Fs in

Article pubs.acs.org/JAFC

Occurrence and Estimated Dietary Intake of PCBs and PCDD/Fs in Functional Foods Enriched with Omega‑3 from Spain Á ngel García-Bermejo, Laura Herrero, María-José González, and Belén Gómara* Department of Instrumental Analysis and Environmental Chemistry, Institute of General Organic Chemistry (IQOG-CSIC), Juan de la Cierva 3, 28006 Madrid, Spain S Supporting Information *

ABSTRACT: The polychlorinated biphenyl (PCB), polychlorinated dibenzo-p-dioxin, and dibenzofuran (PCDD/F) contents of six functional foods enriched with omega-3 were characterized. All the samples analyzed showed concentration levels below the maximal levels established by Regulation EC 1259/201120. PCB concentrations were higher than those of PCDD/Fs; oil supplements were the most contaminated samples [1.8 pg of WHO-TEQ/g of lipid weight (lw)] followed by chicken eggs (1.3 pg of WHO-TEQ/g of lw), cow’s milk (0.23 pg of WHO-TEQ/g of lw), biscuits (0.15 pg of WHO-TEQ/g of lw), soy milks (0.11 pg of WHO-TEQ/g of lw), and soy lecithin (0.049 pg of WHO-TEQ/g of lw). The most abundant non-dl-PCBs were PCBs 52 and 101 in cow’s milk, soy products, and biscuits, while in chicken eggs and oil supplements, they were PCBs 153 and 138. PCBs 118 and 105 were the most frequent dl-PCBs in all samples. Only oil supplements presented quantifiable concentrations for almost all PCDD/Fs, OCDD and OCDF being the most abundant. The estimated daily intake was 2.7 pg of WHO-TEQ/day for chicken eggs, 0.91 pg of WHO-TEQ/day for cow’s milk, 0.45 pg of WHO-TEQ/day for soy milks, and 0.44 pg of WHO-TEQ/day for biscuits. For oil supplements, it was more variable, but always higher. KEYWORDS: polychlorinated biphenyls, polychlorinated dibenzo-p-dioxins and dibenzofurans, enriched foodstuffs, omega-3, human exposure, daily intake

■

through the diet in Spain during the past decade.5−8,11−16 However, no studies of commercially available omega-3-enriched foods have been reported, and only one has examined oil-based health supplements. Therefore, it is interesting to evaluate the presence of these contaminants in the aforementioned commercially available enriched food products, to study if their composition could represent a risk to the health of consumers. The aim of this work was to characterize different types of commercially available functional foods enriched with omega-3 PUFAs based on their PCB and PCDD/F concentrations and profiles. An estimation of the daily intake of those pollutants due to the consumption of these enriched foodstuffs is also reported. Additionally, results were compared with those of similar surveys performed recently in Spain to determine if enriched foodstuffs could pose a human health risk.

INTRODUCTION Polychlorinated biphenyls (PCBs), polychlorinated dibenzo-pdioxins (PCDDs), and polychlorinated dibenzofurans (PCDFs) are among the most well-known and researched families of persistent organic pollutants (POPs). They are ubiquitous in the environment and highly toxic to humans and the environment.1 Because of their high persistence and lipophilic nature, they bioaccumulate in the fatty tissues of animals, are concentrated in the food chain, and thus reach and accumulate in the human body, causing adverse health effects. The toxicity of these compounds is associated with the negative effects on development, the immune and reproductive systems, brain development, and learning ability and promotes the growth and development of cancer cells.2−4 It is widely accepted that dietary intake is the main route of human exposure to these contaminants for the general population.5−8 In addition, foods with a high fat content, mainly those of animal origin such as meat, fish, eggs, milk, and their derived products, are some of the major contributors to the intake of these pollutants. In addition, today, foods enriched with omega-3 polyunsaturated fatty acids (PUFAs) are increasingly present in consumer baskets, because of their beneficial properties for health. However, this enrichment generally occurs through the addition of fish oils to foodstuffs, and it is known that fish with a high omega-3 PUFA content can also contain relatively high concentrations of PCBs and polychlorinated dibenzo-p-dioxins/dibenzofurans (PCDD/Fs).9,10 Thus, it is hypothesized that these types of foods may be potentially more contaminated than conventional ones. On the other hand, several researchers have studied the levels of PCBs and PCDD/Fs in foodstuffs and human exposure © XXXX American Chemical Society

