Lihua Xiao, Altaf A. Lal, and Jianlin Jiang
Consumption of contaminated water has been implicated as a major source of Cryptosporidium infection in various outbreak investigations and case control studies. Surveys conducted in various regions of the United States demonstrated the presence of Cryptosporidium oocysts in 67-100% of wastewaters, 24-100% of surface waters, and 17-26.8% of drinking waters (1-4). The identity and human infective potential of these waterborne oocysts are not known, although it is likely that not all oocysts are from human-infecting Cryptosporidium species. Likewise, the source of the oocyst contamination is also not fully clear. Farm animals and human sewage discharge are generally considered to be the major sources of surface water contamination with C. parvum (5). Because Cryptosporidium infection is common in wildlife, it is conceivable that wildlife can also be a source for Cryptosporidium oocysts in waters (4). The presence of host-adapted Cryptosporidium spp. and genotypes makes it possible to develop molecular tools to assess the human infection potential and source of Cryptosporidium oocysts in water.
Currently, the identification of Cryptosporidium oocysts in environmental samples is largely made by the use of immunofluorescent assay (IFA) after concentration processes (Environmental Protection Agency [EPA] recommended information collection rule [ICR] method or method 1622/1623 or similar techniques). Because IFA detects oocysts from all Cryptosporidium parasites, the species distribution of Cryptosporidium parasites in environmental samples cannot be assessed. Although many surface water samples contain Cryptosporidium oocysts, it is unlikely that all these oocysts are from human-pathogenic species or genotypes, because only five genotypes of Cryptosporidium parasites (the C. parvum human and bovine genotypes, C. meleagridis, C. canis, and C. felis) are responsible for most human infections (6). Information on the source of C. parvum contamination is necessary for effective evalu ation and selection of management practices for reducing C. parvum contamination of surface water and the risk of cryptosporidiosis. Thus, identification of oocysts to species and genotype levels is of public health importance.
Molecular techniques, especially polymerase chain reaction (PCR) and PCR-related methods, have been developed and used in the detection and differentiation of Cryptosporidium parasites for many years (22). Most of these techniques were designed for the analysis of fecal samples from humans and domestic animals. Earlier PCR methods (7-10) do not have the ability to differentiate species and thus can only be used for determination of the presence or absence of Cryptosporidium parasites. The primer sequences of these techniques (except for that described in ref. 8) are mostly based on undefined genomic sequences from C. parvum bovine isolates. These sequences tend to be more polymorphic than structural and housekeeping genes; therefore the primers based on them are unlikely to amplify DNA efficiently from Cryptosporidium spp. (such as C. muris, C. baileyi, C. serpentis, C. canis, and C. felis) and from Cryptosporidium genotypes (such as the fox, skunk, and opossum) that are more distant from the C. parvum bovine genotype.
Several PCR-restriction fragment length polymorphism (RFLP)-based tools have been developed for the detection and differentiation of Cryptosporidium parasites at the species level (11-17). All these techniques are based on the small subunit (SSU) rRNA gene. Unfortunately, primers for some of these techniques (11,13,14) used conserved sequences of eukaryotic organisms. Therefore, these primers also amplify DNA from organisms other than Cryptosporidium (18). A PCR-RFLP analysis of the internal transcribed spacers of the rRNA gene can also differentiate several Cryptosporidium parasites (19). Nucleotide sequencing-based approaches have also been developed for the differentiation of various Cryptosporidium spp. (20,21).
Various PCR based techniques have also been developed for the differentiation of C. parvum human and bovine genotypes (22). Both genotypes of C. parvum have been identified in humans, but the human genotype (genotype 1 or anthroponotic genotype) has been almost exclusively found in humans, whereas the bovine genotype (genotype 2 or zoonotic genotype) infects humans, ruminants, and some other animals. As a result, the human genotype was recently renamed C. hominis. Many of the genotyping tools used in these studies, however, cannot detect and differentiate other genotypes of C. parvum and non-C. parvum Cryptosporidium spp. This has limited the utility of these tools in the analysis of environmental samples. Even for clinical samples, their usefulness is compromised by the failure to detect C. canis, C. meleagridis, and C. felis in fecal samples from infected humans. Using newer molecular tools, we and others have found these parasites in AIDS patients as well as in children (6,23-26).
