Food Chemistry 169 (2015) 464–470

Contents lists available at ScienceDirect

Food Chemistry journal homepage: www.elsevier.com/locate/foodchem

Analytical Methods

Determination of toxic heavy metals and speciation of arsenic in seaweeds from South Korea Naeem Khan a, Keun Yeoung Ryu a, Ji Yeon Choi a, Eun Yeong Nho a, Girum Habte a, Hoon Choi b, Mee Hye Kim b, Kyung Su Park c, Kyong Su Kim a,⇑ a b c

Department of Food and Nutrition, Chosun University, Gwangju 501-759, Republic of Korea Food Contaminants Divisions, Food Safety Evaluation Department, Ministry of Food and Drug Safety, Cheongwon 363-951, Republic of Korea Korea Institute of Science and Technology, Seoul 136-791, Republic of Korea

a r t i c l e

i n f o

Article history: Received 19 October 2013 Received in revised form 5 May 2014 Accepted 7 August 2014 Available online 14 August 2014 Keywords: Seaweeds Toxic heavy metals As speciation ICP-MS ICP-OESLC-ICP-MS

a b s t r a c t This study aimed at determining the levels of toxic heavy metals including As, Pb, Cd, Al, Hg and As species, such as, As-III, As-V, MMA, DMA, AsB, and AsC in various edible species of seaweeds from South Korea. ICP-MS was used for determination of As, Pb and Cd, ICP-OES was used for Al, DMA was used for Hg, and LC-ICP-MS was used for As speciation. The analytical methods were validated by linearity, detection limits, precision, accuracy and recovery experiments, obtaining satisfactory results in all cases. From the results toxic heavy metals were found in the decreasing order of: Al > As > Pb–Cd > Hg. Generally concentrations of all analysed heavy metals and both organic and inorganic species of As were very low compared to PTWIs specified by JECFA and EC. Their contribution to the overall intake by the subject seafoods was found very low and thus would not pose any threat to consumers. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction South Korea is surrounded by the sea on the eastern, western and southern sides. A large variety of industrial complexes and municipal cities are located in coastal areas and around 33% of the national population lives along the entire coast comprising around 2,900 km2 (Kim, Lee, Oh, & Kahng, 2000). Due to such geographical location, seafoods contribute a significant part to their daily foods (Hsieh & Jiang, 2012). Seaweeds are well known seafoods, most popular in China, Japan and Korea, although they are also used in other Asian countries, and in countries where there are ethnic Asian communities. Seaweeds are proved to be the rich sources of mineral elements than other usual edible plants and therefore often recommended as food supplements to help meet daily intake of essential mineral and trace elements (Rupérez, 2002). According to the Food and Agriculture Organization (FAO), Fisheries and Aquaculture Department statistics for the year 2011, the total world aquaculture production of seaweeds was 21.0 million tones, with 20.80 million tones only from Asia. China was the largest producer with 11.5 million tones of seaweeds followed by Indonesia (3.9 million tones), Philippines (1.8 million tones) and Republic of Korea (444,300 tones) (FAO, 2013). The harvest of cultured ⇑ Corresponding author. Tel.: +82 62 230 7724; fax: +82 62 224 8880. E-mail address: [email protected] (K.S. Kim). http://dx.doi.org/10.1016/j.foodchem.2014.08.020 0308-8146/Ó 2014 Elsevier Ltd. All rights reserved.

