Use of Volatile Compound Metabolic Signatures in Poultry Liver to

ARTICLE pubs.acs.org/est

Use of Volatile Compound Metabolic Signatures in Poultry Liver to Back-Trace Dietary Exposure to Rapidly Metabolized Xenobiotics Philippe Berge,† J
er
emy Ratel,† Agnes Fournier,‡ Catherine Jondreville,‡ Cyril Feidt,‡ Brigitte Roudaut,§ Bruno Le Bizec,|| and Erwan Engel*,† †

INRA, UR370 QuaPA, MASS Team, Saint-Genes-Champanelle, France UR AFPA, USC340 INRA - Nancy Universit
e, France § ANSES, Fougeres, France ONIRIS, Nantes, France

)

‡

bS Supporting Information ABSTRACT: The study investigated the feasibility of using volatile compound signatures of liver tissues in poultry to detect previous dietary exposure to different types of xenobiotic. Six groups of broiler chickens were fed a similar diet either noncontaminated or contaminated with polychlorinated dibenzop-dioxins/-furans (PCDD/Fs; 3.14 pg WHO-TEQ/g feed, 12% moisture), polychlorinated biphenyls (PCBs; 0.08 pg WHO-TEQ/g feed, 12% moisture), polybrominated diphenyl ethers (PBDEs; 1.63 ng/g feed, 12% moisture), polycyclic aromatic hydrocarbons (PAHs; 0.72 μg/g fresh matter), or coccidiostats (0.5 mg/g feed, fresh matter). Each chicken liver was analyzed by solid-phase microextraction - mass spectrometry (SPME-MS) for volatile compound metabolic signature and by gas chromatography - high resolution mass spectrometry (GC-HRMS), gas chromatography - tandem mass spectrometry (GC-MS/MS), and liquid chromatography - tandem mass spectrometry (LC-MS/MS) to quantify xenobiotic residues. Volatile compound signature evidenced a liver metabolic response to PAH although these rapidly metabolized xenobiotics are undetectable in this organ by the reference methods. Similarly, the volatile compound metabolic signature enabled to differentiate the noncontaminated chickens from those contaminated with PBDEs or coccidiostats. In contrast, no clear signature was pointed out for slowly metabolized compounds such as PCDD/Fs and PCBs although their residues were found in liver at 50.93 ((6.71) and 0.67 ((0.1) pg WHO-TEQ/g fat, respectively.

1. INTRODUCTION It is well established that food-producing animals are exposed to toxic xenobiotics via environment and feeds and that xenobiotics entering the animals are transferred to edible tissues, thus representing a chemical human health hazard.1,2 Most current approaches to assessing contamination levels in foods are based on high performance analytical methods designed to determine the concentration of targeted xenobiotics present down to trace levels in the food. These techniques are costly to set up and operate, which consequently limits practical application. Alternative analytical approaches have been proposed that consist in seeking indirect markers of exposure to xenobiotics, e.g. metabolites such as lipids, amino acids, simple sugars, or cofactors.3
5 Exposure to xenobiotics, and the ensuing development of pathologies, gives rise to biological stresses that alter the metabolic profile of body tissues, thereby generating specific metabolic signatures.6
8 Clearly marked correspondence of biomarkers with type of contamination has been reported.9 r 2011 American Chemical Society

Metabolic signatures are currently obtained by techniques based on mass spectrometry,4,10
12 NMR spectrometry,5,6,13 or 2D gel electrophoresis.14 In fish exposed to crude oil for 30 days, Aas et al.15 found that the PAH concentration in the liver peaked after 3 days exposure and then decreased toward the end of exposure period, whereas PAH metabolites increased concurrently and continuously in bile. In their review paper, Miller and Ramos16 evidenced that the rapid decrease in liver PAH content after animal PAH exposure is mainly due to hepatic oxidation of these compounds in hydroxy-PAHs. Thus the possibility that some xenobiotics may be metabolized and thus become practically undetectable by direct measurement methods argues strongly for seeking persistent metabolic signatures of exposure to xenobiotics in animal food products. Received: March 4, 2011 Accepted: June 25, 2011 Revised: May 24, 2011 Published: July 13, 2011 6584

dx.doi.org/10.1021/es200747h | Environ. Sci. Technol. 2011, 45, 6584–6591

Environmental Science & Technology Volatile compounds occur in the bodies of humans and animals and their presence is mainly the result of their biological activity. In a study by Xue et al.,11 positive levels of 19 volatile compounds among the total of 47 detected in blood were found to differ significantly between healthy and liver cancer patients. Recent works by Vasta et al.,17 Sivadier et al.,18 and Ratel and Engel19 showed that volatile compounds in animal tissues and fluids could provide evidence of differences in their conditions of production. These compounds may thus be regarded as potential biomarkers of any deviation of the metabolism in response to nutrition, pathology, or exposure to xenobiotics. The objective of this work was to evaluate the feasibility of a novel rapid approach based on a nontargeted analysis of volatile compounds in poultry liver to detect previous dietary exposure to different types of xenobiotic. The xenobiotics under study include the main environmental micropollutants susceptible of being found in poultry, including PCDD/Fs, PCBs, PAHs,20 and PBDEs.21 Coccidiostats were also considered, as several reports of the European food safety agencies indicate that their use for coccidiosis prevention could be strongly restricted and even prohibited as a feed additive in a near future.22

