Académique Documents
Professionnel Documents
Culture Documents
discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/257837886
CITATIONS
READS
13
133
4 authors, including:
Alessandra L. Oliveira
University of So Paulo
42 PUBLICATIONS 396 CITATIONS
SEE PROFILE
All content following this page was uploaded by Alessandra L. Oliveira on 21 August 2014.
The user has requested enhancement of the downloaded file. All in-text references underlined in blue
are linked to publications on ResearchGate, letting you access and read them immediately.
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
a r t i c l e
i n f o
Article history:
Received 24 April 2013
Received in revised form 6 August 2013
Accepted 14 August 2013
Available online 29 August 2013
Keywords:
Eugenia uniora
Pressurised uid extraction
High performance liquid chromatography
Centrifugal partition chromatography
High resolution mass spectrometry
a b s t r a c t
Brazilian cherry seeds are a waste product from juice and frozen pulp production and, the seeds composition was investigated to valorize this by-product. Compounds separation was performed with ethanol
by pressurised uid extraction (PFE). Here we determine the effect of temperature (T), static time (ST),
number of cycles (C), and ush volume (VF) on the yield, composition and total phenolic content (TPC)
of the seed extracts. T, ST and their interaction positively inuenced yield and TPC. Extracts were fractionated by high performance liquid chromatography (HPLC) and centrifugal partition chromatography (CPC).
The collected fractions characterizations were made by electrospray ionisation mass spectrometry
(ESI/MS) and high resolution mass spectrometry (HRMS) indicated the presence of ellagic acid pentoside
and deoxyhexose, quercitrin and kaempferol pentoside. All of these compounds have antioxidant
properties and normally are found in plant extracts. These results conrm that Brazilian cherry seed
extract is a potentially valuable source of antioxidants.
2013 Elsevier Ltd. All rights reserved.
1. Introduction
The Brazilian cherry (Eugenia uniora L.) is a traditional crop in
Southern and Southeastern Brazil, but, in recent years, regular cultivation for commercial purposes has begun in the Northeast. It is a
globular berry, 1.55.0 cm in diameter, with seven to ten longitudinal grooves. During maturation, the exocarp varies from green
to yellow, orange, red or dark red (Bezerra, Silva and Lederman,
2000). The fruit has 69% pulp, 31% seeds and is about 85% water
(Guimares, Holanda, Maia, & Moura F, 1982; Silva, 2006). Among
the tropical fruits, the Brazilian cherry has one of the higher levels
of carotenoids (225.9 lg/g) and vitamin C content of 29.4 mg/
100 g. It also contains large amounts of calcium phosphorus and
anthocyanins. The carotenoids are phytouene, b-carotene, f-carotene, b-cryptoxanthin, c-carotene, lycopene, and rubixanthin.
Lycopene represents 32% of the total carotenoids. Among the phenolic compounds, the dark red Brazilian cherry has 22.50 mg/100 g
of anthocyanins (Bezerra et al., 2000; Filho et al., 2008).
Recently, the volatile compounds present in the Brazilian cherry
leaves, whose biological effects have already been proved, were
found in the fruit as selina-1,3,7(11)-trien-8-one for example (Oliveira, Lopes, Cabral, & Eberlin, 2006). An extract obtained by super Corresponding author. Tel.: +55 19 3565 4268; fax: +55 19 3565 4284.
E-mail address: alelopes@usp.br (A.L. Oliveira).
0308-8146/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodchem.2013.08.065
decreases the viscosity of the solvent enabling an improved penetration of the matrix and enhances diffusivity of the solvent. Moreover, high pressure keeps the solvent in the liquid state and forces
the solvent through the matrix. The reduced time of extraction
avoids a possible thermal degradation (Kaufmann & Christen,
2002; Wang & Weller, 2006). PFE can be performed using static
or dynamic methods or a combination of both. This method is
widely used as an extraction technique for sample preparation,
to identify the presence of minor components. More recently, this
extraction technique has also been used to obtain extracts enriched
with bioactive compounds (Claude, Morin, Lafosse, Belmont, &
Haupt, 2008; Pl et al., 2007). In this paper, we seeked to optimise
the PFE procedure by examining the impact of altering four important variables: extraction temperature (T), static time (ST), number
of cycles (C) and the volume of solvent in the extractor in each cycle (VF). In addition, chemical analyses were performed to detect
biological activity, as well as analyse composition of these extracts
using thin layer chromatography (TLC), high performance liquid
chromatography (HPLC) coupled with UV, evaporative light scattering detectors (ELSD), and diode-array detection (DAD), centrifugal partition chromatography (CPC) and HPLC followed by mass
spectrometry coupled with electrospray ionisation (ESI/MS).
