Developments and Applications of Cyclic Cell Penetrating

Home , Simply Amazing

DEVELOPMENTS AND APPLICATIONS OF CYCLIC CELL PENETRATING

PEPTIDES

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

in the Graduate School of The Ohio State University

Ziqing Qian

Graduate Program in Chemistry

The Ohio State University

2014

Dissertation Committee:

Professor Dehua Pei, Advisor

Professor Venkat Gopalan

Professor Dennis Bong

Ziqing Qian

2014

ABSTRACT

Cell penetrating peptides (CPP) have been featured as a powerful delivery vector for the intracellular delivery of membrane-impermeable cargoes. This dissertation primarily focuses on the development, characterization, and application of a new class of CPP: cyclic cell penetrating peptides. Prompted by long-standing interests of the Pei group in developing cyclic peptides as biological tools and potential therapeutics, I describe our exploration of the permeability properties of the cyclic peptides, leading to the discovery that cyclic peptides with a short sequence motif rich in arginine and hydrophobic residues display significantly higher cellular permeability than their linear counterparts. The resulting cyclic CPP also translocated into the mammalian cell interior at significantly higher efficiencies than canonical arginine-rich linear CPPs. Subsequent internalization mechanistic investigations involving model lipid vesicles and pharmacological inhibitors, as well as genetic mutations, which perturb individual internalization steps, have shown conclusively that cyclic CPPs enter cells through endocytosis and are capable of escaping from early endosomes. To explore the utilities of cyclic CPP as cytoplasmic delivery vehicles, we developed endocyclic, exocyclic, and bicyclic delivery methods to deliver monocyclic peptides, bicyclic peptides, and proteins into cells. Biologically active cargoes, such as fluorogenic phosphatase substrates, phosphatase inhibitors, green

fluorescent protein, and protein tyrosine phosphatase 1B were efficiently delivered by

cyclic CPP. Additionally, we explored a potentially general strategy to deliver linear

peptidyl ligands into mammalian cells through reversible, disulfide bond mediated

cyclization. Cell permeable fluorogenic caspase substrates and cell-permeable CAL-PDZ

domain inhibitors were developed using this method.

The last chapter focuses on the structure-based development of a peptidyl ligand for selectively disrupting calcineurin/NFAT protein-protein interaction (PPI). Through

relatively minor structural modifications, we were able to improve the calcineurin-

binding affinity of peptide VIVIT by ~200-fold. The resulting ligand ranks among the

most potent calcineurin inhibitors reported to date, positioning it for future development

as an efficient – and less toxic – therapeutic alternative to cyclosporin A and FK506.

iii

DEDICATION

Dedicated To My Family

ACKNOWLEDGMENTS

My advisor, Dr. Dehua Pei, has been an excellent mentor. He is full of knowledge,

wisdom, enthusiasm, and kindness. Trained as an organic chemist, I am indebted to him

for his encouragement, excellent guidance and mentoring in conducting this

interdisciplinary research. His dedication and philosophy to science has greatly shaped

my own growth as a scientist along the way. I am especially grateful for the opportunity

that I was given to develop my own individuality. For everything you have done for me,

Dr. Pei, I thank you.

Dr. Tao Liu was also extremely helpful when I joined the Pei group. He is very

knowledgeable and selfless, and I thank him for all of his assistance and guidance in

getting my graduate career started and for our ongoing friendship. I also want to thank Dr.

Qing Xiao, Dr. Punit Upadhyaya, Dr. Thi Trinh, Dr. Ryan Hard, Wenlong Lian, Bisheng

Jiang, Patrick Dougherty, Yujing Zhai and all other current and former Pei group members for their help and support. I have been incredibly fortunate to spend my time

among so many talented and diligent labmates with a strong team spirit. Additionally, I

have been very blessed to have the opportunities in collaborating and working with many

extraordinary professors and scientists over the country. I am indebted to Dr. Jonathan

LaRochelle, Dr. Estelle Cormet-Boyaka, Dr. Chris Hadad, Dr. Dennis Bong, Dr. Venkat

Gopalan, Dr. Dmitri Kudrysshov, Dr. Dean Madden, Dr. Amy Barrios, Dr. Patrick Hogan, v

Dr. Shawn Li and all other personnel involved for their generous guidance and input in

my research.

I thank my committee members, Dr. Dennis Bong and Dr. Venkat Gopalan for their

instrumental assistance during my graduate life at OSU. I also want to thank Dr. Jian

Zhou for providing me a solid synthetic chemistry training. I want to acknowledge friends, scientists and staff I have gotten to know and work with. I have always appreciated all your help and generous support in research, classes and extracurricular activities.

Finally, and most importantly, I would like to thank my family for their unconditional support. My parents, Zhengyi and Yongke, have always been inspiring and supporting of my endeavors. I really appreciate all the effort they have made in educating and caring for me. I would like to thank my brilliant and beautiful wife, Yisha

Yao, for her faith in me and for her unwavering love and encouragement. I also wish to thank Yisha’s parents, Qi and Yinxiang. They are both simply amazing individuals and provide me with unending encouragement and support.

VITA

June 2009 ...... B.S. Chemistry, East China Normal University

2009 to present ...... Graduate Teaching and Research Associate,

Department of Chemistry and Biochemistry,

The Ohio State University

Publication

1. Wenlong Lian, Bisheng Jiang, Ziqing Qian, and Dehua Pei. “Cell-permeable bicyclic peptide inhibitors against intracellular proteins.” J. Am. Chem. Soc. [Online early access]. DOI: 10.1021/ja503710n. 2. Punit Upadhyaya, Ziqing Qian, Nurlaila A. A. Habir, and Dehua Pei. “Direct Ras inhibitor identified from a structurally rigidified bicyclic peptide library” Tetrahedron [Online early access]. DOI: 10.1016/j.tet.2014.05.113. 3. Ziqing Qian, Jonathan R. LaRochelle, Bisheng Jiang, Wenlong Lian, Ryan L. Hard, Nicholas G. Selner, Rinrada Luechapanichkul, Amy M. Barrios, and Dehua Pei. “Early endosomal escape of a cyclic cell-penetrating peptide enables effective cytosolic cargo delivery.” Biochemistry 2014, 53(24), 4034-4046. 4. Anamitra Ghosh, Hariharan Saminathan, Arthi Kanthasamy, Vellareddy Anantharam, Huajun Jin, Gautam Sondarva, Dilshan S. Harischandra, Ziqing Qian, Ajay Rana, and Anumantha G. Kanthasamy. “The Peptidyl-prolyl isomerase Pin1 upregulation and proapoptotic function in dopaminergic neurons.” J. Biol. Chem. 2013, 288(30), 2494–2501

vii

5. Ziqing Qian, Tao Liu, Yu-Yu Liu, Roger Briesewitz, Amy M. Barrios, Sissy M. Jhiang, and Dehua Pei. “Efficient delivery of cyclic peptides into mammalian cells with short sequence motifs.” ACS Chem. Biol. 2013, 8(2), 423-431 6. Varun Dewan, Tao Liu, Kuan-Ming Chen, Ziqing Qian, Yong Xiao, Lawrence Kleiman, Kiran V. Mahasenan, Chenglong Li, Hiroshi Matsuo, Dehua Pei, and Karin Musier-Forsyth. “Cyclic peptide inhibitor of HIV-1 capsid-human lysyl-tRNA synthetase interaction” ACS Chem Biol. 2012, 7(4), 761-769 7. Tao Liu, Ziqing Qian, Qing Xiao, and Dehua Pei. “High-throughput screening of one-bead-one-compound libraries: identification of cyclic peptidyl inhibitors against calcineurin/NFAT interaction.” ACS Comb. Sci. 2004, 15(2), 300-306. 8. Miao Ding, Feng Zhou, Zi-Qing Qian, and Jian Zhou. “Organocatalytic Michael addition of unprotected 3-substituted oxindoles to nitroolefins.” Org. Biomol. Chem. 2010, 8(13), 2912-2914. 9. Zi-Qing Qian, Feng Zhou, Tai-Ping Du, Bo-Lun Wang, Miao Ding, Xiao-Li Zhao, and Jian Zhou. “Asymmetric construction of quaternary stereocenters by direct organocatalytic amination of 3-substitued oxindoles” Chem. Commun. 2009, 28(44), 6753-6755.

Fields of Study

Major Field: Chemistry

viii

TABLE OF CONTENTS

ABSTRACT ...... ii DEDICATION ...... iv ACKNOWLEDGMENTS ...... v VITA vii TABLE OF CONTENTS ...... ix LIST OF TABLES ...... xiii LIST OF FIGURES ...... xiv LIST OF ABBREVIATIONS ...... xvii Chapter 1: Introduction ...... 1

1.1 Cell Penetrating Peptides ...... 1 1.2 Classification of Cell Penetrating Peptides ...... 2 1.2.1 Cationic Cell Penetrating Peptides...... 3 1.2.2 Amphipathic Cell Penetrating Peptide ...... 4 1.2.3 Hydrophobic Cell Penetrating Peptides ...... 6 1.3 Applications of Cell Penetrating Peptides ...... 6 1.3.1 CPPs for Small Molecule Delivery ...... 7 1.3.2 CPPs for Peptide Delivery ...... 8 1.3.3 CPPs for Protein Delivery ...... 10 1.3.4 CPPs for Nucleic Acids Delivery ...... 12 1.4 Cell Penetrating Peptide Internalization Mechanism ...... 14 1.4.1 Endocytosis ...... 15 1.4.2 Direct Translocation ...... 19 1.5 Endosomal Entrapment of Cell Penetrating Peptides ...... 20 Chapter 2: Discovery and Mechanism of Cyclic Cell Penetrating Peptides ...... 22

2.1 Introduction ...... 22 2.2 Results ...... 25 2.2.1 Peptide Cyclization and Hydrophobicity Synergistically Enhance Cellular Association ...... 25 2.2.2 Efficiency and Kinetics of Peptide Internalization ...... 28 2.2.3 Subcellular Distribution of Internalized Peptides ...... 31

2.2.4 cFΦR4 Enters Cells via Endocytosis ...... 34

2.2.5 cFΦR4 Escapes from Early Endosomes into the Cytoplasm ...... 38

2.2.7 Stability and Cytotoxicity of cFΦR4 ...... 42 2.3 Discussion ...... 43 2.4 Conclusion ...... 47 2.5 Experimental Section ...... 47 2.5.1 Materials ...... 47 2.5.2 Peptide Synthesis and Labeling ...... 48 2.5.3 Cell Culture ...... 50 2.5.4 Quantification of Peptide Cellular Association ...... 50 2.5.5 Synthesis of 4-MU Labeled Peptides and 4-MU Internalization Assay ...... 51 2.5.6 Confocal Miscroscopy ...... 53 2.5.7 MLV translocation assay ...... 55 2.5.8 Preparation of Small Unilamellar Vesicles ...... 56 2.5.9 Fluorescence Polarization ...... 56 2.5.10 Serum Stability Test ...... 57 2.5.11 Cytotoxicity Assay ...... 57 2.6 Acknowledgements ...... 58 Chapter 3: Cargo Capacity of Cyclic Cell-Penetrating Peptide ...... 59

3.1 Introduction ...... 59 3.2 Results ...... 60

3.2.1 Endocyclic Delivery of Peptidyl Cargos with cFΦR4 ...... 60

3.2.2 Endocyclic Delivery of Biologically Active Peptides with cFΦR4 ...... 62

3.2.3 Exocyclic Delivery of Peptidyl Cargos with cFΦR4 ...... 67

3.2.4 Exocyclic Delivery of Proteins with cFΦR4 ...... 72

3.2.5 Bicyclic Delivery of Peptidyl Cargos with cFΦR4 ...... 77 x

3.3 Discussion ...... 83 3.4 Conclusion ...... 85 3.5 Experimental Section ...... 86 3.5.1 Materials ...... 86 3.5.2 Peptide Synthesis and Labeling ...... 87 3.5.3 Cell Culture ...... 89 3.5.4 Quantification of CPP Cellular Association ...... 89 3.5.5 Confocal Microscope ...... 90 3.5.6 Flow Cytometry ...... 92

3.5.7 Preparation of cFΦR4–Protein Conjugates ...... 93 3.5.8 Immunoblotting...... 94 3.5.9 Staurosporin-Induced Apoptosis Assay ...... 95 3.5.10 Serum Stability Assay ...... 96 3.6 Acknowledgements ...... 96 Chapter 4: Intracellular Delivery of Linear Peptides by Reversible Cyclization of Cell-

Penetrating Peptides ...... 97

4.1 Introduction ...... 97 4.2 Results ...... 98 4.2.1 Synthesis and Uptake Studies of Disulfide-Bond Cyclized Peptide ...... 98 4.2.2 Delivery of Fluorogenic Caspase Substrate ...... 103 4.2.3 Development of Cell Permeable CAL-PDZ Domain Inhibitor ...... 107 4.3 Discussion ...... 110 4.4 Conclusion ...... 111 4.5 Experimental Section ...... 112 4.5.1 Materials ...... 112 4.5.2 Peptide Synthesis ...... 112 4.5.3 Cell Culture ...... 114 4.5.4 Confocal Microscopy ...... 115 4.5.5 Flow Cytometry ...... 115 4.5.6 Peptide Stability Assay ...... 116 4.5.7 In Cellulo Fluorimetric Assay ...... 116

4.5.8 In Vitro Fluorimetric Assay ...... 117 4.5.9 Fluorescence Polorization ...... 117 4.5.10 SPQ Intracellular Chloride Concentration Assay ...... 118 4.6 Acknowledgements ...... 119 Chapter 5: Structure-Based Optimization of a Peptidyl Inhibitor against Calcineurin-

NFAT Interaction ...... 120

5.1 Introduction ...... 120 5.2 Results and Discussion ...... 122 5.2.1 Substitution of tert-Leucine (Tle) for Valine ...... 122 5.2.2 Incorporation of of Cys(ΨMe,MePro) as cis-Pro Analog ...... 123 5.2.3 Molecular Modeling...... 129 5.2.4 Inhibition of Nuclear Translocation of NFAT ...... 131 5.3 Discussion ...... 134 5.4 Conclusion ...... 135 5.5 Experimental Section ...... 135 5.5.1 Materials ...... 135 5.5.2 Cell Culture ...... 136 5.5.3 Peptide Synthesis, Labeling and Conjugation ...... 136 5.5.4 Fluorescence Anisotropy ...... 138 5.5.5 Proteolytic Stability Assay ...... 139 5.5.6 Molecular Modeling...... 139 5.5.7 GFP-NFAT Translocation Assay ...... 141 5.6 Acknowledgements ...... 141 REFERENCE ...... 142

xii

LIST OF TABLES

Table 1. Representative Cell Penetrating Peptides...... 3

Table 2. Sequences and Cellular Association of Linear and Cyclic Peptides ...... 27

Table 3. Sequences of CPPs Studied in Chapter 2 ...... 35

Table 4. Sequences and Cellular Association Efficiency of Monocyclic Peptides ...... 61

Table 5. Sequences of Peptides Studied in Chapter 3 ...... 64

Table 6. Kinetic Activites of Recombinant PTPs toward pCAP-Containing Peptides..... 65

Table 7. Kinetic Activities of PTP1B and cFΦR4-PTP1B against pNPP ...... 76

Table 8. Sequences of Peptides Studied in Chapter 4 ...... 99

Table 9. In vitro Activity of Fluorogenic Substrate against Recombinant Caspase-3. ... 104

Table 10. Sequences and Dissociation Constants of Peptidyl Ligands ...... 122

xiii

LIST OF FIGURES

Figure 1. Applications of CPPs as delivery vectors...... 7

Figure 2. Proposed mechanisms of CPP internalization...... 16

FITC FITC Rho Dex Figure 3. Structures of cFΦR4 , FΦR4 , cFΦR4 and cFΦR4 ...... 24

Figure 4. Comparison of the cellular association efficiencies of FITC-labeled linear and

cyclic peptides...... 27

Figure 5. 4-MU based CPP internalization assay...... 30

Figure 6. Live-cell confocal images of MCF-7 cells treated with FITC-labeled CPPs. ... 32

FITC Figure 7. Live-cell confocal images of A549 cells treated with cFΦR4 ...... 34

+ + Figure 8. cFΦR4 enters cells through endocytosis, localizes to Rab5 and Rab7

endosomes, and releases from early endosomes into cytoplasm...... 37

Figure 9. Confocal images of MLVs after incubation with 1 μM free TAMRA dye (red)

FITC FITC FITC FITC and 1 μM dextran (control), R9 , FΦR4 , or cFΦR4 for 30 min...... 40

Figure 10. Binding of cFΦR4, R9, and Tat to SUV (A) and heparan sulfate (B)...... 41

Figure 11. Serum stability and cytotoxicity studies of cyclic CPP...... 43

Figure 12. Scheme showing the points along the endocytic pathway where cFΦR4, R9, and Tat escape into the cytoplasm...... 46

xiv

Figure 13. Structures showing cargo attachments for endocyclic (A), exocyclic (B), and bicyclic delivery (C) of cargos (shown in red) with cFΦR4...... 59

Figure 14. Structures of cyclic-pCap, bicyclo-PTP1B, cFΦR4-PCP, and cFΦR4-CHO as examples of endocylic, bicyclic, and exocyclic delivery methodologies...... 65

Figure 15. Endocyclic delivery of phosphatase fluorogenic substrate with cFΦR4...... 66

Figure 16. Flow cytometry of MCF-7 cells treated with pCAP-containing peptides...... 67

Figure 17. Exocyclic peptides delivery with cFΦR4...... 68

Figure 18. Exocyclic delivery of pCAP-containing peptides with cFΦR4...... 70

Figure 19. Protection of Jurkat cells from staurosporine-induced apoptosis by caspase inhibitor DEVD-H (CHO), its peptide conjugates LCPP-CHO and cFΦR4-CHO, or prodrug FMK...... 72

Figure 20. Intracellular delivery of GFP with cFΦR4...... 75

Figure 21. Scheme showing the synthesis of (A) CPP-S-S-GFP and (B) CPP-PTP1B. .. 76

Figure 22. Intracellular delivery of PTP1B with cFΦR4...... 77

Figure 23. Bicyclic peptide delivery with cFΦR4...... 80

Figure 24. Evolution of a cell-permeable bicyclic PTP1B inhibitor.229 ...... 81

Figure 25. Delivery of F2Pmp containing bicyclic peptide with cFΦR4...... 83

Figure 26. Comparison of the serum stability of monoPTP1B and bicycloPTP1B...... 83

Figure 27. Scheme showing the reversible cyclization strategy for delivering linear peptidyl cargos into cells. GSH, glutathione...... 98

Figure 28. (A) Synthesis of disulfide-bond cyclized peptide. (B) Synthesis of thioether- bond cyclized peptide. (C) Structures of FITC-labeled peptides 1 and 2...... 100 xv

Figure 29. Disulfide-bond cyclization enhance cellular uptake...... 101

Figure 30. Evaluation of cytoplasmic delivery efficiencies of 1-PCP and 2-PCP...... 102

Figure 31. Comparison of the proteolytic stability of peptides 1 and 2...... 103

Figure 32. Structures of caspase fluorogenic substrates 3-7...... 105

Figure 33. Time-dependent release of coumarin product by Jurkat cells...... 106

Figure 34. Evaluations of cell-permeable CAL-PDZ domain inhibitor...... 108

Figure 35. CAL-PDZ domain binding studies (A) and HeLa cell uptake studies (B) of control peptides 9 and 10...... 110

Figure 36. Binding modes of VIVIT and ZIZIT-cisPro to Calcineurin...... 124

Figure 37. Binding assay of VIVIT, ZIZIT, and ZIZIT-cisPro to CN...... 125

Figure 38. Synthesis of peptide ZIZIT-cisPro...... 126

Figure 39. Competitions for binding to CN by inhibitors VIVIT, ZIZIT, and ZIZIT- cisPro...... 128

Figure 40. Comparison of serum stability of VIVIT and ZIZIT-cisPro...... 128

Figure 41. MD simulation of ZIZIT-cisPro with CN...... 130

Figure 42. Preparation and sequences of R11 conjugated CN inhibitors...... 133

Figure 43. Inhibition of nuclear translocation of GFP-NFAT by CN inhibitors...... 133

xvi

LIST OF ABBREVIATIONS

α alpha Ac acetyl Abu 2-aminobutyric acid AU arbitrary units Amc 7-amino-4-methylcourmarin β beta B β-alanine Boc tert-butoxycarbonyl Baf bafilomycin °C degree Celsius CN calcineurin CsA cyclosporine A CPP cell penetrating peptide CF cystic fibrosis CFTR cystic fibrosis transmembrane conductance regulator CAL CFTR-associated ligand CoA coenzyme A DCM dichloromethane DIC diisopropylcarbodiimide DIPEA diisopropylethylamine DMAP 4-(N,N-dimethylamino)pyridine DMF N,N-dimethylformamide DMSO dimethyl sulfoxide Dex dexamethasone Dyna dynasore DPBS Dulbecco’s phosphate buffered saline EIPA N-ethyl-isopropyl amiloride EDTA ethylenediaminetetraacetic acid xvii equiv. Equivalent F fluoride Fmoc 9-fluorenylmethoxycarbonyl FA fluorescence anisotropy FACS fluorescence-activated cell sorting FP fluorescence polorization

F2Pmp L-4-(phosphonodifluoromethyl)phenylalanine FITC fluorescein isothiocyanate FBS fetal bovine serum FU fluorescence unit GST Glutathione S-Transferase GFP green fluorescence protein GSH Glutathione S-Transferase hr hour HMBA 4-hydroxymethylbenzoic acid HATU 2-(1H-7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate HBTU 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate HEPES N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid HOBt 1-hydroxybenzotriazole HPLC high performance liquid chromatography J L-2,3-diaminopropionic acid

KD Dissociation Constant μ micro M moles per liter 4-MU 4-methylumbelliferone MBCD methyl-β-cyclodextrin MLV multilamellar vesicle MFI mean fluorescence intensity miniPEG 8-amino-3,6-dioxaoctanoic acid xviii

MALDI-TOF matrix assisted laser desorption ionization/time of flight mass spectrometry Nle L-norleucine NMR nuclear magnetic resonance NEMO NF-κB essential modulator NFAT nuclear factor of activated T cells Nal (Φ) L-2-naphthylalanine PPI protein-protein interaction PTD protein transduction domains pCAP L-phosphocoumaryl aminopropionic acid pY phosphotyrosine PTP protein tyrosine phosphatase PyBOP benzotriazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate Rho rhodamine B siRNA small interfering RNA SUV small unilamellar vesicle SPQ 6-methoxy-N-(3-sulfopropyl) quinolinium SPDP succinimidyl 3-(2-pyridyldithio)propionate SASA solvent accessible surface area 5(6)-SFX 5(6)-fluorescein-6(5)-carboxamidohexanoic acid, succinimidyl ester TFA trifluoroacetic acid TIPS triisopropylsilane TAMRA tetramethylrhodamine THF tetrahydrogenfuran Tm trimesoyl TCEP tris(carboxylethyl)phosphine Tle (Z) L-tert-leucine Wort wortmannin

Standard one- or three-letter codes are used for amino acids. xix

Chapter 1: Introduction

1.1 Cell Penetrating Peptides

The plasma membrane serves as a physical barrier for all living systems.

Unfortunately, it also presents a major obstacle for the development of therapeutics, diagnostics, and probes using biologics, such as peptides, protein, and oligonucleotides.

The hydrophobic core of cellular membranes impedes the passage of hydrophilic molecules. Multiple methods including microinjection, electroporation and viral-based vectors have been developed to deliver biologics into cell interior. However, these strategies suffer from high invasiveness, toxicity, and low efficiency. On the other hand, nature has evolved ingenious solutions to overcome the challenge. In 1988,1,2 two

research groups independently discovered that HIV trans-activator of transcription (TAT)

internalizes into mammalian cells and activates viral replication. The active component of

this protein, Tat peptide (RKKRRQRRR, Table 1), was identified and characterized. The

ability of this highly polar and positively-charged peptide to cross the mammalian cell

membrane immediately attracted attention. Combined with the discovery of the 16-mer

peptide penetratin (Table 1), derived from the third helix homeodomain of

Antennapedia,3 peptides showing similar cell-penetrating capacities have been discovered

or rationally designed to serve as a new type of non-invasive molecular vectors. These 1

versatile peptides, which are capable of transporting a wide varieties of cargos across the cell membrane, are collectively called “cell-penetrating peptides (CPPs)” or “protein transduction domains (PTDs)”.

Since the initial discovery of the Tat peptide in the late 1980s, numerous other CPPs

derived from natural proteins or rational design have been reported.4-9,26 CPPs have been

used to deliver small molecules,10,11 nucleic acids,12,13 siRNAs,14-17 proteins,18-20 and nanoparticles21-23 both in vitro and in vivo through either covalent attachment or

noncovalent association. Many CPPs display minimal toxicity and immunogenicity at

physiologically relevant concentrations.24,25

1.2 Classification of Cell Penetrating Peptides

Different criteria have been used to classify CPPs. For example, CPPs could be

divided into subgroups based on the origin of their sequences. One CPP subgroup is

derived from natural protein transduction domains, such as penetratin and Tat. Other

CPPs were partially or completely designed, such as transportan,27 Pep-1,28 loligomers,29 and arginine-rich peptides26. For the hundreds of CPPs collected from publications and

patents, they have also been conveniently grouped into three subgroups on the basis of

their amino acid sequences and physical chemical properties: cationic, amphipathic, and

hydrophobic CPPs (Table 1).30

Table 1. Representative Cell Penetrating Peptides.

Peptide Sequence Origin Ref. Cationic Tat peptide RKKRRQRRR HIV trans-activator of transcription [1,2] Penetratin RQIKIWFQNRRMKWKK Antennapedia homeodomain [3] R9 RRRRRRRRR Designed [33]

Amphipathic Transportan GWTLNSAGYLLGKINLKALAALAKKIL Galanin and mastoparan [27] P1 MGLGLHLLVLAAALQGAWSQPKKKRKV Human immunoglobulin and SV40 [40] MPG GALFLGFLGAAGSTMGAWSQPKKKRKV HIV gp41 and SV40 [41] Pep-1 KETWWETWWTEWSQPKKKRKV Tryptophan-rich cluster and SV40 [28] MAP KLALKLALKALKAALKLA Chimeric [45] SAP (VRLPPP)3 N-terminal domain of γ-zein [48]

Hydrophobic MTS AAVALLPAVLLALLAP Kaposi-fibroblast growth factor [51] PFV PFVILI C105Y [52]

1.2.1 Cationic Cell Penetrating Peptides

CPPs that have a relative high abundance of positively charged amino acids (lysines

or arginines) are classified as cationic CPPs, and are the most widely investigated among

all CPPs. Studies carried out with Tat and penetratin revealed that cell-penetrating

activities does not depend on the peptide backbone, but rather on the number and position

of positively charged residues.26 For example, the Tat peptide is composed of highly

basic residues: six arginines and two lysines. Replacement of any of the basic residues

with an alanine decreases its cell-penetrating activity. It is empirically known that the

cationic nature of the peptide facilitates interactions with the cell surface.

Thermodynamic studies of CPPs with various anionic species present on the surface of

cell membranes, including phospholipids and proteoglycans, have revealed the

importance of the electrostatic interactions during the initial step in cationic CPP internalization.31,32

To assess the contribution of lysines and arginines to the cell penetration, homooligomers of arginine, lysine, and histidines were synthesized and compared.

Polyarginine peptides of 7-9 residues showed significantly higher efficiency than polyhistidines or polylysine peptides, or even the Tat peptide.33 Wender and coworkers

hypothesized that the differences between arginine and lysine may originate from the

ability of the arginine side chain, the guanidinium group, to form bidentate hydrogen

bonds with cell surface phosphates, carboxylates, and sulfates.34 Consistent with their

hypothesis, synthetic, non-peptidic oligomers that display guanidinium groups, such as branched chain arginine polymers,26 guanidinium-rich oligocarbamates,35 polyguanidino

dendrimers,36 guanidinium-rich oligocarbonates,37 guanidinylated carbohydrates,38 and

others39 are all active “CPPs”. Typically, the most commonly used cationic CPPs include

R9, Tat, and penetratin (Table 1).

1.2.2 Amphipathic Cell Penetrating Peptide

Amphipathic CPPs, such as transportan (Table 1),27 contain both polar (hydrophilic)

and nonpolar (hydrophobic) residues. Peptides with amphipathic character derived from

either the primary sequence or the secondary structure are further defined as primary or

secondary amphipathic peptides. To elaborate, primary amphipathic peptides contain

sequential hydrophobic and hydrophilic domains, while secondary amphipathic peptides

display their amphipathic properties under certain conformational states, which allow the

proper positioning of the polar and nonpolar residues.

Primary amphipathic CPPs are normally chimeric peptides containing a common

hydrophilic nuclear localization sequence (NLS), PKKKRKV, and a hydrophobic domain

derived from a signal peptide like P1 (MGLGLHLLVLAAALQGA-WSQ-

PKKKRKV)40, a fusion peptide like MPG (GALFLGFLGAAGSTMGA-WSQ-

PKKKRKV)41, or a tryptophan-rich sequence like Pep-1 (KETWWETWWTE-WSQ-

PKKKRKV)28 (Table 1). Many of these peptides bear a WSQ sequence as a linker

connecting the hydrophilic and hydrophobic domains. Several other protein derived

primary amphipathic CPPs include pVEC (LLILRRRIRKQAHAHSK) from murine

42 vascular endothelial-cadherin protein , ARF1-22 (MVRRFLVTLRIRRACGPPRVRV)

43 from p14ARF protein , and BPrPr1-28 (MVKSKIGSWILVFVAMWSDVGLCKKRP)

from bovine prion protein44 (Table 1).

Secondary amphipathic CPPs adopt either an α-helical structure or a β-sheet

structure. In the first case, hydrophilic residues are placed in positions i, i+3/i+4, i+7 and

so on, so that highly hydrophobic and hydrophilic patches form on opposite sides of the

molecule. This is exemplified by the model amphipathic peptide (MAP) family, including

peptide MAP (KLALKLALKALKAALKLA, Table 1), whose cellular uptake efficiency

closely correlates with its α-helicity.45 In the second case, amphipathic β-sheet CPPs are

based on the assembly of alternating hydrophobic and hydrophilic amino acids which are exposed to the solvent, such as vT5 (DPKGDPKGVTVTVTVTVTGKGDPKPD) from viral protein45.

A special group of secondary amphipathic CPPs are the proline-rich peptides based

on the polyproline II helix conformation.47 Some proline-rich amphipathic peptides are

derived from the N-terminal domain of γ-zein and bactenecin-7, such as Trimer peptide

48,49 (VRLPPP)3 (Table 1). Synthetic derivatized amphipathic polyprolines with excellent

cell penetrating activity have also been prepared and evaluated based on the proline pyrrolidine template.50

1.2.3 Hydrophobic Cell Penetrating Peptides

Peptides containing mostly non-polar residues are classified as hydrophobic peptides.

Limited by aqueous solubility and toxicity, long hydrophobic amino acids sequences are more commonly conjugated with cationic residues for enhanced uptake and delivery.

Only a few hydrophobic sequences have been established as CPPs, such as Kaposi fibroblast growth factor cell-membrane permeable motif AAVALLPAVLLALLAP51 and short hydrophobic peptide PFVILI52 (Table 1).

1.3 Applications of Cell Penetrating Peptides

Small molecules, peptides, protein, and nucleic acids are all capable of influecing

cellular function and serving as biological tools or therapeutic agents. However, many of

these molecules display no or little biological activities because of their limited

membrane permeability. Thus, CPPs have been widely utilized to facilitate the cellular

uptake of these molecules. This section will focus on introducing prototypical cases of

CPPs as cellular Trojan horses for delivering bioactive cargos. 6

Figure 1. Applications of CPPs as delivery vectors.

1.3.1 CPPs for Small Molecule Delivery

In drug discovery, many small molecule drug candidates that exhibit desirable

activities in vitro do not possess the sufficient lipophilicity that is mandated for

membrane partitioning while allowing necessary aqueous solubility.53 In other cases, small molecule therapeutics display poor cellular responses because they are substrates of the efflux system. These scenarios have limited the application and efficacy of many compounds and therefore, improving translocating efficiency is a top priority. Many small molecules have shown improved activity when conjugated with CPPs. Polyarginine conjugates of cyclosporine A can penetrate skin, delivering a therapeutically significant amount of immunosuppressive drug as potential treatment for psoriasis and other skin 7

inflammatory disorders.10 Polyarginine conjugates with hydrophobic drug Paclitaxel showed greatly increased water solubility, cellular uptake, and activity against drug- resistant cancer cell lines.54 In another example, Tat or penetratin conjugates with

doxorubicin are capable of overcoming drug-resistance caused by P-glycoprotein

mediated drug efflux.55,56 Similar therapeutic benefits were also observed in the study of

CPP-methotrexate conjugates.57 Cell penetrating peptides are proposed to divert the

conjugated small molecular cargos away from efflux pathways, providing an efficient

method to counteract resistance.

1.3.2 CPPs for Peptide Delivery

Many peptides possess biological activities that make them potential therapeutics.

Applications of potential peptide drugs are diverse, ranging from cancer treatment to

antibiotic and antiviral agents. Peptides have also been utilized in recent years as building

blocks in the fields of medical chemistry and nanomedicine. The rise of peptide drugs is

accompanied with challenges in their membrane permeability. One pronounced issue

when aiming for intracellular targets is the peptides’ hydrophilic nature and the resulting

poor permeability to cell membranes.58 To overcome this limitation, various approaches

have been investigated, including prenylation,59 pepducins,60 α-helical stapling

peptides,61 and N-methylation.62 Unfortunately, none of these approaches provide a

general solution, necessitating extensive case-by-case investigation. CPP, on the contrary,

potentially offers a general method for delivering peptides with low toxicity and synthetic

complexity.

An attractive feature of peptide ligands is their ability to target protein-protein

interactions, which are challenging to small molecules as the required interaction surface

is too large. Conjugation of these ligands with CPP has yielded impressive biological

tools and potential therapeutics. For example, Tat improved uptake and activity of a

tumor suppressor p53-derived peptide to restore p53 functionnality in cancer cells.63

Similarly, penetratin was used to enhance the uptake of PNC-28, a peptide derived from

MDM-2 binding domain of p53. The resulting conjugate was shown to block pancreatic cancer cell proliferation in vivo.64 Another interesting peptide inhibitor, VIVIT, blocks

the specific interaction of calcineurin and its substrate NFAT (nuclear factor of activated

T cells) in vitro.65 Oligoarginine-conjugated VIVIT provided immunosuppression in

transplanted mice and may serve as alternative therapeutics for Tacromilus.66

Deregulation of proper apoptotic pathways has been directly or indirectly associated

with many diseases, such as cancer. Thus, peptides or protein fragments that regulate

apoptosis are under development as potential therapeutics. CPPs were used to enhance

the uptake of a peptide derived from the N-terminus of the Smac protein, which

antagonizes the anti-apoptotic activity of IAPs (inhibitors of apoptosis).67 Cell permeable

Smac peptide sensitizes cells to chemotherapy induced apoptosis and induce tumor regression in vivo.68,69 Another example is based on the BH4 domain of Bcl-2/Bcl-xL

interaction, which is essential for preventing apoptotic signaling in mitochondria. A

peptide derived from BH4, when conjugated with Tat, can prevent apoptosis and

mitochondrial dysfunction in vivo.70 BH3 peptides from the Bcl-2 protein family are responsible for Bak/Bcl-xL interaction that possesses anti-apoptotic function. The BH3

peptide has been investigated for use in anti-cancer therapeutics, but is not permeable

enough to achieve a sufficient biological response. Conjugation of BH3 peptides to

amphipathic CPP, penetratin, results in a potent apoptosis inducer as a potential

anticancer reagent.71

CPPs have also been widely used to deliver peptides for targeting transcription

factors. A rationally designed phosphotyrosine containing peptide was found to prevent

the dimerization of transcription factor Stat6 and inhibit its activity. Fused to the Tat-

derived CPP PTD4, the cell permeable inhibitor deactivated Stat-6 function in vitro and ex vivo and demonstrated its therapeutical potential for allergic airway diseases.72

Activation of transcription factor Stat-3 is critical for tumor cell survival and proliferation.

