Question

This is a question for Toxicology, it asks about the absorbtion, biotransformation and excretion of a chemical in the human body,

Chemical B is found in smoke from burned plant materials and fossil fuels. It is a fairly large polycyclic aromatic hydrocarbon and has a very high Kow. Although high levels are found commonly in certain tissues, no adverse effect has been determined in humans or laboratory animals. For the following questions about the metabolism of this chemical, assume that exposure to it is oral, the pH of the stomach is 1.0, the pH of the blood is 7.4, the pH of the small intestine is 8.5, and the pH of the urine is 6.8.

1. What factor will determine the rate of absorption and distribution of Chemical B? ___________

2. Why can humans carry high levels of Chemical B in their tissues without exhibiting any problems? ___________________________________________________________________

3. Is this chemical likely to be excreted via the urinary tract? ______________________________

4. Will changing the pH of the urine or blood facilitate excretion of Chemical B? _______________

Answer 1

The occurrence of PAH in foods is influenced by the same physicochemical characteristics that determine their absorption and distribution in man. These are their relative solubility in water and organic solvents, which determine their capacity for transport and distribution between different environmental compartments and their uptake and accumulation by living organisms. The transportation of PAH in the atmosphere is influenced by their volatility. The chemical reactivity of PAH influences adsorption to organic material or degradation in the environment. All these factors determine the persistence and capacity of PAH to bioaccumulate in the food chain. PAH are lipophilic and generally have a very poor aqueous solubility. PAH accumulate in lipid tissue of plants and animals. PAH will not tend to accumulate in plant tissues with a high water content and limited transfer from the soil to root vegetables will occur. The rate of transfer varies widely and is also influenced by soil characteristics, the plant and the presence of co-pollutants. PAH adsorb strongly to the organic fraction of soils and do not penetrate deeply into most soils, therefore limiting both leaching to groundwater and availability for uptake by plants. Some PAH are semi volatile but most of them tend to adsorb on organic particulate matter. Heavier PAH preferentially associate with particulate matter so atmospheric fall out is a principal route of contamination . PAH with 5 or more aromatic rings are found predominantly on particulates, (usually on small (< 2.5 µm) particles such as fly ash and soot). PAH with 2 or 3 rings are almost entirely in the vapor phase, those with 4 rings being in an intermediate position.Consequently, vegetables with large leafs, grazing cattle and poultry which may ingest particulate matter from soil are susceptible to contamination by PAH adsorbed to particles. The waxy surface of vegetables and fruits can concentrate low molecular mass PAH mainly through surface adsorption. PAH concentrations are generally greater on plant surface (peel, outer leaves) than on internal tissue. Careful washing may remove up to 50% of the total PAH. Particle bound PAH are easily washed off the surface whereas those in the waxy layer are less efficiently removed, washing may alter the apparent high to low molecular mass PAH profile.

Degradation PAH are chemically stable and very poorly degraded by hydrolysis but are susceptible to oxidation and photo-degradation in light. PAH half lives in air range from a few hours to days and estimated PAH half lives in soils vary from several months to several years. Abiotic degradation may remove 2-20% of two and three ring PAH in contaminated soils . PAH with 4 or more aromatic rings persist in the environment but they are often strongly adsorbed to organic matter. Following degradation, oxidized reaction products may be formed which tend to react with biological components. Reaction with nitrogen oxides and nitric acid in the atmosphere can form nitro derivatives, which could contaminate foods. Thus although parent compounds cannot always be detected in PAH contaminated foods, degradation products or derivatives, some of which have significant toxicity, may be present. The halflives in soil and air depends on various parameters (e.g. type of adsorption onto particles, molecular weight) and range from hours to days for air and months to years for soil.

Biodegradation The most significant information on biodegradation is summarized below

Contamination of food with PAH during processing and smoking Processing procedures, such as smoking and drying, and cooking of food is commonly thought to be the major source of contamination by PAH. Depending on a number of parameters: time, fuel used, distance from the heat source and drainage of fat, type (grilling, frying, roasting), cooking results in the production in the food of a number of compounds including PAH. Although not precisely known, it is likely that there are several mechanisms of formation of PAH such as melted fat that undergoes pyrolysis when dripping onto the heat and pyrolysis of the meat due to the high temperature . A comparison of PAH levels in duck breast steaks, undergoing various processing and cooking treatments for 0.5 hour to 1.5 hours, showed that charcoal grilled samples without skin contained the highest amount of total PAH (320 µg/kg), followed by charcoal grilling with skin (300 µg/kg), smoking (210 µg/kg), roasting (130 µg/kg), steaming (8.6 µg/kg) and liquid smoke flavouring (0.3 µg/kg). For PAH that are classified as carcinogenic (IARC class 1 or 2 A and B), the trend was the same with the exception that smoked samples contained the highest amount (35 µg/kg). In addition, the highest amounts of total and carcinogenic PAH were observed after smoking of duck breast samples for 3 hours (53 µg/kg) . Contamination of water may lead to intake of PAH through drinking water and cooked foods. The levels are usually below 1 ng/L in drinking water but can be higher when asphalt or coal tar coating of storage tanks and water distribution pipes are used.

