Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)
Article
Identification and quantification of flavonoids from two Southern Italy cultivars of Allium cepa L. Var. Tropea (red onion) and Montoro (copper onion) and their capacity to protect human erythrocytes from oxidative stress Idolo Tedesco, Virginia Carbone, Carmela Spagnuolo, Paola Minasi, and Gian Luigi Russo J. Agric. Food Chem., Just Accepted Manuscript • Publication Date (Web): 12 May 2015 Downloaded from http://pubs.acs.org on May 13, 2015
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 35
Journal of Agricultural and Food Chemistry
1
Identification and quantification of flavonoids from two Southern Italy cultivars of Allium
2
cepa L. Var. Tropea (red onion) and Montoro (copper onion) and their capacity to protect
3
human erythrocytes from oxidative stress
4 5
Idolo Tedesco*, Virginia Carbone*, Carmela Spagnuolo*, Paola Minasi and Gian Luigi Russo§
6 7
Institute of Food Sciences, National Research Council, 83100, Avellino, Italy
8 9
*Equal contribution
10 11
§
12
Dr. Gian Luigi Russo
13
Istituto Scienze dell’Alimentazione – Consiglio Nazionale delle Ricerche
14
Via Roma 64
15
83100 – Avellino
16
Italy
17
Phone: +39 0825 299331
18
Fax: + 39 0825 781585
19
E-mail:
[email protected]
To whom correspondence should be addressed:
20
1 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 2 of 35
21
ABSTRACT
22
Onions (Allium cepa) are consumed worldwide and represent an important source of dietary
23
phytochemicals with proven antioxidant properties, such as phenolic acids, flavonoids,
24
thiosulfinates and anthocyanins. Epidemiological and experimental data suggest that a regular
25
consumption of onions is associated with a reduced risk of degenerative disorders. Therefore, it is
26
of interest to investigate the biological properties of different varieties of onions. Here, we
27
characterized for the first time a variety of onion, called “Ramata di Montoro” (coppery onion from
28
Montoro) grown in a niche area in Southern Italy and compared its phenolic profile and antioxidant
29
properties to a commercial ecotype of red onion, namely Tropea, also present in Southern Italy. An
30
analytical method based on high performance liquid chromatography coupled with UV detector and
31
mass spectrometry was used to separate and characterize the phenolic fraction (anthocyanins and
32
flavonols) extracted from both coppery and red types. The main compounds detected in the two
33
ecotypes were: quercetin and quercetin glucosides, isorhamnetin glucosides, kaempferol glucoside
34
and, among anthocyanins, cyanidin glucosides. Tropea ecotype onion showed a higher content of
35
flavonols (632.82 mg/kg fresh weight) than the Montoro type onion (252.91 mg/kg fresh weight).
36
Accordingly, the antioxidant activity of the former was 2.8-fold higher compared to the latter. More
37
pronounced were the differences existing between the 4 anthocyanins detected in the two ecotypes,
38
with those in the Tropea ecotype onion present at concentrations between 20 and 230-fold higher
39
than in the Montoro type onion. Both extracts reduced LDL oxidation of about 6-fold and protected
40
human erythrocytes from oxidative damage induced by HClO of about 40%. In addition, as a
41
consequence of HClO treatment, GSH concentration in erythrocytes was reduced of about 50% and
42
pre-treatment with onion extracts induced a recovery of GSH level of about 15-22%. Qualitative
43
differences highlighted in the chemical composition of the two phenolic extracts, especially the total
44
content of anthocyanins which was 30-fols higher in Montoro type onion compared to Tropea
45
ecotype can be associated to the protective effects measured against oxidative damage induced in
46
human erythrocytes. 2 ACS Paragon Plus Environment
Page 3 of 35
47
Journal of Agricultural and Food Chemistry
ABBREVIATIONS
48 49
ROS, reactive oxygen species; HOCl, hypochlorous acid; RNS, reactive nitrogen species; COM,
50
Coppery Onion from Montoro; RTO, Red Tropea onion; RBCs, red blood cells; LDL, low density
51
lipoprotein; GSH, glutathione; HPLC, High Performance Liquid Chromatography; QE, equivalent
52
of quercetin; FW, fresh weight;
53 54
3 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
55
Page 4 of 35
INTRODUCTION
56 57
Onions are the second most produced vegetable crop after tomatoes with about 3,642,000 ha grown
58
annually worldwide and a production of 53.6 Mt 1, 2. Their large distribution is probably due to the
59
versatile culinary uses as raw food or different modes of cooking (baked, boiled, braised, grilled,
60
fried, roasted, etc.). In addition, onions are considered among the healthier vegetables due to the
61
high levels of non-nutrient compounds, such as phenolic acids, flavonoids, thiosulfinates and
62
anthocyanins. For these reasons, it is of interest to study and characterize the chemical and
63
biological properties of new varieties, including those grown in niche geographical areas, which are
64
produced in limited amount, but may represent a potential source of new biological compounds.
