Urinary alpha1-microglobulin excretion as biomarker of renal toxicity in trichloroethylene-exposed persons

OBJECTIVES

Concern on human renal toxicity and carcinogenicity of trichloroethylene is based on findings of increased incidences of renal cell cancers in persons with long-lasting and high occupational exposures to this solvent. The full tumour development is likely to require promotional stimuli, by repetitive episodes of high peak exposures to trichloroethylene, leading to nephrotoxicity. This process is visualised by the excretion of tubular marker proteins in the urine of exposed persons. For this purpose, surveillance of alpha1-microglobulin excretion is being suggested by the Deutsche Forschungsgemeinschaft.

METHODS

The present study assessed the applicability of alpha1-microglobulin as a biomarker of proximal tubule damage in the prevention of nephrotoxicity by trichloroethylene exposures. For this purpose, alpha1-microglobulin excretions were assessed in trichloroethylene-exposed and non-exposed subgroups of both cases (diseased with renal cancer) and controls (not diseased with renal cancer) of a recent case-control study.

RESULTS

The median of alpha1-microglobulin excretions in non-exposed persons was below the detection limit, but it was clearly elevated in exposed persons. The Wilcoxon rank sum test showed a significant difference ( P=0.0090). Consistent with the underlying concept, renal cell cancer cases who had been exposed to trichloroethylene had higher alpha1-microglobulin excretions than non-exposed cases ( P=0.0005) and also had higher alpha1-microglobulin excretions than exposed controls ( P=0.0004). Of 20 trichloroethylene-exposed renal cell cancer cases only three (15%) displayed a normal alpha1-microglobulin excretion of <5 mg/l. By contrast, 41 (52%) out of 79 non-exposed renal cancer cases showed normal excretions of the biomarker.

CONCLUSION

The present data are in agreement with the concept of pathogenesis of renal cell cancers developing under high (suggested: >500 ppm peak exposures) and long-term (several years) exposure to trichloroethylene. They also visualise the potential value of alpha1-microglobulin excretion as a routine biomarker of renal toxicity that may be used in the medical surveillance of trichloroethylene-exposed persons.

Chromosomal genotoxicity of nitrobenzene and benzonitrile

In order to investigate the chromosomal genotoxicity of nitrobenzene and benzonitrile, we studied the induction of micronuclei (MN) by these test compounds in V79 cells, as well as effects on the formation and stability of microtubules and on motor protein functions. No cytotoxicity was seen in V79 cell cultures in terms of Neutral red uptake after 18 h treatment with up to 1 mM nitrobenzene or 1 mM benzonitrile. Subsequently, a concentration range up to 100 micro M was used in the experiments on induction of MN. Both test compounds exhibit a weak, but definitely positive test result compared to the solvent (DMSO) control. Minimal effect concentrations of nitrobenzene and benzonitrile appeared as low as 0.01 micro M, and no-effect-concentrations were between 0.001 and 0.005 micro M. Clearly enhanced MN rates were found at 0.1 micro M and higher. Both, nitrobenzene and benzonitrile, induced mostly kinetochor (CREST)-positive micronuclei, thus characterising the chromosomal effects as aneugenic. In cell-free assays, a slight effect on tubulin assembly was observed at 1 mM nitrobenzene without addition of DMSO. Higher concentrations (5 mM) led to secondary effects. In presence of 1% DMSO, nitrobenzene exerted no detectable effect on tubulin assembly up to the solubility limit in water of about 15 mM. For benzonitrile in presence of DMSO, a clear dose-response of inhibition of tubulin assembly at 37 degrees C was seen above the no-effect-concentration of 2 mM, with an IC(50) of 13 mM and protein denaturation starting above a level of about 20 mM. The nature of the effects of nitrobenzene and benzonitrile on the association of tubulin to form microtubules was confirmed by electron microscopy. Treatment by either 5 mM nitrobenzene or 13 mM benzonitrile plus 1% DMSO left the microtubular structure intact whereas 5 mM nitrobenzene, in absence of DMSO, led to irregular cluster formations. The experiments demonstrate that both nitrobenzene and benzonitrile, in millimolar concentration ranges, may lead to interference with tubulin assembly in a cell-free system. The functionality of the tubulin-kinesin motor protein system was assessed using the microtubule gliding assay. Nitrobenzene affected the gliding velocity in a concentration-dependent manner, starting at about 7.5 micro M and reaching complete inhibition of motility at 30 micro M, whereas benzonitrile up to 200 micro M did not affect the kinesin-driven gliding velocity. The micronucleus assay data demonstrate a chromosomal endpoint of genotoxicity of nitrobenzene and benzonitrile. Aneugenic effects of both compounds occur at remarkably low concentrations, with lowest-effect-concentrations being 0.1 micro M. This points to the relevance of interactions with the cellular spindle apparatus.

