Detection of indicator polychlorinated biphenyls (I-PCBs) and polycyclic aromatic hydrocarbons (PAHs) in cow milk from selected areas of Dhaka, Bangladesh and potential human health risks assessment

In this study, the levels of indicator PCBs congeners and PAHs compounds were reported in cow milk from selected areas in Dhaka, Bangladesh, and the potential human health risks were assessed. A total of 100 milk samples were collected and analyzed using gas chromatography-tandem mass spectrometry (GC-MS/MS). Method validation was performed using recovery performance, linearity, limit of detection (LOD), and limit of quantification (LOQ) assays. PCBs congeners, including PCB No. 52 (2,2´,5,5´-tetrachlorobiphenyl), PCB No. 101 (2,2´,4,5,5´-pentachlorobiphenyl), PCB No. 153 (2,2,4,4,5,5–hexachlorobiphenyl), and PCB No 209 (Decachloro-1,1′-biphenyl perchlorobiphenyl) were detected, whereas PCB No 28 (2,2 ´,4–trichlorobiphenyl), PCB No. 138 (2,2´,3,4,4´,5´–hexachlorobiphenyl), and PCB No 180 (2,2´,3,4,4´,5,5´–heptachlorobiphenyl) were not detected in the analyzed milk samples. Among the 16 PAHs compounds, benzo (a) anthracene and chrysene were detected in milk samples. The Σ hazard risk index (HI) of all detected PCBs congeners was below the limit set by the European Food Safety Authority, which indicates limited health risks to animals and humans in the study area. However, the presence of PCBs and PAHs in milk samples from industrial areas may negatively affect human health, and further detailed studies are required to ensure public health safety.


Introduction
Biogenic and anthropogenic sources introduce pollutants into the environment. However, human activities are primarily responsible for their massive discharge into the atmosphere. Polychlorinated biphenyls (PCBs), hexachlorocyclohexane (HCH), polybrominated diphenyl ethers (PBDEs), polyaromatic hydrocarbons (PAHs), pesticides, and heavy metals are anthropogenic inorganic and organic pollutants that are dispersed throughout the atmosphere, hydrosphere, and lithosphere. They may be transformed into other compounds that are even more toxic to flora and fauna [1]. Because of their industrial uses, several contaminants, including PCBs and PAHs, are mixed with the environment, which is very hazardous and is retained in the environment, posing a severe threat to both public health and the ecosystem [2][3][4][5][6][7][8][9].
PCBs are long-lasting contaminants that result from the defective combustion of organic and synthetic materials [10]. PCBs are synthetic organic compounds found in most industrial and consumer items. The production of PCBs was banned in the USA in 1977 [11,12]. Because of their persistence and bioaccumulation, PCBs are categorized as persistent organic pollutants (POPs) [13]. The major concern is the presence of PCBs in consumer items such as plasticizers in paints, plastics, and rubber products. As PCBs accumulate in adipose tissue, they are typically found in animal-derived foods. PCB exposure is commonly caused by eating fish; however, other foods with lower PCB levels that are consumed more frequently, such as beef, dairy, and chicken products, also contribute to PCB exposure [11]. PCBs are causative agents of cancer, immune system diseases, and malfunctions in the reproductive, neurological, and other biological systems [14]. PCBs are widely used as pollutant markers because of their presence in biotic and abiotic conditions [15].
International Agency for Research on Cancer (IARC) and Joint Food and Agriculture Organization/World Health Organization (FAO/WHO), Scientific Committee on Food (SCF), and European Food Safety authority (EFSA).
Although the presence of PCBs and PAHs in fat-rich foods, such as milk, is a major health concern, not much work has been done so far in Bangladesh. This study aimed to determine the levels of PCBs and PAHs in cow milk from selected areas in Bangladesh and assess possible human health risks.

