Introduction

Contamination of water resources due to sanitary conditions has been a risk factor for population health. Water plays an important role in the route of transmission of biological agents such as protozoa, bacteria and viruses, with emphasis on Cryptosporidium spp., causing diarrhea worldwide (Hlavasa et al. 2005; King et al. 2015). Even in countries with high standards of sanitation, river flooding can cause epidemics of gastroenteritis, when sewage systems overflow and pollute water ways used by humans (Gertler et al. 2015).

Cryptosporidium spp. are a genus of protozoa whose transmission occurs mainly through water, including treated water (Tzipori and Ward 2002; Sunderlanda et al. 2007; Xiao 2010).

The first description of the genus Cryptosporidium spp. and the identification of species of Cryptosporidium muris were made from samples obtained from gastric glands of mice, in 1907 (Tyzzer 1907). In 1912, the second species was described through an observation of the intestine of laboratory mice and this was identified as Cryptosporidium parvum (Tyzzer 1912).

Phylogenetic studies were carried out by several groups of researchers (Tzipori and Griffiths 1998; Morgan and Thompson 1998; Xiao et al. 1998, 1999; Adl et al. 2005) in order to standardize the taxonomy of Cryptosporidium, but there are still disagreements about some species that cause serious health problems. Among these species, C. hominis, C. parvum are related to clinical events in the infected individuals and, occasionally, a few other species, such as C. muris, C. suis, C. andersoni and some different genotypes of Cryptosporidium coming from deer and monkeys (Xiao and Feng 2008), which differ significantly in their molecular sequences but, as yet, have not been ascribed species status (Plutzer and Karanis 2009). Literature also shows that domestic and wild mammals can be sources for human infection and water contamination (Feng et al. 2008).

Lake et al. (2007) showed that C. parvum and C. hominis differ in their epidemiology, being C. hominis mostly associated with areas with high socioeconomic status individuals, children under the age of four and urban areas, while C. parvum risk factors were rural areas and association with water reservoirs, watersheds and groundwater sources. Thus, it is important removal of Cryptosporidium and Giardia by treatment process.

Detection of C. parvum and C. hominis in environmental samples is a very important step to reduce and/or prevent the presence of these pathogens in drinking or recreation water. The presence of oocysts from these coccidia in the environment suggests that humans and animals can acquire the infection through different transmission routes (Xiao et al. 2001; Guy et al. 2003; Franco et al. 2012). These pathogens can be transmitted by ingesting contaminated food and/or water with infected animal or human feces (Gomes et al. 2004; Appelbee et al. 2005).

In 1984, the first cryptosporidiosis outbreak in Texas (USA) occurred and killed 79 people. Later there were others: one in 1987, in Georgia (USA), affecting 13,000 people; and the largest outbreak that infected 403,000 people in Milwaukee (USA), in the year of 1993, caused by the presence of Cryptosporidium in the icy water of the region. In 1996, reports showed that 8705 individuals were affected in Saitama (Japan), where Cryptosporidium was detected in treated and untreated water (Mark and John 1994; Mackenzie et al. 1994; Karanis et al. 2007).

The most employed technique used for staining fecal smears for the general detection of parasites in feces is those of staining, which can be used to detect other coccidia, such as Isospora and Cyclospora. However, this technique can present limiting factors, such as the size of oocysts (4–8 μm), which may be confused with colored organic matter, thus demonstrating the importance of a well-trained technician (Fahey 2003; Gonzáles-Ruiz and Bendall 1985). This technique was used in combination with morphometry for Cryptosporidium detection in untreated water by Santos et al. (2010).

Immunological methods may offer some advantages over optical microscopy for detection of Cryptosporidium spp. oocysts. For example, direct immunofluorescence technique (DIF), which uses monoclonal antibodies conjugated with fluorescein isothiocyanate, which recognize specific epitopes present on the oocysts’ surface, has been widely used. This technique has high specificity (96–100 %) and sensitivity (98.5–100 %) in Cryptosporidium spp. oocysts detection on environmental samples (Jex et al. 2008), but it has the same disadvantage of leading to a large number of false negatives, depending on the number of oocysts in the sample.

