Gli1/ DNA interaction is a druggable target for Hedgehog‐dependent tumors
Patched Receptors
0301 basic medicine
570
Kruppel-Like Transcription Factors
610
cancer: Gli inhibitors; Gli1–DNA interaction; Hedgehog
Receptors, Cell Surface
Mice, SCID
Molecular Dynamics Simulation
Gli1-DNA interaction
Mice
03 medical and health sciences
Mice, Inbred NOD
Cerebellum
cancer
Animals
Embryonic Stem Cells
Cancer
Cancer; Gli inhibitors; Gli1-DNA interaction; Hedgehog
Mice, Knockout
Mice, Inbred BALB C
Gli inhibitor
Articles
DNA
Fibroblasts
Embryo, Mammalian
Isoflavones
Mutation
Mutagenesis, Site-Directed
Cancer; Gli inhibitors; Gli1-DNA interaction; Hedgehog; Animals; Cerebellum; DNA; Embryo, Mammalian; Embryonic Stem Cells; Female; Fibroblasts; Glioblastoma; Isoflavones; Kruppel-Like Transcription Factors; Mice; Mice, Inbred BALB C; Mice, Inbred NOD; Mice, Knockout; Mice, SCID; Molecular Dynamics Simulation; Mutagenesis, Site-Directed; Mutation; Receptors, Cell Surface; Receptors, G-Protein-Coupled; Signal Transduction; Molecular Biology; Biochemistry, Genetics and Molecular Biology (all); Immunology and Microbiology (all); Neuroscience (all)
Gli inhibitors
Female
Glioblastoma
Hedgehog
DOI:
10.15252/embj.201489213
Publication Date:
2014-12-05T06:06:29Z
AUTHORS (21)
ABSTRACT
Article4 December 2014Open Access Source Data Gli1/DNA interaction is a druggable target for Hedgehog-dependent tumors Paola Infante Center Life [email protected], Istituto Italiano di Tecnologia, Rome, Italy Search more papers by this author Mattia Mori Romina Alfonsi Department of Molecular Medicine, University La Sapienza, Francesca Ghirga Dipartimento Chimica e Tecnologie del Farmaco, Federica Aiello Chemistry and Industrial Chemistry, Pisa, Sara Toscano Cinzia Ingallina Mariangela Siler Danilo Cucchi Agnese Po Evelina Miele Davide D'Amico Gianluca Canettieri Enrico De Smaele Experimental Elisabetta Ferretti Isabella Screpanti Gloria Uccello Barretta Maurizio Botta Biotechnology, Pharmacy, Siena, Sbarro Institute Cancer Research Temple University, Philadelphia, PA, USA Bruno Corresponding Author Alberto Gulino Pasteur, Fondazione Cenci-Bolognetti - IRCCS Neuromed, Pozzilli, Lucia Di Marcotullio Information Infante1,‡, Mori1,‡, Alfonsi2,‡, Ghirga3, Aiello4, Toscano1, Ingallina1, Siler2, Cucchi2, Po2, Miele1, D'Amico2, Canettieri2, Smaele5, Ferretti5, Screpanti2, Barretta4, Botta6,7, 3, 1,2,8,9 2 1Center 2Department 3Dipartimento 4Department 5Department 6Department 7Sbarro 8Istituto 9IRCCS ‡These authors contributed equally to work *Corresponding author. Tel: +39 649255657; Fax: 649255660; E-mail: protected] or 649255129; 649912781; The EMBO Journal (2015)34:200-217https://doi.org/10.15252/embj.201489213 PDFDownload PDF article text main figures. Peer ReviewDownload summary the editorial decision process including letters, reviewer comments responses feedback. ToolsAdd favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Hedgehog signaling essential tissue development stemness, its deregulation has been observed in many tumors. Aberrant activation result genetic mutations pathway components other Smo-dependent independent mechanisms, all triggering downstream effector Gli1. For reason, understanding poorly elucidated mechanism Gli1-mediated transcription allows identify novel molecules blocking at level, representing critical goal tumor biology. Here, we clarify structural requirements Gli1 binding DNA Glabrescione B as first small molecule zinc finger impairing activity interfering with DNA. Remarkably, consequence robust inhibitory effect on activity, inhibited growth cells vitro vivo well self-renewal ability clonogenicity tumor-derived stem cells. identification highlights their relevance pharmacologic interference Gli signaling. Synopsis Structural features are explored characterise effective Hh-inhibitor Gli1-dependent Binding consensus sequence requires specific previously uncharacterized residues. B, natural that binds these amino acids impairs DNA, inhibiting transcriptional activity. Functionally, shows antitumor efficacy both vivo. Introduction Inappropriate reactivation (Hh) developmental responsible formation progression several human cancers through aberrant regulation functional properties cancer (i.e. self-renewal, survival, metastatic spread, neoangiogenesis) (reviewed