Analysis of Butadiene Urinary Metabolites by Liquid Chromatography-Triple Quadrupole Mass Spectrometry

Journal of Analytical Toxicology, Vol. 28, April 2004 Analysisof Butadiene Urinary Metabolites by Liquid Chromatography-TripleQuadrupole Mass Spectrometry Jacob D. McDonald*, William E. Bechtold, Jennifer R. Krone, Walter B. Blackwell, Dean A. Kracko, and Rogene F. Henderson Love~ace Respiratory Research Institute, 2425 Ridgecrest Drive SE, Albuquerque, New Mexico 87108 Abstract 1,3-Butadiene (BD) is a monomer produced in petrochemical production facilities and from several combustionsources. The United States Environmental Protection Agency has defined BD as a probable human carcinogen. Methods for assessing exposure and internal dose are therefore of critical interest, and one technique is the measurementof urinary metabolites. Here we describe methods for measuringtwo urinary metabolites, N-acetyI-S-(3,4-dihydroxybutyl)-c-cysteine(referred to as MI) and an isomeric mixture of the regio- and stereoisomers (R)l(5)-N-acetyI-S-(1-(hydroxymethyl)-2-propen-yl)-t-cysteine and (R)/(S)-N-acetyI-S-(2-hydroxy-3-butenyl)-L-cysteine (referred to as MII). The method is based on isolation of the metabolites by solid-phaseextraction and measurement using liquid chromatography and triple quadrupole massspectrometry (LC-MS3).The LC-MS3 allowed good selectivity with minimal sample preparation. Assayaccuracy was within 10% or better, with substantialimprovement in accuracy accompanyingthe commercial availability of deuterated internal standardsfor both compounds. Assayprecision and linearity passedrigorous validation criteria, and precision-basedlimits of quantitation values were 12 and 1 ng/mL for MI and MII, respectively.Data are shown from analysisof human urine from occupationally exposed individuals and rat urine from BD exposuresconducted to investigaterodent metabolic profiles. Both of these data sets clearly show that this assaycan discern previouslydescribed relationshipsbetween BD exposureand the production of MI/MII. Introduction Chronic inhalation of 1,3-butadiene (BD) causes cancer in rats and mice, with mice being more sensitive than rats (1). The International Agencyfor Research on Cancer has classified BD as category 2A, "Probably carcinogenic to humans" (2). The highest concentrations of human BD exposures occur in occupational settings, as it is a high commodity chemical used in the production of synthetic rubber and various plastics. * Author to whom correspondenceshouldbe addressed.E-mail:. 168 Workers in styrene-BD rubber plants are at increased risk of developing leukemia relative to the general population (3). Lower concentration exposures also occur from cigarette smoke (4), exhaust from stationary combustion of fossil fuels (5), automotive exhaust (6), and wood smoke (7). BD metabolism has been studied extensively in animals and humans. The general route of metabolism, including differences in species (8), has been quantitatively described (8-12). In all species, the first step in BD metabolism is an epoxidation to form butadiene monoxide. The monoepoxidecan be coupled with glutathione and converted to the corresponding butenol mercapturic acid (MII). Alternatively, the epoxide can be hydrolyzed to butenediol, which presumably is oxidizedbefore reaction with glutathione and results in a butenediol mercapturic acid (MI) (3). These two metabolites have been excreted in urine at concentrations that are proportional to BD exposure concentrations (12); therefore, they are useful compounds for biological monitoring of BD exposure (biomarkers). The two metabolites have the additional advantages of reflecting relative activities of epoxide hydrolase and glutathione transferase; thus, species with high activities of epoxide hydrolase form more M1 than MII. In humans, over 99% of the urinary MI + MII mixture is MI (12). This paper describes an assay to measure MI and MII in urine samples using high-pressure liquid chromatograpic separation and triple quadrupole mass spectrometry (LC-MS3) after isolation by solid-phase extraction. This assay is highly selective and less labor intensive than methods requiring more extensive sample preparation such as derivatization. The assay has been implemented in the analysis of BD-exposed humans and rodents in several studies, but this is the first time the method is reported in detail. Our data from the MI and MII analyses were briefly described and compared to a gas chromatographic-MS method requiring compound derivatization in the results reported by Albertini et al. (12). Since that time, the inclusion of deuterated internal standards to improve the accuracy of the assay and instrumental method modifications have heightened the sensitivityapproximately fivefold.Here, the current method is reported, and the assay performance verification demonstrates the accuracy, recovery, selectivity, precision, linearity, Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission. Journal of Analytical Toxicology, Vol. 28, April 2004 and sensitivity. Select data are presented from studies of exposed humans and rodents that show the utility of this method to assess exposure and study metabolic profiles. Materials and Methods Reagentsand standards Analytical-grade ethyl acetate and methanol were acquired from Fisher Scientific. Purified water (PW) was created in house by distillation of deionized water. Potassium chloride, sodium chloride, urea, and hydrochloric acid (Aldrich)were the highest grades available. Nitrogen gas was industrial grade. During the first phase of this method development, the MI standard was acquired from Peter J. Boogaard (Shell Laboratories, The Hague, The Netherlands) and was determined to be greater than 95% pure by HPLC and NMR. The MII standard was synthesized in-house using the procedure described by Sabourin et al. (13). The purity of the MII standard was greater than 95% by HPLC. More recently, standards for MI, MII, and deuterated MI/MIIhave been obtained commercially (Toronto Research Chemicals, Toronto, ON, Canada). Synthetic urine Synthetic urine was created for method verification tests and establishment of calibration standards. Real urine is not used to create standards because MI is produced endogenously, as evidenced by MI in control samples (12); therefore, it is not possible to find "MI-free" human or rodent urine. MII has also been observed in some control subjects, both rodents and humans, as described later in this paper. Synthetic urine was created to represent normal human physiological levels of salt and pH as reported by Tietz (14). The normal physiological mix is described in Table I. This resulted in 0.128M NaC1, 0.06M KC1,0.05M NaH2PO4,0.18M urea, and pH of 5.5--6. As part of verifying the method performance for assay accuracy, the effect of salt content on the assay was tested by looking at 2 and 3 times physiologically normal. These mixes were created by decreasing the dilution of the synthetic mix by factors of 2 or 3. Calibrati (...truncated)