A key T. cruzi
strategy to survive in a mammalian host is its ability to actively invade a wide variety of non-phagocytic host cells. The acute phase of Chagas disease is characterized by intense parasitaemia and tissue parasitism involving the infection of heart, skeletal and smooth muscle cells, as well as liver, fat, and brain cells 
. Furthermore, during the chronic phase, there are several reports on differences of T. cruzi
tropism to host tissue, which is associated with the pathogenesis of Chagas disease 
. Therefore, an important aspect to understanding T. cruzi
infection is the identification of molecular components of both parasite and host cells that play a role in the infection of a broad range of cell types. In this regard, several studies have investigated changes in gene expression during T. cruzi
infection. However, these studies have focused mainly on the modulation of gene expression of the host cells 
. Although several trypomastigote surface proteins have been implicated in host-cell recognition/invasion 
, so far there is no clear association between a T. cruzi
expression profile and its ability to invade/proliferate in a given host cell.
In our previous work on the molecular characterization of the MASP family 
, we speculated that these proteins may be involved in host–parasite interactions because of their surface localization on the infective circulating trypomastigote forms. Moreover, the MASP family is highly polymorphic and can be secreted by the parasite, thus contributing to a large T. cruzi
polypeptide repertoire that could be exposed to the host cells and the host's immune system 
. In fact, it has been shown recently that a MASP protein is able to induce endocytosis in Vero cells 
, a process whereby the trypomastigote forms of the parasite actively invade host cells 
. As an attempt to investigate whether MASP members could be implicated in interactions with specific cell types, we investigate the MASP expression profile in trypomastigotes derived from epithelial (LLC-MK2) and myoblast (L6) cell lines. We selected these cell types because LLC-MK2 is widely used for maintaining T. cruzi
in in vitro
culture, whereas myoblasts give raise to muscle cells, which are target by T. cruzi
during acute and chronic phase of Chagas disease 
. We did not detected significant changes in the MASP expression profile between these two host cells after 4 passages in tissue culture. However, differential expression of MASP genes were detected by sequencing and by qRT-PCR analyses between trypomastigote forms derived from these two host cells after a larger number of tissue culture passages: MASP2, MASP14 and MASP16 were significantly more expressed in myoblasts compared to epithelial cells after 14 passages (). This is an indirect evidence that different MASP genes may be implicated in the interaction with distinct cell types. As indicated before, it was recently demonstrated that one MASP member (named MASP52, Tc00.1047053504239.220) is secreted by trypomastigote forms upon contact with non-phagocytic Vero cells and is able to induce endocytosis 
. This MASP gene was not sampled in our sequenced clones. Nevertheless, an association between the selection of a MASP profile and the infectivity of L6-derived parasites is suggested by the invasion assay experiment (Figure S3
). Whether peptides derived from MASP2, MASP14 and MASP16 are involved in myoblast recognition/invasion and/or parasite proliferation/survival within myoblast cells remains to be investigated.
We have also investigated whether the transition from in vivo
to in vitro
infection would affect the MASP expression profile. To this end, the same trypomastigote population that was recovered from passage 2 in mice was used to infect myoblast and epithelial cells, and after four passages in each cell type, the MASP expression profile was analyzed. We found noticeable differences in the MASP family expression profile when tissue-culture and bloodstream forms were compared (, and ). Significant differential expression of the genes MASP2, MASP16 and MASP27 was observed by qRT-PCR in tissue-culture derived trypomastigotes after four passages (Tryp.4p.llcmk2 and Tryp.4p.l6) compared to bloodstream trypomastigotes after two passages (BloodTryp.2p) (). These observations suggest that selective pressure driven by tissue culture or in vivo
infection may induce distinct MASP expression profiles. Differences in infections using bloodstream- and tissue culture-derived trypomastigotes have already been reported. Specifically, it has been shown that the rate of in vitro
infection of distinct host cell types does not correlate with the level of parasitaemia of experimentally infected mice 
. We also found a heterogeneous MASP expression profile when bloodstream trypomastigotes recovered from mice after 2 and 10 passages were compared: MASP2 and MASP16 were significantly more expressed in bloodstream forms after 10 passages compared with bloodstream forms from passage 2, and the inverse was observed for MASP16 (). In addition, a large number of MASP pseudogenes were expressed in bloodstream forms after 10 passages, which was not observed for any other library (). Sequencing a larger number of clones of this library did not significantly change this profile. Pseudogenes may contribute to the diversity of the sequence repertoire through recombination. Indeed, the involvement of pseudogenes in the generation of variant surface glycoprotein (VSG) diversity has already been described in T. brucei
, and it was hypothesized that this mechanism may explain the large number of VSG pseudogenes in the T. brucei
. Whether there is a selective pressure imposed by the vertebrate host immune system to increase the sequence diversity of the MASP family in the circulating trypomastigote forms at the expenses of generating pseudogenes remains to be investigated.
