Article pubs.acs.org/JAFC
Reduction of Carcinogenic 4(5)-Methylimidazole in a Caramel Model System: Influence of Food Additives Seulgi Seo, Mi-Hyun Ka, and Kwang-Geun Lee* Department of Food Science and Biotechnology, Dongguk University-Seoul, 3-26 Pil-dong, Jung-gu, Seoul, 100-715, Korea ABSTRACT: The effect of various food additives on the formation of carcinogenic 4(5)-methylimidazole (4-MI) in a caramel model system was investigated. The relationship between the levels of 4-MI and various pyrazines was studied. When glucose and ammonium hydroxide were heated, the amount of 4-MI was 556 ± 1.3 μg/mL, which increased to 583 ± 2.6 μg/mL by the addition of 0.1 M of sodium sulfite. When various food additives, such as 0.1 M of iron sulfate, magnesium sulfate, zinc sulfate, tryptophan, and cysteine were added, the amount of 4-MI was reduced to 110 ± 0.7, 483 ± 2.0, 460 ± 2.0, 409 ± 4.4, and 397 ± 1.7 μg/mL, respectively. The greatest reduction, 80%, occurred with the addition of iron sulfate. Among the 12 pyrazines, 2ethyl-6-methylpyrazine with 4-MI showed the highest correlation (r = −0.8239). KEYWORDS: carcinogenic 4(5)-methylimidazole, caramel model system, food additives
■
INTRODUCTION 4(5)-Methylimidazole (4-MI) is a simple nitrogen-containing heterocyclic compound that has been known to be formed by the Maillard reaction since the mid-1900s.1 4-MI has received much attention from many researchers and regulatory agencies; the National Toxicology Program (NTP, 2007) reported on the carcinogenicity of 4-MI in caramel colors in 2007, and the Office of Environmental Health Hazard Assessment (OEHHA) within the California Environmental Protection Agency has listed 4-MI as a chemical known for causing cancer.2,3 Thus, many carcinogenicity studies on 4-MI have been performed. For example, 14-week toxicity studies of 4-MI showed the induction of thyroid lesions in both Fischer 344/N rats and B6C3F1 mice, and exposure to 4-MI increased the incidences of alveolar/bronchiolar neoplasm in male and female B6C3F1 mice, which indicated the carcinogenic activity of 4MI.4,5 In addition, some studies show that 4-MI is a neurotoxic agent and has the capability to inhibit a cytochrome P450 isoenzyme that catalyzes the oxidation of many known or suspected carcinogens of low molecular mass in the human liver.6,7 At present, 4-MI is expected to be classified as group 2B (possibly carcinogenic in humans) by IARC Monographs (IARC, 2012), and the World Health Organization has recommended a limit on the content of 4-MI at 200 mg/kg for caramel color III (WHO, 1975).8,9 Caramel color is one of the main sources of 4-MI for humans. It is widely used as a food additive in the food industry for coloring in a variety of foods and drinks including baked goods, beers, pickles, sauces, jam, and confectionary. Caramel colors are a complex mixture of compounds produced by heating carbohydrates with reactants, and they are classified into four classes, based on application and the reactants used in manufacturing, as follows: class I, plain caramel or caustic caramel; class II, caustic sulfite caramel; class III, ammonia caramel, and class IV, sulfite ammonia caramel (EFSA, 2011).10 4-MI is found in the ammonia caramel (class III) and sulfite ammonia caramel (class IV) that the Maillard reaction generates as a major part of the reaction. © 2014 American Chemical Society
The Maillard reaction, which refers to a nonenzymatic browning reaction, occurs during the condensation between carbonyl compounds, usually reducing sugar- and nitrogencontaining compounds, such as the amino acids of a protein.11 4-MI was reported to be formed in the Maillard reaction system that consists of D-glucose and ammonia (class III) for the first time in the early 1960s,12 and there have been several subsequent studies on the formation pathway of 4-MI in the Maillard reaction system. According to a previous study of Debus−Radziszewski imidazole synthesis, methylglyoxal formed by glucose degradation reacts with formaldehyde that has originated from the Strecker reaction, leading to the formation of an imidazole ring.13 The development of the Maillard reaction is influenced by the presence of other reactive compounds such as oxidizing or reducing agents. In particular, metal ions can form complexes with the Maillard reaction products by oxidizing Amadori compounds and catalyzing the further interactions of these compounds. For instance, Cu2+ combines with Amadori compounds, forming unstable complexes that produce hydroxyl radicals and dicarbonyl compounds through degradation.14 Furthermore, it is known that sulfite species can inhibit the Maillard reaction. In the case of glucose, the Maillard reaction is inhibited as sulfite reacts with the 3,4-dideoxyhexosulos-3-ene formed by the decomposition of glucosylamines, which is an important intermediate in color formation.15 So far, the interaction of other food components in the Maillard reaction has been studied in the reports, but not with formation of 4-MI. Thus, taking into account the complexity of real food systems, the influence of other components on the formation of 4-MI needs to be verified. The Maillard reaction is largely responsible for the color and flavor in processed and cooked foods. Starting with a condensation between a reducing sugar and an amino Received: Revised: Accepted: Published: 6481
April 28, 2014 June 10, 2014 June 16, 2014 June 16, 2014 dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
Journal of Agricultural and Food Chemistry
Article
pyridine (50:30:20, v/v) and 120 μL of IBCF for derivatization by hand for 30 s. Subsequently, 500 μL of a 1.0 M sodium bicarbonate solution and 500 μL of isooctane were added, and the mixture was mixed for 4−5 s. Then, 2 μL of the upper layer was injected into the GC−MS system. The determination of 4-MI was conducted using an Agilent gas chromatograph interfaced to an inert 5975 mass selective detector (Palo Alto, CA) with an electron impact (EI) ionization chamber and equipped with a 7683B Series injector/autosampler. A DB-5 ms column (30 m × 0.25 mm i.d. × 0.25 μm film thickness; J&W Scientific, Folsom, CA) was used. The injection was carried out in splitless mode at 270 °C. Helium was applied as a carrier gas with a constant flow of 1.3 mL/min. The oven temperature was programmed as follows: An initial temperature of 80 °C was maintained for 1 min and then elevated to 280 °C at 30 °C/min and then held for 1.83 min. The temperature of the MS transfer line was 280 °C. The mass spectrometer was operated by electron impact at 70 eV, in selective ion monitoring (SIM) mode. Ions at m/z 68, 82, 109, and 182 were used for 4-MI and those at 95, 109, and 196 for 2-EI with a dwell time of 100 ms each. The temperatures of the ion source and quadrupole were 230 and 150 °C, respectively. Agilent Chemstation was used to control the GC−MS system. Color Measurement from the Caramel Model System. Color parameters of the prepared solutions from various caramel model systems were monitored by a Nippon Denshoku NE 4000 colorimeter, using the 1976 CIELAB system. The sample solution (200 μL) was diluted with 10 mL of distilled water, and the colors were measured in transmittance mode. The parameters were L* (brightness; L* = 0 for black, and L* = 100 for white), a* (red-green component; −a* = greenness and +a* = redness) and b* (yellow−blue component; −b* = blueness and +b* = yellowness). The color differences were calculated by the expression: ΔE = [(ΔL*)2 + (Δa*)2 + (Δb*)2]1/2.20 To evaluate color changes, ΔE was calculated with L*, a*, and b* values of the caramel model systems with different food additives added, with respect to those in glucose/ammonium hydroxide 1 M model system. Analysis of Pyrazines in the Caramel Model System. Each sample (3 g) was weighed into a 20 mL headspace vial and sealed with an unlined crimp cap containing polytetrafluoroethylene (PTFE)/ silicone septa (Agilent). The assayed SPME device was divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS, 50/30 μm thickness) fiber (Supelco, Bellefonte, PA). After the samples were equilibrated for 5 min at 50 °C, the SPME fiber was inserted to extract chemicals for 30 min at 50 °C under magnetic stirring. Analyses were performed using a GC−MS system equipped with a DB-Wax column (60 m × 0.25 mm i.d. × 0.25 μm film thickness; J&W Scientific, Folsom, CA). The SPME fiber was inserted into the injection port of the GC, which was in splitless mode, and maintained at 260 °C for 5 min. The oven temperature program started at 50 °C, increased by 3 °C/min to 220 °C, and stayed at 220 °C for 10 min. The temperature of the transfer line was 280 °C, and the mass spectrometer was operated by electron impact at 70 eV. Quantification was performed by a GC that was equipped with a flame ionization detector under the same conditions that were described for the GC− MS system. The quantitation of identified pyrazines was carried out with undecane as the internal standard. Pyrazines were identified through comparison with the Kovats gas chromatographic retention index I, by comparing the mass spectra with their authentic compound and through Wiley 275 libraries. Statistical Analysis. All experiments were performed in triplicate. The data were analyzed using one-way ANOVA tests, and Duncan’s new multiple range test was used for comparison of significant differences (p < 0.05) between the replicates (n = 3).
