Sous-Vide Non-Enzymatic Browning of Glucosamine at Different

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Food and Beverage Chemistry/Biochemistry

Sous-vide non-enzymatic browning of glucosamine at different temperatures Prinjiya Dhungel, Yuliya Hrynets, and Mirko Betti J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.8b01265 • Publication Date (Web): 16 Apr 2018 Downloaded from http://pubs.acs.org on April 16, 2018

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 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 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.

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 34

Journal of Agricultural and Food Chemistry

Sous-Vide Non-Enzymatic Browning of Glucosamine at Different Temperatures Prinjiya Dhungel,a Yuliya Hrynetsa and Mirko Bettia

a

Department of Agricultural, Food and Nutritional Science, University of Alberta 410 Agriculture/Forestry Centre Edmonton, AB T6G 2P5 Canada *Corresponding Author: Dr. M. Betti E-mail: [email protected] Tel: (780) 248-1598

ACS Paragon Plus Environment


Page 2 of 34

1

ABSTRACT

2

Sous-vide is increasingly popular method of cooking under controlled conditions of temperature

3

and time inside vacuumed pouches to preserve the nutritional and sensory qualities of food.

4

Sous-vide non-enzymatic browning of glucosamine (GlcN) was investigated at 50, 60 and 70°C

5

for 12 h. Changes investigated were pH, color, level of browning and in the concentration of the

6

key Maillard and caramelization reaction products, including α-dicarbonyls and pyrazines. The

7

concentrations of undesired 4-methylimidazole (4-MEI), 2-acetyl-4(5)-tetrahydroxybutyl

8

imidazole (THI) and 5-hydroxymethylfurfural (5-HMF) were also determined. Six types of

9

caramels were produced of unique composition with no detectable levels of 4-MEI. GlcN

10

caramels produced under vacuum were more acidic and lighter in color, containing significantly

11

less flavorful diacetyl, but more fructosazine (FR) as compared to non-vacuum caramels. THI

12

concentration was well below the toxicity levels for all studied caramels. Principal component

13

analyses showed that the incubation temperature played a key role in determining the

14

composition of caramels.

15

16 17 18 19

KEYWORDS: Glucosamine, Sous-vide, Caramel, Vacuum, Fructosazine, Alpha-dicarbonyls, Diacetyl, 2-Acetyl-4(5)-tetrahydroxybutyl imidazole, 4-Methylimidazole, 5Hydroxymethylfurfural.

20 21 22 23 24 25 26 27 28 29 2 ACS Paragon Plus Environment

Page 3 of 34

30


INTRODUCTION

31

Glucosamine (GlcN, 2-amino-2-deoxy-D-glucose) is a monosaccharide obtained by the

32

tandem hydrolysis–deacetylation of chitin, an abundant biopolymer found in the exoskeletons of

33

crustaceans, insects and the cell walls of fungi. GlcN also can be classified as a Heyns product

34

resulting from the reaction between fructose and ammonia or amino acid, however in much

35

lower yields as compared to the thermochemical degradation of chitin. For example, by using

36

optimized extraction conditions a 96−98% yield of glucosamine hydrochloride can be achieved

37

from marine shrimps’ chitin within 4 h.1

38

In previous studies2,3 GlcN is shown to be an unstable amino sugar which rapidly

39

degrades generating a plethora of compounds, including α-dicarbonyls (α-DCs). Among the

40

major α-DCs, 3-deoxyglucosone (3-DG), glucosone (G), methylglyoxal (MGO), glyoxal (GO)

41

and diacetyl (DA) can be generated from GlcN at as low as 25°C. Increasing the temperature to

42

37 or 50°C can speed up the degradation process; one kilogram of GlcN can generate up to 5.7

43

gram of α-DCs by heating at 37°C for 12 days. Besides being pivotal precursors of colour,

44

flavour and aroma in foods, some of these α-DCs have some other important activities, such as

45

“reductone”

46

deoxyglucosone.3 From the health point of view, endogenously formed α-DCs (ie.

47

methylglyoxal, glyoxal and 3-deoxyglucosone) resulting from glucose-derived modification of

48

proteins have been shown to induce diabetes.7 The mechanism to generate these α-DCs from

49

GlcN has been proposed.2 Glucosone, for instance, may be generated via an oxidative

50

mechanism, while 3-DG is formed through the 1,2-enolization of GlcN. Isomerization of GlcN to

51

Amadori compounds through an interconversion reaction provides another several mechanisms

52

of some α-DCs formation. Along with α-DCs, GlcN generates non-volatile polyhydroxyalkyl

glucosone4 or

antibacterial

methylglyoxal,5 glyoxal,5 diacetyl6

and

3-

3 ACS Paragon Plus Environment


Page 4 of 34

53

pyrazines, namely fructosazine (FR) (2,5-bis(D-arabino-tetrahydroxybutyl)pyrazine) and

54

deoxyfructosazine

55

trihydroxybutyl) pyrazine); one kilogram of GlcN can generate up to 370 g of these pyrazines

56

when heated at 50°C for 2 days. These pyrazines have been recognized as flavouring agents and

57

have been identified in roasted peanuts, caramel and soy sauce.8 Fructosazine also possesses

58

some antimicrobial activity against heat-resistant E. coli AW 1.7 in moderate acidic conditions.9

59

These molecules are formed as a result of the symmetric cyclocondensation of two GlcN

60

molecules

61

tetrahydroxybutyl)dihydropyrazine)]. The latter oxidizes to form fructosazine or dehydrates to

62

generate a deoxyfructosazine.2 In addition to their application as food ingredients, these

63

molecules are gaining interest for their uses in human therapeutics such as in the treatment of

64

type II diabetes, the prevention of atherosclerosis, and in the prevention of the pathological

65

cartilage degradation and other inflammatory diseases.10,11 Hence, GlcN can not only develop

66

flavourful caramel solutions due to diacetyl and pyrazines production, but also has the potential

67

to become a functional food/ingredient due to the bioactivity of FR and DOFR.

which

(DOFR)

follow

(2-(D-arabino-tetrahydroxybutyl)-5-(D-erythro-2,3,4-

dehydration

form

dihydrofructosazine

[2,5-bis(D-arabino-

68

Since GlcN produces flavour, antimicrobial, antioxidant and bioactive health

69

compounds11 in addition to the fact that it can be purified by the uncomplicated deacetylation of

70

chitin, gives us opportunities to research its various applications and to carefully study the non-

71

enzymatic browning mechanism of this compound under different conditions.