Occupational exposure and urological cancer

Occupational exposure is definitely a major cause of cancer. In the field of urology, the urinary bladder is the most important target. A classical cause of bladder cancer is exposure to carcinogenic aromatic amines, especially benzidine and beta-naphthylamine. Such exposures were related to work places in the chemical industry, implying production and processing of classical aromatic amines, and in the rubber industry. Occupational bladder cancer has also been observed in dyers, painters and hairdressers. Even some occupations with much lower exposures to carcinogenic aromatic amines, like coke oven workers or workers in the rubber industry after the ban on beta-naphthylamine, are at risk. In these occupations, exposure to complex mixtures of substances containing combustion products (e.g. polycyclic aromatic hydrocarbons) or nitrosamines is common. Renal cell cancer has been observed as an occupational disease in cases of very high exposure to trichloroethylene having led to narcotic or prenarcotic symptoms. Occupationally related cancers of the prostate or the testes appear currently not relevant.

Di(2-ethylhexyl)phthalate (DEHP) metabolites in human urine and serum after a single oral dose of deuterium-labelled DEHP

Human metabolism of di(2-ethylhexyl)phthalate (DEHP) was studied after a single oral dose of 48.1 mg to a male volunteer. To avoid interference by background exposure the D4-ring-labelled DEHP analogue was dosed. Excretion of three metabolites, mono(2-ethyl-5-hydroxyhexyl)phthalate (5OH-MEHP), mono(2-ethyl-5-oxohexyl)phthalate (5oxo-MEHP) and mono(2-ethylhexyl)phthalate (MEHP), was monitored for 44 h in urine and for 8 h in serum. Peak concentrations of all metabolites were found in serum after 2 h and in urine after 2 h (MEHP) and after 4 h (5OH-MEHP and 5oxo-MEHP). While the major metabolite in serum was MEHP, the major metabolite in urine was 5OH-MEHP, followed by 5oxo-MEHP and MEHP. Excretion in urine followed a multi-phase elimination model. After an absorption and distribution phase of 4 to 8 h, half-life times of excretion in the first elimination phase were approximately 2 h with slightly higher half-life times for 5OH- and 5oxo-MEHP. Half-life times in the second phase-beginning 14 to 18 h post dose-were 5 h for MEHP and 10 h for 5OH-MEHP and 5oxo-MEHP. In the time window 36 to 44 h, no decrease in excreted concentrations of 5OH- and 5oxo-MEHP was observed. In the first elimination phase (8 to 14 h post dose), mean excretion ratios of MEHP to 5oxo-MEHP and MEHP to 5OH-MEHP were 1 to 1.8 and 1 to 3.1. In the second elimination phase up to 24 h post dose mean excretion ratios of MEHP to 5oxo-MEHP to 5OH-MEHP were 1 to 5.0 to 9.3. The excretion ratio of 5OH-MEHP to 5oxo-MEHP remained constant through time at 1.7 in the mean. After 44 h, 47% of the DEHP dose was excreted in urine, comprising MEHP (7.3%), 5OH-MEHP (24.7%) and 5oxo-MEHP (14.9%).

Challenging dogma: thresholds for genotoxic carcinogens? The case of vinyl acetate

Although many questions remain unanswered, the general principle of the sequence of events leading to cancer after exposure to genotoxic carcinogens has become increasingly clear. This helps to understand the parameters that influence the shape of the dose-effect curve for carcinogenesis, including metabolic activation and inactivation of carcinogens, DNA repair, cell cycle control, apoptosis, and control by the immune system. A linear dose-response relationship with no observable threshold seems to be a conservative but adequate description for the carcinogenic activity of many genotoxic carcinogens, such as aflatoxin B1, the tobacco-specific nitrosoketone NNK, and probably N,N-diethylnitrosamine. However, extrapolation models connecting the high-level risk to the zero intercept have clearly resulted in overestimations of risk. Vinyl acetate is an example that is discussed extensively in this review. At extremely high and toxic doses, vinyl acetate is carcinogenic in rats and mice and causes chromosomal aberrations. In tissues of contact, vinyl acetate is converted to acetic acid and acetaldehyde. Only when threshold levels are achieved do critical steps in the mechanism ultimately leading to cancer become active, namely pH reduction in exposed cells of more than 0.15 units leading to cytotoxicity, damage to DNA, and regenerative proliferation. Consistent with the known exposure to endogenous acetic acid and acetaldehyde, tissues sustain a certain level of exposure without adverse effects. Physiological modeling shows that the conditions necessary for carcinogenesis are in place only when threshold levels of vinyl acetate are exceeded. The example of vinyl acetate underlines the importance of toxicological research that unequivocally identifies genotoxic carcinogens acting through a threshold process.

