QUALITY OF WELL WATER INTENDED FOR WATERING PIGS: ANALYSES OF BACTERIOLOGICAL PARAMETERS AND PESTICIDE CONTENT

Water used for watering pigs, originating from fi ve wells located in households in a settlement in Vojvodina, was inspected for its quality by the analysis of bacteriological parameters and pesticide content. Five samples were taken from each well at monthly intervals (n=25). In all water samples from three wells, coliform bacteria, including thermotolerant coliforms and Escherichia coli, indicators of faecal contamination, were repeatedly detected. In the water of all of the fi ve wells organophosphate pesticides were detected, whose maximum allowed concentrations are not defi ned by the current Serbian Regulations on the quality of drinking water. Given the quantity of water consumed by pigs on a daily basis, water contaminated with coliform bacteria and organophosphate pesticides poses a risk to their health. Diminishing resources of drinking water and the growing environmental pollution, renders the assessment of the quality of water intended for domestic animals necessary as part of the integrated system of management in agricultural holdings.


INTRODUCTION
Besides balanced diet, the most important prerequisite for animal health protection and welfare, and optimum production is the provision of ample fresh drinking water (Smith, 2020). Generally, a domestic animal can survive feed deprivation for two months, but without water merely a week. Insuffi cient watering leads to stress, poor performance and, eventually, to dehydration (Landefeld and Bettinger, 2003). Depending on animal species, category and age, water make up 44-88% of their body mass and is necessary for ingestion and feed digestion, dissolving and transport of inorganic and organic matter (both within the body and in order of excretion), biochemical processes, maintaining homeostasis, metabolite excretion, thermoregulation, reproduction and other vital functions of the organism (Manu and Baidoo, 2020). Fattening pigs (body weight 20-110 kg) need 3-12 L of water on a daily basis, depending on body weight (weaners 3-5 L, growers 5-7 L and fi nishers 9/12 L), boars and dry sows 12-15 L, and lactating sows even up to almost 50 L (24-45 L) (Dawson, 2020). Daily water intake may be considered a sensitive indicator of animal health and welfare, and is also closely related to feed consumption (AHDB, 2019; Bigelow and Houpt, 1988). Th e quality of water, infl uencing feed intake, feed conversion and growth rate, is of high importance for animals. Water intended for consumption must be of corresponding physical and chemical quality, and without harmful and toxic substances. Th e quality of water used for watering animals should correspond to the one which is used by humans (Radivojević, 2004), which is in the Republic of Serbia defi ned in the Regulations on the quality of drinking water (Offi cial Gazette of the FRY Nos. 42/98 and 44/99, and Offi cial Gazette of the RS No. 28/2019).
Water quality is infl uenced by a range of factors: sand or sludge particles, increased concentrations of chemical elements, contamination with microbes, pesticides, radionuclides etc. Th e most frequent microbial contaminants of drinking water are bacteria originating from soil and plants, septic tanks and sewers or fl ood waters, which regularly contain large numbers of microorganisms due surface layers of soil being washed away, or insects and rodents which may enter wells. Surface and underground waters are oft en contaminated with pesticides due to their wide use in plant protection, which are commonly used in higher concentrations/ quantities and more frequently than recommended by their instructions on use. Th ere is a scarcity of data on the microbiological quality of water and pesticide concentrations in water intended for animal watering. Given that water intake in pigs corresponds to 10% of their body mass, and is twice as much as ingested feed, it is surprising that so little attention has been devoted to research into the quality of water in comparison to that of feed. Th is prompted us to set the aim of this work -assessment of hygienic correctness of well water which is used for watering pigs by assessing bacteriological parameters and pesticide content.