compound, the Maillard reaction leads to brown color melanoidins and flavor compounds including pyrazines, pyrroles, furans, oxazoles, and so on.16,17 There have been many research studies into the Maillard reaction’s products in terms of color and flavor. Various analytical methods to determine the presence of 4MI have been developed, based on gas chromatography−mass spectrometry (GC−MS), high-performance liquid chromatography (HPLC), and liquid chromatography−tandem mass spectrometry (LC−MS/MS).17−19 These methods require expensive equipment and specialized technical skills. Therefore, development of an index such as color parameters or volatile compounds would be useful for determining the presence of 4MI. The specific aim of this study was to examine the effect of food additives in the formation of 4-MI and determine the correlation among color parameters, pyrazines as dominant flavor compounds, and 4-MI. A caramel model system was chosen that consisted of glucose and ammonium hydroxide (as class III) with food additives such as metal ions (Fe2+, Ca2+, Zn2+, Mg2+) and amino acids (cysteine, tryptophan), each at different levels.
■
MATERIALS AND METHODS
Reagents. 4-Methylimidazole (4-MI, purity ≥98%), 2-ethylimidazole (2-EI, purity ≥98%), D-glucose (purity ≥99.5%), sodium sulfite (purity ≥98%), ammonium hydroxide solution (purity ≥29%), acetic acid (purity ≥99.7%), bis-2-ethylhexylphosphate (BEHPA, purity ≥97%), and isobutyl chloroformate (IBCF, purity ≥98%) were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO). Isooctane, acetonitrile (MeCN), pyridine, and isobutyl alcohol (purity ≥99%) were Lichrosolv quality and purchased from Merck (Darmstadt, Germany). Magnesium sulfate (MgSO4·7H2O) and zinc sulfate (ZnSO4·7H2O) were purchased from Junsei Chemical Co. (Tokyo, Japan), and iron sulfate (FeSO4) was purchased from Kanto Chemical Co. (Tokyo, Japan). Calcium sulfate (CaSO4·0.5H2O) was purchased from Yukari Pure Chemical Co. (Osaka, Japan), and tryptophan (purity ≥99%) and cysteine (purity ≥99.5%) were purchased from Junsei and Fluka (Neu-Ulm, Germany), respectively. All reagents were analytical grade. Standard stock solutions of 4-MI and 2-EI used as an internal standard (IS) were prepared in 0.1 M HCl at a concentration of 20 and 1 mg/kg, respectively. All the solutions were kept at 4 °C until used. Sample Preparations of the Caramel Model Systems. Glucose and ammonium hydroxide both were dissolved in 30 mL of HPLC-grade water in a swing top bottle, each at a concentration of 1 M. To compare the influences of food additives on the formation of 4MI, all additives were added to glucose/ammonium solution individually at three distinct concentrations: 0.1, 0.05, and 0.01 M. The food additives were Fe2+ (as FeSO4), Ca2+ (as CaSO4), Zn2+ (as ZnSO4·7H2O), Mg2+ (as MgSO4·7H2O), cysteine, sodium sulfite, and tryptophan. The caramel model system solutions were heated at 100 °C for 2 h in an oven to instigate the Maillard reaction and cooled under running cold water to room temperature immediately in order to stop the reaction progress. All samples were stored at 4 °C, until they were analyzed. Analysis of 4(5)-Methylimidazole. The procedure to extract 4MI from a sample was carried out in the manner specified in a previous study.17 Each sample (100 μL) was added to 4 mL of phosphate buffer that was extracted from 8 mL of 0.1 M BEHPA in chloroform by mixing for 1 min and centrifugation at 3500 rpm for 10 min. A 7.5 mL portion of the chloroform phase was transferred to a second vial containing 4 mL of 0.1 M HCl and mixed again by vortexing for 1 min. After centrifugation at 3500 rpm for 5 min, a 1 mL aliquot of the aqueous phase was added with 100 μL of 2-EI used as IS, which was placed into a vial and mixed with 1 mL of MeCN−isobutyl alcohol−
■
RESULTS AND DISCUSSION
Measurement of 4-MI in Caramel Model System Solution. As mentioned earlier, 4-MI was not detected in caramel colors I or II but was identified in both colors III and IV. Thus, glucose and ammonium hydroxide, which create 6482
dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
Journal of Agricultural and Food Chemistry
Article
Table 1. Level and Reduction Rate of 4(5)-Methylimidazole Formed in Caramel Model System Solutions (n = 3)a model systems 1 M glucose/1 M ammonium hydroxide 1 M glucose/1 M ammonium hydroxide
concn of additives (M) sodium sulfite iron sulfate
556 583 110 269 452 483 506 576 656 552 652 460 507 581 409 457 556 397 444 609
0.1 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01
magnesium sulfate
calcium sulfate
zinc sulfate
tryptophan
cysteine
a
amount of 4-MI (μg/mL) ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
1.3 2.6 0.7 1.8 2.9 2.0 1.4 1.3 3.4 1.7 2.6 2.0 0.5 5.7 4.4 2.3 1.8 1.7 0.8 0.6
reduction (%)
g gh a b de ef f gh j g ij e f gh cd e g c de hi
0 −5 80 52 19 13 9 −3 −18 1 −17 17 9 −4 26 18 0 29 20 −10
Values followed by the same letter in a given column are not significantly different at p < 0.05.
Table 2. Color Measurement of Caramel Model System Solution by Colorimetrya model system 1 M glucose/1 M ammonium hydroxide 1 M glucose/1 M ammonium hydroxide
concn of additives (M) sodium sulfite iron sulfate
0.1 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01
magnesium sulfate
calcium sulfate
zinc sulfate
tryptophan
cysteine
L* 17.9 15.9 12.0 15.5 14.6 20.4 21.2 18.7 16.4 16.5 17.4 10.1 16.0 16.9 16.8 17.7 17.4 19.1 18.6 18.2
a* o g d f e s t q i j m c h l k n m r q p
1.3 3.6 3.8 2.3 3.0 −0.8 −1.6 0.6 2.6 2.7 1.9 3.9 2.0 2.4 2.0 1.2 1.8 −0.1 0.5 1.0
ΔEb
b* g m m i l d c f jk k h m h i h g h e f f
22.7 22.6 17.2 20.2 20.6 18.8 16.2 21.8 23.3 23.3 23.2 15.2 21.9 22.9 22.7 22.4 23.2 20.4 21.7 22.7
ij i e g g f d h k k k c h ijk ij i jk g h ij
0 3.1 8.5 3.7 4.3 5.0 7.9 1.4 2.1 2.1 1.0 11.2 2.2 1.5 1.4 0.4 0.9 2.9 1.5 0.3
a
Values followed by the same letter in a given column are not significantly different at p < 0.05. bGlucose/ammonium hydroxide model system was used as a control group for calculation of color differences.
