380 nm excitation ratio
and emitted fluorescence 510 nm, respectively, at 37°C. The [Ca2 +]i was calculated using a standard curve of fura 2 values as a
function of calcium concentration as determined using calibrated solutions of calcium.*P < .05 and ** P < .01 for difference from
0 mmol/L glucose, #P < .05 and ###P < .001 for difference from 20 mmol/L glucose. Glime indicates glimeperide; Glu, glucose.
634 N U TR IT ION RE S EA RCH 3 2 ( 2 0 12 ) 62 6 –6 36
The TSE-treated groups had shown significant antihypergly-
cemic effects associated with increase in plasma insulin
activity. The reduced glucose levels suggested that TSE might
either promote glucose uptake by inhibiting hepatic gluconeo-
genesis or by absorption of glucose into the muscle and
adipose tissues either through the stimulation of a regenera-
tion process and revitalization of the remaining β cells to
release insulin or through the insulin mimetic response.
Because TSE is reported to have antidiabetic action, it
was of interest to analyze whether TSE affects islet [Ca2+]i
and insulin release. In the pancreatic β cell, the oscillations in
[Ca2 +]i may be of particular physiologic importance. An
elevated concentration of glucose within the β cell ultimately
leads to membrane depolarization and an influx of extracel-
lular calcium. The resulting increase in [Ca2 +]i is thought to be
one of the primary triggers for exocytosis of insulin-contain-
ing secretory granules [28]. The present results indicate that
the TSE increased Islet [Ca2 +]i significantly in the presence or
absence of 20 mmol/L D-glucose. The data are consistent with
the role of adenosine triphosphate and adenosine diphos-
phate as the connecting linkages between increased glucose
metabolism and the ionic events that lead to release of
insulin. Borman and Melzig [29] reported that flavonoids
might be responsible for inhibition of metallopeptidases that
are involved in degradation of neuropeptides. The ability of
TSE to mobilize intracellular Ca2 + in pancreatic islets both in
the absence and presence of glucose might be due to the
presence of flavonoids, suggesting inhibition of degradation of
neuropeptides in the cells of the islets of Langerhans and in
the basolateral surfaces of pancreatic acinar cells involved in
increase of [Ca2 +]i and enhanced insulin secretion [30].
The Maillard reaction (excess blood glucose + hemoglobin =
HbA1c) occurs between lipids and glucose, which results in
glycation of phospholipids in the cell membrane or the
organelle under hyperglycemic conditions. These glycated
lipids are causative agents of oxidative stress (lipid peroxida-
tion) and are related to exacerbation of chronic diabetic
complications [31]. In the present study, administration of
the extract for 4 weeks tended to significantly bring down the
level of HbA1c in both TSE-treated diabetic groups. This might
have been due to improved glycemic control produced by
the extract.
The destruction of β cells in diabetes is mediated by
changing expression levels of antiapoptotic or proapoptotic
proteins. The STZ could induce NO formation and further lead
to potential mitochondrial membrane changes, hence the
release of cytochrome C, which triggers apoptosis [32]. In the
present study, the diabetic rats showed significant increase in
NO and decrease in insulin secretions and vice versa in
normal rats. The TSE treatment to diabetic rats shows
decrease in NO but significant increase in insulin secretions.
Our results demonstrate that TSE suppresses STZ-induced
apoptosis significantly. These data suggest that TSE possesses
potent antiapoptotic effects when β cells have STZ-induced
impairment. Correlations could be found in our study between
the total phenolic content and NO scavenging ability of the
extract. The results are in agreement with those reported by
Komutarin et al [8], where the seed coat extract of tamarind
seeds suppressed NO in the murine macrophage cell line and
freshly isolated peritoneal macrophages.
Significant changes are observed after TSE treatment in the
distribution of insulin, glucagon, and somatostatin-positive
cells in the islets of Langerhans. The number of insulin-
positive cells decreases markedly in both the TSE-treated and
an untreated diabetic rat when compared with control
animals, but the decrease is much greater in the untreated
rats. The effect may be due to prevention of β-cell death by
decreasing the oxidative stress caused by STZ in diabetic rats.
