Formation and Reduction of Furan in Maillard Reaction Model

Jun 9, 2014 - Analysis of Furan Using SPME–Gas Chromatography–Mass ..... Zeng , S.; Bai , X. Correspondence analysis of antioxidant activity and U...
0 downloads 0 Views 723KB Size
Article pubs.acs.org/JAFC

Formation and Reduction of Furan in Maillard Reaction Model Systems Consisting of Various Sugars/Amino Acids/Furan Precursors Heera Cho and Kwang-Geun Lee* Department of Food Science and Biotechnology, Dongguk UniversitySeoul, 30, Pildong-ro-1-gil, Jung-gu, Seoul 100-715, Republic of Korea ABSTRACT: To investigate the effect of the temperature on the formation of furan, various furan models were conducted at 90, 121, and 150 °C. A total 15 models, including alanine (ALA), serine (SER), ribose (RIB), RIB/ALA, RIB/SER, glucose (GLU), GLU/ALA, GLU/SER, sucrose (SUC), SUC/ALA, SUC/SER, furoic acid (FUR), GLU/FUR, acetaldehyde (ACET), and GLU/ ACET, were prepared. The maximum level of furan was detected in the GLU/SER and GLU/ALA models at a molar ratio of 0.5:0.5. The formation of furan was proportional to the temperature in all models. The RIB/SER model generated the greatest amount of furan among the 11 models ranging from 2.1 to 4931.9 ng mL−1 under all temperature conditions. Among the precursor models, the FUR model formed the greatest amount of furan ranging from 1058.2 to 13 927.9 ng mL−1 at all temperatures. KEYWORDS: model system, furan, Maillard reaction HS-SPME, GC−MS



INTRODUCTION Furan (C4H4O) is a heterocyclic and highly volatile compound with a boiling point of 31.4 °C. Furan has received attention since 1995 because it was classified as “possibly carcinogenic to humans” by the International Agency for Research on Cancer.1 Data have shown furan induces liver and kidney toxicity in rats and mice.2,3 The Food and Drug Administration (FDA) announced that furan was present in a large number of thermally processed foods.4 High amounts of furan have been detected in a variety of foods and beverages, including coffee, canned fish and vegetables, jarred source, fruit and vegetable juices, and bakery products.5,6 Furan is generated in food from multiple routes. Many researchers have reported that primary sources, such as sugars, amino acids, ascorbic acid, and polyunsaturated fatty acids, generate furan through thermal degradation, the Maillard reaction, and oxidation.7,8 In particular, the Maillard reaction between the carbonyl groups of sugars and the amino groups of amino acids are regarded as major pathways of furan formation. Yaylayan proposed a pathway of furan formation with sugars and amino acids.7 On the basis of this pathway, the degree of furan formation was evaluated using various heating sources. Ribose (RIB) as pentose, glucose (GLU) as hexose, and sucrose (SUC) as a disaccharide were chosen to determine the effect on the number of carbons. Serine (SER) and alanine (ALA) were used as amino group donors. In this study, single and binary primary source models were used. Single and binary models refer to one and two substrates, respectively. For example, a GLU or SER model is a single model and GLU/SER or GLU/ALA models are binary models. In this study, to determine the optimal molar ratio of a binary model, the concentration of furan of the GLU/SER and GLU/ALA models was compared at various molar ratios. There are many precursors for furan formation, and they have occasionally been used as food additives. To confirm the effect of various precursors on furan formation, the heating © 2014 American Chemical Society

process of the model including sugars, amino acids, and precursors was carried out in the same manner. Among furan precursors, we chose acetaldehyde (ACET) and 2-furoic acid, which are classified as edible synthetic flavoring substances.9,10 Several factors influence the formation of furan. Many studies have demonstrated that furan is formed in model systems during various heating processes.7 Fan et al. reported that the heating time, temperature, and pH affect the formation of furan in food.11−13 The aim of this study was to investigate the effect of the temperature and various ingredients in furan formation models. Each model was heated at 90, 121, and 150 °C. Eleven and four furan models were selected for clarification of the effect of the temperature and precursors on furan formation, respectively.