72

Caramel colors account for more than 80% of food colorants used in food and beverages

73

industries. Depending on the reactants used in the manufacturing process, the industrial caramel

74

color is classified into four classes. Among these, Caramel Colors III and IV are produced by

75

heating reducing sugars in the presence of ammonium compounds (III) and, additionally, sulfite


Page 5 of 34


76

(IV).13 During caramelization these ammonium compounds serve as a source of nitrogen for a

77

series of undesired neo-formed food contaminants, including a group of toxic imidazoles, such as

78

4-methylimidazole (4-MEI) and 2-acetyl-4(5)-(tetrahydroxybutyl)imidazole (THI). In 2011,

79

4(5)-MEI was identified as being “possibly carcinogenic to humans”,14 while THI was reported

80

to exert immunosuppressive effects;14,15 these caused regulatory agencies to focus on their

81

presence in foods and beverages.16 The allowable level of 4-MEI is restricted to 250 ppm by the

82

European Commission,17 whereas California’s Proposition 65 law requires that beverages

83

containing 4-MEI concentrations corresponding to an exposure of 29 µg/day carry a warning

84

label.18 For Class III Ammonia Caramel, a Joint FAO/WHO Expert Committee on Food

85

Additives specified a maximum level for THI of 25 mg/kg caramel colour, whereas the European

86

Commission sets the maximum at 10 mg/kg on an equivalent color basis (i.e., a color intensity of

87

0.1 absorbance unit).19 Hydroxymethylfurfural (HMF), is another characteristic heterocyclic

88

product of non-enzymatic browning and is a ubiquitous food contaminant. The formation of

89

HMF from sugar dehydration or by caramel color addition is a potential issue. GlcN can be

90

easily deaminated,2 and released ammonia can react with the α-DCs produced during enolization

91

and retroaldolization reactions, which in turn can possibly generate 4-MEI and THI. Our

92

intention, so far, has been the production of GlcN caramel solution using moderate temperatures

93

(40-60°C) so that the production of these toxicants can be minimized.

94

Sous-vide, is the increasingly popular method of cooking using vacuumized pouches at

95

mild temperatures and long time in a circulated water bath.20 Oxygen-free atmosphere is intended

96

to prevent the oxidation processes and thus help preserving not only the nutritional quality of

97

food but also to improve its sensory qualities in terms of aroma, flavor, and texture.20,21 An

98

oxygen-free environment would also reduce the reaction between triplet oxygen (3O2) and other



Page 6 of 34

99

excited molecules in the triplet state thus reducing the browning.22 Up to now there is no study

100

that has examined the non-enzymatic reaction of Heyns compound in a vacuum. Therefore, the

101

objective of this research was to study the chemistry of non-enzymatic browning of GlcN under

102

vacuum condition (sous-vide technology) at different temperatures (50, 60 and 70°C) and

103

evaluate the physico-chemical properties, and the generation of α-DCs, hydroxylalkylpyrazines

104

and alkylimidazoles. From a practical point of view, it is of interest to understand if mild

105

temperatures under vacuum conditions increases or decreases the production of certain aromatic

106

molecules (i.e. the butter- and caramel-like diacetyl odorant), while minimizing the production of

107

the undesirable heterocyclic compounds (i.e. 4-MEI, THI and HMF).

108

MATERIALS AND METHODS

109

Chemicals. D-glucosamine hydrochloride (GlcN, ≥99%), HPLC grade solvents (methanol,

110

formic acid), o-phenylenediamine (OPD; 99.5%), glucosone (G, 2-keto-D-glucose; ≥98%),

111

methylglyoxal (MGO, 2-oxopropanal, 40% in H2O) and glyoxal (GO; (ethanedial, 40% in H2O),

112

4(5)-methylimidazole (4(5)-MEI, 98%), 5-(hydroxymethyl)furfural (HMF ≥ 99%) and an

113

ammonia assay kit were from Sigma-Aldrich (St. Louis, MO, USA). 3-deoxyglucosone (3-DG;

114

3-deoxy-D-erythrohexosulose; ≥ 95%) and 2-acetyl-4(5)-tetrahydroxybutyl imidazole (THI, ≥

115

95%) were from Cayman Chemical (Ann Arbor, MI, USA). Diacetyl (DA; 2,3-butanedione;

116

99%) was from Acros Organics (NJ, USA). Fructosazine (FR) and deoxyfructosazine (DOFR)

117

were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Sodium 1-octanesulfonate (99%)

118

was from Alfa Aesar (Ward Hill, MA, USA). Ammonium hydroxide and potassium dihydrogen

119

phosphate (99.4%) were from Fisher Scientific (NJ, USA). SPE tC-18 Sep-Pak Vac 6 cc

120

columns were from Waters (Milford, MA, USA). Poly(vinylidene fluoride) (PVDF) syringe

121

filters (0.22 µm) and filtration membranes (0.1 µm) were from Millipore (Billerica, MA, USA).


Page 7 of 34


122

The buffers and solutions were prepared with Milli-Q purified distilled water (Millipore,

123

Bedford, MA, USA).

124

Experimental Design. A 3 × 2 factorial design was planned to study the influence of the

125

temperature (50, 60, 70°C), “level of oxygen” (vacuum vs. non-vacuum) during 12 h incubation

126

on physico-chemical properties, generation of α-DCs, and heterocyclic compounds (non-volatile

127

polyhydroxylalkyl pyrazines, HMF, THI and 4-MEI) production during GlcN non-enzymatic

128

browning. For each treatment, 3 vacuum bags containing GlcN solutions were incubated in the

129

water bath circulator at the three different temperatures. Three independent trials were conducted

130

at three different days, resulting in a total number of observation of 54 (9 observations per

131

treatment; n = 9).

132

Preparation of GlcN Solutions. GlcN caramels were obtained by heating aqueous solutions of

133

GlcN (150 g/L) for 12 h. Prior to incubation the pH of the solutions was adjusted to 7.0 ± 0.01

134

with 1 M NaOH. Ten mililiters of the solutions were transferred to the vacuum sealing pouches

135

with oxygen barrier (FoodSaver, Brampton, ON, Canada) and heat-sealed (FoodSaver Vacuum

136

Sealer V4420, Brampton, ON, Canada) to make vacuum condition, whereas GlcN solutions were

137

sealed without vacuum in vacuum sealing plastic bags to make non-vacuum samples. The bags

138

were completely submerged and randomly placed in the water bath circulator (Haake SC100,

139

Thermo Scientific, Waltham, MA, USA) and incubated at 50, 60 and 70°C. The pH was not

140

adjusted during incubation. After retrieval, the bags were cooled on ice and transferred to screw

141

cap tubes. An aliquot of each of the GlcN caramel solutions were immediately tested for

142

respective analyses.

