Differentiation of Cyclospora sp. and Eimeria spp. by Using the Polymerase Chain Reaction Amplification Products and Restriction Fragment Length Polymorphisms
Karen C. Jinneman(1,*), June H. Wetherington(2), Ann M. Adams(1), Janelle M. Johnson(2), Bradley J. Tenge(1), Ngoc-Lan Dang(1), and Walter E. Hill(1)
Seafood Products Research center(1) and Seattle District Laboratory(2)
Food and Drug Administration, Bothell, WA 98041
* Corresponding author: firstname.lastname@example.org
The Laboratory Information Bulletin (LIB) is a tool for the rapid dissemination of laboratory (or information) which appear to work. The user must assure him/herself by appropriate validation procedures the LIB methods and techniques are reliable and accurate for his/her intended use. Reference to any commercial materials, equipment, or process does not in anyway constitute approval, endorsement, or recommendation by the U.S. Food and Drug Administration.
Cyclospora cayetanensis has of late gained prominence as a waterborne and foodborne pathogen (Ortega et al., 1993; Huang et al., 1995). Recent outbreaks (Chambers et al., 1996), as well as the finding of C. cayetanensis on vegetables (Roxas et al., 1996) have stimulated interest in the development of methods to detect this organism when recovered from foods.
A polymerase chain reaction (PCR) based test to detect Cyclospora has been developed (Relman et al., 1996 and Yoder et al., 1996). This method amplifies a region of the Cyclospora 18S ribosomal RNA gene, but also yields an amplicon of the same molecular size with the closely related genus of coccidian parasite, Eimeria (Relman et al., 1996). Eimeria spp. may infect a wide range of non-human animal hosts. The ability to distinguish between an amplified product from Cyclospora and Eimeria spp. will be important for the application of this PCR based test to food and environmental samples.
We sought to develop a method to differentiate the amplicons produced from these two genera and examined the nucleotide sequences of these amplified regions, as deposited in GenBank®. Although these nucleotide sequences of Cyclospora sp. and Eimeria spp are 94-96 percent similar, there are specific nucleotide differences in the amplified segment that can be used to differentiate these amplicons by digestion with the restriction endonuclease Mnl I.
CYCLOSPORA AND EIMERIA STRAINS:
Cyclospora sp. oocysts were received in 2.5% potassium dichromate from Dr. Ynés Ortega (University of Arizona, Tucson, AZ). Cyclospora sp. DNA was received from Drs. Norman Pieniazek and Susan Slemenda (CDC, Atlanta, GA) via Dr. Gene LeClerc (CFSAN, Washington, DC). Oocysts from three strains of Eimeria tenella (E. tenella #10, E. tenella #80, and E. tenella Merck) and one strain of E. mitis were received frozen in distilled water from Dr. Patricia Allen (USDA, Beltsville,MD).
PCR TEMPLATE DNA PREPARATION:
PCR template DNA was released from oocysts by a freeze and thaw method (Johnson et al., 1995). Briefly, 5-50 oocysts were washed 4 times with an equal volume of 1X PCR Buffer I (Perkin-Elmer, Norwalk, CT) by centrifugation at 14,000 rpm for 3 minutes and resuspension in the same volume of 1X PCR Buffer I. Next, 20 µl of a 6 percent resin matrix (Instagene, BioRad, Hercules, CA) was added to the suspension and vortexed well. This mixture was then subjected to six freeze/thaw cycles of 2 minutes in liquid nitrogen followed by 2 minutes in a 98°C water bath. The thawed preparation was mixed by vortexing, then centrifuged for 3 minutes at 14,000 rpm. The supernatant was retained for analysis. In the case of the prepared Cyclospora DNA template, 1 µl of a 1:100 dilution was used directly in the PCR amplification.
