Polychlorinated biphenyls and polychlorinated dibenzo-p-dioxins and furans in imported canned fish in Nigeria and risk assessment

ABSTRACT Concentrations of polychlorinated biphenyls (PCBs) and polychlorinated dibenzo-p-dioxins and furans (PCDD/Fs) were measured in imported canned fish such as mackerel, sardine and tuna to evaluate the risk relating to human consumption of these products. Gas chromatography-mass spectrometry (GC/MS) was used to evaluate the concentrations of PCBs and PCDD/Fs in the samples. The 28 PCB concentrations in the canned mackerel, tuna and sardine ranged from 0.33 to 9.48 ng g −1, <LOQ to 8.8 ng g−1 and <LOQ to 15.1 ng g−1, respectively, while the 14 PCDD/Fs concentrations varied from 0.06 to 4.70 ng g−1 for mackerel, 0.72 to 9.43 ng g−1 for tuna and not detected to 22.0 ng g −1 for sardines. Health risk analysis suggests that ingestion of these samples could lead to adverse non-carcinogenic and carcinogenic risks over a lifetime.


Introduction
Polychlorinated biphenyls (PCBs) and polychlorinated dibenzo-p-dioxins and furans (PCDD/Fs) are ubiquitous halogenated persistent organic pollutants (POPs) that have attracted much apprehension due to their toxicity and adverse health effects (Leong et al. 2014b). Human exposure to these compounds can lead to carcinogenesis, mutations, and impairment of the endocrine, immune and reproductive systems. PCBs and PCDD/Fs are among the compounds catalogued as the "dirty dozen" POPs in the Stockholm Convention of 22 May 2001 (Lewandowski et al. 2014). PCBs have been intentionally produced in the past for diverse industrial applications such as their use in power capacitors, electrical transformers, coolants, carbonless copy paper, lubricants, paints, and plasticisers, among others (Ahmed et al. 2016;Babalola et al. 2017;Aganbi et al. 2019;Iwegbue et al. 2019Iwegbue et al. , 2020Iwegbue et al. , 2022Irerhievwie et al. 2020;Aziza et al. 2021). Unlike PCBs, PCDD/Fs are inadvertent products of anthropogenic activities such as chemical production, manufacture of pesticides (e.g. pentachlorophenol), emissions from incinerators and vehicle exhausts, and disposal of sewage sludge (Cai et al. 2016). In total, 12 of the 209 congeners of PCBs are referred to as dioxin-like PCBs [dl-PCBs] because of their ability to rotate and take up the coplanar structure of PCDDs, and that they possesses similar chemical and toxicological characteristics as 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD) (Baars et al. 2004;Cai et al. 2016). Similarly, only 17 of the 210 PCDD/Fs exhibit significant toxicological effects and have been selected as "priority pollutants" that require regular monitoring in the environment (WHO 2006).
Aquatic organisms such as fish can accumulate PCBs and PCDD/Fs in their tissues several-fold higher than the surrounding water and sediments, and humans can be exposed to these substances through consumption of these aquatic foods. Apart from occupational exposures, the consumption of contaminated fish and fish products constitutes one of the major sourses of human exposure to these compounds (Drábová et al. 2011;Leong et al. 2014a;Leong and Majid 2017). In addition, food ingestion was projected to account for more than 95% of the total intake of these compounds by non-occupationally exposed people (Leong et al. 2014b). Fish and fish products are essential items of the human diet as they provide good sources of protein, amino acids, omega-3 fatty acids, micro-and macronutrients, and vitamins (Miklavčič et al. 2011). Fish and fish products decrease the death rate for coronary diseases, and the risk of stroke from blood clotting and inflammation (Iwegbue et al. 2015a). They also prevent different types of ailments such as circulatory disorders, cardiac arrhythmias, and autoimmune diseases (Storelli 2008).
Fishes caught for commercial purposes are canned, preserved, or processed into other products thereby making them available for consumers far away from the marine environment (Iwegbue et al. 2009(Iwegbue et al. , 2015bTesi and Iniaghe 2020). Canned fish is a major component of the global fish trade, and can be a possible route for contaminant distribution and exposure, as opposed to the well-known long-range atmospheric transport or through ocean currents and associated contamination of the local/regional food web. In Nigeria, approximately 60% of the fish consumed are imported. Presently, PCBs and PCDD/Fs are not on the list of contaminants routinely checked in imported foods by the Nigerian regulatory authority. Since canned fish in the Nigerian market are almost exclusively imported, there is a need for continuously monitoring the occurrence and concentrations of PCBs and PCDD/Fs in these canned fish to safeguard consumers from exposure to these compounds.
There are a number of studies on the PCB and PCDD/F concentrations in canned fish (Kipčić and Vukušić 1991;Usydus et al. 2008;Miklavčič et al., 2011;El Morsy et al. 2013;Perelló et al. 2015;Squadrone et al. 2016;Saktrakulkla et al. 2020). However, there are only few reports on the concentrations of PCBs in imported fish and fish products (Babalola et al. 2017;Tesi and Iniaghe 2020;Wangboje and Okotie 2021). These studies were limited to PCB concentrations in one fish type, and provided no information on PCDD/Fs. Thus, this study evaluates the concentrations and risk resulting from exposure to PCBs and PCDD/Fs by human consumption of three predominant types of canned fish (mackerel, sardine and tuna) in the Nigerian market. This study provides information necessary to evaluate the temporal trends following the global ban on the production, trade and applications of PCBs. In addition, it provides relevant facts necessary for food quality management and consumers to make informed decisions about food preferences to reduce risk.

