Browning Potential of C6-Î±-Dicarbonyl Compounds under Maillard

Subscriber access provided by University of Newcastle, Australia

Article

The Browning Potential of C6-#-Dicarbonyl Compounds under Maillard Conditions Paul T. Haase, Clemens Kanzler, Julia Hildebrandt, and Lothar W. Kroh J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.6b04512 • Publication Date (Web): 15 Feb 2017 Downloaded from http://pubs.acs.org on February 22, 2017

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 30

Journal of Agricultural and Food Chemistry

The Browning Potential of C6-α α-Dicarbonyl Compounds under Maillard Conditions

Paul T. Haase‡,* Clemens Kanzler‡, Julia Hildebrandt, and Lothar W. Kroh

Institut für Lebensmitteltechnologie und Lebensmittelchemie, Lebensmittelchemie und Analytik, Technische Universität Berlin, Gustav-Meyer-Allee 25, TIB 4/3-1, D-13355 Berlin, Germany *(P.T.H.) Phone: +49-30-31472583. Fax: +49-30-31472585. E-mail: [email protected]

‡

both authors contributed equally to this work

1 ACS Paragon Plus Environment


1

Abstract

2 3

In this work, the three major C6-α-dicarbonyl compounds glucosone (GLUC),

4

1-deoxyglucosone (1-DG), and 3-deoxyglucosone (3-DG) were synthesized and examined

5

under Maillard conditions (aqueous solutions with addition of L-alanine at 130 °C and pH

6

5/8). For the first time, the resulting color formation, antioxidant activity, and generation of

7

short chained α-dicarbonyls was investigated and compared to incubations of D-glucose and

8

D-fructose.

9

generation of α-dicarbonyl compounds, and a synergistic effect on the antioxidant activity

10

could be observed for the 1-DG/GLUC combination. Despite their common degradation

11

products, different extinctions could be measured, with 3-DG showing the strongest color

12

formation followed by GLUC and 1-DG. The analyzed α-dicarbonyl compounds have no

13

direct impact on the formation of color but are precursors for the most of the colored

14

compounds. The main difference between the three substances is their ability to form different

15

heterocyclic degradation products, such as pyranones (1-DG), furanones (1-DG), furans

16

(GLUC and 3-DG), and the corresponding N-heterocycles in presence of amino components.

17

This seems to be the main reason for their varying browning potential and antioxidant

18

activity.

An additive effect on the formation of color, an antagonistic effect on the

19 20

Keywords

21

Browning; Maillard reaction; C6-α-dicarbonyl compounds; antioxidative activity; Folin-

22

Ciocalteu assay

23 24

Introduction 2 ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30


25

The Maillard reaction, included in the non-enzymatic browning, is a very complex reaction

26

sequence which takes place whenever a reducing carbohydrate reacts with an amino group of

27

an amino acid or a protein, respectively. Since the first reports by Maillard,1 this field has

28

been of increasing interest in the area of food chemistry, because of its influence on aroma,

29

taste, texture, and color of the affected food. Today, the most of the important steps taking

30

place during the Maillard reaction, like the formation of Amadori rearrangement products

31

(ARPs) or α-dicarbonyl compounds, are well described in literature. On the other hand there

32

are still aspects that remain unknown, e.g. which of the C6-α-dicarbonyl compounds has the

33

strongest impact on the formation of color or how the different α-dicarbonyls react in

34

combination in terms of antioxidant activity, browning or dicarbonyl formation. In the last

35

100 years, various authors described and summarized the common pathways of the Maillard

36

reaction2–4 and today, there is consensus about that the three C6-α-dicarbonyls (glucosone, 1-

37

deoxyglucosone, and 3-deoxyglucosone) are key intermediates of the Maillard reaction. The

38

formation of the brownish color is associated with the appearance of macromolecules, the

39

melanoidins, in consequence of condensation and polymerization reactions. They derive from

40

many different substances, like α-dicarbonyls, furans, and pyrroles,3 which are generated

41

during the Maillard reaction. Because of their intensive color it is likely that large

42

melanoidins possess a chromophoric backbone. Tressl et al.5,6 suggested that pyrroles can

43

form large polymers with partly conjugated π-bonds, which contribute to the browning. It has

44

been shown that other small chromophoric molecules such as 2-[2-(furyl)methylidene]-4-

45

hydroxy-5-methyl-2H-furan-3-one have an impact on the impression of color as well.7–9 Some

46

studies discuss the relation between the formation of color and the antioxidant activity of

47

Maillard reaction mixtures.10–12 In other publications, the formation of color has been

48

associated

49

3-deoxyglucosone,13,14 3-deoxypentosulose,15 methylglyoxal,16 and glyoxal17 and their

directly

with

the

relevant

Maillard

products


1-deoxyglucosone,13


Page 4 of 30

50

reaction(s) with amino compounds. In contrast, recent studies from Pfeifer et al.18 have shown

51

that 3-deoxyglucosone has only little impact on the browning of foods under certain

52

conditions (10 mmol/L 3-DG and 100 mmol/L L-alanine in water or phosphate buffer at 100

53

or 130 °C).

54

Therefore, in this work the behavior of the three main C6-α-dicarbonyl compounds

55

(glucosone, 1-deoxyglucosone, and 3-deoxyglucosone) has been investigated in Maillard

56

model systems to determine their browning potential and possible relations between color,

57

antioxidant activity and the formation of degradation products with α-dicarbonyl structure.