five times. The recoveries that were measured for samples spiked with 4-MI at two concentration levels (20 and 100 μg/ mL) ranged from 84 to 111%. The precision of the interday and intraday analysis, expressed as the relative standard deviation (RSD), were 0.9% and 4.9−6.5%, respectively. Table 1 shows the amount of 4-MI formed in the caramel model system solutions that include each food additive at a level of 0.1, 0.05, and 0.01 M. A glucose and ammonium hydroxide (1 M) model system was used as a control. As each group was compared to the glucose/ammonium hydroxide model system, a reduction of 4-MI was observed in most caramel model systems with the 0.1 M food additives. In
colors III through heating, were used as a control group of the caramel model system in the present study. In a previous study, the Maillard reaction was investigated by using a model system that consisted of a sugar and an amine compound, and 4-MI was found in the glucose/ammonium hydroxide model system at a level that ranged from 490 to 710 μg/mL.21 An absolute calibration curve was obtained (y = 0.7958x + 2.6264, R2 = 0.9937) using 4-MI standard solutions with various concentrations (1, 5, 10, 20, 50, and 100 μg/mL). The recovery and precision of the caramel model system solution of glucose/ammonium hydroxide (1 M) with added sodium sulfite (0.1 M) were determined, and each test was replicated 6483
dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
Journal of Agricultural and Food Chemistry
Article
particular, there was a significant reduction of 4-MI formation in the caramel model system with 0.1 M Fe2+, but the amount of 4-MI increased in the caramel model system with the addition of 0.1 M Ca2+. When the D-glucose and ammonium hydroxide were heated, the amount of 4-MI formed was 556 ± 1.3 μg/mL. The addition of 0.1 M sodium sulfite increased the 4-MI formation by up to 583 ± 2.6 μg/mL (5%), whereas the formation level and reduction rate of 4-MI through the addition of food additives were 110 ± 0.7 (80%) in 0.1 M Fe2+, 483 ± 2.0 (13%) in 0.1 M Mg2+, 460 ± 2.0 (17%) in 0.1 M Zn2+, 409 ± 4.4 (26%) in 0.1 M tryptophan, and 397 ± 1.7 μg/mL (29%) in 0.1 M cysteine. The degree of contribution to the reduction of 4MI from food additives had the following order: Fe2+ > cysteine > tryptophan > Zn2+ > Mg2+. The formation levels of 4-MI ranged from 110 ± 0.7 (0.1 M Fe2+) to 452 ± 2.9 (0.01 M Fe2+), from 483 ± 2.0 (0.1 M Mg2+) to 576 ± 1.3 (0.01 M Mg2+), from 656 ± 3.4 (0.1 M Ca2+) to 652 ± 2.6 (0.01 M Ca2+), from 460 ± 2.0 (0.1 M Zn2+) to 581 ± 5.7 (0.01 M Zn2+), from 409 ± 4.4 (0.1 M tryptophan) to 556 ± 1.8 (0.01 M tryptophan), and from 397 ± 1.7 (0.1 M cysteine) to 444 ± 0.8 μg/mL (0.01 M cysteine). As a higher concentration of food additive was added, apart from the case of Ca2+, the amount of 4-MI created by the Maillard reaction solution decreased, while an especially low concentration of food additive was not sufficient to reduce the formation of 4-MI. Furthermore, the addition of 0.01 M of Ca2+, Mg2+, Zn2+, cysteine, and tryptophan led to a higher level of 4-MI than the control caramel model system, which was the Maillard reaction solution of 1 M glucose and 1 M ammonium hydroxide without additives. These results clearly suggest that some food additives at an appropriate concentration have a positive effect on the reduction of 4-MI in the Maillard reaction. This effect of the food additives on the formation of 4-MI in the Maillard reaction system was not previously reported. As food additives could be used in the manufacturing of foods, adding these food additives may be useful for reducing the formation of 4-MI during the Maillard reaction process. Color Differences in the Caramel Model System Solutions. Since the Maillard reaction induces a brown color with formation of melanoidins, color is one of the important factors when considering how much 4-MI is formed during the Maillard reaction. Table 2 shows the results obtained for three series of color differences: L*, a*, and b*. Significant differences according to the Duncan post hoc test are indicated by superscript letters. The L* values of the caramel model systems with added 0.1 M food additives including 0.1 M sodium sulfite decreased compared to glucose/ammonium hydroxide (1 M) model system, except for Mg2+ and cysteine. Contrarily, values of a* increased in the same caramel model systems, denoting the rise of redness characteristics when adding food additives. The b* values did not show a significant change for any caramel model system. Although the measurement of color from the Maillard reaction model systems has not been reported, the decrease of L* and the increase of a* were observed in the foods due to the brown pigment formation that was induced by the Maillard reaction.22,23 In the studies of Wang et al. and Ramonaitytė et al., the absorbance of the Maillard reaction model system showed increases at 420 and 450 nm, respectively, with addition of metal ions as a result of the Maillard reaction, indicating that the addition of metal ions accelerated the formation of the final browning compounds.24,25
In order to determine the correlation between the 4-MI concentration and color, a linear regression analysis was performed on each L*, a*, b*, and ΔE. The total color difference, expressed as ΔE, showed a lack of negative correlation with the 4-MI concentration (r = −0.4826). On the other hand, L*, a*, and b* values had no significant correlation with 4-MI (p < 0.05). Pyrazines Formed in Caramel Model System Solutions. Table 3 lists the 12 kinds of pyrazine detected in the study of the caramel model system solutions. The majority of the pyrazines detected were 2-methylpyrazine, with a peak area ratio (PAR) ranging from 2.27 to 6.29, followed by 2,6dimethylpyrazine ranging from 0.84 to 2.56 and 2,3dimethylpyrazine ranging from 0.10 to 0.33. It is generally known that pyrazines are among the typical products of the Maillard reaction between amino compounds and carbonyl compounds.26 All of the 12 pyrazine compounds confirmed in the present study were consistent with the previous literature that had investigated the pyrazines that could be derived from the Maillard reaction.26−29 It was difficult to find a relatively substantial relationship between 4-MI and the pyrazines formed in the greatest amounts, whereas 2-ethyl-6-methylpyrazine was among the 12 kinds of pyrazine compounds that showed a correlation with the concentration of 4-MI. Figure 1 showed a negative correlation (r = −0.8239) between 4-MI and 2-ethyl-6methylpyrazine, indicating that 2-ethyl-6-methylpyrazine is formed in inverse proportion to the 4-MI concentration (p < 0.05). When the 4-MI declined by 80% with the addition of Fe2+ into the glucose/ammonium hydroxide 1 M solution, the content of 2-ethyl-6-methylpyrazine increased from 0.03 to 0.21 PAR. Figure 2 shows the proposed formation pathway of 4-MI and 2-ethyl-6-methylpyrazine from methylglyoxal on the basis of the previous studies.13,21,26,30,31 Methylglyoxal, as the sugar-derived reactive intermediate during the Maillard reaction, reacts with ammonia. Interestingly, the nitrogen-containing reactive intermediates formed from methylglyoxal with ammonia can produce either 4-MI by interacting with formaldehyde or 2ethyl-6-methylpyrazine by reacting with ethylglyoxal. Therefore, if the nitrogen-containing reactive intermediates generated 4MI with formaldehyde, the formation of 2-ethyl-6-methylpyrazine might be reduced. Further study that elucidates a more accurate correlation between 4-MI and 2-ethyl-6-methylpyrazine will be needed. 4-MI investigated in the present study has received much attention from regulatory agencies as a possible carcinogenic compound. Therefore, it is important to find a way to reduce the 4-MI produced by the Maillard reaction. The results of the present study demonstrate that the addition of food additives such as metal ions and amino acids at an appropriate level could reduce the formation of 4-MI. With current results various food products such as a bakery product could be applied to reduce the amount of 4-MI. However, more detail application study would be carried out in the future. In addition, a correlation between 2-ethyl-6-methylpyrazine and the 4-MI concentration was determined. Since the analysis of 4-MI requires expensive technology and specialized knowledge, the correlation between color, pyrazine levels, and 4-MI levels could be useful for the evaluation of the 4-MI formed during the Maillard reaction. 6484
dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
0.23 0.05
0.03 0.01 0.01 0.01 0.02 0.04 0.05 0.01 12
Compounds content = peak area of each compound/peak area of undecane (internal standard). bConcentration of food additives (M).
0.11 0.07
Figure 1. Correlation between 4(5)-methylimidazole and 2-ethyl-6methylpyrazine.
Figure 2. Proposed formation pathway of 4(5)-methylimidazole and 2-ethyl-6-methylpyrazine.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: 82-2-2260-3370. Fax: 82-2-2285-3370. E-mail: [email protected]. Funding
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (Grant No.: 2012R1A1A2007551) and R&D Convergence Center Support Program, Ministry for Food, Agriculture, Forestry and Fisheries, Republic of Korea Notes
The authors declare no competing financial interest.