The enzyme-linked immunosorbent assay (ELISA) is an indirect test used for qualitative determination of Cryptosporidium spp. antigens in feces. It is simple to perform and does not require direct observation (Ungar 1990; Anusz et al. 1990). In this technique, antibodies are used, which are not species-specific, being unable to distinguish between Cryptosporidium spp. species. Furthermore, it has low sensitivity and needs sample concentration.

Because of this, a variety of molecular techniques based on the polymerase chain reaction (PCR) has been developed for Cryptosporidium spp. detection in environmental samples (Soba et al. 2006; Soldan et al. 2006; Trotz-Williams et al. 2006; Santos et al. 2010).

Real-time PCR is a new technique that uses fluorescence to allow amplicon continuous monitoring throughout the reaction (Monis et al. 2005). It has been shown to be very useful in Cryptosporidium spp. detection and monitoring (Araújo et al. 2005; Carvalho-Almeida et al. 2005; Meireles et al. 2006; Francino et al. 2006; Huber et al. 2007; Souza et al. 2007; Thomaz et al. 2007; Volotão et al. 2007; Araújo et al. 2007; Gonçalves et al. 2006).

The aim of this study was to evaluate and compare immunological techniques of antigen detection and real-time PCR for the detection and differentiation of Cryptosporidium spp. in treated water samples of an area of the Central-West Region of Brazil.

Materials and Methods

Place of Study and Sample Collection

The treated water samples were collected in previously sterilized bottles, with a volume of 5 L. Samples were collected directly from the taps, at the entrance of the residences, in four points of the city of Goiânia. A total of 32 samples were collected during the second semester of 2013 (August–November). The first point was to the north, located at latitude 16°35′07.82″S and longitude 49°16′50.28″O; the second point was to the south, latitude 16º42′14.43″S and longitude 49°16′17.32″O; the third point was to the east, with latitude 16°39′43.02″S and longitude 49°13′39.95″O, and the fourth point was to the west, longitude 16°41′00.91″S and latitude 49°18′25.02″W (Fig. 1).

Fig. 1
figure 1

First point was to the north (P1), located at latitude 16°35′07.82″S and longitude 49°16′50.28″O; south (P2), latitude 16°42′14.43″S and longitude 49°16′17.32″O; east (P3), with latitude 16°39′43.02″S and longitude 49°13′39.95″O; and west (P4), longitude 16°41′00.91″S and latitude 49°18′25.02″W

Full size image

Sample Concentration

The samples were sent for analysis to the Protozoology Unit of the Institute of Tropical Pathology and Public Health (IPTSP/UFG) and the Laboratory of Molecular and Genetic Diagnosis (LDGM-ICB/UFG). Samples were concentrated through the 0.45-μm porosity nylon membrane technique, with 47 mm in diameter and positively charged, according to Silva et al. (2010). The concentrate was transferred to 1.5-mL microcentrifuge tubes and stored at −80 °C for later analysis (immunological and molecular) (Fig. 2).

Fig. 2
figure 2

Immunological and molecular methods used for the detection of Cryptosporidium spp. in treated water in the Central-West Region of Brazil

Full size image

Detection of Cryptosporidium spp. Antigen by DIF and ELISA

For the Cryptosporidium spp. investigation, the DIF technique was employed (kit MeriFluor® anti-Cryptosporidium/Giardia—Meridian Bioscience, Cincinnati, OH, USA) and, simultaneously, the quantification of oocysts was made (DIF quantitative) from 10 µL of the concentrate, according to the protocol developed by Palmateer et al. (1999).

As a comparative measurement, ELISA assay (kit RIDASCREEN®—Cryptosporidium) was performed using 10 µL of the concentrate. The DIF and ELISA tests were performed according to manufacturer instructions.