65
Onions are among the richest sources of dietary flavonoids. In bulbs, only compounds belonging to
66
the flavonols, the anthocyanins, and the dihydroflavonols have been detected. As reviewed 3, yellow
67
onions contain 270–1187 mg/kg FW of flavonols, whereas red onions contain 415–1917 mg/kg FW
68
of flavonols. Quercetin derivatives (quercetin 4′-glucoside and quercetin 3,4′-diglucoside) are the
69
predominant flavonols in all onion cultivars. The anthocyanins of red onions (approximately 10% of
70
the total flavonoid content) are mainly cyanidin glucosides acylated with malonic acid or
71
nonacylated 3. Storage of onions for 6 weeks in different conditions resulted in a decrease of 64-
72
73% of total anthocyanins which paralleled with a reduction in the total antioxidant activity of 29-
73
36% 4. To this regards, it is worthwhile to mention that onions are considered among the major
74
sources of antioxidant compounds 5 whose level of activity is related to the different cultivars and
75
colors 6-10.
76
The scientific literature of the last twenty years or so has been influenced by the syllogism that
77
since free radicals damage cellular structures, their scavenging by antioxidants is health protective.
78
As a consequence, an increased uptake of antioxidants, such as those deriving from a fruits and
79
vegetables-rich diet, may increase resistance to reactive species induced pathologies. Although this
80
concept probably overestimates the antioxidant capacity of phytochemicals and, possibly, generated 4 ACS Paragon Plus Environment
Page 5 of 35
Journal of Agricultural and Food Chemistry
81
misinterpretations in the field 11, 12, biochemical and genetic studies on cellular and animal models
82
on the mechanism(s) of action of phytochemicals provide a functional explanation of how and why
83
a diet rich in fruits and vegetables can protect against degenerative diseases 13, 14.
84
Therefore, scientists in this field are constantly in search of novel agronomic species and/or local
85
varieties possessing enhanced protective effects against oxidative damage in cellular and animal
86
models. In the present work, we focused our attention on a local variety of onion, so-called “Ramata
87
di Montoro” (coppery onion from Montoro area; COM), whose cultivation was born and passed
88
down through generations in the area of Montoro Inferiore, a village located in the province of
89
Avellino (Southern Italy). We compared the antioxidant properties of polyphenols isolated from
90
COM to a better characterized, commercial ecotype, namely red Tropea onion (RTO), a variety of
91
red onion grown in a specific area of Calabria, a region in Southern Italy 15.
92
In the present study, an analytical method based on high performance liquid chromatography
93
(HPLC) coupled with UV detector and mass spectrometry was used to separate and characterize the
94
phenolic fraction (anthocyanins and flavonols) extracted from cultivated bulbs of COM compared
95
to RTO ecotype. Polyphenolic extracts from the two onions were evaluated for their antioxidant
96
activity. Subsequently, we investigated the protective role of phenolic extracts from RTO and COM
97
against low density lipoprotein (LDL) oxidation and hypochlorous acid-induced oxidative damage
98
in normal red blood cells (RBCs) and their interference with the antioxidant defense systems active
99
in erythrocytes. Oxidized LDL contain a diverse set of toxic species, such as aldehydes, oxysterols
100
and lipid peroxides 16. They are considered as an important initial step in the development of
101
atherosclerosis; therefore, lowering LDL levels by independent mechanisms is considered
102
protective against the likelihood of atherosclerotic events 17-19. Previous studies indicated that
103
phytochemicals, such as phenolic acids, flavonoids and anthocyanins, inhibited LDL oxidation 20, 21
104
22, 23
105
exposed to continuous oxidative stress deriving from endogenous24, or exogenous sources 25. For
106
these reasons, RBCs have developed efficient enzymatic and not-enzymatic antioxidant defenses to
. RBCs have been selected as the study model since, for their physiological role, they are
5 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 6 of 35
107
preserve themselves from oxidative damage. In addition, due to their mobility, erythrocytes can be
108
considered as ideal antioxidant scavengers acting throughout the circulation and reducing the
109
damaging mass effect of elevated levels of ROS on different tissues 24-26. Therefore, we established
110
a novel in vitro model using RBCs isolated from whole blood, treated with HClO. Millimolar
111
concentrations of HClO generates osmotic fragility and formation of transient membrane pores with
112
the consequent hemolysis of erythrocytes 27. Treatment with HClO cannot be seen as an
113
experimental adaptation to allow the functionality of our assay on RBCs, but it is very close to
114
physiological processes; in fact, following inflammatory tissue injury and microbial killing,
115
neutrophils and monocytes generate hyper-production of HClO in the millimolar range 28.
