Laboratory of Translational Immunology, Faculty of Medical Sciences, University of Campinas,
E-Mails: firstname.lastname@example.org (N.S.G.S.); email@example.com (T.C.C.);
firstname.lastname@example.org (D.F.); email@example.com(V.C.C.)
Department of Food and Nutrition, School of Food Engineering, University of Campinas
13083-862 Campinas-SP, Brazil; E-Mails: firstname.lastname@example.org (C.B.B.C.);
Agronomic Institute of Campinas, Theodureto de Almeida Camargo, 1500, Vila Nova,
Department of Internal Medicine, Faculty of Medical Sciences, University of Campinas, Rua Tessália
Tel.: +55-19-3289-3709; Fax: +55-19-3521-8925.
Academic Editors: Antonio Segura-Carretero and David Arráez-Román
Received: 17 June 2015 / Accepted: 8 September 2015 / Published: 9 October 2015
Abstract: Medical and folklore reports suggest that Eugenia uniflora (E. uniflora) is a
functional food that contains numerous compounds in its composition, with anti-inflammatory,
antioxidant and anti-diabetic effects. In the present study, we investigated the best solvents
(water, ethanol and methanol/acetone) for extracting bioactive compounds of E. uniflora
leaves, assessing total phenols and the antioxidant activity of the extracts by
2,2′-Azinobis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) and Oxygen Radical
Absorbance Capacity (ORAC) assays, identifying hydrolysable tannins and three phenolic
compounds (ellagic acid, gallic acid and rutin) present in the leaves. In addition, we
evaluated the incidence of diabetes, degree of insulitis, serum insulin, hepatic glutathione
and tolerance test glucose in non-obese diabetic (NOD) mice. Our results suggest that the
aqueous extract presents antioxidant activity and high total phenols, which were used as a
type 1 diabetes mellitus (DM-1) treatment in NOD mice. We verified that the chronic
consumption of aqueous extract reduces the inflammatory infiltrate index in pancreatic islets,
maintaining serum insulin levels and hepatic glutathione, and reducing serum lipid
peroxidation as well as the risk for diabetes.
Keywords: Eugenia uniflora; type 1 diabetes mellitus; NOD mice; antioxidant
The Eugenia L. genus belongs to the family Myrtaceae and has more than 500 species, of which 400
are native to Brazil and are used as medicinal plants. Eugenia uniflora (E. uniflora), popularly known as
Surinam cherry, belongs to this genus and is found in regions with tropical and subtropical climates,
where it is prized for its fruit . Due to the therapeutic activities of E. uniflora tea, made from its leaves,
this plant has been studied for its effectiveness in treating various diseases and its application in folk
medicine as an antioxidant, hypotensive, anti-inflammatory and hypoglycemic agent .
The main mechanism of action of antioxidants include radical scavengers and suppressors that
neutralize or eliminate reactive oxygen species (ROS)/nitrogen (RNS) and the binding of metal ions,
which are necessary for the production of oxidizing species . Antioxidants may be classified as
endogenous (glutathione peroxidase, catalase and superoxide dismutase) and also as exogenous from
our diet, such as Vitamins A, C and E, minerals and flavonoids, among others . The presence of these
endogenous or exogenous components may be essential for the complex control of oxidative stress and
cell damage. Phenolic compounds derived from plant sources are widely studied antioxidant compounds
and act in the neutralization of free radicals, helping to control the oxidative stress that occurs in the
pancreatic islets of diabetic rats . These compounds are divided into: phenolic acids,
phenylpropanoids, flavonoids, condensed and hydrolyzed tannins [5,6]. Metabolic diseases, such as
diabetes mellitus, are characterized by a decrease in endogenous antioxidants in the body, leading to an
increase in free radicals. Thus, dietary supplementation with antioxidant compounds (present in
NOD mice are used as an experimental DM-1 model because they spontaneously develop a similar
disease to that observed in humans. Diabetes onset in NOD mice starts between the 12th and 24th weeks
of life. Polydipsia, polyuria, high glycosuria, hyperglycemia and insulin deficiency are observed in these
animals, accompanied by a rapid loss of weight [7,8]. Among mechanisms proposed to break immune
tolerance in DM-1, the genetic predisposition of an individual, together with environmental factors such
as stress and diet, seems to contribute to the inflammatory autoimmune response. As such, researchers
was the identification of phenolic compounds with antioxidant and anti-inflammatory properties in
Samples selected to compose this project are not on the list of endangered species, according to
IBAMA-CITES (Convention on International Trade in Endangered Species of Wild Fauna and Flora).