Markers of genetic susceptibility in human environmental hygiene and toxicology: the role of selected CYP, NAT and GST genes

Inherited genetic traits co-determine the susceptibility of an individual to a toxic chemical. Special emphasis has been put on individual responses to environmental and industrial carcinogens, but other chronic diseases are of increasing interest. Polymorphisms of relevant xenobiotic metabolising enzymes may be used as toxicological susceptibility markers. A growing number of genes encoding enzymes involved in biotransformation of toxicants and in cellular defence against toxicant-induced damage to the cells has been identified and cloned, leading to increased knowledge of allelic variants of genes and genetic defects that may result in a differential susceptibility toward environmental toxicants. "Low penetrating" polymorphisms in metabolism genes tend to be much more common in the population than allelic variants of "high penetrating" cancer genes, and are therefore of considerable importance from a public health point of view. Positive associations between cancer and CYP1A1 alleles, in particular the *2C I462V allele, were found for tissues following the aerodigestive tract. Again, in most cases, the effect of the variant CYP1A1 allele becomes apparent or clearer in connection with the GSTM1 null allele. The CYP1B1 codon 432 polymorphism (CYP1B1*3) has been identified as a susceptibility factor in smoking-related head-and-neck squameous cell cancer. The impact of this polymorphic variant of CYP1B1 on cancer risk was also reflected by an association with the frequency of somatic mutations of the p53 gene. Combined genotype analysis of CYP1B1 and the glutathione transferases GSTM1 or GSTT1 has also pointed to interactive effects. Of particular interest for the industrial and environmental field is the isozyme CYP2E1. Several genotypes of this isozyme have been characterised which seem to be associated with different levels of expression of enzyme activity. The acetylator status for NAT2 can be determined by genotyping or by phenotyping. In the pathogenesis of human bladder cancer due to occupational exposure to "classical" aromatic amines (benzidine, 4-aminodiphenyl, 1-naphthylamine) acetylation by NAT2 is regarded as a detoxication step. Interestingly, the underlying European findings of a higher susceptibility of slow acetylators towards aromatic amines are in contrast to findings in Chinese workers occupationally exposed to aromatic amines which points to different mechanisms of susceptibility between European and Chinese populations. Regarding human bladder cancer, the hypothesis has been put forward that genetic polymorphism of GSTM1 might be linked with the occurrence of this tumour type. This supports the hypothesis that exposure to PAH might causally be involved in urothelial cancers. The human polymorphic GST catalysing conjugation of halomethanes, dihalomethanes, ethylene oxide and a number of other industrial compounds could be characterised as a class theta enzyme (GSTT1) by means of molecular biology. "Conjugator" and "non-conjugator" phenotypes are coincident with the presence and absence of the GSTT1 gene. There are wide variations in the frequencies of GSTT1 deletion (GSTT1*0/0) among different ethnicities. Human phenotyping is facilitated by the GST activity towards methyl bromide or ethylene oxide in erythrocytes which is representative of the metabolic GSTT1 competence of the entire organism. Inter-individual variations in xenobiotic metabolism capacities may be due to polymorphisms of the genes coding for the enzymes themselves or of the genes coding for the receptors or transcription factors which regulate the expression of the enzymes. Also, polymorphisms in several regions of genes may cause altered ligand affinity, transactivation activity or expression levels of the receptor subsequently influencing the expression of the downstream target genes. Studies of individual susceptibility to toxicants and gene-environment interaction are now emerging as an important component of molecular epidemiology.

Genotoxicity--threshold or not? Introduction of cases of industrial chemicals

Many industrially and environmentally important industrial carcinogens display effects that lead them to be viewed and regulated as 'genotoxic compounds'. Some of these chemicals cause experimental tumours only at high or toxic doses. The current view is that non-threshold principles should be applied for risk assessments and to define permissible exposure values. The toxicological impact of underlying mechanisms is frequently not well investigated and understood. The classification of carcinogens is now in a state of discussion. In Germany, the 'MAK-Commission' has issued new recommendations to distinguish between 5 groups of proven and suspected carcinogens. This proposal includes a category of 'substances with carcinogenic potential for which genotoxicity plays no or at most a minor role'. Another category comprises 'substances with carcinogenic and genotoxic potential, the potency of which is considered so low that, provided that the MAK-value is observed, no significant contribution to human cancer risk is to be expected'. There is also a number of apparently genotoxic carcinogens where the existence of 'practical thresholds' is at least debated. One outstanding example is vinyl acetate, which must be viewed against the background of discussions on other industrial high-volume chemicals like formaldehyde, acrylonitrile, acrylamide and trichloroethylene. Main arguments in favour or against thresholds of carcinogenicity of these individual compounds are summarised. Current instruments of regulation should be adjusted to allow adequate consideration of carcinogenic effects of chemicals that are practically relevant at high doses only. Also, research into this field is encouraged.