Area of study
A total of 100 raw milk samples from 10 sampling sites were collected from the dairy firms of Mohammadpur and Hazaribag in Dhaka. The samples were collected between October 2020 and February 2021. Hazaribag is used in the leather industry. The site selection was based on industrial and agricultural activities. Many leather and tanning industries are located in Hazaribag. Mohammadpur is adjacent to the Turag River. The water of the Turag River is contaminated by industrial waste. Moreover, several waste dumping sites are located near both sampling areas, where the wastes are burned regularly. Milk samples were collected from dairy farms and local shops. Raw milk samples were used in this analysis. About 500 mL of milk samples were collected in plastic polythene bags, transported to the laboratory on ice box and stored at 4ºC and analyzed within 24 h of sample collection.  7) and a cocktail mix of 16 PAHs compounds containing Napthalene, 2-methylnapthalene, 1-methylnapthalene, Acenapthylene, Acenapthalene, Fluorene, Phenanthrene, Anthracene, Fluranthene, Pyrene, Benzo (A) Anthacene, Chrysene, Benzo (B) Fluoranthene, Benzo (K) Fluranthene, Benzo (A) Pyrene, Indeno (1, 2, 3-CD) Pyrene and Dibenz (A, H) Anthracene brought from Sigma Aldrich (Darmstadt, Germany). Acetonitrile (Advent Chembio, India) and other chemicals including n-hexane, magnesium sulfate, sodium acetate, and NaCl were purchased from Merck (Darmstadt, Germany).

Extraction and cleanup
Sample extraction was performed using the quick, easy, cheap, effective, rugged, and safe (QuEChERS) method [39][40][41] with slight modifications for PCBs analysis. Samples were brought to normal temperature, and 5 mL of milk was transferred to a 50 mL polypropylene tube. After dilution with ultrapure water, the samples were extracted with 10 mL hexane: acetone (1:1, v/v). Anhydrous magnesium sulfate (4 g) and sodium chloride (1 g) were added to the mixture and shaken for 5 min at 175 rpm. Thereafter, the mixture was centrifuged at 3500 rpm for 5 min. Approximately 10 mL of the supernatant was poured into a glass tube, and the solution was dried using a nitrogen stream at 40ºC the solution was dried. Subsequently, the samples were cleaned, and the eluates were evaporated using a nitrogen stream at 40 • C. Finally, the extracts were dissolved in 100 µL of isooctane for GC-MS/MS analysis.
For PAHs analysis, samples were first homogenized, and 5 g of homogenized samples were placed in a conical tube and shaken for 1 min. A mixture of hexane and acetonitrile (1:1 v/v) was prepared, 15 mL of the mixture was added, and the mixture was shaken for a few seconds. Subsequently, magnesium sulfate (6 g) and sodium acetate (1.5 g) were added to the mix [42]. The solubility of the PAHs was reduced by adding NaCl to the mixture. The mixture was then centrifuged at 2500 g for 5 min. The supernatant was removed, and PAHs were isolated through solid-phase extraction (SPE). Subsequently, activated silica gel and 1 g of Na 2 SO4 were loaded onto a glass chromatographic column, and n-hexane was used. The concentrated extracts were dissolved in 5 mL of n-hexane, loaded into the column, and eluted with 50 mL of n-hexane. Finally, the eluents were concentrated and dissolved in n-hexane (0.5 mL of n-hexane for GC-MS/MS analysis [43].

Analysis of PCBs congeners and PAHs compounds in milk samples
PCBs congeners were analyzed using a gas chromatograph coupled with a MS (Thermo Scientific, USA). Analytical separations were performed using a Trace GOLD™ TG-5MS GC Column (0.25 mm × 0.25 µm X 0.25 m). Helium was used at a constant pressure (18.29 psi) at 1.6 mL/ min flow rate. The injection port temperature was 250 • C, and the initial oven temperature was 60 • C, which was further increased to 220ºC at 5ºC/min heating ramp. The detector temperature was approximately 320 • C during the sample analysis. The sample injection volume was 2 µL.
For the PAHs analysis, helium was used as the carrier gas at 1.8 mL/ min flow rate. The injection volume was 1 µL in split-less mode. The ion source and quadrupole temperatures were 280 • C and 180 • C, respectively, whereas the injector and transfer line temperatures were 280 • C and 310 • C, respectively. The column temperature was initially set at 35 • C for 1 min, then increased to 200 • C at a ramp rate of 25 • C/min, and finally ramped to 310 • C at 8 • C/min and kept for 3.5 min. Both of PCBs and PAHs analyte detection was carried out using MS at 70 eV ionization energy in selected ion monitoring (SIM) acquisition mode.

