New paradigm for drug design: Design and synthesis of novel biologically active peptides that are agonists at opioid receptors and antagonists at cholecystokinin receptors

Item Type text; Dissertation-Reproduction (electronic)

Authors Agnes, Richard S

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 29/09/2021 20:04:47

Link to Item http://hdl.handle.net/10150/280340 NEW PARADIGM FOR DRUG DESIGN: DESIGN AND SYNTHESIS OF

NOVEL BIOLOGICALLY ACTIVE PEPTIDES THAT ARE AGONISTS AT

OPIOID RECEPTORS AND ANTAGONISTS AT CHOLECYSTOKININ

RECEPTORS

by

Richard Sario Agnes

Copyright © Richard Sario Agnes 2003

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF CHEMISTRY

In Partial Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2003 UMI Number: 3106966

Copyright 2003 by Agnes, Richard Sario

All rights reserved.

UMI

UMI Microform 3106966 Copyright 2004 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 9

THE UNIVERSITY OF ARIZONA ® GRADUATE COLLEGE

As members of the Final Examination Committee, we certify that we have read the dissertation prepared by Sario Agnes

entitled NEW PARADIGM FOR DRUG DESIGN: DESIGN AND SYNTHESIS OF NOVEL

BIOLOGICALLY ACTIVE PEPTIDES THAT ARE AGONISTS AT OPIOID

RECEPTORS AND ANTAGONISTS AT CHOLECYSTOKININ RECEPTORS

and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy

Victor J Date F-- Robert R. Bates Date

Rich^d S. Glass Date

Vicki U,ysQ«rki ^ Date/ I

S. Scott Saavedra Date ^

Final approval and acceptance of this dissertation is contingent upon the candidate's submission of the final copy of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requireme

Disserta®Lon Director ^ictor J. Hruby Date 3 STATEMENT BY THE AUTHOR

The dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University library to be made available to borrowers under rules of the library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the copyright holder.

SIGNED: 4 ACKNOWLEDGEMENTS

A PhD research project would be very difficult to complete without the kind help of many individuals. I would like to address my appreciation to some of the people without whom this dissertation would never have been possible. I would like to thank Dr. Victor J. Hruby, my research advisor, for his encouragement, friendship, support, understanding, and mentorship throughout the years. Special thanks for the members of my dissertation committee. Dr. Richard Glass, Dr. Robert B. Bates, Dr. Scott Saavedra, Dr. Vicki Wysocki, and the late Dr. David F. O'Brien, for their scientific expertise, encouragement, and support. I would like to thank the core members of the CCK/opioid project, especially to Dr. Frank Porreca and Dr. Josephine Lai, whose insight and vision, along with Dr. Hruby, made this project possible. Thanks to Dr. Yeon Sun Lee and Dr. Balaz Hargaittai for all the discussions and help. Thanks to Peg Davis, Dr. Shou Wu Ma, and Dr. Todd Vanderah for performing the biological assays. Benjamin A. Fish is greatly appreciated for his peptide synthesis work. I would like to express my gratitude to all the members of the Hruby group, Paolo Greico, Andrew Burritt, Preeti Balse, Josue-Alfaro Lopez, Scott Cowell, Toru Okayama, Guoxia Han, Jung Mo Ahn, Josef Vagner, Xue Jun Tang, Xuyuan Gu, John Ndungu, Isabel Alves, Min Ying Cai, Ruben Vardanyan, Zhanna Zhilina, Jinfa Ying, Katalin Kover, Chiyi Xiong, and Junyi Zhang, for their help, support, and friendship. I would like to thanks to all the fnends I have met during these years and in particular, Matthew Golden, Martina Bowen, Anne Runge, Jeimet Lee, Eric Ross, Lisa, Domenic Tiani and Matt Lynn, for their support, encouragement, and friendship. Special thanks to Cheryl McKinley and Margie Colie for their help, support and encouragements. I would to thank Arpad Somogyi and the facility staff for performing the mass spectrometry analysis and Wallace Clark for performing the amino acid analysis. This research was supported by NIDA Grant DA 12394. Thanks to Pfizer Global Research and Development for a 2002 Organic Chemistry Diversity Fellowship. DEDICATION

To my mother

Maximina Sario Agnes

To my father

Ricarte A. Agnes

To my brother and sister

Gilbert Agnes and Joanne Agnes

To my grandmother

Juana Agnes

For all their encouragements, support, and prayers 6 TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS 4

LIST OF FIGURES 13

LIST OF TABLES 15

ABSTRACT 18

CHAPTER 1. INTRODUCTION 19

1.1 New paradigm in drug discovery 19

1.2 Neuropathic pain and roles of CCK and opioid systems 23

1.3 Design concepts for agonists and antagonists 24

1.4 Structure-activity relationships of opioid agonists 30

1.4.1 Importance of N-terminal Tyr at position 1 30

1.4.2 Importance of D-amino acids at position 2 31

1.4.3 Importance of the aromatic residues at position 4 31

1.4.4 Substitution at position 5 32

1.4.5 Cyclic analogues of enkephalin 33

1.5 Structure-activity relationships of cholecystokinin 33

1.5.1. Substitutions for Met residues at positions 28 and 31 35

1.5.2 Modifications of Trp residue 36

1.5.3 Cyclic analogues of CCK 36

1.5.4 Structure activity of non-peptide cholecystokinin antagonists 37 7 TABLE OF CONTENTS—CONTINUED

1.6 Overlapping pharmacophores of opioids and cholecystokinin 38

1.7 Rationale for a single molecule targeting multiple receptors 40

1.8 Goals 41

CHAPTER 2. DESIGN, SYNTHESIS, AND STRUCTURE-ACTIVITY

RELATIONSHIPS OF CCK/OPIOID PEPTIDES: MODIFICATIONS

OF THE LINEAR SEQUENCE 42

2.1 Results: design of linear peptides 42

2.2 Results and discussion: synthesis of linear analogues of CCK/opioid

peptides 46

2.2.1 General methods 46

2.2.2 Results: Coupling of Trp"^ and N-MeNle^ 46

2.2.3 Results: Cleavage of linear peptides from the solid support 49

2.2.4 Results: Incomplete deprotection of Trp 50

2.3 Results: Structure-activity relationships of linear CCK/opioid analogues

at CCK and opioid receptors 51

2.3.1 Truncation of A^-terminal Asp 52

2.3.2 Substitution at position 2 60

2.3.3 Substitution with Nle^ 61

2.3.4 Substitution with D-Trp"^ 63

2.3.5 Double substitution with D-Trp"* and Nle^ 65 8 TABLE OF CONTENTS—CONTINUED

2.4 Discussions 66

2.4.1 Assessment of the binding affinity and functional assays 66

2.4.2 RSA504 is the most potent analogue of CCK at CCK-A

receptors 67

2.4.3 SAR along the peptide backbone at positions 4 and 5 68

2.4.4 Role of D-amino acids at position 2 69

2.4.5 Similarities in opioid and CCK SAR supports the overlapping

pharmacophore hj^othesis 70

2.5 Conclusions 71

2.6 Experimental Section 73

2.6.1 Abbreviations 73

2.6.2 Materials 73

2.6.3 General methods for peptide synthesis 75

2.6.4 Purification 77

CHAPTER 3. DESIGN, SYNTHESIS, AND STRUCTURE-ACTIVITY

RELATIONSHIPS OF CCK/OPIOID PEPTIDES:

SUBSTITUTION OF BULKY RESIDUES 79

3.1 Results: Design ofpeptides with substitution of bulky residues 79

3.2 Results: Structure-activity relationships 84

3.2.1 Substitution of Nal residues at position 4 91 9 TABLE OF CONTENTS—CONTINUED

3.2.2 Substitution of 5'-phenyltryptophan at position 4 93

3.2.3 Substitution of dihydrotryptophan at position 4 94

3.2.4 Substitution of D-homophenylalanine at position 2 96

3.3 Discussions 97

3.4 Conclusions 99

CHAPTER 4. DESIGN, SYNTHESIS, AND STRUCTURE-ACTIVITY

RELATIONSHIPS OF CCK/OPIOID PEPTIDES: CYCLIC

DISULFIDE ANALOGUES 100

4.1 Results: Design of cyclic disulfide analogues of CCK/opioid peptide.. 100

4.2 Results and Discussion: Synthesis of the cyclic disulfide analogues of

CCK/opioid peptide ligands 106

4.3 Results: Structure-activity relationships of the cyclic disulfide

analogues Ill

4.4 Discussions 121

4.4.1 Assays 121

4.4.2 Comparison of cyclic disulfide CCK/opioid peptide analogues

to opioid ligands 122

4.4.3 Substitution with D-Trp in cyclic disulfide CCK/opioid peptide

analogues 123

4.5 Conclusions 124 10 TABLE OF CONTENTS—CONTINUED

4.6 Experimental Section 125

4.6.1 General method for the synthesis of linear peptides for solution

phase cyclization 125

4.6.2 General method for potassium ferricyanide disulfide

cyclization 125

4.6.3 General method for air oxidation in disulfide cyclization 126

4.6.4 General method for DMSO assisted air oxidation in disulfide

cyclization 127

4.6.5 General methods for cyclic disulfide synthesis using on-resin

cyclization 127

CHAPTER 5. DESIGN, SYNTHESIS, AND STRUCTURE-

ACTIVITY RELATIONSHIPS OF CCK/OPIOID

PEPTIDES: LACTAM ANALOGUES 133

5.1 Results: Design of lactam analogues 133

5.2 Results: Synthesis of lactam analogues 137

5.3 Results: Structure-activity relationships 143

5.4 Discussions 154

5.4.1 Assessment of the bioassays 155

5.4.2 Role of position 2 in CCK-antagonist properties 156

5.4.3 Comparison to opioid ligands 157 11 TABLE OF CONTENTS—CONTINUED

5.4.4 Differential opioid agonist and CCK antagonist activities 158

5.4.5 Comparison of cyclic lactam to cyclic disulfide 159

5.4.6 Compatibility of D-Trp with non-methylated residue at

position 5 in cyclic lactam 160

5.5 Conclusions 161

5.6 Experimental section 162

5.6.1 Abbreviations 162

5.6.2 Materials 162

5.6.3 General method for peptide synthesis 164

5.6.4 General method for cyclization via lactam bridge 166

5.6.5 Purification 167

CHAPTER 6. SOLUTION STRUCTURE OF A CYCLIC DISULFIDE

ANALOGUE OF CCK/OPIOID PEPTIDES BY NMR: A

COMAPATIVE STUDY WITH OPIOID AND

CHOLECYSTOKININ LIGANDS 169

6.1 Background 169

6.2 Results: NMR chemical shift assignments and coupling constants for

RSAlOlc 170

6.3 Results: Temperature dependence study 175

6.4 Results: Side-chain conformations 177 12 TABLE OF CONTENTS—CONTINUED

6.5 Results: Molecular dynamics 183

6.6 Results: Comparison to CCK-8 and DPDPE 186

6.7 Experimental section 189

CHAPTER 7. CONCLUSIONS AND FUTURE WORK 191

7.1 Conclusions 191

7.2 Future work 194

REFERENCES 196

APPENDIX A: Methods for color tests and bioassay data 214

APPENDIX B: Rf from thin layer chromatography, retention times and k' values from

RP-HPLC, and amino acid analysis, high resolution mass spectrometry

data 220

APPENDIX C: NMR spectra and temperature coefficients 235 13 LIST OF FIGURES

Figure 1.1 The current model for drug discovery of therapeutics for the treatment of

diseases 20

Figure 1.2 Proposed paradigm for drug discovery of therapeutics for the treatment of

diseases 22

Figure 1.3 The relationship between the levels and/or release of CCK 24

Figure 1.4 The peptide backbone 27

Figure 1.5 Conformations along the peptide backbone displaying trans- and cis-

amide bonds 27

Figure 1.6 Definitions of (p, \|/, and co angles 28

Figure 1.7 Some possible templates for cyclic disulfide and lactam analogues 29

Figure 1.8 Stereoviews of proposed bioactive conformations of DPDPE (top) and

CCK (bottom) at delta and opioid receptors 39

Figure 1.9 Tifluadom, a nonpeptide benzodiazepine ligand with interacts with opioid

and CCK receptors 40

Figure 2.1 Design of CCK/opioid analogues 44

Figure 2.2 Solid phase synthesis of linear analogues of CCK/opioid peptides 48

Figure 2.3 Reduction of indole ring to indoline 50

Figure 2.4 Incomplete deprotection of N'-Boc protected Trp results to N'-carbamic

acid 51

Figure 2.5 Structures of phenylethylester analogues of CCK 69

Figure 3.1 Examples of nonpeptide CCK antagonists 81 14 LIST OF FIGURES—CONTINUED

Figure 3.2 Structures of bulky analogues substituted at positions 4 and 2 82

Figure 4.1 Stereoview of lowest energy conformation of RSA504 103

Figure 4.2 Design of cyclic disulfide analogues 104

Figure 4.3 Synthesis of cychc peptide by K3Fe(CN6) 107

Figure 4.4 Synthesis of cyclic peptide by DMSO assisted air oxidation 108

Figure 4.5 Synthesis of cyclic peptide by on-resin cyclization with Tl(tfa)3 109

Figure 5.1 Design of lactam analogues 135

Figure 5.2 A scheme for synthesis of conformationally constrained CCK/opioid

analogue with Alloc protected Lys^ and Ally protected Glu^ 139

Figure 5.3 Side reaction during cyclization with HBTU 140

Figure 5.4 Structure of JMV-320 156

Figure 6.1 Newman projections of the side chain rotamer populations of a-amino

acids about the x torsional angle 178

Figure 6.2 The stereoview of the lowest energy structure of RSAlOlc 184

Figure 6.3 Structure of RSAlOlc demonstrating 7.4 A distance between the centroids

ofTyr^ andTrp'^ 185

Figure 6.4 Superimposed structures of RSA 101c and CCK-8 187

Figure 7.1 Proposed structures determining the effect of N-methylation in lactam and

cyclic disulfide analogues of CCK/opioid peptides 194

Figure 7.2 Proposed structure incorporating P-Me Trp substituted in lactam

analogues 195 15 LIST OF TABLES

Table 2.1 Linear CCK/opioid peptides based on the [desAsp°] SNF-9007 45

Table 2.2 Binding affinities of CCK/opioid peptides at the human CCK-A and CCK-

B receptors 54

Table 2.3 Functional activities of linear CCK/opioid peptides at the CCK

receptors 55

Table 2.4 Binding affinities of linear CCK/opioid peptides at opioid receptors 56

Table 2.5 Functional activities of CCK/opioid peptides at opioid receptors 57

Table 2.6 Functional activities of linear CCK/opioid peptides at the MVD and

GPI 58

Table 2.7 CCK functional activities of linear CCK/opioid peptides at the

unstimulated GPI/LMMP 59

Table 3.1 CCK/opioid analogues with bulky residue substituted at

positions 4 and 2 83

Table 3.2 Binding affinities of CCK/opioid peptide analogues at the human CCK-A

and CCK-B receptors 85

Table 3.3 Functional activities of CCK/opioid peptide analogues at the CCK

receptors 86

Table 3.4 Binding affinities of CCK/opioid peptide analogues at the opioid

receptors 87 16 LIST OF TABLES—CONTINUED

Table 3.5 Functional activities of CCK/opioid peptide analogues at the opioid

receptors 88

Table 3.6 Functional activities of CCK/opioid peptide analogues at the

MVD and GPI 89

Table 3.7 Functional activities of CCK/Opioid peptide analogue at the unstimulated

GPl/LMMP 90

Table 4.1 Cyclic disulfide analogues designed to interact with CCK and opioid

receptors 105

Table 4.2 Binding affinities of CCKVopioid peptides at the opioid receptors 113

Table 4.3 Functional activities of cyclic disulfide CCK/opioid peptide analogues at

the opioid receptors 114

Table 4.4 Functional activities of cyclic disulfide CCK/opioid peptide analogues at

the MVD and GPI 115

Table 4.5 Binding affinities of cyclic disulfide CCK/opioid peptide analogues at the

CCK receptors 116

Table 4.6 Functional activities of cyclic disulfide CCK/opioid peptide analogues at

the CCK receptors 117

Table 4.7 Functional activity of CCK/opioid peptides at the GPI/LMMP 118

Table 5.1 Lactam analogues 136

Table 5.2 Various conditions used for the lactam bridge formation 142

Table 5.3 Binding affinities of CCK/opioid peptides at the opioid receptors 145 17 LIST OF TABLES—CONTINUED

Table 5.4 Functional activities of CCK/opioid peptides at the opioid receptors....146

Table 5.5 Binding affinities of CCK/opioid peptides at the CCK receptors 147

Table 5.6 Functional activities of CCK/opioid peptides at the CCK receptors 148

Table 5.7 Functional activities of CCK/opioid peptides at the MVD and GPI tissue

assays 149

Table 5.8 Functional activities of CCK/opioid peptides at the GPI/LMMP 150

Table 6.1 chemical shifts for RSAlOlc 172

Table 6.2 Coupling constants for RSAlOlc 173

Table 6.3 Comparison of RSAlOlc chemical shift non-equivalence of Gly Ha to

reported literature values for analogous peptides 174

Table 6.4 Temperature coefficients of NHa for RSAlOlc in DSMO-Je 176

Table 6.5 Calculated side chain rotamer populations (%) about the Ca-Cp bond (xi)

in RSAlOlc 179

Table 6.6 Calculated rotamers for Tyr and Trp to analogous peptides in

DMSO-4 182

Table 6.7 The dihedral angles of the minimized mean for RSAlOlc 185

Table 6.8 Comparison of backbone dihedral angles of RSAlOlc to reported dihedral

angles of analogous peptides 188 18 ABSTRACT

We now know from genomics that many disease states lead to changes in expressed proteins (adaptation/plasticity). Therefore, drug design and discovery based on normal states and single targets often is inadequate or even counter-indicated. Therefore, the "system changes" that have occurred must be considered in any treatment for the disease. Such "systems changes" are clearly evident in neuropathic pain where opioids can actually heighten pain. In these pain states there are increased levels of neurotransmitters such as cholecystokinin (CCK) in which both the peptides and their receptors are increased in pain states.

To effectively treat diseases involving "systems changes" a new paradigm for the design of compounds was proposed. In this new approach single peptide or peptidomimetic molecules are designed to interact with multiple receptor targets. For the treatment of pain, a series of linear and cyclic peptides and peptidomimetics were designed based on the overlapping pharmacophores of opioid and CCK ligands. The

CCK/opioid analogues were synthesized and evaluated for their biological activities.

Several of the CCKVopioid analogues were found to simultaneously interact with opioid receptors as agonists and CCK receptors as antagonists. In addition, the lead compounds have been tested in several pain models and were found to be promising in the treatment of neuropathic pain. Further, the structure-activity relationships of these novel peptides have provided new insights into the requirements for binding and bioactivity at opioid and CCK receptors, as well as the overlapping pharmacophores of CCK and enkephalin. 19 CHAPTER 1

INTRODUCTION

1.1 New paradigm in drug discovery

In recent years, there is growing evidence that Ugands or drugs behave differently in normal states than in pathological states. Thus drugs that are effective in normal states are not effective in pathological disease states. Figure 1.1 summarizes the current paradigm in drug design and discovery which can be described as "one disease-one- receptor-one drug" format. ^Once the biological activity (e.g. disease) is identified, the causative agent (e.g. gene/ cell/ tissue/ protein/ receptor) associated with the disease is identified. From this, biological assays assessing binding affinities and functional assays

(e.g. secondary messenger systems) are developed. An assay is developed to assess binding and functional activities. Often, these assays utilize cells that are "normal."

These become targets for the assessment of potential drugs.

There are a few problems with the current paradigm for drug discovery for the treatment of diseases. First, the current model does not reflect our emerging understanding of diseases, especially in diseases concerning the central nervous system where systemic adaptations (or changes) occur, maybe caused by several interacting systems. Second, in vitro and in vivo bioassays often use normal cells and animals, which do not accurately reflect bioactivities during the disease state. Finally, since the 20 disease is caused by more than one receptor, drug design targeting only one receptor would be ineffective.

IDENTIFY BIOLOGICAL ACTIVITY/ DISEASE

IDENTIFY THERAPEUTIC TARGET/ OR CAUSATIVE AGENT: ONE GENE, PROTEIN, ENZYME OR RECEPTOR

DEVELOP AN ASSAY BINDING AND ACTIVITY

DRUG DESIGN AND SYNTHESIS

Figure 1.1. The current model for drug discovery of therapeutics for the treatment of diseases.

To have more effective drugs for treatment of diseases, a new paradigm for drug discovery has been proposed. Figure 1.2 outlines the general ideas involved. It has been proposed that (1) a new paradigm is necessary for drug discovery, in particular where 21 adaptive changes occur in the central nervous system; (2) potential drugs should be designed for optimal activity in the disease-state to overcome the adaptation that is involved in the disease; (3) bioassays, particularly for in vivo testing, should reflect the disease states; and (4) rational drug design requires binding and activity at more than one target with one ligand. Because disease state may involve several interacting systems,

design of drugs may consider several targets, such that the drug is designed to interact with more than one target. 22

IDENTIFY BIOLOGICAL ACTIVITY AND DISEASE

IDENTIFY SYSTEM CHANGES: MULTIPLE TARGETS: GENES, ENZYMES, RECEPTORS, AND PROTEINS

DEVELOP AN ASSAY BINDING AND FUNCTIONAL ASSAYS AND IN VIVO ASSAYS FOR DISEASE STATES

DRUG DESIGN CONSIDER MULTIPLE TARGETS

Figure 1.2. A proposed paradigm for drug discovery of therapeutics for the treatment of diseases.

The main goal of this project is to apply this new approach to drug discovery in the design and synthesis of novel biologically active compounds for the treatment of pain in the disease state. 23 1.2 Neuropathic pain and roles of CCK and opioid systems

As mentioned earUer, the new paradigm for the drug discovery would be particularly applicable to diseases that involve adaptive changes in the central nervous system. Neuropathic pain, which is an abnormal pain (e.g. hyperalgesia and allodynia) elicited by peripheral nerve injury, is such a disease. Neuropathic pain is one of the most difficult pains to treat, as current treatment with alkaloids have not been efficacious.

Although the mechanism underlying neuropathic pain is not completely understood, recent studies reveal that several changes, including increased activities of neuromodulators and neurotransmitters, occur in the central nervous system. ^ Similar

observations are found in opioid tolerance in which treating pain with opioids actually

intensifies pain. ^

One of these neuromodulators in the disease state of pain is cholecystokinin

(CCK), which has long been considered as an "antiopioid" peptide because it diminishes

opioid induced analgesia. The role of CCK and its interaction with opioids and in pain

pathways have been reviewed. Briefly, CCK and endogenous opioids and their

corresponding receptors have overlapping distribution in the central nervous system.

Acute and chronic administration of an opioid agonist increases levels of CCK and

CCK receptors in the spinal cord. Administration of CCK diminishes the acute

effects of opioid agonists. Administration of a CCK antagonist enhanced the acute

1^10 effects of morphine. " Opioid antagonist have blocked the effects of CCK antagonist

action. 19CCK antagonists have also diminished opioid tolerance and dependence. 20 22 24

NORMAL

CCK level/ release Morphine analgesia

NEUROPATHY

Figure 1.3. The relationship between the levels and/or release of cholecystokinin (CCK) and the changes in the effectiveness of morphine in neuropathic pain models, as adapted from Stanfa et al. ^

1.3 Design concepts for agonists and antagonists

Design of peptide receptor agonists and antagonists are well reviewed^^"^^ and only limited, relevant concepts for the design of the opioid/CCK analogues will be discussed.

Most naturally occurring peptide hormones and neurotransmitters are agonists.

Agonists are peptides whose binding interaction with their respective receptors results in an active response of the cell (i.e. the secondary message system is activated) or organ 25 (function of tissue). Modifications of peptide agonists are often a result of a desire to obtain: (1) increased potency, (2) increased specificity and selectivity for receptor subtypes, and (3) enhanced drug-like properties (increased stability against protease degradation, permeability to blood brain barrier, etc.). In contrast, antagonists are molecules which inhibit the active response of the cell. In most cases, the most effective way to do this is to develop peptide analogues that are competitive antagonists of the natural ligand at the receptor.

The design of agonists often starts with truncation. This sequential removal of N- and/or C-terminal residues allows the determination of the minimum required residues

(primary structure) that allow binding and activity (important for antagonists).

Next, the relative importance of particular functional groups in the primary peptide sequence can be determined by carrying out various amino acid scans— specifically glycine and alanine scans. In terms of exploring the importance of a side- chain group for biological activity, it is often useful to modify the steric, aromatic, hydrophobic, hydrophilic, and acid-base properties of key side chain moieties. With these substitutions, it is just as important to take note of the residues that do not drastically reduce biological activity, because these residues will be important in further modifications, functionalization, as well as key sites for the construction of cyclic structures (which will be described later in detail).

After determining the minimum sequence and key amino acid residues that are

required for agonist activities, the next step is to establish the biologically active

conformation of the peptide. Small modifications of the peptide backbone 26 ((|), (p, CO angles) by introducing a local conformational constraint give important insights into the structural basis of agonist biological activity. Figure 1.4. These can be achieved by modifications of at the a-carbon, the amide nitrogen of the peptide backbone, and the

side chain moiety (e.g. (3-carbon).

In a D-amino acid scan each L-amino acid is replaced by its corresponding D-

amino acid. D-amino acids can stabilize or destabilize secondary structures such as

reverse turns and helices. This can provide insights about the important side chain

chirality of amino acid residues in the peptide, as well as certain conformations that

might be important for the peptide's biological activity.

Another design consideration is the use of N-methylation scan. N-Methylation in

peptide design has several potential advantages related to the conformational and

physical-chemical consequences of such substitution. It reduces the energy difference

between cis and trans amides, providing accessibility for the cis peptide bond

conformation. Figure 1.5. N-Methylation also eliminates the proton donating ability of

the amide bond. Both D-amino acid substitution and N-methylation allow for a more

stable backbone by preventing degradation due to enzymes. N-Methylated analogues of

CCK have been studied. 27

Figure 1.4. The peptide backbone.

trans-axmdQ bond cw-amide bond

Figure. 1.5. Conformations along the peptide backbone displaying trans- and cis- amide bonds. 28

Figure 1.6. Definitions of the (j), (p, co angles.

Although introduction of local constraints can lead to peptides with interesting biological activities, their use often is rather limited without further constraint in the peptides global conformation. The main approach is by cyclization which can induce (or stabilize) secondary structures. A useful strategy is to choose amino acid residues (from

amino acid scans) that have been shown to be tolerant to substitutions. There are several

types of cyclization templates (e.g. head-to-tail, side chain-to-side chain, side chain-to- end, and backbone to side chain) and types of fiinctional groups forming the ring

structure (e.g. lactam and disulfides. Figure 1.7). 29 0 0 0 0 H2N-C14C-N-CI4C-N-CHC-N-CI-IC-OH I ^ R ^ R H I CHo

cyclic disulfide

0000 O O o o II II II II H2N-CI-lC-N-CI-tC-N-CI-tC-N-Ch4C-0H HzN-CHC-N-CHC-N-CI-IC-N-CHC-OH 1 Hi Hi H I H H R H (CH2) (CH2) (CH2) (CH2) -NH HN- \ 0^ O

side chain-to-side chain cyclic lactams

R 0 0 0 O R o 1 I II HN-CHC-N-CHC-N-CHC-OH H2N-CHC-N-CHC-N-CHC'^'' n H R Hi I Hi H ° ^ (CH2) (CH2)

' NH O^

side chain-to-end terminal cyclic lactams

Figure 1.7. Some possible templates for cychc disulfide and lactam peptide analogues. 30

1.4 Structure-activity relationships of opioid agonists

The structure-activity relationships of enkephalin analogues at the opioid receptors have been reviewed extensively. Other key concepts important for the design of CCK/opioid peptide analogues based on the enkephalins will be discussed.

Hughes and Kosterlitz^' isolated the peptides, [Leu^]-enkephalin (Tyr^-Gly^-Gly^-

Phe'^-Leu^-OH) and [Met^]-enkephalin (Tyr^-Gly^-Gly^-Phe'^-Met^-OH), from porcine brain. The compounds have similar the same biological activity at tissue preparations such as electrically stimulated mouse vas deferens (MVD) and guinea pig ileum (GPI) assays. MVD is populated with delta-opioid receptors, while GPI is populated with mu- opioid receptors. Extensive structure-activity relationships were done to identify the chemical moieties and structural requirements essential for enkephalin activity.

1.4.1 Importance of N-terminal Tyr at position 1

The structural requirements at the Tyr^ position are expUcit for biological activity

in both the MVD and GPI. For the most part, only a few modifications are tolerated at

the N-terminal end of enkephalin. Truncation of the Tyr^ leads to inactive analogues.

The para-hydroxy of Tyr is important for potency. Thus, when substitution of Phe (i.e. p-

OH is removed) and other para-substituted Phe were placed in this position, there is loss

of potency and binding affinity. Removal of the N-terminal amino group leads to 31 inactive analogues in the delta and mu opioid receptors. Replacement of Tyr by other amino acids such as Gly, D-Ala, Ala, His, and Sar also yielded inactive analogues.

Contraction of the side chain of Tyr by omission of the methylene group and extension of the peptide chain by adding a methylene group lead to a loss of activity. Configuration is important and is precise for activity since replacement of Tyr^ with D-Tjt^ shows no activity. It has been proposed that the free N-terminal amine is important for interactions with the opioid receptors.

1.4.2 Importance of D-amino acids at position 2

When Gly^ in [Leu^J-enkephalin methyl ester is replaced with D-amino acid residues such as D-Ala, D-Met, D-Ser, and D-Thr, the potency increases significantly in the MVD and GPI tissue assays. Introduction of L-Ala caused a significant loss in potency when compared to the reference compounds. Substitution with D-Pro or Pro also resulted in a drastic drop in potency. N-Methylation of Gly (Sar analogue) resulted in a drop in potency. It is thought that D-amino acid substitution increases stability and protects from enzymatic degradation. ^^Interestingly, D-amino acids are found also in opioids such as dermorphin^^ and deltorphin"^*^"^^, from amphibians.

1.4.3 Importance of the aromatic residues at position 4 32 Several replacements of Phe'^ in enkephalin are often possible. Although enkephalin analogues with Gly'' and Ala"^ are inactive^®'"^^, substitutions with other aromatic residues are often tolerated. In cyclic enkephalin, para-halogen Phe substituted analogues, the potencies are very high and selective for delta opioid receptors. Position

4 can be substituted with another aromatic residue such as Trp"*^, but not with Tjr.

Aromatic residues at position 4 are not fully required to interact with opioid receptors.

For example, Phe"^ has been, particularly in mu receptors, replaced successfully with non-

aromatic side chain groups such as cyclohexylalanine (Cha). "^^Substitution of D-amino

acids such as D-Ala, D-Phe, and D-Trp at position 4 of Met-enkephalin afforded

analogues that were virtually inactive"'^''^^''^^, suggesting a conformational requirement at

position 4.

1.4.4 Substitutions at position 5

Position 5 of enkephalin can be substituted by a variety of residues. Leu^ and

Met^ can be replaced with alkyl groups, esters, or alcohols resulting in enhanced potency

at mu receptors. One interesting example is DAGO (Tj^-D-Ala-Gly-N-MePhe-

NHCH2CH2OH), which has N-methylated Phe at position 4. Replacement of position 5

in [D-Ala^] and [D-Met^]enkephalin with Pro resulted in enhanced potency at the mu

receptors. When the Pro^ analogue was compared to [Nva^]enkephalin (i.e. the

proline ring was opened and position 5 is not N-alkylated), the potency at mu receptors 33 was slightly improved. N-Methylation in position 5 of [Leu^] enkephalin had little

effect on the affinity when compared to [Leu]^.

1.4.5 Cyclic Analogues of enkephalin

Cyclic analogues of enkephalin have been studied to examine the conformational

requirements for the opioid receptors. Side chain to C-terminal end and side chain to side

chain cyclic lactam analogues of enkephalin has shown potent agonist properties at

opioid receptors. ^"^'^^Generally these lactam analogues were shown to be potent and

selective for the mu opioid receptors. Potent and delta opioid receptor selective cyclic

enkephalin analogues were developed from side chain to side chain disulfide bridges.

These enkephahn and enkephalinamide analogues were generally substituted with D-

Cys^ and D- or L-Cys^ residues. Additional local and global conformational restrictions

were introduced by substitution with Pen and/or D-Pen residues at positions at 2 and 5.

More recently, cyclic analogues were developed from lanthionine which showed

high potency at both delta and mu opioid receptors.

1.5 Structure-activity relationships of cholecystokinin

Cholecystokinin (CCK) and its structure and fiinctions are well reviewed in the

literature. Key concepts important for the design of CCK/opioid peptide analogues

based on the enkephalin model will be discussed. 34 CCK is a neuropeptide found both in the periphery and in the central nervous system. It was first isolated from the porcine duodenum as a 33 amino• acid• peptide.• 63 A number of biologically active variants were then subsequently reported, but the C- terminal CCK octapeptide amide (CCK26-33 or CCK-8, H-Asp^^-Tyr(S03H)^^-Met^^-

Gly^^-Trp^°-Met^^-Asp^^-Phe^^-NH2) is the most abundant peptide present in the brain.

CCK-8 interacts with nanomolar affinities with two different receptors designated as CCK-A and CCK-B (also denoted as CCK-1 and CCK-2, respectively). ^^These receptors belong to the G protein-coupled receptors (GPCRs) characterized by seven transmembrane domains. CCK-A receptors are located mainly in the periphery but are also found in some regions of the brain, ^^'^^while CCK-B receptors are predominantly located in the central nervous system (brain and spinal cord), ^^but also are found in the stomach and vagus nerve. Studies with selective CCK receptor antagonists have revealed that although both CCK-A and CCK-B may exist in the brain, CCK-B receptors predominate in the spinal cord of rodents whereas the CCK-A receptor subtype predominates in primates.

CCK-8 has been shown to be involved in numerous physiological functions such '7C\ *7"^ 1 A. 7^ • as satiety (feeding behavior), anxiety and panic attacks, ' memory, ' and analgesia.

76-78 35 1.5.1 Substitutions for Met residues at positions 28 and 31

It has been demonstrated that replacement of both Met^^ and Met^^ residues by the isosteric amino acid norleucine (Nle) provides sulfated CCK analogues with virtually identical biological activities to those with Met residues. These analogues should be resistant to enzymatic degradation since Nle is an unnatural amino acid. Further, peptide synthesis using Nle for Met allows for an efficient synthesis of CCK analogues since side reactions associated with Met are avoided. ^^When Met^^ and Met^^ were substituted with

Nle^® and Nle^\ such as in sulfated Boc[Nle^^,Nle^']CCK27-33, the binding and activities at CCK-A and CCK-B receptors were nearly the same as sulfated CCK-8. In attempts to make an enzyme resistant analogue, N-methylated analogues were synthesized. (Boc-

Tyr(S03H)-gNle-mGly-Trp-N-MeNle-Asp-Phe-NH2) led to a potent and CCK-B selective agonist. Unsulfated CCK-8 analogues also were designed with N- methylnorleucine (N-MeNle) residues at positions 28 and 31. The N-MeNle substitutions led to CCK analogues with exceptional potency and selectivity for CCK-B receptors. Of

particular interest is the [N-MeNle^^'^^]CCK26-33 which exhibited both high potency and

selectivity CCK-B against CCK-A receptors. NMR studies revealed that there is a

cis/trans isomerism about the N-methylnorleucine residue that maybe related to high

selectivity. In an attempt to stabilize the suggested cw-amide bond between the Trp and

Nle residues of CCK, a 1,5-disubstituted tetrazole ring was used as a cis amide bond

surrogate. These analogues resulted in a drastic loss in binding affinity, presumably

due the added steric bulk introduced. Further exploration of position 28 led to the 36 substitution with Phe resulting in SNF-9007 (Asp-Tyr-D-Phe-Gly-Trp-N-MeNle-Asp-

Phe-NH2) which displayed potent and selective activities at CCK-B receptors as well as weak binding affinities at delta opioid receptors. Interestingly, SNF-9007 displayed analgesic properties in several in vivo assays.