143

Color Measurements and pH. The color of GlcN caramel solutions was determined using a

144

tristimulus colorimeter (Minolta CR-400, Konica Minolta Sensing Americas, Inc., Ramsey, NJ)



Page 8 of 34

145

according to Hong et al.2 The instrument was calibrated before each series of measurements

146

using a white tile plate (L* = 32.80; a* = 14.51; b* = 15.19). Chromaticity results are expressed

147

in L*, a* and b* coordinates. Chroma (C*) and hue angle (H°) were calculated using the

148

following formulas, C* = (a*2 + b*2)1/2 and H° = arctan (b*/a*), respectively.

149

Spectrophotometric measurements were conducted with the absorbance recorded using a 1 cm

150

quartz cuvette at 420 nm on a Spectramax M3 multi-mode microplate reader (Molecular

151

Devices, Sunnyvale, CA).

152

A pH meter (UB-10, Ultra basic pH meter, Denver Instrument, Bohemia, NY, USA) was used to

153

monitor the pH of the GlcN caramel solutions.

154

HPLC and Mass Spectrometric Analysis of Free α-Dicarbonyl Compounds. For solid phase

155

extraction, pre-column derivatization, MS identification and HPLC quantitation of G, 3-DG,

156

MGO, GO and DA the previously published method was used.1,2 Concentrations of individual α-

157

DC were determined by the external standard method. Standard curves were constructed using

158

five different concentrations of the standards. The correlation coefficients for all calibration

159

curves were R2 ≥ 0.99. The average limits of detection (LODs) were calculated as 3.6 ± 0.4(G),

160

1.8 ± 0.3(3-DG), 1.3 ± 0.08(GO), 0.5 ± 0.0 (MGO) and 0.6 ± 0.0 µg/mL (DA) and the average

161

limits of quantitation (LOQs) were 10.9 ± 1.3(G), 5.5 ± 1.3(3-DG), 4.2 ± 0.7 (GO), 1.6 ± 0.1

162

(MGO) and 1.8 ± 0.1 µg/mL (DA) where signal-to-noise ratios (S/N) were 3.3:1 and 10:1 for

163

LOD and LOQ, respectively.

164

Analysis of Heterocyclic Compounds. Fructosazine (FR) and Deoxyfructosazine (DOFR).

165

HPLC and MS/MS. Analysis of non-volatile FR and DOFR were performed using the same

166

method as described before.2 To quantitate FR and DOFR the standard curves (five points) were


Page 9 of 34


167

constructed with an R2 ≥ 0.99. The LODs and LOQs for FR were 1.40 ± 0.00 and 4.24 ± 0.03

168

µg/mL, respectively and for DOFR were 0.05 ± 0.01 and 0.15 ± 0.08 µg/mL, respectively.

169

HMF, THI and 4-MEI. Identification. HPLC-MS/MS analyses were used to identify the presence

170

of HMF, THI and 4-MEI in GlcN caramels. HPLC with tandem mass spectrometric (MS)

171

detection was conducted on a HPLC-DAD-ESI/MS instrument equipped with an electrospray

172

ionization (ESI) source interfaced to a QTRAP 4000 mass spectrometer (AB Sciex, ON,

173

Canada). LC was run on an Agilent 1200 HPLC system (Agilent, Palo Alto, CA, USA) with a

174

degasser, a quaternary pump, a thermostated autosampler, and a UV-visible detector. The MS

175

procedures were followed as described by Wang et al.23 and were performed using the reversed-

176

phase column Ascentis Express Peptide ES-C18 (150 × 4.6 mm, 2.7 µm particle size; Sigma-

177

Aldrich). The samples were eluted with (A) 0.05% ammonia in water and (B) 5% acetonitrile

178

with a gradient programmed as: 5% B (0 min), 5% B (3 min), 40% B (5 min), 5% B (5.1 min)

179

and 5% B (6.5 min) at flow rate of 0.3 mL/ min. The sample injection volume was 10 µL. The

180

mass spectrometer was operated in a selected reaction monitoring (SRM) mode. The effluent

181

from the LC was directly introduced with a heated ESI probe operated in the positive mode into

182

the mass spectrometer. The acquisition was performed at spray voltage 3000 V, capillary

183

temperature 350ºC, sheath and aux gas pressure 30 and 10, respectively.

184

Quantitation. The concentrations of HMF and THI were determined as described by Ciolino.24

185

The chromatographic separations were performed using an Agilent 1100 system (Agilent

186

Technologies, Inc., Santa Clara, CA, USA) consisting of a G-1312 binary pump, a G-1328A

187

injector, a G-1322A degasser, and a G-1315A photodiode array detector (PDA), equipped with

188

an Ascentis Express ES-C18 column. The mobile phase was a binary mixture of (A) 0.05 M

189

potassium dihydrogen phosphate and 0.005 M sodium octane sulfonate, adjusted to a pH of 3.0 ±



Page 10 of 34

190

0.01, and (B) 100% methanol. The mobile phase consisted of 92.5:7.5 A:B and detection was

191

performed at 285 nm. The injection volume was 10 µL and flow rate 0.5 mL/min. The analytes

192

were filtered with a PVDF syringe filter (13 mm, 0.22 µm; Millipore Millex, Billerica, MA,

193

USA). The quantitation was achieved using a 5-points standard curves with an R2 ≥ 0.99. The

194

LODs were determined as 1.4 ± 0.06 µg/mL (THI), 0.7 ± 0.03 µg/mL (HMF) and the LOQs were

195

4.2 ± 0.2 µg/mL (THI) and 2.1 ± 0.17 µg/mL (HMF). Data acquisition and processing were

196

performed with Agilent ChemStation software.

197

Ammonia Detection Assay. The ammonia was detected using commercially available kit

198

(Sigma, St. Louis, MO, USA), according to the manufacturer’s instructions. The assay is based

199

on the reaction of ammonia with α-ketoglutaric acid and reduced nicotinamide adenine

200

dinucleotide phosphate (NADPH) in the presence of L-glutamate dehydrogenase to form L-

201

glutamate and oxidised NADP+. The oxidation of NADPH to NADP+ results in a decrease in the

202

absorbance at 340 nm that is proportional to the concentration of ammonia.

203

Statistical Analysis. The data was analysed as a 3 × 2 factorial ANOVA using the PROC

204

MIXED procedure of SAS (v. 9.3, SAS Institute Inc., Cary, NC, USA). The model tested the

205

main effect of vacuum and temperature as well their interaction and used the day of trial

206

replication as a random variable. Tukey’s honestly significant difference (p < 0.05) multiple-

207

range test was conducted to determine differences between the means. A principal component

208

analysis (PCA) was conducted using OriginPro 8.6 software (OriginLab Corporation, MA, USA,

209

2012).

