Safety and quality of the green tide algal species Ulva prolifera for option of human consumption: A nutrition and contamination study

Chemosphere 210 (2018) 1021e1028

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Safety and quality of the green tide algal species Ulva prolifera for option of human consumption: A nutrition and contamination study Juan-Ying Li a, Fengyuan Yang a, Ling Jin b, Qian Wang a, Jie Yin a, Peimin He a, Yiqin Chen a, * a b

College of Marine Ecology and Environment, Shanghai Ocean University, Shanghai, 201306, China Department of Civil and Environmental Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


U. prolifera as a low-fat, high-protein food resource.
Acceptable health risks due to heavy metals, pesticides and PAHs in U. prolifera.
Different PAH composition profiles in the floating and grown-on-raft U. prolifera.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 23 April 2018 Received in revised form 13 July 2018 Accepted 14 July 2018 Available online 20 July 2018

This study sampled U. prolifera and surface seawater from the same locations where green tide broke out in the southern Yellow Sea, in both the year 2016 and 2017. The revealed nutritive components of U. prolifera samples characterized U. prolifera as a high-protein, high-Fe, high ratio of unsaturated lipid acids and low-fat seaweed food, with an ideal ratio of essential and nonessential amino acids. The concentrations and health risk assessment of major micropollutants (heavy metals, pesticides and polycyclic aromatic hydrocarbon (PAHs)) in U. prolifera were also analyzed, respectively. The results showed that the Target Hazard Quotient values of five heavy metals (<1.0  101) and the total hazard index of 13 pesticides (<1.5  108) were lower than the unity, respectively, and the incremental lifetime cancer risk values of PAHs (<7.4  107) were lower than the USEPA limit (1.0  106). It suggested that consuming U. prolifera is safe as a food-source option, with PAHs causing relatively higher risks. PAHs from the sites closer to the shore were also found more originated from pyrolysis. We further confirmed the PAH congeners were partly in equilibrium between seawater and U. prolifera. It suggested the possibility that the food safety-risk turned to be above the USEPA limit was not high regardless of the sample collecting time. However, the sources of PAHs and their contributions to the accumulation in U. prolifera need further investigation. This study favored that U. prolifera of the green tide from the southern Yellow Sea has a potential for a nutritious-food production. © 2018 Elsevier Ltd. All rights reserved.

Handling Editor: Jim Lazorchak Keywords: Green tides U. Prolifera Food-resource option Heavy metals Pesticides PAHs

1. Introduction

* Corresponding author. E-mail address: [email protected] (Y. Chen). https://doi.org/10.1016/j.chemosphere.2018.07.076 0045-6535/© 2018 Elsevier Ltd. All rights reserved.

Green tides usually refer to the abnormal proliferation of floating macroalgae in the phylum of Chlorophyta (Alstyne et al.,

1022

J.-Y. Li et al. / Chemosphere 210 (2018) 1021e1028

2015; Wang et al., 2015). There have been reports of occurrence of green tides at the Cadiz Bay and Palmones estuaries in Spain (Palomo et al., 2004), Tokyo Bay in Japan (Yabe et al., 2009), the Brittany coast in France (Perrot et al., 2007), the northeast Pacific coast (Alstyne et al., 2006), the South Korean coast (Kim et al., 2004), and the southern coast of England (Taylor et al., 2001). As the largest green tide events in the world to date, green tides in the Yellow Sea happened in ten consecutive years from 2007 (Hu et al., 2014). The maximum distribution area of the floating macroalgae mats of each event ranged from 1.0  104 to 6 .0 104 km2, which was approximately one-tenth of the total area of the Yellow Sea (Wang et al., 2015). The large-scale green tides can cause negative impacts on the local ecosystem, and are destructive to the coastal landscapes and mariculture, incurring significant economic losses (Wang et al., 2018). Morphological and molecular analyses have revealed Ulva prolifera as the dominant species of green tides in the Yellow Sea (Liu et al., 2013). The precaution of the green tides is requiring more investment. Meanwhile, removing the massive floating mats from the Yellow Sea by human and machines has been at considerable expense to the government of China every year. Effective utilization of the massive amount of the collected U. prolifera has therefore become a focus of research. U. prolifera can be processed into bio-oil and bio-char products, as a viable eco-friendly, green feedstock substitute for biofuels and chemicals production (Zhuang et al., 2012). Nowadays, the yield of U. prolifera is approximate 20% of wet weight for bio-oil and 40% for bio-char (Wang et al., 2018; Zhou et al., 2010; Chow et al., 2013). The utilization rate of U. prolifera can be enhanced through developing high value products such as food, feed ingredients (Michalak and Chojnacka, 2009). Being a low-calorie food rich in vitamins, minerals and dietary fiber, U. prolifera has been used as raw material or additives for fertilizers and animal feeds for decades (Liu et al., 2013). Large-scale production of seaweed feed has been achieved (Sun et al., 2011). U. prolifera has also received a lot of attention as functional food ingredients, being used widely for centuries in Asia, mostly as food to provide nutrition and a characteristic taste (Lordan et al., 2011; Collins et al., 2016). Fresh dried seaweed is extensively consumed, especially by people living in coastal areas in China. Moreover, polysaccharide, as the major component of U. prolifera, was found to be of high hypolipidemic activities, potent immunomodulatory properties and antibacterial activities (Zhao et al., 2016). Processing U. prolifera into functional ingredients of food for human consumption and the related food safety have consequently gained increasing attention (Zhao et al., 2016; Michalak et al., 2017). To the best of our knowledge, the nutritional composition of U. prolifera has not been systematically analyzed and reported, however. The cell wall of U. prolifera contains functional groups such as hydroxyl, carboxyl, amino, thiol and phosphate, and thus the cell wall is negatively charged and can attract heavy metal ions in water through ion exchange (Flores et al., 2015). The equilibrium between U. prolifera and water can be achieved very soon when the concentration of heavy metals in the seawater is not high (Sheng et al., 2004). Besides, U. prolifera consists of filamentous shoots with monostromatic hollow cylinders and have high ratio of surface to volume (S/V), which enables them to absorb particles-bounded polycyclic aromatic hydrocarbons (PAHs) and pesticides abundantly when exposed to contaminated aquatic environment (Zhang et al., 2017). Several kinds of seaweed were reported to be contaminated by toxic compounds such as heavy metals, pesticides and PAHs (Besada et al., 2009; Kamala-Kannan et al., 2008; Rubio et al., 2017). Marine macroalage were even selected as novel materials for the removal and monitoring of heavy metals from water (Khan et al., 2015; Khaled et al., 2014; Huan et al., 2018). In comparison, the reports regarding the uptake rate of PAHs and