58

Due to improvements in the synthesis of α-dicarbonyls (Glomb and Pfahler,19 Hellwig et al.,20

59

and Usui et al.21) the pure compounds could be used for these studies. For the first time these

60

substances could be compared to each other directly and were combined, to give a more

61

detailed view on their reaction behavior in binary α-dicrabonyl model systems with or

62

without

63

α-dicarbonyl compounds in combination with L-alanine was investigated together with the

64

browning of model systems with UV/VIS-spectroscopy and their antioxidant activity with the

65

Folin-Ciocalteu-reagent (FCR) assay. In addition, relations between the generation or

66

degradation of α-dicarbonyls and the color formation were studied as well as the correlation

67

between the antioxidant activity and the color formation.

L-alanine.

The formation of α-dicarbonyls from carbohydrates or the named

68 69

Materials and Methods

70

Materials. The following chemicals were commercially available: benzaldehyde (for

71

synthesis), p-toluidine (99 %) (Merck, Hohenbrunn, Germany),

72

2,3-O-isopropylidene-D-erythronolactone (98 %), o-phenylendiamine (OPD) (Fluka/Sigma –

73

Aldrich, Munich, Germany), D-glucose (99.5 %), acetic acid (100 %), hydrochloric acid (1

74

mol/L) (Roth, Karlsruhe, Germany), tetrahydrofuran (extra dry), t-butyllithium (1.6 4 ACS Paragon Plus Environment

L-alanine

(99 %),

M

in

Page 5 of 30


75

pentane), ethylvinylether (99 %), diethyl ether (99.5 %), phenylhydrazine (for synthesis),

76

phenylhydrazide (98 %), D-fructose (99 %), Dowex 50 W x 8 (H+ - form, 50 - 100 mesh)

77

(Acros Organics, Geel, Belgium/Fischer, Nidderau, Germany), ethyl acetate (HPLC-grade)

78

(VWR, Fontenay-sous-Bois, France). 1-deoxyglucosone was synthesized as described by

79

Glomb and Pfahler.19 Glucosone was synthesized according to Usui et al.21 with the following

80

modifications: the phenylhydrazone derivative was washed with cold water (500 mL) and

81

cold ethanol (400 mL) after recrystallization in pyridine. For the second part of the reaction

82

more filtration steps when reducing the solvent to 100 mL (each time the amount of solid

83

affected a good evaporation) and more extraction steps with diethylether and ethyl acetate

84

were performed. The final work-up was carried out as described by Hellwig et al.20

85

3-deoxyglucosone was synthesized according to Hellwig et al.20 with the following

86

modifications: the benzoylhydrazone derivative was extensively washed with diethylether.

87

For the second part of the reaction, the crude reaction mixture was not stirred with Serdolit

88

MB-2 (first step), but with Amberlite IRA-402 (OH-) (second step); the final work-up was

89

identical to glucosone; column chromatography on a silica gel 100 C18 phase with

90

acetonitrile/water (v/v, 19/1) was performed for purification.

91

HPLC Analysis of α-Dicarbonyl Compounds. For HPLC analysis of α-dicarbonyl

92

compounds glucosone, 1-deoxyglucosone, 3-deoxyglucosone, 3-deoxypentosone, 1,4-

93

dideoxyglucosone, glyoxal, methylglyoxal and diacetyl the same method as published before

94

was used (C18 column; water/methanol gradient).12,22

95

Carbohydrate Reaction Models. Model reactions under typical Maillard conditions were

96

conducted. For that purpose aqueous solutions of D-glucose or D-fructose (0.2 mol/L) with or

97

without (supplementary data)

98

adjusted to (5.0 ± 0.1) or (8.0 ± 0.1) with HCl (0.1 mol/L) or NaOH (0.1 mol/L). The

99

solutions were sealed in ampoules and were thermally treated in a thermo block (Behr Labor

L-alanine

(0.2 mol/L) were prepared. The pH value was



Page 6 of 30

100

Technik, behrotest ET2) at (80 ± 1) °C, (100 ± 1) °C (supplementary data), or (130 ± 1) °C for

101

a defined reaction time (0, 30, 60, 120, 180, and 300 min). For direct comparison to the

102

α-dicarbonyl reaction models, additional carbohydrate reaction models were prepared in a

103

concentration of 0.02 mol/L at 130 °C. α-Dicarbonyl compounds were trapped with OPD to

104

form stable quinoxalines. All experiments, except the model reactions containing 0.02 mol/L

105

of a carbohydrate were carried out as duplicate (all results are given as means ± standard

106

deviation).

107

α-Dicarbonyl Reaction Models. For α-dicarbonyl reaction models, aqueous solutions of

108

GLUC, 1-DG, and 3-DG (0.02 mol/L) with or without (supplementary data)

109

(0.2 mol/L) were used. When combined, only 0.01 mol/L of each α-dicarbonyl was used

110

respectively. The reactions were conducted under the same conditions as the carbohydrate

111

reaction models. All α-dicarbonyl experiments were carried out as dublicate (all results are

112

given as means ± standard deviation).

