Article Cite This: J. Agric. Food Chem. XXXX, XXX, XXX−XXX
pubs.acs.org/JAFC
Analysis and Occurrence of Organophosphate Esters in Meats and Fish Consumed in the United States Lijun Han,†,‡ Yelena Sapozhnikova,*,‡ and Alberto Nuñez‡ †
College of Science, China Agricultural University, Beijing 100193, China U.S. Department of Agriculture, Agricultural Research Service, Eastern Regional Research Center, 600 East Mermaid Lane, Wyndmoor, Pennsylvania 19038, United States
‡
Downloaded via GUILFORD COLG on July 18, 2019 at 10:42:38 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: Organophosphate esters (OPEs) are chemicals extensively used as plasticizers and flame retardants in commercial and consumer products. In this study, we developed and validated a method for the analysis of 13 common OPEs in meat (chicken, pork, and beef) and fish (catfish and salmon) to study their occurrence in those foods in the United States. The method was based on QuEChERS extraction with acetonitrile and automated robotic cleanup of the extracts, followed by low pressure gas chromatography-tandem mass spectrometry (GC-MS/MS) analysis for 8 of the OPEs and ultrahigh-pressure liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS) analysis for 13 OPEs, including 8 overlapping OPEs. The developed method was validated in the muscle tissues at four spiking levels (5, 10, 20, and 40 ng/g). The background levels in the laboratory environment and materials presented a challenge for accurate quantification at low ng/g levels. UHPLC-QOrbitrap MS analysis was utilized to pinpoint the source of their contamination. OPEs were found in the water used in the liquid chromatography (LC) mobile phase, and flow injection analysis with organic mobile phase was suggested as an alternative to avoid OPEs contamination in LC-MS/MS analysis. The validated method was applied to the analysis of 68 real-world meat and fish samples from the U.S. markets by three instrumental methods. Tris(2-chloro-isopropyl) phosphate (TCPP), tri-n-butyl phosphate (TnBP), and triphenyl phosphate (TPP) were found in meat, and TCPP and TPP were measured in fish samples. The sum of median OPE concentrations (averaged for the three instrumental methods) measured in the meat and fish samples were 6.2 and 8.7 ng/g wet weight, respectively. No regulations on the maximum residue levels of OPEs permitted in food were found for the U.S. or other countries. KEYWORDS: organophosphate esters, analysis, monitoring, meats, fish
■
INTRODUCTION Organophosphate esters (OPEs) are high-volume production chemicals used as flame retardants and plasticizers in many consumer and commercial products, including furniture, textile, upholstery, electronics, paints, polyvinyl chloride (PVC) plastics, building materials, polyurethane foams, lubricants, and hydraulic fluids.1 Some OPEs have been used as flame retardants since the 1960s, but their worldwide production volumes have tripled after the ban of brominated flame retardants in the past decade.2 When used as flame retardants or product additives, OPEs are not covalently bound to the products, resulting into easy release and a ubiquitous presence in the surrounding environments. Indeed, OPEs have been reported in environmental samples, such as indoor dust, air, wastewater, marine and river waters, groundwater, soils, and sediments.1,3−5 Because of their ubiquitous nature, human exposure is expected to occur via dermal contact, inhalation/ ingestion, and dietary intake. So far, OPEs have been reported in human breast milk, urine, and placenta.6−9 Some OPEs, especially chlorinated ones, are relatively persistent in the environment,1,10 and most have octanol−water partition coefficients (log Kow) values >3,5 suggesting potential bioaccumulation. Structural similarity of OPEs with common organophosphate pesticides (malathion, diazinone, chlorpyrifos, etc.) implies similar potential toxicological effects. Most published reports thus far focused on OPEs levels in air, water, © XXXX American Chemical Society
soil, sediments, biota, and humans, and several recent publications have emerged on their occurrence in foods. Thus, Campone et al. analyzed salmon, cod, and mussels from Swedish markets for 13 OPEs, and none were detected.11 Guo et al. tested powder milk products in China for 9 OPEs, and triethyl phosphate (TEP) was measured at 0.4 ng/g.12 Similarly, TEP was the most frequently detected among other OPEs and was measured at 2.2 ng/g in the whole fish homogenate from Niagara, Lake Ontario, Canada.13 In another study by Ding et al., chicken, pork, beef, vegetable, tofu, eggs, milk, and cereal from Eastern China were analyzed for 10 OPEs and their measured sums ranged 1.1−9.6 ng/g.14 Lorenzo et al. reported concentrations of 9 OPEs at 3.3− 53.0 ng/g in brown trout and European eel from Spain.15 The most comprehensive study to date was conducted on the levels of 8 OPEs in 12 food groups in Sweden, including cereal, pastries, meat, fish, dairy products, eggs, fats and oils, vegetables, fruits, sugar, and beverages.16 The results of this study suggested that human exposure through dietary intake is Special Issue: 55th North American Chemical Residue Workshop Received: Revised: Accepted: Published: A
March 8, 2019 June 24, 2019 June 27, 2019 June 27, 2019 DOI: 10.1021/acs.jafc.9b01548 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry Table 1. Chemical Names and Structures of the Selected OPEs at UHPLC Elution Ordera
B
DOI: 10.1021/acs.jafc.9b01548 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry Table 1. continued
*LC-MS/MS detection only.