Interaction of metal salts with cytoskeletal motor protein systems

Interactions of chemicals with the microtubular network of cells may lead to genotoxicity. Micronuclei (MN) might be caused by interaction of metals with tubulin and/or kinesin. The genotoxic effects of inorganic lead and mercury salts were studied using the MN assay and the CREST analysis in V79 Chinese hamster fibroblasts. Effects on the functional activity of motor protein systems were examined by measurement of tubulin assembly and kinesin-driven motility. Lead and mercury salts induced MN dose-dependently. The no-effect-concentration for MN induction was 1.1 microM PbCl(2), 0.05 microM Pb(OAc)(2) and 0.01 microM HgCl(2). The in vitro results obtained for PbCl(2) correspond to reported MN induction in workers occupationally exposed to lead, starting at 1.2 microM Hg(II) (Vaglenov et al., 2001, Environ. Health Perspect. 109, 295-298). The CREST Analysis indicate aneugenic effects of Pb(II) and aneugenic and additionally clastogenic effects of Hg(II). Lead (chloride, acetate, and nitrate) and mercury (chloride and nitrate) interfered dose-dependently with tubulin assembly in vitro. The no-effect-concentration for lead salts in this assay was 10 microM. Inhibition of tubulin assembly by mercury started at 2 microM. The gliding velocity of microtubules along immobilised kinesin molecules was affected by 25 microM Pb(NO(3))(2) and 0.1 microM HgCl(2) in a dose-dependent manner. Our data support the hypothesis that lead and mercury genotoxicity may result, at least in part, via disturbance of chromosome segregation via interaction with cytoskeletal proteins.

The cytochrome P-450 isoenzyme CYP2E1 in the biological processing of industrial chemicals: consequences for occupational and environmental medicine

The importance of the isoform CYP2E1 of the human cytochrome P-450 superfamily of enzymes for occupational and environmental medicine is derived from its unique substrate spectrum that includes a number of highly important high-production chemicals, such as aliphatic and aromatic hydrocarbons, solvents and industrial monomers (i.a. alkanes, alkenes, aromatic and halogenated hydrocarbons). Many polymorphic genes, such as CYP2E1, show considerable differences in allelic distribution between different human populations. The polymorphic nature of the human CYP2E1 gene is significant for inter-individual differences in toxicity of its substrates. Since the substrate spectrum of CYP2E1 includes many compounds of basic relevance to industrial toxicology, a rationale for metabolic interactions of different CYP2E1 substrates is provided. In-depth research into the inter-individual phenotypic differences of human CYP2E1 enzyme activities was enabled by the recognition that the 6-hydroxylation of the drug chlorzoxazone is mediated by CYP2E1. Studies on CYP2E1 phenotyping have pointed to inter-individual variations in enzyme activities. There are consistent ethnic differences in CYP2E1 enzyme expression, mostly demonstrated between European and Japanese populations, which point to a major impact of genetic factors. The most frequently studied genetic polymorphisms are the restriction fragment length polymorphisms PstI/ RsaI (mutant allele: CYP2E1*5B) located in the 5'-flanking region of the gene, as well as the DraI polymorphism (mutant allele: CYP2E1*6) located in intron 6. These polymorphisms are partly related, as they form the common allele designated CYP2E1*5A. Striking inter-ethnic differences between Europeans and Asians appear with respect to the frequencies of the CYP2E1*5A allele (only approximately 5% of Europeans are heterozygous, but 37% of Asians are, whilst 6% of Asians are homozygous). Available studies indicate a wide variation in human CYP2E1 expression, which are very likely based on complex gene-environment interactions. Major inter-ethnic differences are apparent on the genotyping and the phenotyping levels. Selected cases are presented where inter-ethnic variations of CYP2E1 may provide likely explanations for unexplained findings concerning industrial chemicals that are CYP2E1 substrates. Possible consequences of differential inter-individual and inter-ethnic susceptibilities are related to individual expressions of clinical symptoms of chemical toxicity, to results of biological monitoring of exposed workers, and to the interpretation of results of epidemiological or molecular-epidemiological studies.

Renal cell cancer risk and occupational exposure to trichloroethylene: results of a consecutive case-control study in Arnsberg, Germany

BACKGROUND

German studies of high exposure prevalence have been debated on the renal carcinogenicity of trichloroethylene (TRI).

METHODS

A consecutive hospital-based case-control study with 134 renal cell cancer (RCC) cases and 401 controls was conducted to reevaluate the risk of TRI in this region which were estimated in a previous study. Exposure was self-assessed to compare these studies. Additionally, the job history was analyzed, using expert-based exposure information.

RESULTS

The logistic regression results, adjusted for age, gender, and smoking, confirmed a TRI-related RCC risk in this region. Using the database CAREX for a comparison of industries with and without TRI exposure, a significant excess risk was estimated for the longest held job in TRI-exposing industries (odds ratio (OR) 1.80, 95% confidence interval (CI) 1.01-3.20). Any exposure in "metal degreasing" was a RCC risk factor (OR 5.57, 95% CI 2.33-13.32). Self-reported narcotic symptoms, indicative of peak exposures, were associated with an excess risk (OR 3.71, 95% CI 1.80-7.54).