■
REFERENCES
(1) Tsuchida, H.; Komoto, M. Chemical studies on the reaction products of methylpentose and ammonia. Agric. Biol. Chem. 1967, 31, 185−189. (2) National Toxicology Program (NTP, 2007). Toxicology and Carcinogenesis of 4-Methylimidazole (CAS No. 822-36-6) in F344/N Rats ad B6C3F1Mice (Fee Studies); NTP Technical Report Series, No. 535, NIH Publication No. 07-4471; U.S. Department of Health and Human Service, NTP: Research Triangle Park, NC, 2007; pp 1−274. (3) 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 12; http://oehha.ca.gov/prop65/law/adopt020812.html (accessed Aug 20, 2013).
a
0.04
0.02
0.10 0.03
0.10 0.03 0.01 0.13 0.04 1.42 0.11 0.04 0.42 0.12 0.02 0.13 0.05 0.02 0.18 0.07 0.01 0.14 0.05 0.02 0.19 0.06 0.01 0.17 0.07 0.03 0.25 0.04 0.02 0.23 0.13 0.05 0.39 0.07 0.03 0.33 0.21 0.05 0.57 0.09 0.04 0.10 0.03 0.01 0.12 0.05
0.11 0.04 0.01 0.15 0.03 0.01
1.55 2.56 0.95
2.07
0.14 0.05 0.02 0.19 0.05 0.00
0.10 0.02
0.11 0.03 0.01 0.13 0.06
0.09 0.02 0.02 0.10 0.07
4.25 0.12 0.05 0.51 0.15 0.03
0.86 0.01 0.08 0.01 1.04 1.04
0.01
0.32 0.05
0.23 0.08 0.03 0.34 0.07
1.72 0.01 0.18 0.06 2.41
2.41 0.01 0.21 0.06
0.25 5.06 0.29 6.11 0.25 5.41
0.15 3.02 0.26 0.85 0.02 0.10 0.03 0.01 0.13 0.04 0.12 3.30
0.21 3.88 0.37 0.96 0.20 4.19 0.37 1.04 0.11 2.72 0.12 4.33
unsubstituted 2-methyl 2,5-dimethy2,6-dimethyl 2-ethyl 2,3-dimethyl 2-ethyl-6-methyl 2-ethyl-5-methyl 2-ethyl-3-methyl ethenyl 2,5-dimethyl-3ethyl 2,6-diethyl 1 2 3 4 5 6 7 8 9 10 11
0.15 3.11 0.30 0.84
0.11 2.27
0.10 3.26
0.19 4.13 0.36 1.02
0.18 4.38
0.18 4.25 0.32 0.96
0.18 4.12
0.24 5.77 0.34 2.47
0.27 6.29 0.34 2.30
0.27 5.95 0.27 1.85 0.02 0.37 0.09 0.04 0.33 0.10 0.02
0.15 3.38 0.25 0.89
0.05 0.05 0.1 0.01 0.05 0.01 0.05 0.1 0.01 0.05 0.1 0.01 0.05 0.1 0.1b pyrazines
magnesium sulfate calcium sulfate iron sulfate
Article
no.
glucose/ammonium hydroxide
sodium sulfite
Table 3. Amounts of Pyrazine Compounds Identified in Caramel Model Systems
peak area ratioa
0.1
zinc sulfate
cysteine
0.01
0.1
tryptophan
0.01
Journal of Agricultural and Food Chemistry
6485
dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
Journal of Agricultural and Food Chemistry
Article
(24) Wang, H.; Zhou, Y.; Ma, J.; Zhou, Y.; Jiang, H. The effects of phytic acid on the Maillard reaction and the formation of acrylamide. Food Chem. 2013, 141, 18−22. (25) Ramonaitytė, D. T.; Keršienė, M.; Adams, A.; Tehrani, K. A. The interaction of metal ions with Maillard reaction products in a lactose−glycine model system. Food Res. Int. 2009, 42, 331−336. (26) Hwang, H.; Hartman, T. G.; Rosen, R. T.; Lech, J.; Ho, C. Formation of pyrazines from the Maillard reaction of glucose and lysine-α-amine-15N. J. Agric. Food Chem. 1994, 42, 1000−1004. (27) Lee, K. G.; Jang, H.; Shibamoto, T. Formation of carcinogenic 4(5)-methylimidazole in caramel model systems: A role of sulphite. Food Chem. 2013, 136, 1165−1168. (28) Shibamoto, T. Heterocyclic compounds in browning and browning/nitrite model systems: Occurrence, formation mechanisms, flavor characteristics and mutagenic activity. Instrum. Anal. Foods 1983, 1, 229−278. (29) Yu, A.; Tan, Z.; Wang, F. Mechanistic studies on the formation of pyrazines by Maillard reaction between L-ascorbic acid and Lglutamic acid. LWT-Food Sci. Technol. 2013, 50, 64−71. (30) 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. (31) Yaylayan, V. A.; Haffenden, L. W. Mechanism of imidazole and oxazole formation in [13C-2]-labelled glycine and alanine model systems. Food Chem. 2003, 81, 403−409.
(4) Chan, P.; Mahler, J.; Travlos, G.; Nyska, A.; Wenk, M. Induction of thyroid lesions in 14-week toxicity studies of 2- and 4methylimidazole in Fischer 344/N rats and B6C3F1 mice. Arch. Toxicol. 2006, 80, 169−180. (5) Chan, P. C.; Hills, G. D.; Kissling, G. E.; Nyska, A. Toxicity and carcinogenicity studies of 4-methylimidazole in F344/N rats and B6C3F1 mice. Arch. Toxicol. 2008, 82, 45−53. (6) Patey, A. L.; Shearer, G.; Knowles, M. E.; Denner, W. H. B. Ammonia caramel: Specifications and analysis. Food Addit. Contam. 1985, 2, 107−112. (7) Hargreaves, M. B.; Jones, B. C.; Smith, D. A.; Gescher, A. Inhibition of p-nitro-phenolhydroxylase in rat liver microsomes by small aromatic and heterocyclic molecules. Drug Metab. Dispos. 1994, 22, 806−810. (8) IARC Monographs. 4-Methylimidazole. Monographs.iarc.fr/ ENG/Monographs/vol101/mono101-015.pdf (accessed Oct 16, 2012). (9) WHO. Toxicological Evaluation of Some Food Colours, Enzyme, Flavour Enhancers, Thickening Agents, And Certain Food Additives; International Program on Chemical Safety World Health Organization, 1975; http://www.inchem.org/documents/jecfa/jecmono/ v06je13.htm (accessed Aug 31, 2012). (10) EFSA Panel on Food Additives and Nutrient Sources added to Food (ANS). Scientific opinion on the reevaluation of caramel colours (E 150a,b,c,d) as food additives. EFSA J. 2011, 9 (3), 2004. (11) Nursten, H. The chemistry of nonenzymic browning. The Maillard Reaction: Chemistry, Biochemistry and Implications; Royal Society of Chemistry: London, UK, 2005; pp 5−30. (12) Komoto, M. Chemical studies on the reaction products of glucose and ammonia. III. Behaviors of glucosylamine in browning reaction of glucose and aqueous ammonia. Nippon Nogei Kagaku Kaishi 1962, 36, 403−407. (13) Jang, H. W.; Jiang, Y.; Hengel, Matt.; Shibamoto, T. Formation of 4(5)-methylimidazole and its precursors, α-dicarbonyl compounds, in Maillard systems. J. Agric. Food Chem. 2013, 61, 6865−6872. (14) Horikawa, H.; Okada, M.; Nikamura, Y.; Sato, A.; Iwamoto, N. Production of hydroxyl radicals and alpha-dicarbonyl compounds associated with Amadori compound−Cu2+ complex degradation. Free Radical Res. 2002, 36, 1059−1065. (15) Keller, C.; Wedzicha, B. L.; Leong, L. P.; Berger, J. Effect of glyceraldehyde on the kinetics of Maillard browning and inhibition by sulphite species. Food Chem. 1999, 66, 495−501. (16) Serpen, A.; Gökmen, V. Evaluation of the Maillard reaction in potato crisps by acrylamide, antioxidant capacity and color. J. Food Compos. Anal. 2009, 22, 589−595. (17) Cunha, S. C.; Barrado, A. I.; Faria, M. A.; Fernandes, J. O. Assessment of 4-(5-)methylimidazole in soft drinks and dark beer. J. Food Compos. Anal. 2011, 24, 609−614. (18) Thosen, M.; Willumsen, D. Quantitative ion-pair extraction of 4(5-)methylimidazole from caramel colour and its determination by reversed-phase ion-pair liquid chromatography. J. Chromatogr., A 1987, 211, 213−221. (19) Yanaguchi, H.; Masuda, T. Determination of 4(5)-methylimidazole in soy sauce and other foods by LC−MS/MS after solidphase extraction. J. Agric. Food Chem. 2011, 59, 9770−9775. (20) Calvo, C. Physical characterization and nutrient analysis. Optical properties. In Handbook of Food Analysis; Nollet, L. M. L. , Ed.; Marcel Dekker: New York, 2004; Vol. 1, pp 1−19. (21) Moon, J. G.; Shibamoto, T. Formation of carcinogenic 4(5)methylimidazole in Maillard reaction systems. J. Agric. Food Chem. 2011, 59, 615−618. (22) Bosch, L.; Alegría, A.; Farré, R.; Clemente, G. Fluorescence and color as markers for the Maillard reaction on milk−cereal based infant foods during storage. Food Chem. 2007, 105, 1135−1143. (23) Farroni, A.; Buera, M. P. Colour and surface fluorescence development and their relationship with Maillard reaction markers as influenced by structural changes during cornflakes production. Food Chem. 2012, 135, 1685−1691. 6486
dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
Reduction of Carcinogenic 4(5)-Methylimidazole in a Caramel Model System: Influence of Food Additives Seulgi Seo, Mi-Hyun Ka, and Kwang-Geun Lee* Department of Food Science and Biotechnology, Dongguk University-Seoul, 3-26 Pil-dong, Jung-gu, Seoul, 100-715, Korea ABSTRACT: The effect of various food additives on the formation of carcinogenic 4(5)-methylimidazole (4-MI) in a caramel model system was investigated. The relationship between the levels of 4-MI and various pyrazines was studied. When glucose and ammonium hydroxide were heated, the amount of 4-MI was 556 ± 1.3 μg/mL, which increased to 583 ± 2.6 μg/mL by the addition of 0.1 M of sodium sulfite. When various food additives, such as 0.1 M of iron sulfate, magnesium sulfate, zinc sulfate, tryptophan, and cysteine were added, the amount of 4-MI was reduced to 110 ± 0.7, 483 ± 2.0, 460 ± 2.0, 409 ± 4.4, and 397 ± 1.7 μg/mL, respectively. The greatest reduction, 80%, occurred with the addition of iron sulfate. Among the 12 pyrazines, 2ethyl-6-methylpyrazine with 4-MI showed the highest correlation (r = −0.8239). KEYWORDS: carcinogenic 4(5)-methylimidazole, caramel model system, food additives
■