A 20-mer peptide was identified from a yeast-two-hybrid screening to target dimerization of Stat-3. The selected Stat-3 peptide inhibitor was conjugated to nonaarginine and showed inhibition of Stat-3 signaling and growth inhibition of human myeloma cells.73

Another example of CPP-mediated modulation of transcription activity involved the inhibition of NEMO (NF-κB essential modulator). Penetratin was conjugated with a rationally designed peptide that is inhibitory of normal trimerization step of NEMO subunits. The product inhibits NF-κB activity in human restinoblastoma cells and serves as a potential anti-inflammation and anti-cancer agent.74

1.3.3 CPPs for Protein Delivery

The first example of using CPP to deliver a protein in vivo was established by

Dowdy and coworkers in 1999, who showed that a Tat-β-galactosidase fusion protein

was delivered into into all tissues including brain following intraperitoneal injection into mice.75,76 The same group established a facile bacterial expression vector from which proteins fused with N-terminal Tat CPP sequence can be easily expressed.77 In this method, a cDNA encoding the protein of interest is first cloned in-frame with an N- terminal CPP tag, (His)6 tag, or other appropriate sequences. The recombinant proteins

can be expressed and purified from E. coli. Based on this technique, a large number of

Tat fusion proteins spanning 15 to 121 kDa in size have been purified and delivered into

mammalian cells.18,58

One major application of CPP-fusion protein delivery is the development of novel

anti-cancer reagents. In this category, tumor suppressors and protein inhibitors of cyclin-

dependent kinases (p53, p27, pRb, p16, p15, p21, and merlin protein) have been

delivered into both primary and transformed cells.18 Tat-fusion protein also successfully

delivered antioxidant enzymes to treat disorders pertaining to oxidative stress. This

includes a human Cn, Zn-superoxide dismutase and human liver catalase.78,79 Two other

widely investigated delivery targets are the green fluorescent protein (GFP) and the DNA recombinase Cre.80,81 Both of them have been used to study the internalization pathways

of Tat-fusion proteins,82,83 and to examine delivery efficiency of other newly developed

protein transduction domains.84,85 In some other cases, the antibodies need to be delivered

into cells for their clinical effects.86 Genetically engineered Tat-fusion proteins have been

employed to facilitate this task. For example, Tat CPP sequence has been expressed with the B domain of staphylococcal protein A (SpA), which binds to the Fc fragment of IgG.

The heterodimer of SpA and IgG is capable of delivery IgG in a time and dose-dependent

manner.87

Other CPPs (penetratin, polyarginine, VP22, etc.) have also been used to deliver

biologically active protein into mammalian cells. Illustrated by tumor suppressor p53

where polyarginine fused p53 protein effectively internalized into cancer cells,

translocated into the nucleus, and enhanced cisplatin-induced apoptosis in oral cancer

cells.88 The entire p53 protein has also been effectively delivered into human

osteosarcoma cells by fusing with polypeptide VP22, where it induces apoptosis.89 In a recent development, the N-terminal domain of Clostridium botulinum C2 was found to be an efficient delivery vehicle into eukaryotic cells. The fusion of this internalization sequence C21N and streptavidin was capable of delivering biotinylated p53 protein without obvious cytotoxicity.90

1.3.4 CPPs for Nucleic Acids Delivery

Administering exogenous nucleic acids presents as an exceptionally promising strategy for treating many diseases. One of the biggest challenges, which arises from the

highly negatively charged character of oligonucleotides, is that it renders the

macromolecule impermeable to plasma membranes. Conventional methods, for example,

lipofectamine formulation, to deliver these highly negatively charged macromolecule

suffer from poor efficiency and cellular toxicity. Therefore, CPPs have been widely

exploited as a versatile vector for in vitro and in vivo nucleic acid delivery.12,14,91 In

general, four classes of nucleic acid cargos have been delivered into mammalian cells

using CPPs: DNA plasmid,92,93 antisense oligonucleotides,94 decoy DNA,95 and small

interfering RNA (siRNA).96 A brief overview of the progresses and challenges of CPP-

assisted siRNA delivery is provided below.

SiRNA is a class of double-stranded RNA (20-25 base-pairs) involved in the RNA

interference pathway. Technologies based on siRNA have generated a great deal of

interest in both biomedical research and therapeutic applications given its ability to knock

down, in essence, any gene of interest.97,98 To improve the cellular uptake of siRNA, they

have been crosslinked with Tat peptide through a carbamate linker and delivered into

cultured mammalian cells. The efficiency of Tat-siRNA complex in silencing the target

GFP gene is comparable to the canonical lipofectamine technology.99 CPPs, penetratin and transportan, have been conjugated through a releasable disulfide linker using a 5’- thiol of the sense strand of siRNA to target luciferase or GFP reporters.17 Once the CPP-

siRNA complex reaches the cytosol, the disulfide bond is reduced and RNA molecules,

without any possible intereference from CPP, are generated. The releasable disulfide

bond linker has also been used in penetratin-siRNA complexes targeting endogenous

caspse-3 in primary neuronal cultures, resulting in increased resistance to apoptosis.100 In the above examples, no purification steps were performed following the conjugation procedure, so it is difficult to determine whether covalently attached CPP or nonspecifically bound CPP is responsible for the siRNA delivery. Two subsequent publications showed that HPLC purified CPP-siRNA conjugates required a much higher concentration to achieve similar levels of cellular responses. Additionally, the sites of conjugation of the CPPs to siRNAs also affect cellular activities.101,102

As an alternative to covalent conjugation, non-covalent CPP-siRNA complexes,

which are formed through electrostatic and hydrophoblic interactions have been reported

for various applications. Chimeric amphipathic CPP MPG and its analogues were used

for cellular delivery of multiple exogenous siRNA as a non-covalent complex into

cultured mammalian cells.103 Another chimeric CPP bearing oligoarginine and a peptide

derived from rabies virus glycoprotein enabled the transvascular delivery of siRNA

across blood-brain barrier through non-covalent CPP-siRNA complex.104 In an in vivo

study, antibody conjugated nonaarginine and antiviral siRNA were able to form a non-

covalent complex that significantly suppresses HIV infection in humanized mice.105

Despite many successful examples of siRNA delivery using electrostatically attached

CPP, a large excess of CPP molecules is required for efficient delivery (50:1 to 100:1 molar ratio).103 This large excess of CPP may induce unknown side effects, especially for

in vivo application. Another obstacle in CPP-mediated siRNA delivery is the entrapment

of CPP-cargo complex in endosomes and its inability to cross endosomal membranes.

This so-called “endosomal entrapment” or endosomal escape deficiency, in many cases,

seriously affects CPP delivery applications and will be discussed in details later.

1.4 Cell Penetrating Peptide Internalization Mechanism

Despite three decades of investigation about the mechanism of CPPs’ internalization behavior, the precise mode of action remains elusive. The reason for the difficulty appears to be that the precise mechanism depends on CPP identity, cargo type, concentration of CPP-cargo complex, cell line, and many other factors.106,107 Early 14

studies of the most commonly used cationic CPPs, Tat and nonaarginine, suggested that

these peptides enter the cell by a passive, energy-independent process.108,109 Yet, those experiments were often conducted using confocal microscopy or flow cytometry on fixed cells, and it was later demonstrated that fixation of cells with methanol/formaldehyde altered the intracellular distributions of CPPs.110 Since then, there is a general consensus

that combination of multiple energy-dependent endocytic pathways contributes to the

internalization of most CPP and CPP-conjugates, especially when the cargo is a

macromolecule.111,112 This does not mean that direct translocation cannot mediate any

CPPs’ internalization. In fact, some studies have demonstrated the involvements of both

endocytic and direct translocating pathways in a mutually complementary fashion.113-115

1.4.1 Endocytosis

Endocytosis is a highly regulated energy-dependent process of internalization of

materials from extracellular matrix into cells. As the major cellular uptake pathway for

most CPP-conjugates, especially in case of cationic CPPs,116 endocytosis of CPPs

includes two steps, endocytic entry followed by endosomal escape. The first step of

endocytosis occurs through various pathways that can be classified into clathrin-mediated

endocytosis,117 caveolae and/or lipid-raft-mediated endocytosis,118,119 macropinocytosis,83,120 and phagocytosis. Following the initial endocytic entry from the

plasma membrane, extracellular materials including CPPs are sorted into distinct

membrane compartments, endosomes.121 As the second step of CPPs’ endocytic journey,

CPPs or CPP-conjugates must escape from endosomal compartments in order to avoid

degradation in lysosome and to exert desired biological activities. The optimal conditions

for emdosomal escape are unfortunately not yet clear. We will focus here on three

canonical CPPs, Tat, oligoarginine, and penetratin, and summarize the various findings

regarding to their internalization mechanisms.

Figure 2. Proposed mechanisms of CPP internalization.

Macropinocytosis describes the actin rearrangements that lead to membrane ruffling

and protrusion followed by vesicle invagination. These large endocytic vesicles with

diameter normally greater than 1 μm are called macropinosomes.122 Macropinocytosis was demonstrated to be implicated in the internalization of polyarginines, Tat, and to a 16

less extent of penetratin.82,83,120,123 These observations relied on the inhibition of cellular

uptake or the inhibition of macropinosome formation by the treatment of an amiloride

analogue, 5-(N-ethyl-N-isopropyl)amiloride (EIPA), and of cytochalasin D, which inhibit

F-actin polymerization. It was also demonstrated that rearrangement of the actin

cytoskeleton is induced after treatment with oligoarginine. In a recent study that

investigating the Tat peptide, following the initial internalization, internalized Tat was

proposed to interact with cytoskeleton and trigger the actin rearrangements and

macropinocytosis.124 On the contrary, arginine-rich CPPs were proposed to bind and

cluster proteoglycans on the cell surface first followed by intracellular GTPase Rac

activation, actin skeleton rearrangements, and macropinocytosis. Using proteoglycan-

deficient cells, the indispensable roles of proteoglycan for the induction of

macropinocytosis has been demonstrated.120 How membrane-associated proteoglycans

transmit the CPP binding signal to induce actin organization is yet unclear. Taken

together, these results strongly suggest that macropinocytosis plays a significat role in

endocytic entry of cationic CPPs.

Clathrin- and caveolae-mediated endocytic pathways are mediated by concave

structures on plasma membranes, namely, clathrin-coated pits (about 100 nm in diameter)

or flask-shape caveolae invagination (about 50 nm in diameter). Both pathways have

been recognized to contribute to the uptake of cationic CPPs. Multiple studies

demonstrated the involvement of clathrin-dependent endocytosis based on uptake

inhibition either by hyperosmolar medium or chlorpromazine.117,125 While others

supported the importance of caveolae-mediated endocytosis evidenced by Tat peptide’s

colocalization with caveolae marker caveolin-1 and the uptake inhibition by methyl-β-

cyclodextrin.118,119 Interestingly, in the study of Tat peptide internalization by Chinese

hamster ovary (CHO) cells, both nystatin and filipin III, known caveolin-dependent

endocytosis inhibitors, did not affect the uptake, suggesting that caveolae-dependent

endocytosis might be less critical.117 Similar contradictory results were also observed

when studying the uptake of 10 μM Tat peptide. Neither clathrin knock-down nor

caveolin knock-out affect the uptake of Tat peptide.126 This report, among others, clearly

suggests that the precise internalization mechanism of CPPs could differ due to some

critical experimental conditions, such as cell type, cargo property, incubation time,

concentration, and cell density.106,107,112 Above a certain threshold, endocytosis- independent mechanisms become possible, while at lower concentrations the uptake is

primarily endocytic.126

It should be noted that previous studies have relied on the use of pharmacological

inhibitors and colocalization studies between CPP-fluorophore and protein markers to

access the contributions of various endocytic pathways to CPP uptake. These methods

may provide misleading information due to poor inhibitor selectivity, cell line

dependence and possible cytotoxicity.127 Although many studies have provided evidence

to support one particular internalization pathway to be more convincing than another,

macropinocytosis, clathrin-mediated endocytosis, and caveolae/lipid-raft mediated

endocytosis or even direct translocation could all be involved simultaneously.

1.4.2 Direct Translocation

Direct translocation across the plasma membrane is an energy-independent process.

Unlike passive diffusion across the membrane, direct translocation harnessed by CPPs is driven by membrane potential. As described above, early observations of efficient internalization of cationic peptides at 4 °C is now attributed to the artifacts caused by fixation and strong cell surface absorption of the peptides. However, direct translocation mechanisms were still evidenced using artificial model systems, using live-cell confocal microscopic imaging at 4 °C or in the presence of endocytosis inhibitors.111,114,128-130

Various models have been proposed to explain direct translocation of CPPs, such as inverted micelle formation,131 transient-pore formation,132 adaptive translocation,133 among others.134,135.

The inverted micelle formation model was first proposed to explain the behavior of

penetratin.131 In this hypothesis, both positively-charged and hydrophobic residues are

involved in membrane binding, which subsequently induce destabilization and

reorganization of the membrane structure into an inverted micelle that encapsulates the

CPPs. The inverted micelle can then release the CPPs on the intracellular side. The model

was supported by a molecular modeling simulation and other studies on membrane

deformation upon CPP-membrane association.136-138 Transient-pore formation process is

also induced by the hydrogen bond and electrostatic interaction between peptide side

chains and phospholipids.140,154 The accumulation of the peptides was proposed to cause

membrane thinning followed by transient toroidal pore formation that allows for the

diffusion of cationic CPPs across the membrane.139 Studies with other CPPs corroborated 19

a similar process using fluorescence and calorimetric methods.128,140-142 In the adaptive

translocation model, the guanidinium/phosphate binding is proposed to provide the

oligoarginine peptides sufficient hydrophobicity to enable adaptive diffusion into and

across the membrane. This diffusion is driven by the membrane potential across the

plasma membrane.34 It is worth noting that most of these demonstrated pathways are

focusing on cell penetrating peptides themselves or with small cargos (e.g. fluorophore)

attached. The cargo capacity of direct translocation mechanism is yet unclear.

1.5 Endosomal Entrapment of Cell Penetrating Peptides

Endocytic pathway, as discussed above, is generally accepted as the major

internalization route of CPPs and CPP-conjugates especially at low CPP concentrations.

No matter which endocytic pathways are employed for the uptake, the internalized CPPs

and CPP-conjugates would find themselves in vescicles known as endosomes that mature

from early endosomes to late endosomes before fuse into lysosome where active

enzymatic degradation processes take place. Therefore, CPP-conjugated cargos, which

cannot escape before the fusion of the endosome with the lysosome, will not be able to

reach the cytosol or nucleus and will exert minimal biological activities.

Many studies demonstrated CPP-cargos’ ability to escape from endosomal

compartments based on the functionality of the biomolecular cargo. For example, Tat and

recombinase Cre conjugate (Tat-Cre) is able to activate the expression of a luciferase or

GFP.83 Several mechanisms, similar to the ones that explain direct translocation, have been proposed to explain the ability of CPPs to cross endosomal membranes. The largest 20

notable differences for the proposed mechanisms to explain endosomal escape are the

emphasis on the pH gradient143-145 across the endosomal membrane and a requireminet of

a minimal local concentration146 for the CPPs to rupture endosomal membranes.

Nonetheless, it is generally accepted that efficiency of CPPs’ endosomal escape is often

poor without endosomal escape devices such as fusogenic agent, membrane-disruptive

agents, lysosomotropic agnets and others.147-149 In the case of Tat-Cre and many others,

only a few copies of cargo molecules are needed to exert a biological response. However,

when many copies of cargo molecules were required for activity, CPPs, in many

examples, failed to deliver detectable responses.150-152 Microscopy-based methods have also revealed the punctate intracellular distribution of fluorescent CPP-conjugates, suggesting endosomal entrapment and low endosomal escape efficiency.153 It is generally agreed that endosomal escape is a major bottleneck for efficient CPP-based cellular delivery. Although many chemical and biological agents have been developed to improve endosomal escape,155,156 better physical descriptions of endosomal escape process and

development of CPPs with efficient endosomal esacpe are still open challenges for future

research.

Chapter 2: Discovery and Mechanism of Cyclic Cell Penetrating Peptides

2.1 Introduction

Cyclic peptides are a privileged and yet underexploited class of compound for drug

discovery.157 Compared to their linear counterparts, cyclic peptides have reduced

conformational freedom, which makes them more resistant to proteolysis and allows

them to bind to their molecular targets with higher affinities and specificity.158-160 Cyclic peptides are widely produced in nature and exhibit a broad range of biological activites.

In addition, our group and others have developed technologis to rapidly synthesize and

screen large libraries of cyclic peptides aginast molecular targets or live cells.161-165 A major limitation to the broader application of cyclic peptides as therapeutic agents or research tools is their poor membrane permeability. Most of the clinically used cyclic peptides act on extracellular targets. It is generally believed that cyclic peptides have poor membrane permeability becauase the peptide backbone interacts strongly with water molecules through formation of hydrogen bonds. Desolvation of the peptide bound waters, which is necessary for the peptides to traverse the hydrophobic region of a lipid bilayer, creates a large energy barrier. Some cyclic peptides can lower the desolvation energy by forming intramolecular hydrogen bonds (e.g., a β-turn structure) and are thus more membrane permeable.166 Nα-methylation of the peptide backbone, which elimates

its hydrogen bonding donor capability and increases the overall hydrophobicity of

peptides, also improves their membrane permeability.62,167 Cyclosporine A, a rare

example of cyclic peptide drugs that inhibit intracellular targets (i.e., calcineurin),168 contains Nα-methylation at seven out of its 11 peptide bonds. However, most cyclic

peptides do not form stable intramolecular hydrogen bonds, whereas Nα-methylation may

interfere with target binding and generates a mixture of stereoisomers containing cis/trans

peptide bonds, some of which may have undesired biological activites.

Limited studies to compare the membrane permeability of linear versus cyclic

peptides have led to conflicting conclusions.171-173 While incorporation of certain

unnatural amino acids169 and other chemical moieties37,170 have been investigated heavily

to develop impressive transporters with increased stability and delivery efficiency, little is

known about the effect of cyclization upon cell-penetrating activity. The integration of cyclization and cell-penetrating elements of canonical arginine-rich CPPs [e.g., Tat and nonaarginine (R9)] has been inadequately explored. Recently, we discovered that some

cyclic peptides containing as few as three Arg residues were internalized by mammalian

cells.174 This finding prompted us to further investigate the factors that influence the

membrane permeability of cyclic peptides, with the goal of identifying a group of cyclic

cell penetrating peptides. Herein we report that cyclization greatly enhances the

membrane permeability of short sequence motifs (5-6 amino acids) rich in arginine and

hydrophobic residues [e.g., cyclo(FΦRRRR) (cFΦR4), where Φ is L-2-naphthylalanine].

The resulting cyclic CPP translocates into the cytoplasm and nucleus of cultured human

cells at significantly higher efficiency than canonical arginine-rich linear CPP

demonstrated by live-cell confocal microscopic imaging and a newly developed peptide

internalization assay. To gain insight into the mechanism of action, we investigated the

internalization mechanism of cFΦR4 through the use of model membrane vesicles and pharmacologic agents as well as genetic mutations that perturb various internalization

events. Our data show that cFΦR4 initially binds directly to the plasma membrane

phospholipids and enters cultured cells through multiple endocytic pathways. Unlike

canonical CPPs, which cannot escape endosomal compartments efficiently, cFΦR4 escapes from the early endosomes into the cytosol. In addition, cFΦR4 exhibits superior

stability against proteolysis over linear CPPs but minimal cytotoxicity. cFΦR4 therefore provides an excellent transporter for cytosolic cargo delivery as well as a suitable system for investigating the mechanism of early endosomal escape.

FITC FITC Rho Dex Figure 3. Structures of cFΦR4 , FΦR4 , cFΦR4 and cFΦR4 .

2.2 Results

2.2.1 Peptide Cyclization and Hydrophobicity Synergistically Enhance Cellular

Association

Our previously reported cell permeable cyclic peptides contained 3-5 Arg residues

and at least one aromatic hydrophobic residue [e.g., L-2-naphthylalanine (Nal or Φ)].174

Other investigators have also noted the ability of hydrophobic residues to enhance the membrane transduction activity of CPPs.175,176 To assess the effect of peptide cyclization

and hydrophobicity on cellular uptake, we synthesized fluorescein isothiocyanate (FITC)-

FITC FITC labeled linear and cyclic forms of tetraarginine (R4 ), hexaarginine (R6 ), and

FITC FITC FFRRRR (F2R4 ), along with linear nonaarginine (R9 ) and cyclo(AAAARRRRQ)

FITC (cA4R4 ) as controls (Table 2, peptides 1-10). All of the cyclic peptides contained a C-

terminal Glu residue for peptide cyclization. A Lys residue was added to the Glu side

FITC chain and the Lys side chain was labeled with FITC (e.g. cFΦR4 ; Table 2, peptide 11;

Figure 3). We initially employed the method of Holm et al.177 to assess the amount of

cellular association of these peptides. Each peptide (5 μM) was incubated with MCF-7 human breast cancer cells for 1 h and the cells were washed with buffer and briefly treated with trypsin to remove any peptides bound to the cell surface. The cells were then lysed with NaOH and the amount of fluorescence associated with the cells (which corresponds to the sum of both internalized and any surface-bound peptides) was

FITC quantitated. As expected, R9 efficiently translocated into the cells and the amount of

9,33 FITC internalized peptide was similar to the previously reported values. R4 had

FITC FITC essentially no cellular association, whereas R6 was 17-fold less active than R9

FITC (Figure 4). Replacement of two N-terminal Arg residues of R6 (peptide 5) with Phe

FITC (F2R4 or peptide 7) increased the activity by 2.5-fold (Figure 4). Surprisingly, while

FITC FITC cyclization of peptides R4 and R6 had no measurable effect (Figure 4, compare

FITC peptides 2 vs 3 or 4 vs 5), cyclization of F2R4 increased its cellular association by 14-

FITC fold (compare peptides 6 vs 7). The resulting peptide cyclo(FFRRRRQ) (cF2R4 ,

FITC peptide 6) was ~2-fold more active than R9 . To ascertain that the increased cellular association was not simply due to the higher proteolytic stability of cyclic peptides, we

FITC compared the activities of linear (r6 , peptide 9) and cyclic D-hexaarginine peptides

FITC FITC FITC (cr6 , peptide 8). Although r6 had ~3-fold higher activity than R6 , cyclization of

FITC r6 did not further increase its cellular association. Moreover, replacement of two Arg

FITC residues of cyclic R6 (cR6 ) with four Ala residues did not increase the amount of

peptides associated with the cells (Figure 4, compare peptides 4 and 10). These results

suggest that cyclization and hydrophobicity act synergistically to enhance the ability of

Arg-rich CPPs to associate with and/or translocate cross the cell membrane. To test whether the cellular association efficiency could be further improved by increasing hydrophobicity, we replaced one of the Phe residues in peptide 6 with 2-naphthylalanine

FITC to produce peptide 11, cyclo(FΦRRRRQ) (cFΦR4 ). Indeed, peptide 11 bound

to/internalized MCF-7 cells 23-fold more efficiently than the corresponding linear

FITC FITC peptide (FΦR4 or peptide 12) and was 13-fold more active than R9 (Figure 4).

Similar cellular association results have been obtained when conducted with HEK293

cells.178

Table 2. Sequences and Cellular Association of Linear and Cyclic Peptides

Peptide Abbreviationa Peptide Sequenceb Cellular Association (%)c No. FITC 1 R9 Ac-RRRRRRRRR-(Q-OAll)-K(FITC)-NH2 100 FITC 2 cR4 cyclo(RRRRQ)-K(FITC)-NH2 2.5 ± 0.5 FITC 3 R4 Ac-RRRR-(Q-OAll)-K(FITC)-NH2 2.3 ± 0.8 FITC 4 cR6 cyclo(RRRRRRQ)- K(FITC)-NH2 5.8 ± 1.1 FITC 5 R6 Ac-RRRRRR-(Q-OAll)-K(FITC)-NH2 5.8 ± 1.0 FITC 6 cF2R4 cyclo(FFRRRRQ)-K(FITC)-NH2 190 ± 20 FITC 7 F2R4 Ac-FFRRRR-(Q-OAll)-K(FITC)-NH2 14 ± 1 FITC 8 cr6 cyclo(rrrrrrQ)-K(FITC)-NH2 21 ± 3 FITC 9 r6 Ac-rrrrrr-(Q-OAll)-K(FITC)-NH2 21 ± 2 FITC 10 cA4R4 cyclo(AAAARRRRQ)-K(FITC)-NH2 3.2 ± 0.2

FITC 11 cFΦR4 cyclo(FΦRRRRQ)-K(FITC)-NH2 1260 ± 140

FITC 12 FΦR4 Ac-FΦRRRR-(Q-OAll)-K(FITC)-NH2 55 ± 10 aFor cellular association assays, all cyclic peptides contained a K(FITC) attached to the side chain of the invariant Gln (Q), and all linear peptides carried a main-chain allyl group protected Gln (Q-OAll) and then b c a K(FITC). r, D-arginine; Φ, L-2-naphthylalanine. Cellular association efficiencies were relative to R9 (100%).

Figure 4. Comparison of the cellular association efficiencies of FITC-labeled linear and cyclic peptides.

2.2.2 Efficiency and Kinetics of Peptide Internalization

Because the method of Holm et al. does not distinguish internalized versus cell

surface-bound CPPs, we modified the method of Wender et al.179 to determine the

amount of peptides internalized by the cells. In this method, a “caged” fluorophore 4-

methylumbelliferone (4-MU) was attached to cyclic peptide cFΦR4 and linear control peptides (FΦR4 and R9) via a disulfide linkage (Figure 5A compound 1). Compound 1 is

non-fluorescent and stable in the oxidizing extracellular environment. Upon entering the

cell, however, the disulfide bond would be cleaved by the intracellular thiols such as

glutathione (GSH) to produce peptide 2 and thiol 3. Under the physiological pH, thiol 3

would undergo spontaneous decomposition to release 4-MU as a fluorescent product. To

conjugate cell penetrating peptides with 4-MU, free cysteine containing peptides obtained from standard solid-phase peptide synthesis were linked with labeling reagent 5 through efficient disulfide exchange reaction (Figure 5B). The synthetic details to obtain CPP-4-

MU conjugates and caged 4-MU reagent 5 are described in the experimental section. To verify the rapid response of conjugate 1 to reducing environment (1 mM dithiothreitol in phosphate buffer saline, pH 7.4) resulted in the complete release of 4-MU in less than 10 min (Figure 5C).

As expected, incubation of MCF-7 cells at 37 ºC in the presence of the three peptides all resulted in time-dependent increases in fluorescence intensity (Figure 5D).

However, the three peptides were internalized with different efficiency and kinetics.

Peptides R9 and cFΦR4 showed similar internalization rates during the first 45 min; after

that the internalization of R9 slowed down before reaching a plateau value of 1500 28

fluorescence units, whereas the translocation of the cyclic peptide continued for several

hours. Linear FΦR4 showed both lower initial rate and slightly lower final intracellular

concentration than R9 (~80%). As a control, incubation of the peptides in pH 7.4

phosphate buffer saline (PBS) produced only slight fluorescence increases, due to background hydrolysis of the carbonate ester of compound 1. When MCF-7 cells were

incubated with these peptides at 4 ºC, the rates of fluorescence increase were similar to

the background rate (PBS), indicating the absence of significant peptide internalization

(Figure 5D). These results show that cFΦR4 is translocated into mammalian cells ~2-fold

more efficiently than R9, one of the most potent CPPs reported to date. The lack of

significant translocation at 4 °C suggests that energy-dependent endocytosis may be the

primary mechanism of their uptake. Further, the amount of 4-MU released should largely

represent the amount of CPPs in the cytoplasm and nucleus. The endosomal and lysosomal environments are more oxidizing than that of the cytoplasm, disfavoring the reduction of disulfides in these organelles.180 There are many examples of unreduced

disulfides in the endosome,181-184 although enzymatic reduction of disulfides in the

endosomes of antigen-presenting cells has also been reported.185 Background hydrolysis

of the carbonate ester linkage of 1 would also lead to 4-MU release; however, the

hydrolysis rate decreases with pH and should be minimal in the acidic environments of

endosomes (pH 5-7) and lysosomes (pH ~4.8) (Figure 5E).178

Figure 5. 4-MU based CPP internalization assay. (A) Scheme showing the reduction of the disulfide bond of compound 1 and the release of fluorescent product 4. GSH, glutathione. (B) Scheme showing the synthesis of compound 1 with cysteine containing peptides and caged 4-MU labeling reagent 5. (C) Release of 4-MU from compound 1 by dithiothreitol. (D) Time-dependent internalization of indicated peptides (5 µM) by MCF-7 cells at 37 or 4 ºC. (E) Background hydrolysis of the compound 1 at different pH relative to that of dithiothreitol treatment (100%).

2.2.3 Subcellular Distribution of Internalized Peptides

To obtain further evidence that the cyclic peptides were internalized by mammalian

FITC cells, we carried out live-cell confocal microscopy with peptides 1, 11, and 12 (R9 ,

FITC FITC FΦR4 , and cFΦR4 , respectively). MCF-7 cells were treated with the FITC-labeled peptides (5 μM) for 2 h in the presence of 2% fetal bovine serum (FBS) and immediately

FITC imaged by confocal microscopy without fixation. Cells treated with peptides R9 and

FITC cFΦR4 showed strong green fluorescence, whereas cells incubated with linear

FITC FΦR4 had little intracellular fluorescence under the same imaging conditions (Figure

6). A small amount of fluorescence was detectable at the cell surface when cells were

incubated with the linear peptide for 1 h or the incubation was carried out in the absence

of serum. Presumably, peptide FΦR4 underwent proteolytic degradation during the 2-h

incubation period. The mammalian serum contains high levels of chymotrypsin

activity186 and can rapidly degrade linear peptides containing aromatic residues but not

the corresponding cyclic peptides.160 There were notable differences between peptides

FITC FITC FITC R9 and cFΦR4 . While peptide R9 generated primarily punctate fluorescence,

which is in agreement with numerous earlier reports and indicative of endosomal

4 FITC entrapment of the internalized peptide, cFΦR4 resulted in more diffused signals,

consistent with substantial cytoplasmic and nuclear distribution of the latter peptide.

Figure 6. Live-cell confocal images of MCF-7 cells treated with FITC-labeled CPPs.

FITC (A) Treatment with cFΦR4 I, DIC; II, cFΦR4 distribution in cells. (B) Treatment with

FITC FITC FΦR4 I, DIC; II, FΦR4 distribution in cells. (C) Treatment with R9 I, DIC; II, R9 distribution in cells.

To determine the subcellular distribution of the internalized cyclic peptides, A549

FITC cells (a lung cancer cell line) were incubated simultaneously with cFΦR4 (5 μM) and rhodamine B-labeled dextran (an endocytosis marker, dextranRho), and then the cell

permeable nuclear stain DRAQ5 and examined by live-cell confocal microscopy. The

FITC internalized cyclic peptide cFΦR4 was distributed throughout the cell and accumulated to a greater extent within nucleoli, whereas dextranRho remained largely in the cytoplasm as punctate fluorescence (Figure 7). Within the cytoplasmic region,

FITC cFΦR4 and the endocytosis marker showed overlapping and punctate fluorescence, 32

suggesting that a fraction of the internalized cyclic peptides were still in the endosomes.

All of the cells did not have the same peptide uptake efficiency; ~50% of the cell population exhibited much stronger fluorescence than the rest of the cells (Figure 7).

Importantly, the same cell population also internalized the endocytosis marker more efficiently. When the experiment was carried at 4 °C or after the cells had been pretreated for 1 h with sodium azide (which depletes cells of ATP), the cells failed to take up either the cyclic peptide or the endocytosis marker Figure 7B and 7C). These results strongly suggest that the cyclic peptides entered the cells through energy-dependent internalization mechanism, although minor contributions by other mechanisms cannot be ruled out. The accumulation of fluorescence in the nucleus indicates that the internalized peptides were able to escape from the endosomes. The more diffused fluorescence

FITC FITC throughout cFΦR4 treated cells and the greater accumulation of cFΦR4 in the

FITC FITC nucleoli as compared to R9 also suggest that cFΦR4 is more efficient in endosomal

FITC escape than R9 , in agreement with the 4-MU assay data (Figure 5).

FITC Figure 7. Live-cell confocal images of A549 cells treated with cFΦR4 .

FITC (A) Treatment with cFΦR4 I, nuclear stain with DRAQ5; II, red fluorescence of

Rho FITC dextran ; III, green fluorescence of cFΦR4 ; IV, a merge of I-III. (B) Same as (A)

but in the presence of NaN3. (C) Same as (A) but at 4 ºC.

2.2.4 cFΦR4 Enters Cells via Endocytosis

The failure of cell entry by cFΦR4 at 4 °C or in the presence of sodium azide suggests that energy-dependent endocytic processes mediate its cell entry. However, other cyclic CPPs have been reported to enter cells by direct membrane translocation.175,176 To further test the role of endocytosis and intracellular trafficking

during the cellular uptake of cFΦR4, we employed rhodamine B labeled cyclic peptide

Rho cFΦR4 to further explore the intracellular distribution. Rhodamine B was chosen

because of its desirable fluorescence properties, such as resistance to photobleaching and

insensitivity to pH changes.187 We examined the extent of colocalization between

Rho cFΦR4 (Table 3, compound 1; Figure 3) and GFP-tagged Rab proteins that have been

demonstrated to localize to vesicles of the endocytic system. Rab5 is a small Rab family

GTPase that primarily localizes to early endosomal membranes, whereas Rab7

predominantly localizes to the membranes of late endosomes.121,188 HeLa cells

Rho overexpressing GFP-Rab5 or GFP-Rab7 were incubated with 1 μM cFΦR4 for 30 min

and examined by spinning-disk confocal microscopy (Figure 8A). A large fraction of

rhodamine B fluorescence overlapped with that of Rab5-GFP (RGFP, rhodamine = 0.703) and

Rho Rab7-GFP (RGFP, rhodamine = 0.591), indicating that cFΦR4 is present in both early

(Rab5+) and late (Rab7+) endosomes.

Table 3. Sequences of CPPs Studied in Chapter 2

Peptide Abbreviation Peptide Sequencea No. Rho 1 cFΦR4 cyclo(FΦRRRRQ)-K(Rho)-NH2

Dex 2 cFΦR4 cyclo(FΦRRRRQ)-K(Dex)-NH2 Dex 3 Tat H-K(Dex)-GRKKRRQRRRPPQY-NH2 FITC 4 Tat Ac-YGRKKRRQRRR-K(FITC)-NH2

5 cFΦR4 cyclo(FΦRRRRQ)

6 R9 H-RRRRRRRRR-OH 7 Tat H-YGRKKRRQRRR-OH 8 Antp H-RQIKIWFQNRRMKWKK-OH aΦ, L-2-naphthylalanine; Rho, rhodamine B; Dex, dexamethasone; FITC, fluorescein isothiocyanate.

To further delineate the endocytic events required for the cytosolic delivery of

189-191 cFΦR4, we made use of a novel GR-GFP translocation assay and examined the

effects of various endocytic inhibitors on the cytosolic delivery of Dex-labeled peptides

Dex Dex cFΦR4 and Tat (Figure 3; Table 3, compounds 2 and 3). In resting cells, GR

(glucocorticoid receptor) is maintained in the cytoplasm through interactions with host chaperones.192 Upon cytoplasmic delivery of Dex or Dex-peptide conjugates, association

with the GR activates translocation from the cytoplasm into the nucleus.193 Thus the ratio

of GR-GFP fluorescence intensity in the nucleus to that in the cytoplasm, or the

translocation ratio (TR), provides a semi-quantitation of the cytosolic Dex or Dex-peptide

concentration. It should be noted that this assay is ideal for comparing the cytosolic

concentrations of a Dex derivative under different conditions (e.g., in the absence vs

Dex presence of an endocytosis inhibitor); for different Dex derivatives (e.g., cFΦR4 and

TatDex), the TR may also be affected by their solubility, metabolic stability, and/or

binding affinity to the GR.189,190 Incubation of HeLa cells expressing a GR-GFP fusion

Dex protein with 1 μM cFΦR4 increased the TR from 1.17 ± 0.23 in untreated cells to 3.50

Dex ± 0.66, confirming that a significant amount of cFΦR4 reached the cytoplasm.

Dex Treatment of HeLa cells before and during cFΦR4 incubation with the cell-permeable

dynamin inhibitor Dynasore (Dyna), the cortical actin remodeling inhibitor N-ethyl-

isopropyl amiloride (EIPA), or the cholesterol-sequestering agent methyl-β-cyclodextrin

Dex (MBCD) decreased the cFΦR4 TR to 2.35 ± 0.75, 1.86 ± 0.46, and 1.56 ± 0.39,

respectively (Figure 8B).194-196 These results strongly support our previous hypothesis

that cFΦR4 enters cells predominantly through endocytosis, and the inhibition pattern

suggests the involvement of multiple endocytic mechanisms during the uptake of cFΦR4.

+ + Figure 8. cFΦR4 enters cells through endocytosis, localizes to Rab5 and Rab7 endosomes, and releases from early endosomes into cytoplasm.

Rho (A) Colocalization between Rab5-GFP (left) or Rab7-GFP (right) with cFΦR4 (red) HeLa cells stained with Hoescht 33342 (blue). (B) Translocation of GR-GFP after

Dex treatment with 1 μM Dex or cFΦR4 in the presence and absence of 80 μM dynasore (Dyna), 50 μM N-ethyl-isopropyl-amiloride (EIPA), 5 mM methyl-β-cyclodextrin (MBCD), 200 nM wortmannin (Wort), or 200 nM bafilomycin (Baf). (C) Translocation

Dex of GR-GFP after treatment with 1 μM Dex or cFΦR4 upon overexpression of WT Rab5 or Rab5(Q79L). * p ≤ 0.01; ** p ≤ 0.001; ns, not significant; two-tailed t-test.