Samples and sample collection
The canned fish samples investigated consisted of 18 brands of canned sardines, and six each of mackerel and tuna. They were collected from retail shops in Delta State, Nigeria. For each brand, a composite sample was obtained by mixing 5-7 samples with diverse dates of manufacture and batch numbers. The choice of brands analysed was based on market popularity and price ranges. Before analysis, the oil in these canned fish was drained by making use of a stainless steel sieve. Facts about the origin, and net and drained weights of each sample analysed, are given in Supplementary Material Table S1.

Chemicals and reagents
Acetone, dichloromethane and n-hexane, all of chromatographic grade, were obtained from Sigma-Aldrich (St. Louis, MO, USA). High purity grade silica gel and anhydrous sodium sulphate (Na 2 SO 4 ), ACS reagent, ≥ 99.0%, and alumina were products of Merck (Darmstadt, Germany). Mixed standard solutions containing 28 PCBs and 14 PCDD/Fs were products of AccuStandard Inc. (New Haven, CT, USA), while mixtures of deuterated standards containing six 13 C 12 -PCBs and eight 13 C 12 -PCDD/Fs were products of Cambridge Isotope Laboratories Inc. (Tewksbury, MA, USA).

Extraction and analysis
PCBs and PCDD/Fs were extracted from the samples following the US EPA method 3540C (US EPA, 1996). A 5.0 g mass of the composite sample of the fish was homogenised with an equal amount (5.0 g) of Na 2 SO 4 and spiked with surrogate standards (200 ng) of 13 C 12 -labelled PCBs and PCDD/Fs and subjected to Soxhlet extraction with 150 mL of a 1:1:1 v/v mixture of acetone/ DCM/n-hexane at 65°C for 18 h. The extract was concentrated to 2 mL by rotary evaporation and purified on a column containing 2.0 g each of anhydrous Na 2 SO 4 , silica gel and alumina. The PCBs and PCDD/Fs were eluted from the column with a 40 mL aliquot of 3:1 (v/v) mixture of n-hexane/DCM. The eluent was evaporated to ca. 2 mL under a gentle flow of nitrogen gas and 2 µL of the extract was injected in splitless mode into the gas chromatograph (Agilent 7890A) at an injector port temperature of 250°C, that was interfaced with a 5975C mass selective detector (MSD), all from Agilent Technologies (Palo Alto, CA, USA). The MSD was operated in selected ion monitoring mode (SIM) with an electron impact energy of 35 eV. The separation of these compounds was accomplished on a Rtx-5 MS column (60 m × 0.25 µm × 0.25 mm; Restek Corporation, USA). For PCB analysis, the oven temperature of the GC was programmed as follows: initial temperature was held at 120°C for 2 min, raised to 180°C at 20°C min −1 and finally to 270°C at 2°C min −1 . Similarly, for PCDD/ Fs, the initial temperature was fixed at 130°C for 2 min, increased to 200°C at 30°C min −1 and further to 235°C at 3°C min −1 and finally to 300°C at 6°C min −1 . The carrier gas was helium (99.999% purity) with a flow velocity of 1.5 mL min −1 .