116
Based on these observations, we firstly chemically characterized the polyphenolic contents of RTO
117
and COM ecotypes, the latter representing a new variety, while the former was used as a reference,
118
being already described and characterized. Subsequently, we compared their protective effect
119
against oxidative damage taking advantage of two different assays: capacity to reduce LDL
120
oxidation and protection of human RBCs from healthy subjects from oxidative insult caused by
121
HClO treatment.
122
6 ACS Paragon Plus Environment
Page 7 of 35
123
Journal of Agricultural and Food Chemistry
MATERIAL AND METHODS
124 125 126 127
Chemicals.
128
Ciocalteu’s reagent; 2,4,6-Tris(2-pyridyl)-s-triazine (TPTZ); glutathione (GSH); phtaldialdehyde,
129
trichloroacetic acid (TCA); hydrogen peroxide (H2O2); lipoprotein low density from human plasma
130
(LDL); cupric sulfate anhydrous; xylenol orange; ferrous ammonium sulfate; butylated
131
hydroxytoluene (BHT) were purchased from Sigma-Aldrich (Milan, Italy). Isoquercitrin (quercetin-
132
3-O-glucoside) was obtained from Fluka (Buchs SG, Switzerlandand). Cyanidin-3-O-glucoside
133
chloride (Kuromanin chloride); delphinidin-3-O-glucoside chloride (myrtillin chloride); quercetin-
134
3,4'-di-O-glucoside; quercetin-4'-O-glucoside (Spiraeoside); kaempferol-3-O-glucoside were
135
purchased from Extrasynthese (Genay, France). Ferric chloride, sodium carbonate, sodium
136
hypochlorite were from Carlo Erba (Milan, Italy). Phosphate-buffered saline (PBS) tablets were
137
purchased from Invitrogen (S. Giuliano Milanese, Milan, Italy). HPLC grade water (18.2 mΩ) was
138
prepared using a Millipore Milli-Q purification system (Millipore Corp., Bedford, MA, USA). All
139
other chemicals used were of research highest purity grade.
Methanol and formic acid were obtained from Merck (Darmstadt, Germany). Quercetin; Folin
140 141
Onion collection and sample treatment. Cultivated bulbs of onion, Allium cepa L. Tropea
142
ecotype, were collected in Tropea (Calabria, Italy), while cultivated bulbs of brown onion, Allium
143
cepa L. Montoro type, were kindly provided by a local Committee promoters of Cipolla Ramata di
144
Montoro (www.cipollaramatadimontoro.it) located in Montoro (Campania, Italy). The name Cipolla
145
Ramata (coppery onion) derives from the color of the external skin of the bulb. This variety is
146
cultivated in a niche geographical area in Southern Italy (17 areas within Avellino and Salerno
147
districts).
148
For each variety two bulbs were finely chopped and 10 g were extracted with 20 mL of 80%
149
aqueous methanol, at room temperature, on an horizontal shaker. After 10 min samples were placed 7 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
150
in an ultrasonic bath and sonicated for 15 min and then filtered through filter paper. Extracts were
151
then dried in rotary evaporator (LaboRota 4000 /HB Efficient, Heidolph) and stored at -20°C until
152
being used.
Page 8 of 35
153 154
Analysis of total phenolic content and antioxidant capacity. Dry extracts were suspended in PBS
155
to obtain a stock solution of 50 mg/mL (w/v) and kept at 4°C and in the dark. This stock solution
156
was used in all biological assays performed in the present study 29 and for the analysis of total
157
phenolic content and antioxidant capacity. The amount of total phenols was determined according
158
to the Folin-Ciocalteu’s procedure 30, 31 using quercetin as a reference standard. Briefly, 50 L of
159
Folin–Ciocalteu’s reagent and 500 L of distilled water were added to 10 L of suitable aqueous
160
dilution of the stock solution (50 mg/mL) (w/v). The reaction mixture was mixed and allowed to
161
stand for 2 min. Finally, 0.1 mL of sodium carbonate and 0.34 mL of distilled water were added and
162
the solution was incubated in the dark for 120 min. The samples absorbance was measured at 760
163
nm. The results were expressed as micromolar equivalent of quercetin (QE) in 1 mg/mL of onion
164
extracts. All measurements were carried out in triplicates.
165
The antioxidant capacity of different onion extracts were determined by a ferric reducing
166
antioxidant power (FRAP) assay, as reported 32 and results calculated as micromolar QE in 1
167
mg/mL of the two onion extracts.
168
Ascorbic acid (AA) content in onion extracts was determined by FRASC method 33. Samples were
169
treated with 10 IU ascorbate oxidase/mL, diluted in FRAP solution (20 mM FeCl3 and 10 mM
170
TPTZ in acetate buffer, pH 3.5) and finally incubated for 5 min at 37°C. The absorbance was
171
measured at 595 nm and micromolar concentrations of AA were calculated using a standard curve.