■

MATERIALS AND METHODS

Reagents and Standards. All solvents used were of Pestipur quality and were purchased from SDS (Peypin, France), except n-hexane (Merck, Darmstadt, Germany). Sulfuric acid was of pro-analysis quality (Merck). Anhydrous sodium sulfate was obtained from J. T. Baker (Deventer, The Netherlands) and silica gel 60 from Merck. SPE cartridges of Supelclean Envi-Carb (graphitized carbon pack, 250 mg, 3 mL tubes; Supelco, Palo Alto, CA) were used for final fractionation of the PCBs according to their planarity and for separating them from PCDD/Fs. Received: Revised: Accepted: Published: A

February 20, 2017 April 7, 2017 April 8, 2017 April 9, 2017 DOI: 10.1021/acs.jafc.7b00785 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

290 °C (5 min) at a rate of 7 °C/min. For dl-PCBs, the most intense transitions selected for the SRM method were those from the molecular cluster to the loss of two chlorine atoms. In the case of PCDD/Fs, the loss of the COCl• group was the most abundant transition and the second transition was selected considering the highest sensitivity and selectivity.25 The iLOQs ranged from 0.23 to 2.5 pg/μL for dioxins and from 0.16 to 2.1 pg/μL for dl-PCBs. For the non-dl-PCBs, a similar optimization of the instrumental method was conducted. Table S1 summarizes the final transitions selected and the optimal CID (collision-induced dissociation) voltages for each transition to take place. Precision, linearity, and instrumental limits of detection (iLODs) for the non-dl-PCBs were also studied in the same way that they were for dl-PCBs.25 The precision results obtained are listed in Table S2. Relative standard deviation (RSD) values were always 0.994. iLODs, which were calculated as the concentration corresponding to 3 times the standard deviation (SD) of the signal of three replicate injections of a standard solution close to the limit of detection, ranged from 0.11 to 0.20 pg/μL (Table S2). Quantification was based upon relative response factors in accordance with the isotope dilution method. Relative response factors were measured for each individual compound by the analysis of six different calibration solutions for both PCBs and PCDD/Fs. Finally, the results were expressed as World Health Organization-toxic equivalent values (WHO-TEQs).1 Quality Control Criteria. All analyses that followed the quality criteria such as blanks, recoveries, and parallel analysis complied with analytical standards as recommended by the U.S. Environmental Protection Agency (U.S. EPA) Methods 161326 and 1668A27 and the European Union (EU) Commission in the directive for measuring dioxins and PCBs in food.28 A procedural blank was performed in each batch of samples. To minimize possible interference in blanks, all the glassware, chemicals, solvents, and equipment used during extraction, purification, and cleanup stages and the analytical instrumentation were routinely checked. Recovery rates of the 13C12-labeled PCB and PCDD/F compounds added to samples before the extraction step (WP-LCS and EPA1613-LCS) were >60% in all cases.

