Detection of Cryptosporidium spp. oocysts and Giardia spp. cysts in surface water destined for public supply in the state of Goiás, Brazil

The protozoa Cryptosporidium spp. and Giardia spp. are etiological agents responsible for the transmission of gastroenteritis, mainly due to the consumption of contaminated water. Their (oo)cysts are resistant to adverse environmental conditions, as well as to most conventional water treatment processes. Thus, the objective of this work was to evaluate the occurrence of (oo)cysts of these protozoans in surface water collected for human consumption in the state of Goiás, Brazil. Fifteen samples of raw water were collected to assess the occurrence of (oo)cysts of the protozoa using the Membrane Filtration method, in addition to turbidity and pH analyses. Recovery rates in tests with ultrapure water reached the USEPA (2012) criteria for Giardia (78.1% ± 0%) and for Cryptosporidium (60.6% ± 32.6%); however, recovery in raw water was lower due to turbidity. All samples (n = 15) were positive for Cryptosporidium spp. oocysts, with a maximum concentration of 250 oocysts/L, demonstrating that these protozoa are disseminated in the aquatic environment of the state of Goiás and pose a risk to public health. Due to the use of water sources for public consumption, it is recommended that public authorities and sanitation companies act to preserve and maintain water courses, carry out periodic monitoring of treatment plants that supply the Cerrado, Santana and São Manoel streams and improve existing treatment technologies. The results did not allow to infer whether animal load and grazing area promote an increase in contamination of the lotic aquatic systems.


INTRODUCTION
Waterborne diseases are a relevant public health concern. Indeed, around 829,000 deaths from diarrhea worldwide each year (WHO, 2019) have been related to the use of water contaminated by microorganisms from untreated domestic sewage (CHENG et al., 2009) or surface runoff from livestock farms (IMRE et al., 2017).
Among the protozoa transmitted by water, the genera Cryptosporidium (4.0-6.0 μm in diameter) and Giardia (8-18 μm in length by 5-15 μm in width), stand out as being widely distributed in developed and developing countries (COTRUVO et al., 2004). The outbreaks caused by these etiological agents are associated with severe diarrhea, and have been well documented (Table 1).
The occurrence of these outbreaks can be attributed to the high resistance of these organisms to the environment (KARANIS, 2011) and their high infectivity (THOMPSON, 2004).
Means of transmission begins with the ingestion of (oo)cysts via the fecal-oral route, through contact between people and between people and other animals; the ingestion of water contaminated by host feces (humans and other non-human animals), recreational activities, and contaminated surface water, in addition to eating contaminated food (RYAN;FAYER;XIAO, 2014;KARANIS, 2016).
Several studies have reported the occurrence of these protozoa in surface water destined for public supply in Brazil (ALMEIDA et al., 2015;DIAS et al., 2008;FREITAS et al., 2015;SATO et al., 2013;TOLEDO et al., 2017), indicating a risk for infection by gastrointestinal diseases to consumers and highlighting the need for special attention from public authorities.
Temporal and seasonal variation in the occurrence of these protozoa in the environment and in water supply systems justifies the importance of their monitoring since the absence of fecal indicators, such as E. Coli, does not guarantee the absence of Cryptosporidium spp. and Giardia spp., and, thus, are not considered standard indicators (HORMAN et al., 2004). Therefore, the objective of the present study was to analyze the occurrence of (oo)cysts of Cryptosporidium spp. and Giardia spp. in surface waters destined for public supply in the state of Goiás, Brazil.

Investigated area
The study area encompasses 15 surface water sources for public water supply in the state of Goiás, as described in Table 2 and spatially distributed as shown in Figure 1. The state is located in the Center-West Region of Brazil and has a territorial extension of 340,106,492 km 2 bordering the states of Mato Grosso do Sul, Mato Grosso, Tocantins, Bahia, Minas Gerais and Distrito Federal.
Water sources were selected according to the occurrence of Escherichia Coli in raw water, with an annual geometric average greater than 1,000 most probable number (MPN)/100 mL, as established by Annex XX of Consolidation Ordinance No. 5 (BRASIL, 2017), in at least four years between 2009 and 2018. Monitoring raw water quality using only indicators (E. Coli) is not sufficient to assess the occurrence of specific pathogens (HORMAN et al., 2004), and so the occurrence of pathogens in water courses must be assessed.

