Levels of persistent organochlorine and polychlorinated biphenyls in Nile tilapia (Oreochromis niloticus) from three cage aquaculture farms on the Volta Basin of Ghana: Implications for human health


 Increased agrochemical based agricultural activities along the Volta Basin and aquacultural practices from the cage farms have raised concerns about the potential ecological risk to the aquatic ecosystem. To assess this, caged tilapia samples were analysed for polychlorinated biphenyls (PCBs) and organochlorine pesticide (OCP).Fish samples in acetone/hexane (2:1v/v) solvent medium were sonicated on ultrasonic bath, shaken electronically, cleaned by solid-phase extraction and analysed by GC/ECD for OCPs and GC/MS for PCBs.The levels of OCPs reported in the fish ranged; ND – 2.310 µg/kg (fish farm A), <LOD – 4.260 µg/kg (fish farm B), <LOD – 6.109 µg/kg (fish farm C) and ND – 0.878µg/kg (control). The highest concentration of 6.109 µg/kg was encountered for δ-HCH in the muscles of Tilapia from fish farm C, whiles p,p'-DDE recorded the lowest in farm A. The levels of the OCPs detected in the cage tilapia were below the MRL proposed by the EU. Six PCBs congeners; PCB 18, PCB 28, PCB 52, PCB 101, PCB153 and PCB 180 were detected in the cage tilapia. The concentrations of the indicator congeners ranged from 0.288 ng/g to 0.931 ng/g dw, 0.042 ng/g to 1.321 ng/g dw, 0.260 ng/g to 10.657 ng/g, and ND-0.298ng/g dw in fish farm A, B, C and control respectively. The highest mean concentrations of PCB 153 (8.524 ± 1.5960ng/g) was found in farm Estimation of the EDIs and HQs for the pesticides do not present potential health risk to the consumers of the cage tilapia fish from the studied farms.


25
Organochlorine pesticides (OCPs) and polychlorinated biphenyls (PCBs) are classified as persistent organic 26 pollutants (POPs) by the Stockholm Convention of the United Nations Environment Programme due to their high 27 persistence, toxicity and bioaccumulation to humans and wildlife (WHO 2009). Because of their great persistence 28 and bioaccumulation, they can travel vast distances in the atmosphere and be detected far from where they are not 29 used or manufactured. Their semi-volatility and chemical stability cause them to undergo long-range movement, and 30 therefore, studies have shown their presence in the Arctic and Antarctic regions (Taiwo, 2019).

31
OCPs and PCBs have been manufactured and applied in many items in recent years, including insecticides, flame 32 retardants, transformer oil, coolants, antifouling agents, and building materials (Mwakalapa et al. 2018). Past  OCPs and PCBs (Fu et al. 2018). Pesticides may enter and pollute the aquatic ecosystems in many ways, including 35 spray-drift during pesticide application, accidental spillage or unauthorised recycling of their products (Cox & 36 Surgan, 2006). Contamination of the aquatic ecosystem has become a real concern for fish and other aquatic life that

61
The Volta Basin, which contributes the most to Ghana's inland fisheries, with around 80% of the overall yield, is   Pesticide grade ethyl acetate, distilled water (HPLC grade), acetone, and hexane (both of analytical grades) were 93 supplied by Fisher Scientific (Loughborough, UK), silica gel, and sodium sulphate were purchased from E. Merck 94 (Germany). Sigma-Aldrich Chemicals USA provided disposable solid-phase florisil cartridges (500mg/6mL).

100
The study included 52 cage tilapias (Oreochromis niloticus) bought from three of the fish farms A= 19, B =16, and 101 C =17 on the Volta Basin. Two of the tilapia fishes were pulled together as a composite working sample.

102
Additionally, twenty (20) samples of wild Oreochromis niloticus were collected as controls from the upstream of the 103 Volta Basin where aquaculture is not practised. All samples of fish were preserved in an airtight bag at 4 ° C and 104 conveyed to the Ghana Atomic Energy Laboratory for further study.

105
The morphometric data of the Tilapia taken were almost the same, while the fillets were removed using a stainless-   In order to avoid interferences, the clean-up system is important for the pesticide residues analysis in a sample.

122
Before the clean-up, silica gel was activated by heating it moderately to about 150℃ in an air-tight oven. This 123 process removes water content in the gel and increases its adsorptive capacity significantly. The glass separating 124 column (20cm) was packed with three layers. The agents were arranged with the activated charcoal at the top, 125 followed by 2g of anhydrous granulated Na 2 SO 4 and the bottom packed with 4g of activated silica gel (90% < 45 126 µm). The activated charcoal removes colouration, the anhydrous Na 2 SO 4 serves as a demoisturizer and the silica gel, 127 removes co-extractants. Prior to cleaning, 10 mL n-hexane was used to condition the columns. The extracts were 128 loaded inside the column. The concentrate in the flask was rinsed with 5ml hexane and added to the column again.