Detection and Differentiation of Cryptosporidium spp. by Real-Time PCR

DNA extraction was performed by MagMAX™ commercial kit (Ambion) and following the manufacturer’s recommendations. The extracted DNA and in natura samples were sent for molecular diagnosis by real-time PCR at the Institute of Microbiology Paulo de Góes in the Federal University of Rio de Janeiro (UFRJ). When appropriate, a new DNA extraction was performed using the FastDNA® kit (MP Biomedicals).

The systematic approach to performing the real-time PCR assay consisted of a combination of a duplex assay for the detection of Cryptosporidium spp. and C. parvum and a singleplex one for the detection of C. hominis (Jothikumar et al. 2008). In the duplex assay, target sequences were amplified using a specific pair of primers, based on the gene sequence that encodes the small subunit of the ribosomal RNA, JVAF (5′-ATGACGGGTAACGGGGAAT-3′) and JVAR (5′-CCAATTACAAAACCAAAAAGTCC-3′), and genus-specific probe with fluorescent dye and suppressor JVAP18S (CY5—5′-CGCGCCTGCTGCCTTCCTTAGATG-3′—BHQ2). In this same assay, the primers and probe used for C. parvum were, respectively, JVAGF (5′-ACTTTTTGTTTGTTTTACGCCG-3′), JVAGR (5′-AATGTGGTAGTTGCGGTTGAA-3′) and JVAGP2 (FAM—5′-ATTTATCTCTTCGTAGCGGCG-3′—BHQ1). In the singleplex assay, the primers used were JVAGF and JVAGR and the JVAGP1 probe (FAM—5′-ATTTATTAATTTATCTCTTACTTCGT-3′—BHQ3). The total volume of each reaction was 20 µL.

For each duplex reaction, we used 0.5 µL of the solution to 10 pmoL/µL of each primer, 1 µL of 2 pmoL/µL solution of each probe, 1 µL of ultrapure water and 10 µL of Platinum® Quantitative PCR SuperMix-UDG with ROX (Invitrogen). DNA samples were diluted 1:20 to avoid inhibition, and all reactions were run in triplicate, using at 5 µL of DNA and ultrapure water as nontarget control (NTC).

For the singleplex reaction, we used 0.5 µL of the solution to 10 pmoL/µL of each primer, 2 µL solution of 2 pmoL/µL probe, 1.2 µL of ultrapure water, 0.8 µL of MgCl2, 10 µL of Platinum® Quantitative PCR SuperMix-UDG with ROX (Invitrogen) and 5 µL of purified DNA solution at dilution of 1:20.

At the DNA amplification, using the thermocycler 7500 System (Applied Biosystems), there was an initial activation step at 50 °C for 2 min and pre-denaturation at 95 °C for 2 min, followed by 45 cycles that included a denaturation step for 10 s at 94 °C, hybridization for 33 s at 55 °C and extension for 20 s at 72 °C (Jothikumar et al. 2008).

Microbiological Diagnosis

An aliquot of the sample collected (100 mL) was used to qualitative detection of total/fecal coliforms, in accordance with the kit Alfakit-Tecnobac recommendations, which has a minimum detection limit of 60 CFU/100 mL. These data were compared with that obtained from the detection of Cryptosporidium spp. oocysts using the techniques described above.

Statistical Analysis

The positivity results were presented by f (absolute frequency) and % (percentage). The Spearman correlation test was used to analyze the correlation between the used techniques, sensitivity and specificity values, positive and negative predictive values and concordance tests. We also calculated the cost/benefit for each technique. All analyses were fixed at 95 % confidence, considering p < 0.05.

Results and Discussion

The methodologies used in this study enabled the observation of a high positivity for Cryptosporidium, demonstrating that these are useful tools in monitoring treated water, since the treatment processes are not able to completely eliminate the parasite (Pereira et al. 2008). Thus, Cryptosporidium monitoring in treated water could reduce the risks of contamination and outbreaks like those that occurred in developed countries. Therefore, researches with the goal of discovering new removal indicators of Cryptosporidium spp. oocysts have been carried out, examples of which are the use of polystyrene beads and algae (Le Chevallier and Norton 1992; Huck et al. 2002).