Quality control and statistics
Several quality assurance/control procedures including method banks, recoveries of 13 C-labelled PCBs and PCDD/Fs, and spiked sample matrices were adopted to evaluate the performance of the analytical method. The recovery of the 13 C 12 -labelled PCBs and PCDD/Fs varied from 80.5 to 96.8% and from 79.3 to 98.9%, respectively. In addition, the sample matrix was spiked with known standards of the target compounds and reanalysed following the same steps. The recoveries for PCBs and PCDD/Fs from the spiked sample matrix ranged from 78.9 to 98.7% and 85.2 to 96.3%, respectively. All samples were analysed in triplicate, with precision ranging from 4 to 10% relative standard deviation (RSD). An external calibration method was employed to quantify the PCB and PCDD/ F concentrations in the samples. The seven-points calibration curves for PCBs and PCDD/Fs yielded linear regression coefficient (r 2 ) values of 0.9991 to 0.9998. The limits of detection (LOD) and quantification (LOD) were measured as 3 and 10 times the signal to noise ratio of the blank, respectively. The LODs and LOQs of the PCBs varied from 0.01 to 0.03 ng g −1 and 0.03 to 0.09 ng g −1 , respectively, while LODs and LOQs of the PCDD/Fs varied from 0.01 to 0.02 and from 0.03 to 0.06 ng g −1 , respectively. In this study, 14 of the 17 priority congeners of the PCDD/Fs were investigated. The exclusion of OCDD, 2,3,4,6,7,8 HxCDF and OCDF is related to their omission in the supplier's standard mixture used for calibration. The differences in the PCB and PCDD/F concentrations within each group were evaluated by using the Kruskal-Wallis test, while the normality of the results was evaluated by using the Shapiro-Wilk test, and the Duncan multiple range test was used for comparison of the means of the different groups. The statistical tests at p < 0.05 were accomplished with SPSS version 24.0 software (SPSS Inc., Chicago, IL, USA).

Evaluation of toxic equivalency concentrations
TEQ concentrations of dl-PCBs and PCDD/Fs in these samples were calculated from TEQ ¼ P  Table S2.

Estimation of daily intake and risk
The daily intake (DI) of PCBs and PCDD/Fs relating to the ingestion of these brands of canned fish was calculated from DI (pg kg −1 bw day −1 ) = (C × IngR/bw). The risk arising from lifetime ingestion of PCBs and PCDD/ Fs in the canned fish was evaluated by calculating the chronic daily intake (CDI) from CDI ing = [(C × IngR × EF × ED)/(bw × AT nc )] × 10 −6 , where C represents the PCB and PCDD/Fs concentrations in the canned fish; bw depicts the average body weight (80 kg for adults and 15 kg for children); IngR depicts the ingestion rate of 20.8 g, derived from the average per capita value of 7.6 kg for fish consumption in Nigeria (Iwegbue et al. 2015a(Iwegbue et al. , 2015b; EF depicts the exposure frequency (day/yr; 365 days); ED represents the exposure duration of 6 years for children and 20 years for adults and AT nc depicts the averaging time for non-carcinogenic effect = ED × 365. The Hazard Index HQ ing = CD Ing /RfDo, where RfDo is the oral reference dose (mg kg −1 day −1 ). Finally, the total cancer risk = (C × IngR × EF × ED × CF × SFO)/(bw × AT ca ), where SFO represents the oral slope factor (mg kg −1 day −1 ); AT ca represents the averaging time for carcinogens = LT × 365 (US DOE (United States Department of Energy) 2011) and LT is the lifetime in years (Environment Agency 2009). In this study, a 55 year' lifetime was assumed based on the average life expectancy for Nigeria (WHO 2018). RfDo and SFO values for PCBs and PCDD/Fs are given in Supplementary Material Table S2.