The selected congeners for this study were the most toxic of the three families of compounds (PCBs and PCDD/Fs), which have a toxic equivalence factor (TEF) assigned by the World Health Organization (WHO),1 and the most abundant PCBs in technical mixtures.17,18 The following congeners were chosen for the determinations: the 17 2,3,7,8substituted PCDD/F congeners,19 four non-ortho-substituted PCBs (77, 81, 126, and 169), and eight mono-ortho-substituted congeners (105, 114, 118, 123, 156, 157, 167, and 189), both of them called dioxinlike PCBs (dl-PCBs), and the six indicator PCBs gathered in the legislation (ICES-6; 28, 52, 101, 138, 153, and 180).20 PCB congeners are named according to the IUPAC nomenclature.21 Standard solutions of PCDD/Fs (EPA1613 CS0.5, CSL, CS1−CS4, LCS, and ISS) and dl-PCBs (WP CS1−CS6, LCS and ISS) were purchased from Wellington Laboratories Inc. (Guelph, ON), and standards of individual native non-dl-PCBs were purchased from Dr. Ehrenstorfer (Augsburg, Germany). These solutions were used for calibration, quantification, and analytical recovery calculations. Sample Collection. Six different types of commercially available functional foods enriched with omega-3 PUFAs, i.e., cow’s milk (n = 3), chicken eggs (n = 3), soy milks (n = 2), soy lecithin (n = 1), biscuits (n = 2), and oil dietary supplements (n = 9), were purchased in different supermarkets in Madrid, Spain, between 2010 and 2012 and were analyzed for PCB and PCDD/F determinations. Samples were analyzed individually and then the results grouped to evaluate median levels and ranges in each food type. Sample Preparation. The sample treatment of the different types of foodstuffs was already described in a previous work.22 Briefly, cow’s milk, chicken eggs, soy products, and biscuits were lyophilized, extracted, and purified according to the method previously described by Bordajandi et al.,23 while for oil supplement samples, the treatment described by Bernardo et al.24 was followed. After extraction and purification, the extracts were subjected to a final fractionation/cleanup step on Envi-Carb SPE cartridges.23 The first fraction (35 mL) contains mono-ortho- and non-dl-PCBs and the second fraction (20 mL) the non-ortho PCBs, and the third fraction (60 mL) was eluted in reverse flow for PCDD/Fs. The extracts were rotary evaporated until a volume of approximately 1−2 mL was reached. The final extracts were transferred to conical bottom injection vials, evaporated to dryness under a gentle nitrogen stream, and reconstituted with the appropriate injection standard: first fraction, 100 μL of WP-ISS; second fraction, 10 μL of WP-ISS; third fraction, 5 μL of 1613-ISS and 5 μL of nonane. All 20 samples were processed in batches, one batch for each type of matrix, including a procedural blank. All samples within a batch were extracted, purified, and concentrated in parallel, and all the final extracts were injected and analyzed by gas chromatography coupled to triplequadrupole tandem mass spectrometry [GC−QqQ(MS/MS)] just after they had been obtained. Instrumental Analysis by GC−QqQ(MS/MS). Instrumental determination of PCBs and PCDD/Fs was performed on a TRACE GC Ultra gas chromatograph (Thermo Fisher Scientific, Milan, Italy) equipped with a triple-quadrupole analyzer (TSQ Quantum XLS, Thermo Fisher Scientific, Bremen, Germany) that was operated in positive electron ionization mode (EI+) and in the SRM (selective reaction monitoring) detection mode with a resolution of a 0.7 Da peak width. The equipment was controlled using the Xcalibur data system. Chromatographic conditions as well as the different parameters affecting MS/MS detection of dl-PCBs and PCDD/Fs have been published elsewhere.25 Briefly, injections (1 μL) were performed in the programmable temperature vaporization (PTV) mode, and separation was performed in a capillary HP-5MS column [30 m × 0.25 mm (inside diameter), 0.25 μm film thickness] purchased from Agilent Technologies (Palo Alto, CA). Helium was used as the carrier gas at a constant flow rate of 1.2 mL/min, and the temperatures of the transfer line and the MS source were set to 300 and 240 °C, respectively. The collision gas (Ar) pressure was set to 1.5 mTorr for all the SRM experiments. The PTV program was as follows: 90 °C, held for 0.05 min, heated to 300 °C at a rate of 10 °C/s, held for 1.5 min, heated to 330 °C at a rate of 10 °C/s, and held for 35 min (splitless time of 1.5 min). The oven temperature was programmed from 90 °C (2 min) to 160 °C at a rate of 15 °C/min, then to 225 °C at a rate of 4 °C/min, and then to

■

RESULTS AND DISCUSSION PCB and PCDD/F Concentrations in Food Samples Enriched with Omega-3 PUFAs. Concentrations [median and range in picograms per gram of fresh weight (fw)] of each compound, as well as the total sum of PCBs and PCDD/Fs [expressed in both picograms per gram of fw and picograms per gram of lipid weight (lw)], and the TEQ (picograms of WHO-TEQ per gram of lw)1 of dl-PCBs, PCDD/Fs, and the sum of dl-PCBs and PCDD/Fs in the food samples studied are listed in Tables 1 and 2. The lipid content of the samples is also included in Table 1. All the samples analyzed showed concentration levels below the maximal levels established by Regulation EC 1259/2011.20 In addition, as expected, total PCB concentrations were always higher than total PCDD/F concentrations in all types of foodstuffs analyzed. Among studied PCBs, the non-dl-PCBs made the largest contributions to the total PCBs, with a median contribution of 92%, while for dl-PCBs, that was 8%. The highest ΣICES-6 PCB concentrations were found in oil supplements (median of 6820 pg/g of fw, range of 1759−58900 pg/g of fw), followed by soy lecithin (3119 pg/g of fw), chicken eggs (median of 2355 pg/g of fw, range of 1762−2953 pg/g of fw), biscuits (median of 1429 pg/g of fw, range of 681−2178 pg/g of fw), cow’s milk (median of 724 pg/g of fw, range of 604−992 pg/g of fw), B

DOI: 10.1021/acs.jafc.7b00785 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

C

range 39−70 228−476 136−263 57−111 55−111 27−56 604−992 0.51−2.6 0.022−0.12 6.9−22 0.36−1.2 21−59 0.74−1.7 0.020−0.17 1.9−7.7 0.32−1.4 3.0−11