The standard method known to the detection of Cryptosporidium spp. in water is the method 1623 of the Environmental Protection Agency (EPA—USA), that used indirect immunofluorescence (DIF) as diagnostic. So, we used DIF in comparison with other tests, as ELISA and real-time PCR.

Of the 32 analyzed samples, 50.0 % (16/32) were positive for Cryptosporidium using real-time PCR technique, 28.1 % (9/32) using ELISA technique and 56.3 % (18/32) using the DIF technique.

In comparison between tests, it is observed in Table 1 that there is a correlation between the DIF and ELISA tests (p = 0.001), with 23 (71.9 %) concordant samples and 9 (28.1 %) discordant ones, for these techniques.

Table 1 Comparison of DIF and ELISA test results (positive and negative)
Full size table

It can be observed in Table 2 that there is no correlation of results between PCR and DIF tests (p = 0.699), because there were 17 (53.1 %) concordant samples and 15 (46.9 %) discordant ones.

Table 2 Comparison of DIF and PCR test results (positive and negative)
Full size table

DIF and real-time PCR showed higher positivity than the enzyme-linked immunosorbent assay (ELISA). This result is due to the fact that ELISA, although standardized by researchers and technicians and used for oocysts control in water, has a disadvantage of being slower and having low sensitivity (Stancari 2013; Ungar 1990). Thus, the DIF has the advantage of high sensitivity and specificity, and it is simple to perform, being effective for the detection of Cryptosporidium spp. in treated water (Jex et al. 2008).

PCR detected the specific species C. hominis, reinforcing that correct identification of the protozoan species, especially in this water, is of great importance to assess the risk of human infection, suggesting contamination of source waters by human sewage (or contamination after water treatment depending on the integrity of the distribution system).

In samples in which PCR was negative and DIF was positive, interfering action on the DIF could also have occurred, such as organic and inorganic substances capable of binding to antibodies and promoting a false-positive result (Rodgers et al. 1995).

Although DIF is recommended as the gold standard, it has some limitations, such as the use of antibodies that are not species-specific. The variations of PCR have been developed and standardized to detect species of Cryptosporidium in water (Soba et al. 2006; Soldan et al. 2006; Trotz-Williams et al. 2006; Hashimoto et al. 2006, Keshavarz et al. 2009; Doi 2009; Pilai 2009).

In all, 5.6 % of the samples were positive for bacterial indicators: 25 % (2/8) of samples were positive to the point P1, 75 % (6/8) for P2, 75 % (6/8) for P3 and 50 % (4/8) for P4. Correlation of microbiological indicators with the presence of Cryptosporidium spp. in analyzed samples is shown in Table 3. On the association of detection of Cryptosporidium spp. according to each used technique and the presence of fecal and/or total coliforms as indicators, it was observed that they are not correlated with the presence of Cryptosporidium spp. in the analyzed samples corroborating data (Ahmed et al. 2014). The largest discrepancy between the tests for the detection of Cryptosporidium and coliforms was observed with ELISA (40.63 %), and the higher concordance was observed between the results of real-time PCR, with 68.75 % agreement. The coliform presence in treated water is of concern, because it indicated a high percentage of samples outside the standards recommended by the Ordinance MS (Ordinance No. 2,914), which could expose consumers of this water to health problem risks.

Table 3 Microbiological correlation indicators at the presence of Cryptosporidium spp. in relation to the employed techniques
Full size table

Conclusions

There was no statistically significant difference between the techniques analyzed (DIF or RT-PCR) for the detection of Cryptosporidium spp. in environmental samples. From the methodology applied in this study, it was found that treated water that supplies the city of Goiânia (four regions) is contaminated with Cryptosporidium spp., denoting the importance of implementing techniques for protozoa detection, in the assessment of water quality for human consumption.