2.2.5 cFΦR4 Escapes from Early Endosomes into the Cytoplasm

The presence of intense, diffuse fluorescence throughout the cytoplasm and nucleus

of cells treated with fluorescently labeled cFΦR4 indicates that cFΦR4 efficiently escapes

from the endosome (ref. 189 and vide infra). In an attempt to understand why cFΦR4

possesses this unusual property among CPPs, we again made use of the GR-GFP

translocation assay to examine the effect of downstream endocytic perturbations on the

Dex cytoplasmic delivery of cFΦR4 . Pretreatment of HeLa cells with the endosomal

197 Dex vesicular ATPase inhibitor bafilomycin (Baf) prior to the addition of cFΦR4

decreased the TR to 2.66 ± 0.62, suggesting that endosomal acidification facilitates

Dex + cFΦR4 release into the cytoplasm. Blocking the maturation of Rab5 vesicles by

pretreating cells with the phosphatidylinositol 3-kinase inhibitor wortmannin (Wort)198 had only minor effect on the TR (from 3.50 ± 0.66 to 3.15 ± 0.55), supporting that

Dex + cFΦR4 releases from early Rab5 endosomes. To further test whether endosomal

maturation is required for cytosolic delivery, we overexpressed GTPase-inactive Rab5

mutant, Rab5(Q79L), which halts endosomal maturation at the Rab5+ stage.199

Rab5(Q79L) overexpression significantly reduced the TR for TatDex, which has

previously been shown to release from late endosomes,189 but had no effect on either free

Dex Dex or cFΦR4 (Figure 8C), confirming that cFΦR4 releases from early endosomes into

the cytoplasm. Interestingly, miniature proteins containing a penta-arginine motif on an

α-helix, another system shown to efficiently escape from the endosome, are also released

189 from the early endosome. R9 was previously shown to exit the endocytic pathway at a

point later than the miniature proteins but prior to Tat.189 It appears that compared to

other cationic CPPs, cFΦR4 is less dependent on endosomal acidification for release and

thus able to exit from the less acidic early endosomes.

2.2.6 Cyclic Peptides Bind to Membrane Phospholipids but Do Not Undergo Direct

Translocation

A potential explanation for the apparently conflicting results on the cellular uptake

efficiency of cFΦR4 (relative to R9) as measured by the different methods is that a

significant fraction of the cyclic CPPs bound tightly to the plasma membrane but did not

enter the cells. To test this hypothesis and whether the cyclic CPPs can translocate across

FITC FITC membranes in an energy-independent manner, we incubated cFΦR4 , FΦR4 , and

FITC R9 with multi-lamellar vesicles (MLVs) containing 90% phosphatidylcholine (PC)

and 10% phosphatidylglycerol (PG) (w/w) and monitored peptide internalization by

confocal microscopy. Under our assay conditions, no fluorescence could be detected

FITC FITC inside the vesicles treated with R9 , FΦR4 , or the free dyes [FITC-labeled dextran

(dextranFITC) and tetramethylrhodamine (TAMRA)] (Figure 9). Interestingly, the green

FITC fluorescence of cFΦR4 was immediately and almost completely quenched as soon as

the peptide and vesicle solutions were mixed. In contrast, the red fluorescence of

TAMRA added to the same solution was not affected by the lipids and showed that the vesicles were intact. The same results were obtained with MLVs of more physiologically relevant mammalian cell membrane compositions [45% PC, 20% phosphatidylethanolamine (PE), 20% phosphatidylserine (PS), and 15% cholesterol

FITC (w/w)]. The simplest explanation is that most of the cFΦR4 was bound to the vesicle surface and the resulting high local concentration in the membrane caused internal quenching of the fluorescence. These results suggest that cFΦR4 binds directly to the

phospholipids with high affinity but cannot cross the membrane in an energy-independent

manner.

Figure 9. Confocal images of MLVs after incubation with 1 μM free TAMRA dye (red)

FITC FITC FITC FITC and 1 μM dextran (control), R9 , FΦR4 , or cFΦR4 for 30 min. Top panel, fluorescein channel; bottom panel, TAMRA channel.

To verify the membrane binding capabilities of various CPPs, we prepared small

unilamellar vesicles (SUV) containing neutral phospholipids, which mimic the plasma

membrane of mammalian cells [45% PC, 20% PE, 20% sphingomyelin, and 15%

cholesterol (w/w)] and tested them for binding to FITC-labeled cFΦR4, R9, and Tat

FITC (Table 2) by a fluorescence polarization (FP) assay. cFΦR4 bound to the neutral

FITC SUVs with an EC50 value (lipid concentration at which half of cFΦR4 is bound) of 2.1

FITC ± 0.1 mM (Figure 10A). R9 showed much weaker binding to the artificial membrane

FITC (EC50 >10 mM), whereas Tat did not bind at all. We next tested the CPPs for binding

to heparan sulfate, which was previously proposed to be the primary binding target of

32,120,200-203 FITC FITC arginine-rich CPPs. As expected, R9 and Tat both bound to heparan

sulfate with high affinity, having EC50 values of 144 and 304 nM, respectively (Figure

FITC 10B). Under the same condition, cFΦR4 showed no detectable binding to heparan sulfate. Taken together, these observations explain why the cell surface-bound cyclic

CPPs were not observed by confocal microscopy. In contrast, the membrane-bound CPPs are readily detected by the method of Holm et al., because cell lysis with NaOH disrupts the cell membranes, releasing the bound CPPs into solution. Linear FΦR4 and R9 interact

with the vesicle/cell membranes with much lower affinity than that of cFΦR4; thus these linear peptides would be more easily removed from cell during trypsin treatment.

Figure 10. Binding of cFΦR4, R9, and Tat to SUV (A) and heparan sulfate (B). 41

Additionally, our results are in agreement with the previous observations that non- amphipathic cationic CPPs (e.g. Tat and R9) bind tightly with cell surface proteoglycans

(e.g. heparan sulfate) but only weakly with membrane lipids.32 The insufficient number

of positive charges of cFΦR4 is likely responsible for its lack of strong electrostatic

interaction with heparan sulfate. On the other hand, the amphipathic nature and the more

rigid cyclic structure of cFΦR4 should facilitate its binding to neutral lipid membranes.

These data, together with the inhibition pattern by various endocytic inhibitors described above, strongly suggest that cFΦR4 binds directly to the plasma membrane phospholipids

and is internalized by all of the endocytic mechanisms in a piggyback manner.

2.2.7 Stability and Cytotoxicity of cFΦR4

The relative stability of cFΦR4, R9, Tat, and Antp (Table 3, compounds 5-8) against

proteolytic degradation was determined by incubating the CPPs in 25% human serum at

37 °C and following the disappearance of the full-length peptides by reversed-phase

HPLC. The cationic tryptophan-containing peptide, Antp, was least stable among the four

CPPs; it was degraded at a half-life of <20 min and was completely digested after 2 h

(Figure 11A). R9 and Tat were slightly more stable than Antp, having half-lives of ~30 min. In contrast, cFΦR4 was remarkably stable against serum proteases. There was less than 10% degradation after 6 h of incubation; after 24 h of incubation in the serum, >70%

of cFΦR4 remained intact. Numerous other studies have also demonstrated that

cyclization of peptides greatly increases their proteolytic stabilities.160 The potential

cytotoxicity of cFΦR4 was assessed by MTT assays with five different human cell lines 42

(HEK293, MCF-7, A549, H1650, and H1299). After 24 or 48 h of incubation with up to

50 μM cFΦR4, there was no significant growth inhibition for any of the cell lines (Figure

11B and Figure 11C). After 72 h, a slight growth inhibition (up to 20%) was observed only at 50 μM (Figure 11D). Thus, cFΦR4 is relatively nontoxic to mammalian cells.

Figure 11. Serum stability and cytotoxicity studies of cyclic CPP.

(A) Comparison of the serum stability of cFΦR4, Tat, R9, and Antp. (B) Cytotoxicity of

cFΦR4. The indicated cell lines were treated with DMSO (control), 5 μM, or 50 μM cFΦR4 for 24 h and the percentage of live cells was determined by MTT assay. (C) Same as (B) but for 48 h. (D) Same as (B) but for 72 h.

2.3 Discussion

This work, along with several other recent reports,175,176,204 explores the properties of cyclic cell-penetrating peptides. Our results reveal that cyclic peptides containing a 43

proper combination of Arg and hydrophobic residues bind directly to the membrane

phospholipids with high affinity indicated by SUV binding assays. Presumably,

interactions between the Arg side chains and the phosphate groups of phospholipids

anchor the CPPs on the plasma membrane surface, while the hydrophobic side chains of

Phe and Nal residues insert into the hydrophobic region of the membrane and stabilize the peptide-membrane complex by hydrophobic effect. Membrane insertion also helps forming positive membrane curvature, a component of negative Gaussian curvature, which features negative and positive membrane curvatures in two orthogonal dimensions.124 All major cellular uptake mechanisms (including endocytosis) involved the formation of negative Gaussian curvatures. According to this model, addition of cyclic CPPs to cells would result in the rapid accumulation of the CPPs on the plasma membrane and the efficiency of peptide internalization would depend on the endocytosis rate. Since the rate of endocytosis varies greatly depending on factors such as the cell growth condition (e.g., lower pinocytosis rate at high confluency)205 and the stage in the cell cycle (e.g., higher pinocytosis rate at interphase than metaphase),206 this may explain

the varying CPP uptake efficiencies of different cells observed by confocal microscopy.

As suggested by the MLV translocation assay, the high peptide concentration within the membrane likely caused fluorescence internal quenching, rendering the membrane-bound peptides undetectable by confocal microscopy. These peptides, which are presumably on the outer leaflet of the membrane, would also evade detection by the 4-MU assay.

However, the membrane-bound peptides can be detected by the conventional uptake assay,177 which measures the fluorescence yield after cell lysis. It is yet unclear whether

the membrane-bound peptides are limited to the cytoplasmic membrane, or also present

on internal membranes (e.g., endosomes).

Following the initial membrane binding behavior, several lines of evidence indicate

that cFΦR4 enters cells through multiple endocytic mechanisms, including its failure to

enter cells at 4 °C or in the presence of sodium azide, partial overlap between the

Rho FITC fluorescence puncta of cFΦR4 and the fluid phase endocytic marker dextran ,

Rho colocalization of cFΦR4 and endosomal proteins Rab5 and Rab7, and decreased

Dex cFΦR4 uptake upon administration of endocytic inhibitors. The minimal effect of the

PI3K inhibitor wortmannin and the Rab5 Q79L mutation on the cytoplasmic delivery of

+ cFΦR4, in addition to the strong colocalization observed between cFΦR4 and Rab5 endosomes, suggest that cFΦR4 escapes from early endosomes (Figure 12). In

comparison, Tat has been demonstrated to enter cells through endocytosis and release

189 from late endosomes, while R9 escapes endosomes prior to Rab7 recruitment. Early

endosomal release offers significant advantages, especially for peptide and protein cargos,

since it minimizes cargo degradation by late endosomal and lysosomal proteases and

denaturation caused by acidification during endosomal maturation.

Figure 12. Scheme showing the points along the endocytic pathway where cFΦR4, R9, and Tat escape into the cytoplasm.

Recent studies suggested that the topology of Arg residues can influence the stage at

which a cationic peptide/protein departs the vesicular trafficking pathway to gain

cytosolic access.189 Cyclization of a polyarginine peptide undoubtedly affects the

topology of its Arg residues. Future work will be necessary to examine this hypothesis

and elucidate the molecular mechanism of cyclic CPP uptake and endosomal escape.

Additional advantages of cyclic CPPs include their improved metabolic stability and low

cytotoxicity. Since this study only examined a limited number of peptide sequences, it

remains to be determined whether additional peptide motifs can be found as efficient

membrane transporters.

2.4 Conclusion

We report the discovery, characterization, and mechanism studies of a class of cyclic

cell penetrating peptides. Short cyclic peptide motifs rich in arginine and hydrophobic

residues [e.g., cyclo(FΦRRRRQ), cFΦR4] bound to the plasma membrane of mammalian

cultured cells and were subsequently internalized through endocytic pathways into the

cells. Confocal microscopy and a newly developed peptide internalization assay

demonstrated that cyclic peptides containing these transporter motifs were translocated

into the cytoplasm and nucleus at efficiencies 2-5-fold higher than that of nonaarginine

(R9). Furthermore, this novel cyclic cell penetrating peptide unlike canonical arginine-

rich CPPs was capable of escaping from early endosome. This is supported by confocal microscopic colocalizatoin assay and GR-GFP translocation assay. Coupled with cyclic

CPP’s excellent metabolic stability and minimal cytotoxicity, cFΦR4 was proved to be an excellent transporter and a suitable system for investigating the mechanism of endosomal escape.

2.5 Experimental Section

2.5.1 Materials

Reagents for peptide synthesis were purchased from Advanced ChemTech

(Louisville, KY), NovaBiochem (La Jolla, CA), or Anaspec (San Jose, CA). Rink resin

LS (100-200 mesh, 0.2 mmol/g) was purchased from Advanced ChemTech. Fluorescein

isothiocyanate (FITC), dexamethasone (Dex), and rhodamine B-dextran were purchased 47

from Sigma-Aldrich (St. Louis, MO). Cell culture media, fetal bovine serum, penicillin-

streptomycin, Dulbecco’s phosphate-buffered saline (DPBS) (2.67 mM potassium

chloride, 1.47 mM potassium phosphate monobasic, 137 mM sodium chloride, 8.06 mM

sodium phosphate dibasic.), Lipofectamine 2000 and 0.25% trypsin-EDTA were purchased from Invitrogen (Carlsbad, CA). Nuclear staining dye DRAQ5TM was purchased from Thermo Scientific (Rockford, IL), while cell proliferation kit (MTT) was purchased from Roche (Indianapolis, IN). Anti-phosphotyrosine (pY) antibody (clone

4G10) was purchased from Millipore (Temecula, CA). 1-palmitoyl-2-oleoyl-sn-glycero-

3-phosphocholine (PC), 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho(1’-rac-glycerol)

(sodium salt) (PG), 1-palmitoyl-2-oleoyl-sn-glycero-3-phophoethanolamine (PE), sphingomyelin (Brain, Porcine), and cholesterol were purchased from Avanti Polar

Lipids (Alabaster, AL). Heparan sulfate (HO-03103, Lot#HO-10697) was from Celcus

Laboratories (Cincinnati, OH).

2.5.2 Peptide Synthesis and Labeling

Peptides were synthesized on Rink Resin LS (0.2 mmol/g) using standard Fmoc chemistry. The typical coupling reaction contained 5 equiv of Fmoc-amino acid, 5 equiv of 2-(7-Aza-1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate

(HATU) and 10 equiv of diisoprpylethylamine (DIPEA) and was allowed to proceed with shaking for 75 min. After the addition of the last (N-terminal) residue, the allyl group on the C-terminal Glu residue was removed by treatment with tetrakis(triphenylphosphine)- palladium (0) [Pd(PPh3)4], triphenylphosphine (PPh3), formic acid and diethylamine (1,

10, 10, 10 equiv, respectively) in anhydrous tetrahydrofuran (THF) overnight. The N-

terminal Fmoc group was removed by treatment with 20% piperidine in DMF and the

peptide was cyclized by treatment with benzotriazole-1-yl-oxy-tris-pyrrolidino-

phosphonium hexafluorophosphate (PyBOP)/HOBt/DIPEA (5, 5, and 10 equiv) in DMF

for 3 h. The peptides were deprotected and released from the resin by treatment with

82.5:5:5:5:2.5 (v/v) TFA/thioanisole/water/phenol/ethanedithiol for 2 h. The peptides

were triturated with cold ethyl ether (3x) and purified by reversed-phase HPLC on a C18 column. The authenticity of each peptide was confirmed by MALDI-TOF mass spectrometry. FITC labeling was performed by dissolving the purified peptides (~1 mg each) in 300 µL of 1:1:1 DMSO/DMF/150 mM sodium bicarbonate buffer (pH 8.5) and incubating with 10 µL of FITC in DMSO (100 mg ml-1). After 20 min at room

temperature, the reaction mixture was subjected to reversed-phase HPLC on a C18 column to isolate the FITC-labeled peptide. To generate rhodamine- and Dex-labeled peptides (Figure 3), an Nε-4-methoxytrityl-L-lysine was added to the C-terminus. After

the solid phase peptide synthesis, the lysine side chain was selectively deprotected using

1% (v/v) trifluoroacetic acid in DCM. The resin was incubated with Lissamine

rhodamine B sulfonyl chloride/DIPEA (5 equiv each) in DMF overnight. The peptides

were fully deprotected, triturated with diethyl ether, and purified by HPLC. The Dex-

labeled peptide was produced by incubating the resin with dexamethasone-21-

thiopropionic acid/HBTU/DIPEA (5, 5, and 10 equiv) in DMF for 3 h.189 The peptide

was then deprotected, triturated, and purified by HPLC.

2.5.3 Cell Culture

HEK293, A549, HeLa, and MCF-7 cells were maintained in medium consisting of

DMEM, 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. HEK293 and

A549 cells were grown in DMEM supplemented with 10% FBS and 1%

penicillin/streptomycin. H1650 and H1299 cells were grown in RPMI-1640

supplemented with 10% FBS and 1% penicillin/streptomycin. Cells were cultured in a

humidified incubator at 37 °C with 5% CO2. For HeLa cells transfection, cells were

seeded onto 96-well plate at a density of 10,000 cells/well. Following attachment, cells

were transfected with plasmids encoding Rab5-green fluorescent protein fusion (Rab5-

GFP; a gift from Pietro Di Camilli), Rab7-GFP (Addgene plasmid #28047, Qing Zhong),

glucocorticoid receptor (C638G)-GFP fusion (GR-GFP),190 DsRed-Rab5 WT (Addgene

plasmid #13050, Richard Pagano) or DsRed-Rab5Q79L (Addgene plasmid #29688,

Edward DeRobertis) following Lipofectamine 2000 manufacturer protocols.

2.5.4 Quantification of Peptide Cellular Association

Approximately 1 x 104 MCF-7 cells were seeded in 12-well culture plates (BD

Falcon) in 1 ml of media and cultured for two days. On the day of experiment, 500 µL of

an FITC-labeled peptide solution (5 µM) in HKR buffer (5 mM HEPES, 137 mM NaCl,

2.68 mM KCl, 2.05 mM MgCl2•6H2O, 1.8 mM CaCl2•2H2O and 1 g/L glucose, pH 7.4) was added to the cells after aspirating the media and incubated at 37 °C in 5% CO2 for 1 h. The peptide solution was removed by aspiration, and the cells were gently washed with

HKR buffer (2 x 1 ml) and treated with 200 μL of 0.25% (w/v) trypsin-EDTA for 10 min.

After that, 1 mL of HKR buffer was added and the cells were transferred to a microcentrifuge tube and pelleted by centrifugation. The cell pellet was lysed in 300 µL of 0.1 M NaOH and the FITC fluorescence yield of the cell lysate was determined at 518 nm (with excitation at 494 nm) on a Molecular Devices Spectramax M5 plate reader. For each peptide, a standard line was generated by plotting the fluorescence intensity as a function of peptide concentration in 0.1 M NaOH and the amount of peptides taken up by cells (in pmol) was calculated by using the standard line. The amount of cellular proteins in each well was quantitated by a detergent compatible protein assay (Bio-Rad). Finally, the cellular association efficiency of the FITC-labeled peptide (in pmol of peptide internalized/mg of cellular protein) was calculated by dividing the amount of internalized peptide by the amount of protein in the lysate. The experiments were performed twice and in triplicates each time.

2.5.5 Synthesis of 4-MU Labeled Peptides and 4-MU Internalization Assay

To prepare the caged 4-MU molecule for peptide labeling (Figure 5B),178 a solution

of 4-S-(2’-thiopyridyl)mercaptobutanol (129 mg, 0.6 mmol) in 2 mL of DCM was added

dropwise to a 2 mL of cold DCM containing pyridine (48.3 µL, 0.6 mmol). The resulting

solution was added dropwise over 30 min to a solution of triphosgene (0.2 mmol, 59.4

mg) dissolved in 5 mL of cold DCM, which had been chilled on an ice/brine bath. The

mixture was stirred for 1 h at -10 oC. The DCM was evaporated in vacuo to obtain the

crude chloroformate product. A solution of pyridine (96.6 µL, 1.2 mmol) in 2 mL of cold

THF was added dropwise to an ice-cold solution of 4-methylumbelliferone sodium salt

(119 mg, 0.6 mmol) in 5 mL of THF. The cocktail was then added dropwise into the flask

containing the crude chloroformate product at -10 oC. After 1 h, the reaction mixture was washed with 30 mL of 1% TFA solution. The aqueous layer was back extracted with 30 mL of DCM and the combined organic phases were evaporated in vacuo. The crude product was purified by flash chromatography on a silica gel column eluted with 10 % ethyl acetate in hexane to produce a slightly yellow solid (82 mg, 33% yield). 1H NMR

(400 MHz, CDCl3): δ 8.46-8.43 (m, 1H), 7.74-7.57 (m, 3H), 7.22-7.05 (m, 3H), 6.27-6.25

(s, 1H), 4.26 (t, 2H, J = 6.08 Hz), 2.84 (t, 2H, J = 6.72 Hz), 2.42 (s, 3H), 1.85-1.82 (m,

+ 4H) HRMS: calculated for C20H20NO5S2 (M+H ) 418.0783; found 418.0784.

Peptides R9, FΦR4, and cFΦR4 were synthesized by standard Fmoc chemistry to contain a cysteine added to either the C-terminus or the side chain of invariant Gln

residue (Figure 5B). After deprotection, the peptides were purified by HPLC; the fraction

containing the desired peptide was neutralized with pH 8.0 250 mM Tris buffer and

directly added to a solution of 2 equiv of caged 4-MU compound in a minimal volume of

THF. The resulting mixture was shaken under argon protection for 2 h. Then the reaction

mixture was concentrated in vacuo and fractionated by HPLC on a C18 column (eluted

with a linear gradient of acetonitrile in water containing 0.05% TFA) to isolate the

labeled peptides. The authenticity of each conjugated peptide was confirmed by MALDI-

TOF mass spectrometry.

To assess the 4-MU releases of compound 1 (Figure 5) at different pH conditions

with or without reducing environments, 1 μM 4-MU conjugated nonaarginine peptide

was incubated at 37 °C in 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2.0 mM

KH2PO4 buffer of different pH with or without 1 mM dithiothreitol. The release of 4-MU

was measured in absorbance at 360 nm on a Perkin Elmer Lambda 20 UV/Vis

spectrometer (Figure 5C) or on a Molecular Devices Spectramax M5 plate reader (with

excitation and emission wavelengths at 360 and 450 nm, respectively) (Figure 5E).

MCF-7 cells (~1 x 104 cells per well) were seeded in a 96-well plate and cultured for

36 h. The media were removed and the cells were washed with PBS and incubated with

100 µL of 4-MU-labeled peptides in pH 8.0 PBS (5 µM) at 37 or 4 °C. The fluorescence

intensity in each well was recorded every 10 (37 ºC) or 20 min (4 °C) over a 150-min

period on a Molecular Devices Spectramax M5 plate reader (with excitation and emission

wavelengths at 360 and 450 nm, respectively). Experiments were performed in

quadruplicates and a control experiment was performed without cells. The average

fluorescence intensity, after subtraction of the background signal derived from the no-cell

control, was plotted as a function of time.

2.5.6 Confocal Miscroscopy

Approximately 5 x 104 MCF-7 cells were seeded in 35-mm glass-bottomed

microwell dish (MatTek) containing 1 mL of media and cultured for one day. MCF-7

cells were gently washed and treated with the FITC-labeled peptide (5 µM) in growth

media containing 2% serum for 2 h at 37 °C in the presence of 5% CO2. The peptide- containing media was removed and the cells were washed with DPBS three times. The cells were imaged using green and phase contrast channels on a Visitech Infinity 3 Hawk

2D-array live cell imaging confocal microscope (with a 60x oil immersion lens) at 37 °C

in the presence of 5% CO2. FITC-labeled peptides were excited at 488 nm and detected

with a 520-530 nm band-pass filter.

A549 cells (~5 x 104 cells) were similarly seeded in microwell dishes containing 1

mL of phenol-red free media and cultured overnight. On the day of experiment, the

growth medium was removed and the cells were gently washed with DPBS twice. The

cells were incubated with FITC-labeled peptide (5 µM) and rhodamine B-dextran (1 mg

-1 mL ) in growth media containing 2% serum for 2 h at 37 °C in the presence of 5% CO2.

The medium was removed and the cells were gently washed with DPBS twice and incubated for 10 min in 1 mL of DPBS containing 5 μM DRAQ5. The cells were again washed with DPBS twice and imaged as described above. In the case of inhibitor added condition, the cells were preincubated with 75 mM sodium azide in serum-free media for one hour to deplete ATP synthesis before incubation with cyclic peptide. In the case of low temperature study, the cells were preincubated at 4 oC for 30 min. Subsequent live-

cell confocal microscopy was performed as described above.

To examine the co-localization between rhodamine-labeled cyclic peptide

Rho + + (cFΦR4 ) and Rab5 or Rab7 endosomes, HeLa cells transfected with Rab5-GFP or

Rab7-GFP were plated (200 μL, 104 cells/well, 96-well glass bottom MatriPlates) the day

prior to the experiment. On the day of experiment, HeLa cells were treated with 1 μM

Rho cFΦR4 in DMEM media supplemented with 300 nM Hoescht 33342 for 30 min. After

that, the cells were washed with HKR buffer and imaged using a PerkinElmer LiveView

spinning disk confocal microscope.

For GR translocation assay, HeLa cells transfected with GR-GFP were plated as

described above.51 The cells were treated for 30 min with DMEM media containing 1 μM

Dex or Dex-peptide conjugate and 300 nM Hoescht 33342 and imaged using a Zeiss

Axiovert 200M epifluorescence microscope outfitted with Ziess Axiocam mRM camera and an EXFO-Excite series 120 Hg arc lamp. The translocation ratio (the ratio of mean

GFP intensity in the nuclear and surrounding regions) for individual cells was measured

as described before.189 To examine the effect of endocytosis inhibitors, transfected HeLa cells were pretreated for 30 min with clear DMEM containing the inhibitors before incubation with Dex or Dex-peptide conjugates. To test whether Rab5 activity is required for endosomal escape, HeLa cells were transfected with GR-GFP and DsRed-Rab5 WT or DsRed-Rab5Q79L before treatment with Dex or Dex-peptide conjugate and imaged as

described above.

2.5.7 MLV translocation assay

MLVs were prepared as previously described.207 In each experiment, 10 µL of the

lipid solution (18 mM) was mixed with 10 µL of PBS containing 3 µM TAMRA and 10

µL of PBS of 3 µM FITC-labeled peptide. After incubation for 30 min, 4 µL of the

mixture was spotted onto a glass coverslip and the slide was imaged on a confocal

microscope (Olympus Model FV1000) using a 60x oil objective. MLVs of 5-20 µm in

diameter and spherical shape were selected for imaging. Images were recorded at the

focal plane at which the vesicles had the largest area in the fluorescein detection mode.

The vesicles were imaged using a 488 nm laser and 520 nm bandpass filter for

fluorescein and a 543 nm laser with a 580 nm bandpass filter for TAMRA as previously

described.207

2.5.8 Preparation of Small Unilamellar Vesicles

SUVs were prepared by modifying a previously reported procedure.208 A proper

lipid mixture was dissolved in chloroform in a test tube. The lipid mixture was dried

gently by blowing argon over the solution, and kept in a desiccator overnight. The dried

lipids were rehydrated in DPBS to final total lipid concentration of 10 mM. The

suspension was rigorously mixed by vortexing and sonication on ice until it became clear.

Typical preparation yields homogeneous solution containing vesicles with average

diameter between 50-80 nm and polydipersity index smaller than 0.15 as determined by

dynamic light scattering measurements using Zeta Sizer Nano Series (Malvern,

Brookhaven, CT). The SUV solution was stored in fridge and used for fluorescence polarization experiments on the same day.

2.5.9 Fluorescence Polarization

Typically, experiments are performed by incubating 100 nM FITC-labeled peptide with varying concentrations of heparan sulfate (0-5,000 nM) in DPBS for 2 h at room temperature. The FP values were measured on a Molecular Devices Spectramax M5 plate reader, with excitation and emission wavelengths at 485 and 525 nm, respectively. EC50 were determined by plotting the FP values as a function of heparan sulfate concentrations and fitted to a four-parameter logistic curve with GraphPad PRISM ver.6 software. To

obtain the EC50 value of CPP with lipid membranes, the FP experiment was similarly

conducted using 100 nM FITC-labeled peptide with increasing concentrations of SUV

solutions (0-10 mM) in DPBS.

2.5.10 Serum Stability Test

The stability tests were carried by modifying a previously reported procedure.160

Diluted human serum (25%) was centrifuged at 15,000 rpm for 10 min, and the supernatant was collected. A peptide stock solution was diluted into the supernatant to a final concentration of 5 µM for cFΦR4 and Antp and 50 µM for peptides R9 and Tat and

incubated at 37 °C. At various time points (0-6 h), 200-µL aliquots were withdrawn and

mixed with 50 µL of 15% trichloroacetic acid and incubated at 4 ºC overnight. The final

mixture was centrifuged at 15,000 rpm for 10 min in a microcentrifuge, and the

supernatant was analyzed by reversed-phase HPLC equipped with a C18 column (Waters).

The amount of remaining peptide (%) was determined by integrating the area underneath

the peptide peak (monitored at 214 nm) and compared with that of the control reaction

(no serum).

2.5.11 Cytotoxicity Assay

MTT assays were performed to evaluate cyclic peptide’s cytotoxicity against several

mammalian cell lines.209 One hundred μL of MCF-7, HEK293, H1299, H1650, A549

(1×105 cells/mL) cells were placed in each well of a 96-well culture plate and allowed to

grow overnight. Varying concentrations of the peptide (5 or 50 μM) were added to the

each well and the cells were incubated at 37 °C with 5% CO2 for 24 to 72 h. Ten μL of

MTT stock solution was added into each well. Addition of 10 μL of the solution to the

growth medium (no cell) was used as a negative control. The plate was incubated at

37 °C for 4 h. Then 100 μL of SDS-HCl solubilizing buffer was added into each well,

and the resulting solution was mixed thoroughly. The plate was incubated at 37 °C for

another 4 h. The absorbance of the formazan product was measured at 570 nm using a

Molecular Devices Spectramax M5 plate reader. Each experiment was performed in

triplicates and the cells without any peptide added were treated as control.

2.6 Acknowledgements

Colocalization analysis and GR-GFP translocation assay were performed by Dr.

Jonathan R. LaRochelle (from Dr. Alanna Schepartz’s group at Yale University). Part of

cellular association assay was performed by Dr. Tao Liu (Dr. Dehua Pei’s group at The

Ohio State University). SUVs were prepared by Dr. Ryan L. Hard (Dr. Dehua Pei’s group at The Ohio State University).

Chapter 3: Cargo Capacity of Cyclic Cell-Penetrating Peptide

3.1 Introduction

As discussed from the previous chapter, small amphipathic cyclic peptides such as cyclo(FΦRRRRQ) (cFΦR4, where Φ is L-2-naphthylalanine) are internalized by

mammalian cells in an energy-dependent manner, and enter the cytoplasm and nucleus

with higher efficiencies than that of nonaarginine (R9), one of the commonly employed

CPPs.178 Moreover, unlike canonical arginine-rich CPPs, a significant amount of cyclic

CPPs escapes from early-stage endosomes. The ability of cFΦR4 to bypass the acidic late endosome and degradative lysosome offers many exciting applications. In this chapter, we explored three (endocyclic, exocyclic, and bicyclic) complementary methods to effectively deliver a wide variety of cargos, including linear peptides, monocyclic peptides, bicyclic peptides, and proteins into mammalian cells (Figure 13).

Figure 13. Structures showing cargo attachments for endocyclic (A), exocyclic (B), and

bicyclic delivery (C) of cargos (shown in red) with cFΦR4. 59

3.2 Results

3.2.1 Endocyclic Delivery of Peptidyl Cargos with cFΦR4

One potential strategy of cargo delivery is to insert the cargo directly into the cFΦR4

(endocyclic delivery, Figure 13A). Because cFΦR4 internalization involves initial

binding to the plasma membrane phospholipids, one could expect that the ring size of the

CPPs should have a profound effect on the membrane-binding ability and consequently

the cellular association/uptake efficiency. Linear peptides are flexible in solution; binding

to a membrane decreases their conformational freedom and the associated entropy loss

weakens the overall stability of the peptide-membrane complex. On the other hand, cyclic

peptides are more conformationally constrained; a cyclic peptide of the proper sequence

may preorganize into the binding conformation in solution and thus bind to the

membrane with enhanced affinity. This is indeed the case for cFΦR4, which apparently

binds to the MLVs and SUVs. As the ring size of a cyclic peptide increases, however, its

conformational freedom increases and the entropic advantage for membrane binding

decreases, eventually approaching that of a linear peptide.

To determine what ring sizes confer efficient cellular association/uptake, we

synthesized and tested a series of FITC-labeled cyclic peptides that all contained the

same transporter motif FΦRRRR but an increasing number of Ala residues (Table 4,

peptides 1-8). We chose Ala to expand the ring size because the addition of Ala residues

should not significantly change the overall hydrophobicity of the peptide. Indeed, 60

peptides 1-8 all had virtually identical retention times when analyzed by reversed-phase

HPLC (C-18 column). As predicted, peptides of small ring sizes (e.g., 7-, 8-, and 9-amino

acids) associated with MCF-7 cells most efficiently, with activities 8-13-fold higher than

that of R9 (Table 4). The medium sized peptides (10- to 13-amino acids) were less active,

but still more active than R9. Further increase in the ring size appeared to reduce cellular

association, as peptide 8 (14-aa) was 2-fold less active than R9. We conclude that the observed difference in cellular association was primarily due to the different ring sizes, although we cannot rule out potential contributions by other factors such as small differences in hydrophobicity and/or proteolytic stability.

Table 4. Sequences and Cellular Association Efficiency of Monocyclic Peptides

Peptide Abbreviationa Peptide Sequencea Cellular Association (%)b No. FITC 1 cFΦR4 cyclo(FΦRRRRQ)-K(FITC)-NH2 1260 ± 140

FITC 2 cAFΦR4 cyclo(AFΦRRRRQ)-K(FITC)-NH2 820 ±210

FITC 3 cA2FΦR4 cyclo(AAFΦRRRRQ)-K(FITC)-NH2 980 ± 210

FITC 4 cA3FΦR4 cyclo(AAAFΦRRRRQ)-K(FITC)-NH2 210 ± 70

FITC 5 cA4FΦR4 cyclo(AAAAFΦRRRRQ)-K(FITC)-NH2 200 ± 40

FITC 6 cA5FΦR4 cyclo(AAAAAFΦRRRRQ)-K(FITC)-NH2 390 ± 140

FITC 7 cA6FΦR4 cyclo(AAAAAAFΦRRRRQ)-K(FITC)-NH2 240 ± 100

FITC 8 cA7FΦR4 cyclo(AAAAAAAFΦRRRRQ)-K(FITC)-NH2 47 ± 5 aΦ, L-2-naphthylalanine. bCellular association efficiencies were obtained with MCF-7 cells and relative to R9 (100%).

3.2.2 Endocyclic Delivery of Biologically Active Peptides with cFΦR4

We next examined whether short sequence motifs such as FΦR4 could serve as

general transporters of biologically active cyclic peptides into cells. Phosphopeptides,

which have extremely poor membrane permeability, were chosen as cargo to rigorously

test the efficacy of our transporters. To provide a simple readout for peptide

internalization, we used phosphocoumaryl aminopropionic acid (pCAP) as a

phosphotyrosine analogue. pCAP is non-fluorescent; however, dephosphorylation of pCAP or pCAP-containing peptides by protein tyrosine phosphatases (PTPs) releases a

highly fluorescent coumarin derivative (CAP).210,211 Therefore, we synthesized

cyclo[(pCAP)FΦRRRRQ] (Figure 14, cyclic-pCap; Table 5, peptide 1), the

corresponding linear peptide Ac-(pCAP)FΦRRRRQ (linear-pCap; Table 5, peptide 2),

and Ac-(pCap)BRRRRRRRRR (R9-pCap; B, β-Ala; Table 5, peptide 3) and tested their

cellular uptake by MCF-7 cells. Treatment of the cells with 5 μM cyclic-pCap for 60 min

resulted in blue fluorescence in essentially all of the cells (Figure 15A). Furthermore, the

blue fluorescence was present in both the cytoplasm and the nucleus. In contrast, cells

treated with the linear-pCap had much weaker fluorescence under the same conditions

(Figure 15B). When cells were pretreated with 1 mM sodium pervanadate, a potent PTP

inhibitor, followed by 5 μM cyclic-pCap, the amount of CAP fluorescence in the cells

was dramatically reduced (Figure 15C). Cells treated with R9-pCap also showed weak

fluorescence signal, similar to those of linear-pCap and cells pretreated with sodium pervanadate (Figure 15D). Finally, MCF-7 cells were treated with the above peptides (50

μM) and the amount of peptide uptake was quantitated by fluorescence-activated cell sorting (Figure 16). The results show that cyclic-pCap was most efficiently taken up by the MCF-7 cells, reaching a mean fluorescence intensity (MFI) of 714 arbitrary units

(AU), whereas linear-pCap and R9-pCap had 303 and 161 AU, respectively. In vitro assays against a panel of purified recombinant PTPs (PTP1B, TCPTP, SHP-2, RPTPα, and VHR) showed that all three peptides were readily dephosphorylated by these PTPs

(Table 6). These results indicate that the FΦR4 motif was able to deliver the cyclic phosphopeptide into the mammalian cells and subsequent dephosphorylation of pCAP by endogenous PTPs generated the intracellular fluorescence. Since all known PTPs have their catalytic domains localized in the cytoplasm or nucleus, these data provide further support that the cyclic CPPs can efficiently escape from the endosome and deliver cargos into the cytoplasm and nucleus of mammalian cells. Cyclic-pCap should provide a useful probe for monitoring the intracellular PTP activities.