Numerous attempts have been made to apply PCR techniques in detecting Cryptosporidium oocysts in water samples (27-40). Most of these studies used water samples seeded with Cryptosporidium oocysts, and various successes were reported. One major obstacle is the presence of PCR inhibitors in waters, which are coextracted with DNA and inhibit PCR amplification of the target DNA. This has greatly reduced the sensitivity of PCR detection of oocysts in various water samples. The PCR inhibitors can be removed by an immunomagnetic separation (IMS) procedure (8), which is also a component of method 1622 or 1623 recommended for the detection of Cryptosporidium oocysts in water by the US EPA. Still, the sensitivity (5-10 oocysts in reagent-grade water) of the single-round PCR format adopted in most of these techniques has limited the usefulness of PCR in the analysis of water samples. This is because most water samples are likely to have fewer than five Cryptosporidium oocysts. The usefulness of PCR techniques, nevertheless, has been demonstrated by the detection of Cryptosporidium oocysts in water samples from the 1993 outbreak in Milwaukee, using a Cryptosporidium genus-specific PCR technique (8).
We have successfully used one such technique, the SSU rRNA-based nested PCR-RFLP method, in conjunction with IMS for the detection and differentiation of Cryptosporidium oocysts in storm water, raw surface water, and wastewater. It is the only technique that has been extensively used in the analysis of water samples and specimens from animals and humans (16,17,41,42). In one study, we analyzed 29 storm water samples collected from a stream that contributes to the New York City water supply system after storms and found 12 wildlife genotypes of Cryptosporidium parasites in 27 samples. Twelve of the 27 PCR-positive samples had multiple genotypes. Six of the genotypes were traced to sources (C. baileyi from birds, an unnamed species from snakes, two genotypes from opossums, according to data at the time of publication of the paper, a genotype from deer, and a genotype from muskrat by as shown by more recent data). The rest were wildlife genotypes that have never been found in humans or domestic animals, suggesting that wildlife was a major contributor to Cryptosporidium oocyst contamination in storm water in the area studied. This finding was consistent with the environmental setting (the catchment area was forested and isolated from agricultural activities) of the sampling site (41).
In another study we conducted, analysis of raw surface water samples collected from different locations (Maryland, Wisconsin, Illinois, Texas, Missouri, Kansas, Michigan, Virginia, and Iowa) in the United States with the SSU rRNA-based PCR-RFLP technique produced quite different results. A total of 49 surface water samples were analyzed, and 16 of the samples produced positive PCR amplification. Only four Cryptosporidium genotypes (C. parvum bovine genotype, C. parvum human genotype, C. andersoni, and C. baileyi) were found, all of which are genotypes commonly found in farm animals and/or humans, indicating that humans and farm animals are major sources of Cryptosporidium oocyst contamination in these waters. Similar results were also obtained from wastewater samples. Three Cryptosporidium genotypes (C. parvum bovine genotype, C. andersoni, and C. muris) were found in 7 of 34 grab samples (10 mL each) of raw wastewater collected from a treatment plant in Milwaukee. As expected, the diversity of Cryptosporidium parasites found in source waters and wastewaters was much lower than that in storm waters (42).
Promising results in the genotyping of Cryptosporidium parasites in water samples have also been generated in studies conducted by others. Using heat shock protein (HSP)70 sequence analysis of cell culture/PCR-amplified products, Di Giovanni et al. (29) have found six sequence types of C. parvum in raw surface water samples and filter backwash water samples. Comparison of these sequences with the HSP70 sequences we collected from various Cryptosporidium spp. and C. parvum genotypes indicates that these sequences were from three genotypes (bovine, mouse, and human genotypes) of C. parvum (21), suggesting that farm animals, rodents, and humans are responsible for Cryptosporidium oocyst contamination in these waters. Similarly, studies conducted in Northern Ireland and Japan have shown the presence of the bovine genotype of C. parvum in surface water (43,44). The molecular tools (HSP70 and TRAP-C2-based PCR) used in these studies, however, have probably limited the range of Cryptosporidium parasites found in water. The spectrum of Cryptosporidium oocysts found in surface water samples from Massachusetts using a PCR-sequencing analysis of the SSU rRNA gene is more similar to what we had found previously in surface water (12).