523
524
Table 1
Matrix of the central composite design (CCD) 24 to study the effect of T, ST, C and VF on the extract yield (Y) and TPC-coded and real variables.
Assay
T (C)
C (no)
VF (%)
ST (min)
Y (%)
TPC (g GAE/100 g)
TPC (ppm)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25a
26a
27a
1 (50)
+1 (70)
1 (50)
+1 (70)
1 (50)
+1 (70)
1 (50)
+1 (70)
1 (50)
+1 (70)
1 (50)
+1 (70)
1 (50)
+1 (70)
1 (50)
+1 (70)
0 (60)
0 (60)
0 (60)
0 (60)
0 (60)
0 (60)
a (40)
+a (80)
0 (60)
0 (60)
0 (60)
1 (2)
1 (2)
+1 (4)
+1 (4)
1 (2)
1 (2)
+1 (4)
+1 (4)
1 (2)
1 (2)
+1 (4)
+1 (4)
1 (2)
1 (2)
+1 (4)
+1 (4)
0 (3)
0 (3)
0 (3)
0 (3)
a (1)
+a (5)
0 (3)
0 (3)
0 (3)
0 (3)
0 (3)
1 (80)
1 (80)
1 (80)
1 (80)
+1 (120)
+1 (120)
+1 (120)
+1 (120)
1 (80)
1 (80)
1 (80)
1 (80)
+1 (120)
+1 (120)
+1 (120)
+1 (120)
0 (100)
0 (100)
a (60)
+a (140)
0 (100)
0 (100)
0 (100)
0 (100)
0 (100)
0 (100)
0 (100)
1 (4)
1 (4)
1 (4)
1 (4)
1 (4)
1 (4)
1 (4)
1 (4)
+1 (8)
+1 (8)
+1 (8)
+1 (8)
+1 (8)
+1 (8)
+1 (8)
+1 (8)
a (2)
+a (10)
0 (6)
0 (6)
0 (6)
0 (6)
0 (6)
0 (6)
0 (6)
0 (6)
0 (6)
6.70
11.20
6.93
7.44
7.01
7.44
6.81
10.94
7.76
11.63
7.06
13.20
6.42
12.00
8.62
12.07
6.06
8.42
7.50
8.50
6.93
13.38
6.78
14.27
9.07
7.90
8.36
1.00
1.33
0.61
0.79
0.99
0.91
0.71
1.21
0.57
1.00
0.72
1.68
0.80
1.60
1.17
1.42
0.42
1.03
0.52
0.80
0.76
1.68
0.98
1.20
1.30
1.16
1.20
53.95
43.33
55.31
48.95
86.96
74.27
65.82
95.13
49.17
56.04
64.05
32.61
37.23
78.27
47.25
40.40
48.23
40.16
48.64
57.63
65.54
74.14
82.39
62.32
84.01
85.03
84.89
Central point of the experimental design (CCD), GAE = gallic acid equivalent, Y% extractg
100.
seedsg
In order to verify the presence of phenolic compounds in the extracts, another TLC methodology was applied. The mobile phase
was ethyl acetate:acetic acid:formic acid:water (100:11:11:26,
v/v/v/v) with the same stationary phases used to identify sugars,
the plate was developed with Neu-PEG (Wagner & Bladt, 1996).
Neu reagent was obtained by mixing 1 g of diphenyl boric acid ethylamino ester in 100 mL of methanol and PEG reagent was a solution of polyethylene glycol (PEG) 4000 at 5% in ethanol.
