Bioconjugation Protocols: Strategies and Methods, 2nd Edition (Methods in Molecular Biology Vol 751)

Methods

in

Molecular Biology™

Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK



For further volumes: http://www.springer.com/series/7651

Bioconjugation Protocols Strategies and Methods

Second Edition

Edited by

Sonny S. Mark Pacific Biosciences, Menlo Park, CA, USA

Editor Sonny S. Mark, Ph.D. Pacific Biosciences Menlo Park CA, USA [email protected]

ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-61779-150-5 e-ISBN 978-1-61779-151-2 DOI 10.1007/978-1-61779-151-2 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2011928240 © Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of nformation storage and retrieval, electronic adaptation, computer software, or by similar or ­dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, ­neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Humana Press is part of Springer Science+Business Media (www.springer.com)

Preface I am very pleased to present the second edition of Bioconjugation Protocols: Strategies and Methods, part of the excellent Methods in Molecular Biology™ book series. This current volume builds on the outstanding first edition originally conceived and developed by Prof. Christof M. Niemeyer at Technische Universität Dortmund (Germany). The first edition of this book aimed to address the deficiencies of many of the conventional approaches to the synthesis of chemically modified biomolecular conjugates that lack efficient means to control the stoichiometry of conjugation, as well as the specific site of attachment of the conjugated moiety. In keeping with that aim, this updated and expanded second edition of Bioconjugation Protocols further explores newer approaches that overcome the limitations of classical synthetic methods. In addition, a number of protocols collected in this new volume clearly reflect how insightful techniques and innovative approaches in bioconjugate chemistry can be derived from the seamless interplay between the fields of organic synthesis, surface biotechnology, nanobioscience, and materials science and engineering. It is thus my sincere hope that this revised edition of Bioconjugation Protocols continues to serve as a highly useful and practical reference for scientists of all disciplines confronting the challenges of semisynthesizing novel types of biomolecular reagents and/or biofunctionalizing surfaces and structures of unique interest for a variety of applications ranging from biomedical diagnostics to therapeutics and to biomaterials. The book is divided into five main parts, with Chaps. 1–24 in Parts I, II, and III describing the most recent, leading-edge approaches developed by researchers to prepare semisynthetic conjugates of native/modified biomacromolecules (proteins, nucleic acids, lipids, and carbohydrates). In Part IV, Chaps. 25–31 present methods for the preparation of biofunctionalized inorganic surfaces and polymer thin-film structures. And finally, Chaps. 32–36 in the last part of this book (Part V) specifically focus on procedures for the biofunctionalization of various types of metallic/semiconductor nanoparticles and other nanostructures (magnetic nanoparticles, quantum dots, carbon nanotubes, and silicon nanowire devices). I am most grateful to our distinguished group of international scholars who have generously and enthusiastically contributed their valuable time, tireless efforts, and expertise to make this reference volume truly unique and relevant. I would also like to express my thanks to Prof. John Walker, the Methods in Molecular Biology™ Series Editor at Springer Publishing, for his excellent editorial guidance during all stages of this book project. And finally, I sincerely thank my dear family and friends for their boundless patience and understanding and for kindly providing me with their support and encouragement during the production of this work. Menlo Park, CA

Sonny S. Mark

v

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v xi

Part I  Protein Conjugates   1  Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements Using Genetically Encoded Ketone Functionalities . . . . . . . . . . . . . . . . . . . . . . . . 3 Edward A. Lemke   2  Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins and Small Molecules Using Microbial Transglutaminase . . . . . . . . . . . 17 Ahmed Besheer, Thomas C. Hertel, Jörg Kressler, Karsten Mäder, and Markus Pietzsch   3  Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids . . . . . . . . . . . . . 29 Sukanta Dolai, Wei Shi, Bikash Mondal, and Krishnaswami Raja   4  Dye/DNA Conjugates as Multiple Labels for Antibodies in Sensitive Fluorescence Immunoassays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Qin Zhang, Shengchao Zhu, and Liang-Hong Guo   5  Chemoselective Modification of Viral Proteins Bearing Metabolically Introduced “Clickable” Amino Acids and Sugars . . . . . . . . . . . . . . . . . . . . . . . . . 55 Partha S. Banerjee and Isaac S. Carrico   6  Preparation of Peptide and Other Biomolecular Conjugates Through Chemoselective Ligations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Mathieu Galibert, Olivier Renaudet, Didier Boturyn, and Pascal Dumy   7  New Fluorescent Substrates of Microbial Transglutaminase and Its Application to Peptide Tag-Directed Covalent Protein Labeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Noriho Kamiya and Hiroki Abe   8  Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides: Strategies and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Anna Mero, Chiara Clementi, Francesco M. Veronese, and Gianfranco Pasut   9  Extending the Scope of Site-Specific Cysteine Bioconjugation by Appending a Prelabeled Cysteine Tag to Proteins Using Protein Trans-Splicing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Tulika Dhar, Thomas Kurpiers, and Henning D. Mootz

Part II Nucleic Acid Conjugates 10  Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo . . . . . . . 145 Andrea Masotti and Francesco Pampaloni

vii

viii

Contents

11  Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gwladys Pourceau, Albert Meyer, Jean-Jacques Vasseur, and François Morvan 12  Site-Specific DNA Labeling by Staudinger Ligation . . . . . . . . . . . . . . . . . . . . . . . Samuel H. Weisbrod, Anna Baccaro, and Andreas Marx 13  Improved Cellular Uptake of Antisense Peptide Nucleic Acids by Conjugation to a Cell-Penetrating Peptide and a Lipid Domain . . . . . . . . . . . . Takehiko Shiraishi and Peter E. Nielsen 14  Synthesis of Oligonucleotide–Peptide Conjugates for Biomedical and Technological Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anna Aviñó, Santiago Grijalvo, Sónia Pérez-Rentero, Alejandra Garibotti, Montserrat Terrazas, and Ramon Eritja 15  Amphiphilic DNA Block Copolymers: Nucleic Acid-Polymer Hybrid Materials for Diagnostics and Biomedicine . . . . . . . . . . . . . . . . . . . . . . . . Jan Zimmermann, Minseok Kwak, Andrew J. Musser, and Andreas Herrmann

167

195

209

223

239

Part III Glycosyl and Lipid Conjugates 16  Chemically Selective Liposome Surface Glyco-functionalization . . . . . . . . . . . . . . Hailong Zhang, Yong Ma, and Xue-Long Sun 17  Bioconjugation Using Mutant Glycosyltransferases for the Site-Specific Labeling of Biomolecules with Sugars Carrying Chemical Handles . . . . . . . . . . . . Boopathy Ramakrishnan, Elizabeth Boeggeman, Marta Pasek, and Pradman K. Qasba 18  Lipid-Core-Peptide System for Self-Adjuvanting Synthetic Vaccine Delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mariusz Skwarczynski and Istvan Toth 19  Coupling Carbohydrates to Proteins for Glycoconjugate Vaccine Development Using a Pentenoyl Group as a Convenient Linker . . . . . . . . . . . . . . Qianli Wang and Zhongwu Guo 20  Conjugation of LPS-Derived Oligosaccharides to Proteins Using Oxime Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Joanna Kubler-Kielb 21  Site-Specific Chemical Modification of a Glycoprotein Fragment Expressed in Yeast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Junpeng Xiao and Thomas J. Tolbert 22  On-Resin Convergent Synthesis of a Glycopeptide from HIV gp120 Containing a High Mannose Type N-Linked Oligosaccharide . . . . . . . . . . . . . . . Rui Chen and Thomas J. Tolbert 23  Design and Synthesis of Novel Functional Lipid-Based Bioconjugates for Drug Delivery and Other Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rupa R. Sawant and Vladimir P. Torchilin

269

281

297

309

317

329

343

357

Contents

ix

Part IV  Biofunctionalization of Surfaces and Thin Films 24  Chemical Functionalization and Bioconjugation Strategies for Atomic Force Microscope Cantilevers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Magnus Bergkvist and Nathaniel C. Cady 25  Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces . . . . . Edith H.M. Lempens, Brett A. Helms, and Maarten Merkx 26  Fabrication of Dynamic Self-Assembled Monolayers for Cell Migration and Adhesion Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nathan P. Westcott and Muhammad N. Yousaf 27  DNA Detection Using Functionalized Conducting Polymers . . . . . . . . . . . . . . . . Jadranka Travas-Sejdic, Hui Peng, Hsiao-hua Yu, and Shyh-Chyang Luo 28  Preparation and Dynamic Patterning of Supported Lipid Membranes Mimicking Cell Membranes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stefan Kaufmann, Karthik Kumar, and Erik Reimhult 29  Enzyme Immobilization on Reactive Polymer Films . . . . . . . . . . . . . . . . . . . . . . Ana L. Cordeiro, Tilo Pompe, Katrin Salchert, and Carsten Werner 30  Characterization of Protein–Membrane Binding Interactions via a Microplate Assay Employing Whole Liposome Immobilization . . . . . . . . . . . . . Matthew D. Smith and Michael D. Best 31  A Bioconjugated Phospholipid Polymer Biointerface with Nanometer-Scaled Structure for Highly Sensitive Immunoassays . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kazuki Nishizawa, Madoka Takai, and Kazuhiko Ishihara

381 401

421 437

453 465

477

491

Part V  Biofunctionalization of Nanostructures 32  Purification, Functionalization, and Bioconjugation of Carbon Nanotubes . . . . . John H.T. Luong, Keith B. Male, Khaled A. Mahmoud, and Fwu-Shan Sheu 33  Functional Integration of Membrane Proteins with Nanotube and Nanowire Transistor Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aleksandr Noy, Alexander B. Artyukhin, Shih-Chieh Huang, Julio A. Martinez, and Nipun Misra 34  Single-Step Conjugation of Antibodies to Quantum Dots for Labeling Cell Surface Receptors in Mammalian Cells . . . . . . . . . . . . . . . . . . . . . . Gopal Iyer, Jianmin Xu, and Shimon Weiss 35  A Practical Strategy for Constructing Nanodrugs Using Carbon Nanotubes as Carriers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wei Wu and Xiqun Jiang 36  Design and Synthesis of Biofunctionalized Metallic/Magnetic Nanomaterials . . . Eun-Kyung Lim, Seungjoo Haam, Kwangyeol Lee, and Yong-Min Huh

505

533

553

565 583

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 597

Contributors Hiroki Abe  •  Department of Applied Chemistry, Graduate School of Engineering, Kyushu University, Fukuoka, Japan Alexander B. Artyukhin  •  School of Natural Sciences, University of California, Merced, CA, USA Anna Aviñó  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Anna Baccaro  •  Department of Chemistry, Konstanz Research School Chemical Biology, University of Konstanz, Konstanz, Germany Partha S. Banerjee  •  Department of Chemistry, State University of New York, Stony Brook, NY, USA Magnus Bergkvist  •  College of Nanoscale Science and Engineering, University at Albany, Albany, NY, USA Ahmed Besheer  •  Department of Pharmacy, Ludwig-Maximilians-University Munich, Munich, Germany Michael D. Best  •  Department of Chemistry, The University of Tennessee, Knoxville, TN, USA Elizabeth Boeggeman  •  Structural Glycobiology Section, Center for Cancer Research Nanobiology Program, National Cancer Institute, Frederick, MD, USA; Basic Science Program, SAIC-Frederick, Inc., Frederick, MD, USA Didier Boturyn  •  Département de Chimie Moléculaire, Université de Grenoble, Grenoble, France Nathaniel C. Cady  •  College of Nanoscale Science and Engineering, University at Albany, Albany, NY, USA Isaac S. Carrico  •  Department of Chemistry, State University of New York, Stony Brook, NY, USA Rui Chen  •  Department of Chemistry, Indiana University, Bloomington, IN, USA Chiara Clementi  •  Department of Pharmaceutical Sciences, University of Padova, Padova, Italy Ana L. Cordeiro  •  Leibniz Institute of Polymer Research Dresden, Max Bergmann Center of Biomaterials Dresden, Dresden, Germany Tulika Dhar  •  Fakultät Chemie – Chemische Biologie, Technische Universität Dortmund, Dresden, Germany Sukanta Dolai  •  Department of Chemistry, The City University of New York at the College of Staten Island, Staten Island, NY, USA Pascal Dumy  •  Département de Chimie Moléculaire, Université de Grenoble, Grenoble, France

xi

xii

Contributors

Ramon Eritja  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Mathieu Galibert  •  Département de Chimie Moléculaire, Université de Grenoble, Grenoble, France Alejandra Garibotti  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Santiago Grijalvo  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Zhongwu Guo  •  Department of Chemistry, Wayne State University, Detroit, MI, USA Liang-Hong Guo  •  Research Center for Eco-environmental Sciences, Chinese Academy of Sciences, Beijing, P. R. China Seungjoo Haam  •  Department of Chemical and Biomolecular Engineering, Yonsei University, Seoul, Korea Brett A. Helms  •  Laboratory of Chemical Biology, Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands Andreas Herrmann  •  Department of Polymer Chemistry, The Zernike Institute for Advanced Materials, University of Groningen, Groningen, The Netherlands Thomas C. Hertel  •  Institute of Pharmacy, Martin Luther University, Halle Wittenberg, Germany Shih-Chieh Huang  •  School of Natural Sciences, University of California, Merced, CA, USA Yong-Min Huh  •  Department of Radiology, Yonsei University, Seoul, Korea Kazuhiko Ishihara  •  Department of Materials Engineering, Center for NanoBio Integration, The University of Tokyo, Tokyo, Japan Gopal Iyer  •  Department of Chemistry and Biochemistry, University of California at Los Angeles, Los Angeles, CA, USA Xiqun Jiang  •  Laboratory of Mesoscopic Chemistry, Department of Polymer Science and Engineering, College of Chemistry and Chemical Engineering, Nanjing University, Nanjing, People’s Republic of China Noriho Kamiya  •  Department of Applied Chemistry, Graduate School of Engineering, Kyushu University, Fukuoka, Japan Stefan Kaufmann  •  Laboratory for Surface Science and Technology, Swiss Federal Institute of Technology Zurich (ETH Zurich), Zurich, Switzerland Jörg Kressler  •  Institute of Chemistry, Martin Luther University, Halle Wittenberg, Germany Joanna Kubler-Kielb  •  Program on Developmental and Molecular Immunity, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA Karthik Kumar  •  Laboratory for Surface Science and Technology, Swiss Federal Institute of Technology Zurich (ETH Zurich), Zurich, Switzerland

Contributors

Thomas Kurpiers  •  Ascendis Pharma GmbH, Heidelberg, Germany Minseok Kwak  •  Department of Polymer Chemistry, The Zernike Institute for Advanced Materials, University of Groningen, Groningen, The Netherlands Kwangyeol Lee  •  Department of Chemistry, Korea University, Seoul, Korea Edward A. Lemke  •  Structural and Computational Biology Unit, European Molecular Biology Laboratory, Heidelberg, Germany Edith H.M. Lempens  •  Laboratory of Chemical Biology, Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands Eun-Kyung Lim  •  Department of Chemical and Biomolecular Engineering, Yonsei University, Seoul, Korea Shyh-Chyang Luo  •  Yu Initiative Research Unit, RIKEN Advanced Science Institute, Saitama, Japan John H.T. Luong  •  Biotechnology Research Institute, National Research Council Canada, Montreal, QC, Canada Yong Ma  •  Department of Chemistry, Cleveland State University, Cleveland, OH, USA Karsten Mäder  •  Institute of Pharmacy, Martin Luther University, Halle Wittenberg, Germany Khaled A. Mahmoud  •  Biotechnology Research Institute, National Research Council Canada, Montreal, QC, Canada Keith B. Male  •  Biotechnology Research Institute, National Research Council Canada, Montreal, QC, Canada Julio A. Martinez  •  School of Natural Sciences, University of California, Merced, CA, USA Andreas Marx  •  Department of Chemistry, Konstanz Research School Chemical Biology, University of Konstanz, Konstanz, Germany Andrea Masotti  •  Gene Expression – Microarrays Laboratory, IRCCS - Bambino Gesù Children’s Hospital, Rome, Italy Maarten Merkx  •  Laboratory of Chemical Biology, Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands Anna Mero  •  Department of Pharmaceutical Sciences, University of Padova, Padova, Italy Albert Meyer  •  Institut des Biomolécules Max Mousseron, Université Montpellier, Montpellier, France Nipun Misra  •  School of Natural Sciences, University of California, Merced, CA, USA Bikash Mondal  •  Department of Chemistry, The City University of New York at the College of Staten Island, Staten Island, NY, USA Henning D. Mootz  •  Institut für Biochemic, University of Muenster, Münster, Germany François Morvan  •  Institut des Biomolécules Max Mousseron, Université Montpellier, Montpellier, France Andrew J. Musser  •  Department of Polymer Chemistry, The Zernike Institute for Advanced Materials, University of Groningen, Groningen, The Netherlands

xiii

xiv

Contributors

Peter E. Nielsen  •  Department of Cellular and Molecular Medicine, The Panum Institute, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark Kazuki Nishizawa  •  Department of Materials Engineering, Center for NanoBio Integration, The University of Tokyo, Tokyo, Japan Aleksandr Noy  •  School of Natural Sciences, University of California, Merced, CA, USA Francesco Pampaloni  •  Physical Biology Group Institute for Cell Biology and Neurosciences Frankfurt Institute for Molecular Life Sciences (FMLS) , Goethe University Frankfurt, Frankfurt am Main, Germany Marta Pasek  •  Structural Glycobiology Section, Center for Cancer Research Nanobiology Program, National Cancer Institute, Frederick, MD, USA Gianfranco Pasut  •  Department of Pharmaceutical Sciences, University of Padova, Padova, Italy Hui Peng  •  Key Laboratory of Polar Materials and Devices, College of Information Science and Technology, East China Normal University, Shanghai, China Sónia Pérez-Rentero  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Markus Pietzsch  •  Institute of Pharmacy, Martin Luther University, Halle Wittenberg, Germany Tilo Pompe  •  Leibniz Institute of Polymer Research Dresden, Max Bergmann Center of Biomaterials Dresden, Dresden, Germany Gwladys Pourceau  •  Institut des Biomolécules Max Mousseron, Université Montpellier, Montpellier, France Pradman K. Qasba  •  Structural Glycobiology Section, Center for Cancer Research Nanobiology Program, National Cancer Institute, Frederick, MD, USA Krishnaswami Raja  •  Department of Chemistry, The City University of New York at the College of Staten Island, Staten Island, NY, USA Boopathy Ramakrishnan  •  Structural Glycobiology Section, Center for Cancer Research Nanobiology Program, National Cancer Institute, Frederick, MD, USA; Basic Science Program, SAIC-Frederick, Inc., Frederick, MD, USA Erik Reimhult  •  Laboratory for Surface Science and Technology, Swiss Federal Institute of Technology Zurich (ETH Zurich), Zurich, Switzerland Olivier Renaudet  •  Département de Chimie Moléculaire, Université de Grenoble, Grenoble, France Katrin Salchert  •  Lausitz University of Applied Sciences, Senftenberg, Germany Rupa R. Sawant  •  Department of Pharmaceutical Sciences, Center for Pharmaceutical Biotechnology and Nanomedicine, Northeastern University, Boston, MA, USA Fwu-Shan Sheu  •  NUSNNI-Nanocore Institute, National University of Singapore, Singapore, Singapore; Department of Electrical and Computer Engineering, National University of Singapore, Singapore, Singapore Wei Shi  •  Department of Chemistry, The City University of New York at the College of Staten Island, Staten Island, NY, USA

Contributors

xv

Takehiko Shiraishi  •  Department of Cellular and Molecular Medicine, The Panum Institute, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark Mariusz Skwarczynski  •  School of Chemistry and Molecular Biosciences, The University of Queensland, St. Lucia, QLD, Australia Matthew D. Smith  •  Department of Chemistry, The University of Tennessee, Knoxville, TN, USA Xue-Long Sun  •  Department of Chemistry, Cleveland State University, Cleveland, OH, USA Madoka Takai  •  Department of Materials Engineering, Center for NanoBio Integration, The University of Tokyo, Tokyo, Japan Montserrat Terrazas  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Thomas J. Tolbert  •  Department of Chemistry, Indiana University, Bloomington, IN, USA Vladimir P. Torchilin  •  Department of Pharmaceutical Sciences, Center for Pharmaceutical Biotechnology and Nanomedicine, Northeastern University, Boston, MA, USA Istvan Toth  •  School of Chemistry and Molecular Biosciences, The University of Queensland, St. Lucia, QLD, Australia Jadranka Travas-Sejdic  •  Polymer Electronics Research Centre, Department of Chemistry, The University of Auckland, Auckland, New Zealand Jean-Jacques Vasseur  •  Institut des Biomolécules Max Mousseron, Université Montpellier, Montpellier, France Francesco M. Veronese  •  Department of Pharmaceutical Sciences, University of Padova, Padova, Italy Qianli Wang  •  Department of Chemistry, Wayne State University, Detroit, MI, USA Samuel H. Weisbrod  •  Department of Chemistry, Konstanz Research School Chemical Biology, University of Konstanz, Konstanz, Germany Shimon Weiss  •  Departments of Chemistry & Biochemistry and Physiology, University of California at Los Angeles, Los Angeles, CA, USA; California NanoSystems Institute, Los Angeles, CA, USA Carsten Werner  •  Leibniz Institute of Polymer Research Dresden, Max Bergmann Center of Biomaterials Dresden, Dresden, Germany; Center of Regenerative Therapies Dresden, Dresden, Germany Nathan P. Westcott  •  Department of Chemistry, The Carolina Center for Genome Sciences, The University of North Carolina, Chapel Hill, NC, USA Wei Wu  •  Laboratory of Mesoscopic Chemistry, Department of Polymer Science and Engineering, College of Chemistry and Chemical Engineering, Nanjing University, Nanjing, People’s Republic of China Junpeng Xiao  •  Department of Chemistry, Indiana University, Bloomington, IN, USA Jianmin Xu  •  Department of Chemistry and Biochemistry, University of California at Los Angeles, Los Angeles, CA, USA

xvi

Contributors

Muhammad N. Yousaf  •  Department of Chemistry, The Carolina Center for Genome Sciences, The University of North Carolina, Chapel Hill, NC, USA Hsiao-hua Yu  •  Yu Initiative Research Unit, RIKEN Advanced Science Institute, Saitama, Japan Qin Zhang  •  Department of Occupational Health, West China School of Public Health, Sichuan University, Chengdu, People’s Republic of China Hailong Zhang  •  Department of Chemistry, Cleveland State University, Cleveland, OH, USA Shengchao Zhu  •  Research Center for Eco-environmental Sciences, Chinese Academy of Sciences, Beijing, People’s Republic of China Jan Zimmermann  •  Department of Polymer Chemistry, The Zernike Institute for Advanced Materials, University of Groningen, Groningen, The Netherlands

Part I Protein Conjugates

Chapter 1 Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements Using Genetically Encoded Ketone Functionalities Edward A. Lemke Abstract Studies of protein structure and function using single-molecule fluorescence resonance energy transfer (smFRET) benefit dramatically from the ability to site-specifically label proteins with small fluorescent dyes. Genetically encoding the unnatural amino acid (UAA) p-acetylphenylalanine is an efficient way to introduce commercially available fluorescent tags with high yield and specificity. This protocol describes the expression in Escherichia coli of proteins containing this UAA in response to the amber stop codon TAG. Proteins were purified with high yield and subsequently labeled with the hydroxylamine derivative of Alexa Fluor® 488 functioning as a fluorescent donor dye. The proteins were then labeled via maleimide coupling chemistry at a unique cysteine with the acceptor dye Alexa Fluor® 594 to yield a dual-labeled protein ready for subsequent smFRET observation. Key words: Amber stop codon, Genetically encoded, Synthetase, Unnatural amino acid, Oxime ligation, Maleimide coupling, Single-molecule, FRET, Protein dynamics, Site-specific labeling

1. Introduction Labeling proteins with small fluorescent dyes has long been an established tool to make molecular structure and function visible. Among the plethora of different fluorescence techniques, the observation of fluorescence resonance energy transfer (FRET) between a fluorescent donor dye and a fluorescent acceptor dye has emerged as a widely applied tool in biology. FRET occurs if the emission spectrum of a donor dye overlaps with the absorption spectrum of a proximal acceptor dye. Since the efficiency of energy transfer is distance-dependent, an exact measurement of the FRET efficiency can be used to study distances between

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_1, © Springer Science+Business Media, LLC 2011

3

4

Lemke

20 and 100 Å. Due to the pioneering work of Ha, Deniz and coworkers, it has also become possible to observe fluorescence energy transfer on the single-molecule level using highly sensitive detection methods (1, 2). In order to turn such a single-molecule FRET (smFRET) measurement into a structural biology tool to study molecular function, the labeling sites must be precisely known. Recombinantly expressed proteins are typically composed of 20 naturally occurring amino acids (AA). Of those AA, lysine and cysteine have chemical handles (the side-chain amine group of lysine and the thiol group of cysteine) for which common labeling strategies exist to react them efficiently with commercially available dye derivatives that are suitable for smFRET studies. Due to the rare occurrence of cysteine in most proteins, many smFRET studies on proteins entail encoding two surface accessible cysteines into the protein sequence (at labeling sites X and Y). The cysteines are then reacted with maleimide-reactive dyes, typically leading to randomly labeled species, as reaction sites X and Y are essentially chemically indistinguishable. Such a random labeling strategy creates a heterogeneous sample containing a mixture of donor (D) and acceptor (A) dye species attached at positions X and Y (such as XD/YA, XA/YD). If, for example, XA and YA have different photophysical properties, a FRET measurement could yield distinct signals for each species. This can be detrimental for data interpretation and can compromise the unique capability of single-molecule studies to look beyond the average and detect subpopulations, which has greatly contributed to the technique’s success in yielding an unbiased view of even complex biological processes. As a consequence, a randomly labeled, intrinsically heterogeneous sample can cause severe downstream problems in data analysis and interpretation of biologically relevant subpopulations. Genetically encoding the unnatural amino acid (UAA) p-acetylphenylalanine (pAcPhe) (Fig.  1a) using amber suppression technology has created an orthogonal route to achieve sitespecific dual-labeling of proteins with dyes suitable for single-molecule observation (3, 4). Here, the expression host, E. coli, was engineered with an additional aminoacyl-tRNA synthetase/suppressor tRNA pair (aaRS/tRNACUA). This pair was derived from Methanocaldococcus jannaschii and modified in such a way that the synthetase specifically recognizes pAcPhe and aminoacetylates only its cognate tRNACUA (5). This leads to incorporation of this specific UAA during the translation process whenever the TAG codon occurs. E. coli (and many other organisms) tolerates this reprogramming of one of its rare codons. Transforming a plasmid encoding for the pAcPheaaRS/tRNACUA pair into E. coli thus extends the capability of the host to genetically encode the 21st amino acid pAcPhe. Furthermore, the system is

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

a

c

b

O

ara

ClAmp

COOH

d 34 26 17 10

pAcPhe

crude

M

+

POI

tRNACUA

H2N

5

pure

-

+

aaRS

pAcPhe

ara aaRS

Amp

Fig. 1. (a) Structure of p-acetylphenylalanine (pAcPhe). (b) Plasmid pEVOLpAcPhe encoding ClAmp resistance, tRNACUA, and two copies of pAcPheaaRS (one under control of an l-arabinose-inducible promoter). (c) pBAD expression vector encoding Amp resistance and the protein of interest (POI) (T4 Lysozyme in this protocol) under control of an l-arabinose-inducible promoter. (d) Coomassie-stained SDS-PAGE gel showing expression of T4 Lysozyme 38TAG 157C in the presence (+) and absence (−) of pAcPhe after the first (crude) purification step, as well as the final purified 38pAcPhe 157C T4 Lysozyme protein product.

orthogonal to the existing translational machinery in E. coli, i.e., it does not cross-react or interfere with any of the natural aaRS/tRNA pairs. Expression of a protein of interest (POI) using this 21st UAA only requires a few additional steps compared to standard expression procedures. First, a TAG mutation is introduced by standard site-directed mutagenesis at the desired labeling site and cloned into a suitable expression vector. For transformation, this vector has to carry a resistance marker that is compatible with the expression host as well as with the plasmid harboring the pAcPheaaRS/ tRNACUA pair (termed pEVOLpAcPhe; see Fig. 1b) (5). The vector pEVOLpAcPhe carries a gene encoding for chloramphenicol (ClAmp) resistance and was optimized for high yield expression in E. coli. This plasmid encodes two genes for pAcPheaaRS; because one of the synthetase genes is under direct control of an arabinose-inducible promoter, high yield expression of a protein containing pAcPhe requires induction with l-arabinose. The UAA pAcPhe is membrane permeable and thus can be made available for the translation process by simply adding it to the growth medium prior to induction. As the actual incorporation of UAAs into proteins is carried out by the host translation machinery in vivo, the supplied UAA does not need to be stereochemically pure, as pAcPheaaRS recognizes only l-amino acids. And finally, as amber codon suppression is not necessarily 100% efficient, the POI should contain a C-terminal purification handle (such as a histidine tag), so that it can be easily discriminated during the purification process from truncated mutants. The following protocol describes the efficient incorporation of pAcPhe into a single-cysteine mutant of T4 Lysozyme. While protein expression differs from protein to protein, most steps specific

6

Lemke O

SH SO3

SO3

+

O

H2N

NH2

oxime ligation

C Alexa488

H2N

=

_

O

O HN

O

O

HN

H2N

O

O

Alexa488

O N

SH

CH3 H3C

N

CH3

_

=

COH

CH2SO3

Alexa594

_

CH2SO3

O O

HN O

O

N

O

O N

CH3

N

H3C

O

maleimide coupling

Alexa488

CH3 O

N

Alexa594

O N O S

Fig. 2. T4 Lysozyme 38pAcPhe 157C is first labeled with Alexa Fluor® 488 hydroxylamine using oxime ligation at pH 4, and then Alexa Fluor® 594 C5 maleimide is coupled to the cysteine residue at pH 7.4.

to incorporation of pAcPhe will be common to all proteins, as pointed out in the text. Subsequently, the cysteine and the pAcPhe are site-specifically labeled with a commercially available smFRET dye pair (Fig. 2).

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

7

2. Materials 1. E. coli TOP10 electrocompetent cells (Invitrogen Corp., Carlsbad, CA). 2. Glycerol. 3. Luria–Bertani (LB) growth medium. 4. LB agar plates with 50 mg/L ampicillin and 33 mg/L ClAmp. 5. 20% (w/v) l-arabinose, filter-sterilized through a 0.2-mm syringe filter. 6. pAcPhepEVOL (available from Prof. Peter G. Schultz, The Scripps Research Institute, San Diego, CA). 7. 157Cys T4 Lysozyme-6His gene cloned into pBAD-MCS (Invitrogen) using NcoI and KpnI restriction sites. 8. QuikChange® site-directed mutagenesis kit (Agilent Technologies, Santa Clara, CA) or other site-directed mutagenesis kit. 9. Lysis buffer: 20  mM sodium phosphate–HCl, pH 7.6, 500 mM NaCl, 1 mM imidazole, 1 mM PMSF, complete protease inhibitor cocktail (e.g., Roche Diagnostics Corporation, Indianapolis, IN). 10. Mono S ion exchange chromatography column (GE Healthcare, Piscataway, NJ). 11. Fast protein liquid chromatographic (FPLC) equipment. 12. Wash buffer: 20  mM sodium phosphate–HCl, pH 7.6, 500 mM NaCl, 10 mM imidazole, and 1 mM PMSF. 13. Elution buffer: 20  mM sodium phosphate–HCl, pH 7.6, 500 mM NaCl, 500 mM imidazole, and 1 mM PMSF. 14. Mono S buffer A: 50 mM MES–HCl, pH 6.3, 50 mM NaCl, and 5 mM b-mercaptoethanol. 15. Mono S buffer B: 50 mM MES–HCl, pH 6.3, 500 mM NaCl, and 5 mM b-mercaptoethanol. 16. Oxime labeling buffer: 50 mM sodium acetate–HCl, pH 4.0, 150 mM NaCl, and 4 M guanidine HCl. 17. Maleimide labeling buffer: 20 mM sodium phosphate–HCl, pH 7.4, 150 mM NaCl, and 4 M guanidine HCl. 18. Ni-NTA Superflow resin (Qiagen, Valencia, CA). 19. Alexa Fluor® 488 C5-aminooxyacetamide, bis(triethy-lammonium) salt (Alexa Fluor® 488 hydroxylamine) (Invitrogen) (see Note 1). 20. Alexa Fluor® 594 C5 maleimide (Invitrogen). 21. 1  M pAcPhe HCL stock solution. Filter-sterilize through a 0.2-mm syringe filter (e.g., available from Synchem, IL, USA).

8

Lemke

22. Dimethylsulfoxide (anhydrous). 23. Acetonitrile (anhydrous). 24. Dithiothreitol (DTT). 25. Polypropylene chromatography column with filter (Qiagen). 26. NuPage denaturing SDS-PAGE gel (4–12% Bis–Tris Gel) (Invitrogen). 27. SDS-PAGE MOPS running buffer. 28. SDS-PAGE loading dye. 29. SDS-PAGE molecular weight markers. 30. Centricon® centrifugal filter units (10  kDa MWCO) (Millipore, Billerica, MA) or 10 kDa MWCO dialysis tubes. 31. UV–visible spectrophotometer. 32. Mass spectrometer.

3. Methods High-yielding protein expression greatly simplifies subsequent protein purification. For most smFRET experiments, the final purity of the sample is pivotal, because otherwise impurities could get labeled as well and, therefore, contribute to fluorescent background. The pAcPheaaRS/tRNACUA pair is highly efficient in selective incorporation of pAcPhe in E. coli and several proteins have already been expressed in milligram quantities from a 1-L culture (5). Misincorporation of a naturally occurring amino acid instead of pAcPhe would yield protein that does not contain a ketone handle, and that consequently is missing a labeling site. Such impurities are highly undesirable, and the described strategy for UAA incorporation was developed to allow for the preparation of pure modified protein. The unique side-chain ketone functionality of pAcPhe is inert and nontoxic to E. coli, but reacts efficiently with hydroxylamine dye derivatives by forming a stable oxime bond. 3.1. Transformation of E. coli and Starting Culture

1. Perform site-directed mutagenesis (e.g., a QuikChange® ­reaction) to mutate a codon in your protein sequence to TAG. In this example, the codon for amino acid 38 in the 157C T4 Lysozyme was mutated to TAG and confirmed by DNA sequencing. This generates the pBAD plasmid encoding 38TAG 157C T4 Lysozyme (see Note 2). 2. Transform E. coli TOP10 cells with the pAcPhepEVOL (ClAmp resistance, inducible with l-arabinose) and pBAD-38TAG 157C T4 Lysozyme (ampicillin resistant, inducible with l-arabinose) plasmids (see Note 3).

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

9

3. Spread transformed colonies on LB agar plates with 50 mg/L ampicillin and 33 mg/L ClAmp and grow overnight at 37°C (from here on, this antibiotic mixture is referred to as Amp/ ClAmp). Pick an individual colony and grow overnight in LB medium containing Amp/ClAmp at 37°C. Make a 30% glycerol stock from this overnight culture and store at −80°C. Start all further experiments from this glycerol stock by scratching off a small amount and growing an overnight culture in double-selective medium (Amp/ClAmp) (see Note 4). 3.2. Protein Expression

1. Inoculate two 2-L shake flasks (each containing 500 mL liquid LB medium with Amp/ClAmp) with 5 mL from an overnight seed culture and allow cells to grow at 37°C with constant shaking. Add pAcPhe to a final concentration of 1  mM to one of the flasks (+) at an optical density of OD600 = 0.2–0.3. The other flask serves as a control (−). 2. Let the cells continue to grow until mid-log phase (OD600 = 0.4–0.6), and then induce recombinant protein expression by adding l-arabinose at a final concentration of 0.02% (w/v). 3. Transfer cells to a 30°C shaker and shake overnight for 16 h (see Note 5). Pellet cells by centrifugation at 4,000 × g for 20 min. Discard the supernatant and freeze the cell pellet at −80°C.

3.3. Protein Purification (Part I)

1. Resuspend the cell pellet in 3  volumes of lysis buffer, and then lyse the cells using sonication. 2. Centrifuge the lysate at 16,000 × g for 60  min. Collect the supernatant and incubate with 2 mL of Ni-NTA beads (preequilibrated in lysis buffer) for 2  h at 4°C (under constant mixing using a nutator agitation device). 3. Collect the Ni-NTA resin by passing the lysate through a polypropylene column so that the Ni-NTA beads form a bed above the filter at the bottom of the column. Wash the column with 20 bed volumes of wash buffer. 4. Elute the target protein from the column using three bed volumes of elution buffer.

3.4. Expression Test Using SDS-PAGE

1. Take a sample from the (+) and the (−) eluates and run on an SDS-PAGE gel. An exemplar gel is shown in Fig. 1. In the (+) pAcPhe sample lane, an enriched product at the molecular weight of T4 Lysozyme is evident. For comparison, the (−) pAcPhe sample lane contains the negative control where no pAcPhe was added. Besides minor impurities in both samples, substantial expression of T4 Lysozyme is only visible in the (+) UAA sample. This is a strong indication that the incorporation of pAcPhe was successful (see Notes 6 and 7).

10

Lemke

3.5. Protein Purification (Part II)

Only the (+) UAA expression sample crudely purified in Subheading 3.3 will contain pAcPhe. Therefore, only the pAcPhecontaining sample (+) is used for further purification. 1. Transfer the sample to a Centricon® concentration device with a molecular weight cutoff of 10 kDa, concentrate and wash three times with Mono S buffer A (alternatively, use dialysis for buffer exchange). At this point, a white precipitate forms (see Note 8). Discard the white precipitate material since it originates from protein that is not soluble at pH 6.3 and does not contain the target protein T4 Lysozyme. 2. Concentrate the protein sample to 1  mL and load onto a Mono S column at a flow rate of 0.5 mL/min using an FPLC purification system. After a stable baseline is reached, apply a salt gradient at a flow rate of 1 mL/min (from 100% Mono S buffer A to 100% Mono S buffer B over 30 min). T4 Lysozyme typically elutes around a salt concentration of 60% Mono S buffer B. Analyze the collected FPLC fractions on SDS-PAGE gels (see Note 9). 3. Pool the fractions containing purified T4 Lysozyme together and concentrate in a Centricon® filter device. At this point, exchange the buffer using three repeated washes with oxime labeling buffer. Measure the concentration of protein on a UV–visible spectrophotometer (see Note 10). The yield of 38pAcPhe 157C T4 Lysozyme is typically >15  mg from a 0.5-L expression culture. Confirm successful purification by mass spectrometry (see Note 7).

3.6. Oxime Ligation of T4 Lysozyme Containing pAcPhe with Axela 488 Hydroxylamine Dye

1. Mix 50 mL of 200 mM 38pAcPhe 157C T4 Lysozyme protein with 50  mL of 1  mM Alexa Fluor® 488 hydroxylamine dye (i.e., fivefold molar excess of dye) in pH 4 oxime labeling buffer (see Note 11). Allow the reaction to proceed for 36 h at 37°C in the dark. 2. Wash the labeled protein three times with maleimide labeling buffer. 3. Efficient labeling is verified by SDS-PAGE (see Note 9), UV–visible spectrophotometry (see Note 10), and mass spectrometry (see Note 7).

3.7. Aliquoting Alexa Fluor ® 594 C5 Maleimide for Multiple Reactions

Maleimide-reactive dyes are typically supplied commercially by companies in 1-mg quantities. However, for most single-molecule FRET studies a couple of nanomoles of dual-labeled protein is sufficient, and thus it is desirable to aliquot the maleimide dye for separate reactions. In contrast to most hydroxylamine dye derivates, the maleimide functional group is not stable in water. To allow for multiple reactions out of a single 1-mg sample of Alexa Fluor® 594 C5 maleimide dye, carefully aliquot the dye

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

11

under inert gas (either provided by constant nitrogen or argon flow, or by using a glove box). 1. Solubilize 1 mg of Alexa Fluor® 594 C5 maleimide in anhydrous acetonitrile (see Note 12) 2. Aliquot the Axela Fluor® 594 C5 maleimide solution, e.g., in 25-nmol fractions into dry microcentrifuge tubes 3. Lyophilize the aliquots and store at −80°C. 3.8. Coupling of Alexa Fluor ® 594 C5 Maleimide Dye to the Cysteine Residue of T4 Lysozyme

1. Wash the Alexa Fluor® 488-labeled protein three times with maleimide buffer + 10 mM DTT, and then subsequently wash three more times with maleimide labeling buffer (see Note 13). 2. Set up the following labeling reaction using a protein:Alexa Fluor® 594 dye labeling ratio of 1:1.5. Prepare a solution containing 10  nmol of protein in 80  mL of maleimide labeling buffer. Next, solubilize 15  nmol of Alexa Fluor® 594 C5 maleimide in 20  mL of DMSO and pipette the dye solution into the protein solution while gently vortexing the protein in a microcentrifuge tube. This instantaneous mixing procedure prevents any formation of precipitate. Allow the reaction to proceed for 1 h at room temperature (or overnight at 4°C) and then quench the reaction by adding DTT to a final concentration of 50 mM or proceed immediately to the next step. 3. Exchange the protein into water either by dialysis or by repeated washes in a Centricon® device and subsequently lyophilize for long-term storage (in the dark) at −80°C. 4. Analyze the final sample by using SDS-PAGE (see Note 9), UV–visible spectrophotometry (see Note 10), and mass spectrometry (see Note 7). Typically, site-specific dual-labeling efficiencies are larger than 90% for each labeling site. The lyophilized T4 Lysozyme will be resolubilized best in a denaturing buffer, at which point it is now ready for subsequent single-molecule FRET studies, such as those described in Lemke et al. (3) or Brustad et al. (4).

4. Notes 1. To prepare 1 mM Alexa Fluor® 488 hydroxylamine solution, dissolve 1 mg of dye in pH 4 oxime labeling buffer. This solution can be stored frozen in the dark at −20°C and thawed prior to use. 2. Ideally, only surface-accessible mutants that are unlikely to interfere with protein folding, structure, and function should be chosen as amber suppression sites. Furthermore, it is

12

Lemke

imperative that the stop codon used for termination of the protein sequence is not TAG, but either TAA or TGA. 3. If expression problems occur, make a new glycerol stock of transformed E. coli TOP10 cells (Subheading  3.1, step 2). Further, the pEVOL plasmid should be analyzed for unwanted mutations specifically in the tRNA and aaRS coding regions. 4. Protein expression conditions typically need to be adapted for every given POI. The pAcPhepEVOL system is compatible with most commercially available E. coli hosts. Only a few things need to be considered: (a) the antibiotic selection markers of the plasmids used must be unique and compatible with each other and with the host; (b) the host must support expression with l-arabinose; and (c) it is advisable that the UAA pAcPhe is added 30 min prior to induction. This gives sufficient time for the tRNACUA to become charged with pAcPhe. pAcPhe is a very stable UAA and can in principle also be added to the medium right from the beginning. 5. Expression at 30°C was chosen for this particular protein. Other growth conditions and temperatures can be used for other proteins. Different expression conditions can also affect amber suppression. However, stable, high-yielding amber suppression using pAcPheaaRS/tRNACUA has been observed across a wide spectrum of common expression conditions. 6. The (+)/(−) pattern is a strong indication for successful incorporation of the UAA pAcPhe. However, the following points may be considered if problems occur. The presence of a band in the (−) pAcPhe lane does not necessarily report a failure of the experiment. In the presence of pAcPhe, the binding pocket is occupied by pAcPhe due to the high affinity of the synthetase (pAcPheaaRS) for this specific UAA. In the absence of UAA, it is possible for the binding pocket of the synthetase to be filled by an unwanted substrate that is then charged to the tRNACUA. In such a case, both the (+) and the (−) lanes will show expression of the target protein, but pAcPhe is only incorporated in the (+) UAA case. The (+)/(−) band ratio is also dependent on various factors, including type of growth medium, growth conditions, and concentration of pAcPhe in the medium. Increasing the pAcPhe concentration to 5 mM can increase the yield of the target protein incorporating pAcPhe. Independent of the appearance of the (+)/(−) ratio, high-resolution mass spectrometry must be used to verify sufficient (>90%) incorporation of pAcPhe in the (+) UAA experiment (see Note 10). If the (+)/(−) ratio is not satisfactory, the following steps can be taken: perform protein expression in different growth media (such as TYT, TB) at different temperatures and vary the expression time. As the pAcPheaaRS/ tRNACUA pair is evolved from a wild-type tyrosine synthetase,

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

13

phenylalanine (F) and tyrosine (Y) incorporation are the most likely types of misincorporations. Often, expression in minimal media (lacking Y/F) can dramatically increase specificity. 7. Most essential is the confirmation of successful incorporation of pAcPhe at the specific protein site using high-resolution mass spectrometry, such as electrospray ionization or matrixassisted laser desorption/ionization. The expected mass from the protein can be calculated using standard protein mass calculators (such as the peptide mass calculator available via the http:// www.expasy.ch webpage). Because most calculators do not easily allow for modified unnatural residues, the most straightforward way to do this is to substitute a tyrosine (Y) at the position of the expected pAcPhe incorporation site and add 26 Da, which reflects the mass increase when substituting the tyrosine hydroxyl with a ketone functionality. To remove further uncertainties in the mass calculation due to, e.g., protein modifications (such as acetylation), one should also compare the molecular weight to a sample encoding tyrosine at the mutation site. Additionally, the protein sample can be digested with trypsin and peptide masses be analyzed. In our hands, the exact mass of most of the unmodified peptides was confirmed, but in our specific case the peptide containing pAcPhe was not detected (SPXLNAAK with X = pAcPhe; expected mass (M + H) = 889.46 Da). However, after the first labeling reaction, the same mass spectrometric analyses were repeated and a mass increase of the whole protein of 673 Da was verified, confirming labeling with Alexa Fluor® 488 hydroxylamine. The labeled peptide SPXLNAAK could now also be identified (with X = pAcPhe + Alexa Fluor® 488 hydroxylamine; (M + H) = 1,562 Da). After the second labeling reaction with Alexa Fluor® 594 C5 maleimide, a further mass increase of the whole protein of 886  Da was also confirmed. After tryptic digestion, this labeled peptide was also identified (TGCWDAYK + Alexa Fluor® 594; (M + H) = 1,830 Da). 8. The Mono S purification step is specific for T4 Lysozyme. Other proteins may require a different purification strategy to yield pure protein. Make sure that the molecular weight cutoff of the centrifugal filter or dialysis devices is compatible with the molecular weight of the target protein. 9. SDS-PAGE analysis is used to confirm the expression of the POI. When using a fluorescence scanner, this can also be used to rapidly confirm fluorescence labeling. For detecting a fluorescently-labeled band, the gel should be run in the dark and analyzed on a fluorescence scanner prior to Coomassie staining (otherwise Coomassie can quench the fluorescence signal). After the fluorescence scan, the gel can be stained with Coomassie. The fluorescence image should show a band

14

Lemke

clearly overlaying the Coomassie-stained band. Additional fluorescence bands can pinpoint labeled impurities in the gel (fluorescence detection is very sensitive to even minor impurities) or free dye still present in the sample. 10. After the purification of the protein, the concentration of the protein should be measured using a standard UV–visible spectrophotometer. If the signal-to-noise of the UV–visible absorbance spectrum is not satisfactory, increase the concentration of the protein. Furthermore, exchanging the protein into a low-salt buffer (such as water) can yield a better signal, as salts often contribute to a high level of background absorption in the UV range. The total concentration of pure protein can be determined by measuring the absorbance A280 at the wavelength of 280 nm on a UV–visible spectrophotometer. Using the extinction coefficient of e280, T4L ~ 26,000 cm−1 M−1, the concentration of pure protein can be calculated according to c protein = A280 / e 280,protein . After each labeling reaction, the labeling efficiency cdye/cprotein is determined by calculating the concentration of fluorescent dye and the concentration of protein in the sample as given by c dye = Adye / e dye and c protein = (A280 - corr280 ·Adye ) / e 280,protein , respectively, where edye is the extinction coefficient for the dye at its maximal absorption wavelength, lmax. Most fluorescent dyes also absorb light at 280 nm, which has to be corrected for when calculating the protein concentration using the corr280 parameter. Typically, the extinction coefficient for a given dye, as well as the correction factor (corr280), is available from the manufacturer’s manual. For the specific dyes used in this example protocol, the values are as follows: (1) Alexa Fluor® 488: e495 ~ 71,000 cm−1 M−1, lmax = 495 nm, and corr280 = 0.11; and (2) Alexa Fluor® 594: e594 ~ 90,000 cm−1 M−1, lmax = 590 nm, and corr280 = 0.56. Typical labeling efficiencies obtained for this example protocol are >90%. As the two labeling sites are orthogonal to each other, the order of labeling can in principle be reversed, i.e., one may first perform the maleimide reaction and then perform the oxime ligation. 11. The oxime ligation is mainly dependent on pH, concentration of educts, time, and temperature. Labeling efficiencies can be increased by using higher concentration of educts (i.e., protein and/or dye), incubation at 37°C for a longer time period, or raising the temperature to, e.g., 65°C. The optimum of the reaction is around pH 4. Thus, if a higher pH is needed, then the temperature, concentration, and incubation time should be optimized accordingly. Furthermore, the additional catalysts can also be used (6). 12. Many maleimide-reactive dyes are not readily soluble in low amounts of acetonitrile. Increasing the amount of acetonitrile

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

15

as well as extensive vortexing and sonication on ice can assist solubility. The dye must not be exposed to ambient air or water. 13. The presence of residual thiol-containing contaminants is the most typical cause for low labeling efficiencies. Furthermore, b-mercaptoethanol used in previous purification steps can block the thiol group of the free cysteine. To avoid this, extensively wash or dialyze the protein prior to labeling with buffer containing 10 mM DTT, followed by extensive washes in fresh, deoxygenated thiol-free buffer to remove even trace amounts of thiol-containing compounds.

Acknowledgments I thank all members in the Deniz and Schultz laboratories at The Scripps Research Institute, especially Dr. Brustad for the good collaborations. The critical proofreading of this protocol by Dr. VanDelinder is also very much appreciated. Finally, I want to thank my laboratory members for their productive discussions, and the EMBL and the Emmy Noether Program of the DFG for funding my laboratory. References 1. Ha, T., Enderle, T., Ogletree, D. F., Chemla, D. S., Selvin, P. R., and Weiss, S. (1996) Probing the interaction between two single molecules: fluorescence resonance energy transfer between a single donor and a single acceptor. Proc. Natl. Acad. Sci. USA 93, 6264–6268. 2. Deniz, A. A., Dahan, M., Grunwell, J. R., Ha, T., Faulhaber, A. E., Chemla, D. S., Weiss, S., and Schultz, P. G. (1999) Single-pair fluorescence resonance energy transfer on freely diffusing molecules: observation of Forster distance dependence and subpopulations. Proc. Natl. Acad. Sci. USA 96, 3670–3675. 3. Brustad, E. M., Lemke, E. A., Schultz, P. G., and Deniz, A. A. (2008) A general and efficient method for the site-specific dual-labeling of

proteins for single molecule fluorescence resonance energy transfer. J. Am. Chem. Soc. 130, 17664–17665. 4. Lemke, E. A., Gambin, Y., Vandelinder, V., Brustad, E. M., Liu, H.-W., Schultz, P. G., Groisman, A., and Deniz, A. A. (2009) Microfluidic Device for Single-Molecule Experiments with Enhanced Photostability. J. Am. Chem. Soc. 131, 13610–13612. 5. Young, T. S., Ahmad, I., Yin, J. A., and Schultz, P. G. (2009) An Enhanced System for Unnatural Amino Acid Mutagenesis in E. coli. J. Mol. Biol. 15, 361–367. 6. Dirksen, A., Hackeng, T. M., and Dawson, P. E. (2006) Nucleophilic catalysis of oxime ligation. Angew. Chem. Int. Ed. Engl. 45, 7581–7584.

Chapter 2 Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins and Small Molecules Using Microbial Transglutaminase Ahmed Besheer, Thomas C. Hertel, Jörg Kressler, Karsten Mäder, and Markus Pietzsch Abstract Hydroxyethyl starch (HES) is a water-soluble, biodegradable derivative of starch that is widely used in biomedicine as a plasma volume expander. Due to its favorable properties, HES is currently being investigated at the industrial and academic levels as a biodegradable polymer substitute for polyethylene glycol. To date, only chemical methods have been suggested for HESylation; unfortunately, however, these may have negative effects on protein stability. To address this issue, we have developed an enzymatic method for protein HESylation using recombinant microbial transglutaminase (rMTG). rMTG enzyme is able to catalyze the replacement of the amide ammonia at the g-position in glutamine residues (acyl donors) with a variety of primary amines (acyl acceptors), including the amino group of lysine (Lys). To convert HES into a suitable substrate for rMTG, the polymer was derivatized with either N-carbobenzyloxy glutaminyl glycine (Z-QG) or hexamethylenediamine to act as an acyl donor or acyl acceptor, respectively. Using SDS-PAGE, it was possible to show that the modified HES successfully coupled to test compounds, proving that it is accepted as a substrate by rMTG. Overall, the enzymatic approach described in this chapter provides a facile route to produce biodegradable polymer–drug and polymer–protein conjugates under relatively mild reaction conditions. Key words: Hydroxyethyl starch, Biodegradable polymer, Recombinant microbial transglutaminase, Polymer–drug conjugates, Polymer–protein conjugates

1. Introduction The surface modification of therapeutic proteins by coupling them to water-soluble polymers imparts a number of advantages, such as increased water solubility, increased circulation time, reduced levels of aggregate formation, reduced immunogenicity,

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_2, © Springer Science+Business Media, LLC 2011

17

18

Besheer et al.

and increased stability against proteolytic digestion. The gold standard for this approach is PEGylation (coupling to poly(ethylene glycol), PEG), and many types of PEGylated proteins are already commercially available (1). However, PEG is not biodegradable, raising concerns about its fate and effects after chronic use in large doses. This has motivated academia and industry to search for biodegradable substitutes, such as albumin and hydroxyethyl starch (HES). The latter is a semisynthetic biodegradable polymer that is currently used commercially as a blood plasma volume expander. Its favorable properties, such as high water solubility, low hypersensitivity, and the possibility to tailor its molar mass and biodegradation rate, have recently attracted much attention to HES as a promising substitute for PEG. Consequently, the polymer has been investigated for the stabilization of nanoparticles (2) and for protein conjugation in order to evaluate HESylation as a potential alternative to conventional PEGylation of proteins (3). There are several chemical approaches to couple biomacromolecules to polymers; however, many of them can have detrimental effects on proteins, which are usually quite sensitive to their surrounding environment. Accordingly, enzymatic methods have been proposed and tested as gentle alternatives to chemicalbased coupling strategies. Among these, enzymatic coupling using microbial transglutaminase (MTG) has earned a significant degree of attention in the development of site-specific conjugation strategies (4). MTG catalyzes the replacement of the amide ammonia at the g-position in glutamine residues (acyl donors) with a variety of primary amines (acyl acceptors), including the e-amino group of lysine (5). Furthermore, MTG has a number of advantageous properties over eukaryotic TG – including being a calcium-independent enzyme and having fewer substrate specificity requirements (5) – and has thus found application in the food industry for crosslinking meat and fish products (5). In biomedical applications, Sato et al. provided a clear demonstration of the power of using MTG for site-specific protein–polymer conjugation when they used the enzyme to couple alkylamine derivatives of PEG selectively to a glutamine residue (Gln74) of recombinant human interleukin-2 (6). Fontana et  al. have recently reviewed the relationship between the catalytic activity of MTG and the substrate structural characteristics, and concluded that both the primary structure (viz., the presence of nearby hydrophobic residues) and tertiary structure (viz., chain flexibility) of the substrate are important factors for site specificity (7). In this chapter, we describe protocols for the modification of HES with hexamethylenediamine (HMDA) as well as with N-carbobenzyloxy glutaminyl glycine (Z-QG) to act as substrates for MTG (both as acyl acceptor and acyl donor, respectively). In addition, we present examples of the reaction of the modified

19

Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins

HES polymer with model compounds to demonstrate the feasibility of this conjugation strategy. To carry out these conjugation reactions, we utilize a highly purified recombinant MTG (rMTG) carrying a polyhistidine tag (His-tag) at the C terminus, which has recently been developed in our laboratory (8, 9). The incorporation of a His-tag into rMTG facilitates the workup and separation of the products, and further provides the possibility of immobilizing the enzyme without loss of activity (10).

2. Materials 2.1. Modification of HES to Carry a Lysine-Like Residue (HES 70-Amine)

1. HES (70  kDa) (HES 70) (Serumwerke Bernburg AG, Bernburg, Germany). 2. Tosyl chloride. 3. HMDA. 4. Triethylamine. 5. 50 mM borate buffer, pH 10.

2.2. Modification of HES to Carry a Glutamine Residue (HES 70-GQ-Z)

1. HES 70 (Serumwerke Bernburg AG). 2. Dicyclohexylcarbodiimide (DCC). 3. N-Hydroxysuccinimide (NHS). 4. 4-(Dimethylamino)pyridine (DMAP). 5. N-Carbobenzyloxy glutaminyl glycine Bubendorf, Switzerland). Store at −20°C.

2.3. Reaction of HES 70-Amine with Dimethylcasein

(Bachem

AG,

1. Recombinant microbial transglutaminase (rMTG) (see Notes 1–3). 2. HES 70-Amine, as prepared in Subheading 3.2. 3. N,N-Dimethylcasein (DMC). 4. Human serum albumin (HSA), 20% (w/v). Store at 4°C. 5. Dithiothreitol (DTT). 6. 50 mM Tris–HCL buffer, pH 8. 7. Sample buffer (2×): Dissolve 1.21  g of Tris–HCl, 2.5  g of SDS, 50 mg of bromophenol blue, 10 g of glycerol, and 42 g urea in 95 ml of ultrapure water. Adjust the pH to 8.0 with HCl, and bring up to 100 ml with ultrapure water.

2.4. Reaction of HES 70-GQ-Z with Monodansyl Cadaverine

1. Recombinant microbial transglutaminase (see Notes 1–3). 2. HES 70-GQ-Z, as prepared in Subheading 3.3. 3. MDC. 4. N,N-Dimethylcasein.

20

Besheer et al.

5. HES 70 (Serumwerke Bernburg AG). 6. 50 mM Tris–HCL buffer, pH 8. 7. Sample buffer (2×) with b-mercaptoethanol: Prepare as described in Subheading 2.3, item 7, and then add 10 ml of b-mercaptoethanol/ml sample buffer before use. 2.5. SDSPolyacrylamide Gel Electrophoresis

1. Mighty Small gel electrophoresis unit (Hoefer, Inc., Holliston, MA). 2. Separating gel buffer: Add 18.18 g of Tris–HCL, 0.4 g of SDS, and 100 ml of 10% (w/v) NaN3 to 80 ml distilled water. Adjust the pH to 8.8 with 4N HCl (approximately 6 ml), and then bring up to 100 ml with water. Store at room temperature. 3. Stacking gel buffer: Add 6.06 g of Tris–HCL, 0.4 g of SDS, and 100  ml of 10% (w/v) NaN3 to 70  ml distilled water. Adjust to pH 6.8 with 4 N HCl, and then bring up to 100 ml with water. Store at room temperature. 4. 30% (w/v) acrylamide/bisacrylamide solution: Dissolve 29.1 g of acrylamide and 0.9 g of bisacrylamide in distilled water and bring up to 100 ml. CAUTION: Acrylamide is a neurotoxin and a suspected carcinogen when unpolymerized, so use extreme care and wear gloves and goggles when handling. 5. N,N,N,N ’-Tetramethylethylenediamine (TEMED). 6. 10% (w/v) Ammonium persulfate (APS): Dissolve 100 mg of APS in 1  ml of distilled water. Prepare immediately before use. 7. Water-saturated isobutanol: Shake equal volumes of water and isobutanol in a glass bottle and allow the phases to separate. Recover the organic (top) layer and store at room temperature. 8. Running buffer (10×): Mix 30.28 g of Tris–HCL, 144 g of glycine, 10 g of SDS, and 1 ml of 10% (w/v) NaN3 to distilled water and bring to 1 l. Store at room temperature. 9. Prestained molecular weight marker (Fermentas GmbH, Germany): b-Galactosidase (116.0  kDa), BSA (66.2  kDa), ovalbumin (45.0 kDa), lactate (35.0 kDa), restriction endonuclease Bsp98I (25.0 kDa), b-lactoglobulin (18.4 kDa), and lysozyme (14.4 kDa). 10. Staining solution: Dissolve one PhastGel™ Blue R (Coomassie R 350 stain) tablet (GE Healthcare) in a solution containing 50 ml of acetic acid, 100 ml of isopropanol, and 150 ml of distilled water. Accelerate the dissolution of the tablet by stirring. Filter the staining solution before use. 11. Destaining solution: Add 20 ml of isopropanol to 10 ml of acetic acid, and then bring up to 100 ml with distilled water.

Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins

21

3. Methods 3.1. Modification of HES to Carry a Lysine-like Residue (HES 70-Amine) (see Note 4)

1. Dry 1 g of HES 70 (5.4 mM of the anhydroglucose units, AGU) at 105°C for 2 h. 2. Dissolve the dried HES 70 in a solution containing 10 ml of dry DMF and 1 ml of triethylamine at 60°C. 3. Dissolve tosyl chloride (0.5 g, 2.6 mM) in 1 ml of dry DMF. 4. Cool both solutions on ice to 0°C and protect from light. 5. Add the tosyl chloride solution gradually to the HES 70 solution and stir at 0°C for 1 h. 6. Precipitate the polymer solution in 100 ml of cold acetone, filter, and wash with another 100 ml of acetone. 7. Dissolve the precipitate in water and dialyze against distilled water for 3 days (6–8 kDa MWCO), then lyophilize. 8. From the prepared HES tosylate, dissolve 200 mg in 30 ml DMF/borate buffer, pH 10 (1:2, v/v). 9. Add an excess of HMDA (500 mg, 4.3 mM) dissolved in 10 ml of DMF/borate buffer, pH 10 (1:2, v/v), and stir overnight. 10. Precipitate the polymer in 200 ml of isopropanol/methanol (1:1, v/v), filter, and wash with 100 ml of the precipitating solvent. 11. Dry the precipitate at room temperature for 2 days. 12. Characterize the HES 70-amine product by 1H NMR (D2O): d = 1.27 (broad, 4H, –NH–(CH2)2–(CH2)2–(CH2)2–NH2), 1.5 (broad, 4H, –NH–CH2–CH2–(CH2)2–CH2–CH2–NH2), 5.1–5.7 (broad, 1H, HC– anomeric carbon of AGU).

3.2. Modification of HES to Carry a Glutamine Residue (HES 70-GQ-Z) (see Note 5)

1. Dissolve 185 mg (0.55 mM) of Z-QG, 114 mg (0.55 mM) of DCC, 66 mg (0.55 mM) of DMAP, and 64 mg (0.55 mM) of NHS in 2 ml of dry DMSO. Leave the mixture to react for 24 h under stirring at 400 rpm. 2. Dry 1 g of HES 70 (5.4 mM of the AGU) at 105°C for 2 h, and then dissolve the dried HES 70 in 10 ml of dry DMSO. 3. Filter the solution of activated Z-QG (Z-QG succinimidyl ester) to remove the insoluble byproduct of the reaction, dicyclohexylurea. 4. Add the filtrate from step 3 directly to the HES 70 solution (obtained from step 2) and stir for 6 h. 5. Dialyze the polymer solution against distilled water for 3 days (6–8 kDa MWCO), filter, and then lyophilize. 6. Characterize the HES 70-GQ-Z product by 1H NMR (D2O): d = 2.26 (broad, 2H, –CH2–CO–NH2), 5.03 (broad, 2H, –O–CH2–C6H5), (broad, 1H, HC– anomeric carbon of AGU), 7.31 (broad, 5H, –O–CH2–C6H5).

22

Besheer et al.

3.3. Reaction of HES 70-Amine with Dimethylcasein (see Note 6)

The reaction of HES 70-amine with DMC (step 1) can be performed simultaneously with appropriate control experiments, such as those described in steps 2 and 3 below. 1. Add 100 ml of HES 70-amine (5 mg/ml) to 100 ml of DMC (5 mg/ml), both dissolved in 0.1 M Tris–HCL buffer (pH 8). Next, add 100 ml of rMTG (15 U/ml) in glycerol to the mixture and incubate for 1 h at 37°C. 2. Positive control experiment (see Note 7): Add 200 ml of 0.2% (w/v) HSA in 0.1 M Tris–HCL buffer (pH 8) with 10 mM DTT to 100 ml of rMTG (15 U/ml) in glycerol. Incubate the reaction mixture at 37°C for 1 h. 3. Negative control experiment: Add 200 ml of DMC (5 mg/ml) in 0.1 M Tris–HCL buffer (pH 8) to 100 ml of rMTG (15 U/ ml) in glycerol. Do not add HES 70-amine. Incubate the reaction mixture at 37°C for 1 h. 4. Withdraw samples of the reaction mixture at 10, 20, 30, and 60 min, and mix with an equal volume of 2× sample buffer. Boil each sample at 99°C for 3 min. Analyze the samples by SDS-PAGE as described in Subheading 3.5 below (see Note 8). Stain the resulting gel with Coomassie Blue. Figure  1 shows an example of an SDS-PAGE gel image obtained for the coupling of HES 70-amine to DMC using rMTG.

3.4. Reaction of HES 70-GQ-Z with MDC

The reaction of HES 70-GQ-Z with MDC (step 1) can be performed simultaneously with appropriate control experiments, such as those described in steps 2, 3, and 4 below. 1. Add 25 ml of 5 mM MDC in 0.1 M acetic acid to 75 ml of a 1% (w/v) solution of HES 70-GQ-Z in 0.1  M Tris–HCL buffer (pH 8). Next, add 100 ml of rMTG (15 U/ml) in glycerol to the mixture and incubate at 37°C for 1 h. 2. Positive control experiment: Add 25  ml of 5  mM MDC in 0.1  M acetic acid to 75  ml of DMC (5  mg/ml) in 0.1  M Tris–HCL buffer (pH 8). Next, add 100 ml of rMTG (15 U/ ml) in glycerol to the mixture and incubate at 37°C for 1 h. 3. Negative control experiment 1: Add 25 ml of 5 mM MDC in 0.1 M acetic acid to 75 ml of a 1% (w/v) solution of unmodified HES 70 in 0.1 M Tris–HCL buffer (pH 8). Next, add 100 ml of rMTG (15 U/ml) in glycerol to the mixture and incubate at 37°C for 1 h. 4. Negative control experiment 2: Add 25 ml of 5 mM MDC in 0.1  M acetic acid to 75  ml of a 1% (w/v) solution of HES 70-GQ-Z in 0.1  M Tris–HCL buffer (pH 8). Next, add 100 ml of glycerol buffer (do not add rMTG) to the mixture and incubate at 37°C for 1 h.

Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins

23

Fig. 1. SDS-PAGE analysis of the coupling of HES 70-amine to DMC using rMTG. Lane 1: Protein molecular weight markers, Lane 2: HES 70-amine + DMC (no rMTG), Lane 3: HES 70-amine + rMTG + DMC after 10 min, Lane 4: 20 min, Lane 5: 30 min, Lane 6: 60 min, Lane 7: DMC alone, Lane 8: DMC + rMTG after 60 min, Lane 9: HAS + DTT, and Lane 10: HSA + DTT + rMTG after 60 min. The gel was stained with Coomassie blue. Reproduced with permission from ref. 3 © 2009 John Wiley & Sons, Inc.

5. Withdraw samples (15  ml) from each experiment and mix with 2× sample buffer (15 ml). Boil at 99°C for 3 min. Analyze the samples by SDS-PAGE as described in Subheading 3.5. 6. Examine the gels under UV light (365 nm excitation filter; 520 nm emission filter). Figure 2 shows an example of a fluorescence gel image obtai­ ned for the coupling of HES 70-GQ-Z to MDC using rMTG. 3.5. SDSPolyacrylamide Gel Electrophoresis

The following procedure describes the use of a Mighty Small electrophoresis unit from Hoefer, Inc., but can be easily modified for use with other mini-vertical gel electrophoresis systems. 1. Assemble the gel sandwich stack (comprised of one notched alumina or glass plate, one rectangular glass plate and two spacers) and slide it into the casting clamp assembly. 2. With the middle screws lightly tightened, align the plates and spacers of the gel sandwich stack so that they protrude slightly (~1  mm) from the bottom of the casting clamp assembly. Next, secure the gel sandwich stack in place by tightening all the remaining screws until they are finger-tight.

24

Besheer et al.

Fig.  2. Fluorescence image of the SDS-PAGE gel obtained for the coupling of HES 70-GQ-Z to MDC using rMTG. Lane 1: HES 70-GQ-Z + rMTG + MDC, Lane 2: HES 70 + rMTG + MDC, Lane 3: HES 70-GQ-Z + MDC (no rMTG), and Lane 4: DMC + rMTG + MDC. From top to bottom, the arrows point to HES 70-GQ-Z coupled to MDC; the band of DMC coupled to MDC; and uncoupled MDC. The chemical structure of MDC is also shown below the fluorescence gel image. Reproduced with permission from ref. 3 © 2009 John Wiley & Sons, Inc.

3. Place the clamp assembly in the casting cradle, with the screws facing outward. In this position, the gel will be visible through the rectangular glass plate. Ensure that there is a good seal between the bottom of the gel sandwich stack and the rubber gasket in the casting cradle.

Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins

25

4. Prepare the separating gel in the following order (stir the mixture before and after adding APS to ensure homogeneous polymerization): Distilled water (3.3 ml), separating gel buffer (2.5  ml), acrylamide/bisacrylamide (4.2  ml), TEMED (10 ml), and APS (20 ml). 5. Pour the separating gel into the sandwich stack. Fill the sandwich stack until the gel solution reaches a few centimeters below the top of the rectangular glass plate so that there is sufficient room left for pouring the stacking gel (step 9). 6. Overlay the separating gel with 500  ml of isopropanol and allow the gel to polymerize for 30 min. 7. Remove the isopropanol by simply inverting the whole gel casting assembly and discarding the drained liquid into an organic solvent waste container. 8. Prepare the stacking gel in the following order (stir the mixture before and after adding APS to ensure homogenous polymerization): Distilled water (3  ml), stacking gel buffer (1.25  ml), acrylamide/bisacrylamide (0.75  ml), TEMED (10 ml), and APS (8 ml). 9. Pour the stacking gel onto the separating gel until the gel solution reaches the top of the glass plate. Insert the comb and allow the gel to polymerize. 10. Prepare the running buffer by diluting 30 ml of 10× concentrated running buffer stock solution with 270 ml of distilled water. 11. Transfer the polymerized gel sandwich stack to the electrophoresis unit and secure it in place. 12. Connect the coolant ports of the electrophoresis unit to a circulating (cold) water bath. 13. To aid in loading the samples into the gel, wet the transparent well-locating decal and apply it to the front of the glass plate so that the appropriate edge outlines the sample wells. 14. Fill the upper buffer chamber with running buffer and then remove the comb. 15. Load the samples into the wells of the gel (load 10 ml of the samples and 5 ml of the protein molecular weight markers). 16. Fill the lower buffer chamber with running buffer and install the safety lid onto the electrophoresis unit. 17. Connect the electrophoresis unit to a power supply and run the gel at 40 mA (constant current mode) for ~6 min through the stacking gel, and then run the gel at 30 mA for ~40 min through the separating gel. (When running two gels, run at 80  mA for 6  min through the stacking gel and 60  mA for 40 min through the separating gel.)

26

Besheer et al.

18. After the bromophenol blue tracking dye has reached the end of the gel (or has passed completely through the gel), turn off the power and disconnect the electrophoresis unit from the power supply. Also, stop the circulating water bath. 19. To take the gel out of the sandwich stack, remove the spacers first and then carefully pry the aluminum backing plate with a spatula to strip the gel from the glass plate. 20. Briefly rinse the gel with distilled water. 21. To stain the protein bands, gently agitate the gel in staining solution for ~1 h. 22. Rinse the gel briefly with destaining solution, and then agitate the gel in destaining solution for 30 min. Replace with fresh destaining solution and agitate the gel for an additional 30 min (at this point, the gel background should not contain any blue color). 23. Analyze the stained gel using a gel imaging system.

4. Notes 1. The recombinant biocatalyst (inactive pro-TG) can be produced in soluble form using Escherichia coli and IPTG as an inductor as described in detail in ref. 9. A very important point to note is the temperature shift prior to induction (from 37°C to 24°C) to prevent the formation of inclusion bodies. This step reduces the transcription and translation velocity, delivering the pro-enzyme in limited amounts and enabling the protein to fold properly. Alternatively, the autoinduction medium (9) can be used at an incubation temperature of 28°C. 2. Inactive pro-TG enzyme can be purified using metal affinity chromatography and is activated prior to use by removal of the pro-sequence using a suitable protease, such as dispase (8). The endogenous protease from Streptomyces mobaraensis can also be used instead; however, this enzyme is not commercially available and must be prepared in the laboratory. 3. In principle, commercially available transglutaminase can be used instead of the recombinant enzyme. However, commercial preparations of transglutaminase can in some cases be relatively crude and may contain proteases (which may not be visible on a Coomassie-stained SDS-PAGE gel). Such proteases can cause unwanted degradation of the protein substrate. 4. MTG is known to display a high degree of selectivity for glutamine (Gln) as the acyl donor, while the acyl acceptor can either be a lysine (Lys) residue or simply a primary amine

Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins

27

group attached to an alkyl chain having at least four carbon atoms. Accordingly, we used HMDA for the modification of HES, as well as MDC as a model acyl acceptor, where both have primary amine groups attached to alkyl chains with six and five carbon atoms, respectively. 5. Due to the fact that DCC can lead to the oxidation of alcohols in DMSO, the coupling of Z-QG to HES 70 is carried out in two steps: First, the activation of Z-QG using DCC and NHS is performed. Following this, the coupling of HES 70 to the activated peptide is achieved. 6. DMC is derived from casein by methylation (i.e., blocking) of all the lysine amino groups. Consequently, DMC can act as an acyl donor (i.e., presenting only glutamine residues) and is a useful compound for performing test conjugation reactions with acyl acceptors. 7. HSA functions as a positive control, where it is used to show that the purified rMTG has retained its bioactivity. In the native form, HSA does not act as a substrate for MTG and must first be denatured by reduction using DTT. 8. Since both HES 70-amine and DMC are polyvalent compounds, their crosslinking with rMTG can lead to the formation of very large aggregates, which can be seen as bands at the top of the stacking gel, as shown in Fig. 1. References 1. Duncan, R. (2003) The dawning era of polymer therapeutics. Nature Rev. Drug Discovery 2, 347–360. 2. Besheer, A., Vogel, J., Glanz, D., Kressler, J., Groth, T., Mäder, K. (2009) Characterization of PLGA nanospheres stabilized with amphiphilic polymers: Hydrophobically modified hydroxyethyl starch vs pluronics. Mol. Pharm. 6, 407–415. 3. Besheer, A., Hertel, T.C., Kressler, J., Mäder, K., Pietzsch, M. (2009) Enzymatically catalyzed HES conjugation using microbial transglutaminase: Proof of feasibility. J. Pharm. Sci. 98, 4420–4428. 4. Sato, H. (2002) Enzymatic procedure for sitespecific pegylation of proteins. Adv. Drug. Delivery Rev. 54, 487–504. 5. Yokoyama, K., Nio, N., Kikuchi, Y. (2004) Properties and applications of microbial transglutaminase. Applied Microbiol. Biotech. 64, 447–454. 6. Sato, H., Hayashi, E., Yamada, N., Yatagai, M., Takahara, Y. (2001) Further studies on the

site-specific protein modification by microbial transglutaminase. Bioconj. Chem. 12, 701–710. 7. Fontana, A., Spolaore, B., Mero, A., Veronese, F.M. (2008) Site-specific modification and PEGylation of pharmaceutical proteins mediated by transglutaminase. Adv. Drug Delivery Rev. 60, 13–28. 8. Marx, C. K., Hertel, T. C., Pietzsch, M. (2008) Purification and activation of a recombinant histidine-tagged protransglutaminase after soluble expression in E. coli and partial characterization of the active enzyme. Enz. Microbial. Tech. 42, 568–575. 9. Marx, C. K., Hertel, T. C., Pietzsch, M. (2007) Soluble expression of a pro-transglutaminase from Streptomyces mobaraensis in Escherichia coli. Enz. Microbial. Tech. 40, 1543–1550. 10. Cass, A. E. G., Zhang, J. K. (2001) A Study of His-tagged alkaline phosphatase immobilization on a nanoporous nickel-titanium dioxide film. Anal. Biochem. 292, 307–310.

Chapter 3 Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids Sukanta Dolai, Wei Shi, Bikash Mondal, and Krishnaswami Raja Abstract We present here a general methodology for significantly increasing the number of dye/drug molecules that can be attached per protein molecule. As a demonstration of this approach, poly(acrylic acid) (PAA)based near-infrared fluorescence (NIRF) dye- and glucose-incorporated novel copolymers were synthesized, which were further employed for bioconjugation to avidin and bovine serum albumin (BSA). In this method, azide-terminated poly(tert-butyl acrylate) was synthesized via atom transfer radical polymerization (ATRP). Subsequent deprotection was performed to yield poly(acrylic acid) (PAA) possessing a reactive chain-end. A one-pot sequential amidation of the PAA with the amine derivatives of a nearinfrared fluorescent dye (ADS832WS) and glucose produced NIRF dye-incorporated water-soluble copolymers. End-group modifications were performed to produce alkyne/biotin-terminated copo­lymers, which were further employed to generate dye-incorporated polymer–protein hybrids via the biotin– avidin interaction with avidin or by “click” bioconjugation with azide-modified BSA. Key words: Protein–polymer hybrids, Copolymers, Near infrared fluorescence dye, “Click” bioconjugation

1. Introduction Polymer–protein hybrids are a newly emerging class of bioconjugates with several applications in biotechnology, biopharmaceutical chemistry, and other life science areas (1–14). Drug- and dyelabeled conjugates are a multibillion dollar industry; however, by using the current technologies, the number of copies of cytotoxic drugs/dyes that can be chemically conjugated to a single protein is very limited. For example, in the case of Mylotarg® (a calicheamicin–antibody conjugate), the drug to antibody ratio is 2:1. A fundamental limitation in the detection and therapeutic efficiency of imaging agent/drug-labeled proteins such as antibodies arises Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_3, © Springer Science+Business Media, LLC 2011

29

30

Dolai et al.

from the fact that only a limited number of these molecules can be attached per protein molecule, and extensive modification of proteins with several copies of a drug/dye would cause deactivation of the protein’s functional properties. On the other hand, the synthesis of copolymer–protein hybrid materials, for labeling applications, has so far been restricted to only a few types of acrylate/methacrylate monomers due to differences in solubility between monomers; for instance, the preparation of a copolymer, containing a monosaccharide-derived acrylate (hydrophilic) and taxol acrylate (hydrophobic), would be very challenging. The widely different reactivity of the monomers and the lack of reactivity of many biologically relevant acrylates are other factors that are responsible for the lack of diversity of copolymer–protein hybrids (e.g., the acrylate of the anticancer and antiAlzheimer’s drug candidate curcumin can be synthesized but cannot be polymerized via free radical polymerization methods because the molecule is a radical scavenger (15)). We have recently developed a general methodology for significantly increasing the number of dye/drug molecules that can be attached per protein molecule (14). Herein, we present detailed practical procedures for (1) the synthesis of well-defined living copolymers containing reactive chain-end and functional sidechain pendant groups in which the chain end and side chains possess orthogonal reactivity, (2) the attachment of a number of dye and glucosamine molecules to the functional polymer side-chains, and (3) the attachment of the polymers (via the reactive polymer chain end) with proteins to produce the final bioconjugates.

2. Materials 2.1. Synthesis of Biotin-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Biotin-terminated poly(acrylic acid) (PAA) polymer. This polymer is synthesized according to the procedure previously reported in ref. 14. 2. Near infrared absorption dye (ADS832WS) (American Dye Source, Inc., Baie D’Urfé, Quebec). 3. 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC⋅HCl). 4. N-Hydroxybenzotriazole (HOBt). 5. N,N-Dimethylformamide (DMF). 6. Triethylamine. 7. d(+)-Glucosamine. 8. Milli-Q water (Millipore). 9. Dialysis tubing (3.5 kDa MWCO). 10. Sephadex™ LH-20 gel filtration chromatography medium (GE Healthcare).

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

2.2. Synthesis of Avidin–Poly (NIRF)–Poly(Glu) Polymer Conjugate

1. Biotin-terminated Subheading 3.1).

poly(NIRF)–poly(Glu)

polymer

31

(see

2. Avidin. 3. 0.1 M Phosphate-buffered saline (PBS), pH 7.4. 4. Dialysis tubing (50 kDa MWCO). 5. Milli-Q water (Millipore).

2.3. Size-Exclusion Chromatography Analysis of Avidin– Polymer Conjugate

1. Fast protein liquid chromatography (FPLC) system (AKTA™ Explorer system from GE Healthcare) equipped with a multiple wavelength detector. 2. HiPrep™ Sephacryl™ S-200 HR 26/10 gel filtration column (GE Healthcare). 3. 0.1 M PBS, pH 7.4.

2.4. SDS-PAGE Analysis of Conjugate

1. PAGE gel 4–20% SDS Cassette Gel precast gels, 12-well (Fisher Scientific). 2. PAGEgel DTT reducer, 10× (Fisher Scientific). 3. PAGEgel LDS Sample buffer, 4× (Fisher Scientific). 4. PAGEgel SDS Standard Run buffer, 20× (Fisher Scientific). 5. PAGEgel Two-Color SDS Marker (Orange/Blue) (Fisher Scientific). Ready-to-use in a 1× LDS sample buffer. 6. GelCode Blue coomassie gel stain reagent (Fisher Scientific). 7. FisherBiotech FB1000 electrophoresis power supply. 8. VWR™ rocking platform. 9. Odyssey® Infrared Imaging System (LI-COR Biosciences, Lincoln, NE).

2.5. Synthesis of Azide-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Azide-terminated poly(acrylic acid) (PAA) polymer. This polymer is synthesized according to the procedure previously reported in ref. 14 (see Note 1).

2.6. Synthesis of Alkyne-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Azide-terminated Subheading 3.5).

2. All other required materials are the same as those listed in Subheading 2.1, items 2–10. poly(NIRF)–poly(Glu)

polymer

2. 3-Prop-2-ynoxyprop-1-yne (dipropargyl ether). 3. Copper sulfate pentahydrate (CuSO4⋅5H2O). 4. Sodium ascorbate. 5. Tert-Butanol (t-BuOH). 6. Tetrahydrofuran (THF). 7. Milli-Q water (Millipore).

(see

32

Dolai et al.

8. Dialysis tubing (10 kDa MWCO). 9. Sephadex™ LH-20 gel filtration chromatography medium (GE Healthcare). 2.7. AzideModification of BSA (Protein Model)

1. Bovine serum albumin (BSA). 2. 2,5-Dioxopyrrolidin-1-yl-5-azidopentanoate (NHS-azide). This reagent can be synthesized according to the procedure previously reported in ref. 14. 3. Dimethyl sulfoxide (DMSO). 4. 0.1 M Tris buffer, pH 8.0. 5. Dialysis tubing (10 kDa MWCO).

2.8. “Click” Bioconjugation of BSA–Azide with Alkyne-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Alkyne-terminated Subheading 3.6).

poly(NIRF)–poly(Glu)

polymer

(see

2. Azide-modified BSA (see Subheading 3.7). 3. Tris(2-carboxyethyl)phosphine (TCEP). 4. Tris((1-phenyl-1H-1,2,3-triazol-4-yl)methyl)amine (TPTA). TPTA is a polytriazolylamine ligand which stabilizes Cu(I) toward disproportionation and oxidation, thus enhancing its catalytic effect in alkyne-azide cycloaddition (“click”) reactions. 5. Copper sulfate pentahydrate (CuSO4 ⋅ 5H2O). 6. 0.1 M Tris buffer, pH 8.0. 7. Dimethylformamide (DMF). 8. 0.1 M PBS, pH 7.4. 9. Dialysis tubing (50 kDa MWCO).

2.9. Size-Exclusion Chromatography Analysis of BSA– Poly(NIRF)–Poly(Glu) Polymer Conjugate

1. All the required materials are the same as those listed in Subheading 2.3.

2.10. SDS-PAGE Analysis of BSA– Polymer Conjugate

1. All the required materials are the same as those listed in Subheading 2.4, items 1–9.

3. Methods Using the procedures described below, a one-pot sequential amidation of the PAA with the amine derivative of a nearinfrared fluorescent dye (NIRF) (ADS832WS) and glucose produced NIRF dye-incorporated water-soluble copolymers.

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

33

End-group modifications were also performed to produce alkyne/ biotin-terminated copolymers, which were further employed to generate dye-incorporated polymer–protein hybrids via the biotin–avidin interaction with avidin or by “click” bioconjugation (16) with azide-modified BSA. With these general methods, two fundamental limitations in the synthesis of bioconjugates are overcome: (1) the basic restriction in the diversity of copolymers that can be synthesized for producing bioconjugates and (2) the limitation in the number of dyes/drug molecules that can be attached per protein molecule. 3.1. Synthesis of Biotin-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Dissolve the biotin-terminated poly(acrylic acid) (PAA) polymer (76  mg, 1.055  mmol), near-infrared absorption dye (ADS832WS) (498 mg, 0.531 mmol), EDC ∙ HCl (213 mg, 1.11  mmol), and HOBt (153  mg, 1.148  mmol) in DMF (1.5 mL) in a round-bottom flask. Place a magnetic stir bar inside the flask, and then cap it with a rubber septum. 2. Add triethylamine (0.03 mL, 0.17 mmol) via a syringe. 3. Purge the flask with N2 gas and stir the reaction mixture (using a magnetic stirrer) for 48 h at room temperature under a slow continuous flow of N2 gas. 4. After stirring for 48 h, prepare a mixture of d(+)-glucosamine (230 mg, 1.067 mmol) and EDC ∙ HCl (200 mg, 1.043 mmol) in a mixture of Milli-Q water (2 mL) and DMF (1 mL), and then add this solution to the previous reaction mixture via a syringe (see Note 2). 5. Stir the mixture for another 72 h at room temperature under a continuous flow of N2 gas. 6. After stopping the reaction, transfer the mixture into a dialysis bag (3.5  kDa MWCO) and dialyze extensively against deionized water for 24 h at room temperature (see Note 3). 7. Filter the dialyzed reaction mixture via gravity filtration and lyophilize to yield a green-brown cotton-like crude product. 8. Prepare a size-exclusion chromatography column by suspending the Sephadex™ LH-20 gel filtration medium in Milli-Q water and packing it inside a clean chromatographic column. 9. Dissolve the crude product obtained in step 7 in a minimum volume of Milli-Q water and load it onto the top of the Sephadex™ column. Elute the sample from the column with water and collect the eluted fractions. Combine the fractions containing the biotin-terminated poly(NIRF)–poly(Glu) polymer and lyophilize. 10. Characterize the purified product using nuclear magnetic resonance (NMR) spectroscopy, Fourier transform infrared (FT-IR) spectroscopy, and size-exclusion chromatography (SEC) techniques.

34

Dolai et al.

3.2. Synthesis of Avidin–Poly (NIRF)–Poly(Glu) Polymer Conjugate

1. Dissolve avidin (1  mg) in 0.25  mL of PBS buffer, pH 7.4. Also separately dissolve biotin-terminated poly(NIRF)– poly(Glu) polymer (obtained in Subheading 3.1) (5.5 mg) in 0.65  mL PBS buffer (pH 7.4), and then slowly add this solution to the avidin solution. 2. Incubate the resulting mixture for 2 h at room temperature. 3. After incubation, transfer the mixture into a dialysis bag (50 kDa MWCO) and dialyze extensively against deionized water for 24 h at 4°C (see Note 3). 4. Analyze the protein–polymer hybrid by size-exclusion FPLC (Subheading 3.3) and SDS-PAGE (Subheading 3.4).

3.3. Size-Exclusion Chromatography Analysis of the Hybrid

Perform the size-exclusion FPLC analysis procedure using the following samples: avidin protein (control), biotin-terminated poly(NIRF)–poly(Glu) polymer (control, obtained from Subheading 3.1), and the avidin–poly(NIRF)–poly(Glu) polymer conjugate (obtained from Subheading 3.2). 1. Attach the HiPrep™ Sephacryl™ S200 HR 26/10 gel filtration column to a FPLC system equipped with a sample injection loop of 1 mL, and set the detector to monitor at 260, 280, and 700 nm wavelengths. 2. Prepare a sufficient volume of running buffer (~4 L of PBS, pH 7.4) for the entire run and place it in the FPLC buffer reservoir. 3. Wash the column with PBS, pH 7.4 for at least two column volumes (i.e., about 640 mL), using a flow rate of 1 mL/min. 4. Set up the following chromatography method file: Flow rate = 1  mL/min, total run = 1.5 column volumes; perform column equilibration for 0.2 column volumes before sample injection, sample injection volume = 1 mL. 5. Start the chromatography run, and load 1 mL of the sample (~5  mg/mL) into the injector module via a syringe; make sure there are no air bubbles. 6. Compare the results of the FPLC analysis for the different samples. Representative FPLC chromatogram profiles for avidin, biotin-terminated poly(NIRF)–poly(Glu) polymer, and the avidin– poly(NIRF)–poly(Glu) polymer conjugate are shown in Fig. 1.

3.4. SDS-PAGE Analysis of Avidin– Polymer Conjugate

Perform the SDS-PAGE analysis procedure using the following samples: Avidin protein (control), biotin-terminated poly(NIRF)– poly(Glu) polymer (control, obtained from Subheading 3.1), and the avidin–poly(NIRF)–poly(Glu) polymer conjugate (obtained from Subheading 3.2). The instructions described below assume

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

35

Fig.  1. Size-exclusion FPLC analysis of avidin, biotin-terminated poly(NIRF)–poly(Glu) copolymer, and the avidin–polymer conjugate. The avidin–polymer conjugate eluted earlier (due to its higher molecular weight) in comparison to the avidin control sample and the polymer control sample. Reproduced with permission from ref. 14 © 2007 American Chemical Society.

the use of Fisher Scientific PAGEgel SDS electrophoresis reagents, but are easily adaptable for use with other standard SDS-PAGE gel systems. 1. Load a precast SDS cassette gel into the electrophoresis unit, and insert a dummy plate on the other side of the apparatus. Ensure that the cassette gel and the dummy plate are mounted tightly. 2. Prepare 1× running buffer by taking 60  mL of the SDS Standard Run buffer 20× stock solution and mixing it with 1,140 mL of Milli-Q water. Add the 1× SDS Standard Run buffer to the upper (inner) chamber of the electrophoresis apparatus, and confirm that there is no leakage. Next, fill the chamber with enough running buffer to cover the cassette gel completely. 3. Wash each well in the cassette gel with running buffer using a sample loading tip and a pipettor. 4. Prepare the avidin (control) sample: Add 30  mL of avidin (~3 mg/mL), 4 mL of DTT reducer, and 10 mL of LDS sample buffer to an Eppendorf tube and mix by vortexing. Place the sample in a float and boil it in a water bath for 4  min. Remove the sample and place it aside on ice.

36

Dolai et al.

5. Prepare the polymer (control) sample: Add 30 mL of biotinterminated poly(NIRF)–poly(Glu) polymer (~3  mg/mL), 4 mL of DTT reducer, and 10 mL of LDS sample buffer to an Eppendorf tube and mix by vortexing. Place the sample aside on ice. 6. Prepare the avidin–polymer conjugate sample: Add 30 mL of avidin–poly(NIRF)–poly(Glu) polymer conjugate (~3  mg/ mL), 4 mL of DTT reducer, and 10 mL of LDS sample buffer to an Eppendorf tube and mix by vortexing. Place the sample in a float and boil it in a water bath for 4 min. Remove the sample and place it aside on ice. 7. Prepare a blank sample: Add 30 mL of PBS, pH 7.4 and 10 mL of LDS sample buffer to an Eppendorf tube and mix by vortexing. Place the sample aside on ice. 8. Carefully load the protein/polymer samples (i.e., avidin control, polymer control, and polymer conjugate), protein molecular weight markers, and the blank solution into the wells in the cassette gel by placing the tip of the loading pipette at the bottom of each well and slowly drawing the tip upward as you fill up the wells. Load the blank solution between each of the sample wells. 9. Fill the lower (outer) chamber of the electrophoresis apparatus with running buffer up to the halfway mark on the cassette gel. 10. Place the top safety cover onto the electrophoresis apparatus and connect it to a power source. Turn the power source on, and set the voltage to 90 V. Start the electrophoresis run; the tracking dye in the samples should begin to travel downward. Stop the power when you see the tracking dye reaches the bottom of the gel cassette. 11. Disconnect the power supply and remove the top safety cover of the electrophoresis apparatus. Slowly remove the cassette gel. Remove the outer plastic cover of the cassette gel using a spatula, and carefully place the gel inside a rectangular container. Rinse the gel carefully with Milli-Q water. 12. After rinsing, place the gel onto the scanner bed of an infrared imaging system. Put a small amount of water on top of the gel and on the scanner plate so that the gel does not dry out. Scan the gel at medium resolution to acquire an infrared fluorescence image (EX l = 800  nm for NIRF dye ADS832WS). 13. Place the gel back into the container after scanning and add a sufficient volume of GelCode Blue coomassie stain reagent to completely cover the gel. Place the gel onto a rocking platform and agitate it for 30 min at room temperature.

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

37

Fig.  2. SDS-PAGE analysis of avidin (lane a) and avidin–poly(NIRF)–poly(Glu) polymer conjugate (lane b). The formation of the conjugate is indicated by the higher molecular weight bands present in the SDS-PAGEgel; the conjugate band that glowed when imaged with an infrared scanner (left ) is also visible following Coomassie staining (right ), thus confirming the presence of both polymer and protein at the same position. Reproduced with permission from ref. 14 © 2007 American Chemical Society.

14. Discard the liquid stain and wash the gel three times with Milli-Q water. 15. Compare the image of the coomassie-stained gel with the infrared fluorescence image of the scanned gel to confirm the formation of the avidin–polymer conjugate (Fig. 2). 3.5. Synthesis of Azide-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Dissolve the azide-terminated poly(acrylic acid) (PAA) polymer (72  mg, 1.003  mmol), near-infrared absorption dye (ADS832WS) (100 mg, 0.531 mmol), EDC∙HCl (213 mg, 1.11  mmol), and HOBt (153  mg, 1.148  mmol) in DMF (1.5 mL) in a round-bottom flask. Place a magnetic stir bar inside the flask, and then cap it with a rubber septum. 2. Add Triethylamine (0.03 mL, 0.17 mmol) via a syringe. 3. Purge the flask with N2 gas and stir the reaction mixture (using a magnetic stirrer) for 48 h at room temperature under a slow continuous flow of N2 gas.

38

Dolai et al.

4. After stirring for 48 h, prepare a mixture of d(+)-glucosamine (190 mg, 0.9 mmol) and EDC ∙ HCl (200 mg, 1.043 mmol) in a mixture of Milli-Q water (2 mL) and DMF (1 mL), and then add this solution to the previous reaction mixture via a syringe (see Note 2). 5. Stir the mixture for another 72 h at room temperature under a continuous flow of N2 gas. 6. After stopping the reaction, transfer the mixture into a dialysis bag (10  kDa MWCO) and dialyze extensively against deionized water for 24 h at room temperature (see Note 3). 7. Filter the dialyzed reaction mixture via gravity filtration and lyophilize to yield a green-brown cotton-like crude product. 8. Proceed to purify the product by gel filtration chromatography as described in Subheading 3.1, steps 8 and 9. 9. Characterize the purified product using NMR, FT-IR, and SEC techniques. 3.6. Synthesis of Alkyne-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Dissolve azide-terminated poly (NIRF)–(Glu) polymer (obtained in Subheading 3.1) (60 mg, 3 mmol) in 3 mL of a mixture of t-BuOH, THF, and Milli-Q water (1:1:1, v/v/v) in a round-bottom flask. Place a magnetic stir bar inside the flask, and then cap it with a rubber septum. 2. Add 10  mL of propargyl ether (9.14  mg, 31.5  eq) via a syringe. 3. Add sodium ascorbate (4  mg, 0.02  mmol) to the reaction vessel while stirring (using a magnetic stirrer). Finally, add CuSO4 ⋅ 5H2O (5 mg, 0.02 mmol) to the reaction mixture. 4. Purge the reaction vessel with N2 gas and continue to stir the mixture for 24 h at room temperature. 5. After stopping the reaction, transfer the mixture into a dialysis bag (10  kDa MWCO) and dialyze extensively against deionized water for 24 h at room temperature (see Note 3). 6. Proceed to purify the product by gel filtration chromatography as described in Subheading 3.1, steps 8 and 9. 7. Characterize the purified product using NMR, FT-IR, and SEC techniques.

3.7. Modification of the Amine Surface Groups of BSA with NHS-Azide

1. Dissolve 5 mg of BSA in 1 mL of PBS buffer, pH 7.4 in a small vial and gently agitate on a vortex mixer. Do not excessively shake the mixture in order to avoid the formation of foam. 2. Next, measure 5 mg of NHS-azide (10 eq to each modifiable lysine group on BSA) in 100 mL of DMSO. 3. Add the NHS-azide dropwise to the BSA solution while agitating on a vortex. Add the NHS-azide slowly enough so that nothing precipitates out. Mix the components thoroughly and put the

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

39

vial containing the mixture on a rocking platform and let it shake gently for 3 h at room temperature. 4. Transfer the mixture to a 10-kDa MWCO dialysis membrane and dialyze against 0.1  M Tris buffer, pH 8.0 for 48  h to remove excess azide linker (see Note 3). Transfer the dialyzed solution to a container and store it at 4°C. 3.8. Conjugation of the AlkyneTerminated Poly(NIRF)–Poly(Glu) Polymer with AzideModified BSA

1. Add 20  mg of alkyne-terminated poly(NIRF)–poly(Glu) (obtained from Subheading 3.6) to a vial containing 1.8 mL of azide-modified BSA (obtained from Subheading 3.7). The starting volume of this reaction mixture should be about 1.8 mL. 2. Calculate the amount of TCEP, TPTA, and copper sulfate required in order to make the concentration of these reagents in the final reaction mixture (about 2  mL) to be 4, 4, and 2 mM, respectively. 3. Add the required amount of TCEP to the vial containing the alkyne-terminated poly(NIRF)–poly(Glu) polymer and azidemodified BSA. Mix the resulting solution over a vortex. 4. Dissolve the required amount of TPTA in 200  mL of DMF. The amount of DMF should not exceed more than 10% (v/v) of the total reaction volume (if the DMF amount exceeds 10%, there is a chance that protein precipitation will occur.) Slowly add the TPTA solution to the reaction mixture dropwise while agitating on a vortex. Ensure that no protein precipitates out. If the solution appears cloudy, stop adding TPTA and wait until the solution clears before continuing to add the reagent (see Note 4). 5. Add the required amount of copper sulfate to the reaction mixture and agitate on a vortex. Note that it is necessary to add the TPTA reagent to the reaction mixture before the addition of copper sulfate. 6. Incubate the reaction mixture on an agitator at 4°C for 16 h. Next, transfer the reaction mixture to a 50-kDa MWCO dialysis membrane and dialyze against PBS buffer, pH 7.4 for 48 h to remove excess reagents and byproducts (see Note 3). 7. Characterize the protein–polymer hybrid by size-exclusion FPLC (Subheading 3.9) and SDS-PAGE (Subheading 3.10).

3.9. Size-Exclusion Chromatography Analysis of the BSA–Polymer Conjugate

Perform the size-exclusion FPLC analysis procedure, as described in Subheading 3.3, using the following samples: azide-modified BSA protein (control), alkyne-terminated poly(NIRF)–poly(Glu) polymer (control, obtained from Subheading 3.6), and the BSA– poly(NIRF)–poly(Glu) polymer conjugate (obtained from Subheading 3.8). Compare the results of the FPLC analysis for the different samples. Representative FPLC chromatogram profiles for azide-modified BSA, alkyne-terminated poly(NIRF)–poly(Glu) polymer, and the BSA–poly(NIRF)–poly(Glu) polymer conjugate are shown in Fig. 3.

40

Dolai et al.

Fig. 3. Size-exclusion FPLC analysis of azide-modified BSA (dashed line), alkyne-terminated poly(NIRF)–poly(Glu) polymer (dotted line), and the BSA–poly(NIRF)–poly(Glu) polymer “click” conjugate (solid line). The conjugate eluted earlier than both the azide-modified BSA control sample and the polymer control sample. (The unmodified BSA eluted at the same volume as the azide-modified BSA; hence it was omitted from the chromatogram.) Reproduced with permission from ref. 14 © 2007 American Chemical Society.

Fig. 4. SDS-PAGE analysis of the BSA–poly(NIRF)–poly(Glu) polymer conjugate (lanes 4 and 8 ), a mixture of azide-modified BSA and polymer without “click” catalyst reagents (lane 6 ) and unmodified BSA (lane 10 ). The conjugate bands in lanes 4 and 8 that glowed when imaged with an infrared scanner (right ) are also visible following Coomassie staining (left ), thus confirming the presence of both polymer and protein at the same positions. Conversely, the control samples in lanes 6 and 10 do not glow at all when imaged with an infrared scanner. To confirm that the copolymer is indeed chemically bonded to the protein, we incubated the mixture of alkyne-terminated polymer and azide-modified BSA in the same ratio but without any “click” catalyst reagents, and then dialyzed the mixture with a 50-kDa MWCO membrane; in this case, no higher molecular weight conjugate bands are observed (lane 6 ). Reproduced with permission from ref. 14 © 2007 American Chemical Society.

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

3.10. SDS-PAGE Analysis of the Conjugate

41

Perform the SDS-PAGE analysis procedure as described in Subheading 3.4 using the following samples: unmodified/azidemodified BSA protein (controls), alkyne-terminated poly(NIRF)– poly(Glu) polymer (control, obtained from Subheading 3.6), and the BSA–poly(NIRF)–poly(Glu) polymer conjugate (obtained from Subheading  3.8). Compare the results of the SDS-PAGE analysis for the different samples. Representative coomassiestained gel images and infrared fluorescence gel images for unmodified/azide-modified BSA, alkyne-terminated poly(NIRF)– poly(Glu) polymer, and the BSA–poly(NIRF)–poly(Glu) polymer conjugate are shown in Fig. 4.

4. Notes 1. Azide-terminated poly (tert-butyl acrylate) is synthesized via atom transfer radical polymerization (ATRP). Subsequent deprotection is performed to yield poly(acrylic acid) (PAA) possessing a reactive chain-end. When synthesizing the poly(tert-butyl acrylate) via ATRP, make sure that the reaction vessel is oxygen-free, as singlet oxygen species act as radical scavengers and will stop the polymerization process. Also, do not perform the polymerization reaction at temperatures higher than 70°C with azide-terminated initiator, as higher temperatures could destroy the azide functionality. After the purification of the polymer, confirm the presence of the azide group in the FT-IR spectrum around 2,100 cm−1. 2. When performing the amidation reaction between the biotin-terminated poly(acrylic acid) (PAA) polymer with NIRF-NH2 and glucosamine using EDC⋅HCl and HOBt, remember to add the NIRF-NH2 first and let it react for 48 h; then add the glucosamine to the same reaction vessel. It is also possible to add both reagents at the same time; however, due to their reactivity differences, the glucosamine will react much faster than the NIRF-NH2. Hence, the final amount of NIRF-dye loaded onto the polymer will be much lower than expected. 3. When performing dialysis, replace the dialysis solution every 2–3  h. Typically, five to six buffer exchanges using a total volume of 500 mL of buffer are needed to effectively dialyze the samples. 4. While performing the “click” bioconjugation reaction using the copper sulfate, sodium ascorbate, and TPTA reagent, remember to add sodium ascorbate or TPTA prior to the addition of the copper sulfate.

42

Dolai et al.

Acknowledgments We sincerely thank Dr. K. B. Sharpless for providing us with the TPTA “click” ligand. The National Science Foundation (CHE0723028) and PSC-CUNY provided financial support for this work. References 1. Hermenson, G. T. (2008) Bioconjugate Techniques, 2nd ed., Academic Press, San Diego, CA. 2. Wu, A. M., and Senter, P. D. (2005) Arming antibodies: prospects and challenges for immunoconjugates. Nat. Biotechnol. 23, 1137–1146. 3. Raja, K., McDonald, R., Tuck, S., Rodriguez, R., Milley, B., and Traquina, P. (2007) Onepot synthesis, purification, and formulation of bionanoparticle-CpG oligodeoxynucleotide hepatitis B surface antigen conjugate vaccine via tangential flow filtration. Bioconjugate Chem. 18, 285–288. 4. Raja, K. S., Wang, Q., Gonzalez, M. J., Manchester, M., Johnson, J. E., and Finn, M. G. (2003) Hybrid virus-polymer materials. 1. Synthesis and properties of PEG-decorated cowpea mosaic virus. Biomacromolecules 4, 472–476. 5. Kulkarni, S., Schilli, C., Grin, B., Mueller, A. H. E., Hoffman, A. S., and Stayton, P. S. (2006) Controlling the aggregation of conjugates of streptavidin with smart block copolymers prepared via the RAFT copolymerization technique. Biomacromolecules 7, 2736–2741. 6. Sengupta, S., Raja, K. S., Kaltgrad, E., Strable, E., and Finn, M. G. (2005) Virus-glycopolymer conjugates by copper(I) catalysis of atom transfer radical polymerization and azide-alkyne cycloaddition. Chem. Commun. 34, 4315–4317. 7. Sengupta, S., Kuzelka, J., Singh, P., Lewis, W. G., Manchester, M., and Finn, M. G. (2005) Accelerated bioorthogonal conjugation: a practical method for the ligation of diverse functional molecules to a polyvalent virus scaffold. Bioconjugate Chem. 16, 1572–1579. 8. Hou, S., Sun, X. L., Dong, C. M., and Chaikof, E. L. (2004) Facile synthesis of chain-end functionalized glycopolymers for site-specific bioconjugation. Bioconjugate Chem. 15, 954–959.

9. Vazquez-Dorbatt, V., and Maynard, H. D. (2006) Biotinylated glycopolymers synthesized by atom transfer radical polymerization. Biomacromolecules 7, 2297–2302. 10. Lele, B. S., Murata, H., Matyjaszewski, K., and Russell, A. J. (2005) Synthesis of uniform protein-polymer conjugates. Biomacromolecules 6, 3380–3387. 11. De, P., Li, M., Gondi, S. R., and Sumerlin, B. S. (2008) Temperature-regulated activity of res­ ponsive polymer-protein conjugates prepared by grafting-from via RAFT polymerization. J. Am. Chem. Soc. 130, 11288–11289. 12. Bontempo, D., Heredia, K. L., Fish, B. A., and Maynard, H. D. (2004) Cysteine-reactive polymers synthesized by atom transfer radical polymerization for conjugation to proteins. J. Am. Chem. Soc. 126, 15372–15373. 13. Bontempo, D., and Maynard, H. D. (2005) Streptavidin as a macroinitiator for polymerization: in situ protein-polymer conjugate formation. J. Am. Chem. Soc. 127, 6508–6509. 14. Shi, W., Dolai, S., Averick, S., Fernando, S. S., Saltos, A. J., L’Amoreaux, W., Banerjee, P. and, and Raja, K. S. (2009) A general methodology toward drug/dye incorporated living copolymer-protein hybrids: (NIRF dye-glucose) copolymer-avidin/BSA conjugates as prototypes. Bioconjugate Chem. 20, 1595–1601. 15. Shi, W., Dolai, S., Rizk, S., Hussain, A., Tariq, H., Averick, S., L’Amoreaux, W., El Idrissi, A., Banerjee, P. and, and Raja, K. S. (2007) Synthesis of monofunctional curcumin derivatives, clicked curcumin dimer, and a PAMAM dendrimer curcumin conjugate for therapeutic applications. Org. Lett. 9, 5461–5464. 16. Kolb, H. C., Finn, M. G., Sharpless, K. B. (2001) Click Chemistry: Diverse Chemical Function from a Few Good Reactions. Angew. Chem. Int. Ed. 40, 2004–2021.

Chapter 4 Dye/DNA Conjugates as Multiple Labels for Antibodies in Sensitive Fluorescence Immunoassays Qin Zhang, Shengchao Zhu, and Liang-Hong Guo Abstract Fluorescence immunoassays are widely used in life science research, medical diagnostics, and environmental monitoring due to the intrinsically high specificity, simplicity, and versatility of immunoassays as well as the availability of a large variety of fluorescent labeling molecules. However, the sensitivity of immunoassays needs to be improved further to meet the ever-increasing demands of the new proteomics era. We have developed a novel and simple method to increase immunoassay sensitivity by attaching multiple fluorescent labels on an antibody with a dye/DNA conjugate. Our strategy is to use a DNA fragment as a molecular carrier to attach multiple fluorescent dyes to an antibody at a single site. The dye/DNA conjugate is not presynthesized, but rather formed in situ as part of the immunoassay. Our results demonstrate that by using a 219-bp DNA fragment in conjunction with SYBR Green I fluorescent DNAbinder, the sensitivity of both direct and competitive fluorescence immunoassays is improved by orders of magnitude, reaching a lower detection limit of 1.9 pg/mL for 17b-estradiol. Key words: Fluorescence immunoassay, Antibody labeling, Multiple labels, Dye/DNA conjugate

1. Introduction Immunoassays are widely used in life science research, medical diagnostics, and environmental monitoring due to their high specificity, simplicity, and versatility. The acquirement of detection signal usually entails labeling of antibodies or antigens with specific types of signal molecules or groups, such as radioisotopes, fluorescent dyes, metal complexes, and enzymes (1–3). For different types of labels, there are different detection methods such as colorimetric (3), chemiluminescent, electrochemiluminescent, and amperometric (4–6) detection. Each label and associated detection method has its own pros and cons. For example, although chemiluminescence is a highly sensitive detection Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_4, © Springer Science+Business Media, LLC 2011

43

44

Zhang, Zhu, and Guo

method, it is not suitable for microarray detection, since the light signal must be produced in the liquid phase and the lightemitting species is not confined to the surface (7). Furthermore, while immuno-PCR provides an amazing improvement in immunoassay sensitivity up to 10,000-fold (8), a series of complex and separate procedures need to be carried out for DNA amplification and quantification. Fluorescence thus remains a desirable method to satisfy the demands of protein microarray detection in terms of label stability, assay robustness, detection sensitivity, and multiplexity. In traditional fluorescence immunoassays, the antibody is labeled with a very limited number of fluorescent dyes such as FITC, Cy-3, and Cy-5 so as to minimize nonspecific binding and loss of antibody reactivity. Consequently, these immunoassays suffer from inferior sensitivity. In recent years, organic dendrimers (9), quantum dots (10), fluorophore-doped silica (11), and latex beads (12) have been employed as multiple labels on an antibody or antigen to improve immunoassay sensitivity. Although sensitivity is enhanced tremendously, these materials bear some problems such as high cost, nonuniformity, poor dispersion in aqueous solutions, and difficulty in protein attachment. We have developed a simple and easy method of attaching multiple fluorescent labels at a single site on antibodies (13). Our strategy is to first link a DNA sequence to an antibody, and then use the DNA as a molecular carrier to attach a large number of DNA-binding fluorescent dyes to the antibody (Fig. 1.). We have demonstrated the utility of the dye/DNA conjugate label in both direct and competitive fluorescence immunoassays. In a direct immunoassay for the detection of goat anti-mouse antibody, mouse IgG was adsorbed to the surface of a 96-well plate, and was recognized by a biotinylated goat anti-mouse antibody in

Fig. 1. (a) Schematic illustration of the antibody multiple labeling strategy using a biotinterminated DNA to carry a large number of fluorescent DNA binders (SYBR Green I), and streptavidin to link the DNA to a biotinylated antibody. (b) Schematic illustration of the conventional approach of using FITC-streptavidin to label an antibody. Reproduced with permission from ref. 14 © 2008 from Elsevier B.V.

Dye/DNA Conjugates as Multiple Labels for Antibodies

45

solution. A 219-bp double-stranded oligonucleotide terminated with a biotin was linked to the antibody through the biotin/ streptavidin/biotin interaction. Multiple labeling of the antibody was achieved after SYBR Green I (a fluorescent DNA binder) was added into the solution and bound to the oligonucleotide at high ratios. By comparison with fluorescein-labeled streptavidin, the assay with the dye/DNA label achieved a 150-fold lower detection limit (14). In a competitive immunoassay for the detection of 17b-estradiol, the antigen was again firstly adsorbed to a 96-well plate. Biotinylated anti17b-estradiol antibody and free 17b-estradiol were added into the well. After the immunoreaction, the antibody was multiply labeled with the dye/DNA conjugate by the procedure described above. In this case, the detection limit was lowered by approximately 200-fold over that achievable with fluoresceinlabeled streptavidin, and reached 1.9 pg/mL 17b-estradiol (14). The novel multiple-labeling method described herein uses readily available chemical and biochemical reagents, and is simple to implement. It thus holds great potential for the sensitive detection of protein microarrays.

2. Materials 2.1. Preparation of Biotinylated and FITC-Labeled Streptavidin and Antibody

1. BT-NHS solution: 20  mg Biotinyl-N-hydroxysuccinimide (BT-NHS) (Sigma, St. Louis, MO) is dissolved in 10 mL dry DMF (N,N-dimethylformamide). The solution is freshly prepared for each labeling experiment (see Note 1). 2. FITC solution: 50  mg Fluorescein isothiocyanate (FITC) (Amresco, Solon, OH) is dissolved in 10 mL dry DMSO (analytical purity) in a dark tube. The solution is freshly prepared for each labeling experiment (see Note 2). 3. Carbonate buffer: 50  mM Na2CO3, 50  mM NaHCO3, pH 9.15 (adjusted with HCl). 4. Protein storage buffer: 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.5 mM KH2PO4 (adjust to pH 7.4 with HCl if necessary), containing 0.1% (w/v) BSA and 0.1% (w/v) NaN3. Store at 4°C. 5. D-Salt™ polyacrylamide desalting column (Pierce, Rockford, IL). Store at 4°C. 6. Desalting column elution buffer: 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.5 mM KH2PO4, pH 7.4 (see Note 3). 7. Desalting column storage buffer: 0.05 M NaCl, 0.02% (w/v) NaN3. 8. Microcon YM-3 centrifugal filter device (molecular weight cutoff: 3,000 kDa) (Millipore, Bedford, MA). 9. HABA/avidin reagent (Sigma, St. Louis, MO). Store at 4°C.

46

Zhang, Zhu, and Guo

2.2. Preparation of Biotinylated Double-Stranded DNA Oligonucleotide

1. Design of short-strand DNA oligonucleotides: A 30-bp BT-­DNA (biotinylated double-stranded DNA oligonucleotide) fragment is designed according to the principles of primer design (see Note 4). Strand A: biotin-5¢-TTT TTT TTT GCG GGT AAC GTC AAT ATT AAC TTT ACT CCC-3¢; strand B: 5¢GGG AGT AAA GTT AAT ATT GAC GTT ACC CGC-3¢. 2. PCR primer design for long-strand DNA oligonucleotides: a pair of PCR primers is designed according to the principles of primer design in order to obtain a 219-bp BT-DNA (see Note 5). 3. Hybridization buffer (2× SSC): 0.3 M NaCl, 30 mM sodium citrate, pH 7.0. 4. 1× PCR mixture: 5 mL 10× PCR buffer (Takara Bio, Japan), 1.25  U Taq polymerase, 0.2  mM dNTP mixture, 0.4  mM each of primers, 10  ng template DNA, 1.5  mM MgCl2. Prepare a 1× PCR mixture freshly for each experiment. 5. DNA gel electrophoresis buffer (0.5× TBE): 45  mM Tris– borate, 1 mM EDTA, pH 8.0. 6. DNA fragment purification kit (Takara Bio). 7. DNA storage buffer (10× TE): 10  mM Tris–HCl, 1  mM EDTA, pH 8.0.

2.3. Optimization of Fluorescent Immunoassay Using the Dye/DNA Conjugate Labels

1. Coating buffer: 50 mM Na2CO3, 50 mM NaHCO3, pH 9.6 (adjusted with HCl). 2. Tris–NaCl buffer: 50 mM Tris, 50 mM NaCl, adjust to pH 8.0 with HCl. 3. Blocking buffer: 1% (w/v) BSA in Tris–NaCl buffer. Store at 4°C (see Note 6). 4. Reaction buffer: 0.1% (w/v) BSA in Tris–NaCl buffer. Store at 4°C. 5. Washing buffer: 0.1% (v/v) Tween 20 is added into Tris– NaCl buffer and stored at 4°C. A fresh solution is prepared for each experiment. 6. Detection buffer: 50  mM Tris–HCl and 50  mM NaCl, adjusted to pH 8.0 with HCl. 7. SYBR Green I solution: SYBR Green I (SG1) is dissolved in DMSO and its concentration determined using a UV–visible spectrometer. The dye solution is stable for up to half a year if stored at −20°C in the dark. SG1 dissolved in DMSO is diluted freshly with the detection buffer to the desired concentration (see Note 7). 8. 17b-Estradiol 6-(O-carboxymethyl) oxime:BSA (E2-BSA) (Sigma, St. Louis, MO). Dissolve in protein storage buffer at the desired concentration and store at −20°C. A working solution of E2-BSA is diluted freshly with the coating buffer. 9. Luminescence spectrometer.

Dye/DNA Conjugates as Multiple Labels for Antibodies

47

3. Methods 3.1. Preparation of Biotinylated Goat Anti-mouse Antibody

1. React goat anti-mouse antibody with BT-NHS at a molar ratio of 5.6 BT-NHS:1 antibody in carbonate buffer for 4 h at room temperature (RT) with magnetic stirring. 2. After the reaction, unbound BT-NHS and the antibody are separated by gel filtration on a D-Salt™ polyacrylamide desalting column (see Note 8). 3. Antibody concentration is determined by measuring the absorbance at 280 nm on a UV–visible spectrophotometer. 4. Optional: If necessary, the collected antibody solution is concentrated in a microconcentrator (Microcon YM-3) by centrifuging at 12,000 × g for 30 min at 4°C. The final antibody concentration is determined as above. 5. The biotin labeling ratio on the antibody is determined by using the HABA/avidin reagent (see Note 9). 6. The prepared biotinylated goat anti-mouse antibody (BT-Ab) sample should be aliquoted and stored at −20°C.

3.2. Preparation of FITC-Labeled Streptavidin and Antibody

1. Mix the antibody or streptavidin with FITC at a molar ratio of 5.6 FITC:1 protein in carbonate buffer for 4 h at RT with magnetic stirring (see Note 10). 2. After the reaction, the remaining free FITC is separated from the protein by gel filtration on a D-Salt™ polyacrylamide desalting column. 3. Protein and FITC concentrations are determined by measuring the absorbance at 280 and 494  nm, respectively, on a UV–visible spectrophotometer (see Note 11). 4. Optional: If necessary, the collected protein solution is concentrated in a microconcentrator (Microcon YM-3) by centrifuging at 12,000 × g for 30  min at 4°C. The final protein and FITC concentrations are determined as above. 5. FITC-labeled protein samples should be aliquoted and stored in the dark at −20°C.