The method we are introducing here is a PCR-RFLP analysis of the SSU rRNA gene, for which extensive data are available on genetic heterogeneity (16,17). This method builds on existing procedures (filtration of water samples, elution and concentration of filtrates, IMS isolation of Cryptosporidium oocysts) used in the popular Cryptosporidium detection method 1622/1623 and involves DNA extraction, PCR, and restriction digestion. Because of the multicopy nature of the SSU rRNA gene (five copies per sporozoite in the bovine genotype of C. parvum) and the nested format of the PCR, this technique is one of the most sensitive PCR tools for the detection of Cryptosporidium oocysts. It amplifies and detects all known and unknown Cryptosporidium parasites and differentiates the most common species and genotypes. This method can supplement other detection methods such as 1622 and 1623 to provide information on the distribution of Cryptosporidium species and genotypes in water samples and the likely source of contaminations.
1. Dynabeads® anti-Cryptosporidium kit (for the isolation of Cryptosporidium oocysts only; cat. no. 730.01 for 10 tests or cat. no. 730.11 for 50 tests), or Dynabeads® GC-Combo kit (for the isolation of both Cryptosporidium oocysts and Giardia cysts; cat. no. 730.02 for 10 tests or cat. no. 730.12 for 50 tests) (Dynal, Oslo, Norway).
2. Dynal Magnetic Particle Concentrators (Dynal MPC): Dynal MPC-S (cat. no. 120.20) and Dynal MPC-1 (product No. 120.01).
2.2. Supplies for DNA Extraction
QIAamp DNA mini kit: cat. no. 51304 (50 tests) or cat. no. 51306 (250 tests) (Qiagen, Valencia, CA).
2.3. Supplies for PCR-RFLP
1. Primary PCR primers.
a. Forward (F1): 5'-TTCTAGAGCTAATACATGCG-3'.
b. Reverse (R1): 5'-CCCATTTCCTTCGAAACAGGA-3'.
2. Secondary PCR primers.
a. Forward (F2): 5'-GGAAGGGTTGTATTTATTAGATAAAG-3'.
b. Reverse (R2): 5'-CTCATAAGGTGCTGAAGGAGTA-3'.
3. 10X PCR buffer with 15 mMMg2+, (cat. no. N808-0129, PE Applied Biosystems, Foster City, CA).
4. 100 mM dNTP (cat. no. U1240, Promega, Madison, WI). To make a 1.25 mM working solution, add 12.5 ^L of each dNTP to 950 ^L of distilled water. Store the working solution at -20°C before use.
5. Taq polymerase (cat. no. M2665, Promega).
7. SspI (cat. no. R0132L, New England BioLabs, Beverly, MA).
9. Ddel (cat. no R0175L, New England BioLabs).
3.1. Immunomagnetic Separation of Cryptosporidium Oocysts From Water Pellets
1. Process 10 L of water samples through the filtration, elution, and concentration steps, following method 1622 or 1623 of the US EPA (see Note 1).
2. Wash the concentrated water pellets in 15-mL polypropylene tubes twice with distilled water by centrifugation at 1500g for 10 min.
3. Equilibrate in 10X Buffer A and 10X Buffer B from the Dynabeads kit to room temperature.
4. Add 1 mL of 10X Buffer A and 1 mL of 10X Buffer B into the 15-mL tube containing washed sample. Use only 0.5 mL of the water concentrate if the pellet is bigger than 0.5 mL in volume.