Moreover, TLC was used in order to determine the antioxidant
activity of compounds. For this analysis, the developer used was
a solution of 0.05% 2,2-diphenyl-1-(2,4,6-trinitrophenyl) hydrazyl
radical (DPPH) in methanol sprinkled on the plate after elution
and drying. The compounds with anti-oxidant properties reacted
with DPPH radical generating white spots on the plate.
The regression coefcients, which showed no statistical signicance (P > 0.05), were not considered in the model, but were added
to the pure error. For the linear model (Eq. (1)), T, ST and the
525
Fig. 1. T and ST RSA for the PFE extract yield for the linear and quadratic models (CCD 24).
526
Fig. 2. Analysis of Brazilian cherry seed extract by TLC. TLC was performed with a Silica Gel 60 F254 plate using Brazilian cherry seed extract as the source material as
described in the Materials and Methods section. (a) Ninhydrin, Molish, and VS1 revelation. The mobile phase was acetonitrile:water (75:25, v/v). 1: Sucrose, 2: Rafnose, 3:
Ext. Ultra sonic (water/methanol/formic acid, 8:0.5:1.5, v/v/v), 4: PFE extracted with water/methanol/formic acid (8:0.5:1.5, v/v/v), 5: PFE extracted with ethanol, 6:
Phloridzin dehydrate, 7: Rutin, 8: Avicularin. (b) Neu-PEG revelation. The mobile phase was ethyl acetate:acetic acid:formic acid:water (100:11:11:26, v/v/v/v), 366 nm
wavelength. 1: Tannins isolated from crude ethanol PFE (5000 ppm), 2: Crude ethanol PFE (5000 ppm), 3: Puried ethanol PFE (10,000 ppm), 5: Puried ethanol PFE
(10,000 ppm), 4, 6 and 7: HPLC F2 of puried ethanol PFE (1000 ppm), (c) and (d): Same mobile phase and samples used in (b) Neu-PEG revelation (track 1, 2, 3, 4 and 5) and
DPPH (0.05%) revelation (track 10 , 20 , 30 , 50 ).
Fig. 3. Components in Brazilian cherry seed extracts detected by HPLC monitored with a UV detector. (a): crude (blue line) and puried (red line) extracts, LiChrospher RP-18
(100 4.6 mm 5 lm), mobile phase water (A) /methanol (B) both acidied with 0.1% formic acid, at 95% A for 5 min, 9550% A in 5 min, 500% A in 20 min, 0% A for 5 min,
095% A in 5 min, 95% A for 10 min, with UV detection at 254 nm, (b): puried extract from Brazilian cherry seeds, LiChrospher C18 (150 4.6 mm 5 lm), mobile phase
water (A) /methanol (B) both acidied with 0.1% formic acid, at 95% A for 5 min, 9550% A in 5 min, 500% A in 20 min, 0% A for 5 min, UV detection at 360 nm (blue line) and
254 nm (red line). (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)
527
the ion at m/z 255, also typical of quercetin (Sandhu and Gu,
2010). The phenolic compound quercetin 3-O-b-rhamnoside may
be suggested in this fraction. Unfortunately, due to inadequate separation of the peaks by HPLC, mass spectrometry was not effective.
Thus, it was decided to separate the peaks by CPC and then perform HPLC/HRMS analysis.
Fig. 4. HPLC chromatogram of the puried ethanol extract of Brazilian cherry seeds (F1 and F2 HPLC fractions) and of its two CPC fractions (CPC F3 tubes 1214, and CPC F6
tubes 4954). Column: LiChrospher C18 (150 4.6 mm 5 lm), with a mobile phase water (A) /methanol (B) both acidied with 0.1% formic acid, at 95% A for 5 min, 9550%
A in 5 min, 500% A in 20 min, 0% A for 5 min, UV detection at 254 nm.
528
Table 2
HPLC/ESI/HRMS data and tentative identication of the main compounds present in the PFE extracts of Brazilian cherry seeds.
