Heterocyclic Amines in Several Types of Cooked Meat and Chicken Dishes

Food Sci. Technol. Res., 14 (2), 169 175, 2008 Heterocyclic Amines in Several Types of Cooked Meat and Chicken Dishes which Form Part of the Korean Diet Cheon-Ho jo 1, 3, Young-Eun Sim 1, Hyo Min Lee 2, Taikyung ryeom 2, Seung-Woon myung 1* 1 Department of Chemistry, Kyonggi University, Kyeonggi-do, 443-760, South Korea 2 Risk Management Research Department National Institute of Toxicological Research, Korea Food & Drug Administration, Seoul, 122-704, South Korea 3 Hazard Substances Analysis Department, Center for Food & Drug Administration, Busan, 608-829, South Korea Received August 29, 2007; Accepted December 19, 2007 Heterocyclic amines (HAs) are formed in protein-rich foods during high temperature cooking such as frying and grilling. The amount of HAs formed in several of the most frequently eaten food dishes in Korea such as Bulgogi (roasted beef), Galbi (grilled rib of beef), Samgyupsal (grilled boned rib of pork), fried chicken, charcoal grilled chicken and electrically grilled chicken, was determined. The ground-up samples were extracted under both acidic and basic conditions, cleaned on SPE cartridges, and the concentrations of fifteen HAs determined using electrospray ionization LC/MS. Limits of detection (LODs) and quantification (LOQs) were 0.99 ng g -1, and 3.33 ng g -1 for both MeIQ and Trp-P-1, and 0.51 ng g -1 and 1.67 ng g -1 for the remaining HAs. The absolute recoveries of the raw meat samples at a concentration of 3.33 ng g -1 (n=3) ranged between 60 and 100% for most of the HAs. The precision of inter-day assays was 1.0-15.3%. The range of inaccuracy for inter-day assays was -18.2 to 18.3% (bias) at a concentration of 25 ng g-1. Norharman, Harman, MeAαC and MeIQx were the amines the most frequently found in the cooked dishes, with concentration ranges of 17.4-261.0, 21.6-267, 3.61-27.7, and 6.45-20.8 ng g -1, respectively. While PhIP was the most frequently found HAs in the cooked pork (Samgyupsal) (5.2-17.2 ng g -1 ) and chicken breast (9.3-66.1 ng g -1 ), being present at high concentrations, it was detected in only one of the ten samples of cooked beef. Trp-P-2 was found in a few of the foods and its concentration was less than 5 ng g -1. In the investigation of the HAs contents according to the cooking method for the chicken breast, the highest amounts and frequencies of HAs, were found in the electrically grilled chicken breast, with 4,8-Di- MeIQx, 7,8-DiMeIQx and AαC being especially abundant. Since Bulgogi, Galbi, Samgyupsal and cooked chicken breast are the most popular meals among Korean people, regular consumption of such items may contribute substantially to their dietary intake of HAs. Keywords: High performance liquid chromatography/mass spectrometry, Heterocyclic amines, Muscle meats Mutation, Korean diet Introduction The contribution of the heterocyclic amines (HAs) to the mutagenic activity detected in some cooked meat and fish products has been known for 30 years (Nagao et al., 1977) and these chemicals are one of the major mutagens found in cooked foods. HAs are formed during the high temperature *To whom correspondence should be addressed. E-mail: swmyung@kyonggi.ac.kr cooking of protein-rich foods such as meats, poultry and fish. At present, more than 20 HAs have been isolated and characterized from different cooked foods. Their formation is highly dependent on various factors such as the cooking temperature, cooking method, type of food, fat and moisture content, and the presence of additives. In general, cooking at high temperature (e.g. frying, oven broiling, grilling, barbecuing) forms very high levels of heterocyclic amines Skog et al., 2000). The main precursors for HAs in foods are

