Toxicologically Relevant Aldehydes Produced during the Frying

Subscriber access provided by Weizmann Institute of Science

Article

Toxicologically Relevant Aldehydes Produced during Frying Process are Trapped by Food Phenolics Rosario Zamora, Isabel Aguilar, Michael Granvogl, and Francisco J. Hidalgo J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.6b02165 • Publication Date (Web): 20 Jun 2016 Downloaded from http://pubs.acs.org on June 28, 2016

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 31

Journal of Agricultural and Food Chemistry

1

Toxicologically Relevant Aldehydes Produced during Frying Process

2

are Trapped by Food Phenolics

3

Rosario Zamora, Isabel Aguilar, Michael Granvogl, and Francisco J. Hidalgo*,

4

5

Utrera km 1, Campus Universitario – Edificio 46, 41013-Seville, Spain

6

7

Straβe 34, D-85354 Freising, Germany

Instituto de la Grasa, Consejo Superior de Investigaciones Científicas, Carretera de

Lehrstuhl für Lebensmittelchemie, Techniche Universität München, Lise-Meitner-

8

9

10

11

12

*Corresponding author:

Francisco J. Hidalgo

13

Phone:

+34954611550

14

Fax:

+34954616790

15

e-mail:

[email protected]

1 ACS Paragon Plus Environment


Page 2 of 31

16

ABSTRACT

17

The lipid-derived carbonyl trapping ability of phenolic compounds under common food

18

processing conditions was studied by determining the presence of carbonyl-phenol

19

adducts in both onions fried in the laboratory and commercially crispy fried onions.

20

Four

21

crotonaldehyde, or (E)-2-pentenal were prepared and characterized by 1H and

22

nuclear magnetic resonance (NMR) spectroscopy and high performance liquid

23

chromatography coupled to high resolution mass spectrometry (HPLC-HRMS). The

24

synthesized compounds were 2-(3,4-dihydroxyphenyl)-3,5,8-trihydroxy-9,10-dihydro-

25

4H,8H-pyrano[2,3-f]chromen-4-one (4), 2-(3,4-dihydroxyphenyl)-3,5,8-trihydroxy-10-

26

methyl-9,10-dihydro-4H,8H-pyrano[2,3-f]chromen-4-one (5), 2-(3,4-dihydroxyphenyl)-

27

3,5-dihydroxy-8-methyl-4H,8H-pyrano[2,3-f]chromen-4-one

and

2-(3,4-

28

dihydroxyphenyl)-8-ethyl-3,5-dihydroxy-4H,8H-pyrano[2,3-f]chromen-4-one

(10).

29

When onions were fried in fresh rapeseed oil spiked with acrolein, crotonaldehyde, and

30

(E)-2-pentenal (2.7 µmol/g of oil), adduct 10 was the major compound produced and

31

trace amounts of adducts 4 and 5, but not of adduct 9, were also detected. In contrast,

32

compound 4 was the major adduct present in commercially crispy fried onions.

33

Compound 10 was also present to a lower extent, and trace amounts of compound 5, but

34

not of compound 9, were also detected. These data suggested that lipid-derived

35

carbonyl-phenol adducts are formed in food products under standard cooking

36

conditions. They also pointed out to a possible protective role of food polyphenols,

37

which might contribute to the removal of toxicologically relevant aldehydes produced

38

during deep-frying, assuming that the formed products are stable during food

39

consumption in the human organism.

carbonyl-phenol

adducts

produced

between

quercetin


(9),

and

acrolein, 13

C

Page 3 of 31


40

KEYWORDS: Aldehydes, Carbonyl-phenol reactions, Deep-frying, Fried onions,

41

Lipid oxidation, Quercetin, Reactive carbonyls

42



43

INTRODUCTION

44

Lipid oxidation is a major concern during food production because it causes potential

45

safety problems and consumer rejection as a consequence of the changes in flavor,

46

texture, appearance, and nutritional quality of food products.1–3 All these changes are

47

based on numerous complex interrelated reactions induced by oxygen in the presence of

48

initiators.4 Although these reactions also occur at low temperatures, reaction rates

49

increase with reaction times and temperatures. For that reason, the nature and

50

concentrations of the compounds present in frying oils as a consequence of its thermal

51

degradation during heating have been a matter of great interest for a long time, as some

52

of these compounds are toxicologically relevant and can be ingested directly from the

53

degraded oil or via the fried food.5,6 To this respect, recent studies of Granvogl et al.