Water samples
Water from fi ve dug wells located in the same settlement in the Autonomous Province of Vojvodina (Serbia) was subjected to research. Th e wells were located in separate households which grow and fatten pigs and the meat product use for their own needs and sell them on the local market. Th e wells were dug, (natural waters, open spring), and the water was obtained by turning on the taps. Th e water is used for watering pigs, maintaining the hygiene in pig houses and as technical water in households. Th e water was sampled fi ve times, in monthly intervals, from October 2019 to March 2020. Before each sampling, the plastic opening of the tap was disinfected with 70% ethanol and the water allowed to fl ow for 3-5 minutes at a steady rate. Th e water was sampled in sterile 500 mL glass bottles fi lled up to 1-2 cm below the top. Th e bottles were closed and transported to the laboratory at temperature <5C. Bacteriological assessment began on the arrival of water samples and the remaining samples kept for pesticide tests in a refrigerator.

Bacteriological assessment
In water samples the presence of the following bacteria were determined: the total number of coliform bacteria, Escherichia coli, faecal coliforms, intestinal enterococci, Pseudomonas aeruginosa and Clostridium perfringens. All analyses were done using the membrane fi ltration method, with sterile membrane fi lters with a sieve, diameter 47 mm and pore diameter 0.45 μm (Filter-Lab Gridded MCE Membrane Filter, Filtros Anoia, S.A., Barcelona, Spain). For each analysis 100 mL of water was fi ltered. Bacterium isolation and identifi cation was done with the following methods: -Enumeration of Escherichia coli and coliform bacteria (ISO 9308-1:2014). Th e fi lters were placed on the surface of Chromogenic Coliform Agar (ISO, CM 1205, Oxoid Ltd., Basingstoke, UK), and the plates incubated for 24 h at 37°C. To determine the total numbers of thermotolerant (faecal) coliforms, the plates were incubated for 24 h at 44°C. For species confi rmation, the oxidase test (negative) and the indole test (positive in E. coli) were used.
-Detection and enumeration of Pseudomonas aeruginosa (ISO 16266-2:2018). Th e membrane fi lters were placed on the surface of Pseudomonas Cetrimide agar (CM0579, Oxoid Ltd, Basingstoke, UK), and the plates incubated at 37°C for 44 h. For confi rmation the oxidase test and the failure of growth at 4°C aft er 5-day incubation were used.
Aft er the incubation ceased, the plates were inspected, confi rmation tests for identifi cation performed and the colonies characteristic of the targeted species counted. Th e results were expressed as the absence of bacteria, real number in 100 mL (CFU/100 mL) of water, or too numerous to determine their exact number (too numerous to count -TNTC).

Pesticide content
Water was tested for the presence of organochlorine pesticides (aldrin, dieldrin, endrin, endosulfan I, endosulfan II, endosulfan sulphate, p,p'-DDD, p,p'-DDT, p,p'-DDE, heptachlor, heptachlor epoxide, cis-chlordane, transchlordane, alfa BHC, beta BHC, lindane, delta BHC, methoxychlor, endrin aldehyde) and organophosphate pesticides (thionazin, sulfotep, phorate, dimethoate, disulfoton, methyl parathion, parathion). Th e samples were prepared by liquid-liquid extraction: mixing with an organic solvent (methylene chloride), separation of the organic phase, concentration of the extracts by vaporisation in nitrogen fl ow and reconstitution of the dry residue with nhexane (EPA 1699/2007). For the preparation of calibration solution, mixtures of standard solutions of organochlorine pesticides were used (PPM-808C-1, Agilent) and mixtures o f standard solutions of organophosphate pesticides (ERO-008, Supelco Analytical). To eliminate the infl uence of the matrix on the results, a calibration through matrix was done according to SANTE 12682 (EC 2019) document. As a blank sample, distilled water without pesticides was used. Th e analyses were done with the method of gas chromatography coupled to mass detector. Pesticide content was determined in Agilent system GC/MS. System GS 7890B was connected to the mass spectrometer of the mass selective detector 5977A. Th e mass spectrometer worked at the EI mode at 70 eV. A capillary column (30 m × 0.25 μm fi lm HP-5 M-thickness) which contains 5% diphenyl and 95% dimethylpolysiloxane (HP-5MS, Agilent Technologies, Inc., Santa Clara, CA, USA) was deployed. Data collection and processing was done using MassHunter Soft ware (Agilent, Santa Clara, CA, USA).