3.3. Preparation of Biotinylated Short Double-Stranded DNA Oligonucleotides

1. The two 30-bp complementary oligonucleotides are dissolved in 2× SSC buffer, heated at 95°C for 5 min on a thermocycler, and then allowed to naturally cool down to room temperature. 2. The concentration of the hybridized oligonucleotides is determined by measuring the absorbance at 260 nm. 3. The prepared BT-DNA should be aliquoted and stored at −20°C.

48

Zhang, Zhu, and Guo

3.4. Preparation of Biotinylated Long Double-Stranded DNA Oligonucleotides

1. A tube containing a freshly prepared PCR mixture is loaded into a thermocycler. 2. PCR: Two 219-bp complementary oligonucleotides are heated at 94°C for 3 min, followed by 30 PCR cycles. Each PCR cycle consists of a denaturation at 94°C for 30 s, annealing at 56°C for 30 s, and extension at 72°C for 1 min. After 30 PCR cycles, the mixture is reacted at 72°C for 3 min for the final chain extension. The solution is then ramped down to 4°C and held at that temperature until sample analysis (see Note 12). 3. To verify the desired product, the PCR mixture is separated by electrophoresis on a 1.2% (w/v) agarose gel in 0.5× TBE buffer. The gel is stained with 0.5 mg/mL ethidium bromide and visualized under UV illumination. An example of the results produced is shown in Fig. 2. 4. The PCR product is purified with a DNA fragment purification kit.

Fig. 2. Gel electrophoresis image of a 219-bp long double-stranded DNA oligonucleotide PCR product. Lanes 1 and 2 : PCR product; lane 3 : DNA molecular weight standards (DL2000 DNA markers). The PCR product was loaded onto a 1.2% agarose gel in TBE buffer, and the gel was stained with 0.5  mg/mL ethidium bromide. Reproduced with permission from ref. 14 © 2008 from Elsevier B.V.

Dye/DNA Conjugates as Multiple Labels for Antibodies

49

3.5. Optimization of Fluorescent Immunoassays Using the Dye/DNA Conjugate Labels

1. A white 96-well plate is coated with 100 mL of FITC-labeled mouse IgG dissolved at various concentrations in coating buffer by incubating overnight at 4°C.

3.5.1. Optimization of Mouse IgG Concentration

3. The plate is blocked with 300  mL of blocking buffer overnight at 4°C, and then washed as above.

2. The plate is then washed three times with 300 mL of washing buffer.

4. Fluorescence intensity is measured on a luminescence spectrometer with 495 nm excitation, 525 nm emission, 5 nm slit width, and 515 nm cutoff filter. Based on the fluorescence intensity vs. IgG concentration curve, the optimal mouse IgG concentration can be obtained. In this example protocol, the optimal mouse IgG concentration was determined to be 30 mg/mL.

3.5.2. Optimization of BT-Ab Concentration

1. A white 96-well plate is coated with 100  mL of 30  mg/mL mouse IgG solution. 2. 100  mL of FITC-Ab of different solution concentrations is added into the well and reacted at 37°C for 2 h with shaking. 3. After washing, fluorescence is measured as described above. In this example protocol, the optimal antibody concentration of 20  mg/mL is obtained from the fluorescence intensity vs. FITC-Ab concentration curve.

3.5.3. Optimization of Streptavidin Concentration

1. After plate coating with the optimal concentration of mouse IgG (Subheading 3.5.1) and immunoreaction with the optimal concentration of BT-Ab (Subheading 3.5.2), 100 mL of FITC-SA of different concentrations is added into the well and reacted at 37°C for 2 h. 2. Fluorescence intensity is then measured, and the optimal concentration of streptavidin is obtained as above. In this example protocol, the optimal concentration of streptavidin of 5 mg/mL is obtained.

3.5.4. Optimization of BT-DNA and Dye Concentration

1. After plate coating with the optimal concentration of mouse IgG, immunoreaction with the optimal concentration of BT-Ab, and affinity reaction with the optimal concentration of streptavidin, 100 mL BT-DNA of different concentrations is added to the well and reacted at 37°C for 1 h with shaking. 2. After washing, 100  mL SYBR Green I (SG1) of different concentrations is added into the plate and incubated for 10 min at RT with shaking. 3. The fluorescence intensity is measured and the optimal concentration of BT-DNA and SG1 is determined.

50

Zhang, Zhu, and Guo

In this example protocol, the optimal concentration of BT-DNA was determined to be 88 nM, and the optimal concentration of SG1 was determined to be 574 nM. 3.6. Direct Immunoassay of Goat Anti-mouse Antibody Using the Dye/DNA Conjugate Labels

1. A white 96-well plate is coated with 100  mL of 30  mg/mL mouse IgG by incubating overnight at 4°C, and washed three times with washing buffer. 2. The plate is then blocked with 300 mL of 1% BSA in PBS, pH 7.4 by incubating overnight at 4°C, and washed as above. 3. 100  mL BT-Ab of various concentrations is added into the IgG coated well and incubated at 37°C for 2 h with shaking. The plate is again washed. 4. 100 mL of 5 mg/mL streptavidin is added into the well and reacted at 37°C for 2  h with shaking, followed by plate washing. 5. BT-DNA (88 nM) is added and incubated at 37°C for 1  h with shaking, followed by plate washing. 6. SYBR Green I is added into the plate and reacted for 10 min at RT with shaking, followed by plate washing. 7. Fluorescence intensity is measured with 495  nm excitation, 520 nm emission, a 5-nm slit width, and a 515-nm cutoff filter. 8. The lower limit of detection (LOD) is calculated from the concentration curve. An example concentration curve is shown in Fig. 3.

Fig. 3. Fluorescence immunoassay for the detection of goat anti-mouse IgG on mouse IgG coated 96-well plates using the 219-bp DNA/SYBR Green I conjugate label (filled diamond ), or the 30 bp oligonucleotide/SYBR Green I conjugate label (filled triangle). Each data point is the average of three replicate measurements. Reproduced with permission from ref. 14 © 2008 from Elsevier B.V.

Dye/DNA Conjugates as Multiple Labels for Antibodies

3.7. Competitive Immunoassay of 17b-Estradiol Using the Dye/DNA Conjugate Labels

51

1. A white 96-well microplate is coated with 100 mL of 10 mg/ mL E2-BSA by incubating overnight at 4°C, and washed (see Note 13). 2. The plate is then blocked with 300 mL of the blocking buffer by incubating overnight at 4°C, and washed. 3. 50 mL of 20 mg/mL biotinylated E2 antibody and 50 mL E2 of various concentrations are added into the coated well, incubated at 37°C for 2 h with shaking, and then the plate is washed (see Note 14). 4. 100 mL of 5 mg/mL streptavidin is added into the well and incubated at 37°C for 2  h with shaking. The plate is then washed. 5. 88 nM biotin-labeled 219 bp DNA is added into the well and reacted at 37°C for 1  h with shaking. The plate is then washed. 6. 574 nM SG1 is added into the well and reacted for 7 min at RT with shaking, followed by plate washing. 7. Fluorescence intensity is measured under the same conditions as described above in Subheading 3.6 and the LOD is calculated from the competition curve (see Note 15). An example of the competition curve is shown in Fig. 4.

4. Notes 1. Because the NHS group is very easily hydrolyzed, BT-NHS should be stored frozen in an air-tight container with desiccants. For the same reason, silica gel desiccants are used to keep the DMF solvent dry. 2. To avoid photolysis, FITC should be stored in the dark. 3. The composition of the desalting column elution buffer depends on the storage buffer for the labeled protein. Typically, adding 0.05  M NaCl is helpful in the elution process. 4. The design of short-sequence DNA oligonucleotides conforms to several principles. (1) There is no special restriction to the type of bases; however, intrachain hairpins must be avoided. (2) The length is one of the decisive factors in assay sensitivity. The appropriate range of lengths is between 20 and 60 bp, which can bind enough DNA probes and also reduce spatial hindrance and the cost of nucleic acid synthesis. (3) In this example protocol, Oligo d(T)9 is a homo-oligomeric deoxyribonucleotide used as a linker to reduce spatial hindrance between biotin and streptavidin binding.

52

Zhang, Zhu, and Guo

Fig. 4. (a) Competitive fluorescence immunoassay for the detection of E2 on E2-BSAcoated 96-well plates using the long chain DNA/dye conjugate-SA labeled anti-E2 antibody (filled diamond  ) and FITC-SA labeled anti-E2 antibody (filled square). (b) An expanded view of (a) at low E2 concentrations. Each data point is the average of three replicate measurements. Reproduced with permission from ref. 14 © 2008 from Elsevier B.V.

5. The principles of primer design for long DNA oligonucleotides are the same as that for short DNA oligonucleotides. In this example protocol, a 219-bp DNA sequence is synthesized by PCR and used in fluorescence immunoassays. Whether even longer DNA sequences could improve fluorescence sensitivity is unknown. If the sequence is too long, a random coil may form and disturb the binding of the DNA with fluorescent dyes.

Dye/DNA Conjugates as Multiple Labels for Antibodies

53

6. Other blocking reagents such as 1% (w/v) gelatin can also be used. 7. SYBR Green I was chosen in our work due to its high DNAbinding affinity and excellent fluorescence characteristics. If necessary, other DNA-binding fluorescent dyes such as SYTOX Orange may also be used. However, the optimal concentration and reaction conditions will need to be optimized accordingly. 8. Because both BT-NHS and antibody are colorless, the volume of each collected fraction should be kept as low as possible for good resolution. The collected fractions that contain protein are identified by measuring the absorbance of each fraction at l = 280 nm. 9. According to the manufacturer’s instructions (available from Sigma), the biotin content can be calculated based on the binding of the dye HABA to avidin and the ability of biotin to displace the dye in stoichiometric proportions. 10. The general procedure is the same as that for BT-NHS labeling (Subheading 3.1). Note, however, that all the reactions are conducted in the dark. 11. Because FITC also absorbs at 280 nm, the measured absorbance at 280 nm needs to have the FITC contribution subtracted out before the protein concentration can be calculated correctly. 12. The stated PCR conditions were optimized from a series of pilot studies. If the primer and target sequence are changed, the PCR conditions may need to be adjusted. 13. The optimal concentration of E2-BSA to be used is determined by performing experiments similar to those conducted to determine the optimal mouse IgG coating concentration (Subheading 3.5.1). 14. The concentration of biotinylated E2 antibody employed is the EC50 value (the antibody concentration producing half of the peak value of fluorescence intensity) in the fluorescence vs. antibody concentration curve. 15. All the results shown are from at least three replicates. The lower LOD is calculated based on three times the standard deviation above the blank value.

Acknowledgments This work was supported by the Knowledge Innovation Program of the Chinese Academy of Sciences (KSCX2-YW-G-059) and the National Basic Research Program of China (2006CB403303).

54

Zhang, Zhu, and Guo

References 1. Berson, S. A., Yalow, R. S. (1959) Assay of plasma insulin in human subjects by immunological methods. Nature 184, 1648–1649. 2. Li, T. M., Parrish, R. F. (2002) Fluorescence and Immunodiagnostic Methods. In Topics in Fluorescence Spectroscopy. Volume 3: Biochemical Applications. Lakowicz, J. R. (Ed.). Springer US, New York. pp 273–287. 3. Blackburn, G. F., Shah, H. P., Kenten, J. H., Leland, J., Kamin, R. A., Link, J., Peterman, J., Powell, M. J., Shah, A., Talley, D. B. (1991) Electrochemiluminescence detection for development of immunoassays and DNA probe assays for clinical diagnostics. Clin. Chem. 37, 1534–1539. 4. Marquette, C. A., Blum, L. J. (2006) State of the art and recent advances in immunoanalytical systems. Biosens. Bioelectron. 21, 1424–1433. 5. Kricka, L. J. (1999) Chemiluminescence and Bioluminescence. Anal. Chem. 71, 305-308. 6. Bange, A., Halsall, H. B., Heineman, W. R. (2005) Microfluidic immunosensor systems. Biosens. Bioelectron. 20, 2488–2503. 7. Schweitzer, B., Predki, P., Snyder, M. (2003) Microarrays to characterize protein interactions on a whole-proteome scale. Proteomics 3, 2190–2199. 8. Ong, K. K., Jenkins, A. L., Cheng, R., Tomalia, D. A., Durst, H. D., Jensen, J. L., Emanuel, P. A., Swim, C. R., Yin, R. (2001)

Dendrimer enhanced immunosensors for biological detection. Anal. Chim. Acta 444, 143–148. 9. Zhou, M., Roovers, J., Robertson, G. P., Grover, C. P. (2003) Multilabeling Biomo­ lecules at a Single Site. 1. Synthesis and Characterization of a Dendritic Label for Electrochemiluminescence Assays. Anal. Chem. 75, 6708–6717. 10. Chan, W. C. W., Nie, S. (1998) Quantum dot bioconjugates for ultrasensitive nonisotopic detection. Science 281, 2016–2018. 11. Wang, L., Wang, K., Santra, S., Zhao, X., Hilliard, L. R., Smith, J. E., Wu, Y., Tan, W. (2006) Watching silica nanoparticles glow in the biological world. Anal. Chem. 78, 646–654. 12. Bangs, L. B. (1996) New developments in particle-based immunoassays: Introduction. Pure Appl. Chem. 68, 1873–1879. 13. Zhang, Q., Guo, L. H. (2007) Multiple labeling of antibodies with Dye/DNA conjugate for sensitivity improvement in fluore­­ scence immunoassay. Bioconjugate Chem. 18, 1668–1672. 14. Zhu, S. C., Zhang, Q., Guo, L. H. (2008) Part-per-trillion level detection of estradiol by competitive fluorescence immunoassay using DNA/dye conjugate as antibody multiple labels. Anal. Chim. Acta 624, 141–146.

Chapter 5 Chemoselective Modification of Viral Proteins Bearing Metabolically Introduced “Clickable” Amino Acids and Sugars Partha S. Banerjee and Isaac S. Carrico Abstract The inherent difficulty of performing chemical modifications of proteins in a truly site-specific fashion is often compounded by the need to work within complex biological settings. In order to alleviate this complication, targets can be “prelabeled” metabolically with unnatural residues, which allow access to highly selective bioorthogonal reactions. Due to their small size, permissibility within biosynthetic pathways and access to reactions with high specificity, azides provide excellent bioorthogonal handles. This two-step labeling process is emerging as a highly effective means to modify therapeutic proteins. In this chapter, we take this strategy a step further and apply chemoselective ligation to remodel the surfaces of adenoviruses. Despite the large number of ongoing clinical trials involving these complex mammalian viruses, new methods for their facile, flexible surface modification are necessary to drive the development of next-generation therapeutics. Here we demonstrate the modification of azides on adenoviral surfaces via a straightforward chemoselective protocol based on copper-assisted “click” chemistry. This method provides access to a wide array of effector functionalities without sacrificing infectivity. Key words: Chemoselective modification, Bioorthogonal, CuAAC, “Click” chemistry, Azide, Unnatural amino acid, Azido sugar, Alkyne probe, Adenovirus

1. Introduction The specific chemical modification of proteins can enable fundamental physiological and biochemical studies as well as allow the production of next-generation therapeutics. While a number of strategies allow selective chemical modification, one of the most conceptually straightforward approaches is the introduction of selectively reactive functional groups such as azides, aldehydes, alkynes, etc., within a target protein (1–3). This can be accomplished using a number of methodologies, including synthetic/semisynthetic Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_5, © Springer Science+Business Media, LLC 2011

55

56

Banerjee and Carrico

protein synthesis, enzymatic modification, and the metabolic introduction of unnatural amino acid resides. The latter method has the distinct advantage of enabling the production of modifiable proteins within the native cellular environment (4). In addition, the metabolic introduction of unnatural amino acids and monosaccharides can demonstrate high efficiency and does not require the use of engineered cell lines (5). The resulting flexibility thus allows the exploration of new target proteins of interest quickly regardless of cell type. However, the incorporation of unnatural residues into cellular, nontarget proteins has the potential to perturb cellular physiology. Mammalian viruses have tremendous therapeutic potential as gene delivery, oncolytic and vaccine agents (6); but unfortunately, the inability to efficiently tune their targeting properties and modulate the host immune response has limited their success in these areas. Genetic methods have been widely used to modify viral properties, but in many instances such types of manipulations have led to a propensity for loss in infectivity and limited access to effector functionality (7, 8). In order to sidestep these issues, we have developed an approach to specifically introduce unnatural amino acids and monosaccharides onto the surface of mature, infectious, adenovirus particles. These modified viral vectors demonstrate no loss in either production or infectivity, and can be selectively modified with a wide range of functionalities.

2. Materials All chemicals and reagents were obtained from commercial sources and used without further purification unless otherwise noted. 2.1. Cell Culture, Virus Infection, and Harvesting

1. 100-mm cell culture dishes, sterile. 2. DMEM complete medium: Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) bovine calf serum. 3. DMEM (−Met/−Cys): DMEM minus l-methionine, minus l-cysteine, supplemented with 10% (v/v) bovine calf serum. 4. AHA stock solution (25×): l-Azidohomoalanine (AHA) (Invitrogen, Carlsbad, CA) (see Note 1) is dissolved in DMEM (−Met) medium to make a 100-mM solution. Ideally, the solution should be made fresh before use, but it can also be stored at 4°C overnight. 5. Methionine stock solution (25×): l-methionine is dissolved in DMEM (−Met) medium to make a 100-mM solution. Make fresh before use.

Chemoselective Modification of Viral Proteins

57

6. Cysteine stock solution (100×): l-cysteine dissolved in DMEM (−Met) medium to make a 200-mM solution. Make fresh before use. 7. AHA labeling medium: DMEM (−Met/−Cys) supplemented with AHA stock solution and cysteine stock solution to a final concentration of 4 mM AHA and 2 mM cysteine. 8. Methionine (control) labeling medium: DMEM (−Met/−Cys) supplemented with methionine stock solution and cysteine stock solution to a final concentration of 4 mM methionine and 2 mM cysteine. 9. Ac4GalNAz stock solution (1,000×): Peracetylated N-azidoacetylgalactosamine (Ac4GalNAz) (Invitrogen) (see Note 1) is dissolved in methanol at a concentration of 50 mM and stored at 4°C. 10. Adenovirus type 5 (Ad5) containing a GFP or luciferase transgene (Vector BioLabs, Eagleville, PA). Maintain the viruses at −20°C in 5 mM Tris–HCl buffer (pH 8.0) containing 0.5  mM MgCl2, 50  mM NaCl, 25% (v/v) glycerol, and 0.05% (w/v) BSA. 11. Human embryonic kidney (HEK) 293 cells, maintained in DMEM supplemented with 10% (v/v) bovine calf serum, 2  mM glutamine, 100  U/mL penicillin, and 100  mg/mL streptomycin (see Notes 2 and 3). 12. TD buffer: 137 mM NaCl, 5 mM KCl, 0.7 mM Na2HPO4, and 25 mM Tris–HCl, pH 7.5. Store at room temperature (RT). 13. TC solution (200×): 180  mM CaCl2 and 210  mM MgCl2. Store at RT. 14. Cesium chloride (CsCl) solutions of 1.25 g/mL, 1.35 g/mL, and 1.40 g/mL in TD buffer. Sterile filter and store at RT. 15. Beckman Ultra Clear 3.5-in. centrifuge tubes. 16. Beckman ultracentrifuge equipped with a SW 41 and SW 60 rotor. 17. Virus storage buffer: PBS (pH 7.2) containing 0.5 mM CaCl2, 0.9 mM MgCl2, and 10% (v/v) glycerol. Store at −20°C. 2.2. Azide–Alkyne Cycloaddition or “Click” Chemistry and Virus Purification

1. Copper(I) bromide (CuBr) is dissolved in deoxygenated dimethylsulfoxide (DMSO) to make a 50-mM stock solution. Make fresh every time before use. 2. 2 M Tris–HCl buffer, pH 8.0. Store at RT. 3. Bathophenanthroline disulfonic acid disodium salt (MP Biomedicals, Solon, OH) is dissolved in water to make a 50-mM stock solution. Store at RT. 4. 10  mM Alkyne-TAMRA probe [also known as Click-iT® Tetramethylrhodamine (Invitrogen)] in water. Store at −20°C.

58

Banerjee and Carrico

5. Virus storage buffer: PBS (pH 7.2) containing 0.5 mM CaCl2, 0.9 mM MgCl2, and 10% (v/v) glycerol. Store at −20°C. 6. Centri-Sep™ spin columns (Princeton separations, Adelphia, NJ). 7. Ethylenediaminetetraacetic acid (EDTA). 2.3. SDSPolyacrylamide Gel Electrophoresis

1. Laemmli sample buffer (2×): 125  mM Tris–HCl (pH 6.8), 4% (w/v) SDS, 20% (v/v) glycerol, 10% (v/v) 2-mercaptoethanol, and 0.004% (w/v) bromophenol blue. 2. Resolving buffer (4×): 1.5 M Tris–HCl, pH 8.7. Store at RT. 3. Stacking buffer (8×): 1 M Tris–HCl, pH 6.8. Store at RT. 4. 40% (w/v) acrylamide/bisacrylamide solution (29:1, w/w) (3.3% Cross-linker). 5. N,N,N,N ¢-tetramethylethylenediamine (TEMED). 6. 10% (w/v) SDS solution in water. Store at RT. 7. 10% (w/v) Ammonium persulfate solution in water. Prepare fresh. 8. Running buffer (1×): 187  mM glycine, 19  mM Tris–HCl, 3.5 mM SDS in water. Store at 4°C. 9. Prestained protein molecular weight markers. 10. Fluorescent gel scanner equipped with the required excitation and emission filters. For imaging TAMRA-labeled adenovirus particles, we used the following filter setup: Ex 532 nm; Em 580 nm, BP 30.

3. Methods In order to diminish the level of toxicity due to the incorporation of unnatural amino acids within cellular proteins, metabolic labeling of adenovirus with AHA (Subheading 3.1) is carried out over a 6-h time period during the viral growth phase. This was done expressly to maximize overlap with adenoviral structural protein expression, specifically 18–24  h postinfection (9). Although we have not explored this, it is expected that longer labeling times may lead to reduced levels of virus production and infectivity. Although metabolic labeling with peracetylated sugars has demonstrated cellular toxicity (10, 11), the concentrations needed for the efficient labeling of adenovirus particles (Subheading 3.2) are well below toxic levels. Furthermore, the metabolic conversion of Ac4GalNAz to the donor sugar, UDP-GlcNAz, is relatively slow. As a result, we chose to incubate the infected host cells with Ac4GalNAz throughout the viral production period. It is worth noting that no losses in either the viral production yield or infectivity

Chemoselective Modification of Viral Proteins

59

of the resulting particles was observed using either of the metabolic labeling protocols described in this chapter. The kinetics of chemical modification via copper-catalyzed azide–alkyne cycloaddition (CuAAC) are not well understood and can vary greatly with changes to the composition of the catalyst and the concentrations of either reactant. Notably, under the conditions described herein (Subheading  3.4), this “click” reaction proceeds with high yield (~90%) within a 12-h period. However, incomplete removal of the cytotoxic copper catalyst after the ligation reaction can in some instances pose a potential problem, especially for functional assays within animal model systems. Furthermore, it should be noted that the presence of copper may also impede the mass spectral analysis of modified proteins. Two alternative chemistries are available that allow copper-free and highly selective azide modifications. The first, Staudinger ligation, is well characterized and reagents can be obtained commercially; however, this reaction is slower under the described conditions. Alternatively, ring strain-promoted cycloaddition reactions are significantly faster than the Staudinger ligation. Although the cyclooctyne reagents required for strain promoted reactions have been difficult to synthesize, they have recently become commercially available (Jena Bioscience) making this reaction significantly more approachable. Notably, it has been reported that “click” and Staudinger reactions are hampered by the presence of urea and ionic detergents such as SDS, respectively (12). 3.1. Production of Azidoho­moalanineLabeled Adenovirus

1. For virus production, near-confluent 293 cells grown on 100mm culture dishes are used (see Note 2). A batch of ten culture dishes can generally be used for the production of a single type of virus. The cells in each dish of a batch of ten dishes should be treated similarly, and their combined volumes after cell lysis used for purification of the virus. Before infection, determine the cell count by removing the growth medium from the culture dish and loosening the cells with 1  mL of trypsin–EDTA per dish (see Note 3). Count the number of cells using a hemocytometer and calculate the amount of virus required for infection, depending on the average cell count per dish. For infection, use virus at a multiplicity of infection (MOI) of 10 PFU/cell. For normal adenovirus type 5 (Ad5) obtained from commercial or academic sources, the infectivity index – defined as the ratio of infectious virions (determined by the plaque assay method and expressed as PFU per mL) to the total number of physical virus particles per mL – is typically 1:20 but can vary depending on the serotype or source. 2. Prepare a sufficient volume of infection buffer containing 2% (v/v) serum, 1× TC solution, infective Ad5 and TD buffer to give 1 mL per dish (e.g., for 20 dishes, make up to 22 mL of

60

Banerjee and Carrico

infection buffer using TD buffer to bring the solution up to the final total required volume). 3. Near-confluent 293 cells seeded on 100-mm culture dishes are first infected with adenovirus using the prepared infection buffer. Infection is carried out by slow removal of the DMEM complete growth medium, followed by addition of 1 mL of infection buffer per 100-mm dish and incubating at 37° for 1 h with periodic shaking (see Note 4). 4. Cover the cells with 10 mL of fresh DMEM complete medium and incubate at 37°C for 18 h. 5. After 18-h postinfection (see Note 5), carefully remove the growth medium covering the infected cells and wash each culture dish with 5 mL of TD buffer (see Note 6). If necessary for cell-line stability, preincubate the cells with the wash buffer at 37°C for 10 min. 6. Remove the TD buffer wash solution and fill the culture dish with AHA- or methionine (control)-supplemented labeling medium and incubate for 6 h (see Note 5). For 100-mm culture dishes, a minimum of 5 mL of labeling medium per dish should be used (see Note 7). 7. After 24-h postinfection (i.e., 6 h after addition of the labeling medium), remove the supplemental medium carefully (see Note 8) and cover the cells with fresh DMEM complete growth medium (10 mL per dish). Incubate the cells for an additional 18–20 h at 37°C. 3.2. Production of GlcNAz-Labeled Adenovirus

1. Follow steps 1–3 from Subheading 3.1.

3.3. Purification of Metabolically Labeled Adenoviral Particles

1. Remove the cells (which should have become loose and mostly dislodged from the culture dish by this time) from the culture dish, and centrifuge at 2,000 × g and remove excess medium (see Note 9). Resuspend the cell pellet in 8 mL of TD buffer for every ten dishes of infected cells.

2. Cover the cells with 10 mL of fresh DMEM complete growth medium and then add 10 mL of Ac4GalNAz stock solution. Incubate the cells at 37°C for 44 h.

2. Lyse the cells by repeated freeze–thaw cycles (at least three times) in liquid N2. Centrifuge at 2,000 × g to remove cell debris. 3. Prepare a CsCl density gradient for virus purification with 2.0  mL of 1.4  g/mL CsCl at the bottom and 2.5  mL of 1.25 g/mL CsCl on top in an ultracentrifuge tube (see Note 10). Top these with the soluble cell extract obtained from step 2 above and then centrifuge the samples in a swinging bucket rotor at 126,444 × g (average) for 1 h at 15°C in an ultracentrifuge (see Note 11).

Chemoselective Modification of Viral Proteins

61

4. After completion of ultracentrifugation, the virus particles should be visible as a thick white band at the junction of the two CsCl solutions. Make a small perforation at the bottom of the tube with a sterile needle to allow the solution to come out drop by drop, and collect the thick white virus band (see Notes 12–14). 5. Purified virus extracts should be obtained by rebanding of the collected virus sample using 1.35 g/mL of CsCl in an ultracentrifuge spinning at 125,812 × g (average) for 18 h at 15°C (see Notes 15 and 16). The virus can be collected as described above (see Note 17). 6. For storage of the virus particles over long periods of time, dialysis of the collected virus extract should be performed against virus storage buffer. Store the dialyzed virus samples at −20°C with 10% (v/v) glycerol. 3.4. Chemical Labeling of Azide-Modified Virus with Alkyne Probes Using CopperAssisted “Click” Chemistry

While this protocol describes the use of a Cu(I) bathophenanthroline disulfonic acid catalyst under deoxygenating conditions (Fig. 1), the use of other ligands under atmospheric conditions is well documented and provides robust labeling, albeit at slightly lower levels. 1. To a solution of 2 × 1012 viral particles, add Tris–HCl buffer, pH 8.0 at a final concentration of 100 mM (see Note 18). 2. Add bathophenanthroline disulfonic acid disodium salt at a final concentration of 3 mM to the above solution.

Fig. 1. Schematic illustration of virions metabolically modified with unnatural substrates and their subsequent chemical labeling with alkyne probes using two different chemistries, the copper-assisted “click” reaction (top) and the Staudinger ligation reaction (bottom).

62

Banerjee and Carrico

3. To the above mixture, add the desired alkyne probe (e.g., alkyne-TAMRA, a rhodium-based fluorophore dye) at a final concentration of 200–500 mM. 4. Weigh out 7.2 mg of CuBr in an Eppendorf tube and cover it with aluminum foil. 5. Place the reaction mixture obtained from step 3, the Eppendorf tube containing CuBr (step 4), and 1  mL of DMSO inside a deoxygenated glove bag for 6 h. 6. After the 6-h deoxygenation period, dissolve the CuBr in the DMSO (1 mL) to make a 50-mM stock solution. 7. Add a sufficient volume of the CuBr stock solution (step 6) to the reaction mixture to give a final concentration of 1  mM CuBr (see Notes 19 and 20), and allow the labeling reaction to proceed overnight inside the glove bag. 8. Remove the reaction solution from the glove bag and add EDTA solution to a final concentration of 10 mM (see Note 20) to stop the labeling reaction. 9. Purify the labeled virus samples using Centri-Sep™ gel filtration spin columns (see Notes 21 and 22) equilibrated with PBS buffer, pH 7.2 containing 0.5  mM CaCl2, 0.9  mM MgCl2, and 10% (v/v) glycerol. 10. Store the labeled virus at −20°C. 3.5. SDS-PAGE Analysis of TAMRALabeled Adenovirus

The procedures described below are designed for use with the BioRad Mini-PROTEAN® 3 gel electrophoresis system; however, modification to suit any gel apparatus is straightforward. 1. Prepare a 0.75-mm thick 10% (w/v) resolving gel by adding 4.8 mL of H2O, 2.47 mL of 40% (w/v) acrylamide, 2.5 mL of 4× resolving buffer, 100 mL of 10% (w/v) SDS, 100 mL of 10% (w/v) ammonium persulfate, and 5  mL of TEMED. Pour the gel mixture between the glass plates of the MiniPROTEAN® 3 gel cassette while leaving space for the stacking gel, and then cover it with a layer of ethanol. Allow the resolving gel to polymerize for 20–30 min. 2. Prepare the stacking gel by adding 3.6 mL of H2O, 0.623 mL of 40% (w/v) acrylamide, 0.63  mL of 8× stacking buffer, 50 mL of 10% (w/v) SDS, 50 mL of 10% (w/v) ammonium persulfate and 5 mL of TEMED. Pour this gel mixture over the resolving gel and insert the comb. Allow the stacking gel to polymerize for 30 min. 3. Add Laemmli sample buffer (1×) to 20 mL of purified labeled virus and boil for 10 min (see Note 23). Centrifuge the samples for 30 s to remove any insoluble precipitates. 4. Once the stacking gel has polymerized, remove the comb. Attach the gel cassette to the electrophoresis cell and add

Chemoselective Modification of Viral Proteins

63

Fig. 2. SDS-PAGE analysis of TAMRA-labeled Ad5 virions obtained by “click” conjugation of alkyne-TAMRA probe to azide-modified adenovirus particles. (a) Fluorescent gel image of virus particles labeled with alkyne-TAMRA produced at two different concentrations of AHA (4 and 32 mM). Standard concentrations of TAMRA were used to determine the number of dye molecules coupled to each virus particle. Based on comparisons with the protein molecular weight markers, the adenovirus Hexon, Penton, and Fiber capsid proteins appear to be labeled with AHA. (b) Fluorescent gel image of Ac4GalNAz-modified virus particles treated with alkyne-TAMRA under CuAAC conditions. The gel image indicates that only the adenovirus Fiber capsid protein is labeled with GalNAz.

running buffer (see Note 24). Load the required number of wells with the viral protein samples prepared in step 3 and use one well to load the prestained molecular weight markers. Run the gel for 1 h at 200 V. Keep the electrophoresis cell covered with aluminum foil during the entire period (see Note 25). 5. After the tracking dye has run out of the gel (see Note 26), stop the electrophoresis and quickly transfer the gel to a fluorescent gel scanner and scan the gel for fluorescence emission at the appropriate wavelength (based on the fluorophore dye label used) (Fig. 2).

4. Notes 1. Azidohomoalanine (an unnatural azide-functionalized methionine analog) and peracetylated N-azidoacetylgalactosamine (an unnatural azido sugar) can either be purchased from commercial sources or easily synthesized as previously described (13, 14). 2. Before infection, confirm that the cells are 80–90% confluent. Cultures containing overconfluent cells can be difficult to infect and may not produce viruses as desired.

64

Banerjee and Carrico

3. Before infection, count the number to cells using a hemocyto­ meter to calculate the correct amount of Ad5 required for infection. 4. During the 1-h infection period, gently shake the culture dish side to side every 15 min while incubating at 37°C. 5. AHA labeling can be carried out anytime between 12- and 24-h postinfection. Labeling for more than 6  h, however, may result in reduced virion production. 6. Before addition of the labeling medium, wash the cells with TD buffer. Care must be taken at this step, as infected cells are susceptible to becoming washed away. 7. At least 5 mL of labeling medium per 100-mL culture dish is necessary for sufficient metabolic incorporation. The use of lesser amounts of labeling medium tends to leave uncovered cells in the culture dish. 8. During removal of the labeling medium, great care must be taken as infection has proceeded for 24  h and cells at this point are more susceptible to loss during medium removal. To reduce the loss of infected cells, it may not be absolutely necessary to remove the entire labeling medium; complete DMEM medium contains excess methionine that competes away any residual AHA that may remain. 9. Centrifugation of the cells can be carried out at speeds of up to 2,000 × g. The use of higher centrifugation speeds, however, may cause rupture of the infected cells. 10. To create a CsCl density gradient, first add the 1.25 g/mL CsCl solution into the Ultra Clear centrifuge tube. Next, carefully pipette the 1.4 g/mL CsCl solution with the pipette positioned at the bottom of the tube. Take care to release the CsCl solution at this step very slowly; quick release of the heavier CsCl solution may cause mixing of the two liquids. 11. After step 3 in Subheading  3.3, the two liquids and their junction should be clearly visible. 12. After the first ultracentrifugation run, be careful as not to collect cellular proteins that will band above the virus solution. In addition, empty virion capsids can be seen as a lighter layer slightly above the thick white virus band. 13. If excess cellular proteins are present that may interfere with the collection of the virus, use a pipette to remove some of the denser proteins. 14. While collecting the virus band, try to avoid collection of excess CsCl before the virus band starts dripping out. Excess CsCl collected at this step may reduce the sharpness of the virus banding in the subsequent re-centrifugation step.

Chemoselective Modification of Viral Proteins

65

15. During rebanding on 1.35  g/mL of CsCl (step 5 in Subheading 3.3), add the virus collected in the previous step to a 4-mL ultracentrifuge tube and top it with the CsCl solution. After 18 h of centrifugation, a thick white band of pure virus will be visible in the middle of the tube. This rebanding step is necessary to remove empty capsids and other cellular proteins that may have been collected with the virus particles after the first ultracentrifugation run. 16. If for some reason the concentration of virus particles collected after the first ultracentrifugation run is very low, it may be difficult to observe rebanding of the virions after the second 18-h ultracentrifugation run. 17. Care must be taken at these stages during metabolic labeling to minimize the loss of cells and virions. Careful removal of the labeling medium, adding another freeze thaw cycle, careful collection of the first virion band during density gradient ultracentrifugation can all cumulatively add up to give a good yield of the labeled virus. 18. During the deoxygenation step for “click” labeling, keep the virus solution to less than 100 mL in order to minimize losses due to evaporation. 19. Upon addition of CuBr to the reaction mixture, the solution will turn dark green. 20. At the end of the reaction, addition of EDTA should remove the coloration noted above. 21. Purification with Centri-Sep™ spin columns should be carried according to the manufacturer’s instructions provided with the kit. 22. Alternatively, purification can also be carried out by dialysis against virus storage buffer. 23. If concerned about the presence of residual copper after purification, forgo the boiling of the proteins after the addition of sample buffer as boiling of samples in the presence of copper can result in the formation of thick streaks of protein on the SDS-polyacrylamide gel electrophoresis (SDS-PAGE) gel and will cause difficulty during fluorescent visualization. 24. SDS-PAGE running buffer can be reused and stored at 4°C for up to 2 weeks. 25. Cover the electrophoresis apparatus with aluminum foil to reduce photobleaching of the fluorophore label. 26. Allow the tracking dye to run off the gel during electrophoresis, as running the gel longer will result in better resolution and will also allow free fluorophore to run out of the gel.

66

Banerjee and Carrico

References 1. Kiick, K. L., Saxon, E., Tirrell, D. A., and Bertozzi, C. R. (2002) Incorporation of azides into recombinant proteins for chemoselective modification by the Staudinger ligation, Proceedings of the National Academy of Sciences of the United States of America 99, 19–24. 2. Carrico, I. S., Carlson, B. L., and Bertozzi, C. R. (2007) Introducing genetically encoded aldehydes into proteins, Nat Chem Biol 3, 321–322. 3. Kristi, L. K., Jan, C. M. v. H., and David, A. T. (2000) Expanding the Scope of Protein Biosynthesis by Altering the Methionyl-tRNA Synthetase Activity of a Bacterial Expression Host13, Angewandte Chemie International Edition 39, 2148–2152. 4. Dieterich, D. C., Link, A. J., Graumann, J., Tirrell, D. A., and Schuman, E. M. (2006) Selective identification of newly synthesized proteins in mammalian cells using bioorthogonal noncanonical amino acid tagging (BONCAT), Proceedings of the National Academy of Sciences 103, 9482–9487. 5. Prescher, J. A., and Bertozzi, C. R. (2005) Chemistry in living systems, Nat. Chem Biol. 1, 13–21. 6. Waehler, R., Russell, S. J., and Curiel, D. T. (2007) Engineering targeted viral vectors for gene therapy, Nat. Rev. Genet. 8, 573–587. 7. Mathis, J. M., Stoff-Khalili, M. A., and Curiel, D. T. (2005) Oncolytic adenoviruses - selective retargeting to tumor cells, Oncogene 24, 7775–7791. 8. Henning, P., Lundgren, E., Carlsson, M., Frykholm, K., Johannisson, J., Magnusson, M. K.,

Tang, E., Franqueville, L., Hong, S. S., Lindholm, L., and Boulanger, P. (2006) Adenovirus type 5 fiber knob domain has a critical role in fiber protein synthesis and encapsidation, J. Gen. Virol. 87, 3151–3160. 9. Manuel, A. F. V. G., and Antoine, A. F. d. V. (2006) Adenovirus: from foe to friend, Reviews in Medical Virology 16, 167–186. 10. Mahal, L. K., Yarema, K. J., and Bertozzi, C. R. (1997) Engineering Chemical Reactivity on Cell Surfaces Through Oligosaccharide Biosynthesis, Science 276, 1125–1128. 11. Kim, E. J., Sampathkumar, S.-G., Jones, M. B., Rhee, J. K., Baskaran, G., Goon, S., and Yarema, K. J. (2004) Characterization of the Metabolic Flux and Apoptotic Effects of O-Hydroxyland N-Acyl-modified N-Acetylmannosamine Analogs in Jurkat Cells, Journal of Biological Chemistry 279, 18342–18352. 12. Agard, N. J., Baskin, J. M., Prescher, J. A., Lo, A., and Bertozzi, C. R. (2006) A Comparative Study of Bioorthogonal Reactions with Azides, ACS Chemical Biology 1, 644–648. 13. Dieterich, D. C., Lee, J. J., Link, A. J., Graumann, J., Tirrell, D. A., and Schuman, E. M. (2007) Labeling, detection and identification of newly synthesized proteomes with bioorthogonal non-canonical amino-acid tagging, Nat. Protocols 2, 532–540. 14. Laughlin, S. T., and Bertozzi, C. R. (2007) Metabolic labeling of glycans with azido sugars and subsequent glycan-profiling and visualization via Staudinger ligation, Nat. Protocols 2, 2930–2944.

Chapter 6 Preparation of Peptide and Other Biomolecular Conjugates Through Chemoselective Ligations Mathieu Galibert, Olivier Renaudet, Didier Boturyn, and Pascal Dumy Abstract The synthesis of molecular conjugates through chemoselective ligations represents a very convenient strategy to prepare complex macromolecules with diverse functional elements. Herein, we describe chemical methods based on the preparation of chemoselectively addressable peptides allowing successive oxime ligations and/or alkyne–azide cycloaddition (“click”) reactions of various biomolecules. This modular synthetic approach can be applied to a broad range of purposes. Key words: Peptide, Carbohydrate, Biomolecular conjugates, Chemoselective ligation, Oxime ligation, Alkyne–azide cycloaddition, Click chemistry

1. Introduction The use of efficient and chemoselective ligations is essential to construct biomolecular conjugates. The major barriers to the chemical ligation of biomolecules arise from their high molecular weight and/or from the incompatibility between the chemistries of peptides, nucleic acids and carbohydrates. To overcome these limitations, the fragment-coupling approach using chemoselective ligations has proved to be particularly useful to tailor proteins (1–3), oligonucleotide (4) or carbohydrate derivatives (5). The use of chemoselective reactions yielding disulfide, thioester, thiazolidine, hydrazone, or oxime bonds proved highly efficient for the preparation of relevant macromolecules such as synthetic vaccines (6, 7), synthetic proteins (8, 9), antiviral drugs (10), or anticancer agents (11). Previously, we successfully exploited oxime bond formation to prepare a diverse range Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_6, © Springer Science+Business Media, LLC 2011

67

68

Galibert et al.

of bioconjugates (12, 13); this reaction benefits from the high reactivity between aminooxy and carbonyl groups. Among other types of chemoselective reactions that have been described, the alkyne–azide cycloaddition (“click”) reaction (14, 15) is a powerful tool to prepare new peptidic derivatives through a triazole linkage (16). Very recently, we reported a novel strategy for the synthesis of biomolecular assemblies using orthogonal oxime bond formation and copper(I)-mediated alkyne–azide cycloaddition (CuAAC) reactions in either a stepwise or a onepot approach (17). We believe that this strategy could be a very convenient method for achieving the synthesis of highly sophisticated bioconjugate assemblies.

2. Materials 2.1. Reagents and Solutions

1. Fmoc-protected amino acids and resins. 2. (Benzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate (PyBOP). 3. 2,4,6-Trinitrobenzenesulfonic acid (TNBS) test reagents: DIPEA, DMF, and TNBS. 4. 4-Pentynoic acid succinimidyl ester. 5. Acetic acid (AcOH). 6. Acetone. 7. Acetonitrile (CH3CN). 8. Celite®. 9. Copper micro-powder. 10. Dichloromethane (CH2Cl2). 11. Diethyl ether (Et2O). 12. Diisopropylethylamine (DIPEA). 13. Dimethyldichlorosilane [Si(CH3)2Cl2]. 14. Dimethylformamide (DMF). 15. Dioxane. 16. Ethanol. 17. Ethyl acetate (EtOAc). 18. Ethyl N-hydroxyacetimidate. 19. HCl. 20. Iodoacetic acid. 21. Kaiser test reagents: Ninhydrin, phenol, pyridine, and potassium cyanide (KCN).

Preparation of Peptide and Other Biomolecular Conjugates

69

22. Magnesium sulfate (MgSO4) (anhydrous). 23. Methanol. 24. Methylhydrazine. 25. N-Hydroxyphthalimide. 26. N-Hydroxysuccinimide. 27. N,N ¢-Dicyclohexylcarbodiimide. 28. NaOH. 29. Pent-4-ynoic acid. 30. Pentane. 31. Piperidine. 32. Saturated aqueous solution of sodium chloride (brine). 33. Silica gel 60 (0.063–0.2 mm/70–230 mesh) (Merck). 34. Silica gel 60 F254 thin layer chromatography (TLC) plates (Merck). 35. Sodium bicarbonate (NaHCO3). 36. Sodium periodate (NaIO4). 37. Sodium sulfate. 38. Sulfuric acid. 39. tert-Butanol (tBuOH). 40. Triethylamine. 41. Trifluoroacetic acid (TFA). 42. Trifluoroethanol (TFE). 2.2. Equipment

1. Glass reaction vessels for manual solid-phase peptide synthesis (SPPS). 2. Handheld UV light source (see Note 1). 3. Nuclear magnetic resonance (NMR) spectrometer (see Note 2). 4. High performance liquid chromatography (HPLC) system equipped with analytical C18 columns (see Note 3). 5. Mass spectrometer (see Note 4).

3. Methods 3.1. Peptide and Functionalized Building Block Synthesis

Some examples of the amino acid building blocks synthesized in this section are shown in Fig. 1.

70

Galibert et al.

O

O O

OEt N

O

N O

1

O

O O

O

O

2

O

O

N

O O

N H

O

3

N

O

OEt N O

NH

O O

NH

O O

O OH

N H

O

O

4

O

O

AcOOAc O

O

A cO

NH

O O

OAc

OAc

AcO O

6

OH

OAc

O

O HO

OH

O

N O

O

HO OH N H

O

5

O

H N

O

OH

N H

7 OH

HO O

O

NH2

O OH

8

Fig. 1. Examples of amino acid and carbohydrate building blocks used in this work.

3.1.1. Solid-Phase Peptide Synthesis (see Note 5)

1. The linear peptides were assembled on 2-chlorotritylchloride resin using standard solid-phase peptide synthesis through an Fmoc/tBu strategy. 2. The coupling reactions were performed using, relative to the resin loading, 1.5–2 eq. of Fmoc-protected amino acid activated in situ with 1.5–2 eq. of PyBOP and 3–4 eq. of DIPEA in DMF (10 mL/g resin) for 30 min. 3. The resin was washed twice with DMF (10 mL/g resin) for 1 min and twice with CH2Cl2 (10 mL/g resin) for 1 min. 4. The completeness of the amino acid coupling reactions was checked by performing Kaiser and TNBS tests.

Preparation of Peptide and Other Biomolecular Conjugates

71

5. Na-Fmoc protecting groups were removed by treatment with piperidine/DMF (1:4, v/v) (10  mL/g resin) for 10  min, three times. 6. The resin was further washed five times with DMF (10 mL/g resin) for 1 min, and twice with CH2Cl2 (10 mL/g resin) for 1 min. 7. The completeness of the deprotection was checked by UV measurement (l = 299 nm, e = 7800 M−1 cm−1). 8. The cleavage and washing solutions were collected together and the volume of the combined solution was adjusted to a known value (V) with methanol. 9. Measurement of the absorbance of the solution obtained from step 8 at l = 299 nm enabled the amount of Fmoc protecting groups released from the cleavage to be calculated according to the Beer–Lambert relation. 3.1.2. Peptide Cleavage

1. Synthetic linear peptides were recovered directly upon repeated acid cleavage of the resins (Subheading  3.1.1) using TFE/ AcOH/CH2Cl2 (2:1:7, v/v/v) (10 mL/g resin) for 2 h. 2. The solvent was removed under reduced pressure and the residue dissolved in a minimum volume of CH2Cl2. 3. Ether was added to precipitate the peptide, then triturated and washed three times with ether to yield crude material without further purification.

3.1.3. Peptide Cyclization (see Note 6)

1. The linear peptide was dissolved in DMF (0.5 mM), and the pH of the resulting solution was adjusted to 8–9 by the addition of DIPEA. 2. The reagent PyBOP (1.2 eq.) was then added, and the solution was stirred at room temperature for 1 h (see Note 7). 3. The solvent was removed under reduced pressure, and the residue dissolved in a minimum volume of CH2Cl2. 4. Ether was added to precipitate the peptide, then triturated and washed three times with ether to yield crude material without further purification.

3.1.4. Synthesis of 4-Pentynoic Acid Succinimidyl Ester 1

1. Dissolve pent-4-ynoic acid (1 eq.) and N-hydroxysuccinimide (1 eq.) in EtOAc/dioxane (1:1, v/v, 1 mM) at 0°C. 2. Add N,N ¢-dicyclohexylcarbodiimide (1 eq.) in one portion at 0°C and stir at room temperature for 5 h. 3. Remove the formed dicyclohexylurea by filtration on a pad of Celite®. Evaporate the solvent to dryness under reduced pressure.

72

Galibert et al.

4. Dissolve the residue in EtOAc and wash the organic layer with 5% (w/v) aqueous NaHCO3 (two times), water (once), and brine (once) by using a separatory funnel. 5. Dry the organic layer over anhydrous MgSO4, filter, and evaporate the filtrate to dryness. 6. Precipitation from CH2Cl2/pentane (1:10, v/v) affords the product quantitatively as a white solid (see Note 8). 3.1.5. N-Hydroxysuccinimidyl 2-(1-Ethoxyethyli­ deneaminooxy)Acetate 2 (18)

1. Dissolve iodoacetic acid in water (1 M) at 0°C and add aqueous NaOH (1 eq., 40% w/w). 2. At room temperature, add ethyl N-hydroxyacetimidate (1.2 eq.) and aqueous NaOH (1.5 eq., 40% w/w) and stir at 80°C for 5 h. 3. Cool at room temperature and then add water (2 volumes). Wash the aqueous mixture with CH2Cl2 (three times). Bring the water phase to pH 2–3 with a 2 M HCl solution. Extract with EtOAc (four times) (see Note 9). 4. Wash the combined organic phases with brine (one time), dry over MgSO4, filter and evaporate to dryness to obtain the product as a colorless oil (73% yield). 5. React the product obtained in step 4 above with N-hydroxysuccinimide by following the procedure described previously in Subheading 3.1.4. 6. Precipitation from Et2O/pentane (1:10, v/v) affords the product quantitatively as a white solid (see Note 10).

3.1.6. Synthesis of Boc-Ser(tBu)-OSu 3

1. Follow the procedure described in Subheading 3.1.4 to react the commercial reagent Boc-Ser(tBu)-OH with N-hydroxysuccinimide. 2. Precipitation from Et2O/pentane (1:10, v/v) affords the product qualitatively as a white solid (see Note 11).

3.1.7. Fmoc-Lys[N-4Pentynoic Acid]-OH 4

1. Dissolve Fmoc-Lys-OH in DMF (1 mM) and adjust the pH to 8–9 by addition of DIPEA. 2. To the above mixture, add compound 1 (1  eq.) (obtained from Subheading 3.1.4) and stir the reaction at room temperature for 2 h. 3. Evaporate the solution to dryness under reduced pressure and dissolve the residue in EtOAc. 4. Wash the organic layer with concentrated citric acid solution (three times), water and brine. 5. Dry the organic layer over anhydrous MgSO4, filter, and evaporate the filtrate to dryness.

Preparation of Peptide and Other Biomolecular Conjugates

73

6. Precipitation from CH2Cl2/pentane (1:10, v/v) affords the product as a white solid (90% yield), which is used without further purification (see Note 12). 3.1.8. Fmoc-Lys[N-EeiAoa]-OH 5 (19)

1. Follow the procedure described in Subheading 3.1.7 to react compound 2 (1 eq.; obtained from Subheading 3.1.5) with Fmoc-Lys-OH. 2. Purify the crude product by column chromatography using CH2Cl2/EtOH (9:1, v/v) as eluent. 3. After evaporation, the desired compound is obtained as a white solid (71% yield) (see Note 13).

3.1.9. Fmoc-Lys[BocSer(tBu)]-OH 6

1. Follow the procedure described in Subheading 3.1.7 to react compound 3 (1 eq.; obtained from Subheading 3.1.6) with Fmoc-Lys-OH. 2. Precipitation from CH2Cl2/pentane (1:10, v/v) affords the product as a white solid (92% yield), which is used without further purification (see Note 14).

3.1.10. Glycosylation Between N-hydroxyphthalimide and a Glycosyl Fluoride: Synthesis of Phthalimido Lactosyl 7 (20) as an Example

1. Add N-hydroxyphthalimide (1 eq.) and triethylamine (1 eq.) to a solution of acetylated lactosyl fluoride in dry CH2Cl2 (1 M) (see Note 15). 2. Add the promoter BF3·Et2O (4 eq.) and stir the solution for 15 min at room temperature (see Note 16). 3. Add CH2Cl2 to the mixture and wash the organic layer with 10% (w/v) aqueous NaHCO3 (three times) and then with water. 4. Dry the organic layer under sodium sulfate, evaporate, and purify the resulting glycosylated derivative 7 by silica gel chromatography (CH2Cl2/EtOAc, 4:1, v/v). 5. Precipitation from CH2Cl2/pentane (1:10, v/v) affords 7 as a white powder (70% yield) (see Note 17).

3.1.11. Hydrazinolysis: Synthesis of Aminooxy Lactosyl 8 (20) as an Example

1. Dissolve the acetylated compound 7 (obtained from Subheading  3.1.10) in a solution of ethanol/methylhydrazine (1:1, v/v, 10 mL). 2. Evaporate the solvent under reduced pressure after stirring the solution overnight at room temperature. 3. Precipitation from MeOH/CH2Cl2 (1:10, v/v) affords the fully deprotected and pure b-aminooxylated lactose as a white amorphous powder (74% yield) (see Note 18).

3.2. Molecular Assembly of Bioconjugates

See (Fig. 2).

74

Galibert et al.

A: Ligation Oxime/Oxime O

O O

O K K

O

O K

A K

K

O

N

Sugar

O

N

O

O

G P

Sugar

O

O

Sugar

Sugar

P G

1)

Peptide

2)

Sugar

CHO

O O

N O K

G K

P

N O K

A K

K

P G

ONH2 O ONH2

O N

Peptide

CHO

2)

Peptide

de

Peptide

P G

N N N

pti

1)

O K

Pe

K

K

O

de

K

A

pti

G P

NH2O

O K

Pe

NH2O

O

de

de

pti

pti

Pe

Pe

B: Ligation Oxime/Click

O

N N N

N3

O N

O K

G P

NO

O

K

O O K

A K

K

P G

Fig. 2. General strategies for biomolecular assembly using oxime bond formation and copper(I)-mediated alkyne–azide cycloaddition (CuAAC; “click”) reactions.

3.2.1. Oxidative Cleavage of Serine Residues to Generate Aldehyde Groups

1. Dissolve the peptide containing serine residues in CH3CN/ H2O (2:1, v/v, 10 mM). 2. Add NaIO4 (10 eq. per serine) and stir the solution at room temperature for 30 min. 3. Purify the oxidized product (containing aldehyde groups) by reversed-phase HPLC (RP-HPLC) and lyophilize the combined fractions.

3.2.2. Oxime/Oxime Ligation (21) (Fig. 2a)

1. Dissolve the peptide containing aldehyde groups in CH3CN/ H2O/TFA (1:1:0.1, v/v/v, 10 mM). 2. Add peptide containing an oxyamine function (2 eq. per aldehyde site) and stir at 37°C overnight. 3. Quench the excess of oxyamine peptide with acetone and then oxidize the crude mixture with NaIO4 by following the procedure described in Subheading 3.2.1.

Preparation of Peptide and Other Biomolecular Conjugates

75

4. After semi-preparative HPLC, dissolve the aldehydic peptide in CH3CN/H2O/TFA (1:1:0.1, v/v/v, 10 mM). 5. Add the aminooxy sugar (2 eq. per aldehyde). 6. Stir reaction mixture at 37°C overnight and purify the final compound by semi-preparative RP-HPLC. 3.2.3. Oxime/CuAAC Ligation (17) (Fig. 2b)

1. Dissolve the peptide containing alkyne and oxyamine moieties in tBuOH/H2O/AcOH (50:45:5, v/v/v, 10 mM). 2. Add the aldehyde-containing peptide (or other aldehydecontaining biomolecule) (1.5 eq. per oxyamine site) and stir at room temperature for 1 h. 3. Adjust the pH to 8 using 10% (w/w) NaHCO3 solution, and then add the azide-containing peptide (or other azide-­ containing biomolecule) (1.5 eq. per alkyne site) and micropowder copper (5 eq.). Stir the reaction mixture overnight at room temperature. 4. Centrifuge the crude solution and purify the supernatant (containing the biomolecular conjugate) by semi-preparative RP-HPLC.

4. Notes 1. UV light is used for TLC spot visualization. Prior to visualization, the TLC plate is heated after treatment with a solution of 10% sulfuric acid in ethanol. 2. 1H and 13C NMR spectra were recorded on a Bruker AC300 spectrometer and the chemical shifts (d) are reported in parts per million (ppm). Spectra were referenced to the residual proton solvent peaks relative to the signal from CDCl3 (d 7.27 and 77.23 ppm for 1H- and 13C-NMR, respectively) or relative to the signal of D2O (d 4.79  ppm for 1H-NMR). Proton and carbon assignments were obtained from GCOSY and GHMQC experiments. The anomeric configuration of the carbohydrate samples was established by determination of the coupling constant (J) between H-1 and H-2. 3. The progress of the peptide synthesis reactions was monitored by reversed-phase HPLC using a Waters or equivalent HPLC system equipped with C18 columns. Analytical HPLC (Nucleosil 120 Å 3 mm C18 particles, 30 × 4.6 mm2) was performed at 1.3  mL/min and preparative HPLC (Delta-Pak 100 Å 15  mm C18 particles, 200 × 25  mm2) at 22  mL/min with UV monitoring at 214 and 250 nm using a linear A–B gradient (buffer A: 0.09% CF3CO2H in water; buffer B: 0.09% CF3CO2H in 90% acetonitrile).

76

Galibert et al.