172 173
HPLC-UV analyses. Extracts from cultivated bulbs of the two different onion types were
174
reconstituted in 1% formic acid and analysed by HPLC-UV/Vis using a HP 1110 Series HPLC
175
(Agilent, Palo Alto, CA, USA) equipped with a binary pump (G-1312A) and an UV detector (G8 ACS Paragon Plus Environment
Page 9 of 35
Journal of Agricultural and Food Chemistry
176
1314A). Individual phenols were separated on a Hypersil BDS C18 column (250 mm x 4.6mm, 5
177
µm) (Thermo, Bellefonte, PA, USA) at a flow rate of 1 mL/min; solvent A was 1% formic acid and
178
solvent B was 1% formic acid in methanol and water (50:50, v/v). After a 5 min hold at 20%
179
solvent B, elution was performed according to the following conditions: from 20% (B) to 80% (B)
180
in 22 min, isocratic elution (80% B) for the next 13 min, from 80% (B) to 95% (B) in 10 min,
181
followed by 15 min of maintenance. Flavonols were monitored at 340 nm while anthocyanins at
182
520 nm. Standard curves for each flavonol standard were prepared over a concentration range of 1–
183
180 µg/mL with six different concentration levels and triplicate injections at each level.
184
Quantification of anthocyanins was performed with external calibration curves generated by
185
repeated injections of a fixed volume of standard solutions of cyanidin-3-O-glucoside over a
186
concentration range of 0.5–20 µg/mL with five different concentration levels and duplicate
187
injections at each level. All samples were prepared and analysed in duplicate. Results were
188
expressed as mg/kg of fresh weight (FW).
189 190
HPLC-ESI-ITMSn analysis. Identification of phenolic compounds in the extracts from cultivated
191
bulbs of the two different onion ecotypes was effectuated by HPLC–ESI/MS and MSn
192
fragmentation analyses using a SURVEYOR MS micro HPLC (Thermo Finnigan, San Josè, CA,
193
USA) coupled with a Finnigan LCQ DECA XP Max ion trap mass spectrometer (Thermo Finnigan,
194
San Josè, CA, USA), equipped with Xcalibur® system manager data acquisition software (Thermo
195
Finnigan, San José, CA, USA). Individual phenols were separated on a Hypersil BDS C18 column
196
(250 mm x 2.1 mm, 5 µm) (Thermo, Bellefonte, PA, USA) at a flow rate of 200 µL/min. The HPLC
197
conditions were as described for the HPLC-UV system. Mass spectra were recorded from mass-to-
198
charge ratio (m/z) 50 to 1200 both in negative and in positive ionization mode. The capillary voltage
199
was set at -13 V, the spray voltage was at 4.5 kV and the tube lens offset was at -15 V in negative
200
ion mode while, in positive ion mode, the capillary voltage was set at 38 V, the spray voltage was at
9 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
201
3 kV and the tube lens offset was at 50 V. The capillary temperature was 275°C. Data were
202
acquired in MS, MS/MS and MSn scanning mode.
Page 10 of 35
203 204
LDL oxidation. The assay was performed incubating 0.1 mL of LDL (200 g protein) with or
205
without onion extracts (0.25 mg/mL; w/v) for 15 min, then oxidation was induced by adding 50 M
206
CuSO4 at 37°C for 15 h. Subsequently, 0.9 mL of FOX reagent (1 mM xylenol orange, 2.5 mM
207
ferrous ammonium sulfate and 4.4 mM BHT in methanol) were added to the samples. After 30 min,
208
samples were centrifuged in a microfuge at maximal speed for 5 min, and absorbance of the
209
supernatants was detected spectrophotometrically at 560 nm. Values were expressed as hydrogen
210
peroxide equivalents 30, 34.
211 212
Blood samples. Whole blood samples were obtained from 16 healthy donors, collected in EDTA-
213
treated tubes and immediately used. All blood samples were obtained after informed consent.
214
Erythrocytes were isolated by centrifugation, 2000xg for 15 min, to remove plasma, platelets and
215
buffy coat and washed twice in PBS. Erythrocyte pellets were suspended in PBS and aliquots of 1.2
216
x105 cells/L were used in all experiments presented.
217 218
Reaction of human erythrocytes with HOCl and hemolysis assay. NaOCl was diluted with PBS,
219
and the pH of the solution was adjusted to 7.4 immediately before use. At this pH, the solution
220
contains approximately equimolar amounts of HClO (hypochlorous acid) and NaOCl, and is
221
referred to hereafter as HClO 35.
