Characterization of Trans-sialidases family proteins from Trypanosoma cruzi

LEIVA MJP; BOURDIN ED; BONTEMPI I; MARCIPAR I; AMADIO AF

Rosario

Congreso; 4to Congreso Argentino de Bioinformática; 2013

Asociación Argentina de Bioinformática

The hemoflagelate parasite Trypanosoma cruzi, which cause Chagas disease, is a protist of the Trypanosomatidae family. This microorganism displays a family of proteins called Trans-sialidase (TS) in its surface. Some fragments of these protein proteins gave the best inmunoprotection against the disease caused by the parasite [1]. TS is a super family of highly polymorphic proteins with over 1400 genes and pseudogenes [2] and the patterns of variation that they present among different strains of the parasite are not yet exactly known. The aims of this work were to collect all the annotated and unannotated sequences in data banks corresponding to this family of proteins and characterize the heterogeneity of the family within each strain and between strains. First, a database search was done using keywords ?Trans sialidase & Trypanosoma cruzi? for annotation fields. Later, the search was complemented with local alignments using BLAST against main databases to obtain all the probably sequences of TS present (both nucleotides and proteins). Finally, CD-Hit [3] was used to eliminate the redundancy in the total set of sequences. With the non-redundant set, multiple sequence alignment were done using MUSCLE and T-COFFEE and highly divergent sequences were discarded since they produce a decrement in the quality of preselected data. We carried out a division into six groups taking into account preserved blocks in the set of proteins considered. Phylogenetics trees were constructed using PHYML to add a intraspecific parameter of similarity for each sub-division. Generated groups were evaluated and classified according to the characteristics presented by an active trans-sialidase (specific motifs, GPI-anchor, repetitive regions and signal peptide). The initial set contained more than 8000 sequences with high redundancy and different sources. After the redundancy elimination strategies and characterization of the groups, we selected a complete dataset of 1597 highly characterized and reliable sequences. This generated dataset was divided into six groups based mostly on sequence length. The most significant groups were 1, 2, 3 and 5. These groups share the VTVxNVxLYNR motif, which defines the belonging to the TS family [2] with a total of 1004 sequences. The group number 4 lacks this important motif, suggesting that is composed by fragments of this protein. The group number 6 have this motif degenerated. The complete dataset, besides this phylogenetic characterization, was used for epitope prediction.