Identification of the Schistosoma mansoni TNF-alpha receptor gene and the effect of human TNF-alpha on the parasite gene expression profile

Abstract

Background: Schistosoma mansoni is the major causative agent of schistosomiasis. The parasite takes advantage of host signals to complete its development in the human body. Tumor necrosis factor-alpha (TNF-alpha) is a human cytokine involved in skin inflammatory responses, and although its effect on the adult parasite's metabolism and egg-laying process has been previously described, a comprehensive assessment of the TNF-alpha pathway and its downstream molecular effects is lacking.

Methodology/principal findings: In the present work we describe a possible TNF-alpha receptor (TNFR) homolog gene in S. mansoni (SmTNFR). SmTNFR encodes a complete receptor sequence composed of 599 amino acids, and contains four cysteine-rich domains as described for TNFR members. Real-time RT-PCR experiments revealed that SmTNFR highest expression level is in cercariae, 3.5 (+/-0.7) times higher than in adult worms. Downstream members of the known human TNF-alpha pathway were identified by an in silico analysis, revealing a possible TNF-alpha signaling pathway in the parasite. In order to simulate parasite's exposure to human cytokine during penetration of the skin, schistosomula were exposed to human TNF-alpha just 3 h after cercariae-to-schistosomula in vitro transformation, and large-scale gene expression measurements were performed with microarrays. A total of 548 genes with significantly altered expression were detected, when compared to control parasites. In addition, treatment of adult worms with TNF-alpha caused a significantly altered expression of 1857 genes. Interestingly, the set of genes altered in adults is different from that of schistosomula, with 58 genes in common, representing 3% of altered genes in adults and 11% in 3 h-old early schistosomula.

Conclusions/significance: We describe the possible molecular elements and targets involved in human TNF-alpha effect on S. mansoni, highlighting the mechanism by which recently transformed schistosomula may sense and respond to this host mediator at the site of cercarial penetration into the skin.

Figure 1. Schematic representation of SmTNFR gene and respective ORF.

Each gray bar represents an exon and its respective localization in genome scaffolds is shown. SmTNFR gene is encoded in the Plus strand. The hatched bar represents the original Smp_168070 in silico gene prediction. The colored bars represent protein conserved domains of the ORF encoded by SmTNFR gene: Signal peptide region is shown in purple; the white dotted vertical line marks the position of an internal Methionine residue near the translation start site; the yellow hatched lines mark the transmembrane helical domain predicted at the signal peptide region; Gly-44 indicates the predicted start of the mature protein, when cleaved at the predicted cleavage site, between Ala-43 and Gly-44; TNFR (green boxes) mark the Cystein Reach Domains, the hallmark of TNFR superfamily; TM, Transmembrane helical domain. Numbered arrows indicate primers used in RACE and PCR experiments. Other three in silico predicted genes in this region of the genome are all encoded in the Minus strand, and they are indicated by Smp_numbers.

Figure 2. Phylogeny tree of the tumor necrosis factor receptor superfamily.

Sequences corresponding to the TNFR conserved domain (cd00185) were (A) aligned using ClustalX, and (B) the evolutionary history of the protein family was inferred using the Neighbor-Joining method, as described under Methods. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (100 replicates) is shown for each tree branch. Accession numbers for the sequences used in the analyses are given under Materials and Methods.

Figure 3. Relative expression levels of SmTNFR among five S. mansoni life cycle stages.

The different bar colors indicate three independent biological replicas. The expression levels of egg stage were arbitrarily set as 1, and values for other stages are represented as fold changes relative to this value, within the corresponding biological replica. Bars represent standard deviation of the mean for three technical replicates within each biological replica. Tubulin was used as internal normalization gene. ANOVA was used for calculating the statistical significance of expression changes among all life cycle stages (p-value <0.0001). Tukey test was used to calculate the significance between two life cycle stages. * p-value <0.01 between cercaria and egg, cercaria and miracidium, cercaria and schistosomulum and cercaria and adult comparisons.

Figure 4. Schematic representation of putative S. mansoni TNF signal transduction pathway.

Orange elements were in silico identified in S. mansoni, blue elements were not found. Symbol (→) indicates covalent modification : protein phosphorylation; (}) indicates non-covalent interaction : association : protein/protein. Relations: (+), stimulatory; (−), inhibitory; (0), neutral; (?), undefined. Adapted from ref. . Inside the nucleus (rectangles) we have highlighted two examples of messages whose transcriptions are described in humans to be mediated by c-Jun.

Figure 5. Effect of human TNF-α on gene expression profile in 3-h old early schistosomula.

Schistosomula were treated for 1 h with 20 ng/ml human TNF-α just 3 h after in vitro mechanical transformation of cercariae. Microarrays were used for large-scale gene expression measurements, and the figure shows a hierarchical clustering of the 548 genes that exhibited significant (q-value ≤0.05) changes in their expression levels, in relation to control non-treated 3-h old early schistosomula. Each horizontal line represents a gene; each column represents one experimental replica. There are two technical replicates for each of three biological replicates. For each gene, color intensity is proportional to the number of standard derivations from the mean expression for that gene, as indicated in the scale bar, and shows genes that were induced (red) or repressed (green) by treatment.

Figure 6. Effect of human TNF-α on gene expression profile in adult worms.

Paired adult worms were treated for 1 h and 24 h with 20 ng/ml human TNF-α. Microarrays were used for large-scale gene expression measurements, and the figure shows the genes that exhibited significant (q-value ≤0.05) changes in their expression levels, in relation to control non-treated paired adults. (A) Hierarchical clustering of 1365 genes with transient changes in their expression levels. (B) Hierarchical clustering of 492 genes with sustained changes in their expression levels throughout the 24 h period of observation. Each horizontal line represents a gene, each column represents one experimental replica. There are two technical replicates for each of three biological replicates, at each treatment time. Expression levels are indicated as the log2 ratio of intensities (Treated/Control) for each treatment time (1 and 24 h). For each gene, color intensity is proportional to the log ratio as indicated in the scale bar, and shows genes that were induced (red) or repressed (green) by treatment.

Figure 7. Most significantly enriched (p = 10⁻²⁹) network of S. mansoni genes that were sustainably altered at 1h and 24h treatment with human TNF-alpha.

In red are the genes that were induced in 1 h and 24 h with respect to their control. In green are genes that were repressed in 1 h and 24 h with respect to their control. Non-significantly altered genes are in grey; in white are the human genes known to belong to the network for which no homolog was found in S. mansoni. MIR124 stands for human microRNA 124. Direct relations are marked by continuous lines, while indirect relations have dashed lines.

Figure 8. Schematic representation of modules for cysteine-rich domains (CRD) and for death domains of selected TNF receptors.

Colored ellipses represent modules or domains detected in the represented TNF receptors. This simplified diagram is intended only to reflect modules organization as proposed in and contains no information relative to size of modules or size of the whole receptor. Solid arrows above the scheme point to the receptors that have TNF-α ligand interaction, and the dotted line arrow indicates a putative interaction of SmTNFR with TNF-α ligand.