polychlorinated biphenyls (pcbs) in commonly consumed ... · from the coastal area of bangladesh:...
TRANSCRIPT
RESEARCH ARTICLE
Polychlorinated biphenyls (PCBs) in commonly consumed seafoodfrom the coastal area of Bangladesh: occurrence, distribution,and human health implications
Md. Habibullah-Al-Mamun1,2& Md. Kawser Ahmed3
& Md. Saiful Islam4& Anwar Hossain1,2
& Masahiro Tokumura5,6 &
Shigeki Masunaga6
Received: 28 March 2018 /Accepted: 5 November 2018 /Published online: 13 November 2018# Springer-Verlag GmbH Germany, part of Springer Nature 2018
AbstractDietary intake is the most important route of polychlorinated biphenyls (PCBs) exposure and seafood is the major dietarycomponent for the coastal populations. It is, therefore, an urgent need to assess the levels of PCBs in seafood. A comprehensivecongener-specific evaluation of PCBs was carried out for the first time in Bangladesh. All 209 congeners of PCBs in 48 seafoodsamples (5 finfish and 2 shellfish species) collected in winter and summer of 2015 were measured by GC-MS/MS. Regardless ofseason and site, the ∑PCBs (ng/g wet weight) in finfish and shellfish were in the range of 6.4–86.2 and 3.8–37.7, respectively.The results were comparable to or higher than those observed in other studies worldwide, particularly from Spain, China, Korea,Thailand, and Hong Kong. No significant seasonal variation was observed in the levels of ∑PCBs in the examined seafood(p > 0.05); however, interspecies differences were significant (p < 0.05). Nonetheless, spatial distribution revealed seafood col-lected from the areas with recent urbanization and industrialization (Chittagong, Cox’s Bazar, and Sundarbans) were morecontaminated with PCBs than the area unaffected by industries (Meghna Estuary). Moderately chlorinated (4–6 Cl) homologsdominated the PCB profiles. The congener profile and homolog composition revealed that the source origin of PCBs in theBangladeshi seafood was related to mixtures of technical PCBs formulations. The dietary exposure assessment revealed that thecoastal residents are sufficiently exposed to the dietary PCBs through seafood consumption which may cause severe health riskincluding dioxin-like toxic effects.
Keywords Polychlorinated biphenyls (PCBs) . Seafood . Health risk . Coastal area . Bangladesh
Responsible editor: Philippe Garrigues
Electronic supplementary material The online version of this article(https://doi.org/10.1007/s11356-018-3671-x) contains supplementarymaterial, which is available to authorized users.
* Md. [email protected]; [email protected]
1 Graduate School of Environment and Information Sciences,Yokohama National University, 79-9 Tokiwadai Hodogaya,Yokohama, Kanagawa 240-8501, Japan
2 Department of Fisheries, University of Dhaka, Dhaka 1000,Bangladesh
3 Department of Oceanography, Earth & Environmental ScienceFaculty, University of Dhaka, Dhaka 1000, Bangladesh
4 Department of Soil Science, Patuakhali Science and TechnologyUniversity, Dumki, Patuakhali 8602, Bangladesh
5 Graduate School of Nutritional and Environmental Science,University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526,Japan
6 Faculty of Environment and Information Sciences, YokohamaNational University, 79-9, Tokiwadai Hodogaya,Yokohama, Kanagawa 240-8501, Japan
Environmental Science and Pollution Research (2019) 26:1355–1369https://doi.org/10.1007/s11356-018-3671-x
Introduction
Polychlorinated biphenyls (PCBs) are a class of chlorinatedorganic compounds that have up to 10 chlorine (Cl) atomsconnected to biphenyl rings and the variation in number andposition of the Cl atoms results in 209 possible configurations,or congeners. PCBs are highly persistent due to their resis-tance to biological and chemical degradation. They are ubiq-uitous in the environment and are susceptible to large-scaledispersal through long-range transport and biologically activemechanisms such as bioaccumulation (Goerke and Weber2001; Gouin et al. 2004). The levels of PCBs are further ac-cumulated in the food chain through biomagnification (Porteand Albaigés 1993). The transfer of PCBs through the foodchain thus poses potential risks to the ecological integrity andto the human health as well (Batang et al. 2016). Twelve PCBcongeners (4 non-ortho: PCBs 77, 81, 126, 169, and 8 mono-ortho: PCBs 105, 114, 118, 123, 156, 157, 167, 189) aretermed as “dioxin-like PCBs (DL-PCBs)” due to their chem-ical structure being similar to 2,3,7,8-tetrachlorodibenzo-p-di-oxin (2,3,7,8-TCDD) and have been recognized as one of themost toxic organic chemicals (McFarland and Clarke 1989).These congeners are generally being monitored worldwidedue to their toxicity potentials to disrupt the endocrine system,neurological, and carcinogenic effects (Van den Berg et al.2006). Conversely, the remaining congeners, referred to asthe non-dioxin-like PCBs (nDL-PCB), exert weak or no effecton Ah-receptors; however, they cause neurotoxicity in thebiological systems (Langer et al. 2003; Simon et al. 2007).For these reasons, since 1970s, the use of PCBs has beenrestricted globally. PCBs are classified as persistent organicpollutants (POPs) under the 2001 Stockholm Convention, andtheir intentional production, distribution, and use are subjectedto strict restriction (UNEP 2001).
Although the global inventory and use of PCBs are exten-sively reported, the status of PCBs in Bangladesh is still un-known. PCBs have never been manufactured in Bangladesh.PCB-containing equipment such as capacitors, transformers,PCB mixtures, lubricating oils, etc. were imported toBangladesh (DoE 2007). There are several suspected localemission sources of PCBs in Bangladesh, for example,PCB-containing equipment, PCB stockpiles, landfill sites,ship breaking industries, etc. (ESDO 2005a, b, 2010; Nøstet al. 2015). PCB compounds are still in use, mostly in theelectrical generating sector in Bangladesh (DoE 2007).According to the national report published by theDepartment of Environment (DoE) of the Government ofBangladesh (GoB), the total content of PCBs in functioningelectrical equipment was estimated at 51.6 MT, and the totalPCBs in the electrical sector requiring destruction was esti-mated at 55.8 MT (DoE 2007).
Bangladesh is an agricultural country that has an irregular580-km-long deltaic marshy coastline which is divided by
many rivers and streams that enter into the Bay of Bengal.The environmental and ecological integrity of the coastalareas of Bangladesh suffers from a number of anthropogenicactivities such as the development of industrial hubs, rapidhuman settlement, tourism and transportation, dumping of e-waste, widespread ship breaking and port activities, excessiveoperation of mechanized boats, deforestation, intensive agri-culture and aquaculture activities, discharges of untreated andsemi-treated land-based sewage, and effluents from variouslarge and small local industries.
Serving as an ultimate sink for POPs, the marine or coastalenvironments are greatly affected by PCBs (Jonsson et al.2003). Although the presence of these contaminants are ubiq-uitous, elevated levels are reported particularly in the industri-alized and densely populated coastal areas (Fowler 1990).Fish and seafood may be used as good bio-indicators to ex-amine the status of contamination and distribution of POPs inthe aquatic environments (Ueno et al. 2003). Furthermore,consumption of the contaminated seafood might be a signifi-cant route of POPs exposure to the local populations. About42 million people (30% of the total population) live in thecoastal area (47,211 km2; 32% of the total land area) ofBangladesh, of which about 5 million are engaged directlyin commercial fishing (BOBLME 2011Seafood is one of themost important dietary components of the coastal populations.Therefore, it is an urgent need to assess the potential healthrisk that the Bangladeshi coastal people are exposed throughPCB-contaminated seafood consumption. Due to the lack ofauthenticate evidences of PCB contamination in theBangladeshi seafood, the present study was initiated. To thebest of our knowledge, this is the first comprehensive study todetermine the levels of all PCB congeners in commonly con-sumed seafood (finfish and shellfish) from the coastal area ofBangladesh. The spatial-temporal distribution and composi-tional pattern of PCBs in the seafood samples were investigat-ed. Moreover, we also compared the concentration of PCBs inseafood with data from other countries. Using the data, weestimated the average daily intake of PCBs via seafood con-sumption by the coastal residents (adults and children) ofBangladesh and preliminarily examined the health effects bycomparing the intake data with the health criteria recommend-ed by the international authorities.
Materials and methods
Study area and collection of samples
In the present study, seafood samples were collected from fourmain fish landing centers located in the four major coastalareas of Bangladesh (Cox’s Bazar, Chittagong, Bhola, andSundarbans; Fig. S1) which contribute approximately 70–80% of the tota l catchment of the seafood (see
1356 Environ Sci Pollut Res (2019) 26:1355–1369
Supplementary Material). A total of 48 seafood samples (5finfish and 2 shellfish species) were collected in winter(January–February; n = 24) and summer (August–September; n = 24) of 2015, of which the finfish species in-cluded Ilish (Tenualosa ilisha), Rupchanda (Pampusargentius), Loitta (Harpadon nehereus), Sole (Cynoglossuslingua), and Poa (Otolithoides pama), whereas the shellfishspecies included shrimp (Penaeus indicus) and crab (Scyllaserrata). The selected species are frequently consumed by thecoastal people of Bangladesh (Raknuzzaman et al. 2016).Rupchanda and Sole were not collected from Bhola andSundarbans area due to species unavailability in the catch-ments during our sampling campaign. The seafood sampleswere wrapped in aluminum foil immediately after collectionand kept in ice-filled airtight insulating box and transported tothe laboratory of Fisheries Department of Dhaka University.To remove the surface adherents, samples were rinsed withdeionized water. The length–weight data was recorded foreach individual of every single species. The edible portionsthat the Bangladeshi people usually consume were collectedfrom the seafood samples and homogenized, weighed, freeze-dried for about 48 h. To ensure the representativeness of sam-ples, the collected edible portions from 10 to 20 individuals ofeach species were composited into a single sample as pooled.The processed samples were brought to the YokohamaNational University, Japan for further chemical analyses.The species-specific information with their biometric data(length, weight, and lipid content) are shown in Table S2.