222
Diluted erythrocytes in PBS (1.2 x105 cells/L) were pre-treated with onion extracts at the indicate
223
concentrations for 45 min and then added with 0.3 mM of HClO for 15 min at 37°C. After
224
incubation with HClO, samples were centrifuged at 2000x g for 2 min and the percent of hemolysis
225
determined spectrophotometrically at 540 nm 24. 10 ACS Paragon Plus Environment
Page 11 of 35
Journal of Agricultural and Food Chemistry
226 227
GSH measurement and catalase assay. In parallel with hemolysis assay, GSH concentration and
228
catalase enzymatic activity were measured. GSH determination was evaluated by a
229
spectrofluorimetric method. Following treatments, erythrocyte pellets were washed with PBS and
230
then proteins were precipitated using TCA (5.0% v/v final concentration in 0.1 M HCl, 10 mM
231
EDTA). Fluorescence of the supernatants was measured at 340 nm (excitation wavelength) and 460
232
nm (emission wavelength). GSH concentration was calculated from a standard curve and refereed
233
as percentage compared to controls (untreated samples) 24.
234
To determine catalase activity, after treatment, erythrocytes were diluted with cold distilled water,
235
ethanol and chloroform (1.5:1:1, v/v) to precipitate the hemoglobin. Samples were shaken
236
vigorously and centrifuged at 13000xg for 3 min and the water-ethanol layer was used. Catalase
237
activity was evaluated as the first-order kinetic constant of the rate of disappearance of H2O2,
238
measured by absorbance at 240 nm, as reported previously 36. Values were expressed as specific
239
activity of the samples and referee as percentage of the control (untreated sample).
240 241
Statistical Analysis. Data are presented as mean values ± standard deviation (s.d.). For LDL
242
oxidation, hemolysis assay, GSH measurement, catalase assay, concentration of individual and total
243
polyphenolics determined by HPLC significance was measured using Student’s t-test. Number of
244
determinations are as reported in figure legends.
245
11 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
246
Page 12 of 35
RESULTS AND DISCUSSION
247 248
Total polyphenol content and antioxidant activity. Total phenolic content of onion extracts
249
indicate that the highest content of total polyphenols was found in Tropea onion ecotype (29.35 M
250
QE). They were 2.3-fold higher than those present in Montoro onion (12.68 M QE) (Table 1). As
251
expected, considering the result of the Folin-Ciocalteu assay, the antioxidant capacity was
252
proportional to the polyphenol content, with RTO showing the highest values which resulted about
253
3-fold higher than COM extract (Table 1).
254
While for COM no previous data were reported in the literature on its total polyphenol content and
255
antioxidant activity to allow a comparison with our results, the information on red onion ecotypes
256
are largely available. Although the use of different protocols of extractions complicates the
257
comparison among data from different articles, our determinations are comparable to those reported
258
by others. As an example, Marotti and Piccaglia 37 measured a total amount of flavonoids in the
259
ecotype Tropea rossa tonda of 762.9 mg/kg, not far from 632.82 mg/kg of total flavonols reported
260
in Table 3. In three red ecotypes from Spain, total flavonols ranged between 211.7-304.3 mg/kg,
261
lower than Tropea onion, but in the same order of magnitude 2. The comparative evaluation of the
262
antioxidant capacity appears more complicated, since the different methods and units used to
263
measure it. However, in general, red onion ecotypes present higher values of antioxidant capacity
264
independently from the scale of color. A good example in this sense is reported by a Turkish group
265
who demonstrated that, in 14 different cultivars differing in color, the red onions had higher
266
antioxidant activities than yellow and white onions although the formers had the richest phenolic
267
contents 15.
268 269
Identification and quantification of polyphenols in the two different onion types extracts.
270
HPLC-UV/Vis chromatograms of extract from cultivated bulbs of Allium cepa L. Montoro type are
271
shown in Figure 1. Nine flavonols and four anthocyanins were identified and quantified (Table 2 12 ACS Paragon Plus Environment
Page 13 of 35
Journal of Agricultural and Food Chemistry
272
and 3). The main flavonols found in this onion type were quercetin-3,4'-di-O-glucoside and
273
quercetin-4'-O-glucoside (spiraeoside), which accounted for ∼85% of the total flavonol content.
274
Results obtained from the HPLC-UV/Vis and ESI–ITMSn analyses of Tropea ecotype onion and the
275
total phenolic content measured by HPLC-UV/Vis (expressed as mg/kg of FW) were also reported
276
in Table 3. RTO showed a higher content of flavonols (632.82 mg/kg FW) than COM (252.91
277
mg/kg FW). Only kaempferol-3-O-glucoside was present in the COM and absent in RTO. More
278
pronounced were the differences existing between the 4 anthocyanins detected in the two ecotypes,
279
with those in RTO present at concentrations between 20 (compound 12 in Table 3) and 230
280
(compound 10 in Table 3)-fold higher than in COM. All these differences, including the significant
281
difference in the anthocyanin content, may justify the high level of antioxidant capacity measured in
282
RTO (Table 1).