2. EXPERIMENTAL SECTION 2.1. Poultry and Exposure to Pollutants. Six groups of 10
13 1-day-old broiler chickens (JA-957 line, Les Couvoirs de l’Est, Willgottheim, France) were reared indoors in individual coops under controlled conditions at the AFPA Research Unit Center (Nancy, France). The xenobiotics were all administered through diet, 1/as diet is known to be the main route of chicken exposure to PCDD/Fs, PCBs, and coccidiostats and has a non-negligible contribution to PAH and PBDE contamination20 and 2/as working with dietary contamination made it easier to experimentally control animal exposure and to compare the metabolic response to each class of these target compounds. After a one-week period of adaptation, the chickens were fed different experimental diets ad libitum. Within each group the chickens were exclusively given a control feed that was either noncontaminated (control group) or contaminated with one of four different types of xenobiotic mixtures regarded as contaminants commonly found in the poultry production chain, namely PCDD/Fs, PCBs, PBDEs, PAHs, and two coccidiostats (i.e., antimicrobial agent used to prevent coccidiosis), nicarbazin and narasin. The xenobiotics were introduced as mixtures. The PAH mixture was made up of phenanthrene (purity g97%, Sigma, St. Louis, MO, USA), pyrene (purity g98%, Sigma), and benzo[a]pyrene (purity g96%, Sigma). These three compounds feature among the 16 priority PAHs listed by the US Environmental Protection Agency (EPA). Phenanthrene is the PAH most commonly found in metabolite form in animal tissue.23 Pyrene is a routine indicator of environmental PAH contamination according to Chahin et al.,24 and its metabolism has been extensively studied (e.g., Haddad et al.25). Benzo[a]pyrene is considered by the US EPA as the reference PAH in terms of toxicity and has long been used as the key marker of PAH contamination. The PBDE mixture was made up of BDE-47 and BDE-99 (purity g99.9%, Sigma). These two PBDEs were selected based on their persistence and prominence in environmental samples and animal tissues26 and to the extensive research already published on their metabolism in liver tissues.27,28 For PCBs, PCDD/Fs, and coccidiostats, the standard mixture of Arochlor 1254 including the PCB congeners listed in Table S1 (Sigma), a standard mixture of PCDD/Fs including the 17 toxic congeners

ARTICLE

(Wellington Laboratories, ON, Canada), and a standard coccidiostat mixture of nicarbazin and narasin (Maxiban, Elanco, Clinton Laboratories, Clinton, Indiana, USA) were respectively used. The control noncontaminated feed was a pelleted mixture of commercial feedstuffs with 5% sunflower oil (INRA, Nouzilly, France). For the contaminations of feeds by PAHs, PBDEs, PCBs, and PCDD/Fs, the sunflower oil added to the control feed was polluted by each of these xenobiotics under agitation (150 rpm at 20 °C). The polluted oil mixed with 3 kg of control feed was gradually added to feed in a large mixer to a weight of 50 kg. For the coccidiostats, the powdered anticoccidial was added directly to the standard premix formulated by INRA (Nouzilly, France) to meet the needs of the chickens with a final concentration of 0.5 g Maxiban/kg feed. The concentrations of the other xenobiotics in each contaminated feed are given in Table S1 (Supporting Information). After seven weeks exposure, the chickens were electrically stunned and slaughtered according to European recommended standard conditions. 2.2. Preparation of Tissue Samples. Immediately after slaughter, liver and abdominal adipose tissues were excised from the chicken carcasses and immersed in liquid nitrogen, wrapped in aluminum foil, vacuum packed, and stored at
80 °C. In a second step, each tissue sample (liver tissues, n = 71, and adipose tissues, n = 23) was ground for 3 min in liquid nitrogen into a fine homogeneous powder using a homemade stainless steel ball mill. For the targeted quantification of the xenobiotics in tissues, a 10 g aliquot of each ground tissue was placed in a glass bottle (VWR International, Fontenay-sous-Bois, France) closed under a nitrogen flow and stored at
80 °C. For the determination of the volatile compound metabolic signature of the exposure to xenobiotics in liver, a 1.2 g aliquot of each ground tissue was placed in a glass vial (VWR International) sealed under nitrogen flow and stored at
80 °C. 2.3. Quantification of the Xenobiotics in Feeds and Tissue Samples. A representative sample of each of the six experimental feeds was placed in a glass bottle and kept at
20 °C. The three sets of samples (liver, adipose tissue, and feed) were subsequently used for the targeted quantification of the xenobiotics by the reference methods. The xenobiotics studied were quantified in all feed samples and in the liver samples of the control group, apart from coccidiostats, which were shown to be absent from the control diet. The PCDD/Fs, PCBs, and PBDEs were quantified by gas chromatography coupled with high-resolution mass spectrometry (GC-HRMS) and the PAHs by gas chromatography coupled with tandem mass spectrometry (GC-MS/MS). The PCDD/Fs, PCBs, PBDEs, and PAHs were assayed by ONIRIS (ENV Nantes, France) which is the French NRL (National Reference Laboratory) for PCDD/Fs, PCBs, and PAHs. PBDEs were also assayed by GC-HRMS in adipose tissues. These analyses were carried out according to Costera et al.29 for PCDD/Fs and PCBs, Antignac et al.30 for PBDEs, and Veyrand et al.31 for PAHs. Coccidiostats (nicarbazin and narasin) were assayed by ANSES (Fougeres, France) by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) in liver samples of chickens given these xenobiotics, according to a protocol derived from Dubreil-Cheneau et al.32 2.4. Volatile Compound Signatures. The volatile compounds of liver samples were extracted by solid-phase microextraction (SPME). A MPS2 multipurpose sampler (GERSTEL, Baltimore, MD) was used to carry out the following successive steps: (i) preheating of the sample in the agitator (500 rpm) for 10 min at 60 °C, (ii) trapping by SPME (75 μm carboxen-polydimethylsiloxane, 23 gauge needle, Supelco) of the volatile compounds for 6585

