Environmental Toxicology and Pharmacology 30 (2010) 224–244

Contents lists available at ScienceDirect

Environmental Toxicology and Pharmacology

journal homepage: www.elsevier.com/locate/etap

Review

Analytical methods for determining metabolites of polycyclic aromatic

hydrocarbon (PAH) pollutants in fish bile: A review
Jonny Beyer a,b,∗ , Grete Jonsson c , Cinta Porte d , Margaret M. Krahn e,1 , Freek Ariese f
a
IRIS – International Research Institute of Stavanger, N-4068 Stavanger, Norway
b
University of Stavanger, N-4036 Stavanger, Norway
c
Stavanger University Hospital, N-4068 Stavanger, Norway
d
CSIC, Environmental Chemistry Dept IDAEA, ES-08034 Barcelona, Spain
e
NOAA Fisheries, Northwest Fisheries Science Center, Seattle, WA 98112 USA
f
Vrije Universiteit Amsterdam, NL-1081 HV Amsterdam, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: The determination of polycyclic aromatic hydrocarbon (PAH) metabolites in bile can serve as a tool
Received 3 May 2010 for assessing environmental PAH exposure in fish. Biliary PAH metabolite levels can be measured
Accepted 22 August 2010 using several analytical methods, including simple fluorescence assays (fixed fluorescence detection or
Available online 20 September 2010
synchronous fluorescence spectrometry); high-performance liquid chromatography with fluorescence
detection (HPLC-F); gas chromatography–mass spectrometry (GC–MS) after deconjugation, extraction
Keywords:
and derivatization of the bile sample, and finally by advanced liquid chromatography-tandem mass spec-
PAH pollution
trometry (LC–MS/MS) and gas chromatography-tandem mass spectrometry (GC–MS/MS) methods. The
Fish
Environmental monitoring
method alternatives are highly different both with regard to their analytical performance towards differ-
Bile metabolites ent PAH metabolite structures as well as in general technical demands and their suitability for different
Biomarkers of exposure monitoring strategies. In the present review, the state-of-the-art for these different analytical methods
is presented and the advantages and limitations of each approach as well as aspects related to analytical
quality control and inter-laboratory comparability of data and availability of certified reference materials
are discussed.
© 2010 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
2. Methods used to analyze for metabolites of PAHs in fish bile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
2.1. Semiquantitative fluorometric screening of PAH metabolite mixtures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
2.1.1. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
2.1.2. Fixed wavelength fluorescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
2.1.3. Synchronous fluorescence spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
2.1.4. HPLC-Fluorescence screening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
2.2. Quantitative chromatographic analysis of PAH metabolites in bile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
2.2.1. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
2.2.2. Preparation and hydrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
2.2.3. Extraction of hydrolyzed PAH metabolites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
2.2.4. Derivatization of hydroxy metabolites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
2.2.5. Instrumental determination of individual PAH metabolites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
2.2.6. Other analytical techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
3. Aspects of analytical quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
3.1. General considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
3.2. Sample stability and storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236

∗ Corresponding author at: IRIS, N-4068 Stavanger, Norway. Tel.: +47 51875507; fax: +47 51875530.
E-mail address: jb@iris.no (J. Beyer).
1
Retired; may be reached through gina.ylitalo@noaa.gov.

1382-6689/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.etap.2010.08.004
 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244 225

3.3. Hydrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236

3.4. Derivatization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
3.5. Internal standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
3.6. Interference and matrix effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
3.7. Method validation and quality control schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
4. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240

1. Introduction or more aromatic rings (and bay-region and fjord-region structural

elements).
Polycyclic aromatic hydrocarbons (PAHs) constitute a diverse Identifying the source(s) of PAH contamination in the envi-
class of hydrophobic organic molecules that are ubiquitous envi- ronment is important to mitigate the pollution or to prevent
ronmental contaminants (Harvey, 1997). Because some common further contamination. Chemical fingerprinting of a PAH mixture
PAHs are known to be potent carcinogens, this contaminant class can lead to the specific identification of the contamination source,
is generally regarded as a high priority for environmental pollution for instance when oil is spilled from an oil tanker (Wang et al.,
regulation and in ecological risk assessment of industrial effluent 1999, 2009; Wang and Fingas, 2003). Environmental forensics
discharges. When investigating “stress” in fish related to exposure, plays a critical role in determining the responsibility for cleanup-
the detection of parent PAHs in tissues may lead to an underesti- related environmental claims. Another source of petrogenic PAH
mation of the exposure level. This is due to the fact that PAHs in contamination of marine environments is oil released from off-
fish are metabolically degradable. The enzymatic biotransforma- shore oil installations, both from accidental spills or in “produced”
tion (bioactivation) of PAHs is also essential for their mode of action water effluents (Utvik, 1999; Utvik et al., 1999). In the offshore oil
as genotoxic carcinogens. Bile represents the major excretion route industry, PAHs are recognized as priority pollutants and significant
for biotransformed (hydroxylated and/or conjugated) PAH metabo- resources have been used to minimize the amount of PAHs released
lites. Therefore, the concentration of biliary PAH metabolites can be to sea during normal operational discharges. Smoke fallout from
used as a marker for assessing the ongoing (or recent) PAH exposure maritime transport vessels and effluents and fallout from land-
in fish samples. based industries (such as Söderberg primary aluminum smelters
PAHs are a subset of the broader class of polycyclic aromatic and coke ovens) contribute significantly to pyrogenic PAH contam-
compounds (PACs) that also include the heterocyclic organic com- ination in many coastal aquatic ecosystems. Other important causes
pounds, i.e., compounds that contain at least one element other are the multiple small combustion sources in the urbanized society,
than carbon (such as sulfur, oxygen or nitrogen) within at least the leaching from creosote-preserved wooden harbor structures,
one ring structure. Ecotoxicological issues related to PAHs and diesel contaminated sites, gasworks and tar impregnation facilities
PACs have been investigated in great detail and findings have been (Pies et al., 2008).
reported in thousands of peer-reviewed scientific papers. In these, In water, the more hydrophobic PAHs will typically adsorb
the attention is clearly focused on PAHs which are referred to strongly to particles and become generally more resistant to
approximately three times more often than the PACs in the liter- bacterial degradation (Cerniglia, 1984). Therefore, sediment con-
ature. Anthropogenic sources of PAHs are normally categorized as centrations of PAHs are measured to provide valuable information
being of pyrogenic or petrogenic type; the former involves pro- about local PAH loads. High concentrations of pyrogenic PAH mix-
cesses of incomplete combustion of organic materials and fuels, tures in sediments have been found in several freshwater, estuarine
whereas the latter involves the environmental release of crude oil and marine regions with heavy vessel traffic or at sites that receive
or petroleum products (Neff, 1979, 1985; Robertson, 1998). In addi- PAH-containing industrial effluents. And at such locations, dele-
tion, natural sources of PAHs such as forest fires, natural oil seeps, terious effects on the aquatic biota are sometimes observed, e.g.
volcanic activity and diagenic conversion of precursor biomolecules (Nikolaou et al., 2009).
in sediments, may sometimes contribute to significantly elevated In certain areas the adverse biological effects associated with
PAH levels in the environment (Fernandes et al., 1997; Opuene et high levels of PAH contamination have been studied and described
al., 2007; Tan et al., 1996). in great detail, with the best example probably being the Puget
Freshwater and marine sediments function as the ultimate sink Sound, Washington, USA, e.g. (Myers et al., 2008, 1994). Many
for PAHs and more studies have therefore been carried out on studies have shown that PAHs are toxic to fish and other aquatic
the environmental fate and effects of PAHs in aquatic ecosystems organisms, e.g. (Anderson and Lee, 2006; Bunton, 1996; Buryskova
in comparison to terrestrial ecosystems. PAH contaminants are et al., 2006; Chen and White, 2004; Lee and Anderson, 2005; Myers
always found as complex mixtures. To distinguish between pet- et al., 1994; van der Oost et al., 2003; Wurl and Obbard, 2004;
rogenic and pyrogenic mixtures, the ratio of alkylated PAHs to Aas et al., 2001). PAH toxicity is highly structure dependent, and
non-alkylated PAHs can be used: i.e., extensive alkylation points to even isomers with the same number of rings may vary from non-
a predominantly petrogenic contamination source and less alky- toxic to extremely toxic depending on different steric positioning
lation indicates a predominantly pyrogenic source. Different types of the benzene rings. The toxicity of PAHs is also closely related to
of combustion may yield different distributions of PAHs, both in enzymatic biotransformation, and organisms that have poor bio-
absolute amounts and in isomer ratios, e.g. coal burning, diesel transformation capacity (e.g., blue mussels) are less vulnerable to
combustion and forest fires all produce different mixtures of PAHs. PAH hazards. In contrast, fish can metabolize PAHs to form reactive
What is often observed at PAH-contaminated locations is a mixed metabolites, which subsequently can bond covalently as adducts to
contribution from both petrogenic and pyrogenic sources (Ye et cellular macromolecules such as DNA, RNA, and proteins. Because
al., 2006). Pyrogenic PAH mixtures are generally more toxic than fish do not have a highly developed DNA repair system, this may
petrogenic mixtures, partly because of the higher relative concen- lead to many forms of lesions and adverse conditions in cells and in
tration of larger, non-alkylated PAHs that contain a number of four the organism, including mutagenesis, teratogenesis, and carcino-
226 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244

genesis (Dipple et al., 1984; Tuvikene, 1995). Strong causal links route (Lee et al., 1972). These findings were important and led to
between environmental PAH concentrations and the incidence of follow-up studies (Statham et al., 1976; Varanasi and Gmur, 1981;
liver neoplasms and liver tumors in fish have been demonstrated Varanasi et al., 1985) and the suggestion of using PAH metabolites
(Baumann and Harshbarger, 1995; Malins et al., 1987, 1988; Myers in bile as a means for determining the PAH exposure in fish (Krahn
et al., 1991; Reichert et al., 1998; Varanasi et al., 1987, 1992; et al., 1986). In many subsequent studies, the analysis for metabo-
Vethaak et al., 1996). In other studies, adverse and ecologically lites in fish bile has been used to monitor for environmental PAH
relevant effects of PAH exposures in fish have been observed in exposure, e.g. (Beyer et al., 1996, 1998; Brown and Steinert, 2004;
the form of delayed growth, reduced survival and increased preva- Budzinski et al., 2004; Escartin and Porte, 1999a,b; Hanson and
lence of developmental malformation at early life history stages Larsson, 2009; Hillenweck et al., 2008; Holth et al., 2009; Kammann,
(Colavecchia et al., 2007; Heintz et al., 2000; Incardona et al., 2004). 2007; Krahn et al., 1986; McDonald et al., 1995; Neves et al., 2007;
The actual exposure of aquatic organisms to PAHs does not only Oikari, 1986; van der Oost et al., 1994a; Vuorinen et al., 2006; Aas et
depend on the sediment concentrations, but also on factors such as al., 2001). Studies have shown that PAH metabolite concentrations
solubility of the particular PAH compound, ingestion of suspended in bile (in contrast with parent PAH levels in tissues) are sensitive
matter, and the preferred habitat and food sources of the species measures and correlate well with the exposure to PAHs (Beyer et al.,
in question. The feasibility of using tissue concentrations of PAH 1997; Collier and Varanasi, 1991; Deshpande et al., 2002; Fuentes-
parent compounds in marine species as a marker for the envi- Rios et al., 2005; Johnson et al., 2007; Krahn et al., 1992, 1984; Aas
ronmental PAH exposure depends on the relative rates of uptake, et al., 2001). The suitability of measuring PAH metabolites in bile
biotransformation, and excretion. Invertebrate filter feeders, such as a means of assessing ongoing and recent PAH exposure in fish
as blue mussel species, Mytilus spp, are highly efficient accumula- has gained wide acceptance following publication of the numerous
tors and bioconcentrators of PAHs. High PAH concentrations were papers described previously. However, it is necessary to consider
measured in blue mussels collected at PAH contaminated locations, certain limitations and confounding factors related to the dynamic
such as the Karmsund strait along the west coast of Norway (Aas nature of bile, the technical challenges related to the analytical
et al., 2001). PAH accumulation in mussels is related to the low methods, as well as the complex nature of PAH exposure and the
biotransformation capability and hence limited elimination capac- resulting biliary metabolite mixtures.
ity for these contaminants (James, 1989). Determination of parent The scope of this review paper is to present various analytical
PAHs in mussels is therefore a feasible means for assessing PAH techniques that can be used to determine metabolites of PAHs in
contamination in marine ecosystems. Fish and other vertebrates, fish bile and to evaluate the knowledge gained from these meth-
on the other hand, rapidly biotransform PAHs and as a result the ods for monitoring and assessing environmental PAH exposure.
PAH accumulation levels in routinely monitored tissues (e.g. mus- Emphasis is put on identifying the advantages and limitations of
cle, liver) are generally very low. These low PAH levels in tissues are each analytical technique (e.g. selectivity, detection and quantifica-
not only difficult to measure, but are also sensitive to confounding tion limits). In addition, the use of the PAH bile metabolite approach
factors and tend to correlate very poorly with exposure (van der is reviewed in relation to different environmental PAH exposure
Oost et al., 1994b). For instance, Budzinski et al. found no increase scenarios: (1) petrogenic and pyrogenic PAH mixtures; (2) acciden-
in PAH levels in the muscle of sole (Solea solea) following the Erika tal oil spills vs. industrial effluents; (3) contaminated sediments vs.
oil spill of 1999, even though PAH metabolite levels in the same fish contaminated particles in water; (4) high exposure vs. low expo-
pointed at a strongly increased PAH uptake (Budzinski et al., 2004). sure; and (5) laboratory vs. field exposures. Moreover, we will
Fish and other aquatic vertebrates possess well-developed discuss the quality control of biliary PAH metabolite determina-
enzyme systems that efficiently convert PAHs to epoxides and tions and suggest ways to improve the intercomparability of PAH
hydroxylated derivatives during phase I metabolism, and these bile metabolite data among laboratories by using certified refer-
products are further converted into highly water-soluble conju- ence materials (CRMs). Finally, we intend to provide the researcher
gates (e.g., glucuronides or sulfates) during phase II metabolism to with sufficient information to be able to choose an appropriate
facilitate excretion (Lech and Vodicnik, 1985). In fish, the liver is the method when addressing environmental PAH contamination in fish
major site for phase I and phase II biotransformation of PAHs, and for monitoring and environmental risk assessment purposes. The
the metabolic products are subsequently stored in the bile in the current review publication is an update of a method review paper
gall bladder. The bile is emptied into the alimentary tract after food which in 2005 was published by ICES at the Techniques in Marine
intake to assist with the breakdown of lipids and other compounds Environmental Sciences (TIMES) website, i.e. (Ariese et al., 2005a).
in the gut. Because of the pH conditions in the intestine, conju-
gated PAH metabolites may become hydrolyzed and hence more
hydrophobic again, causing them to be reabsorbed over the gut wall 2. Methods used to analyze for metabolites of PAHs in fish
and transported directly back to the liver via the portal vein. This bile
phenomenon, known as enterohepatic circulation, plays an essen-
tial role in the recirculation of bile acids from the gut for reuse in 2.1. Semiquantitative fluorometric screening of PAH metabolite
the liver. PAHs and other hydrophobic contaminants may act as mixtures
stowaways in the enterohepatic circulation and will thereby stay
longer in the liver-bile biotransformation–excretion system, hav- 2.1.1. Background
ing more time to exert adverse toxic actions within the organism, PAH molecules are strong fluorophores due to the rigid, aro-
before they are finally eliminated. matic structure with delocalized -electrons that are distributed
As early as 1972, Lee et al. described the uptake, metabolism throughout the molecule. When a PAH molecule absorbs a pho-
and excretion of two radiolabelled PAHs (14 C-naphthalene and ton with energy hex , one of its -electrons is brought to a higher
3 H-benzopyrene) in three species of marine fish (sand goby, energy level, and the molecule is said to be promoted from its
Gillichthys mirabilis; sculpin, Oligocottus maculosus; and sand dab, ground state S0 to an excited singlet state S1 or higher. As this
Citharichthys stigmaeus). They concluded that the path of hydro- electron returns from the S1 level to the ground state, a pho-
carbons through the fish included uptake mainly through the ton with lower energy hem (longer wavelength) is emitted and
gills, metabolism by the liver, transfer of hydrocarbons and their this process is known as fluorescence emission (photolumines-
metabolites to the bile (which acted as a major storage site) and cence). The wavelengths at which the PAH molecule can be excited,
finally excretion, which also involved the urine as an elimination and at which it emits fluorescence are characteristic for each
 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244 227