170 creatinine, which is present in meats and fish in high concentrations, and various amino acids and sugars (Jägerstad et al., 1984). Many of these HAs have been isolated from various protein-rich foods including cooked meats and fish, and some of them have been detected in environmental components such as airborne particles (Manabe et al., 1992), indoor air, cigarette smoke, diesel exhaust particles, and cooking fumes (Thiébaud et al., 1995). HAs have also been detected in fermented foods such as soy sauce and vinegar (Adachi et al., 1991) and PhIP (2-amino-1-methyl-6-phenyl-imidazo[4,5- b]-pyridine) has been identified in a number of beer and wine samples (Richling et al., 1996). Humans are regularly exposed to HAs through diet and the environment. In the Korean diet, meat and poultry dishes are cooked using several methods that provide a suitable flavor. These cooking methods create temperatures high enough to favor the formation of mutagenic compounds such as heterocyclic amines (Skog et al., 1995). Although HAs are formed at low concentration levels, their intake may be dangerous for human health, because of the total amount of cooked meat consumed. The Ministry of Agriculture & Forestry, Republic of Korea, reported that the Korean per capita consumption of meat in 2003 was 6.8 kg of beef, 6.6 kg of chicken and 17.9 kg of pork. Because of the low levels of the HAs that are formed, it is necessary to use a selective and sensitive detection system and to prevent interference from the complex sample matrix (Pais et al., 2000). In this paper, we report the results for HAs in meat and poultry food purchased from Korean food restaurants. The foods which Korean favorite meats are Bulgogi (roasted beef), Galbi (grilled rib of beef), Samgyupsal (grilled boned rib of pork), oil fried chicken, charcoal grilled chicken, electrically grilled chicken were studied. Experimental Chemicals 2-amino-3-methylimidazo[4,5-f]quinoline (IQ), 2-aminodipyrido[1,2-a:30,20-d]imidazole (Glu-P-2), 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ), 2-amino-6-methyldipyrido[1,2-a:30,20-d]imidazole (Glu-P-1), 2-amino-3,8-dimethylimidazo [4,5-f]quinoxaline ( M e I Q x ), 2 - a m i n o - 3, 7, 8 - t r i m e t h y l i m i d a z o [ 4, 5 - f]quinoxaline (7,8-DiMeIQx), 2-amino-3,4,8-trimethylimidazo trimethylimidazo [4,5-f]quinoxaline (4,8-DiMeIQx), 9H-pyrido[4,3-b]indole (Norharman), 1-methyl-9H-pyrido[4,3-b]indole (Harman), 2-amino- 3,4,7,8-dimethylimidazo[4,5-f]quinoxaline (Tri-MeIQx), 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2), C-H. Jo et al. 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1), 2-amino-9Hpyrido[2,3-b]indole (AαC) and 2-amino-3-methyl-9H-pyrido[2,3-b]indole (MeAαC) were obtained from Toronto Research Chemicals (Ontario, Canada). A mixture of the different HAs (about 1 mg ml -1 ) dissolved in methanol was used as a spiking mixture. Water, methanol, and dichloromethane were HPLC-grade and obtained from Burdick & Jackson (Muskegon, MI). 