In-depth comparative analysis of Tritrichomonas foetus transcriptomics reveals novel genes linked with adaptation to feline host

Abstract

Tritrichomonas foetus is a flagellated parasite able to infect cattle, cats, and pigs. Despite its prevalence, feline tritrichomonosis has received markedly less attention than venereal infection, and little information about the molecular mechanisms that participate in feline host infection is available. Through a bioinformatics approach, we integrated public transcriptomic data for three T. foetus isolates and explored the differences at transcript level with a focus on pathogenesis and adaptation processes, particularly for the feline isolate. Our analysis revealed higher abundance levels of predicted virulence factors, such as proteases and surface antigens. Additionally, by a comparative and expression analysis of T. foetus genes, we proposed putative virulence factors that could be involved in feline infection. Finally, we identified a great proportion of predicted transcription factors of the MYB protein family and, by a promoter analysis, we revealed that MYB-related proteins could participate in the regulation of gene transcription in T. foetus. In conclusion, this integrated approach is a valuable resource for future studies of host-pathogen interactions and identifying new gene targets for improved feline tritrichomonosis diagnosis and treatment.

Figure 1

Exploring Tritrichomonas foetus transcriptome assembly. (A) Molecular functions and biological processes assigned to new assembled transcripts in this work; Percent: Counts of gene ontology terms represented as percent values for biological process or molecular functions; BP: Biological Processes, MF: Molecular Functions (B) Enzyme Commission number (EC) annotation; Percent: Counts of EC represented as percent values for each EC class; (C) Principal component analysis for the transcripts abundance (FPKM) data set. Colored points represent a dimensional reduction of transcript expression data for each isolate; (D) A Venn diagram that highlights differences and similarities between isolates at transcript abundance levels.

Figure 2

Specific abundant transcripts in G10/1 isolate are related to Myb-like and BspA-like proteins. (A) A multiple sequence alignment of predicted amino acid sequences for G10/1 MYB transcripts and E. histolytica (XP_648148.1) and T. vaginalis (XP_001307180.1); these are MYB validated proteins. Black arrowheads highlight conserved tryptophan (W) residues characteristic from the MYB family. At bottom, a schematic representation of second structure prediction by JPred algorithm, JNetPRED refers to the consensus prediction. Red bars represent alpha helix and sheets are presented as green arrows; (B) Schematic description of predicted amino acid sequences for G10/1 isolate transcripts related to BspA-like proteins. TpLRR: Treponema pallidum leucine rich repeat.

Figure 3

Analysis of predicted pathogenic and adaptive factors in the three isolates. (A) A correlation heatmap for the Spearman's coefficients for abundance values (FPKM) of predicted virulence factors analyzed in this work; Corr = Correlation coefficients code. Values are coded from light blue (− 1) to deep orange (1). Scatter plots for analyzed factors, those mostly abundant (fc > 4) in G10/1 isolate were highlighted (B) proteases; black arrow head: TfCP7 (C) surface antigens; TSP: tetraspanin proteins (D) MYB proteins; more expressed MYB transcripts (log₂(fc) = 4) on BP ~ PIG are also shown.

Figure 4

Clustering analysis reveals patterns of transcript abundance in G10/1 isolate. Transcript abundance data set was reduced to 458 variables or clusters of transcripts. Variables were represented in a heat-map of 22 × 21 cells. Abundance was represented by the average of each transcription in the cluster. Values were scaled between 0 and 1. Positions of the clusters in the heatmaps are the same for all isolates. WordClouds highlight the most represented annotated transcripts in that section of the heatmap. Distinctive regions for G10/1 isolate are highlighted as “(a)” and “(b)”.

Figure 5

Predicted promoters from Tritrichomonas foetus genes with MYB binding motif. (A) A schematic representation of predicted promoters and the predicted MYB binding motifs from genes related to Hsp70 family and malic enzymes; (B) Representation of predicted promoters and the predicted MYB binding motifs for genes related to virulence and adaptive factors studies in this work.