1.5.2 Modifications on Trp residue

Truncation studies have shown that Trp is an important residue for interaction with either the CCK-A or CCK-B receptor. In sulfated Ac-CCK-7 analogues, when Trp^*^ was substituted with D-Trp, the potency in the binding assays dropped. The NalCZ')^*^ analogue of sulfated Ac-CCK-7 was nearly equipotent as the model compound in CCK receptor binding assays, while the NalCl')^*^ analogue was significantly less potent. N-

MeTrp^*^ analogues were ten-fold less potent than the control compound. Substitution of all diastereoisomers of topographically constrained beta-methyl substituted Trp SI resulted in a loss in binding the CCK receptors. The loss of binding and activities resulting from the modification of Trp^'' suggest a limited tolerance for modification of this residue.

1.5.3 Cyclic analogues of CCK

Structural studies of linear CCK-8 suggested a turn in the structure. ^^To stabilize this turn, cyclic analogues were synthesized by modifying positions 26 and 29 to form 37 lactam bridges, resulting in CCK-B selective analogues. Cyclic analogues of CCK-4 with selectivity• • for CCK-B had been developed as well. 87

1.5.4 Structure-activity of non-peptide cholecystokinin antagonists

Potent and selective cholecystokinin antagonists have been developed in search of a tool to further characterize CCK's role in normal physiology as neurotransmitters and

/: oo neurohormones. Much of the work on these molecules has been reviewed extensively '

(Figure 3.1). Early CCK antagonists were based on dipeptide or peptoids antagonists such as proglumide and benzotript, but both had marginal potency and poor selectivity 89.

Further interests in the development CCK antagonist were based on its possible appUcation as an anti-anxiety drug. A nonpeptide CCK antagonist was developed from the natural product asperlicin, which showed some CCK antagonist properties. ^°To improve its oral bioavailability and to mimic diazepam, an anti-anxiety compound, the core structure of asperhcin was modified, resulting in 1,4-benzodiazepine scaffolds. In

a particular series of 3-alkylbenzodiazepine, it was shown that the D-Trp isomer of the

indolylmethyl group was a significantly more effective CCK antagonist, than to the L-

Trp analogue. A similar observation was made with the configuration of the tryptophan

residue in the dipeptoid CI-988. It was found that the R-configuration of the alpha-

methyl tryptophan resulted in a higher affinity for the CCK-B receptors when compared

to the corresponding L-configuration. Other dipeptides or peptoids with CCK

antagonist properties with improved bioavailabity used alpha-methyl-(R)-tryptophan. 38 1.6 Overlapping pharmacophores of opioids and cholecystokinin antagonists

In the hterature, several studies have alluded to the similarities between opioid and CCK peptide ligands as well as their nonpeptide ligands. Biophysical studies have suggested unsulfated C-terminal CCK-7 might have a biological activity at opioid receptors.

Computational studies of CCK and enkephalin included comparison of lowest energy structures of [Met^]-enkephalin and CCK-7, showing similarities in structure.

The aromatic moiety of Tyr' in [Leu^] enkephalin can be aligned with the Tyr^^ residue of

CCK-7. Similarly, the side chain of Phe"^ in [Leu^Jenkephalin can be aligned with the side-chain indole ring of Trp^° of CCK-7. More recent molecular modeling studies of topographically constrained CCK analogue SNF-9007 showed an agreement between the receptor-bound conformation of CCK-B ligands with the delta opioid receptor template model based on DPDPE. Several modeling studies suggested similarities between the proposed biologically active conformation between CCK and opioid ligands.

Examination of these proposed bioactive conformations demonstrated topographical similarities of the surfaces of the aromatic side chain residues suggesting that, at least in part, delta opioid receptors and CCK-B receptors have overlapping structural and topographical requirements. Figure 1.8 shows similarities in the orientation of the aromatic residues in DPDPE and sulfated CCK-8. 39

Figure 1.8. Stereoviews of the proposed bioactive conformations of DPDPE (top) and

CCK (bottom) at delta opioid and CCK-B receptors. The aromatic groups have similar orientation.

Biological activities of nonpeptide ligands also have suggested similarities in the structures important for binding at CCK and opioid receptors. For example, tifluadom, a benzodiazepine kappa opioid receptor agonist with some binding affinity for mu receptors, binds to CCK receptors and displays CCK antagonist activity as well.

Tifluadom has been found to enhance feeding behavior'''^ and analgesia. 40

O,VO H3C f-NH

F

Figure 1.9. Tifluadom, a nonpeptide benzodiazepine ligand which interacts with opioid and CCK receptors.

1.7 Rationale for a single molecule targeting multiple receptors

The desire to develop a single molecule (rather than relying on the mixture of two different compounds) is crucial for ultimate testing of the hypothesis in humans. Unlike a mixture of two compounds, a single chemical entity allows for the quantification of kinetics for appropriate use. Thus, properties such as time of onset, half-life, duration, and distribution would only be determined for one compound. Furthermore, the determination of side effects and toxicity of a single molecule would be less complicated.

Additionally, a single chemical entity would allow for refinement of the ratio of affinities at each receptor site allowing for a more exact therapeutic profile. 41

1.8 Goals

The goal of this dissertation was to design a single peptide that interacts with opioid receptors as agonist and with CCK receptors as an antagonist.

In Chapter 2, the lead compound SNF-9007 will be modified to increase the binding affinity at opioid receptors, improve on the CCK antagonist activities, and to balance the selectivity between CCK-A and CCK-B receptors. In Chapter 3, bulky analogues of Trp will be substituted at position 4 to increase antagonist properties at CCK receptors. In Chapters 4 and 5, CCK/opioid peptide will be conformationally constrained resulting in cyclic disulfide and lactam analogues, respectively. In Chapter 6, the analysis of the NMR structure of a lead cyclic disulfide analogue will be given. 42 CHAPTER 2

DESIGN, SYNTHESIS, AND STRUCTURE-ACTIVITY RELATIONSHIPS OF

CCK/OPIOID PEPTIDES: MODIFICATIONS OF THE LINEAR SEQUENCE

2.1 Results: design of linear peptides

Our design of peptides that can simultaneously interact with delta (and/or mu) opioid receptors as agonists and at CCK-A and CCK-B receptors as antagonists is based on the hypothesis of overlapping pharmacophores of opioid peptides and CCK ligands.

Previous studies in our laboratory has developed three-dimensional models for the bioactive conformations of ligands for both the delta opioid receptor''^^"''^^ and the CCK 1 08 receptor. From these three-dimensional models, it was observed that there are interesting topographical similarities between the opioid pharmacophores and CCK pharmacophores, particularly in the orientation of the aromatic residues. By accident, a ligand, SNF-9007, which has a potent selective agonist activity at the CCK-B receptor and weak affinity but robust agonist activity at delta opioid receptor, was discovered.

As seen in Figure 2.1, SNF-9007, the model lead compound, the opioid pharmacophore elements at the A'-terminal overlap with the CCK pharmacophore elements at the C- terminal. The initial goal was to redesign SNF-9007 to have mixed delta and mu binding affinity for the opioid receptors and a more balanced binding affinity for the CCK-A and

CCK-B receptors. It is known that the fi^ee A^-terminal tyrosine is important for the 43 potency of peptide ligands at opioid receptors. To improve the delta opioid receptor potency we propose to remove the A'-terminal Asp of SNF-9007. Also, Asp'' can be removed without greatly affecting binding affinity at CCK receptors, as evidenced by the activity of smaller fragments of CCK-8 at CCK-B receptors. ^"^To explore the optimal binding at opioid receptors as well as CCK receptors, a variety of amino acids were substituted at position 2 (analogues 1-7, Table 1.1). The second goal was to have a balanced affinity between CCK-A and CCK-B. SNF-9007 has an N"-methylated Nle (N-

MeNle or MeNle), which was hypothesized to be responsible for the selectivity for the

CCK-B receptor (i.e. low binding affinity at the CCK-A receptor). To improve the balanced binding affinity between the CCK-A and CCK-B receptors, N-MeNle was substituted with Nle (analogues 8 -12). The third goal was to improve the antagonist properties at CCK-A and CCK-B by focusing on Trp"*. Since it has been known that substitution of L-Trp with D-Trp in peptide, peptoid, and non-peptide analogues for CCK receptors lead to antagonist activity at CCK receptors, Ve replaced L-Trp with D-Trp in the N-MeNle series (analogues 13-16). In an attempt to have the simultaneous effect of having antagonists property at CCK receptors as well as a balanced binding affinity between CCK-A and CCK-B receptors, L-Trp was subsequently substituted with D-Trp"^ along with Nle^ (i.e. a double substitution, analogues 17 - 20). 44

CCK Pharmacophore

Asp-Tyr^-DPhe^-Gly^-Trp^-N-MeNle^-Asp^-Phe^-NH2 (SNF-9007)

Opioid Pharmacophore 1. Increase potency at opioid receptors. 2. Improve balanced affinity between CCK-A and CCK-B receptors. 3. Improve CCK receptor antagonist V properties.

Linear Analogues: Tyr-Xxx-Gly-Yyy-Zzz-Asp-Phe-NH2

Xxx = D-Amino acids, Pro, or Gly

Yyy = Trp or D-Trp

Zzz = N-MeNle or Nle

Figure 2.1. Design of CCK/opioid analogues. The linear analogue modifications focus on positions 2, 4, and 5. Position 2 is substituted with various D-amino acids, Pro and

Gly. Position 4 is substituted with D-Trp. Position 5 is substituted with N-MeNle and

Nle. 45 Table 2.1. Linear CCK/opioid peptides based on [desAsp'^] SNF-9007

No. Code Sequence SNF-9007 H-Asp^'-Tyr'^-D-Phe'"-Gly^'-Trp'^-NMeNle"-Asp'^-Phe''-NH2

1 RSA210 H-Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2

2 RSA211 Tyr-D-Ala-Gly-Trp-MeNle-Asp-Phe-NH2

3 RSA203 T yr-D-Nle-Gly-T rp-MeNIe-Asp-Phe-NH2

4 RSA204 Tyr-D-Val-Gly-Trp-MeNle-Asp-Phe-NH2

5 RSA215 T yr-D-Pro-Gly-T rp-MeNle-Asp-Phe-NH2

6 RSA205 Tyr-Pro-Gly-Trp-MeNle-Asp-Phe-NH2

7 RSA218 T yr-Gly-Gly-Trp-MeNle-Asp-Phe-NH2

8 RSA500 H-Asp-Tyr-DPhe-Gly-Trp-Nle-Asp-Phe-NH2

9 RSA501 H-Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2

10 RSA502 H-Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2

11 RSA503 H-Tyr-D-Ala-Gly-Trp-Nle-Asp-Phe-NH2

12 RSA504 H-Tyr-D-NIe-Gly-Trp-Nle-Asp-Phe-NH2

13 RSA601 H-T yr-D-AIa-Gly-D-T rp-MeNle-Asp-Phe-NH2

14 RSA602 H-Tyr-D-Phe-Gly-D-Trp-MeNle-Asp-Phe-NH2

15 RSA603 H-Tyr-D-Val-Gly-D-Trp -MeNIe-Asp-Phe-NH2

16 RSA604 H-Tyr-D-Nle-Gly-D-Trp-MeNle-Asp-Phe-NH2

17 RSA621 H-Tyr-D-Ala-Gly-D-Trp-Nle-Asp-Phe-NH2

18 RSA622 H-Tyr-D-Plie-Gly-D-Trp-NIe-Asp-Phe-NH2

19 RSA623 H-Tyr-D-Val-Gly-D-Trp-Nle-Asp-Phe-NH2

20 RSA624 H-T yr-D-Nle-Gly-D-Trp-Nle- Asp-Phe-NH2 Bold, residues of interest. 46 2.2 Results and discussion: Synthesis of linear analogues CCK/opioid peptides

2.2.1 General methods

The peptide analogues were synthesized manually using standard solid phase peptide synthesis by Fmoc/®u strategy on Rink Amide AM resin. The N"-Fmoc group was deprotected using 20% piperidine in DMF. The N"-Fmoc proctected amino acid was coupled to the growing peptide chain using in situ activation with HBTU and HOBt in the presence of DIEA. The side chain protecting groups were deprotected concomitant with cleavage of the peptide from the resin using 95% TFA/2.5% triisopropylsilane

(TlSy 2.5% H2O. Crude peptides were purified using RP-HPLC followed by lyophilization to give pure peptides in 20-50% yield.

2.2.2 Results: Coupling of N-MeNle^ and Trp''

Following the usual protocol of using 1 hr coupling, the initial syntheses of the peptide analogues presented some difficulties. The mass spectra of crude products often indicated a failed coupling of Trp.

The coupling of Fmoc-N'^-MeNle to Asp was easily completed in less than an hour as monitored by the standard Kaiser test. ^^^However, the deprotection of the Fmoc group from the N-MeNle could not be adequately monitored with the Kaiser test because of the secondary amine of the N-MeNle. The color of the resins was a yellow/orange, 47 instead of the expected dark blue. Completion of the deprotection of Fmoc was substantiated with the chloranil test and the TNBS test (see Appendix for details of these tests). Complete deprotection of Fmoc resulted in red resins on the chloranil test,

''^while the color was dark blue for the TNBS test. Incomplete deprotection of N-

MeNle resulted in failed coupling of the following Trp residue. HN "O ^ /

H3CO O H

Rink Amide AM resin 1. Swell overnight 2. Deprotection ofFmoc 25% Piperidine/ DMF w 3. DMF and DCM wash cycles

4. Coupling with: 3 eq. Fmoc-Phe-OH, 3 eq. HBTU 3 eq. HOBt 6 eq. DIEA

5. Sequential deprotection and washes 6. Sequential coupling and washes: Fmoc-Asp(O^Bu)-OH Fmoc-Nle-OH Fmoc-Trp (N'^-Boc)-©!! Fmoc-Gly-OH Fmoc-D-Nle-OH Fmoc-Tyr(?Bu)-OH f 7. Final Fmoc deprotection 8. Wash cycles with final wash of DCM 9. Cleavage 95% TFA/2.5%TIS/2.5%water V

H-Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2

Figure 2.2. Solid phase peptide synthesis of linear analogues of CCK/opioid peptides.

The structure of Rink amide resin is depicted. The analogue shown is RSA504. 49 Another difficulty was the coupling of Fmoc-Trp(Boc)-OH to the secondary amine of the N-MeNle. The coupling was often sluggish, resulting in positive Kaiser test

(blue resins), even after an hour of reaction. Several coupling conditions were tested including various coupling agents and extended reaction times. Although PyBrop in the presence of DIEA is a recommend coupling agent for N-methylated residues, the coupling with PyBrop for one hour was just was similar to coupling with in situ activated

HBTU-HOBt (4 equivalents relative to resin loading) for two hours. Double coupling procedures were not pursued.

2.2.3 Results: Cleavage of linear peptides from the solid support

Another difficulty was in the cleavage reaction. Initial syntheses of the peptides used 3 hour cleavage times with 95% TFA/ 2.5% TES/ 2.5% water. TES (triethylsilane) is now a common scavenger used in cleavage as a substitute for thiol-based scavengers, which are avoided because of the stench. The initial conditions, however, resulted in a side product with 2 mass units more than the desired mass based on ESI-MS. These extra

2 mass units were hypothesized to be a result of reduction of the indole tryptophan to indoline (Figure 2.3). 50

H H

Figure 2.3. Reduction of indole ring to indoline due to the presence of the

silane based scavenger in the cleavage mixture. R represents the rest of

the peptide

Some reaction conditions were explored and the optimal cleavage condition was determined. To suppress the reduction of the indole ring to indoline, the cleavage protocol was modified by using triisopropylsilane (TIS), in lieu of TES, with shorter 1.5 hours cleavage time.

2.2.4 Results: Incomplete deprotection of Trp

Initial syntheses of these classes of peptides often resulted in low yields.

Analytical HPLC chromatograms of the crude peptides showed two peaks and the mass spectrum of the crude compounds showed a 44 m/z unit greater than expected monoisotopic mass. The unknown peak was further characterized and identified. Using a combination of the analyses of the b and a ions of the MS/MS (ESI) data and HR-MS data (FAB iononization), the extra 44 m/z unit was determined to be a CO2 fragment added to the Trp residue (Figure 2.4). In N'-Boc proctected indole, carbamic acid derivative is formed after cleavage with trifluoro acetic acid. The side product was stable as a solid, but not in solution. Decarboxylation was achieved by incubating the peptide 51 solution in 0.1% TFA for 1 day at room temperature. Optimal isolated yields were achieved by dissolving the crude peptide in 0.1% TFA.

Figure 2.4. Incomplete deprotection of N'-Boc protected Trp resulted in a carbamic acid

2.3 Results: Structure-activity relationships of linear CCK/opioid analogues at

CCK and opioid receptors.

The competition binding assays against radioligands at the delta and mu opioid receptors and CCK-A and CCK-B receptors, as well as the functional assays, [ SJGTP-y-

S assays and phosphoinositide (PI) hydrolysis assays, were performed by Dr. Shou-Wu

Ma and Hamid Badghisi fi-om the laboratory of Dr. Josephine Lai at the Department of

Pharmacology in the University of Arizona. The binding assays at CCK receptors were • 118 • performed as previously described by Lm et al. The [ S]GTP-y-S bindmg assays at opioid receptors are performed, with modifications, as previously described by Hoshota et al. PI assays were performed as previously described by Patel et al.

The opioid agonist activities at the delta and opioid receptors in isolated mouse vas deferens (MVD) and guinea pig ileum (GPI) tissues, respectively, and the CCK 52 agonist and antagonist activities in GPI tissues were determined and performed by Peg

Davis from the laboratory of Dr. Frank Porreca at the Department of Pharmacology in the

University of Arizona. The assays for the bioactivity at the opioid receptors were • • 191 ... performed as previously described in Kramer et al. The assays for the bioactivity at

CCK receptors were performed as previously described by Lucaites et al. The experimental details of the tissue assays are summarized in the Appendix.

The data for the competitive binding assay and the corresponding functional PI at

CCK receptors are summarized in Table 2.2 and Table 2.3, respectively. The data for the competitive binding assay and the corresponding functional [^^S]-GTP-y-S assay at delta and mu opioid receptors are summarized in Table 2.4 and Table 2.5, respectively. The functional opioid agonist assay in the MVD and GPI tissues and CCK agonist and antagonist activities in the GPI/LMMP tissues are summarized in Tables 2.6 and Table

2.7, respectively.

2.3.1 Truncation of A'-terminal Asp

SNF-9007 has poor binding and potency at the mu and delta opioid receptors.

To increase the binding affinity and potency at delta and mu opioid receptors, the Asp was removed to expose the free A^-terminal Tyr' which is important to the pharmacophore of opioid peptide ligands. In a competition binding assay, [desAsp''] SNF-9007 (1), increased the binding affinity by 36 fold to 6.8 nM at the delta opioid receptor and by 38- fold to 136 nM at the mu opioid receptor, respectively, when compared to SNF-9007. In 53 the MVD and GPI assays and in the second messenger assay ([ S]GTP-y-S bmdmg assay), the opioid agonist activity of 1 had a corresponding increase in binding affinity and functional activity at opioid receptors. As expected, removal of Asp did not change the high binding affinity at the CCK-B receptor (Ki = 2.13 nM), while the binding affinity at the CCK-A receptor remained very low, greater than in the micromolar range. The functional activity at the CCK-B receptor was potent (Ki = 15.8 nM) while at the CCK-A receptors, there was no response at 10 |iM. Interestingly, 1 showed weak antagonist activity in the unstimulated GPI/LMMP assay, actually about 8-fold less than SNF-9007

(Ke = 254 nM). 54 Table 2.2. Binding affinities of CCK/opioid peptides at the human CCK-A and CCK-B receptors

Binding Affinity® No. Sequence (Ki, nM) CCK-A CCK-B SNF-9007 Asp-Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 3270 2.1 1 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 >10000 2.10 2 Tyr-D-Ala-Gly-Trp-MeNle-Asp-Phe-NH2 5700 2.1 3 T yr-D-Nle-Gly-T rp-MeNle-Asp-Phe-NH2 3900 0.61 4 T yr-D-V al-Gly-Trp-MeNle-Asp-Phe-NH2 9200 5.2 5 T yr-D-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 7200 2.7 6 Tyr-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 >10000 >10000 7 Tyr-Gly-Gly-Trp-MeNle-Asp-Phe-NH2 870 1.3 8 Asp-T yr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 140 8.0 9 Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 140 14. 10 Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 120 8.1 11 T yr-D-Ala-Gly-Trp-Nle-Asp-Phe-NH2 5700 150 12 Tyr-D-Nle-Gly-Trp-NIe-Asp-Phe-NHa 11. 16. 13 T yr-D-Phe-Gly-D-T rp-MeNle-Asp-Phe-NH2 1100 1.6 14 T yr-D-Ala-Gly-D-T rp-MeNle-Asp-Phe-NH2 32. 1.3 15 Tyr-D-Val-Gly-D-Trp-MeNle-Asp-Phe-NH2 140 1.9 16 T yr-D-Nle-Gly-D-T rp-MeNle-Asp-Phe-NH2 5200 2400 17 Tyr-D-Ala-Gly-D-Trp-Nle-Asp-Phe-NH2 6500 2700 18 Tyr-D-Phe-Gly-D-Trp-Nle-Asp-Phe-NH2 1600 1900 19 Tyr-D-Pro-Gly-D-Trp-Nle-Asp-Phe-NHa 3000 3600 20 Tyr-D-Nle-Gly-D-Trp-Nle-Asp-Phe-NH2 nc nc ^Competition against ['^^I]CCK-8 (sulfated) at hCCK-A and hCCK-B receptors expressed in HEK cell lines. Bold, residues of interest; nc, no competition. 55 Table 2.3. Functional activities of linear CCK/opioid peptides at the CCK receptors

Functional Analysis" No. Sequence (EC50, nM) CCK-A Emax CCK-B Emax Asp-Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 n/d n/d n/d n/d

1 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 ns ~ 16.0 9.3

2 Tyr-D-Ala-Gly-Trp-MeNle-Asp-Phe-NH2 ns ~ 10.0 23

3 T yr-D-Nle-Gly-T rp-MeNle-Asp-Phe-NH2 ns — 20.0 12

4 Tyr-D-Val-Gly-Trp-MeNle-Asp-Phe-NH2 ns — 110 17

5 T yr-D-Pro-Gly-T rp-MeNle-Asp-Phe-NH2 ns — 37.0 13.0 6 Tyr-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 ns 8.3 ns 2.6

7 Tyr-Gly-Gly-Trp-MeNle-Asp-Phe-NH2 nr — 32.0 14 8 Asp-Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 51 7.8 570 54 9 Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 790 13. 3100 17 10 T yr-D-Phe-Gly-Trp -Nle-Asp-Phe-NH2 630 5.3 110 8.0

11 T yr-D-Ala-Gly-T rp-Nle-Asp-Phe-NH2 n/d ~ n/d ~ 12 T yr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 2500 15 1900 10

13 T yr-D-Phe-Gly-D-T rp-MeNle-Asp-Phe-NH2 nr — 19 15

14 T yr-D-Ala-Gly-D-Trp -MeNle-Asp-Phe-NH2 ns — 2.5 14 15 T yr-D-Val -Gly-D-Trp-MeNle-Asp-Phe-NH2 8100 11 8.6 15

16 Tyr-D-Nle-Gly-D-Trp-MeNle-Asp-Phe-NH2 nr — nr —

17 Tyr-D-Ala-Gly-D-Trp-Nle-Asp-Phe-NH2 ns — 190 13 18 T yr-D-Phe-Gly-D-T rp-Nle-Asp-Phe-NH2 7300 12 1200 7

19 Tyr-D-Pro-Gly-D-Trp-Nle-Asp-Phe-NH2 n/d ~ 360 10 20 T yr-D-Nle-Gly-D-Trp-Nle- Asp-Phe-NH2 >10000 3.5 >10000 5.4 "Phosphoinositide (PI) hydrolysis assay in hCCK-A and hCCK-B, expressed receptors in HEK cell lines. Emax = (total PI hydrolysis/basal PI hydrolysis). Bold, residues of interest; n/d, not determined; nr, no response; ns, non-saturable. 56

Table 2.4. Binding affinities of linear CCK/opioid peptides at opioid receptors.

Binding Affinity^ No. Sequence (Ki, nM) hDOR rMOR Asp- Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 250 5200 1 T yr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NHa 6.8 140 2 Tyr-D-Ala-Gly-Trp-MeNle-Asp-Phe-NH2 14 46 3 T yr-D-Nle-Gly-T rp-MeNle-Asp-Phe-NH2 1.6 25 4 Tyr-D-V al-Gly-Trp-MeNle-Asp-Phe-NHi 26 94 5 T yr-D-Pro-Gly-Trp-MeNle-Asp-Phe-NHi 110 3200 6 Tyr-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 1500 >10000 7 T yr-Gly-Gly-Trp-MeNle-Asp-Phe-NH2 20 610 8 Asp-Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 320 6800 9 Tyr-Nle-Gly-T rp -Nle-Asp-Phe-NH2 74 1000 10 T yr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 0.42 79 11 T yr-D-Ala-Gly-Trp-Nle-Asp-Phe-NH2 39 3.3 12 Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 2.9 27 13 T yr-D-Phe-Gly-D-Trp-MeNle- Asp-Phe-NH2 0.55 5.7 14 Tyr-D-Ala-Gly-D-Trp-MeNle-Asp-Phe-NH2 1.9 21 15 Tyr-D-Val-Gly-D-Trp-MeNle-Asp-Phe-NH2 52 180 16 Tyr-D-NIe-Gly-D-Trp-MeNle-Asp-Phe-NH2 300 1200 17 Tyr-D-Ala-Gly-D-Trp-Nle-Asp-Phe-NH2 130 560 18 Tyr-D-Phe-Gly-D-Trp-Nle-Asp-Phe-NH2 190 350 19 Tyr-D-Pro-Gly-D-Trp-Nle-Asp-Phe-NHa >10000 2100 20 T yr-D-Nle-Gly-D-Trp-Nle- Asp-Phe-NHi 20 280 ^Competitive assay against radiolabelled [^H] DPDPE at hDOR and [^HJDAMGO at rMOR. Opioid receptors were expressed from CHO cell lines. Bold, residues of interest. 57 Table 2.5. Functional activities of CCK/opioid peptides at opioid receptors

Functional Activity (Agonist) ^ No. Sequence (EC5Q, nM)

hDOR Emax rMOR Emax

Asp- Tyr-D-Plie-Gly-Trp-Nle-Asp-Phe-NH2 n/d ~ n/d ~

1 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 23 310 2.5 84

2 Tyr-D-Ala-Gly-Trp-MeNle-Asp-Phe-NH2 110 260 18 75

3 T yr-D-Nle-Gly-T rp-MeNle-Asp-Phe-NH2 15 150 110 68

4 Tyr-D-V al-Gly-Trp-MeNle-Asp-Phe-NH2 200 200 63 130

5 T yr-D-Pro-Gly-T rp-MeNle-Asp-Plie-NH2 700 100 1000 200

6 T yr-Pro-Gly-Trp-MeNle-Asp-Phe -NH2 ne ~ ne —

7 Tyr-Gly-Gly-Trp-MeNle-Asp-Phe-NH2 76 91 130 21

8 Asp-Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 1800 34 370 69

9 Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 1000 68 1800 66

10 T yr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 0.90 170 15 200

11 Tyr-D-Ala-Gly-Trp-Nle-Asp-Plie-NH2 7.9 230 78 430

12 Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 4.5 82 0.47 110

13 Tyr-D-Phe-Gly-D-Trp-MeNle-Asp-Phe-NH2 21 34 520 53

14 T yr-D-Ala-Gly-D-Trp-MeNle- Asp-Phe-NH2 27 50 38 46

15 T yr-D-Val -Gly-D-Trp-MeNle-Asp-Phe-NH2 130 170 50 49

16 T yr-D-Nle-Gly-D-Trp-MeNle- Asp-Phe-NH2 n/d n/d 240 57

17 T yr-D-Ala-Gly-D-Trp-Nle- Asp-Phe-NH2 130 84 0.06 110

18 T yr-D-Phe-Gly-D-Trp-Nle- Asp-Phe-NH2 130 160 46 150

19 Tyr-D-Pro-Gly-D-Trp-Nle-Asp-Phe-NH2 230 45 350 230

20 Tyr-D-Nle-Gly-D-Trp-Nle-Asp-Phe-NH2 60 51 200 110 ^[^^S]GTP-y-S binding assay. Emax = (net total bound/basal binding) X 100%. Opioid receptors were expressed from CHO cell lines. Bold, residues of interest; ne, no effect; n/d, not determined. 58

Table 2.6. Functional activities of linear CCK/opioid peptides at the MVD and GPI

Opioid Agonist®, (IC50 ,nM) No. Sequence MVD (delta) GPI (mu) Asp- Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NHa 58.4 176

1 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 65 ± 9.5 100 ±23 2 Tyr-D-Ala-Gly-Trp-MeNle-Asp-Phe-NHi 66± 11 470 ±61

3 Tyr-D-Nle-Gly-Trp-MeNle-Asp-Phe-NH2 9.0 ±0.81 48 ±6

4 Tyr-D-VaI-Gly-Trp-MeNle-Asp-Phe-NH2 640±150 510 ±63''

5 T yr-D-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 890 ±310 13% at 1 laM

6 Tyr-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 5% at 1 |aM 1.2% at 1 [iM

7 Tyr-Gly-Gly-Trp-MeNle-Asp-Phe-NH2 220 ± 16 4200 ± 720

8 Asp-Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 25% at 1 laM' 0% at 1 )a,M

9 T yr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 17% at 1 nM'' 0% at 1 f4,M

10 Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 12 ±2.0 420 ± 68

11 T yr-D-Ala-Gly-Trp-Nle-Asp-Phe-NH2 44.50 ±9.8® 160 ±28

12 Tyr-D-Nle-Gly-Trp-NIe-Asp-Phe-NH2 23 ± 9.7 210±52

13 Tyr-D-Phe-Gly-D-Trp-MeNle-Asp-Phe-NH2 24 ± 4.6 71 ±5.4

14 T yr-D-Ala-Gly-D-Trp -MeNle-Asp-Phe-NH2 63 ±27 150 ±65

15 Tyr-D-Val-Gly-D-Trp-MeNle-Asp-Phe-NH2 310 ±43 120 ± 19

16 T yr-D-Nle-Gly-D-Trp -MeNle-Asp-Phe-NH2 7000±100 8.8% at 1 nM

17 T yr-D-Ala-Gly-D-Trp-Nle- Asp-Phe-NH2 1300± 190 2300 ± 330

18 Tyr-D-Phe-Gly-D-Trp-Nle-Asp-Phe-NH2 170 ±21 2700±180

19 Tyr-D-Pro-Gly-D-Trp-Nle-Asp-Phe-NH2 17% at 1 ^iM 0% at 1 i^M

20 T yr-D-Nle-Gly-D-Trp-NIe- Asp-Phe-NH2 n/d 1000 ^Concentration at 50% inhibition of muscle contraction at electrically stimulated isolated tissues. Bold, residues of interest; n/d, not determined; ^CTAP reversed; "^no shift in DPDPE, ''no shift in DPDPE; ^2.4±0.3 |j,M with protein cocktail inhibitors. 59 Table 2.7. CCK functional activities of linear CCK/opioid peptides at the unstimulated GPl/LMMP

unstimulated GPI/LMMP No. Sequence CCK-Agonist' CCK-Antagonist" (A50) (K,, nM) Asp- Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NHa 0% at 1 iiM' 31.2 1 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NHa 0% at 1 |a,M 250± 160

2 T yr-D-Ala-Gly-Trp-MeNle-Asp-Phe-NH2 0% at 1 )j,M 100 ± 12

3 T yr-D-Nle-Gly-Trp-MeNle-Asp-Phe -NH2 0% at 1 ^M 7.5 ±2.6

4 Tyr-D-Val-Gly-Trp-MeNle-Asp-Phe-NH2 0% at 1 |j,M 450±120

5 T yr-D-Pro-Gly-Trp-MeNle-Asp-Phe -NH2 0% at 1 )a,M 360 ± 48

6 Tyr-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 12% at 1 |a.M 0% at 1 |aM

7 Tyr-Gly-Gly-Trp-MeNle-Asp-Phe-NH2 3.5% at 1 jaM 0% at 1 fj-M

8 Asp-Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 1200 ±240*^ agonist

9 Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 0% at 1 |j,M 0% at 1 |j,M

10 Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 0% at 1 i^M 40 ±20

11 Tyr-D-Ala-Gly-Trp-Nle-Asp-Phe-NH2 0% at 1 )aM 190 ±32

12 T yr-D-Nle-Gly-T rp-Nle-Asp-Phe-NH2 0% at 1 |j,M 190 ±80

13 Tyr-D-Plie-Gly-D-Trp-MeNle-Asp-Phe-NH2 0% at 1 uM 7.0 ± 1.9

14 T yr-D-Ala-Gly-D-Trp -MeNle-Asp-Phe-NH2 16% at 1 ^iM 13 ±4.5

15 Tyr-D-Val-Gly-D-Trp-MeNle-Asp-Phe-NH2 0% at 1 |j,M 0% at 1 )aM

16 Tyr-D-Nle-Gly-D-Trp-MeNle-Asp-Phe-NH2 0% at 1 |a,M 0% at 1 )J,M

17 T yr-D-Ala-Gly-D-Trp-Nle- Asp-Phe-NH2 0% at 1 |a.M 0% at 1 |a,M

18 Tyr-D-Phe-Gly-D-Trp-Nle-Asp-Phe-NH2 0% at |iM 530±130 19 Tyr-D-Pro-Gly-D-Trp-Nle-Asp-Phe-NHa 0% at 1 |j,M 0% at 1 |j,M

20 Tyr-D-Nle-Gly-D-Trp-Nle-Asp-Phe-NH2 0% at 1 |aM 480±170 ''Contraction of isolated tissue relative to initial muscle contraction with KCl. Inhibitory activity against the CCK-8 induced muscle contraction. Ke, concentration of antagonist needed to inhibit CCK-8 to half its activity. '^No response.'^Concentration at 50% activation. Bold, residues of interest. 60 2.3.2 Substitution at position 2

Since removal of Asp improved the binding affinity at the opioid receptors and was well tolerated at the CCK receptors, structural modifications of 1 were further explored. Position 2 was substituted with various D-amino acids (D-Ala 2, D-Nle 3, D-

Val 4, D-Pro 5) as well as Pro 6 and Gly 7 to determine which residue would have an optimal effect for CCK antagonist activity while maintaining potent opioid agonist activity. As expected, the binding affinities at delta and mu opioid receptors dramatically increased the binding affinity at the hDOR and the rMOR when compared to SNF-9007

(Table 2.4). Analogues 2-4 generally retained good binding affinity at both opioid receptors (1.6 - 46 nM), but moderate preference for the delta over the mu opioid receptor (about 3-10 |a/5). The binding affinities at delta opioid receptors were poor when Pro 6 and Gly 7 were placed at position 2 (Ki = 1.5 )j,M and 2.0 |j,M, respectively).