dx.doi.org/10.1021/es200747h |Environ. Sci. Technol. 2011, 45, 6584–6591

Environmental Science & Technology

ARTICLE

corresponds to the sum of its abundance acquired at each MS scan during the GC run. 2.5. Data Processing. Data were processed using the Statistica Software release 8.0 package (Statsoft, Maisons-Alfort, France) and the R software version 2.1.4 (http://www.R-project.org). The 218 mass fragments of each SPME-MS signature were normalized according to a method adapted from Engel et al.:33 first, the A2218 possible logs of ratios between the 218 fragment abundances were calculated, and then the most discriminative ratios were selected by a one-way ANOVA comparing noncontaminated chickens with the five groups of contaminated chickens. The robustness of the one-way ANOVA was ensured by a leave-one-out cross-validation procedure. Principal component analysis (PCA) was performed on the 25 most discriminant log ratios of mass fragment abundances selected in the SPME-MS signatures in order to visualize the structure of the data for each comparison. For the discriminant analysis (DA), only the two most relevant log ratios of mass fragments were selected according to the “best subset” algorithm (Statistica).

3. RESULTS AND DISCUSSION

Figure 1. Detection of chicken exposure to PCDD/Fs by analysis of livers. (A) Increase in PCDD/F concentration in contaminated feed (pg/g 12% moisture basis; n = 1) and liver samples (pg/g fat; n = 10) compared with control samples. PCDD/Fs were quantified by GC-HRMS. Error bars are standard errors. (B) Use of metabolic signatures in volatile compounds generated by SPME-MS for differentiating the noncontaminated “control” chickens from those contaminated with dioxins. First map of normed PCA plotted on the most discriminant log ratios of mass fragments selected in the SPME-MS signatures.

30 min at 60 °C, and (iii) thermal desorption at 280 °C for 2 min in the GC inlet. Further volatile compound analysis was performed by GC-full scan MS (GC: model 6890, Hewlett-Packard, PA, USA; quadrupole MS: model 5973A, Hewlett-Packard) according to Ratel and Engel19 except for the mass range and the scan rate which were set at m/z = 33
250 amu and at 6.2 scans s
1, respectively. The chromatograms obtained in full scan by SPMEGC-MS were converted into virtual SPME-MS fingerprints according to Vasta et al.17 Each fingerprint is a mass spectrum where the abundance of each mass fragment ranging from 33 to 250 amu

3.1. PCDD/Fs and PCBs. Figure 1A presents the average increase in PCDD/Fs found in feed and liver samples of PCDD/F contaminated animals compared with their counterparts from control noncontaminated animals. The plotting confirms that the feed was properly contaminated with the 17 toxic dioxin congeners the concentrations of which were found significantly higher in the deliberately contaminated diet than in the control diet. The PCDD/F level reached 3.14 pg WHO TEQ/g on a 12% moisture basis in the deliberately contaminated diet, which is higher than but within the same range of magnitude as the European Food Safety Authority (EFSA) standard set at 0.7 pg WHO TEQ/g on a 12% moisture basis for the maximum levels of these contaminants.34 Similarly, the concentrations of the 17 congeners were much higher in the liver of contaminated chickens than in the control group, evidencing their transfer from the feed to the liver tissue. The high proportion of octachloro-congeners evidenced by Figure 1A for contaminated feed is consistent with the composition of the PCDD/F mixture deliberately added to the diet. This congener pattern was chosen to mimic the PCDD/F profiles measured by the ONIRIS on feed contaminated during recent French dioxin scares and was consistent with EFSA figures on the maximum levels of PCDD/Fs measured in real-life feed samples.34 Figure S1A (Supporting Information) shows similar trends for Aroclor 1254 PCB congeners, which were found at higher levels in contaminated samples for both feed and liver than in corresponding controls. Dioxin-like PCB levels reached 0.08 pg WHO TEQ/g on a 12% moisture basis in the deliberately contaminated diet, which was lower than EFSA standards set at 0.3 (action level) and 0.5 pg (maximum level) WHO TEQ/g on a 12% moisture basis.34 Previous studies showed that after animals had been exposed to PCDD/Fs and PCBs, these xenobiotics were transferred in the blood to organs such as the liver and adipose tissues.35,36 Stephens et al.37 evidenced that the highest concentrations of PCDD/Fs (on a fat weight basis) were found in the liver tissue after a dietary exposure of chickens to these xenobiotics. These authors suggested that mechanisms other than lipid solubility operated in this tissue, confirming a previous report by Olling et al.38 in lactating cows suggesting cytoplasmic and microsomal binding in the liver. The examination of the ratios between liver PCDD/F increase and feed PCDD/F increase (Figure 1A) confirms previous reports showing that the transfer from feed to liver tends to decrease 6586

