Knockdown of p53 Improves Hematopoietic and Erythroid Differentiation Defects in Mouse Embryonic Stem Cell Models of Diamond Blackfan Anemia

Diamond Blackfan anemia (DBA) is a rare inherited bone marrow failure syndrome characterized by red blood cell hypoplasia, congenital anomalies and cancer predisposition. The disease has been shown to result from haploinsufficiency of large or small ribosomal subunit proteins. The p53 pathway, known to be activated by abortive ribosome assembly, may play a role in the pathogenesis of DBA. Previously, we described murine embryonic stem (ES) cell models of DBA and reported hematopoietic and erythroid defects common to Rps19- and Rpl5-deficient cell lines, as well as a primitive erythropoiesis defect unique to an Rpl5-deficient cell line [Blood 116(21), 877, 2010].

We studied the effects of p53 knockdown on hematopoiesis in our Rps19- and Rpl5-mutant murine ES cell lines created by gene trap technology. Small interfering RNA (siRNA) targeting p53 was transfected into mutant cell lines at the ES cell stage. A non-targeting siRNA served as a negative control. After 24 hours, cells were plated into methylcellulose medium with fetal bovine serum and stem cell factor (SCF) to generate embryoid bodies (EBs). On day 7, EBs were fed with medium containing SCF, interleukin-3 (IL-3), IL-6 and erythropoietin (epo). EBs were scored on day 12 for total quantity and hematopoietic percentage. For secondary differentiation into primitive erythroid colonies, day 5 EBs were disrupted, and individual cells were suspended in a methylcellulose medium containing fetal bovine plasma-derived serum and epo. Primitive erythroid colonies were counted on day 7 of culture. Definitive hematopoiesis assays were performed by disruption of day 7 EBs, followed by suspension of cells in methylcellulose medium containing SCF, IL-3, IL-6 and epo. Definitive hematopoietic colonies were counted on day 10. In an independent set of experiments, we created an isogenic pair of wild-type and mutant DBA ES cells by electroporation of another Rps19- mutant line with a plasmid vector expressing wild-type Rps19 cDNA (wild-type) or an empty vector (mutant).

By immunoblot assays, we detected an increased amount of p53 protein in our Rps19-and Rpl5- mutant cell lines. Following p53 siRNA transfection, we confirmed 82–95% reduction in p53 expression by quantitative PCR, whereas ES cells transfected with non-targeting siRNA did not alter p53 expression. For both Rps19- and Rpl5- mutants, previously shown to have EB formation defects in comparison to parental controls, p53 knockdown significantly improved EB formation, especially hematopoietic-type EBs, compared to mutants treated with non-targeting siRNA. In addition, p53 knockdown in both mutants reversed the definitive hematopoiesis defect by increasing the ratio of erythroid colony to myeloid colony formation. Furthermore, p53 siRNA transfection of the Rpl5- mutant rescued the primitive erythropoiesis defect previously shown by us.

To further explore the mechanistic basis of our findings, we additionally tested the effects of Rpl11 knockdown in our DBA models. The presence of free RPL11 secondary to abortive ribosome assembly has been hypothesized to be responsible for increased p53 in DBA by binding to and inhibiting the p53 inhibitor HDM2 (Mdm2 in mice). Transfection of Rpl11 siRNA into both Rps19- and Rpl5-mutant cell lines at the ES cell stage led to a marked reduction in EB formation, compared to cells transfected with non-targeting siRNA.

Finally, we also extended our analysis to an isogenic pair of Rps19- wild-type and mutant cells. In the mutant line, we confirmed a 5–8 fold rescue of EB formation with siRNA targeting p53 when compared to the non-targeting siRNA. In order to clarify the role of two major downstream effectors of p53, siRNA targeting either Bax or p21 was transfected into the mutant cell line. Surprisingly, neither siRNA was able to rescue the EB formation defect of the mutant cells.

(1) Knockdown of p53 markedly improves erythroid defects of Rps19- and Rpl5-deficient murine ES cell models of DBA, while inhibition of the upstream target Rpl11 causes significant toxicity to cells already haploinsufficient for Rps19 or Rpl5. (2) Knockdown of either Bax or p21 does not recapitulate knockdown of p53, suggesting that neither plays a significant individual role in downstream signaling from p53 in this model. (3) Further exploration of the p53 pathway may provide insights into the pathogenesis of DBA and identify new targets for therapy.

No relevant conflicts of interest to declare.