Chemicals and reagents
Native calibration PCBs standards (BP-MXP Native PCBsolution/mixture of PCB 3, 8, 28, 52, 101, 118, 138, 153,180, 194, 206, and 209) and isotopically labeled internal stan-dards (MBP-MXP Mass-labeled PCB solution/mixture of13C-PCB 3, 8, 28, 52, 101, 118, 138, 153, 180, 194, 206,and 209) containing at least one congener for each homologgroup of PCBs were obtained from Wellington LaboratoriesInc. (Ontario, Canada). A complete set of all 209 PCB conge-ners (C-CSQ-SET Congener Calibration Set containing 209native PCB congeners) were also purchased fromAccuStandard (New Haven, CT, USA). Supelclean™ ENVI-18 solid-phase extraction (SPE) cartridges (12 mL, 2 g) werepurchased from SUPELCO® (PA, USA). All of the Quick,Easy, Cheap, Effective, Rugged and Safe (QuEChERS) ex-traction kits were obtained from Agilent Technologies (SantaClara, CA, USA). All solvents (n-hexane, acetone, methanol,and dichloromethane) used for sample processing and analysiswere PCB and pesticide analysis grade and purchased fromWako Chemical (Osaka, Japan). Milli-Q (> 18.2 MΩ cm) wa-ter generated from an ultrapure water purification system(Millipore, Billerica, MA, USA) was used throughout the ex-periment. Mixed cellulose ester (MCE) membrane filters
(0.45 μm, 47 mm i.d.; A045A047A) were purchased fromADVANTEC® (Tokyo, Japan).
Sample pretreatment
Sample preparation for the extraction of PCBs from seafoodinvolves a modified QuEChERS method previously validatedby Ahmed et al. (2016) followed by use of EMR-Lipid (“en-hanced matrix removal of lipids”) and an additional “saltingout/polishing” step for cleanup validated by Han et al. (2016).Briefly, an aliquot of 5 g of homogenized sample was weighedinto a polypropylene tube (50 mL capacity). Then, 5 mL ofultrapure water was added, the tube was manually shaken andspiked with 100 μL of 50 ng/mL of a mixture of 13C-PCBcontaining at least one congener for each PCB homolog as aninternal standard (IS) for quantification. Ten milliliters of ex-traction solvent (n-hexane:acetone ≈ 1:1, v/v) and two ceramicbars (Agilent p/n 5982–9313) were added and the tube wasshaken by hand vigorously for 5 min. Afterwards, theQuEChERS salts (6 g of MgSO4 and 1.5 g of sodium acetate;Agilent p/n 5982–5755) was added and the tube was imme-diately shaken for 1 min to avoid agglomeration of saltsfollowed by ultrasonic agitation for 20 min. The tube wascentrifuged for 5 min at 3500 rpm and 10 mL of the superna-tant were transferred to a 15 mL centrifuge tube containing 1 gEMR–lipid sorbent (Agilent p/n 5982–1010). One ceramicbar (Agilent p/n 5982–9312) was added and the tube wasshaken vigorously by hand for 1 min followed by centrifuga-tion for 3 min at 4000 rpm. The entire supernatant wasdecanted into a second 15 mL polishing tube containing 2 gmixture of 4:1 (w/w) anhydrous MgSO4:NaCl (Agilent p/n5982–0101), and vortexed immediately to disperse, followedby centrifugation at 4000 rpm for 3 min. The extracts werethen transferred into a glass test tube and evaporated to neardryness under a gentle stream of high-purity nitrogen. Theresidue was finally re-dissolved in 1 mL n-hexane and keptat − 20 °C until GC-MS/MS analysis.
Instrumental analysis
Gas chromatograph–tandem mass spectrometry (GC–MS/MS) analysis was performed using an Agilent 7890A GC,coupled with an Agilent 7000C triple-quadrupoleMS. A com-puter with MassHunter software (version B.05.00412) wasused for data acquisition and processing (AgilentTechnologies, Palo Alto, CA). Chromatographic separationwas achieved on an HT8-PCB column (60 m × 0.25 mm ID,0.25 μm film thickness, Kanto Chemical Co., Inc., Japan)using helium as a carrier gas at a flow rate of 1.0 mL/min.The GC oven temperature was set initially at 120 °C for 1 min,increased to 180 °C for 0 min at 20 °Cmin−1, raised to 210 °Cfor 0 min at 2 °C min−1 and finally held at 310 °C for 3 min at5 °C min−1. The injection volume was set to 1 μL in splitless
Environ Sci Pollut Res (2019) 26:1355–1369 1357
mode. Mass spectrometry was operated in multiple reactionsmonitoring (MRM) mode with a gain factor of 10. Electronimpact (EI) ionization voltage was 70 eV. Nitrogen andhelium were used as collision gas and quench gas in the col-lision cell at constant flows of 1.5 and 2.25 mL/min, respec-tively. The sets of temperatures of transfer line, ionizationsource, and triple quadrupole mass analyzer were 300, 280,and 150 °C, respectively. A solvent delay was set at 3 min.The first (Q1) and the third quadrupole (Q3) were both oper-ated at “wide” resolution mode. Prior to analysis, MS/MS wasauto-tuned with perfluorotributylamine. All of 209 PCB con-geners were separated into 167 peaks, herein termed domains,denoting 135 individual and 32 coeluting congeners. The or-der of chromatograms and peak assignments were set accord-ing toMatsumura et al. (2002). GC–MS/MS conditions and/orparameters for the analysis of PCBs are shown in Table S3. Toidentify the target analytes, the retention times of the peaksdetected in samples were compared with the peaks obtainedfrom a GC-MS/MS run using a standard solution of a mixtureof all 209 PCB congeners. The peak areas of each analyte inthe samples, the mass/area ratio of the internal standard, theresponse factor obtained from the calibration curve, and theoriginal sample weight were used for PCB quantification. Thedetailed calculating equation has also been given in the reportof Yang et al. (2011). The concentrations of PCB congenersand total PCB (∑PCBs) are expressed as nano gram per gramwet weight (ng/g ww). In addition, the values of < LODs wereassigned to zero while calculating the concentration of PCBhomologs and total PCBs.
Quality assurance and quality control
Strict quality control procedure was maintained during theexperiments. The containers and equipment used during thewhole procedure were pre-cleaned with methanol followed byacetone. By taking into account all the steps of the analyticalmethods, the limit of detection (LOD) was calculated as threetimes the signal to noise (S/N) ratio for each compound. TheLODs were in the range of 0.006 to 0.06 ng/g ww. The instru-mental blanks (solvent without internal standard) and proce-dural blanks (Milli-Q water spiked with internal standards)analyzed with every batch of samples gave S/N values lessthan 3 (< LODs). To validate the method, the accuracy andprecision of the methods was determined from the matrixspike recovery studies in triplicate (n = 3) by spiking the targetcompounds into the samples at 10 ng/g, followed by similarextraction and analysis procedure for the actual samples asdescribed in earlier sections. The mean recoveries (accuracy)of spiked PCBs and the relative standard deviation (RSD)(precision) were 57–113% and 2–15%, respectively. The re-sults of the validation studies indicated good and acceptedaccuracy and precision of the analytical methods. The detailed
quality assurance and quality control (QA/QC) data are givenin Table S4.
Determination of lipid content
The lipid content (%) in the fish muscle tissue was determinedby a gravimetric method adopted from Tölgyessy andMiháliková (2016). Briefly, an amount of 5 g of seafood tissuehomogenate in a 50 mL PP centrifuge tube was extracted with10 mL of extraction solvent (n-hexane:acetone ≈ 1:1, v/v)using vigorous hand shaking for 5 min. After addition of 2 gof anhydrous MgSO4 and 0.5 g of NaCl and agitation, theorganic phase was separated by centrifugation at 4000 rpmfor 5 min. Two 1-mL aliquots of the upper organic phase weretransferred into a pre-weighed glass tubes. The extracts werethen air-dried at room temperature and the residues were oven-dried at 105 °C for 1 h to a constant weight. Final residueswere weighed and an average value was taken for calculationof percent lipid content on a wet weight basis.
Data analysis
The IBM SPSS (Version 23.0, IBM Corp., NY, USA) andXLSTAT (Version 2016.02.28451, Addinsoft, NY, USA) soft-ware were used for statistical analyses. Before analysis, thesignificance level was set at p = 0.05 and concentrations lessthan LODs were set to LOD/2 (Succop et al. 2004). The nor-mality of the data set was tested by a statistical distribution testcalled P–P plots. The differences among the concentrations ofPCBs in the seafood samples and seasonal variations weretested by one-way analysis of variance (ANOVA). The spatialvariations of PCBs in seafood were shown by usingMapViewer™ software (Version 8, Golden Software Inc.,CO, USA). The shorthand numbering notation (PCB1–209)of Ballschmiter and Zell (1980) was followed for PCB no-menclature. The WHO 2005 toxic equivalency factor (TEF)(Van den Berg et al. 2006) was used to calculate the total PCBtoxic equivalents (TEQ).