et al, 2012; Amakye 2013; Briscoe Thèrond, Aberger Ruiz I Altaba, 2014). Autocrine/paracrine Shh, Ihh Dhh ligands bind Patched (Ptch) receptor relieving repressive seven-transmembrane protein Smoothened (Smo), which turn activates factors belonging family. proteins harbor five (ZF) region ZF4 ZF5 domains sequence-specific way, whereas ZF1, ZF2 ZF3 phosphate backbone possibly control stability recruitment co-regulatory (Kinzler Vogelstein, 1990; Pavletich Pabo, 1993). A C-terminal provided transactivating function modulation chromatin remodeling induced TFIID TATA box-binding protein-associated factor TAFII31 (Yoon 1998; Bosco-Clément 2013) HAT HDAC (Canettieri 2010; Malatesta 2013), SWI-SNF5 (Jagani 2010) SWI/SNF-like Brg/Brm-associated (Zhan 2011). In behave final effectors oncogenic genes (Aberger Although Hh-driven involve upstream either loss-of-function Ptch1 gain-of-function Smo loss cAMP/PKA-mediated Gα-GNAS suppressor ligand overproduction) (Goodrich 1997; Yauch 2009; He 2014), Smo-independent increased effectors, due high levels activatory mechanisms gene amplification epigenetically driven overexpression, mutation heterozygosity SuFu number post-synthetic modifications such decreased ubiquitination-mediated degradation acetylation PI3K/mTOR/S6K1 kinase-dependent phosphorylation) 1987; Taylor 2002; Dahlén 2004; 2006a, 2011; Wang Mazzà Tang Notwithstanding, functions way it interacts controls still understood. Small have reported represent helpful tools understand Hh/Gli level transducer effector. This allowed inhibitors targeting order growth. However, antagonists currently investigated clinical trials (GDC-0449 recently approved FDA), few identified (Mas Coni 2013a). Therefore, drugs molecular steps underlying would be beneficial wide spectrum patients whose and/or Furthermore, frequently occurring appearance resistance during therapy (Galimberti 2013). lack information accounts low Gli. HPI-1 HPI-4 shown post-translational events processing/activation Smo, increase proteolytic cleavage Gli2-FL repressor form Gli2-R overall (Hyman 2009). Arsenic trioxide (ATO) prevent Gli2 localization primary cilium, thus leading degradation, while binding-dependent inactivation not yet characterized (Kim Beauchamp Similarly, GANT61 inhibits only living cells, suggesting indirectly promoters unelucidated (Lauth 2007). Based knowledge crystallographic structure domain (Gli1ZF) complex (Pavletich 1993), together NMR studies computational experimental mutagenesis, here (GlaB), an isoflavone naturally found seeds Derris glabrescens (Leguminosae), Gli1ZF interferes turned out efficient inhibitor Hh/Gli-dependent vivo, indicating Gli/DNA appealing therapeutic strategy heterogeneous changes cancer. Results To bases complex, established screening protocol, based available X-ray cobalt ion-coordinated 1993) that, although structurally relevant, does provide itself energy system. conformational dynamics using physiological within Zn-coordination system each performing four replicas (MD) simulations. representative was extracted from MD trajectories (Fig 1A, Supplementary Movie S1) further design silico. Figure 1. Structure-based analysis Representative Gli1ZF/DNA extrapolated trajectories. blue cartoon, residues involved (based single point study) magenta sticks, Zn ions gray spheres. Effect mutants affinity predicted silico alanine scanning. ∆∆G calculated along difference between ∆G mutant Gli1ZF-WT. values kcal/mol means MM-PBSA methods ± SEM. activation. Luciferase assay performed HEK293T transfected 12XGliBS-Luc (GliBS, site), pRL-TK Renilla (normalization control), Flag-Gli1 WT indicated mutants. show mean SD three experiments. *P < 0.01; **P 0.05 versus WT. Western blot expression (bottom panel). Specific H-bond interactions K350 site consensus, MD. binding. Double-stranded oligonucleotide containing canonical GliBS (5′–TTGCCTACCTGGGTGGTCTCTCCACTT–3′) mutated used (5′–TTGCCTACCTCCCACTTCTCTCCACTT–3′) probe (P) EMSA recombinant GST-Gli1ZF-WT (Gli1 fragment: aa 242–424), GST-Gli1ZF-K350A GST-Gli1ZF-K340A. graph right indicates ratio (mean arbitrary units experiments) GST-Gli1ZF bound labeled probe/GliBS-free normalized amount GST-Gli1ZF-WT/DNA (as described 4). data online figure. 