Substitution with D-Nle 3 had the best binding affinity for the hDOR (Kj =1.6 nM) (Table

2.2).

In the in vitro MVD and GPI opioid agonist assays (Table 2.7), analogues 2- 4 maintained or slightly decreased agonist activity at both delta and mu opioid receptors.

The importance of a D-amino amino acid in position 2 was further supported when Gly was placed in position 2 which resulted in a significant decrease in activity compared to 1.

Substitution of D-Pro 5 or Pro 6 resulted in significant loss in opioid agonist potency, suggesting that the secondary amide of Pro introduced an unfavorable turn or constraint in the peptide backbone for these enkephalin peptide ligand. 61 When various amino acids were substituted at the position 2, the binding affinities at the CCK-B receptors remained unchanged (Table 2.2), for the most part, when compared to SNF-9007. The binding at the CCK-A receptor, however, remained in the micromolar range. Unlike in the opioid receptors, substitution with D-Pro 5 and Gly 7 did not decrease the binding at the CCK-B receptor (K = 2.7 and 1.3 nM, respectively).

Replacement with Pro 6, however, resulted to a significant loss in binding at the CCK-B receptors, much like the binding at the delta and mu opioid receptors. This suggests that there is a common structural requirement between opioid ligands and CCK ligands that is not favored when Pro is introduced at position 2. The substitutions at position 2, except for Pro 6, produced a corresponding good functional activity at the hCCK-B receptors, while not producing any activation of hCCK-A receptors (Table 2.3). The CCK antagonist activities in the unstimulated GPI/LMMP assay of 2-5 were surprisingly quite good (K^ = 7-450 nM), but analogues with Pro 6 and Gly 7 had CCK agonist activities at high concentrations.

2.3.3 Substitution with Nie®

As mentioned before, one of our aims was to have a balanced selectivity between the two CCK receptor subtypes, CCK-A and CCK-B. We hypothesized that the added constraint produced by N"-methylation at position 5 induces a conformation that was less favorable for binding at the CCK-A receptor. Thus, we modified the CCK-B selective lead compound SNF-9007 by substituting N"-MeNle^ with Nle. 62 The binding affinity at the hCCK-A receptor increased significantly (Table 2.2), which resulted in a more balanced selectivity between the two receptors while maintaining the high binding affinity at the hCCK-B receptor. Analogues [Asp0 , Nle2 ,

Nle^] 8 and [Nle^, Nle^] 9 demonstrated that Nle^ increased the affinity for the CCK-A receptor, thus improving the balanced selectivity between CCK-A and CCK-B receptors

(13:1 ratio) when compared to SNF-9007. Since 8 and 9 both have L-Nle at position 2, the binding affinity and activity at the opioid receptors were very poor. Even when position 2 was substituted with D-Phe 10 and D-Nle 12, the balanced CCK-A/B binding affinity was maintained, with a nearly 1:1 ratio for 12. Interestingly, when position 2 was substituted with a less bulky and hydrophobic residue D-Ala 11, the binding affinity at

CCK-A was very low at 5.7 |a,M. From this it can be suggested that position 2 does play a role in CCK-A and CCK-B receptor selectivity, hiterestingly, [D-Nle^] 12 showed the best binding affinity (K, =11 nM) at the CCK-A receptor for a non-sulfated analogue of

CCK.

Coincidentally and surprisingly, substitution of N"MeNle^ with Nle resulted in analogues with poor activity at both CCK receptor subtypes in the CCK functional assay

(Table 2.3). Relevant to the binding affinity of these peptides, analogues 10-12 were acting as antagonists, or at least as partial agonists. From this it can be suggested that position 2 has a role in the bioactivity of CCK receptors. For the CCK antagonist activity in the unstimulated GPI/LMMP assay, substitution with Nle^ from SNF-9007 did not dramatically change the CCK antagonist activity. 63 Substitution at position 2 with D-Nle was not expected to significantly affect the binding affinity at the delta and mu opioid receptors. As mentioned earlier, 8 and 9 did not have a good binding and activity at the opioid receptors because position 2 had L- amino acids (Table 2.4). [D-Phe^, Nle^] 10, [D-Ala^, Nle^] 11, and [D-Nle^, Nle^] 12 were substituted with D-amino acids at position 2 which resulted in good binding affinity and activity at delta and mu opioid receptors. Interestingly, 11 was more selective for the mu opioid receptor

In the MVD and GPI assays, substitution with Nle from N"MeNle^ resulted in analogues with improved agonist activity at the opioid receptors when compared to SNF-

9007 (Table 2.6). As expected, 8 and 9 showed very poor agonist activity at delta and mu receptors. With D-amino acids at position 2, 10-12 have high potencies at the delta opioid receptor (IC5o= 23-45 ilM) and moderate potencies at the mu receptor (IC5o= 160-

420 nM).

2.3.4 Substitution with D-Trp"*

L-Trp'^ was substituted with D-Trp in the CCK/opioid analogues to introduce antagonist activity at CCK receptors while retaining agonist potency at the opioid receptors. Introducing a D-Trp substitution was expected to lead to analogues with antagonist activity at the CCK receptors. When L-Trp"^ was replaced with D-Trp, the binding affinity at the hCCK-B receptor was maintained in the low nM range. 13-15 (D-

Phe^, D-Ala^, and D-Val^ respectively) all had binding affinities of 1 - 2 nM (Table 2.4). 64 Furthermore, the binding affinities of 13-15 at the CCK-A receptors increased significantly (Ki = 1080, 32, and 142 nM, respectively) when compared to their L-Trp counterparts (1-3), resulting in an unexpected increased balanced selectivity between

CCK-A and CCK-B. Surprisingly, [D-Nle^, D-Trp'^, Nle^] 16 resulted in poor binding affinity at CCK-A and CCK-B (Ki = 5200 and 2400 nM, respectively).

As for the functional activity at the CCK receptors (Table 2.3), D-Trp^ substitution has resulted in good activity at the CCK-B receptors (EC50 =2-19 nM) while the activity at the CCK-A receptors remained very poor or with no response at all.

In contrast, [D-Phe^, D-Trp'^] 13 and [D-Ala^, D-Trp'*] 14 displayed CCK antagonist activity by inhibiting the CCK-8 induced activity at the unstimulated GPI/LMMP (Ke =

6.9 and 13 nM, respectively). [D-Val^, D-Trp'^] 15 had neither agonist nor antagonist activity at the unstimulated GPI.

In the set of peptides with N"-MeNle^ and various D-amino acids at position 2, the agonist activity at MVD and GPI assay was good for 13 (IC50 = 24 and 71 nM, respectively) and 14 (IC50 = 63 and 150 nM, respectively) while moderate for 15 (IC50 =

310 and 120 nM, respectively) (Table 2.6). In comparison to SNF-9007, 13 is 2.5-fold more potent at both MVD and GPI assays. This is surprising since the replacement of

Phe"^ in Met-enkephalin analogues with D-Phe and D-Trp resulted in analogues that were virtually inactive at the opioid• • receptor. 48

In the competitive binding assays for delta and mu opioid receptors involving the substitution of L-Trp"^ with D-Trp"^, the binding affinity of analogues 13-15 increased by 2 to 10-fold compared the L-Trp"* counterparts (Table 2.4). These analogues were 65 moderately selective for the delta opioid receptor. These same D-Trp"^ analogues had moderate activities at the delta and mu opioid receptors. As mentioned before, the high to moderate binding affinity and bioactivity for the opioid receptors are surprising since it was found that D-aromatic amino acid residues in this position resulted in inactive peptides. 48

2.3.5 Double substitution with D-Trp"* and Nle®

Two strategies were combined to obtain the desired effect of increased binding at opioid receptors and at CCK receptors. Since substitution with D-Trp"^ successfully resulted in an antagonist effect at the CCK-A receptor in the unstimulated GPI/LMMP assay, and substitution with Nle^ resulted in a balanced binding affinity between CCK-A and CCK-B, a double substitution was attempted to get an additive result.

[desAsp^]SNF-9007 was modified to include D-Trp^'and Nle^ substitution. Analogues 17 and 18 showed significant decrease in the opioid activity at the delta and mu opioid receptors. The binding affinities at CCK-A and CCK-B were very poor, in the micromolar range (Table 2.2). The corresponding functional activities at these CCK receptors also were very poor. There was neither an agonist nor an antagonist activity for the CCK-A receptors in the unstimulated GPFLMMP assays.

These results also showed that double substitution of D-Trp^and Nle^ was not tolerated at opioid and CCK receptors and did not have an additive effect when compared to a single substitution of D-Trp"^ or Nle^. 66 Taken together, the initial approach to truncate the //-terminal Asp of SNF-9 increased the binding and activity at the opioid receptors without changing the profile at the CCK receptors. Substitution with Nle^ increased the balanced selectivity between

CCK-A and CCK-B receptors. Substitution with D-Trp"^ increased the CCK antagonist activity while also increasing the binding at CCK-A receptors. Simultaneous substitution with D-Trp"^ and Nle^ was not tolerated at opioid and CCK receptors. This suggests that

CCK ligands and opioid ligands have a common backbone requirement for binding the

CCK and opioid receptors.

2.4 Discussions

2.4.1 Assessment of the binding affinity and functional assays

Classical in vitro functional bioassays were performed for the agonist activities of

CCK/opioids in the GPI and MVD. The results of these assays were generally comparable with the radioligand competition binding assays and the [^^S]GTP-y-S functional bioassays. The CCK activity of the CCK/opioids was characterized using in vitro functional studies using GPI/LMMP, which is a commonly used index of CCK-A function. ^^^'^^'^However, there are some differences between the binding affinity data for human CCK receptors when compared to the CCK-A antagonist activity at the unstimulated GPI/LMMP. While the CCK/opioid analogues were shown to be antagonists at the unstimulated GPI/LMMP assays, several of these peptides did not bind 67 to the CCK-A receptors in the cloned receptors. The binding affinities at the hCCK-B receptors were very high and were showing very good agonist properties in the nanomolar range in the PI assays. These differences in the assay need further investigation. It is possible that the antagonist effect at the CCK receptors is an allosteric effect which explains why the analogues do not bind competitively against the radiolabeled sulfated CCK-8.

Even though the in vitro assays need to be investigated, there are several interesting observations that can be made regarding the structure activity relationships, particularly in the antagonist properties of CCK.

2.4.2 RSA504 is the most potent analogue of CCK at CCK-A receptors

As mentioned earlier, 12 (Nle^) had very good binding affinities for the CCK-A receptor. This is quite notable since it is thought that the sulfated tyrosine is necessary for high binding affinity at the CCK-A receptor. When compared to other SNF analogues, 12 has the highest binding affinity (Kj = 11 nM) for a non-sulfated analogue of CCK at the CCK-A receptor. The potency at the CCK functional PI assay was fairly low at 2 )a,M (Table 2.3), which suggests that 12 is an antagonist (based on the ratio between binding affinity and functional activity). When compared to D-Phe^ 10, the binding affinity of 12 is 10-fold higher. This suggests that non-aromatic alkyl side chain group binds better than an aromatic side chain group. This observation applies to binding affinity at the CCK-B receptor and is consistent with other series. 68

2.4.3 SAR along the peptide backbone at positions 4 and 5

From the perspective of the opioid receptors, these structure function activity studies have provided interesting insights to the conformation of opioids at the delta and mu opioid receptors, particularly the relatively high binding affinity of 13 where position

4 has a D-Trp. Schiller has reported that substitution of D-Phe"^ or D-Trp"^ in [Leu^]- enkephalin resulted in analogues with virtually no activity. Also, in the p-MePhe'^ analogues of DPDPE, the (2S,3S) and (2S,3R) isomers have superior potency compared to the (2R) isomers. This observation is qualitatively consistent with the [D-Trp"^,

Nle^] analogues (e.g. 17-20). Because 13 and 14, [D-Trp'^,N-MeNle^] analogues, had high potency at the opioid receptors, it can be suggested that the N-methylation at position 5 allows analogues with D-Trp to be potent at delta and mu opioid receptors.

The initial hypothesis was that the selectivity for CCK-B was due to the added constraint introduced to the backbone by N-methylation. Based on the results, however, substitution from L-Trp to D-Trp (e.g. 13-15) increased the binding affinity at the CCK-

A receptor, while retaining the high affinity at CCK-B receptor. This suggests that both

[D-Trp'*,NMeNle^] and [Trp^,Nle^] analogues have a common structure that is favorable to CCK-A receptors. These positions in CCK have been explored by stabilizing the cis- amide bond between positions 4 and 5, which resulted in low binding affinities to both

CCK receptors. 69

2.4.4 Role of D-amino acid at position 2.

While substitution of [D-Trp"^, Nle^] resulted in a series of peptides (17-20) with drastic loss of binding affinities at both CCK-A and CCK-B receptors, the phenylethyl ester derivatives of CCK-8 showed that a modification of [L-Trp,Nle]-JMV-180 to [D-

Trp,Nle]-JMV-180 (formally known as JMV-179) had high affinity at CCK-B receptors 1 01 and had a full antagonist property (Figure 2.5). From this it can be inferred that [D-

Trp'^,Nle^] substitution is not solely responsible for the drastic loss of binding affinities at

CCK-A and CCK-B receptors. Other residues might be involved in this, particularly the

D-amino acid analogues at position 2.

Boc-Tyr(S03H)-Nle-Gly-Trp-Nle-Asp

JMV-180

Boc-Tyr(S03H)-Nle-Gly-D-Trp-Nle-Asp

JMV-179

Figure 2.5. Structures of phenylethylester analogues of CCK. JMV-180 has mixed

CCK activities, whereas its D-Trp analogue, JMV-179, is a full antagonist. Both analogues have high affinities for CCK-A receptors at pancreatic acini and CCK-B receptors at guinea pig brains. 70

2.4.5 Similarities in opioid and CCK SAR supports the overiapping pharmacophore hypothesis

From the perspective of overlapping pharmacophores between opioid and CCK hgands, these structure function activity results further show that there are common requirements for opioid and CCK ligands to bind to their respective receptors, in particular the peptide backbone at position 4 and 5. As previously reported, substitution of D-Phe or D-Trp in positions 4 of [Leu^]-enkephalin resulted in very poor activity.

Much like the enkephalin peptide, substitution of D-Trp in Ac-CCK-7 resulted in low binding affinities at CCK-A receptors. In the CCK/opioid analogues, substitution of D-

Trp in position 4 resulted in loss of binding affinity at both CCK and opioid receptors only if position 5 was Nle. But this substitution of D-Trp'^ retained high binding affinity at the CCK and opioid receptors when position 5 is a iV-methylated Nle. Because

CCK/opioid ligands had the same structure activity relationship at the CCK and opioid receptors, it can be inferred that CCK and opioids have similar conformations along positions 4 and 5 which is favored for binding. These observations further support the hypothesis that CCK and opioid have overlapping pharmacophores.

No particular conclusion can be suggested as to whether the backbone is of primary importance or the way the backbone orients the side-chain group has more effect on the binding affinity since all of peptides were linear and thus had considerable flexibility of possible structures without a high energy penalty. 71 Because positions 4 and 5 had a tremendous effect to the binding affinity at CCK and opioid receptors, it might be interesting to determine by computation methods structures of the backbones of D-Trp, N-MeNle and Trp, Nle. A comparison can be made if there are common structures between the two. Studies have postulated that R-

CO-D-Xxx, Pro-NHR' stablizes a P-tum type IF. This could be important in the design of biologically active peptides (agonists or antagonists) especially when the structure-activity/selectivity relationships require a D-amino acid and/or N-methylated residue. These linear analogues have already provided an interesting discussion to the nature of these peptides. Despite the introduction of local constraints on the peptide backbones, these linear peptides are still very flexible along the psi, phi, and chi space.

2.5 Conclusions

A series of linear peptides were designed and synthesized to interact with CCK receptors as antagonists and opioid receptors as agonists. The design of the linear peptides was based on our hypothesis that peptide opioid and CCK ligands have overlapping pharmacophoric groups. These compounds were tested for binding and functional activity in human CCK-A and CCK-B receptors as well as human delta opioid receptors and rat mu opioid receptors. These compounds were also tested in vitro for opioid agonist activities in MVD and GPI assays and for CCK antagonist properties in unstimulated GPI/LMMP assays. 72 The structure activity relationships of the CCKVopioid peptides showed interesting biological activities at CCK and opioid receptors. Substitution of Nle^ produced a more balanced activity between CCK-A and CCK-B receptors as seen in compound 13

(RSA504). Also, substitution of D-Trp"^ when position 5 is N-MeNle showed antagonist properties at CCK receptors while maintaining the opioid agonist properties as seen in compound 14 (RSA601). These structure activity relationships support the hypothesis that peptide opioid and CCK ligands have overlapping pharmacophores. 73 2.6 Experimental section

2.6.1 Abbreviations

Abbreviations used for amino acids and designation of peptides follow the rules of the lUPAC-IUB Commission of Biochemical Nomenclature in J. Biol. Chem. 1972,

247, 977-983. Additional abbreviations are used as follows: AAA, amino acid analysis;

Boc, iert-butyloxycarbonyl; Fmoc, 9-fluorenylmethoxycarbonyl; fBu, tert-huiyV, ACN, , acetonitrile (CH3CN), DCM, dichloromethane; DIEA, A'jA^-diisopropylethylamine; DMF,

A^,A'-dimethylfonnamide; ESI-MS, electrospray ionization mass spectrometry; HBTU, 2-

(lH-benzotriazol-l-yl)-l,l,3,3,-tetramethy-uronium hexafluorophosphate; HOBt, N- hydroxybenzotriazole; HR, high resolution; MALDI-TOF MS, matrix-assisted laser desorption ionization/time-of-flight mass spectrometry; PyBrop, Bromo-tris-pyrrollidino- phosphonium hexafluorophosphate; TES, triethylsilane; TIS, triisopropylsilane; TFA, trifluoroacetic acid; RP-HPLC, reversed-phase high performance liquid chromatography;

TLC, thin layer chromatography;

2.6.2 Materials

All peptides analogues were synthesized manually using a general protocol for peptide synthesis with N"-Fmoc/^butyl chemistry. The manual synthesis was employed on a glass reaction vessel fitted with a course fnt with a three-way stopcock which allowed argon gas to pass through to agitate the resin and a vacuum line to remove excess reagents and solvents. Rink Amide AM resin (200-400 mesh, 0.6-0.7 mmol/gram lA substitution) was purchased from Novabiochem (U.S.A). N"-Fmoc-Phe-OH, N'^-Fmoc-

Asp(OtBu)-OH, N"-Fmoc-Trp(N'"-Boc), N"-Fmoc-Pro-OH, N'^-Fmoc-Gly-OH, N"-

Fmoc-Tyr(0-®u)-OH, N"-Fmoc-D-Trp(N'"-Boc) were from American Peptide Co.,

(Sunnyvale, CA); N"-Fmoc-D-Ala-OH, N"-Fmoc-D-Pro-OH, N"-Fmoc-D-Phe-OH, N*^-

Fmoc-D-Val-OH, and N"-Fmoc-D-Nle were purchased from Novabiochem (U.S.A.); N"-

Fmoc-NMe-Nle-OH was from Chemhnpex (Woodale, NJ); HBTU and HOBt were purchased from Quantum Biotechnologies (Montreal, Canada). PyBrop was purchased from Advanced ChemTech (Louisville, KY). DCM, DMF, and TFA were purchased from EM Science (NJ, U.S.A.), piperidine, TES, and TIS were from Aldrich (Milwaukee,

WI), diethyl ether was from Mallinckrodt Baker (Paris, KY). Reagents and solvents were used as packaged and not purified frirther. Acetonitrile and trifluoroacetic acid used for purification were from EM Science (NJ, U.S.A.). The purification of the crude peptides was achieved using a Hewlett-Packard 1100 series HPLC instrument (Agilent-

Technologies) with a Ci8-bonded silica column semi-preparative column (Vydac

218TP1010, 300 A, 1.0 x 25 cm. Separations Group, Hesperia, CA). The separations were monitored at 280 nm with a Hewlett-Packard 1100 series fixed wavelength UV detector or at 220 and 280 nm with a Hewlett-Packard 1100 series multiple variable wavelength UV detector and were integrated with a Hewlett-Packard 3396 series III integrator. Purity of the isolated peptides was assessed with analytical RP-HPLC in two different gradient systems as detected at 230, 254, 280 nm. In all cases, the peptides were greater than 95% pure. The structures of the pure peptides were confirmed by ESI-

MS (Finnigan, Thermoelectron, LCQ classic) and high resolution FAB-MS (JEOL 75 HXllO sector instrument) at the University of Arizona Mass Spectrometry and Protein

Sequencing Facility. The purity of the peptides was checked by TLC on Analtech

(Newark, NJ) silica gel GF plates (250 microns layer thickness) in at least two solvent systems. The AAA was performed using Applied Biosystems model 420A amino acid analyzer with automated hydrolysis (vapor phase at 160°C for 110 min using 6 N HCl, and a precolumn phenylthiocarbamyl-amino acid (PTC-AA) analyzer. Phenylalanine peak was used as the standard. No corrections were made for amino acid decomposition.

University of Arizona Mass Spectrometry and Protein Sequencing Facility.

2.6.3 General Method for Peptide Synthesis

The peptides were synthesized on 0.5 g of Rink Amide AM resin using N"-Fmoc chemistry. The resin was swelled in the reaction vessel with DMF overnight. Initially, the resin was deprotected with 20% (v/v) piperidine in DMF solution (3 + 20 min). The first amino acid, N"-Fmoc-Phe-OH, was coupled to the resin. The following amino acids were coupled sequentially to the growing peptide chain with N"-Fmoc-Asp(Offlu)-OH,

N"-Fmoc-N"-MeNle-OH, N"-Fmoc-Nle-OH, N"-Fmoc-Trp(N'-Boc)-OH, N"-Fmoc-D-

Trp(N'-Boc)-OH, N'-Fmoc-Gly-OH, N"-Fmoc-D-Ala-OH, N"-Fmoc-D-Phe-OH, N"-

Fmoc-D-Nle-OH, N"-Fmoc-Pro-OH, N"-Fmoc-D-Pro-OH, N"-Fmoc-D-Val-OH, or N"-

Fmoc-Tyr(^Bu)OH using standard solid-phase methods. Each coupling reaction was achieved using 3-fold excesses (relative to resin substitution) of amino acid, HBTU, and

HOBt in the presence of 6-fold excess DIEA. The coupling reactions were incubated for 76 one hour. The completeness of the coupling reaction was monitored by a negative Kaiser test, which is indicated by a clear solution and clear resins'The N"-Fmoc-amino acid following the N"-MeNle was activated with PyBrop instead of HBTU. In these cases, 4- fold excesses of the N"-Fmoc-amino acids and coupUng agents were used along with a 8- fold excess of DIEA and the coupling reaction was incubated for two hours. In addition to the Kaiser test, the chloranil test^'^ and the TNBS test'"' were used. The N"-Fmoc protecting group on the amino acid was removed with piperidine (20% in DMF, 1x3 min, 1 X 25 min). The coupling and deprotection steps were each followed by washes with DMF (3x1 min) and DCM (3x1 min). Completion of the Fmoc deprotection was assessed by positive Kaiser test, which is indicated by a deep blue or purple solution and dark blue resins following the test. In the cases of coupling residues to Pro and NMeNle, a chloranil test was used. After the deprotection of the final N"-terminal Fmoc, the resins were washed with DMF. After a final thorough wash with DCM, removing the DMF, the peptidyl resin was dried with vacuum and ambient air.

The dried peptidyl resins was transferred to a bromosilicate scintillation vial. The peptide was cleaved fi-om the resin and the other side chain protecting groups removed by incubation in a cleavage cocktail (10 mL/ grams) of used varied with 95% TFA, 2.5%

TIS, 2.5% water for 1 hr and 30 min. The resin was filtered through a cotton plugged glass pipette. The resins were then washed with additional TFA (~2 mL) for five min.

The TFA-peptide solution was transferred to a 15 mL propylethylene conical centrifuge tube (Falcon). The volume of TFA of the filtrate was reduced with a gentle flow of inert gas to about 1.5 mL. The peptide was precipitated upon slow addition of cold diethyl 77 ether (12 mL). The precipitate was isolated by centrifugation using a bench top centrifuge (Hamilton Bell, VanGuard V65000). The organic solvent was decanted off and discarded. The pellets were then washed three times with cold diethyl ether (12 mL).

The pellet was dried over air, yielding 55-90% of crude peptides.

2.6.4 Purification

The crude peptides were purified by using a Cig semi-preparative column. The crude peptides was loaded into the column at a concentration of 10 mg/mL. For 100 mg dried crude peptide, the peptide was "wetted" with ACN (~1 mL). Aqueous 0.1% TFA was added slowly until peptides precipitated (~6 mL). ACN was added until the crude peptides until they were dissolved (~2 mL). Iterative addition of aqueous 0.1% TFA and

ACN (or ethanol or methanol) was done until peptide was fully dissolved, without exceeding 20% ACN. The solution was allowed to sit at room temperature overnight, or until side product peak is no longer detected by HPLC, to decarboxylate the intermediate carbamic acid on the Trp. The dissolved peptide was filtered through a 0.45 micron cellulose acetate filter (Aerodisc) to remove small inorganic salts. For the Cig semi- preparatative sized column, the maximum loading capacity of 10 mg/mL was injected.

The injection volume was adjusted depending on the detector capacity and resolution of the desired peak. Generally, the gradient used was 20 to 60% ACN in aqueous 0.1%

TFA in 30 min at a flow rate of 3 mL/ min. The gradient was adjusted depending on the resolution of the desired peak from the impurities. After pooling the collected fractions. 78 ACN was removed by rotary evaporation. The aqueous solution was then pooled in a 50 mT, polypropylethylene conical tube (Falcon) vial and frozen for lyophilization. 79 CHAPTER 3

DESIGN, SYNTHESIS, AND STRUCTURE-ACTIVITY RELATIONSHIPS OF

CCK/OPIOID PEPTIDES: SUBSTITUTION OF BULKY RESIDUES

3.1 Results: Design of peptides with substitution of bulky residues

One of our major goals was to obtain a CCK antagonist. In Chapter 2, linear analogues of the CCK/opioid peptides inhibited the activity of CCK-8 in the GPL For example, RSA502 (Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2) demonstrated moderate CCK antagonist property in the unstimulated GPI/LMMP assay (K^ = 40 nM, Table 2.7) while simultaneously displaying potent opioid agonist properties (IC50 = 12. nM at the MVD).

RSA502 also showed high binding affinities at human CCK-B receptors (Ki = 8.1 nM), and only moderate binding at the human CCK-A receptors (Ki = 120 nM), and poor agonist potency at these receptors (EC50 =110 and 630 nM, respectively). RSA502 displayed at least partial agonist property at CCK receptors. To improve the antagonist

CCK receptors properties, while maintaining potent agonist properties at the opioid receptors, it was decided to introduce bulky residues.

Introduction of bulky residues in the CCK/opioid linear peptides was inspired by the common theme of hydrophobic and bulky moieties found in non-peptide CCK antagonists. A brief survey of nonpeptide CCK antagonists is illustrated in Figure 3.1.

89,92,111,129 (j^_ggg other dipeptoid series contain a bulky adamantyl group, an C"- methylated Trp, and another phenyl group. There are also two aromatic groups in the 80 scaffold in the benzodiazepene analogues of CCK antagonists. These common hydrophobic, aromatic groups are also present in benzotript, proglumide, ureidoactamide analogues, and other CCK antagonists (Figure 3.1). Thus to improve the antagonist properties of CCK/opioid analogues at the CCK receptors, while maintaining the opioid agonist properties, bulky, hydrophobic residues were substituted at position 4.

Naphthalanine 2' and 1' (22-24) and 5'-phenyltryptophan (25-26) were substituted into position 4. Position 4 was also substituted with dihydrotryptophan (Dht) 27. Analogue

28 was substituted with D-homophenylalanine (D-Hph) at position 2. Table 3.1 lists the

CCK/opioid peptides substituted with bulky residues. 81

HO^ /O

O 0.^ .N N. N O

O OH

CI-988 Proglumide Horwell's dipeptoid

O H H N O ^O ^ N V SO3H O O

N-CH,

L-365,260, Lotti and Chang benzodiazepene series RP-73870 Pendley's ureidoacetamides series

N N HO P HN OH

Dibutyryl-cGMP Mercedes's CCK-A receptor antagonist

Figure 3.1. Examples of nonpeptide CCK antagonists. NH2 HO O HO' "O

I'-Naphthalanine 2'-Naphthalanme

V/ ,NH2 N HO O H

2',3'-Dihydotryptophan 5'-Phenyltryptophan (racemic mixture)

Homophenylalanine

Figure 3.2. Structures of bulky analogues substituted at position 4 and 2. 83

Table 3.1. CCK/opioid analogues with bulky residues substituted at positions 4 and 2.

No. Code Sequence 22 RSA801 Tyr-D-Phe-Gly-Nal(2')-NleAsp-Phe-NH2

23 RSA802 Tyr-D-Phe-Gly-D-Nal(2')-Nle-Asp-Phe-NH2

24 RSA803 Tyr-D-Phe-Gly-Nal(l')-Nle-Asp-Phe-NH2

25 RSA806 T yr-D-Phe-Gly-Trp(5' Plie)-Nle-Asp-Phe-NH2

26 RSA805 Tyr-D-Phe-Gly-D-Trp(5'Phe)-Nle-Asp-Phe-NH2

27 RSA804 Tyr-D-Phe-Gly-Dht®-NMeNleAsp-Phe-NH2

28 RSA505 Tyr-D-HPh^-Gly-Trp-Nle-Asp-Phe-NH2

^Dht, dihydrotryptophan (racemic mixture). ''D-HPh, D-homophenylalanine. Bold, residues of interest. 84 3.2 Results: Structure-activity relationships

The competition binding assays against radiohgands at the delta and mu opioid receptors and CCK-A and CCK-B receptors, as well as the functional assays, [^^S]GTP-y-

S assays and phosphoinositide hydrolysis (PI) assays, were performed by Dr. Shou-Wu

Ma and Hamid Badghisi from the laboratory of Dr. Josephine Lai at the Department of

Pharmacology in the University of Arizona.

The opioid agonist activities at the delta and opioid receptors in isolated mouse vas deferens (MVD) and guinea pig ileum (GPI) tissues, respectively, and the CCK agonist and antagonist activities in GPI tissues were determined and performed by Peg

Davis from the laboratory of Dr. Frank Porreca at the Department of Pharmacology in the

University of Arizona.

The in vitro assays were performed as described in Chapter 2 and the experimental details of the tissue assays are summarized in the Appendix.

The data for the competitive binding assay and the corresponding functional PI at

CCK receptors are summarized in Table 3.2 and Table 3.3, respectively. The data for the competitive binding assay and the corresponding functional [^^SJ-GTP-y-S binding assay at delta and mu opioid receptors are summarized in Table 3.4 and Table 3.5, respectively.

The functional opioid agonist assay in the MVD and GPI tissues and CCK agonist and antagonist activities in the GPI/LMMP tissues are summarized in Tables 3.6 and Table

3.7, respectively. 85 Table 3.2. Binding affinities of CCK/opioid peptide analogues at the human CCK-A and

CCK-B receptors.

Binding Affinity^ No. Sequence (Kj, nM) CCK-A CCK-B RSA502 Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 120 8.1

RSA210 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 >10000 2.13

22 Tyr-D-Phe-Gly-Nal(2')-Nle-Asp-Phe-NH2 170 100

23 Tyr-D-Phe-Gly-D-Nal('2)-Nle-Asp-Phe-NH2 2100 1800

24 Tyr-D-Phe-Gly- Nal(l ')-Nle-Asp-Phe-NH2 2700 3300

25 Tyr-D-Phe-Gly-Trp(5'Phe)-Nle-Asp-Phe-NH2 >10000 nc

26 Tyr-D-Phe-Gly-D-Trp(5'Phe)-Nle-Asp-Phe-NH2 nc nc

27 Tyr-D-Phe-Gly-Dht-MeNle-Asp-Phe-NH2 nc nc

28 Tyr-D-HPh'-Gly-Trp-Nle-Asp-Phe-NH2 nc nc

^Competition against [^^^I]CCK-8 (sulfated) in expressed hCCK-A and hCCK-B receptors in HEK cell lines. Bold, residues of interest; nc, no competition. 86 Table 3.3. Functional activities of CCK/opioid peptide analogues at the CCK receptors

Functional Analysis^ No. Sequence (EC50, nM) CCK-A •^maxF CCK-B F RSA502 Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 630 5.3 110 8.4

RSA210 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 nr ~ 16. 9.3

22 Tyr-D-Phe-Gly-NaI(2')-Nle-Asp-Phe-NH2 1600 4.2 43 12

23 Tyr-D-Phe-Gly-D-Nal(2')-Nle-Asp-Phe-NH2 nr — 120 4.8

24 Tyr-D-Phe-Gly-Nal(l ')-Nle-Asp-Phe-NH2 3000 6.1 2200 16.

25 Tyr-D-Phe-Gly-Trp(5'Phe)-Nle-Asp-Phe-NH2 nr ~ nr ~

26 Tyr-D-Phe-Gly-D-Trp(5'Phe)-Nle-Asp-Phe-NH2 nr — nr ~

27 Tyr-D-Phe-Gly-Dht-MeNle-Asp-Phe-NH2 nr 2.4 nr 1.2

28 Tyr-D-HPh-Gly-Trp-Nle-Asp-Phe-NH2 300 12. 29. 41.

^Phosphoinositide (PI) hydrolysis assay in hCCK-A and hCCK-B, expressed receptors in HEK cell lines. Emax = (total PI hydrolysis/basal PI hydrolysis). Bold, residues of interest; nr, no response. 87 Table 3.4. Binding affinities of CCK/opioid peptide analogues at the opioid receptors.

Binding Affinity® No. Sequence (Kj, nM) hDOR rMOR RSA502 Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 0.42 79

RSA210 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NHi 6.8 140

22 Tyr-D-Phe-Gly-Nal(2')-Nle-Asp-Phe-NH2 1.8 370

23 Tyr-D-Phe-Gly-D-Nal(2')-Nle-Asp-Phe-NH2 0.12 >10000

24 Tyr-D-Phe-Gly-Nal(l ')-Nle-Asp-Phe-NHi 20. 97

25 Tyr-D-Phe-Gly-Trp(5'Phe)-Nle-Asp-Phe-NH2 790 nc

26 Tyr-D-Phe-Gly-D-Trp(5'Phe)-Nle-Asp-Phe-NH2 950 6200

27 T yr-D-Phe-Gly-Dht-MeNle-Asp-Phe-NH2 200 4800

28 Tyr-D-HPh-Gly-Trp-Nle-Asp-Phe-NH2 0.23 15.

^Competitive assay against radiolabelled [^H] DPDPE at hDOR and [^HJDAMGO at rMOR. hDOR and rMOR were expressed fi-om CHO cell lines. Bold, residues of interest; nc, no competition. 88 Table 3.5. Functional activities of CCK/opioid peptide analogues at the opioid receptors.