compound and this can be used for identification (Sharma and

Schulman, 1999). In mixtures that contain several PAH structures,
the differences in fluorometric properties of each PAH present
can be exploited to obtain a preliminary identification or even a
semiquantitative measure of the PAHs present, without a prior
chromatographic separation of the mixture into individual compo-
nents. The possible use of synchronized excitation of fluorescence
emission spectra of mixtures to achieve this kind of direct PAH
identification was noted first by Lloyd in the early seventies (Lloyd,
1971). Certain PAHs or classes of compounds could be discerned
in the spectrum of the mixture, even though the spectra partially
overlapped. A few years later, Statham et al. investigated the possi-
bility of applying similar principles for screening the PAH content
in fish bile samples as a means for monitoring waterborne xenobi-
otic contaminants (Statham et al., 1976). Further development of
the fluorometric PAH screening in fish bile was then conducted by
Krahn et al. who described an HPLC-F screening method for BaP,
naphthalene and their metabolites in fish bile samples from pol-
luted waterways (Krahn et al., 1984, 1981). Subsequently, Ariese
et al. described a simple synchronous fluorescence spectrometry
assay of 1-OH pyrene in fish bile for use as a rapid screening
technique to monitor PAH exposure in fish (Ariese et al., 1993a).
Eventually, the fixed wavelength fluorescence (FF) approach as an
even easier means of screening for PAH exposures in fish has been
developed by several groups (Lin et al., 1996; Aas et al., 2000b). In
Fig. 1. Synchronous fluorescence spectrometry of non-hydrolyzed bile from fish
general, these fluorescence screening methods all represent tools exposed to three different PAH treatments and unexposed control. The wavelength
for easy, rapid, and semiquantitative assessment of PAH metabo- interval used in scans is 42 nm. Data obtained from Aas et al. (2000b) with permis-
lites in fish. A summary of the essential technical aspects related sion.
to these methods and a review of these techniques in conjunction
with PAH monitoring and aquatic environmental risk assessment as a first-tier screening in field surveys and monitoring studies,
is provided in the following sections. this rather unspecific signal is often sufficient for the investigator
to distinguish contaminated locations from less contaminated or
2.1.2. Fixed wavelength fluorescence uncontaminated ones. However, the fish bile contains various pig-
FF analysis of diluted fish bile is the most rapid and techni- ments and other constituents that are able to absorb the incoming
cally undemanding measurement that produces meaningful PAH or emitted light (inner filter effect) or causes other matrix effects
exposure data in fish, see (Lin et al., 1996; Yang et al., 2003; Aas such as resonance energy transfer and collisional quenching, which
et al., 2000b). The simplicity of the assay has rendered it popu- may lead to signal loss. In addition, fluorescence calibration curves
lar among investigators, although the assay is not fully selective deviate from linearity at high concentrations. As a rule of thumb, the
or quantitative. Bile from fish exposed to PAH mixtures represents total absorbance of the sample (at the wavelength of interest and
a concentrated multifluorophoric sample, attributed primarily to over a pathlength of half the cuvette size) should not be higher than
the PAH metabolites present. The FF method is based on the fact 0.05. These matrix effect problems of bile can, to a large degree, be
that most PAH metabolites are strong fluorophores, whereas most overcome simply by ensuring a sufficient dilution of the bile prior
other compounds in the bile show little or no fluorescence. Dif- to measurement. A 1000- or 2000-fold dilution with a solvent such
ferent PAHs vary considerably in their wavelength optima for light as 50% methanol or 50% ethanol is most often enough to avoid these
excitation and fluorescence emission depending on their structural problems (Aas et al., 2000b). A simple dilution effect study is rec-
properties, see Fig. 1. In general, small PAH metabolite molecules ommended when optimizing the FF method or when some samples
with only a few fused benzene rings (e.g., 2-ring naphthalene or in a set are much more concentrated (dark in color) than the rest. In
3-ring phenanthrene-derived metabolites) require greater excita- dense bile samples with very high PAH metabolite concentrations,
tion energy (i.e. excitation light at shorter wavelength) than do a very high dilution factor (>20,000 times) may be needed to keep
larger PAHs (e.g., 5-ring BaP metabolites). The bile fluorescence the detector signal on scale or within the linear range of the assay
detected at specific excitation/emission wavelength pairs (EEWP) (Lin et al., 1996).
may thus indicate the presence and even the (semi)quantity of The FF procedure is described in (Lin et al., 1996; Aas et al.,
the PAH metabolites of various ring sizes in the analyzed sample. 2000b) and the total assay is briefly described as follows. Bile is
By measuring the fluorescence of hydroxy PAH standard solutions obtained from the fish specimen, mixed well (i.e. mildly sonicated,
and known PAH mixtures in bile at different wavelength pairs, the especially important if samples are taken from the freezer) and cen-
occurrence of various PAH metabolites (or classes of these) in a trifuged. After appropriate dilution in a 50% methanol solvent, the
biliary matrix can be identified. diluted bile sample is analyzed in a quartz cuvette using a fluo-
Method characterization and validation studies have shown that rescence spectrometer. The selection of spectrometer slit widths
PAH fluorescence spectra under conventional, room temperature is a compromise of signal strength and specificity, but spectral
conditions are relatively broad. The signal is thus often not specific band widths of 2.5 nm for both excitation and emission have in
enough to distinguish between closely related isomers. Further- most cases been found suitable. The signal from the blank (due
more, the optimal EEWP of free hydroxy PAHs will not exactly to Raman scattering and traces of impurities in the solvent used
match those of the conjugated metabolites that are the domi- for dilution) should be subtracted. Standard solutions of hydroxy
nant forms in bile samples. The increased PAH fluorescence at PAHs and bile samples from fish exposed to single PAHs have
specific EEWPs is therefore often termed as typical 2&3-ring, 4- both been used to establish optimal FF wavelength pairs for dif-
ring and 5-ring PAH fluorescence (Aas et al., 2000b). Nevertheless, ferent classes of PAH metabolites. Recommendations of specific
228 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244

excitation/emission wavelengths for detection of certain PAHs vary are varied simultaneously, with a fixed wavelength interval (),
slightly among published papers. In general, the following wave- and over a specific wavelength range (Ariese et al., 1993a; Lin et al.,
length pairs have been found suitable for screening: 290/335 nm 1994; Aas et al., 2000b). SFS assays of diluted fish bile often yield
for 2/3-ring PAHs (e.g., naphthalene and phenanthrene metabo- very characteristic fluorescence spectra that are dependent on the
lites), 341/383 nm for 4-ring PAHs (e.g., pyrene metabolites), and particular PAH metabolites present in the assay mixture (e.g. Fig. 1)
380/430 nm for 5/6-ring PAHs (e.g., BaP metabolites) (Krahn et al., and similar spectral features can be observed in bile samples from
1987, 1984; Lin et al., 1996; Aas et al., 2000b). Because of a lack fish exposed to PAHs from petrogenic and pyrogenic sources.
of glucuronidated hydroxy PAH standards and because the signal Different scans optimized for specific classes of PAHs can be pro-
intensities will not be related to a single compound, FF data are duced through variation of the  used in the assay. When the
often expressed in arbitrary fluorescence units, which may be con- target compound has a strong S0 –S1 absorption, a very small 
sidered as acceptable within the framework of a screening tool (e.g., value (typically 5 nm) can be chosen, corresponding to the Stokes’
in discriminating exposed samples from non-exposed), as long as shift. In principle, because the Stokes’ shift is not constant in wave-
all samples are analyzed with the same instrument under identical length units it would be more correct to apply a constant energy
conditions. The FF signal may also be expressed as equivalents of a interval, although the difference is very minor and that approach
selected PAH standard, such as pyrene (Aas et al., 2000b). If FF data has not found widespread use. If, on the other hand, excitation into
are to be compared among different laboratories, the same com- the S2 state is much more efficient, a much larger  will be pre-
pound (whether one of the analytes or a different standard) should ferred, corresponding to the difference between S0 –S2 in excitation
be used for calibration and the instrumental settings (in the case and S1 –S0 in emission. Optimization of  for specific PAH mix-
of FF: wavelengths, spectral slit widths, lamp and detector type) tures should be conducted whenever necessary, but experience has
must be specified. In addition, bile reference materials (Ariese et shown that a  of 42 nm is suitable for many PAH metabolites and
al., 2005b) can be used to determine interlaboratory comparability discriminates to a reasonable degree between bile PAHs of different
(see below). size categories (Fig. 1).
Interestingly, the signal intensity ratios at the three wavelength Similar to FF, SFS analysis of non-hydrolyzed bile samples is
pairs 290/335, 341/383, and 380/430 nm will provide informa- a very rapid technique that normally takes only a few minutes.
tion about the dominant PAH source: a complex mixture of In screening studies, SFS is often conducted as a supplement to
non-alkylated and alkylated naphthalene metabolites is typically FF measurements and both non-hydrolyzed and hydrolyzed bile
found in fish exposed to petrogenic contamination (crude oil or samples can be used. However, in order to save time, and because
marine fuel related), leading to stronger FF fluorescence intensity the conjugated PAH metabolites typically display enhanced fluo-
at 290/335 nm, whereas combustion-type sources (traffic, domes- rescence compared to their hydrolyzed counterparts, it is often
tic heating) yield a stronger FF signal at 341/383 and 380/430 nm recommended that non-hydrolyzed samples are used for PAH
due to the presence of 4 and 5/6-ring PAHs (Lin et al., 1996; Aas screening purposes. As discussed above, a sufficient dilution factor
et al., 2000b). Most of the larger PAH molecules in the bile of PAH- (typically 1000 or 2000-fold with 50% methanol or 50% ethanol)
exposed fish are present in the form of hydrophilic conjugates such is needed to avoid inner filtering and other matrix effects, and
as glucuronides. One of the attractive features of these rapid tech- the solvent blank should be subtracted. As with FF, SFS is used
niques is that the time-consuming hydrolysis step can be omitted to measure a sum parameter rather than individual compounds;
and the conjugated species can be measured directly by fluores- although the technique is somewhat more selective than FF and
cence. However, for quantitation one should be aware of the fact concentrations of predominant metabolites such as conjugated 1-
that the fluorescence intensities of conjugated species are usually OH pyrene can be determined with reasonable accuracy (Ariese et
enhanced relative to the hydroxy PAHs and the optimal excitation al., 1993a). Usually only a single compound is used as the calibrant
and emission wavelengths are slightly shifted to shorter wave- and the results are expressed in “equivalents” of this compound.
lengths. Therefore, when the conjugated species are not available, The method is primarily meant for studies in which all samples are
as is most often the case, the corresponding free hydroxy PAHs analyzed by the same laboratory with the same method. When-
cannot be used for calibration without addressing certain caveats ever data obtained in different laboratories need to be compared,
(Ariese et al., 1993a). the analytical protocol, instrument settings, and calibration method
As a simple way of detecting the PAH exposure of fish, the need to be clearly defined and reference materials (RMs) should be
FF assay approach has been utilized in a number of studies, both exchanged between laboratories to check for data comparability.
in connection with field investigations, e.g. (Beyer et al., 1996, Other types of fluorescence scanning techniques, such as excitation
1998; Cheevaporn and Beamish, 2007; Cormier et al., 2000; Gagnon or emission spectroscopy, can also be utilized for the determina-
and Holdway, 2002; Gorbi et al., 2005; Hanson et al., 2006, 2009; tion of biliary PAH metabolites. For example, as demonstrated by
Haugland et al., 2005; Krca et al., 2007; Neves et al., 2007; Ribeiro Hawkins et al. (2002), metabolite emission spectra collected at a
et al., 2005; Vuorinen et al., 2006; Wang et al., 2008; Webb et al., fixed excitation wavelength of 254 nm can be used for screening
2005; Yang et al., 2003; Aas and Klungsøyr, 1998), and in labora- detection of phenanthrene metabolites in bile and other sample
tory exposure studies, e.g. (Beyer et al., 1997; Boleas et al., 1998; matrices.
Camus et al., 1998; Goanvec et al., 2008; Gravato and Santos, 2003; The SFS approach has been used in a number of studies for
Sandvik et al., 1997; Aas et al., 2000a). Interestingly, a comparable detecting PAH exposure of fish, both in connection with field inves-
fixed wavelength approach can be used to detect PAH metabolites tigations, e.g. (Fenet et al., 2006; Fuentes-Rios et al., 2005; Lyons et
in urine from decapod crustaceans such as crabs (Dissanayake and al., 1999; Schipper et al., 2009; van der Oost et al., 1994a; Zhang
Galloway, 2004; Koenig et al., 2008; Watson et al., 2004), hence et al., 2002), and in laboratory studies, e.g. (Barra et al., 2001;
supplementing the use of fish as study samples for rapid field sur- Hellou and Upshall, 1995; Wilson et al., 2001). Also for studies
veys of PAH pollution. One relevant advantage of using crustacean using crustacean urine, SFS has been effective as a screening tool,
urine is the possibility of obtaining the samples non-destructively. e.g. (Dissanayake and Galloway, 2004; Eickhoff et al., 2003). The
SFS approach has also been utilized for other PAH detecting pur-
2.1.3. Synchronous fluorescence spectrometry poses, such as for screening of BaP in foods with a high fat content
An alternative to using FF detection as a simple means for PAH (Garcia-Falcon et al., 2000), for identification of PAH exposure in tis-
metabolite detection is to use synchronous fluorescence spectrom- sue from marine polychaetes (Giessing et al., 2003; Tairova et al.,
etry (SFS). In SFS, the excitation/emission light detection conditions 2009) and as a selective method for monitoring pyrene in microbial
 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244 229

PAH biodegradation studies (Mahanty et al., 2008), just to mention

a few. Recently, a novel method for screening of non-hydrolyzed
fish bile based on fluorescence excitation–emission matrices com-
bined with parallel factor analysis was reported, suggesting this
assay to have an improved sensitivity in comparison to the stan-
dard SFS approach (Christensen et al., 2009). A similar approach
was demonstrated for PAH metabolites spiked at low levels in urine
(Vatsavai et al., 2008).

2.1.4. HPLC-Fluorescence screening

A reverse-phase HPLC-Fluorescence method for semiquanti-
tative screening of non-hydrolyzed PAH and PAC metabolites
were developed by Krahn (Krahn et al., 1984, 1986). Briefly, non-
hydrolyzed bile (1–5 ␮L) is injected directly onto a C-18 HPLC
column and subsequently eluted using a linear gradient from 100%
water (containing a small amount of acetic acid) to 100% methanol
or acetonitrile as described by (Escartin and Porte, 1999a,c; Krahn
et al., 1984, 1986; Leonard and Hellou, 2001; McDonald et al.,
1995). Because each phase I metabolite may be present as dif-
ferent conjugates, the resulting chromatograms tend to be very
complex with many overlapping peaks. The chromatograms can be
recorded at one or several excitation/emission wavelength pairs for
detecting specific compound classes. Metabolite concentrations are
determined semiquantitatively by using a parent PAH, e.g., naph-
thalene, BaP, as calibrant (Escartin and Porte, 1999a,c; Krahn et al.,
1986; Leonard and Hellou, 2001) or a phenolic standard, such as
1-OH pyrene or 3-OH BaP (Krahn et al., 1986). In practice, inte-
grated areas of peaks eluting after about 7 min (determined for each
chromatographic system) are summed and the sum is expressed
as the amount of the chosen standard (“equivalents”) that would Fig. 2. Chromatograms from HPLC-Fluorescence screening of PAH metabolites in
be present if the area were attributable only to that compound. bile of rock sole from oiled (Exxon Valdez crude) and reference (clean) sites and in
Because a single PAH or PAC compound is used as calibrant to mea- bile of English sole from urban and reference sites using EEWP 260/380 nm (phenan-
sure a large number of metabolites that fluoresce at the selected threne wavelengths).

wavelength, accurate quantitation is not possible, and only rela-

tive concentrations of a class of metabolites are obtained. However, assess PAH and PAC contamination in fish following pollution
judicious choice of the quantitation standard can improve accu- events, such as the Exxon Valdez oil spill (Krahn et al., 1992,
racy. For example, when a metabolite (e.g., 1-OH pyrene) was used 1993b). Estimated concentrations of bile metabolites that fluoresce
as the quantitation standard at 380/430 nm rather than a parent at 380/430 nm have been statistically correlated with the preva-
hydrocarbon (e.g., BaP), the accuracy of the screening determina- lence of certain hepatic lesions, including neoplasms, in English sole
tion improved (reference values were obtained by GC–MS (Krahn from Puget Sound, WA, USA (Krahn et al., 1984, 1986). In addition,
et al., 1987)). Furthermore, HPLC-F can be used to accurately mea- related methods have been used in laboratory exposure studies to
sure individual analytes that are abundant and well-resolved from measure dose- and time-related changes in bile metabolite con-
coeluting compounds (e.g., 1-OH pyrene). As earlier discussed for centrations (Collier and Varanasi, 1991; Stroomberg et al., 2003). A
FF and SFS, when data from different laboratories need to be com- version of the method involving enzymatic hydrolysis, methylene
pared, methods and instrument settings need to be clearly defined. chloride extraction and hexane/potassium hydroxide partitioning
In addition, reference materials should be exchanged to check the with HPLC-Fluorescence analysis, was developed by (Krone et al.,
comparability. 1992). Because this HPLC-F method involves hydrolysis of the polar
Bile screening by HPLC-F has proven to be a powerful tool for fluorescent aromatic metabolites prior to chromatographic analy-
estimating exposure of fish and other marine vertebrates to PAHs, sis, the use of the corresponding hydroxy PAHs as standards and
including nitrogen and sulfur homologs. The assay provides more calibrants would be preferred. The quantitative assays of biliary
information about the molecular composition than the FF and SFS PAH metabolites in fish are reviewed in the following sections.
screening methods. Metabolite patterns are readily discernable and
these patterns can assist in establishing the source of pollution 2.2. Quantitative chromatographic analysis of PAH metabolites in
(Krahn et al., 1993a). For example, the chromatographic pattern bile
from bile of a rock sole (Lepidopsetta bilineata) captured from Prince
William Sound after the Exxon Valdez spill was distinctly different 2.2.1. Background
from that of an English sole (Parophrys vetulus) exposed to contam- A quantitative analysis of biliary PAH metabolites will typically
inants from an urban site (Fig. 2). This difference is most likely due involve a multistep process in which the bile sample is first sub-
to the fact that crude oil contains an array of alkylated 2- and 3-ring jected to various chemical pretreatment steps before the assay
PAHs and PACs, whereas the urban contaminant mixture contains mixture is analyzed using a chromatographic system that pro-
a greater proportion of the larger, non-alkylated pyrogenic PAHs vides a sufficient separation of the analytes. This is followed by
(Krahn et al., 1993a). a detection step that should enable a proper identification and
The semiquantitative HPLC-F method has been widely applied, quantitation of the eluting compounds. The task of analyzing PAH
using fish bile, fish liver and other matrices in connection with metabolites in bile in a quantitative way is quite demanding; both
environmental monitoring programs, e.g. (Escartin and Porte, because of the complex nature of the bile in a PAH exposed (field)
1999a,b,c; McDonald et al., 1995) and in field investigations to specimen (each PAH compound can be biotransformed into sev-
230 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244

eral metabolites), but also because of the different physicochemical

(Ariese et al., 2000)

(Ariese et al., 2000)

(van Schanke et al.,

properties of the different PAH compounds involved in various

(Richardson et al.,

(Ruddock et al.,
types of exposures (i.e. petrogenic and pyrogenic PAH mixtures).

(Jonsson et al.,

(Jonsson et al.,
(Ariese et al.,

2002, 2003)
A range of alternative techniques exists for the preparation and
Reference

quantitative analysis of PAH metabolites in bile samples, and in

2004b)
1993a)

2001)

2001)

2003)
the following sections the essential technical features and practi-
cal knowledge for each of the steps are reviewed. Three tables have
been prepared to provide a reference guide: methods for HPLC-F
Fluorescence

Fluorescence

Fluorescence

Fluorescence

Fluorescence

Fluorescence

Fluorescence

Fluorescence
analysis are found in Table 1; those for LC–MS analysis are given in
Detection

Table 2; and GC–MS methods are listed in Table 3.

2.2.2. Preparation and hydrolysis

During fish sampling in the field or laboratory, bile samples are
obtained and normally frozen without further treatment. Typical

Gradient: ammonium

Gradient: ammonium
Gradient: water and

Gradient: water and

Gradient: water and

acetate buffer, pH 4,
bile volumes range from 5 to 100 ␮L, depending on the size and

acetate buffer, pH 4
containing 1 mg L−1

Gradient: acidified
Acetonitrile:water

ascorbic acid and

water, pH 4, and

feeding status of the fish. Prior to the analysis, the frozen bile sam-
Gradient: water

and acetonitrile

and acetonitrile
Mobile phase

ples are first thawed and briefly homogenized in a sonication bath.

acetonitrile

acetonitrile

acetonitrile
(70:30 v/v)

methanol

methanol

Because metabolites of PAHs exist in bile mainly as conjugates

(e.g., sulfates and glucuronides) and currently only few conju-
gated standards are available, bile samples are first hydrolyzed to
yield free hydroxy metabolites. As described by Hellou and Payne
(1987), Krahn et al. (1987), Kennedy et al. (1989), Steward et al.
Vydac 201TP52 C-18 5 ␮m;

Vydac 201TP52 C-18 5 ␮m;

(1990), van Schanke et al. (2001) and Jonsson et al. (2003), glu-
Vydac 201TP54 C-18;

Vydac 201TP54 C-18;

Vydac 201TP54 C-18;

curonides and sulfates can be simultaneously hydrolyzed by adding

4.6 mm × 250 mm

4.6 mm × 250 mm

4.6 mm × 250 mm

2.1 mm × 250 mm

2.1 mm × 250 mm
3.1 mm × 200 mm

␤-glucuronidase containing arylsulfatase activity and pH 4.8–5.0

acetate buffer, followed by incubation for 1–24 h at 37–40 ◦ C. In
Rosil C-18 HL;
HPLC analysis

RP silica C-18
Xterra RP-18

order to provide information about the relative concentrations

Column

of different conjugates a successive deconjugation can be per-

formed. Accordingly, Kennedy et al. (1989), Steward et al. (1990),
and Willett et al. (2000) separated bile metabolites into groups of
free hydroxy compounds, glucuronides, and sulfates. To prevent
Internal standard

Triphenylamine
Anthracene-d10

oxidation of the hydrolyzed metabolites, Ariese et al. (2000) and

Richardson et al. (2001) added 0.5 and 4% ascorbic acid, respec-
Perylene

tively, to a mixture of the bile and the deconjugation enzymes. An

internal standard is usually added prior to hydrolysis; this is more
critical for GC methods (because of a more complex procedure and
more volatile solvents) than for HPLC methods.
Method reference reports of sample preparation and HPLC-F analysis of hydrolyzed PAH metabolites.