4. Mass spectra were obtained by electron spray ionization (ESI-MS) on a VG Platform II or by chemical ionization (DCI-MS) on a Thermo Finnigan PolarisQ in the positive mode. 5. The assembly of all linear, protected peptides was performed manually in a glass reaction vessel fitted with a sintered glass frit or automatically on a peptide synthesizer (348 W Peptide Synthesizer, Advanced ChemTech). In manual SPPS, the glass reaction vessel allows elimination of excess reagents and solvents under compressed air. Before use, the glass reaction vessel was treated for 12  h (typically overnight) with (CH3)2SiCl2 as lubricant to prevent the resin beads from sticking to the glass inner wall during the synthesis reactions. The glass reaction vessel was then carefully washed with CH2Cl2 until the complete clearance of any trace amounts of acid was achieved. 6. The presence of a b-turn in the peptide structure facilitates the cyclization process. 7. After 5 min, it is necessary to control the pH. The pH should be adjusted to 8–9 by the addition of DIPEA. 8. Characterization of 4-pentynoic acid succinimidyl ester 1: 1H NMR (300 MHz, CDCl3): d = 2.03 (t, 1H, J = 2.4 Hz), 2.60 (td, 2H, J = 2.4, 7.0 Hz), 2.83 (s, 4H), 2.87 (t, 2H, J = 7.0 Hz); 13 C NMR (75 MHz, CDCl3) d = 14.1, 25.6, 30.3, 69.6, 80.3, 167.1, 169.0. 9. After each extraction, the pH is adjusted to 2–3 with a 2 M HCl solution. 10. Characterization of N-hydroxysuccinimidyl 2-(1-ethoxyethylideneaminooxy)acetate 2: 1H NMR (CDCl3, 300  MHz) d = 4.78 (s, 2H), 4.01 (q, 2H, J = 7.2 Hz), 2.84 (s, 4H), 1.98 (s, 3H), 1.28 (t, 3H, J = 7.2 Hz). 11. Characterization of Boc-Ser(tBu)-OSu 3: 1H NMR (300  MHz, CDCl3): d = 5.41 (d, 1H, J = 9.0  Hz), 4.78 (d, 1H, J = 9.0 Hz), 3.92 (m, 1H), 3.66 (m, 1H), 2.82 (s, 4H), 1.46 (s, 9H), 1.20 (s, 9H); ESI-MS (positive mode): calculated for C16H26N2O7: 358.2, found 359.3. 12. Characterization of Fmoc-Lys[N-4-Pentynoic acid]-OH 4: 1 H NMR (300 MHz, CDCl3) d = 7.75 (d, 2H, J = 7.4 Hz), 7.60 (d, 2H, J = 7.4 Hz), 7.38 (t, 2H, J = 7.4 Hz), 7.29 (t, 2H, J = 7.4  Hz), 6.0 (t, 1H, J = 5.6  Hz), 5.70 (d, 1H, J = 7.8  Hz), 4.38–4.36 (m, 3H), 4.12 (t, 1H, J = 6.9  Hz), 3.19 (m, 2H), 2.42 (t, 2H, J = 7.0 Hz), 2.30 (dt, 2H, J = 2.3, 7.0 Hz), 1.92 (t, 1H, J = 2.4 Hz), 1.96 (s, 3H), 1.81 (m,2H), 1.58 (m, 2H), 1.45 (m, 2H), 1.24 (t, 3H, J = 7.1 Hz); 13C NMR (CDCl3, 75  MHz) d = 173.9, 170.0, 156.1, 143.8,143.7, 140.6, 127.6, 127.0, 125.2, 120.0, 83.7, 71.2,

Preparation of Peptide and Other Biomolecular Conjugates

77

65.6, 53.7, 46.6, 38.2, 34.2, 30.4, 28.6, 23.0, 14.2; ESI-MS (positive mode): calculated for C26H28N2O5: 448.5, found 449.1. 13. Characterization of Fmoc-Lys[N-Eei-Aoa]-OH 5: 1H NMR (300  MHz, CDCl3) d = 7.75 (d, 2H, J = 7.4  Hz), 7.60 (d, 2H, J = 7.4  Hz), 7.38 (t, 2H, J = 7.4  Hz), 7.29 (t, 2H, J = 7.4 Hz), 6.50 (t, 1H, J = 5.6 Hz), 5.70 (d, 1H, J = 7.8 Hz), 4.38–4.36 (m, 5H), 4.20 (t, 1H, J = 6.9 Hz), 3.96 (q, 2H, J = 7.1 Hz), 3.34 (m, 2H), 1.96 (s, 3H), 1.81 (m, 2H), 1.58 (m, 2H), 1.45 (m, 2H), 1.24 (t, 3H, J = 7.1 Hz); 13C NMR (75  MHz, CDCl3) d = 174.7, 171.3, 164.4, 156.2, 143.7, 141.3, 127.7, 127.1, 125.1, 119.9, 72.6, 67.1, 62.8, 53.7, 47.2, 38.5, 31.7, 29.1, 22.1, 14.2, 13.9; ESI-MS (positive mode): calculated for C27H33N3O7: 511.2, found 512.1. 14. Characterization of Fmoc-Lys[Boc-Ser(tBu)]-OH 6: 1H NMR (300 MHz, CDCl3): d = 7.68 (d, 2H, J = 7.5 Hz), 7.55 (d, 2H, J = 7.5 Hz), 7.33–7.18 (m, 4H), 6.84 (broad s, 1H), 6.25 (broad d, 1H), 5.62 (broad s, 1H), 4.34–4.06 (m, 5H), 3.63 (m, 1H), 3.36 (m, 1H), 3.20–3.05 (m, 2H), 1.84 (m, 1H), 1.65 (m, 1H), 1.47–1.32 (m, 14H),1.09 (s, 9H); ESI-MS (positive mode): calculated for C33H45N3O8: 611.3, found 611.2. 15. The solution immediately turns orange after the addition of triethylamine. 16. The solution becomes colorless after the addition of the promoter. This decoloration indicates the completeness of the reaction. 17. Characterization of O-(2,3,4,6-tetra-O-acetyl-b-dgalactopyranosyl)-(1 → 4¢)-(2¢,3¢,6¢-tri-O-acetyl-b- d glucopyranosyl)-N-oxyphthalimide 7: 1H NMR (300 MHz, CDCl3): d = 7.86–7.74 (m, 4 H, Har.), 5.34 (dd, 1 H, J4¢,5¢ = 0.9 Hz, J3¢,4¢ = 3.3 Hz, H-4¢), 5.28–5.18 (m, 2 H, H-2, H-3), 5.14 (d, 1 H, J1,2 = 6.9  Hz, H-1), 5.10 (dd, 1 H, J1¢,2¢ = 7.8 Hz, J2¢,3¢ = 10.4 Hz, H-2¢), 4.95 (dd, 1 H, J3¢,4¢ = 3.3 Hz, J2¢,3¢ = 10.4 Hz, H-3¢), 4.53 (d, 1 H, J1¢,2¢ = 7.8 Hz, H-1¢), 4.42 (dd, 1 H, J5,6a = 2.2 Hz, J6a,6b = 12.1 Hz, H-6a), 4.15 (dd, 1 H, J5,6b = 5.8 Hz, J6a,6b = 12.1 Hz, H-6b), 4.11–4.07 (m, 3 H, H-4, H-6¢), 3.88 (td, 1 H, J4¢,5¢ = 0.9 Hz, J5¢,6¢ = 6.8 Hz, H-5¢), 3.78–3.73 (m, 1 H, H-5), 2.16, 2.13, 2.08, 2.06, 2.05, 2.03, 1.94 (7s, 21 H, 7OCOCH3); 13C NMR (75 MHz, CDCl3): d = 170.8 (C=OAc), 170.7 (C=OAc), 170.5 (C=OAc), 170.4 (C=OAc), 170.0 (C=OAc), 169.9 (C=OAc), 169.5 (C=OAc), 163.1 (C=OPht), 135.1 (CHar.), 129.1 (Car.), 124.2 (CHar.), 104.5 (C-1), 101.5 (C-1¢), 76.3 (C-4), 73.3, 73.1 (C-3, C-5), 71.4, 71.1 (C-3¢, C-5¢), 70.4 (C-2), 69.4 (C-2¢), 67.1 (C-4¢), 62.5 (C-6), 61.3 (C-6¢), 21.2 (OCOCH3), 21.1 (OCOCH3), 21.0 (OCOCH3), 20.9 (OCOCH3); ESI-HRMS

78

Galibert et al.

(positive mode): calculated for 804.19631 [M+Na]+, found: 804.19576.

C34H39NO20Na:

18. Characterization of O-(b-d-galactopyranosyl)-(1 → 4¢)-(b-dglucopyranosyl) oxyamine 8: 1H NMR (300  MHz, D2O): d = 4.62 (d, 1 H, J1¢,2¢ = 8.3  Hz, H-1¢), 4.47 (d, 1 H, J1,2 = 7.7  Hz, H-1), 4.03 (dd, 1 H, J5¢,6a¢ = 1.3  Hz, J6a¢,6b¢ = 11.6 Hz, H-6a¢), 3.95 (bd, 1 H, J3,4 = 3.1 Hz, H-4), 3.84 (dd, 1 H, J5¢,6b¢ = 4.6 Hz, J6a¢,6b¢ = 11.6 Hz, H-6b¢), 3.793.63 (m, 7 H, H-3, H-5, H-6, H-3¢, H-4¢, H-5¢), 3.66 (dd, 1 H, J1,2 = 7.7  Hz, J2,3 = 9.8  Hz, H-2), 3.38 (bt, 1 H, J2¢,3¢ = 8.4 Hz, H-2¢); 13C NMR (75 MHz, D2O): d = 105.2 (C-1¢), 103.3 (C-1), 78.6, 75.7, 75.1, 74.8, 72.8, 71.7 (C-2¢), 71.3 (C-2), 68.9 (C-4), 61.4 (C-6¢), 60.4 (C-6). ESIHRMS (positive mode): calculated for C12H23NO11Na: 380.11688 [M+Na]+, found: 380.11633.

Acknowledgements We thank the Université Joseph Fourier (UJF-Grenoble), the Centre National de la Recherche Scientifique (CNRS), and the NanoBio program (Grenoble) for providing support for this work. References 1. Dawson, P. E. and Kent S. B. H. (2000) Synthesis of native proteins by chemical ligation. Annu. Rev. Biochem. 69, 923–960. 2. Hang, H. C. and Bertozzi, C. R. (2001) Chemoselective approaches to glycoprotein assembly. Acc. Chem. Res. 34, 727–736. 3. Carrico, I. S. (2008) Chemoselective modification of proteins: hitting the target. Chem. Soc. Rev. 37, 1423–1431. 4. Venkatesan, N. and Kim, B. H. (2006) Peptide conjugates of oligonucleotides: synthesis and applications. Chem. Rev., 106, 3712–3761. 5. Peri, F. and Nicotra, F. (2004) Chemoselective ligation in glycochemistry. Chem. Commun., 623–627. 6. Verez-Bencomo, V., Fernández-Santana, V., Hardy, E., Toledo, M. E., Rodríguez, M. C., Heynngnezz, L., Rodriguez, A., Baly, A., Herrera, L., Izquierdo, M., Villar, A., Valdés, Y., Cosme, K., Deler, M. L., Montane, M., Garcia, E., Ramos, A., Aguilar, A., Medina, E., Toraño, G., Sosa, I., Hernandez, I., Martínez, R., Muzachio, A., Carmenates, A., Costa, L.,

Cardoso, F., Campa, C., Diaz, M. and Roy, R. (2004) A synthetic conjugate polysaccharide vaccine against haemophilus influenzae type b. Science 305, 522–525. 7. Zeng, W., Jackson, D. C., Murray, J., Rose, K. and Brown, L. E. (2000) Vaccine 18, 1031–1039. 8. Rose, K. (1994) Facile synthesis of homogeneous artificial proteins. J. Am. Chem. Soc. 116, 30–33. 9. Canne, L. E., Ferré-D’Amare, A. R., Burley, S. K. and Kent S. B. H. (1995) Total chemical synthesis of a unique transcription factorrelated protein: cMyc-Max. J. Am. Chem. Soc. 117, 2998–3007. 10. Naicker, K. P., Li, H., Heredia, A., Song H. and Wang, L.-X. (2004) Design and synthesis of aGal-conjugated peptide T20 as novel antiviral agent for HIV-immunotargeting. Org. Biomol. Chem. 2, 660–664. 11. Henry, M. D., Wen, S., Silva, M. D., Chandra, S., Milton, M. and Worland, P. J. (2004) A prostate-specific membrane antigen-targeted

Preparation of Peptide and Other Biomolecular Conjugates monoclonal antibody–chemotherapeutic conjugate designed for the treatment of prostate cancer. Cancer Res. 64, 7995–8001. 12. Garanger, E. Boturyn, D.; Renaudet, O. Defrancq, E. and Dumy, P. (2006) Chemoselectively addressable template: a valuable tool for the engineering of molecular conjugates. J. Org. Chem. 71, 2402–2410. 13. Renaudet, O., Boturyn, D. and Dumy, P. (2009) Biomolecular assembly by iterative oxime ligations. Bioorg. Med. Chem. Lett. 19, 3880–3883. 14. Tørnoe, C. W., Christensen, C. and Meldal, M. (2002) Peptidotriazoles on solid phase: [1,2,3]-triazoles by regiospecific copper(I)catalyzed 1,3-dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem. 67, 3057–3064. 15. Rostovtsev, V. V., Green, L. G., Fokin, V. V. and Sharpless, K. B. (2002) A stepwise huisgen cycloaddition process: copper(I)-catalyzed regioselective ligation of azides and terminal alkynes. Angew. Chem. Int. Ed. 41, 2596–2599. 16. Angell, Y. L. and Burgess, K. (2007) Peptidomimetics via copper-catalyzed azide– alkyne cycloadditions. Chem. Soc. Rev. 36, 1674–1689.

79

17. Galibert, M., Dumy, P. and Boturyn, D. (2009) One-pot approach to well-defined biomolecular assemblies via orthogonal chemoselective ligations. Angew. Chem. Int. Ed. 48, 2576–2579. 18. Duléry, V., Renaudet, O. and Dumy, P. (2007) Ethoxyethylidene protecting group prevents N-overacylation in aminooxy peptide synthesis. Tetrahedron 63, 11952–11958. 19. Foillard, S., Olsten Rasmussen, M., Razkin, J., Boturyn, D. and Dumy, P. (2008) 1-ethoxyethylidene, a new group for the stepwise spps of aminooxyacetic acid-containing peptides. J. Org. Chem. 73, 983–991. 2 0. Renaudet, O. and Dumy, P. (2006) On-bead synthesis and binding assay of chemoselectively template-assembled multivalent neoglycopeptides. Org. Biomol. Chem. 4, 2628–2636. 21. Renaudet, O., Křenek, K., Bossu, I., Dumy, P., Kádek, A., Adámek, D., Vaněk, O., Kavan, D., Gažák, R., Šulc, M., Bezouška, K. and Křen, V. (2010) synthesis of multivalent glycoconjugates containing the immunoactive LELTE peptide: Effect of Glycosylation on cellular activation and natural killing by human peripheral blood mononuclear cells. J. Am. Chem. Soc. 132, 6800–6808.

Chapter 7 New Fluorescent Substrates of Microbial Transglutaminase and Its Application to Peptide Tag-Directed Covalent Protein Labeling Noriho Kamiya and Hiroki Abe Abstract Transglutaminase (TGase) is an enzyme that catalyzes the post-translational covalent cross-linking of Gln- and Lys-containing peptides and/or proteins according to its substrate specificity. We have recently designed a variety of Gln-donor fluorescent substrates of microbial transglutaminase (MTG) from Streptomyces mobaraensis and evaluated their potential use in MTG-mediated covalent protein labeling. The newly designed substrates are based on the relatively broad substrate recognition of MTG for the substitution of the N-terminal group of a conventional TGase substrate, benzyloxycarbonyl-l-glutaminylglycine (Z-QG). It is revealed that MTG is capable of accepting a diverse range of fluorophores in place of the N-terminal moiety of Z-QG when linked via a suitable linker. Here, we show the potential utility of a new fluorescent substrate for peptide tag-directed covalent protein labeling by employing fluorescein-4-isothiocyanate-b-Ala-QG as a model Gln-donor substrate for MTG. Key words: Fluorescent substrate peptide, Peptide tag, Site-specific protein modification, Transglutaminase

1. Introduction The site-specific labeling of proteins with small molecules has been widely employed for probing and/or utilizing protein functions both in vivo and in vitro. Among the current protein labeling strategies available, covalent protein labeling is superior in terms of its robustness, which is a key factor in a range of practical applications. In particular, the use of a short peptide tag that can be post-translationally modified by a variety of means is highly useful for directing site-specific protein modifications because the incorporation of such types of short tag sequences into the

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_7, © Springer Science+Business Media, LLC 2011

81

82

Kamiya and Abe

N- and/or C termini of target proteins can typically be easily achieved by standard genetic manipulations. Moreover, the introduction of a short tag can help minimize perturbations of the intrinsic function of target proteins. To further the development of peptide tag-directed protein modification strategies, we have focused on the use of transglutaminase (TGase) – and more specifically, microbial transglutaminase (MTG) from Streptomyces mobaraensis. MTG catalyzes the post-translational covalent cross-linking of Gln- and Lyscontaining peptides and/or proteins and can also accept amineterminating small probe molecules in place of Lys-containing peptides, which is very useful for introducing a new chemical entity to a substrate protein (1, 2). The practical utility of MTG in biotechnological applications was validated by demonstrations of the site-specific conjugation of a native protein with synthetic polymers (3) and the synthesis of recombinant proteins tagged with genetically introduced substrate peptide tags (4, 5). The basic concept has now been extended to the first example of the enzymatic labeling of recombinant proteins with oligonucleotides chemically modified with a substrate peptide of MTG (6). Several other studies have also revealed that MTG-mediated protein modification can be used to achieve the site-specific immobilization of recombinant proteins to solid surfaces, as well (7–9). In this chapter, we first focus on the synthesis of fluorescein4-isothiocyanate (FITC)-b-Ala-QG (Fig. 1), a new fluorescent probe molecule that can be used in site-specific, covalent protein labeling reactions catalyzed by MTG (10). Following this, we describe how a substrate peptide of MTG consisting of six amino acids (MKHKGS, abbreviated hereafter as K6-tag) can be fused to the N terminus of Escherichia coli alkaline phosphatase (AP) through appropriate genetic modifications to generate N-terminal K6-tagged AP (NK6-AP). Finally, in the last part of this chapter, we describe the covalent labeling of NK6-AP, our model recombinant protein substrate, with the new fluorescent probe molecule FITC-b-Ala-QG.

Fig. 1. Molecular structure of a conventional MTG substrate (Z-QG) and two new substrate peptides (Flc-QG; FITC-bAla-QG) designed in this work.

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

83

2. Materials A useful catalytic property of MTG, which can employ small probe molecules (e.g., cadaverine derivatives) containing a primary amine as substrates, allows the labeling of a specific Gln residue in native and recombinant proteins (11). However, there are a few cases where Gln-containing small fluorescent probes have been used for labeling a specific Lys residue in target proteins. For example, the pioneering work by Fuchsbauer and coworkers demonstrated that chemically labeling the C-terminal carboxylic group of benzyloxycarbonyl-l-glutaminylglycine (Z-QG) with monodansylcadaverine yielded a fluorolabeled Z-QG, and that this designed fluorescent substrate could be recognized by MTG (12). In subsequent reports, Z-QG and Z-QQPL derivatives labeled with fluorophores at the C termini were also prepared for MTG- and human tissue TGase-mediated labeling of IgG antibodies, respectively (13). Because all the existing fluorescent Glndonor substrates of MTG are comprised of the core structure, Z-QG, our laboratory’s first attempt to develop alternative donor substrates focused on simply altering the N-terminal Z moiety with a fluorophore. To this end, compound Flc-QG (Fig. 1) was prepared and applied to MTG-mediated conjugation with NK6-AP; however, no reaction was evident. On the other hand, it was revealed that the introduction of a b-Ala linker between the fluorophore and reactive Gln residue (e.g., FITC-b-Ala-QG; see Fig.  1) dramatically enhances the reactivity of the compound toward MTG. Interestingly, we further discovered that MTG can also accept a diverse range of fluorophores in place of the FITC moiety of FITC-b-Ala-QG (10). The following experimental procedures can be generalized for the use of different pairs of fluorescent substrates and recombinant proteins fused with a MTG-reactive Lys-containing peptide tag (see Note 1). 2.1. Synthesis of FITC-b-Ala-QG by Fmoc Solid-Phase Peptide Synthesis

1. Empty PD10 column (GE Healthcare). 2. H-Gly-Trt (2-Cl) resin, Fmoc-Gln(Trt)-OH, Fmoc-b-Ala-OH. 3. 2-(1H-Benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU). 4. 1-Hydroxybenzotriazole (HOBt). 5. N,N-Dimethylformamide (DMF). 6. Dichloromethane. 7. Trifluoroacetic acid (TFA). 8. Piperidine (PPD) (20%, v/v) solution in DMF (PPD/DMF). 9. Triisopropylsilane (TIPS). 10. Fluorescein-4-isothiocyanate.

84

Kamiya and Abe

11. N,N-Diisopropylethylamine (DIEA). 12. tert-Butyl methyl ether. 13. a-Cyano-4-hydroxycinnamic acid (CHCA). 14. Reversed-phase high-performance liquid chromatography (RP-HPLC) system. 15. RP-HPLC Eluent A: 0.1% (v/v) TFA in water. 16. RP-HPLC Eluent B: Acetonitrile. 17. RP-HPLC column: Inertsil ODS-3, 10 × 250  mm (GL Science Inc.). 18. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometer (MALDI-TOF MS) (Bruker Autoflex III). 19. MALDI-TOF matrix: 10  mg/mL CHCA in acetonitrile/ Milli-Q water/TFA (50:50:0.1, v/v). 2.2. Preparation of NK6-AP Recombinant Protein Modified with a MTG-Specific Peptide Tag

1. Plasmid pET22b(+) (Novagen). 2. Gene fragment encoding E. coli alkaline phosphatase (AP) (we used the phoa gene from plasmid pET20, a kind gift from Dr. Hiroshi Ueda of The University of Tokyo). 3. Restriction enzymes (ApaI, BamHI, HindIII, XhoI). 4. Pyrobest® DNA polymerase (Takara Bio Inc., Japan). 5. E. coli strain JM109. 6. E. coli strain BL21(DE3). 7. Isopropyl-1-thio-b-d-galactopyranoside (IPTG). 8. Ampicillin. 9. HisTrap™ HP chromatography column (GE Healthcare). 10. BCA protein assay kit (Pierce). 11. Luria-Bertani medium (LB medium): Dissolve 10  g bactotryptone, 5 g bacto-yeast extract, and 10 g NaCl in deionized water to a total volume of 1  L. Adjust the pH to 7.0 with sodium hydroxide and sterilize the medium by autoclaving for 20 min at 0.2 MPa before use. 12. Buffer A: 50 mM Tris–HCl (pH 7.4), 1 mM EDTA, and 20% (w/w) sucrose. 13. Buffer B: 20  mM Tris–HCl (pH 7.4), 500  mM NaCl, and 35 mM imidazole. 14. Buffer C: 20 mM Tris–HCl (pH 7.4), 500 mM NaCl, and 500 mM imidazole.

2.3. MTG-Mediated Labeling of NK6-AP with FITC-b-Ala-QG

1. MTG from Streptomyces mobaraensis (705 U/g, in a purified form provided by Ajinomoto Co. Ltd., Japan) (see Note 2). 2. ATTO AE-6500 mini-slab polyacrylamide gel electrophoresis (PAGE) system (ATTO Co., Japan).

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

85

3. SDS–PAGE buffer D (4×, for separating gel preparation): 1.5 M Tris–HCl, pH 8.8. 4. SDS–PAGE buffer E (4×, for stacking gel preparation): 0.5 M Tris–HCl, pH 6.8. 5. SDS–PAGE buffer F (1× running buffer): Dissolve 3.03  g Tris, 14.4 g glycine, and 1 g SDS in deionized water to a total volume of 1 L. 6. 30% (w/v) Acrylamide/bisacrylamide stock solution (29:1 w/w with 3.3% crosslinking) (see Note 3). 7. 10% (w/v) SDS in water. 8. 10% (w/v) Ammonium persulfate (APS) in water. 9. N,N,N ¢,N ¢-Tetramethylethylenediamine (TEMED). 10. SDS–PAGE sample buffer (2×): 12% (v/v) 2-mercaptoethanol, 4% (w/v) SDS, and 20% (v/v) glycerol in 100 mM Tris– HCl, pH 6.8. 11. Molecular Imager FX Pro imaging system (Bio-Rad Inc., USA) equipped for fluorescence imaging. 12. Dimethyl sulfoxide (DMSO). 13. p-Nitrophenyl phosphate (pNPP). 14. Quick-Coomassie brilliant blue (CBB) protein staining kit (Wako Pure Chemical Industries, Ltd., Japan).

3. Methods 3.1. Synthesis of FITC-b-Ala-QG by SPPS

1. FITC-b-Ala-QG was manually synthesized by Fmoc solidphase peptide synthesis (SPPS) (14) using a PD10 column filled with H-Gly-Trt (2-Cl) resin (0.1 mmol). 2. The coupling reactions were conducted with 5 eq. (relative to the resin) of an Fmoc-amino acid activated in situ with 5 eq. of HBTU, 5  eq. of HOBt, and 10  eq. of DIEA in DMF (2.3 mL) for 2–3 h. Removal of the Fmoc-protecting group was achieved by treatment with PPD/DMF for 15 min. 3. The deprotection and cleavage of the target peptide from the resin was achieved by the addition of TFA/TIPS/H2O (95:2.5:2.5, v/v/v) for 1 h. 4. The mixture was filtered to remove the resin, and the filtrate was treated with tert-butyl methyl ether (three to five times the volume of the filtrate) to precipitate the cleaved peptide. 5. The collected solid was lyophilized to yield a yellow powder.

86

Kamiya and Abe

Fig. 2. Verification of the synthesis of FITC-b-Ala-QG by mass spectroscopic analysis.

6. The peptide was purified by RP-HPLC (UV monitoring at 493 nm; flow rate = 5 mL/min; 10–80% Eluent B in 30 min using a linear elution gradient). 7. The RP-HPLC fractions containing the peptide product were collected and lyophilized. 8. The synthesis of FITC-b-Ala-QG was confirmed by MALDITOF MS (detection mode: linear positive). Mass spectrum analysis: C31H29N5O10S; calculated exact mass 663.16; found m/z [M+H]+ 664.2 (Fig. 2). 3.2. Preparation of NK6-AP 3.2.1. Plasmid Vector Construction for NK6-AP Expression in E. coli

1. A DNA fragment encoding alkaline phosphatase (AP) was amplified by the polymerase chain reaction (PCR) using a plasmid encoding AP (pET20-AP) as the DNA template. The overall molecular cloning procedure is shown in Fig. 3 (see Note 4). The primer nucleotide sequences used for this PCR reaction were 5¢-GGG GGG ATC CAC CCC CAG AAA TGC CTG TTC TAG-3¢ (Primer 1) and 5¢-CCC CCA AGC TTC TCG AGT TTC AGC CCC AGA GCG GCT TTC ATG G-3¢ (Primer 2). The resultant gene fragment encoding

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

87

Fig. 3. Cloning of the phoa gene encoding alkaline phosphatase (ALP) into plasmid pET22.

AP (with BamHI and XhoI sites) was digested with the restriction enzymes BamHI and XhoI, and cloned into a bacterial plasmid expression vector, pET22b(+), to generate the plasmid pET22-AP. 2. To attach the K6-peptide tag to AP, a second PCR reaction was conducted using pET22-AP as the DNA template. The overall procedure is shown in Fig. 4 (see Note 5). The primer nucleotide sequences used for this PCR reaction were 5¢GCC AGC CAG ACG CAG ACG CGC CGA GAC AGA-3¢ (Primer 3) and 5¢-GGT GGA TCC TTT ATG TTT CAT GGC CAT CGC CGG CTG GGC AG-3¢ (Primer 4). The resultant gene fragment encoding NK6-AP was again cloned into pET22-AP, but this time by ApaI/BamHI doubledigestion, to produce plasmid pET22-NK6-AP. 3.2.2. NK6-AP Protein Expression in E. coli

1. NK6-AP was expressed by transforming E. coli strain BL21(DE3) cells with the recombinant plasmid pET22-NK6AP obtained from Subheading  3.2.1 above. Transformants carrying the recombinant expression plasmid were grown in

88

Kamiya and Abe

Fig. 4. Construction of plasmid pET22-NK6-AP for the expression of NK6-AP in E. coli.

50  mL of LB medium supplemented with ampicillin (100  mg/L) at 37°C. The resulting overnight seed culture was used to inoculate 1 L of LB medium supplemented with ampicillin (100 mg/L), and the cells were grown at 37°C to an optical density (OD600) of 0.6. To induce the expression of NK6-AP protein, the incubation temperature was lowered to 27°C and IPTG was added to the culture to give a final concentration of 0.1 mM. The culture was then grown further for 24 h at 27°C. Following this, the cells were harvested by centrifugation, washed with Buffer A, and incubated in an ice-cold 5 mM MgCl2 aqueous solution (see Note 6). 2. The lysates were centrifuged (30,000 × g, 30 min) at 4°C and then purified with a Ni-NTA column (HisTrap™ HP) by using the hexahistidine-tag (His-tag) sequence attached to the C terminus of NK6-AP. After equilibrating the Ni-NTA column with Buffer B, the lysate sample was applied and the column was washed with Buffer B.

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

89

3. The purification of NK6-AP was confirmed by SDS–PAGE (see Subheading  3.2.3) and N-terminal protein sequencing (see Note 7). 4. The protein concentration of the purified sample was determined by employing the BCA assay kit according to the manufacturer’s instructions (using bovine serum albumin as the standard). 5. The catalytic activity of NK6-AP was measured by using pNPP as the substrate: To 1 mL of 1 M Tris–HCl buffer (pH 8.0), the hydrolysis of pNPP (1  mM) was initiated by the addition of NK6-AP (30 nM) at 25°C (see Note 8). 3.2.3. SDS–PAGE Analysis of Protein Samples

1. The glass plates for the analysis were scrubbed clean with 20% ethanol prior to use. 2. A monomer solution for a 10% (w/v) acrylamide separating gel was prepared by combining 2.67 mL of 30% (w/v) acrylamide/bisacrylamide solution, 2 mL of Buffer D, 3.25 mL of water, 80 mL of 10% (w/v) SDS, 6 mL of TEMED, and 27 mL of 10% (w/v) APS, followed by gentle but thorough mixing. The mixture was poured between the glass plates, leaving space for a stacking gel, and overlaid with water-saturated n-butanol. The separating gel was allowed to polymerize for about 1 h. 3. The n-butanol was poured off and the top of the gel was rinsed with deionized water. 4. A monomer solution for a stacking gel was prepared by combining 0.4 mL of 30% (w/v) acrylamide/bisacrylamide solution, 0.625 mL Buffer E, 1.44 mL water, 25 mL 10% (w/v) SDS, 4 mL TEMED, and 8.3 mL 10% (w/v) APS, followed by gentle but thorough mixing. The mixture was poured on top of the separating gel, the comb was inserted, and the stacking gel was overlaid with water-saturated n-butanol. The stacking gel was allowed to polymerize for 30 min. 5. Once the gel polymerization had completed, the comb was carefully removed and the wells were rinsed with deionized water. 6. Samples for SDS–PAGE analysis were prepared by mixing 10 mL of the sample with an equivalent volume of 2× sample buffer, followed by heat-treatment at 95°C for 15 min. Before applying to the wells of the gel, the samples were incubated for 30 min at 37°C. 7. Running buffer was added to the upper and lower chambers of the electrophoresis unit, and the samples (10  mL) were loaded into the wells. One well was loaded with prestained protein molecular weight markers (see Note 9).

90

Kamiya and Abe

8. After completing the assembly of the electrophoresis unit and connecting to a power supply, the gel was run at 250 V for 100 min. 9. The gel was stained with CBB using the Quick-CBB protein staining kit according to the manufacturer’s instructions. 3.3. MTG-Mediated Labeling of NK6-AP with FITC-b-Ala-QG

1. A stock solution (20 mM) of the fluorescent substrate FITCb-Ala-QG was prepared in DMSO. 2. The labeling reaction mixture was comprised of NK6-AP protein (0.5 mg/ml) and FITC-b-Ala-QG substrate (1 mM) in 100 mM Tris–HCl buffer (pH 8.0) containing of 5% (v/v) DMSO (see Note 10). 3. The labeling reaction was initiated by the addition of MTG (0.1 U/ml), and then the mixture was incubated at 4°C for 6 h (see Note 11). 4. To follow the time course of the protein labeling reaction, the labeling reaction products were analyzed by SDS–PAGE (see Subheading  3.2.3) (Fig.  5). A small aliquot of the reaction mixture was removed at periodic time intervals and mixed with SDS–PAGE 2× sample buffer to terminate the MTG reaction (see Note 12). 5. After completing the electrophoresis run, the in-gel fluorescence from the FITC-b-Ala-QG label appended to NK6-AP should be analyzed before staining the gel with CBB. The progress of the reaction can be followed by the increase in the fluorescence of the protein bands in the fluorescence image of the SDS–PAGE gel (Fig. 6) (see Note 9). In this work, the fluorescence intensity of the SDS–PAGE gel was measured at

Fig. 5. SDS–PAGE analysis of the MTG-mediated labeling of NK6-AP and wild-type AP with FITC-b-Ala-QG. Left : Gel image after staining with Coomassie brilliant blue stain. Right : Corresponding fluorescence image of the gel before CBB staining.

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

91

Fig. 6. SDS–PAGE analysis of the time course of the MTG-mediated labeling of NK6-AP with FITC-b-Ala-QG. The progress of the labeling reaction was followed by monitoring changes in the fluorescence signal intensities derived from labeled protein fractions. Top: Gel image after staining with Coomassie brilliant blue stain. Bottom: Corresponding fluorescence image of the gel before CBB staining.

room temperature with a fluoroimager system using excitation at 488 nm and a 530 ± 15 nm band-pass emission filter. A quantitative analysis of the measured signal intensities was conducted by using the Quantity One® software provided by the fluoroimager instrument manufacturer. For purposes of following the time course of the protein labeling reaction, the maximum fluorescence signal intensity measured in the gel was defined as 100%. 6. Under the described experimental conditions, the labeling of NK6-AP was observed to be nearly complete within 1 h. To ensure complete labeling, however, we recommend that the labeling reactions should be incubated for 6 h (see Note 13).

4. Notes 1. The simple and straightforward approaches described in this chapter are potentially applicable to labeling any type of recombinant protein; however, it should be noted that nonspecific modification may occur depending on the tertiary structure of substrate proteins. Based on our own investigations with several recombinant proteins (e.g., bacterial alkaline phosphatase (6, 8, 10), dihydrofolate reductase (5), enhanced green fluorescent protein (4), glutathione S-transferase (15), and cytochrome P450 protein components from Pseudomonas putida (16)), it has been revealed that, in general, unstable proteins tend to be recognized and crosslinked

92

Kamiya and Abe

more easily by MTG. However, even though a protein of interest may contain a structurally disordered region, the protein may not be subjected to MTG recognition if there is no Gln or Lys residue present and the overall protein scaffold is stable enough. A recent comprehensive review by Fontana et al. covers the important physical and structural factors of substrate proteins that are susceptible to TGase-mediated site-specific post-translational protein modification (17). 2. MTG can be commercially obtained and purified according to the reported protocol (18). The catalytic activity of MTG was measured by using the colorimetric hydroxamate procedure with Z-QG as described previously in ref.19. One unit (U) of activity is defined as the amount of enzyme that catalyzes the formation of 1  mmol of hydroxamate per minute using l-glutamic acid g-monohydroxamate as the standard. 3. Caution: Be careful in handling the acrylamide/bis solution since acrylamide monomer is a potent neurotoxin. 4. To introduce the K6-tag just before the pelB coding sequence of pET22b(+) in the subsequent step, the BamHI site in the plasmid was used to provide the Gly-Ser-encoding gene sequence (Fig. 4). 5. The PCR reaction with Primer 4 makes it possible to introduce the K6-tag via a BamHI site on the primer. Primer 3 was designed by overlapping the single ApaI site in the plasmid, which makes it easier to purify the PCR product by agarose gel electrophoresis. 6. The plasmid pET22-NK6-AP was designed to produce NK6-AP in the periplasmic space of the host E. coli cells. The enzyme product is isolated from the periplasm by osmotic shock with treatment of Buffer A, and subsequent extraction using an aqueous MgCl2 solution. If the protein yield is low, then cell disruption by gentle sonication may also be applied. 7. To confirm the expression of the protein of interest, the N-terminal sequence should be checked. In the case of purified NK6-AP, the N-terminal amino acid sequence was clearly identified as being MKHKG, suggesting that intrinsic alkaline phosphatase from the E. coli host did not contaminate the purified protein preparation. 8. E. coli alkaline phosphatase exists in a dimeric quaternary structure. For the alkaline phosphatase activity assays, the molar concentration of NK6-AP was calculated based on the monomeric unit. 9. When following the time course of the labeling reaction by the increase in the fluorescence intensity derived from the protein band in the gel, the gel should be analyzed using a

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

93

fluoroimager instrument before staining with any dyes. To visualize the protein molecular weight markers in the analysis, one can use commercial fluorescent molecular weight markers (e.g., DyLight fluorescent protein molecular weight markers from Pierce). 10. Although DMSO was employed in this work as the solvent for preparing stock solutions to increase the solubility of the synthetic substrates, the use of DMSO is not always required. In the case of FITC-b-Ala-QG, the fluorescent substrate can be dissolved in aqueous solutions up to a concentration of 1 mM in 100 mM Tris–HCl buffer, pH 8. 11. To shorten the overall reaction time, the protein labeling reaction can be conducted at higher temperature (e.g., 37°C) and/ or by using a higher MTG concentration (e.g., 1 U/mL). 12. MTG-catalyzed reactions can also be terminated by the addition of N-ethylmaleimide (NEM) to the reaction mixture under approximately neutral pH conditions. The NEM working concentration should be adjusted according to the MTG concentration used in the labeling reaction. In our case, we employed NEM at a final concentration of 1 mM. 13. For the direct identification of the labeling site of NK6-AP, one can conduct peptide mapping analyses with MALDITOF MS (4, 10). In analyzing MTG-mediated protein labeling reactions with a Gln-donor substrate (e.g., Z-QG), digestion of the labeled proteins with a protease whose substrate recognition site includes Lys residues is very useful because this provides different MS patterns before and after MTG-mediated protein labeling with a specific Lys residue on the target protein. In this work, we found that the second Lys residue in the K6-tag (MKHKGS, underlined) was specifically labeled (10). For the control experiments, the MTGmediated labeling of a synthetic N-terminal peptide, MKHKGSTPEMPVLENR, with a Gln-donor substrate should be conducted, which will provide a clear indication of the actual protein labeling site.

Acknowledgments We are grateful to Ajinomoto Co., Inc. (Japan) for providing us with the MTG sample. This work was supported by the Research for Promoting Technological Seeds from the Japan Science and Technology Agency (JST) of Japan, and also in part by a Grantin-Aid for the Global COE Program, “Science for Future Molecular Systems” from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan.

94

Kamiya and Abe

References 1. O’Hare, H. M., Johnsson, K. and Gautier, A. (2007) Chemical probes shed light on protein function. Curr. Opin. Struct. Biol. 17, 488–494. 2. Sunbul, M. and Yin, J. (2009) Site specific protein labeling by enzymatic posttranslational modification Org. Biomol. Chem. 7, 3361–3371. 3. Sato, H., Hayashi, E., Yamada, N., Yatagai, M., Takahara, Y. (2001) Further studies on the site-specific protein modification by microbial transglutaminase. Bioconjugate Chem. 12, 701–710. 4. Kamiya, N., Tanaka, T., Suzuki, T., Takazawa, T., Takeda, S., Watanabe, K. and Nagamune, T. (2003) S-peptide as a potent peptidyl linker for protein crosslinking by microbial transglutaminase from Streptomyces mobaraensis Bioconjugate Chem. 14, 351–357. 5. Tanaka, T., Kamiya, N., Nagamune, T. (2005) N-terminal glycine-specific protein conjugation catalyzed by microbial transglutaminase FEBS Lett. 579, 2092–2096. 6. Tominaga, J., Kemori, Y., Tanaka, Y., Maruyama, T., Kamiya, N. and Goto, M. (2007) An enzymatic method for site-specific labeling of recombinant proteins with oligonucleotides Chem Commun. 401–403. 7. Josten, A., Meusel, M., Spencer, F., and Haalk, L. (1999) Enzyme immobilization via microbial transglutaminase: a method for the generation of stable sencing surface. J. Mol. Catal. B. Enzymol. 7, 58–66. 8. Tominaga, J., Kamiya, N., Doi, S., Ichinose, H., Maruyama, T. and Goto, M. (2005) Design of a specific peptide tag that affords covalent and site-specific enzyme immobilization catalyzed by microbial transglutaminase Biomacromolecules 6, 2299–2304. 9. Wong, L. S., Khan, F. and Micklefield, J. (2009) Selective Covalent Protein Immobilization: Strategies and Applications Chem. Rev. 109, 4025–4053. 10. Kamiya, N., Abe, H., Goto, M., Tsuji, Y. and Jikuya, H. (2009) Fluorescent substrates for covalent protein labeling catalyzed by microbial transglutaminase Org. Biomol. Chem. 7, 3407–3412.

11. Meusel, M. (2004) Synthesis of hapten-protein conjugates using microbial transglutaminase in Methods in Molecular Biology, Bioconjugation Protocols: Strategies and Methods 283, 109–123. 12. Pasternack, R., Laurent, H.-P., Rüth, T., Kaiser, A., Schön, N. and Fuchsbauer, H.-L. (1997) A fluorescent substrate of transglutaminase for detection and characterization of glutamine acceptor compounds Anal. Biochem. 249, 54–60. 13. Mindt, T. L., Jungi, V., Wyss, S., Friedli, A., Pla, G., Novak-Hofer, I., Grüngerg, J. and Schibli, R. (2008) Modification of different IgG1 antibodies via glutamine and lysine using bacterial and human tissue transglutaminase Bioconjugate Chem., 19, 271–278. 14. Garanger, E., Aikawa, E., Reynolds, F., Weissleder, R. and Josephson, L. (2008) Simplified syntheses of complex multifunctional nanomaterials Chem. Commun. 4792–4794. 15. Tanaka, Y., Tsuruda, Y., Nishi, M., Kamiya, N. and Goto, M. (2007) Exploring enzymatic catalysis at a solid surface: a case study with transglutaminase-mediated protein immobilization Org. Biomol. Chem., 5, 1764–1770. 16. Hirakawa, H., Kamiya, N., Tanaka, T., Nagamune, T. (2005) Intramolecular electron transfer in a cytochrome P450cam system with a site-specific branched structure Protein Eng. Des. Sel., 20, 453–459. 17. Fontana, A., Spolaore, B., Mero, A. and Veronese, F. M. (2008) Site-specific modification and PEGylation of pharmaceutical proteins mediated by transglutaminase Adv. Drug. Deliv. Rev., 60, 13–28. 18. Ando, H., Adachi, M., Umeda, K., Matsuura, A., Nonaka, M., Uchio, R., Tanaka, H. and Motoki, M. (1989) Purification and characteristics of a novel transglutaminase derived from microorganisms. Agric. Biol. Chem., 53, 2613–2617. 19. Folk, J. E. and Cole, P. W. (1965) Structural requirements of specific substrates for guinea pig liver transglutaminase J. Biol. Chem. 240, 2951–2960.

Chapter 8 Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides: Strategies and Methods Anna Mero, Chiara Clementi, Francesco M. Veronese, and Gianfranco Pasut Abstract PEGylation, the covalent linking of PEG chains, has become the leading drug delivery approach for proteins. This technique initiated its first steps almost 40 years ago, and since then, a variety of methods and strategies for protein–polymer coupling have been devised. PEGylation can give a number of relevant advantages to the conjugated protein, such as an important in vivo half-life prolongation, a reduction or an abolishment of immunogenicity, and a reduction of aggregation. Furthermore, the technique has demonstrated a great degree of versatility and efficacy – not only PEG–protein conjugates have reached the commercial marketplace (with nine types of derivatives), but a PEG-aptamer and PEGylated liposomes are now also available. Most of this success is due to the development of several PEGylation strategies and to the large selection of PEGylating agents presently at hand for researchers. Nevertheless, this technique still requires a certain level of familiarity and knowledge in order to achieve a positive outcome for a PEGylation project. To draw general guidelines for conducting PEGylation studies is not always easy or even possible because such experiments often require case-by-case optimization. On the other hand, several common methods can be used as starting examples for the development of tailormade coupling conditions. Therefore, this chapter aims to provide a basic introduction to a wide range of PEGylation procedures for those researchers who may not be familiar with this field. Key words: Poly(ethylene glycol) (PEG), PEGylation, PEG–protein conjugate, Protein modification

1. Introduction PEGylation, the covalent attachment of PEG moieties to a therapeutic agent, was first reported by Abuchowski et al. in the 1970s (1) who demonstrated the usefulness of the strategy to improve the therapeutic value of proteins and peptides. Since then, PEGylation has been long studied in the literature and several conjugates have already reached the marketplace. The most

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_8, © Springer Science+Business Media, LLC 2011

95

96

Mero et al.

prominent effect of PEGylation is a prolonged circulation time for therapeutic conjugates due to a decreased rate of kidney clearance and a reduction of proteolysis. Together, these advantages have lead to lower doses of administration and increased compliance by patients (2, 3). PEG is synthesized by the anionic ring opening polymerization of ethylene oxide initiated by nucleophilic attack of a hydroxide or a methoxide ion to the epoxide ring. The polymer can be obtained with a low polydispersity and a low content of impurities. Several derivatives of PEG that vary in molecular weight (300 Da to 40 kDa), structure (linear or branched), and reactive moiety are currently available from a number of different companies (e.g., Iris biotech, Laysan Bio, NOF, etc.). The choice of an appropriate PEG derivative to use will depend on the particular features of a given protein of interest, such as the primary sequence, chemical reactivity of available functional groups, molecular weight, biological activity, and function. In this chapter, we outline several strategies for carrying out the PEGylation of proteins and peptides, and also present a number of methods for characterizing and purifying PEG–protein conjugates that have been used in our own laboratory or published in the literature. 1.1. Characterization of PEGylating Agents

The development of an efficient and successful PEGylation study requires reliable methods for analyzing PEG reagents and PEG– protein conjugates at various stages of the process. In fact, PEG reagents contribute to a substantial portion of the manufacturing costs associated with PEGylated proteins, and their purity impacts the conjugation efficiency and overall product quality. The determination of the exact molecular weight, polydispersity index, presence of reactive and nonreactive impurities, degree of activation, and the presence of PEG dimers in the raw material must all be carefully evaluated.

1.1.1. NMR Spectroscopy of PEGylating Agents

To evaluate the degree of activation and to detect the presence of impurities (4), PEGs can be analyzed by both 1H- and 13C-NMR spectroscopy (Fig. 1). This technique requires a few milligrams of polymer and a suitable deuterated solvent such as dimethyl sulfoxide (DMSO-d6), chloroform (CDCl3), or water (D2O). Usually, CDCl3 or DMSO are preferred for the analysis of those activated PEGs that can be hydrolyzed in D2O.

1.1.2. Determination of PEG Diol Impurities in Methoxy-PEG Batches

The most utilized PEG for protein modification is methoxy-PEG (mPEG), where only one terminal end of the polymer can be activated while the other is capped with a methoxy group, preventing undesired intra- or intermolecular crosslinking (5). During the synthesis of mPEG, the anionic polymerization of ethylene oxide initiated by CH3O– eventually contains PEG diol

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

97

Fig. 1. NMR spectrum of Boc-PEG-NHS in CDCl3.

secondary by-products due to the presence of trace amounts of water. In fact, OH− formed by water present in the reactor can itself initiate the polymerization process, yielding a chain that can grow at both ends. This process leads to the formation of diol PEG impurities (HO–PEG–OH) with the peculiarity of a double-MW value with respect to the MW of the mPEG batch. Size-exclusion chromatography (SEC) analysis, as described in Subheading 1.3.3 below, is suitable for the determination of such PEG diol species in mPEG batches. 1.1.3. Activation Degree of Amino-Reactive PEGylating Agents

The degree of activation of most amino-reactive PEGs can be determined by using a spectroscopic assay based on the so-called Glycyl-Glycine (Gly-gly) test. In this procedure, the degree of PEG activation is determined by reacting an equimolar amount of Gly-Gly with the activated PEG, followed by performing a Snyder and Sobocinsky colorimetric assay of the unreacted dipeptide (6). This assay uses 2,4,6-trinitrobenzenesulfonic acid (TNBS), which reacts stoichiometrically with primary amino groups in an alkaline medium to give a trinitrophenyl derivative absorbing at 420 nm (Fig. 2).

1.1.4. Activation Degree of Thiol-Reactive PEGylating Agents

To determine the degree of activation of thiol-reactive PEG reagents (e.g., PEG-OPSS, PEG-Mal), the polymer is mixed with an equimolar amount of a suitable molecule containing a free thiol (e.g., cysteine (Cys) or glutathione (GSH)). The presence of unreacted thiol groups is determined by the Ellman assay,

98

Mero et al.

Fig. 2. Reaction between 2,4,6-trinitrobenzensulfonic acid and a free primary amine group.

Fig. 3. Reaction between 5,5¢-dithiobis(2-nitrobenzoic acid) and a free thiol group.

Table 1 Hydrolysis half-lives of PEG-NHS species at pH 8, 25°C PEG-NHS ester

Active groups structure

Half-life (min)

PEG–(CH2)4–CO2–NHS

Succinimidyl valerate (SVA)

33.6

PEG–O–CO2–NHS

Succinimidyl carbonate (SC)

20.4

PEG–O2C–(CH2)3–CO2–NHS

Succinimidyl glutarate (SG)

17.6

PEG–O2C–(CH2)–CO2–NHS

Succinimidyl succinate (SS)

9.8

PEG–O–CH2–CO2–NHS

Succinimidyl carboxymethyl (SCM)

0.75

PEG–O–(CH2)–CO2–NHS

Succinimidyl propionate (SPA)

16.5

which is based on the use of 5,5¢-dithiobis (2-nitrobenzoic acid) (DTNB). In this assay, free thiols react with DTNB at neutral pH to give 2-nitro-5-thiobenzoic acid (TNB), a derivative absorbing at 412 nm (Fig. 3) (7). 1.1.5. Reactivity of NHS-Activated PEGs

The reactivity of N-hydroxysuccinimide (NHS)-activated PEGs can be easily evaluated by following the increase in UV absorbance at 260 nm due to the release of the NHS group when the reaction is performed at room temperature in borate buffer. The reactivity of several representative PEG-NHS species, expressed in terms of the hydrolysis rate, is reported in Table 1 (8). It is worth noting that the hydrolysis rate of such acylating

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

99

PEGs depends upon the nature of the chemical group adjacent to the active ester. Another point to be highlighted is that in all cases, the aminolysis rate is always higher than the hydrolysis rate due to the higher nucleophilicity of free amino groups. 1.1.6. Mass Spectrometric Analyses of PEGylating Agents

PEG molecular weight determination can be rather difficult because of polymer polydispersity. For such determinations, matrix-assisted laser-desorption ionization (MALDI) mass spectrometry (MS) has been the technique most often employed because it produces mainly monocharged ions, thus generating mass spectra of low complexity (9–11). The specific operating conditions will depend upon the particular MALDI-TOF instrument employed, but using an acceleration voltage of 20 kV with linear detection is a suitable general method. The MALDI-TOF mass spectrum of a PEG sample typically shows a monomodal distribution of MW values with the main mass signals spaced apart by Dm/z = 44, in agreement with the monomer unit mass of the oxyethylene unit (44.053 g/mol) (Fig. 4) (12). Electrospray ionization mass spectrometry (ESI-MS), a method widely used for protein characterization, has a strong tendency to form multiply charged ions of the samples, thus hampering the analysis of polydisperse PEG polymers with molecular masses above a few kilodaltons.

1.1.7. Determination of Reactive, Low-Molecular Weight Impurities in PEG Batches

The impurities found in raw PEG preparations can come from either the process of synthesis of the polymer or the activation reaction, while others are formed by the degradation of the polymer (e.g., oxidation) or by cleavage of the chain itself. PEG-aldehyde, for example, is easily oxidized in air and low-molecular weight aldehydic impurities can often be found in the raw product.

Fig. 4. MALDI-TOF mass spectrum of PEG 6,000 Da.

100

Mero et al.

MS analysis, reversed-phase high performance liquid chromatography (RP-HPLC), or NMR spectroscopy are all useful methods to detect the presence of these substances (13). 1.2. Protein PEGylation

The PEGylation of proteins can be achieved either by a direct chemical reaction between an amino acid residue and a suitable PEGylating reagent, or by an enzyme-catalyzed linkage. Brief descriptions of several strategies for conjugating PEG molecules to proteins and peptides are presented below.

1.2.1. Random PEGylation at Free Amino Groups

The primary amino groups of proteins are good nucleophiles, and as such are exploited most frequently for PEG coupling by reaction in mildly basic media (pH 8.0–9.5). Lysines, commonly located on protein surfaces, are relatively abundant and easily accessible to reactive PEG reagents (e.g., activated PEGcarboxylates and PEG-carbonates; see Fig.  5a, b). Random PEGylation at these amino acid residues typically yields mixtures of different isomers and different degrees of modification. To a lesser extent, such reactive PEGylating agents can also target other types of nucleophilic groups found in proteins such as the side chains of serine, threonine, tyrosine, and histidine (14, 15). In this case, it is possible to cleave any unstable bonds by treating the resultant conjugate mixture with hydroxylamine, thus improving the homogeneity of the final PEGylated product (16).

Fig. 5. Coupling of protein amines with activated PEG-carbonates (a1, a2), PEG-carboxylates (b) and PEG-aldehyde (c).

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

101

1.2.2. Protein N-Terminal PEGylation

In order to guarantee a higher degree of homogeneity of the product, it is possible to direct the PEG coupling reaction to take place only at the N terminus of a protein. This selectivity is possible by taking advantage of the different pKa values between the e-NH2 of lysine and the a-NH2 of the N terminus. By lowering the pH of the reaction mixture to ~5–6, all the e-amines in a protein will tend to be protonated whereas the a-NH2 group will still be partially present as a free base available for coupling with activated PEG molecules. This method generally gives optimal results when less reactive PEG-aldehydes are used. In these reactions, an unstable Schiff base is initially obtained, which is in turn reduced to a stable secondary amine (Fig. 5c). Several papers describe the modification of primary amines with PEG-acetaldehyde and later with the more stable PEG-propionaldehyde (17, 18). This conjugation method has been successfully exploited for the preparation of several PEG–protein conjugates; among these Neulasta®, an N-terminal mono-PEGylated granulocyte colony-stimulating factor (G-CSF) has demonstrated therapeutic and marketing success (19).

1.2.3. Thiol PEGylation

Cysteine residues are valuable targets for achieving the site-specific modification of proteins or peptides, and are present in the free form at a relatively low natural abundance level compared to the oxidized cystine species. Nevertheless, cysteines – if present – are often found partially or fully buried within the structure of proteins with limited accessibility to chemical reagents (20). Under appropriate conditions, cysteine residues can be modified selectively, rapidly, quantitatively, and either in a reversible or irreversible fashion (21). Furthermore, thanks to its relatively facile coupling chemistry, there are several examples of the insertion of cysteines by genetic engineering at desired positions in a protein sequence for site-specific conjugation (22). PEG-maleimide (PEG-Mal), PEG-vinyl sulfone (PEG-VS), or PEG-iodo acetamide (PEG-IA) derivatives have been used to obtain stable, irreversible thioether bonds between polymers and proteins (Fig. 6). PEG-orthopyridyl disulfide (PEG-OPSS)

Fig. 6. Thiol-reactive PEGs (a) and conjugation of PEG to a free thiol group (b).

102

Mero et al.

is also widely used, and it forms a disulfide linkage with cysteine. This linkage can be cleaved under reducing conditions, whereas PEG-Mal gives stable conjugates. PEG-IA and PEG-VS are both less reactive and infrequently used, whereas PEG-Mal and PEG-OPSS yield quantitative protein modification. Very recently, an interesting strategy was devised to direct PEGylation to protein disulfide bridges as well; in this case, the disulfide link is firstly reduced and then the resulting free thiols are reacted with a special PEG monosulfone reagent to give a stable three-carbon PEGylated bridge. This procedure, although very promising, is not described here further because the relevant PEG monosulfone is currently unavailable commercially. The reader, however, may refer to the literature for further details (23). 1.2.4. PEGylation to Carboxylic Acid Groups

The direct coupling of PEG-NH2 to activated protein carboxylic groups cannot be easily performed because it typically yields intra- or intermolecular linkages between the protein amines. An original solution uses PEG-hydrazide that is reactive at low pH toward carboxyl groups, but does not react with protein amino groups that are protonated under such conditions (24).

1.2.5. PEGylation of Proteins Modified with Aldehydic and Keto Groups

Aldehydic and keto groups, absent in natural proteins, can be exploited for certain types of nucleophilic additions. Aldehyde functional groups can usually be introduced into proteins by the oxidation of an N-terminal threonine or serine residue using sodium periodate. The introduced aldehydic groups can react with an aminooxy-functionalized PEG chain to obtain a conjugate at the N terminus of the protein (Fig. 7). In some cases, when the N terminus is an amino acid other than threonine or serine, a similarly reactive group can be introduced by metal-catalyzed oxidation, although the conditions for this reaction are potentially more damaging to proteins (25).

1.2.6. Selective PEGylation in Structuring or Denaturing Media

Although proteins typically present a defined structural conformation in aqueous solutions, peptides often have a random coil structure. The presence of organic solvents or chaotropic salts can promote structural rearrangements of the peptide and eventually modify the degree of solvent exposure and reactivity of some residues. As a few examples of selective PEGylation in the presence of organic solvents have been reported in the literature, they cannot be considered to be of general applicability.

Fig. 7. Conjugation of PEG to proteins containing aldehydic groups.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

103

In fact, several parameters such as the solvent, PEG/peptide molar ratio, peptide concentration, and the temperature need to be evaluated on a case-by-case basis. For instance, it was reported that in the presence of 60% (v/v) of dimethylformamide (DMF) at high pH, insulin is selectively modified by a NHS-activated PEG at LysB29 only, even though there are three potential sites for conjugation that are present within the peptide sequence (viz., at GlyA1, PheB1, and LysB29) (26). Moreover, growth hormone releasing factor (GRF), bearing the amino acid residues Lys12 and Lys21 within the N-terminal region, has been conjugated to activated PEGs to yield equimolar amounts of two distinct monoPEGylated isomers (i.e., Lys12- and Lys21-conjugates) (27). However, when the same reaction is conducted in trifluoroethanol (TFE) (50% v/v), it was found that 90% of the reaction mixture is composed of the derivative at Lys21. And finally, the lone Cys17 of G-CSF – buried within a hydrophobic region – could be PEGylated only under mild reversible denaturation conditions that preserved the integrity of the disulfide bridges (28). 1.2.7. Amino PEGylation of Peptides by Reversible Protection

For the selective PEGylation of peptides, a different type of strategy involves the reversible protection of specific residues. This procedure is possible for peptides only because they generally contain just a few nucleophilic groups and are more stable than full-length proteins toward harsh chemical treatments. This method involves three steps: (1) protection by suitable reagents of the residues known to be important for the activity and, eventually, the purification of the desired isomers; (2) PEGylation at the level of the lone unprotected, reactive target residue; and (3) removal of all the protecting groups. This method is not suitable for proteins because the harsh methods employed during the protection and/ or deprotection reactions can negatively affect protein integrity. The protected peptides can be obtained by solid-phase synthesis or by chemical modification. For example, a somatostatin analogue, bearing one terminal a-amino group and one lysine residue, was modified by selective pH-driven tert-butyloxycarbonyl (Boc) protection of the first amino group, followed by PEGylation of the second one and deprotection (29). In addition, GRF and salmon calcitonin, prepared by solid-phase synthesis, were fluorenylmethyloxycarbonyl (Fmoc)-protected at the N terminus and at one of the two internal lysine residues, and then selectively conjugated to PEG (30, 31).

1.2.8. Glutamine Enzymatic PEGylation

As an alternative to chemical conjugation, promising selective methods have been proposed that use enzymes to catalyze the covalent attachment of polymers to proteins. Among these, transglutaminase (TGase)-mediated conjugation has drawn significant attention for its high degree of site-specificity. TGase catalyzes an acyl transfer between the g-carboxamide group of a glutaminyl residue

104

Mero et al.

Fig. 8. PEGylation mediated by transglutaminase (TGase).

