Synthetic Activators of Cell Migration Designed by Constructive Machine Learning

Abstract Constructive machine learning aims to create examples from its learned domain which are likely to exhibit similar properties. Here, a recurrent neural network was trained with the chemical structures of known cell‐migration modulators. This machine learning model was used to generate new molecules that mimic the training compounds. Two top‐scoring designs were synthesized, and tested for functional activity in a phenotypic spheroid cell migration assay. These computationally generated small molecules significantly increased the migration of medulloblastoma cells. The results further corroborate the applicability of constructive machine learning to the de novo design of druglike molecules with desired properties.

[1]  Dongwei Kang,et al.  Discovery of non-peptide small molecular CXCR4 antagonists as anti-HIV agents: Recent advances and future opportunities. , 2016, European journal of medicinal chemistry.

[2]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[3]  G. Bemis,et al.  The properties of known drugs. 1. Molecular frameworks. , 1996, Journal of medicinal chemistry.

[4]  R. Bronson,et al.  Impaired B-lymphopoiesis, myelopoiesis, and derailed cerebellar neuron migration in CXCR4- and SDF-1-deficient mice. , 1998, Proceedings of the National Academy of Sciences of the United States of America.

[5]  G. Schneider,et al.  Computer-assisted quantification of motile and invasive capabilities of cancer cells , 2015, Scientific Reports.

[6]  J. Rubin,et al.  Chemokine signaling in cancer: one hump or two? , 2009, Seminars in cancer biology.

[7]  P. Murphy.  The molecular biology of leukocyte chemoattractant receptors. , 1994, Annual review of immunology.

[8]  Michael M. Mysinger,et al.  Structure-based ligand discovery for the protein–protein interface of chemokine receptor CXCR4 , 2012, Proceedings of the National Academy of Sciences.

[9]  G. Hessler,et al.  Artificial Intelligence in Drug Design , 2018, Molecules.

[10]  M. Ashburner,et al.  Gene Ontology: tool for the unification of biology , 2000, Nature Genetics.

[11]  M. Rosenkilde,et al.  Probing the Molecular Interactions between CXC Chemokine Receptor 4 (CXCR4) and an Arginine-Based Tripeptidomimetic Antagonist (KRH-1636). , 2015, Journal of medicinal chemistry.

[12]  Gisbert Schneider,et al.  Tuning artificial intelligence on the de novo design of natural-product-inspired retinoid X receptor modulators , 2018, Communications Chemistry.

[13]  John P. Overington,et al.  ChEMBL: a large-scale bioactivity database for drug discovery , 2011, Nucleic Acids Res..

[14]  Gisbert Schneider,et al.  Automated De Novo Drug Design: Are We Nearly There Yet? , 2019, Angewandte Chemie.

[15]  The Gene Ontology Consortium,et al.  Expansion of the Gene Ontology knowledgebase and resources , 2016, Nucleic Acids Res..

[16]  G. Schneider,et al.  Review : Macromolecular target prediction by self-organizing feature maps , 2016 .

[17]  J. Snyder,et al.  Discovery of tetrahydroisoquinoline-based CXCR4 antagonists. , 2013, ACS medicinal chemistry letters.

[18]  S. Jenkinson,et al.  Synthesis and SAR of novel isoquinoline CXCR4 antagonists with potent anti-HIV activity. , 2010, Bioorganic & medicinal chemistry letters.

[19]  Dominique Schols,et al.  Discovery of novel small molecule orally bioavailable C-X-C chemokine receptor 4 antagonists that are potent inhibitors of T-tropic (X4) HIV-1 replication. , 2010, Journal of medicinal chemistry.

[20]  Gisbert Schneider,et al.  A Computational Method for Unveiling the Target Promiscuity of Pharmacologically Active Compounds. , 2017, Angewandte Chemie.

[21]  Thomas Blaschke,et al.  Molecular de-novo design through deep reinforcement learning , 2017, Journal of Cheminformatics.

[22]  Gisbert Schneider,et al.  Unveiling the target promiscuity of pharmacologically active compounds in silico , 2017 .

[23]  F. Baleux,et al.  CXCR7 heterodimerizes with CXCR4 and regulates CXCL12-mediated G protein signaling. , 2009, Blood.

[24]  N. Heveker,et al.  Development of novel CXC chemokine receptor 7 (CXCR7) ligands: selectivity switch from CXCR4 antagonists with a cyclic pentapeptide scaffold. , 2015, Journal of medicinal chemistry.

[25]  Arun K. Ghosh,et al.  Insights into the Mechanism of Inhibition of CXCR4: Identification of Piperidinylethanamine Analogs as Anti-HIV-1 Inhibitors , 2015, Antimicrobial Agents and Chemotherapy.

[26]  Gisbert Schneider,et al.  De Novo Design of Bioactive Small Molecules by Artificial Intelligence , 2018, Molecular informatics.

[27]  Erik De Clercq,et al.  The AMD3100 story: the path to the discovery of a stem cell mobilizer (Mozobil). , 2009, Biochemical pharmacology.

[28]  Petra Schneider,et al.  Generative Recurrent Networks for De Novo Drug Design , 2017, Molecular informatics.

[29]  George Papadatos,et al.  The ChEMBL bioactivity database: an update , 2013, Nucleic Acids Res..

[30]  E. De Clercq,et al.  Synthesis and SAR of novel CXCR4 antagonists that are potent inhibitors of T tropic (X4) HIV-1 replication. , 2011, Bioorganic & medicinal chemistry letters.

[31]  A. Mantovani,et al.  The chemokine system: redundancy for robust outputs. , 1999, Immunology today.

[32]  Petra Schneider,et al.  Chemically Advanced Template Search (CATS) for Scaffold-Hopping and Prospective Target Prediction for ‘Orphan’ Molecules , 2013, Molecular informatics.

[33]  Gisbert Schneider,et al.  Mind and machine in drug design , 2019, Nature Machine Intelligence.

[34]  E. De Clercq,et al.  Design of novel CXCR4 antagonists that are potent inhibitors of T-tropic (X4) HIV-1 replication. , 2011, Bioorganic & medicinal chemistry letters.

[35]  T. Kipps,et al.  CXCR4: a key receptor in the crosstalk between tumor cells and their microenvironment. , 2006, Blood.

[36]  M. Rosenkilde,et al.  Design, synthesis, and biological evaluation of scaffold-based tripeptidomimetic antagonists for CXC chemokine receptor 4 (CXCR4). , 2014, Bioorganic & medicinal chemistry.

[37]  D. Liotta,et al.  Discovery of novel N-aryl piperazine CXCR4 antagonists. , 2015, Bioorganic & medicinal chemistry letters.

[38]  Trixie Wagner,et al.  Orally bioavailable isothioureas block function of the chemokine receptor CXCR4 in vitro and in vivo. , 2008, Journal of medicinal chemistry.

[39]  Y. Nie,et al.  Dopamine D2 receptor suppresses gastric cancer cell invasion and migration via inhibition of EGFR/AKT/MMP-13 pathway. , 2016, International immunopharmacology.

[40]  Thierry Kogej,et al.  Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks , 2017, ACS central science.

[41]  Henry Ho,et al.  Discovery of Novel CXCR2 Inhibitors Using Ligand-Based Pharmacophore Models , 2015, J. Chem. Inf. Model..

[42]  Petra Schneider,et al.  Identifying the macromolecular targets of de novo-designed chemical entities through self-organizing map consensus , 2014, Proceedings of the National Academy of Sciences.