Pons and Aoki [33] reported that the number of glucagon-
producing α and somatostatin-producing δ cells was increased
in diabetes, and it appears that somatostatin-producing cells
rise in number to compensate for the relative reduction in
insulin-secreting cells. This up-regulation of α cells of the
pancreas controls the glucagon release to maintain glucose
homeostasis by way of the feedback mechanism of somato-
statin. This plausible mechanism supports the results of the
present study, where there is a decrease in glucagon-positive
cells, whereby somatostatin-positive cells were noted in TSE-
supplemented rats.
The GLUT-4 expression is down-regulated when there is
relative insulin deficiency, such as in STZ-induced diabetes
[34]. The GLUT-4 protein level was measured by immunoblot
analysis in the homogenate of skeletal muscles and was
decreased in the diabetic control group. After treatment with
TSE, the contents of GLUT-4 protein and GLUT-4 mRNA were
restored to near normal values in skeletal muscles. The
decrease in GLUT-4 levels is essentially one of the main
reasons of hyperglycemia in the diabetic state, which is due to
decreased uptake of glucose by the skeletal muscles. Resto-
ration of GLUT-4 levels would, therefore, enhance the uptake
of glucose in the skeletal muscles and thus help to combat
hyperglycemic conditions. Glucose unresponsiveness associ-
ated with GLUT-2 impairment is typically demonstrated in
type 2 diabetes [35]. The effects of insulin on SREBP-1c have
been corroborated by in vivo studies showing that SREBP-1c
expression and nuclear abundance were low in the livers of
STZ-induced diabetic rats and markedly increased after
insulin treatment [36]. Therefore, the stimulating action of
TSE on GLUT-2 protein and SREBP-1c mRNA expression might
be due to the insulin secretogogue effect.
Although our data are compatible with the previous
findings on T indica, some limitations in this study may be
considered. The increased cytosolic calcium in the present
investigation is thought to be one of the mechanisms for
insulin release; nevertheless, the possibility of other sites of
TSE action on the insulin-secretary mechanisms cannot be
eliminated, for example, interaction with calmodulin and
protein kinase C. Furthermore, a glucagon-like peptide 1 and
gastrointestinal functions, which also play an important role
in postprandial hyperglycemia, are also criteria that need to be
studied. However, the present study has highlighted the
relationship between β-cell regeneration and TSE, which
represents a challenge for future studies. Better knowledge
of the mechanisms underlying the β-cell function in regener-
ating islets should be of more help for the development of
diabetes treatment.
To conclude, the data presented in this study support and
thus accept our primary hypothesis that the antidiabetic
action of tamarind seeds comprises an array of actions and
provides direct evidence for the involvement of GLUTs (GLUT-
 N U TR IT ION RE S E ARCH 3 2 ( 2 0 12 ) 62 6 –6 3 6
2 and GLUT-4) in controlling glucose homeostasis. In addition,
the β cell–preserving action of TSE along with the insulin
mimectic potential (mediated by increase in cytosolic Ca2 + of
islets) also helps to control hyperglycemic excursions in
diabetes. Furthermore, the study has also provided a plausible
mechanistic explanation for the effect of seeds and fruits of T
indica on lipid metabolism [13,37], as the SREBP-1c mRNA
concentration in the liver was improved. Therefore, polyphe-
nol-rich seeds of tamarind and its aqueous extract could be
added as a supplement in human health care food or drugs
or combined with other hypoglycemic agents. The present
study on the antidiabetic potential of tamarind seeds
provides a strong rationale for the study of compounds
and functionalities involved in its antihyperglycemic action,
thereby offering a foundation for a new herbal drug in
diabetes therapy.
Acknowledgment
The authors acknowledge the help of Panacea Biotech and
Ranbaxy, India, for providing glimeperide and metformin,
respectively. Funding for the study was by the Govt of NCT,
Delhi, India (MH-2203-O.E.1.1[1][1][1][4]). The authors thank
Dhiren Bahl of Wordsway Copyediting for editorial assistance.
REFERENCES
[1] Ndsing JF. Role of hemeoxygenase in inflammation insulin
signaling diabetes and obesity. Mediat Inflamm 2010;359:
732–8.
