A multiinstitutional, transitional epidemiologic study was conducted with a worker population in the Czech Republic to evaluate the utility of a continuum of non-disease biological responses as biomarkers of exposure to 1,3-butadiene (BD)* in an industrial setting. The study site included two BD facilities in the Czech Republic. Institutions that collaborated in the study were the University of Vermont (Burlington, Vermont, USA); the Laboratory of Genetic Ecotoxicology (Prague, the Czech Republic); Shell International Chemicals, BV (Amsterdam, The Netherlands); the University of North Carolina at Chapel Hill (Chapel Hill, North Carolina, USA); University of Texas Medical Branch at Galveston (Galveston, Texas, USA); Leiden University (Leiden, The Netherlands); and the Health and Safety Laboratory (Sheffield, United Kingdom). Male volunteer workers (83) participated in the study: 24 were engaged in BD monomer production, 34 in polymerization activities, and 25 plant administrative workers served as unexposed control subjects. The BD concentrations experienced by each exposed worker were measured by personal monitor on approximately ten separate occasions for 8-hour workshifts over a 60-day exposure assessment period before biological samples were collected. Coexposures to styrene, benzene, and toluene were also measured. The administrative control workers were considered to be a homogeneous, unexposed group for whom a series of 28 random BD measurements were taken during the exposure assessment period. Questionnaires were administered in Czech to all participants. At the end of the exposure assessment period, blood and urine samples were collected at the plant; samples were. fractionated, cryopreserved, and kept frozen in Prague until they were shipped to the appropriate laboratories for specific biomarker analysis. The following biomarkers were analyzed: * polymorphisms in genes involved in BD metabolism (Prague and Burlington); * urinary concentrations of 1-hydroxy-2-(N-acetylcysteinyl)-3-butene and 2-hydroxy-1-(N-acetylcysteinyl)-3-butene (M2 [refers to an isomeric mixture of both forms]) (Amsterdam); * urinary concentrations of 1,2-dihydroxy-4-(N-acetylcysteinyl)-butane (M1) (Amsterdam); * concentrations of the hemoglobin (Hb) adducts N-(1-[hydroxymethyl]-2-propenyl)valine and N-(2-hydroxy-3-butenyl)valine (HBVal [refers to an isomeric mixture of both forms]) (Amsterdam); * concentrations of the Hb adduct N-(2,3,4-trihydroxybutyl)valine (THBVal) (Chapel Hill); * T cell mutations in the hypoxanthine phosphoribosyltransferase (HPRT) gene (autoradiographic assay in Galveston with slide review in Burlington; cloning assay in Leiden with mutational spectra determined in Burlington); and * chromosomal aberrations by the conventional method and by fluorescence in situ hybridization [FISH]), and cytogenetic changes (sister chromatid exchanges [SCEs] (Prague). All assay analysts were blinded to worker and sample identity and remained so until all work in that laboratory had been completed and reported. Assay results were sent to the Biometry Facility in Burlington for statistical analyses. Analysis of questionnaire data revealed that the three exposure groups were balanced with respect to age and years of residence in the district, but the control group had significantly more education than the other two groups and included fewer smokers. Group average BD exposures were 0.023 mg/m3 (0.010 ppm) for the control group, 0.642 mg/m3 (0.290 ppm) for the monomer group, and 1.794 mg/m3 (0.812 ppm) for the polymer group; exposure levels showed considerable variability between and within individuals. Styrene exposures were significantly higher in the polymer group than in the other two groups. We found no statistically significant differences in the distributions of metabolic genotypes over the three exposure groups; genotype frequencies were consistent with those previously reported for this ethnic and national population. Although some specific genotypes were associated with quantitative differences in urinary metabolite concentrations or Hb adduct dose-response characteristics, none indicated a heightened susceptibility to BD. Concentrations of both the M2 and M1 urinary metabolites and both the HBVal and THBVal Hb adducts were significantly correlated with group and individual mean BD exposure levels; the Hb adducts were more strongly correlated than the urinary metabolites. By contrast, no significant relations were observed between BD exposures and HPRT gene mutations (whether determined by the auto-radiographic or the cloning method) or any of the cytogenetic biomarkers (whether determined by the conventional method or FISH analysis). Neither the mutational nor the cytogenetic responses showed any association with genotypes. The molecular spectrum of HPRT mutations in BD-exposed workers showed a high frequency of deletions; but the same result was found in the unexposed control subjects, which suggests that these were not due to BD exposure. This lack of association between BD exposures and genetic effects persisted even when control subjects were excluded from the analyses or when we conducted regression analyses of individual workers exposed to different levels of BD.

Genetic Toxicology Laboratory University of Vermont 655 Spear Street Building C Burlington VT 05405 USA