Table 5. Sequences of Peptides Studied in Chapter 3

Peptide Abbreviation Peptide Sequencea No. 1 cyclic-pCap cyclo[(pCAP)FΦRRRRQ)-K(Rho)-NH2

2 linear-pCap Ac-(pCAP)FΦRRRRE(OAll)-NH2

3 R9-pCap Ac-(pCAP)BRRRRRRRRR-NH2

Rho 4 cFΦR4-R5 cyclo(FΦRRRRQ)-RRRRR-K(Rho)-NH2

Rho 5 cFΦR4-A5 cyclo(FΦRRRRQ)-AAAAA-K(Rho)-NH2

Rho 6 cFΦR4-F4 cyclo(FΦRRRRQ)-FFFF-K(Rho)-NH2

7 cFΦR4-PCP cyclo(FΦRRRRQ)-miniPEG-DE(pCAP)LI-NH2

8 PCP Ac-DE(pCAP)LI-NH2

9 R9-PCP Ac-RRRRRRRRR-miniPEG-DE(pCAP)LI-NH2

10 Tat-PCP Ac-RKKRRQRRR-miniPEG-DE(pCAP)LI-NH2

11 Antp-PCP Ac-RQIKIWFQNRRMKWKK-miniPEG-DE(pCAP)LI-NH2

12 cFΦR4-CHO cyclo(FΦRRRRQ)-DEVD-H 13 LCPP-CHO Ac-AAVALLPAVLLALLAP-DEVD-H

14 FMK Z-VAD(OMe)-CH2F 15 CHO Ac-DEVD-H

Rho 16 bicyclo(FΦR4-A5) bicyclo[Tm(AAAAA)K(RRRRΦF)J]-K(Rho)-NH2

Rho 17 bicyclo(FΦR4-A7) bicyclo[Tm(AAAAAAA)K(RRRRΦF)J]-K(Rho)-NH2

Rho 18 bicyclo(FΦR4-RARAR) bicyclo[Tm(RARAR)K(RRRRΦF)J]-K(Rho)-NH2

Rho 19 bicyclo(FΦR4-DADAD) bicyclo[Tm(DADAD)K(RRRRΦF)J]-K(Rho)-NH2

Rho 20 monocyclo(FΦR4-A5) cyclo(AAAAARRRRΦF)-K(Rho)-NH2

Rho 21 monocyclo(FΦR4-A7) cyclo(AAAAAAARRRRΦF)-K(Rho)-NH2

22 monoPTP1B cyclo(SvP-F2Pmp-HRRRRΦFQ)-K-NH2

23 bicycloPTP1B bicyclo[Tm(SvP-F2Pmp-H)J(FΦRRRR)J]-K-NH2 a pCAP, phosphocoumaryl amino propionic acid; Φ, L-2-naphthylalanine; B, β-alanine; Rho, rhodamine B; Z, benzyloxycarbonyl; CH2F, fluoromethylketone; miniPEG, 8-amino-3,6- dioxaoctanoic acid; J, L-2,3-diaminopropionic acid; Tm, trimesoyl; v, D-valine; F2Pmp, L-4- (phosphonodifluoromethyl)phenylalanine;

Figure 14. Structures of cyclic-pCap, bicyclo-PTP1B, cFΦR4-PCP, and cFΦR4-CHO as examples of endocylic, bicyclic, and exocyclic delivery methodologies.

Table 6. Kinetic Activites of Recombinant PTPs toward pCAP-Containing Peptides

a,b PTP cyclic-pCap linear-pCap R9-pCap cFΦR4-PCP Tat-PCP R9-PCP Antp-PCP

PTP1B 4300 540 499 37100 13800 14700 17400 TCPTP 1240 670 147 2780 560 457 970 SHP2 381 214 50 7400 2290 248 2210 CD45 NDc ND ND 35100 21800 2940 22300 VHR 10400 26000 5900 2460 1460 6240 2030 RPTPα 322 822 236 ND ND ND ND a 212 b -1 -1 Recombinant PTPs were expressed in E. coli and purified as described previously. kcat/KM (M s ) were measured as described elsewhere.212 c ND, not tested.

Figure 15. Endocyclic delivery of phosphatase fluorogenic substrate with cFΦR4. Fixed-cell confocal images of MCF-7 cells treated with pCAP-containing peptides (5 μM each) and nuclear stain DRAQ5. (A) Cells treated with cyclic-pCap and DRAQ5 in the same Z-section. (B) Same as panel A but with linear-pCap. (C) Same as panel A except that the cells were pretreated with 1 mM sodium pervanadate prior to the addition of cyclic-pCap and DRAQ5. (D) Same as panel A but with R9-pCap.

Figure 16. Flow cytometry of MCF-7 cells treated with pCAP-containing peptides. MFI, mean fluorescence intensity.

3.2.3 Exocyclic Delivery of Peptidyl Cargos with cFΦR4

Since endocyclic delivery by cFΦR4 is limited to a heptapeptide or smaller cargos, we tested the ability of cFΦR4 to deliver cargos of varying sizes and physicochemical properties attached to the Gln side chain (Figure 13B, exocyclic delivery). We first covalently attached positively charged (RRRRR), neutral (AAAAA), hydrophobic

(FFFF), and negatively charged peptides [DE(pCAP)LI] to cFΦR4 (Table 5, peptides 4-7).

The first three peptides were labeled with rhodamine B at a C-terminal lysine side chain, and their internalization into HEK293 cells was examined by live-cell confocal

Rho microscopy. Cells incubated for 2 h with 5 μM peptide cFΦR4-A5 (Figure 17A) or

Rho cFΦR4-R5 (Figure 17B) showed evidence of both punctate and diffuse fluorescence, 67

with the latter distributed almost uniformly throughout the cell. In contrast, the fluid

phase endocytic marker FITC-labled dextran (dextranFITC) displayed predominantly punctate fluorescence, indicative of endosomal localization. The diffuse rhodamine fluorescence suggests that a fraction of the peptides reached the cytosol and nucleus of

Rho the cells. cFΦR4-F4 could not be tested due to its poor aqueous solubility.

Figure 17. Exocyclic peptides delivery with cFΦR4. Live-cell images of HEK293 cells treated with rhodamine B-labeled peptides and fluid-

FITC FITC phase uptake marker, dextran . (A) Cells treated with 5 μM cFΦR4-A5 and dextran

in the same Z-section. (B) Same as (A) but with 5 μM cFΦR4-R5.

Peptide cFΦR4-DE(pCAP)LI (cFΦR4-PCP; Figure 14, Table 5, peptide 7) was used to assess the cytoplasmic delivery efficiency of cFΦR4 relative to R9, Tat, and penetratin

(Antp), which are some of the most widely used and active CPPs. It is noteworthy that cFΦR4-PCP carries a net charge of zero at physiological pH. Untagged PCP [Ac-

DE(pCap)LI-NH2] and PCP conjugated to R9 (R9-PCP), Tat (Tat-PCP), or penetratin

(Antp-PCP) (Table 5, peptides 8-11) were also prepared. pCAP is non-fluorescent, but

upon entering the cell interior, should be rapidly dephosphorylated by endogenous

protein tyrosine phosphatases (PTPs) to produce coumaryl aminopropionic acid, a

fluorescent product (CAP, excitation 355 nm; emission 450 nm). 210,211 When assayed

against a PTP panel in vitro, all four CPP-PCP conjugates were efficiently

dephosphorylated (Table 6). This assay detects only the CPP-cargo inside the cytoplasm

and nucleus, where the catalytic domains of all known mammalian PTPs are localized.218

Further, CAP is fluorescent only in its deprotonated state (pKa = 7.8); even if some dephosphorylation occurs inside the endosome (pH 6.5-4.5) or lysosome (pH 4.5), it would contribute little to the total fluorescence.219 Treatment of HEK293 cells with 5 μM cFΦR4-PCP for 60 min resulted in diffuse blue fluorescence throughout the cell,

suggesting that cFΦR4-PCP reached the cell interior, whereas the untagged PCP failed to enter cells under the same condition (Figure 18A). When HEK293 cells were pretreated

with the PTP inhibitor sodium pervanadate for 1 h prior to incubation with cFΦR4-PCP

(5 μM), the CAP fluorescence in the cells diminished to background levels. HEK293 cells treated with R9-PCP, Antp-PCP, or Tat-PCP under identical conditions showed

weak fluorescence, consistent with the poor ability of these peptides to access the cell

interior (Figure 18A). To quantify the relative intracellular PCP delivery efficiency, HeLa

cells were treated with each peptide and analyzed by fluorescence activated cell sorting

(Figure 18B). cFΦR4-PCP was most efficiently internalized by the HeLa cells, with a mean fluorescence intensity (MFI) of 3510 arbitrary units (AU), whereas R9-PCP, Antp-

PCP, Tat-PCP, and untagged PCP produced MFI values of 960, 400, 290, and 30 AU,

respectively (Figure 18C). Again, when cells were treated with cFΦR4-PCP in the

presence of sodium pervanadate, the amount of CAP fluorescence was reduced to near

background levels (70 AU). Thus, cFΦR4 is capable of delivering peptidyl cargos of

varying physicochemical properties into the cytoplasm with efficiencies 4-12-fold higher

than R9, Antp, and Tat.

Figure 18. Exocyclic delivery of pCAP-containing peptides with cFΦR4.

I, untagged PCP; II, cFΦR4-PCP; III, cFΦR4-PCP and Na3VO4; IV, R9-PCP; V, Tat-PCP; and VI, Antp-PCP. (A) Representative live-cell confocal images of HEK293 cells treated with 5 μM peptides. Top panel, nuclear stain with DRAQ5; bottom panel, CAP fluorescence in the same Z-section. (B) Flow cytometry of HeLa cells treated with 0 or 10 μM peptides. (C) CAP fluorescence from (B) after subtraction of background fluorescence (untreated cells). MFI, mean fluorescence intensity. 70

To test the ability of cFΦR4 to deliver bioactive cargo into cells, we conjugated the

cell impermeable inhibitor of caspases 3 and 7, peptidyl aldehyde DEVD-H,213 to the Gln

side chain of cFΦR4 (cFΦR4-CHO) (Figure 19 and Table 5, peptide 12), and compared

the antiapoptotic activity of this conjugate with two previously reported cell-permeable

caspase inhibitors, LCPP-CHO214 and FMK215 (Table 5, peptides 13 and 14), and free

Ac-DEVD-H (CHO; Table 5, peptide 15). To assess protection from apoptosis, Jurkat

cells were pretreated with caspase inhibitors or DMSO for 30 min prior to 24 h

incubation with the kinase inhibitor staurosporine, which induces apoptosis through both

caspase-dependent and independent mechanisms.216 Post treatment, cell viability was

measured with the cell health indicator alamarBlue®.217 As expected, peptide Ac-DEVD-

H did not display any protective effect against staurosporine-induced apoptosis at either

25 or 100 μM, likely due to its inability to enter the cells, while LCPP-CHO and FMK exhibited dose-dependent antiapoptotic activities (Figure 17). At 100 μM concentration,

LCPP-CHO and FMK resulted in 40% and 49% protection against staurosporine-induced apoptosis, respectively. cFΦR4-CHO displayed the greatest protection against apoptosis,

providing 16% and 63% protection at 25 and 100 μM, respectively (Figure 19). Taken

together, the above results demonstrate that cFΦR4 is capable of cytosolic delivery of

bioactive peptide cargos of varying physicochemical properties.

70 CHO 60 LCPP-CHO FMK 50 cFΦR4-CHO 40 30

% Protection 20 10 0 25 μM100 μM

Figure 19. Protection of Jurkat cells from staurosporine-induced apoptosis by caspase inhibitor

DEVD-H (CHO), its peptide conjugates LCPP-CHO and cFΦR4-CHO, or prodrug FMK.

3.2.4 Exocyclic Delivery of Proteins with cFΦR4

To test whether cFΦR4 is capable of transporting full-length proteins into

mammalian cells, we chose GFP because of its intrinsic fluorescence and attached cFΦR4 to its N-terminus through a disulfide bond (Figure 20A). The disulfide exchange reaction is highly specific, efficient, and reversible; upon entering the cytoplasm, the CPP-S-S- protein conjugate is expected to be rapidly reduced to release the native protein. Although cFΦR4 can be directly attached to a native or engineered surface cysteine residue(s) on a

cargo protein, we employed a GFP variant containing a 12-amino acid ybbR tag at its N-

terminus (which was already available in our laboratory) and used phosphopantetheinyl

230 transferase Sfp to enzymatically attach cFΦR4 to the ybbR tag. Indeed, ybbr-tagged

protein was efficiently conjugated with coenzyme A (CoA) and cFΦR4 adduct (Figure

21). The success of protein labeling is confirmed by both native polyacrylamide gel

electrophoresis and mass spectrometric analysis. This method permitted the attachment of

a single cFΦR4 unit to GFP in a site-specific manner. Incubation of HEK293 cells in the

presence of 1 μM cFΦR4-S-S-GFP resulted in accumulation of green fluorescence inside

the cells (Figure 20B). The fluorescence signal was present throughout the entire cell

volume and the intensity increased upon extended incubation. Some of the cells

contained small spots of intense green fluorescence (indicated by arrows in Figure 20B),

which may represent endosomally sequestered cFΦR4-S-S-GFP or aggregated GFP

inside the cell. As expected, untagged GFP was unable to enter cells, as judged by the

lack of detectable cellular fluorescence, whereas Tat-S-S-GFP entered cells less

efficiently than cFΦR4-S-S-GFP (Figure 20B); flow cytometry analysis of HeLa cells

treated with 1 μM protein revealed a 5.5-fold higher total intracellular fluorescence for the latter. Moreover, cells treated with Tat-S-S-GFP showed predominantly punctant fluorescence in the cell periphery with no detactable fluorescence in the nuclear region, suggesting that Tat-S-S-GFP is mostly entrapped in the endosomes, in agreement with

83 previous reports. Thus, with a protein as cargo, cFΦR4 also has substantially higher

efficiency than Tat with regard to both uptake and endosomal escape. Attemps to test R9-

S-S-GFP with cells failed due to precipitation of the conjugate.

To demonstrate the generality of cFΦR4 for protein delivery, we chose next to deliver a functional protein, the catalytic domain of PTP1B (amino acids 1-321), into the cell interior at a level sufficient to perturb the cell signaling process. To show that a non- cleavable linker is also compatible with our delivery method, we conjugated cFΦR4 to

ybbR-tagged PTP1B via a thioether linkage (cFΦR4-PTP1B) (Figure 21). This protein- 73

peptide was constructed through bifunctional linker LC-SMCC. The production of

conjugated peptide was also confirmed by native polyacrylamide gel electrophoresis. In

vitro assay using p-nitrophenyl phosphate as substrate showed that addition of the cFΦR4 tag does not affect the catalytic activity of PTP1B.219 NIH-3T3 cells were incubated for 2

h in the presence of untagged PTP1B or cFΦR4-PTP1B and their global pY protein levels were analyzed by anti-pY western blotting (Figure 22). Treatment of the cells with cFΦR4-PTP1B, but not untagged PTP1B, resulted in concentration-dependent decrease in

pY levels of most, but not all, proteins. The total cellular protein levels, as detected by coomassie blue staining, were unchanged (Figure 22B), indicating that the observed decrease in pY levels was due to dephosphorylation of the pY proteins by cFΦR4-PTP1B

and/or secondary effects caused by the introduction of cFΦR4-PTP1B (e.g., inactivation

of cellular protein tyrosine kinases). Interestingly, different proteins exhibited varying

dephosphorylation kinetics. Several proteins in the 150-200 kD range were completely

dephosphorylated upon the addition of 62 nM cFΦR4-PTP1B, whereas proteins of ~80

kD remained phosphorylated at 500 nM cFΦR4-PTP1B. The changes in the pY pattern

are consistent with the broad substrate specificity of PTP1B212 and very similar to that

caused by overexpression of PTP1B inside the cytosol of mammalian cells.231 Our results

indicate that cFΦR4 is indeed able to deliver PTP1B into the interior of NIH 3T3 cells in

the catalytically active form and to sufficient levels to dramatically perturb the cell

signaling process. cFΦR4 thus provides a powerful tool for introducing other functional

proteins, especially proteins that cannot be genetically expressed (e.g., toxic and

chemically modified proteins), into mammalian cells in order to study their cellular

functions.

Figure 20. Intracellular delivery of GFP with cFΦR4. (A) Structures of CPP-S-S-GFP conjugates. (B) Live-cell images of HEK293 cells after

2-h treatment with 1 μM GFP (I), Tat-S-S-GFP (II), or cFΦR4-S-S-GFP (III) and nuclear stain DRAQ5. All images were recorded in the same Z-section.

S N N S N S Coenzyme A CPP-SH CPP-S CPP-S-S-CoA PBS buffer pH 7.5, 2h PBS buffer, pH 6.0, o/n

VLDSLEFIASKL-GFP ybbr-GFP HO O O CPP O S O Sfp HN S N N P H H HO H HO O O NH2 VLDSLEFIASKL CPP-S-S-GFP B O O O O N N H CPP O N O O H LC-SMCC HN N O O N CPP-Lys H O PBS buffer, pH 7.4, 2h O NH2 CPP-SMCC O

VLDSLEFIASKL-PTP1B ybbr-PTP1B Coenzyme A HO CPP-SMCC-CoA PBS buffer pH 7.5, 2h Sfp

CPP O VLDSLEFIA H SKL HN N O O O N O H S O O N P O NH N 2 H H HO H O OH O

Figure 21. Scheme showing the synthesis of (A) CPP-S-S-GFP and (B) CPP-PTP1B.

Table 7. Kinetic Activities of PTP1B and cFΦR4-PTP1B against pNPP

PTP cFΦR4-PTP1B PTP1B -1 -1 kcat/KM (M s ) 1604 1338 212 pNPP = p-nitrophenyl phosphate; kcat/KM was measured as described elsewhere.

Figure 22. Intracellular delivery of PTP1B with cFΦR4. (A) Western blot analysis of the global pY protein levels of NIH 3T3 cells after treatment with 0-500 nM PTP1B or cFΦR4-PTP1B (B) Same samples as in (A) were analyzed by SDS-PAGE and coomassie blue staining. M, molecular-weight markers. (IB: anti-pY antibody 4G10).

3.2.5 Bicyclic Delivery of Peptidyl Cargos with cFΦR4

In recent years, there has been much interest in cyclic peptides as therapeutic agents and biomedical research tools.157,220 For example, cyclic peptides are effective for

inhibition of protein-protein interactions,221-224 which are challenging targets for

conventional small molecules. A major obstacle in developing cyclic peptide therapeutics

is that the vast majority of them are impermeable to the cell membrane. Our attempt to

deliver cyclic peptides by cFΦR4 by the endocyclic method had only limited success;

increase in the cargo size from 1 to 7 residues led to progressively lower cellular uptake,

likely because the larger, more flexible rings bind more poorly to the cell membrane.178

To overcome this limitation, I collaborated with W. Lian and B. Jiang of the Pei group to explore a bicyclic peptide system, in which one ring contains a CPP motif (e.g., FΦR4)

while the other ring consists of peptide sequences specific for the desired targets (Figure

13C). The bicyclic system should in principle be able to accommodate cargos of any size,

because the cargo does not change the structure of the CPP ring and should have less

impact on its delivery efficiency. The additional rigidity of a bicyclic structure should

also improve its metabolic stability as well as the target-binding affinity and specificity.

The bicyclic peptides were readily synthesized by forming three amide bonds between a

trimesoyl scaffold and three amino groups on the corresponding linear peptide [i.e., the

N-terminal amine, the side chain of a C-terminal diaminopropionic acid (Dap), and the

side chain of a lysine (or ornithine, Dap) imbedded in between the CPP and target-

224 binding motifs]. To test the validity of this approach, we chose FΦR4 in the C-terminal

ring as the CPP moiety and peptides of different lengths and charges (AAAAA,

AAAAAAA, RARAR, or DADAD) as cargo (Table 5, peptides 16-19). For comparison, we also prepared two monocyclic peptides containing FΦR4 as transporter and peptides

A5 and A7 as cargos (Table 5, peptides 20 and 21). All of the peptides (prepared by

Bisheng Jiang) were labeled at a C-terminal lysine side chain with rhodamine B and their

internalization into HEK293 cells was examined by live-cell confocal microscopy.

Treatment of cells with 5 μM peptide for 2 h resulted in efficient internalization of all six

peptides (Figure 23). The intracellular distribution of the internalized peptides was quite

78 different between the bicyclic and monocyclic peptides. While the four bicyclic peptides showed evidence for their presence in both the cytoplasm/nucleus (as indicated by the diffuse rhodamine fluorescence) and the endosomes (as indicated by the fluorescence puncta), the monocyclic peptides exhibited predominantly punctate fluorescence that overlapped with that of the endocytic marker. In all cases, the endocytic marker displayed only punctate fluorescence, indicating that the endosomes were intact in the cells treated with the peptides. These results indicate that the increased structural rigidity of the bicyclic peptides facilitates both the uptake by endocytosis and endosomal release, presumably because of their improved binding to the plasma and endosomal membranes.

The bicyclic system should provide a general strategy for intracellular delivery of bicyclic peptides.

Figure 23. Bicyclic peptide delivery with cFΦR4. Live-cell confocal microscopic images of HEK293 cells treated for 2 h with rhodamine B-labeled bicyclic peptides and endocytosis marker, dextranFITC. The red fluorescence of rhodamine B and the green fluorescence of dextranFITC from the same Z-section and their merged image are shown in each panel. The enlarged images of a typical cell(s) are shown in each case in order to show the intracellular distribution of the internalized

Rho Rho peptides. (A) Cells treated with bicyclo(FΦR4-A5) ; (B) monocyclo(FΦR4-A5) ; (C)

Rho Rho Rho bicyclo(FΦR4-A7) ; (D) monocyclo(FΦR4-A7) ; (E) bicyclo(FΦR4-RARAR) ; and

Rho (F) bicyclo(FΦR4-DADAD) .

To further demonstrate the utilities of bicyclic delivery system, we investigated one cell-permeable and biologically active bicyclic peptide inhibitor against protein tyrosine phosphatase 1B (PTP1B) based on a non-permeable monocyclic PTP1B inhibitor developed by W. Lian using one-bead two-compound library screening technology221.

PTP1B is a prototypical member of the PTP superfamily and plays numerous roles during eukaryotic cell signaling. Because of its role in negatively regulating insulin and leptin

receptor signaling, PTP1B is a valid target for treatment of type II diabetes and

obesity.225,226 A large number of PTP1B inhibitors have been reported,227 however, none

of them have succeeded in the clinic. Designing PTP inhibitors is challenging because

most of the phosphotyrosine (pY) isosteres such as difluorophosphonomethyl

228 phenylalanine (F2Pmp) are impermeable to the cell membrane. Additionally, because all PTPs share a similar active site, achieving selectivity for a single PTP has been

difficult. Based on a monocyclic peptide ligand, which was identified from one-bead two-

compound library screening, we greatly improved the cellular uptake by applying the

bicyclic delivery system in which the FΦR4 motif is placed in one ring whereas the

enzyme-binding sequence constitutes the other ring (Figure 24).229

Figure 24. Evolution of a cell-permeable bicyclic PTP1B inhibitor.229

To convert the monocyclic PTP1B inhibitor monoPTP1B (Table 5, peptide 22;

Figure 24) into a bicyclic peptide, we replaced the Gln residue (used for attachment to the

solid support and peptide cyclization) with L-2,3-diaminopropionic acid (Dap, J) and 81

inserted a second J residue at the junction of CPP and PTP1B-binding sequences (C-

terminal to histidine) (Figure 24). To facilitate labeling with fluorescent probes, a lysine

was added to the C-termini, and the resulting bicyclic PTP1B inhibitor bicycloPTP1B

(Table 5, peptide 23; Figure 24) was obtained following standard bicyclic peptide synthesis, cleavage and HPLC purification.224 As expected, inhibitor bicycloPTP1B has

greatly improved cell permeability over inhibitor monoPTP1B as detected by live-cell

confocal microscopic imaging of A549 cells treated with FITC-labeled inhibitors (Figure

25). The bicycloPTP1B treated cells showed both diffuse fluorescence throughout the

cytoplasm and nucleus as well as fluorescence puncta, indicating that significant amount

of the inhibitor escaped from endocytic pathways. As we demonstrated previously, the

smaller CPP ring of a bicyclic peptide should result in greater structural rigidity and

improved metabolic stability. Incubation of bicycloPTP1B in human serum for 24 h at

37 °C resulted in ~10% degradation, whereas ~90% of monoPTP1B was degraded under

the same condition (Figure 26). Furthermore, Wenlong Lian demonstrated the in vitro

and in vivo activities of bicycloPTP1B including its desired inhibition potency against

PTP1B, excellent selectivity over the highly similar TCPTP (17-fold), phosphatase

inhibition activity when tested with A549 and excellent performance on regulating

PTP1B downstream cell signaling.229 Taken together, inhibitor bicycloPTP1B compares

favorably with the small-molecule PTP1B inhibitors reported to date227 with respect to

potency, selectivity, cell permeability, and stability.

Figure 25. Delivery of F2Pmp containing bicyclic peptide with cFΦR4. Live-cell confocal microscopic images of A549 cells treated for 2 h with FITC-labeled PTP1B inhibitors and endocytic marker. (A) Cells treated with 5 μM bicycloPTP1B and dextranRho in the same Z-section. (B) Same as (A) but with 5 μM monoPTP1B.

100

80 bicycloPTP1B 60 monoPTP1B 40

20 %Peptide Remaining

0 0510152025 Time (h)

Figure 26. Comparison of the serum stability of monoPTP1B and bicycloPTP1B.

3.3 Discussion

We demonstrate that cFΦR4 is effective for endocyclic, exocyclic, and bicyclic delivery of biologically active cargos into the cytoplasm and nucleus of mammalian cells.

For the “endocyclic” delivery method, relatively short peptides (≤7 amino acids) could be inserted into cFΦR4 peptide ring, and the resulting cyclic peptides showed comparable 83

internalization efficiency. Furthermore, incorporation of a phosphocoumaryl

aminopropionic acid (pCAP) residue into cFΦR4 generated a cell permeable, fluorogenic

probe for detecting intracellular protein tyrosine phosphatase activities. For the

“exocyclic” delivery approach, various cargo molecules including linear peptides of

varying charges, pCAP containing peptides, capase inhibitor, and large proteins (e.g.,

PTP1B and GFP) were attached to the glutamine side chain. We found that cFΦR4 is

remarkably tolerant to the size and nature of cargos and efficiently transported all of the

cargos tested into the cytoplasm and nucleus of mammalian cells. For the “bicyclic”

delivery method, we demonstrated the capacity of cFΦR4 of delivering bicyclic peptide

by fusing cFΦR4 and cyclic cargo rings. Bicyclic peptides carrying different various

charges and ring sizes were successfully delivered by cFΦR4. Additionally, incorporation

of PTP1B-binding motif with transporting sequence produced potent, selective,

proteolytically stable, and biologically active inhibitors against the enzyme. Our preliminary studies uncovered a potentially general approach to design cell-permeable bicyclic peptides against targets.

By using a pCAP-containing peptide as cargo/reporter, we show that cFΦR4 is significantly more efficient than R9, Tat, and Antp for cytoplasmic cargo delivery (Figure

18), making cFΦR4 one of the most active CPPs known to date. Although modification of

polybasic CPPs such as addition of hydrophobic acyl groups has previously been reported to enhance cellular uptake by a similar magnitude,170 to our knowledge, these previous studies have not established whether the enhanced uptake translates into a similar increase in the cytoplasmic CPP concentration (Our attempt to directly compare cFΦR4 84 and the acylated CPPs was not successful, because the latter caused extensive cell death during our experiments). The pCAP-based reporter system described in this work should provide a simple, robust method to quantitatively assess the cytoplasmic delivery efficiency of other CPPs. Early endosomal release offers significant advantages, especially for peptide and protein cargos, since it minimizes cargo degradation by late endosomal and lysosomal proteases and denaturation caused by acidification during endosomal maturation. Indeed, both GFP and PTP1B delivered into the cytoplasm by cFΦR4 were in their folded, active forms, as evidenced by the green fluorescence and the ability to dephosphorylate intracellular pY proteins, respectively. Additionally, due to its more rigid structure, cFΦR4 is significantly more stable against proteolytic degradation than linear peptides, and due to its smaller size, cFΦR4 is less expensive to synthesize and potentially less likely to interfere with the cargo function. Direct protein delivery provides a useful research tool, e.g., for studying the cellular function of a protein, as it offers improved temporal control over DNA transfection and subsequent gene expression and allows delivery of chemically modified proteins and proteins whose overexpression causes toxicity. Taken together, cFΦR4 provides a versatile and extremely useful transporter for cytosolic cargo delivery.

3.4 Conclusion

In this chapter, cyclic heptapeptide cyclo(FΦRRRRQ) (cFΦR4, where Φ is L-2- naphthylalanine) was used to deliver a wide variety of molecules including monocyclic,

linear and bicyclic peptides of various charged states, and proteins. Depending on the

nature of the cargos, they may be delivered by endocyclic (insertion of cargo into the

cFΦR4 ring), exocyclic (attachment of cargo to the Gln side chain), or bicyclic

approaches (fusion of cFΦR4 and cyclic cargo rings). The overall delivery efficiency (i.e.,

delivery of cargo into the cytoplasm and nucleus) of cFΦR4 was 4- to 12-fold higher than

those of R9, HIV Tat peptide, or penetratin. The higher delivery efficiency, coupled with

superior serum stability, minimal toxicity, and synthetic accessibility, renders cFΦR4 a

useful transporter for intracellular cargo delivery and a suitable system for investigating

the mechanism of endosomal escape.

3.5 Experimental Section

3.5.1 Materials

Reagents for peptide synthesis were purchased from Advanced ChemTech

(Louisville, KY), NovaBiochem (La Jolla, CA), or Anaspec (San Jose, CA). 2,2’-

Dipyridyl disulfide, Lissamine rhodamine B sulfonyl chloride, fluorescein isothiocyanate

(FITC), coenzyme A trilithium salt, FITC-labeled dextran (dextranFITC), rhodamine B- labeled dextran (dextranRho) and human serum were purchased from Sigma-Aldrich (St.

Louis, MO). Alamar Blue, cell culture media, fetal bovine serum (FBS), penicillin-

streptomycin, 0.25% trypsin-EDTA, Dulbecco’s phosphate-buffered saline (DPBS) (2.67

mM potassium chloride, 1.47 mM potassium phosphate monobasic, 137 mM sodium

chloride, 8.06 mM sodium phosphate dibasic), and live/dead fixable far red dead cell 86

stain kit were purchased from Invitrogen (Carlsbad, CA). PD-10 desalting columns were

purchased from GE-Healthcare (Piscataway, NJ). Anti-phosphotyrosine (pY) antibody

(clone 4G10) was purchased from Millipore (Temecula, CA). Rink resin LS (100-200

mesh, 0.2 mmol/g) was purchased from Advanced ChemTech. LC-SMCC (succinimidyl-

4-[N-maleimidomethyl] cyclohexane-1-carboxy-[6-amidocaproate]) and nuclear staining

dye DRAQ5TM was purchased from Thermo Scientific (Rockford, IL). Caspase inhibitors

Ac-AAVALLPAVLLALLAP-DEVD-H (LCPP-CHO) and Z-VAD(OMe)-CH2F (FMK) were purchased from Santa Cruz Biotechnology (Dallas, TX).

3.5.2 Peptide Synthesis and Labeling

Peptides were synthesized on Rink Resin LS (0.2 mmol/g) using standard Fmoc chemistry. The typical coupling reaction contained 5 equiv of Fmoc-amino acid, 5 equiv of 2-(7-aza-1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate

(HATU) and 10 equiv of diisopropylethylamine (DIPEA) and was allowed to proceed with mixing for 75 min. After the addition of the last (N-terminal) residue, the allyl group on the C-terminal Glu residue was removed by treatment with Pd(PPh3)4, phenylsilane

(0.1 and 10 equiv, respectively) in anhydrous DCM (3 x 15 min). The N-terminal Fmoc

group was removed by treatment with 20% piperidine in DMF and the peptide was

cyclized by treatment with benzotriazole-1-yl-oxy-tris-pyrrolidino-phosphonium

hexafluorophosphate (PyBOP)/HOBt/DIPEA (5, 5, and 10 equiv) in DMF for 3 h. The

peptides were deprotected and released from the resin by treatment with 82.5:5:5:5:2.5

(v/v) TFA/thioanisole/water/phenol/ethanedithiol for 2 h. The peptides were triturated

with cold ethyl ether (3x) and purified by reversed-phase HPLC on a C18 column.

Peptide labeling with FITC was performed by dissolving the purified peptide (~1 mg) in 300 µL of 1:1:1 (v/v) DMSO/DMF/150 mM sodium bicarbonate (pH 8.5) and mixing with 10 µL of FITC in DMSO (100 mg/mL). After 20 min at room temperature, the reaction mixture was subjected to reversed-phase HPLC on a semi-preparative C18 column to isolate the FITC-labeled peptide. To generate rhodamine-labeled peptides, an

Nε-4-methoxytrityl-L-lysine was added to the C-terminus. After the solid phase peptide synthesis, the lysine side chain was selectively deprotected using 1% (v/v) trifluoroacetic acid in DCM. The resin was incubated with Lissamine rhodamine B sulfonyl chloride/DIPEA (5 equiv each) in DMF overnight. The peptides were fully deprotected as described in above, triturated with diethyl ether, and purified by HPLC. Peptide aldehydes were synthesized on H-Asp(OtBu)-H NovaSyn® TG resin (NovaBiochem) as

described above. Final peptide aldehyde release was carried out by treatment with

AcOH/water/DCM/MeOH (10:5:63:21) for 1 h twice. The peptide was then deprotected,

triturated, and purified by HPLC. Bicyclic peptides, phosphocoumaryl aminopropionic

acid (pCAP), and pCAP-containing peptides were synthesized as previously

described.210,211,224 The authenticity of each peptide was confirmed by MALDI-TOF mass

spectrometry.

3.5.3 Cell Culture

HEK293, MCF-7, NIH 3T3 and HeLa cells were maintained in medium consisting

of DMEM, 10% FBS and 1% penicillin/streptomycin. Jurkat cells were maintained in

medium consisting of RPMI-1640, 10% FBS and 1% penicillin/streptomycin. Cells were

cultured in a humidified incubator at 37 °C with 5% CO2.

3.5.4 Quantification of CPP Cellular Association

Approximately 1 x 104 MCF-7 cells were seeded in 12-well culture plates (BD

Falcon) in 1 ml of media and cultured for two days. On the day of experiment, 500 µL of

an FITC-labeled peptide solution (5 µM) in HKR buffer (5 mM HEPES, 137 mM NaCl,

HKR buffer (2 x 1 ml) and treated with 200 μL of 0.25% (w/v) trypsin-EDTA for 10 min.

the cellular association efficiency of the FITC-labeled peptide (in pmol of peptide

internalized/mg of cellular protein) was calculated by dividing the amount of internalized

peptide by the amount of protein in the lysate. The experiments were performed twice

and in triplicates each time.

3.5.5 Confocal Microscope

To examine the internalization of pCAP containing cyclic peptide using fixed-cell

confocal microscopic imaging, ~8,000 MCF-7 cells were seeded in 8-well chamber slides

(Lab-Tek) containing 0.2 mL of media and cultured for one day. On the day of

experiment, the growth medium was removed, and the cells were gently washed with

DPBS twice. The cells were incubated with 0.2 mL of CPP solution (5 µM) at 37 °C for

60 min in the presence of 5% CO2. After removal of the medium, the cells were gently

washed with DPBS twice and incubated for 10 min in 0.2 mL of DPBS containing 5 μM

DRAQ5. The resulting cells were washed with DPBS twice, fixed by treatment with 0.2

mL of 4% paraformaldehyde for 20 min on ice and washed with DPBS (3 times). The

chamber slide was sealed with a coverslip and subjected to confocal imaging on an

Olympus FV1000-Filter confocal microscope (with a 60x oil objective). Coumarin

fluorescence was generated with excitation at 405 nm using a diode laser and detected

using a DAPI emission filter. All images were recorded by using the same parameters.