Results and discussion
Concentration of PCBs in seafood and globalcomparison
PCBs were detected in all of the 48 seafood samples including5 finfish and 2 shellfish species, suggesting the ubiquitousoccurrence and continuous accumulation of these compoundsin the Bangladeshi seafood. The data of PCBs in seafoodsamples are presented in Table 1 and Fig. 1, whereas the de-tailed results are presented in Table S5 and S6. Regardless ofspecies and seasons, a total of 141 domains consist of 114single congeners, 20 double coeluting isomers, 5 triple
1358 Environ Sci Pollut Res (2019) 26:1355–1369
Table1
Concentratio
ns(ng/gww)of
PCBsandfatcontent
(%)in
seafood(finfish
andshellfish)
collected
from
thecoastalareaof
Bangladeshin
thetwoseasons
PCBhomolog
Finfish
Ilish
Rupchanda
Loitta
Sole
Poa
CX
CT
BH
SNCX
CT
CX
CT
BH
SN
CX
CT
CX
Winter
Fat(%)
19.3
21.1
14.5
13.2
2.8
2.1
2.6
2.9
1.9
2.0
3.1
2.2
3.0
Mono-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
Di-CB
1.32
1.05
1.24
1.33
0.38
0.31
0.54
0.41
0.16
0.42
1.04
0.56
0.50
Tri-CB
4.25
3.83
3.62
4.35
1.18
1.22
1.81
1.51
0.68
0.80
1.94
2.12
1.63
Tetra-CB
7.75
13.60
7.62
7.37
1.87
1.43
2.53
2.64
1.03
1.34
2.62
4.22
2.56
Penta-CB
22.84
22.45
17.40
15.70
2.71
2.59
4.80
6.62
2.03
2.14
7.30
9.52
4.65
Hexa-CB
18.36
19.43
13.49
13.69
2.93
2.15
4.07
4.96
1.37
1.51
7.05
7.53
5.27
Hepta-CB
17.54
18.22
11.54
11.48
2.22
1.53
2.53
3.81
1.03
1.57
4.67
5.11
4.25
Octa-CB
0.57
1.03
0.88
0.60
0.08
0.14
0.46
0.31
0.09
0.13
0.39
0.47
0.46
Nona-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
Deca-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
∑DL-PCBsa
8.8
7.9
5.6
5.8
1.1
1.2
2.1
2.5
0.8
0.9
2.7
4.1
2.2
∑iPCBsb
17.2
17.3
15.4
153.1
2.3
4.1
5.6
1.6
2.1
6.9
7.1
4.7
∑PC
Bsc
72.6
79.6
55.8
54.5
11.4
9.4
16.7
20.3
6.4
7.9
2529.5
19.3
Summer
Fat(%)
20.1
22.3
16.2
14.1
3.5
3.0
2.4
3.5
1.9
2.1
3.4
3.8
3.1
Mono-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
Di-CB
2.93
3.12
1.83
2.22
0.48
0.42
0.55
0.44
0.24
0.30
1.30
0.52
0.29
Tri-CB
7.22
4.66
5.83
5.03
1.19
1.66
2.59
2.95
1.12
0.79
5.07
5.80
3.44
Tetra-CB
13.54
16.46
14.33
12.51
2.84
1.94
4.61
4.51
1.15
1.85
5.41
7.18
3.59
Penta-CB
22.89
24.57
18.57
15.29
5.03
2.49
5.05
8.16
1.81
2.90
9.79
11.33
5.30
Hexa-CB
17.14
22.26
15.11
15.88
5.23
2.43
3.74
5.10
1.85
2.50
7.91
8.14
6.60
Hepta-CB
12.09
13.77
12.25
9.91
2.56
1.31
2.28
3.50
1.22
2.48
5.32
5.95
3.61
Octa-CB
2.36
1.35
1.17
0.84
0.13
0.24
0.56
0.32
0.16
0.28
1.06
0.72
0.72
Nona-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
Deca-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
∑DL-PCBs
10.9
13.2
9.5
8.5
1.6
1.4
2.5
3.7
0.7
13.6
4.2
2.6
∑iPCBs
14.7
17.2
14.7
11.2
4.4
2.5
3.6
5.2
2.1
3.1
78.5
4.8
∑PC
Bs
78.2
86.2
69.1
61.7
17.5
10.5
19.4
257.6
11.1
35.9
39.7
23.5
Environ Sci Pollut Res (2019) 26:1355–1369 1359
PCBhomolog
Finfish
Shellfish
Poa
Shrim
pCrab
CT
BH
SN
CX
CT
BH
SN
CX
CT
BH
SN
Winter
Fat(%)
3.7
3.0
2.3
2.0
1.9
1.2
3.0
4.1
4.5
3.6
3.9
Mono-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
Di-CB
0.42
0.38
0.20
0.26
0.17
0.03
0.57
1.43
0.94
0.46
1.34
Tri-CB
1.28
0.94
0.85
1.03
0.63
0.55
1.83
4.26
4.08
2.32
2.90
Tetra-CB
3.11
1.61
1.37
1.90
1.34
1.26
2.12
5.96
5.72
2.26
3.34
Penta-CB
5.26
3.22
2.21
2.71
2.13
1.14
2.80
7.76
10.81
3.73
7.14
Hexa-CB
6.86
4.68
1.97
1.86
1.38
1.02
2.89
7.27
7.64
7.00
6.17
Hepta-CB
4.15
1.86
1.12
1.33
1.60
1.11
2.14
6.48
7.26
4.87
6.03
Octa-CB
0.29
0.15
0.12
0.12
0.05
0.05
0.28
1.12
1.26
0.63
0.86
Nona-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
Deca-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
∑DL-PCBsa
2.5
1.7
1.1
1.2
1.1
0.6
1.7
54.6
3.2
3.7
∑iPCBsb
5.7
4.1
1.8
1.9
1.8
1.1
2.6
6.4
7.3
4.2
5.1
∑PC
Bsc
21.4
12.8
7.8
9.2
7.3
5.2
12.6
34.3
37.7
21.3
27.8
Summer
Fat(%)
3.9
2.9
2.4
2.3
2.0
1.5
2.6
4.0
4.2
3.1
3.4
Mono-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
Di-CB
0.47
0.47
0.21
0.07
0.12
0.02
0.19
1.26
1.02
0.40
0.86
Tri-CB
1.91
1.04
0.78
0.92
0.68
0.49
1.10
3.43
4.52
1.61
1.94
Tetra-CB
6.96
4.08
1.82
1.77
1.67
1.13
1.76
4.55
5.30
1.94
2.38
Penta-CB
6.90
5.25
2.81
2.27
1.45
0.87
1.86
5.32
7.04
2.73
3.86
Hexa-CB
5.66
4.71
2.08
1.52
0.90
0.68
2.24
5.21
6.21
4.17
5.36
Hepta-CB
4.14
2.11
1.39
0.87
1.00
0.62
1.27
4.54
5.36
2.88
6.04
Octa-CB
0.78
0.44
0.14
0.09
0.02
0.02
0.20
0.67
0.62
0.47
0.43
Nona-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
Deca-CB
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
<LOD
∑DL-PCBs
2.7
21.3
11
0.4
1.2
3.2
3.1
1.9
1.7
∑iPCBs
6.4
3.6
21.3
1.3
0.9
1.7
4.6
6.9
3.6
5.1
∑PC
Bs
26.8
18.1
9.2
7.5
5.9
3.8
8.6
2530.1
14.2
20.9
aSu
mof
12dioxin-likePCBs(4
non-orthosubstituted
PCBs:PCB77,81,126,169;
8mono-orthosubstituted
PCBs:PC
B105,114,118,123,156,157,167,189)
bSu
mof
6ICESindicatoror
markerPCBs(ICES6-PC
Bs:PC
B28,52,101,138,153,and180)
(EuropeanCom
mission
2011)
cThe
concentrations
ofcongenerswith
<LODwereassumed
tobe
zero
whilecalculatingthetotalP
CBs;CXCox’sBazar,C
TChittagong,B
HBhola,S
NSu
ndarbans
Tab
le1(contin
ued)
1360 Environ Sci Pollut Res (2019) 26:1355–1369
coeluting isomers, and 2 quadruple coeluting isomers wereresolved in 48 tissue samples. However, the detected PCBswere only the congeners with two to eight Cl atoms. Therecoveries for mono-CB were very low (e.g., 57% forPCB3; Table S4), thus indicating that these compounds mightbe lost at some stages of the sample pretreatment procedure.Congeners with one Cl atom might be evaporated in the pre-treatment procedure, and congeners with nine and ten Clatoms were under their LODs. The total concentrations ofthe PCBs (∑PCBs, sum of detected domains) in the seafoodsamples ranged from 5.2 to 79.6 ng/g ww in winter, and from3.8 to 86.2 ng/g ww in summer. The class distribution ofsamples based on ∑PCBs levels were 17% (> 50 ng/g ww),17% (25–50 ng/g ww), and 66% (< 25 ng/g ww). A positivecorrelation was extracted between the∑PCBs and the numberof detected domains per sample (r = 0.86, p < 0.05). However,the variation in total PCB concentrations might be dependenton some dominant congeners. The sum of 12 dioxin-like PCB(∑DL-PCBs) and 6 ICES (International Council for theExploration of the Sea) indicator/marker PCB congeners(∑iPCBs) ranged from 0.6 to 8.8 ng/g ww (mean 3 ng/gww) and 1.1 to 17.3 ng/g ww (mean 6 ng/g ww) in winter,and from 0.4 to 13.2 ng/g ww (mean 3.5 ng/g ww) and 0.9 to17.3 ng/g ww (mean 5.7 ng/g ww) in summer, respectively.The ∑DL-PCBs and ∑iPCBs together accounted for 26–47%of the ∑PCBs concentrations. Therefore, the DL-PCBs andiPCBs could not comprehensively reflect the contaminationlevel of PCB in seafood, elucidating that the degree of PCBcontamination in the Bangladeshi coastal area might be im-pacted by some other congeners along with these frequentlymonitored PCBs. However, correlations between ∑PCBs and∑DL-PCBs, and ∑iPCBs were strong (r = 0.98 and 0.99, re-spectively) and significant (p < 0.05).