The majority of publications about PAH metabolite analysis

Acetonitrile (80%)

Acetonitrile (70%)

in fish bile focus on the determination of glucuronides and sul-

Methanol (50%)

Methanol (50%)
Ethanol, ≈50 or

Ethanol (84%)

fates, but a significant portion of PAH metabolites can be present

as glutathione conjugates (Varanasi et al., 1987). However, unlike
Ethanol

Ethanol
Solvent

glucuronides and sulfates these conjugates are more likely to be

100%

retained in the liver (Plummer et al., 1980; Varanasi et al., 1987).

Glutathione conjugates in fish bile can be hydrolyzed at ambi-
ent temperature by acidic hydrolysis (pH 0–1) as described by
van Schanke et al. (2001) and Yu et al. (1995), or by addition of
37 ◦ C for 45 min
30 ◦ C for 30 min
37 C for 1 h or

␥-glutamyltranspeptidase followed by incubation at 37 ◦ C for 6 h

37 ◦ C for 2 h

37 ◦ C for 1 h

40 C for 2 h

40 ◦ C for 2 h
Incubation

(Willett et al., 2000).

◦

2.2.3. Extraction of hydrolyzed PAH metabolites

Most reports on HPLC analysis of hydrolyzed bile describe a
simple sample preparation procedure without any extraction step
sulfatase or HCl, pH 0
Enzymatic treatment

␤-glucuronidase and

␤-glucuronidase and

(Jonsson et al., 2003; Richardson et al., 2001; van Schanke et al.,

Sample preparation

2001). For example, hydrolyzed bile samples were diluted 2- to 5-

␤-glucuronidase

␤-glucuronidase

␤-glucuronidase

␤-glucuronidase

␤-glucuronidase

␤-glucuronidase

fold or more with ethanol (Ariese et al., 1993a; Richardson et al.,

2001) or methanol (Jonsson et al., 2003; van Schanke et al., 2001),
sulfatase

sulfatase

sulfatase

sulfatase

sulfatase

sulfatase

sulfatase

followed by centrifugation before injection of the supernatant onto

the HPLC column. Alternatively, extraction of bile samples can be
done as a clean-up step prior to (Leonard and Hellou, 2001; Steward
et al., 1990) or following deconjugation of PAH metabolites (Ariese
Target compounds

et al., 1993a; Leonard and Hellou, 2001; Willett et al., 2000). For
3- to 5-ring PAH
hydroxy PAHs

hydroxy PAHs

hydroxy PAHs

hydroxy PAHs
BaP metabolites

metabolites

metabolites

example, Steward et al. (1990) extracted bile with ethylacetate:

2- to 5-ring

2- to 5-ring

2- to 4-ring

2- to 5-ring

acetone (2:1 v/v) prior to enzymatic treatment in order to remove

Chrysene
1-OH Pyr

unwanted lipophilic compounds from the bile. Extraction is usu-

Table 1

ally performed following enzymatic hydrolysis of bile in order to

transfer the hydroxy PAHs into a volatile organic solvent that is
 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244 231

compatible with gas chromatography analysis (Krahn et al., 1987).

Hydrolyzed metabolites have been successfully extracted from

(Law et al., 1994)

(Zhu et al., 2008)

(Steward et al.,
acidified (pH 4.8–5.0) samples using methylene chloride containing

(Willett et al.,
10% methanol (Hellou and Payne, 1987; Krahn et al., 1992), ethyl
Reference
acetate (Escartin and Porte, 1999a; Jonsson et al., 2003; Solbakken

1990)

2000)
et al., 1980), or a mixture of hexane and methyl-t-butyl ether
(2:1) (Stephensen et al., 2000). The extractions are usually repeated
three to five times and the combined organic phase is dried with
anhydrous sodium sulfate and concentrated to 0.1–0.5 mL (Escartin

LC-APCI-MS/MS

UPLC-APCI+ -MS
and Porte, 1999a; Jonsson et al., 2003), or to dryness for some
Scintillation

LC-APCI-MS
derivatization procedures (Krahn et al., 1992). Law et al. (1994)
Detection

counting

used a solid-phase extraction clean-up step (Bond Elut C-18 car-

tridge containing 500 mg sorbent) to separate the hydrolyzed PAH
metabolites from the bile matrix. The metabolites were eluted with
100% methanol, evaporated to dryness, and the residues dissolved
ammonium acetate
buffer, 10 mM, and

acetonitrile: water
in derivatizing reagent [N,O-bis(trimethylsilyl)acetamide] before
Gradient: water,

Gradient: water
acetonitrile and

injection onto the GC column. Additionally, Bond Elut C-18 columns

and methanol
Mobile phase

were used as an alternative to successive liquid–liquid extractions

acetonitrile
methanol
Gradient:

Gradient:

(Hellou and Payne, 1987; Willett et al., 2000) in order to separate

conjugated from deconjugated metabolites. Furthermore, Willett
et al. (2000) successively extracted the bile sample before and after
treatment with pure ␤-glucuronidase and arylsulfatase, providing
Waters acquity UPLCTM

information about the relative amounts of deconjugated metabo-

S80A C-18 ECQ 5 ␮m;
Marck® LiChroCART®

lites, sulfates, and glucuronides.

Zorbax ODS 5 ␮m;
4.6 mm × 250 mm

4.0 mm × 125 mm

4.6 mm × 250 mm

BEHC18, 1.7 ␮m;

2.1 mm × 50 mm
HPLC analysis

RP-18 5 ␮m;

2.2.4. Derivatization of hydroxy metabolites

Hydroxy metabolites are not especially amenable to analysis by
Column

GC–MS, but chemically modifying the hydroxy group to a less polar

Method reference reports of sample preparation and LC analysis of hydrolyzed PAH metabolites using MS, MS/MS or other detectors.

form via derivatization improves their chromatographic charac-

teristics (Jacob and Grimmer, 1988). Derivatization also prevents
dehydration of dihydrodiols to the two corresponding phenolic
benzanthracene-7,12-

compounds in the GC (Jacob and Grimmer, 1988; Krahn et al., 1987,

Internal standard

1984). Therefore, hydrolyzed bile metabolites (hydroxy PAHs) are

6-OH chrysene

6-OH chrysene

frequently converted to trimethylsilyl (TMS) ethers or acetylates

before GC–MS analysis. There are several commercially available
reagents that easily form TMS ethers with alcohols, of which
dione

bis-(trimethylsilyl)trifluoroacetamide (BSTFA) is the most widely

used (Evershed, 1993). Hydroxy PAHs are derivatized by adding
100–200 ␮L BSTFA to the concentrated organic extract (Jonsson et
methylene chloride

ethyl acetate and

ethyl acetate and

ethylacetate and

al., 2003) or to the dried residues of the metabolites as described

Extracted with

Extracted with

Extracted with

by Krahn et al. (1987) and heating the mixture for 2 h at 60 ◦ C.

Methanol or

Trimethylsilylimidazole (TMSI) in pyridine (1:4) forms TMS ethers

acetone

acetone

acetone
Solvent

with phenolic PAHs as well as with diols and tetrols within 15 min
at room temperature (Day et al., 1991; Jonsson et al., 2004b).
Additional silylating reagents used for this purpose are 1,1,1,3,3,3-
hexamethyldisilazane (SYLON TP) (Solbakken et al., 1980) and
Acid hydrolysis at

N,O-bis(trimethylsilyl)acetamide (BSA) (Law et al., 1994). Acety-

lated hydroxy PAHs have been prepared for GC–MS analysis using
37 ◦ C for 6 h

37 ◦ C for 6 h

37 ◦ C for 6 h
80 ◦ C for 3 h
Incubation

ethereal diazomethane as described by Hellou and Payne (1986)

and Hellou and Payne (1987).

2.2.5. Instrumental determination of individual PAH metabolites

2.2.5.1. HPLC-Fluorescence. When deconjugated hydroxy PAHs are
glutamyltranspeptidase

glucuronosohydrolase
Enzymatic treatment

separated and quantified by means of HPLC-F, an extraction step is

No treatment or acid
Sample preparation

␤-glucuronidase or

possible, but usually only a brief centrifugation step is carried out.

␤-glucuronidase,

␤-d-glucuronide
sulfatase or ␥-

For chromatographic separation, most laboratories apply reversed

hydrolyzed

phase HPLC conditions and an acetonitrile/aqueous buffer gradi-

Glucurase,
sulfatase

ent. Thermostated columns are preferred to avoid slow drifts of

retention times. A selection of typical HPLC approaches using fluo-
rescence detection is listed in Table 1.
In connection with homogeneity and stability tests of PAH
metabolites

metabolites

metabolites

metabolites

metabolites in two fish bile certified reference materials (Ariese

compounds

et al., 2005b; Jonsson et al., 2003) the following measurement con-

ditions were used: deconjugated PAH metabolites in hydrolyzed
Pyrene
Target
Table 2

BaP

BaP

BaP

bile samples were separated using a thermostated Vydac 201TP54

25/4.6 (Aurora, USA) column and an acetonitrile/water gradient.
 232
Table 3
Method reference reports of sample preparation and GC–MS analysis of hydrolyzed PAH metabolites.

Target compounds Sample preparation GC analysis Reference

Enzymatic treatment Incubation Extraction solvent Derivatization Internal standards Column Detection

J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244

products

Phenanthrene metabolites ␤-glucuronidase sulfatase 37 ◦ C for 10 h Ethylacetate TMS-ethers ␣-naphthol SE-54 5% phenyl; MS EI, 70 eV (Solbakken et al.,
20 m × 0.33 mm 1980)
2- to 4-ring PAH-metabolites ␤-glucuronidase sulfatase 40 ◦ C for 4 h Methylene chloride TMS-ethers Hexamethylbenzene J & W DB5 MS 5% MS EI, 70 eV (Krahn et al.,
and isopropanol phenyl;30 m × 0.25 mm 1987, 1984)
2- to 3-ring alkylated and ␤-glucuronidase 37 ◦ C for 24 h Methylene chloride Acetylates n-pentadecane SE-30 MS EI, 70 eV (Hellou and
non-alkylated hydroxy and methanol Payne, 1987)
PAHs
2- to 4-ring alkylated and ␤-glucuronidase sulfatase 40 ◦ C for 3 h Methylene chloride 2,6-dibromophenol J & W DB5 MS 5% MS EI, 70 eV (Krahn et al.,
non-alkylated hydroxy and methanol phenyl;30 m × 0.25 mm 1992)
PAHs
2- to 5-ring alkylated and ␤-glucuronidase sulfatase 40 ◦ C for 2 h Methylene chloride TMS-ethers 2,6-dibromophenol, J & W DB5 MS 5% MS EI, 70 eV (Yu et al., 1995)
non-alkylated hydroxy and methanol 1-naphthol-d8 , phenyl;30 m × 0.25 mm
PAHs phen-d10 , chry-d12 ,
flu-d10 and BaP-d12
2- to 4-ring hydroxy PAHs ␤-glucuronidase sulfatase 40 ◦ C for 2 h Ethylacetate 2,6-dibromophenol HP-5 MS 5% MS EI, 70 eV (Escartin and
hexamethylbenzene phenyl; Porte, 1999a,c)
30 m × 0.25 mm
1- to 4-ring aromatic ␤-glucuronidase sulfatase 37 ◦ C overnight Hexane: Octadecane MSBP5 5% phenyl; MS EI, 70 eV (Stephensen et
compounds methyl-t-butyl 30 m × 0.25 mm al., 2000)
ether (2:1)
2- to 4-ring hydroxy PAHs ␤-glucuronidase sulfatase Acetylates Decachlorobiphenyl GC separation MS EI, 70 eV (Ariese et al.,
2000)
◦
2- to 4-ring hydroxy PAHs ␤-glucuronidase sulfatase 40 C for 2 h Ethylacetate TMS-ethers 2,6-dibromophenol CP-Sil 8 5% phenyl; MS EI, 70 eV (Jonsson et al.,
50 m × 0.25 2003)
Chrysene metabolites ␤-glucuronidase sulfatase 40 ◦ C for 2 h Ethylacetate TMS-ethers 4-Cl-naphthol perylene CP-Sil 8 5% phenyl; MS EI, 70 eV (Jonsson et al.,
50 m × 0.25 2004b)
2- to 4-ring hydroxy PAHs ␤-glucuronidase sulfatase 40 ◦ C for 1 h Ethylacetate TMS-ethers HP-5 MS 5% MS EI, 70 eV (Fernandes et al.,
phenyl; 2008a,b)
30 m × 0.25 mm
 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244 233

The fluorescence excitation/emission wavelength settings were Law et al. (1994) used an HPLC reversed-phase column, followed
programmed for optimal detection of each target analyte: 2- by identification using APCI/MS to characterize conjugated and
OH naphthalene, 325/358 nm; 1-OH phenanthrene, 269/380 nm; unconjugated metabolites in bile and urine of pyrene-exposed trout
1-OH pyrene, 346/384 nm; anthracene-d10 (internal standard), (Oncorhynchus mykiss). In a later study, Willett et al. (2000) deter-
250/400 nm; 3-OH BaP, 385/450 nm. External calibration was used, mined 3-OH BaP, two BaP-dihydrodiols and three BaP-diones in
based on peak heights. A calibration solution (75 ng/mL level) was deconjugated fish bile using HPLC/APCI/MS. However, the detection
analyzed after each 10 injections to check for possible instrument limits were rather poor, ranging between 50 and 250 ␮L/L. Diones
drift. are not easily determined by HPLC-F due to their low fluorescence
HPLC-F methods are widely used for the determination of PAH intensity (van Schanke et al., 2001), which underscores the impor-
metabolites as pollution exposure markers (Fillmann et al., 2004; tance of an alternative detection method for these compounds.
Harman et al., 2009; Hellou and Leonard, 2004; Jonsson et al., Recently, Zhu et al. (2008) reported the use of ultra performance
2004a; Kammann, 2007; Kawamoto et al., 2007; Pikkarainen, 2006; liquid chromatography (UPLC) combined with positive APCI for
Stroomberg et al., 2004; Tairova et al., 2009; Vuontisjarvi et al., analysis of BaP metabolites in fish bile. The analysis time was short-
2004; Vuorinen et al., 2006; Wang et al., 2008; Watson et al., ened significantly (from 60 to 10 min) and the detection limits
2004). Obviously, this popularity may be due to the widespread improved as compared with the method reported by Willett et
availability of HPLC instrumentation in laboratories around the al. (2000): <10 ␮g/L for BaP-diones, BaP-dihydrodiols and hydroxyl
world but may also be due to the suitability of HPLC for analysis BaP and 80 ␮g/L for BaP-7,8,9,10-tetrol (Zhu et al., 2008). Use of
of the more hydrophobic PAH compounds. The method requires no HPLC/ESI/MS for analysis of oxidized PAHs has been reported for a
extraction or derivatization steps, and there is little risk of thermal range of biological study matrices beside fish bile, including clam
decomposition. Fluorescence detection offers excellent sensitivity tissue (Simpson et al., 2002), cricket excreta (He et al., 1998), and
for deconjugated PAH metabolites, especially the larger and more bovine (Ferrari et al., 2002) and human urine (Andreoli et al., 1999).
hydrophobic PAHs, such as BaP. For example, typical limits of detec- However, HPLC/MS users should be aware of the possible problems
tion (LODs) for 3-OH BaP are 2–10 ng/g bile. On the other hand, that may occur because of adsorption, contamination, surface reac-
the chromatographic resolution of HPLC is poorer than that of GC tions, ion suppression or other interferences (Oehme et al., 2002).
and the fluorescence detector does not add sufficient selectivity. Methods involving HPLC–MS/MS combine the advantages of
Therefore, HPLC-F chromatograms tend to be very crowded and liquid chromatography (simple sample preparation, suitability for
in some regions show considerable overlap, in particular for the polar species and therefore no derivatization required) with the
smaller PAH metabolites, such as two-ring PAH structures. This extraordinary selectivity of tandem MS. In recent years, the great
occurs mainly when determining petrogenic PAH metabolites in potential of HPLC–MS/MS techniques has gained considerable
bile from oil-exposed fish, due to the presence of numerous alky- attention for organic chemical analysis in general and in particular
lated isomers. for the characterization of metabolites of endogenous and exoge-
nous organic compounds, a trend that is currently underscored by
2.2.5.2. HPLC/MS and HPLC–MS/MS. HPLC techniques with mass the rapidly emerging field of metabolomics. There is a huge increase
spectrometry (MS) or tandem mass spectrometry (MS/MS) detec- in the number of studies using HPLC–MS/MS for the analysis of
tion are considered to have great potential for the determination metabolites in bile, urine and other biological matrices, but so far
of conjugated, unconjugated and deconjugated PAH metabolites in this field is totally dominated by studies addressing medical and
environmental monitoring. A selection of typical HPLC approaches, human-oriented issues. In ecotoxicology, the use of HPLC–MS/MS
using MS, MS/MS or other detection methods, is listed in Table 2. is still very limited and the existing literature is to a large part
The combination of HPLC separation with the selectivity of an related to endocrine disrupting contaminants and their metabolic
MS or MS/MS detector constitutes a powerful tool for quantita- degradation products, e.g. (Tyler et al., 2009). Just a handful of stud-
tive and qualitative analysis of water-soluble and thermolabile ies addressing the analysis of PAH and PAC contaminants or their
compounds (Ferrer and Barcelo, 1998; Voyksner, 1994), conju- metabolites in fish samples are currently available, e.g. (Fernandez-
gated metabolites (Andreoli et al., 1999; Capotorti et al., 2004; Gonzalez et al., 2008; Nacher-Mestre et al., 2009; Zhu et al., 2008).
Law et al., 1994; Simpson et al., 2002) and PAHs containing one Interestingly, the potential of using fish bile as a matrix in PAH
or more free hydroxy groups (Galceran and Moyano, 1996; Willett and PAC metabolite studies is indirectly suggested in the two last
et al., 2000). HPLC/MS has been used for environmental analysis studies cited, in which it is reported that biological matrices with
for many years, particularly in connection with pesticide residue high fat content represent a challenge for HPLC–MS/MS analyses
analysis in various sample materials (Ferrer and Barcelo, 1998; because large amounts of lipid compounds are co-extracted and can
Niessen, 1999). Hornung et al. (2007) used multiphoton laser interfere with the analysis of PAH metabolites, complicating their
scanning microscopy (MPLSM) in combination with liquid chro- identification and quantification (ibid.). HPLC–MS/MS techniques
matography/mass spectrometry to study the metabolism of BaP in are expected to become particularly relevant in connection with
exposed embryonic and larval medaka (Oryzias latipes), identifying the specific analysis of conjugated metabolites in non-hydrolyzed
BaP-3-glucuronide to be the major metabolic product. However, bile samples. This is not a common part of monitoring studies, but
the number of reports that describe the use of HPLC/MS for the it could be important in special cases, such as studying the sponta-
determination of PAH metabolites in fish bile is still rather lim- neous hydrolysis of pyrene-1-glucuronide into 1-OH pyrene upon
ited, although the technique has potentially many advantages; e.g., long-term storage at ambient temperature (Ariese et al., 2000).
identification of different conjugates and a sensitive determina-
tion of poorly fluorescent PAH compounds (e.g. diones) (Doerge 2.2.5.3. Gas chromatography mass spectroscopy GC–MS. For some
et al., 1993; Koeber et al., 1997; Willett et al., 2000). By means of compounds, gas chromatography rather than liquid chromatogra-
HPLC/MS, both conjugated and unconjugated PAH metabolites can phy is the preferred approach for PAH analysis in environmental
be measured using various ionization techniques, such as electro- samples, or for the parent compounds or their metabolites in biotic
spray ionization (ESI), atmospheric pressure chemical ionization matrices. A review by Poster et al. (2006) summarizes the instances
(APCI), thermospray (TSP), and particle beam (PB). Today, ESI and where GC generally affords greater selectivity, resolution, and sen-
APCI are by far the most frequently used ionization techniques sitivity than HPLC. Generally, GC–MS is the preferred analytical
and they can both be operated in negative and positive ioniza- technique for the small and more volatile PAHs, whereas it is
tion mode (Voyksner, 1994; Willoughby et al., 1998). For example, normally less suited for larger PAH metabolites, such as hydroxy-
234 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244

