Using liquid chromatography–tandem mass spectrometry to quantify monohydroxylated metabolites of polycyclic aromatic hydrocarbons in urine

https://doi.org/10.1016/j.jchromb.2009.02.067Get rights and content

Abstract

We present an assay which employs enzyme digestion and solid phase extraction followed by liquid chromatography–tandem mass spectrometry to simultaneously quantify 16 hydroxylated polycyclic aromatic hydrocarbons (OHPAHs) in 3-ml samples of urine. The analytes consisted of 2-, 3-, and 4-ring OHPAHs, namely, 1- and 2-hydroxynaphthalene (1- and 2-OHNAP), 2-hydroxyfluorine (2-OHFLU), 1-, 2-, 3-, 4-, and 9-hydroxyphenanthrene (1-, 2-, 3-, 4-, and 9-OHPHE), 1-hydroxypyrene (1-OHPYR), 1- and 2-hydroxybenzo(a)anthracene (1- and 2-OHBAA), 3- and 6-hydroxychrysene (3- and 6-OHCHR) and 3-, 7-, and 9-hydroxybenzo(a)pyrene (3-, 7-, and 9-OHBAP). The method was validated using urine samples from steel workers and control subjects. The coefficients of variation of the method for the particular analytes were between 7% and 27% and the limits of quantitation were between 0.002 and 0.010 μg/l urine. The 2- and 3-ring OHPAHs were easily quantified in all subjects. However, 1-OHPYR was the only representative of the 4- and 5-ring metabolites that could be quantified. Pairwise correlations showed that all OHPAHs were highly correlated with each other (0.553  r  0.910) and with 1-OHPYR (0.614  r  0.910), the metabolite most widely accepted as a short-term biomarker of exposure to PAHs. The analyte, 2-OHNAP exhibited the lowest pairwise correlations with the other OHPAHs (0.542  r  0.628), presumably due to confounding by smoking. Metabolites of phenanthrene, an abundant PAH and the smallest to possess a bay region, are promising OHPAHs for characterizing both exposures to PAHs and the various metabolic pathways.

Introduction

Polycyclic aromatic hydrocarbons (PAHs) are organic compounds containing two or more fused aromatic rings that are formed by incomplete combustion of organic materials. Environmental PAH exposure occurs primarily through the inhalation of cigarette smoke and automobile exhaust and through consumption of smoked and barbecued foods [1], [2]. Occupational exposures can occur from the use of PAH-containing products, such as asphalt, and during the production of aluminum, iron, and steel [3], [4], [5]. Human exposures to PAHs have been associated with cancers of the skin, bladder, and lung [6], [7] as well as with adverse reproductive outcomes [8], [9], [10].

The scheme shown in Fig. 1 uses phenanthrene as a prototype PAH to illustrate the main metabolic pathways. The following metabolism is proposed based on the available in vitro and in vivo data. The first step involves the formation of epoxide(s) via cytochrome p450 (CYP) enzymes, notably CYP1A1, 1A2, 1B1 and 3A4 [11], [12], [13]. The epoxide(s) can spontaneously rearrange to phenols, conjugate with glutathione via glutathione-S-transferases, form dihydrodiols via the action of epoxide hydrolase [14], [15], [16], or react with macromolecules to form adducts [17], [18]. A given epoxide can produce more than one isomer of a phenolic or hydroxylated metabolite, which we will refer here to as a hydroxylated polycyclic aromatic hydrocarbon (OHPAH). It has been postulated that OHPAHs can also arise from direct hydroxylation of the parent PAH by the action of CYP enzymes [19]. The major route of elimination of glutathione conjugates and low molecular weight OHPAHs (both free and conjugated) is through urine. Dihydrodiols can be conjugated and eliminated in urine, converted to catechols via dihydrodiol dehydrogenase and oxidized to diolepoxides via the action of CYPs [20], [21]. Catechols can be conjugated and eliminated in urine or oxidized to the corresponding ortho-quinones which can react with cellular macromolecules to form adducts [22]. Redox cycling between PAH diols and quinones can also generate reactive oxygen species [23]. Diolepoxides can be converted to the corresponding tetrols via the action of epoxide hydrolase [24]. Diolepoxides formed in the bay region of PAHs, such as those of phenanthrene and benzo(a)pyrene, are resistant to hydrolysis by epoxide hydrolase due to steric hindrance. This resistance to hydrolysis may explain the potent carcinogenicity of benzo(a)pyrene-7,8-diolepoxide and diolepoxides of certain other PAHs that contain bay regions [25], [26].