CONCLUSIONS

The study supports the human nephrocarcinogenicity of trichloroethylene.

Association between head and neck cancer and microsomal epoxide hydrolase genotypes

Tobacco-associated carcinogens are catalyzed by microsomal epoxide hydrolase (mEH). Combinations of the Y113H and H139R polymorphic EPHX1 variants have been assumed to alter the enzyme activity and thus the risk of squamous cell head and neck cancer (SCCHN). Based on in vitro data, a putative low, medium and high mEH activity has been associated with combinations of these genotypes, and the respective activity categories have been frequently used in the estimation of risks for smoking-related cancers. We investigated the SCCHN risk for EPHX1 genotypes among 280 cases and 289 controls. We could not detect main effects of the EPHX1 genotypes, but a smaller risk of the 139HR genotype in smokers (odds ratio, OR, 0.57; 95% confidence interval, CI, 0.34-0.95). We could not confirm an increase of the SCCHN risk for genotype combinations according to a putative medium and high enzyme activity (OR 1.28, 95% CI 0.84-1.96; OR 0.98, 95% CI 0.58-1.64, respectively), but a significant heterogeneity of the estimated risks for the singular genotypes within these categories among smokers ( P=0.02). Further, p53 mutations among smoking cases were less frequent in the group with a putative high enzyme activity, although insignificant due to small numbers (OR 0.54, 95% CI 0.13-2.17). This supports uncertainties in categorizing genotypes with respect to limited enzyme activity data, especially when taken from in vitro experiments.

Children as a sensitive subgroup and their role in regulatory toxicology: DGPT workshop report

There is increasing discussion that children might be considered as a specific subgroup in public health regulations which could be more sensitive than the average "adult" human being. Differences between children and adults, with regard to susceptibility towards toxicants, may result from a combination of toxicokinetic, toxicodynamic and exposure factors. Kinetic factors are of importance mainly in the early postnatal period, largely as the result of immature elimination systems, i.e. metabolising enzymes and/or renal function. Specific vulnerability may prevail during several time periods, related to the development and maturation of organs (for example, brain, bone, endocrine system). For some substances, it has been shown that children at a specific age are less sensitive than adults. Specific exposures of toddlers to environmental chemicals may be high due to their moving behaviour and hand-to-mouth activities. Existing scenarios and models for exposure of children should be improved, in particular with respect to different ages. The outcome of model calculations must be verified by human biomonitoring analysis. At present, there is ongoing discussion of toxicological test models suitable to delineate human postnatal development. Experience with infant-orientated test systems is scarce (for example in developmental neurotoxicity). In general, tools for predicting toxicological sensitivity of children must be further improved. Regulators should also be aware that reduction of lifestyle-related toxic exposures such as smoking and drug abuse in children and adolescents is now an increasing public health problem in many countries.

Nephrotoxicity and nephrocarcinogenicity of dinitrotoluene: new aspects to be considered

Technical dinitrotoluene (DNT) is a mixture of 2,4- and 2,6-DNT. In humans, industrial or environmental exposure can occur orally, by inhalation, or by skin contact. The classification of DNT as an 'animal carcinogen' is based on the formation of malignant tumors in kidneys, liver, and mammary glands of rats and mice. Clear signs of toxic nephropathy were found in rats dosed with DNT, and the concept was derived of an interrelation between renal toxicity and carcinogenicity. Recent data point to the carcinogenicity of DNT on the urinary tract of exposed humans. Between 1984 and 1997, 6 cases of urothelial cancer and 14 cases of renal cell cancer were diagnosed in a group of 500 underground mining workers in the copper mining industry of the former GDR and having high exposures to explosives containing technical DNT. The incidences of both urothelial and renal cell tumors in this group were 4.5 and 14.3 times higher, respectively, than anticipated on the basis of the cancer registers of the GDR. The genotyping of all identified tumor patients for the polymorphic enzymes NAT2, GSTM1, and GSTT1 identified the urothelial tumor cases as exclusively 'slow acetylates'. A group of 161 miners highly exposed to DNT was investigated for signs of subclinical renal damage. The exposures were categorized semi-quantitatively into 'low', 'medium', 'high', and 'very high'. A straight dose-dependence of the excretion of urinary biomarker proteins with the ranking of exposure was seen. Biomarker excretion (alpha1-microglobulin, glutathione S-transferases alpha and pi) indicated that DNT-induced damage was directed toward the tubular system. New data on DNT-exposed humans appear consistent with the concept of cancer initiation by DNT isomers and the subsequent promotion of renal carcinogenesis by selective damage to the proximal tubule. The differential pathways of metabolic activation of DNT appear to apply to the proximal tubule of the kidney and to the urothelium of the renal pelvis and lower urinary tract as target tissues of carcinogenicity.

