17. :1... v.11. .7 .1 :v I: . 3sz gm...» v . s... 2:. B... 3 E... 5.5... . . h . . {tub-v1 4i. .073..an , . :0 ha. t. .5 . 3|.?.:.au- . b 01..) 13%.. ‘i‘i 1:,5 . . . .... . u. 3.2.fi:-:§.fln c............X . . L s .34-}. ‘00.... I» I... .. . Ct. it, ”um—9.. 2.2. a. :xuf. . fizizlrn! if. We ‘ 255M: 1...}: 5.01. ,Lv \. 5J4“... . I . izut, 3n... $5.935? ; O .5 3‘! 70 it... 503.. )\r c::..‘:& . ii. ). a...) 2.2.. .1123. .331 1.1.2959 13.3»? .55.. .......R:..:..1.. :{152 3.2 1.4:: t .73. :39. L. - u -L. «V2.5. that...“ I.) , $3.63.». A L .n.é.n...L..!.. Lia}! .4...va 3 . :1 L... 9:. n: . g is"... KvIIb-uu. l .. 3m tile ‘15.}. i.» iui :14. s. .. :2 V V; {‘3 011.13..) 3. I» 1.79. L. .3 1 ‘§ 1.- . . .... THES!‘ 11111111'111'1111111111‘111'11111111111111111111 3 1293 01411 This is to certify that the thesis entitled FORMATION AND INHIBITION OF HETEROCYCLIC AROMATIC AMINES IN FRIED GROUND BEEF PATTIES presented by Zsuzsanna Balogh has been accepted towards fulfillment of the requirements for Master's degree in Food Science film“ 4241/ Major progssorq 9—0.3“95— 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution _ '—._4_k 1——.v~ LIBRARY Michigan State University PLACE It RETURN BOX to man this checkout from your rocord. TO AVOID F INES rotum on or before dd. duo. DATE DUE DATE DUE DATE DUE MSU to An Atflnnativo Action/Ema! Oppoitmlty Intuition WM' FORMATION AND INHIBITION OF HETEROCYCLIC AROMATIC AMINES IN FRIED GROUND BEEF PATTIES By Zsuzsanna Balogh A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1995 ABSTRACT FORMATION AND INHIBITION OF HETEROCYCLIC AROMATIC AMINES IN FRIED GROUND BEEF PATTIES By Zsuzsanna Balogh This study was designed to determine the effect of vitamin E and oleoresin rosemary on heterocyclic aromatic amine formation in fried ground beef patties. Patties were fiied at three temperatures (175°C, 200°C, 225°C) for 6 and 10 minutes per side to determine the conditions for optimum heterocyclic aromatic amine (HAA) formation. HAAs were isolated by solid phase extraction and quantitated by high performance liquid chromatography. Greatest concentrations were generated when patties were fi'ied at 225°C for 10 min/side ---- 16.9 ng/g raw meat PhIP [2-amino-l-methyl-6-phenylimidazo (4,5-b) pyridine] and 3.1 ng/g raw meat MeIQx [2-arnino-3,8-dimethy1imidazo (4,5-f) quinoxaline]. Vitamin B, when used at two concentrations (1 and 10% based on fat content) and added directly to the ground beef pattie, reduced PhIP concentrations by approximately 72%. Smaller but more variable reductions were achieved for MeIQx. Oleoresin rosemary also successfully reduced HAA formation, although not to the same extent as vitamin E. The application of vitamin E (1% based on fat content) to the surface of the pattie provided comparable inhibition of HAA formation. The concentrations of five HAAs studied were all significantly reduced (p<0.006). The average reductions ranged from 49% to 77%. To my husband, Zoltan, and my daughter, Krisztina, for their enduring love, patience and kindness. ACKNOWLEDGEMENTS I wish to express my sincere gratitude to my major professor, Dr. J. Ian Gray, for his encouragement, guidance and support throughout my graduate studies, and for his friendship during these years. Appreciation is also extended to Doctors Alden M. Booren, Gale Strasburg and Matthew J. Zabik for serving on my guidance committee, and for their critical review of this thesis. I am also grateful to Dr. James S. Felton and Dr. Mark G. Knize fi'om Lawrence Liverrnore National Laboratory, University of California, for hosting my visit and for the generosity of sharing their experiences in this field. Further acknowledgment is due to Dr. Gian Gross fi'om Nestec Ltd., Lausanne, Switzerland, for providing the 'Trace-Calc' spreadsheet. I also wish to thank Dr. Shaun C. Chen for his assistance and helpful hints with the computer work during this research. Thanks to all my fiiends and colleagues at Michigan State University who have been so helpful and supportive and such good company. Special thanks to Pervaiz Akhtar, Eric Cole, Arti Arora, Dr. Enayat Gomaa, Caroline Saba and Lisa Scranton for their friendship and moral support. Finally, my deepest gratitude to my husband, Zoltan, and to my daughter, Krisztina, for their love, understanding words of encouragement and support for my graduate study. TABLE OF CONTENTS INTRODUCTION .............................................................................................. 1 LITERATURE REVIEW .................................................................................... 4 Formation of heterocyclic aromatic amines ............................................. 4 Types of mutagenic compounds found in fried ground beef ........ 4 Heterocyclic aromatic amines in fn'ed ground beef ..................... 4 Mutagenicity of heterocyclic aromatic amines ............................ 19 Mechanism of heterocyclic aromatic amine formation ................ 20 Factors influencing heterocyclic aromatic amine formation .................... 30 Fat .............................................................................................. 30 Creatine/creatinine ...................................................................... 3 1 Amino acids and dipeptides ........................................................ 33 Sugars ......................................................................................... 34 Cooking time and temperature .................................................... 35 Inhibition of mutagen formation ............................................................. 37 Sugars and other carbohydrates ................................................... 37 Soy protein concentrate .............................................................. 3 8 Defatted glandless cottonseed flour ............................................. 39 Synthetic antioxidants ................................................................. 3 9 Vitamin E ................................................................................... 40 Tea phenolics .............................................................................. 41 Flavone ....................................................................................... 42 Quantitation of mutagens ........................................................................ 42 MATERIALS AND METHODS ......................................................................... 45 Materials ................................................................................................. 45 Methods .................................................................................................. 45 Moisture and fat determination ................................................... 45 Influence of temperature and time of frying on heterocyclic aromatic amine formation in ground beef patties ................................................... 46 Frying of ground beef patties ...................................................... 46 Statistical analysis ....................................................................... 46 Inhibition of heterocyclic aromatic amine formation in fried ground beef patties by direct addition of phenolic antioxidants .......................... 47 Frying of ground beef. ................................................................ 47 Statistical analysis ....................................................................... 47 Inhibition of heterocyclic aromatic amine formation in fried ground beef patties through surface application of vitamin E ............................. 48 Frying of ground beef patties ...................................................... 48 Statistical analysis ....................................................................... 48 Analysis of heterocyclic aromatic amines in fried ground beef patties... 49 Extrelut cartridge assembly ......................................................... 49 Extrelut - PRS tandem (propylsulfonic acid silica) ...................... 49 Removal of interferences from the PRS cartridge ........................ 50 Transfer of heterocyclic aromatic amines from PRS to C 18 ........ 50 Elution of heterocyclic aromatic amines fiom the C18 cartridges 50 Quantitative determinations of heterocyclic aromatic amines .................. 53 High performance liquid chromatographic (HPLC) analysis of heterocyclic aromatic amines ...................................................... 53 Calibration .................................................................................. 53 Analysis ...................................................................................... 53 RESULTS AND DISCUSSION .......................................................................... 55 Optimization of the extraction/quantitation procedures for heterocyclic aromatic amines ..................................................................................... 55 Influence of temperature and time of frying on heterocyclic aromatic amine formation in ground beef patties .................................................. 58 Inhibition of heterocyclic aromatic amine formation in fiied ground beef patties by direct addition of phenolic antioxidants .......................... 64 Inhibition of heterocyclic aromatic amine formation in fried ground beef patties through surface application of vitamin E ............................. 69 Mechanism of heterocyclic aromatic amine formation and inhibition ...... 73 SUMMARY AND CONCLUSIONS .................................................................. 76 FUTURE RESEARCH ....................................................................................... 77 REFERENCES ................................................................................................... 80 vii LIST OF TABLES LITERATURE REVIEW _T_alie_ l. Heterocyclic aromatic amine content of cooked foods .............................. 7 2. Mutagenicities of heterocyclic aromatic amines and typical carcinogens in Salmonella typhimurium (Sugimura andSato, 1982; Sugirnura et al., 1988) ........................................................................................................ 20 3. Heterocyclic aromatic amines produced in model systems from creatin(in)e and amino acids, with or without sugars (Skog, 1993) ............ 23 RESULTS AND DISCUSSION Mg 4. Percent recoveries of heterocyclic aromatic amines in ground beef patties using solid phase extraction ........................................................... 57 5. Heterocyclic aromatic amine concentrations in ground beef patties (ng/g cooked meat) fried using different time/temperature combinations ............ 59 6. Heterocyclic aromatic amine concentrations in ground beef patties (ng/g cooked meat) using difierent time/temperature combinations of frying ..... 60 7. Inhibition of heterocyclic aromatic amine formation in fiied beef patties by the direct addition of antioxidants ........................................................ 65 8. Formation and inhibition of heterocyclic aromatic amines in fried ground beef patties following surface application of vitamin E ............................. 7O viii 9. Concentrations of heterocyclic aromatic amines (ng/ g cooked beef) in fiied patties prepared from fresh ground beef and from the same beef that had been frozen for two months ................................................................ 71 LIST OF FIGURES LITERATURE REVIEW figure 1. Chemical structures of the principal mutagens in cooked foods (Skog, 1993) ....................................................................................... 5 2. A possible pathway of browning in the Maillard reaction through a novel free radical (N amiki and Hayashi, 1981) ................................... 25 3. Postulated reaction route for formation of IQ compounds. R, X, and Y may be H or Me; Z may be CH or N (Jagerstad et al., 1983a) .............. 27 4. Alternative route for formation of IQ compounds. R, X, and Y may be H or Me; Z may be CH or N (Nyhammar, 1986) ................................. 29 MATERIALS AND METHODS figure 5. Extraction of heterocyclic aromatic amines from fried ground beef patties following the method of Gross and Gruter (1992) ................... 51 6. Solid phase extraction to detect heterocyclic aromatic amines from food systems ........................................................................................ 52 RESULTS AND DISCUSSION Figrge 7. Formation of heterocyclic aromatic amines in ground beef patties . fried for 6 minutes at three temperatures .............................................. 61 Formation of heterocyclic aromatic amines in ground beef patties fried for 10 nrinutes at three temperatures ............................................ 