EFFECT OF DIET ON SERUM ALBUMIN AND HEMOGLOBIN ADDUCTS OF 2-AMINO-1-METHYL-6-PHENYLIMIDAZO[4,5-b]PYRIDINE (PhIP) IN HUMANS Cinzia MAGAGNOTTI 1 , Federica ORSI 1 , Renzo BAGNATI 1 , Nicola CELLI 2 , Domenico ROTILIO 2 , Roberto FANELLI 1 and Luisa AIROLDI 1 * 1 Department of Environmental Health Sciences, Istituto di Ricerche Farmacologiche Mario Negri, Milan, Italy 2 Center for Environmental Health “G. Paone”, Consorzio Mario Negri Sud, Santa Maria Imbaro (Chieti), Italy 2-Amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) is the most abundant heterocyclic amine formed in meat and fish during cooking and can be used as a model compound for this class of chemicals possibly involved in human carcinogen- esis. Knowing the exposure to heterocyclic amines is impor- tant for establishing their role in human diseases. Serum albumin (SA) and globin (Gb) adducts were first tested as biomarkers of exposure to PhIP in male Fischer 344 rats given oral doses of 0.1, 0.5, 1 and 10 mg/kg. Blood samples were collected 24 hr after treatment and PhIP released from SA and Gb after acidic hydrolysis was analyzed by gas chro- matography-mass spectrometry or liquid chromatography- tandem mass spectrometry. PhIP-SA and Gb adducts in- creased linearly with the dose. Studies on 35 volunteers with different dietary habits exhibited that diet was a major de- terminant in the formation of both adducts. PhIP-SA adducts were significantly higher in meat consumers than in vegetar- ians (6.7  1.6 and 0.7  0.3 fmol/mg SA; respectively, mean  SE; p  0.04, Mann-Whitney U test). The Gb adduct pattern was quantitatively lower but paralleled SA (3  0.8 in meat consumers and 0.3  0.1 in vegetarians). PhIP-SA ad- ducts were no different in smokers and in non-smokers. The results show for the first time that PhIP-blood protein ad- ducts are present in humans not given the synthetic com- pound. Both biomarkers appear to be suitable for assessing dietary exposure and internal PhIP dose and may be prom- ising tools for studying the role of heterocyclic amines in the etiology of colon cancer and other diseases. Int. J. Cancer 88: 1– 6, 2000. © 2000 Wiley-Liss, Inc. Epidemiological evidence places colon cancer among those closely linked with dietary factors (Doll and Peto, 1981; Potter, 1996). Studies have focused on the role of dietary components and on the identification of chemical carcinogens in foods, thought to be involved in colon cancer etiology (Potter, 1996). Dietary factors showing correlations with this tumor site include high fat and low fiber intake, alcohol and meat consumption (Potter, 1996). Among chemical carcinogens identified in foods, heterocyclic amines have attracted particular interest. These compounds are formed by pyrolysis in the presence of creatine or creatinine, sugar and amino acids during the cooking of meat and fish, and their amount depends on cooking temperature and how well the meat is done (Skog et al., 1998). Heterocyclic amines are potent mutagens in the Ames test and carcinogenic in mice and rats, the colon being one of the target organs (Sugimura, 1997; Wakabayashi et al., 1992). Animal studies have led to the hypothesis that heterocyclic amines might be the etiological agents for human colon cancer and this has been tested (Byrne et al., 1998; Gooderham et al., 1997; Layton et al., 1995; Vineis and McMichael, 1996; Augustsson et al., 1999, 1997; Probst-Hensch et al., 1997). However uncertainty persists, mainly due to the difficulty of assessing exposure; hence the need for specific markers. Individual heterocyclic amines in- take is deduced from information on food consumption and infor- mation on heterocyclic amines content in food items (Augustsson et al., 1997; Byrne et al., 1998; Layton et al., 1995). Among biological markers, urinary heterocyclic amines metabolites appear to be the most widely used (Kidd et al., 1999; Reistad et al., 1997; Stillwell et al., 1997). However they give an indication of recent exposure only, whereas endogenous macromolecule adducts would give information about exposure occurred during the mac- romolecule’s whole lifespan. Biomonitoring of DNA adducts is a significant indicator of exposure to genotoxic agents, but it can be only used to analyze damage in accessible tissues, such as white blood cells. Carcinogen adducts with proteins have been widely measured as a surrogate for DNA damage (Skipper and Tannen- baum, 1990). Blood protein adducts have been used as exposure markers for chemical carcinogens such as 4-aminobiphenyl or benzo(a)pyrene (Pastorelli et al., 1998, 1996; Skipper and Tan- nenbaum, 1990). A number of studies suggest using protein ad- ducts as markers of exposure for different heterocyclic amines (Alexander et al., 1997). Tests with radioactive heterocyclic amines in humans have proved they bind to blood protein and DNA (Dingley et al., 1999, 1998). Recently, human hair analysis for 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) content was proposed as a long-term marker of exposure (Reistad et al., 1999). The aim of our study was to identify a biomarker of heterocyclic amines exposure in humans by using PhIP as a model compound for this class of chemicals. PhIP is one of the most ubiquitous heterocyclic amine present in cooked foods, being detected at the ppb (g/kg) range and it has also been detected in wine and in mainstream cigarette smoke (Augustsson et al., 1997; Byrne et al., 1998; Manabe et al., 1993, 1991; Richling et al., 1997; Skog et al., 1998). When administered with the diet to mice and rats, PhIP increased the tumor incidence at different sites including lympho- mas, mammary gland and small and large intestine adenocarcino- mas (Sugimura, 1997; Wakabayashi et al., 1992; WHO, 1993). Metabolic activation to reactive species that bind to DNA is essential for heterocyclic amines mutagenicity and carcinogenicity (Schut and Snyderwine, 1999). The first step in the metabolism of heterocyclic amines to reactive species is N-oxidation to the N- hydroxy derivative. This occurs primarily in the liver in animals and humans and is catalyzed mainly by the inducible enzyme CYP1A2, though other P450 enzymes might be involved (Boobis et al., 1994; Crofts et al., 1997; Hammons et al., 1997; Lynch et al., 1992; Turesky et al., 1998). N-Oxidation of heterocyclic amines competes with the activity of N-acetyltransferases and sulfotransferases. Whereas N-acetylation and N-sulfation of aro- matic amines in general are detoxifying processes, O-acetylation and O-sulfation of N-hydroxy-heterocyclic amines are regarded as activation steps. N-hydroxy-heterocyclic amines themselves can react with DNA, but the corresponding O-acetyl and O-sulfate derivatives are more reactive electrophiles (Davis et al., 1993; Lin et al., 1992; Schut and Snyderwine, 1999). These reactive species presumably bind to blood proteins (Alexander et al., 1997; Lynch et al., 1991; Lynch et al., 1993; Watkins et al., 1991a,b). We investigated whether PhIP-blood protein adducts could be used as markers of human exposure and internal dose. First, PhIP Grant sponsor: Associazione Italiana per la Ricerca sul Cancro (AIRC). *Correspondence to: Luisa Airoldi, Department of Environmental Health Sciences, Istituto di Ricerche Farmacologiche Mario Negri, Via Eritrea 62, 20157 Milan, Italy. Fax: 39 02 3546277. E-mail: airoldi@irfmn.mnegri.it Received 22 November 1999; Revised 1 March, 13 April 2000; Ac- cepted 25 April 2000 Int. J. Cancer: 88, 1– 6 (2000) © 2000 Wiley-Liss, Inc. Publication of the International Union Against Cancer