Background: Cyclospora cayetanensis is a food-borne intestinal human parasite that causes outbreaks of diarrhea. There is a need for efficient laboratory methods for strain-level characterization to assist in outbreak investigations. By using next generation sequencing, genomic sequences can be obtained and compared to identify potential genotyping markers. However, there is no method available to propagate this parasite in the laboratory. Therefore, genomic DNA must be extracted from oocysts purified from human stool. The objective of this study was to apply optimized methods to purify C. cayetanensis oocysts and extract DNA in order to obtain high-quality whole genome sequences with minimum contamination of DNA from other organisms. Results: Oocysts from 21 human stool specimens were separated from other stool components using discontinuous density gradient centrifugation and purified further by flow cytometry. Genomic DNA was used to construct Ovation Ultralow libraries for Illumina sequencing. MiSeq sequencing reads were taxonomically profiled for contamination, de novo assembled, and mapped to a draft genome available in GenBank to assess the quality of the resulting genomic sequences. Following all purification steps, the majority (81-99%) of sequencing reads were from C. cayetanensis. They could be assembled into draft genomes of around 45 MB in length with GC-content of 52%. Conclusions: Density gradients performed in the presence of a detergent followed by flow cytometry sorting of oocysts yielded sufficient genomic DNA largely free from contamination and suitable for whole genome sequencing of C. cayetanensis. The methods described here will facilitate the accumulation of genomic sequences from various samples, which is a prerequisite for the development of typing tools to aid in outbreak investigations.

BACKGROUND: Cyclospora cayetanensis is an emerging coccidian parasite that causes endemic and epidemic diarrheal disease called cyclosporiasis, and this infection is associated with consumption of contaminated produce or water in developed and developing regions. Food-borne outbreaks of cyclosporiasis have occurred almost every year in the USA since the 1990s. Investigations of these outbreaks are currently hampered due to lack of molecular epidemiological tools for trace back analysis. The apicoplast of C. cayetanensis, a relict non-photosynthetic plastid with an independent genome, provides an attractive target to discover sequence polymorphisms useful as genetic markers for detection and trace back analysis of the parasite. Distinct differences in the apicoplast genomes of C. cayetanensis could be useful in designing advanced molecular methods for rapid detection and, subtyping and geographical source attribution, which would aid outbreak investigations and surveillance studies. METHODS: To obtain the genome sequence of the C. cayetanensis apicoplast, we sequenced the C. cayetanensis genomic DNA extracted from clinical stool samples, assembled and annotated a 34,146 bp-long circular sequence, and used this sequence as a reference genome in this study. We compared the genome and the predicted proteome to the data available from other apicomplexan parasites. To initialize the search for genetic markers, we mapped the raw sequence reads from an additional 11 distinct clinical stool samples originating from Nepal, New York, Texas, and Indonesia to the apicoplast reference genome. RESULTS: We identified several high quality single nucleotide polymorphisms (SNPs) and small insertion/deletions spanning the apicoplast genome supported by extensive sequencing reads data, and a 30 bp sequence repeat at the terminal spacer region in a Nepalese sample. The predicted proteome consists of 29 core apicomplexan peptides found in most of the apicomplexans. Cluster analysis of these C. cayetanensis apicoplast genomes revealed a familiar pattern of tight grouping with Eimeria and Toxoplasma, separated from distant species such as Plasmodium and Babesia. CONCLUSIONS: SNPs and sequence repeats identified in this study may be useful as genetic markers for identification and differentiation of C. cayetanensis isolates found and could facilitate outbreak investigations.

Illumina library preparation methods for ultra-low input amounts were compared using genomic DNA from two foodborne parasites (Angiostrongylus cantonensis and Cyclospora cayetanensis) as examples. The Ovation Ultralow method resulted in libraries with the highest concentration and produced quality sequencing data, even when the input DNA was in the picogram range.

The parasite Cyclospora cayetanensis causes foodborne diarrheal illness. Here, we report draft genome sequences obtained from C. cayetanensis oocysts purified from a human stool sample. The genome assembly consists of 865 contigs with a total length of 44,563,857 bases. These sequences can facilitate the development of subtyping tools to aid outbreak investigations.