It is well established that several T. cruzi
surface proteins are co-expressed at a given time in the parasite population, leading to the phenomenon of antigenic variability 
. Here, we found that, in fact, several MASP genes are co-expressed, although the level of expression of each transcript is very variable in a given library (), indicating a heterogeneous expression of the family. This expression profile was also observed in the proteomic study of the trypomastigote form of the Y strain 
. In this study, 37 unique MASP peptides found in 167 MASP proteins were identified at different expression levels in a parasite population derived from LLC-MK2 host cells. Although the conditions and strains used in both studies were different, it is interesting that two MASP members sampled in our expression libraries from trypomastigotes derived from LLC-MK2 host cell were also represented in this proteomic study (Tc00.1047053508253.10 and Tc00.1047053510163.30). We have previously analyzed the MASP expression in individual trypomastigotes by performing immunofluorescence using non-permeabilized cells and an affinity-purified antibody specific to a MASP subgroup. Only a few trypomastigotes were labeled with the antibody, suggesting that, at least for some MASP proteins, their expression on the surface of trypomastigotes is not uniform in the parasite population 
. The present study added another layer of complexity to the expression of the MASP family since we detected temporal changes in gene expression of the same gene after sequential passages in mice and also in trypomastigotes derived from epithelial and myoblast cells after a large number of in vitro
passages ( and ).
How the parasite modulates the expression of MASP genes during the infection is an open question. We did not detect a correlation between chromosomal location of the MASP genes sampled in our cDNA libraries and their expression levels (data not shown), suggesting that there is no apparent bias regarding the chromosomal location of expressed MASP genes. It is well established that the control of gene expression in Trypanosomatids operates almost exclusively at a post-transcription level, primarily mediated by regulatory elements with the 3′UTR of the transcripts that modulate the mRNA stability by means of interactions with regulatory proteins [reviewed in 54]
. We have previously shown that the 3′UTR of MASP transcripts is highly conserved among the family members 
and therefore may not be involved in the differential expression of the distinct MASP genes. Nevertheless, subtle nucleotide differences in these regions and/or alternative polyadenylation sites among the different transcripts may favor or abolish specific interactions with regulatory proteins. How the parasite changes the expression of the same MASP gene under different in vitro
and in vivo
conditions is also intriguing. In this case, it is possible that the host cell and/or the host immune system may configure a specific MASP expression profile.
Another possible MASP function that may explain the high level of polymorphism of the family would be its involvement in immune evasion mechanisms. In addition to its extreme polymorphism, localization at the trypomastigote surface, and shedding properties, another MASP feature that reinforces this hypothesis is the large repertoire of distinct repetitive motifs of the MASP proteins 
. It has been shown that several parasitic repetitive proteins are targets for strong B-cell responses 
. In fact, in silico
predictions performed by our group on the entire MASP proteome suggest the occurrence of a large repertoire of B-cell epitopes in the family (data not shown). In the present study, we validated these predictions by showing that several peptides derived from MASP-expressed members reacted with sera from acutely infected mice ( and ). The antibody recognition of several MASP peptides supports the interaction of the MASP family with the host immune system during acute T. cruzi
We have also investigated whether the MASP antigenic profile changes during acute infection. Indeed, variable antigenic profiles between the trypomastigotes isolated from sequential passages in experimentally infected mice were observed by immunoblotting and ELISA ( and ). The MASP family, along with other T. cruzi
surface proteins, may contribute to the polyclonal lymphocyte activation that leads to hypergammaglobulinemia and the delayed specific humoral immune response, that are characteristic of the acute phase of Chagas disease. These phenomena are suggested to be an immune evasion mechanism 
. Polyclonal lymphocyte B activation could scatter the immune response, preventing the development of a specific and neutralizing response against the parasite and its complete elimination. T-cell independent responses may contribute to hypergammaglobulinemia 
. Indeed, the SAPA repetitive C-terminal region of the trans-sialidase protein was reported to be a T-cell independent B mitogen and inducer of non-specific Ig secretion 
. It is possible that MASP peptides could mediate both specific T-dependent or unspecific T-independent immune responses, a hypothesis that is partially supported by the differential recognition of MASPs by the two immunoglobulin types (IgM and IgG) and the difference in the antibody affinity levels against each of the synthetic peptides. We speculate that variations in the large repertoire of antigenic peptides derived from the MASP family may contribute to the mechanism of immune evasion during the acute phase of the infection.
This is the first report on the antigenic properties of the MASP family, supported by the description of the antibody recognition of expressed MASP peptides in the acute phase of the experimental infection. MASP expression in bloodstream trypomastigotes is also first described in this study, as well as the differential expression of its members in trypomastigotes derived from distinct host cells and during acute experimental infection. The MASP expression profile is likely to be even more complex than reported here due to the limitations of our approach. The construction of the expression libraries in our investigation was limited by the similarity of the MASP 3′UTRs, and by the limited number of clones sequenced. Nevertheless, this study revealed a much more complex pattern of MASP expression than was previously described 
. The use of an RNA-seq
approach to study the transcription of bloodstream, tissue-culture derived parasites and infected host cells will reveal a comprehensive picture of the expression of genes involved in T. cruzi
–host cell interactions.