Method quality control
Working solutions were prepared by diluting standard stock solutions with n-hexane. For the linearity test, working solutions of several dilutions in the range of 50-200 µg/L were prepared and injected into the GC-MS/MS system. The LOQ and LOD were determined using the signal-to-noise ratio. Method validation was performed using a recovery performance evaluation. The recovery percentage was determined using the following equation: Here, Pi is the recovery percentage, Si is the result of the control, and Ti is the percent recovery of the spiked samples.

Health risk assessment of PCBs congeners
Human exposure to chemicals in food is estimated using three different parameters: substance concentration in food (ng/kg lipid weight (l.w)), food consumption rate (kg), and population body weight (kg).
The estimated daily intake (EDI) is determined through the following equation: In this case, EDI is the estimated daily intake (g day− 1), DFC indicates daily food consumption (kg/day), FFC indicates the food chemical concentration present in the analyzed food (mg/kg), and BW indicates the body weight (kg) of consumers. In this analysis, an average body weight of 70 kg/per adult person was considered for calculating the dietary exposure. The per capita milk consumption per day in the Bangladeshi population is 27.31 g [44].

Hazard Risk Index (HI) of PCBs congeners
The following formula was used for HI calculation: EDI: estimated daily intake of PCBs; ADI: acceptable daily intake through milk consumption. HI > 1 is considered a potential health hazard, and HI< 1 is considered safe for public health [45,46].

Statistical analysis
Data were processed using mass spectrometry. The peak areas were considered for PCB quantification. Analytical data were collected and organized using Microsoft Excel. Mean concentration differences and standard deviations in milk samples, in accordance with the sampling sites, were determined using one-way ANOVA. Samples below the LOD values were considered to have a concentration of 0.0 for statistical analysis. Median, Interquartile range of detected analytes were calculated using Calculator Soup online software.

Method performance
There are two common methods for the detection of PCBs in food samples: GC with an ionizing flame detector and MS [37,47], and liquid chromatography with a fluorescence detector [48][49][50] are used. There are several methods for the detection of PAHs in milk products, including HPLC-FLD [51,52], high-performance liquid chromatography-ultraviolet detection (HPLC-UV) [53][54][55][56][57] and GC-MS [35,58,59] or GC-MS/MS [24]. Because of its sensitivity and accuracy, we used GC-MS/MS in this study to determine PCBs and PAHs in cow milk. The chromatograms of the residual peaks of PCBs and PAHs are shown in Figs. 1 and 2, respectively. The method was validated based on recovery performance. Blank samples were adulterated with 50 ng/g and 100 ng/g PCBs congeners and PAHs compounds, respectively. Later, the adulterated samples were compared with those of the blank samples. In this study, the recovery rate of the PCBs were varied from 77.53% to 92.49%, while LOD and LOQ values were detected in the range of 0.016 ng/g to 0.031 ng/g l.w and 0.059 ng/g to 0.08 ng/g l.w, respectively. Table 1 Table 2.

Detection of PCB residues in milk
The levels of PCBs in cow milk samples collected from local firms in Dhaka, Bangladesh, were determined using GC-MS/MS. In this study, 100 milk samples were collected. PCBs residues including PCB No 52  Tables 3 and 4

Determination of PAHs compounds in milk
The mean values, median, interquartile range of detected PAHs analytes and Percentage of Positive samples of PAHs in the milk samples are presented in Tables 5 and 6