The levels of PCBs in the Bangladeshi seafood were com-pared to other areas around the world and presented in Table 2.To some degree, the results of comparison could at least reflectthe present scenario of PCB contamination in the Bangladeshiseafood, although the numbers of monitored PCB congeners,the sampling time and methods, and the investigated species
among these studies might be different from each other. Thecomparison results revealed that the ∑PCBs concentrationsfor both seasons in this study were comparable to or higherthan those reported in finfish and shellfish from Spain (Perelloet al. 2012); South Korea (Moon et al. 2009); Dalian, Tianjin,and Shanghai in China (Yang et al. 2006); Indonesia(Sudaryanto et al. 2007); Eastern Coast of Thailand(Jaikanlaya et al. 2009); Southern Iran (Mohebbi-Nozaret al. 2014); and Hong Kong (Chan et al. 1999), but lowerthan that from Scheldt estuary (Netherlands–Belgium) (VanAel et al. 2012), Houston Ship Channel, USA (Howell et al.2008), and Hyderabad, India (Ahmed et al. 2016). Besides,the ∑PCBs in half of the seafood samples (50–58%) weregreater than the screening value of 20 ng/g ww proposed bythe USEPA (2010) and were comparable to or at the middlerange of the worldwide PCBs data reported in the case ofedible marine species (Domingo and Bocio 2007).Therefore, the present level of PCBs measured in theBangladeshi seafood is obviously a matter of concern.
Influence of seasons and species on PCB accumulationin seafood
The levels of∑PCBs in finfish were slightly higher in summer(7.6–86.2 ng/g ww) than in winter (6.4–79.6 ng/g ww),whereas in shellfish it was higher to an extent in winter(5.2–37.7 ng/g ww) compared to summer (3.8–30.1 ng/gww), although these differences were not statistically signifi-cant in either cases (p > 0.05). The lipid content of the exam-ined seafood species also varied similarly between the twoseasons, although the differences did not show statistical sig-nificance (p > 0.05). However, the variations might be attrib-uted to the seasonal discrepancies in their physiological activ-ities, feeding behavior, as well as the degree of contaminationof their habitats and these will be explained in details in thefollowing discussions.
The interspecies differences in the levels of ∑PCBs con-centrations among the four coastal areas were statistically sig-nificant (ANOVA, p < 0.05). In particular, significantly higher
0
20
40
60
80
100
CX CT BH SN CX CT CX CT BH SN CX CT CX CT BH SN CX CT BH SN CX CT BH SN
Ilish Rupchanda
Loi�a Sole Poa Shrimp Crab
Finfish Shellfish
sBCP∑fonoitartnecnoC
)w
wg/gn(
Winter Summer
Fig. 1 Concentrations of totalPCB (∑PCBs) in the seafoodsamples (finfish and shellfish)from the coastal area ofBangladesh in winter andsummer. In the figure: CX Cox’sBazar, CT Chittagong, BH Bhola,SN Sundarbans
Environ Sci Pollut Res (2019) 26:1355–1369 1361
levels of ∑PCBs were found in Ilish (54.5–79.6 and 61.7–86.2 ng/g ww in winter and summer, respectively) amongthe finfish species and crab (21.3–37.7 and 14.2–30.1 ng/gww in winter and summer, respectively) within the shellfishspecies (p < 0.05 for both cases). These differences were prob-ably due to the elevated lipid content in these species (Table 1).Batang et al. (2016) reported that PCBs accumulation in fishtissue tend to increase with lipid content. Our analysis revealeda positive and strong correlation between the levels of PCB andthe lipid content in the studied seafood samples (r = 0.95;p < 0.01), whereas the association between lipid-adjustedΣPCB and lipid content was not significant. Furthermore, highcorrelations were also apparent between the concentrations ofPCBs and the length and weight of the examined seafood spe-cies (r = 0.75 and 0.88; p < 0.01). In finfish, irrespective toseasons, the concentrations of ∑PCBs showed the followingtrend: Ilish > Sole > Poa > Loitta > Rupchanda. This patterncorrelates well with fish size (length), weight, and lipid yieldsobtained from the individual fish species which is in well agree-ment with the previous studies reported elsewhere (Larssonet al. 1991; Batang et al. 2016).
Besides the species-specific biometric data, accumulationpotentials of PCBs in fish tissue might also be influenced bythe organisms’ bio-demographic parameters, such as their tro-phic level, feeding behavior, reproductive status, and metabo-lism including uptake and elimination, depth, and habitat loca-tion. Comparatively higher PCBs levels along with larger num-ber of detected domains in Ilish could be explained by thefollowing phenomena: (1) it is a high trophic carnivorous spe-cies that tends to concentrate contaminants to a higher degreethan other species (Das and Das 2004; Miao et al. 2000). (2) Itis an anadromous species which migrates from the sea to the
rivers for spawning. Consequently, this species is habitated todifferent ecosystems (marine, estuarine, brackish, and freshwa-ter) and exposed to various degree of contamination. (3) Ilishhas relatively a larger body surface area with a very thin layer ofskin and a larger gill surface area that facilitate the process ofaccumulation of the contaminants like PCBs from water col-umn into the fish body. McKim and Heath (1983) reported thatPCBs can undergo bioconcentration via gills in fish, althoughthe main uptake is through their diet. In addition, elevated con-centrations of PCB in Sole could be related to their livinghabitats. Since sediment has been recognized as an ultimatesink of organic pollutants like PCBs, there is high potential ofPCB accumulation in the sediment-dwelling fish such as Solewhich predate benthic organisms frequently or consume someorganic particles from sediments.
Within shellfish species, fewer congeners and lower levelsof ∑PCBs in Shrimp might be attributed to their low extract-able lipids (2.1 ± 0.4%) and their pelagic nature and feedingpattern (Voorspoels et al. 2004). Shrimp primarily feed onplanktons of low trophic level (e.g., mysids and amphipods),resulting in their less intense contact with the sediment com-pared to other exclusively benthic species like crab (Oh et al.2001). Furthermore, faster metabolization and elimination ofcertain PCBs was reported in shrimp than some other seafoodspecies (Goerke andWeber 2001; Voorspoels et al. 2004). Theother shellfish species (Crab) contained relatively higher PCBlevels and that can be attributed to their comparatively higherlipid content and body weight as well as their living and feed-ing habit (Voorspoels et al. 2004). In particular, crab is a typ-ical benthic organism, also known as a scavenger that tends tofeed partially on detritus or decaying organic material withhigh pollutant loads (Everaarts et al. 1998; Ip et al. 2005).
Table 2 Global comparison of total PCBs (ng/g ww) in seafood
Location Sampling year Species Na ∑PCBs References
Twelve cities of Catalonia, Spain 2008 Finfish and shellfish 18 0.35–48.8 Perello et al. 2012
South Korea 2005–07 Finfish and shellfish 22 0.2–41 Moon et al. 2009
Dalian, Tianjin and Shanghai in China 2002 Finfish and shellfish –b 0.83–11.4 Yang et al. 2006
Scheldt estuary (Netherlands–Belgium) 2010 Finfish and shellfish 33 3.27–285 Van Ael et al. 2012
Indonesia 2003 Finfish NA 23 Sudaryanto et al. 2007
Eastern Coast of Thailand 2006–07 Shellfish 49 0.05–7.53 Jaikanlaya et al. 2009
Houston Ship Channel, USA 2003 Finfish and shellfish 209 3.44–1596 Howell et al. 2008
Southern Iran 2010–11 Finfish and shellfish 7 1.91–7.05 Mohebbi-Nozar et al. 2014
Hyderabad, India 2013–14 Finfish 45 0.22–118.7 Ahmed et al. 2016
Hong Kong 1997 Finfish 51 < 0.01–94 Chan et al. 1999
Coastal area of Bangladesh 2015 Finfish and shellfish 209 5.2–79.6 (W)c This study3.8–86.2 (S)c
a Number of PCB congenersb Sum of di- to deca-CB homologscW and S represent winter and summer, respectively
1362 Environ Sci Pollut Res (2019) 26:1355–1369
The sediment-dwelling organisms such as crabs are exposedto sedimentary PCBs via several routes, such as direct contactand secondary ingestion of sediment, and tend to exhibithigher contaminant load compared to other organisms(Storelli et al. 2003). Moreover, crabs also possess gills witha relatively larger surface area which could facilitate the accu-mulation of PCBs into their bodies since gills can continuous-ly transfer the pollutants from both water and suspended par-ticles onto its surface that are subsequently distributedthroughout the whole body via blood (Yang et al. 2006).