210 RESULTS AND DISCUSSION 211 Chemico-Physical Characteristics of GlcN Caramel Solutions. As shown in Table 1 a higher 212 temperature, in general, resulted in a greater drop in pH. However, GlcN caramels produced under


Page 11 of 34


213 vacuum were significantly more acidic as compared to the treatments under non-vacuum, and this was 214 more evident at 50°C (4.2 vs. 4.5 in vacuum vs non-vacuum, respectively). Generation of formic and 215 acetic acids during GlcN incubation in the presence of oxygen, causing a decrease in pH, has been 2

216 reported before. Greater acidity of vacuum-treated samples is most likely due to the different 217 degradation pathways leading to the formation of different types and/or concentrations of the major 218 degradation products, which will be discussed in the following parts. 219

Tristimulus colorimetry was used to visualize and integrate different dimensions of the color

220 space. The colorimetric parameters, L* (lightness, black (0)-white (100)), a* and b* representing red221 green and yellow-blue, respectively were determined. The a* and b* values are reported in the Table S1 222 in the Supporting Information. The CIE values of a* and b* were then transformed into the H° and C*. 223 With regard to the main effect of temperature (Table 1), the lightness of GlcN caramels significantly 224 decreased with increased incubation temperatures. This is expected since higher temperatures usually 225 produce darker caramels or Maillard reaction systems.

25

However, changes in L* values not necessarily

226 correlate with the visually observed browning.26 Therefore, the color was also expressed by means of 227 the chroma (C*, metric chroma) and hue angle (H°, chromatic tonality). The latter acquires measures of 228 redness at values near 0° and yellowness near 90°.25 The significant effect of vacuum, temperature and 229 their interactions was found for hue angle values. In general, for all treatments the values were in the 230 range between 13.9 to 53°, indicating orange-red to yellow hue of GlcN caramels. The least (p < 0.05) 231 hue value of 13.9° was observed in non-vacuum GlcN caramels incubated at 50°C followed by the 232 vacuum 60°C treatments (14.5°) representing reddish tonality of these treatments. GlcN solutions 233 incubated under vacuum at 50 and 70°C were also reddish with the hues of 20.1 and 21.4°, respectively. 234 The non-vacuum 70°C GlcN caramels had a hue of 35.5°, denoting orange tonality. The greatest value 235 of hue (p < 0.05) was observed in 60°C non-vacuum samples representing an orange-yellowish tonality.



Page 12 of 34

Chroma is a measure of color saturation or intensity, and is defined by the magnitude of the

236

237 vector at each point designating the departure from dull to more vivid chromatic color (“–“ to “+” 238 values).

27

Non-vacuum GlcN caramels had significantly smaller chroma values, indicating their lower

239 vividness or color intensity as compared to vacuum-produced caramels. Temperature also significantly 240 affected chroma, where increasing temperature significantly decreased chroma values for both vacuum 241 and non-vacuum caramels. These results indicate that vacuum conditions and lower incubation 242 temperature generate caramels with the greatest color vividness. Both caramelization and the Maillard 243 reaction are responsible for the formation of browning compounds (ie. melanoidins) absorbing at 420 244 nm.

28

Results outlined in Table 1 show that caramels produced under vacuum had significantly less

245 absorbance at 420 nm as compared to non-vacuum samples, indicating less browning intensity of 246 vacuum caramels. Melanoidin production occurs with consumption of oxygen,29 therefore it is expected 247 that caramel solutions prepared in the vacuum condition absorbed less at 420 nm. Interestingly, that an 248 increase in temperature did not affect the absorbance at 420 nm in vacuum treatments; whereas 249 temperature affected non-vacuum samples only between 50 and 70°C. These results agree with the 250 study of Kanner and Shapira22 who found less non-enzymatic browning of grape fruit juice packaged 251 with less oxygen. 252 Analysis of α-Dicarbonyl Compounds. Analysis of the α-DCs content of GlcN yielded five major 253 compounds,

supporting previous findings on GlcN browning.1,2 Representative HPLC-UV

254 chromatograms and MS/MS identification of α-DC are shown in the supplementary information file 255 (Figures S1 and S2). 3-DG was found to be the dominating α-DC and its concentration was 1.7-times 256 greater in non-vacuum samples (Table 2). The temperature also had a significant effect on 3-DG, where 257 under non-vacuum conditions higher temperatures resulted in significantly less 3-DG concentration. 258 Under vacuum no significant effect was found between treatments at 50 and 70°C. The non-oxidative


Page 13 of 34


259 mechanism of 3-DG formation from GlcN through 1,2-enolization has been previously proposed.1 A 260 significant decrease of 3-DG with higher temperatures under non-vacuum is most likely due to the 261 progress of the reaction including C3-C4 or C4-C5 cleavage and dehydration generating MGO, GO, 1

262 HMF and/or 3,4-dideoxyglucosone-3-ene (3,4-DGE). Indeed, the presence of not only MGO, GO and 263 HMF, but also 3,4-DGE was identified in GlcN caramels (Figure S2). G is an important α-DC as it acts 264 as a reductone and as a radical scavenger.4 It is therefore important to monitor the level of G in caramel 265 solutions. The concentration of G was significantly greater in vacuum treatments in comparison to non266 vacuum (512 vs 264 mg/L, respectively) and under both conditions its concentration significantly 267 decreased as a function of temperature (Table 2). G was proposed to be generated oxidatively from 268 GlcN,2 therefore finding its greater concentration under vacuum conditions was initially surprising. 269 However, it has been reported that reductone molecules like G consume oxygen during non-enzymatic 270 browning reaction possibly forming melanoidins;

30

therefore in a more oxidative environment (i.e. non-

271 vacuum treatment) G would consume oxygen forming more melanoidins. The results of absorbance at 272 420 nm (Table 1) supports this hypothesis. However, Gobert and Glomb31 while studying the 273 degradation of glucose in the presence of lysine for at 50°C, found greater amount of G under aerated 274 condition. Smuda and Glomb32 also reported greater concentrations of G during incubation of maltose 275 in the presence of lysine under aerated conditions. Possible reasons for discrepancies between these 276 studies might be due to the fact that the glucose/maltose-lysine reaction systems would yield lower 277 quantity of Amadori compound compared to an already formed Heyns compound obtained from chitin 278 hydrolysis and deacetylation. The other reason could be due to another mechanism involved in GlcN 279 degradation under vacuum. In addition, vacuum conditions were achieved differently in the previous 280 studies