a
Cruz Biotechnology) and CDN Isotopes (Pointe-Claire, Quebec, Canada)) and quality control standards, p-terphenyl-d14 and 13Cphenacetin (Accustandard and Sigma-Aldrich), respectively. All standards were ≥92% purity. Individual OPE stock solutions were prepared at 1−2 mg/mL in MeCN for all, except TCEP, TCPP which were prepared in toluene and CDPP and TCrP were provided as 0.1 mg/mL solutions in toluene. A standard mixture of 14 OPEs was prepared at 35 μg/mL in MeCN. The IS mixture containing TCEP-d12 and TPP-d15 was prepared in MeCN at 5 μg/mL. The quality control standard for UHPLC-MS/MS analysis, 13Cphenacetin, was prepared at 4 μg/mL in MeCN, and the quality control standard for low-pressure (LP)GC-MS/MS analysis, pterphenyl-d14, at 0.88 mg/mL was added to an analyte protectant (AP) mixture, containing 25 mg/mL 3-ethoxy-1,2-propanediol, 2.5 mg/mL gulonolactone, 2.5 mg/mL sorbitol, and 1.25 mg/mL shikimic acid (all acquired from Sigma-Aldrich) in 2/1 MeCN/ water with 0.88% formic acid. Mini solid phase extraction (SPE) columns for an automated instrument top sample preparation (ITSP) cleanup were from ITSP solutions (Hartwell, GA) and contained 45 mg of sorbent mixture: anhydrous MgSO4, PSA (primary secondary amine), C18, Carbon X (20/12/12/1, w/w/w/w). Filter vials of 0.2 μm PVDF membranes were from Thomson Instrument (Oceanside, CA). Meat (pork, beef, chicken) and fish (catfish and salmon) samples for method development and validation were acquired from local grocery stores. The samples were partially frozen and comminuted using a Robot Coupe (Ridgeland, MS) Blixer 2 with dry ice and then transferred to glass jars and kept in a −20 °C freezer. Another set of meat and fish samples, including 22 fish (halibut, flounder, tuna, monk fish, mackerel, swai, salmon, tilapia, catfish) and 46 meat (beef, chicken, goat, turkey and pork) samples, was acquired from local supermarkets and supplied by the USDA Food Safety and Inspection Service (FSIS). These samples were used to test the method
equally important to the ingestion of dust exposure. While these reports presented data on occurrence of OPEs worldwide, there is only one recently published study on OPEs occurrence in the U.S. foodstuffs, which concluded that OPEs levels found in the U.S. meats and fish/seafood were higher than in cereals and dairy products.17 Therefore, the goal of this study was to develop and validate a fast and simple method for analysis of the 14 most commonly used OPEs (listed in Table 1) in commonly consumed meat (pork, beef, chicken) and fish (salmon, catfish) in the U.S. and apply the developed method for the analysis of meat and fish samples from the U.S. markets to generate information on their occurrence.