Comparison with the contribution to the PAH intake from other sources The contribution by food ingestion to the total intake of PAH was compared with the intake from drinking water and inhalation of air. The available PAH concentrations in drinking water and in urban atmosphere, as measured in the 1990s in European countries, are reported . As regards drinking water, the available data are few. This is likely due to the very low expected levels, commonly below the analytical limit of detection. The measurements are not easily comparable because of differences in the analytical procedure (particularly regarding the filtration of the sample before extraction) and the coating of the distribution pipes. So, the overall results were approximated to the nearest order of magnitude. The atmospheric measurements reported were limited to the investigations performed during the whole year due to the well-known seasonal variability of PAH concentrations. As regards the measurement of 3- and 4-ring PAH, the table includes only the investigations where vapour phase was also collected, besides particulate phase. Air concentrations generally show a good consistency among different investigations and the typical level of concentration is given as an order of magnitude. Based on concentration data and on the estimated dietary intake the estimated contribution to the total personal PAH intake from the three routes of exposure. Benzo[a]pyrene is the compound showing the highest number of measurements, by any route of exposure, and the results are also quite consistent relatively to the other PAH. This makes it possible to compare the individual mean benzo[a]pyrene intakes . Due to the overall uncertainties associated to the original concentration data and mainly to the mean intakes adopted, the results of this comparison are to be considered as approximate estimates, acceptable as indicative of the actual exposure. As regards the mean benzo[a]pyrene intake by food ingestion, the average value of the value is used in the calculation. The total mean daily intake of benzo[a]pyrene results to be about 0.2 µg/person. The dietary intake accounts for about 90% of it. Drinking-water contributes by less than 1%, thus resulting to be a relatively insignificant route of exposure. The remaining is attributable to air inhalation. The latter contribution was calculated on the basis of a ‘typical’ benzo[a]pyrene concentration in urban air (1 ng/m3 ), commonly measured at road level close to vehicular emissions: the actual personal exposure, however, is expected to be lower, at least for most of the A50 population. Hence, the mean intake by food ingestion is likely to contribute by more than 90%.These benzo[a]pyrene intake figures are consistent with those calculated in an US study: 12 and 140 ng/day as the mean intakes by, respectively, inhalation (based on personal air samples) and ingestion of food (based on prepared food samples) . An evaluation of the relative intakes of other PAH is made difficult by the wide ranges of the estimated ingestion by both food and water, as well as by the low number of available data. Taken overall, the data indicate that food contributes by far most of the intake also for the other PAH, including the lower-molecular ones.