31,32

compared to the current. The significant decrease of G with increased incubation

281 temperatures was presumably the result of its cleavage at C2-C3 or C3-C4 generating GO or



Page 14 of 34

282 hydroxypyruvaldehyde, respectively. GO concentration was significantly greater under vacuum, the 283 same as was one of its precursors, G. Its concentration significantly increased with increased incubation 284 temperature in both vacuum and non-vacuum treatments. MGO and DA were found in significantly 285 greater concentrations in non-vacuum treatments, being on average 1.9 and 1.3-times greater 286 respectively to those found in caramels produced under vacuum. As in the case of GO, the 287 concentrations of MGO and DA significantly increased with higher incubation temperatures (Table 2); 288 suggesting that higher temperatures facilitate the degradation of a long chain α-DCs 3-DG and G into 289 the short chain GO, MGO and DA. A more oxidative environment (non-vacuum) promotes the 290 formation of odorant molecules like diacetyl at a level of 14 ppm and the brown melanoidins. 291 Effect of Vacuum vs Non-vacuum Condition on FR and DOFR Levels. Fructosazine and 292 deoxyfructosazine are the major products of GlcN autocondensation. These molecules posses both 293 bioactive and functional properties. For instance, they possess anti-inflammatory activity against 10

294 diabetes and cartilage degradation.

At the same time, the so-called “colourless caramel” which is a

295 caramel extract containing a relatively large amount of FR and DOFR, can also be used to protect beer 296 from the phenomenon of UV-light induced off-flavour generation.33 The representative HPLC-UV 297 chromatograms used for FR and DOFR quantitation and the MS/MS spectra that was used to verify the 298 identification of these non-volatile pyrazines are shown in the Supplementary Information (Figures S3 299 and S4). Results reported in Table 3 show that GlcN caramel solutions generated under vacuum 300 contained significantly more FR as compared to those produced with no vacuum, however no 301 significant effect on DOFR concentration was found between these treatments. An increase in 302 temperature significantly decreased the concentrations of both FR and DOFR in vacuum and non303 vacuum treatments. FR is generated during GlcN’s double dehydration followed by oxidation, therefore 304 its greater concentration in vacuum samples is surprising. It is possible that FR is degraded in a more


Page 15 of 34


305 oxidative environment, forming new derived pyrazines. A decrease in FR and DOFR concentration with 306 higher temperatures is most likely due to acidification of GlcN caramel solutions (Table 1). Wu et al.34 307 showed that pH 6-8 are favorable for larger FR and DOFR yields; this agrees with our results where the 308 formation of FR and DOFR was greater at 50°C treatments, which were significantly less acidic than 309 treatments at 60 and 70°C. In summary, a lower temperature (50°C) and vacuum condition promote the 310 formation of these bioactive compounds. 311 Changes in Concentration of HMF. The generation of HMF during the Maillard reaction and 312 caramelization (i.e. fructose) is well-established; however, HMF was not identified and quantitated 313 before during caramelization of GlcN. Representative HPLC-UV chromatograms of HMF are shown in 314 Figure 1, where a well resolved peak of HMF was observed and was eluted at the exactly the same time 315 as the HMF standard. Additional HPLC-UV chromatograms for the other treatments can be found in the 316 Supplementary Information (Figure S5). For the unambiguous identification of HMF in GlcN caramels, +

317 HPLC-MS/MS was performed, where HMF yielded protonated molecular ions [M+H] at m/z 127.1 and 318 the major fragment ions at m/z 109.1 and 81.1 (Table 4, Figure 1). This fragmentation pattern 319 corresponded to the one obtained from HMF standard and consistent to the previously reported 320 elsewhere,35 allowing for correct identification. The concentration of HMF was significantly affected by 321 absence of oxygen during GlcN caramelization and was on average 1.8-times less in vacuum conditions 322 (Table 3). An increase in temperature of caramelization from 50 to 70°C resulted in an increased 323 generation of HMF in both vacuum and non-vacuum treatments. As mentioned before, HMF is formed 324 from its precursor, 3-DG, upon dehydration, therefore most likely that higher temperatures of incubation 36

325 favored 3-DG’s dehydration. Grainger et al.

reported a weak negative correlation between pH and

326 HMF and no correlation between MGO and HMF in Manuka honey. In this study a negative correlation 327 (r = -0.78, p < 0.01) was observed between pH and HMF and a positive correlation (r = 0.77, p < 0.01)



Page 16 of 34

328 between 3-DG and HMF formation (Table S2). Shallenberger and Mattick37 showed that at pH 3, the 329 rate of HMF formation from fructose was approximately double from that at pH 4-6; while at pHs 2 and 330 1 it was about ten and nearly forty times as rapid, respectively. In this study the caramels produced 331 under non-vacuum conditions were significantly more acidic that those produced under vacuum (Table 332 1), which could be one of the reasons for the greater HMF generation under non-vacuum. To the best of 333 our knowledge, no HMF limits in caramels has been set. The levels of HMF found in commercial 334 caramels vary considerably and are dependent on caramel type; where for instance Caramel type I range 335 is 700-2700 mg/kg HMF.

17

The results from this study showed that the HMF concentrations in GlcN

336 caramels produced under vacuum or non-vacuum at 50-70°C were well less than the concentrations 337 found in commercial caramels. 338 Identification of THI and 4-MEI. THI is an alkylimidazole formed as a by-product during thermal 339 processing resulted from addition of ammonia caramel colorants, Class III and IV caramel colors.

