Effect of washing, soaking, and cooking methods on perfluorinated compounds in mackerel (Scomber japonicus)

Abstract Perfluorinated compounds (PFCs) are environmental pollutants, and dietary intake is a major route of human exposure to them. We aimed to see the effects of washing, soaking, and cooking (grilling, braising, frying, and steaming) on the change of PFCs in mackerel fillets and PFCs before and after each treatment were analyzed using LC‐MS/MS. Washing resulted in a decrease in the PFC content of mackerel (average 74%) comparing to control. Among the 19 PFCs detected, perfluorobutanoic acid and perfluorotridecanoic acid (PFTrDA) were found to be abundant after washing. Soaking mackerel in sake reduced its PFC content by 51%, whereas soaking in rice‐washed solution reduced by 80% comparing to control. All the four cooking methods were effective in reducing the PFC content of mackerel. The degree by which the PFC content decreased varied with the cooking method: grilling (91%), steaming (75%), frying (58%), and braising (47%) comparing to uncooked sample. In addition, when mackerel was braised with potato, PFCs decreased more in fillet than the ones without potato. PFCs in potato increased after cooking with mackerel. The excessive consumption through the mackerel was 0.1997 ng/kg bw/day and 0.7987 ng/kg bw/day, respectively. These exposure levels were well below the tolerable daily intake values of both compounds (PFOS, 150 ng/kg bw/day; PFOA, 1,500 ng/kg bw/day). The results of this study indicated that employing appropriate pretreatment and cooking methods could be an effective way to reduce the dietary exposure to PFCs in mackerel.


| INTRODUC TI ON
Perfluorinated compounds (PFCs) have been found in air, water, soil, and house dust (Jogsten et al., 2009). They have been widely used in industrial products such as coating agents, varnishes (Prevedouros, Cousins, Buck, & Korzeniowski, 2006), and food packaging materials (Begley et al., 2005). Perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) are the most stable and detectable PFCs (Lin, Panchangam, & Lo, 2009). The halflife of PFOA in human body is reported to be about 4.37 years (Renner, 2003) and chronic exposure to it is reported to cause liver damage by fat metabolism (Vestergren, Cousins, Trudel, Wormuth, & Scheringer, 2008). Also, PFOS shows higher bioaccumulation than PFOA (Haug et al., 2010). European Food Safety Authority (EFSA) (2008) reported the tolerable daily intake (TDI) for PFOS and PFOA as 150 ng/kg bw/day and 1,500 ng/kg bw/ day, respectively. Haug et al. (2010) reported that food is an important source of exposure to PFCs in humans, and the main sources of chronic human exposure to PFCs are contaminated drinking water and food (Jogsten et al., 2009). Several studies have been conducted for determining the concentration of PFCs in food (Haug et al., 2010;Heo, Lee, Kim, & Oh, 2013;Kannan et al., 2002;Martin, Mabury, Solomon, & Muir, 2003) and their exposure levels for human body (Tittlemier, Pepper, & Edwards, 2006;Vestergren et al., 2008). Seafood has been reported to show a higher PFC content than other food items (Haug et al., 2010;Trudel et al., 2008), and Korea is one of the highest seafood consuming countries in the world (Choi et al., 2017). Though the dietary exposure to PFCs needs to be controlled (Jogsten et al., 2009), the study for the dietary exposure in Korean population is very limited. Therefore, it is important to monitor the amount of PFCs and analyze their changes during cooking in foods.
Mackerel (Scomber japonicus) has been the favorite and most popular fish in Korea with an average intake of 3.93 g per person per day (Korea Health Industry Development Institute, 2015;National Institute of Fisheries Science, 2010). It is consumed after various pretreatment (washing and soaking) and cooking method including sushi, grilling, braising, frying, and steaming. Also, mackerel is cooked with other ingredients such as potatoes, zucchini, and carrots according to preference (Kim, Baek, Kim, & Moon, 1998;Korean Food Promotion Institute, 2017). Potatoes contain high dietary fiber, which can be used as an absorbent to react with heavy metals (Kim, Shin, & Hwang, 1989;Lee & Lee, 1993). In Korea, rice-washed solution and sake are commonly used as pretreatments for removing the unpleasant odor of fish. Rice-washed solution has the advantage of being easily available because steamed rice is the main dish of Korea. Hong, Son, Kang, and Noh (2009) have reported that rice-washed solution is a major source of domestic sewage. Therefore, the use of rice-washed solution is an economical approach to remove the unpleasant odor of fish.
The results obtained from the studies carried out till date on the effect of cooking methods and pretreatment on the PFC content of fish are controversial. Del Gobbo et al. (2008) and Luo et al. (2019) have reported that the PFC concentration in fish, swimming crab, or fish cake decreased by cooking but the most effective cooking method to reduce its PFC content was different depending on the fish type. Jogsten et al. (2009) also reported that the PFC content in fish varied with their type and the method of cooking. On the other hand, Vassiliadou et al. (2015) reported that the PFC content in finfish and shellfish increased when they are fried or grilled. Therefore, further investigation is necessary to gain a clear understanding of the effect of cooking as well as washing and soaking methods on the PFC content of seafoods.
In this study, we aimed to investigate the effect of washing, soaking, and cooking methods on the PFC content in mackerel. For this purpose, different washing and soaking conditions as well as four cooking methods (grilling, braising, steaming, and frying) were investigated and the PFC concentrations before and after the washing, soaking, and cooking were analyzed using LC-MS/MS.