Absorption, Distribution, Metabolism and Excretion

Absorption The two major determinants of gastrointestinal absorption are aqueous solubility and lipophilicity, since absorption requires compounds to go into solution in the lumen of the intestine, pass through the cell walls of intestinal cells and be removed to the circulation. PAH are lipophilic compounds with low aqueous solubility. Those considered in this opinion have log Kow values ranging from 3.4 to 7.3 and aqueous solubilities from 0.17 to 31740 µg/L at 25 ˚C. Although aqueous solubility generally decreases as the log Kow value increases, there is considerable variability amongst compounds with similar log Kow values reflecting the influence of molecular structure on aqueous solubility. There are three main routes of absorption in humans, lung and respiratory tract following inhalation of aerosols or particulates containing PAH, dermal following skin contact and gastrointestinal tract following ingestion in water or food. For the purposes of this evaluation only the last route will be considered in detail. Rees and colleagues in 1972 showed rapid absorption of benzo[a]pyrene following intragastric administration in rats with highest levels seen in the thoracic lymph nodes after 3-4 hours. Based on results in whole animals and intestinal sacs, these workers suggested that absorption of benzo[a]pyrene involved two phases, uptake by the mucosa followed by diffusion through the intestinal wall . Laurent and colleagues (2001) described studies on the absorption of two PAH (benzo[a]pyrene (log Kow 6.5, aqueous solubility 3.8 µg/L) and phenanthrene (log Kow 4.6, aqueous solubility 1290 µg/L)) following oral administration to pigs in a lipophilic milieu. Two castrated Large White pigs were catheterised in the portal vein and brachiocephalic artery. Fourteen days post-surgery they were fed 1 litre of milk containing 50 µCi of [7,10- 14C]-benzo[a]pyrene or 15 µCi of [9-14C]-phenanthrene and 10 ml arterial and portal blood samples were collected before administration and then hourly for 6 hours at 9 and 24 hours. A53 Radioactivity was detectable within 1 hour and peaked at 5-6 hours and reached background by 24 hours. The peak radioactivity was higher for phenanthrene despite the 3- fold lower dose. Values in the portal vein were slightly higher than those in the brachiocephalic artery. The peak time corresponds with that observed for milk fat and is much longer than glucose (45 minutes) or protein (30 minutes). No areas under the curves (AUCs) or other pharmacokinetic parameters are reported. Rahman and colleagues (1986) showed the presence of bile increased the intestinal absorption of PAH in Sprague-Dawley rats, absorption of benzo[a]pyrene (log Kow 6.50, solubility 3.8 µg/L) (and 7,12-dimethylbenz[a]anthracene) being affected more than that of anthracene (log Kow 4.5, solubility 78 µg/L) or pyrene (log Kow 5.18, solubility 135 µg/L) . Kawamura and co-workers (1988) demonstrated that the composition of the diet influenced the absorption of co-administered 14C-benzo[a]pyrene in Wistar rats. The radio-labelled benzo[a]pyrene was orally administered in a solution, emulsion or suspension of 200 mg of food or food component and blood samples collected over the first 6 hours and at 24 hours post-administration. The foods and components studied were triolein, soya bean oil, cellulose, bread, rice flake, lignin, water, starch, katsuobushi (dried bonito), ovalbumin, potato flake and spinach. The AUCs for administration in lipophilic foods (triolein and soya bean oil) were 50% and 42% of that following intravenous administration in saline. The AUCs for the other foods tested were 20-25% of that following intravenous administration in saline except for cellulose (around 30%), all were significantly lower than the lipophilic foods. These data suggest that the bioavailability of PAH from food will be in the range of 20-50% and that it increases with increasing content of lipophilic components in the food.

Distribution Distribution of PAH has been studied in rodents and levels in tissues are influenced by several factors; the PAH, route of administration, vehicle, time of tissue sampling after treatment and presence or absence of inducers or inhibitors of hydrocarbon metabolism. However three common traits are observed; there are detectable levels of PAH (probably more accurately PAH-derived material) in almost all organs, those organs rich in adipose tissue act as depots from which material is slowly released and high levels are found in the gastrointestinal tract irrespective of the route of administration. Following intravenous administration benzo[a]pyrene was rapidly removed from the bloodstream with a distribution half-life of less than 1 minute. The rate of clearance of radioactivity after administration of radio-labelled benzo[a]pyrene was increased after pretreatment with inducers of metabolism either benzo[a]pyrene or phenobarbital. A54 Whole body autoradiography was used to study distribution in mice and their fetuses following intravenous administration of 14C-labelled 3-methylcholanthrene to pregnant dams. Radioactivity was widely distributed in maternal tissues and was detected in fetuses showing that it crossed the placenta . Similar results have been obtained following inhalation, intragastric or intravenous administration of benzo[a]pyrene and 7,12-dimethylbenz[a]anthracene to rats and mice . In a small study in humans, samples of milk, placenta, maternal and umbilical cord blood were taken from 24 women and analysed for selected PAH. The highest levels of benzo[a]pyrene, dibenz[a,c]anthracene and chrysene were observed in milk and umbilical cord blood but levels were only above the detection limit in half of the samples. Nevertheless the authors concluded that both fetuses and infants were exposed to PAH which were presumed to be from the maternal diet .PAH in cancer free liver (N=6) and fat (N=10) biopsy samples obtained at autopsy from humans PAH Liver concentration µg/kg wet weight Fat concentration µg/kg wet weight benz[a]anthracene benz[a]anthracene > chrysene, and that their aqueous solubility decreased in a similar order. The group noted that these interactions were occurring at relatively high doses, the significance of these interactions at lower dose levels and for human risk assessment were unclear. A56.