BaP, due to thermal decomposition and poor sensitivity (Jacob and peak area) within 25 serial injections of a number of underiva-
Grimmer, 1988; Jonsson et al., 2003). Nevertheless, diols and tetrols tized hydroxy PAHs and the decrease was more pronounced for
of several four- and five-ring PAHs have been successfully deter- 1-OH pyrene (>60%) than for 2-OH naphthalene (25%) and 1-OH-
mined as methyl- (Simpson et al., 2000), trimethylsilyl- (Day et phenanthrene (30%). At a concentration of 1 ␮g/g, 3-OH BaP could
al., 1991), or pentafluorobenzylbromide (Väänänen et al., 2003) barely be observed in the chromatogram if underivatized. The for-
derivatives using negative chemical ionization GC–MS. Following mation of TMS derivatives, acetylates, or methyl ethers resulted in
enzymatic hydrolysis, extraction and often a derivatization step, more symmetrical and narrower peaks as compared to underiva-
quantitation of individual PAH metabolites can usually be achieved tized compounds, and therefore improved separation and detection
by GC–MS in electron impact (EI) mode. Standard chromato- limits (Jacob and Grimmer, 1988; Jonsson et al., 2003). Furthermore,
graphic conditions are as follows: columns, 20–50 m × 0.25 mm i.d. the tendency of dihydrodiols to decompose in the GC is avoided
crosslinked 5% PH ME siloxane or similar (Escartin and Porte, 1999a; when the hydroxy groups are derivatized (Jacob and Grimmer,
Stephensen et al., 2003), are generally programmed from 80 to 1988; Krahn et al., 1987). However, the derivatization efficiency
120 ◦ C at 15 ◦ C/min and from 120 to 300 ◦ C at 6 ◦ C/min, holding the decreases as the number of fused rings increases, resulting in lower
final temperature for 5–10 min. The carrier gas is helium at 80 kPa. recovery and reduced sensitivity for larger hydroxy PAHs (Jonsson
The injector temperature is in the range of 250–280 ◦ C, and the ion et al., 2003).
source and the analyzer are maintained at 200–220 and 250–280 ◦ C, The GC–MS technique provides detailed information about
respectively. The mass spectra are obtained at 70 eV and selected which particular metabolites are present in the hydrolyzed sample
ion monitoring (SIM) mode is routinely used to reduce the back- and about the potential source of those PAHs. Crude and refined
ground and thus improve the sensitivity. A selection of typical oils are enriched in parent and alkylated naphthalenes, phenan-
GC–MS methods is listed in Table 3. PAH metabolites (derivatized threnes, and dibenzothiophenes, with lower amounts of pyrenes
or underivatized) are identified by comparing retention times and and other large, non-alkylated PAHs. GC–MS analyses of bile from
mass spectra with those of reference standards. However, due to oil-exposed fish often reflect this PAH metabolite pattern (Fig. 3,
the limited number of standards available, compound identifica- top) (Fernandes et al., 2008a,b; Krahn et al., 1987, 1992; Yu et al.,
tion is often based on interpretation of the mass spectra, which 1995). In contrast, the bile of fish exposed to PAHs of pyrolytic origin
are characterized by the presence of the molecular ion and specific is characterized by the presence of a relatively high amount of 1-
fragment ions. The spectra of both derivatized (e.g., TMS-ethers) OH pyrene and the low abundance of alkylated PAHs (Escartin and
and underivatized metabolites contain a prominent molecular ion Porte, 1999a; Fernandes et al., 2008a) (Fig. 3, bottom). Furthermore,
(either a base peak or a peak with over 40% abundance compared to the GC–MS technique allows the detection of a number of nitrogen
the base peak), as well as some other fragment ions. For derivatized containing PAC compounds, namely diphenylamine and several
metabolites the most abundant fragment ions include [M−15]+ and hydroxylated metabolites, 1,2-dihydro-2,2,4-trimethylquinoline,
[M−31]+ , and a strong [M−89]+ ion [–O–Si(CH3 )3 ] that indicates and N-(1-methylethyl)-N9-phenyl-1,4-benzenediamine, in the bile
the presence of a hydroxyl group (Yu et al., 1995). For underiva- of fish exposed to other PAH sources, such as car tire rubber
tized metabolites a major fragment ion is [M−29]+ (Escartin and (Stephensen et al., 2003).
Porte, 1999a; Krahn et al., 1992; Stephensen et al., 2003). In comparison to HPLC-F, GC–MS techniques are more
Quantitation of individual metabolites is based on their GC–MS labor-intensive and more expensive due to the cost of the instru-
response relative to that of an internal standard. The GC–MS mentation as well as the need for more extensive lab facilities
instrument is calibrated by injecting standards of different con- and the increased competence of personnel required for produc-
centrations (i.e., multilevel standards). The molecular ion and a ing good analytical results. The technique is preferred for studies
major fragment ion are used for quantitation, e.g., m/z 144, 115 in which complex target analyte mixtures require the separation
for 1-naphthol; m/z 182, 152 for 9-fluorenol; m/z 194, 165 for 9- power of the GC column, such as the analytical determination of
phenanthrol, and m/z 218, 189 for 1-pyrenol (Escartin and Porte, mixtures of naphthalene and phenanthrene contaminants. This
1999a; Stephensen et al., 2003). Alternatively, the molecular and is especially important for investigations in exposed biota after
major fragment ions corresponding to the trimethylsilyl deriva- marine oil spills or the identification of specific sources of petro-
tives should be used in case of derivatization; e.g. m/z 216, 201 genic PAH contamination (Iqbal et al., 2008a,b; Troisi et al., 2006;
for 1-naphthol; m/z 254, 165 for 9-fluorenol; m/z 266, 251 for 9- Webster et al., 1997).
phenanthrol, and m/z 290 for 1-pyrenol (Fernandes et al., 2008a,b;
Yu et al., 1995). The instrumental LODs of the GC–MS/EI technique 2.2.5.4. Modern gas chromatography approaches. Gas chromatogra-
for underivatized metabolites, calculated for a signal-to-noise ratio phy combined with time-of-flight mass spectrometry (GC/TOFMS)
of 3:1, were at the low pg level (4–9 pg), except for 1-naphthol has been suggested recently as a reliable analytical approach for
(95 pg) and 1-pyrenol (68 pg) (Escartin and Porte, 1999a). For the determination and confirmation of PAHs in complex matrices
underivatized 3-OH BaP, the LOD is much higher, ca. 1000 pg (Nacher-Mestre et al., 2009). The method is based on GC coupled to
(Jonsson et al., 2003). For underivatized metabolites, run-to-run triple quadrupole and TOFMS analyzers in which the ions produced
reproducibility ranged from 2 to 4%, with the exception of 1-OH by the separated analytes are accelerated by an electrical field in
pyrene (19%). The LODs for TMS-derivatives were 1.2, 2.4, and a vacuum tube to the same kinetic energy, with the velocity of
6 pg for 2-naphthol, 1-phenanthrol, and 1-pyrenol, respectively. each ion depending on the mass-to-charge ratio. GC/TOFMS results
For 3-OH BaP, the LOD was 50 pg when analyzed with a clean for identification and quantitation of biliary metabolites of PAHs,
ion source, but worsened with the number of analyzed samples; alkylated PAHs and alkylphenols in fish exposed to an artificial off-
in real samples the LOD was about 3000 ng/g bile (Jonsson et al., shore produced water mixture have also been recently reported
2003). For derivatized molecules, the run-to-run reproducibility (Grung et al., 2009; Harman et al., 2009). Similar to HPLC–MS/MS,
for TMS ethers was better than 5% for 1-naphthol, 2-naphthol, and the use of GC combined with tandem MS may represent potential
1-phenanthrol, and better than 10% for 1-pyrenol (Jonsson, unpub- new avenues to analyze for parent PACs or their metabolites.
lished results).
Not carrying out a derivatization step prior to GC–MS obviously 2.2.6. Other analytical techniques
saves time, but this approach should be followed with caution, 2.2.6.1. Capillary electrophoresis with fluorescence detection. The
especially when large series of samples are to be analyzed. Jonsson applicability of capillary electrophoresis with fluorescence detec-
et al. (2003) reported a decrease in signal intensity (measured as tion (CE-F) was demonstrated by Kuijt et al. (2001). In
 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244 235

Fig. 3. Chromatographic profiles (total ion current) obtained by GC–MS/EI SIM mode from a representative (underivatized) bile sample of fish exposed to PAHs of petrogenic
(top) and pyrolytic origin (bottom).

order to achieve the separation of neutral compounds, a laser excitation (using a tuneable excimer laser/dye laser combi-
cyclodextrin-modified micellar electrokinetic chromatography nation) and time-resolved detection. After enzymatic hydrolysis
mode (CD-MEKC) was used. Both conventional (lamp-excited) and extraction, hydroxy PAH metabolites were methylated using
and laser-induced fluorescence (LIF) were tested. LIF detection at methyliodide and frozen in n-octane to 23 K. In real samples,
266 nm yielded excellent detection limits (2–6 ng/mL), but for com- extremely low LODs down to 0.005 ng/mL bile were obtained, and
plex mixtures, a tuneable laser system would be more selective. 3-OH BaP could even be detected in fish bile from control sites.
When analyzing a complex PAH metabolite mixture in a bile sample Although some will argue that the use of Shpol’skii fluorescence
from oil-exposed plaice (Pleuronectes platessa) a very good separa- spectroscopy could become a tool for routine analyses of PAHs
tion was obtained. The elution order was very different from that and PAH metabolites (Garrigues and Budzinski, 1995; Yu et al.,
of reversed-phase HPLC. The technique was also used to separate 2002), the rather complex spectroscopic setup required is prob-
conjugated metabolites of pyrene. A major advantage would be that ably a major reason for the relatively limited use of the approach in
only minute volumes of bile are required (for instance for the anal- fields outside analytical chemistry. In addition, the limited tuning
ysis of very small fish), but so far the method has not been applied range of most dye lasers limits the ability to determine differ-
to monitoring studies. ent metabolite types in the same sample. However, the Shpol’skii
method is proven to be highly useful for the determination of PACs
2.2.6.2. Shpol’skii spectroscopy. In Shpol’skii spectroscopy normal that are present at very low levels, sulfur- and nitrogen-containing
(straight-chain) alkanes (e.g. n-octane) are used as solvents that compounds, very large metabolites or when isomer-specificity is
at low temperatures form neat crystalline matrices in which the crucial. A new development that simplifies the experimental setup
analyte compounds are embedded prior to fluorometric determi- and strongly reduces the cool-down time is a sample holder with a
nation. This technique can be used to obtain high-resolution spectra fiber optic probe attached that can simply be lowered into a helium
of PAHs and PACs and mixtures (Colmsjö et al., 1987; Garrigues storage vessel (Yu and Campiglia, 2005).
and Ewald, 1987). Ariese et al. (1993b) demonstrated the great
specificity of Shpol’skii fluorescence spectra for hydroxy BaPs, with 2.2.6.3. Supercritical fluid chromatography/atmospheric pressure
spectral bandwidths typically improving from 10 to 0.1 nm. This chemical ionization MS. Supercritical fluid chromatography (SFC)
way, fingerprint fluorescence spectra were obtained that allowed in combination with atmospheric pressure chemical ionization
the identification of all 12 monohydroxy BaP isomers. By means of (APCI) mass spectrometry has been used for the characterization
Shpol’skii spectroscopy, compounds can thus be determined in a of hydroxy PAHs. These compounds were separated on a reversed-
mixture without chromatographic separation. The selectivity and phase column and mass spectra were obtained in both positive and
sensitivity was improved even further by means of site-selective negative ionization modes. Proton addition and proton abstraction
236 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244

were the common routes of ionization and losses of CO and H2 O the total number of analyses is limited due to analytical capacity
occurred. Transformation to quinone structures was observed for constraints, analyzing more individual samples may be more useful
some hydroxy PAHs. Run-to-run reproducibility was good (2–12% than carrying out several replicate analyses per sample.
RSD), but detection limits were at the ng level (Moyano et al., 1997).
3.2. Sample stability and storage
2.2.6.4. Thin-layer chromatography. Tyrpien (1996) used thin-layer
chromatography (TLC) to determine hydroxy PAHs in airborne Sample stability during sampling, shipment, and storage has tra-
particulate matter. The TLC spots could be detected under UV illu- ditionally been a matter of great concern, especially when cooling
mination at  254 and 365 nm at low ng levels after visualization facilities are limited in the field. Relevant experiments on the sta-
with a fluorescent reagent. TLC is a relatively simple analytical tech- bility of PAH metabolites in fish bile were reported (Ariese et al.,
nique that can be optimized for the analysis of hydroxy PAHs by 2005b) in connection with the preparation and characterization of
careful selection of the stationary phase and the mobile phase com- two fish bile certified reference materials. Aliquot non-hydrolyzed
position (Coman et al., 1997). To the best of our knowledge, the bile samples were kept at four different temperatures (−20, +4, +20
method has not yet been applied to fish bile studies. and +40 ◦ C, respectively) and compared to aliquots kept at −70 ◦ C.
Analyses done at several time points during the test period revealed
3. Aspects of analytical quality a surprisingly high level of stability. During a period of 13 months,
no degradation of metabolites was observed in bile kept at −20 ◦ C
Several aspects of analytical quality and quality control (QC) are or +4 ◦ C, and in samples stored at +20 ◦ C only very minor changes
important when PAH metabolite analysis in fish bile is used for in metabolite levels were observed. Even in the harshest, +40 ◦ C
the purpose of environmental monitoring. Reproducibility (preci- treatment, which was included as a worst-case scenario mimicking
sion) and accuracy of the data are crucial, although the required field sampling in the absence of cooling equipment, no signifi-
level of analytical quality depends, of course, on the purpose of the cant decrease of metabolite concentrations was observed after 2
study. Typically, monitoring studies that involve the analysis of PAH months. However, a separate test using non-hydrolyzed bile sam-
metabolites in fish bile samples have been carried out by a single ples showed a significant decrease in the pyrene-1-glucuronide
institute using a single analytical method and an in-house analytical levels after a few months at room temperature. This indicates that
protocol. Obviously, if the purpose of the monitoring is to map the a certain degree of hydrolysis of conjugated metabolites to hydroxy
environmental distribution of PAH pollution by studying relative forms will occur at ambient temperature, but the resulting decon-
differences in PAH metabolite levels between samples obtained at jugated PAH metabolites were apparently not very sensitive to
different sites, short-term reproducibility (minimization of random further degradation. In fact, when the analytical method is aimed at
errors) will be most important. Long-term reproducibility is essen- hydroxy metabolites after enzymatic deconjugation, such sponta-
tial when such monitoring studies are conducted over several years. neous hydrolysis would never be noticed. Although the long-term
Quality requirements will be most stringent for studies involving stability in that study was not specifically tested for all possible PAH
several institutes or different methods – in those instances it is metabolites, the findings led to the conclusion that PAH metabo-
also essential to rule out any significant interlaboratory bias (sys- lites do not seem to be very sensitive to degradation as long as they
tematic error). In the following sections, various aspects of quality are in the fish bile matrix (Ariese et al., 2005b). The good stability is
assessment and control for PAH metabolite analysis and options for relevant when deciding on storage conditions of fish bile samples
implementing of QC measures are discussed. and bile reference materials, for this −20 ◦ C or colder is adequate.
The stability is also important in connection with sampling activity
3.1. General considerations (e.g. during field work) and shipment of bile samples by courier
or express mail: keeping the samples on an ice bath, or having
Fish bile typically contains a very large number of compounds, them exposed to ambient temperatures for shorter periods is not
and quantitation of individual compounds is often difficult due to likely to cause any significant degradation (Ariese et al., 2005b).
chromatographic overlap. Compounds absorbing/emitting in the The stability of metabolites in the bile is probably related to the
same wavelength region may cause interference in the case of flu- high levels of anti-oxidants in the fish bile matrix, although spec-
orescence detection, whereas isomers or fragments with identical ulations about this fall beyond the scope of this paper. It should be
masses may cause interference in the case of MS detection. Other stressed, however, that PAH metabolites in clean solvents are much
potential analytical problems can be related to: (1) poor signal- more prone to degradation than the same compounds in a bile sam-
to-noise ratios (e.g., for some less abundant compounds or for ple. Calibrant solutions should therefore be prepared fresh, and a
samples from unpolluted sites); (2) instability of the analytes in suitable antioxidant (e.g., 0.5% ascorbic acid in water/methanol, or
the sample; (3) instability of the calibrant solutions; (4) incom- 1.5% butylhydroxytoluene (BHT) for nonpolar GC-solvents) should
plete hydrolysis; (5) incomplete derivatization (for GC); (6) random be added for protection. Matching the vial size with the solution
errors from weighing and/or diluting; (7) solvent evaporation; (8) volume is also recommended in order to reduce the amount of air
matrix effects; (9) random variations in injection volume; (10) (oxygen) above the sample. To save time and reduce the weighing
non-linearity of detector response; (11) poor recoveries; or (12) error, a larger batch of calibrant solution could be prepared and
the difficulty of accurately determining recovery correction fac- subdivided over many small vials before freezer storage.
tors. Many of these potential error sources occur in other types of
chemical analyses as well, and many of the standard quality con- 3.3. Hydrolysis
trol measures that can be found in analytical chemistry textbooks
can be applied to PAH metabolite measurements. Therefore, in this In fish bile samples the PAH metabolites are covalently bound
section these measures will only be briefly discussed, with empha- to different kinds of hydrophilic groups (glucuronic acid, sul-
sis on aspects that are specific to PAH metabolite determination. fate, glycine, glucose, glutathione, and others). Analytical methods
Although issues related to sampling protocols and experimental aimed at quantifying individual PAH metabolites will normally
design are beyond the scope of this paper, we would like to stress start with a deconjugation step to liberate/regenerate hydroxy-
that when designing a monitoring program, one should always take lated phase I metabolites. In most studies, this is done by adding a
into account the biological variability and the analytical error. The deconjugation enzyme solution that contains ␤-glucuronidase and
latter should not contribute significantly to the total variability. If aryl sulfatase activity, e.g. (Grung et al., 2009; Ruddock et al., 2002,
 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244 237

