Human SEC23B-deficient hematopoietic stem cell model of CDAII recapitulates erythroid and cell-cycle defects

Technical Summary: Human SEC23B-Deficient Hematopoietic Stem Cell Model of CDAII

Problem Statement
Congenital dyserythropoietic anemia type II (CDAII) is an inherited disorder caused by mutations in the SEC23B gene, a core component of the COPII complex responsible for endoplasmic reticulum-to-Golgi protein trafficking. The disease is characterized by ineffective erythropoiesis, hypoglycosylation of erythroid membrane proteins, and the presence of bi- or multinucleated erythroblasts in the bone marrow. While murine models and immortalized cell lines (e.g., HUDEP-2) exist, they fail to fully recapitulate the human disease phenotype. Murine models often require the deletion of multiple Sec23 paralogs to manifest an erythroid phenotype due to species-specific compensation, and immortalized lines may not accurately reflect in vivo human erythropoiesis. Consequently, the limited availability of primary human disease models hinders the comprehensive understanding of CDAII pathophysiology and the development of novel therapeutic strategies.

Methodology
The authors developed a primary human cellular model of CDAII by disrupting the SEC23B gene in hematopoietic stem and progenitor cells (HSPCs) derived from cord blood using CRISPR/Cas9 gene editing.

• Gene Editing: CD34+ HSPCs were nucleofected with Cas9 ribonucleoproteins (RNPs) targeting SEC23B using single or pooled sgRNAs. Editing efficiency exceeded 60–75%, resulting in frameshift mutations and marked reduction of SEC23B protein.
• In Vitro Differentiation: Edited HSPCs were differentiated into erythroid cells over 14 days using a three-stage cytokine protocol. Phenotypic analysis included flow cytometry for erythroid maturation markers (CD71/CD235a), morphological assessment of binucleation, and apoptosis assays.
• In Vivo Validation: Edited HSPCs were transplanted into sublethally irradiated NBSGW mice to assess human hematopoiesis and engraftment in a physiological setting.
• Transcriptomics: Bulk RNA-Seq was performed on day 9 and day 14 erythroid cultures to identify differentially expressed genes and pathways.
• Functional Validation: To test the mechanistic link between glycosylation defects and erythroid failure, the authors treated wild-type HSPCs with pharmacological inhibitors of O-glycosylation (luteolin, benzyl) and N-glycosylation (swainsonine). Additionally, they knocked out specific cell-cycle regulators (PTTG1, CDKN3, KIF22, NUF2) and other COPII components (SEC13, SEC31A) to assess phenotypic overlap.

Key Results

1. Recapitulation of CDAII Phenotype: SEC23B-knockout (KO) HSPCs successfully mimicked the CDAII patient phenotype both in vitro and in vivo. Key features included:
   • Impaired terminal erythroid differentiation, evidenced by a reduced output of mature CD71-/CD235a+ cells.
   • A significant increase (nearly threefold) in bi- and multinucleated erythroblasts.
   • Defective protein glycosylation.
   • A selective disadvantage for edited cells during differentiation, as evidenced by a progressive decrease in frameshift alleles and enrichment of wild-type alleles, mirroring the ineffective erythropoiesis seen in patients.
2. Transcriptomic Alterations: RNA-Seq analysis revealed distinct transcriptional programs in SEC23B-KO cells:
   • Upregulation: Genes associated with cellular stress, apoptosis, and immature/non-erythroid populations (e.g., CXCR4, HLA-DPB1), suggesting a delay in maturation.
   • Downregulation: Genes critical for terminal erythroid function (gas transport, hypotonic response) and, notably, cell-cycle regulators and mitotic integrity (PTTG1, CDKN3, KIF22, NUF2).
3. Mechanistic Insights:
   • Disruption of cell-cycle regulators (PTTG1, CDKN3, KIF22, NUF2) and the COPII component SEC13 in wild-type HSPCs partially phenocopied the SEC23B defect, including reduced erythroid output, increased apoptosis, and cell-cycle progression defects (specifically G2-M transition perturbations).
   • Pharmacological inhibition of protein glycosylation in wild-type HSPCs reproduced the key features of SEC23B deficiency, including reduced erythroid output, increased apoptosis, multinucleation, and cell-cycle alterations. This suggests that defective glycosylation is a primary downstream mechanism linking COPII dysfunction to ineffective erythropoiesis.

Significance and Claims
The paper claims to have established a robust, physiologically relevant human model of CDAII using primary gene-edited HSPCs. This model faithfully recapitulates the disease's hallmark features, including the accumulation of multinucleated erythroblasts and defective terminal maturation, overcoming the limitations of previous murine and immortalized cell models.

The study provides mechanistic insight by linking SEC23B deficiency to a decoupling of cell-cycle progression and erythroid differentiation. The authors propose that impaired COPII-mediated trafficking leads to defective protein glycosylation, which in turn disrupts the regulation of mitotic integrity and cell-cycle checkpoints, ultimately causing the multinucleated phenotype characteristic of CDAII.

The authors conclude that this human cellular platform serves as a valuable tool for:

1. Deepening the understanding of CDAII pathophysiology, specifically the interplay between glycosylation, transcriptional regulation, and cell-cycle control.
2. Facilitating the identification and evaluation of novel therapeutic strategies, such as those targeting epigenetic regulators (e.g., LSD1) or glycosylation pathways, to restore effective erythropoiesis.

The paper maintains a modest tone regarding future implications, noting that while the model identifies key drivers, further studies (e.g., single-cell approaches and proteomics) are required to fully dissect the specific glycosylation targets and the direct versus secondary effects on mitotic regulators.