AOAC INTERNATIONAL
Collaborative
Study
Draft Protocol
Determination of Polycyclic Aromatic Hydrocarbons (PAHs) in Seafood using Gas Chromatography-Mass Spectrometry: A Collaborative Study
Katerina Mastovska
Covance Laboratories Inc.
671 S. Meridian Road
Greenfield, IN 46140
Phone: 317.467.3645
Fax: 317.467.3790
E-mail:
Wendy R. Sorenson
Covance Laboratories Inc.
3301 Kinsman Boulevard
Madison, WI 53704
Phone: 608.310.8206
Fax: 608.443.1521
E-mail:
Jana Hajslova
Institute of Chemical Technology, Prague
Faculty of Food and Biochemical Technology
Department of Food Chemistry and Analysis
Technická 3, 166 28 Prague 6 Czech Republic
Phone: +420 220 44 3185
E-mail:
Determination of Polycyclic Aromatic Hydrocarbons (PAHs) in Seafood using Gas Chromatography-Mass Spectrometry: A Collaborative Study
Katerina Mastovska
Covance Laboratories Inc.
Wendy R. Sorenson
Covance Laboratories Inc.
Jana Hajslova
Institute of Chemical Technology, Prague
Introduction
Within a European integrated project CONffIDENCE (Contaminants in food and feed: Inexpensive detection for control of exposure), Jana Hajslova’s group at the Institute of Chemical Technology (ICT) in Prague, Czech Republic developed a method for the determination of polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) and polybrominated diphenyl ethers (PBDEs) in fish and seafood using gas chromatography coupled with time-of-flight mass spectrometry (GC-TOFMS). This method was selected for further study as an AOAC collaborative study by the AOAC Stakeholders Panel on Seafood Contaminants (SPSC), which was formed as a response to the seafood contamination resulting from the recent oil spill in the Gulf of Mexico. The analytes for this collaborative study have been narrowed down to include only PAHs and some of the relevant PAH alkyl homologues. Having a rapid method is essential for quick determination of contaminants in food, especially after environmental disasters. The nineteen contaminants found in Table 1 will be studied in this collaborative study.
Table 1. Polycyclic Aromatic Hydrocarbons (PAHs) to be studied in Collaborative Study
PAH Name / PAH Abbreviationanthracene / Ant
benzo[a]anthracene / BaA
benzo[a] pyrene / BaP
benzo[b]fluoranthene / BbF
benzo[k]fluoranthene / BkF
benzo[g,h,i]perylene / BghiP
chrysene / Chr
dibenzo[a,h]anthracene / DBahA
fluoranthene / Flt
fluorene / Fln
indeno[1,2,3-cd]pyrene / IcdP
naphthalene / Naph
phenanthrene / Phe
pyrene / Pyr
3-methylchrysene / 3-MC
1-methylnaphthalene / 1-MN
1-methylphenanthrene / 1-MP
2,6-dimethylnaphthalene / 2,6-DMN
1,7-dimethylphenanthrene / 1,7-DMP
Method Principle. Homogenized seafood samples (10 g sample with a 5 µg/kg addition of 13C-PAH surrogate mixture) are mixed with 5 mL of water and shaken vigorously with 10 mL of ethyl acetate in 50 mL polypropylene centrifuge tube for 1 min. Subsequently, 4 g of anh. magnesium sulphate and 2 g of sodium chloride are added to the mixture to induce phase separation and force the analytes into the ethyl acetate layer. The tube is again shaken for 1min and then centrifuged for 10 min at > 1500 rfc. A 5-mL aliquot of the ethyl acetate layer is taken from the tube, rotary evaporated to the last drop of solvent, re-dissolved in 1 mL of hexane, and cleaned on a manually prepared silica gel SPE column (a Pasteur pipette filled with glass wool, 1 g of silica gel and approx. 0.2 g of sodium sulphate). The column is conditioned with 6mL of hexane:dichloromethane (3:1, v/v) and 4 mL of hexane, followed by application of the 1 mL of the extract in hexane. The analytes are eluted by 10 mL of hexane:dichloromethane (3:1, v/v). The clean extract is carefully evaporated by a rotary vacuum evaporator and the residual solvents are removed by a gentle stream of nitrogen. The residue after evaporation is dissolved in 0.5 mL of isooctane and analyzed by GC-MS.
Purpose. The purpose of this study is to evaluate the method’s intra-laboratory and inter-laboratory performance. The results of this study will be submitted to the AOAC INTERNATIONAL for adoption as an Official Method for the determination of polycyclic aromatic hydrocarbons (PAHs) in seafood.