1 [embj201489213-SourceData-Fig1.pdf] Download figure PowerPoint structure, impact serine basic thermodynamic adduct evaluated delta (ΔG) 1B, Table compared wild-type (Gli1ZF-WT) (ΔΔG). results strongest ΔΔG contribution given H-bonding electrostatic Indeed, K340, K350, R354 K360, K371, R380 K381 strongly impaired ΔG correlate observations Gli1, transiently expressing ectopic different Gli-dependent luciferase reporter Gli-responsive sequence. Mutations completely abrogated K371 K360 did lesser extent. Notably, significant linear correlation obtained (R2 = 0.6918) comparing theoretical value respective percentage cell assays. Mutagenesis suggested above may 1C, S1). We chose K350A K340A mutants, showing highest intermediate binding, respectively (Supplementary S1), test direct An electrophoretic mobility shift (EMSA) equal amounts GST-Gli1ZF-K340A Fig S2) version unable 1D E). Comparison strong confirms it, albeit significantly lower extent 1E, S2). Gli1ZF-K350A Gli1ZF-K340A cell-based assays, 1C), suggest K340 strength. Virtual library interacting whether could regulated molecules, house composed than 800 unique compounds docked toward GOLD program S3) (Verdonk 2003). Literature (Sheng 2006) mutagenesis study were set up docking analyze poses, respectively. able interact least one highlighted selected. then computed MM-GBSA method (Mori 2011) divided heavy atoms efficiency (LE) scoring parameter. six (three vismiones, GlaB, chalcone V94 opioid alkaloid narceine) putatively behaving potential S4A). investigate modulators assay. Whereas GlaB Vismione E similar (a antagonist, Lauth 2007), partially displayed active context 2A, S4B C). Since vismiones quite chemically unstable conditions (Delle Monache, 1985), limiting bioactive specie, focused 2B). 2. Hh Inhibition Gli1-induced control) plus (empty) vector treated increasing concentrations GANT61. Treatment time 24 h, DMSO only. chemical numbering scheme analysis. Smo−/− MEF h control. graphs Ptch1−/− MEFs 48 mRNA determined quantitative real-time PCR (qRT–PCR) endogenous (β2-microglobulin HPRT). Pfkfb3 negative model hyperactivation: MEFs, constitutive consequently Gli1; SuFu−/− release suppression. qRT–PCR β2-microglobulin HPRT expression. Promoter occupancy prevented treatment. Flag-tagged empty vectors, immunoprecipitation (ChIP) carried out. primers encompassing Gli-BS mouse promoter (right, schematic representation). fold difference, relative (pcDNA3) (left panel) Daoy siRNA (siGli1/2) non-specific (siCtr) (right siGli1/2 siCtr. expressed repression control, GAPDH information: All DMSO. Gli1/GlaB associated factor, monitored spectroscopy proton mono-selective relaxation rates (Rms), prove slowing down motion upon (Valensin 1986; Neuhaus Williamson, 1989). end, protons H-1 H-3 C-2 C-4 methoxyl groups chosen ring A, H-11, H-12 H-15 H-8 C (Table 1; Monoselective (Rms, s−1) selected (0.412 mM, 600 MHz, DMSO-d6, 25°C) corresponding (ΔR/Rf, where ΔR Rms-Rf) mixtures GlaB/GST-Gli1ZF K340A/K350A GlaB/GST Proton RING Rf(s−1) ΔR/Rf 0.41 0 0.03 0.02 2-OMe 1.20 0.06 0.63 0.10 4-OMe 1.39 n.d.a 0.36 0.13 0.07 H-11 0.15 0.38 0.22 0.89 0.95 0.23 n.d.b 0.76 0.14 Relaxation rate because signal superimposed water. b measured large linewidth. First, free (Rf) normalize Rms detected GlaB/protein 1). presence GST-Gli1ZF, vicinal (ring B), B) C) S5). Instead, perturbed very likely Gli1ZF. rule any possible sole GST, also mixture, weak unspecific involvement rings determinant GST-Gli1ZF. O-prenyl too informative, broad signals, they might crucial correct positioning B. showed preferably same interface ZF5, noticeable shape pharmacophoric overlapping S6). group C-13 clearly overimposed E, prenyl chain position relevant inhibition. derivatives without chains S7) tested assay, but none them active, reinforcing important biological role residues, 1C–E), GST-Gli1ZF-K340A/K350A double mutant. preserved Gli1ZF-K340A, weaker local suggests adopt conformation Gli1ZF-WT Gli1ZF-K340A. contrast, additional Gli1ZF-K340A/K350A affects capability Gli1ZF, providing pattern spanning non-specifically throughout C, GST 1, summary, directly emphasize K350. key determinants imparing Confirming compound significatively element 2C). Consistently, reduced endoge
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (66)
CITATIONS (151)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....