113

UV/VIS and Browning Measurements. For all UV/VIS measurements a SHIMADZU UV-

114

1650PC UV-VIS spectrophotometer and the appropriate UVProbe 2.21 software by

115

SHIMADZU was used. All samples were diluted 1:20 in water, centrifugalized (HERMLE Z-

116

233MK, 10 min, 20 °C; 14000 rpm) and measured at 420 nm in comparison to water.

117

Folin-Ciocalteu Reagent Assay. The calibration was performed with five gallic acid

118

standards (10-80 µg/mL). The samples were diluted with water to fit the calibration curve.

119

200 µL of the sample, 1000 µL of Folin & Ciocalteus phenol reagent (diluted 1:10 with

120

water), and 800 µL of sodium carbonate solution (7.5 wt%) were mixed. The mixture was

121

incubated for 15 min at 35 °C and measured at 736 nm compared to a blank sample (water

122

instead of sample). To convert the results from gallic acid equivalents (GAE) to trolox

123

equivalents (TE), different samples of trolox were measured with respect to gallic acid.12

124 6 ACS Paragon Plus Environment

L-alanine

Page 7 of 30


125

Results and Discussion

126

Carbohydrate Model Systems. The extinctions at 420 nm of the heated D-glucose (D-glc)

127

and

128

dependent on the chosen pH value and temperature. Model systems at pH 8 and at high

129

temperatures exhibit higher extinctions than at pH 5 and at low temperatures for both

130

carbohydrates. At pH 5 the

131

increased to 0.98 at pH 8. For D-fru the difference between both pH values is not as high as

132

for D-glc but differs between 0.81 (pH 5) and 0.90 (pH 8). These observations coincide with

133

literature.23 Subsequent only examples for pH 5 with L-ala are shown, because this pH value

134

is more relevant for the processing of food.

135

To understand why both carbohydrates exhibit different extinctions at 420 nm under varying

136

conditions, the formation of α-dicarbonyl compounds was investigated. Both sugars produce

137

a comparable range of α-dicarbonyl compounds and only differ in the detected amounts.

138

While the temperature has a significant impact on the formation of α-dicarbonyls, the pH

139

value is of lower relevance although the formation of color varies with the pH value. This

140

indicates that other formed intermediates play a role for the color as well. At 80 °C and a pH

141

value of 5 D-glc only produces glucosone (GLUC) and 3-deoxyglucosone (3-DG), whereas 1-

142

deoxyglucosone (1-DG), glyoxal (GO), and methylglyoxal (MGO) can additionally be found

143

in model systems with D-fru under the same conditions. At pH 8 (80 °C) both carbohydrates

144

degrade to the same α-dicarbonyls namely GLUC, 3-DG, 1-DG, GO, MGO, and 3-

145

deoxypentosone (3-DP). When the temperature of model systems at pH 5 is raised to 100 °C

146

diacetyl (DA) can be detected in the D-fru/L-ala reaction mixtures while 1-DG, GO, and MGO

147

are additionally found in the D-glc/L-ala reaction mixtures. Furthermore, DA can be quantified

148

at pH 8 and 100 °C in both reaction systems (supplementary data). Altogether, with higher

D-fructose

(D-fru) systems in aqueous solution show that the color formation differs

D-glc

model shows a maximal extinction of 0.63 which is



Page 8 of 30

149

temperatures more α-dicarbonyls are formed and they can be found in higher concentrations

150

what is even more evident in model reactions at 130 °C.

151

3-DG is the dominating α-dicarbonyl compound under all conditions and reaches

152

concentrations between 400 and 670 µmol/L in D-fru model incubations. This has been shown

153

by Fiedler et al.

154

below 200 µmol/L. Table 1 also shows that D-fru model reactions form a higher amount of

155

α-dicarbonyls (3-DG and the short-chained compounds) than

156

dicarbonyls can be detected after the first 30 min. Table 1 shows an overview for all detected

157

α-dicarbonyl compounds at 130 °C.

158

If the concentrations of the formed α-dicarbonyls are considered in relation to the extinction

159

at 420 nm, it can be seen that the majority of the dicarbonylic compounds is formed in the

160

first 60 min and remains nearly constant (for D-glc, Figure 1a) or decreases for the time of

161

measurement (for D-fru, Figure 1b). In contrast to this behavior, the formation of color is

162

slower and increases over the whole reaction time for both carbohydrates (see Figure 1).

163

Even if the total amount of α-dicarbonyls in the D-fru model under aqueous conditions is

164

temporarily nearly twice as much as in the

165

reactions differs only around a factor of 1.12. Due to the method, only temporary

166

concentrations at the given time are detected and considering that the α-dicarbonyls are

167

highly reactive intermediate stages, the observed difference in the browning potential is

168

probable dependent on the already degraded α-dicarbonyls. As published before11,12 the

169

antioxidant activity correlates with the formation of color for Maillard reaction mixtures of

170

sugars.

171

α-Dicarbonyl Model Systems. To get a better understanding about how the α-dicarbonyl

172

compounds affect the color formation, the three C6-α-dicarbonyls GLUC, 1-DG, and 3-DG

16

as well. In contrast, the other α-dicarbonyls are formed in concentrations

D-glc

D-glc.