62 Inhibition of heterocyclic aromatic amine formation in fried beef patties by the direct addition of antioxidants ........................................ 66 INTRODUCTION A series of mutagenic and carcinogenic heterocyclic aromatic amines have been found in meat and fish cooked at temperatures over 150°C. The most common heterocyclic aromatic amines identified in fiied ground beef are: IQ (2- amino-3-methy1imidazo[4,5-f]quinoline), MeIQ (2-amino-3,4-dimethylimidazo[ 4,5-flquinoline), MeIQx (2-arnino-3,8-dimethylimidazo[4,S-flquinoxaline), 4,8- DiMeIQx (2-amino-3,4,8-trimethylimidazo[4,5-f]quinoxaline), and PhIP (2- amino-l-methyl-6-phenylimidazo[4,5-b]pyridine. Many of these compounds are multi-potential carcinogens in rodent bioassays (Sugimura and Wakabayashi, 1990; Ohgaki et al., 1991), and one of the most abundant heterocyclic aromatic amines, PhIP, has been shown to induce colon and mammary carcinomas in rats (Ito et al., 1991; Ochiai et al., 1991). Epidemiological studies suggest that the consumption of well-done red meat is associated with a high risk of colon and other cancers (Norrel et al., 1986; Schifl‘man and Felton, 1990; Steineck et al., 1990; Willett et al., 1990). Recent estimates of potential human cancer potency are consistent with an upper-bound cancer risk between 10"3 and 10'4 for an average lifetime cooked-beef intake of 3.3 g/kg/day (or approximately 0.5 lb/day) (Bogen, 1994). Precursors of these mutagenic/carcinogenic compounds in beef are creatine/creatinine, amino acids and sugars (Jagerstad et al., 1983a). However, heterocyclic aromatic amines can also be formed in dry-heated mixtures of amino acids and creatine (Yoshida et al., 1984; Taylor et al., 1987; Knize et al., 1988a; Overvik et al., 1989; Felton and Knize, 1990). Cooking temperature and time are also important factors in the formation of these mutagens (Commoner et al., 1978a; Bjcldanes et al., 1983; Miller and Buchanan, 1983; Overvik et al., 1984; Knize et al., 1985, 1994; Reutersward et al., 1987a,b; Nielsen et al., 1988). The formation of heterocyclic aromatic amines has been suggested to follow the Maillard reaction (Spingam and Garvie, 1979; Shibamoto et al., 1981; Wei, 1981; Powrie ct al., 1982) through vinylpyrazine, vinylpyridine and aldehyde formation (J agerstad et al., 1983b). However, the formation of the free radical, N,N'-disubstituted pyrazine cation, by early carbon fragmentation prior to the Amadori product was demonstrated by Namiki and Hayashi (1981). Thus, another route for the formation of heterocyclic aromatic amines was suggested to proceed via a free radical process (Pearson et al., 1992), for which further support was provided by Milic et a1. (1993). Reports of synthetic phenolic antioxidants inhibiting the formation of these compounds lend credence to this theory (Wang et al., 1982; Barnes et al., 1983; Chen et al., 1992; Faulkner, 1994). Inhibition of the formation of IQ-like compounds by vitamin E was demonstrated by Chen et a1. (1992). However, the reported concentrations of the mutagenic compounds were IOOO-fold higher than those cited by other researchers (Felton et al., 1986a,b; Sugirnura et al.,1988; Thiebaud et al., 1994). Thus, the quantitative data of Chen et al. (1992) are suspect as no confirmatory studies were carried out. Recently, Faulkner (1994) confnmed the inhibition of PhIP formation by vitamin B using both the Salmonella typimurium overall mutagenicity test and an analytical procedure to quantitate the extent of inhibition. This analytical procedure is a challenging one which is reflected in the fact that reported heterocyclic aromatic amine concentrations in beef are based on recoveries of standards compounds that range from as low as 5% to as high as 85% (Jackson et al., 1994; Johansson and Jagerstad, 1994; Knize et al., 1994; Thiebaud et al.,1994). Thus, some modification of the method is necessary to obtain reproducible recoveries of the heterocyclic aromatic amines in cooked meats. The objectives of this study are: (1)To optimize the extraction and analysis of heterocyclic aromatic amines in flied ground beef patties using the Standard Addition Quantitation procedure developed by Gross and Gruter (1992). (2) To compare the formation of heterocyclic aromatic amines in ground beef patties flied at various time/temperature combinations. (3) To evaluate the inhibition of heterocyclic aromatic amine formation by natural phenolic antioxidants (a) by the direct addition of vitamin E and oleoresin rosemary to the beef pattie before flying; (b) through surface application of vitamin E to the beef pattie before flying. LITERATURE REVIEW Formation of heterocyclic aromatic amines Types ofmutagenic compoundsfound in fried ground beef Several mutagenic/carcinogenic substances are produced or introduced into foods during their cooking, processing, and storage. Among the first reported were the polycyclic aromatic hydrocarbons, including benzo[a]pyrene, formed by the pyrolysis of fat which dripped onto the heated coals during the barbecuing of meats (Lijinsky and Shubik, 1964). Another group of compounds was reported by Japanese investigators, who found that heating proteins or amino acids to high temperatures (>300°C) produced several potent mutagens. These were called 'pyrolytic mutagens' (Matsumoto et al., 1977; Nagao et al., 1977). Cooking meat at lower temperatures produces another group of mutagenic compounds (Commoner et al., 1978a; Dolara et al., 1979), often referred to as 'thermic mutagens'. Several have been identified in cooked meat, fish and food grade beef extracts (Table l). Thermic mutagens are heterocyclic aromatic amines and are often called aminoimidazoazaarenes. They Can be broken down into four categories : quinolines, quinoxalines, pyridines, and furopyridines ( Skog, 1993). The chemical structures of the principal mutagens in cooked foods are shown in Figure l. Heterocyclic aromatic amines infried ground beef QUINOLINES: The heterocyclic aromatic amines, 2-amino-3-methylimidazo[4,5-t]- quinoline (IQ) and 2-amino-3,4-dimethylimidazo[4,5-fJ-quinoline (MeIQ), were isolated from the crust of broiled sardines (Kasai et al., 1980a,b; 1981a). Early studies reported the presence of IQ in fried ground beef at concentrations ranging . Quinolines Gig: on, MeIQC /N\CII Quinoxalines NH N=( 2 N=§:{H N\ CH l 3 EN NMeIQx CH3 IQx 4-MCIQX NH =r 2 NH N ==[ [1, N \ CH3 H,C ’N H3C \ 7,3—DiMeiQ, . \N CH Pyridine: 4,8-DiMelQ, PhIP CH N’ 3 “G 9mm m}mfl H30 WQNH, DMIP Furopyridines N H,c >~NH, N\ MeIFP CH3 Figure 1. Chemical structures of the principal mutagens in cooked foods (Skog, 1993). . from 0 to 20 ng/g (Barnes et al., 1983; Felton et al., 1984; Turesky et al., 1988). Reported concentrations for MeIQ are lower (F elton et al., 1986a; Yamaizumi et al., 1986; Gross et al., 1993). QUINOXALINES: The first quinoxaline to be identified in flied ground beef was 2-amino-3,8- dimethylimidazo[4,5-t]-quinoxaline (MeIQx) (Kasai et al., 1981b), followed by 2- amino-3,7,8-trimethylimidazo[4,5-t]-quinoxaline (7,8-DiMeIQx) (Negishi et al., 1984a), and 2-amino-3,4,8-trimethylimidazo[4,5-fJ-quinoxaline (4,8-DiMeIQx) (Grivas et al., 1985). MeIQx has been identified in flied ground beef at concentrations ranging flom non-detectable to 12.3 ng/g (Kasai et al., 1981b; Hargraves and Pariza, 1983; Felton et al., 1984, 1986a; 1992; Wakabayashi et al., 1986; Murray et al., 1988; Sugimura et al., 1988; Turesky et al.,l988, 1989; Knize et al., 1994). The other two quinoxalines, 4,8-DiMeIQx and 7,8-DiMeIQx, are both present in flied ground beef, but in relatively small concentrations: 0 to 3.9 ng/g (Felton et al., 1986a; 1992; Turesky et al., 1988; Murray et a1, 1988; Sugimura et al., 1988; Knize et al., 1994). PYRIDINE: 2-Amino-1-methyl-6-phenylimidazo[4,S-fl-pyridine (PhIP), first isolated flom the crust of flied ground beef by F elton et al. (1986b), is the most abundant heterocyclic aromatic amine in cooked meat, with concentrations ranging from O to 67.5 ng/g (Felton et al. 1986b; 1992; Gross et al., 1989; Gross, 1990; Hayatsu et al., 1991, Knize et al., 1994; Thiebaud et al., 1994). Two other pyridines, 2- amino-n,n,n-trimethy1imidazopyridine (TMIP) and 2-amino-1,6- dimethylimidazopyridine (DMIP) have also been identified in flied meat (Becher et al., 1988; 1989; Felton et al., 1984). Table 1. Heterocych aromatic amine content of cooked foods Food type Mutagen Amount1 Temp2 Time3 Weight4 RefS BEEF Fried MeIQx 0.5 0 43 4,8-DiMeIQx 3.9 275 15 0 38 8-MeIQx 12.3 200 15 0 38 8-MeIQx 4.0 150 6 0 25 IQ 1.9 200 15 0 38 BOUILLON Heated 4,8-DiMeIQx 0.3 0 33 8-MeIQx 0.6 0 33 PhIP 0.3 0 33 EXTRACT Boiled 4,8-DiMeIQx 28.0 1 l4 4,8-DiMeIQx 0.0-3.7 1 39 4,8-DiMeIQx 0.0-4.4 l 9 4,8-DiMeIQx 0.0 l 33 4,8-DiMeIQx 2.5-4.9 l 10 7,8-DiMeIQx 0.0 1 39 8-MeIQx 28.0 1 l4 8-MeIQx 3.1 1 41 8-MeIQx 20.5 1 38 8-MeIQx 8.5-30.0 1 9 8-MeIQx 23.0-69.0 1 40 8-MeIQx 0.0-44.0 1 10 8-MeIQx 0.6 1 33 8-MeIQx 3.1 1 43 8-MeIQx l 1.7-36.4 1 39 AaC 0.0 1 10 IQ 0.0 1 4 l IQ 0.0 1 14 IQ 0.5 1 36 IQ 4.8-6.2 1 40 IQ 0.0-6.2 l 39 IQ 0.0-8.0 l 9 PhIP 3.62 1 15 PhIP 0.0 1 9 PhIP 0.0 l 19 Food type Mutagen Amount1 Temp2 Time3 Weight4 Ref5 PhIP 0.0 1 33 4'-OH-PhIP 21.0 15 6O 1 27 EXTRACT with creatine Heated 4-CH20H-8-Mex 6.2 121 60 1 24 4-CH20H-8-Mex 6.7 160 300 l 24 4-CH20H-8-Mex 7.2 200 300 1 24 FLAVOR 4,8-DiMeIQx 0.0 l 16 4,8-DiMeIQx 0.0 l 16 MeIQx 0.0-12.5 1 16 MeIQx 0.0-4.4 1 16 Roasted 4,8-DiMeIQx 0.0 1 16 MeIQx 0.0-4.4 1 16 Grilled 4,8-DiMeIQx 0.0 1 16 MeIQx 0.0 1 16 GROUND Charbroiled 4,8-DiMeIQx 0.2 6 l 17 4,8-DiMeIQx 0. 1 10 1 17 MeIQ 0.0 6 1 17 MeIQ 0.4 10 1 17 MeIQx 1.0 6 1 17 MeIQx 0.4 10 1 17 IQ 0.0 6 1 17 IQ 0.1 10 1 17 PhIP 0.0 6 1 17 PhIP 0.0 10 1 17 Broiled IQ 0.5 l 44 Grilled AorC 0.0 270 3 1 12 AorC 0.0 270 5 1 12 AaC 0.0 270 7 1 12 4,8-DiMeIQx 0.0 270 3 1 12 4,8-DiMeIQx 0.0 270 5 1 12 4,8-DiMeIQx 0.0 270 7 1 12 MeIQx 0.8 270 3 1 12 MeIQx 2.0 270 5 1 12 MeIQx 0.0 270 7 1 12 PhIP 0.7 270 3 1 12 PhIP 1.4-4.8 270 5 1 12 Food type Mutagen Amount1 Temp2 Time3 Weight4 Ref5 PhIP 0.0 270 7 1 12 Fried AorC 21.0 277 6 l 37 DiMeIQx 4.5 277 6 l 37 4,8-DiMeIQx 0.5 300 6 0 7 4,8-DiMeIQx 0.0 275 5 0 39 4,8-DiMeIQx 0.0 275 10 0 39 4,8-DiMeIQx 3.9 275 15 0 39 4,8-DiMeIQx 0.5-1.2 200 2 32 4,8-DiMeIQx 0.0 250 10 0 9 4, 8-DiMeIQx 0. 12 0 35 4,8-DiMeIQx 0.3 0 25 4,8-DiMeIQx 0.0-0.28 2 5 4,8-DiMeIQx 0.54 250 12 1 25 4,8-DiMeIQx 0.0 150 2 0 26 4,8-DiMeIQx 0.0 150 4 0 26 4,8-DiMeIQx 0. 1 150 6 0 26 4,8-DiMeIQx 0.7 150 10 0 26 4,8-DiMeIQx 0.0 190 2 O 26 4,8-DiMeIQx 0. 10 190 4 0 26 4,8-DiMeIQx 0.55 190 6 0 26 4,8-DiMeIQx 2.6 190 10 0 26 4,8-DiMeIQx 0.0 230 2 0 26 4,8-DiMeIQx 0. 15 230 4 0 26 4,8-DiMeIQx 0.25 230 6 O 26 4,8-DiMeIQx 9.35 230 10 O 26 4,8-DiMeIQx 0.7 225 6 l 29 4.8-DiMeIQx 3. 1 1 25 7,8-DiMeIQx 0.0 275 5 0 39 7,8-DiMeIQx 0.0 275 10 0 39 7,8-DiMeIQx 0.7 275 15 0 39 MeIQ 0.0 300 5.5 O 4 4-MeIQ 0.1 300 6 0 7 MeIQx 1.0 250 6 1 6 MeIQx 16.4 277 6 1 37 MeIQx 1.0 300 5.5 O 4 MeIQx 0.0 O 19 MeIQx 0.0-0.68 2 5 MeIQx 0.3 0 43 MeIQx 1.3-2.4 200 2 32 10 Food type Mutagen Amount1 Temp2 Time3 Weight4 RefS MeIQx 2.7 275 5 0 39 MeIQx 4.2 275 10 0 39 MeIQx 12.3 275 15 0 39 MeIQx 0.5-1.5 0 40 8-MeIQx 0.1 O 6 8-MeIQx 0.45 190 O 14 8-MeIQx 1. l 250 10 1 9 8-MeIQx 1.0 300 6 0 7 8-MeIQx 0.64 0 35 8-MeIQx 0.8 1 25 8-MeIQx 2.95 250 6 1 25 8-MeIQx 0.0 150 2 0 26 8-MeIQx 0.0 150 4 0 26 8-MeIQx 0. 15 150 6 0 26 8-MeIQx 2.7 150 10 0 26 8-MeIQx O. 1 190 2 0 26 8-MeIQx 0.25 190 4 O 26 8-MeIQx 1.3 190 6 0 26 8-MeIQx 5. l 190 10 0 26 8-MeIQx 0.0 230 2 0 26 8-MeIQx 0.4 230 4 0 26 8-MeIQx 1. 1 230 6 0 26 8-MeIQx 8.0 230 10 0 26 8-MeIQx 2.2 225 6 1 29 8-MeIQx 10.8 1 25 IQ 0.5-20.0 240 5 0 1 IQ 0.02 250 6 l 6 IQ 0.5 2 44 IQ 0.0 192 0 18 IQ 0.3 275 5 0 39 IQ 0.3 275 10 0 39 IQ 1.9 275 15 0 39 IQ 0.02 300 5.5 l 4 IQ 0.0 150 2 0 26 IQ 0.0 150 4 0 26 IQ 0.1 150 6 0 26 IQ 1.5 150 10 O 26 IQ 0.1 190 2 0 26 IQ 0.1 190 4 0 26 ll Food type Mutagen Amountl Temp2 Time3 Weight4 RefS IQ 0.45 190 6 0 26 IQ 0.82 190 10 O 26 IQ 0.0 230 2 0 26 IQ 0.15 230 4 0 26 IQ 0.25 230 6 0 26 IQ 1.8 230 10 O 26 IQ 0.0 250 10 1 9 PhIP 15.0 300 5.5 1 4 PhIP 67.5 277 6 1 37 PhIP 1.2 250 10 1 9 PhIP 5.0 1 25 PhIP 0.56 1 15 PhIP 0.0 150 2 0 26 PhIP 0.0 150 4 O 26 PhIP 0.25 150 6 0 26 PhIP 0.9 150 10 0 26 PhIP 0.0 190 2 0 26 PhIP 0.15 190 4 0 26 PhIP 1.9 190 6 0 26 PhIP 6.0 190 10 O 26 PhIP 0.55 230 2 0 26 PhIP 1.35 230 4 O 26 PhIP 4.1 230 6 0 26 PhIP 21.5 230 10 0 26 PhIP 16.4 225 6 1 29 PhIP 21.8 1 25 TMIP 0.5 300 6 0 7 Trp-P-l 0.0 300 6 0 8 Trp-P-l 0.19 0 35 Trp-P-2 0.0 200 0 31 Trp-P-2 0.21 0 35 STEAK Broiled or flied 4,8-DiMeIQx 1.3 190 3 1 10 4,8-DiMeIQx 2.0 190 6.5 1 10 4,8-DiMeIQx 0. 1 225 6 1 33 8-MeIQx 2.1 1 0 35 8-MeIQx 5. 