pesticides by U. prolifera were not found (García-Rodríguez et al., 2012; Pavoni et al., 2003). The key aims of this study were: (1) to assess the nutritive components and the concentrations of the PAHs, pesticides and heavy metals in U. prolifera collected from the southern Yellow Sea during the outbreak of green tides in 2016 and 2017; (2) to evaluate the health risk of U. prolifera by human consumption based on contamination levels; and (3) to investigate the possible sources of the contaminants that would result in high food safety risk and their bioconcentration potentials into U. prolifera. This study was in hope of providing supporting data for the advanced utilization of the green tide algae from the southern Yellow Sea in China. 2. Materials and methods 2.1. Sample collection According to the distribution pattern and growth status of U. prolifera along and off the coast of Jiangsu Province, site-matched samples of U. prolifera and surface water were collected between May and August in 2016 and 2017, respectively, from three Porphyra culturing rafts at the coast of Rudong City (RD1, RD2, and RD3), one at the coast of Lvsi City (LS) and one at the coast of Dafeng City (DF). In 2017, U. prolifera and water samples were collected from three sampling sites following the direction of algae floating in the southern Yellow Sea (SE1, SE2, and SE3) (Fig. 1). In total, 65 U. prolifera samples were collected in this study, with five replicate samples collected from each sampling site in each year. The collected U. prolifera samples (2 kg for each site) were stored in a cool box with ice packs and transferred to a 20  C freezer in the lab until analysis. Water samples (2 L for each site) were collected into pre-cleaned glass bottles and transferred to a 4  C fridge in the lab until analysis. U. prolifera samples were freeze-dried within 48 h and all sample analysis were completed within 7 days. 2.2. Sample analysis Water content was determined via oven-drying at 105  C; crude ash content was determined via heating at 550  C; crude fat content was determined using the Soxhlet extraction method; chlorophyll content was determined using the N, N0 -dimethylformamide extraction method (Kumar et al., 2011). The crude protein, phosphate, iron, copper, manganese, zinc, carbohydrate, fatty acid and amino acid content were determined according to the methods published previously (Ortiz et al., 2006; Kumar et al., 2011). PAHs, heavy metals, organochlorine pesticides (OCPs), pyrethroid pesticides (PEs) and organophosphorus pesticides (OPPs) in the samples (seawater and U. prolifera) were measured following methodologies published previously (Li et al., 2015a, 2015b; Khan et al., 2015). Details of the instrumental method were provided in Supplementary Information. The target compound list was presented in Table S1. 2.3. Quality control Quality control procedures for sample analysis included duplicate analysis, method blanks and spiked samples. For duplicate analysis of each sample, the relative standard deviation (RSD) of the analytical results was less than 15%. For every ten samples analyzed, a blank sample was included to check laboratory contamination. Pre-extracted U. prolifera were used as the blank samples. For recovery analysis, the pre-extracted U. prolifera samples were spiked with heavy metals, pesticides and PAHs of 0.1, 1.0, 10 mg/g dw, 0.5, 5.0, 50 ng/g dw and 0.1, 1.0, 10, 100 ng/g dw,

J.-Y. Li et al. / Chemosphere 210 (2018) 1021e1028

1023

Fig. 1. Map of sampling locations (eight sampling sites: RD1, RD2, RD3, LS, DF, SE1, SE2 and SE3).

respectively. Additional details of blank values, limit of quantification (LOQ) and average recoveries of each compound are provided in Table S2. The LOQ of each compound was defined as the mean value of its concentration in blank samples plus ten times its standard deviation, or the lowest level of the calibration curve, whichever is higher. The average recoveries of the labelled standards ranged from 66% to 110% for all chemicals. Only results that were above the LOQ were used in the data analysis. For summarizing data, results that were below LOQ were imputed as LOQ/2. 2.4. Risk assessment of human consumption of U. prolifera 2.4.1. Risk assessment caused by heavy metals Target Hazard Quotient (THQ) was used for health risk assessment of heavy metals in U. prolifera. THQ represents the target risk factor, which is the ratio of the measured human ingested dose to the reference dose, and it is applicable to single pollutant evaluation (Basim and Khoshnood, 2013). It is assumed that the human ingested dose is equal to the absorbed dose. Higher THQ values indicate higher risks. THQ values of heavy metals were calculated followed by equation (1) (Basim and Khoshnood, 2013):