INTRODUCTION 4(5)-Methylimidazole (4-MI) is a simple nitrogen-containing heterocyclic compound that has been known to be formed by the Maillard reaction since the mid-1900s.1 4-MI has received much attention from many researchers and regulatory agencies; the National Toxicology Program (NTP, 2007) reported on the carcinogenicity of 4-MI in caramel colors in 2007, and the Office of Environmental Health Hazard Assessment (OEHHA) within the California Environmental Protection Agency has listed 4-MI as a chemical known for causing cancer.2,3 Thus, many carcinogenicity studies on 4-MI have been performed. For example, 14-week toxicity studies of 4-MI showed the induction of thyroid lesions in both Fischer 344/N rats and B6C3F1 mice, and exposure to 4-MI increased the incidences of alveolar/bronchiolar neoplasm in male and female B6C3F1 mice, which indicated the carcinogenic activity of 4MI.4,5 In addition, some studies show that 4-MI is a neurotoxic agent and has the capability to inhibit a cytochrome P450 isoenzyme that catalyzes the oxidation of many known or suspected carcinogens of low molecular mass in the human liver.6,7 At present, 4-MI is expected to be classified as group 2B (possibly carcinogenic in humans) by IARC Monographs (IARC, 2012), and the World Health Organization has recommended a limit on the content of 4-MI at 200 mg/kg for caramel color III (WHO, 1975).8,9 Caramel color is one of the main sources of 4-MI for humans. It is widely used as a food additive in the food industry for coloring in a variety of foods and drinks including baked goods, beers, pickles, sauces, jam, and confectionary. Caramel colors are a complex mixture of compounds produced by heating carbohydrates with reactants, and they are classified into four classes, based on application and the reactants used in manufacturing, as follows: class I, plain caramel or caustic caramel; class II, caustic sulfite caramel; class III, ammonia caramel, and class IV, sulfite ammonia caramel (EFSA, 2011).10 4-MI is found in the ammonia caramel (class III) and sulfite ammonia caramel (class IV) that the Maillard reaction generates as a major part of the reaction. © 2014 American Chemical Society
The Maillard reaction, which refers to a nonenzymatic browning reaction, occurs during the condensation between carbonyl compounds, usually reducing sugar- and nitrogencontaining compounds, such as the amino acids of a protein.11 4-MI was reported to be formed in the Maillard reaction system that consists of D-glucose and ammonia (class III) for the first time in the early 1960s,12 and there have been several subsequent studies on the formation pathway of 4-MI in the Maillard reaction system. According to a previous study of Debus−Radziszewski imidazole synthesis, methylglyoxal formed by glucose degradation reacts with formaldehyde that has originated from the Strecker reaction, leading to the formation of an imidazole ring.13 The development of the Maillard reaction is influenced by the presence of other reactive compounds such as oxidizing or reducing agents. In particular, metal ions can form complexes with the Maillard reaction products by oxidizing Amadori compounds and catalyzing the further interactions of these compounds. For instance, Cu2+ combines with Amadori compounds, forming unstable complexes that produce hydroxyl radicals and dicarbonyl compounds through degradation.14 Furthermore, it is known that sulfite species can inhibit the Maillard reaction. In the case of glucose, the Maillard reaction is inhibited as sulfite reacts with the 3,4-dideoxyhexosulos-3-ene formed by the decomposition of glucosylamines, which is an important intermediate in color formation.15 So far, the interaction of other food components in the Maillard reaction has been studied in the reports, but not with formation of 4-MI. Thus, taking into account the complexity of real food systems, the influence of other components on the formation of 4-MI needs to be verified. The Maillard reaction is largely responsible for the color and flavor in processed and cooked foods. Starting with a condensation between a reducing sugar and an amino Received: Revised: Accepted: Published: 6481
April 28, 2014 June 10, 2014 June 16, 2014 June 16, 2014 dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
Journal of Agricultural and Food Chemistry
Article
pyridine (50:30:20, v/v) and 120 μL of IBCF for derivatization by hand for 30 s. Subsequently, 500 μL of a 1.0 M sodium bicarbonate solution and 500 μL of isooctane were added, and the mixture was mixed for 4−5 s. Then, 2 μL of the upper layer was injected into the GC−MS system. The determination of 4-MI was conducted using an Agilent gas chromatograph interfaced to an inert 5975 mass selective detector (Palo Alto, CA) with an electron impact (EI) ionization chamber and equipped with a 7683B Series injector/autosampler. A DB-5 ms column (30 m × 0.25 mm i.d. × 0.25 μm film thickness; J&W Scientific, Folsom, CA) was used. The injection was carried out in splitless mode at 270 °C. Helium was applied as a carrier gas with a constant flow of 1.3 mL/min. The oven temperature was programmed as follows: An initial temperature of 80 °C was maintained for 1 min and then elevated to 280 °C at 30 °C/min and then held for 1.83 min. The temperature of the MS transfer line was 280 °C. The mass spectrometer was operated by electron impact at 70 eV, in selective ion monitoring (SIM) mode. Ions at m/z 68, 82, 109, and 182 were used for 4-MI and those at 95, 109, and 196 for 2-EI with a dwell time of 100 ms each. The temperatures of the ion source and quadrupole were 230 and 150 °C, respectively. Agilent Chemstation was used to control the GC−MS system. Color Measurement from the Caramel Model System. Color parameters of the prepared solutions from various caramel model systems were monitored by a Nippon Denshoku NE 4000 colorimeter, using the 1976 CIELAB system. The sample solution (200 μL) was diluted with 10 mL of distilled water, and the colors were measured in transmittance mode. The parameters were L* (brightness; L* = 0 for black, and L* = 100 for white), a* (red-green component; −a* = greenness and +a* = redness) and b* (yellow−blue component; −b* = blueness and +b* = yellowness). The color differences were calculated by the expression: ΔE = [(ΔL*)2 + (Δa*)2 + (Δb*)2]1/2.20 To evaluate color changes, ΔE was calculated with L*, a*, and b* values of the caramel model systems with different food additives added, with respect to those in glucose/ammonium hydroxide 1 M model system. Analysis of Pyrazines in the Caramel Model System. Each sample (3 g) was weighed into a 20 mL headspace vial and sealed with an unlined crimp cap containing polytetrafluoroethylene (PTFE)/ silicone septa (Agilent). The assayed SPME device was divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS, 50/30 μm thickness) fiber (Supelco, Bellefonte, PA). After the samples were equilibrated for 5 min at 50 °C, the SPME fiber was inserted to extract chemicals for 30 min at 50 °C under magnetic stirring. Analyses were performed using a GC−MS system equipped with a DB-Wax column (60 m × 0.25 mm i.d. × 0.25 μm film thickness; J&W Scientific, Folsom, CA). The SPME fiber was inserted into the injection port of the GC, which was in splitless mode, and maintained at 260 °C for 5 min. The oven temperature program started at 50 °C, increased by 3 °C/min to 220 °C, and stayed at 220 °C for 10 min. The temperature of the transfer line was 280 °C, and the mass spectrometer was operated by electron impact at 70 eV. Quantification was performed by a GC that was equipped with a flame ionization detector under the same conditions that were described for the GC− MS system. The quantitation of identified pyrazines was carried out with undecane as the internal standard. Pyrazines were identified through comparison with the Kovats gas chromatographic retention index I, by comparing the mass spectra with their authentic compound and through Wiley 275 libraries. Statistical Analysis. All experiments were performed in triplicate. The data were analyzed using one-way ANOVA tests, and Duncan’s new multiple range test was used for comparison of significant differences (p < 0.05) between the replicates (n = 3).
compound, the Maillard reaction leads to brown color melanoidins and flavor compounds including pyrazines, pyrroles, furans, oxazoles, and so on.16,17 There have been many research studies into the Maillard reaction’s products in terms of color and flavor. Various analytical methods to determine the presence of 4MI have been developed, based on gas chromatography−mass spectrometry (GC−MS), high-performance liquid chromatography (HPLC), and liquid chromatography−tandem mass spectrometry (LC−MS/MS).17−19 These methods require expensive equipment and specialized technical skills. Therefore, development of an index such as color parameters or volatile compounds would be useful for determining the presence of 4MI. The specific aim of this study was to examine the effect of food additives in the formation of 4-MI and determine the correlation among color parameters, pyrazines as dominant flavor compounds, and 4-MI. A caramel model system was chosen that consisted of glucose and ammonium hydroxide (as class III) with food additives such as metal ions (Fe2+, Ca2+, Zn2+, Mg2+) and amino acids (cysteine, tryptophan), each at different levels.