1 190 3 l 10 8-MeIQx 8.3 190 6.5 1 10 8-MeIQx 0.5 225 6 1 33 12 Food type Mutagen Amountl Temp2 Time3 Weight4 RefS AaC 1.2 0 35 AaC 3.2 190 3 1 10 AorC 8.9 190 6.5 l 10 Glu-P-l 0.0 0 18 G1u-P-2 0.0 O 45 IQ 0.19 0 35 PhIP 15.7 1 15 PhIP 23.5 190 3 l 10 PhIP 48.5 190 6.5 l 10 PhIP 0.6 225 6 1 33 Trp-P-l 53.0 0 45 Trp-P-l 0.21 0 35 Trp-P-2 0.25 O 35 BONITO Grilled 4,8-DiMeIQx 5.4 220 15 l 22 8-MeIQx 5.2 220 15 1 22 Grilled, dried 8-MeIQx 2.5 O 22 CHICKEN Charbroiled 4,8-DiMeIQx O. 1 l 33 8-Me1Qx 0.3 1 33 Broiled 4,8-DiMeIQx 0.81 1 35 MeIQx 2.1 44 8-MeIQx 2.33 l 35 AaC 0.2 1 l 35 PhIP 38.1 1 15 Trp-P-l 0.12 1 35 Trp-P-2 0.18 1 35 Fried Trp-P-l 0.0 300 6 O 8 CONSOMME' Heated 4,8-DiMeIQx 0.0 O 33 8-MeIQx 0.1 O 33 PhIP 0.0 0 33 EEL, roasted Fried, canned 7,8-DiMeIQx 5.3 180 4 l 28 8-MeIQx 1. 1 180 4 l 28 FALUN SAUSAGE 13 Food type Mutagen Amount1 Temp2 Time3 Weight4 Refs Boiled, smoked 4,8-DiMeIQx 0.0 160 2.5 l 17 MeIQ 0.0 160 2.5 1 l7 MeIQx 0.6 160 2.5 1 17 IQ 0.3 160 2.5 1 l7 PhIP 0.0 160 2.5 1 l7 FISH FLOUNDER 4,8-DiMeIQx 0.6 1 17 Smoked MeIQ 0.3 1 l7 MeIQx 0.0-2.9 1 l7 IQ 0.7 l 17 PhIP 0.0 l 17 I-IERRING Fried 4,8-DiMeIQx 0.3 1 17 MeIQ 0.1 1 17 MeIQx 0.6 1 l7 IQ 0.2 1 17 PhIP 0.0 1 17 POLLACK Fried 4-MeIQ 0.03 260 8 1 46 4,8-DiMeIQx O. 1 260 8 1 46 8-MeIQx 6.44 260 8 1 46 IQ 0.16 260 8 1 46 PhIP 69.2 260 8 l 46 SALMON Baked 8-MeIQx 0.0 200 20 1 1 1 8-MeIQx 4.6 200 30 l 1 l 8-MeIQx 3.1 200 40 1 l 1 AaC 0.0 200 20 1 1 1 AorC 0.0 200 30 1 1 1 AaC 0.0 200 40 1 1 1 PhIP 0.0 200 20 l 1 l PhIP 18.0 200 30 1 1 1 PhIP 5.9 200 40 1 1 1 Broiled, flesh 4-MeIQ 0.6-2.8 1 44 Broiled, skin 4-MeIQ 1.1-1.7 1 44 Broiled 4-MeIQ O. l-O.9 l 3 MeIQ 1.4-5.0 1 1 1 Broiled, flesh IQ 0.3-1.8 1 44 14 Food type Mutagen Amountl Temp2 Time3 Weight4 RefS Broiled, skin IQ 1.1-1.7 l 44 Broiled IQ 0.2-0.4 1 3 PhIP 1.7-23.0 1 1 1 Charbroiled 8-MeIQx 0.0 270 4 1 1 1 8-MeIQx 0.0 270 6 1 1 1 8-MeIQx 0.0 270 9 1 1 l 8-MeIQx 0.0 270 12 1 1 1 AorC 2.8 270 4 1 1 1 AaC 6.9 270 6 1 1 1 AaC 73.0 270 9 1 l l AaC 109.0 270 12 1 l 1 PhIP 2.0 270 4 1 l 1 PhIP 6.2 270 6 l 1 l PhIP 69.0 270 9 1 1 1 PhIP 73.0 270 12 1 1 l Cooked 4,8-DiMeIQx 0.2 150 9 1 17 MeIQ 1.0-1.6 150 9 l 17 MeIQx 0.6 150 9 1 17 IQ 0.6 150 9 1 17 PhIP 2. 7-3 .3 150 9 l 17 Fried 8-MeIQx 1.4 200 3 1 l 1 8-MeIQx 5.0 200 6 1 1 l 8-MeIQx 4.7 200 9 1 1 l 8-MeIQx 3.7 200 12 1 1 1 AorC 0.0 200 3 1 1 1 AaC 4.6 200 6 1 l 1 AaC 8.0 200 9 l l l AaC 9.0 200 12 1 l l PhIP 1.7 200 3 1 1 1 PhIP 23 .0 200 6 1 1 1 PhIP 14.0 200 9 1 1 1 PhIP 17.0 200 12 1 1 1 Smoked 4,8-DiMeIQx 0.0 0 17 MeIQ 0.0 O 17 MeIQx 1.2-1.4 O 17 IQ 0.3 O 17 PhIP 0.0 O 17 SARDINE Broiled 4-MeIQ 16.6 1 45 15 Food type Mutagen Amount1 Temp2 Time3 Wei ght4 Ref5 4-MeIQ 20.0 0 20 8-MeIQx 0.0 2 45 Glu-P-l 0.0 l 45 IQ 20.0 0 19 IQ 4.9 1 44 IQ 20.0 0 20 Phe-P-l 8.6 1 45 Trp-P-l 13.3 1 45 Trp-P-2 13. 1 1 45 UNSPECIFIED Fried Trp-P-2 0.0 200 1 3 1 Heated 4,8-DiMeIQx 5.4 1 23 MeIQx 5.2 1 23 Smoked, dried 4,8-DiMeIQx 0.08 1 21 MeIQx 0. 8 1 2 1 LAMB, mutton Broiled 4,8-DiMeIQx 0.67 1 35 8-MeIQx 1.01 1 35 AorC 2. 5 l 35 AMaC 0.19 l 35 PhIP 42.5 1 15 Trp-P-2 0. 15 1 35 NIEATBALLS Fried 4,8-DiMeIQx 0.2 1 l7 MeIQ 0.3 1 17 MeIQx 0.7 l 17 IQ 0.2 1 17 PhIP 0.6 l 17 MEAT EXTRACT Boiled 4,8-DiMeIQx 29-36 1 34 8-MeIQx 6.2-28.3 1 34 IQ 1.9-4. 8 l 34 PORK Charbroiled 4,8-DiMeIQx O. 1 O 33 8-MeIQx 0.4 0 33 PhIP 4.2 O 33 Fried Trp-P-l 0.0 3 00 6 0 8 BACON 16 Food type Mutagen Amountl Temp2 Time3 Weight4 Ref5 Fried MeIQx 0.9-18.0 12-16 1 12 4,8-DiMeIQx 0.0-l. 12-16 1 12 PhIP 0.0-53.0 12-16 1 12 AaC 0.0 12-16 1 12 Fried, moderate 4,8-DiMeIQx 1.7-5.1 150 2.5 l 17 Fried, well-done 4,8-DiMeIQx 1.0 150 5 1 17 Fried, moderate MeIQ 0.0 150 2.5 1 17 Fried, well-done MeIQ 1.4-2.0 150 5 1 17 Fried, moderate MeIQx 0.0-5.8 150 2.5 1 l7 Fried, well-done MeIQx 1.4-3.6 150 5 1 17 Fried, moderate IQ 2.3-5.3 150 2.5 1 17 Fried, well-done IQ 95-115 150 5 1 l7 Fried, moderate PhIP 0.2 150 2.5 1 17 Fried, well-done PhIP 1.0 150 5 1 17 BACON, fatty Fried 4,8-DiMeIQx 0.3 225 6 1 33 8-MeIQx 1.2 225 6 1 33 PhIP 2.7 225 6 1 33 BACON, lean Fried 4,8-DiMeIQx 0.2 225 6 1 33 8-MeIQ 0.9 225 6 1 33 PhIP 1.6 225 6 1 33 GROUND Fried 4,8-DiMeIQx 0.6 250 5 0 42 4,8-DiMeIQx 0.24 180 1 2 4,8-DiMeIQx 0.0 O 13 4-MeIQ 0.0 O 13 4-MeIQ 0.02 250 5 0 42 4-MeIQx 0. 1 250 5 0 42 4-MeIQx 1.4 250 5 0 42 4-MeIQx 0.4 180 1 2 4-MeIQx 0.0 0 13 IQ 0.04 250 5 0 42 IQ 0.01 180 1 2 IQ 0.0 0 l3 PhIP 4.5 250 5 0 42 PhIP 1.7 180 1 2 PhIP 0.0 0 13 GROUND, gravy 17 Food type Mutagen Amount1 Temp2 Time3 Weight4 Ref5 Fried 4,8-DiMeIQx 0.9 250 1 2 8-MeIQx 1.5 250 l 2 IQ 0.04 250 1 2 PhIP 10.0 250 l 2 SAUSAGE Fried 4,8-DiMeIQx 0.2 160 6 1 17 MeIQ 0.2 160 6 1 17 MeIQx 0.7 160 6 1 17 IQ 0.1 160 6 1 17 PhIP 0.1 160 6 1 17 Trp-P-l 0.0 300 6 0 8 AorC = 2-amino-9H-pyrido[2,3-b]indole; AMorC = 2-amino-3-methyl-9H-pyrido[ 2,3-b]indole; 4-OH-PhIP = 2-amino-1-methy1-6-(4-hydroxyphenyl)imidazo[4,5-b ]pyridine; 4-CH20H-8-Mex = 2-amino-4-hydroxy-methyl-3,8-dimethylimdazo[ 4,5-1] quinoxaline; Trp-P-l = 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole; Trp-P-2 = 3-amino-1-methyl-5H-pyrido[4,3-b]indole; Glu-P—l = 2-amino-6- methyldipyrido[1,2-a:3',2'-d]imidazole; Glu-P-2 = 2-aminodipyrido[1,2-a:3',2'-d] imidazole. 1 Amount of mutagen formed (ng/g) 2 Temperature of flying (°C) 3 Cooking time (minutes per side), except for sausage (total flying time) 4 Basis for heterocyclic amine concentration. O=cooked weight of food, 1=uncooked weight, 2=unspecified 5 References: 1- Barnes et al., 1983 2- Dragsted, 1992 3- Edmonds et al., 1986 4- Felton eta1., 1986a 5- F elton et al., 1992 6- Felton et al., 1984 7- F elton et al., 1986b 8- Felton etal., 1987 9- Gross et al., 1989 10- Gross, 1990 11- Gross and Gruter, 1992 12- Gross etal., 1993 13- Gry et al., 1986 18 14- Hargraves and Pariza, 1983 15- Hayatsu et al., 1991 16- Jackson et al., 1994 17- Johansson etal., 1994 18- Kasai et al., 1981b 19- Kasai et al., 1981a 20- Kasai et al., 1980a 21- Kato et al., 1986 22- Kikugawa et al., 1986 23- Kikugawa and Kato, 1987 24- Kim et al., 1994 25- Knize, personal communication 26- Knize etal., 1994 27- Kurosaka et al., 1992 28- Lee and Tsai, 1991 29- Lynch et al., 1992 31- Murray et al., 1987 32- Murray etal., 1988 33- Murray et al., 1993 34- Schuirrnann and Eichner, 1991 35- Sugimura et al., 1988 36- Taylor et al., 1985 37- Thiebaud et al., 1994 38- Turesky et al., 1983 39- Turesky et al., 1988 40- Turesky et al., 1989 41- Takahashi et al., 1985 42- Vahl et al., 1988 43- Wakabayashi et al., 1986 44- Yamaizumi et al., 1986 45- Yamaizumi et al., 1980 46- Zhang et al., 1988 FUROPYRIDINES: Because of the complexity of the interactions occurring in cooked meats, it is not surprising that new heterocyclic aromatic amines are still being identified. A methylimidazofuropyrine (MeIFP), with a molecular weight of 202, was isolated flom flied ground beef with added milk and creatinine (Felton et al., 1986a; 19 Becher et al., 1988). Evidence suggests that this mutagen is related to the food mutagen with a molecular weight 216, a amino-dimethylimidazoflrropyridine (Knize etal., 1990). A hydroxy derivative of PhIP, 2-amino- l-methyl-6-(4-hydroxyphenyl) imidazo[4,5-b]pyridine, was detected in broiled beef by Kurosaka et a1. (1992), at concentrations similar to those for PhIP. Recently, a compound similar to 4'-OH- PhIP, containing an exocyclic oxygen atom, was identified as 2-amino-4-hydroxy- methyl-3,8-dimethy1imidazo[4,5-f]quinoxaline (4-CH20H-8-MeIQx) in a beef extract (Kim et al., 1994). Mutagenicitxofheteroqyclic aromatic amines Heterocych aromatic amines are highly active in the Ames Salmonella system (Wakabayashi et al., 1992), in the DNA repair test of Williams, and in almost all tests measuring genotoxicity (Yoshimi et al., 1988). The heterocyclic aromatic amines have specific mutagenic activities toward Salmonella whimurium TA98 and TA100. The mutagenicity of heterocyclic aromatic amines and other typical carcinogens is listed in Table 2. Heterocyclic amines are metabolically activated by cytochrome P450. Metabolic activation of these compounds, in general, involves N-hydroxylation, followed by esterification to an acetyl or sulfate moiety (Okamoto et al., 1981; Saito etal., 1985; Paterson and Chipman, 1987; Snyderwine et al., 1987). Bioassays have revealed them to be toxic and carcinogenic for several specific target organs, including the liver, urinary bladder, pancreas, intestinal tract, colon and mammary gland (Ohgaki et al., 1984; 1986; 1987; 1991; Kato et al., 1988; 1989; Ito et al., 1991; Snyderwine et al., 1993). 20 Table 2. Mutagenicities of heterocyclic aromatic amines and typical carcinogens in Salmonella typhimurium (Sugimura and Sato, 1982; Sugimura et al., 1988). Revertants / ug Compound TA98 TAIOO IQ 433,000 7,000 MeIQ 661,000 30,000 IQx 75,000 1,500 MeIQx 145,000 14,000 4,8-DiMeIQx 183,000 8,000 7,8-DiMeIQx 163,000 9,900 Trp-P-l 39,000 1,700 Trp-P-2 104,200 1,800 Glu-P-l 49,000 3,200 Glu-P-2 1,900 1,200 Om-P-l 56,800 - AorC 300 20 MeAaC 200 120 Aflatoxin B1 6,000 28,000 AF-2 6,500 42,000 4-Nitroquinoline l-oxide 970 9,900 Benzo[a]pyrene 320 660 N-Methyl-N'-nitro-N-nitrosoguanidine 0_00 370 N-Nitrosodiethylamine 0.02 0. 15 N-Nitrosodirnethylamine 0.00 0.23 Mechanismfs) of heterocyclic aromatic amineformation REACI‘ANTS: The mechanism by which heterocyclic aromatic amines are formed during the cooking of meats has not been fully elucidated. It was demonstrated by 21 Yoshida and Okamoto (1980a,b,c) and Yoshida and Fukuhara (1982) that dry heating creatinine with either glucose, fatty acids or various amino acids produced high mutagenic activity, and suggested that these reactants were possible precursors of heterocyclic aromatic amines. Bjeldanes et al. (1982a) reported a positive mutagenic response of the Ames Salmonella assay for extracts of cooked beef, pork, ham, bacon, lamb, chicken, fish and eggs. However, other foods with high protein content, e.g., tofu, milk, cheese, shrimp and organ meats, showed very low or negligible mutagen formation (Bjeldanes et al., 1982b). Jagerstad et al. (1983a) proposed that three naturally occuning substances in meat, creatine/creatinine, flee amino acids and sugars, were the precursors of the imidazoquinoline- and imidazoquinoxaline-type mutagens. It was later demonstrated that chicken and beef contain the same heterocyclic aromatic amines in similar proportions as does flied ground fish, although in smaller amounts. This suggested that heterocyclic aromatic amines in cooked muscle foods all have similar precursors (Knize et al., 1988b; F elton and Knize, 1991). Supporting evidence for creatine/creatinine involvement in the formation of heterocyclic aromatic amines is the low or nonexisting mutagenic activity in foods high in protein but lacking in creatine, e.g., liver and kidney (Reutersward et al., 1987b; Felton and Knize, 1990). Jagerstad et al. (1983a) demonstrated a significant increase in the mutagenic activity of beef when creatine was spread over the surface before flying. Mutagenic activity was detected also in shrimp when treated with creatine before heating (Miller, 1985). Other investigators have demonstrated the importance of creatine/creatinine in the formation of mutagenic activity (Nes, 1986; Becher et al., 1988; Knize et al., 1988a; Overvik et al., 1989; Felton and Knize, 1991). In addition to creatine, flee amino acids and dipeptides play an important role in the formation of heterocyclic aromatic amines. Mutagenic activity was first 22 reported in protein-rich foods, but when beef extracts were subjected to enzymatic proteolysis before boiling, increased mutagenic activity was observed (Taylor et al., 1984; 1985). These results indicate that amino acids and not proteins participate in the formation of heterocyclic aromatic amines. When amino acids were dry heated with creatine at 200°C for 1 hr, mutagenic activity was detected (Overvik et al., 1989). Many studies have indicated that a single amino acid can produce several food mutagens in model reactions, and under similar conditions a specific heterocyclic aromatic amine can be produced flom several arrrino acids (Table 3). The involvement of sugars in heterocyclic aromatic amine formation was proposed by J agerstad et al. (1983a). However, their role in the formation of these compounds remains unclear. Model systems with creatine/creatinine, amino acids and various sugars, have shown that sugars have a substantial impact on the formation of mutagens (Muramatsu and Matsushirna, 1985; Skog and J agerstad, 1990; 1991; Manabe et al., 1992). Many of the mutagens have been identified in reaction systems without sugars (Knize et al., 1988a; Overvik et al., 1989). REACTION ROUTE: Several investigators have proposed the Maillard reaction to be important in the formation of heterocyclic aromatic amines, but without a specified reaction route (Spingam and Garvie, 1979; Shibamoto et al., 1981; Wei, 1981; Powrie et al., 1982). In the Maillard reaction, reducing sugars and amino groups flom either amino acids, peptides or proteins combine to form a glycosylamine, which undergoes an Amadori rearrangement to yield a 1-amino-2-keto sugar (Hodge, 1953). This sugar may then be broken down into 2- and 3- carbon flagrnents by two pathways (3-deoxyhexosone and methyl a-dicarbonyl routes), leading to the formation of a variety of compounds such as aldehydes, ketones and 23 Table 3. Heterocyclic aromatic amines produced in model systems from creatin(in)e and amino acids, with or without sugar (Skog, 1993). Compound IQ MeIQ IQx MeIQx 4.8- DiMeIQx 7.8- DiMeIQx Yieldl nd 2.7 nd nd 4.4 0.9 1.8 4.2 nd 6-7 nd nd nd 4.0 nd 10.0 8.8-17.9 7.0-10.0 9.0 nd 1.9-2.6 4.2 1.5 26.1 nd nd 36.0 30.0 nd 1.1 nd nd Amino acids Sugar pro sly phe SCI ala SCI’ 31y 1ys glu glu glu Heating conditions DEG-H20 DEG-H20 DEG-H20 DEG-H20 DEG-H20 DEG-H20 DEG-H20 DEG-H20 H20 H20 DEG-H20 DEG-H20 H20 Reference Yoshida et al., 1984 Grivas et al., 1986 Felton and Knize, 1990 Felton and Knize, 1990 Knize et al., 1988a Grivas et al., 1985 Knize et al., 1988a Skog and Johanssomunpublished,1993 Skog and Jagerstad, 1993 Jagerstad et al., 1984 Muramatsu and Matsushima, 1985 Muramatsu and Matsushima, 1985 Muramatsu and Matsushima, 1985 Negishi etal., 1985 Grivas et al., 1986 Overvik et al., 1989 Overvik et al., 1989 Overvik et al., 1989 Skog and Jagerstad, 1990 Skog and Jagerstad, 1991 Skog et al., 19923 Johansson et al., 1993 Skog and Jagerstad, 1993 Skog and Jagerstad, 1993 Negishi et al., 1984a; 1985 Grivas et al., 1985 Muramatsu and Matsushima, 1985 Muramatsu and Matsushima, 1985 Muramatsu and Matsushima, 1985 Skog and Jagerstad, 1990 Skog and Jagerstad, 1991 Skog et al., 1992a Skog and Jagerstad, 1993 Johansson et al., 1993 Negishi et al., 1984b Skog and Jagerstad, 1990 Johansson and Jagerstad, unpublished data, 1993 24 4,7,8- TriMeIQx 6.0 ala, thr glu DEG-H20 Skog et al., 1992a P11]? 