Primers were as described by Relman et al. (1996) except that they were synthesized without the sequencing "leader" by GenoSys (The Woodlands, TX) resulting in a predicted PCR product of 294 basepairs rather than 308 basepairs as reported by Relman et al. (1996). The sequences are listed in Table 1.
|CYCF1E (modified)||TACCCAATGAAAACAGTTT||1st PCR - forward|
|CYCR2B (modified)||CAGGAGAAGCCAAGGTAGG||1st PCR - reverse|
|CYCF3E (modified)||CCTTCCGCGCTTCGCTGCGT||2nd PCR - forward|
|CYCR4B (modified)||CGTCTTCAAACCCCCTACTG||2nd PCR - reverse|
PCR AMPLIFICATION CONDITIONS:
Basically, a nested PCR (in which an amplified product serves as the template for a second PCR which uses primers that are closer together) was conducted as described by Relman et al. (1996). Conditions for the first PCR are shown in Table 2 and the thermal cycler parameters are listed in Table 3.
The second round of PCR uses identical components and volumes with the exception that 5 µl each of a 4 mM working solution of primers CYCF3E (modified) and CYCR4B (modified) are used and the annealing temperature is 60°C. For the template, 5 µl of a 10-1 dilution of the first PCR is used.
|10X PCR Buffer I||10.0||50 mM KCl, 10 mM Tris-HCl, pH8.3, 0.001% gelatin, 1.5 mM MgCl2|
|dATP, dCTP, dGTP, dTTP mix||8.0||200 µM each dNTP|
|MgCl2; 25 mM||2.0||0.5 mM (Total 2.0 mM)|
|Primer CYCF1E (modified); 4 mM||5.0||0.2 µM|
|Primer CYCR2B (modified); 4 mM||5.0||0.2 µM|
|AmpliTaq polymerase||0.3||1.5 Units|
|Template DNA||5.0||5-50 oocysts|
|Initial Inactivation/Denaturation: 1 cycle||95°C/5 min|
|Amplification: 35 cycles||Denaturation||94°C/30 sec|
|Final extension: 1 cycle||72°C/10 min|
DNA SEQUENCE ANALYSIS:
DNA sequences were analyzed using the Genetics Computer Group ver. 8.0 suite of nucleotide sequence analysis programs (Devereux et al., 1984) as implemented on the CFSAN VAX. A STRINGSEARCH was conducted to identify Cyclospora entries and yielded the 18S rDNA sequence (accession number U40261) submitted by Relman et al. (1996) which was recently corrected (D. Relman, personal communication). Next a FASTA search using U40261 as a query sequence was conducted to identify closely related sequences. The Cyclospora sp. (U40261), E. tenella (U40264), E. mitis (U40262), and E. nieschulzi (U40263) sequences were then aligned using PILEUP; the output of which was displayed using PRETTY to emphasize sequence differences. Restriction endonuclease sites within the nested PCR amplicon were found using MAP. The restriction sites were compared to find differences between Cyclospora and Eimeria. The restriction endonuclease Mnl I was selected because it provided one unique cut site for Cyclospora sp., a different unique cut site for E. tenella and E. mitis and one cut site in common for all four sequences. The predicted Mnl I restriction fragments produced from digestion of the PCR amplicon are illustrated in Figure 1. E. nieschulzi would be predicted to be cut only at the 872 cut site thereby producing only two restriction fragments of 106 bp and 188 bp.
Figure. 1. Molecular sizes of observed restriction fragments from Cyclospora sp., E. tenella and E. mitis PCR product digested with Mnl I. Base pair cut sites based on Gen Bank sequence #U40261. mCYCF3E and mCYCR4B refer to the modified sequences of the primers listed in Table 1.
RESTRICTION ENDONUCLEASE DIGEST CONDITIONS:
Fifty ml restriction digests were prepared using 10 ml of PCR amplified DNA from each Cyclospora and Eimeria strain, one unit of the restriction endonuclease Mnl I (Amersham Life Sciences Inc., Arlington Heights, IL) and 5 ml 10X Buffer M supplied with the restriction enzyme. A digest including l DNA was also run to demonstrate complete digestion by the enzyme. The restriction digests were incubated 1 hr at 37°C.