In most cases α-

model system, the extinction of both


Page 9 of 30


173

and their binary combinations were examined under the same conditions as the carbohydrates

174

described above. Considering the low quantities formed from

175

α-dicarbonyl compounds were used in a ten times lower concentration. Figure 2a shows the

176

extinctions at 420 nm for GLUC, 1-DG, and 3-DG after heating with L-ala at 130 °C, whereas

177

figure 2b shows the results of their binary combinations. The browning of all α-dicarbonyl

178

reaction mixtures increases over the reaction time reaching exctintions between 0.17 (1-DG)

179

and 0.48 (3-DG) at 300 min (Figure 2a).

180

For 3-DG and GLUC the extinction is nearly twice as high as for 1-DG (see Figure 2a). If

181

combined (see Figure 2b), extinctions between 0.25 (1-DG/GLUC) and 0.43 (3-DG/GLUC)

182

are reached. Assuming, that two α-dicarbonyls show an additive browning the resulting

183

extinction

184

+

could -

be

calculated

with

the

D-glc

and

following

D-fru,

the

formula:

= Ecombination. Hence, the extinction for the combination of GLUC .

and 1-DG with L-ala at 130 °C and 300 min should be

186

result of 0.306 shows that there is a slight difference between the theoretical additive

187

combination and the observed browning for the system GLUC/1-DG. Figure 3 shows the

188

differences of the model reaction and the theoretical result over the reaction time of 300 min.

189

Even if the real values are smaller than the theoretical values all of the time, the difference is

190

too small to speak about an antagonistic effect on browning. Generally the effect of

191

combining two α-dicarbonyl compounds seems to be additive for the extinction at 420 nm

192

(for 1-DG/3-DG or 3-DG/GLUC see supplementary data). This suggests that no additional

193

compounds with an impact on color formation are generated on account of the combination.

194

If the degradation of the α-dicarbonyl compounds is seen in relation to the extinction at

195

420 nm (data not shown) it seems clear that the formation of color must result from the

196

decrease of the measured α-dicarbonyl compound or the respective combinations. For GLUC

+

.

185


= 0.268. The measured


197

the decomposition in the presence of L-ala is very fast (no measureable amount of GLUC after

198

30 min) and the extinction rises faster in comparison to the other α-dicarbonyls but stays

199

nearly constant after 120 min (E420 ~ 0.34). The degradation of 1-DG is as fast as of GLUC

200

but does not result in a high extinction (E420 ~ 0.17) and the color stays nearly constant after

201

1-DG is completely decomposed after 60 min. One of the main degradation products of 1-DG

202

is 2,3-dihydro-3,5-dihydroxy-6-methyl-4(H)-pyran-4-one (DHHM) which is known to form

203

mostly short chained cleavage products that do not contribute to color formation.24 3-DG

204

shows a fast decomposition as well (no measureable amount of 3-DG after 120 min) and in

205

contrast to GLUC and 1-DG a constantly rising extinction at 420 nm over the measured time

206

(E420 = 0.48 at 300 min). When combined the α-dicarbonyl compounds show more or less the

207

same degradation behavior as the single compounds and their extinctions display a similar

208

course as well. Generally the extinctions of the α-dicarbonyl model reactions rise faster than

209

the extinctions of the carbohydrate reactions under identical conditions (0.02 mol/L). These

210

results confirm that not only the direct successors of the α-dicarbonyls are important for the

211

color formation but also products of later stages of the Maillard reaction, because the

212

browning still increases when the α-dicarbonyls are already completely degraded (especially

213

for 3-DG and on a minor level for 1-DG and GLUC). In contrast to sugars, Maillard reaction

214

mixtures of the α-dicarbonyls or their combinations do not show a correlation between

215

antioxidant activity and formation of color. For all examined α-dicarbonyl model systems

216

both measured parameters rise in the first minutes but then the antioxidant activity stagnates

217

or even decreases. The formation of compounds with an antioxidant capacity apparently takes

218

place in the first 30 min of the heating time and is followed by reactions that consume these

219

compounds. Another explanation for the decreasing antioxidant activity could be the

220

formation of condensation products in later stages of the browning and the subsequent loss of

221

antioxidant functional groups such as the reductone structure. For model reactions containing 10 ACS Paragon Plus Environment

Page 10 of 30

Page 11 of 30


222

GLUC or 1-DG which are both compounds showing an antioxidant capacity themselve,12 this

223

seems even more logical due to the fact that they are degraded during the heating time.

224

If the antioxidant activity is analyzed for combined model systems, a synergistic effect can be

225

observed for GLUC/1-DG after 30 min. (figure 4). For the other two combinations only an

226

additive effect can be noticed (see supplementary data). This indicates that 1-DG and GLUC

227

in combination form stable degradation or condensation products that do not lose the

228

reductone structure of their precursors. A possible reaction pathway could be the aldol

229

condensation of furanones formed from 1-DG and furfural formed from GLUC resulting in

230

furanone derivatives with intact reductone ether structure.25

231

To understand the differences between the three C6-α-dicarbonyl compounds, the formation

232

of short chain α-dicarbonyls was analyzed. 1-DG, GLUC, and their combinations form a

233

wider range of α-dicarbonyls (1,4-DDG, DA, GO, MGO, and 3-DP) than 3-DG for which

234

only MGO can be detected. In reaction mixtures of 1-DG MGO is the main α-dicarbonyl

235

whereas 3-DP can be found as dominating species in GLUC incubation.