54

found significant amounts of short chain toxicologically relevant aldehydes, namely 2-

55

propenal and (E)-2-butenal (acrolein and crotonaldehyde, respectively) in both frying

56

oils and fried foods. Acrolein was present to a higher extent compared to

57

crotonaldehyde and the amounts of these compounds were much higher in the frying

58

oils than in the fried foods.7–9 Thus, for example, in rapeseed oil heated at 180 ºC, up to

59

63.2 mg of acrolein/kg of oil and 27.5 mg of crotonaldehyde/kg of oil were found,

60

whereas the amounts of acrolein and crotonaldehyde present in potato chips ranged

61

from 23 to 26 µg/kg and from 12 to 25 µg/kg, respectively, depending on the oil used.8

62

The fact that the amount of lipid-derived aldehydes in fried foods is reduced

63

compared to the content of these aldehydes in the frying oils might be the consequence

64

of a limited absorption of these compounds by the foods, but also of a reaction of the

65

generated aldehydes with other food components. Among these components, the

66

formation of carbonyl-phenol adducts has recently been suggested,10 which would


Page 4 of 31

Page 5 of 31


67

imply a mostly unknown protective role of phenolic compounds in the elimination of

68

these toxicologically relevant aldehydes, assuming the stability of the formed adducts.

69

In an attempt to investigate whether the formed aldehydes in frying oils are

70

incorporated into food products in the form of carbonyl-phenol adducts, this study

71

aimed at the identification of this kind of adducts in onions submitted to frying in

72

rapeseed oil containing acrolein, crotonaldehyde, and (E)-2-pentenal as well as in

73

commercially crispy fried onions. These aldehydes were selected because of both their

74

confirmed formation in oils submitted to frying7–9 and the fact that the reaction

75

mechanism of (E)-2-alkenals with phenolic compounds is now well understood.10 In

76

addition, onion was chosen as a model food because its flavonoid composition and

77

content is well known.11–15 Thus, total flavonoids in red onions are about 140 mg/100 g

78

of fresh weight from which quercetin derivatives are about 85% of total flavonoids. The

79

main quercetin derivative is quercetin 3,4’-O-diglucoside (77.1 mg/100 g) followed by

80

quercetin 4’-O-glucoside (33.8 mg/100 g), whereas the quercetin content is much lower

81

(about 1.31 mg/100 g). However, diglucosides can be hydrolyzed upon maceration,

82

among other processes, and the amount of monoglucosides and free quercetin increases.

83

During the frying process, the quercetin content is reduced by about 20%.12,13

84

MATERIALS AND METHODS

85

Safety. Acrolein and crotonaldehyde are hazardous and should be handled carefully.

86

Food Samples. Rapeseed oil, red onions (Allium cepa L.), and commercially crispy

87

fried onions were purchased at local supermarkets.

88

Chemicals. Acrolein, crotonaldehyde, (E)-2-pentenal, and quercetin were obtained

89

from Sigma-Aldrich (St. Louis, MO) and were of the highest available quality.

90

Sephadex LH-20 was obtained from GE Healthcare Europe (Freiburg, Germany). All



91

other chemicals were purchased from Sigma-Aldrich, Fluka (Buchs, Switzerland), or

92

Merck (Darmstadt, Germany).

93

Preparation of Aldehyde-Quercetin Adducts as Reference Compounds. Keeping

94

in mind that quercetin is the major flavonoid present in onions,12,13 the corresponding

95

adducts between quercetin and acrolein, crotonaldehyde, or (E)-2-pentenal were

96

synthesized and characterized by 1H and

97

spectroscopy and high performance liquid chromatography coupled to high resolution

98

mass spectrometry (HPLC-HRMS). The chemical structures of the isolated compounds

99

are illustrated in Figure 1. These compounds were prepared by dissolving quercetin

100

(200 µmol) and the corresponding aldehyde (400 µmol) in methanol (2 mL; containing

101

290 µmol of triethylamine). The solution was heated under nitrogen at 100 ºC for 3.5 h

102

(acrolein), 10 h (crotonaldehyde), or 96 h ((E)-2-pentenal), respectively. At the end of

103

the heating time, the reaction mixture was fractionated on a Sephadex LH-20 column

104

using methanol/water (80/20, v/v) as eluent at a flow rate of 15 mL/h. Eluted products

105

were detected by mass spectrometry using direct injection.