Functional Activity® (Agonist) No. Sequence (ECso, nM) hDOR Emax rMOR -Lrnax RSA502 Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 21 34 520 53

RSA210 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 23 310 2.5 84.

22 Tyr-D-Phe-Gly-Nal(2')-Nle-Asp-Phe-NH2 44. 27. 20 180

23 Tyr-D-Phe-Gly-D-Nal(2')-Nle-Asp-Phe-NH2 1000 140 3900 110

24 Tyr-D-Phe-Gly-Nal(l')-Nle-Asp-Phe-NH2 21. 26 1300 150

25 Tyr-D-Phe-Gly-Trp(5'Phe)-Nle-Asp-Phe-NH2 nr ~ nr ~

26 Tyr-D-Phe-Gly-D-Trp(5'Phe)-Nle-Asp-Phe-NH2 1300 18 1600 21

27 T )/r-D-Phe-Gly-Dht-MeNle-Asp-Phe-NH2 26. 26 190 36

28 Tyr-D-HPh-Gly-Trp-Nle-Asp-Phe-NH2 48. 64 12. 45

"[^^SJGTP-y-S binding assay. Emax = (net total bound/basal binding) X 100%. Opioid receptors were expressed from CHO cell lines. Bold, residues of interest; nr, no response. 89

Table 3.6. Functional activities of CCK/opioid peptide analogues at the MVD and GPL

Opioid Agonist®, No. Sequence (IC50. nM) MVD(delta) GPI(mu) RSA502 Tyr-D-Phe-Gly-Tip-Nle-Asp-Phe-NH2 12±2.0 420±68

RSA210 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 65±9.5 100±23

22 Tyr-D-Phe-Gly-NaI(2')-NleAsp-Phe-NH2 17±1.5 910±330

23 Tyr-D-Phe-Gly-D-Nal(2')-Nle-Asp-Phe-NH2 275±88 6000±1400''

24 Tyr-D-Phe-Gly-Nal(l ')-Nle-Asp-Phe-NH2 64±8 210±50

25 Tyr-D-Phe-Gly-Trp(5'Phe)-Nle-Asp-Phe-NH2 357±52 n/d

26 Tyr-D-Phe-Gly-D-Trp(5'Phe)-Nle-Asp-Phe-NH2 74±9 n/d

27 T yr-D-Phe-Gly-Trp -Dht-NMeNleAsp-Phe -NH2 88±24 n/d

28 Tyr-D-HFh-Gly-Trp-Nle-Asp-Phe-NH2 140±11 n/d

^Concentration at 50% inhibition of muscle contraction at electrically stimulated isolated tissues, ^ot a mu antagonist. Bold, residues of interest; n/d, not determined; 90 Table 3.7. Functional activities of CCKyopioid peptide analogue at the unstimulated GPI/LMMP.

unstimulated GPI/LMMP CCK- No. Sequence Agonist® CCK-Antagonist'' (Aso) (Ke, nM) RSA502 Tyr-DPhe-Gly-Trp-Nle-Asp-Phe-NH2 0% at 1 )j,M'^ 40±20

RSA210 Tyr-DPhe-Gly-Trp-MeNle-Asp-Phe-NHi 0% at 1 fO-M 250±160

22 Tyr-D-Phe-Gly-Nal(2')-NleAsp-Phe-NH2 0% at 1 [iM 982±361

23 Tyr-D-Phe-Gly-D-Nal(2')-Nle-Asp-Phe-NH2 0% at 1 i^M 0% at 1 |j,M

24 Tyr-D-Phe-Gly-Nal(l')-Nle-Asp-Phe-NH2 0% at 1 |j,M 77±16

25 Tyr-D-Phe-Gly-Trp(5'Phe)-Nle-Asp-Phe-NH2 n/d n/d

26 Tyr-D-Phe-Gly-D-Trp(5'Phe)-Nle-Asp-Phe-NH2 n/d n/d

27 Tyr-D-Phe-Gly-Dht-NMeNIeAsp-Phe-NH2 n/d n/d

28 T yr-D-HPh-Gly-T rp-Nle-Asp-Phe-NH2 n/d n/d

''Contraction of isolated tissue relative to intial muscle contraction with KCl. Mhibitory activity against the CCK-8 induced muscle contraction. Kg, concentration of antagonist needed to inhibit CCK-8 to half its activity. "^No response. Bold, residues of interest; n/d, not determined. 91 3.2.1 Substitution of Nal residues at position 4

It was hoped to introduce antagonist activity at the CCK receptors by replacing

Trp"^ with a bulky analogue of tryptophan. This strategy was introduced by substituting residue 4 with Nal(2'), Nal(r) and D-Nal(2') using RSA502 as a template. In these substitutions, it was hoped the agonist activities at the opioid receptors would be retained or improved.

In the unstimulated GPI/LMMP receptor antagonist assays (Table 3.7), substitution of L-Nal(2') and L-Nal(r) derivatives at position 4 led to analogues that did not have any agonist activities. Substitution with Nal(2') resulted in 22 (Ke - 982 nM) which had a 25-fold loss in CCK antagonist activity compared to RSA502, while substitution with Nal(r) resulted in 24 (Ke = 76 nM) which only had a two-fold loss in antagonist activity. In contrast, when D-Nal(2') 23 was substituted in position 4, there was no antagonist activity, even at high concentrations. These findings suggest that replacement of Trp with Nal as a bulky analogue does not increase the antagonist activity at CCK receptors. Nonetheless, the antagonist activities of 22 and 24 were good.

In the competitive binding assays at the cloned receptors (Table 3.2), substitution of L-Nal(2') and L-Nal(r) lowered the binding affinity at both CCK-A and CCK-B receptors when compared to RSA502. 22 had good binding affinities at both CCK-A and

CKK-B receptors (Ki = 116 nM and 8.1, respectively). Both 23 and 24, however, weakly bound at low micromolar binding affinities at both CCK receptors (K = 2 - 3.3 |aM).

Consistent with analogues with Nle^ which improved balanced selectivity, the binding 92 affinities, particularly with 22, were fairly balanced between CCK-A and CCK-B receptors with a nearly 1.5:1 ratio.

In the PI hydrolysis agonist assays at CCK-A and CCK-B receptors (Table 3.3), the Nal"^ substitution produced interesting activities at the CCK receptors. 22 was nearly two-fold more potent at CCK-B receptors (EC50 = 43 nM) than RSA502 (EC50 =110 nM), but it was less potent at CCK-A receptors (EC50 = 1600 nM). In contrast with Nal(r) substitution, 24 was weakly potent at CCK-A and CCK-B receptors (EC50 = 3000 and

2200 nM, respectively). The D-amino acid D-Nal(2') 23 produced no response at CCK-

A receptors while at CCK-B receptors, 23 displayed surprisingly good activity (EC50 =

43 nM).

In the MVD and GPI assays (Table 3.6), L-Nal(2') and L-Nal(r) substitutions 22 and 24 produced moderate agonist activity at the opioid receptors. The Nal(2')'^ analogue

22 displayed good agonist activity at the MVD (IC50 =17 nM) while producing moderate activity at the GPI (IC50 = 910 nM). Similarly, the analogue Nal(r)'^ 24 showed good agonist activity at the MVD (IC50 = 64 nM) while the activity was moderate at the GPI

(IC50 = 211.7 nM). Compared to RSA502, 22 and 24 had lower agonist activities at both delta and mu opioid receptors. When the D-amino acid was placed in position 4, 23 produced moderate activity at the MVD (IC50 = 280 nM) and the GPI (IC50 = 6000 nM).

These Nal"^ substituted analogues generally retained the activities at the opioid receptors.

In the competitive binding assays at cloned opioid receptors (Table 3.4), the Nal"^ substitution produced some interesting results. Analogues Nal(2') 22 and Nal(r) 24 produced good binding affinities at the hDOR (K = 1.8 and 20 nM, respectively) and 93 moderate binding affinities at the rMOR (Ki - 370 and 97 nM, respectively).

Interestingly, while Nal(r)'^ 22 displayed good selectivity for the delta opioid receptor

(130:1 6/|i), 24 was only five-fold more selective for the delta opioid receptor over the mu opioid receptor. Furthermore, when the D-amino acid D-Nal(2') was substituted at position 4, the binding affinity of 23 at the hDOR was very good (0.12 nM) while 23 did not bind at the rMOR at very high concentrations (>10,000 nM). 23 was very selective for the delta opioid receptors (>106:1, delta/mu).

In the [^^S]GTP-y-S binding assays at the cloned opioid receptors (Table 3.5), the

22 and 24 had good activities at the hDOR (EC5o= 44. and 21. nM, respectively) with moderate activities at the rMOR (EC5o= 20 and 1300 nM, respectively). At the rMOR,

23 showed full agonist activities at both hDOR and rMOR but at very high concentrations

(EC50 = 1000 and 3900 nM, respectively).

3.2.2 Substitution of S'-phenyltryptophan residues at position 4

In the competitive binding assays at the human CCK-A and CCK-B receptors

(Table 3.2), substitution of a 5'-phenyltryptophan residue (25) and its D-isomer 26 at position 4 resulted in analogues which showed drastic losses in the binding affinities at both CCK-A and CCK-B receptors when compared to RSA502. Analogue 25 showed very poor binding at CCK-A receptors (K = >10,000 nM) while it displayed no competition at the CCK-B receptor. Not surprisingly, in the CCK agonist assays at the human CCK-A and CCK-B receptors (Table 3.3), the 5'-phenyltryptophan analogues 25 and 26 did not produce any agonist response. 94 In the competitive binding assays at human delta opioid receptors and rat mu opioid receptors, the 5'-phenyltryptophan substitution at position 4 resulted in a significant loss in binding affinities when compared to RSA502 (Table 3.4). Analogue

25 showed poor binding affinity at hDOR (Ki = 790 nM), while it displayed no competition at the rMOR. Similarly, the D-isomer 26 resulted in poor binding affinities at hDOR and rMOR (Kj = 950 and 6200 nM, respectively).

In the opioid agonist activation assays at the human delta and mu opioid receptors, analogue 25 did not produce any response at both hDOR and rMOR (Table 3.5). In contrast, 26 displayed very poor opioid agonist activities at the hDOR and rMOR (Ki =

1300and 1600 nM, respectively). The opioid agonist activities at MVD and GPI and

CCK antagonist activities at the unstimulated GPI/LMMP were not determined.

3.2.3 Substitution of dihydrotryptophan at position 4

In the previous chapter, the dihydrotryptophan analogue 27 was a peptide synthetic by-product of the RSA210 synthesis during the cleavage reaction. Instead of isolating the by-product, 27 was synthesized by coupling Fmoc-Dht(Boc)-OH instead of

Fmoc-Trp(Boc)-OH. The Fmoc-Dht(Boc)-OH was a racemic mixture and the resulting in peptides with two peaks under analytical RP-HPLC. The two peaks were not separable using a semi-preparative column in RP-HPLC. The compound tested for bioassays contained both isomers of Dht. 95 In the competitive binding assays at the human CCK-A and CCK-B receptors

(Table 3.2), substitution of Dhtat position 4 resulted in a significant loss in the binding affinity at both CCK-A and CCK-B receptors (no competition). Analogue Dht'^ 27 showed no competition at the CCK-B receptor. Not surprisingly, in the CCK agonist assays at the human CCK-A and CCK-B receptors (Table 3.3), analogue 27 did not produce any agonist response.

In the competitive binding assays at human delta opioid receptors and rat mu opioid receptors, the Dht substitution at position 4 resulted in a loss in binding affinities when compared to RSA210 (Table 3.4). Analogue 27 showed poor binding affinity at hDOR (Ki = 200 nM) and rMOR (K, = 4800 nM). In the opioid agonist activation assays at the human delta and mu opioid receptors, analogue 27 surprisingly resulted in potent opioid activities hDOR and rMOR. Analogue 27 showed good delta opioid activities at the hDOR (K, = 26 nM) and moderate mu opioid activity rMOR (Ki =190 nM). It should be noted that the K values corresponded with limited responses at the hDOR and rMOR

(Emax = 26% and 36%, respectively).

The opioid agonist activities at the MVD and GPI, and the CCK antagonist activities at the unstimulated GPI/LMMP were not determined. 96

3.2.4 Substitution of D-homophenylalanine at position 2

In the competitive binding assays at the human CCK-A and CCK-B receptors

(Table 3.2), substitution of D-homophenylalanine residue 28 resulted in a drastic loss in the binding affinity at both CCK-A and CCK-B receptors vi^hen compared to RSA502.

Analogue 28 showed no competition at CCK-A and CCK-B receptors. Surprisingly, in the CCK functional assays at the human CCK-A and CCK-B receptors (Table 3.3), D- 2 2 HPh 28 resulted in moderate gain in agonist activities compared to RSA502. D-HPh 28 displayed moderate agonist activity at the CCK-A receptor (EC50 = 300 nM) and good

activity at the CCK-B receptor (EC50 = 29. nM). It should be noted that analogue 28 has

the highest Emax at the CCK-B receptor for this functional assay of the analogues thus far.

In the competitive binding assays at human delta opioid receptors and rat mu

opioid receptors, D-homophenylalanine substitution at position 4 significantly increased

the binding affinities when compared to RSA502 (Table 3.4). Analogue 28 showed very

good binding affinity at hDOR and rMOR (Kj = 0.20 and 15 nM).

In the opioid agonist activation assays at the human delta and mu opioid receptors,

analogues 27 showed interesting opioid activity. At the hDOR, the functional activity

was moderate (Ki = 48 nM) while the activity was surprisingly better at rMOR (Ki = 12

nM).

The opioid agonist activities at MVD and GPI and CCK antagonists activities at

the unstimulated at the GPI/LMMP were not determined. 97

3.3 Discussions

To improve the antagonist properties of CCK/opioid peptides at CCK receptors, a series of peptides were designed and synthesized with substitution of bulky residues at the four position. These bulky residues were meant to mimic the bulky, hydrophobic moieties of non-peptide CCK antagonists. The initial set of Nal substitution resulted in interesting biological activities.

In general, Nal(2'), D-Nal(2'), and Nal(r) substitution resulted in moderate binding affinities at the CCK receptors. However, the Nal substitution did not improve antagonist activity against CCK-induced activity at the GPI when compared to RSA502.

The best one was Nal(r) 24 (Kg = 77 nM). Despite the potent CCK antagonist activity of Nal(r) 24, it is the Nal(2') 22 analogue that displayed the more potent binding affinity at the human CCK-A and CCK-B receptors. These biological activity profiles are comparable to those of the previously reported Nal(2') analogue of Ac-CCK-7, which was determined to be as potent as Ac-CCK-7 [Ac-Tyr(S03H)-Met-Gly-Trp^°-Met-Asp-

Phe-NH2] in binding to both CCK receptors.

It is noteworthy that when comparing the binding affinity of Nal(2')'^ containing

22 to the binding affinity of RSA502 at the human CCK-A receptor, the potency did not

significantly change (Kj = 170 vs. 120 nM). Considering the sensitivity of Trp^°

substitution in CCK, particularly at CCK-A receptors, this tolerance of Nal substitution

is quite remarkable. In contrast, the binding affinity of Nal(2') 22 at the CCK-B receptor

significantly dropped 12-fold. This resulted in the desired balanced selectivity between 98 CCK-A and CCK-B receptors (nearly 1.5:1, CCK-B/CCK-A) when compared to RSA502

(14:1, CCK-B/CCK-A).

These results that indicate among CCK analogues, features unique to the indole moiety of Trp such as the presence of a free NH are not absolutely required for potent

CCK agonist activity. However, the interactions of the side chain of Trp replacements with the CCK receptor might depend on a subtle interplay of steric and electronic effects.

While Nal substitution introduced hydrophobicity in the peptide, the bulk of the side

chain group also introduced a constraint in chi space resulting in a bias in the topological • • 1 '^0 conformation in position 4.

These properties of Nal may also apply to opioid• • receptors where the Nal 30

substitution in CCK/opioid was generally tolerated, especially at delta opioid receptors.

However, consistent with the results with D-Trp"^ analogues from the previous chapter,

the substitution with D-Nal(2') generally did not have a very good effect on opioid

agonist activity. These results are consistent with enkephalin amide analogues with

1 •11 substitution of Nal(2') at position 4.

Since Nal'' substitution, for the most part, was tolerated at CCK and opioid

receptors, substitution of Nal for Trp may be a viable approach in the initial design for

the screening of CCKVopioid mimetics. This is especially important when reactivity of

the indole moiety of Trp becomes a synthetic issue.

Other substitutions apart from the bulky Trp analogues at position 4 also may be

considered. For example, it would be an interesting to substitute />ara-halogenated Phe

analogues at position 4 in CCK/opioid peptides to determine the role of electronic effects 99 on the aromatic residue. It has been reported that substitution of the Phe(4-Br) at position

30 of Ac-CCK-7 resulted in a surprisingly potent binding affinity and antagonist property at CCK receptors. Also it is known that [Phe(4-Cl)]-DPDPE analogues are more potent than the parent compounds. From the point of view of overlapping pharmacophores of CCK and opioid ligands, para-halogenated Phe substitution would be an interesting approach to obtaining a CCK antagonist and opioid agonist in a single molecule.

3.4 Conclusions

Bulky residues were substituted at position 2 and 4 in an attempt to increase the

CCK antagonist activity of CCK/opioid peptides while maintaining the agonist activities at opioid receptors. In general, substitution of bulky residues at position 4 did not result in an increase in CCK antagonist activity, while the opioid agonist activities were somewhat retained. The modest binding affinities and activities at the CCK receptors suggest that replacement of Trp with Nal residues might be a reasonable initial approach for the design and synthesis of non-peptide CCK/opioid analogues. 100 CHAPTER 4

DESIGN, SYNTHESIS, AND STRUCTURE-ACTIVITY RELATIONSHIPS OF

CCK/OPIOID PEPTIDES: CYCLIC DISULFIDE ANALOGUES

4.1 Results: Design of cyclic disulfide analogues of CCK/opioid peptides

In Chapters 2 and 3, linear peptides based on the overlapping pharmacophores of opioid and CCK hgands were designed and synthesized. Various substitution strategies were explored including single and multiple amino acid substitutions, as well as substitutions that introduced local conformational constraints with D-amino acids and N- methylated residues. Several linear analogues demonstrated simultaneous potent agonist activities at opioid receptors and antagonist activities at CCK receptors. To further examine the required bioactive conformation for binding at opioid and CCK receptors and improve the agonist properties at the opioid receptors and antagonist activities at

CCK receptors, the linear CCK/opioid peptide analogues were modified by cyclization.

Along with improving the biological activities of the peptides, cyclization induces a conformational constraint to the peptide that allows for the use of biophysical methods to define the parameters (e.g. angles of the psi and phi angles of the peptide backbone, orientation and distances between the pharmacophores) that are necessary for the biological activities at the CCK and opioid receptors. This information would be useful in the de novo design of peptide mimics. Another advantage to cyclization is that it 101 allows the peptide to cross the blood brain barrier (BBB), as well as making the peptide more stable to chemical and enzymatic degradation.

In the design of a globally constrained CCK/opioid peptide, a model peptide was determined by virtue of its binding affinity to opioid and CCK receptors. One of the lead compounds, RSA504 (Tyr^-D-Nle^-Gly^-Trp'^-Nle^-Asp^-Phe^-NH2), had nanomolar binding at the delta and mu opioid receptors (Ki = 2.9 and 27.1 nM, respectively) and

CCK-A and CCK-B receptors (Kj = 11.2 and 15.9 nM, respectively), as well as a balanced selectivity between CCK-A and CCK-B receptor types (nearly a 1:1 ratio).

The lowest energy conformation in water was determined by using a Monte Carlo conformational search. As seen in Figure 4.1, in one of the lowest energy conformations the side chain groups of D-Nle^ and Nle^ are facing on the same side and the terminal alkyl groups are in close proximity to each other. Thus, position 2 and 5 were determined to be appropriate sites of substitution for cyclization. These sites of cyclization are consistent with cychc peptide ligands for opioid receptors, particularly for DPDPE, in which positions 2 and 5 were substituted with D-Pen. ^^Also, this design retains the free jV-Tyr\ which is important for agonist activity at opioid receptors. From the perspective of the CCK receptors, the Phe^ C-terminal was not used for cyclization since this is important for the recognition of peptide hgands for CCK receptors. The summary of the design of the cyclic disulfide peptides is depicted in Figure 4.2.

Cyclization was implemented on the leading compound, RSA504. Positions 2 and 5 were substituted with D-Cys and Cys, as well as D-Pen and Pen. Analogues were oxidized to make a side chain to side chain cyclic disulfide bridge. Pen residues also 102 were considered because the p,p-dimethyl group adds topographical constraint to the cyclic peptide. The peptides synthesized and examined for bioactivity are listed in Table

4.1. 103

TYR TYR

TRP TRP

D-NLE D-NLE

Figure 4.1. Stereoview of lowest energy conformation of RSA504 (Tyr 1 -D-Nle2 -Gly 3-

Trp'^-Nle^-Asp^-Phe^-NHi) based on a Monte Carlo conformational search. D-Nle^ and

Nle^ are oriented on the same side and the terminal alkyl groups are in close proximity to each other. These sites were used for substituting residues appropriate for cyclization. 104

CCK Pharmacophore

Asp-Tyr' -DPhe^-Gly^-Trp^-NMeNle^-Asp^-Phe^-NHi (SNF-9007) v_ J

Opioid Pharmacophore 1. Increase potency at opioid receptors. 2. Improve balanced affinity between CCK-A and CCK-B receptors. 3. Improve CCK receptor antagonist V properties.

Linear Analogues: Tyr-DNle-Gly-Yyy-Nle-Asp-Phe-NH2

4. Global conformational constraint Disulfide cyclization V

Cyclic Analogues: Tyr-c[Xxx-Gly-Yyy-Zzz]-Asp-Phe-NH2

Xxx = D-Cys, D-Pen

Zzz = Cys, D-Cys, Pen

Yyy = Trp, D-Trp

Figure 4.2. Design of the cyclic disulfide analogues. Positions 2 and 5 were used as points of cyclization. Position 2 was substituted with D-Cys or D-Pen. Position 5 was substituted with Cys, D-Cys, or Pen. Position 4 was substituted with Trp or D-Trp. 105 Table 4.1. Cyclic disulfide analogues designed to interact with CCK and opioid receptors.

No. Code Sequence

29 RSAIOIC Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2

30 RSA102C Tyr-c[D-Cys -Gly-Trp -D-Cys] - Asp-Phe-NHi

31 RSA104C Tyr-c[D-Pen-Gly-Trp-Cys]-Asp-Phe-NH2

32 RSA103C Tyr-c[D-Cys-Gly-Trp-Pen]-Asp-Phe-NH2

33 RSA121C T yr-c[D-Cys-Gly-D-Trp-Cys]- Asp-Phe-NHa

34 RSA123C Tyr-c[D-Cys-Gly-D-T rp-Pen]- Asp-Phe-NHa

Pen, penicilliamine. 106 4.2 Results and Discussion: Syntheses of the cyclic disulfide analogues of

CCK/opioid peptide ligands

The cyclic disulfide analogues were synthesized using different strategies. The cyclic peptides were synthesized by first constructing the linear peptide on solid support using standard procedures for Fmoc/Bu strategy. The linear peptide was then released

from the resin and oxidized in solution, or the fully protected peptide was cyclized on- resin and then released from the resin. In both strategies, different advantages and disadvantages were observed.

The first cyclization strategy used solution phase potassium ferricyanide

K3Fe(CN)6 assisted cyclization (Figure 4.3). The standard solid phase protocol using

Fmoc/'Bu strategy on Rink Amide resin was used for the synthesis of the linear peptide.

The iV-Fmoc protected Cys had a side chain protecting group of S'-Trt which was easily

removed using acidic conditions. The Trp was protected with an N'-Boc group. The

Fmoc-CysC^-Trt) was coupled to the growing peptide chain using HBTU/HOBt as the

coupling agents. To minimize racemization during coupling reaction, there was no

preactivation step and the coupling reaction was carried out in a 1:1 mixture of DMF and

DCM, which is a less polar solvent system. '^^Recent reports indicate that TMP

(collidine) is the best base for these coupling conditions. ^^^Special attention was given to

the cleavage cocktail. Ordinarily, silane-TFA based cleavage cocktail is used, but

because the thio-side chain protecting group 5'-Trt easily reattaches to the Cys SH, or

alkylates Trp, a more powerful scavenger such like 1,2-ethanedithiol (EDT) was added to 107 the cocktail. When a thiol-based cleavage cocktail was not used, mass spectra showed a mass unit corresponding to Trt. Presumably, the Trt group had alkylated the Trp residue, though MS-MS fragmentation spectra could not positively prove this.

To cyclize the frilly deprotected linear peptide, the linear peptide was dissolved in a basic buffer (pH ~8) solution and slowly added to a solution of potassium ferricyanide (1.2 eq. relative to peptide). The cycUzation was monitored by HPLC and was completed overnight. However, the resulting cyclic peptide had a tendency to precipitate out of the solution and it could not be dissolved easily, even in methanol. This precipitation made it difficult to separate the peptide from the ion exchange resin, which was used to remove excess iron and iron complexes.

H-Tyr(®u)-D-Cys(5'-Trt)-Gly-Trp(Boc)-D-Cys(5'-Trt)-Asp(0®u)-Phe-NH2-Resin

1. Cleavage from resin 95% TFA, 2% EDT, 2% water, 1% TIS y

H-Tyr-D-Cys-Gly-Trp-D-Cys-Asp-Phe-NHa

2. K3Fe(CN6) 3. Ion exchange resin

T

— S- S -] H-Tyr-D-Cys-Gly-Trp-D-Cys-Asp-Phe-NH2

Figure 4.3. Synthesis of cychc peptide by K3Fe(CN6). 108 This problem led us to use a different approach for cyclization. While simple air oxidation was sufficient for the oxidation of the sulfliydryl groups, the reaction was slow, often requiring up to five days. DMSO-assisted oxidation of the linear peptide to the cyclic disulfide was faster (Figure 4.4). the cases where the cyclic peptide was insoluble in aqueous solution, the cyclic peptide was isolated by centrifugation and decantation.

H-Tyr(ffiu)-D-Cys(5-Trt)-Gly-Trp(Boc)-Cys(5-Trt)-Asp(0®u)-Phe-NH2-Resin

1. Cleavage from resin 95% TFA, 2% EDT, 2% water, 1% TIS •

H-Tyr-D-Cys-Gly-Trp-Cys-Asp-Phe-NH2

2. Air oxidation DMSO-assisted

•

— S- S —I H-Tyr-D-Cys-Gly-Trp-Cys-Asp-Phe-NHa

Figure 4.4. Synthesis of cyclic peptide by DMSO-assisted air oxidation.

Alternatively, a faster oxidation and work-up procedure for synthesis of cyclic

disulfide peptide with on-resin cyclization was used (Figure 4.5). This procedure called

for the side chain group of Cys to be protected with as .S-acetamidomethyl (5-Acm) group. 109 which allowed for a concomitant deprotection and oxidation of the sulfhydryl groups on the sohd support. This procedure was also compatible with the presence of Trp when the indole moiety is proctected by N'-Boc. In this protocol, standard Fmoc/'Bu strategy with HBTU/HOBt in situ activation was used. The coupling of Fmoc-Cys-(5'-

Acm) was carried out in DMF/DCM (1:1) with no preactivation to suppress racemization.

138

H-Tyr(^Bu)-D-Pen(5'-Acm)-Gly-Trp(Boc)-Cys(5'-Acm)-Asp(0®u)-Phe-NH2-Resin

1. Tl(tfa3), DMF/anisole (19:1), 12-18 hrs r s- s n H-Tyr-D-Pen-Gly-Trp-Cys-Asp-Phe-NHi-Resin

2. 95% TFA, 2.5% water, 2.5% TIPS

— S- S -] H-Tyr-D-Cys-Gly-Trp-D-Cys-Asp-Phe-NH2

Figure 4.5. Synthesis of cyclic peptide by on-resin cyclization with Tl(tfa)3 110 For on-resin cyclization, the oxidizing agent was carefully chosen. Although, iodine can be used as an agent for the deprotection of S-Acm and as an oxidizing agent, iodine was not ideal because of the possible iodination of the Tyr and Trp residues. In fact, even with a stringent precaution of using acetic acid, the HPLC trace of the crude peptide showed several inseparable peaks. Oxidation with iodine was not used after this.

Another oxidizing agent that was used was thallium trifluoroacetate Tl(tfa)3.

^"^^Once the linear peptide was assembled on the resin, excess Tl(tfa)3 in cold DMF was added. A small amount of anisole was added as a scavenger. The reaction was carried out in an ice bath for at least 12 hours. The oxidation was not monitored. Initially, the solution mixture was dark brown, but the coloration lightens as the reaction progressed.

After the cyclization step, the peptide was cleaved with TFA/TIS cocktail. While the

HPLC trace of the crude compound showed many impurities, the peak corresponding to

the cyclic peptide was isolated with good resolution. The isolated yields varied from 5 to

40%.

The cyclic peptides were purified using reversed-phase high pressure

chromatography. Because some of the cyclic peptides were not easily dissolved in

aqueous 0.1% TFA, alternate organic solvents were used. Of particular interest was the

cyclic peptide RSAlOlc. The crude RSAlOlc was initially dissolved (wetted) in DMSO

and aqueous 0.1% TFA was added slowly to the desired concentration before the cyclic

disulfide peptide precipitated out of the solution. The solvent gradient for the mobile

phase was adjusted to allow for the elution of the DMSO, as detected at 220 nm, and a

clear resolution of the cyclic peptide peak was seen. In the case of RSAlOlc, using a Ill semi-preparative sized Cig column, the gradient was 20-60% CAN in 0.1% aqueous TFA in 60 min with a flow rate of 3 mL/min. RSA102c, RSAlOSc, and RSA104c were soluble in aqueous 0.1%) TFA.

4.3 Results: Structure-activity relationships of the cyclic disulfide analogues

The binding affinities of the CCK/opioid analogues were determined by competition assays against radiolabelled DPDPE and DAMGO using stably transfected hDOR and rMOR, and radiolabelled sulfated CCK-8 for hCCK-A and hCCK-B receptors. The functional activity at the transfected receptors were assessed usmg [ S]-

GTP-y-S binding assays and PI hydrolysis assay for opioid receptors and CCK receptors, respectively. These assays were performed by Dr. Shou-Wu Ma and Hamid Badghisi fi-om the laboratory of Dr. Josephine Lai at the Department of Pharmacology in the

University of Arizona. Opioid agonist activity at delta and mu receptors were determined in vitro using electrically stimulated MVD and GPI, respectively. CCK antagonist was determined in vitro using unstimulated GPI/LMMP against sulfated CCK-8. These assays were performed by Peg Davis from the laboratory of Dr. Frank Porreca at the

Department of Pharmacology in the University of Arizona.

The data for the competitive binding assay and the corresponding ftinctional [^^S]-

GTP-y-S assay at delta and mu opioid receptors are summarized in Table 4.2 and Table

4.3, respectively. The functional opioid agonist assay in the MVD and GPI tissues and

CCK agonist and antagonist activities in the GPI/LMMP tissues are summarized in 112 Tables 4.4 and Table 4.7, respectively. The data for the competitive binding assay and the corresponding functional PI assays at CCK receptors are summarized in Table 4.5 and

Table 4.6, respectively. 113 Table 4.2. Binding affinities of CCK/opioid peptides at the opioid receptors.

Binding Affinity® No. Sequence (Kj, nM) hDOR rMOR Asp- Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 250 5200

Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2(RSA504) 23. 27.

29 Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2 0.2 2.2

30 Tyr-c[D-Cys -Gly-Trp-D-Cys]- Asp-Phe-NHi 3.8 9.2

31 Tyr-c[D-Peii-Gly-Trp-Cys]-Asp-Phe-NH2 0.3 1300

32 Tyr-c[D-Cys-Gly-Trp-Peii]-Asp-Phe-NH2 2.6 1.4

33 T yr-c[D-Cys-Gly-D-T rp-Cys]- Asp-Phe-NH2 57. 43.

34 T yr-c[D-Cys-Gly-D-T rp-Pen]- Asp-Phe-NH2 510 1500

^Competitive assay against radiolabeled [^H] DPDPE at hDOR and [^HJDAMGO at rMOR. Opioid receptors were expressed from CHO cell lines. Bold, residues of interest. 114 Table 4.3 Functional activities of cyclic disulfide CCK/opioid peptide analogues at the opioid receptors.

Functional Activity® (Agonist) No. Sequence (ECso, nM) hDOR Emax rMOR Emax Asp- Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NHa 1.5 74 n/d n/d

Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 5.5 82 0.47 110

29 Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2 780 190 42 69

30 Tyr-c[D-Cy s -Gly-Trp -D-Cys]- Asp-Phe-NH2 200 110 25 110

31 Tyr-c[D-Pen-Gly-Trp-Cys]-Asp-Phe-NH2 14. 160 110 42

32 Tyr-c[D-Cys-Gly-Trp-Peii]-Asp-Phe-NH2 1.8 44 1.7 45

33 Tyr-c[D-Cys-Gly-D-Trp-Cys]-Asp-Phe-NH2 n/d n/d 42 69

34 T yr-c[D-Cys-Gly-D-Trp-Pen] - Asp-Phe-NH2 2200 43 5400 58

''[^^S]GTP-y-S binding assay. Emax = (net total bound/basal binding) X 100%. Opioid receptors were expressed from CHO cell lines. Bold, residues of interest; n/d, not determined. 115 Table 4.4 Functional activities of cyclic disulfide CCK/opioid peptide analogues at the MVD and GPI.