17,38

340 THI was also identified in GlcN caramel solution produced in this study under both vacuum and non341 vacuum conditions. The representative HPLC-UV chromatograms showed the peak of THI eluted 342 before the HMF (Figure 1 C, D). For unambiguous THI identification its MS/MS spectra was acquired 343 and compared to the standard THI solution, where the same mass fragments were found, confirming 344 peak identification. Table 4 shows that significantly more THI was formed under vacuum conditions. 345 On average 3.3 mg/L of THI was formed in vacuum treatments, while 1.3 mg/L was found in non346 vacuum treatments. Higher temperatures also favored the formation of THI under both vacuum and 347 non-vacuum conditions. The greatest concentration of THI was formed in GlcN under vacuum caramels 17

348 produced at 70°C and was 4.7 mg/L; this concentration was still below the recommended level. It has 349 been proposed

39

that the formation of THI during the caramelization of glucose in the presence of

350 ammonia involves the condensation of fructosamine and methylglyoxal, two products formed via the


Page 17 of 34


351 Amadori rearrangement and alkaline degradation of glucose, respectively. To verify this mechanism, the 352 concentration of ammonia was also determined and showed a significantly less generation of ammonia 353 in vacuum treated GlcN caramels at each incubation temperature (Figure S6). Since significantly more 354 THI was formed in samples produced under vacuum, lower ammonia concentration detected in the 355 same treatments suggest its greater involvement in THI formation. Significantly less MGO was also 356 found in vacuum treated samples (Table 2), which may also imply its involvement in THI formation. On 357 the basis of our experimental evidence, we propose the mechanism of THI formation during GlcN 358 browning (Figure 2), where in the first step GlcN molecule condensation with ammonia results in 359 formation of imine form of GlcN existing in equilibrium with its eneamine form. The reaction of 360 enediamine with MGO and further rearrangements will form THI. Depending on which among C1-NH2 361 or C2-NH2 reacts with MGO, 5-THI or 4-THI, is formed, respectively. No 4-MEI was identified in 362 GlcN caramels produced under the conditions of this study (Figure S7). This is very important result as 40

363 4-MEI is a major concern in caramel-containing foods and beverages. A previous study

indicated that

364 GlcN caramel produced at 120°C for 2 h generate 4-MEI; the moderate temperature used in this study 365 allowed to produce 4-MEI-free caramel with a greater content of the aromatic diacetyl. 366 Multivariate Analyses: Principal Components. Pooling the full set of analytical analyses enabled 367 performing a principal component analyses, which showed that data could be distinguished into six 368 separate groups (Figure 3), indicating that each caramel solution was of unique composition. Two 369 principal components were extracted from the data showing 58.68% (PC1) and 29.52% (PC2) of the 370 variation (Figure 3, Table S3); implying that 88.2% of the total variance in the thirteen dependent 371 variables determined could be condensed into two PCs (Table S3). PC1 had relatively large loadings of 372 FR, G, DA, L* and C*, while 3-DG, THI, GO and MGO had large loadings on the PC2 (Table 5). 373 Overall, PC2 was able to discriminate the caramel solutions produced under vacuum (4, 5, and 6) from



Page 18 of 34

374 caramel solutions produced in more oxidative condition (non-vacuum treatments), as the firsts are 375 located in higher quadrants and the latter in the lows (Figure 2). On the other hand, PC1 is separating 376 the caramel solutions based on the non-enzymatic browning temperature (Figure 2). Therefore, FR, G, 377 DA and color characteristics (higher loading coefficients in PC1) can be used as possible markers to 378 discriminate non-enzymatic browning temperature, while 3-DG, THI, GO and MGO (higher loading 379 coefficients in PC2) can be used as markers to discriminate between vacuum and non-vacuum 380 treatments. Also, the small angle between the positions of 50°C non-vacuum and 60°C non-vacuum 381 (caramels 1 and 2, Figure 3) treatment conditions indicate that these two groups have similar response 382 patterns across the GlcN browning. The plot also allows to see associations between the different 383 variables, where for instance HMF, MGO and H° have similar response patterns, and are most strongly 384 distinguished by incubation at 50°C and 60°C under non-vacuum. Likewise, G, C* and FR show similar 385 response pattern and are strongly differentiated by 70°C vacuum condition. G, C*, FR, GO and THI are 386 located on the lower quadrants which indicates the inverse relationship between factor and variable. For 387 example, GO and THI are negatively correlated to 50°C vacuum condition, while G, C* and FR are 388 negatively correlated to 70°C vacuum treatment condition. 389 In conclusion, this study showed that the level of oxygen and temperature of incubation both play 390 significant roles in determining physico-chemical properties and composition of GlcN caramel 391 solutions. Combinations of different temperatures and vacuum/non-vacuum conditions result in GlcN 392 caramel solutions that possess different acidity, browning level, and concentration of flavouring (i.e. 393 pyrazines and diacetyl) and undesirable (HMF, THI) compounds. The treatments generated six unique 394 caramels. In general, GlcN caramels produced under vacuum were slightly more acidic, lighter (less 395 absorbance at 420 nm) and of a more intense color compared to those generated under non-vacuum. In 396 terms of flavor profile, GlcN caramels under vacuum contained 1.3-times less DA, but almost 3-times


Page 19 of 34


397 more FR as compared to non-vacuum caramels. Vacuum caramels also contained less HMF, but more 398 THI, which in either treatment were significantly less from those required by regulations. It was found 399 that FR tend to be heat and oxygen unstable and decreased proportionally with an increased temperature 400 of incubation. The opposite was found for DA, where its concentration increased with increasing 401 temperatures. As for undesired HMF and THI, an increase in the temperature of incubation resulted in 402 an increased concentration of both compounds. No 4-MEI was found in any of the GlcN caramels tested 403 in this study. In accordance with the results, it is suggested to use a lower incubation temperature to 404 minimize the formation of undesired HMF and THI, while still retaining high amount of flavouring 405 agents DA and FR. The caramel solutions obtained in this study, particularly the ones with greater 406 amount of FR, have the potential to be used in beer production against UV light-induced off-flavor 407 generation as FR posses a strong light absorption in the UV-B range. Furthermore, the low pH (due to 3

408 acetic and formic acids production ) and dark colour pave the possibility to produce a “chemical 409 balsamic vinegar” from GlcN. 410 411

REFERENCES

412 413 414

1. Mojarrad, J.S.; Nemati, M.; Valizadeh, H.; Ansarin, M.; Bourbour, S. Preparation of glucosamine from exoskeleton of shrimp and predicting production yield by response surface methodology. J. Agric. Food Chem. 2007, 55, 2246-2250.

415 416 417

2. Hrynets, Y.; Ndagijimana, M.; Betti, M. Studies on the formation of Maillard and caramelization products from glucosamine incubated at 37°C. J. Agric. Food Chem. 2015, 63, 6249-6261.

418 419 420

3. Hrynets, Y.; Bhattacherjee, A.; Ndagijimana, M.; Hincapie Martinez, D.J.; Betti, M. Iron (Fe2+)-catalyzed glucosamine browning at 50°C: Identification and quantification of major flavor compounds for antibacterial activity. J. Agric. Food Chem. 2016, 64, 3266-3275.

421 422

4. Kanzler, C.; Haase, P.T.; Kroh, L.W. Antioxidant capacity of 1-deoxy-D-erythro-hexo-2,3diulose and D-arabino-hexo-2-ulose. J. Agric. Food Chem. 2014, 62, 2837-2844.

