The Short Lef-1 Isoform, Lacking The β-Catenin Binding Domain, Is Not Acting As a Dominant Negative Lef-1 Variant, But Is a Hematopoietic Active Protein With Unique DNA Binding Properties

Studies in the hematopoietic system of human and mice have demonstrated that the Wnt signaling pathway is essential for maintenance, activation and proliferation of normal hematopoietic stem cells. Lef-1 is a member of the Lef-1/T-cell-factor (Tcf) family of transcription factors regulated by the canonical Wnt signaling pathway and generally directly interacts with β-catenin in transcriptional complexes to induce expression of target genes, including the cell cycle regulators cyclin D1 and c-myc. We have previously shown that aberrant expression of Lef-1 perturbs normal hematopoietic stem cell (HSC) and progenitor function and induces acute myeloid leukemia in the murine bone marrow transplantation model and that it is a novel prognostic independent factor in patients with normal karyotype AML (Petropoulos et al. JEM, 2008; Metzeler et al., Blood 2012). Furthermore, we could recently demonstrate that shRNA mediated depletion of Lef-1 drastically compromises the function of long-term repopulating stem cells (Edmaier et al., Leukemia, in press), indicating a vital function of Lef-1 at the level of the most primitive stem cell compartment. So far, all these Lef-1 functions are ascribed to the full-length transcript of Lef-1, which acts as transcriptional mediator of Wnt signaling via its β-catenin binding domain, whereas the shorter Lef-1 isoform, deriving from an intronic promoter is thought to function as a dominant negative variant. To dissect the role of the ‘dominant negative’ Lef-1 isoform we first retrovirally engineered primary murine bone marrow cells to express the N-terminal deleted Lef-1 (Lef-1d56), lacking the β-catenin binding domain in comparison to the full-length Lef-1 (Lef-1WT) and the empty retroviral EGFP control. Loss of binding to β-catenin was validated by Co-IP for Lef-1d56. Of note, Lef-1d56 induced a 2,05fold and significant increase at the level of clonogenic cells in vitro compared to the control (n=3; p<0,05) and did not differ substantially from the activity of Lef-1 WT. β-catenin binding did not impact Lef-1 activity at the level of the short-term repopulating stem cells as documented in the CFU-S assay with almost comparable CFU-S frequencies between the wild-type and the mutant Lef-1 (d56) (73 CFU-S/1x10⁵ and 64 CFU-S/1x10⁵ , respectively, compared to 38 CFU-S/1x10⁵ for the EGFP control; p<0.05). In clear contrast, loss of β-catenin interaction reduced the CRU frequency dramatically (1:597.197(Lef-1 WT) and 1:1.675.238 (Lef-1 d56), respectively and 1: 1.233.152 for the EGFP control; p=0,066), indicating the necessity of Lef-1 to collaborate with β-catenin at the level of the long-term repopulating stem cell. Retroviral expression of Lef-1 d56 in murine bone marrow stem and progenitor cells for 48h induced a distinct gene expression profile and deregulated more genes than Lef-1WT compared to the EGFP control. Differentially expressed genes between Lef-1d56 and Lef-1WT comprised known hematopoietic factors such as like Gata2, Ets1 and genes associated to the G-protein coupled receptor protein signaling. ChIP-Seq in hematopoietic murine cells revealed unique binding sites for Lef-1d56 compared to Lef-1WT with binding of the Lef-1d56 to promoters of Wnt5a and CD81. Taken together, our data clearly indicate that loss of the β-catenin binding site does not convert Lef-1 into an overall dominant negative variant, but creates a ‘neomorphic’ isoform with distinct biological and DNA binding properties. Analyses are ongoing which quantify the expression of LEF-1WT and the β-catenin domain lacking LEF-1 isoform in over 100 patients with normal karyotype AML in comparison to normal human stem and progenitor cells to test potential differences between normal and leukemic cells as well as associations of the two LEF-1 transcripts with the mutational status, gene expression and treatment outcome in this AML patient group.