(acyl donor) and a primary amine (acyl acceptor). The latter can be selected among a variety of amines, including the e-amino group of lysine or an appropriate PEG derivative bearing an amino group (PEG-NH2) (Fig. 8) (32). Among the various prokaryotic and eukaryotic TGases that have been explored for conjugation applications, the most widely used enzyme is microbial TGase, which has a number of advantageous properties over eukaryotic TGases such as a calcium-independence and lower substrate specificity requirements. These properties conveniently allow for the use of TGase as a biochemical reagent on a large scale for industrial applications. In the area of pharmaceutical biotechnology, several proteins have been selectively modified by TGase; these are recombinant human interleukin-2 (IL-2), G-CSF, and human growth hormone. In these examples, the selective conjugation is due to the TGase active site structure, which is accessible only to those glutamines present within flexible regions of the protein substrate (33). 1.3. PEG-Conjugate Purification

Usually PEG coupling reactions, especially random amino PEGylation, yield a mixture of heterogeneous compounds and, therefore, a purification step is always required. Even in the case of a selective PEGylation reaction, a purification step is still needed to eliminate unreacted proteins, excess amounts of polymer, and various by-products. Several purification strategies may be used depending upon the properties of both the protein and the conjugated PEG moiety. Dialysis or ultrafiltration of the reaction mixture can be used to remove low-molecular weight components or to exchange the solvent. The removal of unreacted PEG polymers, non-conjugated proteins, and the separation of different PEGylated products can be achieved by using more specific and elaborate chromatographic techniques equipped with an online detector. Proteins and their conjugates can usually be monitored by measuring the UV absorbance at l = 214 or 280 nm or by fluorescence detection (lEX = 295 nm, lEM = 310 nm). On the contrary, unreacted PEGs, which are nearly transparent in the UV spectrum, can be monitored by using refractive index (RI) detectors (34) or light scattering techniques. Typically, the amount of PEG polymers present in a sample can be estimated by analyzing the collected fractions with an iodine assay (35).

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

105

In Subheadings 3.3 and 3.4, we describe several protocols that can serve as a useful guideline for the purification and analysis of most types of PEGylated proteins. 1.3.1. Ion Exchange Chromatography

When PEG is coupled to a protein’s amino groups, the resulting conjugate may have an isoelectric point (pI) different from that of the starting native protein. Cation exchange chromatography (CEX) is the method of choice for the separation of PEGylated proteins because it exploits differences in charge at the protein surface (36, 37). Effectively, PEG modifies the elution time of proteins in ion exchange chromatography (IEX) separations, either by coupling to the amines or shielding the charges on the protein surface. Even in those situations where the net charge is unaltered with respect to the starting protein, the presence of the PEG moiety may decrease the interaction between the protein and the chromatographic matrix, thus yielding a shorter elution time for the PEG–protein conjugate compared to the starting protein. In the case of CEX separations, the elution order of PEG–protein conjugates is determined mainly by the number of linked PEG chains: Highly PEGylated molecules tend to elute first, followed by less PEGylated isomers and then by the unreacted protein (38); and any unreacted PEG that does not present positive charges elutes in the column void volume. Given that PEG–proteins tend to interact weakly with IEX matrices, special consideration must be given to the buffer conductivity and pH conditions. For this reason, it is convenient to carry out an extensive dialysis of the PEGylation mixture against the buffer that will be used for the CEX step in order to reach the same ionic strength (as measured by conductometry) and pH value of the column equilibration buffer. In addition, the reaction mixture samples should be diluted and filtered to avoid high back pressure and column fouling. Occasionally, a double-step procedure may be suitable, where a first chromatographic step is performed with a large-particle size resin to eliminate the free polymer, followed by a second step in which a resin with higher resolution is employed. The choice of buffers is a very important step for successfully carrying out IEX separations. Usually the equilibration buffer (A) has a low ionic strength, whereas the elution buffer (B) contains a higher concentration of salt (NaCl). In some cases, a change in pH can be also used between buffers A and B. Small increases in the salt concentration or pH of the buffer solution can effectively reduce the strength of the interactions between PEG–proteins and the resin, causing the conjugates to be eluted from the column before the un-PEGylated proteins. A problem that is commonly encountered in the purification of PEGylated proteins is the low capacity of the pores of conventional IEX media; the effective diameter of these pores can sometimes be too small to allow the penetration of

106

Mero et al.

PEG–protein complexes with a large surface area. Consequently, the desired conjugate products can be lost in the column flow-through. This is not generally considered to be an issue in analytical-scale experiments where the sample loading is relatively low (around 1  mg of protein/mL of sample); for preparative-scale purifications, however, a higher loading (e.g., >6 mg/mL) of the PEG– protein mixture may become a limiting factor. Conventional strong cation exchange resins (e.g., Mono S™ (sulfo) or Mono SP™ (sulfopropyl) sepharose (GE Healthcare)) that are usually employed with standard fast protein liquid chromatography (FPLC) or HPLC systems can be loaded at a protein concentration of 5–10  mg/mL at a linear flow rate of 60  cm/h. MacroCap™ SP (GE Healthcare) is a new type of strong cation exchange medium especially designed for the purification of PEGylated proteins and other large biomolecules under high sample loading conditions (12–15  mg/mL). The matrix is highly porous and provides good mass transfer characteristics and improved accessibility to the internal surface area for the adsorption of large molecules. PEGylated proteins can be also purified by anionic exchange chromatography (AEX) (40). In this case, the PEG–protein conjugate is bound to the column at a higher pH value than its isoelectric point (pI), which results in a negative net charge on the molecule. Once again, the charges at the surface of the protein can be shielded by the presence of PEG. As in CEX, it is often convenient to use a linear salt or pH gradient to elute the PEGylated derivatives from the column. 1.3.2. Reversed-Phase Liquid Chromatography

Reversed-phase high performance chromatography (RP-HPLC) is often used for the characterization of PEGylated species due to its high resolution and the possibility of coupling the technique with an online mass spectrometer detector (41). Although the method is very rapid, only a limited amount of sample material can be loaded at once. Furthermore, PEGylated proteins often give wide peaks due to polymer polydispersity, which compromises the resolution of PEGylated species to different extents. The elution conditions employed in RP-HPLC generally require high percentages of acetonitrile (CH3CN) or methanol (MeOH), which can be a concern for protein stability. Other parameters such as the column temperature, the elution gradient profile, and the mobile phase composition must all be carefully evaluated on a case-by-case approach.

1.3.3. Size-Exclusion Chromatography

Size-exclusion chromatography (SEC) separates molecules based on differences in their hydrodynamic volumes. Since PEGylation is often performed with the aim of increasing the size of proteins, it is clear that SEC can be a useful technique for purifying PEG– protein conjugates. SEC, however, has several limitations: it generally gives broad peaks with poor resolution for PEG-conjugates; it

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

107

is a low-throughput technique with relatively high costs, and thus has limited applicability in large-scale processes; and it also cannot separate positional isomers (which have the same mass and hydrodynamic volume). Furthermore, in those cases where there are only small differences in size, unreacted PEG and protein molecules can be co-eluted with monoPEGylated species. Conversely, SEC is a useful method for removing low-molecular weight impurities (e.g., by-products formed by the hydrolysis of functionalized PEG, buffer salts, solvents, and other low-molecular mass reagents), and also serves as an effective tool for obtaining an initial evaluation of the degree of protein modification as well as for determining the presence of aggregates. Typically, dextran- and agarose-based SEC columns are often used in conjunction with FPLC systems, with Sephadex G-75, G-50, and Superose 6 or 12 (42) being the most commonly used media to purify proteins. The choice of a particular column type depends on the molecular weight of the protein. Generally, buffers with a low percentage of organic solvents are used to minimize the occurrence of hydrophobic interactions between proteins in the sample mixture and the column matrix (43–45). One important factor in the use of SEC for PEGylation applications is to take into account that the apparent size of a PEG–protein conjugate is roughly five to ten times larger than that of a globular protein with the same nominal molecular weight. 1.3.4. Ultrafiltration/ Diafiltration

One particularly important nonchromatographic step in the production of pharmaceutical proteins is the ultrafiltration/diafiltration operation. Ultrafiltration/diafiltration are effective processes for exchanging buffers between chromatographic steps and for concentrating the conjugate products to achieve the desired final concentration. Usually, membrane filters with the same nominal molecular weight cutoff value as those used for the native (unmodified) protein should also be employed to concentrate monoPEGylated-protein samples since such types of PEGylated products can potentially escape by the so-called “snake effect” through membranes with larger porosities, causing substantial losses of product (46–48). In the case of branched PEG or multiPEGylated conjugates, larger cutoffs may be used, since in this case the “snake effect” of PEGs is not as likely to be an issue. Regenerated cellulose and polyethersulfone membranes, with their low protein binding and high protein retention characteristics, are the most commonly used filter media and also provide the highest rates of recovery for purified proteins (48). However, polyethersulfone filters are much less efficient for processing PEGylated protein species because of the hydrophobic nature of their membrane surfaces. Indeed, the full application potential of membrane-based technologies for PEGylated products is somewhat limited because the increased size, greater hydrophobicity,

108

Mero et al.

and lower electrostatic interactions of PEG-conjugates with respect to unmodified proteins tend to lead to increased fouling, i.e., the largely irreversible adsorption and/or deposition of proteins on and within membrane filter media (49). 1.4. PEG-Conjugate Characterization

The first issue to be faced with the characterization of PEGconjugates deals with the accurate determination of the amount of attached PEG. Several methods are available such as colorimetric assays, SEC, electrophoresis, and mass spectrometry (MS). However, these methods are not suitable for separating and identifying positional isomers of PEG; for these purposes, IEX is generally better suited. The identification of PEGylation sites within proteins is based mainly on the use of standard protein sequence analysis methods, with peptide mapping and mass spectrometry both falling within the scope of such types of useful techniques. The simplest and most rapid methods available for conducting preliminary characterizations of a PEG–protein conjugate are colorimetric assays. In the case of amino-PEGylation products, the Habeeb assay can be used to quantify the amount of unreacted protein primary amines with TNBS reagent, thus allowing one to indirectly calculate the number of bound PEG chains (50). In a similar approach, Ellman’s assay can be used to determine the presence of any remaining free cysteines following PEG–thiol conjugation (51). Finally, the iodine assay, based on the noncovalent interaction of iodine with the PEG backbone, can be used to obtain both qualitative and quantitative information about the polymer (35). The preceding methods are all described in detail in the following sections. To characterize the protein conformation in the conjugates, various spectroscopic methods may be used. In particular, circular dichroism, fluorescence, and UV spectra can all be applied to analyze PEG–protein conjugates – thanks to the fact that PEG is transparent to light at UV–visible wavelengths. For the interpretation of such spectra, the reader is referred to dedicated books on protein characterization (52).

1.4.1. Bicinchoninic Acid Assay

The protein concentration in a non-PEGylated sample can usually be determined by simply measuring the spectrophotometric absorbance of the aromatic amino acids residues at l = 280 nm. This approach may also be acceptable for measuring the concentration of a PEG–protein conjugate sample after first confirming that the absorbance of the protein is not altered by PEG coupling. Alternatively, a colorimetric assay can be carried out to estimate the protein concentration. The bicinchoninic acid (BCA) assay, for example, is the most commonly used technique because it is less affected than other types of dye-binding assays by the presence of PEG. BCA protein assay kits are commercially available from a number of different suppliers (e.g., Sigma-Aldrich, Bio-Rad,

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

109

Thermo Fisher, etc.); in these assays, the quantitative determination of the protein concentration in a PEGylated sample relies on the generation of a calibration curve using standard solutions of the native protein (53). 1.4.2. Ion Exchange Chromatography

IEX, as previously discussed in Subheading  1.3.1, is the most widely used technique for the fractionation and purification of PEGylated proteins on a preparative scale. IEX is also very useful for analytical purposes because it allows the efficient separation of positional isomers (54). In particular, our laboratory has found that analytical strong cation exchange columns (e.g., TSKgel SP-5PW, 7.5  mm × 7.5  cm (Tosoh); and HiLoad™ 16/10 SP Sepharose™ HP (GE Healthcare)) can provide good results for PEGylated samples.

1.4.3. Reversed-Phase High Performance Liquid Chromatography

RP-HPLC is a useful technique for the determination of purity and species content of PEGylated protein samples. This fractionation method is based on differences in hydrophobicity of the native and PEGylated proteins. PEG is an amphiphilic molecule, and thus PEGylated proteins often exhibit higher retention times on RP-HPLC columns than their un-PEGylated counterparts. The increase is generally dependent on the length and mass of the conjugated polymer (37). RP-HPLC is not only an excellent and robust tool for the fractionation of PEGylated and un-PEGylated species, but is also a useful method to detect protein oxidation, deamidation, or cleavage of the protein backbone (55). In addition, the technique can also be used for the high-resolution analysis of protein fragments and peptides (i.e., peptide fingerprinting) (19, 43), or to separate positional isomers (53) in the case of peptide PEGylation. In analytical RP-HPLC applications, the PEG/protein ratio appears to be the predominant factor affecting the resolution of PEGylated conjugates (37). Typically, reversed-phase columns containing packings such as butyl (C4) or octadecyl (C18) can be employed for the fractionation of PEGylated protein using an elution gradient of H2O/CH3CN (either with or without trifluoroacetic acid (TFA)).

1.4.4. Size-Exclusion Chromatography

SEC, or gel filtration, is widely used to estimate the molecular weight (MW) of native (unmodified) proteins through the use of a standard calibration curve. As PEGylated proteins show a larger hydrodynamic volume than native proteins of the same nominal MW, SEC (and, similarly, sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)) cannot provide an accurate determination of the exact MW of PEG–protein conjugates, but can only be used to monitor a PEGylation reaction and characterize the homogeneity of the conjugate product. An interesting and useful discussion on the effect of PEG on the apparent size of conjugates and their behavior in SEC has recently appeared in the literature (57).

110

Mero et al.

Several brands of gel filtration columns are commercially available in the marketplace such as Superose 6 or 12 (GE Healthcare), TSK-Gel (Tosoh), Zorbax GF-250 (Agilent), and BioSep SEC (Phenomenex). Each type of column has a specified MW fractionation range, and hence the selection of the most appropriate column to use must be done on a case-by-case. Typically, phosphate and HEPES buffers containing a salt (NaCl) gradient are the most often used eluents in SEC applications. The addition of water-miscible organic solvents (e.g., acetonitrile or isopropanol) into the mobile phase can generally improve the overall separation of PEG–protein conjugate mixtures by increasing peak sharpness and reducing peak tailing due to nonspecific adhesion of PEGs to the stationary phase (43–45). SEC-HPLC is also useful for the determination of free PEG in PEGylated samples, but since free PEG itself does not absorb at UV wavelengths, it is necessary to use a refractive index (RI) detector coupled with a second UV detector to reveal simultaneously the protein fractions, PEG, and other low molecular weight by-products (e.g., NHS) that emerge from the column. It should be noted, however, that a RI detector cannot be used for analyses requiring a gradient elution because the changes in the eluent composition modifies the baseline. 1.4.5. Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis

SDS-PAGE is a highly useful technique for determining the purity and molecular weight (MW) of native (unmodified) protein samples. However, SDS-PAGE is not suitable as a direct method to easily evaluate the MW of PEGylated proteins because proper calibration standards are typically unavailable for such analyses. During electrophoresis, the migration rate of PEG-conjugates through the porous gel matrix is significantly slowed by the long and heavily hydrated PEG chains. Consequently, PEGylated proteins usually display apparent MW values on SDS-gels that do not correlate with that of the free protein MW standards (54). Generally, SDS-PAGE can only be used to qualitatively follow the progress of a PEGylation reaction and the subsequent purification procedures. For example, a significant shift in the SDS-PAGE band position for a protein after PEGylation would provide evidence in support of a success PEGylation experiment. Random multi-PEGylation will typically yield several bands on an SDSPAGE gel, with each band migrating slower than that corresponding to the native protein; on the contrary, only a single new band should be expected to appear for monoPEGylated derivatives. In some cases, using PEG molecules of different MWs instead of native protein standards may enable a rough estimation of the apparent MW of the conjugates (58, 59); however, it is then necessary to use a specific iodine staining procedure to reveal the PEG bands since PEG is negative toward conventional protein stains. SDS-PAGE gels of varying degrees of crosslinking

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

111

can usually be employed for such analyses, but attention must be paid to the full range of MWs actually present in a particular PEGylated sample in order to accommodate the migration of all the components in the mixture. 1.4.6. Mass Spectrometry of Conjugates

To evaluate the molecular weight of PEGylated proteins, mass spectrometry (MS) analyses are highly recommended. Typically, MALDI-MS is employed for such determinations, even if these types of spectra are sometimes complicated by PEG polydispersity. For PEGylated proteins, the total MW is easily calculated as the sum of the MW of the native protein and the MW of the conjugated PEG chains (60). Nevertheless, the MS technique does present a few limits, such as (1) poor ionization efficiency with large polymers, which can affect the detection sensitivity as well as the accurate determination of the average MW and (2) degradation of the proteins during sample extraction and evaporation.

1.4.7. General Method for Determination of PEGylation Sites for Proteins Modified with Polydisperse PEGs

The method used for the localization of the sites of PEG conjugation follows the strategy commonly employed for conventional protein sequence determinations. This approach is based on the proteolytic digestion of the native (unmodified) protein and the conjugated protein, followed by a comparison of the two elution patterns obtained by RP-HPLC. Subsequent analysis of each peak in the RP-HPLC fingerprint by ESI-MS or MALDI-MS allows the determination of peptide composition (61, 62) and the site of amino acid PEGylation. The determination is carried out by comparison of the fingerprint of the native protein and that of the conjugate. The peptides that are missing in the conjugate elution pattern represent those sequences that contain the polymer. Typically, trypsin is the most commonly employed proteolytic enzyme; however, other proteolytic enzymes with different digestion specificities may also be used depending on the sequence of the protein under investigation. This is necessary in those cases where the PEGylated peptide fragments contain more than one potential site of PEGylation.

1.4.8. Simplified Method for Determination of PEGylation Sites for Proteins Modified with Monodisperse PEGs

The method described above in Subheading 1.4.8 is an “indirect” analysis because, due to the polydispersity of the polymer, the PEGylated peptide is identified by the disappearance of the corresponding fragment. A simplified procedure for the identification of PEGylation sites can be applied if a monodisperse PEG is used for conjugation. Monodisperse PEG polymers, only recently available in the marketplace, are detectable by ESI-MS, as well. Unfortunately, monodisperse PEGs are available only in low MW forms, making them unsuitable for the half-life prolongation of proteins; however, they can be useful for characterization purposes. A monodisperse PEG was recently employed by our group for the specific modification of several proteins of pharmaceutical

112

Mero et al.

interest mediated by the enzyme TGase (63). ESI-MS and tandem mass spectrometry (MS/MS) of the digested peptides allowed for a “direct” (as opposed to indirect, as above) characterization of the peptide and the identification of the PEGylated site even if more than one is present within the same peptide. It is important to note that both the high-MW polydisperse PEG and the low-MW monodisperse form of the polymer are linked to the same site in the protein sequence since the specificity of the conjugation site is dictated by the TGase enzyme (31, 61). Generally, the analysis procedure requires only a very small amount of sample material due to the possibility of determining both composition and sequence in a single analysis.

2. Materials 2.1. Characterization of PEGylating Agents

1. PEG reagents (see Note 1). 2. Deuterated DMSO-d6, CDCl3, or water (D2O). 3. Borate buffer, pH 8: 0.2 M borate buffer, pH 8. 4. Borate buffer, pH 9.3: 0.1 M borate buffer, pH 9.3. 5. Gly-Gly solution (2 mM): Dissolve 28.5 mg of glycyl-glycine in 100 mL of 0.2 M borate buffer, pH 8. 6. TNBS (1% w/v solution in DMF). 7. Phosphate-ethylenediaminetetraacetic acid (EDTA) buffer: 0.1 M Sodium phosphate buffer, 1 mM EDTA, pH 7. 8. Cys or GSH solutions (2 mM): Dissolve 2.61 mg of Cys or 6.62 mg of GSH in 10 mL of phosphate-EDTA buffer, pH 7. 9. Ellman’s reagent (10 mM): Dissolve 4 mg of DTNB in 1 mL of 0.1 M phosphate-EDTA buffer, pH 7. 10. MALDI-TOF MS matrix: Sinapinic acid (58), dihydroxybenzoic acid (65), or a-cyano-4-hydroxycinnamic acid (46) mixed with acetonitrile/water (1:1, v/v).

2.2. Protein PEGylation

1. Glycine solution: Use 200 mL of glycine (250 mM) in water, pH 7.4 for each milliliter of the reaction mixture. 2. NaCNBH3 solution: Prepare 20 mM NaCNBH3 in the same buffer as that used in the reaction mixture. 3. Hydroxylamine solution: Prepare 2 mL NH2OH, pH 7.3 in water.

2.3. Conjugate Purification

1. CEX-buffer A: 10 mM phosphate buffer, 10 mM NaCl, pH 4.7. 2. CEX-buffer B: 10 mM phosphate buffer, 0.1 mM NaCl, pH 4.7. 3. RP-buffer A: 0.05% (v/v) TFA in water. 4. RP-buffer B: 0.05% (v/v) TFA in acetonitrile (CH3CN).

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

2.4. Conjugate Characterization

113

1. BCA Protein Assay Kit (Pierce): To prepare the BCA working solution, mix 50 parts of reagent A (containing sodium carbonate, sodium bicarbonate, BCA, and sodium tartrate in 0.1 M sodium hydroxide) with one part of reagent B (containing 4% CuSO4·5H2O) according to the manufacturer’s instructions. 2. BaCl2 solution: 5% (w/v) barium chloride in 1N HCl. 3. Iodine solution: Dissolve 1.27 g of I2 in 100 mL of 2% (w/v) KI in water. 4. Phosphate buffer: 20 mM phosphate buffer, pH 7.2. 5. Bicarbonate buffer: 4% (w/v) NaHCO3 buffer, pH 8.5. 6. Tris-Gdn buffer: 6 M guanidine hydrochloride, 50 mM Tris– HCl, pH 9.0. 7. PepClean™ C-18 spin columns (Pierce, Rockford, IL).

3. Methods 3.1. Characterization of PEGylating Agents 3.1.1. NMR Spectroscopy of PEGylating Agents

3.1.2. Analysis of PEG Diol Content in mPEG Batches

1. Dissolve the PEG sample (10–20 mg) (see Note 1) in 0.75 mL of deuterated solvent and perform the NMR analysis according to the instrument manufacturer’s instructions. In the 1H NMR spectrum, the integral values of the reactive group signals can be compared with the integral values of the backbone chain signals (–CH2–CH2–; e.g., PEG 5 kDa, 3.6 ppm, 491H) or with other characteristic signals of the polymer. For example, Fig.  1 shows the NMR spectrum of commercial BocPEG-NHS (5  kDa), where the H signals of the Boc group (–C(CH3)3, 1.4 ppm, 9H) are compared with those arising from the N-hydroxysuccinimide group (–NHS) (–CH2–CH2, 1.2 ppm, 4H) and the backbone chain signals. In this case, the analysis of the integrals indicates that the polymer is activated with NHS at only 60% of the maximum level, since the signal integration value is 2.4 instead of the expected value of 4. 1. Equip an HPLC system with a suitable SEC column (see Subheading  1.4.4 for a discussion on column selection). Equilibrate the column with the desired elution buffer. 2. Solubilize the PEG sample (0.2  mM) in 1  mL of elution buffer. 3. Load 20 mL of the sample solution onto the SEC column. 4. PEG does not absorb at wavelengths suitable for UV–visible detection and can be revealed by employing a refractive index detector, instead.

114

Mero et al.

Table 2 Preparation of test solutions for the TNBS assay Blank

PEG reaction mixture

Gly-Gly standard solution

20 mL of TNBS

20 mL of TNBS

20 mL of TNBS

980 mL of Borate pH 9.3

955 mL of borate pH 9.3

955 mL of borate pH 9.3

25 mL of Sample A

25 mL of Sample G

3.1.3. Evaluation of the Degree of Activation of Amino-Reactive PEGylating Agents

1. Prepare Sample A (PEG reaction mixture) as follows: To 1 mL of 2 mM Gly-Gly solution, add 1 eq. (2 mmol) of the activated PEG polymer (e.g., PEG-NHS). The required amount of activated PEG to add depends on its MW; for example, if the MW of PEG-NHS is 5 kDa, then 10.8 mg of the activated polymer should be added to 1 mL of 2 mM Gly-Gly solution. 2. Let the mixture react for 30 min at room temperature under continuous agitation. 3. Prepare 1 mL of 2 mM Gly-gly solution (sample G) as control. 4. Perform the TNBS assay in duplicate at room temperature according to Table 2. 5. Incubate the reactions for 30 min. 6. Read the absorbance at l = 420  nm using a UV–visible spectrophotometer.



The percentage of activation of the amino-reactive PEGs is calculated by using the following formula: % Activation = [1 - (AbsA - AbsB) / (AbsG - AbsB)] ´ 100% AbsA = Absorbance of PEG reaction mixture AbsG = Absorbance of Gly-Gly standard solution AbsB = Absorbance of the blank solution.

3.1.4. Evaluation of the Degree of Activation of Thiol-Reactive PEGylating Agents

1. Prepare Sample A (PEG reaction mixture) as follows: To 1 mL of 2 mM Cys or GSH solution, add 1 eq. of the thiolreactive PEG molecule. The required amount of thiolreactive PEG to add depends on its MW; for example, if the MW of the PEG is 5 kDa, then 10.2 mg of polymer should be added to 1 mL of 2 mM Cys or GSH solution. 2. Let the mixture to react for 30  min at room temperature under continuous agitation. 3. Prepare 1 mL of 2 mM Cys or GSH (Sample C) as control. 4. Perform Ellman’s assay in duplicate at room temperature according to Table 3. 5. Incubate the reactions for 15 min. 6. Read the absorbance at l = 412  nm using a UV–visible spectrophotometer.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

115

Table 3 Preparation of test solutions for Ellman’s assay Blank

PEG reaction mixture

Cysteine standard solution

50 mL of Ellman’s reagent

50 mL of Ellman’s reagent

50 mL of Ellman’s reagent

1 mL of phosphateEDTA pH 7

970 mL of phosphateEDTA pH 7

970 mL of phosphateEDTA pH 7

30 mL of Sample A

30 mL of Cys or GSH solution

The percentage of activation of thiol-reactive PEGs is calculated by using the following formula:

% Activation = [1 - (AbsA - AbsB) / (AbsC - AbsB)] ´ 100% AbsA = Absorbance of PEG reaction mixture AbsC = Absorbance of cysteine standard solution AbsB = Absorbance of the blank solution.

3.1.5. Half-Life Measurement of NHSActivated PEGs

1. Prepare a solution of NHS-activated PEG (0.2–0.5  mM) in dioxane. 2. Add 50 mL of the PEG-NHS solution to 950 mL of 0.2 M borate buffer (pH 8.0) and immediately read the absorbance at 280 nm every 5 s until a plateau is reached. To evaluate the aminolysis rate, add 50 mL of the PEG-NHS polymer solution to 950 mL of a Gly-Gly solution (Gly-Gly/PEG-NHS polymer, 1:1 molar ratio) in 0.2 M borate buffer (pH 8.0). 3. The absorbance (l = 280 nm) at time point = 0 s is taken to be the blank and corresponds to the NHS already present in the mixture. 4. Plot the resulting absorbance data and determine the half-life of the PEG-NHS reagent.

3.1.6. Mass Spectrometric Analyses of PEGylating Agents

1. Dissolve PEG (20 mg) in 0.1% (v/v) TFA in water (5–10 mL). 2. Mix one volume of saturated matrix solution with one volume of the PEG solution. 3. Load the matrix/PEG mixture (10–20 mL) onto the MALDI sample plate and perform the MS analysis according to the instrument manufacturer’s instructions.

3.1.7. Analysis of Low-Molecular Weight Impurities of PEG-Aldehyde

1. Dissolve PEG-aldehyde (0.25–1 mM) in water. 2. At determined time points, withdraw 100  mL and treat the sample aliquot with 2,4-dinitrophenylhydrazine (DNPH/ PEG-aldehyde, 1:1 eq.).

116

Mero et al.

3. Fractionate the resulting reaction mixture by RP-HPLC on a C8 column (100 × 2.1 mm; flow rate = 0.8 mL/min; mobile phase = H2O/CH3CN/H3PO4, 1:1:1). 4. Collect the fractions, lyophilize, and analyze by RP-HPLC on a C18 column (flow rate = 0.2 mL/min; mobile phase = H2O/ CH3CN) using an ESI-MS detector operating in the negative ion mode to identify low-MW impurities. 5. As an alternative to the above procedure (Steps 1–4), dissolve the PEG-aldehyde in MeOH and infuse the sample into an ion mobility/quadrupole/time-of-flight mass spectrometer at a flow rate of 5 mL/min to analyze for the presence of truncated PEG-aldehyde molecules. 1. Prepare the protein solution (1–5 mg/mL) in 0.1 M borate buffer, pH 8.0–9.0.

3.2. Protein PEGylation 3.2.1. Random PEGylation at Free Amino Groups of a Protein

2. Determine the exact concentration of the protein by UV absorption using its molar extinction coefficient (66). 3. Add the amino-reactive PEG – in small amounts – to the protein solution under gentle stirring. An excess of activated PEG is usually required. The optimum ratio of PEG to each protein amino group may range from 1 to 10, depending on the particular PEG and the reactivity of the amino groups on the protein. Table 4 lists some examples of protein conjugation experiments that have reported in the literature using different ratios of PEG/protein, PEGs with different MWs and different protein concentrations. 4. Incubate the reaction mixture at room temperature for 1–5 h. 5. Quench the reaction with glycine solution and stir for 1 h. 6. Eventually add 1 mL of hydroxylamine solution and stir for 30 min.

Table 4 PEGylation reaction conditions and yields as reported in the literature for different molar ratios of NHS-activated PEG to protein NH2 groups Reaction conditions

Protein concentration Protein MW PEG MW PEG/protein (mg/mL) (kDa) (kDa) molar ratio Yield (%) References

Aqueous, pH 8.5

10

21

5

1–3

85

(42)

Aqueous, pH 8.5

4

23

5

5

50

(67)

Aqueous, pH 8.5

1.5

13.7

5

2.5

55

(67)

Aqueous, pH 8.5

6

10

3

50

(68)

Aqueous, pH 8.5

2

17.3

5

8

60

(42)

Aqueous, pH 9

5

19.3

40

3

60

(69)

141

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

117

7. Dialyze the resulting solution to eliminate low-molecular weight products. 8. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4. 3.2.2. PEGylation at the N Terminus of a Protein Using PEG-Aldehyde

1. Prepare a protein solution at 1–5 mg/mL (as determined by absorbance spectrophotometry) in a buffer with pH ~5.0–6.0. 2. Add PEG-aldehyde (see Note 2) to the protein solution at the desired molar ratio. Depending on the protein properties, a great excess of PEG-aldehyde may be needed. It is advisable to test several different PEG/protein molar ratios to optimize the reaction (e.g., 10–50 eq. with respect to the amount of protein molecules). 3. Incubate the reaction mixture for 1 h, and then add NaCNBH3 solution (20 mM) (add 50 eq. of NaCNBH3 per 1 eq. of PEG). 4. Incubate the reaction mixture further at 4°C for 24 h under gentle stirring. 5. Quench the reaction with glycine solution and stir for 1 h. 6. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4.

3.2.3. Thiol PEGylation

1. Prepare a protein solution of 1–5 mg/mL in 0.1 M phosphateEDTA buffer, pH 7.2. 2. Add PEG-OPSS (see Note 3) or PEG-maleimide PEG-MAL (see Note 4) to the protein solution at a molar ratio 1:1 or 2:1 with respect to the amount of free thiols present. If PEG-VS is used, an excess of 2–10 eq. is recommended. 3. Incubate the reaction mixture at 4°C for 4–24 h (depending on the PEG derivatives used) under gentle stirring. 4. Monitor the disappearance of the free thiols in the reaction mixture by Ellman’s assay (see Subheading 3.4.7). 5. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4.

3.2.4. PEGylation at Protein Carboxylic Groups

1. Prepare a protein solution at 1–5  mg/mL in 0.1  M phosphate buffer, pH 4.0–5.0 (see Note 5); 2. Add 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (2–5 eq. with respect to the amount of protein carboxylic acid groups) and PEG-hydrazide (10–50 eq. with respect to the amount of protein molecules).

118

Mero et al.

3. Incubate the reaction mixture at 4°C for 24 h under gentle stirring. 4. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4. 3.2.5. PEGylation of Proteins Containing Introduced Aldehydic Groups

1. Prepare a solution of protein containing N-terminal serine or threonine residues at 1–5 mg/mL in 1% (w/v) ammonium bicarbonate buffer, pH 8. 2. Add a tenfold molar excess of sodium periodate for 10 min. 3. Add a 2,000-fold molar excess of ethylene glycol to stop the oxidation reaction, and then dialyze the mixture against an acidic buffer solution. 4. Add aminooxy-PEG (10–50  eq.) and adjust the pH of the reaction mixture to 3.6. 5. Incubate the reaction mixture for 20 h at room temperature. 6. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4.

3.2.6. Selective PEGylation of Peptides in Structuring or Denaturing Media

1. Dissolve the peptide (2 mg/mL) in a suitable mixture of H2O and a water-miscible organic solvent (e.g., H2O/DMF, 2:3 v/v; or H2O/TFE, 1:1 v/v); 2. Bring the pH of the peptide solution to 9–10. 3. Add PEG-NHS (1–3 eq. with respect to the peptide) to the peptide solution. 4. Incubate the reaction mixture for 30 min under gentle stirring. 5. Quench the reaction mixture with glycine solution and continue stirring for 1 h. 6. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4.

3.2.7. Amino PEGylation of Peptides by Reversible Protection

1. Protect the most reactive amine groups of the peptide with a suitable procedure (e.g., using t-Boc or Fmoc groups); 2. Purify the desired products by chromatography (e.g., RP-HPLC). 3. Dissolve the protected peptide in DMF, DMSO, or other solvent at a final concentration of 5–10 mg/mL. 4. Add an activated amino-reactive PEG at an excess of 2–10 eq. over the peptide. 5. Incubate the reaction mixture at room temperature for 4 h under gentle stirring. 6. Quench the reaction with glycine solution and continue stirring for 1 h.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

119

7. Dialyze the solution against water to remove the organic solvent and lyophilize. 8. Deprotect the conjugate using a suitable method; for example, Boc removal can be carried out by dissolution in TFA, while the Fmoc group can be cleaved in 20% (v/v) piperidine in DMF. 9. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4. 3.2.8. Enzymatic PEGylation Using Microbial Transglutaminase

1. Prepare a protein solution at 1–5  mg/mL in 0.1  M phosphate buffer, pH 7.0. 2. Add an excess of 2–10 eq. of PEG-NH2 with respect to the amount of protein. 3. Add TGase at an enzyme:substrate ratio of 1:75 (w/w). 4. Incubate the reaction mixture at room temperature for 4 h under gentle stirring. 5. Quench the reaction with a few drops of acetic acid at pH 3. 6. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4.

3.3. Conjugate Purification

3.3.1. Ultrafiltration/ Diafiltration

After quenching the PEGylation reaction mixture, dialysis is typically performed using regenerated cellulose membranes against buffers with low ionic strength at 4°C (see Note 6) for 2 days with continuous stirring. As an alternative to dialysis, ultrafiltration/diafiltration can also be used to exchange the buffer components of the reaction. For both procedures, it is always important to verify that the protein remains stable during all the processing steps. After dialysis or ultrafiltration/diafiltration, the removal of unreacted PEG, unreacted protein, and the separation of the different PEGylated species can be achieved by using several different chromatographic techniques, as described in the following sections. 1. Equilibrate the membrane filter in isopropanol for 45 min to remove any wetting/storage agents. Following this, wash the membrane with water and equilibrate it in the same buffer as that used in the PEG–protein solution mixture. 2. Add the PEG–protein solution and allow the system to equilibrate before pressurizing. 3. Fill the stirring cell with the feed solution and connect it to a reservoir containing pure buffer. 4. Pressurize the system with air and monitor the filtrate flux over time; use small adjustments of the pressure to maintain a constant flux. Samples can be removed periodically from both the collected filtrate and the stirred cell to analyze the solute concentration by SEC-HPLC in order to detect any loss of protein.

120

Mero et al.

3.3.2. Cationic Exchange Chromatography

1. Equilibrate a cation ion exchange column with CEX-buffer A (see Note 7) according to the manufacturer’s instructions. For a TSKgel SP-5PW column (21.5 mm × 15 cm, 10 mm), it is recommended to use a flow rate of 5–8 mL/min. 2. Load the dialyzed reaction mixture onto the column. 3. Elute the PEGylated products by slowly increasing the elution gradient with CEX-buffer B. 4. Analyze the eluted fractions for the presence of PEG by performing an iodine assay, and for the presence of protein by monitoring the UV absorption. Collect and pool the fractions containing the PEGylated products. 5. Concentrate and exchange the buffer against CEX-buffer A. Keep the solution at 4°C for short-term storage or at −20°C for long-term storage (see Note 8).

3.3.3. Reversed Phase HPLC

1. Connect C4 (or C18) RP-HPLC column to the HPLC system. 2. Equilibrate the C4 column (250 × 21.1  mm, 10  mm) with RP-buffer A at a flow rate of 8 mL/min. 3. A column temperature of ~45°C is recommended for running the RP-HPLC procedure. 4. Load the dialyzed reaction mixture onto the column at 5–10 mg/mL (total protein concentration). 5. Elute the column initially with RP-buffer A. 6. Continue the elution using a moderately shallow gradient (1–2% per min) with RP-buffer B (see Note 9). 7. Collect and pool the fractions containing protein (as detected by monitoring the UV absorbance). Check for the presence of PEG by performing an iodine assay. 8. Concentrate and exchange the buffer against an appropriate saline solution and keep at 4°C for short-term storage or −20°C for long-term storage (see Note 8).

3.3.4. Size-Exclusion Chromatography

1. Connect a suitable SEC column to the HPLC system. 2. Equilibrate the SEC column with an appropriate saline buffer for 1  h at a flow rate of 1  mL/min (for a 10 × 300  mm column). 3. Load the dialyzed reaction mixture at 5–10  mg/mL (total protein concentration). 4. Elute the product with the same buffer used for column equilibration. 5. Collect and pool the fractions containing protein (as detected by monitoring the UV absorbance). Check for the presence of PEG by performing an iodine assay.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

121

6. Concentrate the product by ultrafiltration and keep at 4°C for short-term storage or −20°C for long-term storage (see Note 8). 3.4. Conjugate Characterization 3.4.1. BCA Protein Assay

1. Prepare a series of unmodified (i.e., non-PEGylated) protein samples with known protein concentrations (0.2–1.2  mg/ mL) at a final volume of 50 mL. 2. Prepare triplicate samples (50 mL each) of the PEGylated protein with unknown concentrations. 3. To each sample, add 1 mL of BCA working reagent. 4. Incubate the samples for 30 min at 37°C and then read the absorbance at l = 562 nm. 5. Prepare a standard curve by plotting the measured absorbance values versus protein concentration. Using the standard curve, determine the protein concentration of the PEGylated samples.

3.4.2. Ion Exchange Chromatography

1. Pre-equilibrate the column (7.5 × 75 cm, 5-mm particle size) with CEX-buffer A at a flow rate of 1 mL/min. 2. Load 200  mL of the PEG–protein conjugates (1  mg/mL protein) dissolved in CEX-buffer A. 3. Elute the products with CEX-buffer B using a shallow elution gradient. 4. Analyze the eluate with a UV–visible or fluorescence detector at a suitable wavelength. 5. In case further analysis of the samples is desired (e.g., SDS electrophoresis, mass spectrometry), collect the fractions containing the conjugate and use dialysis or ultrafiltration to change the buffer. 6. Lyophilize the conjugate products and store at −20°C until use (see Note 8).

3.4.3. Reversed-Phase HPLC

1. Pre-equilibrate the column with RP-buffer A at a flow rate of 1 mL/min. 2. Load 20 mL of the PEG–protein conjugates (0.1 mg/mL of protein) solubilized or diluted in RP-buffer A. 3. Elute the products with RP-buffer B using an appropriate elution gradient (formic acid can be used instead of TFA if a LC/MS detector is employed). 4. Analyze the eluate with a UV–visible or fluorescence detector at a suitable wavelength. 5. In case further analysis of the samples is desired (e.g., SDS electrophoresis, mass spectrometry), collect the fractions containing the conjugate and use dialysis or ultrafiltration to change the buffer.

122

Mero et al.

6. Lyophilize the conjugate products and store at −20°C until use (see Note 8). 3.4.4. Size-Exclusion Chromatography

1. Equilibrate the size-exclusion column and the refractive index (RI) detector with elution buffer for 1  h at a flow rate of 1 mL/min. 2. Load 20  mL of the PEG–protein conjugates (0.1–0.5  mg/ mL protein). 3. Elute the product with the same elution buffer used to equilibrate the column in Step 1. 4. To analyze the eluate, the system can be connected to two channels: a UV–visible or fluorescence detector and a RI detector.

3.4.5. SDS-Polyacrylamide Gel Electrophoresis and PEG–Protein Detection

1. Run the PEG–protein conjugate sample on an SDS-PAGE gel according to the electrophoresis apparatus manufacturer’s instructions. 2. After separating the PEG–protein conjugates by electrophoresis, soak the resulting SDS-PAGE gel in 20 mL of perchloric acid (0.1 M) for 15 min. 3. Add 5 mL of BaCl2 solution and 2 mL of iodine solution. The brown-stained PEG bands should appear within a few minutes. 4. After 10–15 min, replace the staining solution with H2O and incubate the gel for another 15 min. 5. The iodine-stained gel can also be further stained with Coomassie blue for the detection of proteins.

3.4.6. Mass Spectrometry of PEG–Protein Conjugates

1. Dissolve the PEG–protein conjugate sample (20 mg) in a 0.1% (v/v) TFA aqueous solution (5–10 mL). 2. Mix a saturated matrix solution with the PEG–protein sample solution in the ratio of 1:1 (v/v). 3. Load the mixture (10–20  mL) onto the sample plate and perform the MS analysis procedure after solvent evaporation.

3.4.7. Determination of PEGylation Sites for Proteins Modified with a Polydisperse PEG

1. Separately dissolve the native and PEGylated proteins (200 mg) in Tris-Gdn buffer (pH 9.0) at a final protein concentration of 1 mg/mL. 2. Add tris (2-carboxyethyl) phosphine to the protein solution at a final concentration of 5 mM. Incubate the reaction mixture for 1 h at 37°C. 3. Add iodoacetamide (25 mM) to the reduced protein solution and incubate the reaction mixture for 30 min at 37°C in the dark. 4. Purify the protein samples by RP-HPLC using a C18 column. Dry the collected fractions.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

123

5. Dissolve the reduced and S-carboxamidomethylated samples of native and PEGylated protein in 8  M urea. Next, dilute the samples further in phosphate buffer to obtain a final protein concentration of 0.8 mg/mL and a final urea concentration of 0.8 M. 6. Add trypsin at an enzyme/substrate (E/S) ratio of 1:50 (w/w) and let the proteolysis reaction proceed at 37°C overnight (see Note 10). 7. Fractionate (e.g., use an analytical C4 or C18 column) both the native and PEGylated protein digestion mixtures by RP-HPLC. Collect the products by monitoring the eluate by UV absorbance (l = 214 nm). 8. Compare the elution patterns of the peptides obtained from the modified and native digests. The identity of the peptides that are missing in the PEGylated protein digest can be established by analysis of the corresponding peaks in the non-PEGylated digest. For this purpose, mass spectrometry is employed. 9. The elution pattern obtained from the modified (PEGylated) protein may sometimes show new peaks corresponding to the PEGylated peptides. Their identity can be revealed by MALDI-TOF mass spectrometry, even though this may not be an easy task (due to the polydispersity of PEG). 3.4.8. Determination of PEGylation Sites for Proteins Modified with a Monodisperse PEG

Follow Steps 1–6 in Subheading 3.4.7 above, and then continue with the following procedure: 1. Desalt the native and PEGylated protein digestion mixtures using a PepClean™ C-18 spin column and analyze the products directly by ESI-MS. 2. For both the native protein and PEGylated protein samples, identify all the resulting peptide fragments. Some of the peptides obtained in the PEGylated sample may show an increase in mass corresponding to the conjugation of a single chain of polymer. 3. If more than one available site for conjugation is present within the same fragment, a tandem mass spectrometry (MS/ MS) analysis is performed to determine which particular amino acid is modified.

3.4.9. Degree of Protein Modification by the Habeeb Assay

1. Prepare the native (unmodified) protein sample and the PEGylated derivative at the same protein concentration (0.2– 0.8 mg/mL) in phosphate buffer, pH 7.2. 2. Prepare the TNBS reagent and perform the assay in test tubes (in duplicate) at room temperature according to Table 5. 3. Incubate all the samples in a water bath at 40°C for 2 h, and then add 250 mL of 10% (w/v) SDS and 125 mL of 1N HCl. 4. Read the absorbance of the solution at l = 335  nm using a spectrophotometer.

124

Mero et al.

Table 5 Preparation of test solutions for the Habeeb assay Blank

Native protein

Pegylated protein

250 mL phosphate pH 7.2

250 mL of protein sample

250 mL of PEGylated sample

250 mL bicarbonate pH 8.5

250 mL bicarbonate pH 8.5

250 mL bicarbonate pH 8.5

250 mL TNBS

250 mL TNBS

250 mL TNBS

Table 6 Preparation of test solutions for the indirect Ellman’s assay Blank

Native protein

Pegylated protein

50 mL of Ellman’s reagent

50 mL of Ellman’s reagent

50 mL of Ellman’s reagent

1 mL of phosphate pH 7.2

970 mL of phosphate pH 7.2

970 mL of phosphate pH 7.2

30 mL of protein sample

30 mL of PEGylated sample

The degree (%) of amine substitution is calculated as follows:

% Substitution = [1 - (AP - AB)/(AN - AB)] ´ 100%

AP = Absorbance of the PEGylated protein AB = Absorbance of the blank AN = Absorbance of the native protein. 3.4.10. Degree of Cysteine Modification by the Indirect Ellman’s Assay

1. Prepare the native (unmodified) protein sample and the PEGylated derivative at the same protein concentration (0.2–0.8 mg/mL) in phosphate buffer, pH 7.2. 2. Prepare the Ellman’s reagent and perform the assay in test tubes (in duplicate) at room temperature according to Table 6. 3. Incubate all the samples for 15 min. 4. Read the absorbance spectrophotometer.

at

l = 412 

nm

using

a

The percentage of free SH groups is calculated by applying the following formula:

% Free thiol groups = [(AP - AB)/(AN - AB)] ´ 100

AP = Absorbance of the PEGylated sample AB = Absorbance of the blank AN = Absorbance of the native protein.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

125

Table 7 Preparation of test solutions for the iodine assay Blank

Sample

525 mL of Milli-Q water

500 mL of Milli-Q water

250 mL of BaCl2 solution

250 mL of BaCl2 solution

250 mL of iodine solution

250 mL of iodine solution 25 mL of PEG solution

3.4.11. Qualitative Test for the Presence of PEG by the Iodine Assay

A rapid, qualitative analysis of the total PEG content in a sample can be performed as follows: 1. To a clean tube, add 975 mL of deionized water, 250 mL of BaCl2 solution, and 250 mL of iodine solution. 2. To the above mixture, add 25 mL of the PEG–protein conjugate sample solution. 3. The test is positive if the final mixture forms a dark precipitate, or if it shows increased absorbance at l = 535 nm.

3.4.12. Quantitative Test for the Amount of PEG by the Iodine Assay

A quantitative analysis of the total PEG content in a sample can be performed as follows (see Note 11): 1. Prepare the blank, PEG standard solutions (0.2–0.5  mg/ mL PEG) and unknown sample solutions according to Table 7. 2. Incubate the solutions for 15 min and then read the absorbance at l = 535 nm. 3. Generate a calibration curve by plotting the measured absorbance values (535 nm) versus the known concentration values of the PEG standards. 4. The amount of PEG present in the unknown sample solutions can be determined from comparison of the measured absorbance values against the standard curve generated in Step 3.

4. Notes 1. A limitation of the use of PEG is its hygroscopicity: If not stored under dry conditions, activated PEGs are easily hydrolyzed. It is useful to always verify the degree of activation of new batches of activated PEGs that have been obtained

126

Mero et al.

commercially. Activation values between 70 and 90% are acceptable, but they must be taken into consideration when determining the excess amount of PEG needed for a desired conversion yield. To minimize the levels of deactivation due to hydrolysis, store activated PEG reagents at −20°C under nitrogen and warm the bottle to room temperature before opening. 2. A major limitation of PEG aldehyde derivatives is their susceptibility to air oxidation. Low-temperature storage under an inert atmosphere is mandatory, even if such conditions may not always be effective. 3. When PEG-OPSS is used, careful attention should be paid to avoid the presence of any thiol-reducing agents in all steps of conjugate preparation and purification. 4. When PEG-Mal is used, avoid pH conditions above 7.5 since at higher pH, reaction with primary amine groups can also take place (although at a slower rate compared with free thiol groups). 5. Different aqueous buffers may be employed (e.g., phosphate, borate, HEPES, etc.), but do not use tris (hydroxymethyl) aminomethane (Tris) or any other primary amine-containing buffer components because they will compete with proteins in the PEG coupling reaction. 6. Dialysis is effective if the outside buffer is often changed and the dialysis volume is 500- to 1,000-fold greater than the volume of the sample. 7. The mobile phase in CEX can be a phosphate, acetate, or citrate buffer solution. Specific buffers are provided in Subheading  2.3 as examples, but the optimal buffer to use needs to be considered case-by-case depending on the particular protein studied. 8. Some proteins are also stable at room temperature provided that maintenance of sterility is guaranteed. 9. For RP-HPLC buffer B, CH3CN is typically used, but in some cases a mixture of CH3CN/MeOH or only MeOH may also be suitable. 10. Different enzymes may be used to digest the protein samples; trypsin, chymotrypsin, and V8-protease are the most commonly employed. 11. Iodine assays for the quantitative determination of PEG must be performed after purification of the conjugates from free, unbounded PEG molecules.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

127

References 1. Abuchoswki, A., Van, E., Palczuk, N.C., et al. (1977) Alteration of immunological properties of bovine serum albumin by covalent attachment of polyethylene glycol. J. Biol. Chem. 252, 3578–3581. 2. Jevševar, S., Kunstelj, M., Gaberc Porekar, V. (2010) PEGylation of therapeutic proteins. J. Biotechnol. 5, 113–128. 3. Mero, A., Veronese, F.M. (2008) The impact of PEGylation on biological therapies Biodrugs 22, 315–329. 4. Zalipsky, S. (1995) Functionalized poly (ethylene glycol) for preparation of biologically relevant conjugates. Bioconjug. Chem. 6, 150–165. 5. Veronese, F.M. (2001) Peptide and protein PEGylation: a review of problems and solutions. Biomaterials 22, 405–417. 6. Snyder, S. L. and Sobocinski, P. Z. (1975) An improved 2,4,6,-trinitrobenzenesulfonic acid method for the determination of amines. Anal. Biochem. 64, 284–288. 7. Riddles, P.W., Blakeley, R. L., and Zarner, B. (1979) Ellman’s reagent: 5,5’-dithiobis(2nitrobenzoic acid) a reexaminatio. Anal. Biochem. 94, 75–81. 8. www.laysanbio.com 9. Hanton, S. D. (2001) Mass spectrometry of polymers and polymer surfaces. Chem. Rev. 101, 527–570. 10. Vestling, M. M., Murphy, C. M., Fenselau, C., Dedinas, J., Doleman, M. S., Harrsch, P. B., Kutny, R., Ladd, D. L., and Olsen, M. A. (1992) Techniques in Protein Chemistry III, Academic Press, New York, pp. 477–485. 11. Vestling, M. M., Murphy, C. M., Keller, D. A., Fenselau, C., Dedinas, J., Ladd, D. L., and Olsen, M. A. (1993) A strategy for character­ ization of polyethylene glycol-derivatized proteins: a mass spectrometric analysis of the attachment sites in polyethylene glycol-derivatized superoxide dismutase. Drug Metab. Dispos. 21, 911–917. 12. Montaudo, G., Samperi, F., Montaudo, M. S. (2006) Characterization of synthetic polymers by MALDI-MS. Prog. Polym. Sci. 10, 1016–1020. 13. Zhang H., Zhang, J., Luo, Y., Wilson, J., and Miller, K. (2007) Mass spectrometric analyses of potential impurities in 20 kDa monomethoxy poly (ethylene glycol)-propionaldehyde (PEG aldehyde) raw material. AAPS Annual Meeting & Exposition. 14. Sivakolundu, S. G., and Mabrouk, P. A. (2003) Proton NMR study of chemically modified

horse heart ferricytochrome C confirms the presence of histidine and lysine-ligated conformers in 30% acetonitrile solution. J. Inorg. Biochem. 94, 381–385. 15. Orsatti, L. and Veronese, F. M. (1999) An unusual coupling of poly(ethylene glycol) to tyrosine residues in epidermal growth factor. J. Bioact. Compat. Pol. 14, 429–436. 16. Wylie, D. C., Voloch, M., Lee, S., Liu, Y. H., Cannon-Carlson, S., Cutler, C., Pramanik, B. (2001) Carboxyalkylated histidine is a pHdependent product of PEGylation with SC-PEG. Pharm Res. 18, 1354–1360. 17. Kinstler, O. B., Brems, D. N., Lauren, S. L., Paige, A. G., Hamburger, J. B. and Treuheit, M. J. (1996) Charecterization and stability of N-Terminally PEGylated rhG-CSF. Pharm. Res. 13, 996–1002. 18. Lee, H., Jang, I. H., Ryu, S. H., and Pack, T.G. (2003) N-Terminal site-specific monoPEGylation of epidermal growth factor. Pharm. Res. 20, 818–825. 19. Kinstler, O., Molineux, G., Treuheit, M., Ladd, D. et  al. (2002) Mono-N-terminal poly(ethyleneglycol)-protein conjugates. Adv. Drug Deliv. Rev. 54, 477–485. 20. Arakawa, T., Prestrelski, S. J., Narhi, L. O., Boone, T. C., Kenney, W. C. (1993) Cysteine 17 of recombinant human granulocyte-colony stimulating factor is partially solvent-exposed. J. Protein Chem. 12, 525–531. 21. Colonna, C., Conti, B., Perugini, P., Pavanetto, F., Modena, T., Dorati, R., Iadarola, P., Genta, I. (2008) Site-directed PEGylation as successful approach to improve the enzyme replacement in the case of prolidase. Int. J. Pharm. 24, 230–237. 22. Xian-Hui, H., Pang-Chui, S., Li-Hui, X., and Siu-Cheung, T. (1999) Site-directed polyethylene glycol modification of trichosanthin: Effects on its biological activities, pharmacokinetics, and antigenicity. Life Sic. 64, 1163–1175. 23. Balan, S., Choi, J., Godwin, A., Teo, I., Laborde, C. M., Heidelberger, S., Zloh, M., Shaunak, S., and Brocchini, S. (2007) SiteSpecific PEGylation of Protein Disulfide Bonds Using a Three-Carbon Bridge. Bioconjug. Chem. 18, 61–76. 24. Zalipsky, S., and Meno-Rudolph, S. (1997) Hydrazide derivatives of polyethylene glycol and their bioconjugates. In Polyethylene glycol chemistry and biological applications, ACS sym­ posium series 680 (Harris, J. M., and Zalipsky, S. eds.), pp. 318–341.

128

Mero et al.

25. Gaertner, H. F., and Offord, R. E. (1996) Site-Specific Attachment of Functionalized Poly(ethylene glycol) to the Amino Terminus of Proteins. Bioconjug. Chem. 7, 38–44. 26. Baudys, M., Uchio, T., Mix, D., Wilson, D., and Kim, S. W. (1995) Physical stabilization of insulin by glycosylation. J. Pharm. Sci. 84, 28–33. 27. Esposito, P., Barbero, L., Caccia, P., Caliceti, P., D’Antonio, M., Piquet, G., Veronese, F. M. (2003) PEGylation of growth hormone-releasing hormone (GRF) analogues. Adv. Drug Del. Rev. 55, 1279–1291 28. Veronese, F. M., Mero, A., Caboi, F., Sergi, M., Marongiu, C., and Pasut, G. (2007) SiteSpecific PEGylation of G-CSF by Reversible Denaturation. Bioconjug. Chem. 18, 1824–1830. 29. Morpurgo, M., Monfardini, C., Hofland, L. J., Sergi, M., Orsolini, P., Dumont, J. M., and Veronese, F. M. (2002) Selective Alkylation and Acylation of a and e Amino Groups with PEG in a Somatostatin Analogue: Tailored Chemistry for Optimized Bioconjugates. Bioconjug. Chem. 13, 1238–1243. 30. Youn, Y. S., Na, D. H., Lee, K. C. (2007) High-yield production of biologically active mono-PEGylated salmon calcitonin by sitespecific PEGylation. J. Control Release 117, 371–379. 31. Chae, S. Y., J., C. H., Shin, H. J., Youn, Y. S., Lee, S., Lee, K.C. (2008) Preparation, characterization, and application of biotinylated and biotin-PEGylated glucagon-like peptide-1 analogues for enhanced oral delivery. Bioconjug. Chem. 19, 334–341. 32. Sato, H. (2002) Enzymatic procedure for sitespecific PEGylation. Adv. Drug Delivery. Rev. 54, 487–504. 33. Fontana, A., Spolaore, B., Mero, A., and Veronese, F. M. (2008) Site-specific modification and PEGylation of pharmaceutical proteins mediated by transglutaminase. Adv. Drug Delivery Rev. 60, 13–28. 34. Li, N., Ziegemeier, D., Bass, L., Wang, W. (2008) Quantitation of free polyethylene glycol in PEGylated protein conjugate by size exclusion HPLC with refractive index (RI) detection. J. Pharm Biomed Anal. 48, 1332–1338. 35. Sims, G. E., and Snape, T. J. (1980) A method for the estimation of polyethylene glycol in plasma protein fractions. Anal. Biochem. 107, 60–63. 36. Seely, J. E., Richey, C. W. (2001) Use of ionexchange chromatography and hydrophobic interaction chromatography in the preparation and recovery of polyethylene glycol-linked proteins. J. Chromatogr. A 908, 235–241.

