Mercury bioaccumulation in relation to dietary habits in shing communities along river Swat, Pakistan

Muhammad Aamir Munir University of Peshawar Bushra Khan University of Peshawar Ishaq Ahmed Mian The University of Agriculture Peshawar Muhammad Ra q Institute of Management Sciences Samreen Shahzadi Pakistan Institute of Nuclear Science and Technology Kashif Naeem Pakistan Institute of Nuclear Science and Technology Iqbal Ahmad (  iahmad@gu.edu.pk ) Gomal University https://orcid.org/0000-0001-6529-2823


Introduction
Mercury (Hg) is a widespread heavy metal of great environmental concern (Hsu- Kim et al. 2018, Selin 2018, Tang et al. 2020. Hg exists in elemental, inorganic and or organic form in nature and poses severe threats to the food safety and human health (Hsiao et al. 2011, Sun et al. 2013 due to its toxicity and bioaccumulation in food webs (Dus et al. 2005). The chemical form of Hg largely affects its mobility and toxicity (ATSDR 2013). For instance, the organic form of Hg such as methylmercury (MeHg) is more toxic than the elemental or inorganic form (Hasegawa et al. 2005). The toxic effects of MeHg were rst realized when serious Hg poisoning occurred in Japan near the Minamata bay due to consumption of contaminated sh and sea food (Björkman et al. 2007, Karabedian et al. 2009), as the MeHg was released from an industrial plant into the coastal area of Minamata.
Hg contamination in aquatic ecosystems arise from both natural and anthropogenic origin such as mineral deposits, forest res (Barbosa et al. 2021, Camargo 2002, agricultural runoffs, mining activities, combustion of Hg containing fuels, municipal and industrial wastewater discharges (Moiseenko & Gashkina 2016, Pavlish et al. 2003, Wang et al. 2004, WHO 2005. During the recent decades, urbanization, industrialization, and population growth has intensi ed the contamination of freshwater ecosystems. The sh living in polluted waters tend to accumulate Hg in their tissues and transfer across the food chain (Moiseenko &Gashkina 2016).
Human exposure to Hg occurs mainly through ingestion of the contaminated food and sh (ATSDR 2013, Fakour et al. 2010) and may cause acute or chronic effects ranging from severe disruption of tissue, shock, cardiovascular collapse, autism, gastrointestinal damage, renal and neurological disturbances, dermatitis, fatigue and respiratory problems to life threatening toxicity (Bastos et al. 2006, Eisler 2006, Lee et al. 2003, Riaz et al. 2016, WHO 2005. The Codex Alimentarius guideline levels of MeHg for predatory and non-predatory sh are 1000 µgKg − 1 and 500 µgKg − 1 wet weight respectively (WHO 2007). While, in Japan the established guideline limit for total Hg and MeHg in sh species is 400 µgKg − 1 and 300 µgKg − 1 respectively (WHO 2016). The concentration of Hg in scalp hair is a widely used biomarker to evaluate Hg toxicity and contamination particularly for MeHg exposure from dietary consumption of sh and other foods (Agusa et al. 2007, Anwar et al. 2007, Díez et al. 2008, Feingold et al. 2020, Nuttall 2006 because hair sequesters mercury during its formation. Once incorporated into hair, Hg remains stable for a longer period as it binds to cysteine and can provide a history of MeHg exposure (Nuttall 2006).
The river Swat serve as a major fresh water source for a large portion of the Khyber Pakhtunkhwa province of Pakistan and have a signi cant role in sheries and agricultural sector. The variety of sh found in river Swat provides signi cant contribution to the diet of the rural communities of Swat. This study was conducted to 1) assess the Hg contamination in sh and scalp hair of the shing community of river Swat. 2) Assess the correlation between mercury accumulation in sh and human sh dietary patterns 3). Assess the health effects of Hg exposure on the shing community. To the best of our knowledge no such study is conducted in the region previously.

Study area
District Swat is located between 34° 34' to 35° 55' N latitude and 72° 08' to 72° 50' E longitude. The river Swat originates from Hindukush mountains and extend from Mahodand and Gabral at about 3,000 meters elevation and passes through the valley of Swat (DAWN 2007). Sampling was carried out in June -July 2019 at the up and downstream of the river Swat (Fig. 1).
According to the Fisheries department in Swat, Government of Khyber Pakhtunkhwa, Pakistan, a large variety of sh are found in river Swat. The classi cation of sh based on their habitat/locality and mode of nutrition is given in supplementary information (Table. S1).