The depth at which organisms live is an important factor indetermining the accumulation of contaminants between thespecies. In general, deep water seafood species showed higherconcentrations of POPs including PCBs (de Brito et al. 2002;Storelli et al. 2009). The above statements agree well with ourresults, both Ilish, Sole, and crab being organisms inhabitingdeeper marine or coastal areas. In general, the concentration of∑PCBs in finfish (6.4–79.6 and 7.6–86.2 ng/g ww in winterand summer, respectively) were significantly higher (p < 0.05)than those in shellfish samples (5.2–37.7 and 3.8–30.1 ng/gwwin winter and summer, respectively), elucidating a higher po-tential of PCB bioaccumulation in finfish compared to theshellfish species. Shen et al. (2009) also reported the elevatedlevels of certain POPs including PCBs in fatty fish (finfish)compared to that of shellfish from six coastal provinces inChina. Howell et al. (2008) reported that the average PCBconcentrations in catfish tissue (finfish) exceeded those deter-mined in crab tissue (shellfish) by a factor of approximately six.
Spatial distribution
The spatial variations of PCBs in the examined seafood speciesare presented in Fig. 2. Geographically, the differences in PCBconcentrations among the four coastal areas were not
statistically significant (p > 0.05). However, regarding finfish,the highest mean concentration of ∑PCBs were found in sam-ples from Chittagong in both seasons (32 ng/g ww in winter,37.6 ng/g ww in summer). On the other hand, shellfish samplesfrom Chittagong also exhibited the maximum mean levels of∑PCBs (22.5 ng/g ww in winter, 18 ng/g ww in summer). Ingeneral, the mean concentration of ∑PCBs seemed to have thefollowing trends: Chittagong > Cox’s Bazar > Bhola >Sundarbans, and Chittagong > Cox’s Bazar > Sundarbans >Bhola in finfish and shellfish, respectively (Fig. 2). The con-tamination level of PCBs in seafood is significantly affected byenvironmental factors such as water and sediment, since theaquatic organisms are primarily exposed to the contaminantslike PCBs from water and sediment. Interestingly, the spatialdistribution pattern of PCBs in seafood coincided to a greaterextent with that of water and sediment (data not shown).Furthermore, relationships between levels of ∑PCBs in water-seafood and sediment-seafood were investigated and the resultsrevealed significant correlations for both cases (r = 0.72, r =0.74; p < 0.05). However, our PCB data for water and sedimentsamples elucidated that the PCBs in the aquatic environment ofthe Bangladeshi coastal area were mainly originated from pointsources such as ship breaking industries, port activities, e-wastelandfills and dumping, untreated or semi-treated industrial andmunicipal effluents, etc. along with non-negligible but minorcontribution from non-point/defuse sources such as atmospher-ic wet/dry deposition and long-range transport via ocean cur-rents/tides. Overall, the seafood species collected fromChittagong, Cox’s Bazar, and Sundarbans areas depicted higherlevels of PCB contamination than those from Bhola (MeghnaEstuary). These spatial differences might be attributed to greaterdevelopments in these regions, and thus these compounds areassociated to recent urbanization and intense industrialization.
Homolog composition
The PCB congener and homolog profiles in environmentalmatrices can often provide valuable information on the envi-ronmental source, transport, and fate processes of PCBs.Figure 3 shows the PCB homologs composition (%) in theseafood samples for both seasons (winter and summer).Analysis of variance (ANOVA) revealed that, regardless ofspecies, there was no significant difference in homolog com-positions between the two seasons (p > 0.05). Also, the homo-log composition in seafood species, regardless of source, didnot differ significantly (p > 0.05), although the compositionswere not fully identical. In general, regardless of source (sitesand species) and season, the proportions of the homologs withdifferent degree of chlorination were in the decreasing order ofmedium (tetra- to hexa-CBs) > heavier (hepta- to deca-CBs) >lighter (di- to tri-CBs), contributing 56–78, 13–31, and 6–20%to the ∑PCBs, respectively.
Legends
SummerWinter
40 ng/g ww
FF: FinfishSF: Shellfish
89 E
23 N
Fig. 2 Spatial distribution of average ∑PCBs in seafood (finfish andshellfish) from the coastal area of Bangladesh. Colored area in the insetmap represents the coastal area of Bangladesh
Environ Sci Pollut Res (2019) 26:1355–1369 1363
In general, penta- and hexa-CB was the most prevalenthomologs in the seafood samples, accounting for about halfof ∑PCBs (42–62 and 40–50% of ∑PCBs in winter and sum-mer, respectively), followed by hepta-CB (14–24 and 12–29%of ∑PCBs in winter and summer, respectively) and tetra-CB(10–24 and 11–29% of∑PCBs in winter and summer, respec-tively). Predominance of penta- and hexa-CB homologs werealso reported for seafood from the Eastern Coast of Thailand(Jaikanlaya et al. 2009), Houston Ship Channel, Texas(Howell et al. 2008), South Korea (Moon et al. 2009), BlackSea, Bulgaria (Stancheva et al. 2017), the Mediterranean Sea(Storelli et al. 2009), Gulf of Naples, Southern Italy (Nasoet al. 2005), and coastal fisheries in China (Pan et al. 2016).Particularly, in most of the examined samples, congeners withhigher number of chlorine atoms were accumulated more thanthat of congeners with fewer chlorine atoms, which might bedue to the fact that the lipophilicity of PCB congeners increasewith the number of chlorine atoms substituted to the hydrogenatoms in biphenyl rings increase (Bernes 1998). Additionally,PCB molecules with fewer chlorine are larger enough to havegreater difficulty while passing through the organism’s cellmembranes, and thus tend to less efficient accumulation infish samples (Bernes 1998).
Congener profiles and source characterization
We identified the dominant PCBs in the seafood samples bothby occurrence and abundance. The data of detection rate of eachcongeners/co-elute and their mean concentrations were com-bined to rank them in decreasing order. The top 20 domainswere identified for their significant contributions to∑PCBs andthese were PCB153, 180, 138, 120 + 110, 102 + 93 + 98 + 95,187, 170, 118, 101, 52, 43 + 49, 99, 128, 66, 105, 151, 177, 28,18, and 8 + 5, comprising up to 59–82% of ∑PCBs by species,and sum of them was highly correlated with ∑PCBs (r = 0.98;p < 0.01), well representing the ∑PCBs in the Bangladeshi sea-food. Interestingly, the top PCB congeners identified in thisstudy are frequently reported in the environment media andbiological samples (Jones 1988; Hansen 1998). PCB153, 180,and 138 are generally abundant in finfish and/or shellfish tissue(Miao et al. 2000; Storelli et al. 2009; Xia et al. 2012;Mohebbi-Nozar et al. 2014; Batang et al. 2016). These congeners aremetabolically obstinate against monooxygenase enzyme andtend to eliminate more slowly from the fish body because oftheir high degree of chlorination on the aromatic rings (Storelliet al. 2009). Moreover, the most abundant congeners identifiedin the present study correspond to diverse homolog groups (tri-
a
b
0%
25%
50%
75%
100%
CX CT BH SN CX CT CX CT BH SN CX CT CX CT BH SN CX CT BH SN CX CT BH SN
Ilish Rupchanda
Loi�a Sole Poa Shrimp Crab
Finfish Shellfish
)retniw(
noitisopmoc
%
Deca-CB
Nona-CB
Octa-CB
Hepta-CB
Hexa-CB
Penta-CB
Tetra-CB
Tri-CB
Di-CB
Mono-CB
0%
25%
50%
75%
100%
CX CT BH SN CX CT CX CT BH SN CX CT CX CT BH SN CX CT BH SN CX CT BH SN
Ilish Rupchanda
Loi�a Sole Poa Shrimp Crab
Finfish Shellfish
)rem
mus(noitisop
moc%
Deca-CB
Nona-CB
Octa-CB
Hepta-CB
Hexa-CB
Penta-CB
Tetra-CB
Tri-CB
Di-CB
Mono-CB
Fig. 3 Relative contribution ofPCB homologs (% composition)to the total PCB in theBangladeshi seafood (finfish andshellfish) in winter (a) andsummer (b), 2015. In the figure:CX Cox’s Bazar, CT Chittagong,BH Bhola, SN Sundarbans
1364 Environ Sci Pollut Res (2019) 26:1355–1369
to hepta-CBs), suggestingmultiple sources of PCBs in this area.Sahu et al. (2009) reported that the presence of different conge-ners belonging to different homologs was an indication of var-ious sources of PCBs in marine/coastal environments.Furthermore, the dominant congeners in the seafood samplescontribute major percentages of several commercial PCB mix-tures, e.g., PCB18, 8, 28, 52, 66 contribute 29% in Aroclor1242; PCB52, 66, 18, 28, 49, 118, 101 contribute 30% inAroclor 1248; PCB110, 101, 118, 95, 138, 52, 153 contribute46% in Aroclor 1254; and PCB180, 153, 138, 187, 170, 101,151 contribute 43% in Aroclor 1260 or equivalents (Jones1988; Ishikawa et al. 2007). It indicates that the environmentalPCB burdens in the Bangladeshi seafood might be originatedfrom more than one PCB Aroclors or equivalents and the mostlikely mixtures were Aroclors 1248, 1254, and 1260 orequivalents.