37. Seely, J. E., Buckel, S. D., Green, P. D., Richey, C. W. (2005) Making site-specific PEGylation work. Biopharm Int. 18, 30–35. 38. Wang, Y. S., Youngster, S., Bausch, J., Zhang, R., McNemar, C., Wyss, D. F. (2000) Identification of the major positional isomer of pegylated interferon alpha-2b. Biochemistry 39, 10634–10640. 39. Fee, C. J., Van Alstine, J. M. (2006) PEGproteins: Reaction engineering and separation issues. Chem. Eng. Sci. 61, 924–939. 40. Pabst, T. M., Buckley, J. J., Ramasubramanyan, N., Hunter, A. K. (2007) Comparison of strong anion-exchangers for the purification of a PEGylated protein. J. Chromatogr. A 1147, 172–182. 41. Park, E. J., Lee, K. C, Na, D. H. (2009) Separation of positional isomers of monopoly(ethylene glycol)-modified octreotides by reversed-phase high-performance liquid chromatography. J Chromatogr. A 6, 7793–7797. 42. Clark, R., Olson, K., Fuh, G., Mariani, M., Mortensen, D., Teshima, G., Chang, S., Chu, H., Mukku, V., Canova-Davis, E., Somers, T., Cronin, M., Winkler, M., and Wells, J. A. (1996) Long-acting Growth Hormones Produced by Conjugation with Polyethylene Glycol. J. Biol. Chem. 271, 21969–21977. 43. Foser, S., Schacher, A., Weyer, K. A., Brugger, D., et al. (2003) Isolation, structural characterization, and antiviral activity of positional isomers of monopegylated interferon alpha-2a (PEGASYS). Protein Expr. Purif. 30, 78–87. 44. Gaberc-Porekar, V., Zore, I., Podobnik, B., Menart,V. (2008) Obstacles and pitfalls in the PEGylation of therapeutic proteins. Curr. Opin. Drug Discov. Devel. 11, 242–250. 45. Piedmonte, D. M., Treuheit, M. J. (2008) Formulation of Neulasta(R) (pegfilgrastim). Adv. Drug Del. Rev. 60, 50–58. 46. Fee, C. J (2007) Size comparison between proteins PEGylated with branched and linear poly(ethylene glycol) molecules. Biotechnol Bioeng. 98, 725–731. 47. Edwards, C. K., Martin, S. W., Seely, J., Kinstler, O. et al. (2003) Design of PEGylated soluble tumor necrosis factor receptor type I (PEG sTNF-RI) for chronic inflammatory diseases. Adv. Drug Deliv. Rev. 55, 1315–1336. 48. Molek, J. R., Zydney, A. L. (2006) Ultrafiltration characteristics of pegylated proteins. Biotechnol. Bioeng. 95, 474–482. 49. Kwon, B., Molek, J., Zydney, A. L. (2008) Ultrafiltration of PEGylated proteins: Fouling and concentration polarization effects. J. Memb. Sci. 319, 206–213.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides 50. Habeeb, A. F. S. A. (1966) Determination of free amino groups in protein by trinitrobenzenesulphonic acid. Anal.Biochem. 14, 328–336. 51. Riddles, P. W., Blakeley, R. L., and Zarner, B. (1983) Reassessment of Ellman’s reagent. Methods Enzymol. 91, 49–60. 52. Jiskoot, W., Crommelin, D. (2005) Methods for Structural Analysis of Protein Pharmace­ uticals Biotechnology: Pharmaceutical Aspects. American Assoc. of Pharm. Scientists, Springer, New York. 53. www.piercenet.com 54. Kusterle, M., Jevsevar, S., Gaberc-Porekar, V. (2008) Size of Pegylated Protein Conjugates Studied by Various Methods. Acta Chim. Slov. 55, 594–601. 55. Piedmonte, D. M., Treuheit, M. J. (2008) Formulation of Neulasta(R) (pegfilgrastim). Adv. Drug Del. Rev. 60, 50–58. 56. Lee, K., Moon, S. C., Park, M. O., Lee, J. T., Na, D. H., Yoo, S. D., et al. (1999) Isolation, characterizasion, and stability of positional isomers of mono-PEGylated salmon calcitonins. Pharm. Res. 16, 813–818. 57. Manjula, B. N., Tsai, A., Upadhya, R., Perumalsamy, K., Smith, P. K., Malavalli, A., Vandegriff, K. R., Winslow, M., Intaglietta, M., Prabhakaran, M., Friedman, J. M., and Acharya A. S. (2003) Site-Specific PEGylation of Hemoglobin at Cys-93(b): Correlation between the Colligative Properties of the PEGylated Protein and the Length of the Conjugated PEG Chain. Bioconjug. Chem. 14, 464–472. 58. Fee, C. J., Van Alstine, J. M. (2004) Prediction of the viscosity radius and the size exclusion chromatography behavior of PEGylated proteins. Bioconjug. Chem. 15, 1304–1313. 59. Kurfurst, M. M. (1992) Detection and Molecular-Weight Determination of Polyethylene Glycol-Modified Hirudin by Staining After Sodium Dodecyl-Sulfate Polyacrylamide-Gel Electrophoresis. Anal. Biochem. 200, 244–248. 60. Caccia, D., Ronda, L., Frassi, R., Perrella, M., Del Bavero, E., Bruno, S., Pioselli, B., Abbruzzetti, S., Viappiani, C., and Mozzarelli A. (2009) PEGylation Promotes Hemoglobin

129

Tetramer Dissociation. Bioconjug. Chem. 20, 1356–1366. 61. Caserman, S., Kusterle, M., Kunstelj, M., Milunovic, T. et al. (2009) Correlations between in  vitro potency of polyethylene glycol-protein conjugates and their chromatographic behaviour. Anal. Biochem. 389, 27–31. 62. Cindric, M., Cepo, T., Galic, N., BukvicKrajacic, M. et al. (2007) Structural characterization of PEGylated rHuG-CSF and location of PEG attachment sites. J. Pharm. Biomed. Anal. 44, 388–395. 63. Mero, A., Spolaore, B., Veronese, F. M., and Fontana, A. (2009) TransglutaminaseMediated PEGylation of Proteins: Direct Identification by Mass Spectrometry Using a Novel Monodisperse PEG. Bioconjug. Chem. 20, 384–389. 64. Sergi, M., Caboi, F., Maullu, C., Orsini, G., and Tonon, G. (2009) Enzymatic techniques for PEGylation of biopharmaceuticals. In PEGylated Protein Drugs: Basic Science and Clinical Applications (Milestones in Drug Therapy) (Veronese, F.M., ed.) Birkhauser Verlag, Boston, MA, pp. 75–88. 65. Basu, A., Yang, K., Wang, M., Liu, S., et  al. (2006) Structure-function engineering of interferon-beta-1b for improving stability, solubility, potency, immunogenicity, and pharmacokinetic properties by site-selective mono-PEGylation. Bioconjug. Chem. 17, 618–30. 66. Stoscheck, C. M. (1990) Quantification of protein. Methods Enzymol. 182, 50–69. 67. Yu, P., Zheng, C., Chen, J., Zhang, et  al. (2007) Investigation on PEGylation strategy of recombinant human interleukin-1 receptor antagonist. Bioorg. Med. Chem. 15, 5396–405. 68. Monfardini, C., Schiavon, O., Caliceti, P., Morpurgo, M., Harris, J. M., and Veronese, F. M. (1995) A branched mono­methoxypoly (ethylen glicol) for protein modification. Bioconjug. Chem. 6, 62–69. 69. Bailon, P., Palleroni, A., Schaffer, C. A., Spence, C. L. et al. (2001) Rational design of a potent, long-lasting form of interferon: a 40 kDa branched polyethylene glycol-conjugated interferon alpha-2a for the treatment of hepatitis C. Bioconjug. Chem. 12, 195–202.

Chapter 9 Extending the Scope of Site-Specific Cysteine Bioconjugation by Appending a Prelabeled Cysteine Tag to Proteins Using Protein Trans-Splicing Tulika Dhar, Thomas Kurpiers, and Henning D. Mootz Abstract Incorporating synthetic probes site-specifically into proteins is of central interest in several areas of biotechnology and protein chemistry. Bioconjugation techniques provide a simple and effective means of chemically modifying a protein. In particular, covalent chemical modifications of cysteine residues belong to one of the most important reactions due to the unique reactivity of its thiol moiety and the relatively low abundance of this amino acid in proteins. However, such types of modifications cannot be performed in a regioselective fashion when one or more additional cysteines are present. To address this limitation, we have developed an approach where a short cysteine-containing tag (Cys-Tag) fused to one part of a split intein and modified at its sulfhydryl group can be used to label proteins by trans-splicing with a protein of interest (POI) fused to the other half of the split intein. In this way, it is possible to selectively label a protein containing multiple cysteines. The artificially split Mycobacterium xenopi GyrA intein and the Synechocystis sp. DnaB intein were highly suitable for this purpose and were successfully used for the labeling of several proteins. This approach enables a simple route for labeling proteins by site-specific cysteine bioconjugation with any one of several commercially available cysteine-modifying probes. Key words: Cys-Tag, Split intein, Protein trans-splicing, Protein modification, Bioconjugation

1. Introduction Inteins are internal polypeptide sequences of proteins that can remove themselves autocatalytically without utilizing external co-factors or energy in a process-termed protein splicing (1–3). Besides a few naturally occurring split inteins, many cis-splicing inteins can be artificially split into two parts. These complementary intein halves, IntN and IntC, require first the association and folding of both halves to facilitate protein trans-splicing activity (1–7). Similar to a regular intein, they excise themselves out by breaking two peptide bonds and concomitantly join their ­respective Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_9, © Springer Science+Business Media, LLC 2011

131

132

Dhar, Kurpiers, and Mootz

flanking extein sequences (ExtN and ExtC) by a new peptide bond. This posttranslational reaction has successfully been used for several in vitro applications, such as protein semisynthesis (7, 8), modulation of extein sequences (9), and segmental isotopic labeling (10). Site-specific modification of proteins is an attractive tool for studying protein structure and function, and also for endowing them with new properties. However, each technique has certain advantages and limitations. Therefore, new techniques are required to expand the scope of accessible proteins and modifications. Split inteins combine general and easy applicability and variability with cost effectiveness. Figure 1 shows the concept of adding a CysTag to a protein of interest (POI) using split inteins. This approach not only selectively labels the protein, but the intein excision makes the reaction almost traceless, as well (11, 12). It also has the advantage of leaving other essential cysteines unaffected. In addition, a large number of synthetic probes, e.g., fluorophores, biotin, and PEG coupled to well-established ­functional groups for bioconjugation (like haloacetamides and maleimides) are commercially available from various suppliers. Furthermore, this approach requires only very low concentrations of the POI–intein fusion protein (in the low micromolar range). The trans-splicing reaction is highly selective due to the required association of both

Fig. 1. General scheme for the two-step approach of site-selectively labeling a protein of interest by a combination of cysteine modification and protein trans-splicing. (a) Covalent chemical modification of a cysteine in the Cys-Tag fused to the IntC auxiliary protein (RG = reactive group). (b) Appending the prelabeled Cys-Tag to the protein by protein transsplicing. A native peptide bond is formed between the protein and the Cys-Tag.

Extending the Scope of Site-Specific Cysteine Bioconjugation

133

complementary split intein fragments (in fusion with the POI), allowing the technique to be performed on crude proteins in cell lysates. A prerequisite for this method is that the split intein halfattached to the Cys-Tag must itself be completely free of cysteines. Most commonly known inteins harbor a catalytic cysteine at each splice junction that is required for the protein splicing mechanism to function. In contrast, the artificially split Ssp DnaB intein from Synechocystis sp. has a serine at the (+1) position of the C-terminal intein fragment (11). Similarly, the C-terminal nucleophile in the case of the artificially split Mxe GyrA intein from Mycobacterium tuberculosis is a threonine, thereby rendering it suitable for our cysteine-labeling approach (12) (see Note 1). Inteins were artificially split for several applications before the discovery of a naturally split intein. Many inteins are bi-functional elements consisting of a protein splicing domain and an endonuclease domain responsible for its mobility (intein homing). It was reported that the 275-amino acid endonuclease region could be completely deleted from the Ssp DnaB intein without loss of the protein splicing activity, resulting in a functional mini-intein of 154 amino acids. Furthermore, the Ssp DnaB mini-intein could be split into two halves that reconstitute splicing activity when co-expressed in Escherichia coli (13). Our lab was then able to demonstrate that the separately expressed and purified halves of the intein are active in trans-splicing under native conditions, and are therefore highly suitable for in vitro applications (14). The Mxe GyrA intein is a 198-amino acid native mini-intein consisting of a short linker region in place of the endonuclease domain (15). This intein was chosen and artificially split in our lab on the basis that – being a native mini-intein – it would prove beneficial in terms of solubility and efficiency for in vitro reconstitution. In fact, it proved superior to the Ssp DnaB intein and is our favored intein for the Cys-Tag approach (12) (see Notes 2 and 3). This approach may further be extended to other inteins and tags (see Notes 1 and 4). It can also be used for labeling an internal position of the POI if the POI is split at an appropriate site (16). A second chemical group could be incorporated in the protein after the trans-splicing. This technique can also be applied suitably to obtain N-terminally labeled proteins, as demonstrated for the Psp-GBD Pol intein (16). The following protocol is written based on the use of fusion proteins and conditions recently reported by our laboratory (11, 12). Specifically, it describes the Cys-Tag labeling approach using the split Mxe GyrA intein. However, the methodologies presented in this chapter can be used as a general guideline for devising strategies to append a labeled Cys-Tag to any POI under different conditions. The example we report in detail herein is the ­modification of the nonribosomal peptide synthetase TycA from Bacillus brevis, as illustrated in Fig. 2.

134

Dhar, Kurpiers, and Mootz

Fig. 2. Illustration of a specific example of the Cys-Tag bioconjugation strategy presented in this chapter. (a) Alkylation of the Cys-Tag in auxiliary protein MBP-IntC-Cys-Tag-His6 with 5¢-iodoacetamide fluorescein. (b) Transfer of the Cys-Tag to the protein TycA by protein trans-splicing. Note that in the fusion constructs, the MBP and FKBP moieties serve to assist in maintaining protein solubility and correct protein folding.

2. Materials 2.1. Expression Plasmid Construction ( see Figs. 2 and 3)

1. Plasmid pTK130 (encoding Strep-tag II-TycA-GyrAN-FKBP; protein 1) (12). 2. Plasmid pTK120 (encoding MBP-GyrAC-Cys-Tag-His6; protein 2) (12) (see Note 5). 3. Restriction enzymes: EcoRI, NheI (Fermentas).

2.2. Protein Expression and Purification

1. Escherichia coli BL21 Gold (DE3) cells (Stratagene). 2. LB medium: 5 g/l NaCl, 10 g/l tryptone, 5 g/l yeast extract, pH 7.0. 3. Ampicillin stock solution (50 mg/ml), filter sterilize. 4. Isopropyl-b-thiogalactoside (IPTG) stock solution (400 mM), filter sterilize. 5. Strep-Tactin® affinity chromatography resin. 6. Strep-Tactin® Buffer W: 100 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, pH 8. 7. Strep-Tactin® Buffer E: 100 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, 2.5 mM desthiobiotin, pH 8. 8. Amylose affinity chromatography resin.

Extending the Scope of Site-Specific Cysteine Bioconjugation

135

Fig. 3. Schematic representation of the cloning of any protein of interest (POI) as a fusion construct with the IntN auxiliary protein.

9. Amylose Buffer: 20  mM Tris–HCl, 200  mM NaCl, 1  mM EDTA, pH 7.4. 10. Amylose Elution Buffer: 20 mM Tris–HCl, 200 mM NaCl, 1 mM EDTA, 10 mM maltose, pH 7.4. 11. Ni2+-NTA immobilized metal affinity chromatography resin. 12. Ni2+-NTA Wash Buffer A: 50 mM Tris–HCl, 300 mM NaCl, pH 8.0. 13. Ni2+-NTA Wash Buffer B: 50 mM Tris–HCl, 300 mM NaCl, 250 mM imidazole, pH 8.0. 2.3. Cysteine Labeling

1. Dilution buffer: 50 mM phosphate buffer (pH 7.2) containing 150 mM NaCl. 2. 5-(Iodoacetamido)fluorescein (5-IAF) stock solution (10 mM): Dissolve 1 mg of 5-IAF in 20 ml of DMF, and then dilute with 174.1  ml of dilution buffer to get a final stock solution of 10 mM (see Note 6). 3. Dithiothreitol (DTT) stock solution (10  mM): Dissolve 1.54 mg of DTT in 1 ml of water to give a stock solution of 10 mM (freshly prepared). 4. Hi-Trap™ Sephadex™ G-25 Superfine desalting column (GE Healthcare).

2.4. Protein ­Trans-splicing

1. Splice Buffer: 50  mM Tris–HCl, 300  mM NaCl, 1  mM EDTA, pH 7.0. 2. 4× SDS-loading buffer: 500 mM Tris–HCl, pH 6.8, 8% (w/v) SDS, 40% (v/v) glycerol, 20% (v/v) b-mercaptoethanol, 4 g/l bromophenol blue.

136

Dhar, Kurpiers, and Mootz

3. Methods The Cys-Tag labeling approach can essentially be used for selectively modifying almost any POI. In our hands, this method has allowed us to successfully carry out the modification of several important proteins, such as human growth hormone (hGH), the nonribosomal peptide synthetase tyrocidine A (TycA), beta-lactamase, etc., with all of them containing one or more functionally or structurally important cysteine residues (11, 12). However, certain points are to be kept in mind regarding the nature of the intein. Artificially split inteins tend to behave differently than their intact parent proteins. When artificially split protein fragments are individually expressed, they often display a higher tendency to misfold or aggregate; this could be due to the exposure of hydrophobic patches, which in some cases can result in insolubility. On a similar note, many artificially split inteins require a refolding step under denaturing conditions to become active in protein trans-splicing, thus preventing their application to proteins that cannot be refolded. However, this is not the case for the above-mentioned Ssp DnaB and Mxe GyrA inteins, which can be purified from and reconstituted under native conditions. Another consideration to bear in mind is the nature of the amino acids directly flanking the intein. These amino acids can play a significant role in influencing the efficiency of the splicing reaction. In order to prevent distortions caused by their close proximity to the intein active site, it is generally recommended to keep three to five native residues on each side of the intein when inserting it into a foreign protein sequence. The most critical flanking residue is usually the (−1) amino acid located immediately upstream of the intein. The (+1) catalytic residue, a cysteine, serine or threonine, is required for the protein splicing reaction and is retained in the splice product. To generate the two halves of the Mxe GyrA intein, we split the encoding gene between positions for Arg119 and Gly120 on the DNA level, thereby creating a 119-amino acid IntN fragment and a 79-amino acid IntC fragment. The first two adjacent residues of the natural N-terminal extein sequence (positions −1 and −2) and the first three residues of the natural C-terminal extein sequence (positions +1 to +3) were retained in our constructs (plasmids pTK130 and pTK120, respectively). The resulting IntC is free of any cysteine residues. Additionally, to prevent the larger IntN fragment from potential aggregation and to minimize the exposure of hydrophobic regions which might lead to misfolding, a highly soluble protein like the phage protein gpD or FK506 binding protein (FKBP) domain was added at the C-terminus (12).

Extending the Scope of Site-Specific Cysteine Bioconjugation

3.1. Generating Int N Fusion Construct with POI

137

1. Amplify the gene encoding your POI by PCR with an N-terminal extension for the recognition sequence of NheI and a C-terminal extension for the recognition sequence of EcoRI (see Note 7). 2. Digest the PCR product and plasmid pTK130 with the restriction enzymes NheI and EcoRI. 3. Purify both fragments by agarose gel electrophoresis. 4. Ligate both fragments together with T4 DNA ligase to generate a complete plasmid encoding your POI (see Fig. 3).

3.2. Expression and Purification of Recombinant Protein (see Note 8)

1. Transform one E. coli BL21 (DE3) cell culture with the plasmid encoding the IntN (pTK130) fusion construct. Transform a second (i.e., separate) E. coli BL21 (DE3) cell culture with the plasmid encoding for the IntC (pTK120) fusion construct. 2. For each fusion construct, inoculate 600  ml of fresh LB medium containing 100  mg/ml ampicillin with 6  ml of an overnight seed culture. 3. Grow the cells at 37°C to an OD600 of 0.7. 4. Lower the culture temperature to 25°C and induce protein expression by adding 600 ml of IPTG stock solution to give a final concentration of 0.4 mM IPTG. 5. After 3–5 h, harvest the cells by centrifugation at 10,000 × g for 20 min at 4°C. 6. Resuspend the cells expressing the IntN fusion protein (protein 1) in cold Buffer W, and those expressing the IntC fusion protein (protein 2) in amylose buffer. 7. Either freeze the resuspended cells at −80°C or immediately proceed to disrupt the cells using an emulsifier (high-pressure homogenizer). Keep the cell suspension and the cell lysate on ice. 8. Centrifuge the cell lysate at 30,000 × g for 30 min at 4°C to separate the soluble fraction (supernatant) (see Note 2). 9. For the IntN fusion protein (protein 1), perform protein purification on a Strep-Tactin® affinity column (with gravity feed) according to the manufacturer’s instructions: Load the supernatant extract directly onto a Strep-Tactin® column pre-equilibrated with Buffer W. After the cell extract has completely entered the column, wash the column with three column volumes of Buffer W and then elute the fusion protein from the column using Buffer E. 10. For the IntC fusion protein, perform protein purification first on an amylose column, and then perform a second column purification using Ni2+-NTA as follows: Amylose column purification (gravity feed):  Fill a chromatography column with 5 ml of amylose resin and equilibrate with

138

Dhar, Kurpiers, and Mootz

Amylose Buffer. Load the supernatant extract and wash the column three times with Amylose Buffer. Elute the fusion protein from the column using Amylose Elution Buffer and pool the fractions containing the fusion protein. Perform a buffer exchange (by dialysis) to replace the Amylose Elution Buffer with Ni2+-NTA Buffer A for a second column purification using Ni2+-NTA. Ni2+-NTA column purification (gravity feed):  Fill a chromatography column with 2 ml of Ni2+-NTA resin and equilibrate with Buffer A. Load the protein sample (obtained after amylose column purification) and wash the column with 5 ml of Buffer A. Repeat the wash with 5 ml of Buffer A containing 5 mM imidazole. Wash twice more, once with Buffer A containing 20  mM imidazole and the last wash with Buffer A containing 40 mM imidazole. Finally, elute the fusion protein from the column using Buffer B. 11. Analyze the protein samples by SDS-PAGE: Mix 10 ml of protein sample with 3 ml of 4× SDS-loading dye, and load 10 ml onto the gel to check for purity. 12. Dialyze each of the proteins against splice buffer containing 2 mM DTT and 10% (v/v) glycerol. 13. Calculate the protein concentration of the dialyzed samples using the calculated molecular extinction coefficient at l = 280 nm. 14. Aliquot the purified fusion proteins into Eppendorf tubes, and flash-freeze in liquid nitrogen. Store the proteins at −80°C. 3.3. Cys-Tag Labeling

1. Reduce protein 2 by adding a tenfold molar excess of reducing agent (e.g., to a 100 ml aliquot of a 20 mM protein solution, add 2 ml from a 10 mM DTT stock solution). Incubate the mixture for 10 min at room temperature. 2. Add the labeling reagent (e.g., 5-IAF) at a 25-fold molar excess over protein 2 (e.g., add 5 ml from a 10  mM 5-IAF stock solution). 3. Incubate the mixture at 25°C for 2 h. 4. Quench the labeling reaction with 2–10 mM DTT. 5. Remove excess labeling and reducing agent by gel filtration (optional).

3.4. Protein Transsplicing Reaction

1. Add protein 1 and labeled protein 2 together in equimolar concentrations in a 1.5-ml Eppendorf tube along with 2 mM DTT. Adjust the volume with splice buffer. We suggest using a concentration of 10 mM for each protein (adjust according to the eluted protein concentrations, or perform further concentration if desired).

Extending the Scope of Site-Specific Cysteine Bioconjugation

139

Fig. 4. SDS-PAGE analysis of the trans-splicing reaction of the Cys-Tag strategy described in this chapter. (a) Analysis of the products of the protein trans-splicing reaction between protein 1 and protein 2 on a Coomassie brilliant blue-stained gel (left ) and a UV-illuminated gel (right ). (b) Analysis of the labeled splice product obtained from the reaction mixture after performing two consecutive column chromatography purification procedures. Left: Coomassie-stained gel. Right : UV-illuminated gel. (Lane 1: flow-through of the Ni2+-NTA column; Lane 2: elution fraction of the Ni2+-NTA column; Lane 3: flow-through of the Strep-Tactin® column; Lane 4: elution fraction of the Strep-Tactin® column.).

2. Incubate the reaction mixture at 25°C and remove 10-ml ­aliquots at different time points to monitor the progress of the reaction by SDS-PAGE (see Note 9). 3. Stop the reaction by mixing each aliquot with 3 ml of 4× SDSPAGE loading buffer. 4. Boil the samples for 10 min and load 10 ml onto the SDSPAGE gel. 5. Observe the gel on a UV transilluminator. 6. Stain the gel with Coomassie brilliant blue (see Fig. 4a and Note 10). 7. Scan the gel to determine the relative intensities of the protein bands (see Note 11). 3.5. Purification of Labeled Splice Product (see Fig. 4b and Note 12)

1. Perform the splice reaction as described Subheading 3.4, step 1 for approximately 4 h.

above

in

2. Apply the entire reaction mixture onto a Ni2+-NTA column and perform column purification as described earlier in Subheading 3.2, step 10. 3. Collect the flow through and apply the protein sample onto a Strep®-Tactin column and again perform column purification as described earlier in Subheading 3.2, step 9.

4. Notes 1. There are other known inteins that in principle could be used similarly. For example, the Tli-Pol2 intein has a threonine as the +1 nucleophile (17), whereas the Psp-GBD Pol intein

140

Dhar, Kurpiers, and Mootz

(16) and the Ssp GyrB intein (18) have a serine at this position. 2. A comparison between the split Ssp DnaB and Mxe GyrA inteins has revealed that GyrA has superior properties for protein trans-splicing applications. The product yield with GyrA is up to 70% (as revealed by densitometric analysis), in contrast to only a 40–50% yield with the Ssp DnaB intein, although the product formation is faster in the case of Ssp DnaB. Furthermore, the Mxe GyrA intein shows better solubility and has worked for several difficult proteins expressed in our lab. One of these, human growth hormone (hGH), contains four cysteine residues involved in disulfide bridges. hGH is generally difficult to express in E. coli and is typically found in the inclusion bodies when expressed as a hGH-IntN fusion. In addition, an attempt to refold the fusion protein by renaturation from 8 M urea proved unsuccessful. In order to circumvent this problem, the fusion protein was solubilized with 8  M urea, purified under these denaturing conditions, and then the complementary intein fragment (IntC-CysTag) was added to the 8  M urea solution before dialysis. Next, the reaction mixture was dialyzed against splice buffer at 4°C and then further incubated at 25°C for 2 h. The expected splice product, hGH-CysTag, was formed in high yields and remained fully in solution. Thus, a poorly soluble protein could be successfully labeled using a modified version of our approach with the split GyrA intein, which might be applicable to other such types of proteins as well (11, 12). 3. The two intein systems are orthogonal, and no product formation was seen when the constructs were incubated in either of the two heterologous combinations (12). 4. We have tested the suitability of this approach with several fluorophores, as well as by labeling with polyethylene glycol (PEG) moieties (12). 5. The Cys-Tag can be a short peptide tag as described here, or it can be part of a larger protein domain. The exact sequence of the attached Cys-Tag is variable, and at present has not been fully optimized. In this protocol, we have used a short peptide sequence in protein 2 (T[+1]EAGSCS), but in principle it could be shortened to even just two amino acids (i.e., the +1 nucleophile and the cysteine). 6. Fluorescein must be protected from light exposure in order to avoid photobleaching. 7. In this cloning scheme, if NheI and EcoRI cut within your POI then this strategy cannot be applied. Other techniques such as restriction-free cloning or the use of modified vectors to circumvent these sites may be applied instead.

Extending the Scope of Site-Specific Cysteine Bioconjugation

141

8. The optimum expression and purification conditions must be adjusted according to the specific POI being studied. 9. The splicing reaction can be carried out over a wide range of temperatures. We have found that the highest yields are obtained between 20 and 30°C, and that significant activity is still observed at temperatures as low as 12°C. Even at 4°C, a high yield is observed if the reaction is allowed to proceed for more than 24 h (12). 10. During the splice reaction with the Mxe GyrA intein, an additional protein band will be observed that has a higher molecular weight than both starting proteins. This band appears almost immediately after both proteins are mixed but disappears completely after 2  h. This band corresponds to the branched intermediate of the splicing reaction (12). 11. The product yield of the trans-splicing reaction is unaffected by the modification with the Cys-Tag label when compared with the unmodified product. 12. In the above-mentioned splice reaction, the splice product (obtained from Subheading  3.4) can be selectively purified using two types of affinity chromatography purification since it is the only product in the reaction mixture that contains both the Strep-tag II® and polyhistidine (6×His) affinity tags (see Fig. 4b). Conversely, this strategy can also be applied to remove other components and purify only the splice product if the splice product is the only species in the reaction mixture that contains no affinity tags.

Acknowledgments The author’s would like to thank all members of the Mootz lab for helpful discussions. T.D. acknowledges a Ph.D. stipend from the International Max Planck Research School in Chemical Biology. Funding for this work was provided by the DFG and the Fonds der Chemischen Industrie. References 1. Noren, C. J., Wang, J., and Perler, F. B. (2000) Dissecting the Chemistry of Protein Splicing and Its Applications, Angew Chem Int Ed Engl 39, 450–466. 2. Paulus, H. (2000) Protein splicing and related forms of protein autoprocessing, Annu Rev Biochem 69, 447–496. 3. Gogarten, J. P., Senejani, A. G., Zhaxybayeva, O., Olendzenski, L., and Hilario, E. (2002)

Inteins: structure, function, and evolution, Annu Rev Microbiol 56, 263–287. 4. Wu, H., Hu, Z., and Liu, X. Q. (1998) Protein trans-splicing by a split intein encoded in a split DnaE gene of Synechocystis sp. PCC6803, Proc Natl Acad Sci USA 95, 9226–9231. 5. Mills, K. V., Lew, B. M., Jiang, S., and Paulus, H. (1998) Protein splicing in trans by purified Nand C-terminal fragments of the Mycobacterium

142

Dhar, Kurpiers, and Mootz

tuberculosis RecA intein, Proc Natl Acad Sci USA 95, 3543–3548. 6. Evans, T. C., and Xu, M.-Q. (2002) Mechanistic and Kinetic Considerations of Protein Splicing, Chemical Reviews 102, 4869–4884. 7. Mootz, H. D. (2009) Split inteins as versatile tools for protein semisynthesis, Chembiochem 10, 2579–2589. 8. Muralidharan, V., and Muir, T. W. (2006) Protein ligation: an enabling technology for the biophysical analysis of proteins, Nat Methods 3, 429–438. 9. Southworth, M. W., Adam, E., Panne, D., Byer, R., Kautz, R., and Perler, F. B. (1998) Control of protein splicing by intein fragment reassembly, EMBO J 17, 918–926. 10. Muona, M., Aranko, A. S., and Iwai, H. (2008) Segmental isotopic labelling of a multidomain protein by protein ligation by protein trans-splicing, Chembiochem 9, 2958–2961. 11. Kurpiers, T., and Mootz, H. D. (2007) Regioselective cysteine bioconjugation by appending a labeled cystein tag to a protein by using protein splicing in trans, Angew Chem Int Ed Engl 46, 5234–5237. 12. Kurpiers, T., and Mootz, H. D. (2008) Sitespecific chemical modification of proteins with a prelabelled cysteine tag using the artificially split Mxe GyrA intein, Chembiochem 9, 2317–2325.

13. Wu, H., Xu, M. Q., and Liu, X. Q. (1998) Protein trans-splicing and functional miniinteins of a cyanobacterial dnaB intein, Biochim Biophys Acta 1387, 422–432. 14. Brenzel, S., Kurpiers, T., and Mootz, H. D. (2006) Engineering artificially split inteins for applications in protein chemistry: biochemical characterization of the split Ssp DnaB intein and comparison to the split Sce VMA intein, Biochemistry 45, 1571–1578. 15. Telenti, A., Southworth, M., Alcaide, F., Daugelat, S., Jacobs, W. R., Jr., and Perler, F. B. (1997) The Mycobacterium xenopi GyrA protein splicing element: characterization of a minimal intein, J Bacteriol 179, 6378–6382. 16. Brenzel, S., Cebi, M., Reiss, P., Koert, U., and Mootz, H. D. (2009) Expanding the scope of protein trans-splicing to fragment ligation of an integral membrane protein: towards modulation of porin-based ion channels by chemical modification, Chembiochem 10, 983–986. 17. Saves, I., Ozanne, V., Dietrich, J., and Masson, J. M. (2000) Inteins of Thermococcus fumicolans DNA polymerase are endonucleases with distinct enzymatic behaviors, J Biol Chem 275, 2335–2341. 18. Appleby, J. H., Zhou, K., Volkmann, G., and Liu, X. Q. (2009) Novel split intein for transsplicing synthetic peptide onto C terminus of protein, J Biol Chem 284, 6194–6199.

Part II Nucleic Acid Conjugates

Chapter 10 Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo Andrea Masotti and Francesco Pampaloni Abstract Polyamine polymers are among the commonest polymers used in biomedicine. Among polyamine ­polymers, polyethylenimine (PEI) may be used as an efficient delivery vehicle for nucleic acids (DNA, RNA, etc.) or employed as a versatile imaging probe in vivo. In this chapter, the preparation of various PEI bioconjugates will be fully explained and discussed. Key words: Polyamines, Polyethylenimine, Bioconjugate polymers, DNA delivery, Imaging

1. Introduction Polyethylenimine (PEI) is a versatile organic polymer widely employed in several biomedical applications (1), as well as being an efficient gene delivery vector (Fig. 1) (2, 3). PEIs, in both the linear and branched forms, have been used for the delivery of oligonucleotides (4), DNA, small RNA, and siRNA (5). The delivery efficiency generally varies accordingly to the type of polymer structure (linear or branched), molecular weight and degree of chemical substitution. By functionalizing PEI with alkyl chains (Fig. 2) or with an appropriate dye (Fig. 3), several interesting compounds, have recently been obtained (6, 7). For example, hydrophobic PEI derivatives give vesicular structures, called polycationic liposomes, that strongly interact with DNA. These compounds show a slow rate of release of DNA in vitro. PEI derivatized with a fluorescent dye (i.e., a near-infrared dye) allows one to follow the in vivo distribution of DNA in animal models. A PEI–dye conjugate containing the near-infrared emission fluorescent dye IR-820, an indocyanine derivative, has been obtained (IR820–PEI)

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_10, © Springer Science+Business Media, LLC 2011

145

146

Masotti and Pampaloni

Fig. 1. The structure of branched polyethylenimine (PEI).

Fig. 2. Schematic of the functionalization of branched PEI with hydrophobic chains for the production of polycationic liposomes.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

147

Fig. 3. Structure of the near-infrared (NIR) dye IR820–PEI conjugate.

(Fig. 3); this conjugate appears to be particularly promising for monitoring DNA delivery in vivo. The synthesis of substituted hydrophobic compounds, in conjunction with tagging with fluorescent dyes, allows one to obtain multifunctional delivery vectors with tunable properties. Specifically, the vector’s functional properties can be varied by choosing different hydrophobic grafting molecules, by adjusting the percentage of substitution, and by employing appropriate fluorophores. Tuning the properties of delivery vectors is valuable for many biomedical applications; the preparation and use of these derivatives will be explained and discussed in this chapter.

2. Materials 2.1. Reagents

1. Branched polyethylenimine (PEI) (25 kDa) (Sigma-Aldrich) (see Note 2). 2. Dichloromethane (DCM). 3. Dimethylformamide (DMF). 4. Triethylamine (TEA). 5. Ethyl acetate (AcOEt). 6. Methanol (MeOH). 7. Dimethyl sulfoxide (DMSO). 8. Chloroform (CHCl3).

148

Masotti and Pampaloni

9. Lauryl bromide (97%). 10. Myristyl bromide (97%). 11. Cetyl bromide (97%). 12. Lauroyl chloride (97%). 13. Myristoyl chloride (98%). 14. Palmitoyl chloride (98%). 15. Indocyanine dye IR-820 (Sigma-Aldrich). 16. 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC). 17. Poly-l-aspartic acid (PAA) (15–50 kDa) (Sigma-Aldrich). 18. Hoechst 33258. 19. Tris-Borate-EDTA (TBE) Buffer (1×): 90  mM Tris-borate, 2 mM ethylenediaminetetraacetic acid (EDTA), pH 8.3. 20. NIH-3T3 fibroblast cell line (ATCC). 21. Cell tissue culture plates (24-well and 96-well). 22. Dulbecco’s Modified Eagle’s Medium (DMEM) with 10% (v/v) fetal calf serum. 23. Animal models (e.g., nude mice). 24. CellTiter-Blue™ kit (Promega). 25. Plasmid DNA encoding enhanced green fluorescent protein (pEGFP) (Clontech). 26. Plasmid DNA encoding cytomegalovirus beta-galactosidase (pCMV b-Gal) (Clontech). 27. b-Galactosidase enzyme assay kit. 2.2. Equipment

1. Ultrafiltration apparatus. 2. Regenerated cellulose membranes (10 kDa MWCO). 3. Cellulose dialysis tubing (10 kDa MWCO). 4. Rotary evaporator (rotavap). 5. Lyophilizer. 6. Gel electrophoresis apparatus equipped with an image acquisition system. 7. Nuclear magnetic resonance (NMR) spectrometer. 8. Dynamic light scattering (DLS) instrument. 9. Potentiometric titration apparatus. 10. Laminar flow hood for cell culture. 11. Controlled-environment incubator (37°C) for culturing mammalian cells. 12. Fluorescence microplate reader. 13. Fluorescence microscope.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

149

14. Transmission electron microscope (TEM) and ­Formvar-coated TEM grids. 15. Electrospray ionization (ESI) mass spectrometer (MS). 16. Optical imaging system equipped with near-infrared (NIR) fluorescence detector.

3. Methods The synthesis procedures to obtain various PEI derivatives consist of several steps that we will briefly illustrate in the following sections. We will separately describe the preparation of hydrophobic derivatives (Subheading  3.2) and a fluorescent derivative (Subheadings 3.2 and 3.4). Additional information may be found in the cited publications (see References section) 3.1. Polyethylenimine Solubilization

Branched high-molecular weight polyethylenimine (PEI) is a viscous polymer that is generally highly soluble in water, methanol, ethanol, chloroform, and other nonpolar solvents. The major drawback is the slow dissolution kinetics – branched PEI dissolves quite slowly in water. A convenient procedure to dissolve PEI is the following: 1. Use a spatula or a little spoon to collect a small amount of PEI directly from the product bottle. When ordering PEI, verify that the bottle has a wide-mouth opening. This will simplify the overall collection procedure. 2. Place the viscous polymer material into a small beaker (pretared on an analytical balance), taking care to distribute it onto one side of the beaker. 3. Weight the amount of PEI added to the beaker using an analytical balance. 4. Submerge the PEI with water (or other solvents) and place a magnetic stirrer into the beaker. 5. Stir gently until complete dissolution is achieved. Initially, stirring may become impaired by the high viscosity of the PEI. 6. Wait until the PEI material is no longer visible on the beaker’s sidewall. This indicates that the PEI has become fully dissolved.

3.2. Synthesis of PEI Derivatives by Grafting Reactions

In order to obtain PEI derivatives, the first step is to calculate the overall number of moles of the polymer to be reacted and defining the desired percentage level of substitution. For a given number of moles of grafting agent, an exact amount (moles) of PEI should be employed. However, when working with PEI ­polymers,

150

Masotti and Pampaloni

a precise molar ratio cannot be achieved, since the polymer itself does not have a single, unique molecular weight. Generally the molecular weight of a polymer such as PEI typically follows a Gaussian distribution. For example, branched PEI from SigmaAldrich (No. 408727) has a MW of ~25 kDa by light-scattering analysis or an average MW ~10 kDa by gel permeation chromatography (GPC). To calculate the molarity of a PEI solution, the common practice is to consider –(CH2–CH2–NH)– as the monomer unit (8) with a MW 43.068 Da. From this assumption, the number of monomeric units present in PEI 25  kDa is: ~25,000/43.068 = ~580. Since the monomer unit contains one nitrogen atom, it follows that PEI 25 kDa contains ~580 amino groups. As the calculated ratio of primary:secondary:tertiary amino groups in the polymer is 25:50:25, this implies that in the case of PEI 25 kDa the distribution of primary, secondary, and tertiary amino groups is, respectively, 145:290:145. This number is useful for calculating the desired percentage of substitution. For example, in order to obtain a PEI grafted at a 10% substitution level, ~58 groups have to be functionalized. As an example, we report here a detailed protocol for the functionalization of PEI with lauryl chains at a substitution percentage of 3%. The same protocol may be employed for the synthesis of other hydrophobic derivatives of PEI (at different percentage levels of substitution) by properly adjusting the stoichiometry of the starting reagents. For additional examples, see ref. (6). 1. Prepare a solution of lauryl bromide by adding 1-­bromododecane (0.178 g, 0.72 mmol) in DCM (50 mL) over a period of 3 h to a solution of PEI (1 g, 0.04 mmol) in DCM (100 mL). 2. Stir the resulting mixture at room temperature (RT) for 1 day. 3. Concentrate the solution by bringing down the solvent to a volume of ca. 15–20  mL under reduced pressure using a rotavap. The obtained product has a viscous aspect and a yellowish color. 4. The product may be purified by dialysis or by ultrafiltration (the second option is preferred). Dialyze the product against an ethanol/water (1:1) solution (4 L, with five changes of the solution), or perform ultrafiltration using 1  L of the same ethanol/water mixture with an ultrafiltration apparatus. 5. Dialyze the product again with pure water for 1 day, or perform ultrafiltration using water (3  L) with an ultrafiltration apparatus. 6. Transfer the dense, sticky oil to a round-bottom flask and proceed to lyophilize using standard procedures. Place the flask at −20°C until the product is completely frozen, then

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

151

attach the flask to a lyophilization apparatus. Lyophilize the product until all the water is removed. The final product should be a semicrystalline material. 3.3. Preparation of a Linker-Modified IR-820 Dye Intermediate [ 1]

For the preparation of fluorescent or near-infrared PEI conjugates, the same procedure used for polymer grafting is applied. However, a linker molecule between PEI and the dye must be employed as a spacer (Fig. 4). In fact, PEI is a branched polymer with amino groups localized both on the external surface and in the inner part. The coupling of the dye to the inner amino groups may result in decreased fluorescence. We have also observed that the direct conjugation of PEI with dyes and molecules may decrease the photostability of the dye itself, leading to partial degradation (data not shown). Thus, the attachment of a spacer molecule to the dye is recommended in order to avoid detrimental dye–PEI interactions. One of the most interesting dyes that we have conjugated to PEI is the near-infrared dye IR-820 (7). The IR-820 dye has a heptamethine moiety with heterocyclic nitrogen atoms bearing two

Fig. 4. Schematic of the synthesis of aminohexanoic acid-linked IR-820 ®e intermediate 1 and the NIR dye–polymer conjugate IR820–PEI.

152

Masotti and Pampaloni

alkyl-sulfonate groups. These pendant arms improve ­photostability and also provide a sphere of solvation in water that prevents dye aggregation. IR-820 dye also has a reactive chloride group. Linking an aminohexanoic acid molecule to IR-820 dye was successfully carried out to obtain a near-infrared PEI derivative (compound 1). In principle, any other molecules could be employed for this purpose. Aminohexanoic acid, having both one amino and one carboxylic group, may be used as a heterobifunctional linking molecule. In this case, we exploited the free amino group for dye functionalization to afford a versatile dye intermediate. The synthesis of compound 1 is performed as follows: 1. Place the IR-820 dye (300 mg, 0.283 mmol) and an excess of 6-aminohexanoic acid (126 mg, 1.42 mmol) inside a roundbottom flask. 2. Dissolve the mixture with 10 mL of anhydrous DMF under a nitrogen atmosphere. 3. Add an excess of TEA (198 mL, 1.42 mmol) to the mixture. 4. Using a stirring plate, incubate the flask at 85°C for 3 h. The color of the solution turns from green to blue during this time. 5. The product is recovered by solvent evaporation under reduced pressure, followed by flash chromatography purification on a silica gel column (AcOEt/MeOH from 70/30 to 0/100, v/v). 6. Characterize the linker-modified IR-820 dye molecule 1 by infrared (IR) spectroscopy, nuclear magnetic resonance (NMR) spectroscopy, and mass spectrometry (MS) (see Note 3). 3.4. Conjugation of PEI with Linker-Modified IR-820 [ 1]

The conjugation of PEI with the linker-modified dye molecule 1 follows a slightly different method compared to the alkyl chain grafting procedure (Fig. 4). In fact, since the reaction involves the formation of an amide bond, a peptide-like synthesis reaction should be performed. To this aim, a carboxylic group activator must be employed. We have found that the use of 1-ethyl-3-(3dimethylaminopropyl)carbodiimide (EDC) in the hydrochloride form is well suited to achieve this aim. This carbodiimide is water soluble and is typically employed in the pH 4.0–6.0 range. To obtain the polyethylenimine–dye molecule bioconjugate, the following steps are performed: 1. Dissolve PEI (84  mg, 0.0037  mmol) in anhydrous DMF (5 mL) and stir to the complete dissolution of the polymer. 2. Dissolve the linker-modified IR-8320 dye molecule 1 (19 mg, 0.020  mmol) in a round-bottom flask with 5  mL of DMF under a nitrogen atmosphere at room temperature.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

153

3. Add a catalytic amount of EDC to the PEI solution and ­transfer the mixture to the linker-dye molecule 1 solution. 4. Stir the reaction at room temperature for 12 h. 5. Ultrafilter the reaction mixture on a cellulose membrane (10 kDa MWCO) against methanol. The IR820–PEI conjugate product is used as an aqueous solution after evaporation of the methanol. 6. Characterize the IR820–PEI conjugate by IR spectroscopy and NMR (see Note 4). 7. Store the IR820–PEI conjugate at −20°C protected from light (see Notes 5 and 6). 3.5. Ultrafiltration and Lyophilization of PEI Derivatives

These two procedures are used to completely purify the synthesized PEI derivative compounds (obtained from Subheading 3.2) and allow an easy recovery of the final product in a semicrystalline form. Some of the reported derivatives are reduced to fine powders that can facilitate weighing. A schematic representation of an ultrafiltration apparatus is shown in Fig. 5. 1. Thoroughly mix the PEI or PEI derivative compound with 200 mL of deionized water. 2. Place the resulting mixture into the ultrafiltration cell (Fig. 5). 3. Apply a nitrogen flux (4 atm.) to the top of the apparatus. 4. Collect the permeate solution into a clean beaker or a flask.

Fig. 5. Ultrafiltration apparatus used to purify the PEI derivative compounds after synthesis.

154

Masotti and Pampaloni

5. Collect the retentate solution containing the ultrafiltered polymer product. 6. Lyophilize the retentate solution to obtain a semicrystalline solid product. 3.6. 1H NMR Spectroscopy of PEI Derivatives

To assess the purity of the synthesized PEI derivative compounds (obtained from Subheading 3.2), NMR spectroscopy is the technique of choice. 1H and 13C NMR spectra were recorded on a 300-MHz NMR instrument using default acquisition parameters and standard deuterated solvents (water, chloroform, and reference materials). Several indications are reported to simplify the overall data acquisition procedure. 1. Dissolve a small amount of the derivatized polymer sample in D2O at a pH 14. 2. Place the solution into an NMR sample tube. 3. Acquire a 1H NMR spectrum of the sample according to your instrument set up. Integration of the proton magnetic resonance (1H NMR) spectrum of the product should indicate ~3 mol % of lauryl groups (C12H25; 1–1.6  ppm) per residue mol of ethylenimine unit (C2H4NH; 2.2–3.2 ppm) in the polymer. The substituted polymer may therefore be represented by the stoichiometric formula (C2H4NH)m(C12H25)0.03m, m = 580. The 1H and 13C NMR spectra of the product should be very close to the following example. 1H NMR (D2O): d (ppm) 0.92 (b, CH3), 1.30 (b, –(CH2)8–), 2.90 (b, –CH2CH2N–). 13 C NMR (D2O): d (ppm) 14.4, 23.1, 29.9, 30.2, 32.3, 37.8, 39.5, 44.0, 45.9, 47.7, 49.3, 51.4, 52.0, 53.1, 54.2, 69.2, 75.8.

3.7. Potentiometric Titration and Data Analysis of PEI Derivatives

PEI and PEI derivative polymer samples (obtained from Subheading 3.2) at a concentration of 45 mM (monomer aqueous solution) were titrated in a jacketed cell at 25°C at 0.1  M ionic strength (KCl) using a potentiometric apparatus equipped with a burette for automatic titration to control the addition of acid or base. The concentration of the HCl and NaOH solutions was standardized following reported procedures (9). 1. Place 10–20 mg of the PEI derivative into the potentiometric cell and dissolve it with 25 ml of KCl 0.1 M. 2. After dissolution, begin titrating the sample with HCl since the starting pH of the solution is alkaline. 3. Register and plot the pH values during the titration process for determination of the acid-base profile. As a representative example, in Fig. 6 we show the titration of a PEI derivative obtained by grafting the polymer with lauryl chains at different percentages of substitution.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

155

Fig.  6. Representative titration data for PEI derivatives obtained by grafting PEI with dodecane chains at different percentage levels of substitution.

3.8. Agarose Gel Electrophoresis of DNA Polyplexes and Polyanion Exchange Reaction Assessment

Agarose gel electrophoresis was performed with a common gel electrophoresis apparatus equipped with a power supply. A digital camera and a digital recorder were used in all the described experiments to record the gel images. When preparing polymer/DNA complexes (polyplexes), the N/P ratio should be defined: The N/P ratio indicates the ionic balance between the PEI amino groups (N) and the DNA phosphate groups (P) in their complexes. The ratio is calculated based on considering that 1 mg DNA corresponds to 3 nmol of phosphate, and that 1 ml of PEI (or PEI derivative) at a concentration of 10 mM (monomers) corresponds to 10 nmol of amine nitrogen.

3.8.1. Agarose Gel Electrophoresis of Polymer/DNA Polyplexes

Gel shift assays of PEI and PEI derivatives complexed with DNA (polyplexes) are performed as follows: 1. Prepare a stock solution (10 mg/ml) of the intercalating dye Hoechst 33258 by dissolving the pure solid in a mixture of DMSO/H2O (2:1, v/v). Add 10 ml of dye stock solution to each 50 ml of 0.8% (w/v) agarose gel solution (in 1× TBE) used to cast the electrophoresis gel (see Note 1). 2. Prepare the PEI or PEI derivative solutions (1 mg/mL). 3. Add 1 mg plasmid DNA (e.g., pCMV b-Gal) to each mL of polymer solution. 4. Incubate the polymer/DNA mixture at 37°C for 5 min.

156

Masotti and Pampaloni

Fig. 7. Gel retardation assays (agarose 0.8%, 1× TBE, Hoechst 33258 staining) of ­polymer/ DNA polyplexes. (a) Analysis of polyplexes containing DNA and PEI or PEI derivatives with different percentage levels of substitution. All polyplex complexes were formed at a N/P ratio = 15.5. Samples: Reference DNA (without polymer) (Lane 1); PEI/DNA (Lane 2 ); Polymer/DNA complexes with PEI grafted with cetyl chains at 3% (Lane 3 ), 6% (Lane 4 ), 13% (Lane 5 ), and 30% (Lane 6 ) substitution. (b) Analysis of PEI grafted with myristoyl chains (3% substitution) complexed with DNA at N/P ratio = 15.5 after the addition of poly-aspartic acid (PAA). Samples: DNA molecular weight standards (ladder) (Lane 1); reference DNA (without polymer) (Lane 2 ); Polymer/DNA complexes after PAA addition: 10 min (Lane 3 ), 1 h (Lane 4 ), 2 h (lane 5 ), 3 h (lane 6 ), 18 h (lane 7 ) and 24 h (lane 6 ).

5. Add gel loading buffer to the polymer/DNA mixture. Load the sample onto the agarose gel (prepared in step 1) and run the gel for 45 min at 100 V (Fig. 7a). 3.8.2. Polyanion Exchange Reaction Assessment

In order to check the DNA complexing ability of PEI and the PEI derivatives, poly-l-aspartic acid (PAA) is used to competitively displace DNA from the polyplexes. Agarose gel electrophoresis is performed following the steps below: 1. Prepare the DNA/PEI or DNA/PEI derivative complexes as described above in Subheading 3.8.1. DNA is generally added to the PEI or PEI derivative solution (and not the other way around), as this facilitates the formation of DNA–polymer complexes. 2. Add 2 mg of poly-l-aspartic acid (PAA) for each microgram of DNA in the DNA/PEI or DNA/PEI derivative complexes. Incubate the mixtures at 37°C for 5 min.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

157

3. Add running buffer to each sample and load the gel with solutions collected at 10 min, 1 h, 2 h, 3 h, 18 h, and 24 h after the addition of PAA (Fig. 7b). 3.9. Dynamic LightScattering Analysis of Hydrophobic PEI Derivatives and Their DNA Polyplexes

Dynamic light-scattering (DLS) experiments were performed with a classical setup equipped with a correlator operating with logarithmic sample time spacing. The light source was a 10-mW He–Ne laser (wavelength 632.8 nm) and the data were acquired at a scattering angle of 90°. The size and size distribution of the diffusing particles were found using the standard data analysis program CONTIN, based on fitting a continuous distribution of exponential decay times. The light-scattering data were used to assess whether the PEI derivatives or their DNA complexes (polyplexes) are able to form vesicular structures (polycationic liposomes) due to the presence of hydrophobic chains within the polymeric scaffold. 1. Weight 4–5 mg of a hydrophobic derivative of PEI (e.g., containing lauryl, myristyl, and cetyl chains) in a round-bottom flask. 2. Dissolve the product in a mixture of CHCl3/MeOH (3:1, v/v) (10 ml). 3. Evaporate the solvent with a rotavap until a thin film is obtained on the flask wall. 4. Add distilled water and vortex until the film is completely swelled and detached from the flask wall. 5. Transfer the sample into a light-scattering cuvette and start the data acquisition. As a representative example, Fig. 8a shows the dynamic lightscattering measurements for PEI derivatives containing lauryl, myristyl, and cetyl chains. The light-scattering profile of ­polyplexes obtained by grafting PEI with lauryl chains (3% of substitution) at different N/P ratios: (a) N/P = 15.5, (b) N/P = 11.6, (c) N/P = 7.75, (d) N/P = 3.87 is also reported (Fig. 8b). From these measurements we concluded that functionalizing PEI with hydrophobic chains imparts the polymer with the ability to form vesicular structures resembling liposomes. For this reason, these structures are also called polycationic liposomes.

3.10. Freeze-Fracture Microscopy Analysis

Hydrophobic PEI derivatives giving vesicular structures (as observed by light scattering) may also be examined by freeze-fracture microscopy. Conveniently, the same samples prepared for analysis by light scattering may also be used for freeze-fracture analysis. 1. Soak the PEI derivative sample with 30% (v/v) glycerol, and freeze them in partially solidified Freon 22. 2. Fracture the material in a freeze-fracture device (−105°C, 10−6 mmHg).

158

Masotti and Pampaloni

Fig. 8. Dynamic light-scattering measurements of (a) hydrophobic PEI derivatives (lauryl (a ), myristyl (b ) and cetyl (c )) at 3% substitution and (b) PEI derivatives obtained by grafting PEI with dodecane chains at 3% substitution under different N/P ratios: N/P = 15.5 (a), N/P = 11.6 (b), N/P = 7.75 (c), and N/P = 3.87 (d ).

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

159

Fig. 9. Freeze fracture images of a PEI derivative obtained by grafting PEI with tetradecane chains at 6% substitution outlining the polycation assembly into vesicles.

3. Replicate the sample by vacuum deposition of platinum/­ carbon using an electron beam gun. 4. Extensively wash the replica with distilled water and transfer it to a Formvar-coated TEM grid. 5. Examine the replica using a transmission electron microscope. As a representative example, Fig.  9 shows a freeze-fracture image of a PEI derivative obtained by grafting PEI with ­tetradecane chains at 6% substitution. The outlines of features resulting from the assembly of the PEI derivative into polycationic vesicles are visible. 3.11. Cytotoxicity, Transfection Assays, and Fluorescence Microscopy Analysis

Internalization assays were performed using mouse NIH-3T3 fibroblast cells to evaluate the cytotoxicity of PEI and hydrophobic PEI derivative compounds. The transfection assays were performed using PEI and PEI derivatives complexed with two reporter plasmids: The first plasmid encodes cytomegalovirus beta-galactosidase (pCMV b-Gal), and the second plasmid encodes enhanced green fluorescent protein (pEGFP). Both plasmids were obtained after bacterial amplification and purification from culture medium following standard protocols. Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% (v/v) fetal calf serum was used in all cytotoxicity and transfection assays.

160

Masotti and Pampaloni

Fig.  10. NIH-3T3 cell viability assay results for PEI and a hydrophobic PEI derivative obtained by grafting PEI with lauryl chains at 3% substitution. 3.11.1. Cytotoxicity Assessment of PEI and PEI Derivatives

1. Plate  1 × 105 cells in a 96-well cell culture plate. Allow the cells to grow to 80% confluence. 2. Add gene delivery vehicle test compounds (hydrophobic PEI derivatives) or vehicle controls (unmodified PEI) (suspended in 100 ml medium) to each culture well. 3. Incubate the cells with the test compounds and controls in a mammalian cell culture incubator for 3 h at 37°C. 4. Remove the cell culture plates from the 37°C incubator and add CellTiter-Blue™ reagent (20  ml/well) to a blank well, wells containing treated cells, and wells containing untreated control cells. 5. Shake the cells with the CellTiter-Blue™ reagent for 10 s. 6. Incubate the cells using standard cell culture conditions for 1–4 h. 7. Shake the plate for 10  s and record the fluorescence (EX l = 560 nm, EM l = 590 nm). 8. Follow the manufacturer’s instructions for the CellTiter-Blue™ kit to calculate the results of the cell viability test (Fig. 10).

3.11.2. Transfection Assays with Polymer/DNA Polyplexes

1. Plate  2 × 105  cells in a 24-well cell culture plate. Allow the cells to grow to 80% confluence. 2. Prepare 500 mL of a solution containing 1 mg of pEGFP and 1 mg of pCMV b-Gal in medium to serve as a control for the evaluation of transfection efficiency. 3. In a separate tube, prepare 500 mL of a solution containing 1 mg of pEGFP and 1 mg of pCMV b-Gal complexed with

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

161

2  mg of PEI or equivalent amounts of the different ­hydrophobic PEI derivatives. 4. Incubate the solutions prepared in steps 2 and 3 above at RT for 10 min. 5. Add the DNA and polymer/DNA polyplex solutions (500 mL) to the desired wells in the culture plate containing cells. 6. Incubate the cells for 3 h and then replace the medium with fresh medium. 7. Examine the GFP fluorescence under a fluorescence microscope to confirm that transfection has occurred. 8. Assess b-galactosidase activity in the whole cell extracts by following the instructions provided with the employed b-galactosidase enzyme assay kit (Fig. 11a).

Fig. 11. (a) b-gal expression efficiency of NIH-3T3 cells transfected with PEI and a PEI derivative obtained by grafting PEI with hexadecane chains. The b-gal activity is measured in OD units 24 h after transfection. (b) Fluorescence microscopy of NIH-3T3 cells transfected with PEI and a PEI derivative obtained by grafting PEI with hexadecane chains.

162

Masotti and Pampaloni

3.11.3. Fluorescence Microscopy

1. Perform the procedures described in steps 1–5 in Subheading 3.11.2 above. 2. Harvest the cells 24 h after transfection. 3. Fix the cells and stain with Hoechst 33258 (1 mg/ml in PBS, pH 7.4). Hoechst 33258 fluorescently stains double-stranded DNA in cells. 4. Examine the cells under a fluorescence microscope using 40× magnification (Fig. 11b).

3.12. NIR Optical Imaging

Near-infrared (NIR)-emitting dyes represent an intriguing avenue for extracting biological information from living subjects since they can be monitored with safe, noninvasive optical imaging and contrast techniques (10, 11). Optical imaging represents a rapidly expanding field, with direct applications in ­pharmacology, molecular and cellular biology, as well as diagnostics. For the ­evaluation of DNA delivery using PEI derivatives, we used a standard NIR optical imaging device to image nude mice injected with a solution of fluorescent conjugates (Fig. 12). 1. Inject nude mice (200–250 g each) via the caudal vein with a 2 mg/ml solution of fluorescent conjugates at a 0.015 mg/g dosage. 2. After injection, place the anesthetized animal under the optical imaging instrument. 3. Acquire a photograph of the animal before irradiation to serves as the “blank” image. The default imaging parameters used in this work were as follows: Exposure time = 0.2  s; ­binning = 4; f = 8; field of view (FOV) = 12.8 × 12.8 cm.

