Abstract
The E2A-HLF fusion gene, created by the t(17;19)(q22;p13) chromosomal translocation in pro-B lymphocytes, encodes an oncogenic protein in which the E2A trans-activation domain is linked to the DNA-binding and protein dimerization domain of hepatic leukemia factor (HLF), a member of the proline- and acidic amino acid-rich (PAR) subfamily of bZIP transcription factors. This fusion product binds to its DNA recognition site not only as a homodimer but also as a heterodimer with HLF and two other members of the PAR bZIP subfamily, thyrotroph embryonic factor (TEF) and albumin promoter D-box binding protein (DBP). Thus, E2A-HLF could transform cells by direct regulation of downstream target genes, acting through homodimeric or heterodimeric complexes, or by sequestering normal PAR proteins into nonfunctional heterocomplexes (dominant-negative interference). To distinguish among these models, we constructed mutant E2A-HLF proteins in which the leucine zipper domain of HLF was extended by one helical turn or altered in critical charged amino acids, enabling the chimera to bind to DNA as a homodimer but not as a heterodimer with HLF or other PAR proteins. When introduced into NIH 3T3 cells in a zinc-inducible vector, each of these mutants induced anchorage-independent growth as efficiently as unaltered E2A-HLF, indicating that the chimeric oncoprotein can transform cells in its homodimeric form. Transformation also depended on an intact E2A activator region, providing further support for a gain-of-function contribution to oncogenesis rather than one based on a dominant-interfering or dominant-negative mechanism. Thus, the tumorigenic effects of E2A-HLF and its mutant forms in NIH 3T3 cells favor a straightforward model in which E2A-HLF homodimers bind directly to promoter/enhancer elements of downstream target genes and alter their patterns of expression in early B-cell progenitors.