Among the 12 DL-PCBs studied, PCB81, 105, 114, 118,123, 156, 167, and 189 were detected in all of the samples forboth seasons, whereas PCB77, 126, 157, and 169 were foundin 67, 33, 100, 33% and 71, 33, 96, 25% ofwinter and summersamples, respectively. Regardless of source and season,PCB118 contributed 16–57% to ∑DL-PCB by species,followed by PCB105 (4–43%) and 156 (2–17%). The abun-dance of PCB105, 118, and 156 among the DL-PCBs in fishand shellfish has been reported previously (Bright et al. 1995;Bhavsar et al. 2007; Mezzetta et al. 2011). However, based onthe abundant DL-PCB congeners, we can say that emissionfrom industrial applications might be the major sources of DL-PCBs in the study area (Azari et al. 2007), whereas combus-tion processes contributed to some extent which can be con-firmed by the presence of PCB77 and 126 (marker PCBs forcombustion processes, Kim et al. 2004; Ishikawa et al. 2007)with a contribution of up to 1.7% of ∑PCBs.
Exposure assessment of dietary PCBs from seafoodconsumption
Consumption of seafood has been recognized as one of themajor routes of human exposure to POPs including PCBs(Smith and Gangolli 2002). Storelli et al. (2003) suggestedseafood as a primary vector of these pollutants/toxicants forhuman. Since seafood contributes a major portion in the dietfor the Bangladeshi coastal population, the consumption ofcontaminated seafood can be a potential risk for the con-sumers. To comprehensively evaluate risk exposure, the esti-mated daily intakes (EDI) for PCBs for both the adults andchildren were calculated and presented in Fig. 4. In this study,we evaluated two approaches: (1) EDI of ∑PCBs consideringall 209 congeners where the toxic effects of non-dioxin likePCBs (NDL-PCBs) were also taken into account, and (2) EDIof ∑DL-PCBs aiming to assess health risk from dioxin-liketoxicities. Lower-bound approach (non-detected andcongeners with < LOD were assigned to zero) was followed
in both cases. EDI of∑PCBs and∑DL-PCBs were calculatedaccording to the Eqs. (1) and (2), respectively.
EDIΣPCBs ¼ Cfinfish � Intakefinfishð Þ þ Cshellfish � Intakeshellfishð ÞBW
ð1ÞEDIΣDL−PCBs
¼ TEQfinfish � Intakefinfishð Þ þ TEQshellfish � Intakeshellfishð ÞBW
ð2Þwhere EDI∑PCBs is the estimated daily intake of ∑PCBs(ng/kg bw/day), C is the concentration of ∑PCBs in seafood(ng/g ww), EDI∑DL-PCBs is the estimated daily intake of∑DL-PCBs (pg TEQ/kg bw/day), TEQ is the toxic equivalent con-centrations (pg TEQ/g ww) which were calculated by sum-ming the concentration of each DL-PCB congener weightedby the 2005 WHO toxic equivalent factors (TEFs) (Van denBerg et al. 2006), Intake is the daily consumption of seafood(g/day), and BW is the average body weight (60 kg for adultsand 25 kg for children). The daily consumption data of sea-food for the adults (≥ 18 years) and children (6–17 years) inthe coastal area of Bangladesh were obtained from aquestionnaire-based dietary survey during our sampling cam-paign. The seafood consumption data and the estimated dailyintake (EDI) of ∑PCBs and ∑DL-PCBs by the adults andchildren in the coastal area of Bangladesh are given inTable S7.
The estimated exposure to ∑PCBs (EDI∑PCBs), regardlessof sites, ranged from 42 to 50 ng/kg bw/day for adults and 49to 62 ng/kg bw/day for children. These values were 2–3 timeshigher than the tolerable daily intake (TDI) of 20 ng/kg bw/
a
b
0
20
40
60
80
CX CT BH SN
Coastal area
EDI
sBCP∑)yad/
wbgk/gn(
AdultChildren
Tolerable daily intake(TDI) of 20 ng/kg bw/dayproposed by WHO
0
4
8
12
16
CX CT BH SN
Coastal area
EDI ∑D
L-PC
Bs
)yad/wb
gk/QET
gp(
AdultChildren
Maximum tolerable intakeof 4 pg TEQ/kg bw/dayproposed by WHO
Fig. 4 Estimated daily intake (EDI) of total PCBs (a) and DL-PCBs (b)for adults and children along with WHO guideline values (horizontaldotted lines)
Environ Sci Pollut Res (2019) 26:1355–1369 1365
day proposed by WHO as a health-based guidance value fortotal PCB (WHO 2003). Hence, the risk from dietary intake of∑PCBs through seafood consumption appears to be high forboth the adults and children. Likewise, the estimated dailyintake of DL-PCBs (EDI∑DL-PCBs) from seafood consumptionwere from 1.3 to 10.5 pg TEQ/kg bw/day and 1.6 to 13.7 pgTEQ/kg bw/day for adults and children, respectively. As seenin Fig. 4, EDI∑DL-PCBs for both the adults and children fromChittagong and Cox’s Bazar were higher than the maximumtolerable intake of 4 pg TEQ/kg bw/day recommended byWHO (1998). However, the estimated exposure to DL-PCBsfor both population subgroups frequently exceeded the mostrestrictive value of the tolerable daily intake (TDI; < 1 pgTEQ/kg bw/day) (WHO 1998). In particular, the childrenwere exposed slightly higher to the dietary PCBs in seafoodthan that for the adults, due to their lower average bodyweight. Furthermore, the highest EDIs of ∑PCBs and ∑DL-PCBs were found in Chittagong area for both the adults andchildren (Fig. 4 and Table S7), elucidating that the coastalresidents in the Chittagong area were more prone to adversehealth effects associated with dietary exposure to PCBs fromseafood consumption.
Furthermore, the daily intake calculated in this study mightbe underestimated because the WHO proposed TDI of 1–4 pgTEQ/kg bw/day includes the PCDDs and PCDFs along withthe DL-PCBs. However, several previous studies have proventhat DL-PCBs are the most important chemicals contributingup to 70–90% to the total TEQ intake from seafood consump-tion (Moon and Choi 2009; Focant et al. 2002; Shaw et al.2006; Lee et al. 2007).Moreover, fish and other seafoodmightbe the main food types that have significant contribution toTEQ. For example, finfish and shellfish contributed mostly intwo Finnish studies (> 80%) (Kiviranta et al. 2001, 2004), aHong Kong study (62%) (Wong et al. 2013), a Spanish study(78%) (Bascompta et al. 2002), and a Japanese study (50%)(Tsutsumi et al. 2001). Studies identified that fish and fishproducts were the main contributors to TEQ in children andadults in Italy, Belgium, France, and Norway (Cimenci et al.2013; Fattore et al. 2008; Sirot et al. 2012; Kvalem et al.2009). Reconciling the above facts, we can say that our expo-sure assessment is relevant to examine the potential healthhazard of dietary DL-PCBs from seafood consumption forthe Bangladeshi coastal residents. Most importantly, as a mat-ter of concern, results of this study suggested that theBangladeshi coastal people are exposed to severe health risksfrom dietary PCBs via seafood consumption from bothdioxin-like and non-dioxin like toxicological point of view.
Conclusion
This study resolved the full PCB profiles in several species ofmost frequently consumed seafood from the coastal area of
Bangladesh. Although PCB uses are strictly restricted at pres-ent, the results indicated that there still existed a variety ofPCBs in seafood, and is presumably due to continuing and/or historical usage. The total PCB levels were at the upper ormiddle range compared to that found globally. No significantseasonal variation was reported in the levels of ∑PCBs.However, the levels and composition profiles of PCBs in sea-food varied depending on the location and the species. Theseafood from the Chittagong, Cox’s Bazar, and Sundarbansareas showed higher levels of PCBs than that from theMeghna Estuary. Among the seafood species, Ilish (Hilsashad) in the finfish group and crab (mud crab) in the shellfishgroup showed the highest concentrations of total PCBs, bothof which were collected from Chittagong. The homolog andcongener profile data indicated that the prominent sources ofPCBs in the Bangladeshi coastal areas were derived as relatedto PCB technical mixtures and combustion. According to theresult from the dietary exposure assessment, the dietary in-takes of total PCBs along with dioxin-like PCBs via seafoodconsumption by the adults and children of the Bangladeshicoastal area exceeded the WHO proposed health-based guide-lines, suggesting a significant health risk in terms of toxico-logical effects by both the dioxin-like and non-dioxin-likePCBs. Since this study reports data from a preliminary healthrisk assessment, further investigations are recommendedwhere larger numbers of species, and consumers’ age, gender,occupation, and eating habit need to be taken into consider-ation. Apparently, this study serves as a snapshot for the pres-ent safety of the Bangladeshi seafood and the status of envi-ronmental health of the coastal areas of Bangladesh.
Acknowledgments This study was supported by the FY2016 AsiaFocused Academic Research Grant from the Heiwa NakajimaFoundation (http://hnf.jp/josei/ichiran/2016ichiran.pdf). The authors arealso grateful for financial support for Dr. Md. Habibullah-Al-Mamunfrom the Research Collaboration Promotion Fund provided by GraduateSchool of Environment and Information Sciences, Yokohama NationalUniversity, Japan (Grant No. 65A0516). Furthermore, we are thankful forthe kind help from the members of Dhaka University, Bangladesh, duringthe field sampling.
