Analysis of Babesia rossi transcriptome in dogs diagnosed with canine babesiosis

Abstract

Background: Canine babesiosis is a tick-borne disease causing detrimental health effects on the domestic dogs with huge economic impact on the owners. The most complicated form of canine babesiosis is caused by a pathogenic Babesia rossi parasite. Canine babesiosis induced by B. rossi still remains the cause of mortality and morbidity in South African dogs, yet, the transcriptomic and genomic information of this parasite species is still not available. The transcriptomic and genomic information is essential in the disease development and processes for the design of effective disease control strategies. Consequently, our understanding of the mechanisms underlying the pathogenesis of the different genotypes of B. rossi remains limited. A previous study suggested a relationship between the parasite genotype and the disease phenotype. To date, thirteen B. rossi genotypes have been identified and associated with diverse clinical signs in their hosts. Hence the aim of this study was to sequence RNA from samples representing B. rossi genotypes, 19, 29 and 31, in order to have insight on the overall transcriptome of this parasite and to establish if there would be significant differences among the genotypes. Methodology: To screen for B. rossi positive samples, total DNA was extracted from 20 blood samples collected from sick domestic dogs presented at the Onderstepoort Veterinary Academic Hospital (OVAH). Babesia rossi infections were confirmed using the PCR-Reverse Line Blot (RLB) hybridization assay. Further confirmation of infection status was done by amplification of the B. rossi Erythrocyte Membrane Antigen 1 (BrEMA1) gene in all the DNA samples using qualitative PCR (qPCR), followed by sequencing of PCR products. Subsequently, total RNA was extracted from the 20 B. rossi-infected blood samples collected from the same dogs in which DNA was extracted. Three samples representing B. rossi genotypes 19, 29 and 31 were selected for transcriptome analysis. RNA sequencing was performed using the Illumina HiSeq 2000 to allow transcriptome analysis. De novo assembly was performed independently for all three transcriptomes using the Trinity software. The unigenes generated from specific transcriptome assemblies were subjected to global functional annotation using Blast2GO version 2.8.0 software, followed by KEGG database for annotation of biological pathways, and DAVID version 6.7, for COG classification to predict and classify their functions. Results: The sample representing B. rossi genotype 31 was excluded in the transcriptome analysis due to low RNA mass, which usually compromises the quality of the library used in RNA sequencing.Thus, a total of 26 747 238 and 25 709 627 paired-end reads were obtained from B. rossi genotypes 19 and 29, respectively. De novo transcriptome assembly produced a total of 3019 unigenes, with an average length of 419 bp and N50 of 362 bp in B. rossi genotype 19, and 2727 unigenes with an average length of 441 bp and N50 of 362 in B. rossi genotype 29. A total of 1193 unigenes were common between B. rossi genotype 19 and 29, while 1828 unigenes were exclusively detected in B. rossi genotype 19; and 1534 were specific to B. rossi genotype 29. Between the two B. rossi genotypes, a total of 4553 unigenes were obtained, representing the overall B. rossi transcriptome. From the overall transcriptome, 12.3% (n=558) of the unigenes could be annotated with 53 different gene ontology (GO) functional categories. About 34% (n=1550) of the unigenes represented in the overall transcriptome mapped to 237 KEGG pathways and only 2.5% (114) could be annotated in the COG database. Conclusion: Although, there were no striking differences in the transcriptomes of B. rossi genotypes 19 and 29, this study presents the first transcriptomic resource for B. rossi, which will highly contribute to our genetic understanding of B. rossi and provide a platform for future gene expression studies. Hypothetical proteins identified in this study will require further characterization as they may have a critical role in the biology and pathogenicity of B. rossi parasite.