283
As to anthocyanins, both RTO and COM contained cyanidin glucosides, acylated or not with
284
malonic acid (Table 2 and 3), which are known to be present in various cultivars of red onions 3, 38,
285
39
286
our analyses did not confirm these results. In fact, the HPLC analysis at 340 nm of RTO,
287
specifically used for monitoring flavonols, showed the presence of two very intense peaks at tR
288
25.35 min and tR 32.51 min and minor peaks at tR 26.72 min and tR 36.01 min (data not shown).
289
Compounds present in these fractions produced molecular ions at m/z 627, 465, 641 and 479
290
respectively, in ESI–ITMSn analysis carried out in positive ion mode, and were identified as
291
flavonols quercetin-3,4'-diglucoside, quercetin-4-O-glucoside, isorhamnetin-3,4'-diglucoside and
292
isorhamnetin-4'-glucoside. In addition, the ESI–ITMSn analysis of these fractions carried out in
293
negative ion mode further confirmed the presence of these flavonols (Table 3). In contrast,
294
compounds originating the same m/z values (m/z 627, 465, 641 and 479) were previously identified
295
as delphinidin and petunidin derivatives3, 4. It is important to note that quercetin and delphinidin as
296
well as isorhamnetin and petunidin displayed the same m/z value in the mass spectrum in positive
297
ESI mode (m/z 303 pseudomolecular ion [M+H]+ for quercetin and protonated molecular ion M+
. It has been reported the presence of delphinidin and petunidin derivatives in RTO 3, 4. However,
13 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 14 of 35
298
for delphinidin; m/z 317 pseudomolecular ion [M+H]+ for isorhamnetin and protonated molecular
299
ion M+ for petunidin). Moreover quercetin-3,4'-diglucoside, quercetin-4-O-glucoside, isorhamnetin-
300
3,4'-diglucoside and isorhamnetin-4'-glucoside originated MS/MS spectra in positive ion mode very
301
similar to those of delphinidin (glucosyl-glucoside), delphinidin-3-glucoside, petunidin (glucosyl-
302
glucoside) and petunidin glucoside, respectively, so that, in this case, tandem mass spectrometry
303
does not lead to a definite discrimination of these compounds and accurate HPLC data are crucial to
304
achieve a conclusive result. In order to rule out the presence of at least delphinidin-3-glucoside, the
305
corresponding standard was analyzed by HPLC-UV/Vis under the reported experimental condition.
306
In agreement with the elution order reported by Wu and Prior (2005)40, the retention time of this
307
compound (tR 22.64 min) was lower than that of cyanidin 3-glucoside (tR 25.88) and of any peak in
308
our HPLC chromatogram (Figure 2), thus, definitively ruling out the presence of delphinidin-3-
309
glucoside in RTO.
310 311
Effect of onion extracts on oxidized-LDL. To confirm the antioxidant role of onion extracts and to
312
highlight potential differences between them, we evaluated their ability to reduce the oxidative state
313
of LDL (Figure 3). We incubated commercial LDL with 50 M Cu2+ for 15 h to obtain the
314
oxidized-LDL. Pre-treating isolated LDL for 30 min with 0.25 mg/mL (w/v) of onion extracts
315
resulted in a strong and significant reduction in oxidized LDL (Figure 3). The extracts deriving
316
from COM and RTO were tested at the same final concentration of 0.25 mg/mL, corresponding to
317
3.1 M QE and 7.2 M QE, respectively. Therefore, although the relative polyphenolic amounts of
318
COM was 2.3-fold lower than RTO, they exerted a similar and significant protective effect when
319
compared to oxidized-LDL (Cu2+ bar in Figure 3), while the difference between RTO + Cu2+ and
320
COM+ Cu2+ time points was not significant. These data suggest that the higher antioxidant capacity
321
of RTO, emerging from Table 1, does not reflect an increased protective activity against LDL
322
oxidation compared to COM.
323 14 ACS Paragon Plus Environment
Page 15 of 35
Journal of Agricultural and Food Chemistry
324
Protective effect of onion extract against erythrocyte oxidative damage. Based on the previous
325
results, we hypothesized that the antioxidant activity of the two onion extracts (Table 1 and Figure
326
3) could be protective in maintaining redox homeostasis in human erythrocytes. Based on this
327
model, we stimulated RBCs isolated from 16 healthy donors, with 0.3 mM HClO to obtain an
328
hemolysis of about 25% (Figure 4). Pre-treating with 0.25 mg/mL (w/v; corresponding to 3.1 M
329
QE COM and 7.2 M QE RTO) of the two onion extracts for 45 min resulted in a significant
330
protective effect against HClO-induced hemolysis of 15 and 17% for RTO and COM, respectively
331
(Figure 4A). The selected concentration applied (e.g., 0.25 mg/mL) was not hemolytic per se;
332
moreover, from previous data we deduct that the total concentration of polyphenols applied in our
333
assay was in the same order of magnitude as that reachable in vivo following the ingestion of an
334
amount of onion enough to bring the blood concentration of quercetin near to 1 M 41-43.