236

The extinctions and the summarized α-dicarbonyl concentration of the GLUC, 1-DG and 3-

237

DG model reactions and their combinations are shown in figure 5. The amounts differ

238

strongly with GLUC model reactions reaching a maximal summarized α-dicarbonyl product

239

concentration of 141 µmol/L compared to 225 µmol/L for 1-DG and 90 µmol/L (MGO) for 3-

240

DG (figure 5). In the combined model reactions no additional α-dicarbonyl compounds can be

241

detected and lower overall concentrations (up to 57 µmol/L) are formed. In model reactions

242

containing GLUC the graphs of the extinction and summarized α-dicarbonyl concentration

243

show mirror symmetry after 30 min. This indicates that the degradation of the produced

244

α-dicarbonyls directly results in the formation of color. A possible explanation is that 3-DP

245

and its main degradation product furfural might be involved in the formation of color. For

246

1-DG (figure 5) the highest concentrations of degradation products with α-dicarbonyl 11 ACS Paragon Plus Environment


247

structure are formed after 30 min as well, whereas the extinction at 420 nm slowly rises to

248

0.17 in 120 min and nearly stays constant for the rest of time. However, the concentrations of

249

the short chained α-dicarbonyl compounds decrease after 120 min and do not result in the

250

formation of color. The MGO detected in 3-DG model reactions reaches its maximum

251

concentration after 120 min. Then the amount decreases over the reaction time, whereas the

252

extinction rises over the whole time frame and reaches 0.48 at 300 min. There might be a

253

relation between the degradation of MGO and the color formation, but there is no mirror

254

symmetry indicating that other products have an impact as well (figure 5).

255

For the combinations lower maximum concentrations of the α-dicarbonyl compounds can be

256

detected (see figure 5: up to 57 µmol/L for GLUC/1-DG after 60 min; up to 55 µmol/L for

257

3-DG/1-DG after 120 min; up to 42 µmol/L for GLUC/3-DG after 30 min) and for all

258

combinations the degradation of α-dicarbonyl products seems to be accompanied with

259

formation of color after a certain point in time (e.g. after 30 min for GLUC/3-DG). Generally,

260

the extinctions of the combinations seem to have an additive effect. If the relation between the

261

α-dicarbonyl concentration of the isolated and combined model reactions is considered (see

262

figure 6), it is obvious that the concentrations in the model reaction are smaller than the

263

theoretical values. In contrast, it seems that the effect of combining two α-dicarbonyl

264

compounds is antagonistic for the concentrations of the generated α-dicarbonyl degradation

265

products (see figure 6 and supplementary data). There have to be other reaction pathways

266

which are preferred but do not influence the formation of color.

267

Relevance of the C6-α-Dicarbonyl Compounds for the Properties of Carbohydrate

268

Reaction Mixtures. The color generation depends on the formation of α-dicarbonyl

269

compounds as intermediates and their degradation into other molecules that are factual

270

colored. For the two carbohydrates only small differences of the extinctions can be observed,

271

with D-fru reaching slightly higher values. The pH value does not affect the color formation in 12 ACS Paragon Plus Environment

Page 12 of 30

Page 13 of 30


272

the same way as the temperature, which exhibits a higher impact but in most reactions pH 8

273

shows a bigger extinction than pH 5 (see figure 7). In general, there are too many influences

274

on the color formation, so the differences between the two carbohydrates and between the

275

three α-dicarbonyl compounds (GLUC, 1-DG, and 3-DG) cannot be explained by the

276

generated α-dicarbonyls alone. 3-DG, which forms the highest extinctions at 420 nm (0.48),

277

generates only 90 µmol/L MGO as an identified α-dicarbonyl compound. In contrast, the

278

highest α-dicarbonyl concentrations can be detected in 1-DG model system (225 µmol/L),

279

which does not reach an extinction half as high as the two other α-dicarbonyls (see figure 7).

280

These results confirm that the prediction of the browning potential solely based on C6-α-

281

dicarbonyl compounds is not possible and further intermediates have to be taken into

282

consideration. A possible key to explain the differences in the color might be the formation of

283

heterocyclic compounds.

284

It is known that the major degradation product of 3-DG is the furan 5-(hydroxymethyl)-2-

285

furaldehyde (HMF)18 and the main cleavage product of 3-DG (MGO) is able to form the

286

heterocycle 4-Hydroxy-2,5-dimethyl-3-furanone.26 GLUC can be cleaved into formic acid and

287

C5-molecules which are precursors of furans (e.g. furfural).27,28 HMF and furfural are able to

288

form colored condensation products 1-DG is mainly cleaved into acetic acid and C4

289

molecules, but also the formation of furanones (e.g. acetylformoin or 4-hydroxy-2-

290

(hydroxymethyl)-5-methylfuran-3(2H)-one) and pyranones, such as its major heterocyclic

291

degradation product DHHM, are described in literature.

292

For furanones and pyranones only few reactions to macromolecules are known which might

293

be the reason for the lower extinctions in 1-DG model systems compared to GLUC and 3-DG

294

model reactions.5,29–33 As described in literature,8,25 furanones form colored condensation

295

products with furfural or HMF, but judging from the observation that there is no synergistic



296

effect on color formation in 1-DG/GLUC or 1-DG/3-DG systems, these reactions pathways

297

do not seem to be of great significance under the chosen conditions.