| Sample preparation and cooking
Mackerel were filleted and divided into three equal portions (upper, middle, and tail part) by mass. Each portion was analyzed to determine the PFC concentration in different parts of mackerel.
Composite samples with three mackerel each were used to avoid bias due to interindividual variability.
In order to study the effect of washing and soaking on the PFC concentration in mackerel, ten mackerel were filleted and mixed into a composite sample, which was then divided into six equal portions by mass. One of these portions was left raw. The rice was mixed with tap water in a ratio of 1:2; then, rice grains were rubbed for 1 min in tap water. The milky white suspension was collected, and an equal amount of tap water was poured in and the rice-washing was repeated two times more. The all rice-washed solution was collected and mixed well for further using. Three portions were washed different times, and the remaining two portions were soaked for 15 min in rice-washed solution (100 ml) and sake (200 ml), respectively. For the washing treatment, the fillets were washed (one, two, and three times) under running water for 20 s (at 75 ml/s).
To see the effect of cooking, the composite samples of 10 mackerel were divided into seven equal portions by mass (100 g each).
One of these portions was left raw, two portions were grilled, two portions were braised, and the remaining two portions were fried and steamed, respectively. The recipes were devised from the preparatory experiments and suggestions from Korean rural development administration (Table 1). For cooking, stainless steel pan (SL, 26 cm diameter, 5 cm depth) was used for grilling, braising, and frying and stainless steel steamer pot (SLP, 26 cm diameter) was used for steaming.
To observe the transfer of PFCs from mackerel to ingredients during cooking, the composite samples were divided into three equal portions by mass. One of these portions was left raw, second portion was braised with only seasoning, and the remaining portion was braised with seasoning and potato. Potato which was selected as supplementary ingredient was sliced of 0.5 cm in thickness and braised in an amount of 100 g with mackerel fillet. Sampling procedure for pretreatment and cooking is shown in Figure 1. After cooking and pretreatments, all the samples were stored at −20°C for the analysis of PFCs.

| Analytical method
The samples were extracted and analyzed for PFCs using the enzyme + hexane ion-paring method reported by Bang et al. (2018).
The samples were mixed with the same amount of distilled water, and the resulting mixtures were homogenized using a food processor (Tefal, BL1401 KR, Rumilly, France). After the addition of an internal standard solution mixture (20 μl), the homogenized samples (1 g) were mixed with protease (350 μl) and lipase (350 μl) in a poly-

| Instrumental analysis
An LC-MS/MS system consisting of an Agilent 1100 LC series (Agilent Technologies) equipped with an Imtackt CD-C C18 column (2.0 mm × 150 mm, 3.0 µm particle diameter; Imtakt, Kyoto, Japan), API 4000 spectrometer (Applied Biosystems), and electrospray ionization operating in negative mode was used. Analysis

