Abstract
Introduction: Abnormal eosinophils or eosinophilia in the bone marrow (BM) or peripheral blood (PB) are often found in acute myeloblastic leukemia (AML), especially in core binding factor (CBF) leukemias, including AML with t(8;21)(q22;q22) and inv(16)(p13;q22)/t(16;16)(q13;q22). However, it is still unclear why eosinophilia is frequently found in CBF leukemia. Recently, a Fip1-like1 (FIP1L1)- Platelet derived growth factor receptor (PDGFR) α fusion gene was demonstrated in idiopathic hypereosinophilic syndrome (HES) and chronic eosinophilic leukemia. This observation suggests that the activation of PDGFR may be linked to eosinophil differentiation of malignant cell progenitors. On the other hand, the signaling of PDGFR also has a very important role in malignant cell growth, survival, and differentiation. A two-hit model for the pathogenesis of AML, which seems to be caused by inactivating mutations in transcription factors and genetic lesions in tyrosine kinase resulting in constitutive activation, has been proposed. Moreover, it was shown that additional constitutive activation of RTKs causes leukemia in AML1/ETO- or CBFβ/MYH11-transgenic mice. In view of these findings, we investigated whether genetic abnormality of PDGFRα/β genes was associated with leukemogenesis or eosinophilia of CBF leukemia.
Method: Twenty-two samples of BM or peripheral blood PB from cases of AML with karyotypic abnormality of t(8;21) or inv(16) were obtained from Japanese adult de novo patients (>16 years old). The expression of the FIP1L1-PDGFRα fusion gene and PDGFRα/β genes was studied by RT-PCR. The FIP1L1-PDGFRα fusion gene was detected by nested PCR. For analysis of the JM and TK2 domains of PDGFRα/β, exons 11–13 and 17–21 of PDGFRα and exons 11–14 and 16–20 of PDGFRβ were amplified, respectively. Confirmed cDNA products were subjected to direct sequencing.
Result: Among 22 cases of CBF leukemia with eosinophilia, 11 cases were inv(16) and 11 were t(8;21). Regarding the cases with inv(16), 9 were M4Eo and 2 were M2 subtype in FAB classification. The median percentage of eosinophils in their BM was 10.7% (range 5.6–31.8%). All cases with t(8;21) were diagnosed as M2 subtype and the median percentage of eosinophils in their BM was 1.8% (range 0.2–6.9). In this study, no FIP1L1-PDGFRα fusion gene was found. The PDGFRα gene was expressed in 19 of 22 (86.4%) cases and expression of the PDGFRβ gene was detected by RT-PCR in all tested cases. Association between the expression of the PDGFRα/β genes and the eosinophil blood count was not seen. The sequencing of JM and TK2 domains of cDNA detected three types of single nucleotide alterations in PDGFRα/β genes; one was in codon 603 in the JM domain of PDGFRα in 9 cases, changing GCG to GCA (G2203A), both of which encode the same amino acid, alanine (Ala). The second was in the TK2 domain of PDGFRα, in which an abnormality was detected in the third nucleotide of codon 824 in 7 cases, changing GTC to GTT (C2866T), both of which encode valine (Val). The third was in codon 867 in the TK2 domain of PDGFRβ in 1 case, changing TTA to TTG (A2957G), both of which encode the same amino acid, leucine (Leu). However, all of them were silent changes. Moreover, all three variants were present in cDNA extracted from BM samples of patients in remission and were previously described as single nucleotide polymorphisms (SNPs).
Conclusion: PDGFRα/β genes do not appear to play a significant pathogenetic role in eosinophilia or leukemogenesis of CBF leukemia.
Author notes
Corresponding author