2003). A potential source of measurement error is incompleteness Normally, when PAH metabolite analyses in fish bile are used in
of the deconjugation step. This may be due to catalytically subopti- connection with environmental monitoring, the use of a decon-
mal enzymatic deconjugation protocols or if metabolites other than jugation mixture consisting of ␤-glucuronidase and aryl sulfatase
those hydrolyzed by the added deconjugation enzymes are present activity is considered to be appropriate.
in significant concentrations. In the study reported by Ariese et al.
(2000) a quality test involving 12 participating laboratories was 3.4. Derivatization
made to determine how critical the hydrolysis incubation time was
to the analytical result. The participants received a test solution of As discussed above, when GC methods are used a derivatiza-
pyrene-1-glucuronide to be spiked to “blank” bile, and were asked tion step (e.g., with BSTFA) will normally be performed in order to
to check the 1-OH pyrene yield after different incubation times increase the volatility of the metabolites, especially for the heav-
(their normal incubation time, half that long, and twice that long). ier PAH metabolites, and to improve the chromatographic peak
The results showed no indication of incomplete hydrolysis or losses shape. Such derivatization reactions may not always result in a
upon extended incubation, indicating that in all laboratories the 100% yield, and if the losses are significant, a correction should
hydrolysis had reached a flat plateau value and was not a critical be made. This bias will be partly compensated for if the calibrants
step (Ariese et al., 2000). are also derivatized with the same reagent and under similar con-
When using a deconjugation mixture of ␤-glucuronidase and ditions. Nevertheless, it should be realized that the yield of such
aryl sulfatase, glutathione conjugated PAH metabolites will not derivatization reactions may not be the same in real samples and
be hydrolyzed. Hence, it is necessary to determine how abundant in clean solutions and a recovery check should thus be carried
glutathione conjugates of PAHs are in fish bile (in comparison to out, for example, by spiking a “blank” bile with known amounts
other major metabolites, i.e. glucuronides and sulfate conjugated of PAH metabolites. Alternatively, a hydroxylated compound that
metabolites) and how glutathione abundance varies by species and does not normally occur in fish bile samples could be used as
season. Varanasi et al. and Solbakken et al. were among the first to the internal standard, assuming its yield in the derivatization step
characterize biliary and urinary PAH metabolites in fish. They con- matches that of the analytes. As a rule, even if losses are in princi-
cluded that, although PAH metabolites sometimes were abundant ple corrected for, one should always strive for maximum yield in
in unconjugated hydroxy forms (e.g. for phenanthrene metabo- order to reduce the impact of random errors. Derivatization is not
lites), the conjugated metabolites were typically dominant (often needed for HPLC-based methods or for the fluorescence screening
»90%) and glucuronides were the major conjugated metabolites. assays.
Furthermore, they found that sulfate conjugates were present to
a lesser degree and also that species, gender and season depen- 3.5. Internal standards
dent variability occurred (Solbakken and Palmork, 1981, 1984;
Solbakken et al., 1980; Varanasi and Gmur, 1981; Varanasi et al., In many analytical schemes it has become common practice to
1979, 1983). Glucuronidated PAH metabolites in bile from PAH add a known amount of a suitable internal standard to the sam-
exposed fish have been measured and reported in a large number of ple at the onset of the sample preparation step. A second internal
studies, e.g., Law et al. (1994) who observed that the majority of the standard is in some cases added just prior to the final separa-
pyrene metabolites in the bile of trout (Oncorhynchus mykiss) were tion/detection step. Ideally, an internal standard should resemble
conjugated with glucuronic acid or sulfate. The glucuronidation the analyte(s) as closely as possible and thus all random or sys-
of products from phase I metabolism is achieved by the action of tematic errors that may occur during the analytical procedure (e.g.,
uridine 5-diphospate-glucuronosyltransferase (UDP-GT) (Leaver et weighing/dilution errors, solvent evaporation, incomplete deriva-
al., 2007), whereas enzymes in the glutathione-S-transferase (GST) tization, variations in injection volume or detector response) will
family are responsible for the glutathione conjugation. affect the internal standard to the same extent as the analyte. The
It is unclear to what degree hepatic GSTs contribute to the phase internal standard should normally not be present in real samples,
II metabolism of PAH in the liver and bile duct system of fish. Stud- and should not interfere with the analyte determination.
ies of substrate specificity, gene expression and tissue distribution For HPLC-F determination of hydroxy PAHs, some common error
of plaice GSTs suggest a predominantly “housekeeping role” for sources (e.g., incomplete derivatization, extraction, solvent evap-
hepatic glutathione conjugation, in particular for detoxification of oration) play no role, and internal standards appear to be used
lipid peroxidation products and other reactive and highly cyto- less frequently in comparison to GC–MS analyses (Richardson et
toxic radicals, rather than in metabolism and excretion of PAHs al., 2001; Ruddock et al., 2002; van Schanke et al., 2001). Willett
(Leaver et al., 1992a,b; Leaver and George, 1998; Martinez-Lara et et al. (2000) used 6-OH chrysene as an internal standard for the
al., 2002). However, the relative contribution of GSTs to the total determination of BaP metabolites and Jonsson et al. (2003, 2004b)
conjugation activity could possibly vary considerably between dif- used perylene and triphenylamine for the quantitation of chry-
ferent PAHs and also between different phase I products of the sene metabolites and two- to five-ring metabolites, respectively.
same PAH parent molecule, e.g. with a preference for the most Other possible internal standards for use in HPLC assays are listed
toxic and reactive PAH phase I metabolites for GSH conjugation, in Table 1.
whereas the major path for all other PAHs is glucuronidation. For For quantitative GC–MS methods, the internal standard can be
example, studies with isolated plaice hepatocytes incubated with added before the sample hydrolysis step, as described by Krahn et
BaP showed that glucuronidation was the major phase II path- al. (1987), Escartin and Porte (1999c), and Jonsson et al. (2003). The
way and glutathione conjugation contributed significantly, whilst choice of internal standard depends on the analytical method and
sulfate conjugation activity was very low (ibid.). In special cases the metabolites of interest. For GC–MS quantitation, ␣-naphthol
when the PAH metabolites in a bile sample are present in signif- (Solbakken et al., 1980), 2,6-dibromophenol (Krahn et al., 1987),
icant concentrations as glutathione conjugates it could possibly n-pentadecane (Hellou and Payne, 1987), octadecane (Stephensen
be worthwhile to supplement the enzymatic deconjugation proto- et al., 2000), and 4-chloro-1-naphthol (Jonsson et al., 2004b) have
col to include deconjugation of metabolites bonded to glutathione, been utilized. Yu et al. (1995) used both deuterated hydroxy PAHs
e.g. by adding gamma-glutamyl transpeptidase (GGT). However, and PAHs in addition to 2,6-dibromophenol as internal standards
this will make the deconjugation protocol more demanding, and for GC–MS determination of a number of bile metabolites. Other
it remains uncertain how extensively this extra species will con- internal standards for use in connection with GC methods are listed
tribute to better sensitivity or relevance of the total measurement. in Table 3.
238 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244

It should be realized that using an internal standard does not, Both in the case of recovery correction and in the case of cali-
in all cases, reduce the total error. This is because the addition of a brant solutions in a fish bile matrix, the matrix chosen should be
known amount of internal standard and its determination implies as close as possible to the matrix in real samples. This will not
an extra measurement and will in itself carry a certain error. There- always be possible, particularly because fish bile samples from the
fore, in the absence of major error sources the total error after field may show a broad range of accumulated PAH levels. In such
correction for the internal standard may even become larger. In instances, the matrix correction can only serve as an approxima-
one particular case, we noticed that when using anthracene-d10 as tion. Of course, optimizing the analytical conditions in such a way
internal standard in HPLC-F, the internal standard peak was some- that matrix effects become insignificant and need no correction is
what sharper than those of the analytes and during a long series always the preferred option.
of chromatographic runs, column deterioration affected the peak
width and height of the internal standard more than that of the ana- 3.7. Method validation and quality control schemes
lytes (IVM-Amsterdam, unpublished results). Such cases can only
be detected when the samples or calibrant solutions are analyzed For PAH metabolite analysis in fish bile, the measures that can be
more than once. We recommend that an internal standard should used for optimization of (quantitative) detection methods and for
always be included, but it should be left to the judgment of the quality assessment/quality control are to a large degree comparable
analyst whether or not to apply the correction. to general QA and QC schemes for the analysis of chemical mix-
tures, possibly with the exception of the limited availability of PAH
3.6. Interference and matrix effects metabolite standards. Because of the lack of pure standards, the role
of laboratory reference materials (RMs) becomes more important
Fish bile samples from the field may contain hundreds of PAH in connection with the quality improvement, testing and documen-
metabolites, so a complete chromatographic separation of the tar- tation effort. Ariese et al. (2005b) described the development and
get analytes may not always be possible. The MS detector offers testing of two fish bile reference materials that were made from
the possibility of checking the peak purity based on the mass spec- real bile samples obtained from fish exposed to pyrogenic and pet-
trum and sometimes tandem MS can be used for extra selectivity. rogenic PAH mixtures. These were used in a EU sponsored project
Interferences are a problem for HPLC-F because of its typically aimed at producing two commercially available reference mate-
poorer chromatographic resolution and limited selectivity of the rials containing concentrations of selected PAH metabolites. Such
fluorescence detector. However, simple coelution experiments can certified reference materials (CRMs) can serve as tools for provid-
be conducted to check for possible interferences: add the tenta- ing better intercalibration of biliary PAH metabolite measurements
tively identified analytes to the sample and then check whether among laboratories at both the national and the international level
the target peak coincides exactly with that of the spike. For HPLC, and also can assist in the development of acceptance criteria for use
thermostated columns are strongly recommended in order to keep in environmental risk assessment and monitoring.
the retention times as constant as possible. In addition to real fish bile samples, RMs can also be made from
The matrix constituents of a fish bile sample may influence the pure compounds, calibrant solutions, extracts or spiked bile sam-
quantitative determination of PAH metabolites in various ways. For ples; all depending on the intended QA or QC use. However, once a
the rapid fluorescence screening methods based on SFS or FF, insuf- method has been optimized, QC measurements should preferably
ficient dilution of the bile sample may cause negative bias due to an not be performed on clean calibrant solutions, because such sam-
“inner filter” effect (absorption of excitation and/or emission light). ples are not representative in terms of interferences and therefore
Furthermore, the yield of derivatization reactions may be lower for may give an overoptimistic impression of the method’s perfor-
real samples than for clean solutions. Chromatographic peak shapes mance. Instead, samples (RMs) should be used that have as similar
and the detector response may also be affected by the matrix. There a matrix to real samples as possible. In order to prepare a suit-
are several approaches that correct for these effects, depending on able RM of fish bile, a relatively large volume of bile from a group
the type of calibration used. In the case of external calibration, a cal- of fish exposed to a specific (and relevant) PAH treatment is col-
ibration series is measured at the beginning of an analytical run. In lected, pooled, thoroughly homogenized, split in multiple aliquots,
addition, it is recommended that one calibration solution be mea- and then analyzed a number of times to establish the mean “×”
sured repeatedly (e.g., after every ten injections) in order to check and standard deviation “” for specific analytes. The × and  val-
for slow instrument drift, column deterioration, etc. Each sample ues are essential parameters in constructing a control chart (CC) A
is measured only once and the result is typically compared to the CC based on a standard Shewhart XBAR chart type (Mullins, 1994),
external calibration curve, after ratioing with the internal standard. or a more advanced CC alternative, e.g. (Prabhu et al., 1994), can
If the calibration curve is made in clean solvent, it may be neces- be used. The RM aliquots should be distributed in separate vials or
sary to investigate possible matrix effects in detail and correct the ampoules and stored under conditions of optimal stability, for fish
final data for incomplete recovery. As an alternative, it is also possi- bile preferably −20 ◦ C or colder. At regular intervals or with each
ble to prepare the calibrant solution in a blank matrix, although in analytical series a subsample of the RM is analyzed, and new data
practice it will not be easy to obtain a true blank that does not con- are entered into the CC. Over a certain time period, the long-term
tain measurable levels of PAH metabolites and that has precisely reproducibility of the method that is an essential quality parameter
the same level of interferences as the sample. Another but more will be documented. Typical warning limits for deviance from the
labor-intensive approach is to spike each measured sample with reproducibility are set at x ± 2 levels, and action limits at x ± 3
known quantities of analyte, preferably at two or three levels, and levels. Obtaining a result outside the action limits indicates that
then repeat the measurement. When using this standard addition the whole series measured on that particular day likely needs to be
approach for calibration, the spiked amounts should be similar to repeated. Smaller, but repeated deviations from the average may
the amount in the sample. The detector sensitivity in the presence point at a slow drift in method performance (for instance a slow
of matrix can be derived from the increase in detector response as decrease in the activity of the enzyme used for hydrolysis) and also
a function of the spiked amount. Standard addition is an excellent calls for further investigation.
approach to correct for matrix effects, but a major drawback is the For an RM to be suitable for long-term QC, it needs to meet
fact that each sample needs to be measured repeatedly, e.g. 3–4 a number of requirements. First of all, the RM needs to be rep-
times. Understandably, external calibration is preferred by most resentative in terms of matrix interferences, analyte pattern and
laboratories, in particular for large sample series. content, physical status, and analyte-matrix interactions. In addi-
 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244 239

tion, the homogeneity and long-term stability of the RM needs to ods is available for the determination of PAH metabolite levels in
be tested, because using the same RM for several years is prefer- fish bile samples. The most rapid screening assays (FF and SFS)
able. Furthermore, the analyte content should not be very close are remarkably undemanding analytically, because simple fluores-
to the detection limit, because at that level the random detection cence detection of diluted bile provides a sensitive indication of the
error becomes very large and may obscure other possible sources presence of PAH metabolic products. As shown in some of the cur-
of error. Laboratories that share the same RMs, and possibly use rently reviewed studies, these semiquantitative assays are suitable
similar CC procedures, can easily intercompare the performance of for screening fish bile samples in situ at field locations e.g. where a
their analysis. relevant PAH pollution incident (e.g., an oil spill) has occurred. They
The active collaboration among laboratories on aspects of QC are also suitable, inexpensive analytical methods for the biomon-
will contribute to the success of an interlaboratory comparison itoring of environmental recovery at such locations. However, FF
exercise or laboratory proficiency tests. In an intercomparison exer- and SFS will most often fail to quantify specific metabolites in a
cise, e.g. (Ariese et al., 2005b; Ibe and Kullenberg, 1995; Parris et complex bile matrix. In order to provide selective and quantitative
al., 2002; Rowles et al., 2002; Villeneuve et al., 2004), a central lab- estimates of biliary PAH metabolites, a chromatographic separation
oratory prepares one or several thoroughly homogenized samples of the metabolite constituents in the bile matrix is required. This is
and distributes these, together with the necessary measurement achieved normally by using GC or reverse-phase HPLC procedures,
instructions, to the participating laboratories who analyze the sam- and as discussed in the current review, the former approach is gen-
ple and report the data to the coordinator by a given deadline. erally favorable for smaller, more volatile PAH species whereas the
The results are then statistically analyzed, and the participants can latter is generally best suited for larger, more hydrophobic PAH
check how their results compare to the assigned reference val- species. Prior to a chromatographical separation, the bile sample
ues. The reference values are only known in the case of standard is typically subjected to various forms of pretreatment, includ-
solutions, otherwise the results from an expert laboratory or the ing enzymatic hydrolysis (deconjugation of metabolites), analyte
mean of all participants after technical scrutiny can be used as the extraction and analyte derivatization. As shown in some of the stud-
assigned value. In cases when results show bimodal or multimodal ies discussed presently, the pretreatment procedure is often crucial
distributions, the statistical model of Cofino et al. (2000) can be used for the quality of the analytical result.
to determine the assigned value. This method calculates the point of The detection of the PAH metabolites is perhaps where most
maximum overlap between submitted datasets, without the neces- progress was made recently. Historically, fluorescence detection
sity of rejecting outliers. In the case of a bimodal distribution of has been the most important approach used in HPLC assays,
datasets, the method will produce two possible maxima, rather whereas MS has been the dominating detection technique used
than an average between the maxima. Unless all participants use in connection with GC assays. Currently, the involvement of
exactly the same analytical protocol (and apply it correctly) any sig- advanced tandem mass spectrometers, after HPLC or GC separa-
nificant systematic error in one of the methods will thereby become tions, now enables a better sensitivity for the determination of
apparent and can be discussed in technical meetings, where pos- specific metabolites (e.g. of highly carcinogenic PAHs) that are
sible error sources are openly discussed, and analysts can obtain present at very low concentrations, or metabolites that so far have
useful advice from their colleagues. If appropriate CRMs are not been hidden because of peak interference or other chromatograph-
available or if they are considered as too expensive for regular ical problems. This stronger analytical power will also open ways
use, the production and QC use of a suitable non-certified fish bile for a more specific identification of PAH contamination sources. In
RM may be a good alternative for most laboratories. However, it addition, improved qualitative analysis of conjugated PAH metabo-
should be noted that although QC activities using a non-certified lites by LC–MS/MS offers a promising tool that will lead to improved
RM can be useful for the monitoring of precision and long-term insight in PAH metabolism mechanisms in vivo. It may also con-
reproducibility, systematic errors may remain unnoticed. In that tribute to a better design of PAH biomonitoring guidelines in fish
case the values in the control chart and the reference values may and other biota, i.e. when selecting which metabolites to analyze
show the same deviation from the true value. Such inaccuracies can and monitor.
only be identified by taking part in interlaboratory comparisons PAH pollution almost always involves a complex mixture of PAH
or by using a CRM. The QUASIMEME office (originally located in and PAC species that are rather diverse with respect to physic-
Aberdeen, UK and currently at Wageningen University, the Nether- ochemical properties, and this complexity is also seen for the
lands) has organized several intercomparison exercises for PAH in metabolite composition in the bile from exposed fish. It is therefore
marine biota samples (Law, 2000; Law et al., 2000; Villeneuve et a general technical challenge to analyze real bile samples with opti-
al., 2004), and also intercomparison exercises and proficiency tests mal accuracy and precision. This calls for the collaboration between
for PAH metabolites in fish bile (Ariese et al., 2005b). laboratories on aspects of quality assurance and quality control
of PAH metabolite analysis. In this respect, commercially avail-
able (certified) reference materials that consist of representative
4. Summary fish bile samples with known content of specific (and ecotoxico-
logically relevant) PAH metabolites is obviously a crucial means
Metabolite levels in fish bile are seen by many as excellent for providing increased intercomparability of PAH metabolite data.
markers for the assessment of exposure to those pollutants that However, these reference materials still have limited availability,
are readily metabolized by fish, among them PAHs and PACs. The although an EU-SMT funded project has demonstrated how such
metabolites are accumulated in the bile before further elimina- fish bile reference materials can be produced and, after analyti-
tion occurs, mainly via the alimentary tract. Because of extensive cal characterization, made available for users for the purpose of
metabolism, detection of parent PAHs in tissues may underesti- analytical quality control and data intercomparison. A considerable
mate the exposure level in vivo, and some PAHs present at low levels number of studies employ PAH metabolites in fish bile as markers
may fall under the level of analytical detection. Biliary PAH metabo- for assessment of PAH exposure in fish in specific aquatic environ-
lite concentrations, on the other hand, are proven to be sensitive ments or in laboratory exposure situations. These studies generally
PAH exposure markers, possibly two orders of magnitude more confirm the feasibility, sensitivity and relevance of these PAH expo-
sensitive than tissue parent compound levels, and bile metabo- sure markers, although one should pay attention to factors related
lites should therefore be the markers of choice for monitoring and to environmental compartment (season, water temperature, etc.),
environmental risk assessment studies. A range of analytical meth- the biology of the study specimen (species, feeding status, etc.) and
240 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244