dx.doi.org/10.1021/es200747h |Environ. Sci. Technol. 2011, 45, 6584–6591

Environmental Science & Technology

Figure 2. Detection of chicken exposure to PBDEs by analysis of livers. (A) Increase in PBDE concentration in contaminated feed (ng/g 12% moisture basis; n = 1) and adipose and liver tissue samples (ng/g fat; n = 10) compared with their counterparts from control noncontaminated samples. PBDEs were quantified by GC-HRMS. Error bars are standard errors; (B) Use of metabolic signatures in volatile compounds generated by SPME-MS analysis of livers for differentiating the noncontaminated “control” chickens from those contaminated with PBDEs. B1. First map of normed PCA plotted on the most discriminant log ratios of mass fragments selected in the SPME-MS signatures; B2. Sample discrimination in the plane formed by a DA-selected pair of log ratios of mass fragment abundances Fi with i = m/z for the considered mass fragment.

ARTICLE

with an increasing number of chlorine substituents for both PCDDs and PCDFs,39 which is particularly significant when tetra-, hepta-, and octa-chlorinated congeners are compared. For PCBs no clear relationship between chlorination and accumulation is evidenced in Figure S1A (Supporting Information). Figures 1B and S1B (Supporting Information) present the first map of the PCA processed on the volatile compound signature in the liver of broiler chickens contaminated or noncontaminated with PCDD/Fs and PCBs, respectively. The overlaying of PCDD/F and PCB group plots with the control group plots suggests that the exposure to dioxins and PCBs had no direct or indirect influence on the liver volatile compound composition and that these xenobiotics with closely similar chemical structures are practically nonmetabolized and so accumulate in this organ as previously reported.35
37 Nevertheless, we cannot rule out the possibility that the SPME-MS signatures obtained here may not have been sensitive enough to detect possible effects of dioxins and PCBs on metabolism and thereby on quantitatively minor volatile compounds. Further investigation is thus needed to determine whether the generation of more detailed metabolic signatures based on volatile compound profiles by GC-MS or comprehensive GC  GC-MS techniques is an effective approach to identifying compounds that could be reliable biomarkers of exposure to PCDD/Fs and PCBs. Moreover, it cannot be excluded that the PCB level measured in deliberately contaminated chicken feed was too weak to elicit a metabolic response, despite being higher than the maximum dose of PCBs reported by Nishimura et al.40 in real-life chicken feed samples. 3.2. PBDEs. Figure 2A presents the average increase in BDE47 and BDE-99 found in feed, liver, and adipose tissue samples from BDE-47 and BDE-99 contaminated chickens compared with those from control noncontaminated chickens. The plot confirms that the feed was properly contaminated with these two BDEs, the concentrations of which were found to be significantly higher in the deliberately contaminated diet than in the control diet. Compared to the average concentration of PBDEs found in chicken fat samples on the U.S. market, i.e. 2.9 ng/g of lipid according to Huwe et al.,41 the average PBDE levels evidenced in Figure 2A for fat samples (10.06 ( 1.27 ng/g of lipid) demonstrate that the PBDE dose deliberately added to the feed was realistic. Similarly, the concentration of BDE-99 was much higher in the liver of contaminated chickens than in controls, evidencing its transfer from the feed to this organ. BDE-47 concentration was not found to be significantly increased in the liver of contaminated chickens in contrast to the results obtained in the adipose tissue of the same animals. The increase in BDE-47 observed in the adipose tissue is consistent with earlier literature data showing a bioaccumulation of BDEs in both fat and liver tissues following intake by chickens.21,27 The notably high BDE47 level measured in the liver of the control group may explain the nonsignificance of the increase in this xenobiotic in contaminated liver samples. Several authors have already evidenced that the determination of BDE-47 may be overestimated in blank or control samples due to analytical artifacts including natural contamination of the laboratory environment.42 Here, a low BDE-47 level was found in the analytical blank (data not shown), indicating a very low contamination of the laboratory environment. Nevertheless, a carry-over effect between contaminated samples and control may have occurred during the freeze-drying process, which was the only sample preparation step applied exclusively to liver samples. 6587

dx.doi.org/10.1021/es200747h |Environ. Sci. Technol. 2011, 45, 6584–6591

Environmental Science & Technology

ARTICLE

Figure 4. Detection of chicken exposure to coccidiostats using the metabolic signatures in volatile compounds of livers. The volatile compound signatures were generated by SPME-MS. First map of normed PCA plotted on the most discriminant log ratios of mass fragments selected in the SPME-MS signatures.