1-naphthylamine (internal standard, I.S.) was purchased from Aldrich Co. (WI, USA). Extrelut-20 (diatomaceous earth extraction cartridge) was obtained from Merck (Darmstadt, Germany). The PRS column (propylsulfonic acid) and C 18 Bond Elut were provided by Varian Co. (Harbor City, USA). Sample preparations The samples used in this survey were consisted of 2 beef dishes (Bulgogi (roasted beef)) and Galbi (broiled rib of beef), 1 pork dish (Samgyupsal (boned rib of pork)), and 3 chicken dishes (fried chicken, charcoal barbecue chicken, electrical barbecue chicken). The meat and chicken dishes were purchased from different five Korean food restaurants in district of Suwon, Korea on same day. The cooked samples were frozen and homogenized with a HMF-1000 mixer (Hanil, Seoul, Korea). All samples were stored at -70 in deep freezer. Sample preparation and clean-up were carried out according to the method proposed by Gross (Gross et al., 1990). Briefly, the cooked samples were lyophilized and ground. Three grams of each sample was saponified with 12 ml of 1 M NaOH by sonication for 1 hr and shaking for 2 hr until dissolution. The alkaline solution obtained was introduced into an Extrelut-20 column coupled to a Bond-Elut PRS column which was preconditioned with 5 ml of 0.1 M HCl, 10 ml of water and 5 ml of methanol. The analytes adsorbed on the Extrelut packing were extracted using 80 ml of dichloromethane, which was introduced into the PRS column coupled on-line. The Extrelut column was discarded, and the PRS column was successively rinsed with 15 ml of MeOH: H 2 O (4:6, v/v) and 2 ml of water. The extracted solution was evaporated to a final volume of 1 ml using a rotary evaporator. The PRS column was then coupled to a C 18 column, and the concentrated solution loaded into the C 18 column. The adsorbed HAs were eluted from the C 18 column using 20 ml of 0.5 M AcONH 4 (ph 8). Then, after first rinsing with 5 ml water, the adsorbed HAs were eluted from the C 18 column, using 2 ml of MeOH:NH 4 OH(9:1, v/v). The solvent was evaporated with a stream of nitrogen and the analytes were redissolved in 1 ml of methanol containing 10 ng of internal standard (1-naphthylamine). Finally, 1 μl of the extract was injected into the LC/MS system. Apparatus A lyophilizer with a stoppering tray dryer was obtained from Labconco Co. (Kansas, USA). A Supelco

Heterocyclic Amines in Several Types of Cooked Meat and Chicken Dishes which Form Part of the Korean Diet 171 SPE vacuum manifold (Supelco, PA, USA) was used for the manipulations with the solid-phase extraction cartridges, and the nitrogen evaporator with streaming nitrogen and rotary evaporator for the solvent evaporation were obtained from Eyela Co. (Tokyo, Japan). A vortex mixer (Thermolyne, Iowa, USA) and multi-shaker (Eyela, Tokyo, Japan) were used for mixing the samples and a Raynger surface thermometer (Raynger, Seoul, Korea) was used to monitor the temperature. Instrumentation All analyses were performed on an Agilent Technologies (Palo Alto, USA) model Series 1100 series LC/MSD Trap equipped with a binary pump system and electrospray ionization (ESI). The chromatographic separation of the HAs was carried out on a Zobax 300SB-C 18 column (0.30 150 mm I.D.) with a particle size of 5 μm (Agilent Technologies, Palo Alto, USA). Optimal separation was achieved with a binary mobile phase at a flow rate of 15 μl min -1. Solvent A was a 7.5 mm ammonium formate buffer adjusted to ph 2.8 and solvent B was acetonitrile. The gradient program was 0-40% B for 0-20 min and return to the initial conditions (100% A) for 15 min equilibration. The volume of the sample injected was 1 μl. The