HPLC/CPC
tR (min)
UVvis kmax
(nm)
[M-H]
(amu)
MS/MS
[M] formula
Tentative identication
F1/F3
19.3
19.5
253, 357
253, 360
20.3
20.8
22.2
254, 365
253, 367
263, 316, 353
433.04151
463.08849
447.05704
300.99951
447.09391
417.08456
301
301
301
229
300, 301
285
C19H14O12
C21H20O12
C20H16O12
C14H6O8
C21H20O11
C20H18O10
F2/F6
the second peak at 19.5 min, there were two compounds, a derivative of quercetin hexose and an ellagic acid deoxyhexose with the
same retention time and both with the same maximum absorbance (253 and 360 nm). The quercetin hexoside is a avonol glycoside, also an antioxidant compound present in plants (Lin, Chen,
& Harnly, 2008; Mtt, Kamal-Eldin, & Trrnen, 2003). The product ion at m/z 301 is the quercetin aglycone formed by loss of a
hexoside [MH162] (Yu et al., 2008). Ellagic acid deoxyhexoside, another ellagic acid derivative with the parent ion at m/z
447 that was identied as an antioxidant compound in chestnut
(Sanz et al., 2010). For this compound, the product ion at m/z
301 is the ellagic acid aglycone formed by loss of a deoxyhexoside
[MH146]. Even if preliminary analysis of HPLC F1 by HPLC/ESI/
MS showed more than one component in the single chromatographic peak, the most intense ion spectra at m/z 301 can correspond to the quercetin or to ellagic acid structure. So, this
analysis was insufcient to propose a structure for the different
compounds. The detection of these several compounds in the same
peak was only possible thanks to the sensitivity and precision of
the HRMS which allow to detect several [MH] ions at different
m/z that all lead after fragmentation to the m/z 301 product ion.
According to its exact molecular mass obtained by HRMS, the
corresponding molecular formula peak at 20.3 min could be identied as ellagic acid.
In CPC F6 fraction, two compounds were detected at 20.8 and
22.2 min, quercitrin and a kaempferol pentoside (Table 2). Both
compounds have antioxidant properties and are known constituents of plant extracts. In the HPLC/ESI/MS analysis of HPLC F2,
the ion at m/z 301 has a precursor ion [M-H] at m/z 447, indicating that it was derived from the loss of a rhamnoside (a deoxyhexose) [M-H-146]. It could be that the phenolic compound quercetin
3-O-b-rhamnoside or quercitrin with [M-H] ion at m/z 447 is
present in the Brazilian cherry seed PFE extract, which was conrmed by the HRMS analysis (Table 2) and by HPLC/MS analysis
of quercitrin standard in the same conditions.
The avonol kaempferol, also present in plants, has antioxidant
activity. Like the others, this avonol glycoside is also present in
other seeds, such as in green beans (Price, Colquhoun, Barnes, &
Rodhes, 1998) and lotus seeds (Husam et al., 2010). Its presence
in Brazilian cherry seed extracts was identied tentatively by UV
spectrum (263 and 353 nm) and HRMS analysis. The abundant
product ion at m/z 285 is the kaempferol aglycone that lost a
pentoside [M-H-132] (Table 2).
The tentative identication of the composition of Brazilian cherry seed extracts, obtained by PFE with ethanol as solvent, indicated
the presence of avonol and acids conjugated with sugars. These
components have antioxidant properties, which were conrmed
by the high antioxidant activity of these extracts. Furthermore,
the tannins with a high degree of polymerization, removed from
the crude extract, may also have the same action. Experiments
are now being conducted to examine their possible bio-applications. Also in this study, PFE was optimised at laboratory scale; it
will be interesting to demonstrate that this process, can be used
at large scale to obtain extracts rich in bioactive compounds and
to generate products from agro-processing waste, abundant in Brazil. Moreover, ethanol is a GRAS solvent and has easy accessibility
in Brazil which is one of the largest producers.
Acknowledgement
The authors would like to thank FAPESP (State of So Paulo Research Foundation, Brazil) Project 2008/00148-0 for nancial
support.
References
Aaby, K., Ekeberg, D., & Skrede, G. (2007). Characterization of phenolic compounds
in strawberry (fragaria x ananassa) fruits by different HPLC detectors and
contribution of individual compounds to total antioxidant capacity. Journal of
Agricultural and Food Chemistry, 55(11), 43954406.