THQ ¼

EF$ED$IR$C  103 RfD$BW$AT

EDI ¼ Cbio  Ccon =Bw

(3)

Where Cbio represents the pesticide residue in U. prolifera (ng$g1(dw)); Ccon represents the average daily consumption of U. prolifera and was set as 10 g week1, i.e. 1.43 g d1 (Zhan et al., 2016). Bw represents the body weight, valued 70 kg. The acceptable daily intake (ADI) values for heavy metals and pesticides in edible algae based on national standards of China (GB2762-2017 and GB2763-2014) were presented in Table S3. Hazard Index (HI) is for health risk assessment of combined pesticides.

HI ¼

i X

THQn

(4)

n¼1

When HI is greater than unity there is high likelihood of adverse health implications. The greater the HI value, the greater the corresponding risk (Cui et al., 2015).

(1) 1

Where EF is the exposure frequency, valued at 365 days∙year ; ED is the exposure duration, valued at 30 years; IR is the intake rate of U. prolifera, valued at 1.4 g∙day1; RfD is the oral reference dose valued at 4  102, 4  103, 1  103, 1.5, 3  104 for Cu, Pb, Cd, Cr (WHO/FAO, 2013) and As (Ma et al., 2010), respectively; BW is the body weight, valued at 70 kg; AT is the average age, valued at 70 years (Zhan et al., 2016). 2.4.2. Risk assessment caused by pesticides THQ was also used for health risk assessment of pesticides in U. prolifera.

THQ ¼ EDI=ADI

EDI (Estimated Daily Intake) represents the daily intake of pesticide [ng$(kg$d)1(dw)],

(2)

2.4.3. Cancer risk caused by consumption of PAH-contaminated U. prolifera This study assessed the cancer risk via consuming PAHcontaminated U. prolifera following the toxic equivalent scheme suggested by US EPA (1993). The concentration (Ci) of each cancercausing PAH was converted into BaP toxicity concentration using the toxic equivalency factor (TEF) (Table S4) relevant to BaP and the total BaPeq values (ng∙g1 dw) in samples were calculated.

BaPeq ¼

n X ½Ci  TEFi 

(5)

i

The incremental lifetime cancer risk (ILCR) for the U. prolifera consumption in the studied areas was calculated according to eq. (6).

1024

ILCR ¼

J.-Y. Li et al. / Chemosphere 210 (2018) 1021e1028

GraphPad Prism 6 (GraphPad Software, Inc., USA).

BaPeq $CSFingestion $IR$EF$ED BW$AT

(6) 3. Results and discussion

CSFingestion is the slope factor for cancer risk which has a value of 7.3 [mg∙(kg
day)1]1, calculated based on the cancer inducing potency of BaP. BW is the body weight, valued at 70 kg. AT is the average age, valued at 70 years. EF is the exposure frequency, valued at 365 days∙year1. ED is the exposure duration, valued at 30 years. IR is the intake rate of U. prolifera, valued at 1.43 g∙day1 (Li et al., 2014). 2.5. Bioconcentration of PAHs into U. prolifera from seawater In order to assess the bioconcentration of PAHs into U. prolifera from seawater, the concentrations of each PAH congener in U. prolifera at equilibrium (Cbio-equilibrium), was compared with the measured concentrations in U. prolifera (Cbio-measured), providing the information on the deviation from the equilibrium for the chemicals. The values of Cbio-equilibrium of PAHs could be calculated from dissolved-organic-carbon-concentration(CDOC)-based Cfree-water multiplied by Klipid-water as shown in the following equations (Cui et al., 2013).

Cbioequilibrium ¼ Cfreewater  Klipidwater Cfreewater ¼ Cwater =ð1 þ KDOC  CDOC



(7) (8)

Where Cwater and CDOC were obtained from the experiments in this study. The KDOC and Klipid-water values were cited from the paper published previously (Li et al., 2016). 2.6. Statistics t-test and ANOVA analyses were used to evaluate the inter-year and inter-site difference and inter-species difference, using