3.6 phe glu DEG-H20 Shioya et al., 1987 735.0 phe - Dry Felton and Knize, 1990 560.0 phe glu Dry Felton and Knize, 1990 nd phe - Dry Overvik et al., 1989 nd leu - Dry Overvik et al., 1989 20.9 phe glu DEG-H20 Skog and Jagerstad, 1991 6.4 phe - DEG-H20 Skog and Jagerstad, 1991 < 0.058 phe glu DEG-H20 Manabe et al., 1992 lYield in nmol/mmol creatin(in)e Dry = dry hheating at 180°C or 200°C for 1 hr DEG-H20 = reflux boiling in diethylene glycol/water H20 = heated in water in closed metal tubes at 180°C for 10 or 30 min Amino acids: pro = proline; gly = glycine; phe = phenylalanine; ser = serine; ala = alanine; thr = threonine; 1ys = lysine; tyr = tyrosine; leu = leucine Sugars: fru = fl'uctose; glu = glucose; rib = ribose nd = not determined melanoidin pigments. Pyrazines and pyridines can be produced from the interaction of the a-dicarbonyls from the Maillard reaction with amino acids, the so-called Strecker degradation. The mechanism proposed by Hodge (1953) for the early stages of the Maillard reaction, identifying the Amadori rearrangement as a key step, was questioned by Namiki and Hayashi (1981). They reported the formation of the N,N'-disubstituted pyrazine cation by early carbon flagmentation prior to the Amadori product. They demonstrated that the radical products are formed by the condensation of two molecules of the two-carbon enarninol compounds which might be formed either directly from Schiff base products or indirectly through the reaction of glycolaldehyde with amino compounds (Fig.2). Thus, C2 and C3 flagments are produced prior to the Amadori rearrangement by a reverse-aldol reaction of the glycosylamine, forming glycolaldehyde alkylimines. These compounds could then be oxidized to form glyoxal monoalkylimines, which produced less flee radicals and reacted more slowly than the glycolaldehyde 25 H CH0 H-C=O H-c-o- H-C=O I I II I H C-OH base H-C-OH H-C-OH Hzc-OH I —> (I: —> —-> H-C-OH - (OH H-C=O Glycolardehyde k. 1. Lori- k. RNH, Sugar H-C=N-R I LRNH2 % Hzc-OH H-OH H (V l H-C=N-R H-C=N-R H-C-N-R ' I u H-C-OH H-C-OH H-C-OH l ——> I —> > H-C-OH H-Ci‘on H-C=O :1 iv iv \-OH‘ ha H-C-N-R Schifl’s base H-C-OH R ’ R \+ R I l N N N“ H .. e‘ f - , H ' 9' . Browning I ~- .’ <— O —* (melanoidin) R l l \ R 1 R FreeRadical Figure 2. A possible pathway of browning in the Maillard reaction through a novel free radical (Namiki and Hayashi, 1981). 26 system (Namiki and Hayashi, 1981; Namiki et al.,l983). Glycolaldehyde is very efl'ective in facilitating rapid and extensive radical formation compared to glyoxal. In a later study, Nyhammer (1986) proposed that heterocyclic aromatic amines are formed by an aldol-type condensation between an aldehyde and a pyridine or pyrazine molecule, followed by the cyclic addition of creatine to yield either an imidazoquinoline or an imidazoquinoxaline. Support for the theory of flee radical involvement in heterocyclic aromatic amine formation was provided by Milic et al.(l993). Glucose, aminobutyric acids, and 2,3-diamino-1,4-naphthohydroquinone were heated in a model system and pyridine flee radicals were detected by electron spin resonance. When only glucose and aminobutyric acids were heated, the formation of pyridine flee radicals occurred. To further establish the reaction pathway, 2,5-dimethylpyrazine was heated with creatinine and acetaldehyde, DiMeIQx formation was observed. When 2-methylpyridine was heated instead of 2,5-dimethylpyrazine, MeIQ was formed. The extracts were analyzed by high performance liquid chromatography, direct probe mass spectrometry and nuclear magnetic resonance. A possible reaction route for the formation of heterocyclic aromatic amines via the Maillard reaction was proposed by Jagerstad et al.(1983b). They postulated that creatine formed the arnino-imidazo ring of the heterocyclic aromatic amine molecule by cyclization and water elimination to creatinine, a reaction that takes place when the temperature is raised above 100°C. The irnidazo ring is common to the heterocyclic aromatic amines produced during normal cooking. The other two precursors, sugar and amino acids, were suggested to react following the Maillard reaction, and produce typical Maillard reaction products such as vinylpyrazines, vinylpyridines, and aldehydes (Fig.3). Thus, the quinoline or quinoxaline portion of the heterocyclic aromatic amine compound was assumed to arise flom these 27 / NH,+ CcHrzos RCH\ Hexose C02- Amino acid X Aldehyde N Me Pyridine 0!” on IQ compound "Ni/NE HOOC V N\ Me Creatine " H20 i NH N—"[ N\ Creatinine \3CH or Me Figure 3. Postulated reaction route for formation of IQ compounds R,X,andeaybeHorMe;Zr_naybeCHorN (J agerstad et al. 1983a). 28 latter compounds by aldol condensation. This hypothesis was verified using a model system where creatin(in)e, glycine or alanine, and glucose, dissolved in diethylene glycol containing 14% water, were boiled under reflux at 130°C for 2 hr. The mixture showed high mutagenic activity, whereas heating the reactants two by two produced only weak, if any, mutagenic activity. The addition of synthetic pyridines or pyrazines to the reaction mixture increased the mutagenic activity by about 50% (Jagerstad eta1., 1983a). According to the hypothesis of Jagerstad et al. (1983b), the aldol condensation first occurred between vinylpyrazines or vinylpyridines and aldehydes, followed by ring closure with creatinine. However, a series of tests involving the addition of selected alkylpyridines and/or pyrazines, including 2- methylpyridine or 2-viny1pyridine, to mixtures containing creatinine failed to yield any mutagen formation (Jones and Weisburger, unpublished data, 1988). Another way of formation was proposed by Nyhammar (1986) who assumed that the condensation first occurred between aldehydes and creatinine, which then condensed with a vinylpyrazine or vinylpyridine (Fig.4). Support for the latter reaction route was provided when Jones and Weisburger (1989) reported the formation of different IQ-like mutagenic products through reactions between creatinine and different aldehydes. Thus, aldehydes may be involved in the formation of heterocyclic aromatic amines through a series of chemical reactions involving specific reagents and creatinine. 29 NH2 NH: W N—‘r HOOC ——> N -H o A/N , \/ \Me 2 0 \Me Creatine Creatinine OHC \ ' H20 Y NH, O N + JY N 0 \ X N MC MC Pyridine CH\ or R Pyrazine Creatinine aldehyde , NH, N Ym \ CH, or Me x N R IQ compound Figure 4. Alternative route for formation of IQ compounds R,X,andeay beHoquZmaybeCHorN (Nyhammar, 1986). Factors influencing heterocyclic aromatic amine formation Egg Many studies have investigated the influence of the fat content of the product on mutagen formation (Spingam et al., 1981; Barnes and Weisburger, 1983; 1984; Bjcldanes et al., 1983; Knize et al., 1985; Chen, 1988), although it may be diflicult to distinguish between the physical and chemical effects of fat. These studies revealed that beef patties containing about 15% fat produced the highest mutagenic activity upon flying because heat penetration is increased. When the fat content was increased to above 15%, mutagenicity was slightly reduced. It was proposed that this could be due to diluting the precursors of the mutagens by the fat in the meat (Knize et al., 1985). The influence of different flying fats on the mutagenic activity of pork flied at 200°C for 10.5 minutes was also studied (Nilsson et al., 1986; Overvik et al., 1987). Results showed that there were no differences in mutagenic activity produced by difl‘erent flying fats, but there was higher mutagenic activity than when flying took place without fat. This was probably due to the higher temperature generated at the meat surface when a flying fat was used. The effects of edible oils and fatty acids on the formation of mutagenic heterocyclic amines was also studied by Johansson et a1. (1993). When corn oil or olive oil was added to model systems containing creatinine, glycine and glucose dissolved in water and heated to 180°C for 30 minutes, the yield of MeIQx was almost doubled relative to the yield without fat. However, this increase was not observed when a fatty acid or glycerol was added to the model system. This may be due to the participation of the lipids in the Maillard reaction. Lipids are known to enhance the production of pyrazines and other products in the Maillard reaction (Watanabe and Sato, 1971a,b; Buttery et al., 1977; Parihar et al., 1981; Kawamura, 1983; Amoldi etal., 1987; 1990). It was also proposed that aldehydes 30 31 are formed more rapidly if fat is present in the reaction mixture (Amoldi et al., 1987). Lipids can also produce carbonyl compounds upon autoxidation, which can react with amino compounds (Kawarnura, 1983). Furthermore, lipid hydroperoxides can decompose to form aldehydes, which can react with amino acids and give Schiff base adducts (Gardner, 1979). However, the addition of oxidized linoleic acid or linolenic acid, to the model system (creatinine, glycine, and glucose) produced about the same amount of MeIQx as a mixture with no fat added when heated for 30 min at 180°C (Johansson et al., 1993). The degree of oxidation of the fat had small effects on the formation of MeIQx, no consistent trend was demonstrated between the yield of MeIQx and the degree of oxidation (Johansson and Jagerstad, 1993). When oxidized oleic acid was added to the model system, a decrease in the amount of MeIQx was observed (J ohansson et al., 1993). In a later study, Johansson and Jagerstad (1994) observed that the type of flying fat had a significant influence on the heterocyclic aromatic amine content in the pan residue. For example, the pan residue flom beefburgers flied in butter contained higher amounts of heterocyclic aromatic amines than the pan residue flom beefburgers flied in oil. This might be due to the inhibiting effects of different antioxidants present in the flying oils. C marine/creatinine Investigators have reported the absence, or only very low levels, of mutagenic activity in liver, kidney, cheese, tofu, beans and shrimp when cooked at normal cooking temperatures (F elton and Knize, 1990). However, in other protein- rich foods such as beef, pork, lamb, chicken and fish, the mutagenic activity was considerably higher (Bjcldanes et al., 1982a, b). The absence of mutagenic activity in non-muscle foods and shrimp (invertebrates) can be explained by the lack of creatine. Creatine, in the form of creatine phosphate, is an energy reserve only in 32 vertebrates (Sulser, 1978), and is transformed into flee creatine within 24 hr after slaughter (F abiansson and Reutersward, 1985). It has been established that mutagenic activity in beef is increased if a solution of creatine is spread over the surface before flying (J agerstad et al., 1983a). To facilitate the identification of heterocyclic aromatic amines, several investigators have added creatine to different meat products before flying to enhance their formation (Nes, 1986; Becher et al., 1988; Overvik et al., 1989, F elton and Knize, 1991). These investigations suggest that the content of creatine in meat products is rate-limiting for the formation of heterocyclic aromatic amines. It has been shown that creatine is converted to creatinine during cooking (Lempert, 1959) and that the proportion of creatinine in the crusts and pan residues increases with increasing temperature (Reutersward et al., 1987a). When beef joints were roasted in an oven at 115°C-245°C, the concentration of creatine and creatinine on a dry matter basis was considerably higher in the pan residues than in the crust, indicating a loss flom the meat to the pan (Reutersward et al., 1987a). Results of a later study indicated that there is a transportation of creatin(in)e to the crust as well as a leakage to the pan (Skog et al., 1992b). As a result of this leakage, the mutagenicity of the pan residues can be as high as that of the crust of the meat (Felton et al., 1981; Overvik et al., 1987; Berg et al., 1988, 1990; Knize et al., 1988a; J ohansson and J agerstad, 1994). It was suggested by Jagerstad et al.