Arsenic Speciation in Rice, Mechanisms and Associated Health Risk Through Rice Consumption in Various Districts of Khyber Pakhtunkhwa, Pakistan

Arsenic (As) is one of the toxic metalloids, therefore, can cause health risk in the consumers through consumption of contaminated food and rice. The current study focused on As speciation in rice, bioavailability, mechanisms and its potential human health risk. For this purpose, rice and soil samples were collected from 16 different districts (non-mining and mining sites) of Khyber Pakhtunkhwa (Pakistan). Soil physicochemical characteristic such as texture, electrical conductivity (EC), organic matter (OM), pH, iron (Fe) and phosphorus (P) were determined. Total arsenic (AsT) concentrations were analyzed using ICP-MS, while the arsenite (As3+), arsenate (As5+), arsenobetine (BAs), dimethylarsenic (DMA) and monomethyl arsenic (MMA) were determined by HPLC–ICP-MS method. The highest AsT (0.28 mg/kg) was observed in the rice sample of DI Khan District and lowest (0.06 mg/kg) in Shangla District. However, these findings were found within the permissible limits set by various authorities. Furthermore, results showed higher concentrations of inorganic As (Asi) than organic As (Aso) species in rice. The estimated daily intake (EDI) and incremental lifetime cancer risk (ILTCR) were used to evaluate the potential human health risk for As consumption in rice. Results revealed that the rice samples collected from the district having mining activities had higher value of As (0.28 mg/kg of AsT) as compared to non-mining (0.072 mg/kg of AsT). The highest ILTCR value (0.00196) was observed for rice collected from mining districts. This study revealed that mining activities have great influence on the As contamination of soil and grown rice. This study recommends the use of geo-sorbents as soil amendments in districts having mining activities for the purpose to lower As availability in soil and its bioaccumulation in growing rice that will help to keep lower the potential risk.

The As mobilization and bioavailability were in uenced by several factors including speciation, pH, reducing environment (Eh), Fe (iron) (Yamaguchi et al. 2011) and its uptake in the rhizosphere of soil-plant system (Anawar et al. 2018; Kowalczyk and Latowski 2018). Flooded paddies soils had the characteristics that resulted in the mobilization of As, thus enhancing its bioavailability and uptake in rice plant (Yamaguchi et al. 2011).
Furthermore, there are a number of characteristics (both biological and physical) of soil and rice, also its chemical characteristics have a great in uence on the bioavailability of As in soil. The uptake and toxicity of As varies with varieties of rice (Aqeel et al. 2014;Pravalprukskul et al. 2018). The As i species may be converted into organic arsenic (As o ) species through methylation governed by microbes' actions in the paddy eld (Zhu et al. 2017;Afroz et al. 2019). Tetramethylarsonium (TMA) was mostly found in rice grain of China emanating from the contaminated paddy soil that had average account 5.9% of the total arsenic (As T ) concentration (Meharg and Zhao 2012). It is because of the anoxic environment of the paddies in which rice is grown, rice can uptake As 3+ more e ciently than other cereals (Williams et al. 2007). Accumulation capability of rice to store As is almost double than other grain plants like barley and wheat (Su et al. 2010). The contamination of As in rice plant and grains result in food chain contamination and could pose serious public health concern (Lei et al. 2013).
The exposure of As (Long-term exposure) contamination through water and food lead to a wide range of lesions such as melanosis, leucomelanosis and keratosis (Rahman et al. 2009b). Higher As contamination results in high blood pressure, neurological effects, obstetric problems, diabetes mellitus, respiratory diseases and blood diseases such as cancers mostly skin, bladder and lungs (Ahmed et al. 2016). Prolong exposure of As is not only carcinogenic, but it may also disturb nervous system ( (Rehman et al. 2016). However, no systematic study conducted to investigate the speciation in rice, mechanisms involved in the formation of these species and their uptake by the rice and its associated human health risk. This study will determine As speciation not only in rice but also soil of the same sites so that it shows the mechanisms involved in its bioavailability and uptake by rice plant. This study will also evaluate the potential chronic risk and cancer risk associated with the ingestion of As contaminated rice. Lower and Dir Upper) which are rice producing areas as shown in the location map (SI). The sites were selected on the basis of rice production in each district and low rice production sites were excluded. Whole study area was divided into two groups mining and non-mining. This was done so as to study the effect of mining activities on the As contamination in both soil and rice.

Samples collection and pre-treatment
Rice grain samples (1 kg) were collected in polyethene zip-bags from the eld having replicates (n=6-10) in each study area district. From base of each uprooted rice, surface soil (0-15 cm) samples were collected in polyethene zip-bags and numbered accordingly. Global position system (GPS, Garman eTrex 30) was used to measure the latitudes and longitudes of each selected sampling points. All rice (n=135) and soil samples (n=135) were taken to laboratory for further analytical procedures.