seaweeds from coastal waters removes nearly a million tones of proteins, with around 150,000 metric tones of nitrogen annually (Troell et al., 2003). In Korea, seaweeds are eaten raw, cooked or processed. Also many cosmetic and pharmaceutical products contain seaweed polysaccharides namely agars, alginates and carrageenans (Chung, Beardall, Mehta, Sahoo, & Stojkovic, 2011). The significant alterations of industrial development in the recent past, lead to an increased discharge of chemical effluents into the environment, resulting to damage of aquatic life in many countries around the world. Heavy metals discharged into the marine environment can damage marine species and whole ecosystem, due to their accumulative behaviour and well known toxicity (Sivaperumal, Sankar, & Viswanathan-Nair, 2007). The impacts of toxic elements like, Al, Pb, Hg, Cd and As on human health and the environment are of great interest, especially from aquatic products prospective (Uluozlu, Tuzen, Mendil, & Soylak, 2007; Verstraeten, Aimo, & Oteiza, 2008; Yabanli, Alparslan, & Baygar, 2012). These elements can be very damaging even at low levels when ingested over a long period of time. Even the essential metals are known to produce toxic effects when their intake is excessive to their recommended levels (Celik & Oehlenschlager, 2007). Aluminium is proposed to be involved in the pathophysiology of neurodegenerative disorders (Parkinsonism dementia, Alzheimer’s disease). Pb is a widespread environmental hazard, and the

N. Khan et al. / Food Chemistry 169 (2015) 464–470

neurotoxic effects of Pb are a major public health concern (Verstraeten et al., 2008). Mercury is another dangerous xenobiotic, especially its vapours and water soluble salts. It has the ability to accumulate in the internal organs of living organisms (Boszke, Siepak, & Falandysz, 2003). In the past century, the anthropogenic inputs of mercury into the environment also significantly increased and therefore monitoring these inputs and estimating their many possible adverse impacts is very important, along with other toxic elements across the food chain, especially seafoods (Konieczka, Misztal-Szkudlin´ska, Namies´nik, & Szefer, 2010). Similarly research studies have indicated that arsenic is abundant in seafoods at concentrations as high as several hundred lg/g. Therefore, it is essential to know the concentrations of individual arsenic species, along with total arsenic, to realise the level of toxicity in countries, such as China, Japan, Taiwan, and Korea, where food from marine sources constitutes a major part of the diet (Hsieh & Jiang, 2012). This study was designed to monitor the exposure of the Korean population to arsenic, cadmium, mercury, aluminium and lead from seaweeds, due to increasing concern about the intake of toxic contaminant elements in foods, by determining their levels in all edible species, commercially available to consumers all around South Korea. Furthermore, the speciation of the important organic arsenics were considered including monomethylarsonic acid (MMA), dimethylarsinic acid (DMA), arsenobetaine (AsB), and arsenocholine (AsC) and inorganic species such as arsenite (As-III), and arsenate (As-V), in order to know about their contamination levels if any in the food subjects. The provisional tolerable weekly intakes (PTWIs) recommended by Joint FAO/WHO Expert Committee on Food Additives (JECFA) and European Commission (EC) was used for all analytes to compare and determine any toxicity from seaweeds to the consuming population.

2. Materials and methods 2.1. Instrumentation A microwave system (Multi-wave 3000, Anton Paar, Graz, Austria) was used to digest the samples. It was programmable for time and power between 600 and 1400 W, and equipped with 16 high pressure PTFE (polytetrafluoroethylene) vessels (MF 100), digestion tubes. The quadrupole, inductively coupled plasma mass spectrometer (ICP-MS) used for determination of As, Cd and Pb, was Elan DRC II (Perkin-Elmer SCIEX, Norwalk, CT, USA). It had a high efficiency sample introduction desolvating system equipped with a quartz cyclonic spray chamber and an additional mixing peristaltic pump (APEX-IR, Omaha, NE, USA). The operating conditions were: forward power 1.35 kW, argon gas flow rate 16.00 L/min (plasma); 1–1.3 L/min (auxiliary), 1–1.07 L/min (nebulizer). The argon gas utilised was of spectral purity (99.9998%). Before each experiment, the instrument was tuned for daily performance, using Elan 6100 DRC Sensitivity Detection Limit Solution, PerkinElmer Pure (N8125034) USA. A Varian Model 730-ES simultaneous CCD, inductively coupled plasma-optical emission spectrometer (ICP-OES), (Wyndmoor, PA, USA), was used for determination of Al in the samples. This spectrometer had a SeaSpray concentric nebulizer (Glass Expansion, Pocasset, MA) and cyclonic spray chamber. Operating conditions for ICP-OES instrument were: forward power 1.3 kW, argon gas flow rate 16.00 L/min (plasma); 1.5 L/min (auxiliary), 0.94 L/mi, (nebulizer), wavelength 396.153 nm. The background correction wavelength was selected manually at appropriate background position for the analyte peak, after scanning a blank, a standard solution and a sample solution in the programmed wavelength range.