(1983a) that creatine and/or creatinine are essential precursors in the formation of heterocyclic aromatic amines, thus confirming the observations of previous investigators (Reutersward et al., 1987a, b; Overvik et al., 1989) that the mutagenic activity of cooked foods is related to its creatine/creatinine content. Vikse and Joner (1993) reported that the correlation between creatine/creatinine and mutagenicity was observed only when a creatinase treatment was applied. The average decrease in creatine concentration was 65%, 33 and this resulted in a decrease of approximately 73% in the mutagenic response flom the meat extract. However, the differences in the normal creatine and creatinine contents of meat (flom 16 different animal species) did not explain the varying mutagenic activity in the extracts of flied meat; thus, the relationship between them is not a simple one. The results of a study by Jackson et a1. (1994) also indicate the lack of a direct relationship between the creatine/creatinine content and mutagenic activity in beef flavors. However, flavor extracts containing heterocyclic aromatic amines also had high mutagenic activity and high levels of creatine and creatinine. Further studies are needed to establish the relationships between creatine and creatinine, mutagenic activity and heterocyclic aromatic amine formation in meat products. Amino acids and dipeptides Mutagenic activity was first reported in protein-rich foods (Commoner et al., 1978a). However, when proteins instead of amino acids were used in model systems, no mutagenic activity was detected (J agerstad et al., 1983a). When beef extracts were subjected to enzymatic proteolysis before boiling, increased mutagenic activity was observed (Taylor et al., 1984; 1985). These results indicate the participation of amino acids and not proteins in the mutagen-forming reactions. When creatine and glucose were refluxed with amino acids at 128°C for 2 hr, most of the latter compounds produced mutagenic activity (Jagerstad et al., 1983b). When several amino acids were dry heated with creatine at 200°C for 1 hr, mutagenic activity was also detected (Overvik et al., 1989). As indicated previously, a specific heterocyclic aromatic amine can be produced flom several difl'erent arrrino acids, while a single amino acid can produce several of these mutagenic compounds. 34 The importance of flee amino acids in heterocyclic aromatic amine formation has also been shown in meat systems. It was demonstrated by Overvik et a1. (1989) that the addition of 15 amino acids to pork before flying enhanced mutagenic activity by 1.5 to 43 times. Ashoor et a1. (1980) reported that only proline, when added to ground beef, increased the mutagenic activity. On the other hand, no significant mutagenicity was detected after flying kidney or liver, although these two organs contain about twice the amount of flee amino acids as does muscle meat (Reutersward eta1., 1987a). However, as pointed out previously, the creatine content of organ meats is very small, which would explain the low mutagenicity in flied organ meats. When the dipeptide, camosine, was refluxed with creatinine and glucose, at 128°C for 2 hr, mutagenic activity similar to that produced flom flee amino acids was developed (Reutersward et al., 1987b; Overvik et al., 1989). These observations demonstrate the importance of not only amino acids but also dipeptides in the production of mutagenic activity and perhaps in the formation of heterocych aromatic amines. 544% The role of sugars in the formation of heterocyclic aromatic amines is not clear. Different model systems with creatin(in)e, amino acids and sugars (glucose, fluctose, ribose, galactose, arabinose, erythrose) have demonstrated the importance of sugars in the formation of heterocyclic aromatic amines (Muramatsu and Matsushima, 1985; Skog and Jagerstad, 1990; 1991; Manabe et al., 1992). Although not obligatory when added to model systems containing amino acids and creatinine, the quantities of heterocyclic aromatic amines are increased and different products are formed when sugars are involved. For example, when phenylalanine was heated with creatine at 180°C for 10 minutes, PhIP was detected as a single mutagen, but when glucose was added, the amount of PhIP 35 increased three-fold and at least two other mutagens (MeIQx and 4,8-DiMeIQx) were formed (Skog and Jagerstad, 1991). When sugars other than glucose were heated with phenylalanine and creatinine, PhIP formation was again observed. Erythrose was the most active sugar in this reaction (Manabe et al., 1992). These studies suggest that there might be two pathways for the formation of heterocych aromatic amines: (l) with sugars, via the interaction of Maillard reaction products with creatinine (J agerstad et al., 1983a; Nyhammar, 1986; Pearson et al., 1992); and (2) without sugars, via the reaction of creatinine with breakdown products of amino acids (Felton et al., 1986a). The role of sugars in the formation of heterocyclic aromatic amines is complex. Model system studies show that mutagens can be formed by dry heating creatinine with different amino acids without sugars. For example, dry heating of phenylalanine with creatinine produced PhIP (Overvik et al., 1989; F elton and Knize, 1990; Skog and Jagerstad, 1991); dry-heating either serine, alanine or tyrozine with creatinine produced MeIQx (Overvik et al., 1989). However, the incorporation of the carbon label into MeIQx when [U-14C]-labeled glucose, glycine and creatinine were heated together, suggests that glucose is a precursor in the formation of heterocyclic aromatic amines (Skog et al., 1992b). Cooking time and temperature The effect of temperature on mutagen formation in cooked ground beef was first described by Commoner et al.(l978b). A number of investigators have subsequently shown that mutagen production increases with the temperature of cooking (Spingam and Weisburger, 1979; Hatch et al., 1982; Bjcldanes et al., 1983; Chen, 1988; Knize et al., 1994). Cooking methods that employ higher 36 heating temperatures generally induce greater heterocyclic aromatic amine formation than low temperature methods (Murray et al., 1993; Knize et al., 1994). Several researchers have observed that there is a progressive increase in the mutagenic activity of cooked products with increasing cooking time (Commoner et al., 1978a; Bjcldanes et al., 1983; Miller and Buchanan, 1983; Overvik et al., 1984; Knize et al., 1985; Knize et al., 1994). However, there is a lag period of 2 to 4 minutes during the flying of ground beef patties when no mutagenicity is observed. This is the time required for the crust surface to reach a temperature above 100°C. Results published by Knize et a1. (1994) show the increase in heterocyclic aromatic amine formation in ground beef patties with increasing time and/or temperature of flying. There was no mutagen formation at 150°C after 2 or 4 minutes of flying, and PhIP and DiMeIQx were not detected in patties flied at 190°C for 2 minutes. At each temperature/time combination of flying, PhIP was present in the highest concentration, demonstrating again that it is the most abundant heterocych aromatic amine in flied ground beef. Although the concentrations of PhIP in the beef patties flied at 230°C for 10 minutes, were approximately 10 times greater than those of MeIQx, PhIP took a longer time to form. A longer time and a higher temperature are necessary to produce the initial 20% of the PhIP formed in flied beef patties compared to MeIQx. To study the rate of PhIP formation, a model system study was conducted by Knize et a1. (1994), and compared to the production of the same compound in the flied ground beef. Results of the heating of 0.05M phenylalanine and 0.05M creatinine in 80% diethylene glycol, at 150°C or 200°C for 10 minutes, suggest that the rate of formation of PhIP in meat and in the model systems is similar. PhIP 37 formation in the simple model system was analogous to that in the more complex ground beef system. Inhibition of mutagen formation Sugars and other carbohydrates Sugars are naturally occurring substances in meat systems and have a substantial impact on the formation of heterocyclic aromatic amines. Studies by Taylor et a1. (1986) revealed that when glucose was added to beef-stock supernatant at a concentration four times greater than that of creatine, mutagenic activity was decreased. A more comprehensive study on the effect of sugar on heterocych aromatic amine formation was carried out by Skog and J agerstad (1990). They demonstrated that excess amounts of sugar in model systems inhibited the formation of heterocyclic aromatic amines. When sugars were present in equimolar or greater amounts than the creatin(in)e concentration, the formation of mutagens was almost completely inhibited. When the glucose concentration was about half the molar concentration of creatine, the mutagenicity was the highest. The mechanism behind the inhibitory effect is not known, but there was a decrease in the recovery of creatine and creatinine as the glucose concentration increased. This phenomenon was observed only when amino acids were present in the reaction mixture, indicating a reaction between Maillard reaction products such as 5-hydroxymethyl-2-furfural (HMF) and creatinine. The creatinine would be less available to form heterocyclic aromatic amines (Skog and J agerstad, 1990). The inhibitory effect of sugars added in excess has also been studied in meat systems. When different carbohydrates were added to beef patties before flying, the mutagenic activity of the crust was dependent on the type of the carbohydrate added. Inhibition (flom 40% to 70%) of the mutagenic activity was demonstrated with glucose and pure lactose or lactose flom milk powder added at 38 concentrations up to 4%. The greatest inhibitory effect was achieved by golden breadcrumbs when added in combination with glucose or lactose. Furthermore, the mutagenic activity decreased when a mixture of potato starch and glucose was added to the beef patties (Skog et al., 1992b). After flying, a major portion (90%) of the initial concentrations of creatine and creatinine was still present in the crust of the beef patties. This demonstrated again that the role of creatine/creatinine as a rate limiting factor in the formation of heterocyclic aromatic amines is not fully understood. Sgt) protein concentrate The prevention of mutagen formation in flied beef patties by the addition of soy protein concentrates was reported by Wang et a1. (1982). The mutagenicity of the control beef patties (i.e., without added soy protein concentrate) was over 25,000 revertants per 50 g beef. When soy protein concentrate was added at a level of 24%, total inhibition was achieved. These results clearly showed that the reduction of mutagenicity by soy protein concentrates occurs during the cooking process, as separately cooked patties when mixed with soy protein concentrates, did not have reduced mutagenicity. Overall mutagenicity of the flied ground beef was determined by the Ames assay using Salmonella typhimurium TA98 and the rat liver microsomal flaction for metabolic activation (Ames et al., 1975). Most of the reduction in mutagenicity was attributed to volumetric effects such as through the reduction of interactions among the beef components, and by the reduction of the amount of beef that came into contact with the heating surface. However, some consideration was given to chlorogenic acid which is a naturally occurring polyphenolic antioxidant in soy protein concentrate (Smith and Circle, 1978; Pratt and Birac, 1979; Rappaport et al., 1979) in soy protein concentrates. Wang et a1. (1982) demonstrated that this compound, when added directly