Investigating a Plasmodium falciparum erythrocyte invasion phenotype switch at the whole transcriptome level

Abstract

The central role that erythrocyte invasion plays in Plasmodium falciparum survival and reproduction makes this process an attractive target for therapeutic or vaccine development. However, multiple invasion-related genes with complementary and overlapping functions afford the parasite the plasticity to vary ligands used for invasion, leading to phenotypic variation and immune evasion. Overcoming the challenge posed by redundant ligands requires a deeper understanding of conditions that select for variant phenotypes and the molecular mediators. While host factors including receptor heterogeneity and acquired immune responses may drive parasite phenotypic variation, we have previously shown that host-independent changes in invasion phenotype can be achieved by continuous culturing of the W2mef and Dd2 P. falciparum strains in moving suspension as opposed to static conditions. Here, we have used a highly biologically replicated whole transcriptome sequencing approach to identify the molecular signatures of variation associated with the phenotype switch. The data show increased expression of particular invasion-related genes in switched parasites, as well as a large number of genes encoding proteins that are either exported or form part of the export machinery. The genes with most markedly increased expression included members of the erythrocyte binding antigens (EBA), reticulocyte binding homologues (RH), surface associated interspersed proteins (SURFIN), exported protein family 1 (EPF1) and Plasmodium Helical Interspersed Sub-Telomeric (PHIST) gene families. The data indicate changes in expression of a repertoire of genes not previously associated with erythrocyte invasion phenotypes, suggesting the possibility that moving suspension culture may also select for other traits.

Figure 1

Erythrocyte invasion history of parasite cultures used for primary analyses. Efficiency of invasion into neuraminidase treated erythrocytes by all biological replicates of static cultured baseline parasites (week 0) and their respective suspended cultures were monitored at regular intervals by invasion assays. Baseline cultures were confirmed to be SA-dependent, while suspended parasites gradually acquired SA-independent invasion phenotype with increasing length of time in culture. Schizonts were harvested at baseline, and weeks 8 and 6, respectively for suspended W2mef (a) and Dd2 (b). Broken lines indicate suspended replicates which were not sequenced.

Figure 2

Genes differentially expressed between static and suspended W2mef. Volcano plot showing the extent and significance of differentially expressed genes between suspended and baseline cultures of W2mef. Insert is a zoom in of gene expression excluding the extremely significant genes Rh4 and eba165. Grey: Benjamini-Hochberg adjusted p value < 0.01, red: absolute log2 fold change >2. Genes with symbols have no known function. Black: Benjamini-Hochberg adjusted P value > 0.01 (considered not significant).

Figure 3

Baseline and static cultures have similar gene expression patterns in comparison to suspended W2mef. Differential expression analyses were conducted for baseline vs suspended (baseline; vertical axis) and then static vs suspended (static; horizontal axis). A spearman rank correlation plot comparing the differential gene expression between baseline and suspended culture (vertical axis) as against static and suspended cultures (horizontal axis) was then constructed using the log2 fold change values from each comparison. ρ is the coefficient of correlation. Red dotted lines indicate the significance threshold which was set at 4-fold increase in gene expression.

Figure 4

Multi-gene families are differentially expressed between static and suspended W2mef. Differential expression analyses conducted between baseline and suspended W2mef showed increased expressed of specific gene families across the genome, including (a) the eba and Rh gene families, (b) SURFIN genes, (c) exported protein family 1 (EPF1) genes and PHIST genes. Red dotted lines indicate the significance threshold which was set at 4-fold increase in gene expression.

Figure 5

Pattern of differential gene expression is similar between Dd2 and W2mef. (a) Differential gene expression analyses between baseline and suspended Dd2 shows increased expression of Rh4, eba165 in addition to other genes to be highly expressed in suspended Dd2. Grey: Benjamini-Hochberg adjusted P value < 0.01, red: absolute log2 fold change >1. Genes with symbols have no known function. Black: Benjamini-Hochberg adjusted P value > 0.01 (considered not significant). (b) A Spearman rank correlation plot of differential gene expression between suspended vs baseline W2mef (horizontal axis) and suspended vs baseline of Dd2 (vertical axis) shows similarities in the expression pattern between suspended cultures of the two strains. ρ is the coefficient of correlation.