106

13

C nuclear magnetic resonance (NMR)

2-(3,4-Dihydroxyphenyl)-3,5,8-trihydroxy-9,10-dihydro-4H,8H-pyrano[2,3-

107

f]chromen-4-one (4). HRMS, m/z 357.06154 (M+ – 1), error 0.2 ppm. 1H NMR (500

108

MHz, DMSO-d6): δ 1.97 (br, 2, H9), 2.88 (br, 2, H10), 3.38 (br, OH), 5.58 (br, 1, H8),

109

6.18 s (s, 1, H6), 6.90 (d, 1, J = 8.5 Hz, H5’), 7.61 (dd, 1, J = 2.1 Hz, J = 8.5 Hz, H6’),

110

7.73 (d, 1, J = 2.1 Hz, H2’), 9.47 (br, OH), 12.30 (br, OH).

111

DMSO-d6): δ 14.85 (C10), 26.87 (C9), 93.29 (C8), 98.89 (C6), 101.26 (C13), 104.20

112

(C11), 115.32 (C2’), 116.18 (C5’), 120.43 (C6’), 122.64 (C1’), 136.56 (C3), 145.64

113

(C3’), 147.26 (C2’), 148.27 (C4’), 153.10 (C14), 158.59 (C12), 158.78 (C5), 176.51

114

(C4).


13

C NMR (125.7 MHz,

Page 6 of 31

Page 7 of 31


115

2-(3,4-Dihydroxyphenyl)-3,5,8-trihydroxy-10-methyl-9,10-dihydro-4H,8H-

116

pyrano[2,3-f]chromen-4-one (5). HRMS, m/z 371.07763 (M+ – 1), error 0.9 ppm. 1H

117

NMR (500 MHz, DMSO-d6): δ 1.38 (d, 3, J = 7.0 Hz, H1”), 1.94 (br, 2, H9), 3.38 (br,

118

OH), 3.41 (br, 1, H10), 5.50 (br, d, 1, J = 7.9 Hz, H8), 6.18 s (s, 1, H6), 6.91 (d, 1, J =

119

8.5 Hz, H5’), 7.60 (dd, 1, J = 2.2 Hz, J = 8.5 Hz, H6’), 7.73 (d, 1, J = 2.2 Hz, H2’), 9.52

120

(br, OH), 12.38 (br, OH).

121

(C1’), 56.49 (C9), 92.76 (C8), 98.64 (C6), 104.38 (C11), 105.74 (C13), 115.27 (C2’),

122

116.20 (C5’), 120.25 (C6’), 122.61 (C1’), 136.56 (C3), 145.71 (C3’), 147.31 (C2),

123

148.27 (C4’), 153.44 (C14), 158.88 (C12), 159.16 (C5), 176.51 (C4).

13

C NMR (125.7 MHz, DMSO-d6): δ 19.02 (C10), 21.74

124

2-(3,4-Dihydroxyphenyl)-3,5-dihydroxy-8-methyl-4H,8H-pyrano[2,3-f]chromen-4-

125

one (9). HRMS, m/z 353.0625 (M+ – 1), error 4.8 ppm. 1H NMR (500 MHz, DMSO-d6):

126

1.42 (d, 3, J = 6.6 Hz, H1”), 3.36 (br, OH), 5.15 (m, 1, H8), 5.83 (dd, 1, J = 3.2 Hz, J =

127

10.0 Hz, H9), 6.23 s (s, 1, H6), 6.87 (dd, 1, J = 1.1 Hz, J = 10.0 Hz, H10), 6.91 (d, 1, J

128

= 8.5 Hz, H5’), 7.62 (dd, 1, J = 2.2 Hz, J = 8.5 Hz, H6’), 7.73 (d, 1, J = 2.2 Hz, H2’),

129

9.52 (br, OH), 12.49 (br, OH), 12.65 (br, OH), 12.99 (br, OH). 13C NMR (125.7 MHz,

130

DMSO-d6): δ 21.56 (C1”), 72.88 (C8), 98.83 (C6), 101.63 (C13), 104.48 (C11), 115.21

131

(C2’), 116.00 (C10), 116.25 (C5’), 120.58 (C6’), 122.36 (C1’), 124.55 (C9), 136.61

132

(C3), 145.68 (C3’), 147.57 (C2), 148.39 (C4’), 150.47 (C14), 159.24 (C12), 160.71

133

(C5), 176.57 (C4).

