The RNA Editing Enzyme ADAR1 Is Required for Survival of Differentiating Hematopoietic Progenitor Cells in Adult Mice.

Adenosine Deaminases Acting on RNA (ADAR) are RNA-editing enzymes converting adenosine residues into inosine (A-to-I) in many double-stranded RNA substrates including coding and non-coding sequences as well as microRNAs. Disruption of the ADAR1 gene in mice results in fetal liver, but not yolk sac, defective erythropoiesis and death at E11.5 (Wang Q et al, Science 2000). Subsequently, a conditional knockout mouse model confirmed these findings and showed massively increased cell death in the affected organs (Wang Q et al, JBC 2004). However, the actual impact of ADAR1 absence on definitive or adult hematopoiesis has not been examined. To define the role of ADAR1 in adult hematopoiesis, we first examined the expression of ADAR1 in different hematopoietic stem/progenitor cell subsets isolated from bone marrow by real-time RT-PCR. ARAR1 was present in hematopoietic stem cells (HSCs) at relatively low level and increased in hematopoietic progenitor cells (HPCs). A series of functional hematopoietic assays were then undertaken. A conditional deletion of ADAR1 was achieved by transducing Lin⁻ or Lin⁻cKit⁺ bone marrow cells from ADAR1-^lox/lox mice with a MSCV retroviral vector co-expressing Cre and GFP. PCR analysis confirmed the complete deletion of ADAR1 in the transduced cells within 72 hours after the transduction. This system allowed us to evaluate the acute effect of ADAR1 deletion in a specific hematopoietic cell population. Following 4 days of in vitro culture after transduction, the absolute number of Lin⁻ Sca1⁺ cells in the Cre transduced group was similar to the input number; however the differentiating Lin⁺ cells significantly decreased whereas both the Lin⁻Sca1⁺ and Lin⁺ cells in the vector (MSCV carrying GFP alone) transduced group increased during culture. Moreover, the colony forming cell (CFC) assay showed much fewer and smaller colonies that contained dead cells from the gene deleted group as compared to those from the control group (p<0.001). The TUNEL assay showed a dramatic increase of apoptosis in the Lin⁺ population but not in the Lin⁻ cells. Given the mixed genetic background of the ADAR1-^lox/lox mice, repopulation of the transduced hematopoietic cells in vivo was examined in immunodeficient mice. Sublethally irradiated (3.5 Gy) NOD/SCID-γc^null recipient were transplanted with either 1.5 × 10⁵ Cre or vector transduced Lin⁻ ADAR1-^lox/lox cells. Multi-lineage engraftment in peripheral blood was monitored monthly. While the vector transduced cells were able to constitute more than 90% in multiple lineages of the peripheral blood at 1 to 3 months, Cre-transduced cells were virtually undetectable at all the time points (n=9 to 13, p<0.001). A similar result was found in the hematopoietic organs, including the bone marrow, spleen and thymus. Interestingly, however, the Lin⁻Sca1⁺cKit⁺ cell population was preserved in the Cre transduced group despite the very low level of total donor-derived cells in the bone marrow (n=6 to 7, p<0.01). Consistently, the single cell culture experiment demonstrated that there was no significant difference between ADAR^−/− and wild-type HSCs in terms of survival and division during the first 3 days of culture. Taken together, our current study demonstrates nearly absolute requirement of ADAR1 for hematopoietic repopulation in adult mice and it is also suggested that ADAR1 has a preferential effect on the survival of differentiating progenitor cells over more primitive cells.