the composition of the bile matrix being studied (inner filter effect, Boleas, S., Fernandez, C., Beyer, J., Tarazona, J.V., Goksøyr, A., 1998. Accumulation
chromatographic overlap, etc.). The quality of biliary PAH metabo- and effects of benzo(a)pyrene on cytochrome P450 1A in waterborne exposed
and intraperitoneal injected juvenile turbot (Scophthalmus maximus). Marine
lite data also depends on the selection of an appropriate analytical Environmental Research 46, 17–20.
technique, based on the particular type of PAH contamination (e.g., Brown, J.S., Steinert, S.A., 2004. DNA damage and biliary PAH metabolites in flatfish
pyrogenic, petrogenic) and the level of analytical specificity needed. from Southern California bays and harbors, and the Channel Islands. Ecological
Indicators 3, 263–274.
Finally, we hope that the information included in the current paper Budzinski, H., Mazeas, O., Tronczynski, J., Desaunay, Y., Bocquene, G., Claireaux, G.,
will assist researchers in this method selection process. 2004. Link between exposure of fish (Solea solea) to PAHs and metabolites:
Application to the “Erika” oil spill. Aquatic Living Resources 17, 329–334.
Bunton, T.E., 1996. Experimental chemical carcinogenesis in fish. Toxicologic Pathol-
Conflict of interest ogy 24, 603–618.
Buryskova, B., Hilscherova, K., Blaha, L., Marsalek, B., Holoubek, I., 2006. Toxicity
and modulations of biomarkers in Xenopus laevis embryos exposed to polycyclic
None. aromatic hydrocarbons and their N-heterocyclic derivatives. Environmental
Toxicology 21, 590–598.
Camus, L., Aas, E., Børseth, J.F., 1998. Ethoxyresorufin-O-deethylase activity and fixed
Acknowledgements wavelength fluorescence detection of PAHs metabolites in bile in turbot (Scoph-
thalmus maximus L.) exposed to a dispersed topped crude oil in a continuous
Financial support for different parts of the current work was flow system. Marine Environmental Research 46, 29–32.
Capotorti, G., Digianvincenzo, P., Cesti, P., Bernardi, C.P., Guglielmetti, G., 2004.
provided by the EU – Standards, Measurements and Testing pro-
Pyrene and benzo[a]pyrene metabolism by an Aspergillus terreus strain isolated
gram (contract SMT4-CT98-2250), the Norwegian Research Council from a polycyclic aromatic hydrocarbon polluted soil. Biodegradation 15, 79–85.
(projects 653879, 152449/720 and 153898/S40) and Total E&P Cerniglia, C.E., 1984. Microbial metabolism of polycyclic aromatic hydrocarbons.
Norge AS (TCP project). M.O. Sydnes (IRIS), G. Ylitalo and J. Stein Advances in Applied Microbiology 30, 31–71.
Cheevaporn, V., Beamish, F.W.H., 2007. Cytochrome P450 1A activity in liver and
(NOAA) are acknowledged for their valuable critical comments on fixed wavelength fluorescence detection of polycyclic aromatic hydrocarbons
the manuscript. in the bile of tongue-fish (Cynoglossus acrolepidotus, Bleeker) in relation to
petroleum hydrocarbons in the eastern Gulf of Thailand. Journal of Environ-
mental Biology 28, 701–705.
References Chen, G.S., White, P.A., 2004. The mutagenic hazards of aquatic sediments: a review.
Mutation Research-Reviews in Mutation Research 567, 151–225.
Aas, E., Baussant, T., Balk, L., Liewenborg, B., Andersen, O.K., 2000a. PAH metabolites Christensen, J.H., Tomasi, G., Strand, J., Andersen, O., 2009. PARAFAC modeling of
in bile, cytochrome P4501A and DNA adducts as environmental risk parameters fluorescence excitation-emission spectra of fish bile for rapid en route screening
for chronic oil exposure: a laboratory experiment with Atlantic cod. Aquatic of PAC exposure. Environmental Science & Technology 43, 4439–4445.
Toxicology 51, 241–258. Cofino, W.P., van Stokkum, I.H.M., Wells, D.E., Ariese, F., Wegener, J.W.M., Peerboom,
Aas, E., Beyer, J., Goksøyr, A., 2000b. Fixed wavelength fluorescence (FF) of bile as a R.A.L., 2000. A new model for the inference of population characteristics from
monitoring tool for polyaromatic hydrocarbon exposure in fish: an evaluation experimental data using uncertainties. Application to interlaboratory studies.
of compound specificity, inner filter effect and signal interpretation. Biomarkers Chemometrics and Intelligent Laboratory Systems 53, 37–55.
5, 9–23. Colavecchia, M.V., Hodson, P.V., Parrott, J.L., 2007. The relationships among CYP1A
Aas, E., Beyer, J., Jonsson, G., Reichert, W.L., Andersen, O.K., 2001. Evidence of uptake, induction, toxicity, and eye pathology in early life stages of fish exposed to oil
biotransformation and DNA binding of polyaromatic hydrocarbons in Atlantic sands. Journal of Toxicology and Environmental Health – Part a – Current Issues
cod and corkwing wrasse caught in the vicinity of an aluminium works. Marine 70, 1542–1555.
Environmental Research 52, 213–229. Collier, T.K., Varanasi, U., 1991. Hepatic activities of xenobiotic metabolizing
Aas, E., Klungsøyr, J., 1998. PAH metabolites in bile and EROD activity in North Sea enzymes and biliary levels of xenobiotics in English sole (Parophrys vetulus)
fish. Marine Environmental Research 46, 229–232. exposed to environmental contaminants. Archives of Environmental Contam-
Anderson, J.W., Lee, R.F., 2006. Use of biomarkers in oil spill risk assessment in the ination and Toxicology 20, 462–473.
marine environment. Human and Ecological Risk Assessment 12, 1192–1222. Colmsjö, A.L., Zebuhr, Y.U., Ostman, C.E., 1987. The effects of deuteration on Shpolskii
Andreoli, R., Manini, P., Bergamaschi, E., Mutti, A., Franchini, I., Niessen, W.M.A., spectra of some polyaromatic compounds. Chemica Scripta 27, 253–256.
1999. Determination of naphthalene metabolites in human urine by liquid Coman, V., Marutoiu, C., Puiu, S., 1997. Optimization of separation of some polycyclic
chromatography-mass spectrometry with electrospray ionization. Journal of aromatic compounds by thin-layer chromatography. Journal of Chromatography
Chromatography A 847, 9–17. A 779, 321–328.
Ariese, F., Beyer, J., Jonsson, G., Porte Visa, C., Krahn, M.M., 2005a. Review of analytical Cormier, S.M., Lin, E.L.C., Fulk, F., Subramanian, B., 2000. Estimation of exposure
methods for determining metabolites of polycyclic aromatic compounds (PACs) criteria values for biliary polycyclic aromatic hydrocarbon metabolite concen-
in fish bile. ICES-TIMES Techniques in Marine Environmental Sciences (39), 41. trations in white suckers (Catostomus commersoni). Environmental Toxicology
Ariese, F., Beyer, J., Wells, D., 2005b. Two fish bile reference materials certified for and Chemistry 19, 1120–1126.
PAH metabolites. Journal of Environmental Monitoring 7, 869–876. Day, B.W., Naylor, S., Gan, L.S., Sahali, Y., Nguyen, T.T., Skipper, P.L., Wishnok, J.S., Tan-
Ariese, F., Beyer, J., Wells, D., 2000. Preparation and certification of fish bile refer- nenbaum, S.R., 1991. Gas chromatographic-mass spectrometric analysis of diols
ence materials for exposure monitoring of PAHs in the aquatic environment, and tetrols from reactions of polycyclic aromatic hydrocarbon epoxides with
2nd progress report. Institute for Environmental Studies/IVM, Amsterdam, the hemoglobin. Journal of Chromatography. Biomedical Applications 562, 563–571.
Netherlands. Deshpande, A.D., Huggett, R.J., Halbrook, R.A., 2002. Polycyclic aromatic hydrocarbon
Ariese, F., Kok, S.J., Verkaik, M., Gooijer, C., Velthorst, N.H., Hofstraat, J.W., 1993a. metabolites in the bile of a territorial benthic fish, oyster toadfish (Opsanus tau)
Synchronous fluorescence spectrometry of fish bile – a rapid screening method from the Elizabeth River, Virginia. Archives of Environmental Contamination
for the biomonitoring of PAH exposure. Aquatic Toxicology 26, 273–286. and Toxicology 42, 43–52.
Ariese, F., Kok, S.J., Verkaik, M., Hoornweg, G.P., Gooijer, C., Velthorst, N.H., Hof- Dipple, A., Moschel, R.C., Bigger, C.A.H., 1984. Polynuclear aromatic carcinogens. In:
straat, J.W., 1993b. Chemical derivatization and Shpolskii spectrofluorometric Searle, C.E. (Ed.), Chemical Carcinogens, Second ed. American Chemical Society,
determination of benzo[a]pyrene metabolites in fish bile. Analytical Chemistry Washington, DC, pp. 41–163.
65, 1100–1106. Dissanayake, A., Galloway, T.S., 2004. Evaluation of fixed wavelength fluorescence
Barra, R., Sanchez-Hernandez, J.C., Orrego, R., Parra, O., Gavilan, J.F., 2001. Bioavail- and synchronous fluorescence spectrophotometry as a biomonitoring tool of
ability of PAHs in the Biobio river (Chile): MFO activity and biliary fluorescence environmental contamination. Marine Environmental Research 58, 281–285.
in juvenile Oncorhynchus mykiss. Chemosphere 45, 439–444. Doerge, D.R., Clayton, J., Fu, P.P., Wolfe, D.A., 1993. Analysis of polycyclic aromatic
Baumann, P.C., Harshbarger, J.C., 1995. Decline in liver neoplasms in wild brown hydrocarbons using liquid chromatography-particle beam mass spectrometry.
bullhead catfish after coking plant closes and environmental PAHs plummet. Biological Mass Spectrometry 22, 654–660.
Environmental Health Perspectives 103, 168–170. Eickhoff, C.V., Gobas, F., Law, F.C.P., 2003. Screening pyrene metabolites in the
Beyer, J., Sandvik, M., Hylland, K., Fjeld, E., Egaas, E., Aas, E., Skåre, J.U., Goksøyr, hemolymph of Dungeness crabs (Cancer magister) with synchronous fluo-
A., 1996. Contaminant accumulation and biomarker responses in flounder rescence spectrometry: method development and application. Environmental
(Platichthys flesus L) and Atlantic cod (Gadus morhua L) exposed by caging to Toxicology and Chemistry 22, 59–66.
polluted sediments in Sørfjorden, Norway. Aquatic Toxicology 36, 75–98. Escartin, E., Porte, C., 1999a. Assessment of PAH pollution in coastal areas from the
Beyer, J., Sandvik, M., Skåre, J.U., Egaas, E., Hylland, K., Waagbø, R., Goksøyr, A., 1997. NW mediterranean through the analysis of fish bile. Marine Pollution Bulletin
Time- and dose-dependent biomarker responses in flounder (Platichthys flesus 38, 1200–1206.
L) exposed to benzo[a]pyrene, 2,3,3 ,4,4 ,5-hexachlorobiphenyl (PCB-156) and Escartin, E., Porte, C., 1999b. Biomonitoring of PAH pollution in high-altitude moun-
cadmium. Biomarkers 2, 35–44. tain lakes through the analysis of fish bile. Environmental Science & Technology
Beyer, J., Aas, E., Borgenvik, H.K., Ravn, P., 1998. Bioavailability of PAH in effluent 33, 406–409.
water from an aluminium works evaluated by transplant caging and biliary Escartin, E., Porte, C., 1999c. Hydroxylated PAHs in bile of deep-sea fish. Relationship
fluorescence measurements of Atlantic cod (Gadus morhua L.). Marine Environ- with xenobiotic metabolizing enzymes. Environmental Science & Technology
mental Research 46, 233–236. 33, 2710–2714.
 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244 241