Figure 3. Detection of chicken exposure to PAHs by analysis of livers. (A) Increase in PAH concentration in contaminated feed (ng/g fresh matter; n = 1) and liver samples (10 ng/g fresh matter; n = 10) compared with their counterparts from control noncontaminated samples. PAHs were quantified by GC-MS/MS. Error bars are standard errors; (B) Use of metabolic signatures in volatile compounds generated by SPME-MS for differentiating the noncontaminated “control” chickens from those contaminated with PAHs. B1. First map of normed PCA plotted on the most discriminant log ratios of mass fragments selected in the SPME-MS signatures; B2. Sample discrimination in the plane formed by a DA-selected pair of log ratios of mass fragment abundances Fi with i = m/z for the considered mass fragment.

The clear distinction of the PBDE groups from the control group in Figure 2B1 indicates that changes in metabolism occurred in response to dietary contamination by PBDEs and that these changes were perceptible through the volatile compound signature. The combination of only two ratios of mass fragments selected by DA allowed a good differentiation of PBDE-contaminated chickens from control noncontaminated ones (Figure 2B2). This confirms that the biological signature generated in the liver by exposure to PBDEs was significantly different from that found in the liver of noncontaminated chickens. These results are consistent with several previous reports showing that BDE-47 and BDE-99 generate a quick metabolic response of the liver including oxidation, bromination, and debromination reactions,21,27 which may thus change the volatile compound profile in this organ. 3.3. PAHs. Figure 3A shows that the feed was properly contaminated with the three PAHs the average concentrations of which were found to be significantly higher in the deliberately contaminated feed than in the control feed. The feed contamination level (0.72 μg/g fresh matter) was 5-fold higher than the average PAH level measured in real-life chicken feed by Loutfy et al.,43 thus demonstrating that the PAH dose given to the chicken was significant and realistic. By contrast, Figure 3A is also evidence that none of these three PAHs was detected at a significantly higher concentration in the liver of contaminated chickens than in the control group. This can be explained by a rapid metabolic detoxification of these compounds. Animal tissues are reputed to have the capacity to metabolize PAHs quickly following dietary exposure.44 The enzymatic systems that metabolize PAHs are widely distributed in the tissues of both humans and animals, and the liver is the organ that exhibits the highest capacity to metabolize PAHs.45 Figure 3B1 points out that changes in animal metabolism occur in response to dietary contamination by PAHs and that these changes are perceptible through the clearly distinct volatile compound signatures found in the PAHs in contaminated and control noncontaminated groups. Figure 3B2 confirms this by showing that a combination of only two ratios of mass fragments selected by DA allowed a good discrimination of PAHs 6588

dx.doi.org/10.1021/es200747h |Environ. Sci. Technol. 2011, 45, 6584–6591

Environmental Science & Technology

Figure 5. Applicability of metabolic signatures in volatile compounds from chicken livers for revealing dietary exposure to xenobiotics. First map of normed PCA plotted on the most discriminant mass fragment log ratios selected in the metabolic signature differentiating the three data sets: [1] control, PCDD/Fs and PCBs, [2] PAHs, and [3] PBDEs.

contaminated chickens from controls. The finding that the exposure to a xenobiotic, i.e. PAHs, can generate a metabolic response in a target organ despite this xenobiotic being undetectable by reference methods in this organ could thus pave the way to new approaches for revealing animal exposure to environmental contamination by rapidly metabolized xenobiotics. 3.4. Coccidiostats. The measurement made by LC-MS/MS confirmed the contamination of the liver by both coccidiostats with average levels of 28765 ((7500) and 150 ((80) μg/kg tissue for dinitrocarbanilide and narasin, respectively. Although narasin was administered together with nicarbazin, a different pattern for narasin and dinitrocarbanilide (the marker residue for nicarbazin) was observed, due to their different pharmacokinetic properties. EFSA studies46 showed that nicarbazin as dinitrocarbanilide was absorbed to a large extent by chickens and was accumulated in the liver, where the highest content was found. In another study, Sweeney et al.47 showed that orally administered narasin was rapidly metabolized by chickens and excreted in their feces. Figure 4 shows that chickens given coccidiostats were clearly discriminated from controls by their volatile compound signatures in the liver. However, we note that the average slaughter weight of the chickens from the coccidiostat group was significantly lower than that of the chickens from the control group (1.4 ( 0.3 kg vs 2.1 ( 0.2 kg; P < 0.05). By contrast, no significant difference in average slaughter weight was observed between the control group and the groups contaminated with PCDD/Fs, PCBs, PBDEs, or PAHs. The drop in slaughter weight indicates that the level of coccidiostat administration may have exceeded the toxicity threshold in the coccidiostat group. The difference in the volatile compound signatures between the control and coccidiostat group should thus be interpreted carefully. For this reason, the data relating to the coccidiostat group were excluded from the subsequent chemometric treatments. 3.5. Specificity of Volatile Compound Signatures. A normed PCA was performed on the mass fragment ratios discriminating the metabolic signatures obtained from the three resulting data