MATERIALS AND METHODS

Chemicals. L-ALA, D-SER, D-RIB, D-GLU, D-SUC, 2-furoic acid (FUR), and ACET were provided by Sigma-Aldrich Chemical (St. Louis, MO). Water and methanol were obtained from J.T. Baker (Phillipsburg, NJ). Furan (+99%, purity) and d4-furan (>99 atom % D) were purchased from Sigma-Aldrich. Stock solutions of furan and d4-furan were prepared weekly by dissolving standard solutions in cold methanol at a concentration of 100 μg mL−1. The stock solutions were further diluted in cold water to give a working solution with a concentration of 1 μg mL−1. The stock and working solutions were stored in headspace vials closed with silicone−polytetrafluoroethylene (PTFE) septa and aluminum seals (Supelco, Bellefonte, PA) at 4 °C until analysis. Method Validation for the Furan Analysis. The furan analytical method was validated using linearity, limit of detection (LOD), limit of quantification (LOQ), and repeatability. The calibration curve was drawn using the ratio of the area of furan versus the area of d4-furan). A 10 μL aliquot of internal standard (1 μg mL−1) was spiked in 10 mL Received: Revised: Accepted: Published: 5978

April 4, 2014 June 3, 2014 June 9, 2014 June 9, 2014 dx.doi.org/10.1021/jf501619e | J. Agric. Food Chem. 2014, 62, 5978−5982

Journal of Agricultural and Food Chemistry

Article

Figure 1. Effect of the molar ratio of amino acid to sugar on the furan formation. Within same figure, lower case alphabetical letters indicate spastically differences (p < 0.05, Duncan’s multiple range test). Analysis of Furan Using SPME−Gas Chromatography−Mass Spectrometry (GC−MS). A SPME fiber was pushed through a vial septum. The fiber was pushed out of the housing and exposed to the headspace (1.5 cm in depth) of the vial. Extraction was carried out at 50 °C for 20 min using a heating block. After extraction, the fiber was pulled into the housing and the SPME device was removed from the vial and inserted into the injection port of a gas chromatograph for thermal desorption analysis (4 cm in depth). The fiber was desorbed for 5 min at 230 °C to eliminate any residual volatile compounds. The SPME fiber was a 75 μm Carboxen−polydimethylsiloxane (CAR−PDMS, Supleco, Bellefonte, PA) fiber. The extraction was carried out at 50 °C for 20 min on a heating block, and desorption was carried out at 230 °C for 5 min. GC−MS analyses were performed with an Agilent Technologies 6890 N GC system coupled with an Agilent Technologies 5975 mass selective detector (Palo Alto, CA) equipped with a HP-PLOT Q capillary column (HP-PLOT Q, 15 m × 0.32 mm inner diameter, 20 μm film thickness, J&W Scientific, Folsom, CA). Operational conditions were as follows: carrier gas, highpurity helium (99.999%); flow rate, 1 mL min−1; and injector temperature, 230 °C. The oven temperature was programmed from 50 °C (held for 5 min) to 230 °C (held for 2 min). The mass spectrometer was operated in electron impact ionization mode (70 eV) and selected ion monitoring mode. The m/z (mass/charge) of 68 for furan and m/z of 72 for d4-furan were used as quantification ions, and the m/z of 39 for furan and 42 for d4-furan were used as confirmation ions. Measurement of Ultraviolet (UV) Absorbance in Each Model System. A total of 1 mL of each sample was transferred to a cuvette. The absorbance of each sample was measured with a UV spectrometer (UV-1800, Shimadzu, Tokyo, Japan). The wavelength was set to 420 nm. Water was used as the control. All experiments were repeated 3 times. Statistical Analysis. Results are presented as the mean ± SD. Data were analyzed by one-way analysis of variance (ANOVA), and multiple comparisons were performed with Duncan’s test. The SAS 9.3 program was used for the statistical analysis (SAS Institute, Cary, NC). Statistical significance was set at p < 0.05.

of water in a 20 mL headspace vial with a syringe to prepare the calibration curve. Seven different amounts (5, 10, 20, 50, 100, 500, and 1000 μL) of furan (1 μg mL−1) were spiked into the vial. All experiments were performed at