298

Some of these heterocycles are known to have an antioxidant capacity34 and therefore it is

299

possible that their chromophore polymers might have this attribute as well. This could be the

300

reason for the observed synergistic effect on the antioxidant activity when both α-dicarbonyls

301

with a reductone structure (1-DG and GLUC) are combined.

302

Of course in presence of amino compounds the formation of N-derivatives of the different

303

heterocycles has to be taken in account, but most of the mentioned pathways are described for

304

O- and N-heterocycles.

305

In Conclusion, these results indicate that the degradation products of different C6-α-

306

dicarbonyls inhibit or intensify certain pathways during the Maillard reaction leading to

307

different properties of the respective reaction mixtures. It could be shown that the α-

308

dicarbonyls do not affect each other when the formation of color is compared to the single

309

model reactions, but lower amounts of dicarbonyl cleavage products could be found in all

310

three combinations. The 1-DG/GLUC system formed more antioxidants resulting in higher

311

antioxidant activities of the reaction mixtures. Therefore, the combined model systems help to

312

understand, how the reducing abilities of complex Maillard mixtures are preserved for the

313

most part in course of the reaction, even when the antioxidants GLUC and 1-DG decompose.

314

Contrary, in our previous study12 we could only find a decreasing antioxidant activity with

315

decreasing concentrations of the named α-dicarbonyl compounds when these were used

316

isolated. This knowledge might help to control the formation of antioxidants during food

317

processing and increase the oxidative stability of heat treated foods depending on the used

318

conditions in the future, especially in beverages such as beer and coffee or in baked goods. In

319

future studies, heterocyclic intermediates and their condensation products should be


Page 14 of 30

Page 15 of 30


320

investigated to provide more insight in the reasons for different reactivity of GLUC, 1-DG

321

and 3-DG.



Page 16 of 30

322

Abbreviations Used

323

GLUC,

324

3-deoxypentosone; 1,4-DDG, 1,4-dideoxygulucosone; GO, glyoxal; MGO, methylglyoxal;

325

DA, diacetyl;

326

dihydroxy-6-methyl-4(H)-pyran-4-one

glucosone;

1-DG,

L-ala, L-alanine;

1-deoxyglucosone;

3-DG,

3-deoxyglucosone;

3-DP,

HMF, hydroxymethylfurfural; DHHM, 2,3-dihydro-3,5-

327 328

Acknowledgement

329

We want to thank Prof. T. Hofmann, M. Ilse and O. Frank for the temporary collaboration.

330


Page 17 of 30


331

Literature Cited

332

(1) Maillard, L. C. Action of Amino Acids on Sugars. Formation of Melanoidins in a

333 334 335 336 337 338 339

Methodical Way. CR ACAD SCI II C 1912, 66–68. (2) Hodge, J. E. Dehydrated Foods, Chemistry of Browning Reactions in Model Systems. J. Agric. Food Chem. 1953, 1, 928–943. (3) Beck, J.; Ledl, F.; Sengl, M.; Severin, T. Formation of acids, lactones and esters through the Maillard reaction. Z. Lebensm.-Unters -Forsch. 1990, 190, 212–216. (4) Hellwig, M.; Henle, T. Baking, ageing, diabetes: a short history of the Maillard reaction. Angewandte Chemie (International ed. in English) 2014, 53, 10316–10329.

340

(5) Tressl, R.; Wondrak, G. T.; Garbe, L.-A.; Krüger, R.-P.; Rewicki, D. Pentoses and

341

Hexoses as Sources of New Melanoidin-like Maillard Polymers. J. Agric. Food Chem.

342

1998, 46, 1765–1776.

343 344

(6) Tressl, R.; Wondrak, G. T.; Krüger, R.-P.; Rewicki, D. New Melanoidin-like Maillard Polymers from 2-Deoxypentoses. J. Agric. Food Chem. 1998, 46, 104–110.

345

(7) Frank, O.; Hofmann, T. On the Influence of the Carbohydrate Moiety on Chromophore

346

Formation during Food-Related Maillard Reactions of Pentoses, Hexoses, and

347

Disaccharids. Helv. Chim. Acta 2000, 83, 3246–3261.

348

(8) Hofmann, T. Characterization of the most intense coloured compounds from Maillard

349

reactions of pentoses by application of colour dilution analysis. Carbohydr. Res. 1998,

350

313, 203–213.

351

(9) Hofmann, T. Characterization of precursors and elucidation of the reaction pathway

352

leading to a novel coloured 2H,7H,8aH-pyrano[2,3-b]pyran-3-one from pentoses by

353

quantitative studies and application of 13C-labelling experiments. Carbohydr. Res. 1998,

354

313, 215–224.

355 356 357

(10)

Yilmaz, Y.; Toledo, R. Antioxidant activity of water-soluble Maillard reaction

products. Food Chem. 2005, 93, 273–278. (11)

Manzocco, L.; Calligaris, S.; Mastrocola, D.; Nicoli, M. C.; Lerici, C. R. Review of

358

non-enzymatic browning and antioxidant capacity in processed foods. Trends Food Sci.