to 39 ground beef patties, succesfully inhibited the development of mutagenicity in the flied beef. Defatted glandless cottonseed flour Defatted glandless cottonseed flour added at a level of 5% (w/w) to ground beef before flying also reduced the mutagencity of the cooked meat (Rhee et al., 1987). The magnitude of the reduction in mutagenicity tended to be much greater than the meat dilution effect by the glandless cottonseed flour. Glandless cottonseed ingredients are effective naturally-occurring antioxidants and have been shown to retard lipid oxidation in various soy and meat products (Rhee et al., 1981; Ziprin et al., 1981). Flavones (mainly quercetin derivatives) are the major flavonoids present in cottonseed. However, the investigators did not clearly define whether the reduction in mutagenicity was a result of volumetric effects or through the antioxidant properties of glandless cottonseed components. Synthetic antioxidants The effects of synthetic phenolic antioxidants on mutagen formation in cooked ground beef was first described by Wang et a1. (1982). They showed that butylated hydroxyanisole (BHA) successfully reduced the mutagenic activity of flied ground beef patties. BHA also reduced the mutagenicity when it was directly added to the testing mixture of beef extract, S-9 mix and the bacterial culture in the Ames bioassay. The inhibitory effects of BHA, propyl gallate and tertiary butylhydroquinone were more intensively studied by Chen et al. (1992). These antioxidants reduced the overall concentrations of IQ-like compounds (IQ, MeIQx and DiMeIQx) in flied ground beef by approximately 80-90% when added at 0.1% of the fat. At this low level of addition, the volumetric effect was insignificant. More recently, Faulkner (1994) reported a reduction of mutagen formation in flied ground beef on adding BHA to the patties before flying. When BHA 40 (0.1% based on the fat content) was added to ground beef, a significant (p<0.005) reduction in the mutagenicity, flom 7000 revertants/ 100g raw meat revertants to 2800 revertants/100g raw meat, was achieved. The PhIP concentrations were reduced flom 2.4 ng/g to 1.2 ng/g. Vitamin E Much research is currently focused on the use of naturally occurring ingredients as antioxidants because of growmg concerns about the safety of synthetic antioxidants and a general consumer perception that natural is better (Gray and Crackel, 1992). Vitamin E is an effective monophenolic antioxidant in lipids and lipid-containing foods because it efl‘ectively scavenges peroxy radicals (Niki, 1987). The inhibitory effects of vitamin E on mutagen formation in flied ground beef were studied by Chen et a1. (1992). When vitamin E ( 1% based on the fat . content) was added to ground beef patties, the concentration of IQ-like compounds (IQ, MeIQx and DiMeIQx) was reduced by 50%. The reported concentrations of the mutagenic compounds were 1000-fold higher than those published by other researchers (Felton et al., l986a,b; Sugimura et al., 1988; Thiebaud et al., 1994). Therefore, the quantitative data of Chen et a1. (1992) are questionsble as no confirmatory studies were carried out. However, these data did establish for the first time the inhibitory effects of vitamin E on the formation of heterocyclic aromatic amines. In a recent study, Faulkner (1994) confirmed the inhibition of PhIP formation by vitamin E (1% based on fat content). PhIP concentrations in ground beef patties were reduced by 80% by adding vitamin E (1% based on fat content) to the ground beef before flying. The overall mutagenicity was reduced by 70%, flom 7000 revertants/ 100 g raw meat to 1900 revertants/100 g raw meat. This was the first reported study of the simultaneous use of both the Salmonella 41 Whimurium overall mutagenicity test and an analytical procedure to quantitate the extent of inhibition of specific heterocyclic aromatic amines. Tea phenolics Tea polyphenolic compounds, particularly epigallocatechin gallate, epicatechin gallate and epigallocatechin, have been established as potent antioxidants (Sorata et al., 1984; Chen et al., 1990; Sichel et al., 1991; Ho et al., 1992; Terao et al., 1994). Namiki and Osawa (1986) evaluated the antioxidant activities of different polyphenols including cr-tocopherol, propyl gallate, catechin, epicatechin, gallocatechin, epigallocatechin, epicatechin gallate, and epigallocatechin gallate, in rabbit blood cells subjected to peroxidation. Peroxide generation was suppressed to the same extent by a-tocopherol, propyl gallate, epicatechin gallate and epigallocatechin gallate. The effect of black tea (BT) and green tea (GT) and the polyphenols, theaflavine gallate (TFG) and epigallocatechin gallate (EGCG), on the formation of heterocyclic aromatic amines was studied by Weisburger et a1. (1994). They demonstrated substantial reduction in heated model systems containing creatinine, glycine and glucose. Black tea had a substantial effect in lowering the formation of both MeIQx and PhIP (flom 9350 to 7340 revertants/plate and flom 6530 to 2070 revertants/plate, respectively). The effect of green tea in reducing the formation of mutagenic compounds was effective only for PhIP (flom 6530 to 2180 revertants/plate). When EGCG was added to the model system, there was an 80% reduction in the mutagenic activity for both MeIQx and PhIP. Considerable inhibition of MeIQx and PhIP formation (83% and 74%) was achieved when TFG was applied. In a separate experiment, in which the mutagenicity was lowered to 38% by EGCG, HPLC analysis showed a 15% decrease in PhIP formation. The relationship between antioxidant activity and antimutagenicity of green tea, pouchong tea, oolong tea and black tea was investigated by Yen and Chen 42 (1995). The antimutagenic effect of tea extracts on IQ toward Salmonella typhimurium TA98 and TA100 was correlated with their reducing power and scavenging efl‘ect on the hydroxyl radical. All tea extracts exhibited antioxidant activity and reducing power. The antioxidant effect of tea extracts was well correlated to their antimutagenicity in some cases, but varied with the mutagen and antioxidative properties. flame F lavone is a naturally occuring flavonoid that is present in edible and medicinal plants (Brown, 1980). The inhibitory effects of flavone on the formation of heterocyclic aromatic amines was studied by Lee et a1. (1992) in a glycine, creatine and glucose model system. Results showed a decrease in the mutagenic activity of MeIQx and 7,8-DiMeIQx by 31.1% and 27.8%, respectively. The total mutagenic activity of the heated glycine/creatine/ glucose mixture was decreased by the addition of flavone in a dose-related response relationship. Quantitation of mutagens The extraction/detection of heterocyclic aromatic amines has been a challenging undertaking because of the small concentrations (low ng/g levels) of these compounds, the diversity of the different mutagens formed under same reaction conditions, and the complexity of the food samples to be analyzed. In the early 1980's, two procedures were used to extract organic material flom cooked and uncooked meat samples. First, an extraction procedure described by Commoner et al. (1978a) used dilute acid and ammonium sulfate to precipitate proteins, followed by pH adjustment and solvent extraction of the organic constituents. The second procedure used acetone to extract the organic constituents directly flom the cooked protein foods. These organic extracts were then separated into basic, neutral, and acidic flactions (Felton et al., 1981). 43 A method which utilized Amberlite XAD-2 resin to isolate mutagenic activity flom an initial aqueous acid extract of flied beef was developed by Bjcldanes et al. (1982c). However, thin layer chromatographic profiles of the mutagenic extracts isolated by this method indicated poor recoveries for the difl‘erent mutagens. An improved method for the isolation and characterization of new mutagens flom flied ground beef was described by F elton et a1. (1984). Mutagens were separated by aqueous/acid extraction flom the beef, XAD adsorption, acid/neutral/base-liquid/liquid extraction, preparative reverse phase HPLC, normal phase HPLC, and analytical reverse phase HPLC. The identification was carried out by low and high resolution mass spectrometry, ultraviolet absorption spectroscopy and nitrite sensitivity assays. The mutagenicity of each flaction was monitored by the Salmonella assay described by Ames et a1. (1975). Simple methods for quantifying mutagenic heterocyclic aromatic amines in food products were deve10ped by Gross (1990). The solid-phase extraction procedure included a copper phthalocyanine (CPC) tandem extraction, and a propylsulfonyl silica gel (PRS) tandem extraction, both followed by further clean- up. The method was improved by Gross and Gruter (1992) to allow the purification of the entire range of heterocyclic aromatic amines. This procedure involved tandem extraction with diatomaceous earth and an ion exchange resin (PRS), followed by clean-up with a C18 column, and subsequent separation and identification of the heterocyclic aromatic amines on HPLC using a photodiode array UV detection system. The advantage of this solid-phase extraction procedure is a simpler and more rapid sample preparation prior to chromatographic analysis. Peak confirmation is a crucial problem when working with such low levels of heterocyclic aromatic amines since co-elution with other compounds can occur. Thus, HPLC retention times alone do not provide unequivocal identification of 44 these compounds. Some researchers have combined HPLC directly with mass spectrometry (MS) to identify heterocyclic aromatic amines. Yamaizumi et al. (1986) identified MeIQ and IQ in broiled salmon by using HPLC - thermospray mass spectrometry. Data show that MeIQx, IQ, and DiMeIQx were identified in beef extracts and flied beef by using HPLC-MS analysis at concentrations ranging flom 0.3 to 52 ng/g (Turesky et al., 1988). The most sensitive approach for heterocyclic aromatic amine analysis is that devised by Murray et al. (1988). Using cooked meat samples spiked with heavy-isotope-labeled standards, samples were dissolved in dilute hydrochloric acid, washed with dichloromethane and then extracted into ethyl acetate. Dried extracts were derivatized with 3,5-bis-trifluoromethylbenzyl bromide and analyzed by gas chromatography - electron-capture negative ion chemical ionization mass spectrometry. Lean minced beef patties were found to contain 1.0 to 2.4 ng/g MeIQx and 0.5 to 1.2 ng/g DiMeIQx. MA T ERIALS AND METHODS Materials Freshly ground beef (85% lean) was obtained flom local supermarkets and used within one hour of purchase, or stored at -20°C until required for flying. Samples of meat were taken randomly for fat and moisture determinations. Vitamin E (dl-cr-tocopherol) was purchased flom Sigma Chemical Company (St Louis, MO), while oleoresin rosemary was donated by Kalsec Inc. (Kalamazoo, MI). Extrelut-20 columns were obtained flom EM Separations (Gibbstown, NJ). Bond-Elut PRS (500 mg) and C18 (100 mg) cartridges were purchased flom Varian, Inc. (Harbor City, CA). All solvents were HPLC or glass-distilled reagent grade. The heterocyclic aromatic amine standards (IQ, MeIQ, MeIQx, 4,8- DiMeIQx and PhiP) were obtained flom Toronto Research Chemicals (Toronto, Canada). The heterocyclic aromatic amine standard (F EMA - Flavour and Extracts Manufacturers' Association) and the internal standard, caffeine, were kind gifts flom Dr. Mark Knize, Lawrence Liverrnore National Laboratory, University of California, CA. The FEMA standard contained IQ, MeIQ, MeIQx, 4,8-DiMeIQx and PhIP, each at 0.5 ng/ul. Methods Moisture and fat determination Moisture and fat contents were determined by the AOAC official methods (AOAC, 1992). 