[2] Charron MJ, Brosius FC, Alper SL, Lodish HF. A glucose
transport protein expressed predominantly in insulin-re-
sponsive tissues. Proc Natl Acad Sci 1989;86:2535–9.
[3] Shepherd PR, Kahn BB. Glucose transporters and insulin
action: implications for insulin resistance and diabetes
mellitus. N Engl J Med 1999;341:248–57.
[4] Bell GI, Kayano T, Buse JB, Burant CF, Takeda J, Lin D.
Molecular biology of mammalian glucose transporters. Dia-
betes Care 1990;13:198–208.
[5] Shimomura I, Bashmakov Y, Ikemoto S, Horton JD, Brown MS,
Goldstein JL. Insulin selectively increases SREBP-1c mRNA in
the livers of rats with streptozotocininduced, diabetes. Proc
Natl Acad Sci 1999;96:13656–61.
[6] Prentki M, Matschinsky FM. Ca2 +, cAMP, and phospholipid-
derived messengers in coupling mechanisms of insulin
secretion. Physiol Rev 1987;67:1185–248.
[7] Hadjivassiliou V, Green MHL, James RFL, Swift, Clayton HL,
Green IC. Insulin secretion DNA damage and apoptosis in
human and rat islets following exposure to nitric oxide,
peroxynitrite and cytokines. Nitric Oxide 1998;2:429–41.
[8] Komutarin T, Butterworth AL, Keil D, Chitsomboon B, Suttajit
M, Meade BJ. Extract of seed coat of Tamarindus indica L.
Inhibits nitric oxide production by murine macrophages in
vitro and in vivo. Food Chem Toxicol 2004;42:649–59.
[9] Tsuda T, Watanable M, Ohshima K, Yamamoto A, Kawakishi
S, Osawa T. Antioxidative components isolated from the seed
of Tamarind (Tamarindus indica L). Agric Food Chem 1994;42:
2671–4.
[10] Joyeux M, Mortier F, Flurentin J. Screening of antiradical,
antilipoperoxidant and hepatotoprotective effects of nine
plant extracts used in Caribbean folk medicine. Phytother Res
1995;9:228–30.
635
[11] Rimbau V, Cerdan C, Vila R, Iglesia J. Antiinflammatory
activity of some extracts from plants used in traditional
medicines of North-African Countries (II). Phytother Res
1999;13:128–32.
[12] Iyer SR. Tamarindus indica Linn. In: Warrier PK, Nambiar VPK,
Kutty CR, editors. Indian Medicinal Plants, Vol. 5. Orient
Longman Limited; 1995. p. 235–6.
[13] Maiti R, Das UK, Ghosh D. Attenuation of hyperglycemia and
hyperlipidemia in streptozotocin induced diabetic rats by
aqueous extract of seed of Tamarindus indica. Biol Pharm Bull
2005;28:1172–6.
[14] Khillare B. Orientation training course on research method-
ology in reproductive biomedicine. Department of Repro-
ductive Biomedicine, National Institute of Health and Family
Welfare, Government of India; 2000.
[15] Weir GC, Clore EE, Zma-Chinsky CJ, Bonnier-weir S. Islet
secretion in new experimental model of non-insulin depen-
dent diabetes. Diabetes Metab Rev 1981;30:590–4.
[16] Hsu SM, Raine L, Fanger H. Use of avidin-biotin peroxidase
complex (ABC) in immunoperoxidase techniques: a compar-
ison between ABC and unlabelled antibody (PAP) procedures.
J Histochem Cytochem 1981;29:577–80.
[17] Matsuno T, Sasaki H, Nakagawa K, Nakagawa K, Ishido N,
Matsuda H, et al. Fas antigen expression and apoptosis in
kidney allografts. Transplant Proc 1997;29:177–8.
[18] Lacy PE, Kostianovsky M. Method for the isolation of intact
islets of Langerhans from the rat pancreas. Diabetes 1967;16:
35–9.
[19] Thomas PJ, Gaspers LD, Pharr C, Thomas JA. Continuous
measurement of mitochondrial pH gradients in isolated
hepatocytes by difference ratio spectroscopy. Arch Biochem
Biophys 1991;288:250–61.
[20] Suh YH, Kim Y, Bang JH, Choi KS, Lee JW, Kim WH, et al.