359

Tech. 2000, 11, 340–346.

360 361

(12)

Kanzler, C.; Haase, P. T.; Kroh, L. W. Antioxidant Capacity of 1-Deoxy-d-erythro-

hexo-2,3-diulose and d-arabino-Hexo-2-ulose. J. Agric. Food Chem. 2014.



362

(13)

Yaylayan, V. A.; Kaminsky, E. Isolation and structural analysis of Maillard polymers:

363

caramel and melanoidin formation in glycine/glucose model systems. Food Chem. 1998,

364

63, 25–31.

365 366 367

(14)

Cämmerer, B.; Jalyschko, W.; Kroh, L. W. Intact Carbohydrate Structures as Part of

the Melanoidin Skeleton. J. Agric. Food Chem. 2002, 50, 2083–2087. (15)

Hollnagel, A.; Kroh, L. W. 3-Deoxypentosulose: An α-Dicarbonyl Compound

368

Predominating in Nonenzymatic Browning of Oligosaccharides in Aqueous Solution. J.

369

Agric. Food Chem. 2002, 50, 1659–1664.

370

(16)

Fiedler, T.; Moritz, T.; Kroh, L. W. Influence o a-dicarbonyl compounds to the

371

molecular weight distribution of melanoidins in sucrose solutions. Eur. Food Res.

372

Technol. 2005, 223, 837–842.

373

(17)

Hofmann,

T.

4-Alkylidene-2-imino-5-[4-alkylidene-5-oxo-1,3-imidazol-2-

374

inyl]azamethylidene-1,3-imidazoline - A novel colored substructure in melanoidins

375

formed by maillard reactions of bound arginine with glyoxal and furan-2-carboxaldehyde.

376

J. Agric. Food Chem. 1998, 46, 3896–3901.

377 378 379 380 381 382 383

(18)

Pfeifer, Y. V.; Haase, P. T.; Kroh, L. W. Reactivity of Thermally Treated α-

Dicarbonyl Compounds. J. Agric. Food Chem. 2013, 61, 3090–3096. (19)

Glomb, M. A.; Pfahler, C. Synthesis of 1-deoxy-D-erythro-hexo-2,3-diulose, a major

hexose Maillard intermediate. Carbohydr. Res. 2000, 329, 515–523. (20)

Hellwig, M.; Degen, J.; Henle, T. 3-Deoxygalactosone, a “New” 1,2-Dicarbonyl

Compound in Milk Products. J. Agric. Food Chem. 2010, 58, 10752–10760. (21)

Usui, T.; Yoshino, M.; Wantnabe, H.; Hayase, F. Determination of Glyceraldehyde

384

Formed in Glucose Degradation and Glycation. Biosci., Biotechnol., Biochem. 2007, 71,

385

2162–2168.

386

(22)

Pfeifer, Y. V.; Kroh, L. W. Investigation of Reactive -Dicarbonyl Compounds

387

Generated from the Maillard Reactions of L-Methionine with Reducing Sugars via their

388

Stable Quinoxaline Derivatives. J. Agric. Food Chem. 2010.

389 390 391 392

(23)

Ashoor, S. H.; Zent, J. B. Maillard Browning of Common Amino Acids and Sugars.

Journal of Food Science 1984, 49, 1206–1207. (24)

Kim, M.-O.; Baltes, W. On the Role of 2,3-Dihydro-3,5-dihydroxy-6-methyl-4( H )-

pyran-4-one in the Maillard Reaction. J. Agric. Food Chem. 1996, 44, 282–289.


Page 18 of 30

Page 19 of 30

393


(25)

Ledl, F.; Severin, T. Bräunungsreaktionen von Pentosen mit Aminen. Z Lebensm

394

Unters Forch (Zeitschrift für Lebensmittel-Untersuchung und -Forschung) 1978, 167,

395

410–413.

396

(26)

Wang, Y.; Ho, C.-T. Formation of 2,5-dimethyl-4-hydroxy-3(2H)-furanone through

397

methylglyoxal: a Maillard reaction intermediate. J. Agric. Food Chem. 2008, 56, 7405–

398

7409.

399 400 401

(27)

Rakete, S.; Klaus, A.; Glomb, M. A. Investigations on the Maillard reaction of

dextrins during aging of Pilsner type beer. J. Agric. Food Chem. 2014, 62, 9876–9884. (28)

Usui, T.; Yanagisawa, S.; Ohguchi, M.; Yoshino, M.; Kawabata, R.; Kishimoto, J.;

402

Arai, Y.; Aida, K.; Watanabe, H.; Hayase, F. Identification and Determination of a-

403

Dicarbonyl Compounds Formed in the Degradation of Sugars. Biosci., Biotechnol.,

404

Biochem. 2007, 71, 2465–2472.

405

(29)

Hofmann, T.; Heuberger, S. The contribution of coloured Maillard reaction products

406

to the total colour of browned glucose/l-alanine solutions and studies on their formation.

407

Z. Lebensm.-Unters -Forsch. 1999, 208, 17–26.

408

(30)

Hofmann, T. Identification of Novel Colored Compounds Containing Pyrrole and

409

Pyrrolinone Structures Formed by Maillard Reactions of Pentoses and Primary Amino

410

Acids. J. Agric. Food Chem. 1998, 46, 3902–3911.

