Role of GATA-1s in early hematopoiesis and differences between alternative splicing in human and murine GATA-1

To the editor:

We read with interest the description by Hoeller et al of an exon 2 GATA-1 mutation leading to severe transient myeloproliferative disease (TMD),¹  in which the authors speculated that the severe phenotype might reflect loss of both full-length (FL) GATA-1 and the shorter isoform, GATA-1s, due to the position of the mutation in codon 2. Although almost all GATA-1 mutations in children with Down syndrome are in exon 2^2-4  and lead to loss of GATA-1FL, exon 2 mutations are also predicted to leave GATA-1s protein production unaffected,⁴  by translation of an alternatively spliced mRNA comprising exons 1/3/4/5/6.²  In humans, alternative splicing producing an exon 1/2/3/4/5/6 mRNA for GATA-1FL and exon 1/3/4/5/6 splice variant for GATA-1s has been demonstrated in adult bone marrow CD34⁺ cells.²  Using exon 1 and 3 primers, we have also found that both variants are consistently expressed in all normal cord blood and second-trimester fetal liver and bone marrow cells we have tested (n = 12; Figure 1). It seems likely, therefore, that exon 2 mutations would still allow expression of GATA-1s mRNA, which may be difficult to detect at the protein level due to the relative insensitivity of commercial GATA-1 antibodies in immunohistochemical reactions. Interestingly, and by contrast, although mice produce the 2 Gata-1 isoforms by alternative translation of a single mRNA,⁵  Gata-1s transcripts have not been reported in murine tissues and, in our experience, exon 1 and 3 primers consistently amplify only the FL transcript in mice (Figure 1B) despite evidence of Gata-1s protein production (Figure 1C). Further investigation of the functional consequences of different GATA-1 mutations may shed further light on the enigmatic role of GATA-1s in early hematopoiesis and differences between alternative splicing in human and murine GATA-1.