Fig. 12. Optical images of nude mice subjected to tail vein injection of various polymer and polymer/DNA complexes. The images show the distribution of the IR820–PEI vector alone, the IR820–PEI/DNA complex and PEI/IR820–DNA. A threedimensional (3D) reconstruction of a mouse body indicating the organs of interest (brain, lungs, and liver) is also shown.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

163

4. Acquire NIR optical images for 3 h at a rate of 1 image/min. The default imaging parameters used in this work were as follows: Excitation filter ICG (710–760  nm); emission filter ICG (810–875 nm); exposure time = 1 s; binning = 16, f = 2, FOV = 12.8 × 12.8 cm.

4. Notes 1. Replacing ethidium bromide (EtBr) with Hoechst 33258 in the agarose gel. For the agarose gel electrophoresis procedure, a slight modification has been introduced with respect to the classical setup. Generally, EtBr is added to the agarose gel during gel casting. As a result, the nucleic acids incorporate the dye during electrophoresis and become fluorescent under UV illumination. However, some authors have reported that when DNA/EtBr interacts with PEI, an almost complete quenching of the dye fluorescence is observed (12). This quenching effect is reduced by ~50% when a minor groove-dye such as Hoechst 33258 is employed. For this reason, we incorporated Hoechst 33258 instead of EtBr during gel casting. 2. Commercially available high-molecular weight PEI polymers may contain different amounts of water. Linear PEI (25 kDa) is available as a pure crystalline solid, while the branched form is a viscous liquid containing traces of water. PEI with a molecular weight of 500 kDa is generally available as a 50:50 mixture with water. Several other different formulations for PEI may be found commercially; therefore, carefully check the amount of water present in the compound and calculate the exact concentration of the polymer accordingly. 3. The linker-modified IR-820 dye molecule 1 shows the following characteristics: IR (film): 1,718, 1,740 cm−1. 1H NMR (DMSO-d6): d (ppm) 8.10 (m, 3H), 7.89 (m, 4H), 7.66 (d, J = 12  Hz, 2H), 7.50 (m, 4H), 7.31 (m, 2H), 5.67 (d, J = 12 Hz, 2H), 3.96 (m, 4H), 3.05 (m, 4H), 2.53 (m, 4H), 1.96 (m, 4H), 1.86 (s, 12H), 1.76 (m, 6H), 1.68 (s, 6H), 1.44 (m, 4H). 13C NMR (DMSO-d6): d (ppm) 177.2, 166.6, 165.0, 162.5 (2C), 140.9, 130.2, 129.9 (4C), 128.4 (4C), 127.3 (4C), 123.3 (4C), 121.7 (4C), 110.8 (2C), 93.7, 51.0 (2C), 50.2, 48.8 (2C), 38.2, 37.1, 36.0, 35.9, 30.9, 30.7, 27.7 (4C), 27.5, 26.4, 25.7, 25.2 (2C), 22.6 (2C). The elemental analysis calculated for C52H62N3NaO8S2 is: C, 66,15; H, 6,62; N, 4,45; Na, 2,43; O, 13,56; S, 6,79%. Found: C, 66, 23; H, 6,87; N, 4,47; S, 6,77%. The ESI-MS spectra resulted in a peak with m/z = 921.56 relative to [M-Na]−. 4. The conjugation reaction between PEI and the linker-modified dye molecule 1 yields the derivative IR820–PEI, which shows

164

Masotti and Pampaloni

the following characteristics: IR (CHCl3): 1,670  cm−1. The 1 H NMR spectrum (CDCl3) of IR820–PEI was poorly resolved, but the signal integration indicated the presence of 1 mol % of IR820 (CH2; 0.9 ppm) per residue mol of ethylenimine unit (C2H4NH; 2.2–3.2 ppm) in the polymer. The substituted polymer may be represented by the stoichiometric formula (C2H4NH)m(C52H61N3NaO7S2)0.01m, m = 580. 5. When storing the dye–polymer conjugate, dye photobleaching processes should be prevented. Hence, we recommend wrapping aluminum foil all around the container (bottle, flask, vials, etc.) used to store the product and keeping the product at −20°C. 6. A brief comment regarding collateral NIR dye oxidation: Bouteiller et al. recently described the synthesis and the photophysical properties of two novel near-infrared (NIR) ­cyanine dyes, NIR5.5-2 and NIR7.0-2. These two dyes are water soluble and are potential substitutes for the commercially available Cy5.5 and Cy7.0 dyes (13). It has been found that meso-chlorine derivatives of stabilized Cy7.0-like dyes conjugated to amino acids are not stable in Milli-Q grade water; in particular, the cleavage of the amino acid residue and subsequent conversion of the dye into an alpha-hydroxy ketone (or related hydroperoxide) has been observed. The decomposition mechanism is similar to the one that has been recently suggested by Toutchkine et al. (14) to explain the photodecomposition of merocyanine dyes.

Acknowledgments The Ministero della Salute and MIUR (Ministero dell’Istruzione, dell’Università e della Ricerca) are gratefully acknowledged for providing financial support for this work. FP also thanks Ernst H.K. Stelzer for support and many interesting discussions. References 1. Vicennati, P., Giuliano, A., Ortaggi, G., and Masotti, A. (2008) Polyethylenimine in medicinal chemistry. Curr. Med. Chem. 15, 2826–2839. 2. Eliyahu, H., Barenholz, Y., and Domb, A. J. (2005) Polymers for DNA delivery. Molecules 10, 34–64. 3. Park, T. G., Jeong, J. H., and Kim, S. W. (2006) Current status of polymeric gene delivery systems. Adv. Drug Deliv. Rev. 58, 467–486.

4. Lemkine, G. F., and Demeneix, B. A. (2001) Polyethylenimines for in  vivo gene delivery. Curr. Opin. Mol. Ther. 3, 178–182. 5. Jere, D., Jiang, H. L., Arote, R., Kim, Y. K., Choi, Y. J., Cho, M. H., Akaike, T., and Cho, C. S. (2009) Degradable polyethylenimines as DNA and small interfering RNA carriers. Expert Opin. Drug Deliv. 6, 827–834. 6. Masotti, A., Moretti, F., Mancini, F., Russo, G., Di Lauro, N., Checchia, P., Marianecci, C.,

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo Carafa, M., Santucci, E., and Ortaggi, G. (2007) Physicochemical and biological study of selected hydrophobic polyethyleniminebased polycationic liposomes and their complexes with DNA. Bioorg. Med. Chem. 15, 1504–1515. 7. Masotti, A., Vicennati, P., Boschi, F., Calderan, L., Sbarbati, A., and Ortaggi, G. (2008) A novel near-infrared indocyanine dye-polyethylenimine conjugate allows DNA delivery imaging in  vivo. Bioconjug. Chem. 19, 983–987. 8. Boussif, O., Lezoualc’h, F., Zanta, M. A., Mergny, M. D., Scherman, D., Demeneix, B., and Behr, J. P. (1995) A versatile vector for gene and oligonucleotide transfer into cells in culture and in  vivo: Polyethylenimine. Proc. Natl. Acad. Sci. USA. 92, 7297–7301. 9. Gans, P., and O’Sullivan, B. (2000) GLEE, a new computer program for glass electrode calibration. Talanta 51, 33–37. 10. North Atlantic Treaty Organization. Scientific Affairs Division. (1998) Near-infrared dyes for high technology applications. In: Daehne, S., Resch-Genger, U., and Wolfbeis, O.S. (Eds.)

165

Proceedings of the NATO Advanced Research Workshop on Syntheses, Optical Properties and Applications of Near-Infrared (NIR) Dyes in High Technology Fields, Trest, Czech Republic, September 24–27, 1997. Kluwer, Dordrecht, London. 11. Mishra, A., Behera, R. K., Behera, P. K., Mishra, B. K., and Behera, G. B. (2000) Cyanines during the 1990s: A review. Chem. Rev. 100, 1973–2012. 12. Wiethoff, C. M., Gill, M. L., Koe, G. S., Koe, J. G., and Middaugh, C. R. (2003) A fluorescence study of the structure and accessibility of plasmid DNA condensed with cationic gene delivery vehicles. J. Pharm. Sci. 92, 1272–1285. 13. Bouteiller, C., Clave, G., Bernardin, A., Chipon, B., Massonneau, M., Renard, P. Y., and Romieu, A. (2007) Novel water-soluble near-infrared cyanine dyes: Synthesis, spectral properties, and use in the preparation of internally quenched fluorescent probes. Bioconjug. Chem. 18, 1303–1317. 14. Toutchkine, A., Nguyen, D. V., and Hahn, K. M. (2007) Merocyanine dyes with improved photostability. Org. Lett. 9, 2775–2777.

Chapter 11 Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition Gwladys Pourceau, Albert Meyer, Jean-Jacques Vasseur, and François Morvan Abstract A glycomimetic oligonucleotide conjugate bearing four galactose residues on a mannose core is ­synthesized using oligonucleotide solid-phase synthesis and Cu(I)-catalyzed azide-alkyne 1,3-dipolar cycloaddition (CuAAC, or “click” chemistry). To achieve this purpose, new building blocks (including the solid support and phosphoramidites) are synthesized and used on a DNA synthesizer to generate a tetraalkyne oligonucleotide, which is then conjugated with a galactose azide derivative by click chemistry to afford the desired 3¢-tetragalactosyl-mannose oligonucleotide conjugate. The procedures described in this chapter provide a general approach for the synthesis of novel glycoconjugates that can be immobilized to a DNA chip via DNA-directed immobilization to study, for example, their multivalent interactions with lectins in cellular targeting/uptake, etc. Key words: Solid-phase automated oligonucleotide synthesis, 1,3-Dipolar cycloaddition, Click chemistry, Glycoconjugate, Glycocluster, Glycomimetic, Tetraalkyne oligonucleotide, Galactose azide

1. Introduction Interactions between carbohydrates and sugar-binding proteins – known as lectins – are involved in many biological processes and play a leading role in cellular recognition events such as those involved in viral and bacterial adhesion (1–3). For example, Pseudomonas aeruginosa (PA) is an opportunistic pathogen that is largely involved in the morbidity and mortality of cystic fibrosis and immunocompromised patients. This clinically important bacterium expresses two soluble lectins, PA-IL and PA-IIL, which specifically recognize d-galactose and l-fucose, respectively (4). Treatment

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_11, © Springer Science+Business Media, LLC 2011

167

168

Pourceau et al.

with the corresponding saccharide derivatives has been shown to be potent against acute pneumonia in mice models (5); however, in order to find highly effective inhibitors of PA, the synthesis of novel glycomimetics of galactose is required (6). Furthermore, it is necessary to be able to characterize the interactions between glycomimetics and lectins to understand the relevant biological processes and develop new therapeutic strategies. To achieve this purpose, researchers need to adequately address several different challenges. First, the great structural diversity of monosaccharides and their large number of branching sites generally make polysaccharide synthesis procedures difficult to perform. Moreover, the interactions between lectins and carbohydrates are relatively weak and require multipresentation of the relevant residues to be significant (7); hence, many types of glycoclusters have been designed and reported in literature. To overcome the synthesis difficulties and to take advantage of the so-called cluster effect, a number of simple carbohydrate derivatives have been conjugated with various scaffolds such as linear (8) or cyclic (9) peptides, calixarenes (10, 11), oligosaccharides (12), pentaerythritol (13), and cyclodextrins (14); however, very few examples of glycoclusters based on DNA oligonucleotide chemistry currently exist (15–20). Thanks to the phosphoramidite method developed by Beaucage and Caruthers in 1981 (21), the use of solid-phase oligonucleotide synthesis and the development of automated DNA synthesizers has grown considerably. We thus decided to take advantage of these efficient synthetic tools and combined them with methods based on Cu(I)-mediated azide-alkyne 1,3-dipolar cycloaddition (CuAAC) (22, 23) to synthesize new glycocluster conjugates using a non-nucleosidic alkyne phosphoramidite and an azide solid support. In addition, the glycocluster conjugates can be immobilized to a DNA chip via DNA-directed immobilization (DDI) (24, 25) to study, for example, their multivalent interactions with lectins in cellular targeting/uptake, etc. In this chapter, we describe a strategy to synthesize a novel 3¢-tetragalactose oligonucleotide conjugate using a tetraalkyne mannose core that is conjugated with a galactose azide derivative by CuAAC. To achieve the synthesis of the conjugate, it is necessary to synthesize previously the following: (1) a solid support exhibiting an azide group and a protected hydroxyl group (26); (2) 1-O-propargyl-a-d-mannopyranose (27, 28) for its introduction onto the solid support by CuAAC, thus providing a solidsupported mannose core with four free hydroxyls; (3) a pentynyl phosphoramidite (26) to introduce alkyne groups into the mannose core; and (4) a galactose azide derivative (29) for its introduction four times by a second round of CuAAC reactions onto the tetraalkyne mannose core present at the 3¢-end of the oligonucleotide synthesized by solid phase. Using these specific building blocks, we describe herein the synthesis of a 3¢-tetragalactose oligonucleotide conjugate; however, by modifying the relevant

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

169

procedures accordingly, it is obviously possible to introduce other types of saccharides using the corresponding azide derivatives. Since the CuAAC reaction is highly efficient, similar yield levels can be expected in such cases regardless of the particular type of saccharide azide derivative that is utilized. Moreover, by following similar procedures, non-saccharidic azide derivatives could also be used to afford the synthesis of new oligonucleotide conjugates exhibiting other types of biomolecules such as peptides, amines/polyamines, lipids, and biotinylated ligands (30).

2. Materials 2.1. Equipments

1. Rotary evaporator. 2. Silica gel (0.04–0.06 mm particle size). 3. Vacuum desiccator with phosphorous pentoxide (P2O5). 4. UV–Visible spectrophotometer. 5. Microwave synthesizer with 0.2–0.5-mL vials, caps, and crimper. 6. ABI 394 DNA synthesizer with synthesis columns (Applied Biosystems). 7. SpeedVac® vacuum concentrator system. 8. Matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometer. 9. Reversed-phase high-performance liquid chromatography (HPLC) system. 10. Reversed-phase C18 (5 mm) column (150 × 4.6-mm; for analytical HPLC). 11. Reversed-phase C18 (15  mm) column (300 × 7.8-mm; for preparative-scale HPLC purification). 12. NAP™-10 disposable columns prepacked with Sephadex™ G-25 (GE Healthcare).

2.2. Synthesis of Carbohydrate Derivatives 9 and 19

1. Peracetylated d-mannose. 2. Peracetylated d-galactose. 3. Propargyl alcohol. 4. Boron trifluoride etherate (BF3∙OEt2). 5. 1,4-Cyclohexanedimethanol. 6. p-Toluenesulfonyl chloride. 7. Sodium azide (NaN3). 8. Sodium iodide (NaI).

170

Pourceau et al.

2.3. Synthesis of Modified Solid Support 6

1. 1,1,1-Tris(hydroxymethyl)ethane. 2. 4,4¢-Dimethoxytrityl chloride. 3. Sodium hydride (NaH). 4. Sodium iodide (NaI). 5. 1,6-Dibromohexane. 6. Tetramethylguanidinium azide (TMGA). 7. 4-Dimethylaminopyridine (DMAP). 8. N-(3-Dimethylaminopropyl)-N ¢-ethylcarbodiimide (DEC). 9. Hydroquinone-O,O ¢-diacetic acid (“Q-linker”). 10. Long-chain alkylamine controlled-pore glass (LCAA-CPG, 500 Å, 135 mmol amino group/g loading). 11. Acetic anhydride. 12. Cap A solution: Acetic anhydride/dry pyridine/THF, 10/10/80 (v/v/v). 13. Cap B solution: 10% (v/v) N-Methylimidazole in dry THF.

2.4. Synthesis of O-Pent-4-ynyl-O ¢(2-cyanoethyl)N,N ¢-diisopropylphos­ phoramidite 12 2.5. DNA Synthesis

Pent-4-yn-1-ol 1. Diisopropylammonium tetrazolide (DIAT). 2. 2-Cyanoethyl-(N,N,N¢,N¢-tetraisopropyl)phosphorodia­ midite. 1. Standard 2-cyanoethyl deoxyribonucleoside phosphoramidites can be obtained from commercial suppliers. 2. Activator solution: 0.3 M 5-Benzylthio-1H-tetrazole (BMT) in dry acetonitrile. 3. Oxidation solution: 0.1  M Iodine in 70/10/20 (v/v/v) ­tetrahydrofuran (THF)/pyridine/water. 4. Cap A solution: Acetic anhydride/dry pyridine/THF, 10/10/80 (v/v/v). 5. Cap B solution: 10% (v/v) N-Methylimidazole in dry THF. 6. Deblocking solution: 3% (v/v) Dichloroacetic acid in dry CH2Cl2.

2.6. Click Reactions

1. Copper sulfate (CuSO4). 2. Sodium ascorbate.

2.7. Dry Solvents

1. Pyridine. 2. Methylene chloride (CH2Cl2). 3. Dry and extra dry acetonitrile (CH3CN). 4. Triethylamine (Et3N). 5. Tetrahydrofuran.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

2.8. “For Synthesis” Quality Solvents

171

1. Pyridine. 2. Methylene chloride (CH2Cl2). 3. Acetonitrile (CH3CN). 4. Triethylamine (Et3N). 5. Cyclohexane. 6. Ethyl acetate (EtOAc). 7. Methanol (MeOH). 8. Toluene. 9. Acetone. 10. Petroleum ether (Ep). 11. Concentrated aqueous ammonia solution (NH4OH). 12. Dimethylformamide (DMF).

2.9. Buffers and Other Solutions

1. Saturated aqueous sodium bicarbonate solution. 2. Saturated aqueous sodium chloride solution (brine). 3. 0.05 M Triethylammonium acetate, pH 7 (TEAAc). 4. 0.1 M p-Toluenesulfonic acid in acetonitrile (CH3CN). 5. Sulfuric acid (H2SO4) stain solution: Ethanol (EtOH)/H2SO4 (90:10, v/v). 6. Vanillin stain solution: Add 5  g of vanillin and 25  mL of H2SO4 to 225 mL of MeOH/H2O (1:1, v/v).

3. Methods The first few sections described below outline the synthesis, purification, and characterization/quantification of the following building blocks: the mannose-core solid support 10 (Subheading 3.1); the O-pent-4-ynyl-O ¢-(2-cyanoethyl)-N,N-diisopropylphosphoramidite 12 (Subheading 3.2); the 3¢-(tetrapentynylphosphodiester mannosyl) oligonucleotide 14 (Subheading  3.3); and the 1-O-[4(azidomethyl)cyclohexyl-1-methyl]-2,3,4,6-tetra-O-acetylb-d-galactopyranose 19 (Sub­heading 3.4.1).The conjugation of the galactosyl azido derivative 19 by CuAAC with the 3¢-(tetrapentynylmannose)-oligonucleotide intermediate 14 to generate the final 3¢-tetragalactose oligonucleotide conjugate product 20 (containing a mannose core) is described in Subheading 3.4.2. 3.1. Preparation of Mannose-Core Solid Support 10

The synthesis of the mannose-core solid support 10 is described below in Subheadings  3.1.1–3.1.3. The synthesis procedure requires first the preparation of the solid support azide 6 (26) and the 1-O-propargyl-b-d-mannopyranose 9 (27, 28), which are then reacted together by microwave-assisted CuAAC to provide 10 (Fig. 1).

172

Pourceau et al. HO

HO

DMTrCl

HO

Pyr

OH

Br (CH2)6 Br

DMTr O

1

OH

Br (CH2)6 O

THF NaH

DMTr O

2

3 NH2 N

O HO2C N3 (CH2)6

O

N

N3 (CH2)6

O

DMTr

O

CH3CN

N3

DEC pyr DMAP

O

OH 4

O

O

O

5

CO2H

O

O

DMTr

OH

OAc

CO2H

AcO O

AcO AcO NH2

CPG

N3 (CH2)6

OAc 7

DEC pyr DMAP

BF3. Et2O

OH

CH2Cl2

O

DMTr

O

OAc

O O

O

O

HN

AcO O

AcO AcO

8

O

6

O NH4OH MeOH

OH HO

CuSO4 Na Ascorbate H2O/MeOH MW

HO HO 9

OH HO HO HO

O

N

N

N (CH2)6

O

O

DMTr 10

O

O O

O

O

HN O

Fig. 1. Synthesis of the solid support mannose core 10.

O O

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition 3.1.1. Preparation of Azido Solid Support 6 3.1.1.1. Dimethoxytritylation of Tris(hydroxymethyl) ethane 1

173

1. In a 500-mL round-bottom flask, coevaporate 3.6  g (30 mmol, 2 eq.) of 1,1,1-tris(hydroxymethyl)ethane 1 three times, each with 50  mL of dry pyridine. Dissolve the final evaporated residue in 250 mL of dry pyridine. 2. While stirring, add 5.1 g (15 mmol, 1 eq.) of 4,4¢-dimethoxytrityl chloride under an argon (Ar) atmosphere. 3. Stir the reaction mixture overnight. Monitor by thin layer chromatography (TLC) (see Notes 1–4) and stain with sulfuric acid solution. (TLC: CH2Cl2/MeOH/Et3N, 99:1:1 v/v/v; Rf DMTrCl = 0.7; Rf2 = 0.15.) 4. Stop the reaction with 10  mL of MeOH to hydrolyze any remaining 4,4¢-dimethoxytrityl chloride which did not react. Concentrate the mixture to dryness on a rotary evaporator under reduced pressure and dissolve the residue in 100 mL of CH2Cl2. 5. Pour the reaction mixture into a 250-mL separatory funnel and wash the organic layer twice with 100 mL of saturated aqueous sodium bicarbonate solution. 6. Collect the organic layer, dry it by adding 10 g of anhydrous sodium sulfate, and then filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure and coevaporate with toluene to remove pyridine residues; a yellow oil is obtained. 7. Dissolve the crude product in a minimal volume of CH2Cl2 and apply the solution to a 5-cm diameter chromatography column containing 120 g of silica gel equilibrated firstly with 250 mL of CH2Cl2/Et3N, (99:1, v/v) and then with 250 mL of CH2Cl2 (see Note 5). 8. Gradually increase the concentration of acetone (from 1 to 50%, v/v) in CH2Cl2. 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain a foam. 11. Check the purity of the synthesized compound by 1H, NMR and MS.

C

13

2-[(4,4¢-Dimethoxytrityl)oxymethyl]-2-methylpropane-1,3diol 2: Yield 53%. TLC (CH2Cl2/MeOH/Et3N, 99:1:1 v/v/v): Rf = 0.15. 1H NMR (CDCl3, 400 MHz): 0.87 (3H, s), 2.42 (2H, br s), 3.16 (2H, s), 3.69 (2H, d, J = 25.2  Hz), 3.64 (2H, d, J = 25.2 Hz), 3.82 (6H, s), and 6.79–7.49 (13H, m). 13 C NMR (CDCl3, 100 MHz): 17.4, 41.1, 55.2, 67.2, 68.2, 86.3, 113.3, 126.9, 128.0, 130.0, 135.8, 144.7, and 158.5. HRMS FAB (positive mode, nitrobenzylic alcohol) m/z: calcd for C26H31O5 [M + H]+ 422.2093; found 422.2098.

174

Pourceau et al.

3.1.1.2. Alkylation of 1-O-[(4,4¢Dimethoxytrityl) oxymethyl]-2methylpropane-1,3-diol 2

1. In a 50-mL round-bottom flask, coevaporate 844  mg (2  mmol, 1  eq.) of 2 three times, each with 10  mL of dry CH3CN. Dissolve the final evaporated residue in 25 mL of dry THF. 2. While stirring at 0°C, add 240 mg (6 mmol, 3 eq.) of 60% (w/w) sodium hydride (NaH) in mineral oil, 30  mg (0.2  mmol, 0.1  eq.) of sodium iodide (NaI), and 1.54  mL (10 mmol, 5 eq.) of 1,6-dibromohexane. 3. Stir the mixture at room temperature overnight. Monitor by TLC (see Notes 1, 3, 4) and stain with sulfuric acid solution. (TLC: cyclohexane/acetone, 1:1 v/v; Rf2 = 0.5; Rf3 = 0.6.) 4. Dilute the reaction mixture with 20 mL of CH2Cl2 and then carefully stop the reaction with 3 mL of water to hydrolyze the excess NaH. 5. Pour the reaction mixture into a 250-mL separatory funnel and wash the organic layer twice with 100 mL of saturated aqueous sodium bicarbonate solution. 6. Collect the organic layer, dry it by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure to obtain a yellow oil. 7. Dissolve the crude product in a minimal volume of 99:1 (v/v) cyclohexane/Et3N and apply the solution to a 5-cm diameter chromatography column containing 75 g of silica gel equilibrated with the same solvent. 8. Gradually increase the concentration of ethyl acetate (from 0 to 50%, v/v) in cyclohexane, still containing 1% (v/v) of Et3N (see Note 5). 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain a foam. 11. Check the purity of the synthesized compound by 1H, NMR and MS.

C

13

1-O-(4,4¢-Dimethoxytrityl)-2-(6-bromohexyloxymethyl)-2methylpropane-1,3-diol 3: Yield 69%. TLC (cyclohexane/ acetone, 1:1 v/v): Rf = 0.6. 1H NMR (CD3CN, 300 MHz): 0.89 (3H, s), 1.34–1.54 (6H, m), 1.78–1.88(2H, m), 2.67 (1H, t, J = 5.5 Hz), 2.98 (2H, s), 3.36–3.49 (8H, m), 3.79 (6H, s), and 6.87–7.48 (13 H, m). 13C NMR (CD3CN, 75 MHz): 16.8, 24.7, 26.6, 27.3, 28.9, 32.0, 32.2, 33.8, 33.9, 54.5, 65.0, 66.1, 70.7, 73.8, 85.0, 112.5, 126.3, 127.4, 127.7, 129.7, 136.0, 136.1, 145.2, and 158.2. HRMS ESI (positive mode) m/z: calcd for C32H41O5BrNa [M + Na]+ 607.2035; found 607.2083.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition 3.1.1.3. Azidation

175

1. In a 25-mL round-bottom flask equipped with a condenser and a CaCl2 guard, dissolve 1.17  g (2  mmol, 1  eq.) of 1-O-(4,4¢-dimethoxytrityl)-2-(6-bromohexyloxymethyl)-2methylpropane-1,3-diol 3 in 5 mL of dry CH3CN. 2. While stirring, add 949.2 mg (6 mmol, 3 eq.) of TMGA. 3. Stir the mixture 3 h at 50°C. Monitor by TLC (see Notes 1, 3, 4, 6) and stain with sulfuric acid solution. (TLC: cyclohexane/EtOAc, 85:15 v/v; Rf3 = 0.46; Rf4 = 0.41.) 4. Stop the reaction with 50  mL of CH2Cl2 and wash with 70 mL of brine. 5. Pour the reaction mixture into a 250-mL separatory funnel and extract the aqueous layer twice with 50 mL of CH2Cl2. 6. Collect the organic layers, dry them by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate the mixture on a rotary evaporator under reduced pressure. 7. Dissolve the crude product in a minimal volume of 99:1 (v/v) cyclohexane/Et3N and apply the solution to a 5-cm diameter chromatography column containing 30 g of silica gel equilibrated with the same solvent. 8. Gradually increase the concentration of EtOAc (from 0 to 30%, v/v) in cyclohexane, still containing 1% (v/v) of Et3N (see Note 5). 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain a foam. 11. Check the purity of the synthesized compound by 1H, NMR and MS.

C

13

1-O-(4,4¢-Dimethoxytrityl)-2-(6-azidohexyloxymethyl)-2methylpropane-1,3-diol 4: Yield 52%. TLC (cyclohexane/EtOAc, 85:15 v/v): Rf = 0.41. 1H NMR (CD3CN, 300 MHz): 0.90 (3H, s), 1.29–1.59 (8H, m), 2.67 (1H, br s), 2.99 (2H, s), 3.28 (2H, t, J = 6.9  Hz), 3.36–3.44 (6H, m), 3.77 (6H, s), and 6.86–7.49 (13H, m). 13C (CD3CN, 75 MHz): 16.8, 25.1, 25.9, 28.1, 28.9, 40.6, 50.8, 54.5, 65.0, 66.1, 70.7, 73.8, 85.1, 112.6, 126.3, 127.4, 127.8, 129.8, 136.0, 145.3, and 158.2. HRMS ESI (positive mode) m/z: calcd for C32H41N3O5Na [M + Na]+ 570.2944; found 570.2955. 3.1.1.4. Anchoring of 1-O-(4,4¢Dimethoxytrityl)-2(6-azidohexyloxymethyl)2-methylpropane-1,3-diol 4 on LCAA-CPG Beads

1. In a 25-mL round-bottom flask, coevaporate 344  mg (0.63  mmol, 1  eq.) of 1-O-(4,4¢-dimethoxytrityl)-2-(6azidohexyloxymethyl)-2-methylpropane-1,3-diol 4 and 15.4 mg (0.13 mmol, 0.2 eq.) of 4-DMAP three times, each time with 10  mL of dry pyridine. Dissolve the final evaporated residue in 5 mL of dry pyridine.

176

Pourceau et al.

2. While stirring, add 170.5  mg (0.75  mmol, 1.2  eq.) of hydroquinone-O,O ¢-diacetic acid (“Q-linker”) (31) and 120.2 mg (0.63 mmol, 1 eq.) of 1-(3-dimethylaminopropyl) 3-ethylcarbodiimide (DEC) under an Ar atmosphere. 3. Stir the mixture at room temperature overnight. Monitor by TLC (see Notes 1–4) and stain with sulfuric acid solution. (TLC: CH2Cl2/MeOH, 99:1 v/v: Rf4 = 0.36; Rf5 = 0.) 4. Pour the reaction mixture into a 500-mL separatory funnel with 150 mL of water and 150 mL of CH2Cl2. Extract the aqueous layer twice with CH2Cl2, each time with 150 mL. 5. Collect the organic layers, dry them by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure. 6. Dissolve the crude product in a minimal volume of 98.5:1:0.5 (v/v/v) CH2Cl2/pyridine/MeOH and apply the solution to a 5-cm diameter chromatography containing 15 g of silica gel equilibrated with the same solvent. 7. Gradually increase the concentration of MeOH (from 0.5 to 10%, v/v) in CH2Cl2, still containing 1% (v/v) of pyridine (see Note 5). 8. Monitor the collected fractions by TLC and combine those containing the pure compound. 9. Evaporate to dryness on a vacuum evaporator. (Yield: 23%). 10. In a sealed tube, coevaporate 500 mg of LCAA-CPG (70 mmol of NH2, 1 eq.) and 10 mg (70 mmol, 1 eq.) of DMAP three times with dry pyridine, each time with 10 mL. 11. Add 120 mg of 5 (140 mmol, 2 eq.) dissolved in 4 mL of dry pyridine, and then add 120 mg (0.7 mmol, 10 eq.) of DEC and 80 mL of dry Et3N. Gently shake the mixture overnight at room temperature. 12. Filter off the CPG beads on a frit and wash them with 10 mL of CH2Cl2. 13. Dry the recovered CPG beads in a vacuum desiccator over P2O5 for 30 min at room temperature (see Note 7). 14. In a sealed tube, add successively the CPG beads, 5  mL of commercial Cap A solution and 5 mL of commercial Cap B solution. Gently shake the mixture for 1  h at room temperature. 15. Filter off the CPG beads on a frit and wash them three times, each time with 10  mL of CH2Cl2. Dry the recovered CPG beads in a vacuum desiccator over P2O5 for 30 min at room temperature, affording 6.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

177

16. Accurately weigh out ~10  mg of 6. Transfer the weighed amount of 6 into a 10-mL volumetric flask and fill up to 10 mL with 0.1 M p-toluenesulfonic acid solution in CH3CN. Transfer 1 mL of the resulting solution to a clean 10-mL volumetric flask and add 9  mL of the same p-toluenesulfonic solution. 17. Measure the absorbance of the mixture at 498 nm (A498) and calculate the loading in micromoles per gram of CPG (mmol/g CPG) according to the following formula: Loading(mmol / g CPG ) = ( A 498 ´ 10 ´ 10 ´ 106 ) / (70, 000 ´ m ) where m is the mass of 6, “70,000” is the extinction coefficient of the cation DMTr, and the two “10” values correspond to the total volume of the solution (mL) and the dilution factor. In our hands, the loading of the solid support 6 was calculated to be ~61 mmol/g CPG (see Note 8). 3.1.2. Preparation of 1-O-Propargyl-a-dmannopyranose 9 3.1.2.1. Glycosylation of 2,3,4,6-O-Tetraacetyl-a-dmannopyranose 7

1. In a 50-mL round-bottom flask, coevaporate 1.9 g (4.8 mmol, 1 eq.) of peracetylated d-mannose 7 three times with 10 mL dry CH3CN. Dissolve the final evaporated residue in 20 mL of dry CH2Cl2 with 420 mg (7.5 mmol, 1.5 eq.) of propargyl alcohol. Cool the reaction mixture at 0°C. 2. While stirring at 0°C, add 3.14 mL (25 mmol, 5 eq.) of boron trifluoride etherate (BF3∙OEt2) under an Ar atmosphere. 3. Stir the mixture for 36 h at room temperature. Monitor by TLC (see Note 4) and stain with sulfuric acid solution. (TLC: Ep/EtOAc, 4:6 v/v; Rf7 = 0.48; Rf8 = 0.60.) 4. Dilute the reaction mixture with 10 mL of CH2Cl2 and then carefully stop the reaction with 30 mL of saturated aqueous sodium bicarbonate solution to hydrolyze any remaining BF3∙OEt2 that did not react. 5. Pour the reaction mixture into a 250-mL separatory funnel and extract the aqueous layer twice with 100 mL of CH2Cl2. 6. Collect the organic layers, dry them by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure to obtain a yellow oil. 7. Dissolve the crude product in a minimal volume of 6:4 (v/v) cyclohexane/EtOAc and apply the solution to a 2.5-cm diameter chromatography column containing 40  g of silica gel equilibrated with the same solvent. 8. Gradually increase the concentration of EtOAc (from 40 to 60%, v/v) in cyclohexane. 9. Monitor the collected fractions by TLC and combine those containing the pure compound.

178

Pourceau et al.

10. Evaporate to dryness on a vacuum evaporator, dissolve the residue in a minimum of EtOAc, and then crystallize the white powder by adding petroleum ether. 11. Check the purity of the synthesized compound 8 (27) by 1H NMR. 1-O-Propargyl-2,3,4,6-tetraacetyl-a-d-mannopyranose 8: Yield 66%. TLC (Ep/EtOAc 4:6, v/v): Rf = 0.60. 1H NMR (CDCl3, 200 MHz): 1.65 (s), 2.03–2.21 (12H, 4 s), 2.51 (1H, t, J = 2.4 Hz), 4.07–4.38 (5H, m), and 5.07–5.41 (4H, m). 3.1.2.2. Hydrolysis of Acetyl Groups

1. In a 25-mL round-bottom flask, dissolve 503.6 mg, (1.3 mmol) of 1-O-propargyl-2,3,4,6-O-tetraacetyl-a-d-mannopyranose 8 in 14 mL of a mixture of NH4OH/MeOH (1:1, v/v). 2. Stir the mixture at 35°C for 2 h. Monitor by TLC (see Note 4) and stain with sulfuric acid solution. (TLC: Cyclohexane/ EtOAc 4:6 v/v; Rf 8 = 0.60; Rf  9 = 0). 3. Stop the reaction by evaporation of NH3, and concentrate on a rotary evaporator under reduced pressure. 4. Pour the reaction mixture into a 250-mL separatory funnel, wash the organic layer twice with 50 mL of water, and combine the aqueous layers. 5. Concentrate on a rotary evaporator under reduced pressure, coevaporate three times with 10  mL of CH3CN to remove water, and then dry the mixture in a vacuum desiccator over P2O5 overnight at room temperature. 1-O-Propargyl-a-d-mannopyranose 9: Yield 100%. HRMS ESI (positive mode): m/z calcd for C9H15O6 [M + H]+ 219.0869; found 219.0876.

3.1.3. Click Reaction Between Azido Solid Support 6 and 1-O-Propargyl-a-dmannopyranose 9

1. In a 2-mL microcentrifuge tube, prepare a 0.1-M stock solution of 1-O-propargyl-a-d-mannopyranose 6 by dissolving 21.8 mg in 1.0 mL of degassed water. 2. In separate 0.5-mL microcentrifuge tubes, prepare a 0.1-M solution of copper sulfate (CuSO4) (3.2  mg in 200  mL of degassed water) and a 0.5-M solution of sodium ascorbate (9.9 mg in 100 mL of degassed water). Mix 4 mL of the CuSO4 solution with 4  mL of the sodium ascorbate solution in a 0.5-mL microcentrifuge tube (see Note 9). 3. To a 0.2- to 0.5-mL microwave vial equipped with a micro stir bar and containing 16.4 mg (1 mmol, 1 eq.) of the azido solid support 6, add 50  mL of 1-O-propargyl-a-dmannopyranose solution (5 mmol, 5 eq.), 42 mL of degassed water, 100 mL of MeOH, and the mixture (8 mL) of sodium ascorbate (2 mmol, 2 eq.) and CuSO4 (0.4 mmol, 0.4 eq.) prepared in step 2.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

179

4. Flush with Ar and seal the microwave vial with a cap using the crimper. Place the vial in the microwave synthesizer for 30 min at 70°C (see Note 10). 5. Transfer the beads and supernatant into a synthesis column using micropipettes (see Note 11). Connect a 2-mL syringe to the bottom of the opened column and remove the supernatant. Close the synthesis column, connect a second 2-mL syringe to the top, and wash the beads by pushing back and forth three times with 2 mL of water, three times with 2 mL of MeOH, and three times with 2 mL of CH3CN. Perform each wash for 1 min. Flush with dry nitrogen (N2) for 3 min. 6. Dry in a vacuum desiccator over P2O5 for 1 h at room temperature (see Note 12). 3.2. Preparation of O-Pent-4-ynylO ¢-(2-cyanoethyl)N,N-diisopropyl­ phosphoramidite 12

The pentynyl phosphoramidite (26) is prepared in a single procedure, starting from commercially available reagents as follows (Fig. 2): 1. Dry 420 mg (5 mmol, 1 eq.) of pent-4-yn-1-ol over molecular sieves (3 Å) overnight. 2. In a 50-mL round bottom flask, coevaporate 428  mg (2.5 mmol, 0.5 eq.) of DIAT three times with 10 mL of dry CH3CN. Dissolve the final evaporated residue in 15  mL of dry CH2Cl2. Add successively the pent-4-yn-1-ol (with molecular sieves) and 1.5 g (5 mmol, 1 eq.) of 2-cyanoethyl(N,N,N¢,N¢tetraisopropyl)-phosphorodiamidite. Equip the round-bottom flask with a CaCl2 guard. 3. Stir the mixture at room temperature for 4  h. Monitor by TLC (see Notes 1, 3, 4) and stain with vanillin solution. (TLC: Cyclohexane/CH2Cl2/Et3N, 5:4:1 v/v/v; Rf11 = 0.34; Rf12 = 0.65.) 4. Stop the reaction by adding 20 mL of CH2Cl2. OH 11

iPr2N

P

CH2Cl2

OCne

N N

iPr2N

N

N

iPr2NH2

O iPr2N

P

OCne

12

Fig. 2. Synthesis of O-pent-4-ynyl-O ¢-(2-cyanoethyl)-N,N-diisopropylphosphoramidite 12.

180

Pourceau et al.

5. Pour the reaction mixture into a 250-mL separatory funnel and wash the organic layer twice with 200 mL of brine. 6. Collect the organic layer, dry it by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure to obtain a colorless oil. 7. Dissolve the crude phosphoramidite product in a minimal volume of 75:20:5 (v/v/v) cyclohexane/CH2Cl2/Et3N and apply the solution to a 2.5-cm diameter chromatography column containing 45 g of silica gel equilibrated with the same eluent (see Note 5). 8. Purify the crude product by isocratic chromatography using the same eluent. Analyze the collected fractions by TLC and combine those containing the pure compound. Evaporate to dryness on rotary evaporator to obtain a colorless oil. 9. Check the purity of the synthesized compound by 1H, 31 P NMR and MS.

C,

13

O-Pent-4-ynyl-O¢-(2-cyanoethyl)-N,Ndiisopropylphosphoramidite 12: Yield 80%. TLC (cyclohexane/ CH2Cl2/Et3N, 5:4:1  v/v/v): Rf = 0.65. 1H NMR (CDCl3, 200 MHz): 1.18–1.21 (12H, d, J = 6.4 Hz), 1.77–1.97 (3 H, m), 2.27–2.36 (2H, m), 2.62–2.69 (2H, m), and 3.55–3.88 (6H, m); 13C NMR (CDCl3, 75 MHz): 15.3, 20.4, 29.2, 42.1, 58.2, 58.4, 61.9, 68.9, 83.7, and 117.6; 31P NMR (CDCl3, 81 MHz): 148.8. HRMS ESI (positive mode) m/z: calcd for C14H26O2N2P [M + H]+ 285.1732; found 285.1732. 3.3. Synthesis of 3 ¢-(Tetrapentynyl­ phosphodiester mannosyl) Oligonucleotide 14 3.3.1. Elongation Cycle

This section describes the automated synthesis of an oligonucleotide bearing a tetrapentynylphosphodiester mannosyl at its 3¢end starting from the solid support 10, and using the pentynyl phosphoramidite 12 and commercially available nucleoside phosphoramidites on an ABI DNA synthesizer (Fig. 3). 1. Prepare a 0.3 M solution of BMT in anhydrous CH3CN (12 g in 200 mL). Place the bottle at position #9 in the DNA synthesizer using the change bottle procedure. Place the Cap A and Cap B solutions, the deblocking solution, and the oxidation solution at bottle position #11, 12, 14, and 15, respectively. 2. Prepare 0.075  M solutions of commercially available phosphoramidites in extra dry CH3CN and place the corresponding bottles on the DNA synthesizer. 3. Prepare a 0.2 M solution of phosphoramidite 12 in extra dry CH3CN and place it at position #5 using the change bottle procedure.

181

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition 10 1) 12

2) I2 /H2O

BMT CH3CN

3) Ac2O

O O P OCne O O O O P OCne O O O P N CneO O O N N O P OCne O O

(CH2)6

O

DMTr

13

O

O O

O

O

1-SPOS

HN O

2-NH4OH

O O P O- O O P O O OO O O P N -O O O N N O P OO O

(CH2)6

O O-

ACA CCC AAT TCT O P

O

OH

O 14

Fig. 3. Synthesis of 3¢-(tetrapentynylphosphodiester mannosyl) oligonucleotide 14.

4. Program a modified elongation cycle for the ABI 394 DNA synthesizer as shown in Table 1 (see Note 13). 5. Enter the sequence of the desired oligonucleotide: 5¢-5T-3¢. In this program, 5 refers to the position of the modified phosphoramidite 12 and T refers to the solid support. 6. Pack a synthesis column with 1 mmol of the modified solid support 10 (16.4 mg). Place the column on the synthesizer. 7. Run the synthesis program corresponding to the reaction of 12 on 10 activated with BMT, then the oxidation of the resulting phosphite triester by means of the oxidation solution, and then the final capping step, affording 13.

182

Pourceau et al.

Table 1 Modified elongation cycle for the incorporation of pentynyl phosphoramidites using an ABI 394 DNA synthesizer Step

Reagents and solvents

Wash

CH3CN

10

Wash

CH2Cl2

35

Wash

CH3CN

20

Coupling

0.2 M amidite 12 + 0.3 M BMT in CH3CN

Wait

Time (s)

6 180

Wash

CH3CN

Coupling

0.2 M amidite 12 + 0.3 M BMT in CH3CN

Wait

15 6 180

Wash

CH3CN

15

Capping

Cap A and Cap B solution

10

Wait

15

Wash

CH3CN

Oxidation

Oxidation solution

Wait Wash

10 8 13

CH3CN

20

8. Program a standard elongation cycle for the ABI 394 DNA synthesizer as shown in Table  2. The standard elongation cycle is available through the synthesizer’s program library. 9. Enter the sequence of the desired oligonucleotide: 5¢-ACA CCC AAT TCT T-3¢. In this program, the first 3¢-T refers to the support 13. 10. Run the synthesis program with “Trityl off.” The elongation cycle corresponds to (1) a detritylation step; (2) coupling of the nucleoside phosphoramidite; (3) oxidation of the phosphite triester; and (4) capping (see Note 14). 3.3.2. Deprotection/ Cleavage and Analysis by HPLC and MALDITOF MS 3.3.2.1. Deprotection/ Cleavage

1. Transfer 8.2 mg (~0.5 mmol) of the solid-supported 3¢-tetrapentynyl-mannose oligonucleotide into a sealed HPLC vial. Add 3  mL of concentrated aqueous ammonia (NH4OH) solution and place the vial into a dry bath for 2 h at 65°C (see Note 15). 2. Recover the supernatant and evaporate the ammonia using a SpeedVac® vacuum concentrator system.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

183

Table 2 Standard oligonucleotide elongation cycle for an ABI 394 DNA synthesizer Step

Reagents and solvents

Time (s)

Wash

CH3CN

10

Wash

CH2Cl2

35

Detritylation

Deblocking solution

43

Wash

CH3CN

20

Coupling

0.075 M commercial amidites + 0.3 M BMT in CH3CN

 6

Wait

20

Wash

CH3CN

15

Capping

Cap A and Cap B solution

10

Wait

15

Wash

CH3CN

10

Oxidation

Oxidation solution

 8

Wait Wash

3.3.2.2. Analysis by HPLC and MALDI-TOF MS

13 CH3CN

20

1. Dissolve the crude oligonucleotide 14 in 500 mL of pure water to determine its amount by UV spectrophotometry as follows: Withdraw 20 mL of the solution and add it to 1 mL of pure water. Read the absorbance at 260 nm using a UV–visible spectrophotometer (see Note 16). (Isolated amount: 178 nmol.) 2. Characterize 14 by MALDI-TOF mass spectrometry (matrix: 2¢,4¢,6¢-trihydroxyacetophenone, THAP) and by reversedphase HPLC (gradient: 8–32% (v/v) of CH3CN in 0.05 M TEAAc for 20 min). (HPLC RT = 10.26 min, lmax = 262 nm, e = 124,700 L/mol/cm, MALDI-TOF MS (negative mode) m/z calcd for C155H211N44O92P16 [M − H]−: 4658.202; found 4659.02.) 3. Freeze-dry the solution. The oligonucleotide 14 can be stored for several months in a freezer at −20°C under an Ar atmosphere.

3.4. Conjugation of Galactosyl Azide 19 to Tetraalkyne Oligonucleotide 14

This section describes the synthesis of the galactosyl azido derivative 19 (29) (Fig. 4), which is eventually conjugated in solution by CuAAC to the 3¢-(tetrapentynyl-mannose)-oligonucleotide 14 (Fig. 5) affording the final tetragalactosyl-mannose ­oligonucleotide conjugate product 20 after HPLC purification and ammonia treatment to remove the acetyl protecting groups (see Note 17).

184

Pourceau et al. HO

HO

Tos-Cl

H

DMAP Et3N

H OH

HO

NaN3 NaI

H

H

DMF

H

CH2Cl2

H

OTos

15

17

16

N3

AcO OAc

BF3. Et2O

O AcO 18

OAc

CH2Cl2

OAc

AcO OAc AcO

N3

H O

H

O

OAc

19

Fig.  4. Synthesis of 1-O-[4-(azidomethyl)cyclohexyl-1-methyl]-2,3,4,6-tetra-O-acetylb-d-galactopyranose 19.

14 19

CuSO4 Na Ascorbate H2O/MeOH

NH4OH

O HO

OH

O

N

N

N

O PO OO O O P O OP O O O N -O O O N N (CH2)6 O P OO O

H N N N

O

OH

O

H

H N N N

O

H

HO OH HO

HO

H

HO OH

H

H O

HO

O

N H

N

HO OH

O

OO ACA CCC AAT TCT O P O

HO OH

OH

O

OH

N 20

OH

Fig. 5. Synthesis of the tetragalactosyl-mannose oligonucleotide conjugate 20 by 1,3-dipolar cycloaddition.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

185

3.4.1. Preparation of 1-O-[4-(Azidomethyl) cyclohexyl-1-methyl]2,3,4,6-tetra-O-acetyl-bd-galactopyranose 19

1. In a 50-mL round-bottom flask, coevaporate 1.4 g (10 mmol, 2  eq.) of 1,4-cyclohexanedimethanol 15 and 61  mg (0.5 mmol, 0.1 eq.) of DMAP three times, each with 25 mL of dry pyridine. Dissolve the final evaporated residue in 15 mL of dry CH2Cl2.

3.4.1.1. Prepare [4-(p-Toluenesulfonylmethyl) cyclohexyl]methanol 16

2. While stirring, add 950 mg (5 mmol, 1 eq.) of p-toluenesulfonyl chloride and 1.05 mL (7.5 mmol, 1.5 eq.) of dry Et3N under an Ar atmosphere. 3. Stir the mixture during 3 h. Monitor by TLC (see Notes 2–4) and stain with sulfuric acid solution. (TLC: CH2Cl2/MeOH, 95:1 v/v; Rf TosCl = 0.9; Rf15 = 0.2; Rf16 = 0.55.) 4. Stop the reaction with 50 mL of saturated aqueous sodium bicarbonate solution. 5. Pour the reaction mixture into a 250-mL separatory funnel and extract the aqueous layer twice with 100 mL of CH2Cl2. 6. Collect the organic layers, dry them by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure. A pink oil is obtained. 7. Dissolve the crude in a minimal volume of CH2Cl2 and apply the solution to a 2.5-cm diameter chromatography column containing 35 g of silica gel equilibrated with CH2Cl2. 8. Gradually increase the concentration of MeOH (from 0 to 5%, v/v) in CH2Cl2. 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain a pink oil. 11. Check the purity of the synthesized compound by 1H, NMR and MS.

13

C

[4-(p-Toluenesulfonylmethyl)cyclohexyl]methanol 16: Yield 72%. TLC (CH2Cl2/MeOH 95:5  v/v): Rf = 0.55. 1H NMR (CDCl3, 300  MHz): 0.85–1.95 (11H, m), 2.47 (3H, s), 3.44–3.95 (2H, m), 3.83–3.95 (2H, m), 7.36 (2H, d, J = 8.1 Hz), and 7.80 (2H, d, J = 8.3 Hz). 13C NMR (CDCl3, 75 MHz): 22.0, 25.2, 25.3, 28.8, 35.0, 37.7, 38.1, 40.5, 66.3, 68.7, 73.5, 75.6, 128.3, 130.2, 133.4, and 145.0. HRMS ESI (positive mode) m/z: calcd for C15H23O4S1 [M + H]+ 299.1317; found 299.1305. 3.4.1.2. Prepare [4-(Azidomethyl)cyclohexyl] methanol 17

1. In a 100-mL round-bottom flask equipped with a condenser, dissolve 1 g (3.4 mmol, 1 eq.) of [4-(p-toluenesulfonylmethyl) cyclohexyl]methanol 16 in 25 mL of DMF. 2. While stirring, add 2.1 g (13.7 mmol, 4 eq.) of sodium iodide (NaI) and 890  mg (13.7  mmol, 4  eq.) of sodium azide (NaN3) at room temperature.

186

Pourceau et al.

3. Stir the mixture for 3 h at 75°C. Monitor by TLC (see Notes 3 and 4) and stain with vanillin solution. (TLC: Cyclohexane/ EtOAc 7:3, v/v; Rf16 = 0.1; Rf17 = 0.3.) 4. Remove the solvent under reduced pressure and dissolve the resulting residue in 100 mL of CH2Cl2. 5. Pour the reaction mixture into a 250-mL separatory funnel and wash the organic layer twice with 50  mL of saturated aqueous sodium bicarbonate solution and twice with 50 mL brine. 6. Collect the organic layer, dry it by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure to obtain a yellow oil. 7. Dissolve the crude in a minimal volume of 93:7 (v/v) cyclohexane/EtOAc and apply the solution to a 2.5-cm diameter chromatography column containing 50 g silica gel equilibrated with the same solvent. 8. Gradually increase the concentration of EtOAc (from 7to 60%, v/v) in cyclohexane. 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain 17 as colorless oil. 11. Check the purity of the synthesized compound by 1H, 13C NMR, MS and Fourier transform infrared spectroscopy (FTIR) (see Note 18). [4-(Azidomethyl)cyclohexyl]methanol 17 Yield 79%. TLC (cyclohexane/EtOAc 7:3  v/v) Rf: 0.29. 1H NMR (CDCl3, 300  MHz): 0.85–1.95 (11H, m), 3.15–3.26 (2H, m), and 3.47–3.57 (2H, m). 13C NMR (CDCl3, 75  MHz): 25.1, 26.3, 28.8, 30.0, 35.5, 37.9, 38.3, 40.4, 55.4, 58.0, 66.0, and 68.5. HRMS ESI (positive) m/z: calcd for C8H16 N3O [M + H]+ 170.1293; found 170.1315. FTIR: 3343, 2922, 2856, 2524, 2096, 1451, 1378, 1345, 1262, 1132, 1097, 1036, 992, 955, 941, 885, 825, 655, 588, and 556. 3.4.1.3. Preparation of 1-O-[4-(Azidomethyl) cyclohexyl-1-methyl]2,3,4,6-tetra-O-acetyl-bd-galactopyranose 19

1. In a 50-mL round-bottom flask, coevaporate 414  mg (1.05 mmol, 1.05 eq.) of peracetylated d-galactose 18 three times with 10 mL dry acetonitrile. Dissolve 18 in 10 mL of dry CH2Cl2 with 171 mg (1 mmol, 1 eq.) of [4-(azidomethyl) cyclohexyl]methanol 17. Cool the reaction mixture to 0°C. 2. While stirring at 0°C, add 630 mL (5 mmol, 5 eq.) of boron trifluoride etherate (BF3∙OEt2) under an Ar atmosphere. 3. Stir the mixture 4 h at room temperature. Monitor by TLC (see Note 4) and stain with vanillin solution. (TLC:

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

187

Cyclohexane/EtOAc, 1:1  v/v; Rf18 = 0.27; Rf17 = 0.33; Rf19 = 0.40.) 4. Dilute the reaction mixture with 10 mL of CH2Cl2 and stop the reaction carefully with 30 mL of saturated aqueous sodium bicarbonate solution to hydrolyze the excess BF3∙OEt2. 5. Pour the reaction mixture into a 250-mL separatory funnel and extract the aqueous layer twice with 100 mL of CH2Cl2. 6. Collect the organic layers, dry them by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure to obtain a yellow oil. 7. Dissolve the crude in a minimal volume of cyclohexane (see Note 19) and apply the solution to a 2.5-cm diameter chromatography column containing 25 g of silica gel equilibrated with cyclohexane. 8. Gradually increase the concentration of EtOAc (from 0 to 30%, v/v) in cyclohexane. 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain 19 as colorless oil. 11. Check the purity of the synthesized compound by 1H, NMR and MS.

C

13

1-O-[[4-(Azidomethyl)cyclohexyl]methyl]-2,3,4,6-tetra-Oacetyl-b-d-galactopyranose 19: Yield 21%. TLC (cyclohexane/ EtOAc, 1:1  v/v): Rf = 0.29. 1H NMR (CDCl3, 200  MHz): 0.85–2.08 (22H, m), 3.05–3.35 (3H, m), 3.65–3.86 (2H, m), 4.00–4.17 (2H, m), 4.34–4.39 (1H, m), 4.94 (1H, dd, J = 3.2 Hz, J = 10.5  Hz), 5.14 (1H, dd, J = 7.8  Hz, J = 10.5  Hz), and 5.32 (1H, d, J = 3.2  Hz). 13C NMR (CDCl3, 75  MHz): 19.9, 20.0, 20.3, 24.3, 24.4, 25.3, 25.5, 28.0, 28.2, 29.1, 29.2, 34.2, 34.8, 37.0, 37.4, 54.7, 57.1, 60.5, 66.3, 68.2, 69.9, 70.2, 72.3, 74.8, 100.9, 168.7, 169.5, 169.6, and 169.7. HRMS ESI (positive) m/z: calcd for C22H34N3O10 [M + H]+ 500.2244; found 500.2244. 3.4.2. Click Reaction of Galactosyl Azido Derivative 19 with 3¢-(Tetrapentynylmannose)-oligonucleotide 14 in Solution

1. In a 2-mL microcentrifuge tube, prepare a 1-mM solution of tetraalkyne mannose-oligonucleotide 14 (178  nmol in 178 mL of pure water). 2. In a 2-mL microcentrifuge tube, prepare a 0.1-M stock solution of 19 by dissolving 49.9 mg in 1 mL of MeOH. 3. In separate 0.5-mL microcentrifuge tubes, prepare a 0.04-M solution of copper sulfate (CuSO4) (6.4  mg in 1  mL of degassed water) and a 0.05-M solution of sodium ascorbate (9.9 mg in 1 mL of degassed water). Mix 5 mL of the CuSO4

188

Pourceau et al.

solution with 20  mL of the sodium ascorbate solution in a 0.5-mL microcentrifuge tube (see Note 9). 4. To a 0.2- to 0.5-mL microwave vial equipped with a micro stir bar, add 100  mL of the oligonucleotide solution 14 (100 nmol, 1 eq.), 113 mL of MeOH, 12 mL of 19 (1.2 mmol, 12 eq.), and the mixture (25 mL) of sodium ascorbate (1 mmol, 10 eq.) and CuSO4 (200 nmol, 2 eq.) prepared in step 3. 5. Flush with Ar and seal the microwave vial with a cap using the crimper. Place the vial in the microwave synthesizer for 45 min at 70°C. (See Note 20). 6. Desalt the resulting solution on a NAP™-10 column and then concentrate using a SpeedVac® vacuum concentrator system. 7. Dissolve the residue in 300 mL of pure water for characterization by MALDI-TOF mass spectrometry and by reversedphase HPLC (gradient: 5–60% (v/v) CH3CN in 0.05  M TEAAc for 20 min). (HPLC RT = 15.33 min, MALDI-TOF MS (negative mode) m/z: calcd for C243H343N56O132P16 [M − H]−: 6656.20; found 6653.95.) 3.4.3. Purification and Characterization of the Final Tetragalactosyl-Mannose Oligonucleotide Conjugate Product 20 3.4.3.1. Purification

3.4.3.2. Deprotection of Galactose Hydroxyls and Characterization

1. Purify the crude product obtained from Subheading 3.4.2 by preparative reversed-phase HPLC using a gradient of 32–44% (v/v) CH3CN in 0.05 M TEAAc for 15 min. 2. Pool the fractions containing the protected oligonucleotide glycoconjugate and evaporate to dryness. Coevaporate the residue further ten times, each with 1.5 mL of pure water to remove volatile buffer salts. (MALDI-TOF MS (negative mode) m/z: calcd for C243H343N56O132P16 [M − H]−: 6656.20; found 6656.74.) 1. In a 2-mL microcentrifuge tube, add successively the protected tetragalactose oligonucleotide conjugate obtained after HPLC purification and 1 mL of concentrated aqueous ammonia solution. Stir the reaction mixture at room temperature for 2 h. 2. Evaporate the ammonia using a SpeedVac® vacuum concentrator system. 3. Dissolve the residue in 300 mL of pure water and extract the acetamide three times with 500 mL of CH2Cl2. 4. Determine the amount of 20 by absorbance spectrophotometry at l = 260 nm (see Subheading 3.3.2). 5. Characterize the pure tetragalactose oligonucleotide conjugate product 20 by MALDI-TOF mass spectrometry and reversed-phase HPLC (gradient: 6–24% (v/v) CH3CN in 0.05 M TEAAc for 20 min) (Fig. 6). (HPLC RT = 13.83 min, lmax = 262  nm, e = 124,700  L/mol/cm. MALDI-TOF MS (negative mode) m/z: calcd for C211H311N56O116P16 [M − H]−: 5983.615; found 5983.36. Isolated amount: 23 nmol.)