Bagetti, M., Facco, E. M. P., Rodrigues, D. B., Vizzotto, M., & Emanuelli, T. (2009).
Antioxidant capacity and composition of pitanga seeds. Cincia Rural, 39(8),
25042510.
Berthod, A., Hassoun, M., & Ruiz-Angel, M. (2005). Alkane effect in the Arizona
liquid systems used in countercurrent chromatography. Analytical &
Bioanalytical Chemistry, 383, 327340.
Bezerra, J. E. F, Silva, J. F, Jr, & Lederman, I. E. (2000). Pitanga (Eugenia uniora L.) Srie
Frutas Nativas. Jaboticabal, So Paulo, BR: Funep.
Claude, B., Morin, P., Lafosse, M., Belmont, A.-S., & Haupt, K. (2008). Selective solidphase extraction of a triterpene acid from a plant extract by molecularly
imprinted polymer. Talanta, 75(2), 344350.
Filho, G. L., De Rosso, V. V., Meireles, M. A. A., Rosa, P. T. V., Mercadante, A. Z.,
Oliveira, A. L., et al. (2008). Supercritical CO2 extraction of carotenoids from
pitanga fruits (Eugenia uniora L.). The Journal of Supercritical Fluids, 46(1),
3339.
Guimares, F. A., Holanda, L. F. F., Maia, G. A., & Moura F, J. A. (1982). Estudos
analticos e fsicos em polpa e semente de pitanga (Eugenia uniora L.). Cincia e
Tecnologia de Alimentos, 2, 208215.
Husam, M. K., Huang, D., Xie, B., He, H., Yang, E., Tian, B., et al. (2010). Flavonols of
lotus (Nelumbo nucifera, Gaertn.) seed epicarp and their antioxidant potential.
European Food Research & Technology, 231(3), 387394.
Kaufmann, B., & Christen, P. (2002). Recent extraction techniques for natural
products: Microwave-assisted extraction and pressurised solvent extraction.
Phytochemical Analysis, 13, 105113.
Lhuiller, A., Fabre, N., Moyano, F., Martins, N., Claparols, C., Fourast, I., et al. (2007).
Comparison of avonoid proles of Agauria salicifolia (Ericaceae) by liquid
chromatography-UV diode array detectionelectrospray ionisation mass
spectrometry. Journal of Chromatography A, 1160(12), 1320.
Liang, X.-m, Jin, Y., Feng, J.-t, & Ke, Y.-x. (2011). Identication and structure
elucidation of compounds from herbal medicines. In W. J. H. Liu (Ed.).
Traditional Herbal Medicine Research Methods: Identication, Analysis, Bioassay,
and Pharmaceutical and Clinical Studies (pp. 448). Hoboken, NJ, USA: John Wiley
& Sons.
Lin, L.-Z., Chen, P., & Harnly, J. M. (2008). New phenolic components and
chromatographic proles of green and fermented teas. Journal of Agricultural
and Food Chemistry, 56(17), 81308140.
Mtt, K. R., Kamal-Eldin, A., & Trrnen, R. (2003). High-performance liquid
chromatography (HPLC) analysis of phenolic compounds in berries with diode
array and electrospray ionization mass spectrometric (MS) Detection: Ribes
Species. Journal of Agricultural and Food Chemistry, 51(23), 67366744.
Malaman, F. S., Moraes, L. B., West, C., Ferreira, N. J., & Oliveira, A. L. (2011).
Supercritical uid extracts from the Brazilian cherry (Eugenia uniora L.):
Relationship between the extracted compounds and the characteristic avour
intensity of the fruit. Food Chemistry, 124(1), 8592.
Michel, T., Destandau, E., & Elfakir, C. (2011). On-line hyphenation of centrifugal
partition chromatography and high pressure liquid chromatography for the
fractionation of avonoids from Hippopha rhamnoides L. berries. Journal of
Chromatography A, 1218(9), 61736178.
Oliveira, A. L., Kamimura, E. S., & Rabi, J. A. (2009). Response surface analysis of
extract yield and avour intensity of Brazilian cherry (Eugenia uniora L.)
529