Research Article: Replication Protein A Presents Canonical Functions and Is Also Involved in the Differentiation Capacity of Trypanosoma cruzi

Date Published: December 16, 2016

Publisher: Public Library of Science

Author(s): Raphael Souza Pavani, Marcelo Santos da Silva, Carlos Alexandre Henrique Fernandes, Flavia Souza Morini, Christiane Bezerra Araujo, Marcos Roberto de Mattos Fontes, Osvaldo Augusto Sant’Anna, Carlos Renato Machado, Maria Isabel Cano, Stenio Perdigão Fragoso, Maria Carolina Elias, Carlos A. Buscaglia. http://doi.org/10.1371/journal.pntd.0005181

Abstract: Replication Protein A (RPA), the major single stranded DNA binding protein in eukaryotes, is composed of three subunits and is a fundamental player in DNA metabolism, participating in replication, transcription, repair, and the DNA damage response. In human pathogenic trypanosomatids, only limited studies have been performed on RPA-1 from Leishmania. Here, we performed in silico, in vitro and in vivo analysis of Trypanosoma cruzi RPA-1 and RPA-2 subunits. Although computational analysis suggests similarities in DNA binding and Ob-fold structures of RPA from T. cruzi compared with mammalian and fungi RPA, the predicted tridimensional structures of T. cruzi RPA-1 and RPA-2 indicated that these molecules present a more flexible tertiary structure, suggesting that T. cruzi RPA could be involved in additional responses. Here, we demonstrate experimentally that the T. cruzi RPA complex interacts with DNA via RPA-1 and is directly related to canonical functions, such as DNA replication and DNA damage response. Accordingly, a reduction of TcRPA-2 expression by generating heterozygous knockout cells impaired cell growth, slowing down S-phase progression. Moreover, heterozygous knockout cells presented a better efficiency in differentiation from epimastigote to metacyclic trypomastigote forms and metacyclic trypomastigote infection. Taken together, these findings indicate the involvement of TcRPA in the metacyclogenesis process and suggest that a delay in cell cycle progression could be linked with differentiation in T. cruzi.

Partial Text: Trypanosoma cruzi is the etiological agent of Chagas disease that infects 8 to 10 million people worldwide. Alternating between mammalian and insect hosts, the parasite faces changing environmental conditions, including thermal shifting, nutritional availability, and osmotic and oxidative stresses (for review [1]). Based on its success to establish chronic infections, one can infer that T. cruzi possesses adaptive mechanisms to respond to environmental changes. A complex life cycle most likely compensates for the variations in extracellular conditions. T. cruzi has four developmental stages, differing in shape, metabolism, replicative and infective capacity. T. cruzi epimastigotes are a non-infective life cycle stage of the parasite that proliferate by binary fission in the guts of Triatoma infestans insects. These epimastigotes then transform into the infective, non-proliferative metacyclic trypomastigotes forms in the insect hindgut. When the insect vector bites a mammalian host, they eliminate the infective forms in their feces. This allows the parasites to penetrate the wounded skin and enter into the mammalian host’s circulatory system. Within the bloodstream, the metacyclic trypomastigotes infect mammalian cells and transform into replicative, spherically shaped amastigotes. Amastigotes proliferate inside the infected cells until they transform into non-replicative trypomastigotes. The life cycle is completed when an insect vector bites an infected mammalian host and takes up trypomastigotes within the blood that then transform into epimastigotes inside the insect gut ([2]). Although it has been previously described that some stressors, such as acidic pH and starvation, trigger the transition from one form to another [3], the molecular bases involved in this response remain to be elucidated, such as which molecules are sensors or transducers of these differentiation pathways.

In this manuscript, we performed in silico, in vitro and in vivo analysis of T. cruzi RPA-1 and RPA-2 subunits. We could observe, by alignment analysis, that both TcRPA-1 and TcRPA-2 indeed present OB-fold domains, including residues that are important for RPA-DNA interaction in mammalian and fungi RPA [20][35]. However, in vitro analysis demonstrated that TcRPA-1 can interact with single stranded DNA while TcRPA-2 cannot, although we can not exclude the possibility that the RPA-2 tridimensional structure obtained after refolding of insoluble protein could be the cause of lacking of RPA-2-ssDNA interaction. According to our molecular model of the predicted tridimensional structure of TcRPA-1, OBF1 and OBF2 are the major OB-folds responsible for RPA-1-DNA interactions, forming an open binding channel between these two OB-fold domains. In vitro assay showed interaction only between ssDNA and OBF1. It is important to note, however, that OBF2 may need OBF1 to increase its affinity for ssDNA as occurs in mammalian cells ([8]). In contrast, OBF3 of TcRPA-1 seems to be unable to interact with DNA because the residues involved in DNA binding are buried within the protein. These results differ from the crystallographic structure of hRPA-1, where the residues of DBD-C are exposed to solvent and hRPA can interact with DNA via multiple binding modes, which can include DBD-C and also DBD-D (present in RPA-2) in some binding modes [8]. Our molecular model of TcRPA-2 suggests that this protein does not have the capacity to interact with DNA due to an exclusive insertion, present in trypanosomatids RPA-2 inside OBF/DBD-D, which is unstable and adopts multiple positions that very often occlude the DNA binding channel during MD simulations. In conclusion, we suggest that T. cruzi RPA interacts with DNA only through TcRPA-1. Regarding the third subunit of RPA, we performed a thorough and low stringent in silico search in the local files of annotated proteins and we have found a bona fide candidate for T. cruzi RPA-3 (TcCLB.507011.150). However, further investigation is necessary in order to demonstrate that it is indeed part of RPA complex.

Source:

http://doi.org/10.1371/journal.pntd.0005181