Emissions of PAHs from a given source are distributed between the gas phase (compounds such as naphthalene, phenanthrene and fluorene) and the particulate phase [compounds such as pyrene, chrysene, benz(a)anthracene and benzo(a)pyrene (BAP)]. These mixed-phase emissions coupled with the possibility of multiple exposure routes from air, water, and food have complicated our ability to monitor human exposures to PAHs. Consequently, biomarkers have been widely used to characterize PAH exposures, with urinary OHPAHs receiving considerable attention [27], [28], [29].

Although 1-hydroxypyrene (1-OHPYR) has been used widely as a urinary biomarker of PAH exposure, it gives little insight into the overall metabolism of PAHs. More importantly, pyrene itself is not carcinogenic and does not possess a bay region, a feature closely associated with carcinogenicity. Hence, investigators have turned their attention to biomarkers of BAP to more accurately assess the exposure and metabolism of carcinogenic PAHs [27], [30], [31], [32]. However, due to the low abundance of particle-bound BAP in air and the fact that BAP metabolites are predominantly eliminated in feces via biliary excretion, investigators have been unsuccessful in measuring BAP-related OHPAHs in urine. In recent years, hydroxylated metabolites of abundant gas-phase PAHs such as naphthalene (OHNAP) and phenanthrene (OHPHE) have gained attention as biomarkers of PAH exposure [33], [34], [35].

Assays have been developed using gas chromatography–mass spectrometry or high performance liquid chromatography with fluorescence detection to simultaneously measure OHPAHs derived from multiple PAHs [36], [37], [38], [39], [40], [41]. More recently, liquid chromatography–tandem mass spectrometry (LC–MS/MS) has been introduced to eliminate the need for derivatization of OHPAHs for gas chromatography and to overcome the lack of specificity in fluorescence detection-based methods [42]. However, the application of LC–MS/MS technique in biomonitoring has not been demonstrated. Here, we describe a LC–MS/MS method to simultaneously detect 16 OHPAHs representing 2–5 ring PAHs, namely, naphthalene (1- and 2-OHNAP), fluorene (2-OHFLU), phenanthrene (1-, 2-, 3-, 4-, and 9-OHPHE), pyrene (1-OHPYR), benz(a)anthracene (1- and 2-OHBAA), chrysene (3- and 6-OHCHR) and BAP (3-, 7-, and 9-OHBAP). The method is validated using urine samples from steel workers and control subjects.

Section snippets

Chemicals and supplies

1-OHNAP, [2H8]1-hydroxynaphthalene ([2H8]1-OHNAP) (98% isotope purity), 2-OHNAP, 2-OHFLU, and 1-OHPYR were obtained from Aldrich Chemical Company (Milwaukee, WI). 3-, 4-, and 9-OHPHE, were purchased from Promochem (Wesel, Germany), while 2-OHPHE, 3-OHCHR, [13C6] 3-hydroxychrysene ([13C6]3-OHCHR), 6-OHCHR, 1-OHBAA, [13C6]1-hydroxybenz(a)anthracene ([13C6]1-OHBAA), 2-OHBAA, 3-OHBAP, 7-OHBAP, and 9-OHBAP were purchased from Midwest Research Institute (Kansas City, MO). [13C6]3-hydroxyphenanthrene