3.1. Nutritional composition of U. prolifera The nutritional composition of U. prolifera of the green tides from the southern Yellow Sea of China was analyzed (Table S5, S6). The two-year-averaged results were quite consistent cross all the eight sampling sites with relative standard deviation (RSD) values of less than 15%, except for Ca (RSD ¼ 61%) and Fe (RSD ¼ 46%). The two major components of the water-free U. prolifera samples were carbohydrates and proteins, which were approximately 50% and 30% of the total weight, respectively; the lipid content was only approximately 0.5%. In comparison, three other edible seaweeds, Caulerpaceae (C). veravelensis, C. scalpelliformis and C. racemosa reported elsewhere (Kumar et al., 2011) showed significantly higher lipid content (2.80 ± 0.19% dw; t-test, P < 0.001), lower protein content (10.50 ± 0.91% dw; t-test, P < 0.001) and lower Fe content (0.20 ± 0.07%; t-test, P ¼ 0.011). While all four seaweeds were high in water and carbohydrate content, and low in Zn and Mn content, U. prolifera was slightly lower in Ash content and higher in total chlorophyll content (Fig. 2a and b). The results characterized U. prolifera as a high-protein, high-Fe and low-fat seaweed food. A variety of lipid acids were found in U. prolifera and most of them were unsaturated lipid acids, accounting for 55% of the total (Table S7). More specifically, polyunsaturated lipid acids account for 43%. C16:0 (palmitic acid) has the highest level among saturated lipid acids, accounting for 24% of the total saturated. C18:1T, trans-9 (elaidic acid) has the highest level among polyunsaturated lipid acids, accounting for 38% of the total polyunsaturated. As compared in Fig. 2c, U. prolifera had a higher ratio of unsaturated lipid acids which are essential lipid acids benefiting human health (Kumar et al., 2011). A total of 17 amino acids were found (Table S8), including 8

Fig. 2. Comparison of the nutritional compositions of different edible seaweeds (the columns with different letters (a, b, c, d) have levels of a significant difference (t-test, P < 0.05)) (for the figure C, the t-test was based on the percentage of unsaturated lipid acids).

J.-Y. Li et al. / Chemosphere 210 (2018) 1021e1028

essential amino acids and 9 nonessential amino acids, which gave a ratio of 28.93%～35.92% between essential amino acids and total amino acids (EAA/TAA), and 64.08%～71.07% between essential and nonessential amino acids (EAA/NEAA). The values were within the ideal ratios (EAA/TAA40%,EAA/NEAA60%) recommended by the Food and Agriculture Organization/World Health Organization (FAO/WHO, 1991). As compared in Fig. 2d, U. prolifera had both a higher content of essential amino acids and nonessential amino acids than the 73 seaweed samples (including nine red algae, six ~a green algae and three brown algae) investigated by Astorga-Espan et al. (2016). Those seaweeds included Adenocystis, Lessonia, Macrocystis, Ceramium, Heterosiphonia, Iridaea, Gigartina, Mazzaella, Nothogenia, Polysiphonia, Porphyra, Sarcothalia, Ballia, Cladophora, Codium, Enteromorpha, Monostroma and Ulva. The findings were supporting U. prolifera could be a good protein resource. Further evaluation of U. prolifera in terms of polysaccharides is needed as a high-value food ingredient. 3.2. The concentrations of heavy metals in U. prolifera and health risks This study tested five heavy metals including copper, lead, cadmium, chromium and arsenic in U. prolifera samples collected from five sampling sites between May and August in 2016, and from three more sampling sites during the same time period in 2017. Considering all the sampling sites, the two-year-averaged level for Cu, Pb, Cd, Cr and As was 1.7e10, 0.064e2.3, 0.0068e1.6, 1.6e9.7 and 0.66e0.93 mg/g dw, respectively, with negligible inter-year differences (t-test, P ¼ 0.22, 0.08, 0.56, 0.49 and 0.23, respectively) (Table S9). Compared with the heavy metal content in seaweeds of other areas, the content of lead, cadmium and arsenic in U. prolifera in this study were at much lower levels, while the level of copper and chromium content was similar (Kumar et al., 2011; Flores et al., 2015; Giusti, 2001; Caliceti et al., 2002) (Table S10). The food safety-risk of the U. prolifera were calculated using the THQ values, using the two-year-averaged value for each heavy metal and each sampling site. The results showed that the THQ values of all five heavy metals were much lower than unity (<101). Arsenic had higher THQ values than the other four target heavy metals for all eight sampling sites, while Cr had lower THQ values.

1025

The finding suggested that consuming U. prolifera is quite safe with respect to its heavy metal contents (Fig. 3a). 3.3. The concentrations of pesticides in U. prolifera and health risk The concentrations of three OCPs, four PEs and six OPPs in U. prolifera from eight sampling sites of two years were manifested P P OCPs, PEs and in Table S11. The two-year-averaged levels of P OPPs were 26e45, 32e154, and 13e21 ng/g dw, respectively, with negligible inter-year differences (t-test, P ¼ 0.34, 0.19 and 0.73, respectively). The ƩOCPs result was within the range of the concentrations in other seaweeds reported in China (Qiu et al., 2017; Peng et al., 2014); the ƩPEs result was approximately one order of magnitude lower than the content in the seaweeds from Iskenderun Bay, Turkey (Polat et al., 2017) (Table S10). The food safety-risk on the investigated 3 OCPs, 4 PEs and 6 OPPs in U. prolifera was evaluated using the total hazard index (HI), showing the values were far lower than unity. The HI values of OPPs were the highest among three pesticide groups for all eight sampling sites (Fig. 3b). The low variety, concentration and risk value of pesticides found in this study suggested that U. prolifera from the shallows of north Jiangsu, southern Yellow Sea was safe for consumption with respect to its pesticide content. 3.4. The concentrations of PAHs in U. prolifera and health risk The concentrations of 16 PAHs in U. prolifera from five sampling sites in 2016 and three more sampling sites in 2017 were manP ifested in Table S12. The two-year -averaged PAHs levels ranged 1 from 134 to 984 ng g dw, with negligible inter-year differences (ttest, P ¼ 0.064). Among all 16 PAH congeners, FLA, PYR, CHR, BaA, BbF and BkF were found in all sampling sites, while FLU, IPY and BghiP were only detected in some of the sites. Compared with the seven seaweeds from the lagoon of Venice (Figure S1), U. prolifera contained a much higher level of PAHs, possibly due to the differences in influencing factors of environment quality between the two areas (Pavoni et al., 2003). High potential carcinogenic PAHs (BaA, BbF, BkF, BaP, DBA and P IPY) accounted for 20e92% of the PAHs in U. Prolifera, ranging 1 from 0.19 to 414 ng g dw (Table S12). The inter-year variation of