Enterohepatic cycling of PAH Evidence for enterohepatic cycling was found after intratracheal administration of 1 µg/kg bw tritiated benzo[a]pyrene to bile-duct cannulated Sprague-Dawley rats. Studies in rats and rabbits showed the in vivo persistence of benzo[a]pyrene metabolites as a result of enterohepatic cycling. Chipman and colleagues demonstrated that bile was the major route of excretion in bile-duct cannulated rats in the initial 6 hours after intravenous administration of 3 µmol/kg bw 14C-labelled benzo[a]pyrene, biliary excretion accounting for 60% of the dose whilst urinary excretion was 3%. These workers showed that mutagenic or potentially mutagenic derivatives may be excreted via bile into the intestine . The gastrointestinal microflora have been shown to hydrolyse glucuronic acid conjugates of biliary PAH metabolites resulting in the formation of potentially reactive compounds . Other metabolites such as thio-ether conjugates have also been reported to undergo enterohepatic cycling . A study of the pharmacokinetics and bioavailability in rats following oral or intravenous administration of 2-15 mg/kg bw 14C-labelled pyrene provided strong evidence of enterohepatic cycling .

Metabolism Studies of metabolic pathways in whole animals have largely been restricted to the simpler compounds whereas hepatic homogenates, microsomes, cultured cells and explants are the principal method for studying the metabolism of larger more complex compounds. Metabolism and excretion has been studied in whole animals for naphthalene, anthracene, phenanthrene, pyrene, benz[a]anthracene and chrysene and to a lesser degree with benzo[a]pyrene, dibenz[a,h]anthracene and 3-methylcholanthrene. The metabolism of benzo[a]pyrene is described in detail below. The metabolism of PAH has been studied in a number of human cells and tissues including bronchus, colon, mammary cell aggregates, keratinocytes, monocytes, lymphocytes and bronchial macrophages. Whilst similar metabolites are formed in many of the in vitro tissue or cell preparations, both the relative levels and rate of formation are tissue or cell, species and strain of animal specific. The individual variability is marked with around 75-fold variation in the extent of PAH activation as measured by DNA adduct formation reported in human bronchus, mammary cell aggregates and macrophages. A57 What can metabolism do Metabolism of xenobiotics is divided into two phases; phase 1 generally involves alteration of the structure of the compound to increase polarity and phase 2 addition of polar groups. Although metabolism was originally considered to be the inactivation or detoxification of foreign compounds, it has become increasingly clear that the actual situation is more complex and can result in either increased or decreased reactivity of the compound. Further phase 1 metabolism can occur when the initial introduction of a polar group does not alter physicochemical properties markedly and reactive metabolites can also occur from rearrangement or metabolism following conjugation. The general scheme of PAH metabolism involves oxidation to a range of primary (epoxides, phenols, dihydrodiols) and secondary (diol epoxides, tetrahydrotetrols, phenol epoxides) phase 1 metabolites followed by conjugation to phase 2 metabolites with glutathione, glucuronic acid or sulphate. Studies using simple PAH such as naphthalene demonstrated that similar phase 1 metabolites were produced following incubation with hepatic microsomes or tissue homogenates as after administration to animals. The metabolism of PAH has been extensively studied in vitro often in microsomal fractions of rat liver but also in many other tissue preparations. These in vitro methods are considered to adequately represent the in vivo metabolism of PAH.

Summary of absorption, distribution, metabolism and elimination of PAH

The toxicokinetic data indicates that the absorption of PAH following oral administration is affected by the other components of the diet. The evidence suggests that the oral bioavailabilty of PAH ranges from around 40-50% in highly lipophilic milieus to 20-25% in less lipophilic milieus and mixed food. The absorbed PAH are widely distributed particularly to fatty tissues. PAH are associated with toxic effects at the site of first contact and systemically. All PAH can be metabolised by phase 1 enzymes to electrophilic metabolites under experimental conditions. These metabolites are able to react with proteins and DNA, but these metabolites can also be deactivated by further metabolism. Effects are related to the extent of electrophile production and detoxification in vivo. Thus reactive metabolites are not formed to a significant extent from anthracene in vivo whereas for benzo[a]pyrene they represent a significant or major route of metabolism and this is reflected in the data on the genotoxicity and carcinogenicity of these compounds. In using the metabolism data for the risk assessment of an individual PAH certain questions require an answer. Firstly the ability of the compound to produce an electrophile has to be established. The degree to which this represents a major route of metabolism and the characteristics (reactivity, stability, inactivation) of the electrophile require elicitation. Finally the possible effects of route of administration, species specificity or dose dependency on the metabolism needs addressing. In general all PAH have the potential to generate electrophiles, and all probably do produce electrophiles at some concentration or in some circumstances. However several PAH do not produce these to an appreciable A67 extent in vivo e.g. anthracene whereas for other compounds they represent a significant or major route of metabolism e.g. benzo[a]pyrene.