optimal working parameters for the mass spectrometer were: capillary voltage, 3.5 kv; nebulizer temperature, 325 ; drying gas flow, 5 L min -1 (N 2 ); skimmer, 40 V; capillary exit offset, 137.7 V. The mass range for data collection in full scan mode was m/z 50-300. Method validation A commercial meat and chicken extract free from HAs was used as raw material. The calibration curves and precision were obtained by spiking the raw meat with the stock solution. The concentrations of the spiked analytes in the raw meat were in the range 1.67-167 ng g -1 (n=3). The calibration curves were made from the peak area ratio of the HAs to the internal standard (1-naphthylamine). The absolute recoveries were calculated by comparing the amounts of the compounds present in final solution after covering the whole method (described in section of sample preparation ) with the spiked meat with the amounts that were injected without the sample preparation procedure. And three sets of absolute recovery samples were analyzed. Run-to-run precision was evaluated by injection three replicates of HAs standard solution at low concentration level 3.33 ng g -1, medium concentration level 25.0 ng g -1 and high concentration level 167 ng g -1 on the same day. Day-today precision was evaluated injection three daily replicates on three different days. Results and Discussion Validation Calibration The calibration curves were linear in the range of 1.67-167 ng g -1 for IQ, Glu-P-1, MeIQx, Glu-P-2, 7,8-DiMeIQx, 4,8-DiMeIQx, Tri-MeIQx, Trp-P-2, MeAαC, PhIP, Harman, Norharman and AαC, and 3.33-167 ng g -1 for MeIQ and Trp-P-1. The calibration range, coefficient of correlation and equations corresponding to the calibration curves, consisting of five points for the HAs, are shown in Table 1. Recoveries, LODs, LOQs and precision The limits of detection (LODs) was defined as the lowest concentration at which the signal to noise ratio were 3, and the limits of quantitation (LOQs) was defined as a concentration where the accuracy and precision lies within 20% limits. Table 1. Validation for the quantitation of the HAs in the various cooked foods. Compounds Calibration curve R 2 LOD * LOQ ** Recovery(%)±SD(%RSD) (3.33 ng g -1 ) IQ 1.67 ~ 167 y = 0.0259x - 0.0068 0.9973 0.51 1. 67 83.0 ± 1.0 (11.2) Glu-P-1 1. 67 ~ 167 y = 0.0313x + 0.0556 0.9988 0.51 1. 67 85.6 ± 0.5 (3.7) MeIQx 1. 67 ~ 167 y = 0.0729x + 0.0561 0.9954 0.51 1. 67 110 ± 1.6 (7.6) Glu-P-2 1. 67 ~ 167 y = 0.0188x + 0.0648 0.9994 0.51 1. 67 112 ± 0.5 (4.4) 7,8-DiMeIQx 1. 67 ~ 167 y = 0.1085x + 0.3308 0.9987 0.51 1. 67 103 ± 2.5 (5.8) 4,8-DiMeIQx 1. 67 ~ 167 y = 0.1054x + 0.2815 0.9989 0.51 1. 67 111 ± 2.5 (5.8) MeIQ 3.33 ~ 167 y = 0.0344x - 0.1336 0.9951 0.99 3.33 88.8 ± 34.2 (5.4) *** Tri-MeIQx 1. 67 ~ 167 y = 0.0911x + 0.0663 0.9981 0.51 1. 67 98.5 ± 0.5 (6.0) Trp-P-2 1. 67 ~ 167 y = 0.0318x + 0.0701 0.9974 0.51 1. 67 59.1 ± 0.7 (2.4) MeAαC 1. 67 ~ 167 y = 0.0296x + 0.0737 0.9982 0.51 1. 67 59.8 ± 0.5 (4.4) Trp-P-1 3. 67 ~ 167 y = 0.0241x + 0.3772 0.9987 0.99 3.33 108 ± 1.1 (9.4) PhIP 1. 67 ~ 167 y = 0.0435x + 0.1474 0.9991 0.51 1. 67 71.5 ± 2.0 (8.43) Harman 1. 67 ~ 167 y = 0.0316x + 0.0562 0.9998 0.51 1. 67 66.5 ± 1.7 (11.2) Norharman 1. 67 ~ 167 y = 0.0246x - 0.005 0.9983 0.51 1. 67 78.6 ± 2.9 (30.3) AαC 1. 67 ~ 167 y = 0.0233x + 0.0674 0.9976 0.51 1. 67 61.0 ± 3.0 (30.5) * LOD : Limit of Detection (at S/N > 3), ** LOQ : Limit of Quantification (at S/N>10 and 20% < RSD), *** at 25.0 ng g -1.