465

Mercury analyser (MA-2000; NIC, Japan) was used to determine the content of total mercury in the samples. It was an automatic direct mercury analyser, with sample preheating furnace, decomposing furnace, mercury-gold amalgamation chamber and Silicon UV photodetector. The instrument was also attached with auto sample changer (NIC BC-1). A Shimadzu High Performance Liquid Chromatography System (HPLC, LC-VP series, Kyoto, Japan), was coupled to ICP-MS for quantification of six As species. The quadrupole mass analyser was operated in the single ion monitoring mode (m/z 91) for detecting arsenic and dynamic reaction cell (DRC) mode using oxygen. The instrumental specification and various operating conditions followed for LC-ICP-MS were as mentioned in Table 1. 2.2. Reagents The analytical reagent grade concentrated HNO3 (70%) was obtained from Dong Woo Fine-Chem Co., Ltd. Iksan, Korea. For calibration curves of metals, standard stock solutions of 100 mg/L, for all analytes were purchased from AnApure KRIAT Co, Ltd. Daejeon, Korea. The certified reference material (NMIJ CRM 7405-a), Hijiki Seaweed, was obtained from National Metrology Institute of Japan/National Institute of Advanced Industrial Science and Technology (NMIJ/AIST), Ibraki, Japan. Sodium (meta) arsenite [NaAsO2; As(III); 99.0%] and sodium arsenate dibasic heptahydrate [Na2HAsO4.7H2O; As(V); 99.9%] were purchased from Sigma– Aldrich (St. Louis, MO, USA). Dimethylarsinic acid [(CH3)2AsO(OH); DMA; 98%], monosodium acid methane arsonate sesquihydrate [(CH3)AsO(OH)2; MMA, 99.0%], arsenobetaine [[(CH3)3As]+CH2COO ; AsB, 97%], and arsenocholine [[(CH3)3As]+CH2CH2OH, Br ; AsC, 95%] were purchased from Strem Chemicals (Newburyport, MA, USA), Chem Service (West Chester, PA, USA), Fluka (Buchs, Switzerland), and Wako Pure Chemical Industries (Osaka, Japan), respectively. Ultrapure deionised water with a resistivity of 18.2 MO cm was obtained from a Milli-Q Plus water purification system (Millipore, Bedford, MA, USA). 2.3. Apparatus All containers were thoroughly cleaned with a detergent solution, rinsed with metal free water, and soaked for overnight, or longer in a covered acid bath containing 10% HNO3 (v/v) solution. These were rinsed several times with de-ionised metal free water and dried in drying oven (HB-502M, Hanbaek Co., Ltd. Korea). All plastic containers, polypropylene flasks, pipette tips, PTFE vessels digestion tubes and reagents that came into contact with samples or standards were checked for contamination. 2.4. Samples preparation and digestion Samples (198) belonging to five species of edible seaweeds commercially available to consumers were purchased from super markets all around South Korea. These species included laver (Porphyra tenera) (53), seatangle (Laminaria japonica) (45), sea mustard (Undaria pinnatifida) (58), hijiki (Hizikia fusiforme) (27) and gulf weed (Sargassum fulvellum) (15). All the samples were purchased at different time intervals during April to October 2012, in triplicate thus making a total of 594 samples analysed. In each variety, the samples belonged to different packaging and brands. These were rinsed with tap water followed by deionised distilled water and air dried in a clean fume hood. For determination of moisture, each sample was dried at 105 °C in oven (HB502M, Han Back, Korea), until constant weight was achieved. These were powdered in a blender (MR 350CA, Braun, Spain), properly labelled and stored in polyethylene bags in refrigerator (MICOM CFD-0622, Samsung, Korea) at 20 °C until analysis.

