Trypanosomatid EST: a neglected information resource regarding flagellated protozoa?

Transcriptome-wide analysis of the Trypanosoma cruzi proliferative cycle identifies the periodically expressed mRNAs and their multiple levels of control

Trypanosoma cruzi is the protozoan parasite causing American trypanosomiasis or Chagas disease, a neglected parasitosis with important human health impact in Latin America. The efficacy of current therapy is limited, and its toxicity is high. Since parasite proliferation is a fundamental target for rational drug design, we sought to progress into its understanding by applying a genome-wide approach. Treating a TcI linage strain with hydroxyurea, we isolated epimastigotes in late G1, S and G2/M cell cycle stages at 70% purity. The sequencing of each phase identified 305 stage-specific transcripts (1.5-fold change, p≤0.01), coding for conserved cell cycle regulated proteins and numerous proteins whose cell cycle dependence has not been recognized before. Comparisons with the parasite T. brucei and the human host reveal important differences. The meta-analysis of T. cruzi transcriptomic and ribonomic data indicates that cell cycle regulated mRNAs are subject to sub-cellular compartmentalization. Compositional and structural biases of these genes- including CAI, GC content, UTR length, and polycistron position- may contribute to their regulation. To discover nucleotide motifs responsible for the co-regulation of cell cycle regulated genes, we looked for overrepresented motifs at their UTRs and found a variant of the cell cycle sequence motif at the 3' UTR of most of the S and G2 stage genes. We additionally identified hairpin structures at the 5' UTRs of a high proportion of the transcripts, suggesting that periodic gene expression might also rely on translation initiation in T. cruzi. In summary, we report a comprehensive list of T. cruzi cell cycle regulated genes, including many previously unstudied proteins, we show evidence favoring a multi-step control of their expression, and we identify mRNA motifs that may mediate their regulation. Our results provide novel information of the T. cruzi proliferative proteins and the integrated levels of their gene expression control.

Transcription of long hypothetical orfs in Trypanosoma cruzi: the epimastigote stage uses trans-splicing sites that generate short 5' UTRs

We mapped the 5' UTR for five long hypothetical orfs from Trypanosoma cruzi; each one having a length of more than 10,000 bp. Our aim was to verify the constraints to the length of the 5' UTR and to identify the sites of alternative trans-splicing in the epimastigote stage of three T. cruzi strains. We used reverse transcription PCR to amplify the 5' UTR and demonstrated the transcription of all selected genes as well as additional trans-splicing sites in two of these genes. We observed that the length of the 5' UTR in these genes has a limit, in contrast to previous reports that indicated a trend for longer genes to display a proportionally long 5' UTR. The maximum length of the 5' UTR for the long genes analyzed in the present work is approximately 3% of the orf and, on average, is 1% of the orf length. The poly-pyrimidine tracts used as trans-splicing signal are in the range of 17-53 bases within a distance of 6-59 nt to first spliced-leader acceptor site. T. cruzi populations may use both signals differentially. We conclude that the limit for the 5' UTR length in long genes is determined primarily by the distance to neighboring genes.

The composition of untranslated regions in Trypanosoma cruzi genes

We collected the UTRs from Trypanosomacruzi genes that have been experimentally mapped and are publicly available, and made a comprehensive analysis of their composition features including sequence length, G+C content and relationship to ORF, composition of the most frequent words, and distribution of Simple Sequence Repeats (SSR). T. cruzi UTRs exhibit range length of 10-400bp for 5' UTR and 17-2800 for 3' UTR. Both UTRs display mean G+C content of 40%. Ratios between the UTR and protein coding segments show that the 5' UTR is limited to a maximum of 20% of the total length in the final transcript. The 5' UTR most frequent words in the range 4-12 bases are almost exact complement to the 3' UTR respective words. SSR in 3' UTR are longer than in 5' UTR and are mostly derived from TA/AT, TG/GT, and TTA/ATT. SSR accounts up to 20% of the nucleotide composition in 5' UTR and up to 90% in the 3' UTR.