Modeling Fanconi Anemia Using Human Induced Pluripotent Stem Cells By Reversible Complementation

Modeling of Fanconi anemia (FA) using human induced pluripotent stem cells (iPSCs) has been hindered by the requirement for an intact FA DNA repair pathway for effective reprogramming and maintenance of pluripotency in patient-derived iPSCs. Temporary complementation of FA pathway defects could permit effective reprogramming, with removal of complementation upon directed differentiation to hematopoietic stem and progenitor cells (HSPCs) to model the hematopoietic phenotypes of FA. In this study, we used three FA patient-derived iPSC lines engineered with doxycycline inducible complementation of disease-causing FANCA mutations. We maintained complementation with doxycycline exposure in the pluripotent state and during morphogen directed differentiation of hemogenic endothelium (HE) within embryoid bodies, which possesses the capacity to differentiate to HSPCs. Upon initiation of the endothelial-to-hematopoietic transition (EHT) in purified iPSC-derived HE, doxycycline was either removed or maintained in culture to either inactivate or sustain FANCA expression. Morphologic EHT and emergence of CD34+CD45+ immunophenotypic human HSPCs were not affected by the status of FANCA expression, which allowed for the isolation of otherwise isogenic FANCA-expressing and FANCA-deficient HSPCs for disease modeling. FANCA-deficient HSPCs were unable to form FANCD2/γH2AX foci relative to FANCA-expressing HSPCs in response to genotoxic stress, confirming impairment of the function of the FA pathway. We found that uncomplimented, FANCA-deficient HSPCs showed impaired cell cycle progression, increased apoptosis, and markedly decreased colony forming activity in methylcellulose compared to complemented, FANCA-expressing HSPCs. Unexpectedly, we also found that FANCA-deficient HSPCs showed accelerated erythroid differentiation compared to cells with an intact FA pathway. RNA sequencing analysis showed enrichment of signatures of normal HSPCs in complemented HSPCs, while uncomplimented HSPCs showed signatures of more mature hematopoietic cells and cells undergoing erythroid differentiation. Together, these findings demonstrate the utility of reversible complementation in modeling FA with patient-derived iPSCs, and provide an inexhaustible source of otherwise isongenic complimented and uncomplimented human FA HSPCs for use in the investigation of new therapies.