Special studies on cardiovascular effects of PAH

It has been hypothesised that a mechanism could be that PAH from cigarette smoke tars or combustion products could cause endothelial injury and changes in smooth muscle cells leading to clonal expansion of these in the arterial walls and thereby might contribute to the development of arteriosclerosis. It is unequivocal that tobacco smoking is a major risk factor for cardiovascular disease and there is also some evidence that occupational exposure to combustion products including PAH might also be associated with an increased risk of cardiovascular disease . Following induction with 3-methylcholanthrene microsomes from rat aorta transformed benzo[a]pyrene into active carcinogenic and toxic metabolites . In humans, PAH adducts have been detected in the endothelium of the internal mammary artery of smokers. DNA adducts has also been detected by 32P-postlabelling in smooth muscle cells of human abdominal aorta affected by arteriosclerotic lesions. The levels of adducts were correlated to current smoking and other atherogenic risk factors. In individuals with the null GSTM1 had higher adduct levels in both non-smokers and smokers . It is not established whether the DNA adducts are formed from PAH. Similar adducts in the aorta was also observed in smoke exposed rats . PAH adducts determined by immunohistochemistry were also found in the endothelium of microvessels in the muscular layer of large blood vessels . In human umbilical vein endothelial cells cultured in vitro, and induced with βnaphthoflavone, benzo[a]pyrene induced DNA damage as measured by alkaline single cell gel electrophoresis . Benzo[a]pyrene and 7,12-dimethylbenz[a]anthracene caused a decreased secretion of newly synthesised collagen from bovine arterial smooth muscle cells in vitro without reducing the cell number. Total cellular DNA was decreased and the relative collagen secretion increased. When the cells had been preincubated with platelet factors similar effects were observed. Benzo[a]pyrene and 7,12-dimethylbenz[a]anthracene could also induce cell death and release of mitogenic factors, dependingen on the concentration and mode of administration. Dimethylbenz[a]anthracene induced proliferation of smooth muscle celles. This effect was independent of the Ah-receptor . Glutathione transferase (GST) negative human fibroblasts produced relatively more collagene than GST positive fibroblasts . Although PAH related adducts have been observed in blood vessels in humans and some effects of PAH on vascular cells in vitro, this does not prove that the increased cardiovascular risk following tobacco smoking and occupational exposure to combustion products is due to exposure to PAH. A141