45 Influence of temperature and time of frying on heterocyclic aromatic amine formation in ground beef patties Frying of ground beef patties: A teflon-coated electric flying pan without a lid was used to fly the ground beef patties. The temperature control of the flying pan was set at 175°C or 200°C and the surface temperature was measured by a thermocouple (Pacific Transducer Co., Los Angeles, CA) during the flying. Before flying, the flying pan was preheated to the selected temperature. Ground beef purchased flom three different sources (sold as 85% lean beef) contained different levels of fat (14.8%, 22.7% and 17.5%). Patties weighing 100 g were formed in a petri dish (9 cm dia. x 1.5 cm thickness), and flied for 6 or 10 minutes on each side at each of the two temperatures. Two ground beef patties were flied in the flying pan, and blended together in a Waring laboratory blender (Dinamics Co. of America, New Hartford, CT) at 70 rpm for 2 minutes. Two samples were taken flom the blended patties and subsequently analyzed for heterocyclic aromatic amine content. The beef patties flied at 225°C were prepared flom 15% fat ground beef obtained flom a local supermarket (only one source). Two ground beef patties, weighing 100 g each and formed in a petri dish, were flied for 6 and 10 minutes per side, then blended and combined. Heterocyclic aromatic amine concentrations at 225°C were based on three replications. During the study, four sub-samples were taken and analyzed for each replicated experiment. Statistical analysis: Statistical analysis of the concentrations of heterocyclic aromatic amines in meat flied at 175°C and 200°C for 6 or 10 minutes was based on two replications and three different sources of meat. Two ground beef patties were flied for each experimental replication under similar conditions. Extraction, analysis and quantitation were carried out in duplicate for each replicated experiment, with four 46 47 sub-samples analyzed. The results were analyzed by a statistical computer program (MSTAT-C) developed at Michigan State University (Department of Crop and Soil Sciences) by a three-way analysis of variance (AN OVA), and f-values were calculated (manually) for specific comparisons between mean values for the different sources of meat. To compare the heterocyclic aromatic amine concentrations at all three temperatures on an equal fat basis (~15%), meat flom only one source was compared. Thus, concentrations flom one replication (at 175°C and 200°C) were compared to results flom three experimental replications (at 225°C). Manually calculated t-values (Gill, 1988) were compared to determine significant differences between mean values of heterocyclic aromatic amine concentrations. Inhibition of heterocyclic aromatic amine formation in fried ground beef patties by direct addition of ph en olic antioxidants Flying of ground beef patties: Vitamin E (1% or 10% based on fat content) and oleoresin rosemary (1% or 10%) were dissolved in 1 ml corn oil and added directly as separate treatments to the ground beef patties two hours prior to flying. Ground beef patties mixed with 1 ml corn oil served as the control. Patties weighing 100 g were formed in a petri . dish (9 cm dia. x 1.5 cm thickness), and flied for 10 minutes per side at 225°C in a teflon-coated electric flying pan. Two patties were flied for each replication, and three experimental replicates were analyzed for each treatment. Statistical analysis: Statistical analysis of the heterocyclic aromatic amine concentrations in the flied ground beef patties was based on three replicates for all five treatments. All treatments were flom the same source of meat and flied under the same conditions. For each replicate, four sub-samples were analyzed (two for concentration and two 48 for recovery). Therefore, the mean value contains two data points. The results were analyzed by a statistical computer program (MSTAT-C) by a one-way analysis of variance (ANOVA). The Bonferoni's t test was used to determine the significance of the treatment compared to the control, the effect of the different treatments, and the interaction between them (Steel and Torrie, 1980). Inhibition of heterocyclic aromatic amine formation in fried ground beef patties through surface application of vitamin E Frying of ground beef patties: Vitamin E (1% based on the fat content) was dissolved in 1 ml corn oil and spread on the surface of ground beef patties 30 minutes before flying. As a control, 1 m1 corn oil was applied to the surface of the patties. The patties were flied for 10 minutes at 225°C in a teflon-coated flying pan. Seven experimental replications were analyzed for both treatments for heterocyclic aromatic amine concentration. Statistical analysis: All analyses were performed with seven experimental replications for both control and antioxidant treatments. Only one source of meat was used to form the ground beef patties. Two flied patties were combined for each treatment and four sub-samples analyzed. The results were analyzed by the Student's t test to determine the significance of the vitamin E treatment compared to the control. To analyze statistically the heterocyclic aromatic amine concentrations in flied patties prepared flom ground beef (flesh and after two months of freezer storage), t-values were calculated (manually) and compared. These manual calculations were necessary as there were only three replications for the flesh (unfrozen) meat compared to seven experimental replications for meat which had been flozen for two months. Analysis of heterocyclic aromatic amines in fried ground beef patties (Figure 5) A flied ground beef sample (25 g) was weighed into a glass beaker and homogenized thoroughly with 75 g 1N NaOH in a Ultra Turrax (Tekmar Co., Cincinnati, OH) high speed mixer for 1 to 2 minutes. Four 16 g aliquots of the homogenized mixture (equal to 4 g meat) were removed and two were spiked with a mixture of the heterocyclic aromatic amines in 50 ul methanol (i.e., 250 ng of each heterocyclic aromatic amine). Each sample was mixed thoroughly with one package of Extrelut diatomaceous earth to give a flee-flowing homogenous, lump- flee powder. Extrelut cartridge assemblL: A small paper filter was placed in the bottom tip of the column and the column body was inserted. The bottom of the column was then covered with a small amount of diatomaceous earth and the Extrelut mix added. A large paper filter was subsequently placed over the sample. Extrelut - PRS tandem (propylsulfonic acid silica): The PRS cartridges were filled with dichloromethaneztoluene (95:5 v/v) and a slight positive pressure was applied until the solvent passed through the cartridge (Figure 6). The car1ridge was filled again with the dichloromethaneztoluene solvent system, and a needle was assembled at the end of the cartridge for flow reduction. The Extrelut cartridges were filled with the same solvent, when it passed to the bottom of the cartridge, the PRS and Extrelut cartridges were coupled. A 40 ml aliquot of the dichloromethaneztoluene solvent mixture was allowed to flow through the tandem columns. The extraction was stopped by separating the Extrelut column flom the PRS column. The dichloromethaneztoluene was discarded and needles were removed from the PRS cartridges. 49 Removal of interferences from the PRS cartridge: PRS cartridges were transferred to a Visiprep vacuum (Supelco), and the cartridges were dried for 10-15 minutes under maximum vacuum. The PRS cartridges were then connected to a peristaltic pump and successively rinsed at about 1.5-2.0 ml/min with 6 ml of 0.1 M hydrochloric acid, 15 ml of methanol: 0.1M hydrochloric acid (4:6) , and 2 ml of water. T ran_sfer of heterocyclic aromatic aminesfrom PRS to C18: The C18 cartridges were slowly (by gravity) rinsed with 2 ml of MeOH , followed by 5 ml of water, and kept wet until used. The C18 and PRS cartridges were then connected. To transfer the heterocyclic aromatic amines flom the PRS column to the C18 column, a volume of 20 ml of 0.5N ammonium acetate buffer (pH 8.0) was pumped through each tandem at a flow of approximately 1.5-2.0 ml/min. The PRS cartridges were then discarded; and the C18 columns were rinsed with 1 ml of water. Elution of heterocyclic aromatic aminesfrom the C18 cartridges: The C18 cartridges were dried completely by applying vacuum (Visiprep) for 30 minutes. The heterocyclic aromatic amines were eluted slowly flom the C18 cartridges with 1.0 ml of methanolzammonia (9: 1) by applying gentle overpressure through a plastic syringe directly into the microvials. The solvent was evaporated in a 40°C water bath using a stream of nitrogen, and the samples were refligerated until required for HPLC analyses. 50 51 Antioxidant treatments mixed in 1 ml corn oil, added to 100 g of ground beef; patties formed in a petri dish (9 cm dia. x 1.5 cm thickness) U Fry two patties for each replicate, and grind the patties together in a blender U Add 75 g 1N NaOH to 25 g fried beef sample and homogenize U Divide into 4x16g samples, two of which are spiked with 5011.1 of standard mixture (250 ng/each compound) U Mix 16g sample with Extrelut-20, fill column with mixture, attach to PRS column containing 2 ml dichloromethane:toluene (95:5) U Rinse columns with dichloromethane:toluene, collect 40 ml of solvent mixture, detach Extrelut column, and dry PRS under vacuum (10-15 minutes) U Add 6 ml of 0.1M HCI to PRS followed by 15 ml MeOH:0.1M HCI (4:6), and then 2 ml water U Condition C18 column with 1 ml of MeOH followed by 10 ml of water U Attach PRS column to C18, rinse with 20 ml of ammonium acetate (pH 8.0), discard PRS , and rinse C18 withl ml of water U Dry C18 column by applying high vacuum for 25-30 minutes U Elute heterocyclic aromatic amines with 1.0 ml of MeOHzcc. ammonia (9:1) by applying gentle overpressure through a plastic syringe U Evaporate solvent to dryness under nitrogen in a 40°C water bath and refrigerate until injection Figure 5. Extraction of heterocychc aromatic amines in flied ground beef patties following the method of Gross and Gruter (1992). 52 7 l- 49 food sample homogenized in Extrelut-20 ‘ 1N NaOH, mixed with diatomaceous earth E 011,012 : Toluene (95:5) ' Propylsulfonic acid silica (PRS) p P 0.1N HCI:CH,0H 0.5N ammonium acetate (pH 8.0) C18 1ml CH,OH:NH, —> —> HPLC (9:1 ) HAAs Figure 6:. Solid phase extraction to detect heterocyclic aromatic amines from ° food systems. - Quantitative determinations of heterocyclic aromatic amines To measure extraction efficiency and to quantitate the heterocyclic aromatic amines in the fried patties, the standard addition method of Gross and Gruter (1992) was used. Quadruplicate determinations (four sub-samples) were carried out with two unspiked samples and two samples spiked with 50 ul of the heterocyclic aromatic amine standard mixture. The extraction efficiencies were thus calculated for each analyte as the slope of the linear regression line: added analyte concentration versus measured analyte concentration. Quantitative measures of heterocyclic aromatic amines were corrected for incomplete analyte recovery. High performance liquid chromatographic (HPLC) analysis of heterocyclic aromatic amines Calibration: Five aliquots (5, 10, 15, 20, and 25 ul) of the two stande heterocyclic aromatic amine mixtures (containing 0.5 ng/ul and 5 ng/ul of each compound) and the caffeine internal standard (5 ng/ul caffeine) were injected prior to the analysis of the sample extracts. The response linearity of the HPLC was checked by performing a linear regression calculation: ul standard solution injected versus peak area. A positive slope with a correlation coefficient of 0.99 was obtained in each case. Analysis: Beef pattie extracts were redissolved in 50ul methanol (containing 5 ng/ ul caffeine) and an aliquot of 20ul was injected through a Rheodyne 7125 (50 pl loop) injector (Rheodyne Inc., Cotati, CA) into a Waters 510 HPLC, equipped with a Waters 991M photodiode array UV detector (Millipore, Milford, MA). The heterocyclic aromatic amines were detected as follows: 254 nm for IQ and MeIQ, 53 54 262 nm for MeIQx and 4,8-DiMeIQx, and 316 nm for PhIP. In order to improve peak shape, triethylamine (1.4 ml/liter) was added to HPLC grade water. The mobile phase was vacuum filtered through a 0.45 pm membrane and adjusted to pH 3.2 with dilute phosphoric acid (above pH 3.2, Glu-P-l and MeIQ co-elute). Acetonitrile was used as the second mobile phase. A reversed phase silica HPLC column (TSK-Gel ODS-8OTM column; 4.6 mm ID x 25 cm; Tosoh Haas; Montgomeryville, PA) protected by a Supelguard LC-8-DB (Supelco) precolumn was used to separate the heterocyclic aromatic amines. The particle size of the column packing material was 5 um in diameter. The mobile phase flow rate was set at l ml/minute. The acetonitrile concentration in the mobile phase was increased from 8% to 17% during the first 10 minutes and then to 25% in another 10 minutes. Percent acetonitrile was then increased to a concentration of 55% in 10 minutes. By this time, all heterocyclic aromatic amines were eluted from the column. Over a 5-minute period, the percent of acetonitrile was increased to its maximum concentration of 80%, and this was necessary to clean the column of other unwanted compounds. At the end of the 45 minute elution period, the acetonitrile was decreased to its original concentration of 8% and the column was allowed to equilibrate for 10-15 minutes before the next injection. All separations were carried out at ambient temperature. RES UL T S AND DISCUSSION Optimization of the “fiction/quantitation procedures for heterocyclic aromatic amines The extraction and quantitation of hetrocyclic aromatic amines in fried ground beef patties was achieved using the Standard Addition Quantitation procedure developed by Gross and