The identification of samples was performed by the Agronomic Institute of Campinas.
refrigeration at 8 °C, and then submitted to a process for extracting phenolic compounds, using three
different solvents at the same concentration (0.04 mg/mL), i.e., water, ethanol and methanol/acetone.
The aqueous extract was prepared using 1 g of sample and 25 mL of water in an autoclave system for
20 min/121 °C, according to our previous study . The tea made by the process of autoclaving was
filtered and stored at 8 °C. The ethanol extract was adapted from Spagolla et al. (2009) . One gram
of sample was suspended in 15 mL of 60% ethanol and maintained for 1 h at 70 °C. The sample was
filtered, the supernatant suspended in 10 mL of 60% ethanol in the residue and the material was extracted
again for 1 hour. The sample was stored at 8 °C. The preparation of methanol/acetone extract was
adapted from Larrauri et al. (1997) . One gram of sample was suspended in 10 mL of 50% methanol
at room temperature (24 °C), for 60 min. The sample was then centrifuged at 9948× g for
20 min/22 °C. The supernatant was added to 10 mL of 70% acetone for another 60 min, at room
temperature, and the sample was centrifuged again. The supernatants of the three prepared extracts were
filtered and stored at 8 °C for further chemical analysis.
2.2.1. Determination of Antioxidant Activity of Extracts
Extracts (aqueous, ethanol and methanol/acetone) of E. uniflora were submitted to four different
assays for antioxidant activity analysis at an initial concentration of 0.04 g/mL. DPPH radical
scavenging ; capacity to reduce metal iron FRAP [13,14]; inhibition of free radical ABTS [11,15]
and by peroxyl radical capture ORAC .
2.2.2. Determination of Total Phenolic Compounds
Total phenolic compounds in the extracts were determined by the Folin-Ciocalteu method with some
modifications . Gallic acid was used to construct the standard curve for phenol analysis, varying the
concentrations from 0.016 to 0.1 mg/mL. The absorbance was read using a spectrophotometer
(Spectramax 190, Molecular Devices, Sunnyvale, CA, USA) at 725 nm, and the results were expressed
as gallic acid equivalents (GAE mg/g sample).
2.2.3. Identification of Hydrolysable Tannins
were added in 5 mL of distilled water and 100 μL of 2% ferric chloride. The presence of tannins was
determined by the precipitate color and formation, where green precipitates indicate condensed tannins
and blue hydrolysable tannins.
The quantification of hydrolysable tannins was determined according to Brune et al. (1992)  with
some modifications by Schons and Macedo et al. (2011) , using the ferric ammonium reagent
FAS-reagent. The standard curve was performed using tannic acid (Sigma, Aldrich, St. Louis, MO,
USA) at concentrations between 0.01 and 0.4 mg/mL; after 15 min, the absorbance of the reaction
product was read using a spectrophotometer (Spectramax 190) at 578 nm. Results were expressed as
tannic acid equivalent (mg of tannic acid/g sample).
Validation methodology employed the external standards of the Food Analysis Laboratory,
FEA-Unicamp. The extracts were diluted in 80% methanol and additional identification of antioxidant
flavonoids-gallic acid, vanillic acid, ellagic acid, p-coumaric acid, ferulic acid, kaempferol, resveratrol,
quercetin, catechin, epicatechin and rutin, were performed using the standard stock solution
(10 mg/mL HPLC grade) to obtain a 5-point calibration curve. The HPLC system used was HPLC
(Agilent Technologies 1100, Santa Clara, CA, USA), coupled to a diode array detector (DAD) (Agilent
G1315B) with a flow rate of 0.70 mL/min, at room temperature. The chromatographic column used in
the reversed-phase was the Eclipse XBB-C18 (Agilent Technologies). Detection was performed at 210,
254, 300 and 340 nm for the identification of phenolic compounds. The mobile phase consisted of 1%
orthophosphoric acid in ultrapure water (A) and 100% acetonitrile (B). The elution gradient used was as
follows: 0 min: 95% A and 5% B, 10 min: 75% A and 25% B, 25 min: 60% A and 40% B, followed by
a linear increase of solvent A until 35 min. Phenolic compounds were identified in the chromatograms
of the injected samples and compared with the phenolic standards used, with regard to relative retention
time (RT), peak area percentage and spectral data.