Parasitological analysis
The collections were made at the entrance of the treatment plants in August 2019, which is considered the dry period (May to September) in regions with a humid tropical climate, according to the classification of Köppen (1948). A sample of twenty liters of raw water was collected from each water source and stored in a polyethylene container that was previously cleaned with Tween 80 (0.1%).
The Membrane Filtration (MF) method was chosen due to the ease of implementation and the possibility of preserving (oo)cysts for future infectivity assays (MACIEL; SABOGAL-PAZ, 2016). Furthermore, the method has greater recovery efficiency than the cartridge filtration method (HSU et al., 2001), and reduced cost of implantation and analysis than standard methods recommended by USEPA (2012).
This reduced cost is due to the fact that most water treatment stations (WTS) already have the necessary equipment to perform the technique, which makes its implementation feasible in developing countries, and due to its wide use in research carried out in different regions of the world to evaluate the occurrence of these protozoa in water supply sources (BAUTISTA et al., 2018;NISHI et al., 2009;ALMEIDA et al., 2015).
Prior to quantifying (oo)cysts of the protozoa in raw water, the analytical quality control of the method employed was tested in triplicate to verify its reliability and sensitivity regarding the recovery of (oo)cysts in ultrapure and in natura raw waters from the Santana stream, with turbidity of 8.20 Nephelometric Turbidity Unit (NTU).
The samples were contaminated with 10 μL of protozoan inoculum from Merifluor ® Kit (Meridian Bioscience Diagnostics, Cincinnati, USA), with the average concentration of protozoans being quantified in two 10 μL aliquots of the inoculum.
The results were discussed in terms of descriptive statistics. (1) Where: To evaluate the contamination of the method, 1.0 L of ultrapure water was analyzed in duplicate, with subsequent concentration of in natura raw water, whose results should indicate the absence of cysts/ oocysts (CANTUSIO NETO, 2008).
Membrane Filtration was performed by the following steps: • concentration of (oo)cysts of protozoa through filtration of 2.0 L of each sample through a mixed cellulose esters membrane (Millipore ® ), with a diameter of 47 mm and nominal porosity of 3.0 μm; • elution to recover the (oo)cysts by scraping and washing the material retained on the membrane with elution solution Tween 80 (0.1%); • centrifugation and concentration of the material resulting from the elution by double centrifugation process for 15 minutes with a rotational force of 1,500 × g, as recommended by USEPA (2012); • slide preparation using the direct Immunofluorescence Assay (FA) technique with Merifluor ® Kit, according to the manufacturer's guidelines for the simultaneous detection of oocysts and cysts, with aliquots of 10 μL for each slide well and three aliquots for each sample; • reading of slides using a Zeiss Axio Imager M2 microscope for epifluorescence with 400X magnification, 450-490 nm excitation filter, and 520 nm barrier filter for detection and enumeration of (oo)cysts.
Fluorescence defined by bright apple-green-color was adopted to identify (oo)cysts: absence of pores or appendages; 8-18 μm in length (3)

Analysis of physical parameters
The variables turbidity and pH were analyzed according to Standard

Validation of the membrane filtration method
Results regarding the performance of the MF method with samples contaminated with Giardia spp. cysts and Cryptosporidium spp. oocysts are provided in Table 3 for ultrapure water, and   Recovery outside the range recommended by USEPA (2012) for raw water was reported by Canale (2014) and Maciel (2006). The latter associated the method in question with immunomagnetic separation (IMS) and obtained mean recoveries of 31.5% cysts and 5.7% oocysts.
To improve performance in recovery efficiency with raw water samples, a more voluminous aliquot can be added to each slide well, once that according to Maciel (2006), the variability of tests may be associated with low aliquot representativeness(10 μL) in relation to the final sediment volume (250-1,000 μL), mainly for techniques that do not use IMS.
In addition, a rotation energy greater than 1,500 × g (15 minutes) could be adopted, as verified by Clancy et al. (2000), who achieved best recovery performance for Cryptosporidium spp. with a gravitational force of 2,170 × g and a centrifugation time of 15 minutes.