ARTICLE

sets: control, PCDD/Fs and PCBs [set 1], PAHs [set 2], and PBDEs [set 3]. In the first map of this PCA (Figure 5), the plots of PAHs and PBDEs groups are superimposed. This result evidence the difficulty differentiating chickens contaminated with PAHs from those contaminated with PBDEs on the basis of their volatile compound signature in the liver, which suggests that exposure of chickens to these two types of xenobiotic have similar impacts on the rough SPME-MS signature and so on the most abundant volatile compounds in this organ. Further investigation is thus required to evaluate whether the use of more resolutive metabolic signatures, e.g. signatures based on volatile compound profiles generated by GC-MS or GC  GC-MS techniques, can reveal more subtle differences in the metabolic responses of chickens to PAH and PBDE exposure. In addition, these techniques will be needed to identify robust biomarkers of exposure to a particular xenobiotic responsible for the differences observed between overall metabolic signatures in noncontaminated chickens and those in chickens contaminated with PBDEs, PAHs, or coccidiostats, and most probably also with PCDD/Fs or PCBs. By opening up the prospect of using these high-resolution techniques to improve the selectivity, sensitivity, and specificity of the approach, the present study might pave the way to a new generation of monitoring methods which are not based on the measurement of xenobiotic residues or their parent metabolites.

’ ASSOCIATED CONTENT

bS

Supporting Information. This section comprises 1/Table S1 giving the concentration of the xenobiotics quantified by reference techniques in poultry diet and 2/Figure S1 presenting the detection of chicken exposure to PCBs by analysis of livers. This material is available free of charge via the Internet at http:// pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: +33(0)4 73 62 45 89. E-mail: erwan.engel@clermont. inra.fr.

’ ACKNOWLEDGMENT This study was supported by the European Commission, Contract No. FP6 - 518451. SigmaChain (2006) Developing a Stakeholder
s Guide on the vulnerability of food and feed chains to dangerous agents and substances. Available at http://www. sigmachain.eu. We would like to acknowledge Patrick BLINET and Philippe MARCHAND for technical advices. ’ REFERENCES (1) Jones, K. C.; de Voogt, P. Persistent organic pollutants (POPs): state of the science. Environ. Pollut. 1999, 100, 209–221. (2) Andr
ee, S.; Jira, W.; Schwind, K. H.; Wagner, H.; Schw€agele, F. Chemical safety of meat and meat products. Meat Sci. 2010, 86, 38–480. (3) Morris, M.; Watkins, S. M. Focused metabolomic profiling in the drug development process: advances from lipid profiling. Curr. Opin. Chem. Biol. 2005, 9, 407–412. (4) Madalinski, G.; Godat, E.; Alves, S.; Lesage, D.; Genin, E.; Levi, P.; Labarre, J.; Tabet, J. C.; Ezan, E.; Junot, C. Direct introduction of biological samples into a LTQ-Orbitrap hybrid mass spectrometer as a tool for fast metabolome analysis. Anal. Chem. 2008, 80, 3291–3303. (5) Eveillard, A.; Lasserre, F.; De Tayrac, M.; Polizzi, A.; Claus, S.; Canlet, C.; Mselli-Lakhal, L.; Gotardi, G.; Paris, A.; Guillou, H.; Martin, 6589