Cytochrome P450 1B1, a new keystone in gene-environment interactions related to human head and neck cancer?

Alcohol consumption and tobacco smoking are major causes of head and neck cancers, and regional differences point to the importance of research into gene-environment interactions. Much interest has been focused on polymorphisms of CYP1A1 and of GSTM1 and GSTT1, but a number of studies have not demonstrated significant effects. This has mostly been ascribed to small sample sizes. In general, the impact of polymorphisms of metabolic enzymes appears inconsistent, with some reports of weak-to-moderate associations, and with others of no elevation of risks. The classical cytochrome P450 isoenzyme considered for metabolic activation of polycyclic aromatic hydrocarbons (PAH) is CYP1A1. A new member of the CYP1 family, CYP1B1, was cloned in 1994, currently representing the only member of the CYP1B subfamily. A number of single nucleotide polymorphisms of the CYP1B1 gene have been reported. The amino acid substitutions Val432Leu ( CYP1B1*3) and Asn453Ser ( CYP1B1*4), located in the heme binding domain of CYP1B1, appear as likely candidates to be linked with biological effects. CYP1B1 activates a wide range of PAH, aromatic and heterocyclic amines. Very recently, the CYP1B1 codon 432 polymorphism ( CYP1B1*3) has been identified as a susceptibility factor in smoking-related head-and-neck squamous cell cancer. The impact of this polymorphic variant of CYP1B1 on cancer risk was also reflected by an association with the frequency of somatic mutations of the p53 gene. Combined genotype analysis of CYP1B1 and the glutathione transferases GSTM1 or GSTT1 has pointed to interactive effects. This provides new molecular evidence that tobacco smoke-specific compounds relevant to head and neck carcinogenesis are metabolically activated through CYP1B1 and is consistent with a major pathogenetic relevance of PAH as ingredients of tobacco smoke.

Comparative assessment of endocrine modulators with oestrogenic activity. II. Persistent organochlorine pollutants

Risk assessments of synthetic chemicals with oestrogen-like activity must take into account the high dietary levels of natural endocrine modulators in food. In view of current regulations of the European Union, a hygiene-based margin of safety (HBMOS) for xeno-oestrogens was defined as a quotient of estimated human daily intakes weighted by relative rodent in vivo potencies of the compounds. Such comparisons of intakes and potencies of natural isoflavones, with short half-lives, with those of polychlorinated organic pollutants (POP) displaying significant toxicokinetic accumulation, deserves the special consideration of toxicokinetics. For slowly accumulating compounds such comparison is much more favourable when based on comparative blood and tissue levels, not on scenarios of daily exposures. Observing these principles, the present communication extends the HBMOS concept to POP, using o,p'-DDT, the oestrogenic component of DDT mixtures, as a prototype. An HBMOS of 137 is derived for o,p'-DDT indicative of a sufficient margin of safety to ensure the absence of risk to human health due to its hormonal action, under exposure conditions now prevailing in Western countries.

Is multiple chemical sensitivity a clinically defined entity?

In 1996 a WHO/IPCS Workshop has suggested to use as an appropriate descriptor of MCS the broader term "Idiopathic Environmental Intolerances (IEI)", in order to incorporate "a number of disorders sharing similar symptomatologies". Research was strongly encouraged. The following points have been put forward as a precondition to define MCS as a clinical entity: (a) establishment of diagnostic criteria, (b) identification of pathogenic mechanisms, together with, (c) an explanation of relationship between exposures and symptoms. Against this background, progress made in the fields of sensory physiology and neurobehaviour research must be debated. In particular, recent results on processing of cognitive stimuli have to be considered. IEI/MCS patients exhibited differences vs. controls in their reactions to intranasal challenge, consistent with changes in cognitive processing of suprathreshold chemosensory information. Trait anxiety and focus of attention have clearly been identified as major components in eliciting neurobehavioural MCS symptoms. Hence, the question as to whether MCS should be regarded as a clinically defined entity remains controversial, but important progress can be noticed in elucidating and defining the nature of this phenomenon, by a combined effort of several disciplines (toxicology and behavioural toxicology, psychology and psychophysiology, and clinical medicine). The new situation will call for a re-evaluation of traditional positions.