Fish sample collection and preparation
The sh species (n=24) were caught from up and downstream of the river Swat with the help of local shermen (Fig. 1). The upstream sh included Salmo trutta fario (Brown Trout) and Schizothorax plagiostomus (Swati sh). Downstream sh included Swati sh, Crossocheilus diplochilus (Spena dega) and Garra gotyla gotyla (Tora dega). Brown trout is carnivorous while Swati sh, Tora Dega, and Spena Dega are Omnivorous. The sh species were measured for its length and weight, packed in clean polythene zip lock bags with identi cation code and transferred to the laboratory in iceboxes.
In laboratory, the sh were washed with distilled water. Fish llet (5 g), excised from the pectoral region was thawed, washed with deionized water and acid-wet digested (Ahmad et al. 2015). Brie y, the 5 g sh llet were taken in Pyrex glass tube and 2 ml of HNO 3 -HClO 4 (1:1) was added, followed by 2 mL conc. HCl and the mixture was digested overnight in a digestion block at ambient temperature. Finally, 5 mL of conc. H 2 SO 4 was added and heated to 200 C o for 20 minutes until the digests were clear (Voegborlo &Adimado 2010). The extracts were cooled at room temperature, ltered into centrifuge tubes, diluted to 20 mL with double deionized water and analyzed for T-Hg on the Atomic Absorption Spectrometer (GBC 932 plus) via Hydride Generation technique. All the samples were analyzed in triplicates.

Hair sample collection and preparation
Scalp hair samples (n=77) (42 upstream and 35 downstream) were collected from the shing communities including children (age ≤17 years) and adults (age ≥ 18 years) of both genders with prior written consent. The hair was cut from the occipital part of the head closest to the scalp using stainless steel scissors (Ahmad et al. 2018, Ohno et al. 2007). Female enumerators were involved to collect hair samples from female participants following social and cultural values of the study area. The hair samples were packed in separate polyethylene zip-lock bags and coded before transferring to laboratory for further analysis. In laboratory, the hair samples were thoroughly washed with distilled water and acetone solution using ultrasonic bath to remove the dirt and external contaminants (Ahmad et al. 2018, Al-Amodi et al. 2017). The clean hair samples were cut into small pieces and oven dried overnight at 50 C o . Dried hair samples (0.25 g) were weighed into clean Pyrex tubes, 2.0 mL of 1:1 HNO3: HClO4 was added, followed by the addition of 2 mL conc. HCl and the mixture was digested overnight in a digestion block at ambient temperature. Finally, 5 ml of concentrated H 2 SO 4 was added and the tubes were heated at 200 °C for 20 minutes until the digests were clear (Koseoglu et al. 2017, Voegborlo &Adimado 2010. The digested solution was cooled at room temperature, ltered into centrifuge tubes, and diluted to 20 mL with double deionized water (Ahmad et al. 2018 The inclusion criteria for the questionnaire survey were interesting as only household involved in shing were included. For the purpose, shing villages were identi ed by computing the number of shing licenses issued by the sheries department Swat, Government of KPK, Pakistan. In upstream area, the villages selected for household survey and sampling were Koz Kalay, Tanqar, Shagai, Jupin. Whereas, in downstream area, the selected villages for sampling were Garho, Aligrama, and Maam Derai (Fig. 1). A total of 21 randomly selected households were surveyed among which 11 were from the up and 10 from the downstream area. All the households were within 6 Km distance from the river Swat.

Statistical data analysis
Statistical data analysis was carried out using Multilinear regression and Group mean difference t-test.
Descriptive and multivariate statistical analysis were performed using XLSTAT 2017 and OriginPro 2018.

Results And Discussion
Hg concentration in sh The mean Hg concentration in upstream sh trutta fario (Brown Trout) and Schizothorax plagiostomus (Swati sh) were 34.7±18 µgKg -1 and 29.4±15 µgKg -1 ( Hg concentration of 336.73 µgKg -1 was found in Tora dega sh from downstream area. Generally, the concentration of Hg in downstream sh species was high than upstream. Previously, (Diringer et al. 2015) observed high Hg concentration in downstream sh species in Madre de Dios River. However, Michalak et al. (2014) and Anwar et al. (2017) observed no signi cant variation in up and downstream sh. In our study the high Hg concentration observed in downstream sh can be associated with anthropogenic inputs. Our survey revealed that most of the agricultural, municipal, and industrial wastewater is directly discharged into the river Swat. Besides a large proportion of the solid wastes (hospital, industrial, municipal) generated were also dumped into the river Swat.

