Prenatal origin of GATA1 mutations may be an initiating step in the development of megakaryocytic leukemia in Down syndrome

Recently, somatic mutations in exon 2 of the transcription factor GATA1 gene have been detected in essentially all Down syndrome (DS) megakaryocytic leukemia (AMkL) and transient myeloproliferative disorder (TMD) cases.¹  This is the most specific genetic abnormality other than trisomy 21 in DS AMkL cases and is likely linked to the estimated 500-fold higher risk of DS children to develop AMkL compared with non-DS children.²  In this study, GATA1 mutations were screened in hematopoietic tissues from DS fetuses and infants that had no pathologic evidence of leukemia to establish the stage of development at which GATA1 mutations may arise.

DS liver and/or bone marrow samples were obtained from archival autopsy specimens (ages, 10 days to 10 months) from the Department of Pediatric Pathology, Children's Hospital of Michigan and fetal liver tissue blocks, from therapeutically terminated pregnancies (gestational ages, 18-23 weeks) from the Department of Pathology, Hutzel Women's Hospital. The research protocol was approved by the institutional review boards of Wayne State University and the University of Chicago. Following deparaffinization according to the manufacturer's instructions (Qiagen, Valencia, CA), genomic DNA was isolated by standard techniques. All fetuses and infants were confirmed to have trisomy 21 by standard karyotype analysis. Screening for GATA1 exon 2 mutations was performed by single-strand polymorphism assay (SSCP) as previously described.³  Altered migration products were excised, and DNA was eluted and amplified by PCR and then sequenced.

GATA1 mutations were detected in 2 of 9 liver samples from fetuses (gestational ages, 21 and 23 weeks) and in 2 of 5 DS infant autopsy bone marrow samples (ages, 4 and 6 months; Table 1). It is unknown whether the fetuses would have survived to term without the development of TMD and/or AMkL, though the detection of GATA1 mutations in hematopoietic tissues obtained postnatally suggests that mutations may exist in the absence of leukemia and are likely early leukemogenic events in DS. Although no mutations were detected in 60 peripheral blood samples of healthy DS children, this does not exclude the possibility that mutations were present though below the sensitivity of our assay or detectable only in bone marrow hematopoietic cells.

The uniform detection of GATA1 mutations in DS children with TMD and AMkL suggests that trisomy 21 may be associated with an increased mutation rate in DS as reported by Finette et al.⁴  Increased expression of the cystathionine-β-synthase gene (localized to 21q22.3) in the fetal liver⁵  and the known origin of TMD in the fetal liver⁶  may result in an increased mutation rate due to a “functional folate deficiency” state.⁷  Mutations may arise in other unidentified genes in DS tissues, though GATA1 mutations may confer a proliferative advantage allowing for the expansion and survival of GATA1 mutant-containing clones.¹  A case of identical twin DS infants with TMD with the same GATA1 mutations suggests that the GATA1 mutation arose in utero.^1,8  Furthermore, GATA1 mutations have been detected at birth in Guthrie newborn screening cards from DS infants who later developed AMkL.⁹  Since peripheral blood smears were not examined in this latter study, one cannot rule out the possibility that the infants had TMD.

Although the sample size of our study is small, the frequency of GATA1 mutations detected in the fetal and neonatal liver/bone marrow samples appears higher than the incidence of TMD or AMkL in DS children. These results demonstrate that the acquisition of GATA1 mutations can occur prenatally, mutations can exist in the absence of leukemia and are likely early steps in a multistep process of leukemogenesis, and additional genetic events and/or environmental exposures are likely necessary for the full development of leukemia in DS.