AGAROSE GEL ELECTROPHORESIS:
Restriction endonuclease digests were analyzed on 4% NuSieve 3:1 or GTG agarose (FMC, Rockland, ME) gels prepared with Tris Borate EDTA (TBE) buffer. Ten ml of a restriction endonuclease digest was mixed with 2 ml of loading buffer (0.25% bromphenol blue, 0.25% xylene cyanole and 30% glycerol) and the entire volume was loaded into a well on the gel. Alternate lanes contained a molecular size standard ladder (BioMarker Low, BioVentures 101, Murfreesboro, TN). Gels were electrophoresed at 5 volts/cm for 3 hr or until the first dye front was approximately 1 cm from the end of the gel. The gel was post-stained in TBE containing 1 mg/ml ethidium bromide (Sigma Chemicals, St. Louis, MO) for 10 to 15 min and destained in deionized water for 1 to 5 min. The gel was placed on a UV transilluminator and photographed with Type 667 (1 second at f4.5) and Type 665 film (50-60 seconds at f4.5). The negatives from Type 655 film were developed as per manufacturer's instructions (Polaroid Corp., Cambridge, MA).
Images were acquired by transmission densitometry from the Polaroid Type 665 film negatives with a gel scanner (Sharp JX-325 with film scan unit) as *.tif bitmap images (Adobe Photoshop v. 3.0), Figure 2. Gel analysis software (RFLPScan 3.0 (beta) Scanalytics Inc., Billerica, MA) was used to analyze gels with user interactive algorithms for the band (peak) detection and relative peak area (% integrated optical density, OD) calculations. Band detection parameters of lane width of 39 - 29.66, band height threshold of 6 - 4, and smoothing operator pixel length of 2 - 50, allowed for the automatic detection of all peaks (Table 5). The molecular size standards were entered and properly associated with the standard peaks for the lanes containing the standards. The lanes were calibrated using the RFLPScan "de-smile" method with external lane standards and log piecewise linear regression. Band data were exported to EXCEL 5.0 (Microsoft, Redmond WA), Table 5. The band position analysis was also performed by measuring the band migration distances from the bottom of the gel well to the nearest 0.25 mm and using the SeqAid II program ver. 3.81 (D.J. Roufa, Manhattan, KS).
RESULTS AND DISCUSSION:
Distinct RFLP banding patterns were discernible by visual examination of the gel photograph (Figure 2). The molecular basepair sizes of the restriction fragments generated by restriction endonuclease digestion of the PCR products from each of the Cyclospora and Eimeria strains tested were all within ± 5% of the sizes predicted from the GenBank® sequence data (Tables 4 and 5). Only data from the RFLPScan analysis is compiled in Table 5. DNA fragment size data calculated by using SeqAid were also within ± 5 % of the predicted sizes (data not shown). DNA fragment size estimates were obtained from duplicate experiments with the exception of the E. tenella data which was done in triplicate.
These 2 closely related organisms were easily distinguished by the RFLP pattern of the PCR-amplified products. Although no E. nieschulzi strains were available to be tested they would be predicted to cut at only one site. A third distinct RFLP pattern consisting of only 2 fragments would be produced and thus could also be distinguished from the human pathogen, Cyclospora. It is not known whether other Eimeria spp. would also produce a PCR product of the same molecular size and, if so, what the RFLP pattern would be.
Figure 2. Restriction fragments from Mnl I digestion of PCR amplified products from Cyclospora and Eimeria spp.. A. Gel lanes are as follows: Molecular size standards in lanes 1, 3, 5, 7, 9, and 11, Cyclospora sp.(U of A), lane 2; Cyclospora (CDC), lane 4; Eimeria tenella #10, lane 6; E. tenella #80, lane 8; E. tenella Merck, lane 10 and E. mitis, lane 12. B. Composite gel drawing illustrating RFLP patterns for Cyclospora sp., lane 4, molecular size standards, lane 5, and E. tenella, lane 6.
|Organism||Predicted Fragment Sizes: |
|Predicted Fragment Sizes:|
|Predicted Fragment Sizes: |
|E. tenella or E. mitis||127||106||62|
The PCR method developed by Yoder et al., 1996, provides a sensitive molecular technique which may be applied to a variety of food and environmental samples for the detection and identification of Cyclospora spp. The PCR procedure does not produce an amplified fragment from other closely related coccidian species such as Cryptosporidium parvum and Toxoplasma gondii (Relman et al., 1996). However, a PCR product of the same molecular size is generated from Eimeria spp. by this PCR procedure. The Eimeria genera is composed of a number of species each of which infects different host species which include birds and mammals (Noble and Noble, 1971; Levine, 1985).