2.4. Biological Assays: NOD Mice Treatment with E. uniflora Leaves
2.4.1. Experimental Design
Female NOD mice used in this study were obtained from the Multidisciplinary Center for Biological
Research, University of Campinas (Cemib-Unicamp) . Procedures involving animals and their care
were conducted in accordance with guidelines and recommendations established by the Brazilian
Committee for Animal Experimentation—COBEA (protocol: 2824-1).
All animals were treated from the 4th until the 26th week of life, and body weight and fasting blood
glucose levels were monitored weekly, beginning at 10th week of life until the end of the protocol, using
glucometer strips (Medisense Optium
, Abbott Diabetes Care Inc., Alameda, CA, USA).
kept under specific pathogen free (SPF) conditions, with controlled light, temperature and humidity in
the Laboratory of Translational Immunology (LTI) facilities. The groups were divided into: Group
(0.06 g/100mL of filtered water standards established by the IC50-DPPH analysis). Based on previous
trials of antioxidant activity and microbiological control, it was established that the aqueous extract of
exchanges would offer animals every two days. For each new leaf extract, the DPPH technique was
performed to determine the concentration to be used. The second experimental group was classified as
untreated, and composed of 30 NOD mice. Thus, from these groups, a third subgroup was designated as
the acute diabetic group, and composed of untreated and treated mice that became diabetic during the
time of protocol study. In addition to the experimental design, another 36 NOD female and 12 BALB/c
mice (healthy control mice that do not develop diabetes)  were used for glucose tolerance test.
2.4.2. Diagnosis of Acute Diabetes and Animal Sacrifice
At the end of 26 weeks or mice diabetes onset (blood glucose above 250 mg/dL for two consecutive
days), the animals were sacrificed by intraperitoneal anesthesia (ketamine hydrochloride: 150mg/kg and
xylazine hydrochloride: 10mg/kg, both Vetbrands, Paulínia, Brazil) and peripheral blood was collected
by cardiac puncture for serum separation. Liver was removed and homogenized in 5% trichloroacetic
acid (TCA) and frozen at −80°C to further analysis of reduced glutathione (GSH). After this procedure,
the pancreas was removed and included into capsules containing tissue freezing medium (Triangle
Biomedical Sciences, Durham, NC, USA), frozen in liquid nitrogen and stored at −80 °C for
2.4.3. Histological Analysis
The fixed pancreas were submitted to cryosections (Cryostat CM 1850, Leica Biosystems, Wetzlar,
Germany) for morphological analysis of pancreatic islets and insulitis degree classification
(inflammatory cell infiltration). A series of 15 consecutive cuts of 5 μm were placed on silanized slides
(Methacryl-oxypropyl-Methoxysilane, Sigma). The slide numbers 1, 15, 16, 30, 31 and 45 were stained
using hematoxylin and eosin technique (HE) and analyzed by optical microscopy (Nikon
The analysis of the islets was carried out according to the criteria established by Signori et al. (1989) ,
and adapted by Ventura et al. , where grade 0 is characterized by absence of inflammatory cell
infiltrate, grade 1 was less than 25%, grade 2 when the islet has a cell infiltration of between 25% and
80%, grade 3 represents higher than 80% and less than 100%, while islets completely overtaken by
cellular infiltration are classified as islets grade 4 (Figure 1). The insulitis index was determined
according to Leiter (2001) , considering 300 islets/group (5 mice/group).
2.4.4. Detection of Serum Insulin Levels
Serum insulin was determined by rat/mouse insulin ELISA kit, according to the instructions of the
, Billerica, MA, USA). After four hours of fasting, mice were sacrificed and
spectrophotometer (Spectramax 190) at 450 nm and 590 nm, and the curve drawn using four
infiltration. (A) Grade 0 (0%); (B) Grade 1 (<25%); (C) Grade 2 (25%–80%); (D) Grade 3
(>80%) and (E) Grade 4 (100%).
