The TEL gene mapped on 12p13 is rearranged by chromosomal translocations in a variety of human leukemias and is fused to various partner genes mainly encoding receptor type and non-receptor type tyrosine kinases and transcription factors. Using 3′ RACE method, we have cloned novel chimeric cDNAs TEL/protein tyrosine phosphatase receptor type R (PTPRR) that is generated by inv(12)(p13q13) found in an acute myeloid leukemia case. TEL is a member of ETS family transcription factors and contains the homodimerizing or heterodimerizing helix-loop-helix domain at the N-terminus and the DNA-binding ETS domain at the C-terminus. TEL works as a tumor suppressor. On the other hand, PTPRR is a brain-specific receptor type protein tyrosine phosphatase with a single intracellular catalytic domain. PTPRR is the first tyrosine phosphatase that was identified as a fusion partner for TEL. The chromosomal rearrangement joined exon 4 in the TEL gene and exon 7 in the PTPRR gene and produced ten isoforms of TEL/PTPRR cDNAs through alternative splicing. Among them, only one expresses TEL/PTPRR chimeric protein that consists of the HLH domain from TEL and the catalytic domain from PTPRR and the others express truncated forms of TEL. Because reciprocal PTPRR-TEL mRNA as well as wild-type-PTPRR mRNA was not detected by RT-PCR analysis, TEL-PTPRR is likely to be involved in leukemogenesis. To clarify molecular mechanisms by inv(12), we first molecularly analyzed the TEL/PTPRR chimeric cDNA and the N-terminally truncated TEL cDNA containing exons 1–4 (tTEL). When transiently expressed in NIH3T3 cells, both TEL/PTPRR and tTEL were equally contributed to the cytoplasm and the nucleus. Moreover, both isoforms were found to associate with wild-type-TEL by immunoprecipitation assays. Therefore, we speculated that TEL/PTPRR and tTEL affect wild-type-TEL’s functions in the nucleus through heterodimerizing with it. As expected from their structures, both isoforms did not bind to the ETS binding consensus sequence (EBS). In reporter assays with (ETS)3tkLuc, they did not repress transcription through EBS, while wild-type-TEL did. Interestingly, both TEL/PTPRR and tTEL dominantly interfered with the wild-type-TEL-induced transcriptional repression in a dose-dependent manner. On in vitro phosphatase assays using p-NPP as a substrate, both isoforms were shown to completely lack phosphatase activities. To assess the biological properties of TEL/PTPRR, we employed human megakaryoblastic leukemia cell line UT7/GM that proliferate completely depending on granulocyte macrophage-colony stimulating factor (GM-CSF). Overexpression of TEL/PTPRR resulted in factor-independent proliferation in UT7/GM cells. Interestingly, phosphorylated STAT3 remained after GM-CSF withdrawal in the TEL/PTPRR-overexpressing cells, while it rapidly declined in the mock cells. All these data collectively suggest that appearance of inv(12) causes leukemic transformation in human myeloid cells, at least partly through inactivating tumor-suppressive functions of wild-type-TEL. Considering that STAT3-mediated signals are maintained even after the cytokine depletion when TEL/PTPRR is overexpressed, TEL/PTPRR may regulate JAK/STAT signals through unknown mechanisms.

Author notes

Corresponding author

Sign in via your Institution