Fig. 3. U. Prolifera (a) Estimated THQ values of targeted heavy metals; (b) Estimated HI values of targeted pesticides; (c) Estimated ILCR values of target PAHs; and (d) Composition profiles of PAHs. All data used were the mean values of two years for RD1, RD2, RD3, DF, LS.

1026

J.-Y. Li et al. / Chemosphere 210 (2018) 1021e1028

these six congeners (BaA, BbF, BkF, BaP, DBA and IPY) were negligible (t-test, P ¼ 0.45), and therefore the two-year mean values were used for health risk calculation. The ILCR values of the 16 USEPA priority PAHs were calculated based on formula (6), and the calculated results were compared with 1  106, the acceptable limit stipulated by USEPA (Fig. 3c). The ILCR values were all lower than the USEPA limit, suggesting that the consumption of U. Prolifera would not result in unacceptable food safety risk. However, U. Prolifera from the sites closer to shore with more human activities should be consumed with more cautions for their ILCR values were approximately 10 times as high as the values of the sampling sites in the sea (Fig. 3c). Statistically, neither the variation in the PAHs residues and compositions in U. prolifera within the three RD sampling sites (RD1, RD2 and R3) was significant (ANOVA, P ¼ 0.17), nor was the variation within the three off-shore sites (SE1, SE2 and SE3) significant (ANOVA, P ¼ 0.98). However, the variations turned significant when considering all the five nearshore sites and all the eight sampling sites, with P ¼ 0.0077, P < 0.001 (ANOVA), respectively. Analysis on the composition profile of PAHs in U. Prolifera from eight sampling sites (Fig. 3d) found that PAHs in U. Prolifera from RD, DF and LS sites were dominated by four-ring congeners, which accounted for 83e97% of the total PAHs, and two-ring species came secondly, accounting for only 2.2e8.2% of the total. However, tworing congeners were dominant for sites SE1, SE2 and SE3, accounting for 74e77% and followed by four-ring species, which accounted for 23e27% of the total. Six-ring species were only found at site RD3. PAHs generated from petrogenic sources were believed to be PAH isomers primarily with less aromatic rings (two to three rings) and those from pyrogenic sources were mostly with medium to high ring numbers (four to six rings) (Lang et al., 2015; Li et al., 2015a; Thomas et al., 2014). This suggests that PAHs in U. Prolifera from RD, DF and LS were more likely from pyrogenic sources than those in U. Prolifera from SE1, SE2 and SE3. Pyrogenic and petrogenic sources were further confirmed by the diagnostic ratios of LMW/HMW as shown in Table S13. The offshore sampling sites (SE1, SE2 and SE3) are on the busy navy channels in the southern Yellow Sea, where the petroleum leakage is likely to occur frequently. In comparison the nearshore sampling sites (RD, DF and LS) were more likely to be influenced by contamination sources from factories inland.

Fig. 4. Equilibrium between the PAHs in the seawater and the U. Prolifera (data presented as mean ± standard deviation).

equilibrium concentrations and one five-ring PAH congener showing lower accumulation. This conservative estimate based on the thermodynamic potential for bioaccumulation did not take metabolism into account. However, the general agreement between predicated and measured concentrations suggested the possibility that the food safety-risk turned to be above the USEPA limit was not high regardless of the sample collecting time. As mentioned above, the two-ring PAH congeners were the major component of the PAHs in floating U. prolifera and the PAHs of three- and four-ringed PAHs (PHE, ANT, FLA, PYR, CHR, BaA, BbF, BkF) were mainly in near-shore U. prolifera. Two-ring PAHs have the highest vapor pressures among all PAH congeners, so that they tend to be associated with the vapor phase in air (Abdel-Shafy and Mansour, 2016). Studies proved that algae can accumulate (i.e. bioconcentrate) PAHs from low concentrations in the ambient medium (Neff, 1982; Eisler, 1987). Therefore, the higher accumulation of two-ring PAHs in U. prolifera could be due to other sources such as air apart from seawater, especially in the open sea, and it resulted in an underestimation in the prediction by the thermodynamic equilibrium method via Cfree-water and Klipid-water values. Green algae were also found to be able to transform highmolecular-weight PAHs such as five-ring PAHs (Kirso and Irha, 1998). It could be one reason for the lower accumulation of DBA and in U. prolifera than the predictions.