Analytical data and occurrence pattern
The summary statistics of PCB and PCDD/F concentrations in the canned fish are presented in Tables 1 and 2, respectively, while the detailed individual results are shown in Supplementary Material Tables S3 and S4. The PCB and PCDD/F concentrations in these canned fish showed significant discrepancies (p <.05) with respect to brands and fish types. The differences in the PCB and PCDD/F concentrations in these canned fish may be related to the bioaccumulation potentials of the individual fish species, fish diet, oil and tomatoe based sauces used, canning procedures and lipid content (Leong et al. 2014a;Tesi and Iniaghe 2020). The P 28 PCB concentrations in the canned fish varied from 0.33 to 9.48, <LOQ to 8.8, and <LOQ to 15.1 ng g −1 for mackerel, tuna and sardines, respectively. None of the 28 PCBs investigated was detected in TN6, SD10, SD11, and SD13. The highest concentrations of P 28 PCBs were observed in MC5, TN1 and SD5 for mackerel, tuna, and sardines, respectively. The P 14 PCDD/F concentrations in the canned mackerel, tuna and sardines varied from 0.06 to 4.70, 0.72 to 9.43, and <LOQ to 22.0 ng g −1 , respectively. The P 14 PCDD/F concentrations in two samples of canned sardines (SD1 and SD3) were below the LOQ, while the     highest concentrations of P 14 PCDD/Fs in these canned fish types were found in MC3, TN5 and SD8 for mackerel, tuna and sardines, respectively. The mean distribution of P 28 PCBs in these canned fish types was in the sequence: mackerel > sardines > tuna, which is in the reverse order of that for PCDD/Fs (tuna > sardines > mackerel). The specified limit of PCBs in fish is 2000 ng g −1 (US FDA 2018). Despite discrepancies in the number of congeners and analytical approaches, the P 28 PCB concentrations in these canned fish did not exceed the US FDA limit. The P 28 PCB concentrations in our samples were lower than those of previous reports on canned fish in Nigeria (<LOD to 668 ng g −1 ; Tesi and Iniaghe 2020), Egypt (8.56 to 208 ng g −1 : El Morsy et al. 2013), Central Adriatic (5.0 to 4175 ng g −1 ; Kipčić and Vukušić 1991), and those of European cat fish from northern Italy (Squadrone et al. 2016). However, the P 28 PCB concentrations in our samples are comparable with those found in canned fish from Polish (12.8 ng g −1 ; Usydus et al. 2008), Catalonian (Spain) (0.21 to 18.57 ng g −1 ; Perelló et al. 2015) and Slovenian markets (<LOD to 12.6 ng g −1 ; Miklavčič et al. 2011), but were lower than those reported in canned tuna from the USA (Saktrakulkla et al. 2020). In contrast, PCDD/F concentrations in these samples surpassed those reported for canned fish from Catalonian markets (Perelló et al. 2015).