To examine the internalization of rhodamine-labeled peptides, 5 x 104 HEK293 cells

were plated in 35 mm glass-bottomed microwell dish (MatTek). On the day of

experiment, the cells were incubated with the peptide solution (5 µM) and 0.5 mg/mL

dextranFITC at 37 °C for 2 h. The cells were gently washed with DPBS twice and imaged

on a Visitech Infinity 3 Hawk 2D-array live cell imaging confocal microscope.

To detect the internalization of pCAP-containing peptides for exocylic delivery

methodology, HEK293 cells were similarly plated and incubated with the peptide

solution (5 µM) at 37 °C for 60 min. After removal of the medium, the cells were gently

washed with DPBS containing sodium pervanadate (1 mM) twice and incubated for 10

min in DPBS containing 5 μM nuclear staining dye DRAQ5. The resulting cells were

washed with DPBS twice and imaged on a spinning disk confocal microscope (UltraView

Vox CSUX1 system).

To monitor GFP internalization, 5 x 104 HEK293 cells were seeded in a 35 mm

glass-bottomed microwell dish and cultured overnight. Cells were treated with CPP-S-S-

GFP (1 µM) at 37 °C for 2 h. After removal of the medium, the cells were incubated in

DPBS containing 5 µM DRAQ5 for 10 min. The cells were washed with DPBS twice and

imaged on a Visitech Infinity 3 Hawk 2D-array live cell imaging confocal microscope.

To detect the uptake of FITC-labeled PTP1B inhibitors, approximately 5 x 104 A549 cells were plated in 35 mm glass-bottomed microwell dish (MatTek). On the day of experiment, the cells were incubated with the peptide solution (5 µM) and 1.0 mg/mL dextranFITC at 37 °C for 2 h. The cells were gently washed with DPBS twice and imaged

on a Visitech Infinity 3 Hawk 2D-array live cell imaging confocal microscope.

3.5.6 Flow Cytometry

To quantify the delivery efficiencies of pCAP-containing peptides for exocyclic

delivery methodology, HeLa cells were cultured in six-well plates (5 x 105 cells per well)

for 24 h. On the day of experiment, the cells were incubated with 10 μM pCAP-

containing peptide in clear DMEM with 1% FBS at 37 °C for 2 h. The cells were washed

with DPBS containing 1 mM sodium pervanadate, detached from plate with 0.25% trypsin, suspended in DPBS containing 1% bovine serum albumin, and analyzed on a BD

FACS Aria flow cytometer with excitation at 355 nm. Data were analyzed with Flowjo software (Tree Star).

To quantify the delivery efficiencies of pCAP-containing peptides for endocyclic delivery methodology, MCF-7 cells were cultured in six-well plates (5 x 105 cells per well) for 24 h. On the day of experiment, the cells were incubated with 50 μΜ pCAP- containing peptide in phenol red-free DMEM and Ham's F12 (1:1) supplemented with 0.5%

FBS. After 15 min, the peptide solution was removed, and the cells were washed with sodium pervanadate containing DPBS, treated with 0.25% trypsin for 5 min, fixed in 2% paraformaldehyde (w/v) at room temperature for 15 min, and stained with LIVE/DEAD

Fixable Far Red Dead Cell Stain Kits (Invitrogen) according to the manufacturer’s instructions. Finally, the cells were resuspended in the flow cytometry buffer and analyzed by flow cytometry, with excitation at 355 nm.

3.5.7 Preparation of cFΦR4–Protein Conjugates

To conjugate CPP and GFP, peptide containing a C-terminal cysteine [CPP-SH, ~10

μmol; Figure 21] was dissolved in 1 mL of degassed PBS and mixed with 2,2’-dipyridyl disulfide (5 equiv) dissolved in acetone (0.5 mL). After 2 h at room temperature, the reaction product CPP-S-S-Py was purified by reversed-phase HPLC. The product was incubated with coenzyme A (2 equiv) in PBS for 2 h. The resulting cFΦR4-S-S-CoA adduct was purified again by reversed-phase HPLC. Green fluorescent protein (GFP) containing an N-terminal ybbR tag (VLDSLEFIASKL) and a C-terminal six-histidine tag was expressed in Escherichia coli and purified as previously described.230 Next, ybbR-

GFP (30 μM), CPP-S-S-CoA (30 μM), and phosphopantetheinyl transferase Sfp (0.5 μM)

were mixed in 50 mM HEPES (pH 7.4), 10 mM MgCl2 (total volume 1.5 mL) and

incubated at 37 °C for 30 min. The labeled protein, CPP-S-S-GFP, was separated from

unreacted CPP-S-S-CoA by passing the reaction mixture through a PD-10 desalting

column.

To conjugate cFΦR4 and PTP1B, ybbr tag linked catalytic domain of PTP1B was

first cloned, expressed, and purified. The gene coding for the catalytic domain of PTP1B

(amino acids 1-321) was amplified by the polymerase chain reaction using PTP1B cDNA

as template and oligonucleotides 5’ ggaattccatatggagatggaaaaggagttcgagcag 3’ and 5’

gggatccgtcgacattgtgtggctccaggattcgtttgg 3’ as primers. The resulting DNA fragment was

digested with endonucleases Nde I and Sal I, and cloned into pET-22b(+)-ybbR.230 This cloning resulted in the addition of an ybbR tag (VLDSLEFIASKL) to the N-terminus of

PTP1B. Expression and purification of the ybbR tagged PTP1B were carried out as

previously described.212

Peptide containing a C-terminal lysine (CPP-Lys, ~10 μmol; Figure 21) was

synthesized on the solid phase, deprotected and released from the support, dissolved in

degassed DPBS (pH 7.4, 1 mL), and mixed with bifunctional linker LC-SMCC (5 equiv)

dissolved in DMSO (0.2 mL). After incubation at room temperature for 2 h, the reaction

product CPP-SMCC (Figure 14) was purified by reversed-phase HPLC equipped with a

C18 column. The product was then mixed with coenzyme A (2 equiv) in DPBS and incubated for 2 h. The resulting CPP-SMCC-CoA adduct was purified again by reversed- phase HPLC. Next, ybbR-tagged PTP1B (30 μM), CPP-SMCC-CoA (30 μM), and phosphopantetheinyl transferase Sfp (0.5 μM) were mixed in 50 mM HEPES (pH 7.4),

10 mM MgCl2 (total volume of 1.5 mL) and incubated at 37 °C for 30 min. The labeled

protein (CPP-PTP1B) was separated from unreacted CPP-SMCC-CoA by passing the

reaction mixture through a PD-10 desalting column.

3.5.8 Immunoblotting

NIH 3T3 cells were cultured in full growth media to reach 80% confluence. The

cells were starved in serum free media for 3 h and treated with different concentrations of

cFΦR4-PTP1B or untagged PTP1B for 2 h, followed by 30 min incubation in media

supplemented with 1 mM sodium pervanadate. The solutions were removed and the cells

were washed with cold DPBS twice. The cells were detached and lysed in 50 mM Tris-

HCl, pH 7.4, 150 mM NaCl, 1% NP-40, 10 mM sodium pyrophosphate, 5 mM iodoacetic

acid, 10 mM sodium fluoride, 1 mM EDTA, 2 mM sodium pervanadate, 0.1 mg/mL

phenylmethanesulfonyl fluoride, 1 mM benzamidine, and 0.1 mg/mL trypsin inhibitor.

After 30 min incubation on ice, the cell lysate was centrifuged at 15,000 rpm for 25 min in a microcentrifuge. The total cellular proteins were separated by SDS-PAGE and transferred electrophoretically to PVDF membrane, which was immunoblotted using anti- phosphotyrosine antibody 4G10.

3.5.9 Staurosporin-Induced Apoptosis Assay

Jurkat cells were re-suspended in serum-free medium at a density of 2 x 105 cells/mL, seeded in 96-well cell culture plate (45 μL/well), and incubated for 1 h.

Apoptosis inhibitor compound (dissolved in DMSO) or DMSO were added to the cells to a final concentration of 25 or 100 μM. The plate was incubated for 30 min and 5 μL of a staurosporine stock solution (10 μM) was added. The final DMSO concentration was 1% for each well. After 24 h of incubation, 5 μL of an Alamar Blue solution was added to each well. After incubation for 2 h, the fluorescence of the reduced dye was measured on a Molecular Devices Spectramax M5 plate reader with excitation and emission wavelengths at 531 and 589 nm, respectively. The percentage of protection against staurosporine-induced apoptosis was calculated as previously described.217 Three independent sets of experiments, each performed in triplicates, were conducted to calculate the average percentage protection.

3.5.10 Serum Stability Assay

The stability tests were carried by modifying a previously reported procedure.160

Diluted human serum (25%) was centrifuged at 15,000 rpm for 10 min, and the supernatant was collected. A peptide stock solution was diluted into the supernatant to a final concentration of 5 µM and incubated at 37 °C. At various time points (0-24 h), 200-

µL aliquots were withdrawn and mixed with 50 µL of 15% (v/v) trichloroacetic acid and incubated at 4 ºC overnight. The final mixture was centrifuged at 15,000 rpm for 10 min in a microcentrifuge, and the supernatant was analyzed by reversed-phase HPLC equipped with a C18 column. The amount of remaining peptide (%) was determined by integrating the area underneath the peptide peak (monitored at 214 nm) and comparing with that of the control reaction (no serum).

3.6 Acknowledgements

Activities of recombinant PTPs towards pCAP-containing peptides were performed by Dr. Nicholas G. Selner and Dr. Rinrada Luechapanichkul (Dr. Dehua Pei’s group at

The Ohio State University). pCAP amino acid was provided from Dr. Amy M. Barrios lab (University of Utah). Immunoblotting and PTP1B inhibitor synthesis were performed by Wenlong Lian (Dr. Dehua Pei’s group at The Ohio State University). Synthesis of bicyclic was performed by Bisheng Jiang (Dr. Dehua Pei’s group at The Ohio State

University).

Chapter 4: Intracellular Delivery of Linear Peptides by Reversible Cyclization of

Cell-Penetrating Peptides

4.1 Introduction

The applicability of linear peptides as drugs is often limited by their susceptibility to

proteolytic cleavage and poor membrane permeability. Cyclization of peptides has

proven highly effective for improving their proteolytic stability.160 Moreover, cyclization

of certain amphipathic peptides (e.g., FΦRRRR) renders them cell permeable through an

active transport mechanism.178,219 As shown in the chapter 3, biologically active cyclic peptides can be delivered into the cytoplasm and nucleus of mammalian cells by incorporating into them these short sequence motifs.178 However, in many circumstances,

binding to a molecular target (e.g., PDZ232,233 and BIR domains234) requires that the

peptidyl ligand exist in its extended conformations (e.g., α-helix and β-strand) and

cyclization may interfere with target binding. Here, we report a potentially general

strategy for delivering linear peptide ligands into mammalian cells through reversible,

disulfide bond-mediated cyclization. When present in the oxidizing extracellular

environment, the peptides exist as macrocycles, which have enhanced stability against

proteolysis and cell permeability. Upon entering the cell (i.e., cytoplasm and/or nucleus),

the disulfide bond is reduced by the intracellular thiols to produce the linear, biologically

active peptides (Figure 27).

Figure 27. Scheme showing the reversible cyclization strategy for delivering linear peptidyl cargos into cells. GSH, glutathione.

4.2 Results

4.2.1 Synthesis and Uptake Studies of Disulfide-Bond Cyclized Peptide

To test the validity of the reversible cyclization strategy, we synthesized peptide N-

3-mercaptopropionyl-FΦRRRRCK-NH2 and cyclized it by forming an intramolecular disulfide bond (Figure 28; Table 8, peptide 1). A linear peptide of the same sequence

(Figure 28C; Table 8 peptide 2) was also synthesized by replacing the N-terminal 3- mercaptopropionyl group with a butyryl group and the C-terminal cysteine with 2- aminobutyric acid (Abu or U). Both peptides were labeled at a C-terminal lysine residue with fluorescein isothiocyanate (FITC) and their cellular uptake was assessed by live-cell confocal microscopy and flow cytometry. HeLa cells treated with the cyclic peptide (5

μM) showed strong, diffuse green fluorescence throughout the entire cell volume, 98 whereas the endocytosis marker, rhodamine-labeled dextran (dextranRho), exhibited only punctate fluorescence in the cytoplasmic region (Figure 29A). The nearly uniform distribution of FITC fluorescence in both cytoplasmic and nuclear regions suggests that the cyclic peptide was efficiently internalized by HeLa cells and like the parent cyclic

3 peptide, cFΦR4, was able to efficiently escape from the endosome. In contrast, cells treated with the linear control peptide showed much weaker intracellular fluorescence under the same imaging condition. Quantitation of the total intracellular fluorescence by fluorescence-activated cell sorting (FACS) gave mean fluorescence intensity (MFI) of

27,100, 5530, and 1200 arbitrary units (AU), for cells treated with the disulfide cyclized peptide, linear peptide, and FITC alone, respectively (Figure 29B).

Table 8. Sequences of Peptides Studied in Chapter 4

Peptide ID Peptide Sequencea

2 3

6 7

10 aAmc, 7-amino-4-methylcourmarin; FITC, fluorescein isothiocyanate; Φ, L-2-naphthylalanine; M, norleucine; U, 2-aminobutyric acid. 99

Figure 28. (A) Synthesis of disulfide-bond cyclized peptide. (B) Synthesis of thioether- bond cyclized peptide. (C) Structures of FITC-labeled peptides 1 and 2. Reagents and conditions: (a) Standard Fmoc/HATU chemistry; (b) piperidine/DMF; (c) 3,3'-dithiodipropionic acid/DIC; (d) β-mercaptoethanol/DMF; (e) modified reagent K; (f) trituration; (g) DMSO/DPBS (pH 7.4). (h) 4-bromobutyric acid/DIC; (i) 1% TFA/DCM; (j) 1%DIPEA/DMF; PG, protecting group. Trt, trityl; Mmt, methoxytrityl. (C) Structures of FITC labeled peptides 1 and 2.

To assess the cytoplasmic/nuclear delivery efficiencies of peptides 1 and 2, we again used the pCAP based assay,178,219 where a negatively-charged pentapeptide, Asp-Glu-

pCAP-Leu-Ile (PCP), was attached to peptides 1 and 2 through a polyethyleneglycol

linker (Figure 30A). pCAP is non-fluorescent but, when delivered into the mammalian

cytoplasm, undergoes rapid dephosphorylation to generate a fluorescent product,

coumaryl aminopropionic acid (CAP).210,211 FACS analysis of HeLa cells treated with 5 100

μM peptide 1-PCP and peptide 2-PCP gave MFI values of 3020 and 700, respectively

(Figure 30B and Figure 30C). Thus, the above results indicate that cyclization of

FΦRRRR through a disulfide bond has a similar effect to the N-to-C cyclization and increases its cellular uptake efficiency by ~5-fold.219

Figure 29. Disulfide-bond cyclization enhance cellular uptake. (A) Live-cell images of HeLa cells treated with 5 μM FITC-labeled peptide 1 (I) or 2 (II), endocytosis marker dextranRho (0.5 mg mL-1), and nuclear stain. (B) Flow cytometry of HeLa cells treated with 5 μM FITC-labeled peptides 1, 2, or FITC alone. Images were recorded in the same Z-section.

101

Figure 30. Evaluation of cytoplasmic delivery efficiencies of 1-PCP and 2-PCP. (A) Structures of pCAP containing peptides 1-PCP and 2-PCP. (B) FACS analysis of HeLa cells treated with 0 or 5 μM peptides 1-PCP, 2-PCP for 2 h. (C) CAP fluorescence from (B) after subtraction of background fluorescence. MFI, mean fluorescence intensity.

Moreover, the disulfide-bond cyclized peptide also showed enhanced resistance against proteolysis. In the peptide proteolytic stability assay,235 peptides were incubated with α-chymotrypsin and trypsin and the remaining peptides were monitored using

102

HPLC overtime. More than half of peptide 1 remain intact after 12 h incubation with proteases, while linear peptide 2 possess half-life around 20 min (Figure 31).

100 1 2 80

40 % Peptide Remain 20

0 0 50 100 150 200 250 300 350 400 Time (min)

Figure 31. Comparison of the proteolytic stability of peptides 1 and 2.

4.2.2 Delivery of Fluorogenic Caspase Substrate

To illustrate the utility of the reversible cyclization strategy, we applied it to deliver specific caspase236 substrates into cells and monitor intracellular caspase activities in real time. Although peptidyl coumarin derivatives have been widely used to detect caspase activities in vitro,237 they are generally not suitable for cell-based applications due to impermeability to the mammalian cell membrane. To generate a cell permeable caspase substrate, we fused a caspase 3/7 substrate, Ac-Asp-Nle-Abu-Asp-Amc238 (Figure 32;

Table 8, peptide 3, where Amc is 7-amino-4-methylcoumarin and Nle is norleucine), with the CPP motif RRRRΦF. The fusion peptide was subsequently cyclized by the addition of a 3-mercaptopropionyl group to its N-terminus, replacement of the C-terminal Abu with a cysteine, and formation of an intramolecullar disulfide bond, to give cyclic peptide 103

4 (Figure 32; Table 8). For comparison, an isosteric but irreversibly cyclized peptide

(Figure 32; Table 8, peptide 5) was synthesized by forming a thioether bond between an

N-terminal bromobutyryl moiety and the C-terminal cysteine (Figure 28B). A linear

control peptide of the same sequence was also prepared as described above (Figure 32;

Table 8, peptide 6). Finally, the caspase 3/7 substrate was conjugated to nonaarginine (R9) to generate a positive control peptide (Figure 32; Table 8, peptide 7). In vitro kinetic analysis revealed that fusion of the caspase 3/7 substrate to RRRRΦF and R9 decreased

its activity by 53% and 72%, respectively, relative to peptide 3 whereas cyclization by thioether formation rendered the peptide completely inactive toward recombinant caspase

3 (Table 9). The activity of peptide 4 toward caspase 3 could not be reliably determined because the caspase assay required a reducing environment, which would cleave the disulfide bond. Given the structural similarity between peptides 4 and 5, we assume that peptide 4 in the cyclic form is also inactive toward caspases, but has similar activity to peptide 6 after reductive cleavage of the disulfide bond.

Table 9. In vitro Activity of Fluorogenic Substrate against Recombinant Caspase-3.

Peptide ID 3 5 6 7

ΔFU/min 159±19 N. D. 74.7±5.5 45.3±6.5

Activity is presented as mean ± standard deviation; N. D. not detected due to low reactivity.

104

Figure 32. Structures of caspase fluorogenic substrates 3-7.

Jurkat cells were pretreated with the kinase inhibitor staurosporin to induce caspase activities and thus apoptosis.239 These cells were then incubated with peptides 3-7 and the

amount of Amc released was monitored at various time points (0-10 h). As expected, the

impermeable caspase substrate (peptide 3) produced little fluorescence increase over the

10-h period (Figure 33). Peptide 4 produced the fastest fluorescence increase, reaching 105

459 fluorescence units (FU), followed by peptides 7 and 6. Peptide 5, which is inactive toward caspase 3, also produced AMC in a time-dependent manner, albeit at a much slower rate (99 FU). We attribute this slow rate of AMC release to hydrolysis by other intracellular proteases and peptidases. Consistent with this interpretation, pretreatment of

240 Jurkat cells with a pancaspase inhibitor Z-VAD(OMe)-CH2F followed by incubation with peptide 4 released AMC at a rate that was essentially identical to that of peptide 5 alone. The simplest explanation of the above observations is that both peptides 4 and 5 entered the cell interior efficiently but only peptide 4 was converted into the linear caspase substrate inside the cells (Figure 33).

Figure 33. Time-dependent release of coumarin product by Jurkat cells. Cell were treated with peptides 3-7 (5 μM) in the absence and presence of 100 μM caspase inhibitor Z-VAD(OMe)-CH2F (FMK).

106

4.2.3 Development of Cell Permeable CAL-PDZ Domain Inhibitor

Many protein-protein interactions (PPIs) are mediated by protein domains binding

short peptides in their extended conformations.241 For example, the PDZ domain is a

common structural domain of 80-90 amino acids found in the signaling proteins of bacteria to man.232,233,242 PDZ domains recognize specific sequences at the C-termini of

their binding partners and the bound peptide ligands are in their extended β-strand

conformation.232,243 It was recently reported that the activity of cystic fibrosis membrane

conductance regulator (CFTR), a chloride ion channel protein mutated in cystic fibrosis

(CF) patients, is negatively regulated by CFTR-associated ligand (CAL) through its PDZ

domain (CAL-PDZ).244 Inhibition of the CFTR/CAL-PDZ interaction was shown to

improve the activity of ∆Phe508-CFTR, the most common form of CFTR mutation,245,246 by reducing its proteasome-mediated degradation.247 Previous library screening,

computer-aided and rational design have identified several peptidyl inhibitors of the

247- CAL-PDZ domain of moderate potencies (KD values in the high nM to low μM range).

249 However, none of the peptide inhibitors were cell permeable, limiting their therapeutic

potential.

Starting with a hexapeptide ligand for the CAL-PDZ domain, WQVTRV248, we designed a disulfide-mediated cyclic peptide by adding the sequence CRRRRF to its N- terminus and replacing the Val at the -3 position with a cysteine (Figure 34A; Table 8 peptide 8). Thus, in peptide 8, the tryptophan residue at the -5 position was designed to serve the dual function of PDZ binding and membrane translocation. To facilitate affinity measurement and quantitation of its cellular uptake, FITC group was added to the N- 107

terminus of peptide 8. FA analysis showed that in the absence of reducing agent, peptide

8 showed no detectable binding to CAL-PDZ domain (Figure 34B). In the presence of 2 mM tris(carboxylethyl)phosphine (TCEP), which is expected to reduce the disulfide bond, peptide 8 bound to the CAL-PDZ domain with a KD value of 489 nM. As expected,

peptide 8 was readily cell permeable; incubation of HeLa cells with 5 μM peptide 8 for 2 h resulted in intense and diffuse fluorescence throughout the entire cell (Figure 34C).

A B 70 60 50 2mM TCEP + 40

FP 2 mM TCEP - 30 20 10

0 1000 2000 3000 4000 CAL-PDZ (nM) C D 50 *

FLU/20sec 20 Δ

I II III 10

0 Control Corr4a Corr4a Corr4a Corr4a + 8 + 9 + 10 Figure 34. Evaluations of cell-permeable CAL-PDZ domain inhibitor. (A) Structure of CAL-PDZ inhibitor 8. (B) Binding of peptide 8 to CAL-PDZ domain in the presence or absence of reducing reagent. (C) Live-cell microscopic images of HeLa cells treated with peptide 8 (5 μM) and DRAQ5 in the same Z-section. I, green fluorescence of internalized peptide 8; II, DRAQ5 nuclear stain; III, overlay of green peptide fluorescence and nuclear stain. (D) CFTR activities in CFBE cells shown as fluorescence increasment rates in SPQ assays in the absence or presense of 20 μM Corr4a for 24 h and 40 μM peptides 8, 9, and 10 for 2 h. Asterisk (*) indicates p < 0.01 relative to control, two-tailed t-test.

108

To demonstrate the biological activity of CAL-PDZ domain inhibitor in a functional

assay, we made use of a halide efflux assay with a halide quenched dye 6-Methoxy-N-(3-

sulfopropyl) quinolinium (SPQ) on bronchial epithelial CFBE cell line, which is

homozygous for the ∆Phe508-CFTR mutation. The SPQ assay was conducted following

established procedure250 with slight modifications and was described in details below.

Briefly, cells stably expressing ΔPhe508-CFTR were first incubated in the absence or presence of 20 μM small molecule corrector Corr-4a251 for 24 h and 40 μM peptide 8 for

2 h. Subsequently, CFTR was activated using 20 μM forskolin/genistein cocktail251 and

the CFTR activity was monitored by measuring the SPQ fluorescence, whose increasing

rate corresponds to CFTR activity. We expected Corr-4a to serve as a chemical

chaperone to improve protein folding and post-translational modification efficiencies,251 while, peptide 8 to prolong the lifetime of CFTR by minimizing its degradation. Indeed, the combination of inhibitor 8 and Corr-4a reproducibly increased the activity of CFTR by 25% as measured by the SPQ assay (Figure 34D), while Corr-4a alone does not produce significant improvements. Additionally, two control peptides were prepared to further support inhibitor 8’s action of mode. Replacement of cysteine residues in inhibitor

8 to Abu residues provides linear peptide 9 (Table 8), which has poorer cell permeablility but similar binding affinity to CAL-PDZ domain. Replacement of free C-terminal carboxylate in inhibitor 8 with an amide group provides cyclic peptide 10 (Table 8), which displays similar cell peremablity but no target binding (Figure 35). Neither peptide

9 nor 10 improved the functions of mutant CFTR in the same halide efflux assay (Figure

34D). Further biological evaluations are being actively pursued to demonstrate the

109 effectivity of cell permeable CAL-PDZ domain inhibitor as potential cystic fibrosis therapy.

Figure 35. CAL-PDZ domain binding studies (A) and HeLa cell uptake studies (B) of control peptides 9 and 10.

4.3 Discussion

Disulfide bond is one of the most common post-translational modfications during protein folding and is the second most widely used covalent linkage used by nature in proteins and peptides. The reversible nature of disulfide bond provides an opportunity to control proteins’ structure and dynamic properties. In peptides, this structural element widely exists due to its strutual constraints, which improves, in most cases, peptides’ potency, selectivity, and stability.252-254 Thus disulfide bonds are widely present in

peptide therapeutics, for example: insulin, vasopressin, and oxytocin.257 Notably disulfide bond-cyclized Tat sequence175 was shown to possess higher cellular permeability than the

linear Tat peptide, and two cysteiene rich peptides255,256 have been identified as cell- penetrating peptides.

110

As discussed in the last two chapters, peptide cyclization has been developed into an engineerable tool to improve the cellular uptake of peptides.178,219 The cyclic peptides with proper combination of arginines and hydrophobic residues are able to efficiently internalize through endocytic pathways and escape the endosome.178 As demonstrated in

Figure 29Figure 30, disulfide bond cyclization also significantly improves the overall uptake and, most importantly, cytoplasmic delivery of short arginine-rich hydrophobic peptides. The absence of punctate fluorescence (Figure 29) suggests effevtive endosomal escape of these molecules. By integrating the disulfide bond cyclizaton and short CPP motif, bioactive linear peptide could be delivered into mammalian cells after minimal sequence alternations as demonstrated in this work. As expected, the cyclization can provide enhanced protolytic resistence to the residues in the cyclic structure (Figure 31).

However, whether the cyclization can protect the residues outside of macrocycle from protease degradation is undetermined.

4.4 Conclusion

Cyclization of peptide ligands has proven highly effective for improving their stability against proteolytic degradation and in some cases their cell permeability.

However, this strategy is not compatible with proteins that recognize peptide ligands in the extended conformations (e.g., β-strand and α-helix). In this work, we developed a general strategy for intracellular delivery of linear peptide ligands, by fusing them with an amphipathic sequence motif (e.g., RRRRΦF, where Φ is L-naphthylalanine) and cyclizing the resulting conjugate through a disulfide bond. The cyclized peptides have 111

enhanced proteolytic stability and membrane permeability; upon entering the

cytoplasm/nucleus of a cell, the disulfide bond is cleaved in the reducing intracellular

environment to release the linear, biologically active peptide. This strategy was applied to generate cell permeable peptides as fluorogenic caspase substrates and inhibitors against the CAL PDZ domain for potential treatment of cystic fibrosis.

4.5 Experimental Section

4.5.1 Materials

Reagents for peptide synthesis were purchased from Advanced ChemTech

(Louisville, KY), NovaBiochem (La Jolla, CA), or Anaspec (San Jose, CA). Rink resin

LS (100-200 mesh, 0.2 mmol/g) was purchased from Advanced ChemTech. DextraneRho, trypsin and α-chymotrypsin were purchased from Sigma-Aldrich (St. Louis, MO). Cell culture media, fetal bovine serum (FBS), penicillin-streptomycin, 0.25% trypsin-EDTA,

and 6-Methoxy-N-(3-sulfopropyl)quinolinium (SPQ) were purchased from Invitrogen

(Carlsbad, CA). Nuclear staining dye DRAQ5TM was purchased from Thermo Scientific

(Rockford, IL). Caspase-3, Human, recombinant protein was purchased from EMD

Chemicals (San Diego, CA).

4.5.2 Peptide Synthesis

Peptides were synthesized on Rink Resin LS (0.2 mmol/g) using standard Fmoc

chemistry. The typical coupling reaction contained 5 equiv of Fmoc-amino acid, 5 equiv

112

of 2-(7-aza-1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate

(HATU) and 10 equiv of diisopropylethylamine (DIPEA) and was allowed to proceed

with mixing for 75 min. The peptides were deprotected and released from the resin by

treatment with 92.5:2.5:2.5:2.5 (v/v) trifluoroacetic acid

(TFA)/water/phenol/triisopropylsilane for 2 h. The peptides were triturated with cold

ethyl ether (3x) and purified by reversed-phase HPLC equipped with a C18 column.

Peptide labeling with fluorescein isothiocyanate (FITC) was performed by dissolving the

purified peptides (~1 mg each) in 300 µL of 1:1:1 DMSO/DMF/150 mM sodium

bicarbonate (pH 8.5) and mixing with 10 µL of FITC in DMSO (100 mg/mL). After 20

min at room temperature, the reaction mixture was subjected to reversed-phase HPLC on a C18 column to isolate the FITC-labeled peptide.

To generate disulfide bond mediated cyclic peptides, the 3,3’-dithiodipropionic acid

(10 equiv) were coupled on the N-terminal using 10 equiv N,N’-Diisopropylcarbodiimide

(DIC) and 0.1 equiv 4-(dimethylamino)pyridine (DMAP) in anhydrous DCM for 2 h after

the removel of the N-terminal Fmoc protection group by treatment with 20% (v/v)

piperidine in DMF. The resin then was incubated in 20% β-mercaptoethanol in DMF for

2 h twice to expose the free thiol. Triturated crude linear peptides were incubated in 5%

DMSO in DPBS buffer (2.7 mM KCl, 1.5 mM KH2PO4, 137 mM NaCl, 9 mM Na2HPO4, pH 7.4) overnight258 followed by trituration and HPLC purification as described above.

To produce thioether mediated cyclic peptides, 4-bromobutyric acid (10 equiv) were

coupled on the N-terminal using 10 equiv DIC and 0.1 equiv DMAP in anhydrous DCM

for 2 h after the removel of the N-terminal Fmoc protection group by treatment with 20%

113

(v/v) piperidine in DMF. 4-methoxytrityl (Mmt) protection group on L-cysteine side

chain was selectively removed using 1% (v/v) TFA in DCM. Thioether formation was

conducted by incubating the resin in 1% DIPEA in DMF under nitrogen protection

overnight.259 The cyclized peptide was then triturated and purified as described above.

Fmoc-Asp(Wang-resin)-AMC (AMC = 7-amino-4-methylcoumarin) from

NovaBiochem were used as solid support to synthesize fluorogenic caspase substrates.

Standard Fmoc chemistry was employed to synthesize the peptide on solid phase. These

peptides were released from the resin by the treatment with 95:2.5:2.5 (v/v)

TFA/phenol/water for 2 h.237

4.5.3 Cell Culture

HeLa cells were maintained in medium consisting of DMEM, 10% FBS and 1%

penicillin/streptomycin. Jurkat cells were maintained in medium consisting of RPMI-

1640, 10% FBS and 1% penicillin/streptomycin. The bronchial epithelial CFBE cell line,

homozygous for the ∆Phe508-CFTR mutation, was maintained in DMEM containing L-

glutamine supplemented with 10% FBS and 1% penicillin/streptomycin. The tissue

culture plates were coated using human fibronectin (1 mg/ml), collagen I bovine (3

mg/ml), and bovine serum albumin (1 mg/ml) Cells were cultured in a humidified

incubator at 37 °C with 5% CO2.

114

4.5.4 Confocal Microscopy

To detect peptide internalization, 1 mL of HeLa cell suspension (5 x 104 cells) was seeded in a 35 mm glass-bottomed microwell dish (MatTek) and cultured overnight.

Cells were gently washed with DPBS twice and treated with FITC labeled peptides (5

µM) and dextranRho (0.5 mg mL-1) in phenol-red free DMEM containing 1% serum for

37 °C for 1 h in the presence of 5% CO2. After removal of the medium, the cells were gently washed with DPBS twice and incubated with 5 µM DRAQ5 in DPBS for 10 min.

The cells were again washed with DPBS twice and imaged on a Visitech Infinity 3 Hawk

2D-array live cell imaging confocal microscope. Images were captured under the same parameters and adjusted under the same setting using MetaMorph (Molecular Devices).

4.5.5 Flow Cytometry

HeLa cells were cultured in six-well plates (5 x 105 cells per well) for 24 h. On the day of experiment, the cells were incubated with 5 μM FITC labeled peptide in clear

DMEM with 1% FBS at 37 °C for 2 h. The cells were washed with DPBS, detached from plate with 0.25% trypsin, diluted into clear DMEM containing 10% FBS, pelleted at 250g for 5 min, washed once with DPBS and resuspended in DPBS containing 1% bovine serum albumin, and analyzed on a BD FACS Aria flow cytometer. Data were analyzed with Flowjo software (Tree Star).

To quantify the delivery efficiencies of pCAP-conjugated peptides, HeLa cells were cultured in six-well plates (5 x 105 cells per well) for 24 h. On the day of experiment, the cells were incubated with 5 μM pCAP-containing peptide in clear DMEM with 1% FBS 115

at 37 °C for 2 h. The cells were washed with DPBS containing 1 mM sodium pervanadate,

detached from plate with 0.25% trypsin, suspended in DPBS containing 1% bovine serum

albumin, and analyzed on a BD FACS Aria flow cytometer with excitation at 355 nm.

4.5.6 Peptide Stability Assay

The stability tests were carried by slightly modifying a previously reported

procedure.235 24 µL 1.5 mM peptide solution were incubated at 37 °C with 30 µL 50 µM

of α-chymotrypsin and 30 µL 50 µM of trypsin in 200 µL of working buffer (50 mM

Tris-HCl, pH 8.0, 100 mM NaCl, 10 mM CaCl2). At various time points (0-12 h), 40 µL

aliquots were withdrawn and mixed with 40 µL of 15% trichloroacetic acid and incubated

at 4 ºC overnight. The final mixture was centrifuged at 15,000 rpm for 10 min in a

microcentrifuge, and the supernatant was analyzed by reversed-phase HPLC equipped

with a C18 column (Waters). The amount of remaining peptide (%) was determined by integrating the area underneath the peptide peak (monitored at 214 nm) and compared

with that of control reaction (no proteases).

4.5.7 In Cellulo Fluorimetric Assay

One hundred µL Jurkat cell suspension (5 x 105 cells/mL) were seeded in 96-well

plate one hour prior to the experiment. Ten µL of staurosporine stock solution (10 µM) was added into half of the wells to induce apoptosis, while 10 µL of media were added to the other wells. After 1 h incubation, caspase-3 fluorogenic substrates were added to the cells to a final concentration of 5 µM. The fluorescence of the released coumarin was

116 measured on the Spectramax M5 plate reader using an excitation and emission wavelengths 360 and 440 nm, respectively, at various times points (0-6 h). The fluorescence unit (FU) increases between induced and uninduced cells were plotted against the time to present caspase-3 activities measured using various fluorogenic substrates in living cell in real-time. Three independent sets of experiments, each performed in triplicates, were conducted.

4.5.8 In Vitro Fluorimetric Assay

Half µL (100 U/µL) caspase-3 enzyme was first incubated with 90 µL of reaction buffer (50 mM HEPES, pH 7.4, 100 mM NaCl, 10 mM DTT) for 30 min in 96-well plate.

Fluorogenic substrates (10 µL, 100 µM) were mixed into the above solutions to start the reactions, and the plate was measured on a Spectramax M5 plate reader (Ex = 360 nm,

Em = 440 nm) (Molecular Devices). Fluorescence units (FU) increase at one-minute intervals was correlated to the release of Amc due to protease activity. Calculate the

ΔFU/min from the linear portion of the reaction curve. Reported values are averages of three trials with the standard deviation indicated.

