Mode and specificity of binding of the small molecule GANT61 to GLI determines inhibition of GLI-DNA binding

Abstract

The GLI genes, GLI1 and GLI2, are transcription factors that regulate target genes at the distal end of the canonical Hedgehog (HH) signaling pathway (SHH->PTCH->SMO->GLI), tightly regulated in embryonic development, tissue patterning and differentiation. Both GLI1 and GLI2 are oncogenes, constitutively activated in many types of human cancers. In colon cancer cells oncogenic KRAS-GLI signaling circumvents the HH-SMO-GLI axis to channel through and activate GLI in the transcriptional regulation of target genes. We have observed extensive cell death in a panel of 7 human colon carcinoma cell lines using the small molecule GLI inhibitor GANT61. Using computational docking and experimental confirmation by Surface Plasmon Resonance, GANT61 binds to the 5-zinc finger GLI1 protein between zinc fingers 2 and 3 at sites E119 and E167, independent of the GLI-DNA binding region, and conserved between GLI1 and GLI2. GANT61 does not bind to other zinc finger transcription factors (KLF4, TFIIβ). Mutating the predicted GANT61 binding sites in GLI1 significantly inhibits GANT61-GLI binding and GLI-luciferase activity. Data establish the specificity of GANT61 for targeting GLI, and substantiate the critical role of GLI in cancer cell survival. Thus, targeting GLI in cancer therapeutics may be of high impact.

Figure 1. Schema of pathways for the aberrant activation of GLI in colon cancer

Figure 2

Human colon carcinoma cell lines were treated for 72 hr, in duplicate, to equimolar concentrations (20 μM) of GDC-0449 (n=5), AZD6244 (n=4), GANT61 (n=7), or were untreated (n=7). Cells were harvested by trypsinization, cell death analyzed by Annexin V/PI staining and FACS analysis, and raw data quantitated using CellQuest software, as described in Materials and Methods. Data represent the Mean, Range, and SD.

Figure 3

Computational docking of GANT61 to GLI1 using the known crystal structure of the five zinc finger GLI1-DNA complex (PDB ID 2GLI) [55]. A: Two dimensional chemical structures of GANT61 and GANT61-diamine drawn in ACD-ChemSketch (Advanced Chemistry Inc.); B: Computational modeling of GANT61-diamine binding predicts binding to GLI1 within the groove between fingers 2 and 3; zinc fingers 1-5 (yellow); C: Predicted GANT61-diamine bound to GLI1 at amino acids E119 (1 H bond) and E167 (2 H bonds), amino acids residues within 3.5°A of any atom of GANT61 are shown as line. D: GLI1 and GLI2 alignment; E119, E167 (red); residues within 3.5 Å of docked GANT61-diamine (blue).

Figure 4

Surface Plasmon Resonance (SPR) of: A: Immobilized DNA with increasing concentrations of GLI1-WT as analyte: B: Immobilized GLI1-WT with determination of GANT61 binding in presence of increasing concentrations of GANT61 (1-50 μM); C: Dissociation Constants (K_D) for the interaction between GLI1-WT and DNA, GLI1-WT and GANT61, or inhibition of GLI1-WT-DNA binding by GANT61; D: Binding of GANT61 (40 μM) to immobilized ΔGLI-WT in contrast to no binding of GANT61 (40 μM) to immobilized KLF4 or TFIIβ. The analyte was exposed to sensor chips for 3 min followed by 5 min in the absence of analyte as described in Materials and Methods. Response Units are shown vs time of incubation (sec).

Figure 5

SPR of: A: Immobilized DNA +/− GLI1 (100 nM) +/− GANT61 at varied concentrations (1-50 μM); B: Maximum RU vs concentration of GANT61.

Figure 6

SPR conducted with immobilized ΔGLI-WT or ΔGLI-DM proteins following site directed mutagenesis of GLI1-WT, PCR amplification and protein purification of ΔGLI-WT without mutation, or ΔGLI-DM with both E119 and E167 sites mutated. Immobilized A: ΔGLI-WT or B: ΔGLI-DM with varied concentrations of GANT61 (10-40 μM) as analyte; C: Maximum Response for binding of GANT61 to ΔGLI-WT or ΔGLI-DM at varied concentrations of GANT61. D: Dissociation Constants (K_D) for binding of GANT61 to ΔGLI-WT or ΔGLI-DM proteins; E: Purification of ΔGLI-WT or ΔGLI-DM proteins on Ni-NTA columns as described in Materials and Methods.

Figure 7

GLI-luciferase reporter assays in HT29 cells under 2 different conditions: A: Co-transfection of GLI-luc, pRLTK, and either pBabe-Puro, GLI1-WT or GLI1-DM cDNA into HT29 cells using Lipofectamine 2000; 24 hr post-transfection, cells were treated with GANT61 (20 μM) for 24 hr, and luciferase activity determined as described in Materials and Methods. B: An HT29-derived stable cell line expressing GLI-luc was transfected with either pBabe-Puro, GLI1-WT or GLI1-DM cDNA, and exposed to GANT61 (20 μM) for 24 hr. Live cell imaging was performed using a Bruker optical and X-ray small animal imaging system, as described in Materials and Methods.