dx.doi.org/10.1021/es200747h |Environ. Sci. Technol. 2011, 45, 6584–6591

Environmental Science & Technology P. G. P.; Pineau, T. Identification of potential mechanisms of toxicity after di-(2-ethylhexyl)-phthalate (DEHP) adult exposure in the liver using a systems biology approach. Toxicol. Appl. Pharmacol. 2009, 236, 282–292. (6) Watkins, S. M.; German, J. B. Toward the implementation of metabolomic assessments of human health and nutrition. Curr. Opin. Biotechnol. 2002, 13, 512–516. (7) Bundy, J. G.; Keun, H. C.; Sidhu, J. K.; Spurgeon, D. J.; Svendsen, C.; Kille, P.; Morgan, A. J. Metabolic Profile Biomarkers of metal contamination in a sentinel terrestrial species are applicable across multiple sites. Environ. Sci. Technol. 2007, 41, 4458–4464. (8) Michell, A. W.; Mosedale, D.; Grainger, D. J.; Barker, R. A. Metabolomic analysis of urine and serum in Parkinson’s disease. Metabolomics 2008, 4, 191–201. (9) Schiedek, D.; Broeg, K.; Barsiene, J.; Lehtonen, K. K.; Gercken, J.; Pfeifer, S.; Vuontisjarvi, H.; Vuorinen, P. J.; Dedonyte, V.; Koehler, A.; Balk, L.; Schneider, R. Biomarker responses as indication of contaminant effects in blue mussel (Mytilus edulis) and female eelpout (Zoarces viviparus) from the southwestern Baltic Sea. Mar. Pollut. Bull. 2006, 53, 387–405. (10) Want, E. J.; Nordstrom, A.; Morita, H.; Siuzdak, G. From exogenous to endogenous: The inevitable imprint of mass spectrometry in metabolomics. J. Proteome Res. 2007, 6, 459–468. (11) Xue, R. Y.; Dong, L.; Zhang, S.; Deng, C. H.; Liu, T. T.; Wang, J. Y.; Shen, X. Z. Investigation of volatile biomarkers in liver cancer blood using solid-phase microextraction and gas chromatography/mass spectrometry. Rapid Commun. Mass Spectrom. 2008, 22, 1181–1186. (12) Woo, H. K.; Go, E. P.; Hoang, L.; Trauger, S. A.; Bowen, B.; Siuzdak, G.; Northen, T. R. Phosphonium labeling for increasing metabolomic coverage of neutral lipids using electrospray ionization mass spectrometry. Rapid Commun. Mass Spectrom. 2009, 23, 1849–1855. (13) Beckonert, O.; Keun, H. C.; Ebbels, T. M. D.; Bundy, J. G.; Holmes, E.; Lindon, J. C.; Nicholson, J. K. Metabolic profiling, metabolomic and metabonomic procedures for NMR spectroscopy of urine, plasma, serum and tissue extracts. Nat. Protoc. 2007, 2, 2692–2703. (14) Oberemm, A.; Meckert, C.; Brandenburger, L.; Herzig, A.; Lindner, Y.; Kalenberg, K.; Krause, E.; Ittrich, C.; Kopp
Schneider, A.; Stahlmann, R.; Richter
Reichhelm, H. B.; Gundert
Remy, U. Differential signatures of protein expression in marmoset liver and thymus induced by single-dose tcdd treatment. Toxicology 2005, 206, 33–48. (15) Aas, E.; Baussant, T.; Balk, L.; Liewenborg, B.; Andersen, O. K. PAH metabolites in bile, cytochrome P4501A and DNA adducts as environmental risk parameters for chronic oil exposure: a laboratory experiment with Atlantic cod. Aquat. Toxicol. 2000, 51, 241–258. (16) Miller, K. P.; Ramos, K. S. Impact of cellular metabolism on the biological effects of benzo[a]pyrene and related hydrocarbons. Drug Metab. Rev. 2001, 33, 1–35. (17) Vasta, V.; Ratel, J.; Engel, E. Mass spectrometry analysis of volatile compounds in raw meat for the authentication of the feeding background of farm animals. J. Agric. Food. Chem. 2007, 55, 4630–4639. (18) Sivadier, G.; Ratel, J.; Bouvier, F.; Engel, E. Authentication of meat products: determination of animal feeding by parallel GC-MS analysis of three adipose tissues. J. Agric. Food. Chem. 2008, 56, 9803–9812. (19) Ratel, J.; Engel, E. Determination of benzenic and halogenated volatile organic compounds in animal-derived food products by onedimensional and comprehensive two-dimensional gas chromatography mass spectrometry. J. Chromatogr., A 2009, 1216, 7889–7898. (20) Kan, C. A.; Meijer, G. A. L. The risk of contamination of food with toxic substances present in animal feed. Anim. Feed Sci. Technol. 2007, 133, 84–108. (21) Pirard, C.; De Pauw, E. Absorption, disposition and excretion of polybrominated diphenyl ethers (PBDEs) in chicken. Chemosphere 2007, 66, 320–325. (22) Castanon, J. I. R. History of the use of antibiotic as growth promoters in European poultry feeds. Poult. Sci. 2007, 86, 2466–2471. (23) Grova, N.; Monteau, F.; Le Bizec, B.; Feidt, C.; Andre, F.; Rychen, G. Determination of phenanthrene and hydroxyphenanthrenes in various biological matrices at trace levels using gas chromatography
mass spectrometry. J. Anal. Toxicol. 2005, 29, 175–181.