5. Resuspend the beads fully by vortexing the vial for 10 s, and add 100 ^L of Dynabeads to the 15-mL tube.
6. Add distilled water to give a final volume of 10 mL.
7. Rotate at 15-20 rpm for 1 h at room temperature.
8. Prepare the 1X dilution of Buffer A. One milliliter of 1X Buffer A will be required for each sample.
9. At the end of incubation, capture the Dynabeads in the 15-mL tube using MPC-1, and decant the solution in the tube
10. Resuspend the Dynabeads with 1 mL of 1X Buffer A, and transfer the suspension into 1.5-mL microfuge tube.
11. Capture the Dynabeads in a microfuge tube using MPC-S, and decant the solution in the tube. The Dynabeads with bound Cryptosporidium oocysts will be used in DNA extraction (see Note 2).
3.2. DNA Extraction Using the QIAamp DNA Mini Kit
1. Add 180 ^L of Buffer ATL to a 1.5-mL microfuge tube containing IMS-isolated Cryptosporidium oocysts, and vortex for 30 s.
2. Freeze-thaw five times at -70°C (or on dry ice) and 56°C.
3. Add 20 ^L of proteinase K to the tube, vortex for 10 s, and incubate at 56°C overnight.
4. Add 200 ^L of Buffer AL to the sample, vortex, and incubate the tube at 70°C for 10 min.
5. Centrifuge at full speed to precipitate the undigested pellet.
6. Transfer the supernatant into a new 1.5-mL tube.
7. Add 200 ^L of ethanol to the sample and vortex for 15 s.
Carefully transfer the mixture to a QIAamp spin column without wetting the rim, and centrifuge the column at 6000g for 1 min.
Place the spin column in a clean 2-mL collection tube, and discard the tube containing the filtrate.
Add 500 ^L of Buffer AW1 without wetting the rim, and centrifuge at 6000g for 1 min. Place the spin column in a clean 2-mL collection tube, and discard the tube containing the filtrate.
Add 500 ^L of Buffer AW2 without wetting the rim, and centrifuge at full speed for 3 min. Place the spin column into a clean 1.5-mL microfuge tube, and discard the tube containing the filtrate.
Add 100 ^L of Buffer AE, and incubate the tube at room temperature for 1 min. Centrifuge the tube at 6000g for 1 min.
Save the filtrate containing DNA and the store the extraction at -20°C.
PCR-RFLP Analysis of the SSU rRNA Gene
a. Preparation of master mixture. For each PCR reaction, prepare the following (see Note 3): 10 ^L 10X Perkin-Elmer PCR buffer, 16 ^L dNTP (1.25 mM), 2.5 ^L F1 primer (40 ng/^L), 2.5 ^L R1 primer (40 ng/^L), 6 ^L MgCl2 (25 mM), 4 ^L Bovine serum albumin (10 mg/mL), 57.5 ^L Distilled water, and 0.5 ^L Taq polymerase, for a total of 99 ^L.
b. Add 99 ^L of the master mixture to each PCR tube.
d. Run the following PCR program: 94°C, 3 min; 35 cycles of 94°C for 45 min, 55°C for 45 min, and 72°C for 1 min; 72°C for 7 min; and 4°C soaking.
a. A. Preparation of master mixture. For each PCR reaction, prepare the following: 10 ^L 10X Perkin-Elmer PCR buffer, 16 ^L dNTP (1.25 mM), 5 ^L F2 primer (40 ng/^L), 5 ^L R2 primer (40 ng/^L), 6 ^L MgCl2 (25 mM), 55.5 ^L distilled water, and 0.5 ^L Taq polymerase for a total of 98 ^L.
b. Add 98 ^L of the master mixture to each PCR tube.
c. Add 2 ^L of the primary PCR reaction to each tube.
d. Run the following PCR program: 94°C, 3 min; 35 cycles of: 94°C for 45min, 58°C for 45min, and 72°C for 1 min; 72°C for 7 min; and 4°C soaking.
e. Run electrophoresis on a 1.5% agarose gel with 20 ^L of the PCR product.
Was this article helpful?