Global, genomic responses of erythrocytes to infectious agents have been difficult to measure, because these cells are e-nucleated. We have previously demonstrated that in vitro matured, nucleated erythroblast cells at the orthochromatic stage can be efficiently infected by the human malaria parasite Plasmodium falciparum. We now show that infection of orthochromatic cells induces change in 609 host genes. 592 of these transcripts are up-regulated and associated with metabolic and chaperone pathways unique to P. falciparum infection, as well as a wide range of signaling pathways that are also induced in related apicomplexan infections of mouse hepatocytes or human fibroblast cells. Our data additionally show that polychromatophilic cells, which precede the orthochromatic stage and are not infected when co-cultured with P. falciparum, up-regulate a small set of 35 genes, 9 of which are associated with pathways of hematopoiesis and/or erythroid cell development. These data unexpectedly predict that blood stage P. falciparum may induce host responses common to infections of other pathogens. Further P. falciparum may modulate gene expression in bystander erythroblasts and thus influence pathways of erythrocyte development.
Overall design
Human primary erythroid cells were differentiated from CD34+ hematopoietic stem cells isolated from growth factor-mobilized peripheral blood (ALL Cells, Inc.). Cells from five donors were cultured until polychromatophilic and orthochromatophilic stages of differentiation and served as uninfected control samples. Of the five donors, three were used to initiate Plasmodium falciparum (3D7) infection at a multiplicity of infection = 5. Infected cells were harvested 24 hours post-infection, and RNA was isolated with Trizol (Invitrogen) and purified with RNeasy columns (QIAGEN) according to manufacturer recommendations. Microarray labeling and hybridizations were done according to Affymetrix protocols using HG U133 plus 2.0 chips. For GenePattern analysis, all samples (5 control and 3 infected samples) were analyzed; for Dchip analysis, only three samples were analyzed (the same 3 donors served as control and infected samples).