Health risk from the consumption milk from the studied area
The EDI and HI values of the PCBs are presented in Table 7. The HI values of PCBs No. 52, No. 101, 153, and 209 were 5.58 × 10 − 6 , 4.61 × 10 − 6 , 4.20 × 10 − 5 , and 4.73 × 10 − 5 , respectively. From the risk analysis, it can be concluded that the milk samples from the study area were safe for consumers, as the HI values were far below the risk level. However, milk samples containing PCBs residues for a longer time may cause health hazards. The results obtained from the method validation were satisfactory for the analysis of PCBs and PAHs. Table 3 shows the mean concentrations (ng/g l.w) of PCBs detected in milk samples. Among the 100 analyzed milk samples, PCB No. 52, 101, 153, and 209 were detected, whereas some PCBs, including PCB No. 28, 138, and PCB No. 180 showed values below the limit of detection. The PCBs concentrations detected in this study were lower than those reported in other studies [60][61][62]. Findings from Japan have indicated that intake of PCB-contaminated foods is toxic to mothers [63]. Milk processing at higher temperatures could lower PCBs concentrations, as it has been reported in several studies that pesticide concentration is reduced after thermal processing [64][65][66]. According to the regulation of 1259/2011, the Committee for European Communities has established a PCB reference limit of 40 ng/g fat [67] According to the regulation of 1259/2011, the committee for European communities has established a reference limit of PCBs of 40 ng/g fat [65]. As there have no established Maximum    [75]. Similarly, in Mexico, the total amount of detected PAHs did not exceed the permissible limit [76]. The analysis of cow milk samples in California identified PCB-101, PCB-118, PCB-138 as dominant PCBs congeners [69]. However, the the sum of all PCB congeners were below than the permissible tolerance level. Very little amount of PCBs congeners were detected in Slovakia [77]. In another study from UK identified 118, 153, 138 and 180 as dominant PCBs congeners and the total values of all PCBs congeners were in the range of 3.4-16.4 ng/g milk fat [78]. From this analysis, it was also found that some of the milk samples contained higher concentrations of PCBs congeners, which might be due to taking contaminated feeds like contaminated grass by cows. Animals are continually exposed to environmental toxins and are able to collect high concentrations of pollutants in their tissues, particularly fat tissues [79]. Therefore, the EU Scientific Committee on Food has fixed values for foods containing several contaminants. [67]. This analysis indicated that approximately 12% of the samples were contaminated with PCBs congeners. However, the concentration of PCBs detected in all milk samples was below the maximum level set by the EU Scientific Committee on Food. A study reported that if the soil is contaminated, it is possible to present PCBs residues in animal-derived foods, including milk, meat, and eggs [80]. While cattle take grass in an open field, they may take up soil contaminated with grass; as a result, contaminants such as PCBs accumulate in their body, and finally those contaminants are transferred to the human body through foods of animal origin.
Studies have reported that amyloidogenesis can be inhibited by polychlorinated biphenyls upon binding to transthyretin in the blood [81]. PCB exposure leads to neurobehavioral changes. PCBs particularly affect the neurotransmitter, dopamine. PCB associated with brominated flame-retardants (BFRs) induces the formation of reactive oxygen species (ROS) in neurons, which are involved in cell death. PCBs and BFRs also have negative effects on the immune system by producing ROS in neutrophils [82]. The endocrine systems of animals and humans can be disrupted by PCBs by controlling natural hormones [83]. PCBs congeners can affect mammalian oocyte maturation and embryonic development. PCBs can interfere with microtubules and alter ooplasm compartmentalization. Ovotoxicity occurred after a miscommunication between the germinal and somatic compartments. Coplanar PCBs can regulate gene expression through aryl hydrocarbon receptor [84]. PCBs can accelerate DNA damage, which can lead to breast cancer. Several chronic inflammatory diseases, such as cardiovascular disease, type 2 diabetes, obesity, hepatic disorders, endocrine dysfunction, and neurological deficits, can develop through PCB exposure [85,86]. Because of their lipophilic nature, PAHs bind to the cell membranes. As a result, structural changes affect normal cell function [87]. PAHs distribution is mainly dependent on fatty acids in conjugation with triacylglycerides, free cholesterol, and phospholipids, and is later distributed through blood capillaries to tissues such as skeletal muscle tissue, adipose tissue, and liver [88,89]. PAHs exposure can lead to lung cancer and intestinal diseases in smokers and nonsmokers, respectively [27,90,91]. PAHs exposure can lead to the development of cancers, including skin, lung, bladder, and gastrointestinal cancers, as well as damage to several organs such as the liver, kidney, and cataracts. Furthermore, the exposure PAHs causes gene mutation, damage of cells as well as cardiopulmonary related mortality [27,92].