Composition of PCBs and PCDD/Fs
The PCB homologue composition of these canned fish is displayed in Figure 1. The mean occurrence pattern followed the order: hexa > penta > octa > hepta > tri > deca > di > tetra > nona-PCBs for mackerel; tetra > penta > hexa > hepta > tri > octa > di = deca-PCBs for tuna; and penta > deca > tetra > hexa > hepta > di > tri = octa > nona-PCBs for sardines. None of the nona-PCBs investigated were found in the tuna samples. The dioxin-like PCBs (dl-PCBs) were detected at concentrations between 0.11 and 6.01 ng g −1 in 63% of these canned fish. The mean concentrations of dl-PCBs in these canned fish followed the sequence of tuna (1.72 ng g −1 ) > mackerel (1.20 ng g −1 ) > sardine (0.59 ng g −1 ). The non-ortho dl-PCB concentrations ranged between 0.10 and 4.73 ng g −1 , whereas those of mono-ortho dl-PCBs varied between 0.01 and 3.77 ng g −1 . On average, the mono-ortho dl-PCBs were the dominant dl-PCBs in these canned fish. The indicator PCBs (i-PCBs) were detected in 70% of the canned fish analysed in a concentration range of 0.17 to 7.17 ng g −1 . The i-PCB concentrations in these canned fish were below the specified limit of 75 ng g −1 (European Commission (EC) 2011). The low chlorinated PCBs (LC-PCBs) were found in 80% of these canned fish in a concentration range of 0.09 to 7.83 ng g −1 , while high chlorinated PCBs (HC-PCBs), from hexa to deca-PCBs, were found in 77% of these canned fish ranging from 0.24 to 9.05 ng g −1 . On average, the LC-PCB concentrations were higher than those of HC-PCBs in these canned fish except for mackerel. The prevalence of LC-PCBs in these canned fish may be related to the fact that they are less hydrophobic with appreciable water solubility and, therefore, more bioavailable than HC-PCBs (Assefa et al. 2018;Mikolajczyk et al. 2022). Figure 2 illustrates the compositions of the PCDD/Fs in these canned fish. The ∑PCDD concentrations varied from 0.01 to 4.09 ng g −1 and captured 4.8-100% of the ∑14 PCDD/Fs, while those of ∑PCDFs varied from 0.09 to 22 ng g −1 and represented 15.9-100% of the ∑14 PCDD/Fs. Nevertheless, the mean concentrations of ∑PCDFs exceeded those of ∑PCDDs in the tuna and sardines. The detection frequency of the individual PCDDs ranged between 33 and 50% with 1,2,3,7,8,2,3,4,6,7, MC1  MC2  MC3  MC4  MC5  MC6  TN1  TN2  TN3  TN4  TN5  SD1  SD2  SD3  SD4  SD5  SD6  SD7  SD8  SD9  SD12  SD14  SD15  SD16  SD17  SD18   MACKEREL  TUNA  samples, the individual concentrations of the PCDD compounds were less than 5.0 ng g −1 . The concentrations of hexa-congeners of dioxins were higher than those of tri-and hepta-congeners in these samples. The detection frequency of the individual PCDFs in these canned fish ranged between 27 and 43% with 1,2, 3,7,1,2,3,7,8,and 1,2,3,4,6,7,8-HpCDF having the highest detection frequency. The concentrations of the individual furans in these samples were less than 19.0 ng g −1 . On average, the concentration of HpCDFs was higher than those of TCDFs, PCDFs and HxCDFs in the mackerels and sardines. Conversely, the PCDFs showed higher mean concentrations than those of TCDFs, HxCDFs and HpCDFs in the canned tuna.
The hazard index (HI) of dl-PCBs and PCDD/Fs relating to the ingestion of these brands of canned fish by adults and children are presented in Supplementary Material Tables S9 and S10, respectively. The HI values of dl-PCBs relating to children's consumption of these canned fish varied from 0.01 to 364, 0.14 to 19.0, and 0.14 to 108 for mackerel, tuna, and sardines, respectively, while those of the adult counterparts ranged from 0.01 to 4.37, 0.03 to 3.56, and 0.03 to 20.2 for mackerel, tuna, and sardines, respectively. The HI values of dl-PCBs relating to both adults and children's consumption of 30% of these canned fish (MC1, MC2, MC5, TN3, TN5, SD5, SD14, SD16, SD17, and SD18) exceeded 1, indicating adverse non-carcinogenic risks from consumption of these brands of canned fish. Similarly, the HI values of PCDD/Fs relating to both adults and children's consumption of 37% of these canned fish (MC2, MC3, TN1, TN2, TN6, SD4, SD9, and SD15 to SD18) were above 1. This suggests that the intake of PCDD/Fs from the aforementioned samples could cause adverse non-carcinogenic effects for human health. PCB-126 and 2,3,7,8-TCDF are responsible for the high HI values of dl-PCBs and PCDD/Fs in these brands of canned fish, respectively.
The TCR values relating to the intake of PCBs from these canned fish by adults and children were in the order of 10 −8 to 10 −3 and 10 −6 to 10 −3 , respectively (Supplementary Material Table S11). The TCR values relating to adults and children's intake of PCBs from MC3, MC4, MC6, TN6, SD1 to SD4, and SD8 to SD13 (i.e. 53% of the samples) were less than the acceptable cancer risk value of 10 −6 . This implies no carcinogenic risk relating to the intake of PCBs from the aforementioned canned fish. However, the TCR values relating to adults and children's intake of PCBs from 47% of these canned fish exceeded the acceptable cancer risk value of 10 −6 . The TCR values resulting from the intake of PCDD/Fs through ingestion of these products by adults and children were in the order of 10 −4 to 10 −1 and 10 −5 to 10 −2 (Supplementary Material Table S12). Apart from MC6, SD1, SD3, SD5, SD8, SD11, SD13, and SD14, the TCR values resulting from both adults and children's intake of PCDD/Fs from the other brands of canned fish exceeded the safe value. This indicates a potential carcinogenic risk from lifetime ingestion of these brands. The HI and TCR values relating to the intake of PCBs and PCDD/Fs from these canned fish by children were higher than those of adults, suggesting that children were at a higher risk than adults. It is worthy to note that the use of the average per capita consumption value may lead to possible overestimation of the risk of these compounds especially for children since they consume lesser amounts of fish than adults. The HI and TCR values relating to intake of PCDD/Fs were higher than those of PCBs.

Conclusion
This study has shown that the canned fish were contaminated with PCBs and PCDD/Fs. The TEQ values of dl-PCBs and PCDD/Fs in these canned fish were above their specified limits. Similarly, the HI and TCR values for PCBs and PCDD/Fs in 47% of these canned fish exceeded acceptable safe limits, which suggest that consumption of these canned fish over a lifetime could result in adverse non-carcinogenic and carcinogenic risks to consumers. There is a need to restrain from excessive consumption of these canned products, and stringent regulatory checks on imports of these products should be instituted, to minimise the exposure of consumers to the hazardous contaminants in these canned fish.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
The author(s) reported there is no funding associated with the work featured in this article.