3.5. Thermodynamic status of PAHs between U. prolifera and surface seawater 4. Conclusions As discussed above, PAHs in U. prolifera posed much higher health risk than investigated heavy metals and pesticides. Therefore, it is important to confirm whether a thermodynamic equilibrium of PAHs between U. prolifera and seawater was achieved. If the equilibrium has not been achieved, the food safety-risk caused by the consumption of U. prolifera with respect to the PAH contamination will be higher than the values calculated in this study and be bound to be above the USEPA limit. In order to assess the bioconcentration of PAHs into U. prolifera from seawater, the concentrations of each PAH congener in U. prolifera at equilibrium (Cbio-equilibrium), was compared with the measured concentrations in U. prolifera (Cbio-measured), providing the information on the deviation from the equilibrium for the chemicals. Cwater and CDOC were obtained from the experiments as shown in Table S14, 15. The KDOC and Klipid-water values were cited from the paper published previously (Li et al., 2016). The ratios (Cbio-measured/Cbio-equilibrium) of PAH congeners were variable but bound to be at equilibrium between seawater and the U. prolifera (Fig. 4), with exceptions of several two-ring PAH congeners (NAP, ANY, ACE, PHE) showing higher accumulation than

For the first time, our study systematically characterized U. prolifera as a high-protein, high-Fe and low-fat alternative food resource, with an idea ratio of essential and nonessential amino acids. In addition, the concentrations and health risk assessment of heavy metals, pesticides and PAHs in U. prolifera and related health risk were analyzed and discussed, respectively. The results confirmed that the THQ values of all the five heavy metals and the total hazard index (HI) of 13 pesticides were much lower than unity (<3  105 and <1.5  108, respectively), and the ILCR values of PAHs were lower than the USEPA limit of 1  106. It suggested that consuming U. prolifera is quite safe with respect to contaminations of heavy metals, pesticides and PAHs. Our results further showed that the PAH congeners were mostly in equilibrium between seawater and U. prolifera. It ruled out that a large variation may be caused by different sampling time points within a year, but the possible source of two-ring PAHs from air should be noted and investigated in the future. The findings favored that U. prolifera of the green tide from the southern Yellow Sea has a potential for a nutritious-food production.

J.-Y. Li et al. / Chemosphere 210 (2018) 1021e1028

Acknowledgements The present research was supported by National Key Research and Development Plan (2016YFC1402105), Shanghai Science and Technology Commission Key Support Fund (18050502100), Shanghai Collaborative Innovation Center Fund (A1-2037-16-000112), Shanghai Ocean University Technology Development Fund (A2-0203-00-100223) and Shanghai Ocean University Doctoral Foundation (A2-0203-00-100352). Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.chemosphere.2018.07.076. References Abdel-Shafy, H.I., Mansour, M.S.M., 2016. A review on polycyclic aromatic hydrocarbons: source, environmental impact, effect on human health and remediation. Egypt. J. Petrol. 25, 107e123. Alstyne, K.L.V., Nelson, A.V., Vyvyan, J.R., 2006. Dopamine functions as an antiherbivore defense in the temperate green alga Ulvaria obscura. Oecologia 2006 (148), 304. Alstyne, K.L.V., Nelson, T.A., Ridgway, R.L., 2015. Environmental chemistry and chemical ecology of ‘‘Green tide’’ seaweed blooms. Integr. Comp. Biol. 55, 518e532. ~ a, M.A., Rodríguez-Galdo  n, B., Rodríguez-Rodríguez, E.M., DíazAstorga-Espan Romero, C., 2016. Amino acid content in seaweeds from the Magellan Straits (Chile). J. Food Compos. Anal. 53, 77e84. Basim, Y., Khoshnood, Z., 2013. Target hazard quotient evaluation of cadmium and lead in fish from Caspian Sea. Toxicol. Ind. Health 32, 215e220. Besada, V., Andrade, J.M., Schultze, F., 2009. Heavy metals in edible seaweeds commercialised for human consumption. J. Mar. Syst. 75, 305e313. Caliceti, M., Argese, E., Sfriso, A., 2002. Heavy metal contamination in the seaweeds of the Venice lagoon. Chemosphere 47, 443e454. Chow, M.C., Jackson, W.R., Chaffee, A.L., Marshall, M., 2013. Thermal treatment of algae for production of biofuel. Energy Fuel. 27, 1926e1950. Collins, K.G., Fitzgerald, G.F., Stanton, C., Ross, R.P., 2016. Looking beyond the terrestrial: the potential of seaweed derived bioactives to treat noncommunicable diseases. Mar. Drugs 14, 60. Cui, L., Ge, J., Zhu, Y., 2015. Concentrations, bioaccumulation, and human health risk assessment of organochlorine pesticides and heavy metals in edible fish from Wuhan, China. Environ. Sci. Pollut. Res. Int. 22, 15866e15879. Cui, X., Mayer, P., Gan, J., 2013. Methods to assess bioavailability of hydrophobic organic contaminants: principles, operations, and limitations. Environ. Pollut. 172, 223e234. Eisler, R., 1987. Polycyclic aromatic hydrocarbon hazards to fish, wildlife, and invertebrates: a synoptic review. U S Fish Wildl. Serv. Biol. Rep. 85, 11. EPA, 1993. Provisional Guidance for Quantitative Risk Assessment of Polycyclic Aromatic Hydrocarbons. Environmental Criteria and Assessment Office (EPA/ 600/R-93/089). Office of Health and Environmental Assessment. Flores, S.R.L., Dobbs, J., Dunn, M.A., 2015. Mineral nutrient content and iron bioavailability in common and Hawaiian seaweeds assessed by an in vitro, digestion/Caco-2 cell model. J. Food Compos. Anal. 43, 185e193. FAO/WHO, 1991. Protein quality evaluation in human diets. Food and Agriculture Organization, Food and Nutrition Paper, pp. 35e36, 51. Rome, Italy. García-Rodríguez, D., Cela-Torrijos, R., Lorenzo-Ferreira, R.A., Carro-Díaz, A.M., 2012. Analysis of pesticide residues in seaweeds using matrix solid-phase dispersion and gas chromatographyemass spectrometry detection. Food Chem. 135, 259e267. Giusti, L., 2001. Heavy metal contamination of brown seaweed and sediments from the UK coastline between the Wear river and the Tees river. Environ. Int. 26, 275e286. Hu, S., Yang, H., Zhang, J., 2014. Small-scale early aggregation of green tide macroalgae observed on the Subei Bank, Yellow Sea. Mar. Pollut. Bull. 81, 166e173. Huan, L., Su, H., Duan, C., Gao, S., Xie, X., Wang, G., 2018. Ulva prolifera (Chlorophyta): a suitable material to remove Cd2þ from aquatic environments. Water Environ. J. 32, 26e33. Kamala-Kannan, S., Batvari, B.P.D., Lee, K.J., 2008. Assessment of heavy metals (Cd, Cr and Pb) in water, sediment and seaweed ( Ulva lactuca ) in the Pulicat Lake, South East India. Chemosphere 71, 1233e1240. Khaled, A., Hessein, A., Abdel-Halim, A.M., 2014. Distribution of heavy metals in seaweeds collected along Marsa-Matrouh beaches, Egyptian Mediterranean Sea. Egypt. J. Aquat. Res. 40, 363e371. Khan, N., Ryu, K.Y., Choi, J.Y., 2015. Determination of toxic heavy metals and speciation of arsenic in seaweeds from South Korea. Food Chem. 169, 464e470. Kim, K.Y., Choi, T.S., Kim, J.H., 2004. Physiological ecology and seasonality of Ulva pertusa on a temperate rocky shore. Phycologia 43, 483e492. Kirso, U., Irha, N., 1998. Role of algae in fate of carcinogenic polycyclic aromatic hydrocarbons in the aquatic environment. Ecotoxicol. Environ. Saf. 41, 83e89.