423 424 425

5. Mavric, E.; Wittmann, S.; Barth, G.; Henle, T. Identification and quantification of methylglyoxal as the dominant antibacterial constituent of Manuka (Leptospermum scoparium) honeys from New Zealand. Mol. Nutr. Food Res. 2008, 52, 483-489. 19 ACS Paragon Plus Environment


Page 20 of 34

426 427

6. Jay, J.M.; Rivers, G.M. Antimicrobial activity of some food flavoring compounds. J. Food Saf. 1984, 6, 129-139.

428 429

7. Vlassara, H.; Bucala, R. Recent progress in advanced glycation and diabetic vascular disease: role of advanced glycation end product receptors. Diabetes. 1996, 45, S65-S66.

430 431 432

8. Henry, N.; Delépée, R.; Seigneuret, J.M.; Agrofoglio, L.A. Synthesis of water-compatible imprinted polymers of in situ produced fructosazine and 2,5-deoxyfructosazine. Talanta. 2012, 99, 816-823.

433 434 435

9. Bhattacherjee, A.; Hrynets, Y.; Betti, M. Fructosazine, a polyhydroxyalkylpyrazine with antimicrobial activity: Mechanism of inhibition against extremely heat resistant Escherichia coli. J. Agric. Food Chem. 2016, 64, 8530-8539.

436 437 438

10. Giordani, A.; Letari, O.; Stefano, P.; Roberto, A.; Walter, P.; Gianfranco, C.; Claudio, R. L. 2,5-bis(tetrahydroxybutyl)pyrazines for the treatment of osteoarthritis and rheumatoid arthritis. 2006. European Patent Application. Bulletin 2006/39.

439 440 441

11. Zhu, A.; Huang, J.B.; Clark, A.; Romero, R.; Petty, H.R. 2,5-Deoxyfructosazine, a Dglucosamine derivative, inhibits T-cell interleukin-2 production better than D-glucosamine. Carbohydr. Res. 2007, 342, 2745-2749.

442 443 444

12. Hong, P. K.; Betti, M. Non-enzymatic browning reaction of glucosamine at mild conditions: Relationship between colour formation, radical scavenging activity and α-dicarbonyl compounds production. Food Chem. 2016, 212, 234-243.

445 446 447 448

13. Elsinghorst, P.W.; Raters, M.; Dingel, A.; Fischer, J.; Matissek, R. Synthesis and application of 13C-labeled 2-acetyl-4-((1 R, 2 S, 3 R)-1,2,3,4-tetrahydroxybutyl) imidazole (THI), an immunosuppressant observed in caramel food colorings. J. Agric. Food Chem. 2013, 61, 7494-7499.

449 450 451 452

14. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans 101 (15), 2012; http://monographs.iarc.fr/ENG/ Monographs/vol101/mono101-015.pdf (accessed March 3, 2018).

453 454 455 456

15. Houben, G. F.; van Dokkum, W.; van Loveren, H.; Penninks, A. H.; Seinen, W.; Spanhaak, S.; Ockhuizen, T. Effects of Caramel Colour III on the number of blood lymphocytes: A human study on Caramel Colour III immunotoxicity and a comparison of the results with data from rat studies. Food Chem. Toxicol. 1992, 30, 427−430.

457 458 459

16. Hengel, M.; Shibamoto, T. Carcinogenic 4(5)-methylimidazole found in beverages, sauces, and caramel colors: Chemical properties, analysis, and biological activities. J. Agric. Food Chem. 2013, 61, 780-789.

460 461 462

17. Aguilar, F.; Dusemund, B.; Galtier, P.; Gilbert, J.; Gott, D.M.; Grilli, S.; Gurtler, R.; Konig, J.; Lambre, C.; Larsen, J.C.; Leblanc, J.C. Scientific opinion on the re-evaluation of caramel colours (E 150 a, b, c, d) as food additives. EFSA J. 2011, 9, 2004. 20 ACS Paragon Plus Environment

Page 21 of 34


463 464 465 466 467

18. Office of Environmental Health Hazard Assessment (OEHHA). Notice of amendment of text title 27, California code of regulations amendment of section 25705 specific regulatory levels: No significant risk levels 4-methylimidazole (4-MEI) [02/08/12]; https://oehha.ca.gov/proposition-65/crnr/notice-amendment-text-title-27-california-coderegulations-amendment-section (accessed January 3, 2018).

468 469 470 471

19. Commission Regulation (EU) No. 231/2012 of 9 March 2012 laying down specifications for food additives listed in Annexes II and III to Regulation (EC) No. 1333/2008 of the European Parliament and of the Council Text with EEA relevance. Off. J. Eur. Communities: Legis. 2012, L83, 1−295.

472 473 474

20. Chiavaro, E.; Mazzeo, T.; Visconti, A.; Manzi, C.; Fogliano, V.; Pellegrini, N. Nutritional quality of sous vide cooked carrots and Brussels sprouts. J. Agric. Food Chem. 2012, 60, 6019-6025.

475 476

21. Creed, P.G. The sensory and nutritional quality of ‘sous vide’ foods. Food Contr. 1995, 6, 45-52.

477 478 479

22. Kanner, J.; Shapira, N. Oxygen-and metal-ion-dependent nonenzymatic browning of grapefruit juice. In Quality Factors of Fruits and Vegetables. Jen, J.J., Ed.; American Chemical Society: Washington, 1989; Vol. 405., 55-64.

480 481 482

23. Wang, L.; Ren, B.; Liu, Y.; Lu, Y.; Chang, F.; Yang, L. 2-Acetyl-4tetrahydroxybutylimidazole and 4-methylimidazole in caramel colours, vinegar and beverages in China. Food Addit. Contam. Part B. 2015, 8, 163-168.

483 484

24. Ciolino, L. A. Determination and classification of added caramel color in adulterated acerola juice formulations. J. Agric. Food Chem. 1998, 46, 1746-1753.

485 486

25. Jing, H.; Kitts, D.D. Redox-related cytotoxic responses to different casein glycation products in Caco-2 and Int-407 cells. J. Agric. Food Chem. 2004, 52, 3577-3582.

487 488 489

26. Rufian-Henares, J.A.; Garcia-Villanova, B.; Guerra-Hernandez, E. Generation of furosine and color in infant/enteral formula-resembling systems. J. Agric. Food Chem. 2004, 52, 5354-5358.

490 491 492

27. Serratosa, M.P.; Lopez-Toledano, A.; Merida, J.; Medina, M. Changes in color and phenolic compounds during the raisining of grape cv. Pedro Ximenez J. Agric. Food Chem. 2008, 56, 2810-2816.