Observations in humans Biomarkers of exposure to PAH Several methods have been developed to assess internal exposure to PAH after exposure in the environment and in workplaces. In most studies, metabolites of PAH were measured in urine, 1-hydroxypyrene being widely used. It should be noted that the concentration or excretion of parent PAH compound or metabolites in body fluids or urine not only is dependent on the external exposure, but also on absorption, biotransformation and excretion, which can vary considerably between individuals. Genotoxic endpoints of PAH have been used, but they are unspecific and will not be further discussed. Adducts of benzo[a]pyrene with DNA in peripheral lymphocytes, and other tissues and with proteins such as albumin have been used as an indicator of the dose of reactive metabolite. Urinary metabolites The metabolites mostly used have been hydroxylated phenanthrenes and 1-hydroxypyrene. Also total thioether excretion has been used, but this is unspecific. 1-Hydroxypyrene 1-Hydroxypyrene, a metabolite of pyrene, has been widely used as urinary biomarker of PAH exposure since 1986. Its advantages are that pyrene is present in all PAH mixtures at relatively high concentrations (2-10%) of the total PAH, and in certain environments the pyrene content of the total PAH is fairly constant. However, the relationship between pyrene and benzo[a]pyrene may vary considerably beteen different exposures. Coal tar contains 2-10 % pyrene and 0.4-0.6 % benzo[a]pyrene. In studies at different workplaces, a strong correlation was found between the pyrene concentrations in air and those of benzo[a]pyrene, other selected PAH, and total PAH. Pyrene is metabolised predominantly to 1-hydroxypyrene, which can be measured easily. In contrast to other PAH metabolites, which are excreted mainly in faeces, 1-hydroxypyrene is excreted in urine. The background concentrations of 1-hydroxypyrene in urine of persons from different countries range from 0.06 to 0.23 µmol/mol creatinine . For nonsmokers and non-occupationally exposed individuals, food accounted for 99% of the total daily pyrene intake . Five volunteers who ate low-PAH meals and high-PAH meals showed 100- to 250-fold increases in benzo[a]pyrene dose, accompanied by a four- to 12-fold increase in 1-hydroxypyrene excretion in urine . Ten volunteers eating charbroiled beef for five days had a 10-80 fold increase in 1-OH-pyrene glucuronide excretion in urine above background returning to background within 24-72 hours . The intake of pyrene from cigarette smoking (12 nmol/day) is about the same as the dietary intake from normal food (9.4 nmol/day). Tobacco smokers who are not otherwise exposed to PAH have A142 about twice the level of 1-hydroxypyrene in their urine as non-smokers. 1-Hydroxypyrene concentrations in the urine of persons occupationally exposed to PAH at various workplaces are usually increased. Dermal uptake can be a significant pathway in many cases. 1-Hydroxypyrene cannot predict exposure to benzo[a]pyrene or other carcinogenic PAH as the relative content of these two compounds can vary considerably. DNA adducts DNA adducts with reactive metabolites (mainly diol epoxides) of benzo[a]pyrene and other PAH have been identified in humans exposed to smoking or living in polluted areas in numerous studies . PAH- DNA adducts has also been detected in peripheral white blood cells following human exposure to charbroiled meat . Cigarette smokers have higher levels of adducts with PAH in their lungs than nonsmokers. As binding of electrophilic PAH metabolites to DNA is thought to be a key step in the initiation of cancer, measurement of DNA adducts could be an indicator of exposure to PAH and also of the dose of the ultimate reactive metabolite. An increased lung cancer risk has been found among smoking individuals with a higher level of aromatic DNA adducts in white blood cells . The methods for measuring DNA adducts include immunoassays with polyclonal and monoclonal antibodies (enzyme-linked immunosorbent assay [ELISA] and ultrasensitive enzymatic radioimmunoassay), 32P-postlabelling, and synchronous fluorescence spectrophotometry. Direct comparisons of adduct levels determined by different techniques may be misleading, however, because different end-points are measured. In the general population the levels of DNA adduct in control subjects range from 0.2 to about 10 adducts per 108 nucleotides in leukocytes . There are many studies on the effect of adduct level in leukocytes of tobacco smokers and also on populations living in areas polluted by industry; the latter with adduct levels up to 5-13 adducts per 108 nucleotides. Eating charcoal-grilled beef resulted in a 1.9-3.8-fold increase above the individual baseline adduct levels in four of 10 subjects . Workers exposed to PAH in general had elevated levels of adducts (5-70 adducts per 108 nucleotides) . DNA-adducts and 1-hydroxypyrene In general, exposures that lead to the excretion of high concentrations of 1-hydroxypyrene in urine also lead to elevated DNA adduct levels in white blood cells and significant correlations have been found. Although the concentrations of PAH that occur under different exposure conditions differ by orders of magnitude, the differences in DNA adduct levels are quite small, in contrast to the results of experiments on excretion of 1- hydroxypyrene. In all populations studied, exposed by inhalation or orally, there is substantial inter-individual variation in PAH-DNA adduct levels, which is greater than that A143 described for 1-hydroxypyrene excretion in urine. This is probably due to differences e.g. in biotransformation, excretion, DNA adduct removal etc. (Autrup, 2000, Lee et al 2002). Such interindividual variation result in a wide overlap in the ranges of values between exposed and unexposed subjects in all studies. Protein adducts Because genotoxic compounds can bind to haemoglobin and serum protein, the assessment of PAH-blood protein adducts has also been considered as a possible marker of exposure to PAH. However, studies show conflicting results using such adducts as a marker of exposure. Biomarkers and exposure via inhalation and diet In many cases exposure to PAH from food is a confounder when using biomarkers in the evaluation of exposure to PAH by inhalation. Although increased levels in urinary 1- hydroxypyrene have been found in people living in polluted areas exposure from ambient air, large changes in atmospheric levels of PAH is not reflected in urinary 1-hydroxypyrene and DNA-adducts. This indicates that ambient air is relatively unimportant in comparison with dietary PAH and tobacco smoking . In a study of forest fire-fighters in the USA levels of PAH-DNA adducts in blood cells were not found to correlate with recent fire-fighting activity, but with recent consumption of charbroiled meat. Surprisingly the PAH DNA-adduct levels were lower in US army personnel fighting oil field fires in Kuwait, possibly because a lower intake of charbroiled meat (Phillips, 1999). 2.9.2 Effects of PAH exposures observed in humans Unfortunately there are almost none published studies on health effects in humans following oral exposure to PAH. In the majority of studies humans have been occupationally exposed to PAH via inhalation and in a few studies the exposure has been dermal. There is little information on human exposure to single, pure polycyclic aromatic hydrocarbons (PAH) except for accidental exposure to naphthalene and some data from defined short-term studies of volunteers which are not relevant for the human exposure to PAH via food. All other reports are on exposure to mixtures of PAH, which also contained other potentially carcinogenic chemicals, in occupational and environmental situations.