2.4.5. Glucose Tolerance Test
The glucose tolerance test (GTT) was performed according to Andrikopoulos et al. (2008) , with
some modifications. Mice (15–28 weeks old) were divided into the following groups: NOD mice treated
for one month with E. uniflora aqueous extract (n = 12), untreated NOD mice (n = 12), acute diabetic
NOD mice (n = 12) and BALB/c (healthy control mice), which do not develop diabetes (n = 12). The
animals were fasted for 4 hours, followed by dosing of glucose at time 0 (blood sample from the caudal
vein) using measuring strips (Medisense Optium
, Abbott Diabetes Care Inc. Alameda, CA, USA). After
blood samples were collected at 10, 30, 60, 90 and 120 min followed by the glucose quantified at
2.4.6. Reduced Glutathione Analysis
Protein concentration and reduced glutathione (GSH) of liver homogenates were determined
according to the Hartree (1972)  and Faure and Lafond (1995) , respectively. Standard curves
(2.5–500 nmol/mL), 100 μL of Tris/EDTA buffer (1mM Tris/2 mM EDTA) and 20 μL of reactive DTNB
10 mM (5,5′-Dithiobis(2-nitrobenzoic acid) were used in GSH analysis. Absorbance was read at 412 nm
in a spectrophotometer (Spectramax 190). The calculation of glutathione was made by determining the
glutathione/total protein ratio and the result was expressed as nmol GSH/mg protein.
2.4.7. Determination of Serum Antioxidant Activity and Lipid Peroxidation Assay
the results were expressed as μM trolox/g sample [11,15].
The concentration of lipid peroxidation product or thiobarbituric acid reactive substances (TBARS)
was determined in the serum of NOD mice, according to Ohkawa et al. (1979)  with modifications .
The samples were mixed with 8.1% sodium dodecyl sulphate (SDS) plus working reagent (TBA, 20%
acetic acid, and 5% sodium hydroxide). After heating at 95°C for 60 min, the samples were maintained
in an ice-bath for 10 min, and centrifuged at 10,000 g for 10 min. The supernatant was read at 532 nm
in Synergy HT, Biotek microplate reader (Winooski, VT, USA). The results are expressed as μmol MDA
2.5. Statistical Analysis
The results were expressed as mean values ± standard error (SE) and were analyzed using the
GraphPad software, version 6.0 (San Diego, CA, USA). All data were submitted to the normality test.
Water consumption data were analyzed by Student T-test analysis. Other comparisons were performed
by one-way ANOVA of variance with Tukey post-test and Kruskal-Wallis test with Dunn’s post-test.
The statistical differences were represented by letters; the same letters located on top of the bars
correspond to the absence of statistical difference, whereas different letters represent statistical
difference among the groups. p < 0.05 was taken as indicative of statistical significance.
DPPH analysis (% scavenge) shows a higher percentage of radical sequestration with statistical
difference between aqueous (39.9 ± 1.9) and methanol/acetone (27.6 ± 2.0) (p = 0.0061) and no
difference with ethanol extract (34.2 ± 1.5). However, in the FRAP assays (μM FeSO
/ g sample), the
(p = 0.0017), but similar to that of methanol/acetone (80.0 ± 1.9). Using the ABTS assays (μM trolox/g
sample), the aqueous extract demonstrated a higher antioxidant activity (35,792 ± 1044), followed by
methanol/acetone (28,960 ± 760) and ethanol extract (24,775 ± 612), with statistical difference between
the aqueous and ethanol extracts (p = 0.0012). The ORAC assays (μM trolox/g sample) did not
demonstrate any statistical difference between the aqueous (1118 ± 83), ethanol (1021 ± 1) and
methanol/acetone (1001 ± 119) extracts. The total phenolic compound analysis (mg GAE/g sample)
showed a higher concentration in the aqueous extract (73.3 ± 1.0), compared with the ethanol extract
(44.8 ± 0.8) (p = 0.0005), but no difference from the methanol/acetone extract was seen (64.8 ± 1.1).
Hydrolyzed tannins were identified in the three extracts of E. uniflora leaves using the qualitative
method. The quantitative method (mg tannic acid/g sample) showed that ethanol extracted greater
amounts of hydrolysable tannins (167.7 ± 1.5), in comparison with the aqueous (154.5 ± 3.2) and
methanol/acetone (154.1 ± 4.2) extracts. There was a statistical difference only between the ethanol and
methanol/acetone extracts (p = 0.0196).