■
MATERIALS AND METHODS
Reagents. 4-Methylimidazole (4-MI, purity ≥98%), 2-ethylimidazole (2-EI, purity ≥98%), D-glucose (purity ≥99.5%), sodium sulfite (purity ≥98%), ammonium hydroxide solution (purity ≥29%), acetic acid (purity ≥99.7%), bis-2-ethylhexylphosphate (BEHPA, purity ≥97%), and isobutyl chloroformate (IBCF, purity ≥98%) were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO). Isooctane, acetonitrile (MeCN), pyridine, and isobutyl alcohol (purity ≥99%) were Lichrosolv quality and purchased from Merck (Darmstadt, Germany). Magnesium sulfate (MgSO4·7H2O) and zinc sulfate (ZnSO4·7H2O) were purchased from Junsei Chemical Co. (Tokyo, Japan), and iron sulfate (FeSO4) was purchased from Kanto Chemical Co. (Tokyo, Japan). Calcium sulfate (CaSO4·0.5H2O) was purchased from Yukari Pure Chemical Co. (Osaka, Japan), and tryptophan (purity ≥99%) and cysteine (purity ≥99.5%) were purchased from Junsei and Fluka (Neu-Ulm, Germany), respectively. All reagents were analytical grade. Standard stock solutions of 4-MI and 2-EI used as an internal standard (IS) were prepared in 0.1 M HCl at a concentration of 20 and 1 mg/kg, respectively. All the solutions were kept at 4 °C until used. Sample Preparations of the Caramel Model Systems. Glucose and ammonium hydroxide both were dissolved in 30 mL of HPLC-grade water in a swing top bottle, each at a concentration of 1 M. To compare the influences of food additives on the formation of 4MI, all additives were added to glucose/ammonium solution individually at three distinct concentrations: 0.1, 0.05, and 0.01 M. The food additives were Fe2+ (as FeSO4), Ca2+ (as CaSO4), Zn2+ (as ZnSO4·7H2O), Mg2+ (as MgSO4·7H2O), cysteine, sodium sulfite, and tryptophan. The caramel model system solutions were heated at 100 °C for 2 h in an oven to instigate the Maillard reaction and cooled under running cold water to room temperature immediately in order to stop the reaction progress. All samples were stored at 4 °C, until they were analyzed. Analysis of 4(5)-Methylimidazole. The procedure to extract 4MI from a sample was carried out in the manner specified in a previous study.17 Each sample (100 μL) was added to 4 mL of phosphate buffer that was extracted from 8 mL of 0.1 M BEHPA in chloroform by mixing for 1 min and centrifugation at 3500 rpm for 10 min. A 7.5 mL portion of the chloroform phase was transferred to a second vial containing 4 mL of 0.1 M HCl and mixed again by vortexing for 1 min. After centrifugation at 3500 rpm for 5 min, a 1 mL aliquot of the aqueous phase was added with 100 μL of 2-EI used as IS, which was placed into a vial and mixed with 1 mL of MeCN−isobutyl alcohol−
■
RESULTS AND DISCUSSION
Measurement of 4-MI in Caramel Model System Solution. As mentioned earlier, 4-MI was not detected in caramel colors I or II but was identified in both colors III and IV. Thus, glucose and ammonium hydroxide, which create 6482
dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
Journal of Agricultural and Food Chemistry
Article
Table 1. Level and Reduction Rate of 4(5)-Methylimidazole Formed in Caramel Model System Solutions (n = 3)a model systems 1 M glucose/1 M ammonium hydroxide 1 M glucose/1 M ammonium hydroxide
concn of additives (M) sodium sulfite iron sulfate
556 583 110 269 452 483 506 576 656 552 652 460 507 581 409 457 556 397 444 609
0.1 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01
magnesium sulfate
calcium sulfate
zinc sulfate
tryptophan
cysteine
a
amount of 4-MI (μg/mL) ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
1.3 2.6 0.7 1.8 2.9 2.0 1.4 1.3 3.4 1.7 2.6 2.0 0.5 5.7 4.4 2.3 1.8 1.7 0.8 0.6
reduction (%)
g gh a b de ef f gh j g ij e f gh cd e g c de hi
0 −5 80 52 19 13 9 −3 −18 1 −17 17 9 −4 26 18 0 29 20 −10
Values followed by the same letter in a given column are not significantly different at p < 0.05.
Table 2. Color Measurement of Caramel Model System Solution by Colorimetrya model system 1 M glucose/1 M ammonium hydroxide 1 M glucose/1 M ammonium hydroxide
concn of additives (M) sodium sulfite iron sulfate
0.1 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01 0.1 0.05 0.01
magnesium sulfate
calcium sulfate
zinc sulfate
tryptophan
cysteine
L* 17.9 15.9 12.0 15.5 14.6 20.4 21.2 18.7 16.4 16.5 17.4 10.1 16.0 16.9 16.8 17.7 17.4 19.1 18.6 18.2
a* o g d f e s t q i j m c h l k n m r q p
1.3 3.6 3.8 2.3 3.0 −0.8 −1.6 0.6 2.6 2.7 1.9 3.9 2.0 2.4 2.0 1.2 1.8 −0.1 0.5 1.0
ΔEb
b* g m m i l d c f jk k h m h i h g h e f f
22.7 22.6 17.2 20.2 20.6 18.8 16.2 21.8 23.3 23.3 23.2 15.2 21.9 22.9 22.7 22.4 23.2 20.4 21.7 22.7
ij i e g g f d h k k k c h ijk ij i jk g h ij
0 3.1 8.5 3.7 4.3 5.0 7.9 1.4 2.1 2.1 1.0 11.2 2.2 1.5 1.4 0.4 0.9 2.9 1.5 0.3
a
Values followed by the same letter in a given column are not significantly different at p < 0.05. bGlucose/ammonium hydroxide model system was used as a control group for calculation of color differences.