Information on the health effects of these mixtures is in practically all cases confined to their carcinogenic potential, for which there is evidence from a number of epidemiological studies, especially for lung cancer and, in some cases, cancers of the skin and of the urinary bladder. Oral exposure to PAH Only one fully reported study on oral PAH exposure and health effects has been identified (Lopez-Abente et al., 2001). In some rural areas in Spain wine has traditionally been stored in leather bottles sealed with a tar-like substance (i.e. pez) obtained through boiling and A144 destilation of fir and pine wood, and which contain PAH. In order to assess this exposure with regard to risk of gastric cancer, 59 cases and 53 controls all of who were residents in the Province of Soria, were selected from a case control multi-center study from Spain on gastric cancer. The multi-center study consisted of 354 incident cases and 354 controls matched by age, sex and place of residence. The exposure to wine stored in tar impregnated leather bottles was assessed by a self-administered questionnaire. A total of 85 questionnaires were returned of which 78, i.e. 38 cases and 40 controls, could be analysed. In the analysis odds ratios (OR) were calculated by logistic regression taking into account also variables that had been found to be associated with gastric cancer in the multi-center study. Several variables for consumption of wine from leather bottles were reported. Although an increased OR was reported, the study population was to small to achieve statistically significant increases. An exception was consumption of more than 2 litres of wine/week, which appeared to be associated with gastric cancer particularly in males OR and 95 % confidence interval = 10.5, 1.13-97.76, p value for trend 0.02. No change in effect estimator was observed upon inclusion of other risk variables in the model. Studies on intake of cooked meat In a published abstract Sinha and co-workers (2001) described the risk of colorectal adenomas and dietary benzo[a]pyrene intake in a case-control study of 146 newly diagnosed cases and 226 controls. In this study dietary intake of red meat, well-done red meat, grilled red meat and exposure to heterocyclic amines such as 2-amino3,8-dimethylimidazo[4,5-f]-quinoxaline (MeIQx) and 2-amino-1-methyl-6- phenylimidazo[4,5-b]pyridine (PhIP) and benzo[a]pyrene were estimated using a food frequency questionnaire with detailed questions also on meat-cooking methods in combination with a heterocyclic amine and benzo[a]pyrene database . Increased risks were found with a high intake of red meat, well-done and grilled and heterocyclic amines. The median (10th and 90th percentile) exposure to benzo[a]pyrene in the controls was 5 ng (0.2-66) ng/day from meat and 73(35-140) ng/day from all foods. In cases, median benzo[a]pyrene intake was 17 (0.5-101) ng/day from meat and 76 (44-163) ng/day from all foods When multivariate analysis were carried out adjusting for MeIQx and established colorectal adenoma risk factors, the odds ratios (95 % confidence interval) for dietary benzo[a]pyrene from meat with the first quintile as the referent group were: 1.5 (0.7-3.4, 2.0 (0.9-4.3), 2.6 (1.2-5.7) and 3.3 (1.5-7.4) for the second, third, fourth and fifth quintile. The p-value for trend was 0.01. Increased risk of colorectal adenomas was also associated with benzo[a]pyrene intake estimated from all foods, p-value for trend 0.01. In a short feeding study ten healthy adults were fed a diet enriched with chargrilled meat for 7 days . The meat contained from 8.35 to 15.64 ng benzo[a]pyrene/ g and also heterocyclic amines: 100 and 50 ng PhIP/g. Whereas the intake of PhIP was estimated to 13-28 µg/day the intake of benzo[a]pyrene was not given. The chargrilled meat intake resulted in an induction of CYP1A enzymes both in the liver and the small intestine. No induction of CYP3A4, CYP3A5 or P-glycoprotein in the small or large intestine and CYP3A4 in the liver was observed. There was an inverse correlation between the level of PAH DNA adducts in peripheral blood mononuclear cells and both liver A145 CYP1A2 activity and enterocyte CYP1A1 protein concentration on day 11. It is not clear whether the enzyme induction was due to heterocyclic amines or PAH as Sinha and coworkers (1994) in a previous study feeding of pan-fried meat containing high levels of heterocyclic amines, but unchanged and low levels of PAH, also found an induction of CYP1A2.

PAH in cigarette smoke Tobacco smoke contains a mixture of PAH in addition to numerous other carcinogens. For example levels of 11 ng per cigarette benzo[a]pyrene were found in mainstream smoke and 103 ng per cigarette in sidestream smoke; the corresponding values were 6.8 and 76 ng per cigarette for benzo[e]pyrene, 20 and 497 ng per cigarette for chrysene and triphenylene, and 13 and 204 ng per cigarette for benz[a]anthracene for mainstream and sidestream smoke, respectivelly. Other data on intake by smoking are reported in the section ‘Intake estimates’. A large volume of literature exists on the effects of tobacco smoke on human lungs (see IARC, 1986). On the basis of large body of studies in many countries, cigarette smoke has been shown to be by far the most important single factor contributing to the development of lung cancer. Other types of cancer caused by cigarette smoking include cancers of the oral cavities, larynx, pharynx, oesophagus, bladder, renal pelvis, renal adenocarcinoma, and pancreas. Based upon studies using implantation into the lungs of rats estimated that PAH with four or more rings were responsible for 83% of the total carcinogenic activity of sidestream smoke .

