Identification of KLF1-E325K as a loss-of-function mutation in iPSC-derived macrophages

Abstract

With over two million red blood cells produced every second, erythropoiesis is one of the bodies most demanding processes. It is therefore unsurprising that multiple hereditary and acquired red blood cell disorders arise from such a complex system. Existing treatments are effective in managing some of these conditions, but few offer long-term cures. Finding new treatments relies on the full understanding of the cellular and molecular interactions associated with the development of red blood cells, which predominantly occurs within erythroblastic islands. Patients with congenital dyserythropoietic anemia type IV (CDA), caused by mutations in the transcription factor KLF1, present with a large number of nucleated erythrocytes in their peripheral blood. Erythroblastic island macrophages play a vital role in promoting red blood cell enucleation, and therefore a contribution of erythroblastic island macrophages to the defective enucleation observed in these patients was investigated. The use of patient-derived induced pluripotent stem cells, often termed ‘Disease-in-a-Dish’ approaches, was an attractive option in this study due to the rarity of the disease and therefore limited availability of primary cells.

A CDA type IV patient-derived iPSC line that has a mutation in KLF1 (E325K) was obtained and used to generate erythroid cells to confirm that CDA erythroid pathology could be recapitulated in vitro. Key erythroid genes such as GYPA, TFRC, SLC4A1, HBA1 and ICAM4 were downregulated in cells generated from CDA patient iPSCs compared to control iPSCs, in agreement with previously a published study. The CDA patient iPSC line was then used to generate macrophages in vitro, however, I found that iPSC-derived macrophages generated from CDA patient and control iPSCs expressed very low levels of KLF1. Based on previous studies, I would expect KLF1 levels to be high in erythroblastic island macrophages, therefore I concluded that the CDA patient iPSC-derived macrophages were not a good model for CDA patient erythroblastic island macrophages.

Activation of wild-type KLF1 in iPSC-derived macrophages had been previously demonstrated to induce a more ‘erythroblastic island macrophage-like’ phenotype. Therefore, I generated a KLF1-E325K inducible activation system in iPSCs. As proof of principle, I generated cells under erythroid differentiation conditions from two inducible KLF1-E325K iPSC lines. These cells recapitulated CDA erythroid cell pathology, with a reduction in populations of CD235a+ cells and CD71+ cells upon KLF1-E325K activation.

I applied the inducible KLF1-E325K activation system to generate macrophages and investigate their phenotype and function. I found that the E325K mutation prevents KLF1-induced macrophage maturation. In an in vitro model of the erythroblastic island, I observed reduced maturation and enucleation of erythroid cells in co-culture with KLF1-E325K macrophages compared to KLF1-WT macrophages. I hypothesised that the reduced function of KLF1-E325K macrophages to promote erythroid cell maturation and enucleation was the result of dysregulation of key erythroblastic island macrophage genes. The KLF1-E325K mutation had been identified to dysregulate several genes in erythroid cells, but whether there was any transcriptional dysregulation in CDA patient macrophages had not yet been investigated. Therefore, to investigate this hypothesis, I generated a transcriptomic dataset.

RNA sequencing of KLF1-E325K macrophages revealed that KLF1-E325K does not induce expression of the same target genes as KLF1-WT, including genes encoding secreted factors previously identified to promote erythroid cell maturation and enucleation such as ANGPTL7. In fact, the transcriptome of KLF1-E325K macrophages was most comparable to macrophages generated from their parental line which express almost no KLF1. In combination with KLF1-E325K macrophages reduced function within the EBI, these data identified KLF1-E325K as a loss-of-function mutation in iPSC-derived macrophages.

Collectively, these findings demonstrate that genetically modified iPSCs can be a valuable tool to study RBC disorders. Additionally, the inducible-activation strategy utilised here could be applied to model and investigate a contribution of EBI macrophages to other RBC disorders, with the goal of finding new druggable targets.