BACKGROUND : Ticks secrete a diverse mixture of secretory proteins into the host to evade its immune response and
facilitate blood-feeding, making secretory proteins attractive targets for the production of recombinant anti-tick
vaccines. The largely neglected tick species, Rhipicephalus zambeziensis, is an efficient vector of Theileria parva in
southern Africa but its available sequence information is limited. Next generation sequencing has advanced
sequence availability for ticks in recent years and has assisted the characterisation of secretory proteins. This study
focused on the de novo assembly and annotation of the salivary gland transcriptome of R. zambeziensis and the
temporal expression of secretory protein transcripts in female and male ticks, before the onset of feeding and
during early and late feeding.
RESULTS : The sialotranscriptome of R. zambeziensis yielded 23,631 transcripts from which 13,584 non-redundant
proteins were predicted. Eighty-six percent of these contained a predicted start and stop codon and were
estimated to be putatively full-length proteins. A fifth (2569) of the predicted proteins were annotated as putative
secretory proteins and explained 52% of the expression in the transcriptome. Expression analyses revealed that
2832 transcripts were differentially expressed among feeding time points and 1209 between the tick sexes. The
expression analyses further indicated that 57% of the annotated secretory protein transcripts were differentially
expressed. Dynamic expression profiles of secretory protein transcripts were observed during feeding of female
ticks. Whereby a number of transcripts were upregulated during early feeding, presumably for feeding site
establishment and then during late feeding, 52% of these were downregulated, indicating that transcripts were
required at specific feeding stages. This suggested that secretory proteins are under stringent transcriptional
regulation that fine-tunes their expression in salivary glands during feeding. No open reading frames were
predicted for 7947 transcripts. This class represented 17% of the differentially expressed transcripts, suggesting a
potential transcriptional regulatory function of long non-coding RNA in tick blood-feeding.
COCLUSIONS : The assembled sialotranscriptome greatly expands the sequence availability of R. zambeziensis, assists
in our understanding of the transcription of secretory proteins during blood-feeding and will be a valuable resource
for future vaccine candidate selection.
Additional file 1: Table S1. Specifications of R. zambeziensis library
preparation procedures and the size and number of sequence reads
before and after quality filtering. Table S2. Putative R. zambeziensis
orthologues of previously characterised R. appendiculatus proteins.
Table S3. Expression proportions of the highest contributing secretory
protein families during different feeding time points. Table S4. Differential
expression analysis between female and male R. zambeziensis ticks. Table S5.
Number of differentially expressed transcripts in the protein classes and
secretory protein families of R. zambeziensis during feeding.
Additional file 2: Table S6. Annotation of the R. zambeziensis
Additional file 3: Figure S1. Gene Ontology (GO) characterisation of
the R. zambeziensis transcriptome. Level 2 GO terms of cellular components,
molecular functions and biological processes were visualised using WEGO
(Web Gene Ontology Annotation Plot). These included 18,436 cellular
components, 20,487 biological processes and 9659 molecular functions.
Figure S2. KOG clustering of R. zambeziensis transcripts. In total, 9620 R.
zambeziensis transcripts were assigned to 25 Eukaryotic Clusters of
Orthologs (KOG) functional categories, of which 3814 were unique KOG
terms. Figure S3. Top 30 most abundant KEGG pathways identified in the
R. zambeziensis transcriptome. Four thousand eight hundred and sixty nine
transcripts were assigned to 338 I. scapularis Kyoto Encyclopedia of
Genes and Genomes (KEGG) pathways. Figure S4. Top 30 Pfam domain
occurrences in the R. zambeziensis predicted proteins. A total of 13,451
Pfam domains were observed in the R. zambeziensis proteins, of which
3601 were unique. Eight thousand and sixty one of the proteins contained
at least one Pfam domain.
Additional file 4: Table S7. Annotation of the predicted proteins of R.