Soil characteristics
Physiochemical characteristics including electric Conductivity (EC), pH, Organic Matter (OM), Iron (Fe), Phosphorus (P) and texture were measured according to standard procedures (Table 1). Soil EC and pH were measured in solution of soil and deionized water (1:2.5) with Accumet XL 60 meter equipped both electrodes (pH and EC) (Muhammad et al. 2011). Mastersizer 2000 (Malvern Instruments Ltd, UK) was used to determine the soil particle size (sand, silt and clay) according to operational manual of the instrument. The quanti ed data was converted into percentage (Li et al., 2018). Weight-loss procedure was used to analyze OM in soil samples (Roper et al., 2019). The concentration of iron (Fe) and phosphourse (P) were measured according to the procedure adopted from the (Kumar et al. 2011) and ).

Digestion of soil and rice grains
Soil and rice samples were dried in air powdered and passed over 2mm mesh and oven dried. Soil samples were grounded manually with the help of mortar and pestle while rice samples were powdered with Vibrating disc mill (Fritsch, Germany). For the digestion of soil, samples weight 0.2 g were taken and 5 mL HCl (hydrochloric acid (12.0 mol/L)) and 5 mL HNO 3 (nitric acid (15.2 mol/L)) was added, the samples were left overnight at room temperature and then digested with the help of block digester at 100 o C for 1 h, followed by at 120 o C for 1 h, and  Followed by the addition of 2 mL of H 2 O 2 (hydrogen peroxide) was added, and the samples were digested using a microwave accelerated reaction system (CEM-Mars,Version 194A05, USA). Initially temperature was raised to 55 0 C for 10 min, then raised to 75 0 C for 10 min, and lastly to 95 0 C for 30 min, and then the samples were cooled at room temperature. The 0.22 μm membrane was used to lter the digested samples and de-ionized water was added to dilute the samples and make it up till 50 mL. ICP-MS (Agilent Technologies, 7500 CX, USA) was used to analyze As T concentrations while Fe was determined with (AAS atomic absorption spectrometer (Perkin Elmer, AAS-PEA-700) and P was determined with ICP-OES (Perkin Elmer Optima 7000 DV, USA) (Khan et al. 2014).
For the extraction and quanti cation of As species ((AsIII), (AsV), (DMA), (MMA) and (BAs)), 200 mg of powdered rice samples were placed in 50 ml polypropylene tubes and 10 ml of 1% (v/v) HNO 3 was added to them. For the sample extraction, Microwave assisted digestion technique was used (Jia et al. 2012). Different species of As in the extracts were analyzed with the help of HPLC-ICP-MS. The selected arsenic species were separated with the help of an anion-exchange column (PRP X-100, Hamilton Company, USA) with the mobile phase of 10 mM (NH 4 ) 2 HPO 4 and 10 mM NH 4 NO 3 (pH 6.2). Total As i was calculated as the sum of AsIII and AsV, while total As o was calculated as the sum of MMA, DMA and BAs. The extraction e ciency was 83.8 to 87.9% (SI) of these As species from reference material (mentioned in Sec 2.5) which is considered as satisfactory. The detection limits for As species ranged from 0.1 to 0.3 μg/kg. For the method validation, the sample solution obtained from digestion in microwave oven was selected due to the lesser time consumption involved and the lower blank values. The analytical method validation was performed by considering the LOD. All the analytical methods and validation is carried out according to Magnusson (2014) guidelines.

Precision and accuracy
For the veri cation of accuracy of the data standards and blanks run on column at the start and with regular 10 samples interval. Soil and rice our certi ed reference materials (GBW07406-GSS-6 and GBW-10045) were used to verify the extraction e ciency and As species stability. These certi ed reference materials were dried at 50 o C for 6 h before microwave digestion. Thus, As content was not affected by the low temperature treatment. HNO 3 (2% of the concentrated acid (65%)) and Milli-Q water were used to wash glass wares and plastic bottles properly. The recovery rates from reference materials were in the range of 92.1 ± 6.3 to 101.3 ± 8.3 %.

Physical and chemical characteristics of Soil
Results of the texture shows that soil varied from sandy loam to loamy sand (Table 1). Higher EC value was measured for Mansehra (7.53 mS/cm) and lower in Upper Dir (0.77 mS/cm). Highest soil OM was observed in DI Khan (4.7%) and lower in Lower Dir (2.0%). Maximum pH value was observed for Banu (8.1) and minimum for DI Khan (7.4). Low pH of the DI soil could be attributed to higher OM. Higher values of P were found in DI Khan (0.47 g/kg) and lower in Charsadda (0.15 g/kg) and that of Fe were in Upper Dir (3.37 g/kg) and lower in Charsadda (2.06 g/kg). Figure 1 represent the values of As i and As o found in the soils of Khyber Pakhtunkhwa. Overall results shows that the percentage of As i is more as compared to As o while in some areas percentage of the As o is higher than the As i .