335
It is worthwhile to note that also in this case, as above for LDL oxidation, the two onion types
336
induced a similar protective effect, even though the 2.31- and 2.86-fold higher amount of
337
polyphenolic content and antioxidant capacity in RTO compared to COM, respectively (Table 1).
338
To better clarify this point, we treated RBCs with equal concentrations of polyphenols (final
339
concentration 2.7 M QE). In this case, the total amount of polyphenols expressed as QE in our
340
assays was of 0.963 g for RTO and 0.978 g for COM. As reported in Figure 4B, we observed
341
that COM protected significantly RBCs against HClO-induced hemolysis, while the limited
342
reduction observed for RTO was not significant compared to the control (HClO bar). Mathematical
343
elaboration of data presented in Figure 4A and 4B (data not shown) indicates that, to increase by
344
1% the protective effect of RTO against HClO, 0.208 µg QE of extract are necessary. In the case of
345
COM, only 0.081 µg QE (2.5-fold less) are sufficient. This calculation further demonstrates the
346
higher efficacy of COM extract compared to RTO.
347
In addition, we can exclude that the presence of AA could contribute to the protection from HClO
348
damage shown in Figure 4. In fact, measuring by FRASC assay the AA amount present in 0.25 15 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 16 of 35
349
mg/mL of COM and RTO extracts, we observed that AA was null in RTO and present at a
350
concentration of 0.277 M in COM. When we repeated the assay in the presence AA in a wide
351
range (0.11-0.55 M), no reduction of HClO dependent hemolysis was observed (data not shown).
352 353
Effects of onion extracts on antioxidant defenses in human erythrocytes. RBCs excellently
354
regulate intracellular oxidative stress; in fact, through the combined activities of GSH peroxidase,
355
catalase, SOD and shunt of the hexose monophosphate they are able to protect themselves against
356
ROS44. We measured GSH level in RBCs isolated from 16 healthy donors, pretreated with 0.25
357
mg/mL of COM and RTO extracts (corresponding to 3.1 M and 7.2 M QE, respectively) for 45
358
min before HClO addition. As a consequence of HClO treatment, we observed an about 50%
359
reduction in GSH concentration present in RBCs. Pre-treatment with onion extracts induced a
360
recovery of GSH level of about 22 and 15% for RTO and COM, respectively (Figure 5A). Data
361
were significant for the former and the trend of increased production of GSH was confirmed in
362
COM. Similarly to diseases associated with increased production of ROS, the presence of a strong
363
electrophile, as HClO, causes a rapid consumption of GSH in RBCs 24, 45 and restoration of the
364
normal erythrocyte GSH concentration has been shown to have positive therapeutic effects45, 46. The
365
protective role of onion extracts against oxidative stress was also confirmed by their effect on
366
catalase activity which increases of 2.3-fold compared to basal level after HClO treatment and
367
decreases after pretreatment with the two onion extracts (Figure 5B).
368
We hypothesized that treatment of whole human blood with a HClO concentration not far from that
369
reachable in vivo following an inflammatory process, induces large hemolysis in RBCs caused by a
370
burst of intracellular ROS. Pre-treatment of whole blood with onion extracts generates a protective
371
effect against HClO-induced hemolysis. We postulated that the bioactive compounds in RTO and
372
COM extracts trigger downstream events leading to increased GSH concentration which, in turns,
373
protects erythrocytes from the subsequent oxidative damage caused by HClO (Figure 5 A and B). 16 ACS Paragon Plus Environment
Page 17 of 35
Journal of Agricultural and Food Chemistry
374
How are the molecules present in the extracts responsible for the de novo GSH synthesis is actually
375
under investigation in our laboratory. A model for GSH synthesis and turnover in the human
376
erythrocyte has been recently published and represents the starting point for future studies 47.
