Differential expression of pathogenic genes of Entamoeba histolytica vs E. dispar in a model of infection using human liver tissue explants

Abstract

We sought to establish an ex vivo model for examining the interaction of E. histolytica with human tissue, using precision-cut liver slices (PCLS) from donated organs. E. histolytica- or E. dispar-infected PCLS were analyzed at different post-infection times (0, 1, 3, 24 and 48 h) to evaluate the relation between tissue damage and the expression of genes associated with three factors: a) parasite survival (peroxiredoxin, superoxide dismutase and 70 kDa heat shock protein), b) parasite virulence (EhGal/GalNAc lectin, amoebapore, cysteine proteases and calreticulin), and c) the host inflammatory response (various cytokines). Unlike E. dispar (non-pathogenic), E. histolytica produced some damage to the structure of hepatic parenchyma. Overall, greater expression of virulence genes existed in E. histolytica-infected versus E. dispar-infected tissue. Accordingly, there was an increased expression of EhGal/GalNAc lectin, Ehap-a and Ehcp-5, Ehcp-2, ehcp-1 genes with E. histolytica, and a decreased or lack of expression of Ehcp-2, and Ehap-a genes with E. dispar. E. histolytica-infected tissue also exhibited an elevated expression of genes linked to survival, principally peroxiredoxin, superoxide dismutase and Ehhsp-70. Moreover, E. histolytica-infected tissue showed an overexpression of some genes encoding for pro-inflammatory interleukins (ILs), such as il-8, ifn-γ and tnf-α. Contrarily, E. dispar-infected tissue displayed higher levels of il-10, the gene for the corresponding anti-inflammatory cytokine. Additionally, other genes were investigated that are important in the host-parasite relationship, including those encoding for the 20 kDa heat shock protein (HSP-20), the AIG-1 protein, and immune dominant variable surface antigen, as well as for proteins apparently involved in mechanisms for the protection of the trophozoites in different environments (e.g., thioredoxin-reductase, oxido-reductase, and 9 hypothetical proteins). Some of the hypothetical proteins evidenced interesting overexpression rates, however we should wait to their characterization. This finding suggest that the present model could be advantageous for exploring the complex interaction between trophozoites and hepatocytes during the development of ALA, particularly in the initial stages of infection.

Fig 1. Histological analysis of the presence of trophozoites.

PCLS from human liver tissue were infected with E. histolytica or E. dispar trophozoites at different times (0, 1, 3, 24 and 48 h). The column entitled “without trophozoites” corresponds to the control. Tissues were stained with PAS. The arrows point out some illustrative trophozoites in representative images. Scale bar = 20 μm.

Fig 2. Immuno-histochemical detection of EhCRT and Ehlect.

PCLS from human liver tissue were inoculated ex vivo with E. histolytica or E. dispar trophozoites at different times (1, 3, 24 and 48 h). Tissues were counterstained with eosin. The arrows point out some illustrative trophozoites in representative images. Scale bar = 20 μm.

Fig 3. Expression levels of genes linked to the immune response in human liver tissue.

After human PCLS were ex vivo infected with E. histolytica or E. dispar trophozoites at different times (1, 3, 24 and 48 h), the relative quantification (RQ) was determined (by qPCR) for the mRNA of some genes encoding for human cytokines. RQ represents a logarithmic scale, expressed as the mean of separate assays. *p<0.05.