■
MATERIALS AND METHODS
Reagents and Materials. HPLC-grade acetonitrile (MeCN), MS grade water, and toluene were from Fisher Scientific (Pittsburgh, PA). Formic acid (>98% purity) and ammonium formate were from SigmaAldrich (St. Louis, MO). Magnesium sulfate and sodium chloride were purchased from UCT (Bristol, PA). Standards of trimethyl phosphate (TMP), triethyl phosphate (TEP), and tri-n-butyl phosphate (TnBP) were from ChemService (West Chester, PA); tris(2-chloroethyl) phosphate (TCEP), tris(2-chloro-isopropyl) phosphate (TCPP), tris(1,3-dichloro-2-propyl) phosphate (TDCPP), cresyl diphenyl phosphate (CDPP), and tricresyl phosphate (TCrP) were from AccuStandard (New Haven, CT); tripropyl phosphate (TPrP), triphenyl phosphate (TPP), 2-ethyl-hexyl-diphenyl phosphate (EHDPP), tris(2-butoxyehtyl) phosphate (TBEP), and tris(2-ethylhexyl) phosphate (TEHP) were from Sigma-Aldrich; tri-isobutyl phosphate (TiBP) was from Santa Cruz Biotechnology (Santa Cruz, CA). Isotopically labeled compounds used as internal standards (IS) TCEP-d12, TPP-d15 (Santa C
DOI: 10.1021/acs.jafc.9b01548 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry Table 2. MRM Parameters of 14 OPEs by UHPLC-MS/MSa no.
analyte
tR, min
1 2 3 4 5 6 7 8 9+10 11 12 13 14 IS IS QC
TMP TEP TCEP TPrP TCPP TDCPP TPP CDPP TiBP+TnBP TBEP TCrP EHDPP TEHP TCEP-d12 TPP-d15 13 C-phenacetin
1.95 4.13 4.95 6.03 6.20 6.89 7.02 7.29 7.34 7.62 7.80 8.09 9.85 4.95 6.97 4.05
MRM 1, m/z 141 183 287 225 327 431 327 341 267 399 369 363 435 299 342 181
→ → → → → → → → → → → → → → → →
109 99 99 99 99 99 152 229 99 299 165 251 99 234 159 110
MRM 2, m/z 141 183 287 225 327 431 327 341 267 399 369 363 435 299 342 181
→ → → → → → → → → → → → → → → →
MRM 3, m/z
79 81 225 183 175 321 215 152 155 199 243 215 113 167 82 139
183 287 225 327 433 327 341 267 399 369 363
→ → → → → → → → → → →
127 125 141 251 99 77 165 211 101 195 152
299 → 130 181 → 93
CE1, V
CE2, V
23 25 29 23 27 35 47 35 23 17 59 46 19 17 55 29
29 53 17 11 17 17 39 43 13 21 39 45 15 21 63 21
CE3, V 15 23 15 13 40 71 43 11 23 65 46 23 37
DP, V
CXP, V
71 34 95 36 121 86 81 77 51 70 91 61 66 136 92 91
6 4 10 4 6 6 8 8 6 6 8 9 9 10 10 6
a
CE - collision energy; DP - declustering potential; CXP collision cell exit potential. was 2 μL. Fixed parameters include 5 kV ion spray voltage and 300 °C source temperature. Flow injection experiments were performed with the same MS/MS parameters and MeCN with 0.1% formic acid at 0.1 mL/min. The run time was 2 min, and the injection volume was 10 μL for 1:5 diluted extracts. High-Resolution Accurate Mass (HRAM) UHPLC-Q-Orbitrap Analysis. Extracts prepared for UHPLC-MS/MS analysis were diluted 1:10 with MeCN. HRAM analysis of OPEs was performed with a Q-Exactive Plus (Thermo Fisher Scientific, Madison, WI) mass spectrometer equipped with an electrospray ionization (ESI) probe (HESI-II) in positive mode. Waters Nano Acquity UPLC was used with a BEH C18 column, 1.7 μm, 1 mm × 100 mm, with a 70 μL/min flow rate and 2 μL injection. Water (A) and MeCN (B) both with 1% formic acid were used in a gradient starting at 60% A until 0.5 min, 50% A at 3 min, 45% A at 4.5 min, 30% A at 8.5 min, and 0% A at 9 min, holding until 13.5 min and returning to 60% A at 14 min and held until 15.25 min. HRAM data acquisition was set up from 0 to 15.25 min with full MS (m/z 120−500) and mDIA similar to the settings described by Wang et al.21 In full MS, the resolution was set at 70 000 FWHM, AGC target was at 3 × 106, and maximum IT was at 200 ms. In mDIA, the resolution was at 17 500 FWHM, AGC target was at 106, maximum IT was at auto, loop count was at 10, MSX count was at 4, isolation window was at 54.0 m/z, and NCE stepped at 10, 30, 60. The capillary temperature was 250 °C, ion spray voltage was 3500 V, and sheath gas was 32 (arbitrary units). The instrument was calibrated with the Pierce LTQ Velos ESI positive ion calibration solution (Thermo Scientific) for a mass accuracy