377
We described for the first time to our knowledge the flavonoid composition of a new ecotype of red
378
onion, the coppery onion, cultivated in a well limited area of about 40 ha within the districts of
379
Avellino and Salerno in Southern Italy. We also characterize the protective effects of COM against
380
oxidative stress experimentally induced on LDL and RBCs. COM was compared to a red onion
381
variety, i.e., RTO selected for its close geographic origin (Southern Italy), its high concentration of
382
polyphenols 15, 37 and larger commercial distribution. From our study an important observation
383
emerged: although total polyphenol concentration and single flavonoid compounds were both
384
significantly higher in RTO compared to COM, the biological activity was comparable, or even
385
more pronounced in COM. This result led us to formulate the following hypotheses: 1. different
386
qualitative composition in the flavonol and anthocyanin families in the two ecotypes; 2. synergistic
387
associations within the pool of flavonoids in COM resulting in an enhanced antioxidant capacity
388
compared to RTO; 3. polyphenols in red and coppery onions are not directly responsible for the
389
protective, antioxidant effect which, in turns, can be attributed to compounds of unknown chemical
390
structure present in relatively comparable amounts in the two extracts; 4. it may be possible that, at
391
least for RBCs, the protective effect of onion extracts resulting in the reconstitution of depleted
392
intracellular GSH pools, occurs towards mechanisms independent from the antioxidant activity of
393
the extracts and involving specific binding to intracellular targets. Hypotheses 3-4 require intense
394
experimental efforts, currently in progress, to be verified.
395
We focused our attention on the different polyphenolic composition between the two ecotypes. Data
396
reported in Tables 2 and 3 indicate that in COM anthocyanins are almost absent (about 0.46% of the
397
total), compared to those present in RTO (4.54%), a difference of about one order of magnitude.
398
This may suggest that in Figure 4B, when we assayed comparable amount of QE, the relative
399
abundance of anthocyanins in RTO could counteract the antioxidant effect of the flavonols present. 17 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 18 of 35
400
However, initial attempts to identify a cause-effect relationship between the ratio
401
flavonols/anthocyanins and protection from oxidative damage in erythrocytes did not provide
402
conclusive answers.
403
Perhaps, we are facing a situation not totally novel regarding the mechanisms of action of phenolic
404
compounds. In fact, several cases exist in the literature demonstrating that flavonoids may exert
405
their biological activities independently from the capacity to scavenge ROS. Quercetin represents a
406
perfect example in this context, as we largely demonstrated in leukemic cell lines resistant to
407
apoptotic stimuli 48-50.
408
In summary, the present work reports for the first time the chemical characterization of polyphenols
409
present in a new variety of onion from Southern Italy and suggests that the phenolic-enriched
410
extract prepared from this ecotype exerts a protective effect against oxidative damage in LDL and
411
human erythrocytes. We also highlighted the possibility that synergistic combination of specific
412
phenolic compounds and/or the contribution of uncharacterized molecules may explain the
413
quantitative differences in biological activities measured when extracts deriving from different
414
onion varieties are compared.
415
18 ACS Paragon Plus Environment
Page 19 of 35
416
Journal of Agricultural and Food Chemistry
AKNOWLEDGEMENTS
417 418
We gratefully thank the Committee promoters of Montoro coppery onion (Comitato dei Promotori
419
della IGP Cipolla Ramata di Montoro; www.cipollaramatadimontoro.it) for providing us with
420
Montoro type onions. This work was partially supported by a grant from the Italian Ministry of
421
Economy and Finance to the National Research Council for the project "Innovazione e Sviluppo del
422
Mezzogiorno - Conoscenze Integrate per Sostenibilità ed Innovazione del Made in Italy
423
Agroalimentare - Legge n. 191/2009” and by Regione Campania in the framework of the “Rete di
424
Spettrometria di Massa della Campania” (RESMAC).
425
19 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
426
Page 20 of 35
FIGURE LEGENDS
427 428
Figure 1. HPLC chromatograms of extract from cultivated bulbs of Allium cepa L. Montoro type
429
recorded at 340 nm (A) and 520 nm (B). Peaks are labelled according to Table 2.
430 431
Figure 2. HPLC chromatograms of extract from cultivated bulbs of Allium cepa L. Tropea ecotype
432
recorded at 520 nm: a, Cyanidin 3-glucoside; b, Cyanidin-3-laminaribioside (Cyanidin 3-
433
glucosylglucoside); c, Cyanidin 3-(6’’-malonylglucoside); d, Cyanidin-3-malonyl-laminaribioside
434
(Cyanidin3-malonylglucosylglucoside) (Panel A) and of delphinidin-3-O-glucoside standard (de-3-
435
O-glu) (Panel B).
436 437
Figure 3. Effect of onion extracts on LDL oxidation. LDL were incubated for 30 min in the
438
presence of 0.25 mg/mL (w/v) of the two different onion extracts before addition of 50 M of Cu2+
439
for 15 h. At the end of treatment, FOX assay was performed to determinate the level of H2O2
440
generated following LDL oxidation. Bar graphs represent the mean ± s.d. Experiments were carried
441
out in duplicates. Symbols indicate significance: p