Opioid Agonist^, (IC5o,nM) No. Sequence MVD GPI (delta) (mu) Asp- Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 58.4 180

T yr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 23±9.7 210±52

29 Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2 0.45±0.01 63±5.5

30 Tyr-c[D-Cys-Gly-Trp-D-Cys]-Asp-Phe-NH2 120±7.5 280±9.1

31 Tyr-c[D-Pen-Gly-Trp-Cys]-Asp-Phe-NH2 9.5±1.8 151±24

32 Tyr-c[D-Cys-Gly-Trp-Pen]-Asp-Phe-NH2 15±2.3 2900±81

33 Tyr-c[D-Cys-Gly-D-Trp-Cys]-Asp-Phe-NH2 2400 ± 500 19% at l^iM

34 Tyr-c[D-Cys-Gly-D-Trp-Peii]-Asp-Phe-NH2 27% at 1|iM 5% at 1 )aM

Concentration at 50% inhibition of muscle contraction at electrically stimulated isolated tissues. Bold, residues of interest; n/d, not determined 116 Table 4,5. Functional activities of CCK/opioid peptides at the GPI/LMMP.

unstimulated GPI/LMMP CCK- CCK- No. Sequence Agonist^ Antagonist (A.5O) (Ke, nM) Asp- Tyr-DPhe-Gly-Trp-MeNle-Asp-Phe-NH2 0% at 1 laM' 31

Tyr-DNle-Gly-Trp-Nle-Asp-Phe-NH2 0% at 1 fiM 190 ±80

29 Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2 0% at 1 |j.M 7.6 ±2.3

30 Tyr-c[D-Cys-Gly-Trp-D-Cys]-Asp-Phe-NH2 0% at 1 (a,M 23 ±9.1

31 T yr-c[D-Peii-Gly-Trp-Cys]- Asp-Phe -NH2 0% at 1 |j,M 19 ±6.7

32 Tyr-c[D-Cys-Gly-Trp-Pen]-Asp-Phe-NH2 0% at 1 i^M 24 ± 7.8

33 T yr-c[D-Cys-Gly-D-Trp-Cys] -Asp-Phe -NH2 0% at 1 |j,M 150 ±58

34 Tyr-c[D-Cys-Gly-D-Trp-Peii]-Asp-Phe-NH2 0% at 1 [iM 0% at 1 )j,M

^Contraction of isolated tissue relative to initial muscle contraction with KCl. Inhibitory activity against the CCK-8 induced muscle contraction. ICe, concentration of antagonist needed to inhibit CCK-8 to half its activity. '^No response. Bold, residues of interest. 117 Table 4.6. Binding affinities of cyclic disulfide CCK/opioid peptide analogues at the CCK receptors

Binding Affinity^ No. Sequence (Kj, nM) CCK-A CCK-B Asp- Tyr-D-Phe-Gly-Trp-MeNle-Asp Phe-NHa n/d 2.1

Tyr-DNle-Gly-Trp-Nle-Asp-Phe-NHa 11.2 15.8

29 Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2 >10000 nc

30 Tyr-c[D-Cys-Gly-Trp-D-Cys]-Asp-Phe-NH2 2500 4900

31 T yr-c[D-Pen-Gly-Trp-Cys]- Asp-Phe-NHi >10000 >10000

32 Tyr-c[D-Cys-Gly-Trp-Peii]-Asp-Phe-NH2 nc nc

33 T yr-c[D-Cys-Gly-D-Trp-Cys]- Asp-Phe-NHa 2300 >10000

34 Tyr-c[D-Cys-Gly-D-Trp-Pen]-Asp-Phe-NH2 nc nc

"Competition against ['^^I]CCK-8 (sulfated) at hCCK-A and hCCK-B receptors expressed in HEK cell lines. Bold, residues of interest; nc, no competition. 118 Table 4.7. Functional activities of cyclic disulfide CCK/opioid peptide analogues at the CCK receptors

Functional Analysis^ No. Sequence (EC50, nM) CCK-A Emax CCK-B Emax

Asp- Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NH2 n/d ~ 1.5 74

Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 2500 15. 1900 10.

29 Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2 nr ~ nr —

30 T yr-c[D-Cys-Gly-Trp-D-Cys]- Asp-Phe-NH2 nr — nr —

31 Tyr-c[D-Pen-Gly-Trp-Cys]-Asp-Phe-NH2 nr — nr ~

32 Tyr-c[D -Cys -Gly-Trp -Pen]- Asp-Phe-NHi nr ~ 5500 ~

33 Tyr-c[D-Cys-Gly-D-Trp-Cys]-Asp-Phe-NH2 ns ns >10000 3.8

34 Tyr-c[DCys-Gly-D-Trp-Peii]-Asp-Phe-NH2 nr ~ nr ~

^Phosphoinositide (PI) hydrolysis assay in hCCK-A and hCCK-B, expressed receptors in HEK cell lines. Emax = (total PI hydrolysis/basal PI hydrolysis). Bold, residues of interest; n/d, not determined; nr, no response; ns, non-saturable. 119

When compared to the linear analogues SNF-9007 and RSA504, most of the cyclic disulfide analogues had significantly increased binding affinities at both delta and mu opioid receptors in the low nanomolar range. As shown in Table 4.2, the binding affinity of cyclized c[D-Cys^,Cys^] containing analogue 29 was significantly increased to an impressive nanomolar range at delta (Ki = 0.2 nM) and mu (Ki = 2.2 nM) opioid receptors, with a nearly a ten-fold selectivity for the delta opioid receptor. However, when a D-Cys was substituted at position 5 to produce cyclized c[D-Cys^,Cys^] 30, the binding affinity slightly decreased compared to 29 while still potent at delta (Ki = 3.8 nM) and mu (Ki = 9.2 nM) opioid receptors. To introduce more constraint in the cyclic disulfide bridge, D-Pen or Pen was substituted at position 2 or 5 to produce cyclic c[D-

Pen^, Cys^] 31 and cyclic c[D-Cys^, Pen^] 32, respectively. the A^-terminal in analogue 31, the binding affinity is very potent at delta (Ki = 0.3 nM), while the binding affinity was very poor at the mu (Ki = 1300 nM) opioid receptors.

Surprisingly, when the bulkier Pen residue was on the C-terminal in analogue 32, the binding affinity was very potent both at delta (Ki = 2.6 nM) and mu (Ki = 1.4 nM) opioid receptors, with a slight preference for the mu receptor. In an attempt to introduce an antagonist property at CCK receptors, the D-Trp was substituted at position 4 producing analogues c[D-Cys^,D-Trp^,Cys^] 33 and c[D-Cys^,D-Trp'',Pen^] 34. However, 33 and

34 resulted in significant losses in binding affinities at both delta and mu opioid receptors when compared to the other cyclic disulfide analogues.

In the [^^S]GTP-y-S assays (Table 4.3), 29 and 30 had moderate activities at hDOR (EC50 = 780 and 200 nM, respectively while having good activities at the rMOR 120 (ECso = 42 and 25 nM, respectively). When D-Pen and Pen residues were substituted in

31 and 32, respectively, the activities at the hDOR increased (EC50 = 14 and 1.8 nM). At the rMOR, the activity of 31 moderately decreased (£€50= 110 nM) while 32 increased

(EC50 = 1.7). When D-Trp was substituted in 34, the activities at the hDOR and rMOR decreased to the micromolar range (EC50 = 2200 and 5400 nM, respectively).

In the in vitro assays for the opioid agonist activity (Table 4.4), the cyclic [D-

Cys^,Cys^] 29 showed excellent agonist activity in the MVD (IC50 = 0.45 nM) while maintaining good agonist activity in the GPI (IC50 = 63 nM). When both positions 2 and

5 were substituted with D-Cys resulting in cyclic analogue 30, the agonist activities significantly decreased in the MVD (IC50 =120 nM) and GPI (IC50 = 280 nM) when compared to 29. When D-Pen or Pen residues were substituted at positions 2 or 5 to produce analogues 31 and 32, respectively, the bioactivity of both 31 and 32 were good in the MVD (IC50 = 9.5 and 15 nM, respectively). The agonist properties of 31 and 32, however, were less potent in the GPI (IC50 = 150 and 2900 nM, respectively). When D-

Trp was substituted at position 4 in cyclic analogues 33 and 34, there was a significant loss in bioactivities in the MVD and GPI.

In the CCK antagonist assay in the unstimulated GPI/LMMP (Table 4.5), the cyclic disulfide analogues 29 - 34 did not have any CCK agonist activities at high concentrations. However, when tested against the sulfated CCK, the cyclic disulfide analogues 29-32 produced potent CCK antagonists properties (Kg = 7.6 - 24 nM). In analogues 33 and 34, where position 4 is D-Trp, there is a significant loss in the CCK- antagonist activity. 121 In the competition binding assays and functional PI hydrolysis assays at the cloned CCK receptors, cyclic disulfide analogues 29 - 34 did not have significant binding affinity and activity at both hCCK-A and hCCK-B receptors (Table 4.6 and 4.7)

4.4 Discussions

4.4.1 Assays

These cyclic disulfide analogues showed potent CCK antagonist properties against sulfated CCK-8 in the unstimulated GPI/LMMP assays. However, in the competitive binding assays against radiolabelled sulfated CCK, the cyclic analogues showed very low binding at both CCK-A and CCK-B receptors. Furthermore, the cyclic disulfide analogues showed no response in the CCK agonist PI assays. The difference between the tissue assays and the whole cell assay might suggest that the CCK antagonist property of the cyclic disulfide analogues is not due to an interaction at the binding site of sulfated CCK. Rather, the cyclic disulfide analogues might be interacting at an allosteric site. Other assays should be explored to support these interesting assays results. For example, the binding affinity of the CCK/opioid ligands could be also be determined in competition with radiolabelled sulfated CCK in vitro with guinea pig brain (GPB) and guinea pig ileum (GPI) assays. 122 4.4.2 Comparison of cyclic disulfide CCK/opioid peptide analogues to opioid

ligands

The cycHc disulfide analogues 29-34 are conformationally restricted by the virtue of the 14-membered {i, «+3] disulfide-containing rings. The Pen-containing analogues are further constrained by the p,p-diniethyl substituents. As mentioned in the results section, the cyclic disulfide analogues 29-34 produced potent agonist properties and potent antagonist CCK activities in the MVD and GPI in vitro tissue assays. Since the design of these disulfide analogues was based on the very potent delta opioid ligand

DPDPE, ^^the potent activities at the MVD were expected. Furthermore, CCK/opioid analogues 29- 32 have similar bioactivities and selectivity at the MVD and GPI, when compared to the corresponding cyclic disulfide enkephalin or enkephalinamide analogues.

56-58,144

In contrast and surprisingly, these novel cyclic disulfide CCK/opioid analogues retained the potent bioactivities in the MVD and GPI even when position 4 is a Trp, which is a bulkier aromatic residue than Phe. In fact, analogue 29 is six-fold more potent than DPDPE. This increased potency at the MVD is consistent with bulkier analogues such as Nal^^' and para-substituted Phe"^"^'^^^ in position 4.

Moreover, these cyclic CCK/opioid analogues have a unique "address" sequence at the C-terminal end of the peptide for the opioid receptors. Thus, these CCK/opioid cyclic disulfide analogues can be compared to a series of highly delta receptor selective ligands in which a Phe^ has been added to c[D-Pen^,L-Cys^]-enkephalin. ^'^^When 123 compared, the charged side chain residue of the -Asp^-Phe^-NH2 does not seem to significantly affect potency at the mu opioid receptors at the GPI, but loses some potency at the delta opioid receptors at the MVD.

4.4.3 Substitution with D-Trp in cyclic disulfide CCK/opioid peptide analogues

D-Trp was introduced again in the cyclic disulfide as an approach to make CCK antagonists. While substitution of D-Trp at position 4 was an effective strategy in introducing antagonist activity in the unstimulated GPI/LMMP assays with the linear series from Chapter 2, it was not an effective approach with the cyclic disulfide analogues. Substitution of D-Trp"^ resulted in 33 and 34 which had dramatic decreases in the agonist bioactivity at the delta and mu receptors. This drop in potency is consistent with the poor potency of (2R,3S) and (2R,3S)-PMePhe analogues of DPDPE in which the potency dropped significantly in the MVD and GPI^^^. This loss in potency was somewhat expected, since the substitution of D-Trp only induced antagonist property when position 5 is N-MeNle but not when position 5 is Nle, detailed in the Discussion in the previous chapter. In these cyclic disulfide analogues, position 5 is not N-methylated

(i.e. position 5 is not N-MeCys or N-MePen). In the case of analogues 33 and 44, the local constraint and conformations between position 4 and 5 may have more role in the bioactive conformation than the global constraint introduced by the side chain to side chain disulfide [/,/+3] cyclization. Thus in fixture studies, it might be interesting to synthesize cyclic disulfide analogues with N-methylation at position 5: 124

Tyr-c[D-Cys-Gly-Trp-N-MeCys]-Asp-Phe-NH2

Tyr-c[D-Cys-Gly-D-Trp-N-MeCys]-Asp-Phe-NH2

The synthetic route for N-methylation of amino acids in the growing linear chain on the solid support is well established.

4.5 Conclusions

Conformationally constrained cyclic disulfide analogues of the CCK/opioid peptides were designed and synthesized in an attempt to make potent agonists at opioid receptors and antagonists at CCK receptors. Among the cyclic disulfide analogues, c[D-

Cys^,Cys^] 29 displayed the best agonist activity potency at opioid receptors in the MVD and GPI, with a high selectivity for the delta opioid receptor (IC50 = 0.45 and 63 nM, respectively). Analogue 29 also demonstrated potent antagonist activity against CCK- induced activity in the GPI/LMMP assay. Analogue 29, as well at the other cyclic disulfide analogues, displayed very poor binding affinities and fiinctional activity at

CCK-A and CCK-B receptors expressed in HEK cells. The difference between the activity in the GPI tissue assays and binding in the isolated expressed CCK receptors may suggest that the antagonist activity at CCK receptors is an allosteric effect. Substitution of D-Trp in position 4 resulted in a significant loss in binding and activity at opioid and

CCK receptors. 125 4.6 Experimental section

4.6.1. General method for the synthesis of linear peptide for solution phase cyclization

Standard tert-hvAyl side chain protected Fmoc-amino acids were used, except for

N"-Fmoc-Cys in which side chain group is protected with 5'-trityl (S-Trt) group. Rink

Amide AM resin was swelled in a reaction vessel with DMF overnight, hiitially, the resin was Fmoc-deprotected with 25% (v/v) piperidine in DMF solution. Most coupling reactions were achieved using a three-fold excess (relative to resin substitution) of Fmoc- amino acid, HBTU, and HOBt in the presence of six-fold excess DIEA in DMF. The coupling of Cys residue was done with Fmoc-Cys(5'-Trt)-OH (3 eq.), HBTU (3 eq.), and

HOBt (3 eq.) in a solution of DMF/DCM (1:1) with no preactivation, and the base used was 2,4,6-trimethylpyridine (TMP, 6 eq collidine). The cleavage cocktail used was 95%

TFA, 2% EDT, 2% water, and 1% TIPS. The cleavage cocktail was added to the dried resins (10 mL/ gram). The cleavage reaction was limited to 1.5 hours. The resins were filtered and washed with TFA. The peptide was precipitated in cold diethyl ether and was isolated by centrifligation.

4.6.2. General method for potassium ferricyanide disulfide cyclization 126 The crude peptide was taken up in about 20 mL of water. ACN was added to aid in dissolution. The dissolved peptide was slowly added to the 0.01 M aqueous solution of K3Fe(CN)6 with an aid of a syringe pump set to dispense at a rate of 3 mL/ hour. The pH of the solution was monitored and adjusted to pH~8.5 by adding concentrated ammonium hydroxide (NH4OH) or a saturated aqueous solution of ammonium acetate.

After the oxidation was completed, as monitored by HPLC, the pH was adjusted to about

5.0 with acetic acid. An ion exchange resin (IRA-68 Amberlite resin, CI- form) was added to remove excess K3Fe(CN)6 and other iron salts. After one hour, the ion exchange was filtered through a course frit ftmnel. The volume was reduced by rotary evaporation. Evaporation of acetic acid was aided by addition of «-butanol. The remaining aqueous solution was frozen and lyophilized. The crude peptide was purified by semi-preparative HPLC. Monitoring by HPLC was achieved by doing the work-up on a small aliquot of the reaction solution.

4.6.3. General method for air oxidation in disulfide cyclization

The crude linear peptide was dissolved to attain a concentration of 0.1 M.

Acetonitrile was added to aid in dissolution. The solution was aerated with ambient air using a pump while the solution was continually stirred rapidly. The pH was monitored and was maintained at basic pH upon addition of concentrated ammonium hydroxide.

The reaction was monitored by HPLC. After the reaction, the volume was reduced by 127 rotary evaporation. The remaining aqueous solution was frozen and lyophilized. The crude peptide was purified by semi-preparative HPLC.

4.6.4 General methods for DMSO-assisted air oxidation in disulfide cyclization

The crude linear peptide was initially dissolved in DMSO. The peptide concentration was adjusted to 0.1 M with the addition 0.1 M ammonium bicarbonate.

The solution was aerated with ambient air using a pump while the solution was continually stirred rapidly. The reaction was monitored by HPLC. After the cyclization, the volume was reduced by rotary evaporation. The remaining aqueous solution was frozen and lyophilized. The crude peptide was purified by semi-preparative HPLC. In the purification, RP-HPLC conditions were adjusted to elute DMSO with the UV detector set at 220 nm.

4.6.5 General methods for cyclic disulHde synthesis using on-resin cyclization

Standard tert-huXyl side chain protected Fmoc-amino acids were used, except for

N"-Fmoc-Cys in which the side chain group was protected with an i'-acetaminomethyl

(5'-Acm) group. Rink Amide AM resin was swelled in a reaction vessel with DMF overnight, hiitially, the resin was Fmoc-deprotected with 25% (v/v) piperidine in DMF solution. Most coupling reactions were achieved using a three-fold excess (relative to resin substitution) of isT-Fmoc-amino acid, HBTU, and HOBt in the presence of six-fold 128 excess DIEA in DMF. The coupling of Cys residue was done with Fmoc-Cys(5'-Acm)-

OH (3 eq.), HBTU (3 eq.), and HOBt (3 eq.) in a solution of DMF/DCM (1:1) with no preactivation, and the base used was 2,4,6-trimethylpyridine (TMP, collidine 6 eq.) After

the final N"-Fnioc-deprotection, the resins were washed thoroughly with DCM and dried.

The dried resins were transferred to a scintillation vial and were swelled in cold solution of DMF/anisole (19:1, v/v). Thallium trifluoroactate, (Tl(tfa)3, 4 eq.) was added to the

vial. The reaction was kept in an ice bath for 12-18 hours. Following the cyclization, the

resins were washed with DMF and DCM. The cleavage cocktail used was 95% TFA, 2%

EDT, 2% water, and 1% TIPS. The cleavage cocktail was added to the dried resins (10

mL/ gram). The cleavage reaction was limited to 1.5 hours. The resins were filtered and

washed with TFA. The peptide was precipitated in cold diethyl ether and was isolated by

centrifugation.

RSA101c: H-Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2

Rink Amide linked peptide H-Tyr-D-Cys(Acm)-Gly-Trp-Cys(Acm)-Asp-Phe-

NH2 was synthesized using standard Fmoc/Bu soUd phase synthesis. In this procedure,

cychzation was done on the resin before cleavage. To the peptide resin, 10 mL of cold

DMF and 0.426 ng of thallium trifluoroacetate and 0.4 mL anisole were added. The

reaction mixture was incubated at 0 for 15 hours. The resin was washed and cleaved

with 95% TFA, 2.5% H2O, and 2.5% TIS. The HPLC trace had one distinct peak for the

desired product. Isolated yield, 27 mg. 129

RSA101 c: H-Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe -NH2

RSAlOlc was cyclized using DMSO assisted oxidation, as described in the general methods. Rink Amide linked peptide H-Tyr(fBu)-D-Cys(5'-Trt)-Gly-Trp(Boc)-

Cys(5'-Trt)-Asp(OfBu)-Phe-NH2 was synthesized using standard Fmoc/Bu solid phase synthesis. The peptide was cleaved from the resin with a cleavage cocktail consisting of

93.5% TFA, 2.5% EDT (1,2-ethanedithiol), 2.5% H2O, and 1.5% TIS. 320 mg crude peptide was isolated by cold ether precipitation and centrifugation.

110 mg of the crude peptide was dissolved in 10 mL 0.1 M ammonium bicarbonate. Ambient air was pumped into the solution overnight. The solution was cloudy and bubbly. 10 mL DMSO was added to aid cyclization. After four days, the water was reduced in volume by evaporation. The white precipitate formed was dissolved upon addition of DMSO. The volume was adjusted suitable for loading to a preparative reverse phase HPLC. HPLC trace showed one major peak. ~10 mg of peptide was isolated after lyophilization. RSAlOlc was not soluble in water but was clearly soluble in DMSO.

RSA102C H-Tyr-c[D-Cys-GIy-Trp-DCys]-Asp-Phe-NH2

This cyclic peptide was prepared by K3Fe(CN)6 cyclization as described above. 130 RSA103C H-Tyr-c[D-Cys-Gly-Trp-Pen]-Asp-Phe-NH2

This was prepared by DMSO-assisted air oxidation, as described in the general methods.

Starting from 0.5 grams of Rink Amide AM resin, sequential coupling of the N"-

Fmoc/Bu proctected amino acids was done. N"-Fmoc-Cys and N"^-Fmoc-Pen were side chain protected with i'-Trt. Cleavage from the resin and iS'-Trt was done with 94%

TFA/1% EDT/2%TIS/2%H20. Isolated crude peptide was 0.335 grams. 100 mg of the

crude linear peptide was dissolved in 100 mL of 0.1 M ammonium bicarbonate. 1 mL of

DMSO was added. After two days of reaction, the cyclic peptide was frozen and

lyophilized. After RP-HPLC and lyophiUzation, 12 mg of cyclic peptide was isolated.

RS A104 c H-Ty r-c [D-Pen-Gly-T rp-Cys]- Asp-Phe -NH2

This was synthesized using on-resin cyclization with Tl(tfa)3, as described in the

general methods above. An example, synthesis started from 0.5 grams of Rink Amide

resin. Linear peptide was constructed by standard coupling of Fmoc/Bu protected amino

acids. Fmoc D-Pen and Fmoc-Cys were side chain protected with S-Acm. The final N-

terminal Fmoc group was not deprotected. The dried peptidyl resin weighed 700 mg.

The resin was swelled with cold solution of DMF and anisole (19:1, v/v) to which

thallium trifluoroacetate was added (4 eq., 0.315 grams). After a 15 hour reaction, the

resins were washed and the N-terminal Fmoc group was deprotected. After cleavage

from the resin and deprotection of the side chain groups, the peptide was precipitated and 131 dried. The crude peptide yield was 105 mg. After RP-HPLC, the collected fraction was

frozen and lyophilized. 5.2 mg of cyclic peptide was isolated.

RSA121cH-Tyr-c[D-Cys-Gly-D-Trp-Cys]-Asp-Phe-NH2

This was prepared by DMSO-assisted air oxidation, as described in the general methods. As an example, the synthesis started from 0.5 grams of Rink Amide AM resin, and sequential coupling of the Fmoc/Bu proctected amino acids was done. Fmoc-Cys and Fmoc-D-Cys were side chain protected with iS-Trt. After cleavage, the isolated crude peptide was 345 mg. 1.06 grams of the crude linear peptide was dissolved in 100 mL of

0.1 M ammonium bicarbonate. 1 mL of DMSO was added. After two days of reaction,

the cyclic peptide was frozen and lyophilized. After RP-HPLC and lyophilization, 4.4

mg of cyclic peptide was isolated.

RSA123C H-Tyr-c[D-Cys-Gly-D-Trp-Peii]-Asp-Phe-NH2

This was synthesized using on-resin cyclization with Tl(tfa)3, as described in the

general methods above. An example, synthesis started from 0.6 grams of Rink Amide

resin. Linear peptide was constructed by standard coupling of Fmoc/^Bu protected amino

acids. N"-Fmoc-D-Cys and N"-Fmoc-Pen were side chain protected with S-Acm. The

final N-terminal Fmoc group was not deprotected. The dried peptidyl resin weighed 981

mg. The resin was swelled with cold solution of DMF and anisole (19:1, v/v) to which 132 thallium trifluoroacetate was added (~4 eq., 0.900 grams). After a 15 hour reaction, the resins were washed and the N-terminal Fmoc group was deprotected. After cleavage

from the resin and deproctection of the side chain groups, the peptide was precipitated and dried. The crude peptide yield was 300 mg. After RP-HPLC, the collected fraction was frozen and lyophilized. 14 mg of cyclic peptide was isolated. 133 CHAPTER 5

DESIGN, SYNTHESIS, AND STRUCTURE-ACTIVITY RELATIONSHIPS OF

CCK/OPIOID PEPTIDES: LACTAM ANALOGUES

5.1. Results: Design of lactam analogues

In the Chapter 4, cyclic disulfide analogues of CCK/opioid peptides showed

interesting biological activities at both CCK and opioid receptors. With the same

reasoning for the design of cyclic disulfide analogues, lactam analogues of CCK/opioid

peptides were designed. Cyclization with lactam bridges provides for a global

conformational constraint, but unlike the cyclic disulfide analogues, lactam analogues

allow for an exploration of ring size.

In the design of the lactam analogues, positions 2 and 5 were used as sites of

substitution for side chain-to-side chain cyclization, with the same rationalization applied

for the cyclic disulfide. These positions were consistent with the side chain-to-end

lactam analogues of [Leu^]-enkephalin for opioid receptors, as well as the side chain-

to-side chain cyclic disulfides of enkephalin (e.g. DPDPE). Further, the same

corresponding substitution sites were used in cyclic lactam analogues for CCK. A

summary of the design of the lactam is depicted in Figure 5.1. Positions 2 and 5 were

substituted with analogues of Lys and Glu, and the side chain groups were coupled to

form a lactam bridge. The initial cyclic analogues [Lys^,Glu^] 35 and [D-Lys^,Glu^] 36

served as the initial structures with an 18-membered ring. c[D-Lys^,D-Glu^] 37 explored 134 a diasteroisomer on the peptide backbone similar to DPDPE. Moreover, the effect of directionality of the lactam bridge was determined with c[D-Glu^,Lys^] 38. In the cyclic analogue [D-Lys^jD-Trp'^jGlu^] 39, D-Trp was introduced in an attempt to obtain an antagonist activity at the CCK receptors, as was seen in the linear analogues. The optimal ring size favorable to both CCK and opioid receptors was explored with smaller cyclic analogues [D-Om^,Glu^] 40, [D-Om^,Asp^] 41, [D-Dap^,Glu^] 42, and [D-

Dap^,Asp^] 43 which have 17-, 16-, 15-, and 14-membered rings, respectively. 135

CCK Pharmacophore

Asp-Tyr'-D-Phe^-Gly^-Tip^-NMeNIe^-Asp^-Phe'-NHz (SNF-9007) ^

Opioid Pharmacophore 1. Increase potency at opioid receptors. 2. Improve balanced affinity between CCK-A and CCK-B receptors. 3. Improve CCK receptor antagonist V properties.

Linear Analogues: Tyr-D-Nle-Gly-Yyy-Nle-Asp-Phe-NH2

4. Global conformational constraint Lactam bridge cyclization V

Cyclic Analogues: Tyr-Xxx-Gly-Yyy-Zzz-Asp-Phe-NHa

Xxx = Lys, D-Lys, D-Glu, D-Om, D-Dap

Zzz = Glu, D-Glu, Asp, Lys

Yyy = Trp, D-Trp

Figure 5.1. Design of lactam analogues. Positions 2 and 5 were used as sites of cyclization. Position 2 was substituted with Lys, D-Lys, D-Glu, D-Om, or D-Dap.

Position 5 was substituted with Glu, D-Glu, Asp, or Lys. Position 4 was substituted with

Trp and D-Trp. 136 Table 5.1. Lactam analogues.

No. Code Sequence Ring size 35 RSA405 Tyr-c[Lys-Gly-Trp-Glu]- Asp-Phe -NH2 18

36 RSA402 Tyr-c[D-Lys-Gly-Trp-Glu]-Asp-Phe-NH2 18

37 RSA406 Tyr-c[D-Lys-Gly-Trp-D-Glu]Asp-Phe-NH2 18

38 RSA407 Tyr-c[D-Glu -Gly-Trp-Lys] - Asp-Phe -NH2 18

39 RSA422 Tyr-c[D-Lys-Gly-D-Trp-Glu]-Asp-Phe-NH2 18

40 RSA404 Tyr-c[D-Orn-Gly-Trp-Glu]-Asp-Phe-NH2 17

41 RSA408 T yr-c[D-Orii-Gly-Trp-Asp]- Asp-Phe -NH2 16

42 RSA409 Tyr-c[D-Dap-Gly-Trp-Glu]-Asp-Phe-NH2 15

43 RSA410 Tyr-c[D-Dap-Gly-Trp-Asp]-Asp-Phe-NH2 14

Om, ornithine; Dap, 2,3-diaminopropionic acid. 137 5.2. Results: Synthesis of lactam analogues

The synthesis of the cychc analogues with lactam bridges was performed using allyl side chain protecting groups. The allyloxycarbonyl (Alloc) group, introduced by

Stevens and Watanabe, ^^"is used for protection of amines and alcohols. The allyl ester group was used to protect the glutamic acid moiety used for cyclic lactam formation.

Due to the mild conditions used for deprotection of Alloc and Allyl groups and their stability under the conditions used for deprotection of N"-Fmoc and N"-Boc groups,

Alloc and Allyl groups are very useful alternatives for orthogonal protection in peptide synthesis. Because of their properties that Alloc and Allyl groups can be deprotected under mild and neutral conditions, they can be used to prepare partially deprotected

CCK/opioid analogues that subsequently can be cyclized on the solid support to produce conformationally constrained analogues. Fully protected CCK/opioid analogues were synthesized with a standard N"-Fmoc/'Bu strategy, in which the amino or carboxylate side chain groups were protected by Alloc and Allyl groups, respectively, at position 2 or

5. While the Alloc and Allyl groups were deprotected, the resulting free amine and carboxylic acid of Lys and Glu were coupled to form a lactam bridge to produce cyclic peptide analogues. Figure 5.2 illustrates the solid phase lactam bridge formation of

CCK/opioid lactam analogues.

The initial synthesis of these cyclic analogues met with some difficulties, and some technical issues were addressed. A standard laboratory approach developed in the

Hruby laboratory was followed. In this protocol an inert environment was required 138 for the success of the Alloc/Allyl deprotection with Pd°. The entire synthesis of the linear peptide was done in the presence of argon. The standard reaction vessel was equipped with a mineral trap to exclude ambient air and the mixing of the reaction with a gentle flow of argon gas. These extra precautions ensured the absence of oxygen in the solvent and also in the matrix of the solid support. In the protocol a 24 eq. excess of phenylsilane was added to the reaction vessel and it was incubated with resins 5 min before adding the catalyst to ensure homogeneity in the solvent and resin matrix.

In the same protocol described by Grieco et al., '^^once the Alloc/Allyl groups had been removed concomitantly with Pd(PPh3)4/PhSiH3 in DCM, the side chain groups were coupled to form the lactam bridge using HBTU/HOBt as coupling agents. In the initial syntheses of these CCK/opioid lactam analogues, the c[D-Lys^,Asp^] analogues resulted in the linear peptides with tetramethylguanidinium (TMG), as indicated by 98 mass units greater than the expected m/z of the cyclic peptide. Similar results were observed by Story et al. ^^^when the synthesis of a 10-membered lactam ring on a solid support was attempted using HBTU. Presumably the TMG-containing peptide resulted

'y from the transfer of the TMG moiety from HBTU to the amino side chain of D-Lys . The formation of these TMG side-products during cyclization hmits HBTU's usefulness for the formation of lactam bridge. To avoid this side reaction, Arttamangkul et al. used

PyBOP/HOBt, even though the cyclization took 4-5 days. 0 O O HN-CH-C- NH-CH C-NHCH-C-N-CH C-N-CH-C-N CH-C-NH-Ch+C-NH- II I O H

Alloc/Allyl deprotection Pd(PPh3)4/PhSiH3/DCM

V OH NH2 (CH2)4 O o (CH2)2 o II II II HN-CHC-NH-CHCNHCH-C-N-CHC-N-CH-C-N-CH-C-NH-CHC-NH II I H \ o H o CH2

Cyclization HBTU/HOBt/DIEA/DMF

HN < O (CH2)4 O O (CH2)2 HN-CHC-NH-CH C-NHCH-C-N -CHC-N-CH-C-N NH-Ch+C -NH I II I u oh"

Figure 5.2. A scheme for synthesis of a conformationally constrained CCKyopioid analogue with Alloc protected Lys^ and Allyl protected Glu^. 140

N-Boc

Fmoc

t-BuO 0' OH

©PFe 1. HBTU, HOBt, DIEA (H3C)2N^N(CH3)2

2. 20%piperidine -N .N 3. TFA N±)

CH3 CHg HBTU

N

O' OH

Figure 5.3. Side reaction during cyclization with HBTU. The TMG moiety of HBTU capped the epsilon amine of Lys. 141 However, in the formation of a larger lactam ring for c[D-Lys2 ,Glu 5 ] and c[D-Lys2 ,D-

Trp'',Glu^] CCK/opioid analogues, the use of HBTU was successful. Because several lactam analogues were to be synthesized, other coupling agents aside from HBTU were investigated. Table 5.2 summarizes the attempted synthesis of lactam analogues using various coupling agents, reaction times, and solvents. The extent of reaction was qualitatively monitored by a Kaiser test. Smaller lactam ring sizes (17-14) were afforded using PyBop/HOBt or PyBop/HOAt, often after 72 hours of reaction.

Interestingly, c[D-Om^, Glu^] formed a ring very fast within ten minutes. Aside from

DMF, solvent mixtures were used such as 20% DMSO/DMF and DMF/DCM/NMP

(1:1:1). ^^^'^^^These solvent mixtures were suggested to aid lactam formation in solid support by destabilizing the unfavorable aggregation of the linear peptide. Based on these experiments, it was observed that lactam formation was more facile when position 5 is Glu, but not when it is Asp. 142 Table 5.2. Various conditions used for the lactam bridge formation.

Sequence Coupling reaction

Tyr-c[D-Lys-Gly-Trp-Asp] -Asp-Phe-NH2 (17) Failed with HBTU/HOBt

Tyr-c[D-Lys-Gly-D-Trp-Asp]-Asp-Phe-NH2 (17) Failed with HBTU/HOBt

Tyr-c[D-Lys-Gly-Trp-Glu]-Asp-Phe-NH2 (18) HBTU/HOBt, 2 hours

Tyr-c[D-Lys-Gly-D-Trp-Glu]-Asp-Phe-NH2 (18) HBTU/HOBt, 2 hours

Tyr-c[D-Asp-Gly-Trp-Lys] -Asp-Phe-NH2 (17) HBTU/HOBt, 2 hours**

Tyr-c[D-0m-Gly-Trp-Glu]-Asp-Phe-NH2 (17) PyBop/HOBt, 30 min

Tyr-c[Lys-Gly-Trp-Glu]-Asp-Phe-NH2 (18) HBTU/HOBt, 2hours

Tyr-c[D-Lys-Gly-Trp-D-Glu]-Asp-Phe-NH2 (18) PyBop/HOBt, 72 hours

Tyr-c[D-Glu-Gly-Trp-Lys]-Asp-Phe-NH2 (18) PyBop/HOBt, 72 hours

Tyr-c[D-0m-Gly-Trp-Asp]-Asp-Phe-NH2 (16) l)PyBop/HOBt, 72 hours

DMF:DCM:NMP (1:1:1)

2)PyBop/HOAt, 72 hours

20% DMSO/DMF

Tyr-c[D-Dap-Gly-Trp-Glu]-Asp-Phe-NH2 (15) PyBop/HOAt, 10 min

DMF:DCM:NMP (1:1:1)

Tyr-c[D-Dap-Gly-Trp-Asp]-Asp-Phe -NH2 (14) l)PyBop/HOAt, 72 hours

20% DMSO/DMF

2)DMF:DCM:NMP (1:1:1)

( ), ring size. 143

5.3 Results: Structure-activity relationships

The binding affinities of the CCK/opioid analogues were determined by competition assays against radiolabelled DPDPE and DAMGO using stably transfected hDOR and rMOR, and radiolabelled sulfated CCK-8 for hCCK-A and hCCK-B receptors. The functional activity at the transfected receptors were assessed using• [ 35S]-

GTP-y-S binding assays and PI hydrolysis assay for opioid receptors and CCK receptors, respectively. These assays were performed by Dr. Shou-Wu Ma and Hamid Badghisi from the laboratory of Dr. Josephine Lai at the Department of Pharmacology in the

University of Arizona. The experimentals are detailed in the Appendix.