134

2-(3,4-Dihydroxyphenyl)-8-ethyl-3,5-dihydroxy-4H,8H-pyrano[2,3-f]chromen-4-one

135

(10). HRMS, m/z 367.0810 (M+ – 1), error 3.6 ppm. 1H NMR (500 MHz, DMSO-d6):

136

0.97 (t, 3, J = 7.3 Hz, H2”), 1.74 (qu, 2, J = 7.3 Hz, H1”), 3.36 (br, OH), 4.98 (m, 1,

137

H8), 5.83 (dd, 1, J = 3.4 Hz, J = 10.1 Hz, H9), 6.23 s (s, 1, H6), 6.89 (dd, 1, J = 1.1 Hz,

138

J = 10.1 Hz, H10), 6.91 (d, 1, J = 8.5 Hz, H5’), 7.61 (dd, 1, J = 2.2 Hz, J = 8.5 Hz,

139

H6’), 7.72 (d, 1, J = 2.2 Hz, H2’), 9.52 (br, OH), 12.49 (br, OH), 12.65 (br, OH), 12.99 7 ACS Paragon Plus Environment


13

140

(br, OH).

C NMR (125.7 MHz, DMSO-d6): δ 9.23 (C2”), 28.39 (C1”), 77.49 (C8),

141

98.75 (C6), 101.69 (C13), 104.39 (C11), 115.21 (C2’), 116.24 (C5’), 116.40 (C10),

142

120.59 (C6’), 122.37 (C1’), 123.20 (C9), 136.52 (C3), 145.66 (C3’), 147.39 (C2),

143

148.35 (C4’), 150.42 (C14), 159.47 (C12), 160.69 (C5), 176.52 (C4).

144

Deep-Frying Experiments. Deep-frying experiments were carried out on rapeseed

145

oil (100 g) containing acrolein, crotonaldehyde, and (E)-2-pentenal (each 2.7 µmol/g of

146

oil), which were added before heating the oil. The oil was firstly pre-heated for 9 min to

147

achieve 160 ºC and then heated in the presence, or absence, of thin slices of onions (5 g)

148

for further 3 min. At the end of the heating time, the heated oils were analyzed for

149

aldehyde contents and the fried onions were analyzed to identify aldehyde-quercetin

150

adducts.

151

Determination of Aldehydes in Heated Oils. Aldehydes were determined by 1H

152

NMR spectroscopy analogously to the analysis of oil components developed by

153

Sopelana et al.16 Briefly, oil samples (200 mg) were diluted with CDCl3 (400 µL),

154

which contained 0.2% of non-deuterated chloroform, and their spectra were obtained by

155

a Bruker Advance III spectrometer (Karlsruhe, Germany) operating at 500 MHz.

156

Quantitation of the three aldehydes was carried out by considering the area of the CHCl3

157

signal (at δ 7.29 ppm) as internal standard. The proton of the aldehyde group appeared

158

as a doublet that was independent for each one of the three analyzed aldehydes and

159

appeared far from other signals in the spectrum. The signals of the aldehydic proton

160

appeared at δ 9.55 (d, J = 7.4 Hz) for acrolein, 9.47 (d, J = 7.9 Hz) for crotonaldehyde,

161

and 9.49 (d, J = 7.7 Hz) for (E)-2-pentenal. Four samples were analyzed for each heated

162

oil and each sample was measured three times. Therefore, the aldehyde content for each

163

analyzed oil was the mean value of twelve determinations.