| Exposure assessment of PFCs
Dietary intake data were obtained from the Korea National Health

| Statistical analysis
The results were expressed as mean ± standard deviations (SDs).
One-way analysis of variance (ANOVA) is used each experimental set, and significance was defined at p < .05 by Duncan's multiple range test using SAS version 8.0 for Windows (SAS Inst).  Figure 2. The results showed that the PFC levels in different parts of the mackerel samples were almost the same (Figure 3). Del Gobbo et al. (2008) reported that in fish,

| PFC accumulation in different parts of mackerel
PFCs tend to accumulate more in liver than in muscle tissue. In addition, the PFC content in the liver and gut of fish is higher than that in its muscle tissue (Heo et al., 2013). Therefore, it is believed that the dietary exposure to PFCs can be reduced by avoiding the consumption of fish liver and gut. Furthermore, the level of PFC accumulation in different parts of the muscle tissue of fish is almost the same.

| Effect of washing and soaking on PFC concentrations
The effect of washing and soaking on the PFC concentration of raw mackerel was investigated, and the results are given in Figure 4.
Washing is widely used to reduce the concentration of heavy metals or contaminants in food. Lee, Choi, and Park (2003) reported that around 20%-38% of heavy metals in oriental medical materials can be removed by simply washing them with water. In addition, Satpathy, Tyagi, and Gupta (2011) have reported that pesticide residues in vegetables can be reduced by washing (in various solutions). In this study, an average reduction of 74% was observed in the PFC concentration in mackerel after washing. A reduction of 74% was observed when the washing was carried out once (W1). When the washing was carried out twice, a reduction of 79% was observed (W2) and a reduction of 67% was observed when the washing was carried out three times (W3). Because of their high solubility in water (Prevedouros et al., 2006), the concentration of PFCs decreased after washing (Figure 4). This concentration reduction was not significantly affected by the washing frequency. In raw material, PFOA which has been reported to cause liver damage was not detected in any of the washed samples. On the other hand, no significant reduction was observed in the concentration of PFBA and PFTrDA after washing.
In Korea, the traditional method to remove the unpleasant odor of fish is to soak it in sake or rice-washed solution (Hong et al., 2009;Woo, Choi, & Jeong, 2006

| Effect of cooking
In this study, the effect of cooking methods on the PFC concentration in fish was investigated. The PFC concentrations of mackerel after cooking are shown in Table 2. The total PFC content in raw mackerel (control, Norwegian) was 0.57 ± 0.14 ng/g, which is lower than that reported by Heo et al. (2013) for Korean mackerel (2.04 ± 0.57 ng/g). Seawater is exposed to PFCs either by direct contamination or by degradation of precursors in water (Herzke, Nygård, Berger, Huber, & Røv, 2009), and hence is considered a source of PFCs. Therefore, the PFC content in fish is affected by marine environment. Haug et al. (2010) (Hwang, 2013), or pesticides (Jegal, Han, & Kim, 2000) from food. Wu et al. (2018) and Luo et al. (2019) reported that a cooking in high-temperature heating with solvent (water or edible oil) would be effective at reducing PFCs with high polarity in food and this was in accordance with our results.
Potatoes are commonly used as supplementary ingredients to make braised mackerel (Kim et al., 1998). When braised with seasoning only, PFC content of mackerel fillet decreased by 41% comparing to raw fillet (Table 3). On the other hand, when it was cooked with potatoes, total PFC content of mackerel decreased by 79%. The addition of potatoes increased the reduction of PFCs compared to cooking with seasonings only. Also, in cooked potatoes, the contents of PFCs increased by 122%, comparing to uncooked potatoes (Table 3). Researches are being carried out to reduce heavy metals exposed to the natural environment by using agricultural products such as chestnut shell (Shin, Cha, Seo, & Kim, 1999), food-fruit and oriental herb's (Kim, Oh, & Baek, 1999), and allium roots (Kim et al., 1998). Fiber, vitamins, proteins, phytin, and phenolic compounds are known as plant components that can react with heavy metals (Kim et al., 1998). Potatoes contain high dietary fiber, pectin, and phenolic compounds (Kim et al., 1989;Lee & Lee, 1993). Luo et al. (2019) reported similar result that PFCs in swimming crab decreased more when it was cooked with Korean radish which contains high fiber. It was considered that PFCs decreased when mackerel was cooked with potatoes because they were also adsorbed by dietary fibers in potatoes. Therefore, the results obtained in this study suggested that cooking was considered as an effective method to reduce the dietary intake of PFCs from mackerel.