Hg concentration in hair
The average Hg concentration in scalp hair of the up and downstream shing community were 658±125 µgKg -1 and 3969±791 µgKg -1 respectively. In upstream area, Hg concentration in scalp hair was above the WHO permissible limit of 1000 µgKg -1 for 19% subjects. Likewise, in downstream area, Hg concentration in scalp hair was above the WHO permissible limit for 51% subjects. Age and gender wise mean Hg concentration in scalp hair of the children (female and male) and adults (female and male) for up and downstream shing communities is presented in Fig. 2.
The horizontal line indicates the WHO guideline limit of 1000 µg Kg -1 Hg for scalp hair. The mean Hg concentrations in scalp hair of the upstream shing community were below the WHO guideline limit of 1000 µg Kg -1 for children and adults (female and male). In contrast the mean Hg concentrations in scalp hair of the downstream shing community exceeded the WHO guideline limit for children and adults (Fig.  2). Independent T-test showed signi cant difference (t = -3.37) in the mean Hg concentration in scalp hair of the up and downstream communities. However, no signi cant difference was observed for the Hg concentration in scalp hair of the male and female. Previously, (Anwar et al. 2007) and  observed high Hg concentration in scalp hair of the female compared to male. In contrast, (Barbosa et al. 2001) reported signi cantly low Hg concentration in scalp hair of the females than male.
No signi cant correlation was observed between Hg concentration in hair and age of the respondents. Similar ndings were reported by (Szynkowska &Pawlaczyk 2007), (Karabedian et al. 2009), and (Michalak et al. 2014). However, a positive correlation between Hg accumulation in hair and age of the respondents was reported by (Wyatt et al. 2017) and ). The regression model indicated that literacy reduces the Hg accumulation in human hair by 12.98 µgKg -1 from the constant, however, the correlation was not signi cant. While Anwar et al. (2017) reported that individual having 11 or more years of education (p for trend = 0.013) have relatively higher concentration of Hg in scalp hair.

Effect of dietary habits on Hg concentration
The multilinear regression indicated that in the upstream area consuming carnivorous sh (Brown trout) daily or 2-3 times a week signi cantly increases the Hg concentration in scalp hair by 3246 and 368 µgKg -1 respectively from the constant value of 515.4 (con dence level 99%) ( Table S-2). Similarly, consuming omnivorous sh (Swati sh) 2-3 times a week signi cantly increases the Hg concentration in scalp hair by 1981 µgKg -1 from the constant value of 2734 µgKg -1 (con dence level 90%). In the downstream area, consuming omnivorous sh (Swati sh, Tora dega and Spina dega) 2-3 times a week signi cantly increases the Hg concentration by 2605 µgKg -1 from the constant value of 1474 (con dence level 95%) ( Table S-2). Previously, (Feingold et al. 2020) and (Díez et al. 2008) found a positive correlation between sh consumption and Hg level in hair. While, (Fakour et al. 2010) observed that 76.4% of the sh consumers have Hg levels exceeding the USEPA recommended level of 1000 μgKg -1 .
The variation of Hg accumulation in human scalp hair in relation to non-sh dietary habits is given in Table S-3. The Regression analysis showed that consuming dry fruits at least 1-2 times a month or 1-2 times a week signi cantly lowers the Hg accumulation in scalp hair by 5347 and 5077 µgKg -1 respectively from the constant value of 8943 (con dence level 90%). Previously, (Wyatt et al. 2017) observed a decrease in hair Hg levels with consumption of grains and fruits. In contrast, (Feingold et al. 2020) did not nd any signi cant trend of hair Hg levels in relation to rice and fruits consumption.

Conclusion
The concentration of Hg in up and downstream sh species were within the WHO permissible limits. However, the mean Hg concentration in downstream sh species was 5 times higher than upstream sh.
Likewise, the mean Hg concentration in scalp hair of the downstream shing community was 6 times higher than upstream shing community and exceeded the WHO guideline limit of 1000 µg Kg − 1 for both children and adults. The Hg concentration observed in different shes of the river Swat is a serious public health concern especially for the shing community. Further studies are recommended to identify the sources of Hg and welfare impact of sh contamination on the shing community of river Swat.

Declarations
Ethical approval and consent to participate This study was conducted with prior approval from Ethical Research Committee, University of Peshawar. Written consent was obtained from all the sampling population involved in this study.

Consent for publication
Not applicable Availability of data and materials Summary of the data generated or analyzed during this study is included in the article. While detailed data of this study are available upon request.
communities upstream and downstream of artisanal and small-scale gold mining. International journal of environmental research and public health 14, 1582 Figure 1 Sampling locations along the river Swat. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Figure 2
Age and gender wise Hg concentration in scalp hair of the shing communities.