ARTICLE

(24) Chahin, A.; Guiavarc’h, Y. P.; Dziurla, M. A.; Toussaint, H.; Feidt, C.; Rychen, G. 1-hydroxypyrene in milk and urine as a bioindicator of polycyclic aromatic hydrocarbon exposure of ruminants. J. Agric. Food Chem. 2008, 56, 1780–1786. (25) Haddad, S.; Tardif, R.; Withey, J.; Krishnan, K. Rate of pyrene metabolism in rat liver post-mitochondrial fractions. Toxicol. Lett. 1997, 93, 177–184. (26) Hites, R. A. Polybrominated Diphenyl Ethers in the Environment and in People: A Meta-Analysis of Concentrations. Environ. Sci. Technol. 2004, 38, 945–956. (27) Hakk, H.; Huwe, J. K.; Murphy, K.; Rutherford, D. Metabolism of 2,2 0 ,4,4 0 -Tetrabromodiphenyl Ether (BDE-47) in Chickens. J. Agric. Food Chem. 2010, 58, 8757–8762. (28) Albina, M. L.; Alonso, V.; Linares, V.; Bell
es, M.; Sirvent, J. J.; Domingo, J. L.; S
anchez, D. J. Effects of exposure to BDE-99 on oxidative status of liver and kidney in adult rats. Toxicology 2010, 271, 51–56. (29) Costera, A.; Feidt, C.; Marchand., P.; Le Bizec, B.; Rychen, G. PCDD/F and PCB transfer to milk in goats exposed to a long-term intake of contaminated hay. Chemosphere 2006, 64, 650–657. (30) Antignac, J. P.; Cariou, R.; Zalko, D.; Berrebi, A.; Cravedi, J. P.; Maume, D.; Marchand, P.; Monteau, F.; Riu, A.; Andre, F.; Le Bizec, B. Exposure assessment of French women and their newborn to brominated flame retardants: Determination of tri- to deca- polybromodiphenylethers (PBDE) in maternal adipose tissue, serum, breast milk and cord serum. Environ. Pollut. 2009, 157, 164–173. (31) Veyrand, B.; Brosseaud, A.; Sarcher, L.; Varlet, V.; Monteau, F.; Marchand, P.; Andre, F.; Le Bizec, B. Innovative method for determination of 19 polycyclic aromatic hydrocarbons in food and oil samples using gas chromatography coupled to tandem mass spectrometry based on an isotope dilution approach. J. Chromatogr., A 2007, 1149, 333–344. (32) Dubreil
Ch
eneau, E.; Bessiral, M.; Roudaut, B.; Verdon, E.; Sanders, P. Validation of a multi-residue liquid chromatography tandem mass spectrometry confirmatory method for 10 anticoccidials in eggs according to Commission Decision 2002/657/EC. J. Chromatogr., A 2009, 1216, 8149–8157. (33) Engel, E.; Ferlay, A.; Cornu, A.; Chilliard, Y.; Agabriel, C.; Bielicki, G.; Martin, B. Relevance of isotopic and molecular biomarkers for the authentication of milk according to production zone and type of feeding of the cow. J. Agric. Food Chem. 2007, 55, 9099–9108. (34) Results of the monitoring of dioxin levels in food and feed; Highlights of EFSA Journal 2010, 8(3), 1385. European Food Safety Authority. www.efsa.europa.eu/fr/scdocs/doc/1385.pdf (accessed month day, year). (35) Maervoet, J.; Chu, S. G.; De Vos, S.; Covaci, A.; Voorspoels, S.; De Schrijver, R.; Schepens, P. Accumulation and tissue distribution of selected polychlorinated biphenyl congeners in chickens. Chemosphere 2004, 57, 61–66. (36) Fernandes, A.; Mortimer, D.; Rose, M.; Gem, M. Dioxins (PCDD/Fs) and PCBs in offal: Occurrence and dietary exposure. Chemosphere 2010, 81, 536–540. (37) Stephens, R. D; Petreas, M. X.; Hayward, D. G. Biotransfer and bioaccumulation of dioxins and furans from soil: chickens as a model for foraging animals? Sci. Total Environ. 1995, 175, 253–273. (38) Olling, M.; Derks, H. J. G. M.; Berende, P. L. M.; Liem, A. K. D.; de Jong, A. P. J. M. Toxicokinetics of eight 13C-labelled polychlorinated dibenzo-p-dioxins and -furans in lactating cows. Chemosphere 1991, 23, 1377–1385. (39) Pirard, C.; De Pauw, E. Uptake of polychlorodibenzo-p-dioxins, polychlorodibenzofurans and coplanar polychlorobiphenyls in chickens. Environ. Int. 2005, 31, 585–591. (40) Nishimura, K.; Miyamoto, S.; Takeda, T.; Ando, M.; Tanabe, S. Dioxin Concentrations in Body Tissues and Egg of Female Chicken. J. Poult. Sci. 2005, 42, 346–355. (41) Huwe, J. K.; Larsen, G. L. Polychlorinated Dioxins, Furans, and Biphenyls, and Polybrominated Diphenyl Ethers in a U.S. Meat Market Basket and Estimates of Dietary Intake. Environ. Sci. Technol. 2005, 39, 5606–5611. 6590

dx.doi.org/10.1021/es200747h |Environ. Sci. Technol. 2011, 45, 6584–6591

Environmental Science & Technology

ARTICLE

(42) Covaci, A.; Voorspoels, S.; de Boer, J. Determination of brominated flame retardants, with emphasis on polybrominated diphenyl ethers (PBDEs) in environmental and human samples
a review. Environ. Int. 2003, 29, 735–756. (43) Loutfy, N.; Fuerhacker, M.; Tundo, P.; Raccanelli, S.; Tawfi Ahmed, M. Monitoring of polychlorinated dibenzo-p-dioxins and dibenzofurans, dioxin-like PCBs and polycyclic aromatic hydrocarbons in food and feed samples from Ismailia city, Egypt. Chemosphere 2007, 66, 1962–1970. (44) Beyer, J.; Jonsson, G.; Porte, C.; Krahn, M. M.; Ariese, F. Analytical methods for determining metabolites of polycyclic aromatic hydrocarbon (PAH) pollutants in fish bile: A review. Environ. Toxicol. Pharmacol. 2010, 30, 224–244. (45) Polycyclic aromatic hydrocarbons; Highlights of WHO guidelines for indoor air quality: selected pollutants, World Health Organization: 2010; p 289.http://www.euro.who.int/__data/assets/pdf_file/0009/ 128169/e94535.pdf (accessed month day, year). (46) Opinion of the Scientific Panel on additives and products or substances used in animal feed (FEEDAP) on the request from the Commission on the efficacy and safety of the coccidiostat Koffogran; Highlights of EFSA Journal, 2003, 16, 1
40. European Food Safety Authority. http://www.efsa.europa.eu/en/efsajournal/doc/contam_op_ej592_ monensin_en.pdf (accessed month day, year). (47) Sweeney, D. J.; Kennington, A.; Donoho, A. L. Tissue residues and metabolism of narasin in chicken. J. Agric. Food Chem. 1996, 44, 2829–2834.

6591

dx.doi.org/10.1021/es200747h |Environ. Sci. Technol. 2011, 45, 6584–6591