493 494

28. Adams, A.; Borrelli, R.C.; Fogliano, V.; De Kimpe, N. Thermal degradation studies of food melanoidins. J. Agric. Food Chem. 2005, 53, 4136-4142.

495 496

29. Oliver, G.; Colicchio, T. The Oxford companion to beer. Oxford University Press. 2011. p. 582.

497 498

30. Serban, A.; Nissenbaum, A. Melanoidin polymers as possible oxygen sinks in the pre-biotic oceans. In Origin of Life: Proceedings of the Third ISSOL Meeting and the Sixth ICOL



499 500

Page 22 of 34

Meeting, Jerusalem, June 22–27, 1980. Wolman, Y. ed. 1981. Springer Science and Business Media.

501 502

31. Gobert, J.; Glomb, M.A. Degradation of glucose: reinvestigation of reactive α-dicarbonyl compounds. J. Agric. Food Chem. 2009, 57, 8591-8597.

503 504

32. Smuda, M.; Glomb, M.A. Novel insights into the Maillard catalyzed degradation of maltose. J. Agric. Food Chem. 2011, 59, 13254-13264.

505 506 507 508

33. Van Der Ark, R.; Blokker, P.; Bolshaw, L.; Brouwer, E.R.; Hughes, P.S.; Kessels, H.; Olierook, F.; Van Veen, M. Beverages and foodstuffs resistant to light induced flavor changes, processes for making the same, and compositions for imparting such resistance. 2013. US. Patent 8445050B2.

509 510 511

34. Wu, S.; Fan, H.; Zhang, Q.; Cheng, Y.; Wang, Q.; Yang, G.; Han, B. Conversions of cellobiose and inulin to deoxyfructosazine in aqueous solutions. CLEAN–Soil, Air, Water. 2011, 39, 572-576.

512 513 514

35. Serra-Cayuela, A.; Castellari, M.; Bosch-Fuste, J.; Riu-Aumatell, M.; Buxaderas, S.; LopezTamames, E. Identification of 5-hydroxymethyl-2-furfural (5-HMF) in Cava sparkling wines by LC-DAD-MS/MS and NMR spectrometry. Food Chem. 2013, 141, 3373-3380.

515 516 517

36. Grainger, M. N.; Owens, A.; Manley-Harris, M.; Lane, J. R.; Field, R. J. Kinetics of conversion of dihydroxyacetone to methylglyoxal in New Zealand mānuka honey: Part IV– formation of HMF. Food Chem. 2017, 232, 648-655.

518 519

37. Shallenberger, R. S.; Mattick, L. R. Relative stability of glucose and fructose at different acid pH. Food Chem. 1983, 12, 159-165.

520 521

38. Mottier, P.; Mujahid, C.; Tarres, A.; Bessaire, T.; Stadler, R. H. Process-induced formation of imidazoles in selected foods. Food Chem. 2017, 228, 381-387.

522 523 524

39. Kröplien, U.; Rosdorfer, J.; Van der Greef, J.; Long Jr, R. C.; Goldstein, J. H. 2-Acetyl-4(5)(1,2,3,4-tetrahydroxybutyl) imidazole: detection in commercial caramel color III and preparation by a model browning reaction. J. Org. Chem. 1985, 50, 1131-1133.

525 526 527 528

40. Yu, P.; Xu, X.B.; and Yu, S.J. Comparative study of the effect of glucosamine and free ammonium on 4-methylimidazole formation. J. Agric. Food Chem. 2015, 63, 8031-8036.

529 530 531 532 533 534 22 ACS Paragon Plus Environment

Page 23 of 34


535 ABBREVIATIONS AND NOMENCLATURE 536 a*, redness; b*, yellowness; 3-DG, 3-deoxyglucosone; 3,4-dideoxyglucoson-3-ene, 3,4-DGE; C*, 537 chroma; CE, collision energy; CXP, collision cell exit potential; DA, diacetyl; DOFR, 538 deoxyfructosazine; DP, declustering potential; EP, entrance potential; ESI, electrospray ionization; 539 FR, fructosazine; G, glucosone; GlcN, glucosamine; GO, glyoxal; H°, hue angle; HMF, 5540 hydroxymethyl-2-furfural;

HPA, hydroxypyruvaldehyde; HPLC, high-performance liquid

541 chromatography; L*, lightness; LOD, limit of detection; LOQ, limit of quantitation; 4-MEI, 4542 methylimidazole; MGO, methylglyoxal; MS, mass spectrometry; ND, not detected; OPD, o543 phenylenediamine; PCA, principal component analyses; PVDF, poly(vinylidene fluoride); SEM, 544 standard error of the means; SPE, solid-phase extraction; THI, 2-acetyl-(4)5-tetrahydroxylbutyl 545 imidazole. 546 547 ACKNOWLEDGMENT 548 Authors would like to thank Yuan Yuan Zhao (University of Alberta, Edmonton) for technical 549 assistance in mass spectrometry experiments. 550 551 FUNDING SOURCES 552 This research was funded by Natural Sciences and Engineering Research Council of Canada 553 (NSERC). 554 555 NOTES 556 The authors declare no competing financial interest.



Page 24 of 34

TABLES Table 1. Changes in pH, lightness (L*), hue angle (H°), chroma (C*) and absorbance at 420 nm during incubation of GlcN solutions at 50, 60 and 70°C under non-vacuum and vacuum conditions for 12 h.

Level of oxygen Non-vacuum Vacuum SEM Temperature 50°C 60°C 70°C SEM Interaction (Level of oxygen*Temperature) Non-vacuum 50°C 60°C 70°C Vacuum 50°C 60°C 70°C SEM Sources of variation Treatment Temperature Interaction

pH

L*

H°

C*

Absorbance at 420 nm

3.4a 3.3b 0.2

24.6b 25.3a 0.3

34.1a 18.7b 1.9

2.6b 7.8a 0.8

0.13a 0.10b 0.002

4.4a 3.1b 2.5c 0.02

27.1a 24.3b 23.6c 0.2

17.0b 33.8a 28.5a 2.5

10.5a 4.0b 1.1c 0.6

0.11b 0.12a 0.12a 0.004

4.5a 3.1c 2.6e 4.2b 3.0d 2.5f 0.01

26.4b 24.1cd 23.4d 27.7a 24.4c 23.7cd 0.2

6.0c 1.1e 0.6f 15.0a 6.9b 1.6d 0.07

0.12bc 0.13ab 0.14a 0.10d 0.11cd 0.10d 0.003

Sous-Vide Non-Enzymatic Browning of Glucosamine at Different

Recommend Documents