Possible impact of human CYP2E1 polymorphisms on the metabolism of acrylonitrile

Case reports of human accidental poisonings point to significant individual differences in human acrylonitrile metabolism and toxicity. A cohort of 59 persons with industrial handling of low levels of acrylonitrile has repetitively been studied from 1994 through 1999 as part of a medical surveillance programme. The analyses included adduct determinations of N-terminal N-(cyanoethyl)valine in haemoglobin and genotypings of the following cytochrome P-450 2E1 (CYP2E1) polymorphisms: G-1259C and C-1019T (two subjects heterozygous), A-316G (three subjects heterozygous), T-297A (15 subjects heterozygous), G-35T (eight subjects heterozygous), G4804A (two subjects heterozygous), T7668A (six subjects heterozygous). N-(Cyanoethyl)valine adduct levels were, if any, only slightly influenced by smoking and mainly determined by the external acrylonitrile exposures. The individual means and medians of N-(cyanoethyl)valine levels over the entire observation period were compared with the CYP2E1 variants (Wilcoxon rank sum test). No influences of the investigated CYP2E1 polymorphisms on the N-(cyanoethyl)valine levels appeared at the 5% level. However, there was a trend, at a level of P approximately 0.1, pointing to higher acrylonitrile-specific adduct levels in persons with the A-316G mutation. Higher adduct levels would be compatible with a slower CYP2E1-mediated metabolism of acrylonitrile and with lower extents of toxification to cyanide.

The enhanced bladder cancer susceptibility of NAT2 slow acetylators towards aromatic amines: a review considering ethnic differences

Human bladder cancer may be caused by exposure to aromatic amines. The polymorphic enzyme N-acetyltransferase 2 (NAT2) is involved in the metabolism of these compounds. Two classical studies on chemical workers in Europe, exposed in the past to aromatic amines like benzidine, unambiguously showed that the slow acetylator status is a genetic risk factor for arylamine-induced bladder cancer. In the former benzidine industry in Huddington, Great Britain, 22 of 23 exposed cases with bladder cancer, but only 57% of 95 local controls without bladder cancer were of the slow acetylator phenotype. In Leverkusen, Germany, 82% of 92 benzidine-exposed chemical workers with bladder cancer were of the slow acetylator phenotype, whereas only 48% of 331 chemical workers who had worked at that plant were of the slow acetylator phenotype. This is in line with several smaller studies, which also show an over-representation of the slow acetylator status in formerly arylamine-exposed subjects with bladder cancer. Some of these studies included also subjects that were exposed to aromatic amines by having applied dyes, paints and varnishes. These European findings are in contrast to a large study on Chinese workers occupationally exposed to aromatic amines. In this study, only five of 38 bladder cancer cases occupationally exposed to arylamines were of the slow acetylator genotype. This is much lower than the ratio of slow acetylators to the general population in China. This points to different mechanisms of susceptibility for bladder cancer upon exposure to aromatic amines between European (Caucasian) and Chinese populations.

Estrogenic isoflavones in rodent diets

Many rodent diets contain components such as soy isoflavones (daidzein and genistein) known to have estrogenic properties. The dietary background of phytoestrogens may modulate some responses to environmental estrogens when these compounds are tested in rodent bioassays. Thus, and since only few data were available on the phytoestrogen content of rodent diets commonly used in European laboratories, it was of interest to analyze the daidzein and genistein contents of our standard animal feeds. Isoflavone contents were determined in seven batches of rodent chow (from two suppliers in Germany, Altromin and Ssniff) by high-performance liquid chromatography, and also analyzed in six rodent diets from the United States. The soy-based rodent diets from Germany contained isoflavone (daidzein plus genistein) concentrations in the range of 0.3-0.55 mg/g feed. These isoflavone contents are similar to those analyzed in the US rodent diets, and similar to values reported by others, including one particular lot of feed (with 0.35 mg isoflavones per g) which produced a large uterotrophic response in immature ovariectomized rats [Environ. Health Perspect., 106 (1998) 369]. Coumestrol was found in a sample of commercial rabbit food at rather high levels (0.27 mg/g), but, this phytoestrogen was not detected (<1 microg/g feed) in any of the other samples we analyzed. The soy components in our rodent diet produce a measurable background of daidzein and genistein in blood of female DA/Han rats, total isoflavones (aglycone plus conjugates) ranging between 90 and 290 ng/ml plasma. The ovariectomized animals kept on this chow, showed no signs of estrogenization of the reproductive tract (uterus, vagina), and responded normally to (xeno-)estrogen administration in a uterotrophic assay [J. Steroid Biochem. Mol. Biol., 73 (2000) 1]. Moreover, ovariectomized Wistar rats on our standard rodent diet (Ssniff R/M H) had lower uterine weights than animals kept on the isoflavone-free (solvent extracted) chow; both groups of rats responded to genistein administration with an increase in uterine weights. These results suggest that--albeit the sensitivity of the rodent uterotrophic assay is not reduced by the use of a diet containing soy isoflavones at commonly encountered levels--attention should be given to a variable dietary phytoestrogen background.