five times. The recoveries that were measured for samples spiked with 4-MI at two concentration levels (20 and 100 μg/ mL) ranged from 84 to 111%. The precision of the interday and intraday analysis, expressed as the relative standard deviation (RSD), were 0.9% and 4.9−6.5%, respectively. Table 1 shows the amount of 4-MI formed in the caramel model system solutions that include each food additive at a level of 0.1, 0.05, and 0.01 M. A glucose and ammonium hydroxide (1 M) model system was used as a control. As each group was compared to the glucose/ammonium hydroxide model system, a reduction of 4-MI was observed in most caramel model systems with the 0.1 M food additives. In
colors III through heating, were used as a control group of the caramel model system in the present study. In a previous study, the Maillard reaction was investigated by using a model system that consisted of a sugar and an amine compound, and 4-MI was found in the glucose/ammonium hydroxide model system at a level that ranged from 490 to 710 μg/mL.21 An absolute calibration curve was obtained (y = 0.7958x + 2.6264, R2 = 0.9937) using 4-MI standard solutions with various concentrations (1, 5, 10, 20, 50, and 100 μg/mL). The recovery and precision of the caramel model system solution of glucose/ammonium hydroxide (1 M) with added sodium sulfite (0.1 M) were determined, and each test was replicated 6483
dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
Journal of Agricultural and Food Chemistry
Article
particular, there was a significant reduction of 4-MI formation in the caramel model system with 0.1 M Fe2+, but the amount of 4-MI increased in the caramel model system with the addition of 0.1 M Ca2+. When the D-glucose and ammonium hydroxide were heated, the amount of 4-MI formed was 556 ± 1.3 μg/mL. The addition of 0.1 M sodium sulfite increased the 4-MI formation by up to 583 ± 2.6 μg/mL (5%), whereas the formation level and reduction rate of 4-MI through the addition of food additives were 110 ± 0.7 (80%) in 0.1 M Fe2+, 483 ± 2.0 (13%) in 0.1 M Mg2+, 460 ± 2.0 (17%) in 0.1 M Zn2+, 409 ± 4.4 (26%) in 0.1 M tryptophan, and 397 ± 1.7 μg/mL (29%) in 0.1 M cysteine. The degree of contribution to the reduction of 4MI from food additives had the following order: Fe2+ > cysteine > tryptophan > Zn2+ > Mg2+. The formation levels of 4-MI ranged from 110 ± 0.7 (0.1 M Fe2+) to 452 ± 2.9 (0.01 M Fe2+), from 483 ± 2.0 (0.1 M Mg2+) to 576 ± 1.3 (0.01 M Mg2+), from 656 ± 3.4 (0.1 M Ca2+) to 652 ± 2.6 (0.01 M Ca2+), from 460 ± 2.0 (0.1 M Zn2+) to 581 ± 5.7 (0.01 M Zn2+), from 409 ± 4.4 (0.1 M tryptophan) to 556 ± 1.8 (0.01 M tryptophan), and from 397 ± 1.7 (0.1 M cysteine) to 444 ± 0.8 μg/mL (0.01 M cysteine). As a higher concentration of food additive was added, apart from the case of Ca2+, the amount of 4-MI created by the Maillard reaction solution decreased, while an especially low concentration of food additive was not sufficient to reduce the formation of 4-MI. Furthermore, the addition of 0.01 M of Ca2+, Mg2+, Zn2+, cysteine, and tryptophan led to a higher level of 4-MI than the control caramel model system, which was the Maillard reaction solution of 1 M glucose and 1 M ammonium hydroxide without additives. These results clearly suggest that some food additives at an appropriate concentration have a positive effect on the reduction of 4-MI in the Maillard reaction. This effect of the food additives on the formation of 4-MI in the Maillard reaction system was not previously reported. As food additives could be used in the manufacturing of foods, adding these food additives may be useful for reducing the formation of 4-MI during the Maillard reaction process. Color Differences in the Caramel Model System Solutions. Since the Maillard reaction induces a brown color with formation of melanoidins, color is one of the important factors when considering how much 4-MI is formed during the Maillard reaction. Table 2 shows the results obtained for three series of color differences: L*, a*, and b*. Significant differences according to the Duncan post hoc test are indicated by superscript letters. The L* values of the caramel model systems with added 0.1 M food additives including 0.1 M sodium sulfite decreased compared to glucose/ammonium hydroxide (1 M) model system, except for Mg2+ and cysteine. Contrarily, values of a* increased in the same caramel model systems, denoting the rise of redness characteristics when adding food additives. The b* values did not show a significant change for any caramel model system. Although the measurement of color from the Maillard reaction model systems has not been reported, the decrease of L* and the increase of a* were observed in the foods due to the brown pigment formation that was induced by the Maillard reaction.22,23 In the studies of Wang et al. and Ramonaitytė et al., the absorbance of the Maillard reaction model system showed increases at 420 and 450 nm, respectively, with addition of metal ions as a result of the Maillard reaction, indicating that the addition of metal ions accelerated the formation of the final browning compounds.24,25
In order to determine the correlation between the 4-MI concentration and color, a linear regression analysis was performed on each L*, a*, b*, and ΔE. The total color difference, expressed as ΔE, showed a lack of negative correlation with the 4-MI concentration (r = −0.4826). On the other hand, L*, a*, and b* values had no significant correlation with 4-MI (p < 0.05). Pyrazines Formed in Caramel Model System Solutions. Table 3 lists the 12 kinds of pyrazine detected in the study of the caramel model system solutions. The majority of the pyrazines detected were 2-methylpyrazine, with a peak area ratio (PAR) ranging from 2.27 to 6.29, followed by 2,6dimethylpyrazine ranging from 0.84 to 2.56 and 2,3dimethylpyrazine ranging from 0.10 to 0.33. It is generally known that pyrazines are among the typical products of the Maillard reaction between amino compounds and carbonyl compounds.26 All of the 12 pyrazine compounds confirmed in the present study were consistent with the previous literature that had investigated the pyrazines that could be derived from the Maillard reaction.26−29 It was difficult to find a relatively substantial relationship between 4-MI and the pyrazines formed in the greatest amounts, whereas 2-ethyl-6-methylpyrazine was among the 12 kinds of pyrazine compounds that showed a correlation with the concentration of 4-MI. Figure 1 showed a negative correlation (r = −0.8239) between 4-MI and 2-ethyl-6methylpyrazine, indicating that 2-ethyl-6-methylpyrazine is formed in inverse proportion to the 4-MI concentration (p < 0.05). When the 4-MI declined by 80% with the addition of Fe2+ into the glucose/ammonium hydroxide 1 M solution, the content of 2-ethyl-6-methylpyrazine increased from 0.03 to 0.21 PAR. Figure 2 shows the proposed formation pathway of 4-MI and 2-ethyl-6-methylpyrazine from methylglyoxal on the basis of the previous studies.13,21,26,30,31 Methylglyoxal, as the sugar-derived reactive intermediate during the Maillard reaction, reacts with ammonia. Interestingly, the nitrogen-containing reactive intermediates formed from methylglyoxal with ammonia can produce either 4-MI by interacting with formaldehyde or 2ethyl-6-methylpyrazine by reacting with ethylglyoxal. Therefore, if the nitrogen-containing reactive intermediates generated 4MI with formaldehyde, the formation of 2-ethyl-6-methylpyrazine might be reduced. Further study that elucidates a more accurate correlation between 4-MI and 2-ethyl-6-methylpyrazine will be needed. 4-MI investigated in the present study has received much attention from regulatory agencies as a possible carcinogenic compound. Therefore, it is important to find a way to reduce the 4-MI produced by the Maillard reaction. The results of the present study demonstrate that the addition of food additives such as metal ions and amino acids at an appropriate level could reduce the formation of 4-MI. With current results various food products such as a bakery product could be applied to reduce the amount of 4-MI. However, more detail application study would be carried out in the future. In addition, a correlation between 2-ethyl-6-methylpyrazine and the 4-MI concentration was determined. Since the analysis of 4-MI requires expensive technology and specialized knowledge, the correlation between color, pyrazine levels, and 4-MI levels could be useful for the evaluation of the 4-MI formed during the Maillard reaction. 6484
dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
0.23 0.05
0.03 0.01 0.01 0.01 0.02 0.04 0.05 0.01 12
Compounds content = peak area of each compound/peak area of undecane (internal standard). bConcentration of food additives (M).
0.11 0.07
Figure 1. Correlation between 4(5)-methylimidazole and 2-ethyl-6methylpyrazine.
Figure 2. Proposed formation pathway of 4(5)-methylimidazole and 2-ethyl-6-methylpyrazine.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: 82-2-2260-3370. Fax: 82-2-2285-3370. E-mail: [email protected]. Funding
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (Grant No.: 2012R1A1A2007551) and R&D Convergence Center Support Program, Ministry for Food, Agriculture, Forestry and Fisheries, Republic of Korea Notes
The authors declare no competing financial interest.
