Independent clones at separable stages of differentiation, bearing different GATA1 mutations, in the same TMD patient with Down syndrome

Somatic mutations of the megakaryocytic and erythroid differentiation factor GATA1 in transient myeloproliferative disorder (TMD) of Down syndrome (DS) occur in utero,^1-5  and are detectable in fetal liver.¹  Multiple GATA1 mutations can occur in the same DS-TMD patient.⁵  The exact cell type and developmental time-frame in which these mutations occur have not yet been identified. If the acquisition of such mutations occurs within a very limited time frame at a specific stage of differentiation, one would expect individual clones from a case of DS-TMD with multiple independent clones to be arrested at a similar stage of differentiation. We present data which argue against this hypothesis, and demonstrate for the first time that multiple independent GATA1 mutations in the same individual can be found in clones physically separable from each other by surface maturation markers (Figure 1A).

A female DS neonate (patient TMD2 in a previous report² ) presented with typical symptoms of TMD; 2 days after birth, a peripheral-blood (PB) sample was taken from which an aliquot was sorted for CD34⁺ and CD34^- blast-cell populations. Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis (Figure 1B) detected the full-length GATA1 transcript as well as a shorter transcript lacking exon 2 (spliceΔ-exon 2) in the patient's presentation sample and the CD34^- cells, whereas her CD34⁺ cells lacked the full-length product. Sequencing of the RT-PCR product from the unsorted presentation sample²  revealed a 34-base pair (bp) duplication in GATA1 exon 2. However, cloning of the RT-PCR products revealed additional mutations, each causing translation of a typically truncated⁶  GATA1 protein. Unsorted and CD34^- cells were found to carry 3 different mutations (dup34, ins8, and dup4; Figure 1Ci-iii), which were present on different chromosomes and represent independent clonal proliferations (not shown). CD34⁺ cells carried spliceΔ-exon 2 as the sole form of GATA1 transcript (Figure 1Civ). Genomic DNA cloning and sequencing from these CD34⁺ cells revealed the ins8 and dup4 mutations, but not the dup34 mutation (not shown). This was confirmed by PCR amplification from genomic DNA using dup34-allele-specific primers (Figure 1D), which amplified the specific product in the presentation sample but not in CD34⁺ cells. Therefore, the dup34 of exon 2, a predominant change in the RNA of the presentation sample,²  could only have been contributed by the CD34^-, and not CD34⁺, GATA1-expressing cells.

In summary, we have demonstrated the presence of several independent clonal expansions at different stages of megakaryocytic differentiation in a single case of DS-TMD. As all of the observed GATA1 mutations result in the same truncated protein,^6,7  it is unlikely that individual mutations could determine the degree of differentiation arrest or the propensity for proliferation. Therefore, the observation that proliferating blasts bearing different GATA1 mutations are detected at different stages of megakaryocytic differentiation may possibly be explained by independent clones that vary in their stage of differentiation at the time of acquisition of their respective mutations, or by further differentiation of the proliferating clone occurring despite these mutations, which occurred at some distance apart during fetal development. Alternatively, the dup34 clone could have acquired an additional mutation, causing the loss of CD34 expression. Hypothetical additional TMD-triggering mutations might explain why trisomy-21 fetuses can show GATA1 mutations early in gestation without overt fetal leukemia.¹  The first explanation is compatible with the predictions reached through mouse models: that GATA1 mutations may occur at various time points in hematopoietic ontogeny, but confer a proliferative advantage only to early, time-limited fetal progenitors.⁸