Scope and Applicability. This study will evaluate the performance of the method when used to determine polycyclic aromatic hydrocarbons (PAHs) in three seafood matrices: shrimp, oysters, and mussels. The PAHs to be determined are anthracene, benz(a)anthracene, benzo(a)pyrene, benzo(b)fluoranthene, benzo(g,h,i)perylene, benzo(k)fluoranthene, chrysene, dibenz(a,h)anthracene, fluoranthene, fluorene, indeno(1,2,3-cd)pyrene, naphthalene, phenanthrene, pyrene, 1-methylnaphthalene, 2,6-dimethylnaphthalene, 1-methylphenanthrene, 1,7-dimethylphenanthrene, and 3-methylchrysene. Benzo(a)pyrene (BaP) is the analyte of main interest and will be at a range of 2 to 50 µg/kg. The concentration range of the other contaminants will vary from 2 to 250 µg/kg. The target limit of quantification (LOQ) estimated as the lowest calibration level (LCL) is 0.5 µg/kg (fresh weight basis) for BaP.
Collaborators. X laboratories have verbally agreed to participate in this study. These laboratories have been identified as having appropriate instrumentation and expertise needed for this method. See Appendix A for a list of participating laboratories and instrumentation.
Study Design. This study will be conducted on three matrices with five different levels of BaP ranging from 2 to 50 µg/kg. Each matrix will have a varying mix of three different BaP levels. The other studied PAHs will be at varying levels from 2 to 250 µg/kg that mimic typical PAH patterns. The fortified analytes in three matrices will be analyzed as blind duplicates at each level of BaP and corresponding other PAH levels. In addition, a blank with no contaminants for each matrix will be analyzed singly.
Test Sample Preparation. The three matrices will be oyster, shrimp, and mussel. Each presumably blank matrix will be homogenized with liquid nitrogen and tested in duplicate by an independent laboratory for potential contamination with the target PAHs. During homogenization, portions of the blank matrices will be spiked with 1,7-dimethylphenanthrene (which will not be in the mixed spike solutions). This will be utilized as a homogenization check throughout the course of the study. The three different matrices will be spiked at three different levels of 1,7-dimethylphenanthrene in the range of 10 to 100 µg/kg. The collaborators will determine 1,7-dimethylphenanthrene along with the other 18 analytes, which will be spiked into 10 g sample portions placed in polypropylene centrifuge tubes. Five different spiking levels will be made at varying PAH concentrations with BaP levels at 2 to 50 µg/kg, resulting in three different duplicate spiked samples per matrix, in addition to a blank. Participants will be supplied with the test samples ready for analysis labeled with unique identification numbers (SFC1 to SFC21). All test samples will be shipped frozen on dry ice with a return receipt document to be returned to the Study Directors. The test samples should be stored in a freezer set to maintain -70 ± 10°C or at least -20 ± 10°C. Test samples will be analyzed after completion of laboratory qualification and practice sample analysis.
Laboratory Qualification. During the laboratory qualification phase, the collaborators will conduct four steps to optimize their GC-MS conditions and familiarize themselves with the method.
(1) The first step will be the GC separation test where participants will analyze a composite PAHs solution by GC-MS(-MS) to obtain a baseline separation of benzo(a)pyrene and benzo(e)pyrene (concentration ratio of 1:5) and at least 50% valley separation for benzo(b)fluoranthene, benzo(j)fluoranthene, and benzo(k)fluoranthene (concentration ratio of 1:1:1).
(2) The second step will be the calibration range test, where participants will prepare calibration standards and obtain normalized calibration curves for the studied PAHs vs. respective labeled internal standards (13C-PAHs). Collaborators will determine the linear range, test for carry-over by injecting a solvent blank after the highest standard, and adjust injection conditions (such as injection volume, number of washes, syringe size, etc.) to achieve low detection limits, acceptable linearity for the tested concentration range, and minimum carry-over. Coefficients of determination (r2) of 0.990 or greater will be required in order for the participant to proceed with practice samples. For each analyte, r2 values and back-calculated concentrations of the calibration standards will be reported to the Study Directors.
(3) The third step will be the lowest calibration level (LCL) spike test, where collaborators will prepare and analyze 7 spiked samples using blank shrimp matrix and a mixed PAH spiking solution which will both be supplied to them. The samples will be spiked at PAH concentrations equivalent to the target lowest calibration level (0.5 µg/kg for BaP) to test instrument sensitivity and lab reproducibility. The shrimp matrix should be stored in a freezer set to maintain -70 ± 10°C or at least -20 ± 10°C. The mixed PAH spiking solution should be stored in refrigerator set to maintain 5 ± 3°C.
(4) The fourth step will be analyzing practice samples. Three practice samples will be supplied to the participants and two of the three samples will be shrimp blank matrix already spiked with two different mixed PAH solutions (BaP levels of 2 - 50 µg/kg, other PAHs at 2 - 250 µg/kg). The third sample will be the NIST Standard Reference Material 1974b, which is a mussel matrix with certified concentrations of incurred PAHs and other organic contaminants. The practice samples will arrive ready for analysis labeled with unique identification numbers (PSFC1 to PSFC3). All practice samples will be shipped frozen on dry ice with a return receipt document to be returned to the Study Directors. The practice samples should be stored in a freezer set to maintain -70 ± 10°C or at least -20 ± 10°C.