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

189

13,83

0

5

10

min

15

20

25

Fig. 6. C18 RP-HPLC profile of the tetragalactosyl-mannose oligonucleotide conjugate 20.

4. Notes 1. Before spotting the reaction mixture, the TLC plates must be neutralized by dipping in a solution of 99:1 (v/v) CH2Cl2/ Et3N and air-dried in order to avoid sample degradation on the plate (due to the acidity of the silica). 2. Before starting the elution, the spot of the reaction mixture must be air-dried to remove pyridine and/or Et3N, which could modify the retention factor (Rf). 3. After the elution, and before the reaction with stain solution, direct visualization using a 254-nm UV lamp can reveal the starting reagents and/or the final compounds. 4. Staining the TLC plate in the stain solution (vanillin or sulfuric acid) reveals the compounds. The plate is slowly heated to remove solvent, dipped in the stain solution, washed with water, stamped with absorbent paper to remove excess of water, and then slowly heated to dry the plate. 5. It is recommended that the chromatography column first be equilibrated with an elution solvent containing Et3N to avoid degradation due to the acidity of the silica. Derivatives bearing a dimethoxytrityl group or phosphoramidite/phosphorodiamidite functions are very sensitive to acidic treatments, so any traces of acid must be neutralized. In our hands, phosphoramidite compounds could not be purified on silica gel columns without first adding Et3N. 6. The Rf values of both compounds are very similar. To overcome this difficulty, it is necessary to elute the TLC plate an

190

Pourceau et al.

additional two or three times after the first elution using the same elution gradient. 7. Before capping the unreacted functions, the loading of the solid support can be determined by a trityl assay as follows: Filter off some of the beads, wash three times with 5 mL of CH2Cl2, dry in a vacuum desiccator over P2O5 for 30  min, and weigh the beads. Calculate the loading of the functionalized solid support as described in step 16. 8. After capping with acetic anhydride in the presence of N-methylimidazole, the loading is typically found to be between 40 and 65 mmol/g of CPG. 9. To prepare these solutions, water must be degassed to avoid degradation of the oligonucleotide due to the oxygen in the presence of copper. Furthermore, it is best to prepare the sodium ascorbate solution just before the reaction to allow the reduction of copper(II) into copper(I) to proceed effectively. The copper sulfate solution can be prepared in advance and stored for several months at −20°C. When both solutions are mixed together, the color turns brown and opaque. 10. We have noticed that a very efficient CuAAC reaction occurs when using microwave (MW) assistance. If no microwave synthesizer is available, the reaction should be performed under vigorous stirring at 70°C for at least 2  h. However, reactions performed without MW assistance are typically less reproducible. 11. While transferring the beads into the column, be careful to ensure that no beads remain on the rim of the column, as this will result in column malfunction and failed reaction steps. 12. It is essential to have the solid support 10 be as dry as possible since trace amounts of water will react with the phosphoramidite 12, leading to less efficient introduction of pentynyl groups on the mannose core. 13. This cycle is derived from the standard elongation cycle that is available in the DNA synthesizer’s program library. The detritylation step must be removed since the mannose hydroxyls are free, but this is also to keep the dimethoxytrityl group present on the solid support for the further synthesis of the oligonucleotide. The coupling time of the phosphoramidite is longer: 180 s instead of 20 s; and a double coupling step is applied to ensure the success of the reaction on the four hydroxyls on the mannose. (Note that three of the hydroxyls are secondary alcohols that are less reactive than the standard 5¢-primary alcohol of a nucleoside.) 14. After the first incorporation, the cycle can be paused and some of the CPG beads can be deprotected (with 1  mL of NH4OH for 15 h at room temperature) to verify the ­successful

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

191

incorporation of the modified phosphoramidite 12 by reversed-phase HPLC and MALDI-TOF mass spectrometry before continuing the elongation. This verification process cannot be carried out before the incorporation of a nucleoside because there is no chromophore on the support, and the mass is too low to be detectable by MALDI-TOF MS. 15. Since the present oligonucleotide sequence has no deoxyguanosine protected with an isobutyryl group, the deprotection step can be limited to 2  h. On the other hand, sequences containing dG must be treated for 5 h at 55°C to yield full deprotection. 16. The extinction coefficient (e) of the oligonucleotide at l = 260 nm is calculated as the sum of the extinction coefficients of the constitutive nucleosides: e(dA) = 15,400  L/mol/cm; e(dG) = 11,500  L/mol/cm; e(dC) = 7,400  L/mol/cm; and e(dT) = 8,700 L/mol/cm. 17. To facilitate the purification of the tetragalactose oligonucleotide conjugate, we use the tetraacetylated galactose azide derivative 19 instead of the unprotected galactosyl moiety in order to confer lipophilicity to the resulting oligonucleotide conjugate, thus allowing for an easier RP-HPLC separation step. After RP-HPLC purification, a supplementary deprotection step that involves the hydrolysis of acetyl groups is required. A final extraction with ethyl ether is then performed to remove the acetamide. 18. FTIR is a useful method to characterize azide derivatives since they exhibit a strong vibration at 2,096 cm−1. 19. The dissolution of the crude with cyclohexane can be difficult; add a few milliliters of ethyl acetate to help with the dissolution process. 20. We have noticed that when CuAAC reactions are performed in solution, the use of MW assistance is less important. In this case, however, the reaction is usually incubated for a longer period of time (2–3  h) to obtain a similar yield level. It is essential that the click reaction be completed within 5 h to ensure that there is no significant degradation of the oligonucleotide moiety. References 1. Varki, A. (1993) Biological Roles of Oligosaccharides – All of the Theories Are Correct. Glycobiology 3, 97–130. 2. Dwek, R. A. (1996) Glycobiology: Toward understanding the function of sugars. Chem. Rev. 96, 683–720.

3. Lis, H., and Sharon, N. (1998) Lectins: Carbohydrate-Specific Proteins That Mediate Cellular Recognition. Chem. Rev. 98, 637–674. 4. Imberty, A., Wimmerova, M., Mitchell, E. P., and Gilboa-Garber, N. (2004) Structures of the lectins from Pseudomonas aeruginosa:

192

Pourceau et al.

insights into the molecular basis for host glycan recognition. Microbes Infect. 6, 221–228. 5. Chemani, C., Imberty, A., de Bentzmann, S., Pierre, M., Wimmerova, M., Guery, B. P., and Faure, K. (2009) Role of LecA and LecB Lectins in Pseudomonas aeruginosa-Induced Lung Injury and Effect of Carbohydrate Ligands. Infect. Immun. 77, 2065–2075. 6. Deguise, I., Lagnoux, D., and Roy, R. (2007) Synthesis of glycodendrimers containing both fucoside and galactoside residues and their binding properties to Pa-IL and PA-IIL lectins from Pseudomonas aeruginosa. New J. Chem. 31, 1321–1331. 7. Lundquist, J. J., and Toone, E. J. (2002) The cluster glycoside effect. Chem. Rev. 102, 555–578. 8. Darbre, T., and Reymond, J. L. (2008) Glycopeptide dendrimers for biomedical applications. Curr. Top. Med. Chem. 8, 1286–1293. 9. Singh, Y., Renaudet, O., Defrancq, E., and Dumy, P. (2005) Preparation of a multitopic glycopeptide-oligonucleotide conjugate. Org. Lett. 7, 1359–1362. 10. Marra, A., Moni, L., Pazzi, D., Corallini, A., Bridi, D., and Dondoni, A. (2008) Synthesis of sialoclusters appended to calix[4]arene platforms via multiple azide-alkyne cycloaddition. New inhibitors of hemagglutination and cytopathic effect mediated by BK and influenza A viruses. Org. Biomol. Chem. 6, 1396–1409. 11. Cecioni, S., Lalor, R., Blanchard, B., Praly, J. P., Imberty, A., Matthews, S. E., and Vidal, S. (2009) Achieving high affinity towards a bacterial lectin through multivalent topological isomers of calix[4]arene glycoconjugates. Chem. Eur. J. 15, 13232–13240. 12. Dubber, M., and Lindhorst, T. K. (2001) Trehalose-based octopus glycosides for the synthesis of carbohydrate-centered PAMAM dedrimers and thiourea-bridged glycoclusters. Org. Lett. 3, 4019–4022. 13. Ortega-Munoz, M., Perez-Balderas, F., Morales-Sanfrutos, J., Hernandez-Mateo, F., Isac-Garcia, J., and Santoyo-Gonzalez, F. (2009) Click Multivalent Heterogeneous Neoglycoconjugates – Modular Synthesis and Evaluation of Their Binding Affinities. Eur. J. Org. Chem., 2454–2473. 14. Garcia-Lopez, J. J., Hernandez-Mateo, F., Isac-Garcia, J., Kim, J. M., Roy, R., SantoyoGonzalez, F., and Vargas-Berenguel, A. (1999) Synthesis of per-glycosylated beta-cyclodextrins having enhanced lectin binding affinity. J. Org. Chem. 64, 522–531.

15. Matsuura, K., Hibino, M., Yamada, Y., and Kobayashi, K. (2001) Construction of GlycoClusters by Self-Organization of SiteSpecifically Glycosylated Oligonucleotides and Their Cooperative Amplification of LectinRecognition. J. Am. Chem. Soc. 123, 357–358. 16. Zatsepin, T. S., and Oretskaya, T. S. (2004) Synthesis and applications of oligonucleotidecarbohydrate conjugates. Chem. Biodiversity 1, 1401–1417. 17. Biessen, E. A. L., Vietsch, H., Rump, E. T., Fluiter, K., Kuiper, J., Bijsterbosch, M. K., and Van Berkel, T. J. C. (1999) Targeted delivery of oligodeoxynucleotides to parenchymal liver cells in vivo. Biochem. J. 340, 783–792. 18. Katajisto, J., Heinonen, P., and Lonnberg, H. (2004) Solid-phase synthesis of oligonucleotide glycoconjugates bearing three different glycosyl groups: Orthogonally protected bis (hydroxymethyl)-N,N¢-bis(3-hydroxypropyl) malondiamide phosphoramidite as key building block. J. Org. Chem. 69, 7609–7615. 19. Katajisto, J., Virta, P., and Lonnberg, H. (2004) Solid-phase synthesis of multiantennary oligonucleotide glycoconjugates utilizing on-support oximation. Bioconjugate Chem. 15, 890–896. 20. Bouillon, C., Meyer, A., Vidal, S., Jochum, A., Chevolot, Y., Cloarec, J. P., Praly, J. P., Vasseur, J. J., and Morvan, F. (2006) Microwave assisted “click” chemistry for the synthesis of multiple labeled-carbohydrate oligonucleotides on solid support. J. Org. Chem. 71, 4700–4702. 21. Beaucage, S. L., and Caruthers, M. H. (1981) Deoxynucleoside phosphoramidites – a new class of key intermediates for deoxypolynucleotide synthesis. Tetrahedron Lett. 22, 1859–1862. 22. Tornoe, C. W., Christensen, C., and Meldal, M. (2002) Peptidotriazoles on solid phase: [1,2,3]-triazoles by regiospecific copper(I)catalyzed 1,3-dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem. 67, 3057–3064. 23. Rostovtsev, V. V., Green, L. G., Fokin, V. V., and Sharpless, K. B. (2002) A Stepwise Huisgen Cycloaddition Process: Copper(I)Catalyzed Regioselective “Ligation” of Azides and Terminal Alkynes. Angew. Chem. Int. Ed. 41, 2596–2599. 24. Wacker, R., and Niemeyer, C. M. (2004) DDI-mu FIA – A readily configurable microarray-fluorescence immunoassay based on DNAdirected immobilization of proteins. Chembiochem 5, 453–459.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition 25. Chevolot, Y., Bouillon, C., Vidal, S., Morvan, F., Meyer, A., Cloarec, J. P., Jochum, A., Praly, J. P., Vasseur, J. J., and Souteyrand, E. (2007) DNA-based carbohydrate biochips: A platform for surface glyco-engineering. Angew. Chem. Int. Ed. 46, 2398–2402. 26. Pourceau, G., Meyer, A., Vasseur, J. J., and Morvan, F. (2009) Azide Solid Support for 3¢-Conjugation of Oligonucleotides and Their Circularization by Click Chemistry. J. Org. Chem. 74, 6837–6842. 27. Hasegawa, T., Numata, M., Okumura, S., Kimura, T., Sakurai, K., and Shinkai, S. (2007) Carbohydrate-appended curdlans as a new family of glycoclusters with binding properties both for a polynucleotide and lectins. Org. Biomol. Chem. 5, 2404–2412. 28. Pourceau, G., Meyer, A., Vasseur, J. J., and Morvan, F. (2009) Synthesis of Mannose and

193

Galactose Oligonucleotide Conjugates by Bi-click chemistry. J. Org. Chem. 74, 1218–1222. 29. Pourceau, G., Meyer, A., Vasseur, J. J., and Morvan, F. (2008) Combinatorial and automated synthesis of phosphodiester galactosyl cluster on solid support by click chemistry assisted by microwaves. J. Org. Chem. 73, 6014–6017. 30. Lonnberg, H. (2009) Solid-Phase Synthesis of Oligonucleotide Conjugates Useful for Delivery and Targeting of Potential Nucleic Acid Therapeutics. Bioconjugate Chem. 20, 1065–1094. 31. Pon, R. T., and Yu, S. Y. (1997) HydroquinoneO,O¢-diacetic acid (Q-linker) as a replacement for succinyl and oxalyl linker arms in solid phase oligonucleotide synthesis. Nucleic Acids Res. 25, 3629–3635.

Chapter 12 Site-Specific DNA Labeling by Staudinger Ligation Samuel H. Weisbrod, Anna Baccaro, and Andreas Marx Abstract Site-specific and chemoselective labeling of DNA is still a difficult task. The Staudinger ligation is a bioorthogonal reaction between azides and phosphines that requires no catalyst to proceed, allowing for mild reaction conditions. The reaction may be extended for site-specific labeling of DNA using azidomodified triphosphates, which can be incorporated site-specifically into DNA strands by DNA polymerases in a template-dependent manner. The azido-modified DNA, in turn, can be labeled by suitable phosphines. This protocol describes (1) the synthesis of an azido-TTP analogue; (2) the enzymatic synthesis of azido-modified DNA; (3) the synthesis of suitable phosphine labels; and (4) the labeling of azido-DNA with biotin–phosphine by Staudinger ligation with approximately 70% conversion. Key words: Site-specific DNA labeling, DNA modification, Staudinger ligation, Phosphine, Modified triphosphate, Azido thymidine, DNA polymerase, Azido-DNA, Biotin

1. Introduction Several conjugation reactions have been used to label DNA sitespecifically; however, truly chemoselective methods are rarely represented (1). Besides Diels Alder reactions and copper-catalyzed azide alkyne 1,3 dipolar cycloadditions, the Staudinger ligation is a type of chemoselective conjugation reaction that has been used for site-specific labeling of DNA (2). The Staudinger ligation reaction occurs between an azide and a phosphine to form an azaylide that can be trapped by an acyl group to form a stable amide bond (3). Since both functionalities exhibit high bioorthogonality and no further reagents are required, the Staudinger ligation has found several in  vivo applications (4) and should allow the attachment of a wide variety of functional labels. Since the direct incorporation of an azido or phosphor(III) functionality by standard phosphoramidite chemistry is not

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_12, © Springer Science+Business Media, LLC 2011

195

196

Weisbrod, Baccaro, and Marx

Fig. 1. Scheme for labeling DNA with biotin by Staudinger ligation. 5-(5-azido-1-pentynyl)-2¢-deoxyuridine (azido-TTP) is first incorporated by a DNA polymerase instead of TTP in a growing DNA strand. The azido label is then reacted with a suitable phosphine label. In this example, biotin–phosphine is attached to the DNA strand.

a­ pplicable, our approach involves the incorporation of azidofunctionalized deoxynucleoside triphosphates by a DNA polymerase and subsequent labeling of enzymatically generated DNA by Staudinger ligation using phosphine labels without by-product formation (see ref. 5 and Fig. 1). In this chapter, the site-specific labeling of short enzymatically generated DNA strands with biotin will be used to illustrate the method.

2. Materials 2.1. Synthesis of Azido-TTP

1. TLC plates Silica gel 60 F254 (Merck, Darmstadt, Germany). 2. Imidazole. 3. tert-Butyldimethylsilyl chloride. 4. 2¢-Deoxy-5-iodouridine. 5. Triethylamine. 6. Pent-4-yn-1-ol. 7. Tetrakis(triphenylphosphine)palladium(0). 8. Copper(I)iodide. 9. N,N-Diisopropylethylamine.

Site-Specific DNA Labeling by Staudinger Ligation

197

10. Methanesulfonyl chloride. 11. Sodium azide. 12. Tetrabutylammonium fluoride solution: 1  M tetrabutylammonium fluoride in THF. 13. Acetic acid. 14. TEAB buffer: 1  M triethylamine is saturated with gaseous CO2 (from dry ice) overnight, pH 7.5. 15. TEAA buffer: 50  mM triethylamine, 50  mM acetic acid, pH 7.0. 16. 1,8-Bis(dimethylaminonaphthalene). 17. Phosphoryl chloride. 18. Trimethyl phosphate. 19. Tri-n-butylamine. 20. Bis (see preparation)-tri-n-butylammonium pyrophosphate (prepared as described in Subheading 3.1.6.1). 21. Amberlite® IR120, H+-resin. 22. Diethylaminoethyl (DEAE) Sephadex® A-25. 23. Reversed-phase medium-pressure liquid chromatography (MPLC) column: Lobar® LiChroprep® RP-18 (Merck). 2.2. Synthesis of Phosphine Labels

1. 1-Methyl 2-iodoterephtalate. 2. Palladium(II) acetate. 3. Diphenylphosphine. 4. Carbonyldiimidazole. 5. N-Boc-4,7,10-trioxa-1,13-tridecanediamine. 6. Hydrogen chloride solution: 4  M hydrogen chloride in dioxane. 7. Biotin.

2.3. Enzymatic Incorporation of Azido-TTP ( 6) into DNA

1. Deoxyribonucleotide-5¢-triphosphates: dATP (100  mM); dCTP (100 mM); dGTP (100 mM); and dTTP (100 mM) (MBI Fermentas, St. Leon-Roth, Germany). 2. Oligonucleotides: Primer: 5¢-GAC CCA CTC CAT CGA GAT TTC TC-3¢ (10  mM); Template: 5¢-GCG CTG GCA CGG GAG AAA TCT CGA TGG AGT GGG TC-3¢ (10 mM) (Metabion, Martinsried, Germany). 3. Pwo DNA polymerase (1  U/ml) (Peqlab, Erlangen, Germany). 4. 210 mM ethylenediaminetetraacetic acid (EDTA) solution. 5. illustra MicroSpinTM G-25 columns (GE Healthcare, Freiburg, Germany).

198

Weisbrod, Baccaro, and Marx

2.4. Labeling of Azido-DNA by Staudinger Ligation

1. Sodium carbonate–bicarbonate buffer solution: 1 M Na2CO3– NaHCO3, pH 9.0.

3. Methods The methods described below outline (1) the chemical synthesis of an azido-functionalized thymidine triphosphate analogue (Fig. 2); (2) the synthesis of appropriate phosphines for Staudinger ligation reaction; (3) enzymatic generation of azido-modified DNA; and (4) the site-specific labeling of azido-modified DNA by Staudinger ligation. 3.1. Synthesis of Azido-TTP

Water-sensitive chemical reactions are carried out in oven-dried glassware under an argon atmosphere. Anhydrous solvents are stored over molecular sieves under argon. Thin-layer ­chromatography

Fig. 2. Synthesis of 5-(5-azido-1-pentynyl)-2¢-deoxyuridine (azido-TTP).

Site-Specific DNA Labeling by Staudinger Ligation

199

(TLC) is performed with alumina plates, and compounds are visualized by UV light (254 nm). Column chromatographic separations of compounds are achieved by using 50–100 times more silica gel than sample material. NMR spectra are recorded with CDCl3 (d = 7.24) or CD3OD (dCD3 = 3.31) as solvent. Electrospray ionization mass spectra (ESI–MS) are obtained in positive or negative ion mode with the samples dissolved in methanol or methanol/water (1:1, v/v). The synthesis steps below are based on a published procedure (6). 3.1.1. 3 ¢,5 ¢-Bis-O(tert-Butyldimethylsilyl)-2 ¢Deoxy-5-Iodouridine (1)

1. Add imidazole (0.96 g, 14.1 mmol) to a stirred solution of tert-butyldimethylsilyl chloride (1.28 g, 8.5 mmol) in anhydrous DMF (6 mL) at 0°C. 2. Add 2¢-deoxy-5-iodouridine (1 g, 2.82 mmol) at 0°C and stir 3 h at room temperature (RT). 3. Add ice-cold water (100 mL) and extract with diethyl ether (3 × 100 mL). Combine the organic phases, dry with magnesium sulfate and remove the solvent under reduced pressure. 4. Purify the crude product by column chromatography (silica gel, EtOAc-hexane, 2:3). Nucleoside 1 (1.61  g, 97%) is obtained as a white foam (Rf = 0.3, EtOAc-hexane, 1:2); 1H NMR (400  MHz, CDCl3): d 0.07, 0.08, 0.15, 0.16 (4s, 12H, 2 (CH3)2Si), 0.88, 0.91 (2s, 18H, 2 (CH3)3CSi), 1.99 (ddd, J = 5.9, 8.1, 13.3 Hz, 1H, 2¢-Ha), 2.30 (ddd, J = 2.2, 5.6, 13.3 Hz, 1H, 2¢-Hb), 3.76 (dd, J = 2.3, 11.5 Hz, 1H, 5¢-Ha), 3.89 (dd, J = 2.3, 11.5  Hz, 1H, 5¢-Hb), 3.99 (d, J = 2.3 Hz, 1H, 4¢-H), 4.38–4.40 (m, 1H, 3¢-H), 6.27 (dd, J = 5.6, 8.1 Hz, 1H, 1¢-H), 8.09 (s, 1H, 6-H), 8.55 (br s, 1H, NH).

3.1.2. 3 ¢,5 ¢-Bis-O(tert-Butyldimethylsilyl)-5(5-Hydroxy-1-Pentynyl)-2¢Deoxyuridine (2)

1. Add triethylamine (482  mL, 3.43  mmol), pent-4-yn-1-ol (476 mL, 5.16 mmol) and tetrakis(triphenylphosphine)palladium(0) (27  mg, 0.17  mmol) to a stirred suspension of nucleoside 1 (1 g, 1.72 mmol) and copper(I)iodide (65 mg, 0.34 mmol) in anhydrous DMF (5 mL) (see Note 1). 2. Stir the orange solution at RT for 16 h until no more starting material is detected by TLC. 3. Add 10 mL of saturated sodium bicarbonate solution. 4. Extract with ethyl acetate (3 × 10 mL), combine the organic phases, dry the organic phase with magnesium sulfate and remove the solvent under reduced pressure. 5. Purify the crude product by column chromatography (silica gel, EtOAc-hexane, 1:1 to 3:2, v/v). Nucleoside 2 (750 mg, 81%) is obtained as an off-white solid (Rf = 0.3, EtOAc-hexane, 3:2); 1H NMR (600  MHz, CDCl3): d = 0.09, 0.1, 0.15, 0.16 (4s, 12H, 2 (CH3)2Si), 0.91, 0.95

200

Weisbrod, Baccaro, and Marx

(2s, 18H, 2 (CH3)3CSi), 1.77–1.86 (m, 2H, CH2CH2CH2OH), 2.00–2.07 (m, 1H, H-2¢a), 2.31 (ddd, J = 13.1, 5.7, 2.5 Hz, 1H, H-2¢b), 2.51 (t, J = 6.8  Hz, 2H, CH2CH2CH2OH), 3.76–3.81 (m, 3H, H-5¢b, CH2CH2CH2OH), 3.91 (dd, J = 11.4, 2.0  Hz, 1H, H-5¢a), 3.97–3.99 (m, 1H, H-4¢), 4.40–4.43 (m, 1H, H-3¢), 6.30 (t, J = 5.7, 5.7 Hz, 1H, H-1¢), 7.93 (s, 1H, H-6), 8.28 (br s, 1H, NH); ESI–MS: m/z = 561.5 [M + Na]+. 3.1.3. 3 ¢,5 ¢-Bis-O-(tertButyldimethylsilyl)-5(5-Methanesulfonatepent1-ynyl)-2 ¢-Deoxyuridine (3)

1. Add N,N-diisopropylethylamine (182 mL, 1.07 mmol) to a stirred solution of nucleoside 2 in anhydrous CH2Cl2 (5 mL) at 0°C and stir for 30 min at 0°C. 2. Add methanesulfonyl chloride (55 mL, 0.71 mmol) dropwise and stir for 40 min at 0°C. 3. Add saturated sodium bicarbonate solution (10  mL) and extract with diethyl ether (3 × 15  mL). Wash the combined phase with brine (15 mL), dry with magnesium sulfate, and remove the solvent under reduced pressure. 4. Purify the crude product by column chromatography (silica gel, EtOAc-hexane, 1:2). Nucleoside 3 (270  mg, 74%, see Note 2) is obtained as a white foam (Rf = 0.8, EtOAc-hexane, 3:1); 1H NMR (600 MHz, CDCl3): d = 0.07, 0.08, 0.13, 0.14 (4s, 12H, 2 (CH3)2Si), 0.89, 0.93 (2s, 18H, 2 (CH3)3CSi), 1.97–2.05 (m, 3H, H-2¢a, CH2CH2CH2OMs), 2.31 (ddd, J = 13.1, 6.0, 2.7 Hz, 1H, H-2¢b), 2.55 (dd, J = 6.8, 6.8 Hz, 2H, CH2CH2CH2OMs), 3.07 (s, 3H, OSO2CH3), 3.77 (dd, J = 11.5, 2.1  Hz, 1H, H-5¢b), 3.89 (dd, J = 11.5, 2.4 Hz, 1H, H-5¢a), 3.96–3.98 (m, 1H, H-4¢), 4.37– 4.41 (m, 3H, H-3¢, CH2CH2CH2OMs), 6.27 (dd, J = 7.3, 6.0 Hz, 1H, H-1¢), 7.94 (s, 1H, H-6), 8.05 (br s, 1H, NH); ESI–MS: m/z = 639.4 [M + Na]+.

3.1.4. 3 ¢,5 ¢-Bis-O(tert-Butyldimethylsilyl)5-(5-Azido-1-Pentynyl)-2 ¢Deoxyuridine (4)

1. Add sodium azide (46 mg, 0.71 mmol) to a stirred solution of nucleoside 3 (70  mg, 0.11  mmol) in anhydrous DMF (2 mL) and stir for 20 h at 35°C. 2. Add saturated sodium bicarbonate solution (5  mL) and extract with ethyl acetate (3 × 5  mL). Dry the combined organic phases with magnesium sulfate, and remove the solvent under reduced pressure. 3. Purify the crude product by column chromatography (silica gel, EtOAc-hexane, 1:3 to 1:1). Nucleoside 4 (61  mg, 95%) is obtained as a white foam (Rf = 0.8, EtOAc-hexane, 1:1); 1H NMR (600  MHz, CDCl3): d = 0.09, 0.1, 0.15, 0.16 (4s, 12H, 2 (CH3)2Si), 0.91 0.95 (2s, 18H, 2 (CH3)3CSi), 1.85 (dddd, J = 6.8, 6.8, 6.8, 6.8 Hz, 2H, CH2CH2CH2N3), 2.03 (ddd, J = 13.3, 7.7, 6.3 Hz, 1H, H-2¢a), 2.31 (ddd, J = 13.3, 5.9, 2.7 Hz, 1H, H-2¢b), 2.51 (t, J = 6.8 Hz,

Site-Specific DNA Labeling by Staudinger Ligation

201

2H, CH2CH2CH2N3), 3.45 (t, J = 6.8 Hz, 2H, CH2CH2CH2N3), 3.78 (dd, J = 11.5, 2.2  Hz, 1H, H-5¢b), 3.91 (dd, J = 11.5, 2.3 Hz, 1H, H-5¢a), 3.98 (ddd, J = 2.2, 2.3, 2.2 Hz, 1H, H-4¢), 4.40–4.43 (m, 1H, H-3¢), 6.3 (dd, J = 7.7, 5.9 Hz, 1H, H-1¢), 7.93 (s, 1H, H-6), 8.27 (br s, 1H, NH); ESI–MS: m/z = 586.6 [M + Na]+. 3.1.5. 5-(5-Azido1-Pentynyl)-2 ¢Deoxyuridine (5)

1. Add tetrabutylammonium fluoride (1  M in THF, 226  mL, 0.23  mmol) to a stirred solution of nucleoside 4 (58  mg, 0.10 mmol) in anhydrous THF (2 mL) at 0°C. 2. Stir and let warm to RT for 16 h. 3. Remove the solvent under reduced pressure. 4. Purify the crude product by column chromatography (silica gel, CH2Cl2–MeOH, 20:1). Nucleoside 5 (32  mg, 93%) is obtained as a white foam (Rf = 0.4, CH2Cl2–MeOH, 20:1); 1H NMR (400 MHz, CD3OD): d = 1.75 (dddd, J = 6.8, 6.8, 6.8, 6.8 Hz, 2H, CH2CH2CH2N3), 2.10–2.26 (m, 2H, H-2¢a, H-2¢b), 2.43 (t, J = 6.8  Hz, 2H, CH2CH2CH2N3), 3.40 (t, J = 6.8  Hz, 2H, CH2CH2CH2N3), 3.67 (dd, J = 12.0, 3.4  Hz, 1H, H-5¢b), 3.75 (dd, J = 12.0, 3.0 Hz, 1H, H-5¢a), 3.84–3.88 (m, 1H, H-4¢), 4.31–4.36 (m, 1H, H-3¢), 6.18 (t, J = 6.6, 6.6 Hz, 1H, H-1¢), 8.19 (s, 1H, H-6); ESI–MS: m/z = 357.9 [M + Na]+.

3.1.6. 5-(5-Azido1-Pentynyl)-2 ¢Deoxyuridine-5 ¢Triphosphate (6)

1. Prepare a solution of tri-n-butylamine (4.77 mL, 20 mmol) in ethanol (40 mL) and cool with ice (solution A).

3.1.6.1. Preparation of Bis (see preparation)Tri-n-Butylammonium Pyrophosphate

3. Dissolve tetrasodium diphosphate decahydrate (4.46  g, 10 mmol) in water (75 mL) and load it onto the Amberlite® column.

2. Fill a column (20 cm × 3 cm) with fresh Amberlite® IR-120 (H+ form) and wash with water until pH 5.

4. Elute the column dropwise with water directly into the icecold stirred solution A until the eluate again reaches pH 5. 5. Remove the solvent under reduced pressure at 30°C. Co-evaporate the residue with anhydrous ethanol and anhydrous DMF (each 3 × 10 mL). Attention: do not heat above 30°C for the evaporation step. 6. Dissolve the residue in anhydrous DMF (total volume 20 mL) and store over molecular sieves (4 Å) at 4°C.

3.1.7. Preparation of Triphosphate (6)

1. Dry in vacuo 1,8-bis(dimethylaminonaphthalene) (28  mg, 0.13 mmol) and nucleoside 5 (29 mg, 86 mmol) overnight in the dark (see Note 3). 2. Dissolve both reagents in anhydrous trimethyl phosphate (1 mL) and cool to 0°C.

202

Weisbrod, Baccaro, and Marx

3. Add dropwise freshly distilled phosphoryl chloride (10.5 mL, 0.11 mmol) and stir for 3 h at 0°C (see Note 4). 4. Simultaneously add 0.5  M Bis (see preparation)-tri-n-butylammonium pyrophosphate in anhydrous DMF (0.86 mL, 0.43 mmol) and tri-n-butylamine (229 mL, 0.86 mmol), and stir for 10 min. 5. Quench with 1 M TEAB buffer (5 mL) and wash the aqueous phase with ethyl acetate (2 × 10  mL). Lyophilize the aqueous phase. 6. Purify the residue by ion-exchange chromatography (DEAESephadex® A-25 (18 g dry)) with a linear gradient of TEAB buffer (0.1–1 M, total volume of 1,500 mL, 2 mL/min flow rate at 4°C). Lyophilize the product-containing fractions (check by ESI–MS, negative mode). 7. Purify by reversed-phase medium-pressure liquid chromatography (RP-MPLC) (RP-18, 40–63  mm) using a gradient of 5% (200 ml), 20% (200 ml) and 40% (200 ml) acetonitrile in 50 mM (TEAA buffer). The triphosphate elutes at 20% acetonitrile in 50 mM TEAA buffer. Lyophilize several times. 8. Store the triphosphate as a 100  mM solution in water at −20°C (see Note 5). Nucleotide 6 (26 mg, 30%) is obtained as an oil; 1H NMR (400  MHz, CD3OD): d = 1.20–1.35 [m, 36H, N(CH2CH3)3], 1.8–1.93 (m, 2H, CH2CH2CH2N3), 2.30–2.40 (m, 2H, H-2¢b, H-2¢a), 2.47–2.55 (m, 2H, CH2CH2CH2N3), 3.13–3.25 [m, H-5¢a, H-5¢b, N(CH2CH3)3], 3.25–3.46 (m, 2H, CH2CH2CH2N3), 4.1–4.2 (m, 1H, H-3¢), 4.56–4.62 (m, 1H, H-4¢), 6.26 (m, 1H, H-1¢), 8.0 (s, 1H, H-6); 31P NMR (162 MHz, MeOD): d = −22.5 (m, 1 P, Pb), −10.6 (d, J = 20.2 Hz, 1 P, Pa), −9.3 (d, J = 20.2 Hz, 1 P, Pg); ESI–MS: m/z = 575.0 [M–H]−. 3.2. Synthesis of Biotin–Phosphine Label (Fig. 3)

Additionally to the remarks in Subheading 3.1, it should be mentioned that the described phosphines are to some extent susceptible to oxidation. As a precaution, store the phosphines under nitrogen or argon at −20°C and use (whenever possible) degassed solutions for the synthesis procedures. Phosphine oxides exhibit 31 P NMR signals at d = 30–40 ppm.

3.2.1. 1-Methyl 2-Diphenylphosphi­ noterephtalate ( 7)

The synthesis is based on a literature procedure (3). 1. Dissolve 1-methyl 2-iodoterephtalate (1.73  g, 5.66  mmol) and palladium(II)acetate (30 mg, 0.13 mmol) in anhydrous methanol (18 mL). 2. Add triethylamine (1.57 mL, 11.32 mmol) and diphenylphosphine (0.98  mL, 5.66  mmol) and reflux the solution overnight.

Site-Specific DNA Labeling by Staudinger Ligation

203

Fig. 3. Synthesis of the biotin–phosphine label.

3. Remove the solvent under reduced pressure, mix the residue with CH2Cl2 and water (200 mL each), separate phases, wash the organic phase with hydrochloric acid (1 M, 100 mL), dry the organic phase with magnesium sulfate, and remove the solvent under reduced pressure. 4. Purify the crude product by column chromatography (silica gel, CH2Cl2–MeOH, 9:1). Phosphine 7 (0.99  g, 48%) is obtained as a yellow solid (Rf = 0.3, CH2Cl2–MeOH, 9:1); 1H NMR (400 MHz, CDCl3): d = 3.68 (s, 3H, CH3), 7.19–7.30 (m, 10H, aryl), 7.58–7.61 (m, 1H, H-3), 7.99–8.06 (m, 2H, H-5, H-6); 31P NMR (CDCl3) d = −3.2. 3.2.2. 1-Methyl 2-Diphenylphosphino4-(N-Boc-4,7,10-Trioxa13-Tridecaneamine) Carbamoyl Benzoate ( 8)

1. Dissolve phosphine 7 (200  mg, 0.55  mmol) in anhydrous CH2Cl2 (6 mL) and anhydrous DMF (2 mL) and stir at RT. 2. Add carbonyldiimidazole (98 mg, 0.6 mmol) and stir for 1 h. 3. Add a solution of N-Boc-4,7,10-trioxa-1,13-tridecanediamine (176 mg, 0.55 mmol) in anhydrous CH2Cl2 (2 mL) and stir for 2 h. 4. Wash the reaction mixture with saturated ammonium chloride solution (10 mL), dry the organic phase with magnesium sulfate and remove the solvent under reduced pressure. 5. Purify the crude product by column chromatography (silica gel, CH2Cl2 to CH2Cl2–MeOH, 19:1).

204

Weisbrod, Baccaro, and Marx

Phosphine 8 (320  mg, 87%) is obtained as a yellow oil (Rf = 0.4, CH2Cl2–MeOH, 19:1); 1H NMR (400 MHz, CDCl3): d = 1.39 (s, 9H, (CH3)3C), 1.67 (p, J = 6 Hz, 2H, CH2), 1.75 (p, J = 6 Hz, 2H, CH2), 3.14 (m, 2H, NCH2), 3.45–3.40 (m, 6H, OCH2, NCH2) 3.45–3.58 (m, 8H, OCH2), 3.69 (s, 3H, CH3), 7.23–7.30 (m, 10H, aryl), 7.37 (dd, J = 3.6, 1.2 Hz, 1H, H-3), 7.71 (dd, J = 8, 1.2 Hz, 1H, H-6), 8.01 (dd, J = 8.0, 3.6 Hz, 1H, H-5); 31P NMR (CDCl3) d = −3.2; ESI–MS: m/z = 667.1 [M + H]+. 3.2.3. 1-Methyl 2-Diphenylphosphino-4(4, 7, 10-Trioxa-13Aminotridecyl) Carbamoylbenzoate (9)

1. Stir phosphine (8) (320  mg, 0.48  mmol) for 2  h at RT in hydrogen chloride in dioxane (5 mL). 2. Remove the solvent under reduced pressure. 3. Purify the crude product by column chromatography (silica gel, CH2Cl2–MeOH, 9:1) (see Note 6). Phosphine 9 (198  mg, 73%) is obtained as a yellow solid (Rf = 0.35, CH2Cl2–MeOH, 9:1); 1H NMR (400 MHz, CDCl3): 1.69–1.81 (m, 2H, CH2), 1.84–1.96 (m, 2H, CH2), 3.04 (m, 2H, NCH2), 3.30–3.6 (m, 8H, NH2, OCH2, NCH2), 3.50 (m, 8H, OCH2) 3.66 (s, 3H, CH3), 7.19–7.30 (m, 10H, aryl), 7.43 (dd, J = 3.7, 1.6  Hz, 1H, H-3), 7.74 (t, J = 5.6  Hz, 1H, NH), 7.91 (dd, J = 8.1, 1.6 Hz, 1H, H-6), 8.00 (dd, J = 8.1, 3.7 Hz, 1H, H-5); 31P NMR (CDCl3) d = −3.0; ESI–MS: m/z = 567.0 [M + H]+.

3.2.4. 1-Methyl 2-Diphenylphosphino-4(4, 7, 10-Trioxa-13Biotinamidotridecyl) Carbamoyl Benzoate (10)

1. Stir carbonyldiimidazole (26 mg, 0.16 mmol) and biotin (32 mg 0.13 mmol) in anhydrous DMF (3 mL) for 2 h at RT. 2. Add phosphine 9 (50 mg, 0.09 mmol) in DMF (1 mL) to the reaction mixture and stir overnight at RT. 3. Remove the solvent under reduced pressure. 4. Add water (5 mL), extract with CH2Cl2 (2 × 5 mL), combine the organic phases, dry the organic phase with magnesium sulfate and remove the solvent under reduced pressure. 5. Purify the crude biotin–phosphine product 10 by column chromatography (silica gel, CH2Cl2–MeOH, 94:6). Biotin–phosphine 10 (33  mg, 47%) is obtained as a yellow solid (Rf = 0.25, CH2Cl2–MeOH, 94:6); 1H NMR (400  MHz, CDCl3): 1.40 (p, J = 6.4  Hz, 2H, CH2), 1.57–1.66 (m, 4H, CH2), 1.72 (p, J = 6.4 Hz, 2H, CH2), 1.79 (p, J = 6.4 Hz, 2H, CH2), 2.15 (t, J = 7.6 Hz, 2H, CH2), 2.70 (d, J = 12.8 Hz, 1H, CH), 2.86 (dd, J = 12.8, 4.8 Hz, 1H, CH), 3.08–3.12 (m, 1H, CH), 3.28 (q, J = 6.4  Hz, 2H, NCH2), 3.42–3.60 (m, 14H, OCH2, NCH2), 3.72 (s, 3H, CH3), 4.25 (dd, J = 7.6, 5.2 Hz, 1H, CH), 4.45 (dd, J = 7.6, 4.8  Hz, 1H, CH), 5.61 (s, 1H, NH), 6.41 (s, 1H, NH), 6.74 (t, J = 5.6 Hz, 1H, NH), 7.23 (t, J = 5.6 Hz, 1H, NH), 7.25–7.34 (m, 10H, aryl), 7.42 (dd, J = 3.6, 1.0 Hz,

Site-Specific DNA Labeling by Staudinger Ligation

205

1H, H-3), 7.78 (dd, J = 8.0, 1.0 Hz, 1H, H-6), 8.05 (dd, J = 8.0, 3.6  Hz, 1H, H-5); 31P NMR (CDCl3) d = −3.0; ESI–MS: m/z = 814.9 [M + Na]+. 3.3. Enzymatic Incorporation of Azido-TTP ( 6) into DNA

Incorporation of the azido triphosphate is illustrated with the primer/template duplex listed in Subheading 2.3. For each occurrence of a nonhybridized “A” in the primer/template duplex, azido-TTP (6) is incorporated for later labeling of the synthesized duplex DNA molecule. In this example protocol, a single azidoTTP is incorporated. Extinction coefficients for the single DNA strands at 260  nm are calculated according to Sambrook and Russell (8) and used to adjust the concentration of the oligonucleotide solutions. For radioactive detection on 12% polyacrylamide gels, the primer strand contained approximately 5% 5¢ 32 P-labeled primer (see Note 7). 1. Prepare a solution (dNTP mixture) containing 2.5 mM dATP, 2.5 mM dCTP, 2.5 mM dGTP, and 2.5 mM azido-TTP (6) in water. 2. Prepare the primer/template duplex by adding primer (2.4 mL), template (3.2 mL) and Pwo reaction buffer (2 mL) to water (9.4 mL) in an Eppendorf tube. 3. Anneal the primer/template DNA strands by heating the Eppendorf tube to 95°C for 5 min and let cool down to RT. Cool the solution further to 0°C (see Note 8). 4. Add the dNTP mixture (2  mL) and Pwo DNA polymerase (1 mL) at 0°C to bring the total reaction volume to 20 mL. Heat to 72°C for 30 min and cool down again to 0°C. 5. Quench the reaction with EDTA solution (1  mL). Remove buffers and excess nucleotides using Microspin™ G-25 columns. The formation of azido-modified DNA can be analyzed by denaturing PAGE with 32P-labeled DNA (Fig. 4).

3.4. Labeling of Azido-DNA by Staudinger Ligation

1. Combine the azido-DNA solution (5  mL, as prepared in Subheading 3.3), Na2CO3/NaHCO3 buffer (2.5 mL) and the desired phosphine label (e.g., biotin–phosphine 10) (2.5 mL, 10 mM in DMF). 2. Heat for 12 h at 60°C (see Note 9). 3. The formation of Staudinger ligation products can be visualized by denaturing PAGE with 32P-labeled DNA (Fig. 4).

4. Notes 1. In order to obtain high yields, exclusion of air and moisture is very important for this reaction. Copper(I) iodide dissolves upon addition of triethylamine. Sometimes catalyst residues

206

Weisbrod, Baccaro, and Marx

Fig. 4. Analysis of synthesis reaction products by denaturing polyacrylamide gel electrophoresis. (a) Enzymatic synthesis of azido-modified DNA: The incorporation of azido-TTP (lane 4) in comparison to TTP (lane 3), without TTP (lane 2), and only the primer/template complex (lane 1) as marker. (b) Staudinger ligation with azido-modified DNA and a biotin–phosphine label.

cannot be removed by normal column chromatography. However, analytically pure compounds can be obtained by RP-columns with linear gradients from water to acetonitrile. 2. Nucleoside 3 should not be stored over long time periods. It is best to use it as quickly as possible in the next step. Alternatively, column chromatographic purification can be omitted and the crude nucleoside can be used directly in the next step. In this case, however, we have observed a lower overall yield. 3. For the triphosphate synthesis procedure, it is very important that all reagents and solvents are clean and dry. Trimethyl phosphate is distilled in vacuo and stored over molecular sieves (4 Å), and phosphoryl chloride is either freshly distilled or distilled and stored under argon. 4. The reaction may be monitored with TLC (n-propanol/ water/conc. ammonia, 3:1:1 v/v/v). Occasionally, it is necessary to add more phosphoryl chloride, but do not add more than 0.5 additional equivalents since twofold phosphorylation should be avoided. 5. The concentration is determined by UV absorption using e = 9,000 M−1 cm−1 (287 nm) of compound 2 (7).

Site-Specific DNA Labeling by Staudinger Ligation

207

6. This compound can be stored over long time periods under argon at −20°C and utilized for the attachment of different labels. 7. Appropriate permissions to work with 32P should be obtained. Radioactively labeled substances should be handled with appropriate safety precautions. As an alternative to using radioactive labels, one may also consider using different detection techniques such as silver staining. Primer extension and Staudinger ligation do not rely on the use of one specific type of detection technique. 8. Using a PCR thermocycler, the following cooling procedure can be carried out in a few minutes: Heat to 95°C for 4 min, then cool down stepwise for 30 s at 65, 55, 40, 25, 20°C and hold at 0°C. 9. Staudinger ligation with biotin–phosphine 10 seems to be very robust towards different reaction conditions. Lower temperatures with elongated reaction times are possible as well as reactions in water (~pH 7) or sodium hydroxide solution (pH 12), resulting in similar yields. The effect of cosolvents has also been investigated: In addition to DMF, ethanol appears to be a suitable co-solvent; however, most important is to achieve good solubility of the phosphine in the organic co-solvent mixture. References 1. Niemeyer, C. M. ed. (2004) Bioconjugation Protocols (Methods in Molecular Biology Series Vol. 283). Humana Press, Totowa, NJ. 2. Weisbrod, S. H. and Marx, A. (2008) Novel strategies for the site-specific covalent labeling of nucleic acids. Chem. Commun. 44, 5675–5685. 3. Saxon, E. and Bertozzi, C. R. (2000) Cell surface engineering by a modified Staudinger reaction. Science 287, 2007–2010. 4. Prescher, J. A., Dube D. H. and Bertozzi C. R. (2004) Chemical remodelling of cell surfaces in living animals. Nature 430, 873–877. 5. Weisbrod, S. H. and Marx, A. (2007) A nucleoside triphosphate for site-specific labeling of

DNA by the Staudinger ligation. Chem. Commun. 18, 1828–1830. 6. Baccaro, A. Weisbrod, S. H. and Marx, A. (2007) DNA conjugation by the Staudinger ligation: new thymidine analogues. Synthesis 13, 1949–1954. 7. Robins, M. J. and Barr, P. J. (1983) Nucleic acid related compounds. 39. Efficient conversion of 5-iodo to 5-alkynyl and derived 5-substituted uracil bases and nucleosides. J. Org. Chem. 48, 1854–1862. 8. Sambrook J., Russell D. W. (2001) Molecular Cloning: A Laboratory Manual. Laboratory Press, Cold Spring Harbor, NY.

Chapter 13 Improved Cellular Uptake of Antisense Peptide Nucleic Acids by Conjugation to a Cell-Penetrating Peptide and a Lipid Domain Takehiko Shiraishi and Peter E. Nielsen Abstract Unaided cellular uptake of RNA interference agents such as antisense oligonucleotides and siRNA is extremely poor, and in vivo bioavailability is also limited. Thus, effective delivery strategies for such potential drugs are in high demand. Recently, a novel approach using a class of short cationic peptides known as cell-penetrating peptides (CPPs) is attracting wide attention for a variety of biologically active molecules. CPP-mediated delivery is typically based on the covalent conjugation of the (therapeutic) cargo to CPPs, and is particularly relevant for the delivery of noncharged RNA interference agents such as peptide nucleic acids (PNAs) and morpholino oligomers. Although chemical conjugation to a variety of CPPs significantly improves the cellular uptake of PNAs, the bioavailability (and hence antisense activity) of CPP–PNA ­conjugates is still highly limited by endocytotic entrapment. We have found, however, that this low ­bioavailability can be significantly improved by chemical conjugation to a lipid domain (“Lip,” such as a fatty acid), thereby creating “CatLip”-conjugates. The cellular uptake of these conjugates is conveniently evaluated using a sensitive cellular assay system based on a splicing correction of a mutated luciferase gene in HeLa pLuc705 cells by targeting antisense oligonucleotides to a cryptic splice site. Further improvement in the delivery of CatLip–PNA conjugates is achieved by using auxiliary agents/treatments (e.g., chloroquine, calcium ions, or photosensitizers) to induce endosomal disruption. Key words: Antisense oligonucleotides, Cellular uptake, Cell-penetrating peptide, Catlip ­conjugates, Peptide nucleic acid

1. Introduction Synthetic oligonucleotides, such as aptamers, siRNAs, antisense oligonucleotides, and ribozymes are widely exploited in drug discovery studies. In the case of RNA interference agents (antisense oligonucleotides, siRNA, etc.) their target molecules are located intracellularly (in the nucleus and/or cytosol) and therefore these Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_13, © Springer Science+Business Media, LLC 2011

209

210

Shiraishi and Nielsen

agents need to be delivered into the cells by penetrating the plasma membrane. However, such oligonucleotide-based agents typically have a negligible ability to pass through the cell ­membrane because of their size, hydrophilic properties and not least polyanionic character (1). In order to improve the cellular uptake of oligonucleotides, several chemical derivatization, conjugation, and complexation methods, as well as physical methods, have been developed. Currently, transfection via cationic lipids or polymers is the most effective method for the delivery of negatively charged oligonucleotides and their derivatives, and many reagents are available for this approach. Such reagents are mainly cationic lipid- or cationic polymer-based compounds exploiting electrostatic interactions with negatively charged oligonucleotides to form transfection complexes. While these types of transfection reagents are quite efficient for cargo delivery to most cells in in vitro culture, their in vivo applications are rather limited largely because of general toxicity. Consequently, highly efficient, safe and robust delivery methods and agents for in vivo use are still very much in demand. New strategies have recently been devised for the delivery of biologically active molecules through the discovery of cell-penetrating peptides (CPPs) (2). CPPs are derived from a variety of different sources (e.g., venom, toxin, or synthetic) but most share common properties such as small size (less than 30 amino acid residues) and an overall cationic charge. They are generally capable of translocating across the cellular membrane through endocytotic pathway(s), and all are able to carry a variety of molecular cargoes into the cell via chemical conjugation or complexation. This method could be particularly useful for noncharged RNA interference agents such as PNAs, which – in contrast to anionic oligonucleotides – do not interact electrostatically with CPPs (3–5). To date, a variety of CPPs have been tested to improve the cellular uptake of RNA interference agents by chemical conjugation, and practically all are able to improve cellular uptake to some (greatly) varying extent without interfering with the agent’s biological (antisense) activity. However, as most CPPs utilize endocytotic pathways as their main route for cellular internalization (6–8), their bioavailability is limited because of endosomal entrapment. Although this limitation may be partly solved by the use of auxiliary reagents/treatments (e.g., chloroquine, Ca2+-treatment, photochemical internalization, etc.), which can aid by opening endo(lyso)somes to release their contents (9–11), research efforts to develop unaided delivery methods are still warranted. We have recently shown that chemical conjugation of a lipidic domain (e.g., a fatty acid such as decanoic acid) to a CPP–PNA construct significantly improves its cellular delivery (bioavailability) (12). For the evaluation of the cellular (nuclear) uptake of CPP–PNA conjugates, a splice-correction reporter system based on the use of the

Improved Cellular Uptake of Antisense Peptide Nucleic Acids

211

HeLa pLuc 705 cell line (13) was employed. HeLa pLuc 705 cells harbor an integrated, mutated luciferase gene that contains an insertion of intron 2 from human b-globin pre-mRNA carrying a cryptic splice site. This cell-based assay system allows one to gain a quantitative estimation of the cellular (nuclear) delivery efficiency of an antisense agent by measuring the level of luciferase activity that results from the correction of aberrant splicing upon targeting the agent to the cryptic splice site. This method has been widely used for studies of CPP delivery efficiency since it displays a low background level and has a positive readout with a wide dynamic range; moreover, the method is particularly suitable for stericblocking agents such as morpholino oligonucleotides (14) and PNAs since these do not induce RNase H activity.

2. Materials 2.1. Cell Culture

1. Cell growth medium: RPMI 1640 medium supplemented with 10% (v/v) fetal bovine serum (FBS) and 1% (v/v) GlutaMAX™ (Gibco). 2. HeLa pLuc 705 cell line. 3. Cell culture flask (T-25). 4. 24-well tissue culture plate. 5. Teflon cell scrapers.

2.2. Preparation of the MBHA Resin for PNA Synthesis by Downloading

1. 4-Mehylbenzhydryl amine (MBHA) resin (1,000  nt). To demonstrate the latter, the polymerase chain reaction (PCR) – a common technique employed in molecular biology as well as in paternity and diagnostic testing applications – was recently used for the first time as a synthetic tool to produce DNA-polymer hybrids; however, DBCs with high molecular weight (>25 kDa) nucleic acid units will not be described here – for additional information, the reader is referred to the relevant literature (11, 12). After establishing effective synthetic routes for DBCs, the structural properties of these materials need to be studied. Amphiphilic DBCs self-assemble into spherical micelles in aqueous solution, as visualized by atomic force microscopy (AFM) and measured by dynamic light scattering (DLS) and fluorescence correlation spectroscopy (FCS) (10). One of the aims in working with such materials is to manipulate the morphologies of the DNA block copolymer aggregates through mild stimuli such as molecular recognition processes induced by, for example, the addition of complementary ODNs. In a recent study, the hybridization of single-stranded DNA block copolymer micelles with long DNA templates containing multiple copies of the micelle

242

Zimmermann et al.

corona complementary sequence was shown to induce the transformation of the spherical micelles into rod-like aggregates (13). In this case, the Watson–Crick motif of base pairing aligned the hydrophobic polymer segments along the DNA double helix, which then resulted in selective dimer formation between the hybridized DNA block copolymer structures. This study for the first time demonstrated that DNA nanostructures, which are typically generated through the base pairing of complementary ODN sequences, can be built up by employing hydrophobic interactions, adding a new tool to the field of nanotechnology for creating novel nanostructures. DBC-based materials can serve not only as tunable, responsive nanoscale objects, but also as three-dimensional nanoscopic scaffolds for organic reactions. In one study conducted by our group (10), the DNA strands of the corona were organized by the hydrophobic interactions of the organic polymer segments in such a fashion that several DNA-templated organic reactions could be performed in a sequence-specific manner both at the surface of the micelles and at the interface between the biological and the organic polymer blocks. These results demonstrated that DNA block copolymer micelles can thus be regarded as programmable nanoreactors. Moreover, DBC self-assembled scaffolds have also been applied for enzymatic reactions (14). By incubating the DBC aggregates with a template-independent DNA polymerase, facile control over the growth of the nanoparticles under mild isothermal conditions in an aqueous medium was achieved. It was possible to gradually increase the micelle size by a factor of up to 2.4, thereby demonstrating control over nanoparticle size by an enzymatic reaction for the first time. In a further step, DBC aggregates have been investigated for their interactions with living cells. Since spherical and rod-shaped DNA nanoparticles composed of exactly the same materials were readily available, we assessed whether different geometries can influence cellular uptake. Interestingly, it was found that the rodshaped nanoparticles were taken up by cells 12 times more efficiently than their spherical counterparts (15). After looking at nondirected internalization, the targeted uptake of DNA nanoobjects was also studied. For this purpose, DBC aggregates were equipped with folate as a targeting ligand for cancerous cells. The functionalization of the particles was carried out in a very straightforward manner, i.e. ligand–ODN conjugates were hybridized with the micelles. The incorporation of fluorescent reporter groups by the same procedure revealed that receptor-mediated endocytotic uptake of the nanoparticles with a diameter of approximately 10 nm was most efficient when the maximum number of ligands was present on the rim of the micelles. The loading of doxorubicin (DOX), an anticancer drug, into the hydrophobic interior of the ligand-containing micelles resulted in efficient cytotoxicity and

Amphiphilic DNA Block Copolymers

243

high mortality among the targeted cancerous cells. The advantage of using DBC carriers compared to other types of polymeric carriers such as dendrimers or conventional block copolymer micelles is that functionalization and size can be altered extremely easily, allowing for the combinatorial testing of various DBC-based drug delivery systems (16). In the following sections, we describe procedures for the synthesis and characterization of multifunctional DBC nanoparticles such as those described above.

2. Materials 2.1. Synthesis of DNA-b-PPO

1. Poly(propylene oxide) monobutyl ether (6,800 Da) (SigmaAldrich, Germany).

2.1.1. Synthesis of the PPO Phosphoramidite (4)

2. 2-Cyanoethyl N,N-diisopropylchlorophosphoramidite (1  g ampoule) (Sigma-Aldrich). Store at −20°C. 3. Dichloromethane (DCM). “Extra dry” quality, with water content less than 0.005% (ACROS organics). 4. N-Ethyldiisopropylamine (DIPEA). Water content less than 0.3%, stored over molecular sieves (Sigma-Aldrich). 5. Molecular sieves (3 Å). 6. Sodium carbonate solution: 5% (w/v) sodium carbonate in water. 7. Brine: Saturated solution of sodium chloride in water. 8. Magnesium sulfate. 9. Rotary evaporator. 10. Glassware and other materials: Three-neck flask (100  mL), septum, bubble counter, magnetic stirrer, stir bar, 1  mL syringe, separation funnel (250  mL), round-bottom flask (250 mL), Parafilm, and NMR tube. All glassware should be dry and clean. 11. Nuclear magnetic resonance (NMR) spectra (31P) were recorded on a Bruker DRX 500 NMR spectrometer. 12. Deuterated solvent for NMR measurements: Tetrahydrofuran (THF)-d8 (Sigma-Aldrich).

2.1.2. DNA Block Copolymer Solid-Phase Synthesis

1. DNA synthesizer: ÄKTA oligopilot 100 plus DNA synthesizer (GE Healthcare) loaded with Unicorn 5.11 control software (GE Healthcare). Synthesis scale: 260 mmol. 2. Solid support: Polystyrene-based dA Primer Support (GE Healthcare). Loading: 200 mmol/g. 3. Nucleoside phosphoramidites: 0.1 M solution in anhydrous acetonitrile (ACN) (Proligo).

244

Zimmermann et al.

4. Poly(propylene oxide) monobutyl phosphoramidite: 0.1  M solution in anhydrous DCM. 5. Activator: 0.3 M solution of 5-(benzylmercapto)-1H-tetrazole (emp Biotech GmbH, Germany) in anhydrous ACN. 6. Oxidizing reagent: Solution of iodine (0.05  M) in water/ pyridine, 90:10 (v/v) (Merck). 7. Detritylation (Proligo).

reagent:

Dichloroacetic

acid

in

toluene

8. Capping A: N-Methylimidazole (20%, v/v) in anhydrous ACN (Capping A, Proligo). 9. Capping B: Acetic anhydride in anhydrous ACN (Capping B1, Novabiochem) and 2,6-lutidine in anhydrous ACN (Capping B2, Novabiochem). Both solutions are mixed (1:1, v/v) to give the final Capping B solution (acetic anhydride (20%) and 2,6-lutidine (30%) in ACN), which is connected to the DNA synthesizer. 10. Diethylamine (DEA) solution: 20% (v/v) in anhydrous ACN (Proligo). 11. ACN: Water content 1, dilute the stock solution until the A260 is