Analysis of gene expression profiles in insulin-sensitive
tissues from pre-diabetic and diabetic Zucker diabetic fatty
rats. J Mol Endocrinol 2005;34:299–315.
[21] Yoshihiko H, Takuya I, Kaoru Y, Nobuo K. Effects of Hachimi-
jio-gan (Ba-Wei-Di-Huang-Wan) on hyperglycemia in strepto-
zotocin induced diabetic rats. Biol Pharm Bull 2007;30:1015–20.
[22] Nepal S, Malik S, Sharma AK, Bhartia S, Narender Kumar,
Siddiqui KM, et al. Abresham ameliorates dyslipidemia,
hepaticsteatosis and hypertension in high fat diet fed rats by
repressing oxidative stress, TNF-α and normalizing NO
production. Exp Toxicol Pathol 2011, http://dx.doi.org/10.1016/
j.etp.2011.01.003.
[23] Nematbakhsh M, Khazaei M. The effect of estrogen on serum
nitric oxide concentrations in normotensive and DOCA salt
hypertensive ovariectomized rats. Clin Chim Acta 2004;344:
53–7.
[24] Karl J, Burns G, Engel WD, Finke A, Kratzer M, Rollinger W.
Development and standardization of a new immunoturbidi-
metric assay HbA1c Assay. Klin Lab 1993;39:991–6.
[25] Lopez JM, Bennett MK, Sanchez HB, Rosenfeld JM, Osborne TE.
Sterol regulation of acetyl coenzyme A carboxylase: a
mechanism for coordinate control of cellular lipid. Proc Natl
Acad Sci 1996;93:1049–53.
[26] Peirson SP, Butler JN, Foster RG. Experimental validation of
novel and conventional approaches to quantitative real-
timePCR data analysis. Nucleic Acids Res 2003;31:e73.
[27] Valdeolmillos M, Nadal A, Soria B, Garcia-Sancho J. Fluorescence
digital image analysis of glucose-induced [Ca2 +]i oscillations in
mouse pancreatic islets of Langerhans. Diabetes 1993;42:1210–4.
[28] Weigle DS. Pulsatile secretion of fuel-regulatory hormones.
Diabetes 1987;36:764–75.
[29] Borman H, Melzig MF. Inhibition of metallopeptidases by
flavonoids and related compounds. Die Pharmazie 2000;55:
129–32.
[30] Adeghate E, Donath T. Distribution of neuropeptide-Y and
vasoactive intestinal polypeptide immunoreactive nerves in
636 N U TR IT ION RE S EA RCH 3 2 ( 2 0 12 ) 62 6 –6 36
normal and transplanted pancreatic tissue. Peptides 1990;11:
1087–92.
[31] Nagai R, Fujiwara Y, Mera K. Usefulness of antibodies for
evaluating the biological significance of AGEs. Ann N Y Acad
Sci 2008;1126:38–41.
[32] Hirst DG, Robson T. Nitrosative stress as a mediator of apoptosis:
implications for cancer therapy. Curr Pharm Des 2010;16:45–55.
[33] Pons P, Aoki A. Differential proliferation of somatostatin and
glucagons cells in rat pancreatic islets submitted to different
stimuli. Ann Anat 1995;177:221–7.
[34] Charron MJ, Katz EB, Olson AL. GLUT 4 gene regulation and
manipulation. J Biol Chem 1999;274:3253–6.
[35] Thorens B, Wu YJ, Leahy JL, Weir GC. The loss of GLUT2
expression by glucose unresponsive beta cells of db/db mice
is reversible and is induced by the diabetic environment.
J Clin Invest 1992;90:77–80.
[36] Valerio A, Cardile A, Cozzi V, Bracale R, Tedesco L, Pisconti A,
et al. TNF-α downregulates eNOS expression and mitochon-
drial biogenesis in fat and muscle of obese rodents. J Clin
Invest 2006;116:2791–8.
[37] Martinello FSM, Soares JJ, Franco AC, Santos A, Sugohara SB,
Garcia C, et al. Hypolipemic and antioxidant activities from
Tamarindus indica L. pulp fruit extract in hypercholesterolemic
hamsters. Food Chem Toxicol 2006;44:810–8.