References
Ahmed MN, Sinha SN, Vemula SR, Sivaperumal P, Vasudev K, Ashu S,Mendu VV, Bhatnagar V (2016) Accumulation of polychlorinatedbiphenyls in fish and assessment of dietary exposure: a study inHyderabad City, India. Environ Monit Assess 188(2):94
Azari MR, Falaki F, Nejad MRM (2007) Assessment of dioxin-like com-pounds released from Iranian industries and municipalities. Tanaffos6:59–64
Ballschmiter K, Zell M (1980) Analysis of polychlorinated biphenyls(PCB) by glass capillary gas chromatography. Composition of tech-nical Aroclor- and Clophen-PCB mixtures. Fresenius' Z Anal Chem302:20–31
Bascompta O, Montana MJ, Marti R, Broto-Puig F, Comellas L, Diaz-Ferrero J (2002) Levels of persistent organic pollutants (PCDD/F
1366 Environ Sci Pollut Res (2019) 26:1355–1369
and dioxin-like PCB) in food from the Mediterranean diet.Organohalogen Compd 57:149–151
Batang ZB, Alikunhi N, Gochfeld M, Burger J, Al-Jahdali R, Al-JahdaliH, Aziz MA, Al-Jebreen D, Al-Suwailem A (2016) Congener-specific levels and patterns of polychlorinated biphenyls in ediblefish tissue from the central Red Sea coast of Saudi Arabia. Sci TotalEnviron 572:915–925
Bernes C (1998) Persistent organic pollutants: a Swedish view of aninternational problem. Swedish Environmental Protection Agency,Stockholm
Bhavsar SP, Fletcher R, Hayton A, Reiner EJ, Jackson DA (2007)Composition of dioxin-like PCBs in fish: an application for riskassessment. Environ Sci Technol 41:3096–3102
BOBLME (2011) Country report on pollution in the Bay of Bengal.Support to sustainable management of the bay of bengal large ma-rine ecosystem (BOBLME) project-Bangladesh, BOBLMEEcology-01. Accessed 1 June 2016
Bright DA, Grundy SL, Reimer KJ (1995) Differential bioaccumulationof non-ortho-substituted and other PCB congeners in coastal Arcticinvertebrates and fish. Environ Sci Technol 29:2504–2512
Chan HM, Chan KM, Dickman M (1999) Organochlorines in HongKong fish. Mar Pollut Bull 39(1–12):346–351
Cimenci O, Vandevijvere S, Goscinny S, Van Den Bergh MA, Hanot V,Vinkx C, Bolle F, Van Loco J (2013) Dietary exposure of theBelgian adult population to non-dioxin-like PCBs. Food ChemToxicol 59:670–679
Das B, Das P (2004) Organochlorine pesticide residues in water, sedi-ment, and muscle of river shad, Hilsa ilisha (Hamilton 1822) fromthe south patches of the bay of Bengal. Bull Environ ContamToxicol 72:496–503
de Brito APX, Takahashi S, Ueno D, Iwata H, Tanabe S, Kubudera T(2002) Organochlorine and butyltin residues in deep-sea organismscollected from the western North Pacific, off-Tohoku, Japan. MarPollut Bull 45:348–361
DoE (2007) Bangladesh National Implementation Plan (NIP) forManagement of Persistent Organic Pollutants (POPs). Departmentof Environment (DoE), Ministry of Environment and Forests,Government of the People’s Republic of Bangladesh, BGD/02/G31/1G/99. http://chm.pops.int/Portals/0/download.aspx?d=UNEP-POPS-NIP-Bangladesh-1.English.pdf. Accessed 1June 2016
Domingo JL, Bocio A (2007) Levels of PCDD/PCDFs and PCBs inedible marine species and human intake: a literature review.Environ Int 33:397–405
ESDO (2005a) Country situation report on persistent organic pollutants inBangladesh. Environment and Social Development Organization(ESDO), Dhaka http://ipen.org/sites/default/files/documents/1bgd_bangladesh_country_situation_report-en.pdf. Accessed 1 June 2016
ESDO (2005b) POPs hotspots in Bangladesh. Environment and SocialDevelopment Organization (ESDO), Dhaka http://www.ipen.org/project-reports/pops-hotspots-bangladesh. Accessed 1 June 2016
ESDO (2010) Study on E-waste: The Bangladesh situation. Environmentand Social Development Organization (ESDO), Dhaka http://www.e sdo . o rg / image s /S t udy%20on%20Ewas t e%20The%20Bangladesh%20Situation%202010.pdf. Accessed 1 June 2016
European Commission (2011) Commission regulation (EU) no1259/2011 of 2 December 2011 amending regulation (EC) no.1881/2006 as regards maximum levels for dioxins, dioxin-likePCBs and non-dioxin-like PCBs in foodstuffs. Off J Eur Union320:18–23
Everaarts JM, den Besten PJ, Hillebrand MTJ, Halbrook RS, Shugart LR(1998) DNA strand breaks, cytochrome P-450-dependentmonooxygenase system activity and levels of chlorinated biphenylcongeners in the pyloric caeca of the seastar (Asterias rubens) fromthe North Sea. Ecotoxicology 7:69–79
Fattore E, Fanelli R, Dellatte E, Turrini A, di Domenico A (2008)Assessment of the dietary exposure to non-dioxin-like PCBs of theItalian general population. Chemosphere 73:278–283
Focant JF, Eppe G, Pirard C, Massart AC, André JE, De Pawe E (2002)Levels and congener distributions of PCDDs, PCDFs and non-orthoPCBs in Belgian foodstuffs: assessment of dietary intake.Chemosphere 48:167–179
Fowler SW (1990) Critical review of selected heavy metal and chlorinat-ed hydrocarbon concentrations in the marine environment. MarEnviron Res 29:1–64
Goerke H, Weber K (2001) Species-specific elimination ofpolychlorinated biphenyls in estuarine animals and its impact onresidue patterns. Mar Environ Res 51:131–149
Gouin T, Mackay D, Jones KC, Harner T, Meijer SN (2004) Evidence forthe “grasshopper” effect and fractionation during long-range atmo-spheric transport of organic contaminants. Environ Pollut 128:139–148
Han L, Matarrita J, Sapozhnikova Y, Lehotay SJ (2016) Evaluation of arecent product to remove lipids and other matrix co-extractives inthe analysis of pesticide residues and environmental contaminants infoods. J Chromatogr A 1449:17–29
Hansen LG (1998) Stepping backward to improve assessment of PCBcongener toxicities. Environ Health Perspect 106:171–189
Howell NL, Suarez MP, Rifai HS, Koenig L (2008) Concentrations ofpolychlorinated biphenyls (PCBs) in water, sediment, and aquaticbiota in the Houston Ship Channel, Texas. Chemosphere 70:593–606
Ip CCM, Li XD, Zhang G, Wong CSC, Zhang WL (2005) Heavy metaland Pb isotopic compositions of aquatic organisms in the Pearl Riverestuary, South China. Environ Pollut 138:494–504
Ishikawa Y, Noma Y, Mori Y, Sakai S (2007) Congener profiles of PCBand a proposed new set of indicator congeners. Chemosphere 67:1838–1851
Jaikanlaya C, Settachan D, Denison MS, Ruchirawat M, van den Berg M(2009) PCBs contamination in seafood species at the eastern coast ofThailand. Chemosphere 76(2):239–249
Jones KC (1988) Determination of polychlorinated biphenyls in humanfoodstuffs and tissues: suggestions for a congener analytical ap-proach. Sci Total Environ 68:141–159
Jonsson A, Gustafsson O, Axelman J, Sundberg H (2003) Global ac-counting of PCBs in the continental shelf sediments. Environ SciTechnol 37:245–255
Kim K-S, Hirai Y, Kato M, Urano K, Masunaga S (2004) Detailed PCBcongener patterns in incinerator flue gas and commercial PCB for-mulations (Kanechlor). Chemosphere 55:539–553
Kiviranta H, Hallikainen A, Ovaskainen ML, Kumpulainen J, VartiainenT (2001) Dietary intakes of polychlorinated dibenzo-p-dioxins, di-benzofurans and polychlorinated biphenyls in Finland. Food AdditContam 18:945–953
Kiviranta H, Ovaskainen MAL, Vartiainen T (2004) Market basket studyon dietary intake of PCDD/Fs, PCBs, and PBDEs in Finland.Environ Int 30:923–932
Kvalem HE, Knutsen HK, Thomsen C, Haugen M, Stigum H,Brantsaeter AL, Froshaug M, Lohmann N, Papke O, Becher G,Alexander J, Meltzer HM (2009) Role of dietary patterns for dioxinand PCB exposure. Mol Nutr Food Res 53:1438–1451
Langer P, Kocan A, Tajtakova M, Petrik J, Chovancova J, Drobna B et al(2003) Possible effects of polychlorinated biphenyls andorganochlorinated pesticides on the thyroid after long-term exposureto heavy environmental pollution. J Occup Environ Med 45:526–532
Larsson P, Hamrin S, Okla L (1991) Factors determining the uptake ofpersistent pollutants in an eel population (Anguila anguilla L.).Environ Pollut 69(1):39–50
Lee KT, Lee JH, Lee JS, Park KH, Kim SK, ShimWJ, Hong SH, ImUH,Giesy JP, Oh JR (2007) Human exposure to dioxin-like compounds
Environ Sci Pollut Res (2019) 26:1355–1369 1367
in fish and shellfish consumed in South Korea. Hum Ecol RiskAssess 13:223–235