Arsenic in soil
The highest value of As i was recorded for 79.6% for Buner, while the lowest was 34.6% for Swat district. Also, for As o , the higher percentage 72.3%, was found in Malakand and the lowest was 20.2% for Buner. This is due to the fact that As 3+ can be converted into organic forms of As by the process of methylation encouraged by microbial activities in soil paddies ).
In soil the mean As i ranged from 2.27-4.54 mg/kg, while it was higher in Upper Dir and lower in Banu district. As far as As o is concerned, it was higher in Swat district (mean=5.73 mg/kg) and lower in Buner (mean=1.11 mg/kg).
Similarly, mean As T was higher in Chitral district (mean=13.72 mg/kg) and lowest in Banu (mean=4.31 mg/kg).
The higher values of As in Chitral and swat was due to the mining activities in those areas. All these concentrations were below normal level of As in soil (

Arsenic in rice
The rice samples collected from 16 different districts of Khyber Pakhtunkhwa show variation in As species concentration and are presented in Fig 2 (S10) of SI. The As i concentration in the rice ranged from 0.05-0.24 mg/kg with high level in DI Khan and low level in Shangla district (Fig 2). Similarly, the As o ranged from 0.002-0.02 mg/kg with maximum concentration in Charsadda and minimun in Mardan. Also, the As T was ranged 0.06-0.28 mg/kg, which was higher in DI Khan and lower in Shangla.
Flooded rice paddies has the characteristics of reducing environment, and this anoxic condition enhances the availability and mobility of As 3+ as compared to that of As 5+  Numerous researches on speciation of As in South and South East Asia found As i which is highly toxic constitute about 42%-91% of As T in rice while in America rice, the major specie of As is DMA (Schoof et al. 1999;). Other researches also showed that As i (75%-90%) is high in different products of rice such as, baby rice, rice crackers, breakfast cereals, rice milk, and other products of rice (Sun et al. 2009).

Relation of Arsenic and soil physicochemical characteristics
Clayey and silty soils have ner texture, more surface area and higher As scavenging capability as compared to sandy soils, which is because of the Fe oxides presence. Hence, clayey soils plants show less toxic effects of As; As phytotoxicity is ve times high in loamy and sandy soils (Quazi et al. 2011).
The highest OM value is present in soils containing dolomite, while those with greater pyrite and phyllosilicates levels showed very less OM content. The results shown that samples with higher OM content, displayed lower As content. Soil OM is responsible for the mobility of As also its complexes (soluble or insoluble) and chemical ) showed that the uptake and accumulation of As by rice plant is affected by both low and high pH. The reason may be that at low pH (pH < 5), Fe-oxyhydroxide compounds which are As-binding species, becoming highly soluble and increase the uptake of As by plants. Bhattacharya et al. 2010 also supported the negative relationship between soil pH and As concentration in rice. On the other hand, many authors supported the idea of a positive relationship between soil pH and As accumulation (Ahmed et al. 2011;Campbel et al. 1985). increasing soil pH (such as pH 8.5) rises the negative surface charges (hydroxyl ions), which facilitates As desorption from Fe-oxides that leads to mobilization of As in the root, which, in turn, increases accumulation of As in the plant (Ahmed et al. 2011). As the pH decreases As 3+ solubility increases with the soil pH range (pH [3][4][5][6][7][8][9], whereas in the case of As 5+ this pattern is reversed. In ooded paddy soils As 3+ predominates. In this regard, for the current study, a decrease in soil pH can enhance the mobility of soil As, which gives the reason why we observed almost equal concentrations of As in rice grain irrespective of the levels of soil As (Sahoo and Kim 2013).
By the application of external P, the uptake of As is reduced and the toxicity symptoms of As 5+ alleviates. Fe-plaques has high a nity towards As 5+ , therefore it plays an important part in decreasing the As uptake in rice. For the sequestration of As, Fe-plague is probably the best solution which ultimatle reduces the transport of As from roots to shoot in rice plant (Liu et al. 2004). Leonardite had a higher e ciency to adsorb As than biochar's because of the iron contents in leonardite which strongly bind to As anions (Dolphen and Thiravetyan 2019).