Laboratory qualification and practice sample results must be approved by the Study Directors before proceeding with the test sample analysis.
Method. The method to be used for this collaborative study is provided as Appendix B. Method performance parameters were demonstrated in the study conducted by the ICT laboratory for 32 PAHs in shrimp and trout. Limit of quantification (LOQ) for BaP was 0.05 µg/kg and 0.05 to 0.25 µg/kg for the other studied PAHs. The target LCL of 0.5 µg/kg for BaP was selected for the collaborative study to routinely meet the fitness-for-purpose LOQ requirement of 1 µg/kg stated by the AOAC Stakeholders Panel on Seafood Contaminants (SPSC).
The recovery and repeatability, expressed as relative standard deviation RSD (%), were determined for six replicates of the two different matrix samples spiked with target analytes at two concentration levels of 1 and 5 μg/kg. For smoked trout, the PAH recoveries ranged from 73% to 97% with RSDs of 2-13%. For shrimp, the PAH recoveries ranged from 73 to 109% with RSDs of 2-15%. The method trueness was verified by analysis of a NIST Standard Reference Material 2977 (mussel tissue) sample, with results being within the certified values. In addition, a NIST Standard Reference Material 1974b (also mussel tissue) will be analyzed by the ICT laboratory in 7 replicates to demonstrate method precision for the analysis of incurred PAHs under single laboratory conditions. These results will aid in the evaluation of the collaborative study results. If the collaborative study proves to be successful, the method will be recommended for Official First Action.
Quality Assurance. The method uses isotopically labeled 13C-PAH surrogate standards that will be added at 5 µg/kg to the samples prior to the extraction process. Quantitation will be based on linear least-squares calibration of analyte signals (peak areas or heights) divided by signals of respective 13C-labelled internal standards plotted versus analyte concentrations. Eight concentration levels will be used for the calibration, corresponding to 0.5, 1, 2, 5, 10, 20, 50 and 100 µg/kg for BaP and other lower-level PAHs and to 1.25, 2.5, 5, 12.5, 25, 50, 125 and 250 µg/kg for higher-level PAHs. Coefficients of determination (r2) should be 0.990 or greater and back-calculated concentrations of the calibration standards should not exceed ±20% of theoretical. If a well-characterized quadratic relationship occurs, then a best-fitted quadratic curve may be employed for calibration. Otherwise, if the back-calculated concentrations exceed ±20% of theoretical, normalized signals of the nearest 2 calibration standards that enclose the analyte signal in the sample will be used to interpolate the analyte concentration.
In addition to reporting r2, back-calculated calibration standard concentrations, and analyte concentrations for each analytical run of practice and test samples, the collaborators will also be required to report ion ratios as a means of verifying identification of the analyte peaks. For each analyte, the sample relative abundance should be within ±20% (arithmetic difference) of the target relative abundance.
A solvent (isooctane) blank will be injected before and after each calibration set. Reagent (procedural) blanks will be analyzed with each set of samples.
During homogenization, portions of the blank matrices will be spiked with 1,7-dimethylphenanthrene, which will serve as a homogenization check of the sample processing step.
Reporting Raw Data. Individual concentration values will be reported for each test sample for a total of three hundred ninety nine (399) data points. The blanks will serve as a negative control. Participants will be asked to also supply PAH and 13C-PAH signals (peak areas or heights) and other parameters as described above in Quality Assurance. Excel forms will be sent electronically to the participants for ease of use. The Study Directors will generate tables of the raw data from the information received.
Analyzing Raw Data. Average analyte concentrations, standard deviation (SR) and relative standard deviation (RSDR) for reproducibility, number of statistical outlier laboratories, HORRAT value (RSDR/predicted RSDR), and percent recovery will be determined for all data (single replicate and blind duplicate) after removal of outliers. The Cochran test will be used to determine outliers for blind duplicates; the Grubbs tests will be used to determine outliers for both single replicates and blind duplicates. In addition, standard deviation (Sr) and relative standard deviation (RSDr) for repeatability will be determined for blind duplicate data.
Miscellaneous. Additional appendices contain the cover letter to participants of this study (Appendix C), a study materials receipt form (Appendix D), and instructions to collaborators (Appendix E).
References
Lucie Drabova, Kamila Kalachova, Jana Pulkrabova, Tomas Cajka, Vladimir Kocourek and Jana Hajslova. “Rapid Method for Simultaneous Determination of Polycyclic Aromatic Hydrocarbons (PAHs), Polychlorinated Biphenyls (PCBs) and Polybrominated Diphenyl Ethers (PBDEs) in Fish and Sea Food Using GC-TOFM,” ICT document, Prague, Czech Republic, 2010.
Appendix A. Collaborative Study Participants