Gruter (1992). This procedure is a challenging one which is reflected in the fact that reported heterocyclic aromatic amine concentrations in beef are based on recoveries of heterocyclic aromatic amine standards that range from as low as 5% to as high as 85% (Jackson et al., 1994; Johansson and Jagerstad, 1994; Knize et al., 1994; Thiebaud et al., 1994). Thus, the first priority of this study was to optimize the extraction procedure to obtain consistent recoveries of the heterocyclic aromatic amines and to reduce the standard deviations of the data obtained. To accomplish this, several short studies were performed to identify critical steps in the extraction process which impacted recoveries, and to devise ways to overcome these challenges. Results of these preliminary investigations indicated that it was essential to optimize the size of the meat sample for a number of reasons: (a) to obtain a powder-like blend upon combining the fried ground beef with the Extrelut refill diatomaceous earth, and to charge the Extrelut-20® extraction column with an appropriately-sized meat sample. (b) to be able to extract sufficient heterocyclic aromatic amines to enable their detection by the HPLC procedure employed; and (c) to avoid overloading the PRS and/or C18 columns in subsequent phases of the extraction procedure. These studies revealed that for maximum recoveries of the heterocyclic aromatic amines in fried beef patties, the optimum weight of the meat sample should be 55 56 approximately 4 g. When larger amounts of ground beef were used, recoveries of PhIP were greatly reduced, generally below 10%. It was also observed that the recovery of each heterocyclic aromatic amine was improved by 5-10% when the C18 column was completely free of solvent before using. This was achieved by applying vacuum to the column for approximately 25-30 minutes, after rinsing the column with solvent. A review of reported heterocyclic aromatic amine concentrations in meat products using the Gross and Gruter (1992) procedure reveals that smaller recoveries are generally obtained for PhIP than for any of the other heterocyclic aromatic amines. Thus, more attention was directed toward improving its recovery in this study, including an investigation of various solvent systems used to elute the heterocyclic aromatic amines fi'om the Extrelut-20® extraction colmnns. It was determined that the dichloromethane /toluene (95:5) solvent system (Knize, personal communication) increased the recovery of PhIP by 5% relative to that achieved by dichloromethane alone. When these changes were introduced into the extraction procedure, the recoveries of the heterocyclic aromatic amines matched the upper range of recoveries published in the literature (J ohansson and Jagerstad, 1994; Knize et al., 1994; Thiebaud et al., 1994). Table 4 shows the range of recoveries obtained for each compound. Recovery experiments involved the addition of 250 ng each of IQ, MeIQ, MeIQx, 4,8-DiMeIQx and PhIP to two of four samples of fiied ground beef before packing the Extrelut-20 cartridge. In this way, peak confumation was more accurate as half of the samples had peaks with retention times identical to those of the added standard heterocyclic aromatic amines. This complex three-step solid phase extraction and clean-up procedure is necessary because the UV absorption maxima for imidazoquinolines and irnidazoquinoxalines are located at wavelengths around 260 run where many other 57 aromatic compounds absorb light (Gross and Gruter, 1992). This means that a sample must be as completely free of interfering compounds as possible to allow the quantification of heterocyclic aromatic amines at the very low concentrations (ng/g) that these are present in fried beef patties. To accomplish this, one has to be very precise when using this procedure. Peak confirmation is a crucial problem when working with such concentrations since co-elution with other co-extracted compounds can occur. To confirm the identity of the heterocyclic aromatic amines, a derivatization procedure developed in our laboratory by Faulkner (1994) is routinely used. This involves derivatization of the heterocyclic aromatic amines to their mono—pentafluoropropionic derivatives, followed by gas chromatographic - mass spectrometric confirmation. In addition, because all of the heterocyclic aromatic amines have characteristic ultraviolet (UV) spectra and high extinction coefficients, a photodiode array UV detection system essentially prevented false peak identification. The use of fluorescence detection also assisted in the confirmation process because of the strong signal generated for PhIP. Table 4. Percent recoveries of heterocyclic aromatic amines from ground beef patties using solid phase extraction1 Compound IQ MeIQ MeIQx 4,8- PhIP DiMeIQx Percent recovery 58.2-84.8 51.7-78.2 66.7-88.5 72.1-92.4 32.1-61.9 1Range of recoveries are based on twelve sample analyses Influence of temperature and time of frying on heterocyclic aromatic amine formation in ground beef patties In order to optimize the formation of heterocyclic aromatic amines in the flied patties prepared from three sources of ground beef, various time/temperature combinations were investigated. This initially involved flying the ground beef patties at two different temperatures (175°C and 200°C-measured by a thermocouple on the surface of the flying pan during the flying of the patties) for 6 and 10 min/side. The purchased ground beef was advertised as containing 15 % fat. This was desirable for the study as it has been reported previously that maximum heterocyclic aromatic amine formation in flied patties occurs at this level of fat (Spingam et al., 1981; Holtz et al., 1985; Knize et al., 1985; Nilsson et al., 1986; Overvik et al., 1987). However, analyses of the three sources of ground beef revealed fat contents of 14.8%, 22.7%, and 17.5%. Results show that the formation of the mutagenic compounds was dependent on the time and temperature of cooking (Table 5). PhIP concentrations in flied beef patties increased significantly with time (p<0.1) and temperature (p<0.05) of cooking. PhIP is the most abundant heterocyclic aromatic amine found in cooked meat with reported concentrations ranging from O to 67 .5 ng/g flied beef (Thiebaud et al., 1994). Although the concentrations of PhIP in the flied patties were approximately 6 times greater than those of MeIQx under these frying conditions, PhIP took a longer time to form (e.g., at 175°C, 6 min: 0.3 ng/g fresh meat MeIQx and 0.4 ng/g flesh meat PhIP; at 10 min: 0.4 ng/g fresh meat MeIQx and 2.6 ng/g flesh meat PhIP). The same phenomenon was reported by Knize et a1. (1994), who indicated that a longer time and a higher temperature are necessary to produce the initial 20% of the PhIP formed in flied beef patties compared to MeIQx. 58 59 Table 5. Heterocyclic aromatic amine contentrations in ground beef 2patties (ng/g cooked meat) flied using different time / temperature combinationsla ,3 Time / IQ MeIQ MeIQx 4,3- PhIP Temperature DiMeIQx 175°C 6 min 03350.38 0.11:0.1a 030.08 0310.221 04:0.16 10 min 0.6i0.23 04:03b 0.41:0. 1f 0.4:0.1b 2.6i0.9f 200°C 6 min 0.810.213 03:50.1c 0.735023 05:01c l.9i0.6g 10 min 1.3:10b 0.7:0.4d 1.44.09h 1.33:1.2d 8.2i5.4h 1 Values are expressed on a raw ground beef basis. Values are based on measured cooking losses for individual patties. 2 Each value represents the mean of two samples per source of ground beef in which duplicate analyses have been averaged 1' standard deviation, i.e., six samples in all. 3 Mean values in columns with different superscripts are significantly different at p<0.1. After completion of this study, a flying pan was purchased that permitted the use of a higher flying temperature (>200°C). Thus, time/temperature combinations of 6 and 10 minutes at 225°C were investigated. Three experimental replications were performed and results are presented in Table 6. Heterocyclic aromatic amine data presented for 175°C and 200°C are those values pertaining to ground beef patties flom the ground beef source containing 14.8% fat. Because of the reported influence of fat concentration on heterocyclic aromatic amine formation, it was desirable to compare the heterocyclic aromatic amine concentrations at all temperatures on an equal fat basis, recognizing that other intrinsic variables (e.g., creatine/creatinine contents and sugar) may also play a major role in heterocyclic aromatic amine formation in meat. Statistical analyses 60 indicated that the concentrations of MeIQx and PhIP were significantly (p<0.05) influenced by the temperature of cooking, while the concentrations of IQ, MeIQ, and PhIP were only significant at p<0.l level. The difficulty in establishing the significance of the data was attributed to the high standard deviations in the results. These data represent heterocyclic aromatic amine formation in one source of meat containing 15% fat. Heterocyclic aromatic amine concentrations at 175°C and 200°C are based on one replication, while those at 225°C are the mean value of three replications. However, each replication involves two sub-samples which provide two data sets per replication Table 6. Heterocyclic aromatic amine concentrations in ground beef patties using different time / temperature combinations of flying1:2:3 Time / IQ MeIQ MeIQx 4,8- PhIP Temperature DiMeIQx 6 min 175°C 0,410.1a 0.1100c 0.310.2a 0,510.0c 0,510.2a 200°C 1.010.1a,b 0310.06.d 0.910.3a.b 0,510.10,d 2.310.3a,b 225°C 1,611.2b 1.211.9d 1.9107b 1.711.0d 7.213.9b 10 min 175°C 0710.56 0.21026 0.51008 0.51006 3.611.3g 200°C 2.410.562f 1210.76,f 2.310.1g,h 2510.06,f 14.011.623.11 225°C 2.811.9f 1.911.9f 3,110.911 2.612.4f 16.917.0h 1Mean values for similar flying times in columns with different superscripts are significantly different at p<0. 1, and for MeIQx and PhIP at p<0.05. 2Ground beef patties contain 15% fat. 3Values are expressed on a raw ground beef basis. Values are based on measured cooking losses for individual patties. 11 °\:S°x ‘-"\\ as \\ Concentrations of HAAs (nglg) Figure 7. Formation of heterocyclic aromatic amines in ground beef patties fried for 6 minutes at three temperatures. 62 Concentrations of HAAs (nglg) Figure 8. Formation of heterocyclic aromatic amines in ground beef patties fried for 10 minutes at three temperatures. 63 Even by analyzing four sub-samples from each replicate, it is hard to overcome the problem of variation of data. Because of this, there are few published reports that include an exhaustive statistical treatment of data. The concentrations of the five heterocyclic aromatic amines detected in the flied ground beef patties fall within the range of concentrations reported by other investigators (Barnes et al., 1983; Turesky et al., 1988; Lynch et al., 1992; Knize et al., 1994; Thiebaud et al., 1994). Reported concentrations for IQ, MeIQ, MeIQx and DiMeIQx are generally smaller then those concentrations reported for PhIP (Felton et al., 1984; Felton et al., l986a; Sugimura et al., 1988; Gross, 1990; Knize et al., 1994). Many studies have addressed the effects of the method of cooking on heterocyclic aromatic amine formation in meats. Mutagenic compounds are formed rapidly when meat is cooked by frying, and more slowly by broiling (Spingam and Weisburger, 1979). Recent studies also show that microwaving meat before flying decreases substantially the formation of heterocyclic aromatic amines (Felton et al., 1994). Besides cooking methods, the most important physical variables affecting heterocyclic aromatic amine formation are cooking time and temperature. Several researchers have observed that there is a progressive increase in the mutagenic activity of cooked meat products with increasing cooking time (Commoner et al., 1978a; Bjcldanes et al., 1983; Miller and Buchanan, 1983; Overvik et al., 1984; Knize et al., 1985; Reutersward et al., 1987a, b; Nielsen et al., 1988). Often there is a lag period of 2 to 4 minutes when no mutagenicity is observed. This is related to the time required for the crust surface to reach a temperature above 100°C. However, there is no pan temperature at which mutagenic activity in meat is not developed (Skog, 1993). Inhibition of heterocyclic aromatic amine formation in fried ground beef patties by direct addition of phenolic gntioxidants When vitamin E (dl-a-toc0pherol) and oleoresin rosemary were added directly to the ground beef before flying, reduction in heterocyclic aromatic amine formation was observed (Table 7, Figure 9). Control patties in all cases contained the greatest concentrations of heterocyclic aromatic amines. Statistical analyses indicated that concentrations of IQ, MeIQx, and PhIP were significantly different (p<0.05) between the different treatments, while the concentration of DiMeIQx was only significant at p