| Dietary PFOA and PFOS exposure assessment
The mean daily intake of mackerel was 0.9 g/day for whole Korean adults aged above 19 years old and 42.5 g/day for consumer only.
The 99th percentile consumption of mackerel was 36.8 g/day for whole Korean adults aged above 19 years old and 185.0 g/day for consumer only. The mean exposure to PFOs using mean intake of

F I G U R E 4
Efficacy of pretreatment on reducing the total PFC concentrations (ng/g) in mackerel. a,b :Different letters indicate significant differences at p < .05 by Duncan's multiple range test. W1; washing once. W2; washing for two times. W3; washing for three times. S15; soaking in sake for 15 min. R15; soaking in ricewashed solution for 15 min. PFBA, perfluorobutanoic acid; PFOA, perfluorooctanoic acid; PFHpA, perfluorheptanoic acid; PFTrDA, perfluorotridecanoic acid; PFBS, perfluorobutane sulfonate; and PFOSA, perfluorooctane sulfonamide for PFOA and 0.03 μg/kg for PFOS was 0.0197 × 10 -3 μg/kg bw/day and 0.0197 ng/kg bw/day, respectively. The excessive exposure to PFOs using large intake of 185.0 g/day (P99 for consumer only) with maximum concentration of 0.07 μg/kg for PFOA and 0.28 μg/kg for PFOS was 0.1997 ng/kg bw/day and 0.7987 ng/kg bw/day, respectively (Table 4). It was lower than the recommended TDI (1,000 ng/ kg b.w. for PFOA and 150 ng/kg bw for PFOS) proposed by the EFSA.
Chub mackerel is the most popular fish in Korea and Japan, and mackerel is consumed more than 800 million tons in Europe (Bae & Lim, 2012;FAO, 2018). Also, Indian mackerel (Rastreliiger Kanagurta) is the most important fishing resources in South-East Asian countries (Wu, Pu, & Sun, 2019). Therefore, it is important to estimate the dietary PFOA and PFOS exposure by mackerel.
Additionally, we evaluated the excessive exposure to PFOs using data of mackerel consumption P99 and maximum concentration across different age groups. The exposure to PFOA across different age groups was estimate to be a minimum of 0.1243 ng/kg bw/ day aged 30-39 years and a maximum of 0.2532 ng/kg bw/day aged 50-59 years. The exposure to PFOS was calculated to be 0.4970-1.0129 ng/kg bw/day. It was shown no significant differences among age groups (Table 5). Consequently, exposure assessment to PFOs through the mackerel consumption according to each age group estimated to be an intake of PFOs lower than the TDI.

| CON CLUS IONS
In this study, the effect of pretreatment (washing and soaking) and cooking (grilling, braising, frying, and steaming) methods on the PFC content of mackerel was investigated. The PFC contents of the raw, pretreated, and cooked mackerel samples were determined. An average reduction of 74% was observed in the PFC content of mackerel when it was washed with water (one, two, and three times).
Soaking in sake and rice-washed solution reduced the PFC content of mackerel by 51% and 80% comparing to control, respectively. All the four cooking methods were effective in reducing the PFC content of mackerel. The grilled, steamed, fried, and braised samples showed a PFC content reduction of 91%, 75%, 58%, and 47% comparing to uncooked sample, respectively. In addition, when mackerel was braised with potato, PFCs decreased more in mackerel fillet TA B L E 2 Effect of cooking methods on the concentration of PFCs (ng/g) in mackerel

| INFORMED CON S ENT
None.

This research was supported by a grant (15162KFDA077) from
Korea Food and Drug Administration (2016).

CO N FLI C T O F I NTE R E S T
All authors declare that they do not have any conflict or interest.

E TH I C A L S TATEM ENT
This study does not involve any human or animal testing.