Evershed, R.P., 1993. Advances in silylation. In: Blau, K., Halket, J. (Eds.), Handbook of Haugland, C., Ugland, K.I., Børseth, J.F., Aas, E., 2005. Polycyclic aromatic hydrocar-
Derivatives for Chromatography, Second ed. John Wiley & Sons Ltd., Chichester, bons in capelin (Mallotus villosus) in the Barents Sea by use of fixed wavelength
pp. 51–108. fluorescence measurements of bile samples. Marine Pollution Bulletin 50,
Fenet, H., Gomez, E., Rosain, D., Casellas, C., 2006. Polycyclic aromatic hydrocar- 102–104.
bon metabolites and 7-ethoxyresorufin O-deethylase activity in caged European Hawkins, S.A., Billiard, S.M., Tabash, S.P., Brown, R.S., Hodson, P.V., 2002. Alter-
eels. Archives of Environmental Contamination and Toxicology 51, 232–236. ing cytochrome p4501a activity affects polycyclic aromatic hydrocarbon
Fernandes, D., Andreu-Sanchez, O., Bebianno, M.J., Porte, C., 2008a. Assessment of metabolism and toxicity in rainbow trout (Oncorhynchus mykiss). Environmental
pollution along the Northern Iberian shelf by the combined use of chemical and Toxicology and Chemistry 21, 1845–1853.
biochemical markers in two representative fish species. Environmental Pollu- He, S.X., Nicholson, R.A., Law, F.C.P., 1998. Benzo[a]pyrene toxicokinetics in the
tion 155, 327–335. cricket following injection into the heamolymph. Environmental Toxicology and
Fernandes, D., Zanuy, S., Bebianno, M.J., Porte, C., 2008b. Chemical and biochemical Pharmacology 6, 81–89.
tools to assess pollution exposure in cultured fish. Environmental Pollution 152, Heintz, R.A., Rice, S.D., Wertheimer, A.C., Bradshaw, R.F., Thrower, F.P., Joyce, J.E.,
138–146. Short, J.W., 2000. Delayed effects on growth and marine survival of pink salmon
Fernandes, M.B., Sicre, M.A., Boireau, A., Tronczynski, J., 1997. Aquatic hydrocar- Oncorhynchus gorbuscha after exposure to crude oil during embryonic develop-
bon distributions in the Seine estuary: biogenic polyaromatics and n-alkanes. ment. Marine Ecology-Progress Series 208, 205–216.
Estuaries 20, 281–290. Hellou, J., Leonard, J., 2004. Polycyclic aromatic hydrocarbons bioaccumulation and
Fernandez-Gonzalez, V., Muniategui-Lorenzo, S., Lopez-Mahia, P., Prada-Rodriguez, biotransformation products in trout exposed through food pellets. Polycyclic
D., 2008. Development of a programmed temperature vaporization-gas Aromatic Compounds 24, 697–712.
chromatography-tandem mass spectrometry method for polycyclic aromatic Hellou, J., Payne, J.F., 1986. Effect of petroleum hydrocarbons on the biliary bile
hydrocarbons analysis in biota samples at ultratrace levels. Journal of Chro- acid composition of rainbow trout (Salmo gairdneri). Comparative Biochemical
matography A 1207, 136–145. Physiology 84C, 257–261.
Ferrari, S., Mandel, F., Berset, J.D., 2002. Quantitative determination of 1- Hellou, J., Payne, J.F., 1987. Assessment of contamination of fish by water-soluble
hydroxypyrene in bovine urine samples using high-performance liquid fractions of petroleum: a role for bile metabolites. Environmental Toxicology
chromatography with fluorescence and mass spectrometric detection. Chemo- and Chemistry 6, 857–862.
sphere 47, 173–182. Hellou, J., Upshall, C., 1995. Monocyclic aromatic-hydrocarbons in bile of flounder
Ferrer, I., Barcelo, D., 1998. LC/MS for trace determination of pesticides in environ- exposed to a petroleum oil. International Journal of Environmental Analytical
mental samples. Analysis Magazine 26, M118–M122. Chemistry 60, 101–111.
Fillmann, G., Watson, G.M., Howsam, M., Francioni, E., Depledge, M.H., Readman, Hillenweck, A., Canlet, C., Mauffret, A., Debrauwer, L., Claireaux, G., Cravedi, J.P.,
J.W., 2004. Urinary PAH metabolites as biomarkers of exposure in aquatic envi- 2008. Characterization of biliary metabolites of fluoranthene in the common
ronments. Environmental Science & Technology 38, 2649–2656. sole (Solea solea). Environmental Toxicology and Chemistry 27, 2575–2581.
Fuentes-Rios, D., Orrego, R., Rudolph, A., Mendoza, G., Gavilan, J.F., Barra, R., 2005. Holth, T.F., Beylich, B.A., Skarphedinsdottir, H., Liewenborg, B., Grung, M., Hylland, K.,
EROD activity and biliary fluorescence in Schroederichthys chilensis (Guichenot 2009. Genotoxicity of environmentally relevant concentrations of water-soluble
1848): biomarkers of PAH exposure in coastal environments of the South Pacific oil components in Cod (Gadus morhua). Environmental Science & Technology 43,
Ocean. Chemosphere 61, 192–199. 3329–3334.
Gagnon, M.M., Holdway, D.A., 2002. EROD activity, serum SDH and PAH biliary Hornung, M.W., Cook, P.M., Fitzsimmons, P.N., Kuehl, D.W., Nichols, J.W., 2007. Tis-
metabolites in sand flathead (Platycephalus bassensis) collected in Port Phillip sue distribution and metabolism of benzo[a]pyrene in embryonic and larval
Bay, Australia. Marine Pollution Bulletin 44, 230–237. medaka (Oryzias latipes). Toxicological Sciences 100, 393–405.
Galceran, M.T., Moyano, E., 1996. Determination of hydroxy polycyclic aromatic Ibe, A.C., Kullenberg, G., 1995. Quality assurance quality control (QA/QC) regime
hydrocarbons by liquid chromatography mass spectrometry – comparison of in marine pollution monitoring programmes: the GIPME perspective. Marine
atmospheric pressure chemical ionization and electrospray. Journal of Chro- Pollution Bulletin 31, 209–213.
matography A 731, 75–84. Incardona, J.P., Collier, T.K., Scholz, N.L., 2004. Defects in cardiac function precede
Garcia-Falcon, M.S., Simal-Gandara, J., Carril-Gonzalez-Barros, S.T., 2000. Analysis of morphological abnormalities in fish embryos exposed to polycyclic aromatic
benzo[a] pyrene in spiked fatty foods by second derivative synchronous spec- hydrocarbons. Toxicology and Applied Pharmacology 196, 191–205.
trofluorimetry after microwave-assisted treatment of samples. Food Additives Iqbal, J., Overton, E.B., Gisclair, D., 2008a. Polycyclic aromatic hydrocarbons in
and Contaminants 17, 957–964. Louisiana rivers and coastal environments: source fingerprinting and forensic
Garrigues, P., Budzinski, H., 1995. Recent analytical advances in Shpolskii spec- analysis. Environmental Forensics 9, 63–74.
troscopy. Trac-Trends in Analytical Chemistry 14, 231–239. Iqbal, J., Overton, E.B., Gisclair, D., 2008b. Sources of polycyclic aromatic hydro-
Garrigues, P., Ewald, M., 1987. High-resolution emission-spectroscopy (Shpol- carbons in Louisiana Rivers and coastal environments: principal components
skii effect) – a new analytical technique for the analysis of polycyclic analysis. Environmental Forensics 9, 310–319.
aromatic-hydrocarbons (PAH) in the environmental-samples. Chemosphere 16, Jacob, J., Grimmer, G., 1988. Capillary gas chromatographical analysis and mass spec-
485–494. trometric identification of polycyclic aromatic hydrocarbon metabolites from
Giessing, A.M.B., Mayer, L.M., Forbes, T.L., 2003. Synchronous fluorescence spectrom- biological materials. Review of Analytical Chemistry 9, 49–89.
etry of 1-hydroxypyrene: a rapid screening method for identification of PAH James, M.O., 1989. Biotransformation and disposition of PAH in aquatic inverte-
exposure in tissue from marine polychaetes. Marine Environmental Research brates. In: Varanasi, U. (Ed.), Metabolism of Polycyclic Aromatic Hydrocarbons
56, 599–615. in the Aquatic Environment. CRC Press, Boca Raton, FL, pp. 69–92.
Goanvec, C., Theron, M., Lacoue-Labarthe, T., Poirier, E., Guyomarch, J., Le-Floch, S., Johnson, L.L., Ylitalo, G.M., Arkoosh, M.R., Kagley, A.N., Stafford, C., Bolton, J.L., Buzi-
Laroche, J., Nonnotte, L., Nonnotte, G., 2008. Flow cytometry for the evaluation tis, J., Anulacion, B.F., Collier, T.K., 2007. Contaminant exposure in outmigrant
of chromosomal damage in turbot Psetta maxima (L.) exposed to the dissolved juvenile salmon from Pacific Northwest estuaries of the United States. Environ-
fraction of heavy fuel oil in sea water: a comparison with classical biomarkers. mental Monitoring and Assessment 124, 167–194.
Journal of Fish Biology 73, 395–413. Jonsson, G., Beyer, J., Wells, D., Ariese, F., 2003. The application of HPLC-F and GC-MS
Gorbi, S., Baldini, C., Regoli, F., 2005. Seasonal variability of metallothioneins, to the analysis of selected hydroxy polycyclic hydrocarbons in two certified fish
cytochrome P450, bile metabolites and oxyradical metabolism in the European bile reference materials. Journal of Environmental Monitoring 5, 513–520.
eel Anguilla anguilla L. (Anguillidae) and striped mullet Mugil cephalus L. (Mugili- Jonsson, G., Sundt, R.C., Aas, E., Beyer, J., 2004a. An evaluation of two fluorescence
dae). Archives of Environmental Contamination and Toxicology 49, 62–70. screening methods for the determination of chrysene metabolites in fish bile.
Gravato, C., Santos, M.A., 2003. Dicentrarchus labrax biotransformation and genotox- Chemosphere 56, 81–90.
icity responses after exposure to a secondary treated industrial/urban effluent. Jonsson, G., Taban, I.C., Jorgensen, K.B., Sundt, R.C., 2004b. Quantitative determina-
Ecotoxicology and Environmental Safety 55, 300–306. tion of de-conjugated chrysene metabolites in fish bile by HPLC-fluorescence
Grung, M., Holth, T.F., Jacobsen, M.R., Hylland, K., 2009. Polycyclic aromatic hydro- and GC-MS. Chemosphere 54, 1085–1097.
carbon (PAH) metabolites in Atlantic cod exposed via water or diet to a synthetic Kammann, U., 2007. PAH metabolites in bile fluids of dab (Limanda limanda) and
produced water. Journal of Toxicology and Environmental Health – Part a – flounder (Platichthys flesus): spatial distribution and seasonal changes. Environ-
Current Issues 72, 254–265. mental Science and Pollution Research 14, 102–108.
Hanson, N., Guttman, E., Larsson, A., 2006. The effect of different holding conditions Kawamoto, T., Yang, M.H., Kim, Y.D., Kim, H., Oyama, T., Isse, T., Matsuno, K., Katoh,
for environmental monitoring with caged rainbow trout (Oncorhynchus mykiss). T., Uchiyama, I., 2007. Effects of lifestyle on urinary 1-hydroxypyrene concen-
Journal of Environmental Monitoring 8, 994–999. tration. Journal of Occupational Health 49, 183–189.
Hanson, N., Larsson, A., 2009. Experiences from a biomarker study on farmed Kennedy, C.J., Gill, K.A., Walsh, P.J., 1989. Thermal modulation of benzo[a]pyrene
rainbow trout (Oncorhynchus Mykiss) used for environmental monitoring in a metabolism by the gulf toadfish, Opsanus beta. Aquatic Toxicology 15, 331–344.
Swedish River. Environmental Toxicology and Chemistry 28, 1536–1545. Koeber, R., Niessner, R., Bayona, J.M., 1997. Comparison of liquid chromatography-
Hanson, N., Persson, S., Larsson, A., 2009. Analyses of perch (Perca fluviatilis) bile mass spectrometry interfaces for the analysis of polar metabolites of
suggest increasing exposure to PAHs and other pollutants in a reference area on benzo[a]pyrene. Fresenius Journal of Analytical Chemistry 359, 267–273.
the Swedish Baltic coast. Journal of Environmental Monitoring 11, 389–393. Koenig, S., Savage, C., Kim, J.P., 2008. Non-destructive assessment of polycyclic aro-
Harman, C., Holth, T.F., Hylland, K., Thomas, K., Grung, M., 2009. Relationship matic hydrocarbon (PAH) exposure by fluorimetric analysis of crab urine. Marine
between polycyclic aromatic hydrocarbon (PAH) accumulation in semiperme- Pollution Bulletin 56, 2003–2008.
able membrane devices and PAH bile metabolite levels in Atlantic cod (Gadus Krahn, M.M., Burrows, D.G., MacLeod, W.D., Malins, D.C., 1987. Determination of
morhua). Journal of Toxicology and Environmental Health – Part a – Current individual metabolites of aromatic compounds in hydrolyzed bile of English sole
Issues 72, 234–243. (Parophrys vetulus) from polluted sites in Puget Sound, Washington. Archives of
Harvey, R.G., 1997. Polycyclic Aromatic Hydrocarbons. Wiley, New York, NY. Environmental Contamination and Toxicology 16, 511–522.
242 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244

Krahn, M.M., Burrows, D.G., Ylitalo, G.M., Brown, D.W., Wigren, C.A., Collier, T.K., Malins, D.C., McCain, B.B., Landahl, J.T., Myers, M.S., Krahn, M.M., Brown, D.W., Chan,
Chan, S.L., Varanasi, U., 1992. Mass spectrometric analysis for aromatic com- S.-L., Roubal, W.T., 1988. Neoplastic and other diseases in fish in relation to toxic
pounds in bile of fish sampled after the Exxon Valdez oil spill. Environmental chemicals: an overview. Aquatic Toxicology 11, 43–67.
Science & Technology 26, 116–126. Martinez-Lara, E., Leaver, M., George, S., 2002. Evidence from heterologous expres-
Krahn, M.M., Myers, M.S., Burrows, D.G., Malins, D.C., 1984. Determination of sion of glutathione S-transferases A and A1 of the plaice (Pleuronectes platessa)
metabolites of xenobiotics in the bile of fish from polluted waterways. Xeno- that their endogenous role is in detoxification of lipid peroxidation products.
biotica 14, 633–646. Marine Environmental Research 54, 263–266.
Krahn, M.M., Rhodes, L.D., Myers, M.S., Moore, L.K., Macleod, W.D., Malins, D.C., McDonald, S.J., Kennicutt, M.C., Liu, H., Safe, S.H., 1995. Assessing aromatic
1986. Associations between metabolites of aromatic compounds in bile and hydrocarbon exposure in Antarctic fish captured near Palmer and McMurdo
the occurrence of hepatic lesions in English sole (Parophrys vetulus) from Puget stations Antarctica. Archives of Environmental Contamination and Toxicology
Sound, Washington. Archives of Environmental Contamination and Toxicology 29, 232–240.
15, 61–67. Moyano, E., McCullagh, M., Galceran, M.T., Games, D.E., 1997. Supercritical fluid
Krahn, M.M., Schnell, J.V., Uyeda, M.Y., Macleod Jr., W.D., 1981. Determination of chromatography-atmospheric pressure chemical ionisation mass spectrome-
mixtures of benzo[a]pyrene, 2,6-dimethylnaphthalene and their metabolites by try for the analysis of hydroxy polycyclic aromatic hydrocarbons. Journal of
high-performance liquid chromatography with fluorescence detection. Analyt- Chromatography 777A, 167–176.
ical Biochemistry 113, 27–33. Mullins, E., 1994. Introduction to control charts in the analytical laboratory – tutorial
Krahn, M.M., Ylitalo, G.M., Buzitis, J., Chan, S.L., Varanasi, U., 1993a. Rapid review. Analyst 119, 369–375.
high-performance liquid chromatographic methods that screen for aromatic Myers, M.S., Anulacion, B.F., French, B.L., Reichert, W.L., Laetz, C.A., Buzitis, J., Olson,
compounds in environmental samples. Journal of Chromatography 642, O.P., Sol, S., Collier, T.K., 2008. Improved flatfish health following remediation of
15–32. a PAH-contaminated site in Eagle Harbor, Washington. Aquatic Toxicology 88,
Krahn, M.M., Ylitalo, G.M., Buzitis, J., Chan, S.L., Varanasi, U., Wade, T.L., Jack- 277–288.
son, T.J., Brooks, J.M., Wolfe, D.A., Manen, C.A., 1993b. Comparison of Myers, M.S., Landahl, J.T., Krahn, M.M., McCain, B.B., 1991. Relationships between
high-performance liquid chromatography/fluorescence screening and gas chro- hepatic neoplasms and related lesions and exposure to toxic chemicals in marine
matography/mass spectrometry analysis for aromatic compounds in sediments fish from the United States west coast. Environmental Health Perspectives 90,
sampled after the Exxon Valdez oil spill. Environmental Science & Technology 27, 7–15.
699–708. Myers, M.S., Stehr, C.M., Olson, O.P., Johnson, L.L., Mccain, B.B., Chan, S.L., Varanasi,
Krca, S., Zaja, R., Calic, V., Terzic, S., Grubesic, M.S., Ahel, M., Smital, T., 2007. U., 1994. Relationships between toxicopathic hepatic-lesions and exposure to
Hepatic biomarker responses to organic contaminants in feral chub (leuciscus chemical contaminants in English sole (Pleuronectes-vetulus), starry flounder
cephalus)-laboratory characterization and field study in the Sava River, Croatia. (Platichthys-stellatus), and white croaker (Genyonemus-lineatus) from selected
Environmental Toxicology and Chemistry 26, 2620–2633. marine sites on the pacific coast, USA. Environmental Health Perspectives 102,
Krone, C.A., Stein, J.E., Varanasi, U., 1992. Estimation of levels of metabolites of 200–215.
aromatic-hydrocarbons in fish-tissues by HPLC/Fluorescence analysis. Chemo- Nacher-Mestre, J., Serrano, R., Portoles-Nicolau, T., Hernandez, F., Benedito-Palos,
sphere 24, 497–510. L., Perez-Sanchez, J., 2009. A reliable analytical approach based on gas chro-
Kuijt, J., Garcia-Ruiz, C., Stroomberg, G.J., Marina, M.L., Ariese, F., Brinkman, U.A.T., matography coupled to triple quadrupole and time-of-flight mass analyzers
Gooijer, C., 2001. Laser-induced fluorescence detection at 266 nm in capillary for the determination and confirmation of polycyclic aromatic hydrocarbons in
electrophoresis – polycyclic aromatic hydrocarbon metabolites in biota. Journal complex matrices from aquaculture activities. Rapid Communications in Mass
of Chromatography A 907, 291–299. Spectrometry 23, 2075–2086.
Law, F.C.P., Meng, J.X., He, Y.T., Chui, Y.C., 1994. Urinary and biliary metabolites of Neff, J.M., 1979. Polycyclic Aromatic Hydrocarbons in the Aquatic Environment:
pyrene in rainbow trout (Oncorhynchus mykiss). Xenobiotica 24, 221–229. Sources, Fates and Biological Effects. Applied Science Publishers Ltd., London,
Law, R.J., 2000. The analysis of polycyclic aromatic hydrocarbons in marine samples. UK.
International Journal of Environment and Pollution 13, 262–283. Neff, J.M., 1985. Polycyclic aromatic hydrocarbons. In: Rand, G.M., Petrocelli, S.R.
Law, R.J., Kelly, C.A., Nicholson, M.D., 2000. The QUASIMEME laboratory performance (Eds.), Fundamentals of Aquatic Toxicology, Methods and Applications. Hemi-
study of polycyclic aromatic hydrocarbons (PAHs): assessment for the period sphere, New York, NY, pp. 416–454.
1996–1999. Journal of Environmental Monitoring 2, 517–523. Neves, R.L.S., Oliveira, T.F., Ziolli, R.L., 2007. Polycyclic aromatic hydrocarbons (PAHs)
Leaver, M.J., Clarke, D.J., George, S.G., 1992a. Molecular studies of the phase-II xeno- in fish bile (Mugil liza) as biomarkers for environmental monitoring in oil con-
biotic conjugative enzymes of marine pleuronectid flatfish. Aquatic Toxicology taminated areas. Marine Pollution Bulletin 54, 1818–1824.
22, 265–278. Niessen, W.M.A., 1999. State-of-the-art in liquid chromatography-mass spectrom-
Leaver, M.J., George, S.G., 1998. A piscine glutathione S-transferase which efficiently etry. Journal of Chromatography A 856, 179–197.
conjugates the end-products of lipid peroxidation. Marine Environmental Nikolaou, A., Kostopoulou, M., Lofrano, G., Meric, S., Petsas, A., Vagi, M., 2009. Levels
Research 46, 71–74. and toxicity of polycyclic aromatic hydrocarbons in marine sediments. Trac-
Leaver, M.J., Scott, K., George, S.G., 1992b. Expression and tissue distribution of plaice Trends in Analytical Chemistry 28, 653–664.
glutathione S-transferase-A. Marine Environmental Research 34, 237–241. Oehme, M., Berger, U., Brombacher, S., Kuhn, F., Kolliker, F., 2002. Trace analysis
Leaver, M.J., Wright, J., Hodgson, P., Boukouvala, E., George, S.G., 2007. Piscine UDP- by HPLC-MS: contamination problems and systematic errors. Trac-Trends in
glucuronosyltransferase 1B. Aquatic Toxicology 84, 356–365. Analytical Chemistry 21, 322–331.
Lech, J.J., Vodicnik, M.J., 1985. Biotransformation. In: Rand, G.M., Petrocelli, S.R. (Eds.), Oikari, A.O.J., 1986. Metabolites of xenobiotics in the bile of fish in waterways
Fundamentals of Aquatic Toxicology, Methods and Applications. Hemisphere, polluted by pulpmill effluents. Bulletin of Environmental Contamination and
New York, NY, pp. 526–557. Toxicology 36, 429–436.
Lee, R.F., Anderson, J.W., 2005. Significance of cytochrome P450 system responses Opuene, K., Agbozu, I.E., Iwuozor, C.C., 2007. Identification of perylene in sediments:
and levels of bile fluorescent aromatic compounds in marine wildlife following occurrence and diagenetic evolution. International Journal of Environmental
oil spills. Marine Pollution Bulletin 50, 705–723. Science and Technology 4, 457–462.
Lee, R.F., Sauerheber, R., Dobbs, G.H., 1972. Uptake, metabolism and discharge of Parris, R.M., Schantz, M.M., Wise, S.A., 2002. NIST Intercomparison Exercise Pro-
polycyclic aromatic hydrocarbons by marine fish. Marine Biology 17, 201–208. gram for Organic Contaminants in the Marine Environment. National Institute
Leonard, J.D., Hellou, J., 2001. Separation and characterization of gall bladder bile of Standards and Technology, Gaithersburg, MD, p. 250.
metabolites from speckled trout, Salvelinus fontinalis, exposed to individual Pies, C., Ternes, T.A., Hofmann, T., 2008. Identifying sources of polycyclic aromatic
polycyclic aromatic compounds. Environmental Toxicology and Chemistry 20, hydrocarbons (PAHs) in soils: distinguishing point and non-point sources using
618–623. an extended PAH spectrum and n-alkanes. Journal of Soils and Sediments 8,
Lin, E.L.C., Cormier, S.M., Racine, R.N., 1994. Synchronous fluorometric measurement 312–322.
of metabolites of polycyclic aromatic hydrocarbons in the bile of brown bullhead. Pikkarainen, A.L., 2006. Ethoxyresorufin-O-deethylase (EROD) activity and bile
Environmental Toxicology and Chemistry 13, 707–715. metabolites as contamination indicators in Baltic Sea perch: determination by
Lin, E.L.C., Cormier, S.M., Torsella, J.A., 1996. Fish biliary polycyclic aromatic hydro- HPLC. Chemosphere 65, 1888–1897.
carbon metabolites estimated by fixed-wavelength fluorescence: comparison Plummer, J.L., Smith, B.R., Ball, L.M., Bend, J.R., 1980. Metabolism and biliary excre-
with HPLC-fluorescent detection. Ecotoxicology and Environmental Safety 35, tion of benzo[a]pyrene 4,5-oxide in the rat. Drug Metabolism and Disposition 8,
16–23. 68–72.
Lloyd, J.B.F., 1971. Synchronized excitation of fluorescence emission spectra. Nature Poster, D.L., Schantz, M.M., Sander, L.C., Wise, S.A., 2006. Analysis of polycyclic aro-
Physical Science 231, 64–65. matic hydrocarbons (PAHs) in environmental samples: a critical review of gas
Lyons, B.P., Stewart, C., Kirby, M.F., 1999. The detection of biomarkers of geno- chromatographic (GC) methods. Analytical and Bioanalytical Chemistry 386,
toxin exposure in the European flounder (Platichthys flesus) collected from the 859–881.
River Tyne Estuary. Mutation Research-Genetic Toxicology and Environmental Prabhu, S.S., Montgomery, D.C., Runger, G.C., 1994. A combined adaptive sample-size
Mutagenesis 446, 111–119. and sampling interval (X)over-bar control scheme. Journal of Quality Technology
Mahanty, B., Pakshirajan, K., Dasu, V.V., 2008. Synchronous fluorescence as a 26, 164–176.
selective method for monitoring pyrene in biodegradation studies. Polycyclic Reichert, W.L., Myers, M.S., Peck-Miller, K., French, B., Anulacion, B.F., Collier, T.K.,
Aromatic Compounds 28, 213–227. Stein, J.E., Varanasi, U., 1998. Molecular epizootiology of genotoxic events in
Malins, D.C.B.B.M., Myers, M.S., Brown, D.W., Krahn, M.M., Roubal, W.T., Schiewe, marine fish: linking contaminant exposure, DNA damage, and tissue-level alter-
M.H., Landahl, J.T., Chan, S.-L., 1987. Field and laboratory studies of the etiol- ations. Mutation Research 411, 215–225.
ogy of liver neoplasms in marine fish from Puget Sound. Environmental Health Ribeiro, C.A.O., Vollaire, Y., Sanchez-Chardi, A., Roche, H., 2005. Bioaccumulation
Perspectives 71, 5–16. and the effects of organochlorine pesticides, PAH and heavy metals in the Eel
 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244 243