Recoveries, limits of quantitation, precision, and linearity

Although we were able to optimize the chromatographic separations for most of the isomeric and isobaric compounds, we were unable to achieve base-line separation for pairs of 2- and 3-OHPHE, 2-OHBAA and 6-OHCHR, and 3- and 7-OHBAP; these pairs of analytes were quantified together. Table 1 lists the OHPAHs and the corresponding LC–MS/MS parameters used for quantitation. Representative chromatograms obtained from urine spiked with the OHPAHs and carried through the assay are presented in Fig. 2A

Discussion

We present a straightforward assay to simultaneously measure 16 OHPAHs in urine, representing 2-, 3-, 4- and 5-ring PAHs. Our assay is more sensitive and precise than those reported for the earlier GC–HR/MS [39], [44], and LC–MS/MS SRM [42] methods. In fact, this assay compares favorably to that reported by Li et al., which uses a fully automated system for liquid extraction of urine followed by GC–HR/MS [36]. Although our method lacks the luxury of full automation, it is simple to perform and

Conclusion

We developed a straightforward LC–MS/MS assay to measure OHPAHs in 3-ml samples of urine. The assay precision ranged from 5% to 27% and the limits of quantitation ranged from 0.002 to 0.010 μg/l of urine. We applied this assay to urine samples from steel workers and control subjects to evaluate its suitability for biomonitoring studies. Whereas all of the investigated 2- and 3-ring OHPAHs were readily detected in both steel workers and control subjects, 1-OHPYR was the only 4- or 5-ring OHPAH

Acknowledgments

This work was supported in part by the National Institute of Environmental Health Sciences through grants P42ES05948, P30ES10126. The authors thank Dr. Sungkyoon Kim for his assistance in statistical analyses.

References (52)

  • L. Pyy et al.

    Sci. Total Environ.

    (1997)
  • P.J. van Bladeren et al.

    J. Biol. Chem.

    (1985)
  • R. Wu et al.

    J. Chromatogr. B Analyt. Technol. Biomed. Life Sci.

    (2005)
  • S. Waidyanatha et al.

    Chem. Biol. Interact.

    (2002)
  • G. Boysen et al.

    Mutat. Res.

    (2003)
  • F.J. Jongeneelen

    Ann. Occup. Hyg.

    (2001)
  • C.J. Smith et al.

    J. Chromatogr. B Analyt. Technol. Biomed. Life Sci.

    (2002)
  • H. Kim et al.

    Int. Arch. Occup. Environ. Health

    (2001)
  • J. Grainger et al.

    Environ. Res.

    (2006)
  • Z. Li et al.

    Environ. Res.

    (2008)
  • L. Campo et al.

    J. Chromatogr. B Analyt. Technol. Biomed. Life Sci.

    (2008)
  • J.A. van de Wiel et al.

    Toxicology

    (1993)
  • J. Angerer et al.

    J. Chromatogr.

    (1992)
  • P.J. Lioy et al.

    Toxicol. Ind. Health

    (1990)
  • M.J. Dennis et al.

    Food Addit. Contam.

    (1991)
  • A. Bjoresth et al.

    PAH in Work Atmospheres: Occurrence and Determination

    (1986)
  • G. Grimmer et al.

    Int. Arch. Occup. Environ. Health

    (1993)
  • P. Boffetta et al.

    Cancer Causes Control

    (1997)
  • C. Tremblay et al.

    Am. J. Ind. Med.

    (1995)
  • K.A. Bocskay et al.

    Cancer Epidemiol. Biomarkers Prev.

    (2005)
  • F. Perera et al.

    Cancer Epidemiol Biomarkers Prev.

    (2005)
  • S.A. Lederman et al.

    Environ. Health Perspect.

    (2004)
  • A.R. Buckpitt et al.

    J. Pharmacol. Exp. Ther.

    (1983)
  • D.M. Jerina et al.

    Biochemistry

    (1970)
  • J. Booth et al.

    Biochem. J.

    (1949)
  • E. Boyland et al.

    Biochem. J.

    (1935)
  • Cited by (103)

    View all citing articles on Scopus
    View full text