Genetic susceptibility to environmental toxicants: the interface between human and experimental studies in the development of new toxicological concepts

The growing knowledge of the genetic polymorphisms of enzymes metabolising xenobiotics in humans and their connections with individual susceptibility towards toxicants has created new and important interfaces between human epidemiology and experimental toxicology. The results of molecular epidemiological studies may provide new hypotheses and concepts, which call for experimental verification, and experimental concepts may obtain further proof by molecular epidemiological studies. If applied diligently, these possibilities may be combined to lead to new strategies of human-oriented toxicological research. This overview will present some outstanding examples for such strategies taken from the practically very important field of occupational toxicology. The main focus is placed on the effects of enzyme polymorphisms of the xenobiotic metabolism in association with the induction of bladder cancer and renal cell cancer after exposure to occupational chemicals. Also, smoking and induction of head and neck squamous cell cancer are considered.

Occupational versus environmental and lifestyle exposures of children and adolescents in the European Union

There is recent international concern about specific exposures of children and adolescents to toxicants. In general, the situation within the European Union appears as follows. (i) OCCUPATIONAL EXPOSURE: Due to regulatory measures, there are almost no toxicologically significant occupational exposure situations of children to chemical toxicants. This contrasts to the situation in developing countries. There is also strict regulation of occupational exposure of adolescents (aged under 18). In consequence, the number of potentially exposed adolescents has been minimised. (ii) ENVIRONMENTAL EXPOSURE: Specific concern is directed towards exposures of infants, especially to neurotoxic heavy metals and carcinogens, and there is much regional differentiation of environmental exposures. (iii) FOOD: Recent research results are indicative of the general progress made in the field of food safety. (iv) INCIDENTAL ACUTE EXPOSURE: Besides drugs, household chemicals are a source of incidental acute intoxications in children. In Germany, there has been a particular focus on ingestion of lamp petroleum oils since 1989. (v)

LIFESTYLE

Paramount problems are associated with increasing consumption of tobacco (mean age of starting smoking in Germany: 13.6 years), alcohol (percentage of addicts at ages 12-24 in Germany 6%) and cannabis among adolescents, calling for new ways of risk communication. In general, it will be necessary to consider children of different ages as separate risk groups.

Integration of mechanistic data in the toxicological evaluation of endocrine modulators

Controversy has arisen concerning the likelihood of adverse health effects due to exposure to hormonally active agents or endocrine modulators such as environmental estrogens. With the aim to improve the basis for their toxicological evaluation, several chemicals of anthropogenic (bisphenol A, octylphenol, o,p'-DDT) and of natural origin (daidzein, genistein) were investigated with regard to their mode of hormonal action and potency as well as toxicokinetics. Experimental toxicodynamic and toxicokinetic data illustrate important points in a comparative assessment of environmental estrogens. A novel concept, the Hygiene-Based Margine of Safety (HBMOS), has been suggested to characterize the relative impact of these potential endocrine modulators on human health: It integrates exposure scenarios (i.a. those generated within the European Existing Chemicals Programme) and in vivo rodent potency data for xenoestrogens and for dietary phytoestrogens. On the basis of these informations, HBMOS values calculated for the alkylphenol and bisphenol A appear sufficiently high to ensure the absence of a practical risk to human health under the present exposure conditions. For slowly accumulating compounds (e.g. DDT) with much longer half-lifes than isoflavones, such comparison should be based on comparative blood levels rather than on scenarios of daily exposures.

Transplacental transfer of the phytoestrogen daidzein in DA/Han rats

Disposition and transplacental transfer of the phytoestrogen daidzein was studied in pregnant DA/Han rats on day 18 of gestation. Daidzein concentrations were determined by HPLC in maternal blood, maternal organs (liver, kidney, uterus), placenta and fetuses (liver and residual tissues) at specific times (5, 10, 20, 40 and 120 min) after intravenous administration of 10 mg/kg body weight. Early after injection, the majority of circulating daidzein was still in the aglycone form; at later time points the majority consisted of conjugates. The initially high isoflavone concentration in maternal plasma (about 25 microg/ml at 5 min) decreased rapidly within the first hour, and after 2 h total daidzein was below 1 microg/ml. Despite its efficient conjugation, daidzein was rapidly distributed in the organism: peak concentrations were attained 10 min after intravenous administration in all tissues analysed, with mean values of about 31 microg/g in maternal liver, 13 microg/g in kidneys and 5 microg/g in the uterus. Placenta contained about one-tenth the hepatic daidzein concentration, and fetal liver about 1/30 the peak concentration of maternal liver (i.e. 1.3 microg/g, which is one-third the placental concentration). Daidzein levels in tissues then declined in parallel with those in maternal blood. The data show that daidzein is transferred across the placenta of DA/Han rats to fetuses. This is indicative of a rapid transfer from the mother to the fetus, but also that efficient hepatic extraction of daidzein from the maternal blood occurs. Since dietary phytoestrogens account for a significant proportion of human exposure to potential endocrine modulators, and since the placenta does not represent a barrier to daidzein or related estrogenic isoflavones, the consequences of these exposures early in life should be examined and monitored carefully.