Codanin-1 Binds to Key Erythroid Genes and Its Knockdown Coupled with Ectopic Mutant Expression Recapitulates the Congenital Dyserythropoietic Anemia Type I (CDA I) Phenotype

Congenital dyserythropoietic anemia type I (CDA I) is an autosomal recessive disorder marked by specific morphological abnormalities of erythroblasts in the bone marrow, resulting in ineffective erythropoiesis and anemia. CDA I principally affects the erythroid lineage, while other non-erythroid hematopoietic lineages are unaffected. Light microscopy of bone marrow shows erythroid hyperplasia and binucleated erythroblasts connected by internuclear bridges. Electron microscopy of bone marrow demonstrates erythroblasts with spongy heterochromatin and invagination of the nuclear membrane. The gene responsible for causing CDA I, CDAN1, encodes for a ubiquitous protein, codanin-1, with unknown function. The goal of our study is to understand how codanin-1 regulates erythropoiesis and how codanin-1 defects result in CDA I.

To determine codanin-1 expression pattern during erythroid differentiation, we analyzed codanin-1 RNA and protein levels in erythroid primary cells and cell lines. RNA-seq conducted in erythroid-induced human primary CD34+ cells revealed codanin-1 RNA expression is maintained in erythroid lineage. Similarly, immunoblotting shows codanin-1 protein is expressed at high levels in human erythroleukemia cell line stimulated down the erythroid lineage. Codanin-1 has been identified as a chromatin-binding protein. Thus, we conducted ChIP-seq analysis to determine the location of DNA binding sites for codanin-1 in K562 cells. The ChIP-seq results identified 2534 binding sites for codanin-1. Interestingly, many of these sites belong to key erythroid genes. Furthermore, 28% of these DNA binding sites overlap with GATA-1 bound sites.

Because putative loss of function mutations in CDAN1 result in CDA I, we studied the effects of codanin-1 gene silencing on erythroid differentiation. shRNA-mediated knockdown reduced codanin-1 mRNA by 50% and protein expression levels by 90%. To assess the effects of codanin-1 perturbation on erythroid differentiation, hemoglobin synthesis was quantified using Drabkin’s reagent in which the absorbance measured at 540 nm is proportional to hemoglobin content in the sample. Erythroid-induced knockdown cells displayed decreased hemoglobin content when compared to non-targeting control shRNA (0.20 versus 0.48, measured in 3 separate experiments; p<0.01). In order to assess binuclearity of the knockdown cells, we stained cytospin slides with Wright Giemsa to assess nuclear morphology. Codanin-1 knockdown cells showed increased binuclearity when compared to control shRNA cells (6% versus 1%, measured in 3 separate experiments; p<0.01). Furthermore, we quantified changes in gene expression induced by codanin-1 gene silencing. Quantitative PCR showed that knockdown of codanin-1 decreased expression of key erythroid genes.

To establish a robust CDA I cell system, the first well-characterized CDAN1 mutation (R1042W) was studied because codanin-1 deletion mutants are embryonic lethal in mice and do not occur in humans. Endogenous codanin-1 knockdown cells were transduced with mutant or wild type codanin-1 and induced down the erythroid lineage. Because CDA I is characterized by abnormal heterochromatin appearance, we conducted electron microscopy studies to examine the ultrastructural features. Studies show that knockdown of endogenous codanin-1 coupled with ectopic expression of mutant codanin-1 results in recapitulation of the CDA I phenotype in the K562 cell line model.

Because CDA I is almost exclusively restricted to the erythroid lineage, understanding its pathophysiology has the potential to identify critical pathways in erythropoiesis. Here we show that the full phenotype of CDA I is recapitulated upon not only the loss of function of the codanin-1 protein but also on the expression of the mutant form of codanin-1. Our analysis demonstrates that codanin-1 binds to important erythroid loci, suggesting that codanin-1 affects the expression of erythroid genes. Further work will focus on studying codanin-1 function in human primary erythroid cells and in animal models.