Opioid agonist activity at delta and mu receptors were determined in vitro using electrically stimulated MVD and GPI, respectively. CCK antagonist was determined in vitro using unstimulated GPI/LLMP against sulfated CCK-8. These assays were performed by Peg Davis from the laboratory of Dr. Frank Porreca at the Department of

Pharmacology in the University of Arizona. The experimental details are found in the

Appendix.

The data for the competitive bmdmg assay and the corresponding functional [ S]-

GTP-y-S assay at delta and mu opioid receptors are summarized in Table 5.3 and Table

5.4, respectively. The data for the competitive binding assay and the corresponding fiinctional PI hydrolysis assays at CCK receptors are summarized in Table 5.5 and Table

5.6, respectively. The functional opioid agonist assay in the MVD and GPI tissues and 144 CCK agonist and antagonist activities in the GPI/LMMP tissues are summarized in

Tables 5.7 and Table 5.8, respectively. 145 Table 5.3. Binding affinities of CCK/opioid peptides at the opioid receptors.

Binding Affinity^ No. Sequence (Ki, nM) hDOR rMOR Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 (RSA504) 2.9 27.1

35 Tyr-c[Lys-Gly-Trp-Glu]-Asp-Phe-NH2 100 7200

36 T yr-c[D-Lys-Gly-Trp-Glu]- Asp-Phe-NHa 0.9 16.

37 Tyr-c[D-Lys-Gly-Trp-D-Glu]Asp-Phe-NH2 14. 25.

38 Tyr-c[D-Glu -Gly-Trp -Lys]- Asp-Phe-NH2 2.1 11.

39 T yr-c[D-Lys-Gly-D-Trp-Glu] - Asp-Phe -NH2 3.0 8.3

40 Tyr-c[D-Orii-Gly-Trp-Glu]-Asp-Phe-NH2 5.9 7.7

41 Tyr-c[D-Orn-Gly-Trp-Asp]-Asp-Phe-NH2 6.8 3.0

42 T yr-c[D-Dap-Gly-T rp-Glu]-Asp-Phe -NH2 n/d n/d

43 Tyr-c[D-Dap-Gly-Trp-Asp]-Asp-Phe-NH2 n/d n/d

^Competitive assay against radiolabelled [^H] DPDPE at hDOR and [^H]DAMGO at rMOR. Opioid receptors were expressed from CHO cell lines. Bold, residues of interest; n/d, not determined. 146 Table 5.4. Functional activities of CCK/opioid peptides at the opioid receptors.

Functional Activity (Agonist/ (EC50, nM) Compound hDOR Emax rMOR Emax Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 (RSA504) 4.5 82. 0.47 107

35 Tyr-c[Lys-Gly-Trp-Glu]-Asp-Phe-NH2 6200 160 30. 80

36 Tyr-c[D-Lys-Gly-Trp-Glu]-Asp-Phe-NH2 31.2 73 32 46

37 T3T -c[D-Lys-Gly-Trp -D-Glu]Asp-Phe -NH2 170 44 18 94

38 Tyr-c[D-Glu -Gly-Trp -Lys]- Asp-Phe -NH2 3.7 31 19 100

39 T yr-c[D-Lys-Gly-D-Trp-Glu]- Asp-Phe -NH2 42 120 48 99

40 Tyr-c[D-Orn -Gly-Trp -Glu]- Asp-Phe -NH2 3.7 39 13 71

41 Tyr-c[D-Orn-Gly-Trp-Asp]-Asp-Phe-NH2 0.3 110 0.81 18

42 Tyr-c[D-Dap -Gly-T rp-Glu]- Asp-Phe-NH2 n/d n/d n/d n/d

43 Tyr-c[D-Dap -Gly-T rp-Asp]- Asp-Phe -NH2 n/d n/d n/d n/d

®[^^S]GTP-y-S binding assay. Emax = (net total bound^asal binding) X 100%. Opioid receptors were expressed from CHO cell lines. Bold, residues of interest; n/d, not determined. 147 Table 5.5. Binding affinities of CCK/opioid peptides at the CCK receptors.

Binding Affinity® No. Sequence (Kj, nM) CCK-A CCK-B Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2(RSA504) 11.2 15.8

35 Tyr-c[Lys-Gly-Trp-Glu]- Asp-Phe-NHa 3300 >10000

36 T yr-c[D-Ly s-Gly-T rp-Glu]- Asp-Phe-NH2 nc nc

37 Tyr-c[D-Lys-Gly-Trp-D-Glu]Asp-Phe-NH2 nc nc

38 Tyr-c[D-Glu-Gly-Trp-Lys]-Asp-Phe-NH2 nc nc

39 Tyr-c[D-Lys-Giy-D-Trp-Glu]-Asp-Phe-NH2 nc nc

40 Tyr-c[D-Orn-Gly-Trp-Glu]-Asp-Phe-NH2 nc nc

41 Tyr-c[D-Orii-Gly-Trp-Asp]-Asp-Phe-NH2 n/d n/d

42 T yr-c[D-Dap-Gly-T rp-Glu]-Asp-Phe-NH2 n/d n/d

43 T yr-c[D-Dap-Gly-Trp-Asp]-Asp-Phe-NH2 n/d n/d

''Competition against ['^^I]CCK-8 (sulfated) at hCCK-A eind hCCK-B receptors expressed in HEK cell lines. Bold, residues of interest; nc, no competition. 148 Table 5.6. Functional activities of CCK/opioid peptides at the CCK receptors

Functional Analysis^ No. Sequence (ECso, nM) CCK-A Emax CCK-B Emax Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 (RSA504) 250 15 1900 10

35 Tyr-c[Lys-Gly-Trp-Glu]-Asp-Phe-NH2 nr ~ nr ~

36 Tyr-c[D-Lys -Gly-Trp -Glu]- Asp-Phe -NH2 nr — nr —

37 Tyr-c[D-Lys-Gly-D-Trp-Glu]-Asp-Phe-NH2 nr ~ nr ~

38 Tyr-c[D-Glu-Gly-Trp-Lys]-Asp-Phe-NH2 9300 2.7 670 19

39 Tyr-c[D-Lys-Gly-Trp-D-Glu]Asp-Phe-NH2 nr — nr ~

40 Tyr-c[D-Oni-Gly-Trp-Glu]-Asp-Phe-NH2 nr ~ nr ~

41 Tyr-c[D-Orii-Gly-Trp-Asp]-Asp-Phe-NH2 n/d n/d n/d n/d

42 Tyr-c[D-Dap-Gly-Trp-Glu]-Asp-Phe-NH2 n/d n/d n/d n/d

43 Tyr-c[D-Dap-Gly-Trp-Asp]-Asp-Phe-NH2 n/d n/d n/d n/d

"Phosphoinositide (PI) hydrolysis assay in hCCK-A and hCCK-B, expressed receptors in HEK cell lines. Emax = (total PI hydrolysis/basal PI hydrolysis). Bold, residues of interest; n/d, not determined; nr, no response. 149 Table 5.7. Functional activities of CCK/opioid peptides at the MVD and GPI tissue assays.

Opioid Agonist®, No. Sequence (IC50, nM) MVD(delta) GPI (mu) Tyr-Nle-Gly-Trp- Nle-Asp-Phe-NHj (RSA501) 16.7% at 1 jiiM 0% at 1 jiM

Tyr-DNle-Gly-Trp-Nle-Asp-Phe-NH2(RSA504) 23 ± 9.6 210±51

35 T yr-c[Lys -Gly-Trp-GIu]- Asp-Phe -NH2 5.3% at 1|J.M 11% at 10 nM

36 Tyr-c[D-Ly s -Gly-T rp-Glu]- Asp-Phe -NH2 43.0± 16 14.0 ±3.11

37 Tyr-c[D-Lys -Gly-T rp-D-Glu]Asp-Phe-NHa 310±72 110±14

38 Tyr-c[D-Glu-Gly-Trp-Lys]-Asp-Phe-NH2 110± 17 108. ±20

39 Tyr-c[D-Lys-Gly-D-Trp-Glu]-Asp-Phe-NH2 39. ±5.7 16. ±4.12

40 T yr-c[D-Orn -Gly-Trp-Glu]- Asp-Phe -NH2 n/d n/d

41 Tyr-c[D-Orn-Gly-Trp-Asp]-Asp-Phe-NH2 n/d n/d

42 Tyr-c[D-Dap-Gly-Trp-Glu]-Asp-Phe-NH2 n/d n/d

43 T yr-c[D-Dap-Gly-Trp-Asp]- Asp-Phe -NH2 n/d n/d

"Concentration at 50% inhibition of muscle contraction at electrically stimulated isolated tissues. Bold, residues of interest; n/d, not determined. 150

Table 5.8. Functional activity of CCK/opioid peptides at the GPI7LMMP.

unstimulated GPI/LMMP No. Sequence CCK- CCK- Agonist® Anagonist'' (A50) (K,, nM) Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2(RSA504) 0% at 1 190 ±80

35 Tyr-c[Lys-Gly-Trp-Glu]-Asp-Phe-NH2 0% at 1 |j,M 0% at 1 |iM

36 Tyr-c[D-Lys-Gly-Trp-Glu]-Asp-Phe-NH2 0% at 1 |aM 25 ±9.8

37 Tyr-c[D-Lys-Gly-Trp-D-Glu]Asp-Phe-NH2 0% at 1 )aM 30±11

38 Tyr-c[D-Glu-Gly-Trp-Lys]-Asp-Phe-NH2 0% at 1 jxM 6.4 ±0.57

39 T yr-c[D-Lys-Gly-D-Trp-Glu] - Asp-Phe -NH2 0% at 1 )aM 1.2 ±0.22

40 Tyr-c[D-Orn-Gly-Trp-Glu]-Asp-Phe-NH2 n/d n/d

41 Tyr-c[D-Orn -Gly-T rp-Asp]- Asp-Phe-NH2 n/d n/d

42 T yr-c[D-Dap-Gly-Trp-Glu]- Asp-Phe -NH2 n/d n/d

43 Tyr-c[D-Dap-Gly-Trp-Asp]-Asp -Phe-NH2 n/d n/d

^Contraction of isolated tissue relative to initial muscle contraction with KCl. Inhibitory activity against the CCK-8 induced muscle contraction. Kg, concentration of antagonist needed to inhibit CCK-8 to half its activity. '^No responseBold, residues of interest; n/d, not determined. 151

The lactam analogue c[Lys^,Glu^] 35 had significant weak binding affinities at both hDOR (Ki = 100 nM) and rMOR (Ki = 7200 nM) when compared to the linear

RSA504 (Table 5.3). This is not surprising since it is known fi^om the structure activity relationships of opioid ligands that D-amino acids at position 2 are important for good potency. As expected, the functional activity at the hDOR was poor (EC50 = 6200 nM).

But at the rMOR, the functional activity was surprisingly higher (EC50 = 30 nM). In the tissue assays, 35 had low opioid agonist activities at MVD and GPI (Table 5.7).

However, 35 resulted in low binding affinities at the cloned CCK-A and CCK-B receptors at the micromolar ranges (Ki = 3300 and >10000 nM, respectively). In the functional assays at the CCK receptors, 35 did not show any responses. In the unstimulated GPI/LMMP assays, 35 did not show any CCK agonist nor antagonist properties (Table 5.8).

When position 2 was substituted with D-Lys resulting in c[D-Lys^,Glu^] 36, the binding affinities were very good at the delta (Ki = 0.9 nM) and mu (Ki = 16 nM) opioid receptors, which was a significant improvement when compared to 36 (Table 5.3). The functional activities of 36 were good at both hDOR and rMOR opioid receptors (EC50 =

31 and 32 nM, respectively) (Table 5.4). Interestingly, while the binding affinity showed a ca. 15-fold selectivity for the delta opioid receptor, the functional activity was nearly the same. In the MVD and GPI assays, 36 displayed a moderate opioid agonist activity at the delta (IC50 = 43 nM) and mu (IC50 = 14 nM) opioid receptors, with a slight preference for the mu opioid receptors (Table 5.7). As for the binding affinity and functional activity at the cloned CCK-A and CCK-B receptor assays, 36 did not show any binding 152 and activity (Table 5.5 and 5.6). On the other hand, in the CCK antagonist assay at the unstimulated GPI/LMMP, 36 showed good CCK antagonist activity (Ke = 25 nM) (Table

5.8).

When D-Glu was introduced at position 5, c[D-Lys^,D-Glu^] 37 showed binding affinities at the hDOR (Kj = 14 nM ) and rMOR (Ki = 25 nM), which were ca. 16- and 2- fold, respectively, less than for c[D-Lys^,Glu^] 36 (Table 5.3). Surprisingly, the functional activity at the opioid receptors was more selective at the rMOR (EC50 =18. nM) than at the hDOR (EC50 = 170 nM) (Table 5.4). When compared to 36, the agonist activity at the delta opioid receptor decreased ca. five-fold, while it increased two-fold at the mu opioid receptor. The opioid agonist properties and CCK antagonist properties of c[D-Lys^,D-Glu^] 37 in the tissue assays had not been determined. Much like 36, 37 was neither competitive with labeled CCK nor did it have any response in the CCK-A and

CCK-B agonist assays (Tables 5.5 and 5.6).

The directionality of the cyclic lactam bridge was examined by reversing the lactam bridge from c[D-Lys^,Glu^] 36 to c[D-Glu^,Lys^] 38. When compared to 36, the binding affinity of c[D-Glu^, Lys^] 38 at hDOR (Ki =2.1 nM) slightly decreased by ca. two fold (Table 5.3). At the rMOR, the binding affinity increased slightly (K = 11 nM), thus showing a five fold selectivity for the delta opioid receptor. In the [^^SJGTP-y-S binding assays, the opioid agonist property of c[D-Glu^,Lys^] 38 increased when compared to c[D-Lys^, Glu^] 36 (Table 5.4). At the rMOR, the functional activity was good (EC50 =18 nM) with a full response while at the hDOR, the agonist potency was very good (EC50 = 3.7 nM), albeit with a limited response (Emax = 31%). On the other 153 hand, the opioid agonist properties of c[D-Glu , Lys ] 38 decreased at the tissue assays when compared to 36. At the MVD, the opioid agonist activity was moderate (IC50 =110 nM) with ca. 3-fold loss in activity compared to 36 (Table 5.7). At the GPI, the opioid agonist activity was also moderate (IC50 = 110 nM) with ca. 8 fold loss in activity compared to 36. c[D-Glu^,Lys^] 38 showed a balanced activity between the delta and mu opioid receptors in these tissue assays. In the unstimulated GPI/LMMP assays (Table

5.8), c[D-Glu^,Lys^] 38 showed a ca. four-fold increase in CCK antagonist activity (Ke =

6.4 nM) when compared to 36. However, the binding affinities of c[D-Glu 2 ,Lys5 ] 38 at the cloned CCK-A and CCK-B receptors showed no competition (Table 5.5).

Interestingly, c[D-Glu,Lys] 38 displayed poor agonist properties at CCK-A and CCK-B receptors (K = 9,300 and 670 nM, respectively) (Table 5.6).

Introduction of D-Trp at position 4 resulted in cyclic lactam analogue c[D-Lys ,

D-Trp'^,Glu^] 39, which was designed to increase the antagonist properties at the CCK receptors, showed interesting biological activities. When compared to the L-Trp c[D-

Lys^,Trp'^,Glu^] 38, c[D-Lys^,D-Trp'^,Glu^] 39 showed a 21-fold increase in CCK antagonist activity (K^ = 1.17 nM) by inhibiting CCK induced contraction at the GPI

(Table 5.8). However, in the hCCK-A and hCCK-B receptors, 39 did not show any competitive binding against the radioligand [ I]-CCK-8 and it did not display any response in the CCK agonist PI hydrolysis assay (Tables 5.5 and 5.6). On the other hand, c[D-Lys^,D-Trp'^,Glu^] 39 had good binding affinities at the hDOR and rMOR (K = 3.0 and 8.3 nM, respectively) which represents a 3 fold loss of binding at the hDOR and a two fold increase at the rMOR when compared to 36 (Table 5.3). Interestingly, c[D- 154 Lys^,D-Trp'^,Glu^] 39 was nearly equipotent at the hDOR and rMOR (EC50 = 42 and 48 nM, respectively) (Table 5.4). However, in the tissue assays, c[D-Lys^,D-Trp'^,Glu^] 39 showed a ca. two fold preference for the mu opioid receptor at the GPI (IC50 = 16. nM) against the MVD (IC50 = 39. nM) (Table 5.7).

The optimal ring size favorable for binding to both CCK and opioid receptors was explored with smaller cyclic analogues c[D-Om^, Glu^] 40, c[D-Om^, Asp^] 41, c[D-

Dap^, Glu^] 42, and c[D-Dap^, Asp^] 43 which has 17-, 16-, 15-, and 14-membered rings, respectively. The bioassays for these compounds were not complete, and only c[D-Om^, c 9 S Glu ] 40 and c[D-Om , Asp ] 41 were tested at the cloned receptors. The 17-membered ring 40 displayed good binding affinities, albeit with low selectivity, at hDOR and rMOR (Kj = 5.9 nM and 7.7 nM), which represents a ca. six-fold and two-fold loss binding affinity at hDOR and rMOR when compared to the 18-membered ring c[D-Lys^,

Glu^] 36 (Table 5.3). The functional activity increased 10 fold (EC50 = 3.7 nM) at the hDOR and ca. 3-fold (EC50 =13 nM) at the rMOR (Table 5.4). This cychc analogue did not show any binding affinity at hCCK-A and hCCK-B receptors, nor did have any agonist response in the hCCK PI hydrolysis assay (Tables 5.5 and 5.6).

5.4 Discussions

The linear analogue of RSA504 (Tyr-D-Nle^-Gly-Trp-Nle^-Asp-Phe-NH2), which interacts at CCK and opioid receptors, was conformationally constrained via a side chain- to-side chain lactam bridge between position 2 and position 5, representing [i,i +3] cyclic lactam. Several examples of peptide antagonists had been successfully developed which 155 were lactams of corticotropic-releasing factors (CRF) melantropins. Selective antagonists with unusual cyclic lactam bridges also had been developed for CCK receptors. ^^^"^^^The design of lactam peptide antagonists was applied to the CCK/opioid receptors, where antagonist properties are desired at the CCK receptors and agonist properties at the opioid receptors. Compounds 35-39 explored the various stereoisomers and regioisomers of 18-membered cyclic lactam analogues. Cyclic analogues 40-43 examined the effect of smaller ring sizes.

5.4.1. Assessment of the bioassays

The bioassays involving the CCK receptors presented interesting and sometimes contraditory results. While the lactam analogues, except for 35, effectively inhibited the contractions induced by CCK-8 at the unstimulated GPI/LMMP, they did not bind competitively against the radioligand [^^^I]-CCK-8, nor did they demonstrate any significant CCK activity in the PI assay at the hCCK-A and hCCK-B receptors in expressed HEK cells. This may suggest that inhibition of the CCK-8 at the GPI tissue by the cyclic lactam analogues may be due to an allosteric effect. In general these results are in contrast to binding affinity and functional activities reported cyclic analogues of CCK reported in literature. ^^''^^For example, (S)-RB 360 bound well to CCK-B receptors in the guinea pig brain and CCK-A receptors in the guinea pig pancreatic membranes and showed potent agonist activity at the rat CCK-B receptors expressed in the CHO cells.

This may suggest interspecies differences between human CCK receptors and CCK receptors in rats. 156

5.4.2. Role of position 2 in CCK-antagonist properties

Even though the cychc analogues did not interact with human CCK receptors expressed in HEK cells, an interesting structure-activity relationship of cyclic lactams can 9 S still be discussed at CCK in relation to the CCK receptors in the GPL First, c[Lys , Glu ]

35 showed no antagonist activity (against CCK-8 at 1 |_iM) GPI/LMMP. Interestingly, when position 2 was substituted with D-Lys or D-Glu, compounds 36-39, these analogues were able to inhibit the contractions induced by CCK-8, even when position 4 is D-Trp for 37. The various stereoisomers and regioisomers of 35, 36, 38 and 39 generally retained good antagonist activity at the CCK receptors, while maintaining the agonist properties at the opioid receptors in the tissue assays.

This is surprising, since a 24-membered ring analogue (JMV-320, Figure 5.4) of

CCK, in which the corresponding positions were replaced by L-Lys residues and whose side chains are bridged by a succinic moiety, possessed selective binding for the CCK-B receptors at guinea pig brain preparations.

Ac-T yr-Lys-Gly-Trp-Lys-Asp-Phe-NH2

Figure 5.4. Structure of JMV-320. 157

This may suggest that certain stereisomers are required for antagonist activity at the CCK receptors of the GPI, depending on the ring size. Because JMV-320 has a larger, flexible ring, the L-Lys at position 2 afforded a conformation favorable for antagonist activity.

This illustrates that small modifications affecting the conformation of the peptide can

have a large effect on the biological activity of the peptides.

5.4.3 Comparison to opioid ligands

From the point of view of opioid receptors, these cyclic lactam analogues

represents a new class of lactam enkephalin analogues with the substitution of Trp at

position 4 and the new "address" sequence -Asp-Phe-NH2 at the C-terminal. Generally,

these lactam analogues showed slight preference for the mu opioid receptors based on the

MVD/GPI agonist assays. This is in contrast to the linear and cyclic disulfide analogues

with D-amino acid residue at position 2, which were generally selective for the delta

opioid receptors in the same assays. Much like the side-chain-to-tail cyclic lactam

[Leu^]-enkephalin analogues from Schiller's lab, the CCK/opioid cyclic analogues have a

preference for the mu receptors at the GPI. ^"^'^^^''^^This observation suggests that the

lactam moiety on the ring may play a significant role in the selectivity for the mu opioid

receptors. 158

5.4.4 Differential opioid agonist and CCK antagonist activities

When D-Trp was substituted at position 4 in 39, the opioid agonist activity and the selectivity at the MVD and GPI remained the same, while the CCK antagonist activity at the GPI increased by ca. 20-fold when compared to 36. This demonstrates the conformational features at position 4 for an 18-membered ring is not crucial for the opioid receptors. However, this conformational change at position 4 is much more important for the CCK antagonist activities at the GPI. Substitution of D-Trp in 39 demonstrates a differential structure activity relationship between opioid agonist and

CCK antagonist, where there is retention of activity at the opioid receptors and a decrease in antagonist activity at the CCK receptors.

A differential structure activity relationship was also observed when the direction of the lactam bridge was modified fi^om c[D-Lys, Glu] 36 to c[D-Glu, Lys] 38. In reversing the directionality of the lactam bridge from 36 to 38, the opioid agonist activity moderately decreased at the MVD and GPI (ca. 2.5 fold and 14 fold, respectively), while

CCK antagonist activity at GPI increased 6-fold.

These subtle modifications of a cyclic lactam had effectively probed for the required specific bioactive conformations that are required for the antagonist properties at

CCK receptors and the agonist properties for the opioid receptors. 159

5.4.5 Comparison of lactam to cyclic disulfide

The lactams 36 and 37 could be compared and contrasted with the cyclic disulfide

RSAlOlc (Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2, 29) and RSA121c (Tyr-c[D-Cys-

Gly-D-Trp-Cys]-Asp-Phe-NH2,33). RSAlOlc is a potent agonist at opioid receptors at the MVD and GPI (IC50 = 0.45 and 63 nM, respectively) and a potent antagonist at the

CCK receptors at the GPI (Kg = 7.6 nM). The 14-cyclic disulfide analogue was very selective for the delta opioid receptors (ca. 120-fold mu/delta). While the 18-membered cyclic lactam analogues were good opioid agonists, the lactam analogue 36 was slightly selective for the mu receptors at GPI (ca. 3-fold delta/mu).

When position 4 of RSlOlc was substituted with D-Trp resulting in RSA121c, both opioid agonist and CCK antagonist activities significantly decreased. For the lactam

39 where position 5 was substituted with D-Trp, the opioid agonist activity was remained potent at both MVD and GPI, while increasing the CCK antagonist activity at the

GPI/LMMP. Because 39 had a larger ring size than the 14-membered cyclic disulfide

RSA121C, which displayed very poor opioid agonist activity at the MVD and GPI (IC50 =

2400 nM and 18% at 1 ^M) and poor CCK antagonist activity at the GPI, the flexibility of the ring might have afforded the orientation of the side chain groups, D-Trp in particular, to be more favorable for the biological activities at opioid and CCK receptors in these tissue assays. The smaller ring size of RSA121c might have limited the 160 conformation of the D-Trp to unfavorable orientations due to the added conformational constraint of a smaller ring.

5.4.6 Is D-Trp compatible with a non-methylated residue at position 5 in a lactam

In Chapter 2, it was shown for the linear analogues of CCK/opioid peptides that a simultaneous substitution of D-Trp at position 4 and a non-N"-methylated residue at position 5 resulted in weak activities at the opioid and CCK receptors. For example, when position 5 of RSA601 (Tyr-D-Phe-Gly-D-Trp-MeNle^-Asp-Phe-NH2, 14) was replaced with Nle^ resulting in RSA622 (Tyr-D-Phe-Gly-D-Trp-Nle^-Asp-Phe-NH2,18), the delta opioid agonist potency decreased significantly from Kj = 0.55 to 190 nM.

Similar drops in activities were observed at mu opioid receptors and CCK-A and CCK-B receptors. In contrast, the 18-membered lactam c[D-Lys,D-Trp,Glu] 37 with D-Trp at position 4, and a non-N"-methylated Glu at position 5, possessed good opioid agonist activities at the delta and mu opioid receptors at the MVD and GPI (IC50 -39 and 16 nM) and remarkable antagonist activities against CCK-8 in the GPI assays (Kg =1.2 nM).

This is in contrast to the bioactivities displayed by cyclic disulfide RSA121c (Tyr- c[DCys-Gly-DTrp-Cys]-Asp-Phe-NH2,33, IC50 = 2400 nM and 18.6% at nM). While it seemed that D-Trp"^ was not compatible with a non-methylated residue at the position 5 in linear and cyclic disulfide analogues for a good bioactivity at the opioid and CCK receptors, it was well tolerated in the cyclic lactam analogue. 161 5.5 Conclusions

Lactam analogues of CCK/opioid were designed and synthesized, exploring the structure activity relationships of different ring sizes, substitution of D-Trp, as well as directionahty of the lactam bridge. Lactam analogues resulted in potent agonist activity at the opioid receptors in the MVD and GPI and potent CCK antagonist activity at the

GPI. One of the most potent cyclic analogues was c[D-Lys^,D-Trp'',Glu^] 39, which displayed potent opioid agonist activity with a slight preference for mu receptors and very potent antagonist activity at the CCK receptors at the GPI. These cyclic lactam analogues, though, did not show binding at the human CCK receptors expressed in cell lines. This may suggest that the CCK antagonist activity at the GPI is an allosteric inhibition. 162 5.6 Experimental section

5.6.1 Abbreviations

Abbreviations used for amino acids and designation of peptide follow the rules of the lUPAC-IUB Commission of Biochemical Nomenclature in J. Biol. Chem. 1972, 247,

977-983. Additional abbreviations are used as follows: Boc, ferf-butyloxycarbonyl;

Fmoc, 9-fluorenylmethoxycarbonyl; tBu, tert-hutyl; Alloc, allyloxycarbonyl; ACN, acetonitrile, DCM, dichloromethane; DMF, A^,A^-dimethylformamide; DMSO, dimethyl sulfoxide; NMP, iV-methyl pyrrolidine; DIPEA, A'jA^-diisopropylethylamine; HBTU, 2-

(lH-benzotriazol-l-yl)-l,l,3,3,-tetramethyluronium hexafluorophosphate; HOBt, N- hydroxybenzotriazole; PyBop, benzotriazole-1-yl-oxy-tris-pyrrolidino-phosphonium hexafluorophosphate; HOAt, A^-aza-benzotriazole; TFA, trifluoroacetic acid; TIS, triisopropylsilane; Dap, 2,3-diaminopropionic acid, RP-HPLC, reversed-phase high performance liquid chromatography; TLC, thin layer chromatography; ESI-MS, electrospray ionization mass spectrometry; MALDI-TOF MS, matrix-assisted laser desorption ionization/time-of-flight mass spectrometry; HR, high resolution; AAA, amino acid analysis.

5.6.2 Materials

All peptides analogues were synthesized manually using a general protocol for peptide synthesis with N"-Fmoc/?-butyl chemistry. Allyl/Alloc side chain protection 163 groups were used for Lys and Glu and Asp. The manual synthesis was employed with a glass reaction vessel fitted with a course frit with a three way stopcock which allowed

argon gas to pass through to agitate the resins and a vacuum hne to remove excess reagents and solvents. The Rink Amide AM resin (200-400 mesh, 0.6-0.7 mmol/gram

substitution) was purchased from Novabiochem (U.S.A). N"-Fmoc-Phe-OH, N"-Fmoc-

Asp(OtBu)-OH, N^-Fmoc-TrpCN'^-Boc), N"-Fmoc-Tyr(ffiu)-OH, N"-Fmoc-D-Trp(N'"-

Boc) were from American Peptide Co., (Sunnyvale, CA); N"-Fmoc-Asp(OAllyl)-OH

and N"-Fmoc-Glu(OAllyl)-OH were purchased from Neosystems (France); N"-Fmoc-

D-Lys(N'^-Alloc)-OH, ]Sr-Fmoc-Lys(N^-Alloc)-OH were purchased from Advanced

ChemTech (Louisville, KY, U.S.A); N'^-Fmoc-D-Glu(Allyl)-OH, >f-Fmoc-D-

Om(N®-Alloc)-OH, N"-Fmoc-D-Dap(N®-Alloc)-OH were purchased from Senn

Chemicals; HBTU and HOBt were purchased from Quantum Biotechnologies (Monfreal,

Canada). PyBop was purchased from Novabiochem (U.S.A). HOAt was purchased from

Aldrich. DCM and DMF were purchased from EM Science (NJ, U.S.A.), trifluoroacetic

acid was from EM Science (NJ, U.S.A.), piperidine, triethylsilane, phenylsilane,

Pd(PPh3)4, and dimethyl sulfoxide were from Aldrich (Milwaukee, WI), diethyl ether was

from Mallinckrodt Baker (Paris, KY). Reagents and solvents were used as packaged and

were not purified further. The solvents for purification were from the following:

acetonitrile and trifluoroacetic acid were from EM Science (NJ, U.S.A.). All the solvents

were used as received without further purification unless otherwise noted. The

purification of the crude peptides was achieved by using Hewlett-Packard 1100 series

HPLC instrument (Agilent-Technologies) with Cis-bonded silica column semi- 164 preparative column (Vydac 218TP1010, 300 A, 1.0 x 25 cm, Separations Group,

Hesperia, CA). The separations were monitored at 280 nm with a Hewlett-Packard 1100 series fixed wavelength UV detector or at 220 and 280 nm with a Hewlett-Packard 1100 series multiple variable wavelength UV detector and were integrated with a Hewlett-

Packard 3396 series III integrator. Purity of the isolated peptide was assessed with analytical RP-HPLC in two different gradient systems as detected at 280 nm, 230 nm, and 254 nm. In all cases, the peptides were greater than 95% pure. The structures of the pure peptides were confirmed by AAA and ESI-MS and high resolution MALDI-MS at the University of Arizona Mass Spectrometry and Protein Sequencing Facility. AAA used an Applied Biosystems model 420A amino acid analyzer with automated hydrolysis

(vapor phase at 160°C for 110 min using 6 N HCl, and a precolumn phenylthiocarbamyl- amino acid (PTC-AA) analyzer. Phenylalanine peak was used as the standard. No corrections were made for amino acid decomposition. The purity of the peptides was checked by TLC on Analtech (Newark, NJ) silica gel GF plates (250 microns layer thickness) in at least two solvent systems.

5.6.3 General method for peptide synthesis

The fully protected resin-bound peptide with allyloxycarbonyl (Alloc) and allyl

(OAll) protection groups for Lys and Glu, respectively, was synthesized in an inert, argon purged environment. The reaction vessel with a course sintered filter fitted with a three- way stopcock for bubbling and filtration was equipped with a mineral oil trap at the top of the glass vessel to exclude ambient air. During the entire synthesis, including the N"- 165 Fmoc deprotection, washing cycles, and coupling steps, the system was maintained in an inert environment by bubbling with argon gas for mixing, which is critical for success in the catalytic allyl deprotection reaction. The fully protected linear peptide was synthesized on 0.5 g of Rink Amide AM resin using N"-Fmoc chemistry. The resin was swelled in the reaction vessel with DMF overnight. Initially, the resin was deprotected with 20% (v/v) piperidine in DMF solution (3 min + 20 min). The first amino acid, N"-

Fmoc-Phe-OH was coupled to the resin. To the growing chain, the following amino acids were coupled sequentially to the peptide chain with N®'-Fmoc-Asp(0®u)-OH, N"-

Fmoc-Asp(OAllyl)-OH, N"-Fmoc-Glu(OAllyl)-OH, N"-Fmoc-D-Glu(OAllyl)-OH, N®-

Fmoc-Trp(N'-Boc)-OH, N"-Fmoc-D-Trp(N'-Boc)-OH, N"-Fmoc-Gly-OH, N"-Fmoc-D-

Lys(N-Alloc)-OH, N"-Fmoc-Lys(N®-Alloc)-OH, N«-Fmoc-D-Glu(OAllyl)-OH, N"-

Fmoc-D-Om(N®-Alloc)-OH, N"-Fmoc-D-Dap(N®-Alloc)-OH, IST-Fmoc-Tyr(^Bu)OH using standard solid-phase methods. Each coupling reaction was achieved using 3-fold excesses (relative to resin substitution) of amino acid, HBTU, and HOBt in the presence of 6-fold excess DIEA. The coupling reactions were incubated for one hour. The completeness of the coupling reaction was monitored by a negative Kaiser test, which is indicated by a clear solution and clear resins following test. ^^^The N"-Fmoc protecting group on the amino acid was removed with piperidine (20% in DMF, 1 x 3min, 1 x 25 min). The coupling and deprotection steps were each followed by washes with DMF (3 x

1 min) and DCM (3x1 min). Completion of the Fmoc deprotection was assessed by

Kaiser test, ^^^which is indicated by a deep blue or purple solution and dark blue resins following the test. After coupling Fmoc-Tyr(ffiu)OH, the general procedure for cyclic 166 lactam cyclization was followed (described below). After the success of the cyclic lactam cyclization, the deprotection of the final N"-terminal Fmoc, the resins were washed with

DMF with a final wash of DCM. After a thorough wash with DCM, removing all the

DMF, the peptidyl resin was dried with vacuum and ambient air.

The dried peptidyl resins were transferred to bromosilicate scintillation vial. The peptide was cleaved from the resin and the other side chain protecting groups removed using a cleavage cocktail (10 ml/ grams) used varied with 95%TFA, 2.5%TIS, and 2.5% water. The resins were incubated in the cleavage cocktail for 1 hr and 30 min. The resins were filtered through a cotton plugged glass pipette. The resins were then washed with additional TFA (~2 mL) for five min. The TFA-peptide solution was transferred to a 15 mL propylethylene conical centrifuge tubes (Falcon). The volume of TFA of the filtrate was then reduced with a gentle flow of inert gas until is about 1.5 mL. The peptide was precipitated upon a slow addition of cold diethyl ether (12 mL). The precipitate was isolated by centrifugation using a bench top centrifuge (Hamilton Bell, VanGuard

V65000). The organic solvent was decanted and discarded. The pellets were then washed three times with cold diethyl ether (12 mL). The pellet was dried over air yielding 50-90% of crude peptides.