172 The LODs and LOQs were 0.99 ng g -1, and 3.33 ng g -1 for both MeIQ and Trp-P-1, and 0.51 ng g -1 and 1.67 ng g -1 for the remaining HAs in the meat samples, respectively (Table 1). The absolute recoveries of the raw meat samples at a concentration of 3.33 ng g -1 (n=3) ranged between 60 and 100% for most of the HAs (Table 1). The inter-assay precision (within three days) of the proposed method is shown in Table 2. Three sets of quality control samples (25.0 ng g -1, n=3) were analyzed along with calibration samples in one batch. The precisions for the interday assay ranged from 1.0 to 15.3% (RSD). The degree of inaccuracy for the inter-day assay was -18.2 to 18.3% (bias) at a concentration of 25.0 ng g -1 (Table 2). C-H. Jo et al. HAs determination In general, DMIP, MeIQx, 4,8-Di- MeIQx, Norharman, Harman, PhIP, Trp-P-1, AαC and MeAαC were the HAs the most frequently found in the analyzed meat dishes, while Glu-P-2, IQ, MeIQ, Glu-P-1, 7,8-DiMeIQx and Trp-P-2 were found in only a few of them (Busquets et al., 2004) (Table 3-5). From the five Galbi (grilled rib of beef) samples, five HAs were identified. Harman and Norharman were found in the high concentration ranges of 16.3-80.0 and 30.8-176.0 ng g -1, respectively. MeAαC was detected at concentrations ranging from 3.3 to 13.2 ng g -1 with a relatively high frequency (60%), and MeIQx and Trp-P-2 were found in sample II at concentrations ranging from 18.2 to 3.8 ng g -1, respectively (Table 3). Table 2. Inter-assay and accuracy of HAs in 25.0 ng g -1. Compounds Bias* (%) Precision(%) Compounds Bias(%) Precision(%) Glu-P-1-13.2 7.1 MeIQx 3.2 6.5 4,8-DiMeIQx -5.7 3.1 MeIQ 18.3 10.4 Tri-MeIQx -14.6 2.0 Trp-P-2 10.2 1.0 MeAαC -4.3 2.1 Trp-P-1 4.8 15.0 PhIP -13.4 3.7 Harman -0.2 1.4 Norharman -18.2 15.3 AαC -17.0 8.0 Glu-P-2-0.6 2.9 7,8-DiMeIQx -13.9 1.5 IQ 4.7 3.5 *Bias = (Xc-Xm)*100/Xc or (Xa-Xm)*100/Xa Xc : values from calibration curve, Xm : measured values, Xa : added values Table 3. HAs content in the cooked Galbi and Bulgogi. HAs Galbi (grilled rib of beef) Bulgogi (roasted beef) Sample Sample IQ ND ND ND ND ND - ND ND ND ND ND - Glu-P-1 - - MeIQx 18.2 ND ~ 18.2 17.5 ND ~ 17.5 Glu-P-2 ND - ND - 7,8-DiMeIQx - 2.8 ND ~ 2.8 4,8-DiMeIQx - 3.4 ND ~ 3.4 MeIQ - ND - Tri-MeIQx - - Trp-P-2 3.8 ND ~ 3.8 - MeAαC 4.9 13.2 3.3 3.3 ~ 13.2 6.4 5.0 15.6 5.0 ~ 15.6 Trp-P-1 ND ND ND - ND ND ND - PhIP - 32.4 ND ~ 32.4 Harman 16.3 80.0 70.5 16.3 ~ 80.0 240.0 106.0 21.6 26.5 21.6 ~ 240.0 Norharman 30.8 ND 176 ND 30.8 ~ 176.0 205.0 17.4 ND 117.0 17.4 ~ 205.0 AαC ND ND - ND ND ND - ND : Not detected.

