Schwarz K, Iolascon A, Verissimo F, et al. . Nat Genet. 2009;41:936-40.Congenital dyserythropoietic anemias (CDA) are characterized by multinuclear erythroid precursors, iron overload, and erythropoietic maturation abnormalities. The most common form, now known as type II, was first described in 1962 and is characterized by symmetrical bi-lobar “dumbbell” nuclei and multinuclearity (nuclear fragmentation). Type II CDA was studied extensively by the team of Jack and Mary Crookston in Toronto, who noted that, like PNH, red blood cells from these patients demonstrated in vitro hemolysis at low pH (Ham test). They coined the term HEMPAS, for Hereditary Erythroblastic Multinuclearity with a Positive Acidified Serum test.1 Later, patients with similar phenotypes of ineffective erythropoiesis were reported, and those without the full HEMPAS picture were classified as either type I CAD (autosomal recessive inheritance) or type III (autosomal dominant). Electron microscopy of CDAII erythrocytes shows structures that were misnamed as double membrane but are in fact cisternae of the endoplasmic reticulum (ER) that lie parallel to the inner surface of the plasma membrane.

Although the genetic locus for HEMPAS/CDAII was mapped to 20q several years ago, the nature of the mutated gene remained elusive. The groups of Iolascon and Heimpel, have now identified mutations in SEC23B, the gene encoding a protein component of the coat protein complex II (COPII) as causative. COPII is responsible for the biogenesis of ER-derived vesicles destined for the cis-Golgi compartment. These protein complexes are thought to be the primary determinants underlying deformation of membranes into vesicles. The authors studied 33 individuals from 28 unrelated families and showed that all had mutations in both SEC23B alleles. These included deletions, missense, nonsense, and splice-site mutations, and were present either as compound heterozygosity or in the five consanguineous families as homozygosity. Remarkably, all 33 patients had a mutation of the same gene, a testimony to careful phenotyping of this rare disease by the authors. Subsequent to publication of this work, another group published similar findings in 13 CDAII patients from 10 additional families,2 nicely confirming the molecular basis of CDAII.

The role of the SEC23B gene product in erythroid biology was explored by silencing the gene in the K562 red cell line and in vivo in zebrafish. While this resulted in typical binucleated erythroid cells, the characteristic duplication of rough ER was not seen in the zebrafish erythrocytes. As for why the phenotype of SEC23B mutants is erythroid-specific in spite of the ubiquitous expression of the gene, the authors showed that during erythroid differentiation SEC23B expression increased much more than that of the related gene SEC23A. Interestingly, a mouse mutation of another gene in the SEC gene family, Sec1511, that also has a wide tissue expression, produces a restricted erythroid phenotype of hypochromic, microcytic anemia due to a defect in transferrin recycling.3,4

Thus, 60 years since the phrase molecular disease was coined for sickle cell anemia, another important hematologic disorder has yielded its molecular secret to genetic analysis. Just like inherited disorders of hemoglobin, red cell membrane components, and red cell enzymes have helped us to understand the structure and function of the mature red cell, the CDAs will help us to unravel a more complex process: the cell biology of erythropoiesis. These newly discovered molecular defects begin to explain how these mutations can cause multinuclearity and membrane abnormalities.

1.
De Lozzio CB, Valencia JI, Acame E: Chromosomal study in erythroblastic endopolyploidy. Lancet. 1962;1:1004-5.
2.
Crookston JH, Crookston MC, Burnie KL, et al. Hereditary erythroblastic multinuclearity associated with a positive acidified-serum test: A test of congenital dyserythropoietic anaemia. Br J Haematol. 1969;17:11-
3.
Bianchi P, Fermo E, Vercellati C, et al. Congenital dyserythropoietic anemia type II (CDAII) is caused by mutations in the SEC23B gene. Hum Mutat. 2009;30:1292-98.
4.
Garrick MD, Garrick LM. Loss of rapid transferrin receptor recycling due to a mutation in Sec15l1 in hbd mice. Biochim Biophys Acta. 2007;1773:105-8.

Competing Interests

Drs. Luzzatto and Prchal indicated no relevant conflicts of interest.