PAH and occupational exposure Many workplaces have atmospheres with heavy loads of PAH. In general, industrial workers using or producing coal or coal products are exposed to mixtures of PAH The first cancer that might be attributed to an occupational exposure was reported by Pott in 1775, who described the susceptibility of English chimney sweeps to scrotal cancer a second was published by Butlin in 1892 . Epidemiological studies have been conducted on workers exposed at coke ovens in coal coking and coal gasification, at asphalt works, at foundries, and at aluminium smelters and to diesel exhaust. Details of the most recent and most important cohort and case-control studies are given in IPCS (1998) . Most important for an evaluation of the possible risk of cancer due to exposure to PAH are studies of workers exposed at coke ovens in coke plants or in coal-gasification processes. Particularly at coke ovens, the PAH concentrations are considerable, with levels up to 200 /m3 total PAH, but values between some tens of /m3 are more common; extreme values up to 1 mg/m3 total PAH and 100-300 µg/m3 benzo[a]pyrene have been reported in the 60s and 70s in eastern European countries . Values reported from coal A146 gasification plants are lower. In most epidemiological studies on PAH exposed workers, however, the concentrations to which workers were exposed are not available. Significantly increased risks of lung cancer was found among coke oven workers including workers in coal-gasification particularly in the large cohort studies (IPCS, 1998). Several epidemiological studies have been performed on the potential risks of handling asphalt. In the meta-analysis of Partanen and Boffetta, increased risks for lung tumours were seen for both pavers and roofers; tumours of the stomach, bladder, and skin and leukaemia were also observed. Workers are exposed dermally to very high concentrations of PAH when impregnating wood with creosote. Increased risk for skin and lip cancer has been observed. Since the work involves some time outdoors, it cannot be ruled out that exposure to sunlight contributed to the risks for cancers of the skin and lip.

The PAH concentrations in iron, steel, and other ferroalloy foundries reach levels of 50 µg/m3 and that of benzo[a]pyrene about 10 µg/m3 . Increased mortality from lung cancer has been observed consistently in many studies of foundry workers . Information on the possible risks for cancer due to exposure to PAH can also be obtained from studies of workers in aluminium plants. During Söderberg electrolysis, workers may be exposed to 3-35 µg/m3 benzo[a] pyrene; workers in the carbon area are exposed to 0.4- 1.2 µg/m3 benzo[a] pyrene. An increased risk for lung cancer, but also urinary bladder cancer has been found in several studies with exposure in a Söderberg potroom. The risk of bladder cancer may be due to exposure not only to PAH but also to aromatic amines, which have been detected in the potrooms.

Increased risks for lung cancer were found in several studies of workers exposed to diesel exhausts (WHO, 1996). In comparison with the occupations described above, the concentrations of PAH to which these workers are exposed are usually relatively low. The benzo[a]pyrene concentrations in automobile repair shops and garages reach about 70 ng/m3 , and truck drivers are exposed to less than 10 ng/m3 . The increased risk is seen for workers in several occupations, which have exposure to PAH in common. Although other carcinogenic chemicals were present, they differed with each occupation. Furthermore, tobacco smoking is a confounder in several studies, but this factor in general cannot explain the excess risk. Airborne high-molecular-mass PAH, which are considered to be the most carcinogenic, are adsorbed mainly onto particulate matter, and it was often difficult to distinguish the toxicological effects caused by particles from those caused by the PAH themselves. A147

PAH exposure from unvented coal combustion in homes Interdisciplinary studies were conducted to investigate exposure to PAH and the high lung cancer rates in a rural county, Xuan Wei, located in Yunnan Province, China . Mortality from lung cancer in this county is five times the Chinese national average, especially among the women, who have the highest rate in China. The mortality rate from lung cancer was correlated with domestic use of 'smoky' coal (medium-volatile bituminous coal with low sulphur and high ash) as fuel for cooking and heating, but not with use of wood or smokeless coal. Monitoring of air during cooking inside the homes showed that women were exposed to extremely high levels of PAH, with a mean benzo[a] pyrene concentration of 14.7 µg/m3 , comparable to the levels to which coke-oven workers were exposed. Urine samples from Xuan Wei residents confirmed that they were exposed to high concentrations of alkylated PAH. Thus, alkylated PAH may play an important role in the etiology of lung cancer ..