Whole Exome Sequencing of a Patient with Atypical Congenital Pure Red Cell Aplasia Has Enabled the Identification of a Novel Key Actor of Erythropoiesis That May be Involved in CDAI Physiopathology

Background: The development of next generation sequencing techniques has brought important insights into the molecular mechanisms of erythropoiesis and how these processes can be perturbed in human diseases. This strategy may be valuable in some hereditary erythroid disorders where a subset of patients does not carry any mutations in the supposed causal gene and for which transgenic mouse models do not recapitulate the phenotype, suggesting that additional genetic events may be involved in pathogenesis.

Here, we report the case of an adult patient presenting with atypical pure red cell aplasia associated with facial dysmorphy and chronic leg ulcers. Whole exome sequencing revealed a heterozygous missense mutation (R725W) in the CDAN1 gene, which has been previously reported in congenital dyserythropoietic anemia type I (CDAI). However, this mutation was also detected in her healthy brother, suggesting that this event alone was not sufficient to explain her phenotype. According to this hypothesis, we found an additional germline heterozygous nonsense mutation (Q732X) in the MMS22L gene, which was not shared by her unaffected relatives.

MMS22L is a protein involved in homologous recombination-dependent repair of stalled or collapsed replication forks. Additionally, MMS22L is able to bind newly synthesized soluble histones H3 and H4 and exhibits a histone chaperone activity. MMS22L loading onto ssDNA during homologous recombination is promoted by the histone chaperone ASF1. Interestingly, CDAN1 acts as a negative regulator of ASF1 by mediating its sequestration in the cytoplasm, which results in the blocking of histone delivery.

Aims: As MMS22L has never been reported in erythropoiesis before, we aimed to investigate the role of MMS22L in human erythropoiesis. Based on the data summarized above, the purpose of this study was also to determine the effect of combined inactivation of MMS22L and CDAN1 on in vivo erythropoiesis, while exploring the functional cooperation between both proteins.

Results: To decipher the role of MMS22L in human erythropoiesis, we assessed the consequences of complete MMS22L inactivation in human cord blood CD34+ progenitors as well as in CD36+ immature erythroblasts using shRNA lentiviruses. This resulted in a severe decrease of cell proliferation and differentiation due to G1 cell cycle arrest, with a slight increase of apoptosis. Interestingly, this phenotype was not observed when MMS22L was inactivated in the granulo-monocytic lineage, in which differentiation was maintained, suggesting that erythroid cells, that are highly proliferative, are more sensitive to MMS22L inactivation.

To better understand the effect of combined CDAN1 and MMS22L haploinsufficiency observed in the proband, we used zebrafish as an in vivo model. Mms22l and cdan1 expression were simultaneously or separately downregulated by about 50% using antisens morpholino oligomers. 48 hours later, zebrafish embryos were stained with o-dianisidine to detect hemoglobin-containing cells. We found that combined knock-down of mms22l and cdan1 resulted in severe anemia, while knock-down of mms22l or cdan1 alone did not lead to any erythroid disorder. This experiment provides a proof-of-concept, indicating that the phenotype of the proband is indeed caused by the combination of both MMS22L and CDAN1 mutations.

Finally, in order to decipher the cooperation between MMS22L and CDAN1 we used the human erythroid UT-7 cell line. We found that CDAN1 inactivation resulted in a severe decrease in MMS22L expression within the nucleus, suggesting that CDAN1 may regulate MMS22L expression or localization. We therefore wanted to confirm these results by assessing MMS22L expression in B-EBV cell lines established from two CDAI patients with CDAN1 compound heterozygous mutations. We found a great decrease in MMS22L expression within the nucleus of the CDAI patients' cells when compared to three control B-EBV cell lines. Based on these results, we suggest that impairment of MMS22L trafficking to the nucleus could be involved in CDA1 physiopathology.

Conclusion: Through comprehensive genetic analysis of a single case with atypical congenital anemia, we demonstrated for the first time that MMS22L, a cell cycle regulator, is essential for the process of erythropoiesis. The crosstalk between MMS22L and CDAN1 is currently under investigation and could bring important new insights into the physiopathology of CDAI.