129
The eluate was concentrated to near dryness using a rotary evaporator at a temperature of 45°C and picked in 2ml  In order to compensate for the relative retention times and the response behaviour, the GC conditions and the 138 detector response were determined. N was used as carrier gas with a flow rate of 1.0 ml/min and 29 ml/min as

152
Residue levels of OCPs and PCBs were quantitatively measured by reference guidelines and residence time 153 estimations. By matching the peak heights of the samples with the respective peak heights of the reference standards 154 for specific concentrations, the residue levels were determined by the external standard procedure. The calculation 155 was performed within the linear range of the detector. To obtain the concentration, the peak areas whose retention 156 times correlated with the standards were subsequently extrapolated to their respective calibration curves. The lowest 157 analyte level that could reliably and consistently provide recovery of 70% or more from the enriched samples 158 described by Koranteng (2015) was used to determine the LOD for OCPs and PCBs in this paper. Standard OCPs 159 and PCBs mix solutions were serially diluted, and the standard deviation of the signals was measured using the 160 lowest concentration whose recovery from fortified samples was greater than 70% and also gave a signal to noise 161 ratio of 1:3. The standard deviation (SD) was multiplied by 3 to get the LOD (i.e., SD x 3 = LOD). The standard 162 deviation for the LOD determination was multiplied by 10 to get the limit of quantification (LOQ) (i.e. SD x 10 = 163 LOQ).The LOD and LOQ for the determined pesticides in sediment were set as 0.12μg/kg and 0.40μg/kg 164 respectively.

166
The analytical system integrated quality control and assurance. Measures were taken to guarantee the reliability of 167 the results. All glass apparatus used for research (extraction and cleaning) has been thoroughly rinsed with detergent 168 and tap water. The glassware was cleaned with distilled water before being meticulously cleaned with analytical 169 grade acetone and dried overnight in an oven set at 70°C. They were taken out of the furnace and placed in dust-free 170 cabinets after cooling. To guarantee the quality of the OCPs and PCB residues, analyses of samples, procedural 171 matrix blanks, and solvent blanks were done in triplicate. Each batch of analytical extracts was analysed 172 simultaneously with procedural recoveries. Furthermore, with each batch of samples, recalibration curves were ran 173 to ensure that the correlation coefficient remained above r 2 > 0.995. Recoveries for internal standards varied from 78 174 percent to 95 percent for OCPs and 80 percent to 94 percent for PCBs, indicating that the approach used was 175 reproducible.

177
In order to assess the normality of the data, the Kolmogorov-Smirnov (K-S) analysis was performed and at the p-178 value less than 0.05, findings were considered to be statistically significant. Descriptive statistics such as the mean 179 and standard deviation (SD) were used for the levels of OCPs, and PCBs. Ranges were computed for the 180 contaminants. One-way ANOVA was used to test the differences in the contaminants from the fish farms and 181 controls where samples were obtained, with a Tukey's post hoc test.

183
The Estimated Daily Intake (EDI) (mg/kg/day) for the PCBs and OCP residues obtained in the various fish 184 samples was calculated for each age category (Children <12 years and adult ≥12 years) using the equation  The hazard quotient (HQ), which is also the ratio of the EDI to the Acceptable Daily Intake (ADI) or reference 191 doses (Rfd), was used to assess non-carcinogenic risk. Using equation 1 and the Rfd, the HQ for non-192 carcinogenic risk is: 194 If HQ ≤ 1 shows no harmful influence on health. However, if the HQ > 1.0, then there is a potential non-195 carcinogenic adverse health effect but not demonstrated.

196
The exposure thresholds, according to USEPA (2012) used for health hazard calculations through the ingestion 197 of fish for children and adults, are shown in Table 1.

209
The target risk of cancer was determined by using the equation: Where CSF is the cancer slope factor of the individual pollutant and TCR is the target cancer risk. The CSF for   (Table 2 and Table 3).   The concentrations of organochlorine pesticides generally varied from ND to 2.310 µg/kg, <LOD to 4.260 µg/kg, 229 <LOD to 6.109 µg/kg and ND to 0.878µg/kg in fish samples from farm A, B, C, and controls, respectively. Farm A. This was expected, as DDE is more stable than DDT and degrades more slowly when exposed to microbes,  on freshwater fish exposed to ∑DDT at concentrations up to 0.500 g /kg ww. According to this argument, the total 255 DDT concentrations detected in adult fish muscle from farms A (4.158g/kg), B (6.730g/kg), C (6.767g/kg), and 256 controls (1.148g/kg) during the study could impair their normal physiological function and growth rate.