466

N. Khan et al. / Food Chemistry 169 (2015) 464–470

Table 1 LC-ICP-MS operating conditions and measurement parameters for arsenic speciation. LC condition Instrument Column Injection volume Column temp. Mobile phase Gradient program

Shimadzu HPLC (LC-VP series, Kyoto, Japan) Hamilton PRP X-100 (4.1  250 mm, 10 lm) 100 lL Room temp. A: 2 mM ammonium bicarbonate in 1% MeOH, pH 8.0 B: 20 mM ammonium nitrate. 20 mM ammonium phosphate in 1% MeOH, pH 9.2 Time (min)

Flow rate (mL/min)

%A

%B

0.0 2.0 2.5 7.5

1.0 1.0 1.5 1.5

100 100 0 0

0 0 100 100

ICP-MS condition Spectrometer Nebulizer RF power (kW) Argon flows (L/min) Optimisation Scanning condition Analysis mode Analytical mass (amu)

Elan 6100 DRC II (Perkin-Elmer SCIEX, Norwalk, CT, USA) Meinhard nebulizer with cyclonic spray chamber 1.35 Plasma (16.00); Auxiliary (1–1.3), Nebulizer (1–1.07) On masses 9 Be, 59 Co, 115In, and 238U Dwell time, 250 ms; sampling 3.9 pts/s; Reading, 3320 DRC mode, O2 0.5 mL/min 91AsO+

For digestion 0.25–0.5 g of samples were directly weighed into PTFE digestion vessels. Then 7.0 mL concentrated HNO3 (70%) and 2.0 mL H2O2 catalyst were added and digested using microwave system. The combustion procedure was as follow: (1) 1000 W at 80 °C for 5 min, (2) 1000 W at 50 °C for 5 min, (3) 1000 W at 190 °C for 20 min, and (4) 0 W for 30 min for cooling. The contents of the tubes were then transferred to 50 mL self standing polypropylene volumetric tubes with plug seal caps (Corning NY, Mexico). These volumes were then diluted to 25.0 g with ultrapure deionised water, labelled accurately and used for analysis of Pb, Cd and As, by ICP-MS and for Al by ICP-OES analysis (Khan et al., 2013).

water. The calibration curves were built on 8 different concentrations, so that concentrations of all analytes in the samples were within linear range of calibration curves. Measurements were carried out using the full quantitative analysis mode. Polyatomic interferences were checked by measuring several isotopes of the toxic heavy metals and checking the isotopic ratio in the digested solution of the samples. The calibration standards were analysed at regular intervals during analysis as samples to monitor the drift of both the instruments. Also ultrapure deionised water blanks were frequently analysed alongside samples to check for any loss or cross contamination. Sample blanks were also prepared by completion of the full analytical procedure above mentioned except without samples and analysed (Khan et al., 2013).