RESULTS AND DISCUSSION
Bacterial pollution of water has traditionally been expressed as the number of coliform bacteria in a certain water volume. Given that they are normal inhabitants of animal and human guts, their presence in water is an indica-  (Barcina et al., 1990), which mean that their presence is not necessarily related to faecal contamination (Leclerc et al., 2001). For this reason, later has been made a distinction of so-called "thermotolerant" coliforms as specifi c indicators of faecal contamination (Leclerc et al., 2001;WHO, 1997), from the total coliforms. Faecal coliform determinations should be complemented with the quantifi cation of enterococci (Cabral, 2010). Non-coliform species such as Pseudomonas aeruginosa (Fig 2.) and bacteria species of the Clostridium genus (Leclerc et al., 2001) have been used as additional indicators of the microbial safety of drinking water. In spite of being less specifi c indicators of faecal contamination than E. coli, Enterococci are more advantageous indicators of contamination: they are more resistant to disinfectants used for drinking waters, they survive better and may be transported further than E. coli and thermotolerant coliforms. For these reasons, enterococci may be detected even when E. coli cannot (Health Canada, 2020).
Th e recommendations by the Australian and New Zealand Environment and Conservation Council set the limit to ˂100 thermotolerant (faecal) coliforms per 100 mL of livestock drinking water (Davis, 2016). By contrast, the US EPA's recommendation for livestock water is <5,000 coliforms per 100 mL, but faecal coliforms ˂1 (Pfost and Fulhage, 2020).
Th e analysis of water samples (n=25) from fi ve wells detected coliform bacteria in 15 samples originating from three wells, but their precise numbers in fi ve samples could not have been determined (Table 1; Figure 1). Th ermotolerant coliforms, a sub-group of total coliforms, were isolated from six water samples, and in fi ve a thermotolerant Escherichia coli was detected, whilst Clostridium perfringens and faecal enterococci, additional indicators of faecal contamination, were not confi rmed. Due to the presence of coliform bacteria, especially of thermotolerant coliforms (including E. coli), the use of water from these wells for watering pigs could be estimated as a biological risk for their health. It could be of especial importance in warm periods of the year owing to the indirect/direct infl uence of temperature on a series of chemical and physical parameters of water and because temperature is the most important factor aff ecting bacterial growth (LeChevallier et al., 1996). Th e current research was done in autumn and winter (October 2019-March 2020), but E. coli and other enteric bacteria, being mesophilic, grow faster alongside the rise in water temperature above 15ºC (LeChevallier et al., 1996). Th us, higher incidence of animal water-borne infection with E. coli has been supposedly related to the increase in the bacterium numbers in water in summer (Hancock et al., 1994).
According to the Regulations on the quality of drinking water (Offi cial Gazette of the FRY, Nos. 42/98 and 44/99, and Offi cial Gazette of the RS No. 28/2019), natural waters in open springs (dug wells) may contain up to 10 CFU/100 mL of coliform bacteria (determined by the method of membrane fi ltration). In comparison with this criterion, the water in three wells (Nos. 1, 2 and 3) was of inadequate microbial quality. However, it remains unclear whether water intended for watering pigs has to abide the guidelines defi ned for water for human consumption. In literature diverse standards of acceptable water quality can be found, for instance: 1,000 total coliforms/100 mL and 50 Escherichia coli/100 mL in water intended for pigs (Edwards, 2018), or that water used for livestock should not contain more than 5,000 coliforms/100 mL (National Research Council, 1998). According to the Alltech factsheet on water quality in pig production, a total of 50 colony forming units (cfu) is acceptable, but numbers beyond 100 cfu/ml require treatment (Epp, 2019). However, E. coli water contamination was confi rmed as a reality by the assessment of 98 dairy farms in the USA, which revealed the average log10-transformed coliform and E. coli concentrations per milliliter of water 1.76 ± 1.25 and 0.98 ± 1.06, respectively (LeJeune et al., 2001). Th e contamination level of the water provided to livestock correlated positively with the proximity of the water containers to the feeders, protection of the water from direct sunlight and higher weather temperatures (LeJeune et al., 2001).      Water pollution with pesticides is a direct consequence of their use in agriculture. Given that they reach underground waters from the surface layers of soil, their quantities are infl uenced by the river water levels and water fl ow. Besides by utilizing organophosphorus pesticides, more frequently people may be exposed to them indirectly via water, food or air, which is why the methods of their detection are being continually improved (Liu et al., 2020).