5.6.4 General method for the cyclization via lactam bridge

In a typical example of cyclic lactam formation on the resin, following assembly of the fiilly protected peptide on the solid support, the resin was washed with DCM (3x2 min) in the presence of argon, and a solution of phenylsilane (PhSiH3) (24 eq.) relative to 167 the resin loading) in DCM allowed to incubate for 5 min. Pd(PPh3)4 (0.25 equiv) was directly added to the reaction vessel for 15 min. Then the resin was washed with DCM (3

X 2 min), and the process was repeated. Complete orthogonal deprotection of Lys was monitored by the Kaiser test. ^^^After washing with DCM (3x2 min), the resin was suspended in DMF, followed by cyclization of the peptide via the free carboxylic acid chain of Glu and the free amino side chain group of Lys by addition of PyBop (6 eq.),

HOAt (6 eq.), and DIEA (12 eq.) for 2 hours. The coupling process was repeated until a negative Kaiser ninhydrin test resulted. The peptide resin was then treated the usual procedures for the final deprotection of N"-Fmoc and cleavage from the resin.

Several solvent mixtures were used in during the lactam cyclization step. These include 1:1 DMF/DCM; 1:5 DMSO/DMF; and 1:1:1 DMF/DCM/NMP.

5.6.5 Purification

The crude peptides were purified using a Cig semi-preparative column. The crude peptides were loaded into the column at a concentration of 10 mg/mL. For 100 mg dried crude peptide, the peptide was "wetted" with ACN (~1 mL). Aqueous 0.1% TFA was added slowly until peptides precipitated (~6 mL). ACN was added until the crude peptides are dissolved (~2 mL). Iterative addition of aqueous 0.1% TFA and ACN (or ethanol or methanol) was done peptide until it was fully dissolved, without exceeding

20% ACN. The solution was allowed to sit at room temperature overnight, or until peak is no longer detected by HPLC, to decarboxylate the carboxamic acid group on the Trp. 168 The dissolved peptide was filtered through a 0.45 micron cellulose acetate filter

(Aerodisc) to remove small inorganic salts. For the Ci8 semi-prepatative sized column, the maximum loading capacity of 10 mg/mL was injected. Injection volume was adjusted depending on the detector capacity and resolution of the desired peak. Generally, the gradient used was 20% to 60% ACN in aqueous 0.1% TFA in 30 min at a flow rate of 3 mL/ min. The gradient was adjusted depending on the resolution of desired peak from the impurities. After pooling the collected fractions, the ACN was removed by rotary evaporation. The aqueous solution was then pooled in a 50 mL propylethylene conical tube (Falcon) vial and frozen for lyophilization. 169 CHAPTER 6

SOLUTION STRUCTURE OF A CYCLIC DISULFIDE ANALOGUE OF

CCK/OPIOID PEPTIDES BY NMR: A COMPARATIVE STUDY WITH OPIOID

AND CHOLECYSTOKININ PEPTIDE LIGANDS

6.1 Background

In this chapter, the conformational and topological structures of a CCK/opioid peptide will be examined and determined. Structural analysis of a CCK/opioid peptide should provide valuable insight into the conformational and topographical requirements for binding and biological activities at both opioid and CCK receptors. It should also provide some suggestion on how CCK/opioid ligands have properties which are both opioid and CCK. It should also provide biophysical properties that are important for the de novo design of non-peptide• analogues of the CCK/opioid peptides.• 1

The structural properties of peptides can be determined by spectroscopic methods, x-ray crystallography, and energy calculations. Here, the CCK/opioid peptide was investigated by one- and two-dimensional nuclear magnetic resonance (NMR) spectroscopy and molecular dynamics simulation. This method should complement reported structures of enkephalin'®^'^^^ and CCK^'^'^^'^determined by NMR methods.

RSAlOlc was chosen for this study because of its interesting biological activities.

RSAlOlc is very potent as an agonist at both delta and mu opioid receptors (IC50 = 0.45 170 and 63. nM at the MVD and GPI, respectively), as well as an antagonist at CCK receptors

(Ke = 7.6 nM against CCK-8 in the GPI/LMMP assay). While RSAlOlc did not compete against radiolabeled CCK in the hCCK-A and hCCK-B receptors, RSAlOlc has showed potent anti-allodynic properties in rats in preliminary in vivo studies. Furthermore,

RSAlOlc is suitable for spectroscopic studies because of its relatively constrained structure compared to the linear analogues. RSAlOlc has a 14-membered, medium ring formed by a side chain-to-side chain cyclization via a disulfide bridge. RSAlOlc also has the advantage of having natural amino acids which do not require modification of the force fields provided in molecular dynamics software packages.

6.2 Results: NMR chemical shift assignments and coupling constants for RSAlOlc

The NMR studies of RSAlOlc were performed using DMSO-Je as a solvent, primarily because RSAlOlc is not very soluble in aqueous solvents. DMSO is a widely used solvent in NMR studies of peptides, including enkephalins and CCK, because its biophysical properties are comparable to extracellular aqueous solutions. In addition, a comparative study of NMR studies of CCK in DMSO and aqueous solvents (and other mixtures of solvents) showed that different solvents do not significantly modify the conformation of CCK.

The chemical shifts assignments for RSAlOlc were made utilizing 2D NMR experiments. The ^H-NMR parameters such as chemical shifts (6) of amide and aliphatic protons, intraresidue geminal (^J) and vicinal (^J) are summarized in Tables 6.1 and 6.2. 171 Non-equivalence of Gly HaS may provide an insight to the chemical environment surrounding those protons. Table 6.1 shows the chemical shifts of Gly Ha. It is interesting to note that there is a large difference in chemical shifts between the two nonequivalent Gly HaS, A6 = 1.0 ppm. Previous studies of DPDPE had reported a chemical shift difference between the two nonequivalent Gly HaS of DPDPE, in

H2O/D2O, to be 0.80 ppm^^^ and, in DMSO, 1.19 ppm^^^, while other analogues of

DPDPE had 0.45 - 0.84 ppm differences. As for the four stereoisomers of TMT-

DPDPE in DMSO, the chemical shift difference between the two nonequivalent Gly HaS ranged fi^om 1.05 - 1.18 ppm. For DPDPE, it was reasoned that the bulky disulfide bond and bulky geminal dimethyl groups of the Pen residues tended to limit backbone flexibility, forcing the carboxyl groups away from the 14-membered ring. As a consequence of this, the Gly Ha was fixed in the deshielding zone of the carbonyl groups, leading to a strong non-equivalence. In the case of the RSAlOlc, since the difference in chemical shift is about ca. 0.2 ppm greater than DPDPE, other factors may have contributed. It can be suggested that the aromatic side chain of Trp"^, or even Phe^, may be oriented in such a way that that a ring current effect is deshielding one of the Gly Ha, causing the large nonequivalence in chemical shift. 172

Table 6.1. 'H chemical shifts for RSAlOlc (T = 298 K, DMSO-i/e)

Chemical shifts (ppm)

NH HW HP HP' H(Ar)

Tyr^ 4.21 2.86 2.79 7.00, 6.67

D-Cys^ 8.62 4.78 3.04 2.73

Gly^ 8.75 4.31/3.24

Trp"^ 8.78 4.48 3.15 2.95 7.12 ,7.30 , 7.04, 6.9, 7.53, 9.37

Cys^ 7.51 4.47 2.94 2.89

Asp® 8.42 4.48 2.69 2.42

Phe^ 8.02 4.31 3.03 2.81 7.30,7.26, 7.19 Table 6.2. Coupling constants for RSAlOlc

Coupling constants (Hz)

JHOHP JHOHP' Jhphp'

Tyr^ 7.2 7.1 13.9

D-Cys^ 2.6 4.2 14.5

Gly^ — — ~

Trp^ 5.1 9.8 15.4

Cys^ 3.6 11.8 11.8

Asp^ 6.1 7.7 16.8

Phe^ 4.6 9.4 13.9 174 Table 6.3. Comparison of RSAlOlc chemical shift non-equivalence of Gly Ha to reported literature values for analogous peptides.

Chemical shift difference of non-equivalence of Gly Reference Hg (ppm) (DMSQ-^6) RSAlOlc:

Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe- 1.01

NH2

Mosberg, Tyr-c[D-Pen-Gly-Phe-D-Pen]-OH 1.19 1990^69

Mosberg, Tyr-c[D-Cys-Gly-Phe-Cys]-NH2 0.8 1984^^^ 175 6.3. Results: Temperature dependence study

Temperature coefficients of the NH chemical shifts may provide insight to possible conformations stabiUzed by hydrogen bonds. Large values of this parameter (>

6 ppb/K) are generally indicative of exposure to and proton exchange with the solvent, while small values (< 3 ppb/K) suggest inaccessibility to the solvent or participation in an intramolecular hydrogen bond. The A6/AT (-ppb/K) values for RSAlOl, as well as comparative data for DPDPE and CCK-8, are summarized in Table 6.4. The large temperature dependence for Gly^, Trp^ and Phe^ indicates that amide protons in these peptides are exposed to solvent. These amide protons may not have a role in stabilizing a turn. The high temperature coefficient for Gly^ is surprisingly higher than in the values for DPDPE and CCK-8. The temperature coefficient for D-Cys^ was moderate at -2.93 ppb/K. These values are within the range of DPDPE (2.6 - 5 ppb/K). ^^^For Cys^, the amide proton had a temperature coefficient of 0 ppb/K. This may suggest that this amide proton had an intramolecular hydrogen bond, as seen typically in cyclic peptides. Also, this might suggest that the Cys^ amide proton might be buried deep in the hydrophobic pocket of the cyclic peptide. This temperature coefficient is similar to the amide proton of D-Pen^ in DPDPE (0.9 ppb/K). The Asp^ amide proton might be involved in stabilizing a turn by hydrogen bonding with a temperature coefficient of -4.5 ppb/K. As mentioned before, RSAlOlc and DPDPE have similarly small coefficients of Cys^ and

D-Pen^, respectively. On the contrary, CCK-8 (ns) ^"^had large temperature coefficients compared to RSAlOlc. Because CCK-8 is a linear peptide, it was expected to have 176 higher temperature coefficient than the rigid, cycUc peptide RSAlOlc. Similar observations and comparisons can be made with [p-MePhe'*]DPDPE'^^ and

[|3MePhe^,Nle^]SNF-9007 (CCK-8 analogue). ^^^Graphs illustrating the temperature dependence of RSAlOl residues are provided in the Appendix.

Table 6.4. Temperature coefficients of NHa for RSAlOlc in DSMO-i/e.

A6/AT (-ppb/K) RESIDUES RSAIOIC DPDPE NS-CCK-8

Asp"

Tyr' ~ ~ 7.6

D-Cys^/D-Pen/p-MePhe 2.9 3.2 4.8

Gly^ 7.5 2.6 2.0

Trp'^/p-MePhe /Trp 5.6 5.1 7.6

CysVD-Pen/Nle 0.0 -0.3 8.8

Asp^ 4.5 ~ 4.0

Phe^ 7.1 .. 4.8

^DPDPE; Tyr-c[D-Pen-Gly-Phe-D-Pen]-OH,

''NS-CCK-8: Asp-Tyr-Met-Gly-Trp-Met-Asp-Phe-NH2 Ill 6.4 Results: Side-chain conformations

The couphng constants ^Jnh and ^Jnap are summarized in Table 2. These coupling constants are related to the backbone conformation and the mean orientation of the side chain groups. The population percentages of the three rotamers around the Ca-Cp bond were calculated with the Pachler parameters. Additional parameters distinguishing aromatic side chain groups (Tyr, Trp, Phe) from the aliphatic side chain groups also were used. '^^The calculated rotamer populations for gauche (-) and trans were assigned based on the stereospecific assignment of the Hp, using 2D-NMR experiments. ^^'^The calculated values are summarized in Table 6.5.

It is interesting that Tyr^ and Trp'' of RSAlOlc had significant gauche (+) conformations at 30 and 25%, respectively. These values are consistent with the gauche

(+) conformations of Tyr^ and Trp'^ residues of the [P-MePhe^]CCK-8 analogue. '^^(For the (2R,3R)-[p-MePhe^,Nle^]SNF-9007, the gauche (+) populations were 28 and 22% for Tyr and Trp, respectively. For (2R,3S)-[P-MePhe^,Nle^]SNF-9007, the gauche (+) populations werelS and 27%.) 178 Figure 6.1. Newman projections of the side chain rotamer populations of a-amino acids about the x torsional angle

-OC OC NH- -OC NH- H R

•^' = -60^ X' = +/-180° x' = +60° gauche (-) trans gauche (+) 179 Table 6.5. Calculated side chain rotamer populations (%) about the Ca-Cp bond (xi) in RSAlOlc.

Rotamer populations (%)

gauche (-) trans gauche (+)

Tyr(l) 35* 34* 31

D-Cys (2) 15 85 0.00

Trp(4) 60 15 25

Cys(5) 84 9.0 7

Asp (6) 32 46 22

Phe (7) 57 10 33

Rotamer populations were calculated from the measured homonuclear coupling constants using the Pachler equations. Pg(-) = (JHaHp(Pro-R) - '"JHaHpVCJHaHp - ""JnaHp); Pt =

(JHaHp(Pro-s)- " ^'^Jhohp); Pg(+) = 1 " Pg(-) - Pt. For the aliphatic residues, the values of ^'^Jaanp = 2.6 Hz and ^^JnaHp = 13.6 Hz were used. For the aromatic residues, the values of ^'^JnaHp = 3.55 Hz and ^^JnaHp = 13.9 Hz were used. Gauche(-) and trans were assigned qualitative by 2D-NMR methods as described in Kover. Cannot distinguish between gauche (-) and trans because the P,P' were not stereochemically assigned. 180

For a more meaningful comparison to reported rotamer populations of the aromatic residues, the rotamer populations for RSAlOlc were calculated using only the

Pachler parameters, which treats all residues the same (i.e. there are no correction values for aromatic residues). These values and the literature values for analogous peptides are summarized in Table 6.7.

The 17% gauche (+) conformation for Tyr' of RSAlOlc is important because it is in great contrast to the gauche (+) conformations of the corresponding Tyr in DPDPE. In

DPDPE, Tyr and Phe had little or no contribution from the gauche (+) conformation.

'^^In another DPDPE study, Tyr^ had 1% gauche (+) conformation. In c[D-Cys^,L-

Cys^]-enkephalinamide, Tyr^ had 4% gauche (+) conformation. ^^^These large differences in the gauche (+) population are interesting and unexpected since DPDPE and RSAlOlc have similar structures; both are 14-membered cyclic disulfides with aromatic residues at positions 1 and 4. Furthermore, a comparison against the potent (2S,3S)-isomer of [P-

Me-p-NOiPhe'^JDPDPE, which was calculated to only have 10% gauche (+) population for Tyr, can also be made. In this case, the increase in the gauche (+) was due to the introduction of a P-methyl substitution. It would be interesting to compare these data with those of a [Phe'^] analogue of RSAlOlc to determine whether the increase in the gauche (+) conformations of Tjo*^ and Trp"* in RSAlOlc (compared to DPDPE) was due to the Trp^ or the Asp^-Phe^-NH2 residues on the C-terminal.

This new access to the gauche (+) conformation of the Tyr at position 1 may offer new insights regarding receptor recognition and activities at the opioid receptors. What role, if any, does this conformation have at opioid receptors? How about at CCK 181 receptors? Regardless, this observation can be further exploited to probe for the common

(or different) orientation of the aromatic residues responsible for the binding affinities and bioactivities at opioid receptors and CCK receptors, as well. 182

Table 6.6. Calculated retainers for Tyr and Trp to analogous peptides in DMSO-t/e-

Rotamer RSAlOlc' DPDPE" DPDPE® CCK-7 (ns)"

Gauche (-)' 42 39 27.0 (73.0) 25

Tyr Trans® 41 60 73.0 (27.0) 58

Gauche (+) 17 3 0 17

Gauche (-)® 65 69 73.0 (27.0) 20

Trp/Phe Trans® 23 17 27.0 (73.0) 63

Gauche (+) 11 14 0 17

''Rotamer population calculated with Pachler parameters. ^DPDPE^^^, Tyr-c[D-Pen-Gly-Phe-D-Pen]-OH, calculated with Pachler parameters. "^DPDPE^^^, calculated according to deLeeuw and Altona^^\ ''CCK-7, non-sulfated^^, calculated from Pachler. ®Gauche (-) and trans might be arbitrary depending on the assignment of the P protons. 183 6.5 Results: Molecular Dynamics

Molecular Dynamics (MD) simulations were performed. From the 2D-NMR experiments, 62 unique distance restraints, 2 dihedral angle restraints, and four stereospecifically assigned P-CH2 groups were determined and included throughout the structure refinement process. The structure of RSAlOlc was calculated using hybridized distance geometry and simulated annealing protocol. The top 20 structures with the lowest restraint energy were selected to represent the solution structures of RSAlOlc. In all structures, the disulfide bond was right-handed. The overall backbone rsmd of the 20 structures aligned for all seven residues was 1.50 A. The lowest energy structure of

RSAlOlc after MD simulation is depicted in Figure 2. The peptide backbone appears to have two consecutive turns seemingly forming a helical-like structure. The same turn

• • • • SA 1 S'7 1 patterns were observed in previous NMR studies with CCK-7. ' ' In those reports, the primary sequence of CCK-8 was sufficient to form P- and y-tums around the Gly-

Trp-Met-Asp and Met-Asp-Phe-NHa sequences. Furthermore, the backbone is folded such that the Tyr^ and Trp'^ aromatic rings were closely oriented, which was supported by the observed NOEs between protons in these 2 aromatic rings. The distance between the centroids of Tyr and Trp was measured to be 7.4 A. The close proximity of the aromatic rings of Tyr and Trp was also observed in NMR studies of CCK-8. ^^'in that study, Tyr and Trp were reported to be perpendicular to each other, in an edge-to-face orientation which is caused by the electrostatic attraction between the positively charged protons of

Tyr and the negatively charged carbon atoms on Trp. '^^Interestingly, this observation 184 that the close proximity of the aromatic residues at position 1 and 4 is consistent with the observed interaction of hydrophobic side chain groups of Tyr^ and Phe'^ in DPDPE.

However, fluorescence studies of CCK-8, CCK- 7 (ns), and CCK had suggested that Tyr and Trp are far apart (R = 15 A). ' These distances are comparable to the distance

between aromatic residues at position 1 and 4 in enkephalin.

i i

Figure 6.2. The stereo view of the lowest energy structure of RSAlOlc after MD

simulation. 185 Table 6.7. The dihedral angles of the minimized mean for RSAlOlc.

Tyr' D-Cys^ Gly" Trp^ Cys' Asp® Phe' phi 156 -90 -58 -88 -83 -151 psi 162.9 31.6 -28.1 -30.8 -58.9 -100.8 149.2

Figure 6.3. Structure of RSAlOlc demonstrating a 7.4 A distance between the centroids of Tyr^ and Trp"^. The dihedral angle of -82.2 degrees of the disulfide bridge is also shown. 186 6.6 Results: Comparison to CCK-8 and DPDPE

The resulting structure of the extensive NOE-restrained molecular dynamics simulation of RSAlOlc was compared to the reported structure of CCK-8 at CCK-A receptor. The backbone of RSAlOlc turn residues (D-Cys^-Gly^-Trp'^-Cys^) of the lowest energy structure can be superimposed reasonably well (rmsd 1.13 A, see Figure

6.3) onto that of the corresponding residues (Met^-Gly'^-Trp5-Met®) of CCK-8 bound with the third extracellular loop of the human CCK-2 receptor. Coincidentally and interestingly, the spatial orientations of the Trp side chain groups of both peptides were very similar to each other. Figure 3 strongly supports our hypothesis that, at least in part of the residues 1-4, opioid and CCK ligands have overlapping pharmacophores.

However, despite the similarities of the residues in the cyclic peptide, the exocyclic residues Asp^ and Phe^ of RSAlOlc do not overlap well with backbone and side chain groups of the corresponding residues in CCK-8. The turns in these residues are obviously different, orienting the side chain groups differently. This difference in the

N-terminal structure of RSAlOlc may account for the lack of binding affinities and activities at the CCK receptors. 187

Figure 6.4. Superimposed structures of RSAlOl (blue) and CCK-8 (red). 188

Table 6.8. Comparison of backbone dihedral angles of RSAlOlc to reported dihedral angles of analogous peptides

RSAlOlc DPDPE' DPDPE" CCK-8' SNF9007''

9 ~ — — — ~ Asp V —~ " 9 67 Tyr W 163 164 150 -23 -49

9 156 111 135 -63 124 D-Pen/D-Cys/D-Phe ¥ 31 14 -29 -7 -8

9 -90 -98 -80 -29 -69 Gly W 28 -18 49 -45 -28

9 -58 -72 -153 -37 -144 Phe/Trp W 30.8 -46 -81 -42 52

9 -88 83 131 -114 47 D-Pen/Cys V -59 -19 81

9 -83 — -- -104 -98 Asp

¥ -101 -- — 71 135

9 -151 -- -119 -134 Phe

¥ 149 -- — 79 a„168 , uloyb , c , d 189 6.7. Experimental section

These experiments were performed in collaboration with Prof. Kati Kover and Mr. Jinfa

Ying. Peptide solution was prepared at a concentration of 9.6 mM by dissolving 3.9 mg of the peptide (Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2) in approximately 0.6 mL in

DMSO-i/eCCambridge Isotopes). Proton NMR spectra were recorded on a Bruker Avance

600 MHz spectrometer (University of Arizona, NMR Facility). Proton spectra were obtained at 290, 293, and 298K. TOCSY'^^ and NOESY^^^ spectra were recorded in phase sensitive mode at 290K with mixing times of 62 ms and 150ms, respectively. The

TOCSY experiment was performed with 4096 data points in^ and 512 data points in f\, and 8-16 scans were collected at each increment. The NOESY experiment was performed with 1048 data points in^ and 512 data points in f\, and 16-64 scans were collected at each increment. ^H-NMR chemical shifts were reported relative to internal

DMSO peak at 2.49 ppm. The qualitative and quantitative analyses of TOCSY and

NOESY spectra were obtained using the program package FELIX2000. The NOE-based distance restraints were obtained fi-om NOESY spectra collected with a mixing time of

150 ms. The NOE cross-peaks were integrated with the Insightll program and they were converted into upper distance bounds. Molecular dynamics was used to facilitate structure determination with the hybridized distance geometry and simulated annealing protocol in DGII program within the Insightll (Accelrys Inc., San Diego) software package on a Silicon Graphics Octane computer. One hundred structures were generated and were refined using simulated annealing AMBER force field in the Discover module 190 of Insightll. The iterative calculation was performed until distance violations did not exceed 0.16 A. CCK-8 structure at CCK-A receptor was obtained from pdb.org database, with code 1DZ6. 191 CHAPTER 7

CONCLUSIONS AND FUTURE WORK

7.1 Conclusions

The design and synthesis of a single compound that interacts with more than one receptor represents a new paradigm in drug discovery particularly in disease states that involve "systemic changes" and adaptations. Current models for drug discovery do not consider this complexity resulting in processes of biological assays that do not accurately reflect the disease state and of drug discovery with only one target. These shortcomings often result in drugs that are often not effective. Thus, the proposed paradigm for drug discovery asserts that, for such disease states, designing potential drugs for muhiple targets must be considered. This new paradigm in drug discovery was applied to treatment of neuropathic pain, which is a disease caused by opioid receptor induced analgesia which also increases the anti-opioid activities of cholecystokinin. Towards this goal, novel biologically active peptides that are agonists at the opioid receptors and antagonist at cholecystokinin were designed and synthesized.

In examining linear analogues with modifications in positions 2, 4, and 5, it was found that there are similarities between the conformational requirements between CCK and opioid ligand. [des-Asp'']-SNF-9007 analogues with substitution of D-amino acids, except for D-Pro, in position 2 resulted in good binding activities at CCK-A and CCK-B receptors and delta and mu opioid receptors. While maintaining binding and potency at 192 delta and mu receptors, substitution to Nle from N-MeNle in position 5 resulted in a balanced selectivity between CCK-A and CCK-B, primarily due to an increased binding affinity at CCK-A receptors. Substitution of D-Trp at position 4 was more tolerated at both opioid and CCK receptors when position was substituted with N-MeNle than Nle.

Similar structure activity relationships between CCK and opioid receptors support the hypothesis that CCK and opioid peptide ligands have overlapping pharmacophores.

In an attempt to improve antagonist activities at CCK receptors, bulky residues were replaced instead of Trp in position 5. Although the binding affinities and potency at delta and mu receptors decreased moderately, these series of peptides showed Trp residue can be reasonably substituted with Nal residues especially in the design of nonpeptide analogues of the CCK/opioid ligands.

CycUc disulfide analogues with substitutions of D-Cys or D-Pen in position 2 and

L-Cys, D-Cys, or D-Pen in position 5 were synthesized. Most of these cyclic disulfide analogues displayed selectivity for the delta opioid receptors and antagonist activities against CCK-8 in the tissue assays. Substitution of bulky, hydrophobic D-Pen^ or Pen^ equally maintained good potency and selectivity at the delta opioid receptors, as well as

CCK antagonist activities in the tissue assays. When D-Trp"^ was substituted, significant loss of opioid agonist and CCK antagonist activities were observed in the tissue assays.

This suggests a strict conformational requirement for Trp in binding at the opioid and

CCK receptors for disulfide bridge analogues.

Lactam analogues were investigated as well by substituting L-Lys^ or D-Lys^ and

Glu^ or D-Glu^ and exploring different conformational requirements for a 18-membered 193 ring. The lactam analogues were agonists at the opioid receptors with selectivity for mu opioid receptor, as well as antagonists against CCK-8 in the tissue assays. In contrast to to the cyclic disulfide analogues, when D-Trp was substituted position 4 in the lactam, the opioid agonist activity was maintained and the CCK antagonist activity increased.

Although the cyclic disulfide and lactam analogue displayed significant antagonist activity against CCK-8 in tissue assays, they did not have any competitive binding affinities at the human CCK-A and CCK-B receptors. This difference suggests a possible allosteric inhibition.

The structure of the cyclic disulfide RSAlOl was determined. When overlapped with a reported structure of CCK-8, the cyclic disulfide had a reasonable similarity with the backbone and the side chain groups. This validates the hypothesis that CCK and peptide opioid ligands have overlapping pharmacophores.

These series of structure-activity relationships suggest that it is possible to design one ligand for multiple targets, validating the drug design aspect of the proposed paradigm for drug discovery. Preliminary in vivo studies of the CCK/opioid peptides for anti-hyperalgesic and anti-allodynic properties in different neuropathic pain models suggest that these peptides are promising compounds for the treatment of neuropathic pain as well as a pharmacological tool for investigating systemic interactions involving

CCK and opioid receptor systems. 194 7.2 Future Work

There are interesting structure activity relationships that can done to further explore the conformational requirements for balanced opioid agonist and CCK antagonist activities. One of these studies involves the role of D-Trp in regard to the different results from the linear peptides, cyclic disulfide, and lactam. For example, it was determined that D-Trp-NMeNle results in a favorable activities, but not with D-Trp-Nle. With the cyclization of the CCK/opioid analogues, introduction of D-Trp in the cyclic disulfide series (RSA121c) resulted in loss of potency at both opioid and CCK receptors in the tissue assays. Whereas in the cyclic lactam analogues, introduction of D-Trp resulted in good potency at opioid and CCK receptors in the tissue assays even though position 5 is not a N-methylated residue. It would be interesting to determine whether N-methylation in position 5 would improve the desired opioid and CCK activities, particularly when position 4 is D-Trp (Figure 7.1).

Tyr-c[D-Lys-Gly-Trp-N-MeGlu]-Asp-Phe-NH2

Tyr-c[D-Lys-Gly-D-Trp-N-MeGlu]-Asp-Phe-NH2

Tyr-c[D-Cys-Gly-T rp-NMeCys] - Asp-Phe -NH2

Tyr-c[D-Cys-Gly-D-T rp-NMeCys] - Asp-Phe-NH2

Figure. 7.1 Proposed structures for determining the effect of N-methylation in lactam and cyclic disulfide analogues of CCKyopioid peptides. 195 In the lactam studies, introduction of D-Trp in the c[D-Lys, Glu] CCK/opioid analogue resulted in nearly equal opioid agonist activity as but the CCK antagonist activity against CCK-8 in the tissue assays increased significantly. These differential change in activity suggested that these is a possible difference in the binding conformation between CCK and opioid receptors. Of particular interest is the orientation of the aromatic side chain groups. It would be interesting to explore this by the incorporation of all four isomers of beta-methyl substituted Trp at position 4 (Figure 7.2).

Tyr-c[D-Lys-Gly-(p-Me)-Trp-Glu]-Asp-Phe-NH2

Figure 7.2. Proposed structure incorporating p-MeTrp substituted in cyclic lactam analogues.

These constrained, cyclic analogues of CCK/opioid peptides can be used as tools in the de novo design of non-peptide ligands with opioid agonist and CCK antagonist properties. 196 REFERENCES

(1) Hmby, V. J.; Agnes, R. S.; Lee, Y.-S.; Davis, P.; Ma, S.-W. et al. Designing peptide dmgs/ligands for pathological states: A new paradigm for design of bioactive peptide hormones and neurotransmitters. In Peptides: The Wave of the Future-, Lebl, M., Houghten, R. A. Eds.; American Peptide Society 2001: San Diego, 2001; pp 969-970.

(2) Classification of Chronic Pain-, lASP Press: Seattle.

(3) Vrinten, D. H.; Kalkman, C. J.; Adam, R. A. H.; Gispen, W. H. Neuropathic pain: a possible role for the melanocortin system? Eur. J. Pharmacol. 2001, 429, 61-69.

(4) Vanderah, T. W.; Ossipov, M. H.; Lai, J.; Malan, T. P., Jr.; Porreca, F. Mechanisms of opioid-induced pain and antinociceptive tolerance: descending facilitation and spinal dynorphin. Pain. 2001, 92, 5-9.

(5) Gardell, L. R.; Wang, R.; Burgess, S. E.; Ossipov, M. H.; Vanderah, T. W. et al. Sustained morphine exposure induces a spinal dynorphin-dependent enhancement of excitatory transmitter release from primary afferent fibers. J. Neurosci. 2002, 22, 6747-6755.

(6) Noble, F.; Roques, B. P. CCK-B receptor: chemistry, molecular biology, biochemistry and pharmacology. Prog. Neurobiol. 1999, 58, 349-379.

(7) Stanfa, L.; Dickenson, A.; Xu, X. J.; Wiesenfeld-Hallin, Z. Cholecystokinin and morphine analgesia: variations on a theme. Trends in Pharm Sci. 1994, 15, 65-66.

(8) Wiesenfeld-Hallin, Z.; de Arauja Lucas, G.; Alster, P.; Xu, X. J.; Hokfelt, T. Cholecystokinin/opioid interactions. Brain Res. 1999, 848, 78-89.

(9) Stengaard-Pedersen, K.; Larsson, L. 1. Localization and opiate receptor binding of enkephalin, CCK and ACTH/beta-endorphin in the rat central nervous system. Peptides. 1981, 2, 3-19.

(10) Ghilardi, J. R.; Allen, C. J.; Vigna, S. R.; McVey, D. C.; Mantyh, P. W. Trigeminal and dorsal root ganglion neurons express CCK receptor binding sites in the rat, rabbit, and monkey: possible site of opiate-CCK analgesic interactions. J. Neurosci. 1992, 12, 4854-4866.

(11) Zhou, Y.; Sun, Y. H.; Zhang, Z. W.; Han, J. S. Increased release of immunoreactive cholecystokinin octapeptide by morphine and potentiation of mu- 197 opioid analgesia by CCKB receptor antagonist L-365,260 in rat spinal cord. Eur. J. Pharmacol. 1993, 234, 147-154.

(12) Zhou, Y.; Sun, Y. H.; Shen, J. M.; Han, J. S. Increased release of immunoreactive CCK-8 by electroacupuncture and enhancement of electroacupuncture analgesia by CCK-B antagonist in rat spinal cord. Neuropeptides. 1993, 24, 139-144.

(13) Zhou, Y.; Sun, Y. H.; Zhang, Z. W.; Han, J. S. Accelerated expression of cholecystokinin gene in the brain of rats rendered tolerant to morphine. Neuroreport. 1992, 3, 1121-1123.

(14) Tseng, L. F.; Collins, K. A. Cholecystokinin administered intrathecally selectively antagonizes intracerebroventricular beta-endorphin-induced tail-flick inhibition in the mouse, Pharmacol. Exp. Ther. 1992, 260, 1086-1092.

(15) Faris, P. L.; Komisaruk, B. R.; Watkins, L. R.; Mayer, D. J. Evidence for the neuropeptide cholecystokinin as an antagonist of opiate analgesia. Science. 1983, 279,310-312.

(16) Watkins, L. R.; Kinscheck, I. B.; Kaufman, E. F.; Miller, J.; Frenk, H. et al. Cholecystokinin antagonists selectively potentiate analgesia induced by endogenous opiates. 5ra/>2 1985, 327, 181-190.

(17) Watkins, L. R.; Kinscheck, I. B.; Mayer, D. J. Potentiation of morphine analgesia by the cholecystokinin antagonist proglumide. 1985, 327, 169-180.

(18) Vanderah, T. W.; Bernstein, R. N.; Yamamura, H. I.; Hruby, V. J.; Porreca, F. Enhancement of morphine antinociception by a CCKB antagonist in mice is mediated via opioid delta receptors. J. Pharm. Exper. Therap. 1996, 278, 212-219.

(19) Ossipov, M. H.; Kovelowski, C. J.; Vanderah, T.; Porreca, F. Naltrindole, an opioid delta antagonist, blocks the enhancement of morphine-antinociception induced by a CCKB antagonist in the rat. Neurosci. Lett. 1994, 181, 9-12.

(20) Xu, X. J.; Wiesenfeld-Hallin, Z.; Hughes, J.; Horwell, D. C.; Hoekfelt, T. CI988, a selective antagonist of cholecystokininB receptors, prevents morphine tolerance in the rat. Br. J. Pharmacol. 1992, 105, 591-596.

(21) Dourish, C. T.; O'Neill, M. F.; Coughlan, J.; Kitchener, S. J.; Hawley, D. et al. The selective CCK-B receptor antagonist L-365,260 enhances morphine analgesia and prevents morphine tolerance in the rat. Eur. J. Pharmacol. 1990, 176, 35-44. 198 (22) Panerai, A. E.; Rovati, L. C.; Cocco, E.; Sacerdote, P.; Mantegazza, P. Dissociation of tolerance and dependence to morphine: a possible role for cholecystokinin. Brain Res. 1987, 410, 52-60.

(23) Hruby, V. J. Designing peptide receptor agonists and antagonists. Nature Reviews. Drug Discovery. 2002, 1, 847-858.

(24) Hruby, V. J.; Agnes, R. S.; Cai, C. Design of peptide agonists. In Meth. Enzymol.; Iyengar, R., Hildebrandt, J. D. Eds.; Academic Press, Inc., 2002; pp 73-91.