Page 8 of 31

Page 9 of 31


164

Determination of Aldehyde-Quercetin Adducts in Fried Onions. Onions fried in

165

the laboratory and commercially crispy fried onions (5 g each) were homogenized in

166

water (20 mL) and hexane (50 mL). The resulting mixture was centrifuged at 7500 g for

167

10 min at room temperature. The organic layer was discarded and other 50 mL of

168

hexane were added. The mixture was homogenized again and centrifuged at 7500 g for

169

10 min. The organic layer was discarded and methanol (80 mL) was added. The mixture

170

was newly homogenized and centrifuged at 7500 g for 10 min. The supernatant was

171

collected and the solid was extracted with methanol/water (80/20, v/v; 100 mL). The

172

resulting mixture was centrifuged at 7500 g for 10 min and the supernatant was

173

collected. Both supernatants were combined, concentrated to about 4 mL using a

174

rotatory evaporator (35 ºC, 16 mbar) and fractionated on a Sephadex LH-20 column

175

using methanol/water (80/20, v/v) as eluent at a flow rate of 15 mL/h. Fractions

176

obtained were analyzed by HPLC-HRMS.

177

NMR Spectroscopy. All NMR spectra were obtained by a Bruker Advance III

178

spectrometer operating at 500 MHz for protons. For 1H spectra, acquisition parameters

179

were: spectral width 10000 Hz, relaxation delay 1 s, number of scans 16, acquisition

180

time 3.277 s, and pulse width 90º, with a total acquisition time of 1 min 17 s. For 13C

181

spectra, acquisition parameters were: spectral width 27500 Hz, relaxation delay 2 s,

182

acquisition time 1.188 s, and number of pulses depended on the concentration of the

183

sample. All experiments were performed at 23 ºC. For structural determinations, COSY,

184

HMQC and HMBC experiments were carried out.

185

Low Resolution MS. To control carbonyl-phenol reactions and the eluates from the

186

LH-20 column during the isolation of reference compounds, a triple quadrupole API

187

2000 mass spectrometer (Applied Biosystems, Foster City, CA) was employed by using

188

an electrospray ionization interface in the negative ionization mode (ESI–). The 9 ACS Paragon Plus Environment


189

nebulizer gas and the curtain gas were set at 19 and 10 (arbitrary units), respectively.

190

The electrospray capillary voltage was set to –4.5 kV, the declustering potential was –

191

50 V, the focusing potential was –400 V and the entrance potential was –10 V.

192

HPLC-HRMS. The HPLC-ESI-MS system consisted of a Dionex Ultimate 3000RS

193

U-HPLC (Thermo Fisher Scientific, Waltham, MA) coupled to a micrOTOF-QII ultra

194

high resolution time-of-flight mass spectrometer (UHR-TOF) with q-TOF geometry

195

(Bruker Daltonics, Bremen, Germany). Chromatographic separation was performed on a

196

Zorbax Eclipse XDB-C18 column (15 cm × 0.46 cm i. d., 5 µm) from Agilent (Santa

197

Clara, CA). As eluent A, a mixture of acetonitrile containing 0.2% formic acid and 4

198

mM ammonium formiate (30/70, v/v) was used. As eluent B, acetonitrile containing

199

0.2% formic acid was employed. The flow rate was 0.5 mL/min in linear gradient mode:

200

0-13 min 7% B, 13-20 min from 7 to 60% B, 20-30 min 60% B, 30-32 min from 60 to

201

90% B, 32-42 min 90% B, 42-45 min from 90 to 7% B. A split post-column with a flow

202

rate of 0.25 mL/min was inserted directly into the mass spectrometer ESI source. The

203

scan range applied was m/z 50-1500 and mass resolving power was always over 18,000

204

(m/∆m). The instrument was operated in the negative ion mode. Mass spectra and data

205

were obtained by broadband Collision Induced Dissociation (bbCID) mode, providing

206

MS and MS/MS spectra simultaneously. Collision energy was estimated dynamically

207

based on appropriate values for the mass and stepped across a ±10% magnitude range to

208

ensure good quality fragmentation spectra. The instrument control was performed using

209

Compass 1.3 for micrOTOF-Q II + Focus Option Version 3.0 (Bruker Daltonics).

210

Statistical Analysis. Statistical differences among the amounts of aldehydes

211

remaining in the oils after heating in the presence or in the absence of onions were

212

evaluated by the Student t-test.17 These comparisons were carried out using Origin,


Page 10 of 31

Page 11 of 31


213

version 7.0 (OriginLab Corporation, Northampton, MA). Significance level was p

Toxicologically Relevant Aldehydes Produced during the Frying

Recommend Documents