4.5.9 Fluorescence Polorization

The full fluorescence polorization (FP) titration experiment was performed by incubating 100 nM fluorophore-labeled peptidyl ligands with varying concentrations (0 -

6 μM) of CAL-PDZ260 in FP buffer (20 mM HEPES, pH 7.4, 150 mM NaCl, 5 mM glutathione, 0.1% (w/v) bovine serum albumin) for 2 h at room temperature. The FP

117

values were measured on a Molecular Devices Spectramax M5 spectrofluorimeter, with

excitation and emission wavelengths at 485 nm and 525 nm, respectively. Equilibrium

dissociation constants (KD) were determined by plotting the fluorescence anisotropy

values as a function of CAL-PDZ concentration. The titration curves were fitted to the

following equation, which assumes a 1:1 binding stoichiometry

++) − ++) −4) + × − 2 = ++ ) − ++ ) −4) 1+ −1 2 where Y is the measured FP value at a given CAL-PDZ concentration x; L is the peptide concentration; Qb/Qf is the correction fact for dye-protein interaction; Amax is the maximum anisotropy when all the peptides are bound to CAP-PDZ, while Amin is the minimum anisotropy when all the peptides are free.

4.5.10 SPQ Intracellular Chloride Concentration Assay

SPQ [6-Methoxy-N-(3-sulfopropyl)quinolinium] assay was utilized to estimate the

transport activity of ΔPhe508-CFTR activity in CFBE cells, as fluorescence of SPQ is

negatively correlated with increasing concentration of intracellular chloride.250,251 CFBE

cells were grown on 96-well plate, which was pre-coated with 1 mg/ml human

fibronectin, 3 mg/ml collagen I bovine, and 1 mg/ml bovine serum albumin, using

DMEM media supplemented with L-glutamine and 10% FBS. Cells were firstly treated

in the presence or absence of 20 μM CFTR corrector Corr-4a for 24 h and 40 μM CAL-

118

PDZ domain inhibitors for 1 h. On the day of the experiment, cells were loaded with SPQ using hypotonic shock at 37°C for 15 min with 10 mM SPQ containing 1:1 (v/v) Opti-

MEM/water solution. Subsequently, the cells were washed and incubated twice for 10 min with fluorescence quenching NaI buffer [130 mM NaI, 5 mM KNO3, 2.5 mM

Ca(NO3)2, 2.5 mM Mg(NO3)2, 10 mM D-glucose, 10 mM N-(2-hydroxyethyl) piperazine-

N′-(2-ethanesulfonic) acid (HEPES, pH 7.4)]. Subsequently, cells were switched to a dequenching isotonic NaNO3 buffer (identical to NaI buffer except that 130 mM NaI was replaced with 130 mM NaNO3) with CFTR activation cocktail (20 μM forskolin and 20

μM genistein). Fluorescence non-specific to CFTR-mediated iodide efflux was measured by incubating the cells with the activation cocktail and the CFTR-specific inhibitor

GlyH101 (10 μM). The effects of CAL-PDZ inhibitors were evaluated by the fluorescence-increasing rate above the basal level. The fluorescence of dequenched SPQ was measured using the plate reader VICTOR X3 (Perkin Elmer) with excitation wavelength at 350 nm and DAPI emission filter. The data was presented as mean ± standard deviation from at least three individual experiments.

4.6 Acknowledgements

CAL-PDZ protein was cloned and purified by Dr. Jeanine F. Amacher (from Dr.

Dean R. Madden’s group at Dartmouth College). SPQ assays were performed by Dr.

Estelle Cormet-Boyaka (The Ohio State University).

119

Chapter 5: Structure-Based Optimization of a Peptidyl Inhibitor against

Calcineurin-NFAT Interaction

5.1 Introduction

Calcineurin (CN) is a protein serine/threonine phosphatase involved in T cell

signaling. Engagement of T cell-surface receptors with ligands (e.g., an antigen-

presenting cell) causes an increase in the cytoplasmic level of calcium, which activates

many calmodulin (CaM)-dependent enzymes including CN. CN dephosphorylates

multiple phosphoserines on nuclear factor of activated T cell (NFAT), a transcription

factor, leading to its nuclear translocation and activation.261,262 The activated NFAT up-

regulates the expression of interleukin 2 (IL-2), which in turn activates T-helper

lymphocytes, induces the production of other cytokines, and stimulates the immune

response. CN is the target of several naturally occurring macrocycles such as

cyclosporine A (CsA) and FK506. These compounds bind to cellular proteins cyclophilin and FKBP12, respectively, and the resulting binary complexes bind to CN and sterically block the access of NFAT and other protein substrates to the CN active site.168 CsA and

FK506 are clinically used as immunosuppressants in postallogenic organ transplant.263

Nevertheless, treatment with these drugs is associated with severe side effects including 120

nephrotoxicity and hepatotoxicity,264 likely because of their indiscriminate inhibition of

CN activity toward all substrates.265-267 Inhibitors that selectively block the CN-NFAT

interaction while affecting only a subset of other CN substrates would provide less toxic

immunosuppressants.

Previous structural and functional analysis of the CN-NFAT interface has identified

a conserved sequence motif among NFAT proteins, PxIxIT (where x is any amino acid), which specifically interacts with a substrate-docking site on CN.268 This interaction is

critical for dephosphorylation of NFAT and a subset of other CN substrates.269-271

Screening of an oriented peptide library identified a tetradecapeptide,

GPHPVIVITGPHEE (VIVIT, Table 10) which binds to the docking site on CN with 25- fold higher affinity than the naturally occurring PxIxIT motif.65 Expression of peptide

VIVIT in mammalian cells effectively blocks the CN-NFAT interaction and its

downstream signaling without directly blocking CN enzymatic activity. Attachment to a

cell-penetrating peptide (R11) renders the peptide cell permeable and active for immunosuppression in transplanted mice.66 This observation has inspired investigators to

develop both peptides and small molecules as selective CN inhibitors.272 However, the

compounds reported to date are only weakly to moderately active in disrupting the CN-

NFAT interaction and require further optimization prior to any potential in vivo

applications. In this work, we used the structural information derived from previous

NMR and X-ray studies273-275 as a guide and carried out a structure-based optimization of

the VIVIT peptide, which led to ~200-fold improvement in the binding affinity and a

highly potent and selective inhibitor against CN (KD = 2.6 nM).

121

Table 10. Sequences and Dissociation Constants of Peptidyl Ligands

a b Peptide Sequence KD (nM)

VIVIT GPHPVIVITGPHEE 477 ± 26

ZIZIT GPHPZIZITGPHEE 43 ± 12

ZIZIT-cisPro GPHPZIZITG-Cys(ΨMe,MePro)-HEE 2.6 ± 0.8

VAVAA GPHAVAVAAGPHEE >20,000

a Me,Me b Z, tert-leucine; Cys(Ψ Pro), 2,2-dimethylthiazolidine. KD values against CN were obtained from fluorescence anisotropy assay using N-terminal 5(6)-SFX labeled peptides.

5.2 Results and Discussion

5.2.1 Substitution of tert-Leucine (Tle) for Valine

The structure of the CN-VIVIT complex273,274 reveals that the PVIVIT core is in an

extended conformation and engages in hydrophobic, van der Waals, and hydrogen

bonding interactions with CN. The side-chains of three highly conserved residues, Pro4,

Ile6 and Ile8, fit snugly into three well-defined hydrophobic pockets, while the side

chains of Val5 and Val7 are largely solvent exposed (Figure 36A). The PVIVIT core also

forms multiple hydrogen bonds between its backbone amides and CN β-strand 14

residues.273,275 We postulated that substitution of Tle for Val5 and Val7 of the peptide

ligand might improve its the potency and/or bioavailability, based on several

considerations. First, the Val5 and Val7 side chains are distant from the hydrophobic

surface formed by the side chain of CN Val328 for optimal van der Waals interaction.

Replacement of the valines with bulkier Tle should result in closer packing between

122

Tle5/Tle7 and Val328 side chains and improved van der Waals interactions between

them. Second, Tle is frequently used as building blocks for peptidomimetic drugs276,277 and organocatalysts,278 because incorporation of Tle can substantially improve the target-

binding affinity and/or protease resistance/bioavailability.279,280 It has been speculated

that the bulky t-butyl group may interfere with solvation of the adjacent peptide bonds

and therefore decrease the amount of desolvation energy associated with target binding

and membrane transport. We therefore replaced both Val5 and Val7 with Tle and named

the resulting peptide “ZIZIT” (where Z = Tle). Peptide ZIZIT was synthesized using

standard solid-phase peptide synthesis chemistry and 2-(7-aza-1H-benzotriazole-1-yl)-

1,1,3,3-tetramethyluronium hexafluorophosphate (HATU) as the coupling reagent. To

our delight, peptide ZIZIT bound to CN with a 10-fold higher affinity than VIVIT (KD values of 43 ± 12 and 477 ± 26 nM, respectively) (Table 10 and Figure 37).

5.2.2 Incorporation of of Cys(ΨMe,MePro) as cis-Pro Analog

The structure of the CN-VIVIT complex273,274 contained a cis peptide bond between

Gly10 and Pro11 of VIVIT (Figure 36A). The β-turn structure permits the formation of an intricate hydrogen bond network among the side chains of Asn330 (of CN) and His12 and Thr8 of the VIVIT peptide.273 Since the trans-configuration of a peptidyl-prolyl

peptide bond is energetically more stable,281 we envisioned that preorganization of the

Gly10-Pro11 peptide bond into the cis-configuration should increase the binding affinity.

2,2-Dimethylthiazolidine [Cys(ΨMe,MePro)] has previously been used as a proline analog;

when it is incorporated into a peptide, the preceding peptide bond is sterically locked into

123 the cis-configuration.282,283 We thus designed peptide ZIZIT-cisPro (Table 10 and Figure

36B) by replacing Pro11 of ZIZIT with Cys(ΨMe,MePro).

Figure 36. Binding modes of VIVIT and ZIZIT-cisPro to Calcineurin. (A) X-ray crystal structure of the CN-VIVIT complex (the image was generated from pdb file: 2P6B273 with Chimera). (B) Binding mode of peptide ZIZIT-cisPro to CN as derived from MD simulations. CN is displayed as the van der Waals surface with the binding surface shaded solid pink and the rest colored gray. Peptide ligands are shown as sticks with carbon, nitrogen, oxygen, and sulfur atoms in yellow, blue, red and green, respectively. Key ligand residues are labeled in black three-letter codes, while CN 124

residues are labeled in blue sing-letter codes. Tle, tert-leucine. (C) A close-up view of the hydrogen-bond network between CN (in cyan) and ligand residues adjacent to the cis- proline analog (in yellow). (D) van der Waals surface contours of Val328 of CN and Tle5 and Tle7 of the peptide ligand.

0.20 0.18 0.16 0.14 0.12 FA 0.10 VIVIT 0.08 ZIZIT 0.06 ZIZIT-cisPro 0.04 0 500 1000 1500 2000 [CN] (nM)

Figure 37. Binding assay of VIVIT, ZIZIT, and ZIZIT-cisPro to CN.

Synthesis of peptide ZIZIT-cisPro is outlined in Figure 38. The sterically hindered secondary amine of 2,2-dimethyl-1,3-thiazolidine-4-carboxylic acid [H-Cys(ΨMe,MePro)-

OH] is poorly reactive and cannot be directly incorporated into peptides through solid-

phase synthesis. Thus, the pseudo-proline was first prepared as the Fmoc-protected

dipeptide, which was readily introduced into peptides using benzotriazol-1-yl- oxytripyrrolidinophosphonium hexafluorophosphate (PyBop) as the coupling reagent.284

The dipeptide, Fmoc-Gly-Cys(ΨMe,MePro)-OH, was prepared in 69% yield by condensing

Fmoc-protected glycyl fluoride and H-Cys(ΨMe,MePro)-OH.284 Because Cys(ΨMe,MePro) is unstable under strongly acidic conditions (e.g. 95% TFA), acid-labile side-chain protecting groups 4-methyltrityl (Mmt), 2-phenylisopropyl (PhiPr), and trityl (Trt) were employed for His, Glu, and Thr residues, respectively. After the fully protected peptide 125

was synthesized on solid phase, these side-chain protecting groups were removed by

treatment with a mildly acidic condition (1% TFA, 5% triisopropylsilane in DCM, 2 h),

which did not significantly damage the Cys(ΨMe,MePro) moiety. The deprotected peptide was released from the solid support by aminolysis with 1:1 (v/v) propylamine/DMF and purified to near homogeneity by reversed-phase HPLC.

Figure 38. Synthesis of peptide ZIZIT-cisPro. Reagents and Conditions: (a) Fmoc-Gly-F/DIPEA; (b) 4-hydroxymethylbenzoic acid (HMBA)/HATU; (c) Fmoc-Gly-OH/DIC; (d) solid-phase Fmoc/HATU chemistry; (e) Fmoc-Gly-Cys(ΨMe,MePro)-OH/PyBop; (f) 1% TFA; and (g) propylamine/DMF.

The binding affinity of ZIZIT-cisPro for CN was determined by fluorescence

anisotropy (FA) assay. Incorporation of the cis-proline analog further increased the

binding affinity of ZIZIT for CN by ~20-fold, producing a highly potent peptidyl

inhibitor against CN (KD = 2.6 nM, Figure 37). The magnitude of affinity improvement is 126

consistent with increasing the cis peptidyl-prolyl bond population from its normal abundance (5-10%) to ~100% in ZIZIT-cisPro.282-284 To demonstrate the improved

ligands still bind to calcineurin at same site as original VIVIT peptide. Competitive

fluorescence anisotropy assays were performed. A solution of 100 nM FITC-labeled

ZIZIT peptide and 150 nM CN protein was prepared at first. The aliquots of the solution

was titrated with increasing concentrations (0-20 μM) of unlabeled competitors. The

decreasing FA values were plotted to competitor concentrations as dose-response curves.

These independent series of titration yielded half maximal inhibitory concentrations (IC50) values for VIVIT, ZIZIT, and ZIZIT-cisPro of 4102 ± 101, 275 ± 93, and 106 ± 90 nM respectively (Figure 39). In addition, the relative stabilities of VIVIT and ZIZIT-cisPro against proteolytic degradation were determined by incubating the peptides in 25% human serum at 37 oC and following the disappearance of the full-length peptides by

HPLC. Over 50% of unnatural amino acid containing peptide, ZIZIT-cisPro, remained

intact after 6 h. In contrast, VIVIT was degraded at a half-life of about 1 h and was almost completely digested after 3 h (Figure 40).

127

Figure 39. Competitions for binding to CN by inhibitors VIVIT, ZIZIT, and ZIZIT-cisPro. Each competitive fluorescence anisotropy assay contained 100 nM FITC-labeled ZIZIT, 150 nM CN, and 0-20 μM unlabeled competitors VIVIT, ZIZIT, and ZIZIT-cisPro. Three parallel assays were conducted and fluorescence anisotropy values were measured and plotted against the competitor concentration.

100 ZIZIT-cisPro

80 VIVIT

40 % Peptide Remain 20

0 01234567 Time (h)

Figure 40. Comparison of serum stability of VIVIT and ZIZIT-cisPro.

128

5.2.3 Molecular Modeling

To gain mechanistic insight into the observed affinity enhancement, we carried out molecular dynamic (MD) simulations on the CN-ZIZIT-cisPro complex. To provide some information about how the new ligand would interact with the CN surface, we proceeded with a molecular docking study; we began by using the available crystal structure of the CN-VIVIT complex (pdbID: 2p6b)273 and replacing the ligand with

ZIZIT-cisPro. Following the construction of ZIZIT-cisPro ligand and energy minimization as detailed below, MD simulations were performed to obtain the docked conformation285. Analysis of the root mean square deviation (RMSD) between the crystal structure and MD protein showed no deviation indicative of sudden, chaotic structural fluctuations (Figure 41A). Further, ZIZIT-cisPro remained associated with the binding site on the CN surface throughout the simulation, as indicated by the number of hydrogen bonds between the ligand and the protein (Figure 41B).

129

Figure 41. MD simulation of ZIZIT-cisPro with CN. (A) RMSD values generated by the GROMACS program “g_rms” following a least- squares fit of calcineurin bound with ZIZIT-cisPro structure over the 20 ns MD trajectory to the crystal structure of calcineurin (pdbID: 2P6B). (B) Hydrogen bonding interactions between CN and ZIZIT-cisPro, calculated in GROMACS using g_hbond with the standard cutoff distances and angle requirements.

ZIZIT-cisPro adopts a virtually identical conformation to that of VIVIT in the crystal structure (Figure 36). The side chains of Ile6 and Ile8 are clearly accommodated in hydrophobic pockets formed by Met329/Met290/Ile303 and Tyr288/Met290/Ile331, respectively. ZIZIT-cisPro engages in the same set of hydrogen bonds with CN as VIVIT.

The Gly10-Cys(ΨMe,MePro)11 peptide bond is indeed in its cis-configuration, thus

130

permitting the formation of the hydrogen bond network between ligand side chains of

His12 and Thr9 and CN residues Arg332 and Asn330 (Figure 36C). The geminal

dimethyl groups of the proline analog are oriented away from the protein surface and do

not appear to experience any steric clashes with any protein residue. In contrast to the

CN-VIVIT structure, in which Val5 and Val7 side chains are solvent exposed,273 the

additional side chain methyl groups in the CN-ZIZIT-cisPro complex result in close

packing of the Tle5 and Tle7 side chains against the side chain of Val328 (Figure 36D).

In fact, the Tle side chains are ~1 Å closer to the Val328 side chain than those of Val5

and Val7. These results suggest that enhanced van der Waals interactions and/or

hydrophobic effects between the Tle side chains and Val328 contribute significantly to the observed high potency of the ZIZIT-cisPro ligand. We also calculated the solvent

accessible surface area (SASA) of both VIVIT and ZIZIT-cisPro peptides when they are

bound to the CN protein using the trajectories derived from the 20 ns MD simulations.

The calculated SASA (in units of 103 Å2) values for VIVIT (2.12 ± 0.03) and ZIZIT-

cisPro (1.97 ± 0.03) indicate that ZIZIT-cisPro peptide is less solvated than VIVIT

peptide in CN-bound states, providing further support that ZIZIT-cisPro engages in

greater van der Waals and/or hydrophobic interactions with the CN protein than the

parent peptide.

5.2.4 Inhibition of Nuclear Translocation of NFAT

To test whether the increased binding affinity of ZIZIT-cisPro translates into

improved efficacy in cellular assays, we conjugated it to a polybasic cell-penetrating

131

peptide, R11. First, peptide ZIZIT-cisPro was modified at its N-terminus with a bifunctional linker succinimidyl 3-(2-pyridyldithio)propionate (SPDP) (Figure 42). The resulting peptide was conjugated to R11, which was synthesized with a C-terminal

cysteine, via a disulfide exchange reaction. We also prepared R11-VIVIT and a negative

control peptide, R11-VAVAA, which contains replacement of three key CN-binding

residues (Ile6, Ile8, and Thr9) with alanine and has no detectable binding to CN as judged

by FA assay (Table 10). HeLa cells stably transfected with a green fluorescent protein-

NFAT1 fusion (GFP-NFAT)286 were treated with the peptides in the absence and presence of ionomycin and the intracellular distribution of green fluorescence was monitored by live-cell confocal microscopy (Figure 43A).66 In control cells (untreated

with either ionomycin or peptide), GFP-NFAT was localized predominantly in the

cytosol with minimal signal in the nuclear region. Treatment of cells for 10 min with ionomycin, which raises the intracellular Ca2+ concentration and activates CN activity,

caused translocation of GFP-NFAT into the nucleus, as observed by time-lapse live-cell

confocal microscopic imaging. However, incubation of cells with 500 nM R11-ZIZIT- cisPro prior to the treatment with ionomycin almost completely blocked the nuclear translocation of GFP-NFAT (~95% inhibition) (Figure 43B). This potency of peptide

66 R11-ZIZIT-cisPro is similar to that of FK506 in the translocation assay. Under the same

conditions, R11-VIVIT resulted in ~65% inhibition of the nuclear translocation. As

expected, peptide R11-VAVAA had no detectable effect on the ionomycin-stimulated

GFP-NFAT translocation.

132

A SPDP, pH 8.0 buffer O Peptide N S Peptide H2N S N r.t. 4 h H

SH O S Peptide Ac-R 11 -Cys-NH 2 S N H Ac-R Cys NH o/n 11 2 B

Me,Me R11-ZIZIT-cisPr o Ac-RRRRRRRRRRRC-SPDP-GPHP ZIZITG-Cys( Ψ Pro)-HEEG-Propyl

R -VIVIT 11 Ac-RRRRRRRRRRRC-SPDP-GPHPVIVITGPHEEG-NH 2

R -VAVAA 11 Ac-RRRRRRRRRRRC-SPDP-GPHAVAVAAGPHEEG-NH 2

Figure 42. Preparation and sequences of R11 conjugated CN inhibitors.

Figure 43. Inhibition of nuclear translocation of GFP-NFAT by CN inhibitors. (A) Time-lapse live cell confocal microscopic imaging of HeLa cells stably transfected with GFP-NFAT after stimulation with ionomycin and in the absence or presence of different CN inhibitors (500 nM). (B) Relative potencies of the CN inhibitors in blocking the nuclear translocation of GFP-NFAT. The increase in fluorescence intensity in the nuclear region after 10 min of stimulation with ionomycin was measured and compared to that of control cells (untreated with CN inhibitor; 100%). Data reported represent the mean ± SD from at least 30 cells. All CN inhibitors contained R11 on their N-termini.

133

5.3 Discussion

CN is a calcium-activated protein serine/threonine phosphatase, which

dephosphorylates and activates NFAT during immune response. CN inhibitors such as

CsA and FK506 are effective for immunosuppression during organ transplantation and

treatment of other immunological diseases, but produce severe side effects because they

block the total phosphatase activity toward all CN substrates. Previous efforts to develop

more selective CN inhibitors resulted in peptide VIVIT, GPHPVIVITGPHEE, which

binds to CN at the substrate-docking site for NFAT and selectively inhibits the

dephosphorylation of NFAT.65 However, the utility of peptide VIVIT was limited by its moderate potency (KD ~500 nM). By replacing the two valine residues of VIVIT with

tert-leucine and the C-terminal proline with a cis-proline analog, we obtained peptide

ZIZIT-cisPro (where Z is tert-leucine), which binds to CN with a KD value of 2.6 nM.

Molecular dynamics analysis suggests that the tert-leucine side chains engage in

improved van der Waals interactions with Val328 of CN. The tert-leucine residues were

also proposed to decrease solvation to their adjacent amides thus less desolvation penalty

was required during peptide uptake and protein binding. In the cell-based assay, peptide

ZIZIT-cisPro exhibited superior activity over the parent peptide (VIVIT) in inhibiting the

nuclear translocation of NFAT. Peptide ZIZIT-cisPro may be further developed into

therapeutic agents as less toxic therapeutics for cardiovascular, immunological and

inflammatory diseases.

134

5.4 Conclusion

Through relatively minor structural modifications, we were able to improve the CN-

binding affinity of peptide VIVIT by ~200-fold. With a KD value of 2.6 nM, ZIZIT-

cisPro ranks among some of the most potent CN inhibitors reported to date.272 Unlike other potent CN inhibitors such as CsA and FK506, which block the dephosphorylation of all CN substrates, ZIZIT-cisPro only affects the CN activity toward a subset of these substrates. Moreover, ZIZIT-cisPro acts independently of immunophilins and is not expected to cause any toxicity associated with depletion of the peptidyl-prolyl isomerase activity. The steric bulk of Tle and the cis-Pro analog also improve the proteolytic stability of the peptide. Peptide R11-ZIZIT-cisPro could be further developed into an

efficacious but less toxic alternative to FK506 and CsA.

5.5 Experimental Section

5.5.1 Materials

Reagents for peptide synthesis were purchased from NovaBiochem (La Jolla, CA),

Anaspec (San Jose, CA), Peptides International (Louisville, KY), or Chem-Impex

International Inc. (Wood Dale, IL). SPDP was obtained from Thermo Scientific

(Rockford, IL). 5(6)-fluorescein-6(5)-carboxamidohexanoic acid, succinimidyl ester

(5(6)-SFX, F-6129) was from Life Technologies (Carlsbad, CA). Glass-bottom dishes

(35-mm) were purchased from MatTek (Ashland, MA). Cell culture media, fetal bovine serum, 0.25% trypsin-EDTA, Dulbecco’s phosphate-buffered saline (DPBS) (2.67 mM

135

potassium chloride, 1.47 mM potassium phosphate monobasic, 137 mM sodium chloride,

8.06 mM sodium phosphate dibasic.), and other chemical reagents were purchased from

Sigma-Aldrich (St. Louis, MO).

5.5.2 Cell Culture

The generation of HeLa cells stably expressing GFP-NFAT has previously been described.286 The cells were maintained with 10% FBS supplemented Dulbecco’s

modified Eagle medium (DMEM) in a humidified incubator at 37 °C with 5% CO2.

5.5.3 Peptide Synthesis, Labeling and Conjugation

Peptides were synthesized on Rink Resin LS (0.2 mmol/g) using standard Fmoc

chemistry. The typical coupling reaction contained 5 equiv of Fmoc-amino acid, 5 equiv

of HATU and 10 equiv of diisopropylethylamine (DIPEA) and was allowed to proceed

with mixing for 1 h. The peptides were deprotected and released from the resin by

treatment with 92.5:2.5:2.5:2.5 (v/v) TFA/phenol/water/triisopropylsilane for 2 h. The

peptides were triturated with cold ethyl ether and purified by reversed-phase HPLC

equipped with a C18 column. The authenticity of each peptide was confirmed by MALDI-

TOF mass spectrometry. Peptide labeling with 5(6)-SFX was performed by dissolving the

purified peptides (~1 mg) in 300 μL of 1:1 (v/v) DMF/150 mM sodium bicarbonate (pH

8.5) followed by the addition of 10 μL of 5(6)-SFX in DMSO (100 mg/mL). After 1 h reaction, the reaction was quenched and purified by HPLC.

136

The dipeptide Fmoc-Gly-Cys(ΨMe,MePro)-OH was prepared by mixing Fmoc-Gly-F

(420 mg, 1.4 mmol)287 with 1 equiv 2,2-dimethyl-L-thiazolidine-4-carboxylic acid

hydrochloride (277 mg, 1.4 mmol) and 2 equiv of DIPEA (0.49 mL, 2.8 mmol) in

anhydrous DCM (20 mL). After 1 h reaction under argon atmosphere, the mixture was

washed with 20 mL of aqueous solution of 10% (w/v) citric acid, dried, and concentrated

in vacuo. The crude product was purified by silica gel column chromatography to give

Me,Me 1 425 mg of Fmoc-Gly-Cys(Ψ Pro)-OH (69% yield). H NMR (250 MHz, CDCl3): δ

7.76-7.73 (m, 2H), 7.59-7.56 (m, 2H), 7.42-7.26 (m, 4H), 5.76 (br, 1H), 4.77-4.75 (m,

1H), 4.36-4.17 (m, 3H), 4.03-3.94 (m, 2H), 3.38-3.29 (m, 2H), 1.89 (s, 3H), 1.84 (s, 3H).

+ ESI-MS: m/z calculated for C23H24N2O5S 440.14, found 463.13 ([M + Na ]).

The synthesis of Cys(ΨMe,MePro)-containing peptide was similarly performed on

Rink Resin LS (0.2 mmol/g), which was first modified with a 4-hydroxymethylbenzoic acid (HMBA) linker. Coupling of the first residue to the HMBA linker was carried out with 5 equiv of N,N’-diisopropylcarbodiimide (DIC), 5 equiv of Fmoc-amino acid, and 5 equiv of hydroxybenzotriazole (HOBt) for 3 h. Fmoc-Gly-Cys(ΨMe,MePro)-OH was

incorporated into the peptide using 2 equiv of the Fmoc-dipeptide, 2 equiv of PyBop, and

2 equiv of HOBt. Fmoc-His(Mmt)-OH, Fmoc-Thr(Trt)-OH, and Fmoc-Glu(O-2-PhiPr)-

OH were coupled to the growing peptide chain using standard Fmoc chemistry. After the peptide synthesis was complete, the resin was treated with 1% TFA and 5% triisopropylsilane in DCM for 2 h. The peptide was then released from the solid support with 1:1 (v/v) propylamine/DMF for 3 h, purified, and labeled as described above.

137

To conjugate a peptide to R11, the peptide containing a free N-terminal amine (~10

μmol) was dissolved in 200 μL of 50 mM phosphate buffer (pH 8.0) and mixed with 1

equiv of SPDP dissolved in 100 μL of DMF. After incubation for 4 h at room

temperature, 1 equiv of Ac-R11-Cys-NH2 (dissolved in water) was added to the mixture.

After incubation for an additional 12 h, the crude product was purified by reversed-phase

HPLC and the peptide identity was confirmed by MALDI-TOF MS analysis.

5.5.4 Fluorescence Anisotropy

The protein for FA experiment was purified as glutathione S-transferase fusion protein from E. coli BL21 cells and purified on a glutathione-Sepharose column as previously described.288 FA experiments were performed by incubating 100 nM

fluorescein-labeled peptide with varying concentrations of CN in 20 mM HEPES (pH

7.4), 150 mM NaCl, 2 mM Mg(OAc)2, and 0.1% bovine serum albumin for 2 h at room

temperature. The FA values were measured on a Molecular Devices Spectramax M5

spectrofluorimeter, with excitation and emission wavelengths at 485 and 525 nm,

respectively. Dissociation constants (KD) were determined by plotting the FA values as a

function of CN concentrations. The titration curves were fitted to the following equation

++) − ++) −4) + × − 2 = ++ ) − ++ ) −4) 1+ −1 2

138

where Y is the FA value at a given concentration x of CN; L is the peptide concentration;

Qb/Qf is the correction factor for fluorophore-protein interaction; Amax is the maximum

FA value when all the peptide are bound to CN; while Amin is the minimum FA value when all of the peptides are free.

5.5.5 Proteolytic Stability Assay

The stability tests were carried by modifying a previously reported procedure.160

Diluted human serum (25%) was centrifuged at 15,000 rpm for 10 min, and the supernatant was collected. A peptide stock solution was diluted into the supernatant to a final concentration of 50 µM of peptides VIVIT and ZIZIT-cisPro and incubated at 37

°C. At various time points (0-6 h), 200-µL aliquots were withdrawn and mixed with 50

µL of 15% trichloroacetic acid and incubated at 4 ºC overnight. The final mixture was centrifuged at 15,000 rpm for 10 min in a microcentrifuge, and the supernatant was analyzed by reversed-phase HPLC equipped with a C18 column (Waters). The amount of

remaining peptide (%) was determined by integrating the area underneath the peptide

peak (monitored at 214 nm) and compared with that of the control reaction (no serum).

5.5.6 Molecular Modeling

Simulation was first prepared by editing the crystal structure of VIVIT/CN (pdbID:

2p6b) in UCSF Chimera software suite289 to include ZIZIT-cisPro structural

modifications. As the GROMACS implemented ffAMBER03290,291 force field did not include the proper sulfur topology, it was generated using ANTECHAMBER292,293 from

139

AmberTools13 via acpype.294 Hydrogen types, C–H bond lengths, bond length, bond

angle (), GROMACS bond angle force (), dihedral partners (, , , ), phase angle ()

and GROMACS dihedral force constant ( ) were calculated and imported from the

topology generated by ANTECHAMBER/GAFF.295 All charges and protonation sites

were calculated at pH 7.4. A complete topology and the requisite GROMACS files for

the protein-ligand system were generated using pdb2gmx with parameters specified to

use the AMBER03 force field. The system was placed within a dodecahedral periodic

box, each dimension 9.091 nm long, filled with copies of 216 equilibrated TIP3P296 water molecules and sufficient counter-ions for system neutralization. The system underwent two steps of energy minimization using steepest decent, first with only solvent relaxing, then the entire system followed by a heating process for which the temperature is increased from 0 to 300K. The system then underwent two more phases of equilibration via MD, first using NVT (isothermal) conditions for 200 ps and then NPT (isothermal- isobaric with pressure control via the Parrinello-Rahman algorithm) conditions for 200 ps. The production MD simulations were performed under NPT conditions for 20 ns, and the final trajectory was used to determine the bound ligand conformation, hydrogen bond patterns, average distance, and RMSD. The solvent accessible surface area (SASA) of

VIVIT and ZIZIT-cisPro peptides was calculated using the vmd sasa plugin297 and averaged over the 20 ns CN-bound MD simulation; the resulting value is reported as the

SASA value.

140

5.5.7 GFP-NFAT Translocation Assay

HeLa cells (7 × 104) that stably transfected with GFP-NFAT were seeded in 35-mm

glass-bottom dishes. On the day of the experiment, the cells were first incubated for 2 h

with 500 nM of R11-ZIZIT-cisPro, R11-VIVIT, or R11-VAVAA in the presence of full

growth media. Afterwards, the cells were incubated with 1 μM ionomycin containing phenol-red free DMEM supplemented with 2% FBS. Time-lapse live-cell confocal microscopic imaging of cells was performed using Visitech Infinity 3 HAWK confocal microscopy with 60× oil objective. The same imaging parameters were used for nuclear fluorescence intensity quantifications (MetaMorph).

5.6 Acknowledgements

The molecular modeling was performed by Patrick G. Dougherty (Dr. Dehua Pei’s group at The Ohio State University) and Dr. Shameema Oottikkal (Dr. Christopher

Hadad’s group at The Ohio State University). GFP-NFAT stably transfected HeLa cell

line was provided by Dr. Patrick G. Hogan group (La Jolla Institute for Allergy and

Immunology).

141

REFERENCE

1. Frankel, A. D., and Pabo, C. O. (1988) Cellular uptake of the tat protein from human immunodeficiency virus. Cell 55, 1189-1193. 2. Green, M., and Loewenstein, P. M. (1988) Autonomous functional domains of chemically synthesized human immunodeficiency virus tat trans-activator protein. Cell 55, 1179-1188. 3. Derossi, D., Joliot, A. H., Chassaing, G., and Prochiantz, A. (1994) The 3rd helix of the Antennapedia homeodomain translocates through biological-membranes. J. Biol. Chem. 269, 10444-10450. 4. Langel, U. (2011) Cell-penetrating peptides: methods and protocols, Humana Press, New York, p xv. 5. Deshayes, S., Morris, M. C., Divita, G., and Heitz, F. (2005) Cell-penetrating peptides: tools for intracellular delivery of therapeutics. Cell. Mol. Life Sci. 62, 1839-1849. 6. Futaki, S. (2005) Membrane-permeable arginine-rich peptides and the translocation mechanisms. Adv. Drug Delivery Rev. 57, 547-558. 7. Stewart, K. M., Horton, K. L., and Kelley, S. O. (2008) Cell-penetrating peptides as delivery vehicles for biology and medicine. Org. Biomol. Chem. 6, 2242-2255. 8. Bechara, C., and Sagan, S. (2013) Cell-penetrating peptides: 20 years later, where do we stand? FEBS Lett. 587, 1693-1702. 9. Goun, E. A., Pillow, T. H., Jones, L. R., Rothbard, J. B., and Wender. P. A. (2005) Molecular transporters: Synthesis of oligoguanidinium transporters and their application to drug delivery and real-time imaging. ChemBioChem 7, 1497-1515. 10. Rothbard, J. B., Garlington, S., Lin, Q., Kirschberg, T., Kreider, E., McGrane, P. L., Wender, P. A., and Khavari, P. A. (2000) Conjugation of arginine oligomers to cyclosporin A facilitates topical delivery and inhibition of inflammation. Nat. Med. 6, 1253-1257. 11. Nori, A., Jensen, K. D., Tijerina, M., Kopeckova, P., and Kopecek, J. (2003) Tat- conjugated synthetic macromolecules facilitate cytoplasmic drug delivery to human ovarian carcinoma cells. Bioconjugate Chem. 14, 44-50. 12. Hoyer, J., and Neundorf, I. (2012) Peptide vectors for the nonviral delivery of nucleic acids. Acc. Chem. Res. 45, 1048-1056. 13. Eguchi, A., Akuta, T., Okuyama, H., Senda, T., Yokoi, H., Inokuchi, H., Fujita, S., Hayakawa, T., Takeda, K., Hasegawa, M., and Nakanishi, M. (2001) Protein transduction domain of HIV-1 Tat protein promotes efficient delivery of DNA into mammalian cells. J. Biol. Chem. 276, 26204-26210. 142

14. Nakase, I., Akita, H., Kogure, K., Graslund, A., Langel, U., Harashima, H., and Futaki, S. (2012) Efficient Intracellular Delivery of Nucleic Acid Pharmaceuticals Using Cell-Penetrating Peptides. Acc. Chem. Res. 45, 1132-1139. 15. Andaloussi, S. E., Lehto, T., Mager, I., Rosenthal-Aizman, K., Oprea, II, Simonson, O. E., Sork, H., Ezzat, K., Copolovici, D. M., Kurrikoff, K., Viola, J. R., Zaghloul, E. M., Sillard, R., Johansson, H. J., Said Hassane, F., Guterstam, P., Suhorutsenko, J., Moreno, P. M., Oskolkov, N., Halldin, J., Tedebark, U., Metspalu, A., Lebleu, B., Lehtio, J., Smith, C. I., and Langel, U. (2011) Design of a peptide-based vector, PepFect6, for efficient delivery of siRNA in cell culture and systemically in vivo. Nucleic Acids Res. 39, 3972-3987. 16. Jeong, J. H., Mok, H., Oh, Y. K., and Park, T. G. (2009) siRNA conjugate delivery systems. Bioconjugate Chem. 20, 5-14. 17. Muratovska, A., and Eccles, M. R. (2004) Conjugate for efficient delivery of short interfering RNA (siRNA) into mammalian cells. FEBS Lett. 558, 63-68. 18. Wadia, J. S., and Dowdy, S. F. (2005) Transmembrane delivery of protein and peptide drugs by TAT-mediated transduction in the treatment of cancer. Adv. Drug Delivery Rev. 57, 579-596. 19. Pooga, M., Kut, C., Kihlmark, M., Hallbrink, M., Fernaeus, S., Raid, R., Land, T., Hallberg, E., Bartfai, T., and Langel, U. (2001) Cellular translocation of proteins by transportan. FASEB J. 15, 1451-1453. 20. Fawell, S., Seery, J., Daikh, Y., Moore, C., Chen, L. L., Pepinsky, B., and Barsoum, J. (1994) Tat-mediated delivery of heterologous proteins into cells. Proc. Natl. Acad. Sci. U. S. A. 91, 664-668. 21. Josephson, L., Tung, C. H., Moore, A., and Weissleder, R. (1999) High-efficiency intracellular magnetic labeling with novel superparamagnetic-Tat peptide conjugates. Bioconjugate Chem. 10, 186-191. 22. Gupta, B., Levchenko, T. S., and Torchilin, V. P. (2005) Intracellular delivery of large molecules and small particles by cell-penetrating proteins and peptides. Adv. Drug Delivery Rev. 57, 637-651. 23. Liu, J., Zhang, Q., Remsen, E. E., and Wooley, K. L. (2001) Nanostructured materials designed for cell binding and transduction. Biomacromolecules 2, 362- 368. 24. Saar, K., Lindgren, M., Hansen, M., Eiriksdottir, E., Jiang, Y., Rosenthal-Aizman, K., Sassian, M., and Langel, U. (2005) Cell-penetrating peptides: a comparative membrane toxicity study. Anal. Biochem. 345, 55-65. 25. Suhorutsenko, J., Oskolkov, N., Arukuusk, P., Kurrikoff, K., Eriste, E., Copolovici, D. M., and Langel, U. (2011) Cell-penetrating peptides, PepFects, show no evidence of toxicity and immunogenicity in vitro and in vivo. Bioconjugate Chem. 22, 2255-2262. 26. Wender, P. A., Mitchell, D. J., Pattabiraman, K., Pelkey, E. T., Steinman, L., and Rothbard, J. B. (2000) The desgin, synthesis, and evaluation of molecules that enable or enhance cellular uptake: peptoid molecular transporters. Proc. Natl. Acad. Sci. U. S. A. 97, 13003-13008.