Comparison of As level in rice of mining and non-mining areas
One of the most important human activities that contaminate our environment with As is mining ). Due to the presence of As in many minerals and other human activities such as mining, As-contaminated waste water irrigation, fertilizers and pesticides application containing As, As became one of the most common contaminants affecting human health via food chains (Zhu et al. 2014). As is present naturally in many ores such as Cu, Pb, Au and Zn ores. Therefore, in mine impacted areas, As is found in soils at higher levels, posing a risk to not only human health but also ecosystem health ). Current study reported high As i , As o and As T in mining areas. In non-mining areas, the As i concentration is higher in Peshawar (0.14 mg/kg) and while lower in Mardan (0.06 mg/ kg). As o was higher in Charsadda (0.02 mg/kg) while lower in Mardan (0.002 mg/kg). While the concentration of As T is higher in Peshawar (0.018 mg/kg) and lower in Mardan (0.07 mg/kg). By comparing these results with the mining areas result it was concluded that for As i the highest value was detected in rice grown in DI Khan (0.24 mg/kg) and lowest in Shangla (0.05 mg/kg). For As o the maximum concentration was detected in the rice of Lower Dir 0.03 mg/kg while the lowest concentration was found in Shangla (0.002 mg/kg). For As T highest value is (0.28 mg/kg) in rice of DI Khan while lowest value is (0.06 mg/kg) in Shangla rice as shown in Fig 3. The overall results depicted that the highest values for the species of As as well as the As T was in the rice of mining areas such as DI Khan. But there are some non-mining areas such as (Peshawar As T = 0.18 mg/kg) which shows a little higher concentration than the mining areas (Buner As T =0.09 mg/kg, Shangla As T = 0.06 mg/kg and Upper Dir As T = 0.08 mg/kg). The reason behind this may be due to higher level of pollution in Peshawar district. This gap need to be lled through further research that why non mining areas show higher concentration of As than the mining areas.
Both the natural and anthropogenic activities are the sources of As (Duan et al. 2013). Among the natural sources, volcanism and bed rocks weathering are important while among anthropogenic sources, metal mining and smelting, the use of As-pesticides, herbicides, feed additives, wood preservatives and irrigation with Ascontaminated water are important sources that results in elevated levels of As in soil (Zhao et al. 2009). One of the main pathways of human exposure to As is the transport of As in soil-plant systems (Dave 2013). Most of the samples in mining sites shows a higher trend of As and its different species as compared to non-mining sites. The concentration of As in soil of mining area was higher (0.624 mg/kg) than the non-mining area (0.096 mg/kg) in Hunan Province, China ). As is a natural constituent some ores such as ores of Cu, Pb, Zn, and Au. This is the reason why As is mostly found at elevated levels, in mining areas soils posing a risk to health of both ecosystem and human ). None of the district of the study area exceeds the acceptable range (0.5 mg/kg) of As T by (WHO 2011). Kwon et al., 2017 also stated that the values of As was high (0.247 mg/kg) in the rice of mining areas of Korea as compared to other lands.

As concentration in different rice types of Khyber Pakhtunkhwa
Arsenic concentration in different types of rice are presented in Fig 4,

Conclusion
The As speciation in rice plays an essential role in evaluating the human's toxicity. This study has focused on As speciation in rice, mechanism and the factors that can effect As uptake by rice. High rice consumption is the major exposure pathway of As in human beings. The rice in Pakistan is cultivated in paddy soils, which is one of the reason of As uptake by rice plants. The As 3+ is the most toxic specie of As, which is highly carcinogenic in nature. Different factors such as redox Fe-Plague effect, pH, OM effect, soil texture effect, phosphate, As bond to Fe-Mn oxides, irrigation practices, seasonal variations and the genotype effect could affect the bioavailability and solubility of As in rice plants. These factors resulted in increasing or decreasing the As uptake by rice plant. The total level of soil As and rice As samples ranged from 13.72 mg/kg to 4.31 mg/kg and 0.28 mg/kg to 0.06 mg/kg respectively. The study analysis showed that there are mainly 5 As species (As(III), DMA, MMA, As(V) and BAs) in the rice samples, with As i form being the most abundant and toxic species with concentrations of 0.23 mg/kg to 0.04 mg/kg. The hazardous effect of As on Khyber Pakhtunkhwa people cannot be neglected. The people of Khyber Pakhtunkhwa are at risk of exposure to adverse effects of As. Based on our study results, along with the risk assessment, we concluded that the values of ILTCR ranged from 4.1E-4 mg/kg/day to1.9E-3 mg/kg/d. While the said results are boosting further eld research studies are needed to explore the sources of As contamination in rice paddies. Also factors that are responsible for the higher level of As in rice grown in non-mining areas.
Declarations Figure 1 Inorganic and organic arsenic concentration in soil of selected Khyber Pakhtunkhwa districts    showing the values of EDI for the AsT, Aso and Asi through consumption of rice collected from selected districts of Khyber Pakhtunkhwa Figure 7 Showing the ILTCR for Asi through consumption of rice in selected districts of Khyber Pakhtunkhwa