The metabolism of PAHs occurs through the cytochrome P450 peroxidase and aldo-keto reductase pathways. These pathways involve diol-epoxides and radical cations, which disrupt DNA and damage cells by binding to DNA and proteins. As a result, carcinogenic, mutagenic, immunosuppressive, and teratogenic damage occurs and tumors finally develop tumors [93][94][95][96]. Benzo(a)pyrene (BaP) is associated with delayed hatching in zebrafish by altering the structure of the hatching enzyme (ZHE1). BaP toxicity also affects the skeletal regions of zebrafish by enhancing oxidative stress, apoptosis, and delaying embryonic development [97]. BaP can increase the weight of mice through b-adrenergic mediated stimulation of adipose tissues lipolysis [98]. Lower concentrations of chrysene can diminish cell viability in MIO-M1 cells as a result of apoptosis, whereas higher concentrations can lead to necrosis. Chrysene also alters mitochondrial function. Several studies have shown that PAH can form DNA adducts, leading to cell proliferation [99][100][101]. Animal cell culture studies have reported chrysene as a mutagenic, carcinogenic, and genotoxic compound [102][103][104]. In male Table 2 Elution order, limit of detection, limit of quantification and recovery rate of 16 PAHs compounds.   [105]. Chrysene was also found to be toxic to rat liver epithelial cells and organisms collected from the sea [106,107]. A study in C. chanos indicated that anthracene and benzo [a] pyrene can increase oxidative stress and are neurotoxic [108]. The number of cultured hamster cells was reduced after treatment with benzo (a)pyrenediones [109]. Almost similar result was observed with Hexachlorocyclohexane (HCH), an insecticide which significantly reduced the hatching and survivability rate of zebra fishes. Besides this effect, morphological deformities were also observed. Moreover, it induced oxidative stress [110]. PAHs are allocated throughout the environment and persist for a longer time [111]. As PAHs are lipophilic compounds they may contaminate accumulate on high fat containing foods like milk. After accumulation in adipose tissue, PAHs may be excreted in milk [48]. Several PAHs compounds exist; however, only 16 PAHs compounds have been studied so far in most of the studies. These results indicate that environmental pollution plays an important role in the contamination of milk by PAHs [112]. There are several ways to contaminate food. PAHs can contaminate vegetables and fruits in contaminated soil. Fish and seafood can be contaminated by polluted oceans. Cattles can take PAHs along with grasses (sometimes soil from grasses) from contaminated soils in industrial areas. Several studies have shown that milk products contain a remarkable amount of contaminants such as PAHs [113]. Different foods can generate variable levels of PAHs depending on several factors, including duration of heat treatment, temperature, types of fuel used during food production, fat content in food, and smoke utilization during food preparation. Among dairy products, the highest concentration of PAHs is generated in smoked cheese because of the smoke used during preparation and the fat content; environmental toxic elements are bio-accumulated in milk and milk products, transferred to the human body, and finally influence human health. Pasteurized milk samples contain lower PAHs than raw and UHT milk samples because of the differences in fat content [87].  Table 5 Detected PAHs compounds (ng/g) in Cow milk.   Other studies on milk and milk products also showed similar results [58]. In several other studies, naphthalene, acenaphthylene, acenaphthene, fluorene, and benzo (b) fluoranthene were detected in milk samples [33,51,58]. In our study, we detected only benzo (a) anthracene and chrysene at lower concentrations than in other studies performed in several countries worldwide. However, further studies are recommended to ensure public health safety.

Conclusion
This study investigated the presence of PCBs congeners and PAHs compounds in milk samples from several parts of Dhaka, Bangladesh using GC-MS/MS. The presence of PCBs in milk samples indicates the presence of environmental pollution. There are several other sources of PCBs contamination and there is a possibility of food contamination through PCBs congeners. The majority of PAHs compounds were not detected in milk samples. Very low amounts of benzo (a) anthracene and chrysene were detected in samples from industrial areas, where more human activities were observed. Therefore, proper detection and monitoring are required to ensure the safety of foods from animal sources. Detected values below the recommended levels indicate that the milk consumed by the population is free from toxicological risks. However, long-term exposure may result in significant exposure and cause a high contamination risk. Therefore, further studies are needed to establish specific MRL values for the Bangladeshi population.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.