(Anguilla anguilla) at the Camargue Nature Reserve, France. Aquatic Toxicology Tyrpien, K., 1996. Analysis of hydroxy-PAH by thin layer chromatography and OPLC.
74, 53–69. Journal of Planar Chromatography 9, 203–207.
Richardson, D.M., Davies, I.M., Moffat, C.F., Pollard, P., Stagg, R.M., 2001. Biliary PAH Utvik, T.I.R., 1999. Chemical characterisation of produced water from four offshore
metabolites and EROD activity in flounder (Platichthys flesus) from a contami- oil production platforms in the North Sea. Chemosphere 39, 2593–2606.
nated estuarine environment. Journal of Environmental Monitoring 3, 610–615. Utvik, T.I.R., Durell, G.S., Johnsen, S., 1999. Determining produced water originating
Robertson, A., 1998. Petroleum hydrocarbons. In: Wilson, S.J., Murray, J.L., Hunt- polycyclic aromatic hydrocarbons in North Sea waters: comparison of sampling
ington, H.P. (Eds.), AMAP Assessment Report: Arctic Pollution Issues. Arctic techniques. Marine Pollution Bulletin 38, 977–989.
Monitoring and Assessment Programme (AMAP), Oslo, Norway, pp. 661–716. van der Oost, R., Beyer, J., Vermeulen, N.P.E., 2003. Fish bioaccumulation and
Rowles, T., Becker, P.R., Christopher, S.J., Kucklick, J.R., Mackey, E.A., Porter, B.J., biomarkers in environmental risk assessment: a review. Environmental Toxi-
Pugh, R.S., Schantz, M.M., Wise, S.A., 2002. Description and Results of the 2000 cology and Pharmacology 13, 57–149.
NIST/NOAA Interlaboratory Comparison Exercise Program for Organic Contam- van der Oost, R., van Schooten, F.-J., Ariese, F., Heida, H., Satumalay, K., Vermeulen, N.,
inants and Trace Elements in Marine Mammal Tissues. National Institute of 1994a. Bioaccumulation, biotransformation and DNA binding of PAHs in feral eel
Standards and Technology, Gaithersburg, MD, p. 155. (Anguilla anguilla) exposed to polluted sediments: a field survey. Environmental
Ruddock, P.J., Bird, D.J., McCalley, D.V., 2002. Bile metabolites of polycyclic aromatic Toxicology and Chemistry 13, 859–870.
hydrocarbons in three species of fish from the Severn Estuary. Ecotoxicology van der Oost, R., Vangastel, L.D.W., Hanraads, M., Satumalay, K., Vanschooten,
and Environmental Safety 51, 97–105. F., Heida, H., Vermeulen, N., 1994b. Biochemical markers in feral roach
Ruddock, P.J., Bird, D.J., McEvoy, J., Peters, L.D., 2003. Bile metabolites of polycyclic (Rutilus-rutilus) in relation to the bioaccumulation of organic trace pollutants.
aromatic hydrocarbons (PAHs) in European eels Anguilla anguilla from United Chemosphere 29, 801–817.
Kingdom estuaries. Science of the Total Environment 301, 105–117. van Schanke, A., Holtz, F., van der Meer, J., Boon, J.P., Ariese, F., Stroomberg, G., van
Sandvik, M., Beyer, J., Goksøyr, A., Hylland, K., Egaas, E., Skaare, J.U., 1997. den Berg, M., Everaarts, J.M., 2001. Dose- and time-dependent formation of bil-
Interaction of benzo[a]pyrene, 2,3,3 ,4,4 ,5-hexachlorobiphenyl (PCB-156) and iary benzo a pyrene metabolites in the marine flatfish dab (Limanda limanda).
cadmium on biomarker responses in flounder (Platichthys flesus L). Biomarkers 2, Environmental Toxicology and Chemistry 20, 1641–1647.
153–160. Varanasi, U., Gmur, D.J., 1981. Hydrocarbons and metabolites in English sole
Schipper, C.A., Lahr, J., van den Brink, P.J., George, S.G., Hansen, P.D., de Assis, H.C.D., (Parophrys vetulus) exposed simultaneously to [3 H] benzol[a] pyrene and [14 C]
van der Oost, R., Thain, J.E., Livingstone, D., Mitchelmore, C., van Schooten, F.J., naphthalene in oil-contaminated sediment. Aquatic Toxicology 1, 49–67.
Ariese, F., Murk, A.J., Grinwis, G.C.M., Klamer, H., Kater, B.J., Postma, J.F., van der Varanasi, U., Gmur, D.J., Treseler, P.A., 1979. Influence of time and mode of exposure
Werf, B., Vethaak, A.D., 2009. A retrospective analysis to explore the applica- on biotransformation of naphthalene by juvenile starry flounder (Platichthys
bility of fish biomarkers and sediment bioassays along contaminated salinity stellatus) and rock sole (Lepidopsetta bilineata). Archives Environmental Con-
transects. ICES Journal of Marine Science 66, 2089–2105. tamination and Toxicology 8, 673–692.
Sharma, A., Schulman, S.G., 1999. Introduction to Fluorescence Spectroscopy. Wiley, Varanasi, U., Reichert, W.L., Stein, J.E., Brown, D.W., Sanborn, H.R., 1985. Bioavail-
New York, NY. ability and biotransformation of aromatic hydrocarbons in benthic organisms
Simpson, C.D., Cullen, W.R., He, T.Y.T., Ikonomou, M.K.J.R., 2002. Metabolism of exposed to sediment from an urban estuary. Environmental Science & Technol-
pyrene by two clam species. Mya arenaria and Protothaca staminea. Chemo- ogy 19, 836–841.
sphere 49, 315–322. Varanasi, U., Stein, J.E., Nishimoto, M., Hom, T., 1983. Benzo(a)pyrene Metabolites in
Simpson, C.D., Wu, M.-T., Christiani, D.C., Santella, R.M., Carmella, S.G., Hecht, S.S., Liver, Muscle, Gonads and Bile of Adult English Sole (Parophrys vetulus). Batelle
2000. r-7,t-8,9,c-10-tetrahydroxy-7,8,9,10-tetahydrobenzo[a]pyrene in human Press, Columbus, OH.
urine by gas chromatography/negative ion chemical ionization/mass spectrom- Varanasi, U., Stein, J.E., Nishimoto, M., Reichert, W.L., Collier, T.K., 1987. Chemical
etry. Chemical Research in Toxicology 13, 271–280. carcinogenesis in feral fish – Uptake, activation, and detoxication of organic
Solbakken, J.E., Palmork, K.H., 1981. Metabolism of phenanthrene in various marine xenobiotics. Environmental Health Perspectives 71, 155–170.
animals. Comparative Biochemistry and Physiology 70C, 21–26. Varanasi, U., Stein, J.E., Reichert, W.L., Tilbury, K.L., Krahn, M.M., Chan, S.-L., 1992.
Solbakken, J.E., Palmork, K.H., 1984. A note on the predominance of non-K-region Chlorinated and aromatic hydrocarbons in bottom sediments, fish and marine
metabolites of phenanthrene found in bony fishes. FiskDir. Skr. Ser. HavUnders. mammals in US coastal waters: laboratory and field studies of metabolism and
17, 499–500. accumulation. In: Walker, C.H., Livingstone, D.R. (Eds.), Persistent Pollutants in
Solbakken, J.E., Palmork, K.H., Neppelberg, T., Scheline, R.R., 1980. Urinary and biliary Marine Ecosystems. Pergamon Press, New York, NY, pp. 83–115.
metabolites of phenanthrene in the coalfish (Pollachius virens). Acta Pharmacol- Vatsavai, K., Goicoechea, H.C., Campiglia, A.D., 2008. Direct quantification of
ogy and Toxicology 46, 127–132. monohydroxy-polycyclic aromatic hydrocarbons in synthetic urine samples
Statham, C.N., Melancon Jr., M.J., Lech, J.J., 1976. Bioconcentration of xenobiotics in via solid-phase extraction-room-temperature fluorescence excitation-emission
trout bile: a proposed monitoring aid for some waterborne chemicals. Science matrix spectroscopy. Analytical Biochemistry 376, 213–220.
193, 680–681. Vethaak, A.D., Jol, J.G., Meijboom, A., Eggens, M.L., ApRheinallt, T., Wester, P.W.,
Stephensen, E., Adolfsson-Erici, M., Celander, M., Hulander, M., Parkkonen, J., vandeZande, T., Bergman, A., Dankers, N., Ariese, F., Baan, R.A., Everts, J.M.,
Hegelund, T., Sturve, J., Hasselberg, L., Bengtsson, M., Förlin, L., 2003. Biomarker Opperhuizen, A., Marquenie, J.M., 1996. Skin and liver diseases induced in floun-
responses and chemical analyses in fish indicate leakage of polycyclic aromatic der (Platichthys flesus) after long-term exposure to contaminated sediments in
hydrocarbons and other compounds from car tire rubber. Environmental Toxi- large-scale mesocosms. Environmental Health Perspectives 104, 1218–1229.
cology and Chemistry 22, 2926–2931. Villeneuve, J.P., de Mora, S., Cattini, C., 2004. Determination of organochlorinated
Stephensen, E., Svavarsson, J., Sturve, J., Ericson, G., Adolfsson-Erici, M., Förlin, L., compounds and petroleum hydrocarbons in fish-homogenate sample IAEA-406:
2000. Biochemical indicators of pollution exposure in shorthorn sculpin (Myox- results from a worldwide interlaboratory study. Trac-Trends in Analytical Chem-
ocephalus scorpius), caught in four harbours on the southwest coast of Iceland. istry 23, 501–510.
Aquatic Toxicology 48, 431–442. Voyksner, R., 1994. Atmospheric pressure ionization LC/MS. New solutions for envi-
Steward, A.R., Kandaswami, C., Chidambaram, S., Ziper, C., Rutkowski, J.P., Sikka, H.C., ronmental analysis. Environmental Science & Technology 28, 118A–127A.
1990. Disposition and metabolic fate of benzo[a]pyrene in the brown bullhead. Vuontisjarvi, H., Keinanen, M., Vuorinen, P.J., Peltonen, K., 2004. A comparison of
Environmental Toxicology and Chemistry 9, 1503–1512. HPLC with fluorescence detection and fixed wavelength fluorescence methods
Stroomberg, G.J., Ariese, F., van Gestel, C.A.M., van Hattum, B., Velthorst, N.H., for the determination of polycyclic aromatic hydrocarbon metabolites in fish
van Straalen, N.M., 2003. Pyrene biotransformation products as biomarkers of bile. Polycyclic Aromatic Compounds 24, 333–342.
polycyclic aromatic hydrocarbon exposure in terrestrial Isopoda: concentration- Vuorinen, P.J., Keinanen, M., Vuontisjarvi, H., Barsiene, J., Broeg, K., Förlin, L., Gercken,
response relationship, and field study in a contaminated forest. Environmental J., Kopecka, J., Kohler, A., Parkkonen, J., Pempkowiak, J., Schiedek, D., 2006. Use
Toxicology and Chemistry 22, 224–231. of biliary PAH metabolites as a biomarker of pollution in fish from the Baltic Sea.
Stroomberg, G.J., Zappey, H., Steen, R., van Gestel, C.A.M., Ariese, F., Velthorst, N.H., Marine Pollution Bulletin 53, 479–487.
van Straalen, N.M., 2004. PAH biotransformation in terrestrial invertebrates – a Väänänen, V., Hämeilä, M., Kontsas, H., Peltonen, K., Heikkilä, P., 2003. Air concen-
new phase II metabolite in isopods and springtails. Comparative Biochemistry trations and urinary metabolites of polycyclic aromatic hydrocarbons among
and Physiology C – Toxicology & Pharmacology 138, 129–137. paving and remixing workers. Journal of Environmental Monitoring 5, 739–746.
Tairova, Z.M., Giessing, A.M.B., Hansen, R., Andersen, O., 2009. 1-Hydroxypyrene as a Wang, X.H., Hong, H.S., Mu, J.L., Lin, J.Q., Wang, S.H., 2008. Polycyclic aromatic hydro-
biomarker of PAH exposure in the marine polychaete Nereis diversicolor. Marine carbon (PAH) metabolites in marine fishes as a specific biomarker to indicate
Environmental Research 67, 38–46. PAH pollution in the marine coastal environment. Journal of Environmental
Tan, Y.L., Kong, A., Monetti, M.A., 1996. Biogenic polycyclic aromatic hydrocarbons Science and Health – Part a – Toxic/Hazardous Substances & Environmental
in an Alaskan Arctic Lake sediment. Polycyclic Aromatic Compounds 9, 185–192. Engineering 43, 219–226.
Troisi, G.M., Bexton, S., Robinson, I., 2006. Polyaromatic hydrocarbon and PAH Wang, Z.D., Fingas, M., Page, D.S., 1999. Oil spill identification. Journal of Chromatog-
metabolite burdens in oiled common guillemots (Uria aalge) stranded on the raphy A 843, 369–411.
East Coast of England (2001–2002). Environmental Science & Technology 40, Wang, Z.D., Fingas, M.F., 2003. Development of oil hydrocarbon fingerprinting and
7938–7943. identification techniques. Marine Pollution Bulletin 47, 423–452.
Tuvikene, A., 1995. Responses of fish to polycyclic aromatic-hydrocarbons (PAHs). Wang, Z.D., Yang, C., Kelly-Hooper, F., Hollebone, B.R., Peng, X., Brown, C.E., Lan-
Annales Zoologici Fennici 32, 295–309. driault, M., Sun, J., Yang, Z., 2009. Forensic differentiation of biogenic organic
Tyler, C.R., Filby, A.L., Bickley, L.K., Cumming, R.I., Gibson, R., Labadie, P., Katsu, Y., compounds from petroleum hydrocarbons in biogenic and petrogenic com-
Liney, K.E., Shears, J.A., Silva-Castro, V., Urushitani, H., Lange, A., Winter, M.J., pounds cross-contaminated soils and sediments. Journal of Chromatography A
Iguchi, T., Hill, E.M., 2009. Environmental health impacts of equine estrogens 1216, 1174–1191.
derived from hormone replacement therapy. Environmental Science & Tech- Watson, G.M., Andersen, O.K., Galloway, T.S., Depledge, M.H., 2004. Rapid
nology 43, 3897–3904. assessment of polycyclic aromatic hydrocarbon (PAH) exposure in decapod crus-
244 J. Beyer et al. / Environmental Toxicology and Pharmacology 30 (2010) 224–244

taceans by fluorimetric analysis of urine and haemolymph. Aquatic Toxicology environmental exposure and effects. Journal of Great Lakes Research 29,
67, 127–142. 116–123.
Webb, D., Gagnon, M.M., Rose, T.H., 2005. Interannual variability in fish biomarkers Ye, B.X., Zhang, Z.H., Mao, T., 2006. Pollution sources identification of poly-
in a contaminated temperate urban estuary. Ecotoxicology and Environmental cyclic aromatic hydrocarbons of soils in Tianjin area, China. Chemosphere 64,
Safety 62, 53–65. 525–534.
Webster, L., Angus, L., Topping, G., Dalgarno, E.J., Moffat, C.F., 1997. Long-term moni- Yu, L.J., Li, Y.Q., Sui, W., 2002. Analysis of PAHs by Shpol”skii fluorescence spectra in
toring of polycyclic aromatic hydrocarbons in mussels (Mytilus edulis) following low temperature. Spectroscopy and Spectral Analysis 22, 819–821.
the Braer oil spill. Analyst 122, 1491–1495. Yu, S.J., Campiglia, A.D., 2005. Direct determination of dibenzo[a.l]pyrene and its
Willett, K.L., Gardinali, P.R., Lienesch, L.A., Di Giulio, R.T., 2000. Comparative four dibenzopyrene isomers in water samples by solid-liquid extraction and
metabolism and excretion of benzo[a]pyrene in 2 species of Ictalurid catfish. laser-excited time-resolved Shpol’skii spectrometry. Analytical Chemistry 77,
Toxicological Sciences 58, 68–76. 1440–1447.
Willoughby, R., Sheehan, E., Mitrovich, S., 1998. A Global View of LC/MS. How to Yu, Y., Wade, T.L., Fang, J., McDonald, S., Brooks, J.M., 1995. Gas chromatographic-
Solve Your Most Challenging Analytical Problem, first ed. Global View Publish- mass spectrometric analysis of polycyclic aromatic hydrocarbon metabolites in
ing, Pittsburg, PA. Antarctic fish (Notothenia Gibberifrons) injected with diesel fuel Arctic. Archives
Wilson, J.Y., Kruzynski, G.M., Addison, R.F., 2001. Experimental exposure of juve- of Environmental Contamination and Toxicology 29, 241–246.
nile chinook (Oncorhynchus tshawytscha) to bleached kraft mill effluent: hepatic Zhang, Y., Zhu, Y., Wang, S., Zheng, F., Ren, F., Kim, S.J., 2002. Direct deter-
CYP1A induction is correlated with DNA adducts but not with organochlorine mination of 1-hydroxypyrene in fish from coastal water by synchronous
residues. Aquatic Toxicology 53, 49–63. fluorimetry. Bulletin of Environmental Contamination and Toxicology 69,
Wurl, O., Obbard, J.P., 2004. A review of pollutants in the sea-surface microlayer 494–500.
(SML): a unique habitat for marine organisms. Marine Pollution Bulletin 48, Zhu, S.Q., Li, L., Thornton, C., Carvalho, P., Avery, B.A., Willett, K.L., 2008. Simultane-
1016–1030. ous determination of benzo[a]pyrene and eight of its metabolites in Fundulus
Yang, X.A., Peterson, D.S., Baumann, P.C., Lin, E.L.C., 2003. Fish biliary PAH heteroclitus bile using ultra-performance liquid chromatography with mass
metabolites estimated by fixed-wavelength fluorescence as an indicator of spectrometry. Journal of Chromatography B 863, 141–149.