1027

Kumar, M., Gupta, V., Kumari, P., 2011. Assessment of nutrient composition and antioxidant potential of Caulerpaceae seaweeds. J. Food Compos. Anal. 24, 270e278. Lang, S.C., Hursthouse, A., Mayer, P., Kotke, D., Hand, I., Schulz-Bull, D., Witt, G., 2015. Equilibrium passive sampling as a tool to study polycyclic aromatic hydrocarbons in Baltic Sea sediment pore-water systems. Mar. Pollut. Bull. 101, 296e303. Li, J., Li, Z., Cui, Y., Hu, Q., Chen, M., Zheng, Y., 2015a. Biomimetic research on bioavailability and bioaccumulation of sediment-associated pyrethroids using solid-phase microextraction. Asian J. Ecotoxicol. 10, 144e152. Li, J.Y., Cui, Y., Su, L., Chen, Y., Jin, L., 2015b. Polycyclic aromatic hydrocarbons in the largest deepwater port of East China Sea: impact of port construction and operation. Environ. Sci. Pollut. Control Ser. 22, 12355e12365. Li, J.Y., Su, L., Wei, F.H., Yang, J.H., Jing, L., Zhang, X.W., 2016. Bioavailability-based assessment of aryl hydrocarbon receptor-mediated activity in Lake Tai Basin from Eastern China. Sci. Total Environ. 544, 987e994. Li, Y., Song, W., Xiao, J., 2014. Tempo-spatial distribution and species diversity of green algae micro-propagules in the Yellow Sea during the large-scale green tide development. Harmful Algae 39, 40e47. Liu, D., Keesing, J.K., He, P., Wang, Z., Shi, Y., Wang, Y., 2013. The world's largest macroalgal bloom in the Yellow Sea, China: formation and implications. Estuar. Coast Shelf Sci. 129, 2e10. Lordan, S., Ross, R.P., Stanton, C., 2011. Marine bioactives as functional food ingredients: potential to reduce the incidence of chronic diseases. Mar. Drugs 9, 1056e1100. Ma, T.J., Lin, B.R., Jia, C.X., 2010. Reclaimed water farming fish heavy metal residues and edible risk assessment. Chinese agricultural science bulletin. Chin. Agric. Sci. Bull. 26, 332e336. Michalak, L., Chojnacka, K., 2009. Edible macroalga Ulva prolifera as microelemental feed supplement for livestock: the fundamental assumptions of the production method. World J. Microbiol. Biotechnol. 25, 997e1005. Michalak, L., Dmytryk, A., Smieszek, A., Marycz, K., 2017. Chemical characterization of Enteromorpha prolifera extract obtained by enzyme-assisted extraction and its influence on the metabolic activity of Caco-2. Int. J. Mol. Sci. 18, 479. Neff, J.M., 1982. Accumulation and release of polycyclic aromatic hydrocarbons from water, food, and sediment by marine animals. In: Richards, N.L., Jackson, B.L. (Eds.), Symposium: Carcinogenic Polynuclear Aromatic Hydrocarbons N the Marine Environment. U.S. Environ. Protection Agency Rep, pp. 282e320, 600/982-013. Ortiz, J., Romero, N., Robert, P., 2006. Dietary fiber, amino acid, fatty acid and tocopherol contents of the edible seaweeds Ulva lactuca, and Durvillaea Antarctica. Food Chem. 99, 98e104. Palomo, L., Clavero, V., Izquierdo, J.J., 2004. Influence of macrophytes on sediment phosphorus accumulation in a eutrophic estuary (Palmones River, Southern Spain). Aquat. Bot. 2004 (80), 103e113. Pavoni, B., Caliceti, M., Sperni, L., 2003. Organic micropollutants (PAHs, PCBs, pesticides) in seaweeds of the lagoon of Venice. Oceanol. Acta 26, 585e596. Perrot, T., Menesguen, A., Dumas, F., 2007. Ecological modelling of Ulva mass blooms on the Britanny