Th e major mechanism of organophosphorus pesticide toxicity is the inhibition of cholinesterase enzyme (ChE), which has been proven for disulfoton to increase in the process of chlorination (Chu, 2020). ChE inhibition results in the accumulation of acetyl choline neurotransmitter and the blocking the nerve impulse transmission. Besides the cessation of neurotransmission, organophosphates and the products of their transformation (oxidation, isomerization and hydrolysis) also exert non-specifi c toxic action. During acute, and especially during long-lasting exposure to this pesticide group, reactive free oxygen species are produced in the organism, which attack lipids, proteins and DNA, lead to cell membrane damages, enzyme inactivation, genetic damages and, fi nally, to cell death. It has been known that chronic exposure of rats to small doses of phorate may impair energy-related metabolism and antioxidant system, and may cause liver, kidney and DNA damage (Sun et al., 2014). Increased use of agrochemicals, especially of pesticides, result in the spreading of harmful chemicals in the environment. It is especially important to determine pesticide quantities in underground and surface waters, which are the primary media for the transport of pesticides with low volatility.
Only scarce information is available on the criteria for water given to livestock (Valente-Campos et al., 2014). Little data on the hygienic characteristics of water intended for animal consumption is probably related to the fact that water supply was ample and inexpensive for considerable part of the world (FAO, 2011). In the European Union, current legislation prescribes that pigs must have permanent access to fresh water (EU, 2009), which is also stated in the Serbian Law on animal welfare (Offi cial Gazette of the RS, No. 41/09). However, water is becoming a critical resource for profi table swine production (Nyachoti and Kiarie, 2011). Natural water resources are progressively diminishing, and environmental pollution is becoming more intense. Generally, it can be stated that testing the quality of water intended for animal consumption is not done routinely by animal producers, who lack knowledge on the negative infl uence of low-quality water on the swine production. Given that various adverse eff ects may be produced by providing animals with water which does not meet adequate quality requirements (Valente-Campos et al., 2014), water quality assessment should be part of integrated system of management of agricultural holdings, especially when problems related to productive performance or health-related problems arise.

CONCLUSION
Th e scarcity of research into the safety and eligibility of well water intended for swine consumption results in the lack of precise knowledge on the infl uence of poor-quality water on pig rearing and health. In this research, coliform bacteria (including thermophilic coliforms and Escherichia coli) and organo-phosphorus pesticides were repeatedly detected in water samples originating from wells. Consequently, such waters can be considered a potential risk to pigs which consumed them. However, precise assessment/judgement cannot be done due to the fact that whether the criteria given in the Regulation on the drinking water quality (Offi cial Gazette of the FRY, Nos. 42/98 and 44/99, and Offi cial Gazette of the RS No. 28/2019) apply also to water intended for animal consumption. In addition to this, in these legislations the maximum admissible concentrations of organophosphorus pesticides have not been defi ned. Due to the progressive decline in available water resources and increasing environment pollution, it is necessary that the quality control of water intended for animal consumption be monitored and become part of the integrated management of agricultural holdings, and precise criteria for the assessment of water safety be established.

ACKNOWLEDGMENTS
Th is work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia and according to innovation voucher number 488.