Matsumura C, Tsurukawa M, Nakano T, Ezaki T, Ohashi M (2002)Elution orders of all 209 PCBs congeners on capillary columnHT8-PCB. J Environ Chem 12(4):855–865
McFarland VA, Clarke JU (1989) Environmental occurrence, abundance,and potential toxicity of polychlorinated biphenyl congeners: con-siderations for a congener-specific analysis. EnvironHealth Perspect81:225–239
McKim JM,Heath EM (1983) Dose determinations for waterborne 2,5,2′,5′-[14C] tetrachlorobiphenyl and related pharmacokinetics in twospecies of trout (Salmo gairdneri and Salvelinus fontinalis): amass-balance approach. Toxicol Appl Pharmacol 68:177–187
Mezzetta S, Cirlini M, Ceron P, Tecleanu A, Caligiani A, Palla G,Sansebastiano GE (2011) Concentration of DL-PCBs in fish frommarket of Parma city (north Italy): estimated human intake.Chemosphere 82(9):1293–1300
Miao XS, Swenson C, Woodward LA, Li QX (2000) Distribution ofpolychlorinated biphenyls in marine species from French FrigateShoals, North Pacific Ocean. Sci Total Environ 257(1):17–28
Mohebbi-Nozar SL, Ismail WR, Zakaria MP, Mortazavi MS, Zahed MA,Jahanlu A (2014) Health risk of PCBs and DDTs in seafood fromsouthern Iran. Hum Ecol Risk Assess 20(5):1164–1176
Moon HB, Choi HG (2009) Human exposure to PCDDs, PCDFs anddioxin-like PCBs associated with seafood consumption in Koreafrom 2005 to 2007. Environ Int 35:279–284
Moon HB, KimHS, Choi M, Yu J, Choi HG (2009) Human health risk ofpolychlorinated biphenyls and organochlorine pesticides resultingfrom seafood consumption in South Korea, 2005-2007. FoodChem Toxicol 47(8):1819–1825
Naso B, Perrone D, Ferrante MC, Bilancione M, Lucisano A (2005)Persistent organic pollutants in edible marine species from theGulf of Naples, Southern Italy. Sci Total Environ 343:83–95
Nøst TH, Halse AK, Randall S, Borgen AR, Schlabach M, Paul A,Rahman A, Breivik K (2015) High concentrations of organic con-taminants in air from ship breaking activities in Chittagong,Bangladesh. Environ Sci Technol 49(19):11372–11380
Oh CW, Hartnoll RG, Nash RDM (2001) Feeding ecology of the com-mon shrimp Crangon crangon in port Erin Bay, Isle of Man, IrishSea. Mar Ecol Prog Ser 214:211–223
Pan H, Geng J, Qin Y, Tou F, Zhou J, Liu M, Yang Y (2016) PCBs andOCPs in fish along coastal fisheries in China: distribution and healthrisk assessment. Mar Pollut Bull 111(1–2):483–487
Perello G, Gómez-Catalan J, Castell V, Llobet JM, Domingo JL (2012)Assessment of the temporal trend of the dietary exposure toPCDD/Fs and PCBs in Catalonia, over Spain: health risks. FoodChem Toxicol 50(2):399–408
Porte C, Albaigés J (1993) Bioaccumulation patterns of hydrocarbons andpolychlorinated biphenyls in bivalves, crustaceans, and fishes. ArchEnviron Contam Toxicol 26:273–281
Raknuzzaman M, Ahmed MK, Islam MS, Habibullah-Al-Mamun M,Tokumura M, Sekine M, Masunaga S (2016) Trace metal contami-nation in commercial fish and crustaceans collected from coastalarea of Bangladesh and health risk assessment. Environ Sci PollutRes 23(17):17298–17310
Sahu SK, Ajmal PY, Pandit GG, Puranik VD (2009) Vertical distributionof polychlorinated biphenyl congeners in sediment core from ThaneCreek area of Mumbai, India. J Hazard Mater 164(2–3):1573–1579
Shaw SD, Brenner D, Berger ML, Carpenter DO, Hong C-S, Kannan K(2006) PCBs, PCDD/Fs and organochlorine pesticides in farmedAtlantic salmon from Maine, eastern Canada and Norway, and wildsalmon from Alaska. Environ Sci Technol 40:5347–5354
Shen H, Yu C, Ying Y, Zhao Y, Wu Y, Han J, Xu Q (2009) Levels andcongener profiles of PCDD/Fs, PCBs and PBDEs in seafood fromChina. Chemosphere 77(9):1206–1211
Simon T, Britt JK, James RC (2007) Development of a neurotoxic equiv-alence scheme of relative potency for assessing the risk of PCBmixtures. Regul Toxicol Pharmacol 48:148–170
Sirot V, Tard A, Venisseau A, Brosseaud A, Marchand P, Le Bizec B,Leblanc JC (2012) Dietary exposure to polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans and polychlorinated biphe-nyls of the French population: results of the second French Total dietstudy. Chemosphere 88:492–500
Smith AG, Gangolli S (2002) Organochlorine chemicals in seafood: oc-currence and health concerns. Food Chem Toxicol 40:767–779
StanchevaM, Georgieva S, Makedonski L (2017) Polychlorinated biphe-nyls in fish from Black Sea, Bulgaria. Food Control 72(B):205–210
Storelli MM, Giacominelli-Stuffler R, Storelli A, Marcotrigiano GO(2003) Polychlorinated biphenyls in seafood: contamination levelsand human dietary exposure. Food Chem 82:491–496
Storelli MM, Losada S, Marcotrigiano GO, Roosens L, Barone G, NeelsH, Covaci A (2009) Polychlorinated biphenyl and organochlorinepesticide contamination signatures in deep-sea fish from theMediterranean Sea. Environ Res 109(7):851–856
Succop PA, Clark S, Chen M, Galke W (2004) Imputation of data valuesthat are less than a detection limit. J Occup EnvironHyg 1(7):436–441
Sudaryanto A, Monirith I, Kajiwara N, Takahashi S, Hartono P,Muawanah OK, Takeoka H, Tanabe S (2007) Levels and distribu-tion of organochlorines in fish from Indonesia. Environ Int 33(6):750–758
Tölgyessy P, Miháliková Z (2016) Rapid determination of total lipids infish samples employing extraction/partitioning with acetone/ethylacetate solvent mixture and gravimetric quantification. FoodControl 60:44–49
Tsutsumi T, Yanagi T, Nakamura M, Kono Y, Uchibe H, Iida T, Hori T,Nakagawa R, Tobiishi K, Matsuda R, Sadaki K, Toyoda M (2001)Update of daily intake of PCDDs, PCDFs, and dioxin-like PCBsfrom food in Japan. Chemosphere 45:1129–1137
Ueno D, Takahashi S, Tanaka H, Subramanian AN, Fillmann G, NakataH, Lam PKS, Zheng J, Muchtar M, PrudenteM, Chung KH, TanabeS (2003) Global pollution monitoring of PCBs and organochlorinepesticides using skipjack tuna as a bioindicator. Arch EnvironContam Toxicol 45:378–389
UNEP (2001) Stockholm Convention on Persistent Organic Pollutants,United Nations Environment Programme UNEP/POPs/CONF/4
USEPA (2010) National recommended water quality criteria, UnitedStates Environmental Protection Agency. Available at 03/12/2016to http://water.epa.gov/scitech/swguidance/standards/criteria/current/index.cfm. Accessed 2 June 2016
Van Ael E, Covaci A, Blust R, Bervoets L (2012) Persistent organicpollutants in the Scheldt estuary: environmental distribution andbioaccumulation. Environ Int 48:17–27
Van den Berg M, Birnbaum LS, Denison M, De Vito M, Farland W,Feeley F, Fiedler H, Hakansson H, Hanberg A, Haws L, Rose M,Safe S, Schrenk D, Tohyama C, Tritscher A, Tuomisto J, TysklindM, Walker N, Peterson RE (2006) The 2005 World HealthOrganization reevaluation of human andmammalian toxic equiva-lency factors for dioxins and dioxin-like compounds. Toxicol Sci 93:223–241
Voorspoels S, Covaci A, Maervoet J, De Meester I, Schepens P (2004)Levels and profiles of PCBs and OCPs in marine benthic speciesfrom the Belgian North Sea and the Western Scheldt estuary. MarPollut Bull 49:393–404
WHO (1998) Assessment of the health risk of dioxins: re-evaluation ofthe tolerable daily intake (TDI). World Health Organization, Genevahttp://www.who.int/ipcs/publications/en/exe-sum-final.pdf.Accessed 6 June 2016
WHO (2003) Concise international chemical assessment document 55.Polychlorinated biphenyls: human health aspects. World HealthOrganization, Geneva www.who.int/ipcs/publications/cicad/en/cicad55.pdf. Accessed 6 June 2016
1368 Environ Sci Pollut Res (2019) 26:1355–1369
Wong WW, Yip YC, Choi KK, Ho YY, Xiao Y (2013) Dietary exposureto dioxins and dioxin-like PCBs of Hong Kong adults: results of thefirst Hong Kong total diet study. Food Addit Contam Part A ChemAnal Control Expo Risk Assess 30:2152–2158
Xia C, Lam JC, Wu X, Xie Z, Lam PK (2012) Polychlorinated biphenyls(PCBs) in marine fishes from China: levels, distribution and riskassessment. Chemosphere 89(8):944–949
Yang N, Matsuda M, Kawano M, Wakimoto T (2006) PCBs and organ-ochlorine pesticides (OCPs) in edible fish and shellfish from China.Chemosphere 63:1342–1352
Yang HY, Xue B, Jin LX, Zhou SS, Liu WP (2011) Polychlorinatedbiphenyls in surface sediments of Yueqing Bay, Xiangshan Bay,and Sanmen Bay in East China Sea. Chemosphere 83:137–143
Environ Sci Pollut Res (2019) 26:1355–1369 1369