The application of this PCR method to the analysis of food and environmental samples could produce an amplified product that could have been derived from either Cyclospora or Eimeria spp. The ability to distinguish between the emerging human pathogen Cyclospora and possible contamination by members of the genus Eimeria will be essential for any analysis of food products. The use of the RFLP protocol described here could provide a simple, useful tool for the confirmation of an amplified product as being from Cyclospora. The limitation of this method is the small number of strains tested. The amplified region from 15 different Cyclospora strains have all yielded an identical genetic sequence (Relman personal communication) which would result in the same Mnl I RFLP pattern observed from the 2 strains tested here. This uniformity provides encouraging support for the utility of this approach for the distinction between Cyclospora and Eimeria. We plan to continue to test additional strains of both Cyclospora and Eimeria spp. and study additional approaches for the application of the PCR - RFLP method for the analysis of food samples.
We thank specifically Marleen Wekell, Ph.D. , John Wiskerchen, and Richard Rude (SEA-DO) and Elizabeth Keville (DFS, Rockville, MD) for their encouragement and support for this project. Ynés Ortega, Ph.D. (University of Arizona, Tucson, AZ) and Patricia Allen, Ph. D. (USDA, Beltsville, MD) kindly donated oocysts of Cyclospora and Eimeria, respectively. David Relman, M.D., Ph.D. (VA Health Care System, Palo Alto, CA) generously provided unpublished data and observations. We are indebted to numerous other members of SEA-DO for their many indirect but essential contributions required within very short time frames.
Chambers, J., Somerfeldt, S., Mackey, L., Nichols, S., Ball, R., Roberts, D., Dufford, N., Reddick, A., Gibson, J. (1996) Outbreaks of Cyclospora cayetanensis infection -- United States, 1996. MMWR 45: 549-551.
Devereux, J., Haeberli, P., and Smithies, O. (1984) A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12: 387-395.
Huang, P., Weber, J.T., Sosin, D.M., Griffin, P.M., Long, E.G., Murphy, J.J., Kocka, F., Peters, C. and Kallick, K.C. (1995) The first reported outbreak of diarrheal illness associated with Cyclospora in the United States. Ann. Intern. Med. 123: 409-414.
Johnson, D.W., Pieniazek, N.J., Griffin, D.W., Misener, L. and Rose, J.B. (1995) Development of a PCR protocol for sensitive detection of Cryptosporidium oocysts in water samples. Appl. Environ. Microbiol. 61: 3849-3855.
Levine, N.D. (1985) Phylum II: Apicomplexa. In Illustrated Guide to the Protozoa, J.J. Lee, S.H. Hunter, and G.C. Boure, (eds.), Society of Protozoologist, Lawrence, KS, pp. 319-350.
Noble, E.R., and Noble, G.A. (1971) Parasitology: The Biology of Animal Parasitology, 3rd ed., Lee and Febiger, Philadelphia. PA, pp. 84-86.
Ortega, Y., Sterling, C.R., Gilman, R.H, Cama, V.A, and Diaz, F. (1993) Cyclospora species -- a new protozoan pathogen of humans. N. Engl. J. Med. 328: 1308-1312.
Relman, D.A., Schmidt, T.M., Gajadhar, A., Sogin, M., Cross, J., Yoder, K.l., Sethabutr, O., and Echeverria, P. (1996) Molecular phylogenetic analysis of Cyclospora, the human intestinal pathogen, suggests that it is closely related to Eimeria species. J. Infect. Dis. 173:440-445.
Roxas, C., Miller, N., Cabrera, L., Ortega, Y., Gilman, R., and Sterling, D. (1996) Vegetables as a potential transmission route for Cyclospora and Cryptosporidium. Abstracts of the Annual Meeting of the American Society for Microbiology, C-102, p.19.
Yoder, K.E., Sethabutr, O., and Relman, D.A. (1996) "PCR-based detection of the intestinal pathogen Cyclospora" in PCR Protocols for Emerging Infectious Diseases, a supplement to Diagnostic Molecular Microbiology: Principles and Applications. D.H. Persing (ed.), ASM Press, Washington, DC, pp.169-176.