Prevalence of (oo)cysts in surface water sources
Data on the occurrence of E. Coli in raw water (frequency and range), the percentage of pasture area and animal load in the catchment hydrographic basins are presented together, with the results of the physical and parasitological parameters, in Table 5. As it turns out, in the present scenario, (oo)cysts are distributed in the aquatic environment of the state of Goiás, suggesting a potential risk to the population's public health.
LD: limit of detection; MPN: most probable number; NTU: Nephelometric turbidity unit. According to Annex XX of Consolidation Ordinance No. 5 (BRASIL, 2017), the E. Coli microbiological parameter is an indicator of (oo)cysts of Cryptosporidium spp. and Giardia spp. in water resources; however, Grott et al. (2016) and Bastos et al. (2004) failed to find a significant correlation between the prevalence of these organisms when analyzing water from surface sources in the states of Santa Catarina and Minas Gerais, respectively.
According to Horman et al. (2004), the lack of statistical correlation can be partly associated with the survival time of each species in the aquatic environment, since Cryptosporidium spp. oocysts can survive for more than six months while cysts can survive for a maximum of two months (KARANIS; KOURENTI; SMITH, 2007).  (IMRE et al., 2017) as, according to Hachich et al. (2004), lotic environments have a higher incidence of protozoa than lotic aquatic systems. Lopes et al. (2017) found Concentrations of 16.7 oocysts/L and 8,3 cysts/L were detected at the Jatobá stream, which had the smallest pasture area (24.16%), five loads of E. Coli higher than that established by the standard and low densities of cattle (1.88 per hectare). These results indicate that the characteristics of land occupation positively influenced the parasitological quality of raw water (DIAS et al., 2008), due to the smaller area of pasture that was reflected in a smaller number of cattle in the region, since studies in other countries reported that a bovine is capable of releasing 50 to 3.9 × 10 5 (oo)cysts/g of feces (FARIZAWATI et al., 2005).
Among the basins studied, the highest cattle load (2.06) was observed for Boa Esperança stream, with an occurrence of 100 and 175 oocysts/L. It is noteworthy that this contamination is intensified in developing countries and rural areas due to untreated human and animal waste .
The results did not allow to infer whether the animal load in the hydrographic basin generated increased concentrations of protozoa.
According to Bagley et al. (1998), water sources that receive contributions of cattle feces and sewage discharge as a result of land use activities have 10 to 100 times higher concentrations of (oo)cysts. However, in the present study, the Barro Alto stream basin, which had the lowest cattle load Silva, D.P. & Scalize, P.S.
(0.77), had an occurrence of 58.3 oocysts/L. This effect may be related to soil management, reducing the carrying of (oo)cysts (CHUAH et al., 2016), and the presence of uninfected cattle in the catchment basins with lower prevalence of protozoa, which depends on the water source given to the animals and their age, since young animals are more frequently contaminated due to having less immunity (DIAS et al., 2008).
The presence of protozoa in supply sources (Figure 1) 2017), since the monitoring of raw water is an essential tool to control and monitor treated water , in addition to preventing outbreaks of gastrointestinal diseases, negative impacts on public health and ensuring public health security, as the water resource is destined for different uses (LEE et al., 2013).
In addition, seasonal monitoring aims to characterize the aquatic environment, since bacteria in the coliform group are not good indicators of the prevalence of protozoa (HORMAN et al., 2004) and due to the underestimation of risk of infection caused by punctual samples (HELLER et al., 2006), as performed in the present study.
The range of variation in the physical parameters of turbidity and pH were 2.11-16.30 NTU and 7.27-8.01, respectively. This range of turbidity values is in concordance with the drought period, as found by Vasco et al. (2011), who reported that the increased flow of Poxim river in the state of Sergipe, due to the rainy period, directly interferes with turbidity.
Evaluating existing studies on the theme revealed that the present study is the first to report the occurrence of Giardia spp. and Cryptosporidium spp. in water sources intended for public water supply in the state of Goiás. Therefore, there is a need for basic sanitation companies to establish routines for monitoring these protozoa, in order to prevent outbreaks of gastrointestinal diseases, as well as urgent action for the preservation and maintenance of the respective water courses.

CONCLUSION
From the results obtained in the present study, it can be concluded that: • The precision and initial recovery tests reached USEPA (2012) criteria for Giardia (78.1 ± 0%) and Cryptosporidium (60.6 ± 32.6%); • The precision and continuous recovery tests with in natura raw water did not present any sample in accordance with the international standard. This was due to the turbidity of samples, which requires more of membranes to filter samples and a greater final pellet volume; • The Olho D'água stream located in Morro Agudo de Goiás, had maximum concentrations of 250 oocysts/L for Cryptosporidium spp. and 116.67 cysts/L for Giardia spp. These values can be associated with the predominance of livestock activities in the hydrographic basin. However, the punctual sampling campaign did not allow to infer whether the animal load in the hydrographic basin generates increased concentrations of protozoa; • The Jabotá stream, with a low predominance of pasture area, presented less contamination of the watercourse by protozoa Cryptosporidium spp. and Giardia spp.; • Cysts and oocysts were detected in the state's aquatic environment at concentrations ranging from < LD to 116.67 cysts/L and between 8.33 and 250 oocysts/L, which represents a potential risk to consumer health considering that the intake of at least 10 cysts of Giardia spp. and one oocysts of Cryptosporidium spp. can initiate the infectious process. Therefore, periodic monitoring of water sources by service delegates is recommended in order to improve existing technologies of treatment plants to ensure sanitary security; • It is recommended that treatment plants that obtain water from Cerrado, Santana, and São Manoel streams monitor the occurrence of protozoa, as the predominance of pasture indicates a risk for cryptosporidiosis outbreaks.