2.5. Extraction of arsenic species 2.7. Analysis of mercury Extraction of arsenic species was performed by taking 1.0 g powdered samples in 50 mL self standing polypropylene volumetric tubes and adding 8 mL 50% methanol solvent in 1% HNO3. It was kept in a sonicator bath (Powersonic 420, Hwasin Technology, Seoul, Korea), at 30 °C for 30 min. After centrifuging at 5980g for 10 min, using a centrifuge MF 300 (Hanil Science Industrial, Inchon, Korea), the supernatant was filtered into 50 mL polypropylene volumetric tube through 0
45-lm pore size nylon membrane filters, (BDH, Leicester, UK). The extraction step was repeated twice and the two supernatants were combined and diluted to the final volume of 20 mL with 50% (v/v) methanol in 1% HNO3. Anion-exchange cartridge, OasisÒ MAX Cartridge (Waters, Milford, MA USA), was conditioned with 10 mL methanol followed by 10 mL 1% HNO3, without allowing the solid phase to dry. After application of 5 mL extract to a cartridge, the elute was collected and additionally eluted with 5 mL 1% HNO3. The two elutes obtained were combined and final volume was made up to 10 mL with 1% HNO3. Arsenic species of the resulting solution were separated and quantified by LC-ICP/MS (Choi, Park, Kim, & Kim, 2011). 2.6. Calibration procedure The external calibration technique was followed for the quantitative analysis of the samples. Standard solutions were prepared by dilution of the stock solutions with 19.6% (w/w) HNO3 (the same percentage of acid present in the samples) in ultrapure deionised

About 50 mg of samples was weighted directly into the sample boat of mercury analyser, along with catalysts; B (Al2O3) and M (Ca(OH)2+Na2CO3), above and below the samples. The samples were preheated for 60 s at 300 °C, for dryness and then heated for 180 s at 850 °C to decompose mercury under a clean airflow, which was collected in a gold amalgamation tube. The tube was then further heated at 155 °C to release mercury into UV-Photodetector and measured at 253.65 nm wavelength. Each sample was analysed in triplicate and Hg concentrations were reported on fresh weight basis in ppm, after considering the moisture contents (Chen & Chen, 2006; Konieczka et al., 2010). 2.8. Quality assurance The analytical methods followed were validated by measuring several quality parameters including detection limits, linearity, precision, accuracy and spike recovery experiments. The limits of detection (LOD) and limits of quantification (LOQ), were calculated with respectively three and ten times the standard deviation of the blank divided by the slope of the analytical curve. Linearity was evaluated from the calibration curves of the toxic elements and arsenic species using a non-weighted least-squares linear regression analysis method (Pacquette, Szabo, Thompson, & Baugh, 2012). Precision was obtained as percent coefficient of variation (CV%) from the relative standard deviation of 10 repeated determinations of one sample (Khan et al., 2014). The accuracy of the

467

N. Khan et al. / Food Chemistry 169 (2015) 464–470

methods were tested by analysing the CRM (NMIJ 7405-a), Hijiki Seaweed, for the determination of As, Pb, Cd, Al and As(V). Analytical quality control of the methods for the all analyte toxic elements and As species was also verified by spiking and then determining the recovery (%) (AOAC, 2012; Khan et al., 2013). 2.9. Statistical analysis Data were reported as mean ± standard deviation of triplicate measurements. Significant differences (p < 0.05) within means were analysed by one way analysis of variance (ANOVA) and Tukey’s honestly significant difference (HSD) test in the SPSS (Statistical Package for Social Sciences) Software Version 20 (IBM, New York, USA). 3. Results and discussion Table 2; enlist analytical methods validation parameters including linearity, LOD, LOQ, precision and accuracy, Table 3; give analysis of CRM (NMIJ 7405-a) Hijiki Seaweed, Table 4; show

Table 2 Validation of analytical methods followed via linearity, LOD, LOQ, precision and spike recovery measurements. Analyte

Linearity (R2)

Toxic elementsa As 0.9997 Pb 0.9999 Cd 0.9996 Al 0.9995 Hg 0.9993 Arsenic speciesb As-III 0.9995 As-V 0.9991 DMA 0.9996 MMA 0.9994 AsB 0.9993 AsC 0.9991

LOD (ppb)

LOQ (ppb)

Precision (CV%)

Spike recovery (%)

0.028 0.013 0.021 0.024 0.060

0.095 0.042 0.070 0.081 0.199

0.91 2.92 1.81 2.64 2.99

102 97.2 100 102 94.8

0.061 0.025 0.028 0.018 0.024 0.060

0.205 0.083 0.093 0.061 0.081 0.199

2.81 3.37 2.11 2.76 2.64 2.99

99.4 94.5 97.7 107 102 94.8

a Values obtained for As, Cd and Pb by ICP-MS, for Al by ICP-OES and for Hg by DMA. b Values obtained by LC-ICP-MS.