coasts. Houille Blanche-Revue Internationale De L Eau 5, 49e55. Peng, Q.C., Song, J.M., Li, C., Yu, L.D., 2014. Risk assessment of ingesting polyunsaturated fatty acids and organochlorine pesticide residues in five seaweeds from Jiaozhou Bay. Oceanol. Limnol. Sinica 45, 80e87. Polat, A., Polat, S., Simsek, A., Kurt, T.T., Ozyurt, G., 2017. Pesticide residues in muscles of some marine fish species and seaweeds of Iskenderun Bay (Northeastern Mediterranean). Environ. Sci. Pollut. Res. 25, 1e9. Qiu, Y.W., Zeng, E.Y., Qiu, H.L., Yu, K.F., Cai, S.Q., 2017. Bioconcentration of polybrominated diphenyl ethers and organochlorine pesticides in algae is an important contaminant route to higher trophic levels. Sci. Total Environ. 579, 1885e1893. lez, G., 2017. Metals in edible seaweed. ChemoRubio, C., Napoleone, G., Luisgonza sphere 173, 572e579. Sheng, P., Ting, Y., Chen, J.P., Hong, L., 2004. Sorption of lead, copper, cadmium, zinc, and nickel by marine algal biomass: characterization of biosorptive capacity and investigation of mechanisms. J. Colloid Interface Sci. 275, 131e141. Sun, W., Zhang, G., Li, X., Gu, X., Mu, C., 2011. Potential utilization of Enteromorpha prolifera as biotic marine resources. Fish. Sci. 30, 588e590. Taylor, R., Fletcher, R.L., Raven, J.A., 2001. Preliminary studies on the growth of selected ‘green tide’ algae in laboratory culture: effects of irradiance, temperature, salinity and nutrients on growth rate. Bot. Mar. 44, 327e336. Thomas, C., Lampert, D., Reible, D., Janssen, E.M.L., Luthy, R.G., 2014. Remedy performance monitoring at contaminated sediment sites using profiling solid phase microextraction (SPME) polydimethylsiloxane (PDMS) fibers. Environ. Sci. Process. Impacts 16, 2049-2049. Wang, Z., Fu, M., Xiao, J., Zhang, X., Song, W., 2018. Progress on the study of the Yellow Sea green tides caused by Ulva prolifera. Haiyang College J. 40, 1e13. Wang, Z., Xiao, J., Fan, S., Li, Y., Liu, X., Liu, D., 2015. Who made the world's largest green tide in China? -an integrated study on the initiation and early development of the green tide in Yellow Sea. Limnol. Oceanogr. 60, 1105e1117. WHO/FAO, 2013. Guidelines for the Safe Use of Wastewater and food stuff, 2, p. 988, 14. Yabe, T., Ishii, Y., Amano, Y., Koga, T., Hayashi, S., Nohara, S., Tatsumoto, H., 2009. Green tide formed by free-floating Ulva spp. at Yatsu tidal flat, Japan. Limnology 10, 239e245. Zhan, Z., Wang, W., Zhang, G., 2016. Risk analysis of aluminium contamination in

1028

J.-Y. Li et al. / Chemosphere 210 (2018) 1021e1028

kelp and laver. J. Food Saf. Qual. 7, 1330e1334. Zhang, C., Lu, J., Wu, J., Luo, Y., 2017. Removal of phenanthrene from coastal waters by green tide algae Ulva prolifera. Sci. Total Environ. 609, 1322e1328. Zhao, C., Yang, C., Liu, B., 2016. Biological activities of green macroalgae Enteromorpha prolifera for potential applications. MOJ Food Process. Technol. 2, 00048.

Zhou, D., Zhang, L., Zhang, S., Fu, H., Chen, J., 2010. Hydrothermal liquefaction of macroalgae Enteromorpha prolifera to bio-oil. Energy Fuel. 24, 4054e4061. Zhuang, Y., Guo, J., Chen, L., Li, D., Liu, J., Ye, N., 2012. Microwave-assisted direct liquefaction of Ulva prolifera for bio-oil production by acid catalysis. Bioresour. Technol. 116, 133e139.