143

27. Pooga, M., Hallbrink, M., Zorko, M., and Langel, U. (1998) Cell penetration by transportan. FASEB J. 15, 1451-1453. 28. Morris, M. C., Depollier, J., Mery, J., Heitz, F., and Divita, G. (2001) A peptide carrier for the delivery of biologically active proteins into mammalian cells. Nat. Biotechnol. 19, 1173-1176. 29. Sheldon, J., Liu, D., Ferguson, J., and Gariepy, J. (1995) Loligomers: design of de novo peptide-based intracellular vehicles. Proc. Natl. Acad. Sci. U. S. A. 92, 2056- 2060. 30. Wang, F., Wang, Y., Zhang, X., Zhang, W., Guo, S., and Jin, F. (2014) Recent progress of cell-penetrating peptides as new carries for intracellular cargo delivery. J. Controlled Release 174, 126-136. 31. Ziegler, A., Blatter, X., Seelig, A., and Seelig, J. (2003) Protein transduction domains of HIV-1 and SIV TAT interact with charged lipid vesicles. Binding mechanism and thermodynamic analysis. Biochemistry 42, 9185-9194. 32. Ziegler, A. (2008) Thermodynamic studies and binding mechanisms of cell- penetrating peptides with lipids and glycosaminoglycans. Adv. Drug Delivery Rev. 60, 580-597. 33. Mitchell, D. J., Kim, D. T., Steinman, L., Fathman, C. G., and Rothbard, J. B. (2000) Polyarginine enters cells more efficiently than other polycationic homopolymers. J. Peptide Res. 56, 318-325. 34. Rothbard, J. B., Jessop, T. C., Lewis, R. S., Murray, B. A., and Wender, P. A. (2004) Role of membrane potential and hydrogen bonding in the mechanism of translocation of guanidinium-rich peptides into cells. J. Am. Chem. Soc. 126, 9506- 9507. 35. Wender, P. A., Rothbard, J. B., Jessop, T. C., Kreider, E. L., and Wylie, B. L. (2002) Oligocarbamate molecular transporters: design, synthesis, and biological evaluation of a new class of transporters for drug delivery. J. Am. Chem. Soc. 124, 13382-13383. 36. Wender, P. A., Kreider, E., Pelkey, E. T., Rothbard, J., and VanDeusen, C. L. (2005) Dendritic molecular transporters: synthesis and evaluation of tunable polyguanidino dendrimers that facilitate cellular uptake. Org. Lett. 7, 4815-4818. 37. Cooley, C. B., Trantow, B. M., Nederberg, F., Kiesewetter, M. K., Hedrick, J. L., Waymouth, R. M., and Wender, P. A. (2009) Oligocarbonate molecular transporters: oligomerization-based syntheses and cell-penetrating studies. J. Am. Chem. Soc. 131, 16401-16403. 38. Elson-Schwab, L., Garner, O., Schuksz, M., Crawford, B., Esko, J., and Tor, Y. (2007) Guanidinylated neomycin delivers large, bioactive cargo into cells through a heparan sulfate-dependent pathway. J. Biol. Chem. 282, 13585-13591. 39. Stanzl, E. G., Trantow, B. M., Vargas, J. R., and Wender, P. A. (2013) Fifteen years of cell-penetrating, guanidinium-rich molecular transporters: basic science, research tools, and clinical applications. Acc. Chem. Res. 46, 2944-2954. 40. Choloin, L., Morris, M. C., Van Mau, N., Mery, J., Divita, G., and Heitz, F. (1999) Synthetic primary amphipathic peptides as tools for the cellular import of drugs and nucleic acids. Curr. Topics Pept. Protein Res. 3, 153-162. 144

41. Morris, M. C., Vidal, P., Chaloin, L., Heitz, F., and Divita, G. (1997) A new peptide vector for efficient delivery of oligonucleotides into mammalian cells. Nucleic Acids Res. 25, 2730-3736. 42. Nam, Y. H., Park, I. S., Hahm, K. S., and Shin, S. Y. (2011) Antimicrobial activity, bactericidal mechanism and LPS-neutralizing activity of the cell-penetrating peptide pVEC and its analogs. J. Pept. Sci. 17, 812-817. 43. Johansson, H. J., El-Andaloussi, S., Holm, T., Mae, M., Janes, J., Maimets, T., and Langel, U. (2008) Characterization of a novel cytotoxic cell-penetrating peptide derived from p14ARG protein. Mol. Ther. 16, 115-123. 44. Magzoub, M., Sandgren, S., Lundberg, P., Oglecka, K., Lilja, J., Wittrup, A., Goran Eriksson, L. E., Langel, U., Belting, M., and Graslund, A. (2006) N-terminal peptides from unprocessed prion protein enters cells by macropinocytosis. Biochem. Biophys. Res. Commun. 348, 379-385. 45. Oehlke, J., Scheller, A., Wiesner, B., Krause, E., Bevermann, M., Klauschenz, E., Meiziq, M., and Bienert, M. (1998) Cellular uptake of an alpha-helical amiphathic model peptide with the potential to deliver polar compounds into the cell interior non-endocytically. Biochim. Biophys. Acta. 1414, 127-139. 46. Oehlke, J., Krause, E., Wiesner, B., Beyermann, M., and Bienert, M. (1997) Extensive cellular uptake into endothelial cells of an amphipathic? – Sheet forming peptide. FEBS Lett. 415, 196-199. 47. Pujals, S., and Giralt, E. (2008) Proline-rich, amphipathic cell-penetrating peptides. Adv. Drug Deliv. Rev. 60, 473-484. 48. Fernandez-Carneado, J., Kogan, M. J., Castel, S., and Giralt, E. (2004) Potential peptide carriers: amphipathic proline-rich peptides derived from the N-terminal domain of γ-zein. Angew. Chem. Int. Ed. 43, 1811-1814. 49. Sadler, K., Eom, K. D., Yang, J.-L., Dimitrova, Y., and Tam, J. P. (2002) Translocating proline-rich peptide from the antimicrobial peptide bactenecin 7. Biochemistry 41, 14150-14157. 50. Fillon, Y. A., Anderson, J. P., and Chmielewski, J. (2005) Cell penetrating agents based on a polyproline helix scaffold. J. Am. Chem. Soc. 127, 11798-11803. 51. Lin, Y.-Z., Yao, S., Veach, R. A., Torgerson, T. R., and Hawiger, J. (1995) Inhibition of nuclear translocation of transcription factor NF-κB by a synthetic peptide containing a cell membrane-permeable motif and nuclear localization sequence. J. Biol. Chem. 270, 14255-14258. 52. Watkins, C. L., Brennan, P., Fegan, C., Takayama, K., Nakase, I., Futaki, S., and Jones, A. T. (2009) Cellular uptake, distribution and cytotoxicity of the hydrophobic cell penetrating peptide sequence PFVYLI linked to the proapoptotic domain peptide PAD. J. Controlled Release 140, 237-244. 53. Lipinski, C., and Hopkins, A. (2004) Navigating chemical space for biology and medicine. Nature 432, 855-861. 54. Dubikovskaya, E. A., Thorne, S. H., Pillow, T. H., Contag, C. H., and Wender, P. A. (2008) Overcoming multidrug resistance of small-molecule therapeutics through conjugation with releasable octaarginine transporters. Proc. Natl. Acad. Sci. U. S. A. 105, 12128-12133. 145

55. Liang, J. F., and Yang, V. C. (2005) Synthesis of doxorubicin-peptide conjugate with multidrug resistant tumor cell killing activity. Bioorg. Med. Chem. Lett. 15, 5071-5075. 56. Mazel, M., Clair, P., Rousselle, C., Vidal, P., Scherrmann, J. M., Mathieu, D., and Temsamani, J. (2001) Doxorubicin-peptide conjugates overcome multidrug resistance. Anticancer Drugs 12, 107-116. 57. Lindgren, M., Rosenthal-Aizman, K., Saar, K., Eiriksdottir, E., Jiang, Y., Sassian, M., Ostlund, P., Hallbrink, M., and Langel, U. (2006) Overcoming methotrexate resistance in breast cancer tumour cells by the use of a new cell-penetrating peptide. Biochem. Pharmacol. 71, 416-425. 58. Torchilin, V. (2009) Intracellular delivery of protein and peptide therapeutics. Drug Discovery Today Technol. 5, e95. 59. Ochocki, J. D., Mullen, D. G., Wattenberg, E. V., and Distefano, M. D. (2011) Evaluation of a cell penetrating prenylated peptide lacking an intrinsic fluorophore via in situ click reaction. Bioorg. Med. Chem. Lett. 21, 4998-5001. 60. Covic, L., Gresser, A. L., Talavera, J., Swift, S., and Kuliopulos, A. (2002) Activation and inhibition of G protein-coupled receptors by cell-penetrating membrane-tethered peptides. Proc. Natl. Acad. Sci. U. S. A. 99, 643-648. 61. Schafmeister, C. E., Po, J., and Verdine, G. L. (2000) An all-hydrocarbon cross- linking system for enhancing the helicity and metabolic stability of peptides. J. Am. Chem. Soc. 122, 5891-5892. 62. Chatterjee, J., Gilon, C., Hoffman, A., and Kessler, H. (2008) N-methylation of peptides: A new perspective in medicinal chemistry. Acc. Chem. Res. 41, 1331- 1342. 63. Snyder, E. L., Meade, B. R., Saenz, C. C., and Dowdy, S. F. (2004) Treatment of terminal peritoneal carcinomatosis by a transducible p53-activating peptide. PLos Biol. 2, e36. 64. Michl, J., Scharf, B., Schmidt, A., Huynh, C., Hannan, R., von Gizycki, H., Friedman, E. K., Brandt-Rauf, P., Fine, R. L., and Pincus, M. R. (2006) PNC-28, a p53-derived peptide that is cytotoxic to cancer cells, blocks pancreatic cancer cell growth in vivo. Int. J. Cancer 119, 1577-1585. 65. Aramburu, J., Yaffe, M. B., Lopez-Rodriguez, C., Cantley, L. C., Hogan, P. G., and Rao, A. (1999) Affinity-driven peptide selection of an NFAT inhibitor more selective than cyclosporine A. Science 285, 2129-2133. 66. Noguchi, H., Matsushita, M., Okitsu, T., Moriwaki, A., Tomizawa, K., Kang, S., Li, S. T., Kobayashi, N., Matsumoto, S., Tanaka, K., Tanaka, N., and Matsui, H. (2004) A new cell-permeable peptide allows successful allogeneic islet transplantation in mice. Nat. Med. 10, 305-309. 67. Du, C., Fang, M., Li, Y., Li, L., and Wang, X. (2000) Smac, a mitochondrial protein that promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 102, 33-42. 68. Arnt, C. R., Chiorean, M. V., Heldebrant, M. P., Gores, G. J., and Kaufmann, S. H. (2002) Synthetic Smac/DIABLO peptides enhance the effect of chemotherapeutic agents by binding XIAP and cIAP1 in situ. J. Biol. Chem. 277, 44236-44243. 146

69. Fulda, S., Wick, W., Weller, M., and Debatin, K. M. (2002) Smac agonist sensitize for Apo2L/TRAIL- or anticancer drug-induced apoptosis and induce regression of malignant glioma in vivo. Nat. Med. 8, 808-815. 70. Sugioka, R., Shimizu, S., Funatsu, T., Tamagawa, H., Sawa, Y., Kawakami, T., and Tsujimoti, Y. (2003) BH4-domain peptide from BcL-xL exerts anti-apoptotic activity in vivo. Oncogene 22, 8432-8440. 71. Holinger, F. P., Chittenden, T., and Lutz, R. J. (1999) Bak BH3 peptides antagonize Bcl-xL function and induce apoptosis through cytochrome c-independent activation of caspases. J. Biol. Chem. 274, 13298-13304. 72. McCusker, C. T., Wang, Y., Shan, J., Kinyanjui, M. W., Villeneuve, A., Michael, H., and Fixman, E. D. (2007) Inhibition of experimental allergic airways disease by local application of a cell-penetrating dominant-negative STAT-6 peptide. J. Immunol. 179, 2556-2564. 73. Nagel-Wolfrum, K., Buerger, C., Wittig, I., Butz, K., Hoppe-Seyler, F., and Groner, B. (2004) The interaction of specific peptide aptamers with the DNA binding domain and the dimerization domain of the transcription factor Stat3 inhibits transactivation and induces apoptosos in tumor cells. Mol. Cancer Res. 2, 170-182. 74. Agou, F., Courtois, G., Chiaravalli, J., Baleux, F., Coic, Y. M., Traincard, F., Israel, A., and Veron, M. (2004) Inhibition of NF-kappa B activation by peptides targeting NF-kappa B essential modulator (nemo) oligomerization. J. Biol. Chem. 279, 54248-54257. 75. Nagahara, H., Vocero-Akbani, A. M., Snyder, E. L., Ho, A., Latham, D. G., Lissy, N. A., Becker-Hapak, M., Ezhevsky, S. A., and Dowdy, S. F. (1998) Transduction of full-length TAT fusion proteins into mammalian cells: TAT-p27Kip1 induces cell migration. Nat. Med. 4, 1449-1452. 76. Schwarze, S. R., Ho, A., Vocero-Akbani, A., and Dowdy, S. F. (1999) In vivo protein transduction: delivery of a biologically active protein into the mouse. Science 285, 1569-1572. 77. Becker-Hapak, M., McAllister, S. S., and Dowdy, S. F. (2001) TAT-mediated protein transduction into mammalian cells. Methods 24, 247-256. 78. Kwon, H. Y., Eum, W. S., Jang, H. W., Kang, J. H., Ryu, J., Ryong Lee, B., Jin, J. H., and Choi, S. Y. (2000) Transduction of Cu, Zn-superoxide dismutase mediated by an HIV-1 Tat protein basic domain into mammalian cells. FEBS Lett. 485, 163- 167. 79. Jin, J. H., Bahn, J. H., Eum, W. S., Kwon, H. Y., Jang, S. H., Kan, K. H., Kang, T. C., Won, M. H., Kang, J. H., Cho, S. W., Park, J., and Choi, S. Y. (2001) Transduction of human catalase mediated by an HIV-1 TAT protein basic domain and arginine-rich peptides into mammalian cells. Free Radic. Biol. Med. 31, 1509- 1519. 80. Caron, N. J., Torrente, Y., Camirand, G., Bujold, M., Chapdelaine, P., Leriche, K., Bresolin, N., and Tremblay, J. P. (2001) Intracellular delivery of a Tat-eGFP fusion protein into muscle cells. Mol. Ther. 3, 310-318. 81. Peitz, M., Pfannkuche, K., Rajewsky, K., and Edenhofer, F. (2002) Ability of the hydrophobic FGF and based TAT peptides to promote cellular uptake of 147

recombinant Cre recombinase: a tool for efficient genetic engineering of mammalian genomes. Proc. Natl. Acad. Sci. U. S. A. 99, 4489-4494. 82. Wadia, J. S., Stan, R. V., and Dowdy, S. F. (2004) Transducible TAT-HA fusogenic peptide enhances escape of TAT-fusion proteins after lipid raft macropinocytosis. Nat. Med. 10, 310-315. 83. Kaplan, I. M., Wadia, J. S., and Dowdy, S. F. (2005) Cationic TAT peptide transduction domain enters cells by macropinocytosis. J. Controlled Release 102, 247-253. 84. Cronican, J. J., Thompson, D. B., Beier, K. T., McNaughton, B. R., Cepko, C. L., and Liu, D. R. (2010) Potent delivery of functional proteins into mammalian cells in vitro and in vivo using a supercharged protein. ACS Chem. Biol. 5, 747-752. 85. Simeon, R. L., Chamoun, A. M., McMillin, T., and Chen, Z. (2013) Discovery and charaterization of a new cell-penetrating protein. ACS Chem. Biol. 8, 2678-2687. 86. Cochet, O., Kenigsberg, M., Delumeau, I., Virone-Oddos, A., Multon, M. C., Fridman, W. H., Schweighoffer, F., Teillaud, J. L., and Tocque, B. (1998) Intracellular experssion of an antibody fragment-neutralizing p21 ras promotes tumor regression. Cancer Res. 58, 1170-1176. 87. Niesner, U., Halin, C., Lozzi, L., Gunthert, M., Neri, P., Wunderli-Allenspach, H., Zardi, L., and Neri, D. (2002) Quantification of the tumor-targeting properties of antibody fragments conjugated to cell-penetrating HIV-1 TAT peptides. Bioconjugate Chem. 13, 729-736. 88. Takenobu, T., Tomizawa, K., Matsushita, M., Li, S. T., Moriwaki, A., Lu, Y. F., and Matsui, H. (2002) Development of p53 protein transduction therapy using membrane-permeable peptides and the application to oral cancer cell. Mol. Cancer Ther. 1, 1043-1049. 89. Phelan, A., Elliott, G., and O’Hare, P. (1998) Intracellular delivery of funtional p53 by the herpesvirus protein VP22. Nat. Biotechnol. 16, 440-443. 90. Fahrer, J., Schweitzer, B., Fiedler, K., Langer, T., Gierschik, P., and Barth, H. (2013) C2-streptavidin mediates the delivery of biotin-conjugated tumor suppressor protein p53 into tumor cells. Bioconjugate Chem. 24, 595-603. 91. Meade, B. R., and Dowdy, S. F. (2007) Exogenous siRNA delivery using peptide transduction domains/cell penetrating peptides. Adv. Drug Delivery Rev. 59, 134- 140. 92. Liu, Z., Li, M., Cui, D., and Fei, J. (2005) Macro-branched cell-penetrating peptide design for gene delivery. J. Controlled Release 102, 699-710. 93. Rudolph, C., Plank, C., Lausier, J., Schillinger, U., Muller, R. H., and Rosenecker, J. (2003) Oligomers of the arginine-rich motif of the HIV-1 TAT protein are caplable of transferring plasmid DNA into cells. J. Biol. Chem. 278, 11411-11418. 94. Astriab-Fisher, A., Sergueev, D., Fisher, M., Shaw, B. R., and Juliano, R. L. (2002) Conjugates of antisense oligonucleotides with the Tat and antennapedia cell- penetrating peptides: effects on cellular uptake, binding to target sequences, and biologic actions. Pharm. Res. 19, 744-754.

148

95. El-Andaloussi, S., Johansson, H., Magnusdottir, A., Jarver, P., Lundberg, P., and Langel U. (2005) TP10: a delivery vector for decoy oligonucleotides targeting the Myc protein. J. Controlled Release 110, 189-201. 96. Endoh, T., and Ohtsuki, T. (2009) Cellular siRNA delivery using cell-penetrating peptides modified for endosomal escape. Adv. Drug Delivery Rev. 61, 704-709. 97. Hannon, G. J., and Rossi, J. J. (2004) Unlocking the potential of the human genome with RNA interference. Nature 431, 371-378. 98. De Fougerolles, A., Vornlocher, H.-P., Maraganore, J., and Lieberman, J. (2007) Interfering with diseases: a progress report on siRNA-based therapeutics. Nat. Rev. Drug Disc. 6, 443-453. 99. Chiu, Y.-L., Ali, A., Chu, C., Cao, H., and Rana, T. M. (2004) Visualizing a correlation between siRNA localization, cellular uptake, and RNAi in living cells. Chem. Biol. 11, 1165-1175. 100. Davidson, T. J., Harel, S., Arboleda, V. A., Prunell, G. F., Shelanski, M. L., Greene, L. A., and Troy, C. M. (2004) Highly efficient small interfering RNA delivery to primary mammalian neurons induces MicroRNA-like effects before mRNA degradation. J. Neurosci. 24, 10040-10046. 101. Turner, J. J., Jones, S., Fabani, M. M., Ivanova, G., Arzumanov, A. A., and Gait, M. J. (2007) RNA targeting with peptide conjugates of oligonucleotides, siRNA and PNA. Blood Cells Mol. Diseases 38, 1-7. 102. Moschos, S. A., Jones, S. W., Perry, M. M., Williams, A. E., Erjefalt, J. S., Turner, J. J., Barnes, P. J., Sproat, B. S., Gait, M. J., and Lindsay, M. A. (2007) Lung delivery studies using siRNA conjugated to TAT (48-60) and penetratin reveal peptide induced reduction in gene expression and induction of innate immunity. Bioconjugate Chem. 18, 1450-1459. 103. Simeoni, F., Morris, M. C., Heitz, F., and Divita, G. (2003) Insight into the mechanism of the peptide-based gene delivery system MPG: implications for delivery of siRNA into mammalian cells. Nucleic Acids Res. 31, 2717-2724. 104. Kumar, P., Wu, H., McBride, J. L., Jung, K.-E., Kim, M. H., Davidson, B. L., Lee, S. K., Shankar, P., and Manjunath, N. (2008) Transvascular delivery of small interfering RNA to the central nervous system. Nature 448, 39-43. 105. Kumar, P., Ban, H. S., Kim, S. S., Wu, H., Pearson, T., Greiner, D., Laouar, A., Yao, J., Haridas, V., Habiro, K., Yang, Y. G., Jeong, J. H., Lee, K. Y., Kim, Y. K., Kim, S. W., Peipp, M., Fey, G. H., Manjunath, N., Shultz, L. N., Lee, S. K., and Shankar, P. (2008) T cell-specific siRNA delivery suppresses HIV-1 infection in humanized mice. Cell 134, 577586. 106. Maiolo, J. R., Ferrer, M., and Ottinger, E. A. (2005) Effects of cargo molecules on the cellular uptake of arginine-rich cell-penetrating peptides. Biochim. Biophys. Acta. 1712, 161-172. 107. Mueller, J., Kretzschmar, I., Volkmer, R., and Boisguerin, P. (2008) Comparison of cellular uptake using 22 CPPs in 4 different cell lines. Bioconjugate Chem. 19, 2363-2374.

149

108. Vives, E., Brodin, P., and Lebleu, B. (1997) A truncated HIV-1 Tat protein basic domain rapidly translocates through the plasma membrane and accumulates in the cell nucleus. J. Biol. Chem. 272, 16010-16017. 109. Futaki, S., Suzuki, T., Ohashi, W., Yagami, T., Tanaka, S., Ueda, K., and Sugiura, Y. (2001) Arginine-rich peptides. An abundant source of membrane-permeable peptides having potential as carriers for intracellular protein delivery. J. Biol. Chem. 276, 5836-5840. 110. Richard, J. P., Melikov, K., Vives, E., Ramos, C., Verbeure, B., Gait, M. J., Chernomordik, L. V., and Lebleu, B. (2003) Cell-penetrating peptides. A reevaluation of the mechanism of cellular uptake. J. Biol. Chem. 278, 585-590. 111. Duchardt, F., Fotin-Mleczek, M., Schwarz, H., Fischer, R., and Brock, R. (2007) A comprehensive model for the cellular uptake of cationic cell-penetrating peptides. Traffic 8, 848-866. 112. Copolovici, D. M., Langel, K., Eriste, E., and Langel, U. (2014) Cell-penetrating peptides: design, synthesis and applications. ACS Nano 8, 1972-1994. 113. Guterstam, P., Madani, F., Hirose, H., Takeuchi, T., Futaki, S., El Andaloussi, S., Graslund, A., and Langel, U. (2009) Elucidating cell-penetrating peptide mechanisms of action for membrane interaction, cellular uptake, and translocation utilizing the hydrophobic couter-anion pyrenebutyrate. Biochim. Biophys. Acta. Biomembr. 1788, 2509-2517. 114. Jiao, C. Y., Delaroche, D., Burlina, F., Alves, I. D., Chassaing, G., and Sagan, S. (2009) Translocation and endocytosis for cell-penetrating peptides internalization. J. Biol. Chem. 284, 33957-33965. 115. Nakase, I., Takeuchi, T., Tanaka, G., and Futaki, S. (2008) Methodological and cellular aspects that govern the internalization mechanisms of arginine-rich cell- penetrating peptides. Adv. Drug Delivery Rev. 60, 598-607. 116. Koren, E., and Torchilin, V. P. (2012) Cell-penetrating petpides: breaking through to the other side. Trends Mol. Med. 18, 850-860. 117. Richard, J. P., Melikov, K., Brooks, H., Prevot, P., Lebleu, B., and Chernomordik, L. V. (2005) Cellular uptake of unconjugated TAT peptide involves clathrin- dependent endocytosis and heparan sulfate receptors. J. Biol. Chem. 280, 15300- 15306. 118. Ferrari, A., Pellegrini, V., Arcangeli, C., Fittipaldi, A., Giacca, M., and Beltram, F. (2003) Caveolae-mediated internalization of extracellular HIV-1 tat fusion proteins visualized in real time. Mol. Ther. 8, 284-294. 119. Fittipaldi, A., Ferrari, A., Zoppe, M., Arcangeli, C., Pellegrini, V., Beltram, F., and Giacca, M. (2003) Cell membrane lipid rafts mediate caveolar endocytosis of HIV- 1 Tat fusion proteins. J. Biol. Chem. 278, 34141-34149. 120. Nakase, I., Tadokoro, A., Kawabata, N., Takeuchi, T., Katoh, H., Hiramoto, K., Negishi, M., Nomizu, M., Sugiura, Y., and Futaki, S. (2007) Interaction of arginine- rich peptides with membrane-associated proteoglycans is crucial for induction of actin organization and macropinocytosis. Biochemistry 46, 492-501. 121. Huotari, J., and Helenius, A. (2011) Endosome maturation. EMBO J. 30, 3481- 3500. 150

122. Jones, A. T. (2007) Macropinocytosis: searching for an endocytic identity and a role in the uptake of cell penetratin peptides. J. Cell. Mol. Med. 11, 670-684. 123. Nakase, I., Niwa, M., Takeuchi, T., Sonomura, K., Kawabata, N., Koike, Y., Takehashi, M., Tanaka, S., Ueda, K., Simpson, J. C., Jones, A. T., Sugiura, Y., and Futaki, S. (2004) Cellular uptake of arginine-rich peptides: roles for macropinocytosis and actin rearrangements. Mol. Ther. 10, 1011-1022. 124. Mishra, A., Lai, G. H., Schmidt, N. W., Sun, V. Z., Rodriguez, A. R., Tong, R., Tang, L., Cheng, J., Deming, T. J., Kamei, D., and Wong, G. C. L. (2011) Translocation of HIV TAT peptide and analogues induced by multiplexed membrane and cytoskeletal interactions. Proc. Natl. Acad. Sci. U. S. A. 108, 16883- 16888. 125. Saalik, P., Elmquist, A., Hansen, M., Padari, K., Saar, K., Viht, K., Langel, U., and Pooga, M. (2004) Protein cargo delivery properties of cell-penetrating peptides. A comparative study. Biochonjugate Chem. 15, 1246-1253. 126. Ter-Avetisyan, G., Tunnermann, G., Nowak, D., Nitschke, M., Herrmann, A., Drab, M., and Cardoso, M. C. (2009) Cell entry of arginine-rich peptides is independent of endocytosis. J. Biol. Chem. 284, 3370-3378. 127. Vercauteren, D., Vandenbroucke, R. E., Jones, A. T., Rejman, J., Demeester, J., De Smedt, S. C., Sanders, N. N., and Braeckmans, K. (2010) The use of inhibitors to study endocytic pathways of gene carriers: optimization and pitfalls. Mol. Ther. 18, 561-569. 128. Saalik, P., Ninep, A., Pae, J., Hansen, M., Lubenets, D., Langel, U., and Pooga, M. (2011) Penetration without cells: membrane translocation of cell-penetrating peptides in the model giant plasma membrane vesicles. J. Controlled Release 153, 117-125. 129. Thoren, P. E., Persson, D., Esbjorner, E. K., Goksor, M., Lincoln, P., and Norden, B. (2004) Membrane binding and translocation of cell-penetrating peptides. Biochemistry 43, 3471-3489. 130. Fretz, M. M., Penning, N. A., Al-Taei, S., Futaki, S., Takeuchi, T., Nakase, I., Storm G., and Jones, A. T. (2007) Temperature-, concentration- and cholesterol- dependent translocation of L- and D-octa-arginine across the plasma and nuclear membrane of CD34+ leukaemia cells. Biochem. J. 403, 335-342. 131. Derossi, D., Calvet, S., Trembleau, A., Brunissen, A., Chassaing, G., and Prochiantz, A. (1996) Cell internalization of the third helix of the antennapedia homeodomain is receptor-independent. J. Biol. Chem. 271, 18188-18193. 132. Matsuzaki, K., Yoneyama, S., Murase, O., and Miyajima, K. (1996) Transbilayer transport of ions and lipids coupled with mastoparan X translocation. Biochemistry 35, 8450-8456. 133. Rothbard, J. B., Jessop, T. C., and Wender, P. A. (2005) Adaptive translocation: the role of hydrogen bonding and membrane potential in the uptake of guanidinium- rich transporters into cells. Adv. Drug Delivery Rev. 57, 495-504. 134. Pouny, Y., Rapaport, D., Mor, A., Nicolas, P., and Shai, Y. (1992) Interaction of antimicrobial dermaseptin and its fluorescently labeled analogues with phospholipid membranes. Biochemistry 31, 12416-12423. 151

135. Verdurmen, W. P., Thanos, M., Ruttekolk, I. R., Gulbins, E., and Brock, R. (2010) Cationic cell-penetrating peptides induce ceramide formation via acid sphingomyelinase: implications for uptake. J. Controlled Release 147, 171-179. 136. Kawamoto, S., Takasu, M., Miyakawa, T., Morikawa, R., Oda, T., Futaki, S., and Nagao, H. (2011) Inverted micelle formation of cell-penetrating peptide studied by coarse-grained simulation: importance of attractive force between cell-penetrating peptides and lipid head group. J. Chem. Phys. 134, 095103. 137. Hirose, H., Takeuchi, T., Osakada, H., Pujals, S., Katayama, S., Nakase, I., Kobayashi, S., Haraguchi, T., and Futaki, S. (2012) Transient focal membrane deformation induced by arginine-rich peptides leads to their direct penetration into cells. Mol. Ther. 20, 984-993. 138. Lamaziere, A., Burlina, F., Wolf, C., Chassaing, G., Trugnan, G., and Ayala- Sanmartin, J. (2007) Non-metabolic membrane tubulation and permeability induced by bioactive peptides. PLos One 2, e201. 139. Herce, H. D., and Garcia, A. E. (2007) Molecular dynamics simulations suggest a mechanism for translocation of the HIV-1 TAT peptide across lipid membranes. Proc. Natl. Acad. Sci. U. S. A. 104, 20805-20810. 140. Herce, H. D., Garcia, A. E., Litt, J., Kane, R. S., Martin, P., Enrigue, N., Rebolledo, A., and Milesi, V. (2009) Arginine-rich peptides destabilize the plasma membrane, consistent with a pore formation translocation mechanism of cell-penetrating peptides. Biophys. J. 97, 1917-1925. 141. Deshayes, S., Plenat, T., Charnet, P., Divita, G., Molle, G., and Heitz, F. (2006) Formationof transmembrane ionic channels of primary amphipathic cell-penetrating peptides. Consequences on the mechanism of cell penetration. Biochim. Biophys. Acta. 1758, 1846-1851. 142. Yandek, L. E., Pokorny, A., Floren, A., Knoelke, K., Langel, U., and Almeida, P. F. (2007) Mechanism of the cell-penetrating peptide transportan 10 permeation of lipid bilayers. Biophys. J. 92, 2434-2444. 143. Fischer, R., Kohler, K., Fotin-Mleczek, M., and Brock, R. (2004) A stepwise dissection of the intracellular fate of cationic cell-penetrating peptides. J. Biol. Chem. 279, 12625-12635. 144. Magzoub, M., Pramanik, A., and Graslund, A. (2005) Modeling the endosomal escape of cell-penetrating peptides: transmembrane pH gradient driven translocation across phospholipid bilayers. Biochemistry 44, 14890-14897. 145. Madani, F., Abdo, R., Lindberg, S., Hirose, H., Futaki, S., Langel, U., and Graslund, A. (2013) Modeling the endosomal escape of cell-penetrating peptides using a transmembrane pH gradient. Biochim. Biophys. Acta. 1828, 1198-1204. 146. Fuchs, S. M., and Raines, R. T. (2004) Pathway for polyarginine entry into mammalian cells. Biochemistry 42, 2438-2444. 147. El-Sayed, A., Futaki, S., and Harashima, H. (2009) Delivery of macromolecules using arginine-rich cell-penetrating peptides: ways to overcome endosomal entrapment. AAPS J. 11, 13-22. 148. Varkouhi, A. K., Scholte, M., Storm, G., and Haisma, H. J. (2011) Endosomal escape pathways for delivery of biologicals. J. Controlled Release 151, 220-228. 152