Heterocyclic Amines in Several Types of Cooked Meat and Chicken Dishes which Form Part of the Korean Diet 173 Seven HAs were found in the five Bulgogi (roasted beef) meat samples, which is the most favored beef dish in Korea. As in the case of the Bulgogi samples, Norharman and Harman were present at relatively high concentrations (17.4-240.0 ng g -1 ) and frequencies (60-80%) (Table 3). MeAαC was found in the concentration range of 5.0-15.6 ng g -1 with a frequency of 60%. MeIQx, 7,8-DiMeIQx, 4,8-DiMeIQx and PhIP were detected in one sample. The presence of multiple HAs at high concentrations in the Bulgogi samples could be explained by the presence of ingredients and additives that probably increase the formation of HAs in the cooking process, because Bulgogi is cooked in a mixture of several spices, fat, oil and water, using ingredients such as onion, scallion, mushroom, and garlic. Although Galbi and Bulgogi are both made from beef, the amount of HAs formed is lower than that for Bulgogi because Galbi is grilled with the spices under the conditions in which the suet of beef is more likely to be discarded during the cooking process. Grilled Samgyupsal (boned rib of pork), which is Korean bacon made from pork, is one of the most favored pork dish in Korea. Harman was found at a high concentration (101.0-267.0 ng g -1 ) in two of the five products, and Norharman and MeAαC were detected in the concentration ranges of 17.9-33.2 and 5.3-9.1 ng g -1 with 80 % frequency, respectively. MeIQx (10.6-20.8 ng g -1 ) and PhIP ( 5.2-17.2 ng g -1 ) were detected in all of the samples with relatively high average concentrations of 13.7 and 12.1 ng g -1, respectively (Table 4). The concentration of HAs in the chicken breast cooked Table 4. HAs content of cooked Samgyupsal (grilled boned rib of pork) products. Compounds Conc. Frequency (%) IQ ND ND ND ND ND 0 - Glu-P-1 0 - MeIQx 13.2 11.3 20.8 12.4 10.6 100 10.6 ~ 20.8 Glu-P-2 ND ND ND ND ND 0-7,8-DiMeIQx 0-4,8-DiMeIQx 0 - MeIQ 0 - Tri-MeIQx 0 - Trp-P-2 0 - MeAαC 6.3 6.0 5.3 9.1 80.0 5.3 ~ 9.1 Trp-P-1 ND ND ND ND 0 - PhIP 17.2 5.5 5.2 15.3 17.2 100 5.2 ~ 17.2 Harman ND ND ND 101.0 267.0 40.0 101.0 ~ 267.0 Norharman 23.2 17.9 21.4 33.2 80.0 17.9 ~ 33.2 AαC ND ND ND ND 0 - ND : Not detected. HAs Fried-chicken in soybean oil Charcoal grilled chicken Electrical grilled chicken sample sample sample IQ ND ND ND ND ND - ND ND ND ND ND - ND ND ND ND ND - Glu-P-1 - - - MeIQx 7.81 6.45 9.69 6.45 ~ 9.69 14.7 6.9 6.87 ~14.7 - Glu-P-2 ND ND ND - ND ND - - 7,8-DiMeIQx - - 17.3 5.30 5.30 ~ 17.3 4,8-DiMeIQx - - 15.3 7.50 7.50 ~ 15.3 MeIQ - - ND ND - Tri-MeIQx - - - Trp-P-2 4.72 ND ~ 4.72 - - MeAαC 10.3 10.5 14.6 3.61 27.7 3.61 ~ 27.7 12.7 15.6 8.1 3.6 3.62 ~ 15.6 8.40 12.2 9.30 8.40 ~ 12.2 Trp-P-1 ND ND ND ND ND - ND ND ND ND - ND ND ND - PhIP 15.3 21.4 12.8 12.8 ~ 21.4 15.5 12.4 30.2 63.2 9.3 9.30 ~ 63.2 31.8 66.1 21.4 13.5 13.5 ~ 66.1 Harman 32.8 27.6 27.6 ~ 32.8 ND 27.0 50.0 30.4 ND 27.0 ~ 50.0 23.9 ND 92.3 ND 68.4 23.9 ~ 92.3 Norharman 24.3 ND 38.1 24.3 ~ 38.1 18.0 ND 79.2 103 21.7 18.0 ~ 103 ND 21.5 261 ND 21.5 ~ 261 AαC ND ND 14.2 ND ~ 14.2 ND ND ND ND - 36.3 17.3 48.3 23.1 58.0 17.3 ~ 58.0 ND : Not detected. Table 5. HAs content for three cooking methods for the chicken breast.