257
Endosulfan is abroad-spectrum insecticide as well as acaricide marketed in Ghana with a trade name thiodan, and is

285
Aldrin and dieldrin were not detected in fish samples, indicating that farmers in the basin do not use them in their 286 farming practises. Heptachlor is a pesticide that disrupts the endocrine system. Adults exposed to heptachlor may 287 experience congenital abnormalities, cancer, particularly hormonal malignancies, a delay in sexual development, 288 and a delay in nervous system development. Heptachlor was detected in all fish samples taken from the farms,

289
indicating that it is less water-soluble. As a result, it accumulates in fish when released into bodies of water.

290
Heptachlor was detected in fish from farms A, B, and C at an average concentration of of2.126µg/kg, 1.335µg/kg 291 and 2.890µg/kg, respectively. Because heptachlor was most likely converted in vivo via microsomal oxidation to 292 heptachlor epoxide, a more hazardous molecule (US National Academy of Science 1992), the average level of 293 heptachlor was lower in Farm B than in farms A and C.

294
Methoxychlor concentrations in cage tilapia fish samples (Table 2) were highest in farm C (3.889 ± 3.343µg /kg, dry 295 weight); however, methoxychlor concentrations from Farm A were below the detection limit. The elevated

298
The allowable limit for methoxychlor, endrin, dieldrin, aldrin, and heptachlor in fish is 0.3 mg kg −1 (300µg/kg), but 299 the value for DDE, DDT, endosulfan, and DDD is 5 mg kg −1 (500µg/kg) according to the US Federal Drug 300 Administration (Ezemonye et al. 2015). As a result, the levels of identified OCPs were found to be lower than the 301 USFDA levels in all cage Nile tilapia.

302
The pesticide loads (sum of all detected pesticides) in the Tilapia from the farms were;11.011 µg/kg, 19.129 µg/kg, 303 25.408 µg/kg and 5.196µg/kg in fish farms A, B, C and controls respectively. The highest pesticide load of 25.408 304 µg/kg was obtained from fish farm C. This could be due to their high lipid content, making them susceptible to 305 bioaccumulation of more of the organochlorine residue.

306
The findings of the current research are compared in Table 4 with those from other regions of the world, as well as 307 with some from local studies in Ghana. Comparison of the mean OCP residue ranges in muscle tissues from the 308 present study with results from other local studies shows that values from this study were lower. Whereas the 309 highest local OCP concentration range in the muscle of fish (0.78 -94.00 µg/kg) was quoted by Koranteng (2015) 310 for fish species from the Afram arm of Volta Lake, that for this study was ND -6.109 µg/kg. Comparing the results, 311 however, to the ranges by some works from other regions of the world; the concentration range for this study was 312 lower. For instance, as shown on Table 4, Abbassy (2018) reported a mean concentration range less than the limit of

317
Lake Tanganyika, Lake Nyasa and Lake Babati in Tanzania. Cage aquaculture tilapia from the Ghanaian aquatic 318 ecosystems can therefore generally be considered relatively less contaminated.

319
In Table 4, the current study's findings are compared to those from other parts of the world and some from Ghanaian 320 studies. When the mean OCP residue ranges in muscle tissues from the current study were compared to data from 321 other local investigations, the present study's values were lower. 10 OCPs <LOD -273 Lake Victoria, L. Tanganyika, L. Nyasa and L. Babati ND-Non detected, <LOD-Below detection limit While Koranteng (2015) reported the highest local OCP concentration range in fish muscle (0.78 -94.00 µg/kg) for 339 fish species from the Afram arm of Volta Lake, the range for this study was ND -6.109 µg/kg. When comparing the 340 results to those of other studies worldwide, the concentration range for this study was lower. For example, Abbassy

341
(2018) reported a mean concentration range less than the limit of detection -6.71µg/kg for fish species from Egypt's    as well as the controls, were less than one (1) using the data set from the current study (Table 6). As a result, the 387 consumption of cage tilapia from the Volta Basin farms poses no non-carcinogenic health risk to the population. On 388 the other hand, children had a much higher non-carcinogenic risk than adults. Eating pesticide-contaminated cage 389 fish put consumers at a higher non-carcinogenic risk than eating wild fish (Controls). The pesticide residue found in 390 processed fish samples that has the potential to cause cancer was analysed for carcinogenic risk.     CSF=Cancer slope factor Table 7 shows the results of the EDI measurement for carcinogenic risk. The EDIs were significantly lower than the were unlikely to represent a carcinogenic risk to both young children and adults.

418
The study discovered a wide variety of PCBs and OCPs residues in the muscle of cage tilapia. The major pesticides