■
REFERENCES
(1) Tsuchida, H.; Komoto, M. Chemical studies on the reaction products of methylpentose and ammonia. Agric. Biol. Chem. 1967, 31, 185−189. (2) National Toxicology Program (NTP, 2007). Toxicology and Carcinogenesis of 4-Methylimidazole (CAS No. 822-36-6) in F344/N Rats ad B6C3F1Mice (Fee Studies); NTP Technical Report Series, No. 535, NIH Publication No. 07-4471; U.S. Department of Health and Human Service, NTP: Research Triangle Park, NC, 2007; pp 1−274. (3) 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 12; http://oehha.ca.gov/prop65/law/adopt020812.html (accessed Aug 20, 2013).
a
0.04
0.02
0.10 0.03
0.10 0.03 0.01 0.13 0.04 1.42 0.11 0.04 0.42 0.12 0.02 0.13 0.05 0.02 0.18 0.07 0.01 0.14 0.05 0.02 0.19 0.06 0.01 0.17 0.07 0.03 0.25 0.04 0.02 0.23 0.13 0.05 0.39 0.07 0.03 0.33 0.21 0.05 0.57 0.09 0.04 0.10 0.03 0.01 0.12 0.05
0.11 0.04 0.01 0.15 0.03 0.01
1.55 2.56 0.95
2.07
0.14 0.05 0.02 0.19 0.05 0.00
0.10 0.02
0.11 0.03 0.01 0.13 0.06
0.09 0.02 0.02 0.10 0.07
4.25 0.12 0.05 0.51 0.15 0.03
0.86 0.01 0.08 0.01 1.04 1.04
0.01
0.32 0.05
0.23 0.08 0.03 0.34 0.07
1.72 0.01 0.18 0.06 2.41
2.41 0.01 0.21 0.06
0.25 5.06 0.29 6.11 0.25 5.41
0.15 3.02 0.26 0.85 0.02 0.10 0.03 0.01 0.13 0.04 0.12 3.30
0.21 3.88 0.37 0.96 0.20 4.19 0.37 1.04 0.11 2.72 0.12 4.33
unsubstituted 2-methyl 2,5-dimethy2,6-dimethyl 2-ethyl 2,3-dimethyl 2-ethyl-6-methyl 2-ethyl-5-methyl 2-ethyl-3-methyl ethenyl 2,5-dimethyl-3ethyl 2,6-diethyl 1 2 3 4 5 6 7 8 9 10 11
0.15 3.11 0.30 0.84
0.11 2.27
0.10 3.26
0.19 4.13 0.36 1.02
0.18 4.38
0.18 4.25 0.32 0.96
0.18 4.12
0.24 5.77 0.34 2.47
0.27 6.29 0.34 2.30
0.27 5.95 0.27 1.85 0.02 0.37 0.09 0.04 0.33 0.10 0.02
0.15 3.38 0.25 0.89
0.05 0.05 0.1 0.01 0.05 0.01 0.05 0.1 0.01 0.05 0.1 0.01 0.05 0.1 0.1b pyrazines
magnesium sulfate calcium sulfate iron sulfate
Article
no.
glucose/ammonium hydroxide
sodium sulfite
Table 3. Amounts of Pyrazine Compounds Identified in Caramel Model Systems
peak area ratioa
0.1
zinc sulfate
cysteine
0.01
0.1
tryptophan
0.01
Journal of Agricultural and Food Chemistry
6485
dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
Journal of Agricultural and Food Chemistry
Article
(24) Wang, H.; Zhou, Y.; Ma, J.; Zhou, Y.; Jiang, H. The effects of phytic acid on the Maillard reaction and the formation of acrylamide. Food Chem. 2013, 141, 18−22. (25) Ramonaitytė, D. T.; Keršienė, M.; Adams, A.; Tehrani, K. A. The interaction of metal ions with Maillard reaction products in a lactose−glycine model system. Food Res. Int. 2009, 42, 331−336. (26) Hwang, H.; Hartman, T. G.; Rosen, R. T.; Lech, J.; Ho, C. Formation of pyrazines from the Maillard reaction of glucose and lysine-α-amine-15N. J. Agric. Food Chem. 1994, 42, 1000−1004. (27) Lee, K. G.; Jang, H.; Shibamoto, T. Formation of carcinogenic 4(5)-methylimidazole in caramel model systems: A role of sulphite. Food Chem. 2013, 136, 1165−1168. (28) Shibamoto, T. Heterocyclic compounds in browning and browning/nitrite model systems: Occurrence, formation mechanisms, flavor characteristics and mutagenic activity. Instrum. Anal. Foods 1983, 1, 229−278. (29) Yu, A.; Tan, Z.; Wang, F. Mechanistic studies on the formation of pyrazines by Maillard reaction between L-ascorbic acid and Lglutamic acid. LWT-Food Sci. Technol. 2013, 50, 64−71. (30) 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. (31) Yaylayan, V. A.; Haffenden, L. W. Mechanism of imidazole and oxazole formation in [13C-2]-labelled glycine and alanine model systems. Food Chem. 2003, 81, 403−409.
(4) Chan, P.; Mahler, J.; Travlos, G.; Nyska, A.; Wenk, M. Induction of thyroid lesions in 14-week toxicity studies of 2- and 4methylimidazole in Fischer 344/N rats and B6C3F1 mice. Arch. Toxicol. 2006, 80, 169−180. (5) Chan, P. C.; Hills, G. D.; Kissling, G. E.; Nyska, A. Toxicity and carcinogenicity studies of 4-methylimidazole in F344/N rats and B6C3F1 mice. Arch. Toxicol. 2008, 82, 45−53. (6) Patey, A. L.; Shearer, G.; Knowles, M. E.; Denner, W. H. B. Ammonia caramel: Specifications and analysis. Food Addit. Contam. 1985, 2, 107−112. (7) Hargreaves, M. B.; Jones, B. C.; Smith, D. A.; Gescher, A. Inhibition of p-nitro-phenolhydroxylase in rat liver microsomes by small aromatic and heterocyclic molecules. Drug Metab. Dispos. 1994, 22, 806−810. (8) IARC Monographs. 4-Methylimidazole. Monographs.iarc.fr/ ENG/Monographs/vol101/mono101-015.pdf (accessed Oct 16, 2012). (9) WHO. Toxicological Evaluation of Some Food Colours, Enzyme, Flavour Enhancers, Thickening Agents, And Certain Food Additives; International Program on Chemical Safety World Health Organization, 1975; http://www.inchem.org/documents/jecfa/jecmono/ v06je13.htm (accessed Aug 31, 2012). (10) EFSA Panel on Food Additives and Nutrient Sources added to Food (ANS). Scientific opinion on the reevaluation of caramel colours (E 150a,b,c,d) as food additives. EFSA J. 2011, 9 (3), 2004. (11) Nursten, H. The chemistry of nonenzymic browning. The Maillard Reaction: Chemistry, Biochemistry and Implications; Royal Society of Chemistry: London, UK, 2005; pp 5−30. (12) Komoto, M. Chemical studies on the reaction products of glucose and ammonia. III. Behaviors of glucosylamine in browning reaction of glucose and aqueous ammonia. Nippon Nogei Kagaku Kaishi 1962, 36, 403−407. (13) Jang, H. W.; Jiang, Y.; Hengel, Matt.; Shibamoto, T. Formation of 4(5)-methylimidazole and its precursors, α-dicarbonyl compounds, in Maillard systems. J. Agric. Food Chem. 2013, 61, 6865−6872. (14) Horikawa, H.; Okada, M.; Nikamura, Y.; Sato, A.; Iwamoto, N. Production of hydroxyl radicals and alpha-dicarbonyl compounds associated with Amadori compound−Cu2+ complex degradation. Free Radical Res. 2002, 36, 1059−1065. (15) Keller, C.; Wedzicha, B. L.; Leong, L. P.; Berger, J. Effect of glyceraldehyde on the kinetics of Maillard browning and inhibition by sulphite species. Food Chem. 1999, 66, 495−501. (16) Serpen, A.; Gökmen, V. Evaluation of the Maillard reaction in potato crisps by acrylamide, antioxidant capacity and color. J. Food Compos. Anal. 2009, 22, 589−595. (17) Cunha, S. C.; Barrado, A. I.; Faria, M. A.; Fernandes, J. O. Assessment of 4-(5-)methylimidazole in soft drinks and dark beer. J. Food Compos. Anal. 2011, 24, 609−614. (18) Thosen, M.; Willumsen, D. Quantitative ion-pair extraction of 4(5-)methylimidazole from caramel colour and its determination by reversed-phase ion-pair liquid chromatography. J. Chromatogr., A 1987, 211, 213−221. (19) Yanaguchi, H.; Masuda, T. Determination of 4(5)-methylimidazole in soy sauce and other foods by LC−MS/MS after solidphase extraction. J. Agric. Food Chem. 2011, 59, 9770−9775. (20) Calvo, C. Physical characterization and nutrient analysis. Optical properties. In Handbook of Food Analysis; Nollet, L. M. L. , Ed.; Marcel Dekker: New York, 2004; Vol. 1, pp 1−19. (21) Moon, J. G.; Shibamoto, T. Formation of carcinogenic 4(5)methylimidazole in Maillard reaction systems. J. Agric. Food Chem. 2011, 59, 615−618. (22) Bosch, L.; Alegría, A.; Farré, R.; Clemente, G. Fluorescence and color as markers for the Maillard reaction on milk−cereal based infant foods during storage. Food Chem. 2007, 105, 1135−1143. (23) Farroni, A.; Buera, M. P. Colour and surface fluorescence development and their relationship with Maillard reaction markers as influenced by structural changes during cornflakes production. Food Chem. 2012, 135, 1685−1691. 6486
dx.doi.org/10.1021/jf502008q | J. Agric. Food Chem. 2014, 62, 6481−6486