174 by three different methods, viz. frying in oil, charcoal grilling and electrical grilling, were investigated (Table 5). From the chicken fried in soybean oil, MeIQx and PhIP were found in the concentration ranges of 6.4-9.7 and 12.8-21.4 ng g -1 with 60% frequency, respectively, and MeAαC was detected in all of the samples at an average concentration of 13.4 ng g -1. Harman and Norharman were found in the concentration ranges of 27.6-32.8 and 24.3-38.1 ng g -1, respectively (Table 5). In the charcoal grilled chicken breast, Harman and Norharman were detected in the concentration ranges of 27.0-50.0 and 18.0-103.0 ng g -1, respectively. PhIP was found in all of the chicken beast samples at an average concentration of 26.1 ng g -1 and MeAαC and MeIQx were found in the concentration ranges of 3.6-15.6 and 6.9-14.7 ng g -1, respectively. In the electrically grilled chicken breast, 7,8-Di- MeIQx and 4,8-DiMeIQx were detected in two samples with concentrations ranging from 5.3-17.3 ng g -1, whereas these HAs were not found in the fried and charcoal grilled chicken breast samples. Also, AαC was found in the concentration range of 17.3 ~ 58.0 ng g -1. Figure 1 displays amount of HAs in chicken with the cooking conditions. The process that we heated the chicken is this: First chickens went through the Oil, then Charcoal and lastly, the Electrical. Interesting thing is that, MeIQx and MeAαC in chicken has been decreased while all the others have been increased. It is assumed that the temperature and the time heated on the chickens have affected the condition of the chicken when it was cooked. Significant risks on human health through long-term exposure of heterocyclic amines are scientifically unclear. However, among the known mutagenic heterocyclic amines in rats and mece, Trp-P-1, Trp-P-2, AαC, MeAαC, glu-p-1, Glu-P-2, IQ, MeIQ, MeIQx and PhIP have been verified to be carcinogenic or mutagenic (Kataoka et al., 1997). Earlier reports have shown low levels of HAs in meat, beef and chicken(klassen et al., 2002, Philippe et al., 2000, Sinha et al., 1998, Skog et al., 1998). The presence of MeIQx Ave. Conc. (ng/g) 60.0 50.0 40.0 30.0 20.0 10.0 0.0 MeIQx 7,8-DiMeIQx 4,8-DiMeIQx Trp-P-2 MeAαC PhIP Harman Norharman AαC Oil fried Charcoal barbecue Electrical barbecue Cooking condition Fig. 1. The amount of HAs in chicken with the different cooking conditions. and PhIP has been reported in meat extract (2.3 and 1.5 ng g -1, respectively), grilled beef (16.4 and 67.5 ng g -1, respectively) and fried chicken (1.0 and 70.0 ng g -1, respectively) (Skog et al., 1998). In this study, MeIQx and PhIP found in meat extract (20.8 and 17.2 ng g -1, respectively), grilled beef (18.2 ng g -1 and not detected, respectively) and fried chicken (9.69 and 21.4 ng g -1, respectively). This investigation showed that cooked beef, pork and chicken in Korea contain several HAs. Their levels would be expected to vary widely depending on the cooking conditions and the type of meat (e.g. the fat and water content, the presence or absence of other ingredients, etc.). These data will be highly valuable for devising cooking methods to be used in the home designed to reduce the concentrations of HAs, for estimating the average daily intake of these HAs from the various cooked meats by the average Korean and for assessing the possible health hazard implications. Reference