Table 3 Validation of analytical methods followed via analysis of CRM (NMIJ 7405-a), Hijiki Seaweed. Analytes

Certified value (ppm)

Observed value (ppm)

Recovery (%)

As Pb Cd Al As(V)

35.8 ± 0.9 0.43 ± 0.03 0.79 ± 0.02 147 ± 7 10.1 ± 0.5

34.6 ± 0.7 0.44 ± 0.06 0.78 ± 0.03 149 ± 5 9.8 ± 0.8

96.7 102.3 98.7 101.4 97.0

concentration (mean ± SD) of analysed toxic heavy metals obtained for each type of samples, Table 5: give concentration (mean ± SD) of six arsenic species, obtained for each type of samples, and Fig. 1; explain the LC-ICP/MS chromatogram for standard solution (5 ppb) of arsenic species, and calibration curves of As(III), As(V), MMA, DMA, AsB, and AsC determined in seaweed varieties from South Korea. Concentrations of all analytes are shown on fresh weight basis, calculated by incorporation of the moisture content. 3.1. Validation of analytical methods The important quality parameters determined for analytical methods validation including detection limits, linearity, precision, accuracy and spike recovery experiments proved that the analytical methods followed were fulfilling the required criteria in accordance to the specifications by Association of Official Analytical Chemists (AOAC) for analysis (AOAC, 2012; Khan et al., 2013). The values of LOD were in the range of 0.013–0.061 (ppb) and the LOQ were 0.042–0.205 (ppb) (Table 2). These LODs and LOQs thus allowed the determination of both toxic elements and As species at the required levels. All calibration curves were prepared with eight standard solutions, including the blank solution. Concentrations in the samples were within linear range of calibration curve and above the established lower linearity limit. The values of correlation coefficient (R2) calculated for all calibration curves were at least 0.9991 (Table 2), which was in accordance to AOAC criteria (AOAC, 2012). Precision is the degree of variability of results, without considering the sample variability. The CV% values obtained were below 3% (0.91–2.99%), thus fulfilling the required criteria for applicable analytical method. The spiking recoveries for all analyte toxic elements and As species were in the range of 94.5–107% which further confirmed that there was no significant loss or gain for each analyte during the digestion/extraction procedures (p > 0.05; Table 2). For the analysed CRM (NMIJ CRM 7405-a), the recovery percentage was in the range of: 96.7–102.3%, (Table 3). Importantly the recovery value means of the analytes for the CRM was found to be within the interval of confidence (p < 0.05) calculated for the values certified, confirming the applicability of the analytical method (Khan et al., 2013). Based on these results for the quality parameters analysed, the analytical methods followed in this research were found very efficient to apply for determination of the analyte toxic metals and As species in edible seaweed samples. 3.2. Toxic heavy metals in seaweeds Concentrations of toxic heavy metals in the analysed five species of commercially available edible seaweeds were found to be variable and largely dependent upon the sample type. The same trend was also reported by a research study on marine algae in Japan (Narukawa, Hioki, & Chiba, 2012). Compared to other heavy metals, Al and As were found high in gulf weed (52.1 and 6.48 ppm), laver (15.5 and 2.07 ppm), hijiki (6.56 and 4.49 ppm),

Table 4 Concentrations of toxic heavy metals in edible seaweed species expressed in ppm (fresh weight basis). Analyte metals

Laver (n = 53)

Seatangle (n = 45)

Sea mustard (n = 58)

Hijiki (n = 27)

Gulf weed (n = 15)

As Pb Cd Al Hg

2.07aA ± 0.53B 0.063a ± 0.040 0.109bc ± 0.066 15.50b ± 9.36