411

(31)

Voigt, M.; Smuda, M.; Pfahler, C.; Glomb, M. A. Oxygen-Dependent Fragmentation

412

Reactions during the Degradation of 1-Deoxy-d-erythro-hexo-2,3-diulose. J. Agric. Food

413

Chem. 2010, 58, 5685–5691.

414

(32)

Nikolov, P. Y.; Yaylayan, V. A. Thermal Decomposition of 5-(Hydroxymethyl)-2-

415

furaldehyde (HMF) and Its Further Transformations in the Presence of Glycine. J. Agric.

416

Food Chem. 2011, 59, 10104–10113.

417

(33)

Heyns,

K.;

Hauber,

R.

Strukturermittlung

spezifisch14C-markierter

418

Sorbosebräunungspolymerisate durch thermische Fragmentierung. Justus Liebigs Ann.

419

Chem. 1970, 733, 159–169.

420

(34)

Kanzler, C.; Haase, P. T.; Schestkowa, H.; Kroh, L. W. Antioxidant Properties of

421

Heterocyclic Intermediates of the Maillard Reaction and Structural Related Compounds. J.

422

Agric. Food Chem.

423 424



425

Figure Captions

426 427

Figure 1: Correlation between the extinction and the summarized concentrations of the

428

α-dicarbonyls for a)

429

mol/L) [M = 2].

430

Figure 2: Extinctions at 420 nm of the model reactions at 130 °C and pH 5 with L-alanine

431

(0.2 mol/L) for a) GLUC, 1-DG, and 3-DG (0.02 mol/L); and b) GLUC/1-DG, 1-DG/3-DG,

432

and GLUC/3-DG (each 0.01 mol/L) [M = 2].

433

Figure 3: Effect of α-dicarbonyl combination of GLUC/1-DG (each 0.01 mol/L) with

434

L-alanine

435

Figure 4: Effect of α-dicarbonyl combination (for 1-DG and GLUC (each 0.01 mol/L)) with

436

L-alanine

437

Figure 5: Relation between the extinction and the summarized concentrations of the

438

α-dicarbonyls (0.02 mol/L) and combinations (each 0.01 mol/L) with L-alanine (0.2 mol/L).

439

Figure 6: Effect of α-dicarbonyl combination of GLUC/1-DG (each 0.01 mol/L) with L-

440

alanine (0.2 mol/L) at 130 °C and pH 5 on generation of α -dicarbonyl compounds.

441

Figure 7: Formation of color at 130 °C after 300 min with L-alanine (0.2 mol/L) for all model

442

reactions (D-fru and D-glc used with a concentration of 0.2 mol/L and 0.02 mol/L ([M = 1]);

443

α-dicarbonyls with 0.02 mol/L and the combinations with 0.01 mol/L each) [M = 2].

444

Figure 8: Possible ways for formation of color from C6-α-dicarbonyl compounds without

445

amino components.

446

Table 1: Overview of the α-dicarbonyl concentrations (in µmol/L) and extinctions at 420 nm

447

for D-glc and D-fru (0.2 mol/L) model reactions with L-alanine (0.2 mol/L) at pH 5/8 and

448

130 °C.

D-glc

and b)

D-fru

(0.2 mol/L) model reactions with L-alanine (0.2

(0.2 mol/L) on extinction at 420 nm at 130 °C and pH 5.

(0.2 mol/L) at 130 °C and pH 5 on the antioxidative activity (mol TE/L) [M = 2].


Page 20 of 30

Page 21 of 30


Table 1: system

pH value

D-fructose + L-alanine

5

analyt/extinction

0

30

60

120

180

300

GLUC

13

50

53

53

38

20

1-DG

0

165

148

132

108

95

3-DG

24

669

669

653

568

505

GO

0

27

37

36

47

47

MGO

0

102

119

135

119

118

DA

0

10

23

35

43

63

0.002

0.044

0.155

0.444

0.591

0.804

GLUC

18

62

56

39

37

29

1-DG

0

159

149

116

104

85

3-DG

24

531

585

558

522

454

GO

0

65

63

59

52

51

MGO

0

139

136

135

125

139

DA

0

36

37

37

34

35

0.002

0.196

0.315

0.596

0.816

0.903

GLUC

0

27

29

36

36

31

1-DG

0

33

42

37

40

43

3-DG

0

297

436

434

426

445

GO

0

62

79

69

60

57

MGO

0

27

36

44

53

62

E420

0.004

0.028

0.147

0.280

0.582

0.635

GLUC

0

49

52

37

35

32

1-DG

0

72

54

49

45

43

3-DG

0

418

492

466

419

384

GO

0

166

110

80

65

56

MGO

0

67

69

74

73

95

E420

0.007

0.235

0.415

0.691

0.807

0.977

E420

8

E420

D-glucose + L-alanine

5

8

time in min



Figure 1:


Page 22 of 30

Page 23 of 30


Figure 2:



Figure 3:


Page 24 of 30

Page 25 of 30


Figure 4:



Figure 5:


Page 26 of 30

Page 27 of 30


Figure 6:



Figure 7:


Page 28 of 30

Page 29 of 30


Figure 8:



TOC:


Page 30 of 30

Browning Potential of C6-Î±-Dicarbonyl Compounds under Maillard

Recommend Documents