Adachi, J., Mizoi, Y., Naito, T., Yamamot, K., Fujiwara, S., Ninomiya, I. (1991). Determination of β-carbolines in foodstuffs by high-performance liquid chromatography and high-performance liquid chromatography-mass spectrometry, J. Chromatogr. A, 538, 331-339. Busquets, R., Bordas, M., Tribio, F., Puignou, L., Galceran, M.T. (2004). Occurrence of heterocyclic amines in several homecooked meat dishes of the Spanish diet, J. Chromatogr. B, 802, 79-86. Gross, G.A. (1990). Simple methods for quantifying mutagenic heterocyclic aromatic amines in food products, Carcinogenesis, 11, 1597-1603. J_gerstad, M., Olsson, K., Grivas, S., Negishi, C., Wakabayashi, K., Tsuda, M., Sato, S., Sugimura, T. (1984). Formation of 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline in a model system by heating creatinine, glycine and glucose, Mutat. Res., 126, 239-244. Kataoka, H. (1997). Methods for the determination of mutagenic heterocyclic amines and their applications in environmental analysis, J. Chromatogr. A, 774, 121-142. Klassen, R. D., Lewis, D., Lau, B. P.-Y., Sen, N. P., (2002), Heterocyclic aromatic amines in cooked hamburgers and chicken obtained from local fast food outlets in the Ottawa region, Food Res.Int., 35, 837-847. Manabe, S., Wada, O., Morita, M., Izumikawa, S., Asakuno, K., Suzuki, H. (1992). Occurrence of carcinogenic amino-α-carbolines in some environmental samples, Environ. Pollut., 75, 301-305. Nagao, M., Honda, M., Seino, Y., Sugimura, T. (1977). Mutagenicities of smoke condensates and the charred surface of fish and meat, Cancer Lett., 2, 221-226. C-H. Jo et al. Pais, P., Knize, M.G. (2000). Chromatographic and related tech-

Heterocyclic Amines in Several Types of Cooked Meat and Chicken Dishes which Form Part of the Korean Diet 175 niques for the determination of aromatic heterocyclic amines in foods, J. Chromatogr. B, 747, 139-169. Philippe A. Guy, Eric Gremaud, Janique Richoz, Robert J. Turesky, (2000). Quantitative analysis of mutagenic heterocyclic aromatic amines in cooked meat using liquid chromatography-atomospheric pressure chemical ionization tandem mass spectrometry, J. Chromatogr. A, 883, 89-102. Richling, E., Herderich, M., Schreier, P. (1996). High performance liquid chromatography-electrospray tandem mass spectrometry (HPLC-ESI-MS-MS) for the analysis of heterocyclic aromatic amines (HAA), Chromatographia, 42, 7-11. Sinha, R., Knize, M. G., Salmon, C. P. (1998) Heterocyclic amine content of pork products cooked by different methods and to varying degrees of doneness, Food Chem. Toxicol., 36, 289-297. Skog, K., Solyakov, A., Jägerstad, M. (2000). Effects of heating conditions and additives on the formation of heterocyclic amines with reference to amino-carbolines in a meat juice model system, Food Chem., 68, 299-308. Skog, K., Steinneck, G., Augustsson, K., Jägerstad, M. (1995). Effect of cooking temperature on the formation of heterocyclic amines in fried meat products and pan residues, Carcinogenesis, 16, 861-867. Skog, K.I., Johansson, M.A., Jägerstad, M.I. (1998). Carcinogenic heterocyclic amines in model systems and cooked foods: A review on formation, occurrence and intake, Food Chem. Toxicol., 36, 879-896. Thiébaud, H.P., Knize, M.G., Kuzmicky, P.A., Hsieh, D.P., Felton, J.S. (1995). Airborne mutagens produced by frying beef, pork and a soy-based food, Food Chem. Toxicol., 33, 821-828.