149. Erazo-Oliveras, A., Muthukrishnan, N., Baker, R., Wang, T. Y., and Pellois, J. P. (2012) Improving the endosomal esacpase of cell-penetrating peptides and their cargos: strategies and challenges. Pharmaceuticals 5, 1177-1209. 150. Medintz, I. L., Pons, T., Delehanty, J. B., Susumu, K., Brunel, F. M., Dawson, P. E., and Mattoussi, H. (2008) Intracellular delivery of quantum dot-cargos mediated by cell penetrating peptides. Bioconjugate Chem. 19, 1785-1795. 151. Pan, C., Lu, B., Chen, H., and Bishop, C. (2010) Reprogramming human fibroblasts using HIV-1 TAT recombinant proteins OCT4, SOX2, KLF4 and c-MYC. Mol. Biol. Reports. 37, 2117–2124. 152. Loison, F., Nizard, P., Sourisseau, T., Le Goff, P., Debure, L., Le Drean, Y., and Michel, D. (2005) A ubiquitin-based assay for the cytosolic uptake of protein transduction domains. Mol. Ther. 11, 205-214. 153. Gillmeister, M. P., Betenbaugh, M. J., and Fishman, P. S. (2011) Cellular trafficking and photochemical internalization of cell penetrating peptide linked cargo proteins: a dual fluorescent labeling study. Bioconjugate Chem. 22, 556-566. 154. Palm-Apergi, C., Lorents, A., Padari, K., Pooga, M., and Hallbrink, M. (2009) The membrane repair response masks membrane disturbances caused by cell- penetrating peptide uptake. FASEB J. 23, 214-223. 155. Pack, D. W., Hoffman, A. S., Pun, S., and Stayton, P. S. (2005) Design and development of polymers for gene delivery. Nat. Rev. Drug Discov. 4, 581-593. 156. Hogset, A., Prasmickaite, L., Selbo, P. K., Hellum, M., Engesaeter, B. O., Bonsted, A., and Berg, K. (2004) Photochemical internalization in drug and gene delivery. Adv. Drug Delivery Rev. 56, 95-115. 157. Driggers, E. M., Hale, S. P., Lee, J., and Terrett, N. K. (2008) The exploration of macrocycles for drug discovery-an underexploited structural class. Nat. Rev. Drug Discov. 7, 608-624. 158. Szewczuk, Z., Gibbs, B. F., Yue, S. Y., Purisima, E. O., and Konishi, Y. (1992) Conformationally restricted thrombin inhibitors resistant to proteolytic digestion. Biochemistry 31, 9132-9140. 159. Hu, X., Nguyen, K. T., Jiang, V. C., Lofland, D., Moser, H. E., and Pei, D. (2004) Macrocyclic inhibitors for peptide deformylase: An SAR study of the ring size. J. Med. Chem. 47, 4941-4949. 160. Nguyen, L. T., Chau, J. K., Perry, N. A., de Boer, L., Zaat, S. A. J., and Vogel, H. J. (2010) Serum stabilities of short tryptophan- and arginine-rich antimicrobial peptide analogs. PLoS One 5, e12684. 161. Millward, S. W., Takahashi, T. T., and Roberts, R. W. (2005) A general route for post-translational cyclization of mRNA display libraries. J. Am. Chem. Soc. 127, 14142-14143. 162. Tavassoli, A., and Benkovic, S. J. (2005) Genetically selected cyclic-peptide inhibitors of AICAR transformylase homodimerization. Angew. Chem. Int. Ed. 44, 2760-2763. 163. Joo, S. H., Xiao, Q., Ling, Y., Gopishetty, B., and Pei, D. (2006) High-throughput sequence determination of cyclic peptide library members by partial Edman degradation/mass spectrometry. J. Am. Chem. Soc. 128, 13000-13009. 153

164. Kawakami, T., Ohta, A., Ohuchi, M., Ashigai, H., Murakami, H., and Suga, H. (2009) Diverse backbone-cyclized peptides via condon reprogramming. Nat. Chem. Biol. 5, 888-890. 165. Kleiner, R. E., Dumelin, C. E., Tiu, G. C., Sakurai, K., and Liu, D. R. (2010) In vitro selection of a DNA-templated small-molecule library reveals a class of macrocyclic kinase inhibitors. J. Am. Chem. Soc. 132, 11779-11791. 166. Rezai, T., Yu, B., Millhauser, G. L., Jacobson, M. P., and Lokey, R. S. (2006) Testing the conformational hypothesis of passive membrane permeability using synthetic cyclic peptide diastereomers. J. Am. Chem. Soc. 128, 2510-2511. 167. White, T. R., Renzelman, C. M., Rand, A. C., Rezai, T., McEwen, C. M., Gelev, V. M., Turner, R. A., Linington, R. G., Leung, S. S. F., Kalgutkar, A. S., Bauman, J. N., Zhang, Y. Z., Liras, S., Price, D. A., Mathiowetz, A. M., Jacobson, M. P., and Lokey, R. S. (2011) On-resin N-methylation of cyclic peptides for discovery of orally bioavailable scaffolds. Nat. Chem. Biol. 7, 810-817. 168. Liu, J., Farmer, J. D., Lane, W. S., Friedman, J., Weissman, I., and Schreiber, S. L. (1991) Calcineurin is a common target of cyclophilin-cyclosporine A and FKBP- FK506 complexes. Cell 66, 807-815. 169. Rueping, M., Mahajan, Y., Sauer, M., and Seebach, D. (2002) Cellular uptake studies with beta-peptides. ChemBioChem 3, 257-259. 170. Pham, W., Kircher, M. F., Weissleder, R., and Tung, C. H. (2004) Enhancing membrane permeability by fatty acylation of oligoarginine peptides. ChemBiochem 5, 1148-1151. 171. Okumu, F. W., Pauletti, G. M., Vander Velde, D. G., Siahaan, T. J., and Borchardt, R. T. (1997) Effect of restricted conformational flexibility on the permeation of model hexappetides across Caco-2 cell monolayers. Pharm. Res. 14, 167-175. 172. Boguslavsky, V., Hruby, V. J., O’Brien, D. F., Misicka, A., and Lipkowski, A. W. (2003) Effect of peptide conformation on membrane permeability. J. Pept. Res. 61, 287-297. 173. Kwon, Y. U., and Kodadek, T. (2007) Quantitiative comparison of the relative cell permeability of cyclic and linear peptides. Chem. Biol. 14, 671-677. 174. Liu, T., Liu, Y., Kao, H. Y., and Pei, D. (2010) Membrane permeable cyclic peptidyl inhibitors against human peptidylproplyl isomerase Pin1. J. Med. Chem. 53, 2494-2501. 175. Lattig-Tunnemann, G., Prinz, M., Hoffmann, D., Behlke, J., Palm-Apergi, C., Morano, I., Herce, H. D., and Cardoso, M. C. (2011) Backbone rigidity and static presentation of guanidinium groups increases cellular uptake of arginine-rich cell- penetrating peptides. Nat. Commun. 2, 453. 176. Mandal, D., Nasrolahi Shirazi, A., and Parang, K. (2011) Cell-penetrating homochiral cyclic peptides as nuclear-targeting molecular transporters. Angew. Chem. Int. Ed. 50, 9633-9637. 177. Holm, T., Johansson, H., Lundberg, P., Pooga, M., Lindgren, M., and Langel, U. (2006) Studying the uptake of cell-penetrating peptides. Nat. Protoc. 1, 1001-1005.

154

178. Qian, Z., Liu, T., Liu, Y. Y., Briesewitz, R., Barrios, A. M., Jhiang, S. M., and Pei, D. (2013) Efficient delivery of cyclic peptides into Mammalian cells with short sequence motifs. ACS Chem. Biol. 8, 423-431. 179. Jones, L. R., Goun, E. A., Shinde, R., Rothbard, J. B., Contag, C. H., and Wender, P. A. (2006) Releasable luciferin-transporter conjugates: Tools for the real-time analysis of cellular uptake and release. J. Am. Chem. Soc. 128, 6526-6527. 180. Go, Y.-M., and Jones, D. P. (2008) Redox compertmentalization in eukaryotics cells. Biochim. Biophys. Acta 1780, 1273-1290. 181. Collins, D. S., Unanue, E. R., and Harding, C. V. (1991) Reduction of disulfide bonds within lysosomes is a key step in antigen processing. J. Immunol. 147, 4054- 4059. 182. Feener, E. P., Shen, W. C., and Ryser, H. J. (1990) Cleavage of disulfide bonds in endocytosed macromolecules - a processing not associated with lysosomes or endosomes. J. Biol. Chem. 265, 18780-18785. 183. Austin, C. D., Wen, X. H., Gazzard, L., Nelson, C., Scheller, R. H., and Scales, S. J. (2005) Oxidizing potential of endosomes and lysosomes limits intracellular cleavage of disulfide-based antibody-drug conjugates. Proc. Natl. Acad. Sci. U. S. A. 102, 17987-17992. 184. Sauer, A. M., Schlossbauer, A., Ruthardt, N., Cauda, V., Bein, T., and Brauchle, C. (2010) Role of endosomal escape for disulfide-based drug delivery from colloidal mesoporous silica evaluated by live-cell imaging. Nano Lett. 10, 3684-3691. 185. Arunachalam, B., Phan, U. T., Geuze, H. J., and Cresswell, P. (2000) Enzymatic reduction of disulfide bonds in lysosomes: characterization of a Gamma-interferon- inducible lysosomal thiol reductase (GILT). Proc. Natl. Acad. Sci. U. S. A. 97, 745- 750. 186. Heng, M. C. Y., Barrascout, C. E., Rasmus, W., O'Brien, W., and Song, M. K. (1987) Elevated serum chymotrypsin levels in a patient with junctional epidermolysis bullosa. Inter. J. Dermat. 26, 385-388. 187. Fernandez-Suarez, M., and Ting, A. Y. (2008) Fluorescent probes for super- resolution imaging in living cells. Nat. Rev. Mol. Cell Biol. 9, 929-943. 188. Rink, J., Ghigo, E., Kalaidzidis, Y., and Zerial, M. (2005) Rab conversion as a mechanism of progression from early to late endosomes. Cell 122, 735-749. 189. Appelbaum, J. S., LaRochelle, J. R., Smith, B. A., Balkin, D. M., Holub, J. M., and Schepartz, A. (2012) Arginine topology controls escape of minimally cationic proteins from early endosomes to the cytoplasm. Chem. Biol. 19, 819−830. 190. Holub, J. M., LaRochelle, J. R., Appelbaum, J. S., and Schepartz, A. (2013) Improved assays for determining the cytosolic access of peptides, proteins, and their mimetics. Biochemistry 50, 9036-6046. 191. Yu, P., Liu, B., and Kodadek, T. (2005) A high-throughput assay for assessing the cell permeability of combinatorial libraries. Nat. Biotechnol. 23, 746-751. 192. Pratt, W. B., and Toft, D. O. (1997) Steroid receptor interactions with heat shock protein and immunophilin chaperones. Endocr. Rev. 18, 306-360. 193. Picard, D., and Yamamoto, K. R. (1987) Two signals mediate hormone-dependent nuclear localization of the glucocorticoid receptor. EMBO J. 6, 3333-3340. 155

194. Macia, E., Ehrlich, M., Massol, R., Boucrot, E., Brunner, C., and Kirchhausen, T. (2006) Dynasore, a cell-permeable inhibitor of dynamin. Dev. Cell 10, 839-850. 195. Koivusalo, M., Welch, C., Hayashi, H., Scott, C. C., Kim, M., Alexander, T., Touret, N., Hahn, K. M., and Grinstein, S. (2010) Amiloride inhibits macropinocytosis by lowering submembranous pH and preventing Rac1 and Cdc42 signaling. J. Cell Biol. 188, 547-563. 196. Rodal, S. K., Skretting, G., Garred, O., Vilhardt, F., van Deurs, B., and Sandvig, K. (1999) Extraction of cholesterol with methyl-beta-cyclodextrin perturbs formation of clathrin-coated endocytic vesicles. Mol. Biol. Cell 10, 961-974. 197. Yoshimori, T., Yamamoto, A., Moriyama, Y., Futai, M., and Tashiro, Y. (1991) Bafilomycin A1, a specific inhibitor of vacuolar-type H+-ATPase, inhibits acidification and protein degradation in lysosomes of cultured cells. J. Biol. Chem. 266, 17707-17712. 198. Li, G., D’Souza-Schorey, C., Barbieri, M. A., Roberts, R. L., Klippel, A., Williams, L. T. and Stahl, P. D. (1995) Evidence for phophatidylinositol 3-kinase as a regulator of endocytosis via activation of Rab5. Proc. Natl. Acad. Sci. U. S. A. 92, 10207-10211. 199. Stenmark, H., Parton, R. G., Steele-Mortimer, O., Lutcke, A., Gruenberg, J., and Zerial, M. (1994) Inhibition of rab5 GTPase activity stimulates membrane fusion in endocytosis. EMBO J. 13, 1287-1296. 200. Rusnati, M., Tulipano, G., Spillmann, D., Tanghetti, E., Oreste, P., Zoppetti, G., Giacca, M., and Presta, M. (1999) Multiple interactions of HIV-1 Tat protein with size-defined heparin oligosaccharides. J. Biol. Chem. 274, 28198-28205. 201. Tyagi, M., Rusnati, M., Presta, M., and Giacca, M. (2001) Internalization of HIV-1 tat requires cell surface heparin sulfate proteoglycans, J. Biol. Chem. 276, 3254- 3261. 202. Ziegler, A., and Seelig, J. (2004) Interaction of the protein transduction domain of HIV-1 TAT with heparan sulfate: binding mechanism and thermodynamic parameters. Biophys. J. 86, 254-263. 203. Goncalves, E., Kitas, E., and Seelig, J. (2005) Binding of oligoarginine to membrane lipids and heparan sulfate: structural and thermodynamic characterization of a cell-penetrating peptide. Biochemistry 44, 2692-2702. 204. Zhao, K., Choe, U. J., Kamei, D. T., and Wong, G. C. L. (2012) Enhanced activity of cyclic transporter sequences driven by phase behavior of peptide-lipid complexes. Soft Matter 8, 6430-6433. 205. Roederer, M., Mays, R. W., and Murphy, R. F. (1989) Effect of confluence on the endocytosis by 3T3 fibroblasts: increased rate of pinocytosis and accumulation of residual bodies. Eur. J. Cell. Biol. 48, 37-44. 206. Raucher, D., and Sheetz, M. P. (1999) Membrane expansion increases endocytosis rate during mitosis. J. Cell Biol. 144, 497-506. 207. Marks, J. R., Placone, J., Hristova, K., and Wimley, W. C. (2011) Spontaneous Membrane-Translocating Peptides by Orthogonal High-Throughput Screening. J. Am. Chem. Soc. 133, 8995-9004.

156

208. Magzoub, M., Eriksson, L. E. G., and Graslund, A. (2002) Conformational states of the cell-penetrating peptide penetratin when interacting with phospholipid vesicles: effects of surface charge and peptide concentration. Biochim. Biophys. Acta 1563, 53-63. 209. Mosmann, T. (1983) Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55-63. 210. Mitra, S. and Barrios, A. M. (2005) Highly sensitive peptide-based probes for protein tyrosine phosphatase activity utilizing a fluorogenic mimic of phosphotyrosine. Bioorg. Med. Chem. Lett. 15, 5124-5145. 211. Stanford, S. M., Panchal, R. G., Walker, L. M., Wu, D. J., Falk, M. D., Mitra, S., Damle, S. S., Ruble, D., Kaltcheva, T., Zhang, S., Zhang, Z. Y., Bavari, S., Barrios, A. M., and Bottini, N. (2012) High-throughput screen using a single-cell tyrosine phosphatase assay reveals biologically active inhibitors of tyrosine phosphatase CD45. Proc. Natl. Acad. Sci. U. S. A. 109, 13972-13977. 212. Ren, L., Chen, X., Luechapanichkul, R., Selner, N. G., Meyer, T. M., Wavreille, A. S., Chan, R., Iorio, C., Zhou, X., Neel, B. G., and Pei, D. (2011) Substrate specificity of protein tyrosine phosphatases 1B, RPTPalpha, SHP-1, and SHP-2. Biochemistry 50, 2339-2356. 213. Garcia-Calvo, M., Peterson, E. P., Leiting, B., Ruel, R., Nicholson, D. W., Thornberry, N. A. (1998) Inhibition of human caspases by peptide-based and macromolecular inhibitors. J. Biol. Chem. 273, 32608-32613. 214. Hashimoto, T., Kikkawa, U., Kamada, S. (2011) Contribution of caspase(s) to the cell cycle regulation at mitotic phase. PLos One 6, e18449. 215. Eckert, A., Steiner, B., Marques, C., Leutz, S., Romig, H., Haas, C., Muller, W. E. (2001) Elevated vulnerability to oxidative stress-induced cell death and activation of caspase-3 by the Swedish amyloid precursor protein mutation. J. Neurosci. Res. 64, 183-192. 216. Chae, H. J., Kang, J. S., Byun, J. O., Han, K. S., Kim, D. U., Oh, S. M., Kim, H. M., Chae, S. W., Kim, H. R. (2000) Molecular mechanism of staurosporine-induced apoptosis in osteoblasts. Pharmacol. Res. 42, 373-381. 217. Okun, I., Malarchuk, S., Dubrovskaya, E., Khvat, A., Tkachenko, S., Kysil, V., Ilyin, A., Kravchenko, D., Prossnitz, E. R., Sklar, L, Ivachtchenko, A. (2006) Screening for caspase-3 inhibitors: a new class of potent small-molecule inhibitors of caspase-3. J. Biomol. Screen. 11, 277-285. 218. Alonso, A., Sasin, J., Bottini, N., Friedberg, I., Friedberg, I., Osterman, A., Godzik, A., Hunter, T., Dixon, J., and Mustellin, T. (2004) Protein tyrosine phosphatases in the human genome. Cell 117, 699−711. 219. Qian, Z., LaRochelle, J. R., Jiang, B., Lian, W., Hard, R. L., Selner, N. G., Luechapanichkul, R., Barrios, A. M. and Pei. D. (2014) Early endosomal escape of a cyclic cell-penetrating peptide enables effective cytosolic cargo delivery. Biochemistry 53, 4034-4046. 220. Marsault, E., and Peterson, M. L. (2011) Macrocycles are great cycles: applications, opportunities, and challenges of synthetic macrocycles in drug discovery. J. Med. Chem. 54, 1961-2004. 157

221. Liu, T., Qian, Z., Xiao, Q., and Pei, D. (2011) High-throughput screening of one- bead-one-compound libraries: identification of cyclic peptidyl inhibitors against calcineurin/NFAT interaction. ACS Comb. Sci. 13, 537-546. 222. Dewan, V., Liu, T., Chen, K. M., Qian, Z., Xiao, Y., Kleiman, L., Mahasenan, K. V., Li, C., Matsuo, H., Pei, D., and Musier-Forsyth, K. (2012) Cyclic peptide inhibitors of HIV-1 capsid-human lysyl-tRNA synthetase interaction. ACS Chem. Biol. 7, 761-769. 223. Wu, X., Upadhyaya, P., Villalona-Calero, M. A., Briesewitz, R., and Pei, D. (2013) Inhibition of Ras-effector interactions by cyclic peptides. Med. Chem. Commun. 4, 378-382. 224. Lian, W., Upadhyaya, P., Rhodes, C. A., Liu, Y., and Pei, D. (2013) Screening bicyclic peptide libraries for protein-protein interaction inhibitor: discovery of a tumor necrosis factor-α antagonist. J. Am. Chem. Soc. 135, 11990-11995. 225. Elchelby, M., Payette, P., Michaliszyn, E., Cromlish, W., Collins, S., Loy, A. L., Normandin, D., Cheng, A., Himms-Hagen, J., Chan, C. C., Ramanchandran, C., Gresser, M. J., Tremblay, M. L., and Kennedy, B. P. (1999) Increased insulin sensitivity and obesity resistance in mice lacking the protein tyrosine phosphatase- 1B gene. Science 283, 1544–1548. 226. Zabolotny, J. M., Bence-Hanulec, K. K., Stricker-Krongrad, A., Haj, F., Wang, Y., Minokoshi, Y., Kim, Y. B., Elmquist, J. K., Tartaglia, L. A., Kahn, B. B., and Neel, B. G. (2002) PTP1B regulates Leptin signal transduction in vivo. Dev. Cell 2, 489- 495. 227. Zhang, S., and Zhang, Z. Y. (2007) PTP1B as a drug target: recent developments in PTP1B inhibitor discovery. Drug Discov. Today 12, 373-381. 228. Burke Jr., T. R., Kole, H. K., and Roller, P. P. (1994) Small molecule interactions with protein-tyrosine phosphatase PTP1B and their use in inhibitor design. Biochem. Biophys. Res. Commun. 204, 129-134. 229. Lian, W., Jiang, B., Qian, Z., and Pei, D. (2014) Cell-permeable bicyclic peptide inhibitors against intracellular proteins. J. Am. Chem. Soc. [Online early access]. DOI: 10.1021/ja503710n. 230. Yin, J., Straight, P. D., McLoughlin, S. M., Zhou, Z., Lin, A. J., Golan, D. E., Kelleher, N. L., Kolter, R., and Walsh, C. T. (2005) Genetically encoded short peptide tag for versatile protein labeling by Sfp phosphopantetheinyl transferase. Proc. Natl. Acad. Sci. U. S. A. 102, 15815-15820. 231. LaMontagne Jr., K. R., Hannon, G., and Tonks, N. K. (1998) Protein tyrosine phosphatase PTP1B suppresses p210 bcr-abl-induced transformation of Rat-1 fibroblasts and promotes differentiation of K562 cells. Proc. Natl. Acad. Sci. U. S. A. 95, 14094-14099. 232. Doyle, D. A., Lee, A., Lewis, J., Kim, E., Sheng, M., MacKinnon, R. (1996) Crystal structures of a complexed and peptide-free membrane protein-binding domain: molecular basis of peptide recognition by PDZ. Cell 85, 1067-1076. 233. Morais Cabral, J. H., Petosa, C., Sutcliffe, M. J., Raza, S., Byron, O., Poy, F., Marfatia, S. M., Chishti, A. H., and Liddington, R. C. (1996) Crystal structure of a PDZ domain. Nature 382, 649-652. 158

234. Wu, G., Chai, J., Suber, T. L., Wu, J. W., Du, C., Wang, X., and Shi, Y. (2000) Structural basis of IAP recognition by Smac/DIABLO. Nature 408, 1008-1012. 235. Frackenpohl, J., Arvidsson, P. I., Schreiber, J. V., and Seebach, D. (2001) The outstanding biological stability of beta- and gamma-peptides toward proteolytic enzymes: an in vitro investigation with fifteen peptidases. ChemBiochem 2, 445- 455. 236. Riedl, S. J., and Shi, Y. (2004) Molecular mechanisms of caspase regulation during apoptosis. Nat. Rev. Mol. Cell Biol. 5, 897-907. 237. Maly, D. J., Huang, L., and Ellman, J. A. (2002) Combinatorial strategies for targeting protein families: application to the proteases. ChemBioChem 3, 16-37. 238. Thornberry, N. A., Rano, T. A., Peterson, E. P., Rasper, D. M., Timkey, T., Garcia- Calvo, M., Houtzager, V. M., Nordstrom, P. A., Roy, S., Vaillancourt, J. P., Chapman, K. T., and Nicholson, D. W. (1997) A combinatorial approach defines specificities of members of the caspase family and granzyme B. Functional relationships established for key mediators fo apoptosis. J. Biol. Chem. 272, 17907- 17911. 239. Belmokhtar, C. A., Hillion, J., and Segal-Bendirdjian, E. (2001) Staurosporine induces apoptosis through both caspase-dependent and caspase-independent mechanisms. Oncogene 20, 3354-3362. 240. Slee, E. A., Zhu, H., Chow, S. C., MacFarlane, M., Nicholson, D. W., and Cohen, G. M. (1996) Benzyloxycarbonyl-Val-Ala-Asp(OMe) fluoromethylketone (Z- VAD.FMK) inhibits apoptosis by blocking the processing of CPP32. Biochem. J. 315, 21-24. 241. Pawson, T., and Nash, P. (2003) Assembly of cell regulatory systems through protein interaction domains. Science 300, 445-452. 242. Lee, H.-J., and Zhang J. J. (2010) PDZ domains and their binding partners: structure, specificity, and modification. Cell Commun. Signal. 8, 8. 243. Songyang, Z., Fanning, A. S., Fu, C., Xu, J., Marfatia, S. M., Chishti, A. H., Crompton, A., Chan, A. C., Anderson, J. M., and Cantley, L. C. (1997) Recognition of unique carboxyl-terminal motifs by distinct PDZ domains. Science 275, 73-77. 244. Wolde, M., Fellows, A., Cheng, J., Kivenson, A., Coutermarsh, B., Talebian, L., Karlson, K., Piserchio, A., Mierke, D. F., Stanton, B. A., Guggino, W. B., and Madden, D. R. (2007) Targeting CAL as a negative regulator of DeltaF508-CFTR cell-surface expression: an RNA interference and structure-based mutagenetic approach. J. Biol. Chem. 282, 8099-8109. 245. Cheng, S. H., Gregory, R. J., Marshall, J., Paul, S., Souza, D. W., White, G. A., O’Riodan, C. R., and Smith, A. E. (1990) Defective intracellular transport and processing of CFTR is the molecular basis of most cystic fibrosis. Cell 63, 827-834. 246. Kerem, B., Bommens, J. M., Buchanan, J. A., Markiewicz, D., Cox, T. K., Chakravarti, A., Buchwald, M., and Tsui, L. C. (1989) Identification of the cystic fibrosis gene: genetic analysis. Science 245, 1073-1080. 247. Cushing, P. R., Vouilleme, L., Pellegrini, M., Boisguerin, P., and Madden, D. R. (2010) A stabilizing influence: CAL PDZ inhibition extends the half-life of ΔF508- CFTR. Angew. Chem. Int. Ed. 49, 9907-9911. 159

248. Roberts, K. E., Cushing, P. R., Boisguerin, P., Madden, D. R., and Donald, B. R. (2012) Computational design of a PDZ domain peptide inhibitor that rescues CFTR activity. PLos Comput. Biol. 8, e1002477. 249. Kundu, R., Cushing, P. R., Popp, B. V., Zhao, Y., Madden, D. R., and Ball, Z. T. (2012) Hybrid organic-inorganic inhibitors of a PDZ interaction that regulates the endocylic fate of CFTR. Angew. Chem. Int. Ed. 51, 7217-7220. 250. Cormet-Boyaka, E., Jablonsky, M., Naren, A. P., Jackson, P. L., Muccio, D. D., and Kirk, K. L. (2004) Rescuing cystic fibrosis transmembrane conductance regulator (CFTR)-processing mutants by transcomplementation. Proc. Natl. Acad. Sci. U. S. A. 101, 8221-8226. 251. Pedemonte, N., Lukacs, G. L., Du, K., Caci, E., Zegarra-Moran, O., Galietta, L. J. V., and Verkman, A. S. (2005) Small-molecule correctors of defective ΔF508- CFTR cellular processing identified by high-throughput screening. J. Clin. Invest. 115, 2564-2571. 252. Fass, D. (2012) Disulfide bonding in protein biophysics. Annu. Rev. Biophys. 41, 63-79. 253. Adessi, C., and Soto, C. (2002) Converting a peptide into a drug: strategies to improve stability and bioavailability. Curr. Med. Chem. 9, 963-978. 254. Gongora-Benitez, M., Tulla-Puche, J., and Albericio, F. (2014) Multifaceted roles of disulfide bonds. Peptides as Therapeutics. Chem. Rev. 114, 901-926. 255. Cascales, L., Henriques, S. T., Kerr, M. C., Huang, Y. H., Sweet, M. J., Daly, N. L., and Craik, D. J. (2011) Identification and characterization of a new family of cell penetrating peptides: Cyclic cell penetrating peptides. J. Biol. Chem. 286, 36932- 36943. 256. Jha, D., Mishra, R., Gottschalk, S., Wiesmuller, K. H., Ugurbil, K., Maier, M. E., and Engelmann, J. (2011) CyLoP-1: A novel cysteine-rich cell-penetrating peptides for cytosolic delivery of cargoes. Bioconjugate Chem. 22, 319-328. 257. Hruby, V. J. (2002) Designing peptide receptor agonists and antagonists. Nat. Rev. Drug Discov. 1, 847-858. 258. Tam, J. P., Wu, C. R., Liu, W., and Zhang, J. W. (1991) Disulfide bond formation in peptides by dimethyl sulfoxide. Scope and applications. J. Am. Chem. Soc. 113, 6657-6662. 259. Roberts, K. D., Lambert, J. N., Ede, N. J., and Bray, A. M. (1998) Efficient synthesis of thioether-based cyclic peptide libraries. Tet. Lett. 39, 8357-8360. 260. Cushing, P. R., Fellows, A., Villone, D., Boisguerin, P., and Madden, D. R. (2008) The relative binding affinities of PDZ partners for CFTR: a biochemical basis for efficient endocytic recycling. Biochemistry 47, 10084-10098. 261. Crabtree, G. R. (1999) Generic signals and specific outcomes: signaling through Ca2+, calcineurin, and NF-AT. Cell 96, 611-614. 262. Rao, A., Luo, C., and Hogan, P. G. (1997) Transcription factors of the NFAT family: regulation and function. Annu. Rev. Immunol. 15, 707-747. 263. Kiani, A., Rao, A., and Aramburu, J. (2000) Manipulating immune responses with immunosuppressive agents the target NFAT. Immunity 12, 359-372.

160

264. Chapman, J. R. (2011) Chronic calcineurin inhibitor nephrotoxicity-lest we forget. Am. J. Transplant. 11, 693-697. 265. Sigal, N. H., Dumont, F., Durette, P., Siekierka, J. J., Peterson, L., Rich, D. H., Dunlap, B. E., Staruch, M. J., Melino, M. R., Koprak, S. L., Williams, D., Witzel, B., and Pisano, J. M. (1991) Is cyclophilin involved in the immunosuppressive and nephrotoxic mechanism of action of cyclosporine A? J. Exp. Med. 173, 619-628. 266. Platz, K. P., Mueller, A. R., Blumhardt, G., Bachmann, S., Bechstein, W. O., Kahl, A., and Neuhaus, P. (1994) Nephrotoxicity following orthotopic liver transplantation. A comparison between cyclosporine and FK506. Transplantation 58, 170-178. 267. Hojo, M., Morimoto, T., Maluccio, M., Asano, T., Morimoto, K., Lagman, K., Shimbo, T., and Suthanthiran, M. (1999) Cyclosporine induces cancer progression by a cell-autonomous mechanism. Nature 397, 530-534. 268. Aramburu, J., Garcia-Cozar, F., Raghavan, A., Okamura, H., Rao, A., and Hogan, P. G. (1998) Selective inhibition of NFAT activation by a peptide spanning the calcineurin targeting site of NFAT. Mol. Cell 1, 627-637. 269. Li, H., Rao, A., and Hogan, P. G. (2011) Interaction of calcineurin with substrates and targeting proteins. Trends Cell Biol. 21, 91-103. 270. Roy, J., and Cyert, M. S. (2009) Cracking the phosphatase code: Docking interactions determine substrate specificity. Sci. Signal. 2, re9. 271. Grigoriu, S., Bond, R., Cossio, P., Chen, J. A., Ly, N., Hummer, G., Page, R., Cyert, M. S., and Peti, W. (2013) The molecular mechanism of substrate engagement and immunosuppressant inhibition of calcineurin. PLoS Biol. 11, e1001492. 272. Sieber, M., and Baumgrass, R. (2009) Novel inhibitors of the calcineurin/NFATc hub - alternatives to CsA and FK506? Cell Commun. Signal. 7, 25. 273. Li, H., Zhang, L., Rao, A., Harrison, S. C., and Hogan, P. G. (2007) Structure of calcineurin in complex with PVIVIT peptide: portrait of a low-affinity signaling interaction. J. Mol. Biol. 369, 1296-1306. 274. Takeuchi, K., Roehrl, M. H., Sun, Z. Y., and Wagner, G. (2007) Structure of the calcineurin-NFAT complex: defining a T cell activation switch using solution NMR and crystal coordinates. Structure 15, 587-597. 275. Li, H., Rao, A., and Hogan, P. G. (2004) Structural delineation of the calcineurin- NFAT interaction and its parallels to PP1 targeting interactions. J. Mol. Biol. 342, 1659-1674. 276. Davies, S. J., Ayscough, A. P., Beckett, R. P., Clements, J. M., Doel, S., Pratt, L. M., Spavold, Z. M., Thomas, S. W., and Whittaker, M. (2003) Structure-activity relationships of the peptide deformylase inhibitor BB-3497: modification of the P2’ and P3’ side chains. Bioorg. Med. Chem. Lett. 13, 2715-2718. 277. Llinas-Brunet, M., Bailey, M. D., Ghiro, E., Gorys, V., Halmos, T., Poirier, M., Rancourt, J., and Goudreau, N. (2004) A systematic approach to the optimization of substrate-based inhibitors of the hepatitis C virus NS3 protease: discovery of potent and specific tripeptide inhibitors. J. Med. Chem. 47, 6584-6594.

161

278. Sigman, M. S., and Jacobsen, E. N. (1998) Schiff base catalysts for the asymmetric Strecker reaction identified and optimized from parallel synthetic libraries. J. Am. Chem. Soc. 120, 4901-4902. 279. Bold, G., Fassler, A., Capraro, H.-G., Cozens, R., Klimkait, T., Lazdins, J., Mestan, J., Poncioni, B., Rusel, J., Stover, D., Tintelnot-Blomley, M., Acemoglu, F., Beck, W., Boss, E., Eschbach, M., Hurlimann, T., Masso, E., Roussel, S., Ucci-Stoll, K., Wyss, D., and Lang, M. (1998) New aza-dipeptide analogues as potent and orally absorbed HIV-1 protease inhibitors: candidates for clinical development. J. Med. Chem. 41, 3387-3401. 280. Perni, R. B., Almquist, S. J., Byrn, R. A., Chandorkar, G., Chaturvedi, P. R., Coutney, L. F., Decker, C. J., Dinehart, K., Gates, C. A., Harbeson, S. L., Heiser, A., Kalkeri, G., Kolaczkowski, E., Lin, K., Luong, Y.-P., Rao, B. G., Taylor, W. P., Thomson, J. A., Tung, R. D., Wei, Y., Kwong, A. D., and Lin, C. (2006) Preclinical profile of VX-950, a potent, selective, and orally bioavailable inhibitor of hepatitis C Virus NS3-4A serine protease. Antimicrob. Agents Chemother. 50, 899-909. 281. Wedemeyer, W. J., Welker, E., and Scheraga, H. A. (2002) Proline cis-trans isomerization and protein folding. Biochemistry 41, 14637-14644. 282. Dumy, P., Keller, M., Ryan, D. E., Rohwedder, B., Wohr, T., and Mutter, M. (1997) Pseudo-prolines as a molecular hinge: reversible induction of cis amide bonds into peptide backbones. J. Am. Chem. Soc. 119, 918-925. 283. Chierici, S., Jourdan, M., Figuet, M., and Dumy, P. (2004) A case study of 2,2- dimethylthiazolidine as locked cis proline amide bond: synthesis, NMR and molecular modeling studies of a δ-conotoxin EVIA peptide analog. Org. Biomol. Chem. 2, 2436-2441. 284. Wohr, T., Wahl, F., Nefzi, A., Rohwedder, B., Sato, T., Sun, X., and Mutter, M. (1996) Pseudo-prolines as a solubilizing, structure-disrupting protection technique in peptide synthesis. J. Am. Chem. Soc. 118, 9218–9227. 285. Luechapanichkul, R., Chen. X., Taha, H. A., Vyas, S., Guan, X., Freitas, M. A., and Pei, D. (2013) Specificity profiling of dual specificity phosphatase vaccinia VH1- related (VHR) reveals two distinct substrate binding modes. J. Biol. Chem. 288, 6498-6510. 286. Gwack, Y., Sharma, S., Nardone, J., Tanasa, B., Iuga, A., Srikanth S., Okamura, H., Bolton, D., Feske, S., Hogan, P. G., and Rao, A. (2006) A genome-wide Drosophila RNAi screen identifies DYRK-family kinases as regulators of NFAT. Nature 441, 646-650. 287. Kaduk, C., Wenschuh, H., Beyermann, M., Forner, K., Carpino, L. A., and Bienert, M. (1995) Synthesis of Fmoc-amino acid fluorides via DAST, an alternative fluoridation agent. Lett. Pep. Sci. 2, 285-288. 288. Kang, S., Li, H., Rao, A., and Hogan, P. G. (2005) Inhibition of the calcineurin- NFAT interaction by small organic molecules reflects binding at an allosteric site. J. Biol. Chem. 280, 37698-37706. 289. Pettersen, E. F., Goddard, T. D., Huang, C. C., Couch, G. S., Greenblatt, D. M., Meng, E. C., and Ferrin, T. E. (2004) UCSF Chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 13, 1605-1612. 162

290. Sorin, E. J., and Pande, V. S. (2005) Exploring the helix-coil transition via all-atom equilibrium ensemble simulations. Biophys. J. 88, 2472-2493. 291. DePaul, A. J., Thompson, E. J., Patel, S. S., Haldeman, K., and Sorin, E. J. (2010) Equilibrium conformational dynamics in an RNA tetraloop from massively parallel molecular dynamics. Nucleic Acids Res. 38, 4856-4867. 292. Wang, J., Wang, W., Kollman, P. A., and Case, D. A. (2006) Automatic atom type and bond type perception in molecular mechanical calculations. J. Mol. Graphics Model. 25, 247-260. 293. Wang, J., Wolf, R. M., Caldwell, J. W., Kollman, P. A., and Case, D. A. (2004) Development and testing of a general AMBER force field. J. Comput. Chem. 25, 1157-1174. 294. Sousa da Silva, A. W., and Vranken, W. F. (2012) ACPYPE-AnteChamberPYthon Parser interfacE. BMC Res. Notes 5, 367. 295. Jojart, B., and Martinek, T. A. (2007) Performance of the general amber force field in modeling aqueous POPC membrane bilayers. J. Comput. Chem. 28, 2051-2058. 296. Jorgensen, W. L., and Madura, J. D. (1983) Solvation and Conformation of Methanol in Water. J. Am. Chem. Soc. 105, 1407-1413. 297. Humphrey, W., Dalke, A., and Schulten, K. (1996) VMD: visual molecular dynamics. J. Mol. Graph. 14, 33-38.

163