(25) Hruby, V. J. Design in topographical space of peptide and peptidomimetic ligands that affect behavior. A chemist's glimpse at the mind—body problem. Acc. Chem. Res.. 2001, 34, 389-397.

(26) Vitoux, B.; Aubry, A.; Cung, M. T.; Marraud, M. N-Methyl peptides. VII. Conformational perturbations induced by N-methylation of model dipeptides. Int. J. Pept. Protein Res. 1986, 27, 617-632.

(27) Ron, D.; Gilon, C.; Hanani, M.; Vromen, A.; Selinger, Z. et al. N-methylated analogs of Ac[Nle28,31]CCK(26-33): synthesis, activity, and receptor selectivity. J. Med. Chem. 1992, 35, 2806-2811.

(28) Opioid Peptides: Biology, Chemistry and Genetics. In The Peptides: Analysis, Synthesis, and Biology, Udenfriend, S., Meienhofer, J. Eds.; Academic Press, Inc, 1984.

(29) Hruby, V. J.; Gehrig, C. Recent developments in the design of receptor specific Medicinal Research Reviews. 1989, 9, 343-401.

(30) Hruby, V. J.; Agnes, R. S. Conformation-activity relationships of opioid peptides with selective activities at opioid receptors. Biopolymers. 1999, 51, 391-410.

(31) Hughes, J.; Smith, T. W.; Kosterlitz, H. W.; Fothergill, L. A.; Morgan, B. A. et al. Identification of two related pentapeptides from the brain with potent opiate agonist activity. Nature. 1975, 258, 577-580.

(32) Buscher, H. H.; Hill, R. C.; Romer, D.; Cardinaux, F.; Closse, A. et al. Evidence for analgesic activity of enkephalin in the mouse. Nature. 1976, 261, 423-425.

(33) Morgan, B. A.; Smith, C. F.; Waterfield, A. A.; Hughes, J.; Kosterlitz, H. W. Structure-activity relationships of methionine-enkephalin. J. Pharm. Pharmacol. 1976, 28, 660-661. 199 (34) Chang, J. K.; Fong, B. T.; Pert, A.; Pert, C. B. Opiate receptor affinities and behavioral effects of enkephalin: structure-activity relationship of ten synthetic peptide analogues. L//e 5'c/. 1976, 18, 1473-1481.

(35) Lu, Y.; Nguyen, T. M.; Weltrowska, G.; Berezowska, I.; Lemieux, C. et al. [2',6'- Dimethyltyrosinejdynorphin A(1-11)-NH2 analogues lacking an N-terminal amino group: potent and selective kappa opioid antagonists. J. Med. Chem. 2001, 44, 3048-3053.

(36) Beddell, C. R.; Clark, R. B.; Lowe, L. A.; Wilkinson, S.; Chang, K. J. et al. A conformational analysis for leucine-enkephalin using activity and binding data of synthetic analogues. Brit. J. Pharm. 1977, 61, 351-356.

(37) Shaw, J. S.; Tumbull, M. J. In vitro profile of some opioid pentapeptide analogues. Eur. J. Pharmacol. 1978, 49, 313-317.

(38) Hambrook, J. M.; Morgan, B. A.; Ranee, M. J.; Smith, C. F. Mode of deactivation of the enkephalins by rat and human plasma and rat brain homogenates. Nature. 1976, 262, 782-783.

(39) Montecucchi, P. C.; de Castiglione, R.; Piani, S.; Gozzini, L.; Erspamer, V. Amino acid composition and sequence of dermorphin, a novel opiate-like peptide from the skin of Phyllomedusa sauvagei. Int. J. Pept. Protein Res. 1981, 17, 275- 283.

(40) Erspamer, V. The opioid peptides of the amphibian skin. Int. J. Dev. Neurosci. 1992, 10, 3-30.

(41) Lazarus, L. H.; Salvadori, S.; Attila, M.; Grieco, P.; Bundy, D. M. et al. Interaction of deltorphin with opioid receptors: molecular determinants for affinity and selectivity. Peptides. 1993, 14, 21-28.

(42) Lazarus, L. H.; Attila, M. The toad, ugly and venomous, wears yet a precious jewel in his skin. Prog. Neurobiol. 1993, 473-507.

(43) Beddell, C. R.; Clark, R. B.; Hardy, G. W.; Lowe, L. A.; Ubatuba, F. B. et al. Structural requirements for opioid activity of analogues of the enkephalins. Proceedings of the Royal Society of London - Series B: Biological Sciences. 1977, 198, 249-265.

(44) Hruby, V. J.; Bartosz-Bechowski, H.; Davis, P.; Slaninova, J.; Zalewska, T. et al. Cyclic enkephalin analogues with exceptional potency and selectivity for delta- opioid receptors. J. Med. Chem. 1997, 40, 3957-3962. 200 (45) Schiller, P. W.; Yam, C. F.; Prosmamie, J. Synthesis, opiate receptor affinity, and conformational parameters of [4-tryptophan]enkephalin analogues. J. Med. Chem. 1978, 21, 1110-1116.

(46) Lord, J. A.; Waterfield, A. A.; Hughes, J.; Kosterlitz, H. W. Endogenous opioid peptides: multiple agonists and receptors. Nature. 1977, 267, 495-499.

(47) Roques, B. P.; Gacel, G.; Foumie-Zaluski, M. C.; Senault, B.; Lecomte, J. M. Demonstration of the crucial role of the phenylalanine moiety in enkephalin analogues for differential recognition of the mu- and delta-receptors. Eur. J. Pharmacol. 1979, 60, 109-110.

(48) Schiller, P. W.; Natarajan, S.; Bodanszky, M. Determination of the intramolecular tyrosine-tryptophan distance in a 7-peptide related to the C-terminal sequence of cholecystokinin. Int. J. Pept. Protein Res. 1978, 12, 139-142.

(49) Handa, B. K.; Land, A. C.; Lord, J. A.; Morgan, B. A.; Ranee, M. J. et al. Analogues of beta-LPH61-64 possessing selective agonist activity at mu-opiate receptors. Eur. J. Pharmacol. 1981, 70, 531-540.

(50) Szekely, J. I.; Ronai, A. Z.; Dunai-Kovacs, Z.; Miglecz, E.; Berzetri, I. et al. (D- Met^, Pro^)-enkephalinamide: a potent morphine-like analgesic. Eur. J. Pharmacol. 1977, 42, 293-294.

(51) Bajusz, S.; Ronai, A. Z.; Szekely, J. I.; Graf, L.; Dunai-Kovacs, Z. et al. A superactive antinociceptive pentapeptide, (D-Met^, Pro^)-enkephalinamide. FEES Lett. 1977, 76, 91-92.

(52) Audigier, Y.; Mazarguil, H.; Cros, J. (D-Ala^, N-Val^) enkephalinamide and (D- Met^, N-Val^) enkephalinamide. Two potent agonists of opiate and enkephalin XQC&^i0X%. FEES Lett. 1980, 110, 88-90.

(53) Morley, J. S. Structure-activity relationships of enkephalin-like peptides. Annu. Rev. Pharmacol. Toxicol. 1980, 20, 81-110.

(54) DiMaio, J.; Lemieux, C.; Schiller, P. W. Structure-activity relationships of cyclic enkephalin analogs. Life Sci. 1982, 31, 2253-2256.

(55) DiMaio, J.; Nguyen, T. M.; Lemieux, C.; Schiller, P. W. Synthesis and pharmacological characterization in vitro of cyclic enkaphalin analogues; effect of conformational constrainsts on opiate receptor selectivity. J. Med. Chem. 1982, 25, 1432-1438. 201 (56) Schiller, P. W.; Eggimann, B.; DiMaio, J.; Lemieux, C.; Nguyen, T. M. D. CycHc enkephalin analogs containing a cystine bridge. Biochem. Biophys. Res. Commun. 1981, 101, 337-343.

(57) Mosberg, H. I.; Hurst, R.; Hruby, V. J.; Gee, K.; Yamamura, H. I. et al. Bis- penicillamine enkephalins possess highly improved specificity toward delta opioid receptors. Proc. Natl. Acad. Sci. U. S. A. 1983, 80, 5871-5874.

(58) Mosberg, H. I.; Hurst, R.; Hruby, V. J.; Gee, K.; Akiyama, K. et al. Cyclic penicillamine containing enkephalin analogs display profound delta receptor selectivities. Life Sci. 1983, 33 Suppl 1, 447-450.

(59) Haaseth, R. C.; Zalewska, T.; Davis, P.; Yamamura, H. I.; Porreca, F. et al. Para- substituted phenylalanine-4 analogues of [L-Ala^]DPDPE: highly selective delta opioid receptor ligands./. Peptide Res. 1997, 50, 171-177.

(60) Svensson, C. I.; Rew, Y.; Malkmus, S.; Schiller, P. W.; Taulane, J. P. et al. Systemic and spinal analgesic activity of a delta-opioid-selective lanthionine enkephalin analog. J. Pharmacol. Exp. Ther. 2003, 304, 827-832.

(61) Rew, Y.; Malkmus, S.; Svensson, C.; Yaksh, T. L.; Chung, N. N. et al. Synthesis and biological activities of cyclic lanthionine enkephalin analogues: delta-opioid receptor selective ligands. J. Med. Chem. 2002, 45, 3746-3754.

(62) Noble, F.; Wank, S. A.; Crawley, J. N.; Bradwejn, J.; Seroogy, K. B. et al. International Union of Pharmacology. XXI. Structure, Distribution, and Functions of Cholecystokinin Receptors. Pharmacol. Rev. 1999, 51, 745-781.

(63) Mutt, v.; Jorpes, J. E. Structure of porcine cholecystokinin-pancreozymin. Eur. J. Biochem. 1968, 6, 156-162.

(64) Vanderhaeghen, J. J.; Signeau, J. C.; Gepts, W. New peptide in the vertebrate CNS reacting with antigastrin antibodies. Nature. 1975, 257, 604-605.

(65) Innis, R. B.; Snyder, S. H. Distinct cholecystokinin receptors in brain and pancreas. Proc. Natl. Acad. Sci. U. S. A. 1980, 77, 6917-6921.

(66) Hill, D. R.; Shaw, T. M.; Woodruff, G. N. Species differences in the localization of'peripheral' type cholecystokinin receptors in rodent brain. Neurosci. Lett. 1987, 79, 286-289.

(67) Hill, D. R.; Campbell, N. J.; Shaw, T. M.; Woodruff, G. N. Autoradiographic localization and biochemical characterization of peripheral type CCK receptors in 202 rat CNS using highly selective nonpeptide CCK antagonists. J. Neurosci. 1987, 7, 2967-2976.

(68) Saito, A.; Sankaran, H.; Goldfine, I. D.; Williams, J. A. Cholecystokinin receptors in the brain: characterization and distribution. Science. 1980, 208, 1155-1156.

(69) Hill, D. R.; Shaw, T. M.; Woodruff, G. N. Binding sites for 1251-cholecystokinin in primate spinal cord are of the CCK-A subclass. Neurosci. Lett. 1988, 89, 133- 139.

(70) Gibbs, J.; Smith, G. P. Cholecystokinin and satiety in rats and rhesus monkeys. Amer. J. Clin. Nutr. 1977, 30, 758-761.

(71) Singh, L.; Lewis, A. S.; Field, M. J.; Hughes, J.; Woodruff, G. N. Evidence for an involvement of the brain cholecystokinin B receptor in anxiety. Proc. Natl. Acad. Sci. U. S. A. 1991, 88, 1130-1133.

(72) Bourin, M.; Malinge, M.; Vasar, E.; Bradwejn, J. Two faces of cholecystokinin: anxiety and schizophrenia. Fundam. Clin. Pharmac. 1996, 10, 116-126.

(73) Griebel, G. Is there a future for neuropeptide receptor ligands in the treatment of anxiety disorders? Ther. 1999, 82, 1-61.

(74) Katsuura, G.; Itoh, S. Preventive effect of cholecystokinin octapeptide on experimental amnesia in rats. 1986, 7, 105-110.

(75) Fekete, M.; Lengyel, A.; Hegedus, B.; Penke, B.; Zarandy, M. et al. Further analysis of the effects of cholecystokinin octapeptides on avoidance behaviour in rats. Eur. J. Pharmacol. 1984, 98, 79-91.

(76) Baber, N. S.; Dourish, C. T.; Hill, D. R. The role of CCK caerulein, and CCK antagonists in nociception. Pain. 1989, 39, 307-328.

(77) Han, J. S. Cholecystokinin octapeptide (CCK-8): a negative feedback control mechanism for opioid analgesia. Prog. Brain Res. 1995, 105, 263-271.

(78) Wiesenfeld-Hallin, Z.; Xu, X. J.; Hokfelt, T. The role of spinal cholecystokinin in chronic pain states. Pharmacol. Toxicol. 2002, 91, 398-403.

(79) Sugg, E.; Shook, J. E.; Serra, M.; Kroc, M.; Yamamura, H. I. et al. Synthesis and biological evaluation of N-(alpha)-hydroxylsulfonyl-[Nle^^'^h-CCK-26-33. Life Sci. 1986, 39, 1623-1129. 203 (80) Hruby, V. J.; Fang, S.; Knapp, R.; Kazmierski, W.; Lui, G. K. et al. Cholecystokinin analogs with high affinity and selectivity for brain membrane receptors. Int. J. Pept. Protein Res. 1990, 35, 566-573.

(81) Boteju, L. W.; Nikiforovich, G. V.; Haskell-Luevano, C.; Fang, S. N.; Zalewska, T. et al. The use of topographical constraints in receptor mapping: investigation of the topographical requirements of the tryptophan 30 residue for receptor binding of Asp-Tyr-D-Phe-Gly-Trp-(N-Me)Nle-Asp-Phe-NH2 (SNF 9007), a cholecystokinin (26-33) analogue that binds to both CCK-B and delta-opioid receptors. J. Med. Chem. 1996, 39, 4120-4124.

(82) Slaninova, J.; Knapp, R. J.; Wu, J. J.; Fang, S. N.; Kramer, T. et al. Opioid receptor binding properties of analgesic analogues of cholecystokinin octapeptide. Eur. J. Pharmacol. 1991, 200, 195-198.

(83) Danho, W.; Tilley, J. W.; Shiuey, S. J.; Kulesha, I.; Swistok, J. et al. Structure- activity studies of tryptophan 30 modified analogs of Ac-CCK-7. Int. J. Pept. Protein Res. 1992, 39, 337-347.

(84) Foumie-Zaluski, M. C.; Belleney, J.; Lux, B.; Durieux, C.; Gerard, D. et al. Conformational analysis of cholecystokinin CCK26-33 and related fragments by IH NMR spectroscopy, fluorescence-transfer measurements, and calculations. Biochemistry. 1986, 25, 3778-3787.

(85) Charpentier, B.; Pelaprat, D.; Durieux, C.; Dor, A.; Reibaud, M. et al. Cyclic cholecystokinin analogues with high selectivity for central receptors. Proc. Natl. Acad. Sci. U. S. A. 1988, 85, 1968-1972.

(86) Charpentier, B.; Dor, A.; Roy, P.; England, P.; Pham, H. et al. Synthesis and binding affinities of cyclic and related linear analogs of CCK8 selective for central receptors. J. Chem. 1989, 32, 1184-1190.

(87) Blommaert, A. G.; Dhotel, H.; Ducos, B.; Durieux, C.; Goudreau, N. et al. Structure-based design of new constrained cyclic agonists of the cholecystokinin CCK-B receptor. J. Med Chem. 1997, 40, 647-658.

(88) Revel, L.; Makovec, F. Update on nonpeptide CCK-B recptor antagonists. Drugs Put. 1998, 23, 751-766.

(89) Hahne, W. F.; Jensen, R. T.; Lemp, G. F.; Gardner, J. D. Proglumide and benzotript: members of a different class of cholecystokinin receptor antagonists. Proc. Natl. Acad Sci. U. S. A. 1981, 78, 6304-6308. 204 (90) Evans, B. E.; Bock, M. G.; Rittle, K. E.; DiPardo, R. M.; Whitter, W. L. et al. Design of potent, orally effective, nonpeptidal antagonists of the peptide hormone cholecystokinin. Proc. Natl. Acad. Sci. U. S. A. 1986, 83, 4918-4922.

(91) Evans, B. E.; Rittle, K. E.; Bock, M. G.; DiPardo, R. M.; Freidinger, R. M. et al. Design of nonpeptidal ligands for a peptide receptor: cholecystokinin antagonists. J. Med. Chem. 1987, 30, 1229-1239.

(92) Chang, R. S.; Lotti, V. J. Biochemical and pharmacological characterization of an extremely potent and selective nonpeptide cholecystokinin antagonist. Proc. Natl. Acad Sci. U. S. A. 1986, 83,4923-4926.

(93) Horwell, D. C.; Hughes, J.; Hunter, J. C.; Pritchard, M. C.; Richardson, R. S. et al. Rationally designed "dipeptoid" analogues of CCK. alpha-Methyltryptophan derivatives as highly selective and orally active gastrin and CCK-B antagonists with potent anxiolytic properties. J. Med. Chem. 1991, 34, 404-414.

(94) Bellier, B.; McCort-Tranchepain, I.; Ducos, B.; Danascimento, S.; Meudal, H. et al. Synthesis and biological properties of new constrained CCK-B antagonists: discrimination of two affinity states of the CCK-B receptor on transfected CHO cells. J. Med Chem. 1997, 40, 3947-3956.

(95) Blommaert, A. G.; Weng, J. H.; Dorville, A.; McCort, I.; Ducos, B. et al. Cholecystokinin peptidomimetics as selective CCK-B antagonists: design, synthesis, and in vitro and in vivo biochemical properties. J. Med. Chem. 1993, 36, 2868-2877.

(96) Schiller, P. W.; Lipton, A.; Horrobin, D. F.; Bodanszky, M. Unsulfated C- terminal 7-peptide of cholecystokinin: a new hgand of the opiate receptor. Biochem. Biophys. Res. Commun. 1978, 85, 1332-1338.

(97) Pincus, M. R.; Carty, R. P.; Chen, J.; Lubowsky, J.; Avitable, M. et al. On the biologically active structures of cholecystokinin, little gastrin, and enkephalin in the gastrointestinal system. Proc. Natl. Acad. Sci. U. S. A. 1987, 84, 4821-4825.

(98) Nikiforovich, G. V.; Hruby, V. J. Models for the A- and B-receptor-bound conformations of CCK-8. Biochem. Biophys. Res. Commun. 1993, 194, 9-16.

(99) Hruby, V. J.; Fang, S. N.; Kramer, T. H.; Davis, P.; Parkhurst, D. et al. Analogs of cholecystokinin26-33 selective for B-t5^e CCK receptors possess .delta, opioid receptor agonist activity in vitro and in vivo: evidence for similarities in CCK-B and .delta.-opioid receptor requirements. IrxPept.: Chem., Struct. Biol, Proc. Am. Pept. Symp., 13th; Hodges, R., Smith, A. J. Eds., 1994; pp 669-671. 205 (100) Nikiforovich, G. V.; Kolodziej, S. A.; Nock, B.; Bemad, N.; Martinez, J. et al. Conformationally readdressed CCK-B/delta-opioid peptide ligands. Biopolymers. 1995, 36, 439-452.

(101) Chang, R. S.; Lotti, V. J.; Chen, T. B.; Keegan, M. E. Tifluadom, a kappa-opiate agonist, acts as a peripheral cholecystokinin receptor antagonist. Neurosci. Lett. 1986, 72,211-214.

(102) James, I. F.; Goldstein, A. Site-directed alkylation of multiple opioid receptors. I. Binding selectivity. Mol. Pharmacol. 1984, 25, 337-342.

(103) Morley, J. E.; Levine, A. S.; Grace, M.; Kneip, J.; Zeugner, H. The effect of opioid-benzodiazepine, tifluadom on ingestive behaviors. Eur. J. Pharmacol. 1983, 87.

(104) Romer, D.; Buscher, H. H.; Hill, R. C.; Maurer, R.; Petcher, T. J. et al. An opioid benzodiazepine. Nature. 1982, 298, 759-760.

(105) Collins, N.; Flippen-Anderson, J. L.; Hasseth, R. C.; Deschamps, J. R.; George, C. et al. Conformational determinants of agonist versus antagonists properties of [D- Pen^' D-Pen^]enkephalin (DPDPE) analogs at opioid receptors. Comparison of X- ray crystallographic structure, solution IH NMR data, and molecular djmamic simulations of [L-Ala^]DPDPE and [D-Ala^]DPDPE. J. Am. Chem. Soc. 1996, 775,2143-2152.

(106) Shenderovich, M. D.; Liao, S.; Qian, X.; Hruby, V. J. A three-dimensional model of the delta-opioid pharmacophore: comparative molecular modeling of peptide and nonpeptide ligands. Biopolymers. 2000, 53, 565-580.

(107) Nikiforovich, G. V.; Hruby, V. J.; Prakash, O.; Gehrig, C. A. Topographical requirements for .delta.-selective opioid peptides. Biopolymers 1991, 31, 941-955.

(108) Nikiforovich, G.; Hruby, V. J. Models for the A- and B- receptor-bound conformations of CCK-8. Biochem. Biophys. Res. Commun. 1993, 194, 9-16.

(109) Takeda, Y.; Hoshino, M.; Yanaihara, N.; Yanaihara, C.; Isobe, J. et al. Comparison of CCK-8 receptors in the pancreas and brain of rats using CCK-8 analogues. Jap. J. of Pharmacolgy. 1989, 49, 471-481.

(110) Hruby, V. J.; Fang, S.; Knapp, R.; Kazmierski, W.; Lui, G. K. et al. Cholecystokinin analogs v^^ith high affinity and selectivity for brain versus peripheral membrane receptors. In Pept.: Chem., Struct. Biol, Proc. Am. Pept. Symp., llth\ Rivier, J. E., Marshall, G. R. Eds., 1990; pp 53-55. 206 (111) Hughes, J.; Boden, P.; Costall, B.; Domeney, A.; Kelly, E. et al. Development of a class of selective cholecystokinin type B receptor antagonist having potent anxiolj^ic activity. Proc. Natl. Acad. Sci. U. S. A. 1991, 87, 6728-6732.

(112) Kaiser, E.; Colescott, R. L.; Bossinger, C. D.; Cook, P. 1. Color test for detection of free terminal amino groups in the solid-phase synthesis of peptides. Anal. Biochem. 1970, 34, 595-598.

(113) Vojkovsky, T. Detection of secondary amines on solid phase. Pep. Research. 1995, 8, 236-237.

(114) Hancock, W. S.; Battersby, J. E. A new micro-test for the detection of incomplete coupling reactions in solid-phase peptide synthesis using 2,4,6- trinitrobenzenesulphonic acid. Anal Biochem. 1976, 71, 260-264.

(115) Coste, Jacques; Dufour, M.-N.; Pantaloni, A.; Castro, B. Drop: A new reagent for coupling N-methylated amino acids. Tetrahedron Lett. 1990, 31, 669-672.

(116) Pearson, Daniel A.; Blanchette, M.; Baker, M. L.; Guindon, C. A. Trialkylsilanes as scavengers for the trifluoroacetic acid deblocking of protecting groups in peptide synthesis. Tetrahedron Lett. 1989, 30, 2739-2742.

(117) Lanzilotti, A. E.; Littell, R.; Fanshawe, W. J.; McKenzie, T. C.; Lovell, F. M. Stereoselective reduction of some indoles with triethylsilane-trifluoroacetic acids. J. Org. Chem. 1979, 44, 4809-4815.

(118) Lin, C. W.; Holladay, M. W.; Barrett, R. W.; Wolfram, C. A.; Miller, T. R. et al. Distinct requirements for activation at CCK-A and CCK-B/gastrin receptors: studies with a C-terminal hydrazide analogue of cholecystokinin tetrapeptide (30- 33). Mol. Pharmacol. 1989, 36, 881-886.

(119) Hosohata, K.; Varga, E. V.; Alfaro-Lopez, J.; Tang, X.; Vanderah, T. W. et al. (2S,3R) beta-methyl-2',6'-dimethyltyrosine-L-tetrahydroisoquinoline-3-carboxylic acid [(2S,3R)TMT-L-Tic-OH] is a potent, selective delta-opioid receptor antagonist in mouse brain. J. Pharmacol. Exp. Ther. 2003, 304, 683-688.

(120) Patel, S.; Chapman, K. L.; Smith, A. J.; Bailey, I.; Freedman, S. B. Are radioligand antagonist/agonist binding ratios in rat pancreas predictive of fimctional efficacy of cholecystokinin receptor agonists and antagonists? Reeul. Pept. 1996, 65, 29-35.

(121) Kramer, T. H.; Davis, P.; Hruby, V. J.; Burks, T. F.; Porreca, F. In vitro potency, affinity and agonist efficacy of highly selective delta opioid receptor ligands. J. Pharmacol. Exp. Ther. 1993, 266, 577-584. 207

(122) Lucaites, V. L.; Mendelsohn, L. G.; Mason, N. R.; Cohen, M. L. CCK-8, CCK-4 and gastrin-induced contractions in guinea pig ileum: evidence for differential release of acetylcholine and substance P by CCK-A and CCK-B receptors. J. Pharmacol. Exp. Ther. 1991, 256, 695-703.

(123) Dal Fomo, G.; Pietra, C.; Urciuoli, M.; van Amsterdam, F. T.; Toson, G. et al. Evidence for two cholecystokinin receptors mediating the contraction of the guinea pig isolated ileum longitudinal muscle myenteric plexus. J. Pharmacol. Exp. Ther. 1992, 261, 1056-1063.

(124) Dunlop, J. CCK Receptor Antagonists. Gen. Pharmac. 1998, 31, 519-524.

(125) Hruby, V. J.; Toth, G.; Gehrig, C. A.; Kao, L. F.; Knapp, R. et al. Topographically designed analogs of [cyclic] [D-Pen^,D-Pen^]enkephalin. J. Med. Chem. 1991, 34, 1823-1830.

(126) Shenderovich, M. D.; Kover, K. E.; Nikiforovich, G. V.; Jiao, D.; Hruby, V. J. Conformational analysis of beta-methyl-para-nitrophenylalanine stereoisomers of cyclo[D-Pen2, D-Pen5]enkephalin by NMR spectroscopy and conformational energy calculations. 1996, 38, 141-156.

(127) Fulcrand, P.; Rodriguez, M.; Galas, M. C.; Lignon, M. F.; Laur, J. et al. 2- Phenylethyl ester and 2-phenylethyl amide derivative analogues of the C-terminal hepta- and octapeptide of cholecystokinin. Int. J. Pept. Protein Res. 1988, 32, 384-395.

(128) Madalengoitia, J. S. A Novel Peptide Fold: A Repeating .beta.II' Turn Secondary Structure. J. Am. Chem. Soc. 2000, 122, 4986-4987.

(129) Makovec, F.; Chiste, R.; Bani, M.; Pacini, M.; Setnikar, I. et al. New glutamic acid derivatives with potent competitive and specific cholecystokinin-antagonistic activity. Arzneim-Forsch Drug Res. 1985, 35, 1048-1051.

(130) Hruby, V. J.; Li, G.; Haskell-Luevano, C.; Shenderovich, M. Design of peptides, proteins, and peptidomimetics in chi space. Biopolymers. 1997, 43, 219-266.

(131) Mierke, D. F.; Said-Nejad, O. E.; Schiller, P. W.; Goodman, M. Enkephalin analogs containing beta-naphthylalanine at the fourth position. Biopolymers. 1990, 29, 179-196.

(132) Tilley, J. W.; Danho, W.; Shiuey, S. J.; Kulesha, I.; Sarabu, R. et al. Structure activity of C-terminal modified analogs of Ac-CCK-7. Int. J. Pept. Protein Res. 1992, 39, 322-336. 208

(133) Liao, S.; Alfaro-Lopez, J.; Shenderovich, M. D.; Hosohata, K.; Lin, J. et al. De novo design, synthesis, and biological activities of high-affinity and selective non- peptide agonists of the delta-opioid receptor. J. Med. Chem. 1998, 41, 4767-4776.

(134) Weber, S. J.; Greene, D. L.; Hruby, V. J.; Yamamura, H. L; Porreca, F. et al. Whole body and brain distribution of [^H]cyclic [D-Pen^,D-Pen^]enkephalin after intraperitoneal, intravenous, oral and subcutaneous administration. J. Pharmacol. Exp. Ther. 1992, 263, 1308-1316.

(135) Greene, D. L.; Hau, V. S.; Abbruscato, T. J.; Bartosz, H.; Misicka, A. et al. Enkephalin analog prodrugs: assessment of in vitro conversion, enzyme cleavage characterization and blood-brain barrier permeability. J. Pharmacol. Exp. Ther. 1996, 277,1366-1375.

(136) Misicka, A.; Hruby, V. J. Optimization of Disulfide Bond Formation. Polish J. of Chem. 1994, 68, 893-899.

(137) White, P. Fmoc-Trp(Boc)-OH: A new derivative for the synthesis of peptides containing tryptophan. In Peptides: Chemsitry and Biology.', Smith, J. A., Rivier, J. E. Eds.; ESCOM: Leiden, 1992; pp 537-538.

(138) Han, Y.; Albericio, F.; Barany, G. Occurrence and minimization of cysteine racemization during stepwise solid-phase peptide synthesis. J. Org. Chem. 1997, 62, 4307-4312.

(139) Angell, Y. M.; Alsina, J.; Albericio, F.; Barany, G. Practical protocols for stepwise solid-phase synthesis of cysteine-containing peptides. J. Peptide Res. 2002, 60, 292-299.

(140) Wallace, T. J. Reactions of Thiols with Sulfoxides. I. Scope of the Reaction and Synthetic Apphcations. J. Am. Chem. Soc. 1964, 86, 2018-2021.

(141) Tam, J. P.; Wu, C.-R.; Liu, W.; Zhang, J.-W. Disulfide bond formations in peptides by dimethy sulfoxide. Scope and applications. J. Am. Chem. Soc. 1991, 113, 6657-6662.

(142) Fujii, N.; Otaka, A.; Funakoshi, S.; Bessho, K.; Watanabe, T. et al. Studies on peptides. CLI. Syntheses of cystine-peptides by oxidation of S-protected cysteine- peptides with thallium(III) trifluoroacetate. Chem. Pharm. Bull. 1987, 35, 2339- 2347. 209 (143) Albericio, F.; Hammer, R. P.; Garcia-Echeverria, C.; Molins, M. A.; Chang, J. L. et al. Cyclization of disulfide-containing peptides in solid-phase synthesis. Int. J. Pept. Protein Res. 1991, 37, 402-413.

(144) Mosberg, H. I.; Hurst, R.; Hmby, V. J.; Galligan, J. J.; Burks, T. F. et al. Conformationally constrained cyclic enkephalin analogs with pronounced deha opioid receptor agonist selectivity. Life Sci. 1983, 32, 2565-2569.

(145) Schwyzer, R. Molecular mechanism of opioid receptor selection. Biochemistry 1986, 25, 6335-6342.

(146) Bartosz-Bechowski, H.; Davis, P.; Slaninova, J.; Malatynska, E.; Stropova, D. et al. Cyclic enkephalin analogs that are hybrids of DPDPE-related peptides and metenkephalin-Arg-Gly-Leu; prohormone analogs that retain good potency and selectivity for delta opioid receptors. J. Peptide Res. 1999, 53, 329-336.

(147) Rajeswaran, W. G.; Hocart, S. J.; Murphy, W. A.; Taylor, J. E.; Coy, D. H. Highly potent and subtype selective ligands derived by N-Methyl scan of a somatostatin antagonists. /. Med. Chem. 2001, 44, 1305-1311.

(148) Miller, S. C.; Scanlan, T. S. Site-selective N-methylation of peptides on solid support. J. Am. Chem. Soc. 1997, 119, 2301-2302.

(149) Rodriguez, M.; Lignon, M. F.; Galas, M. C.; Amblard, M.; Martinez, J. Cyclic cholecystokinin analogs that are highly selective for rat and guinea pig central cholecystokinin receptors. Mo/. Pharmacol. 1990, 38, 333-341.

(150) Stevens, C.; Watanabe, R. J. Amino acid derivatives. I. carboallyloxy derivatives of alpha-amino acids. J. Am. Chem. Soc. 1950, 75.

(151) Grieco, P.; Gitu, P. M.; Hmby, V. J. Preparation of 'side-chain-to-side-chain' cyclic peptides by Allyl and Alloc strategy: potential for library synthesis. J. Pep. Res. 2001, 57, 250-256.

(152) Ahn, J. M.; Gitu, P. M.; Medeiros, M.; Swift, J. R.; Trivedi, D. et al. A new approach to search for the bioactive conformation of glucagon: positional cyclization scanning. J. Med. Chem. 2001, 44, 3109-3116.

(153) Story, S. C.; Aldrich, J. V. Side-product formation during cyclization with HBTU on a solid support. Int. J. Peptide Protein Res. 1994, 43, 292-296.

(154) Arttamangkul, S.; Ishmael, J. E.; Murray, T. F.; Grandy, D. K.; DeLander, G. E. et al. Synthesis and opioid activity of conformationally constrained dynorphin A 210 analogues. 2. Conformational constraint in the "address" sequence. J. Med. Chem. 1997, 40, 1211-1218.

(155) Arttamangkul, S.; Murray, T. F.; DeLander, G. E.; Aldrich, J. V. Synthesis and opioid activity of conformationally constrained dynorphin A analogues. 1. Conformational constraint in the "message" sequence. J. Med. Chem. 1995, 38, 2410-2417.

(156) Zhang, W.; Taylor, J. W. Efficient Sohd-Phase Synthesis of Peptides with Tripodal Side-Chain Bridges and Optimization of the Solvent Conditions for Solid-Phase Cyclizations. Tetrahedron Letters 1996, 37, 2173-2176.

(157) Taylor, J. W. The synthesis and study of side-chain lactam-bridged peptide. Biopolymers 2002, 66, 49-75.

(158) Gulyas, J.; Rivier, C.; Perrin, M.; Koerber, S. C.; Sutton, A. et al. Potent, structurally constrained agonists and competitive antagonists of corticotropin- releasing factor. Proc. Natl. Acad. Sci. U. S. A. 1995, 92, 10575-10579.

(159) Rodriguez, M.; Lignon, M. F.; Galas, M. C.; Amblard, M.; Martinez, J. CycHc cholecystokinin analogues that are highly selective for rat and guinea pig central cholecystokinin receptors. Mo/. Pharmacol. 1990, 38, 333-341.

(160) Rodriguez, M.; Amblard, M.; Galas, M. C.; Lignon, M. F.; Aumelas, A. et al. Synthesis of cyclic analogues of cholecystokinin highly selective for central receptors. Int. J. Pept. Protein Res. 1990, 35, 441-451.

(161) Charpentier, B.; Pelaprat, D.; Durieux, C.; Dor, A.; Reibaud, M. et al. Cyclic cholecystokinin analogs with high selectivity for central receptors. Proc. Natl. Acad Sci. U. S. A. 1988, 85, 1968-1972.

(162) Charpentier, B.; Marseigne, L; Dor, A.; Begue, D.; Durieux, C. et al. Cychc CCK- 8 analogs highly selective for central receptors. Pept: Chem. Biol, Proc. Am. Pept. Symp. 10th 1988, 608-609.

(163) Amblard, M.; Rodriguez, M.; Lignon, M.-F.; Galas, M.-C.; Bemad, N. et al. Modification of receptor selectivity and functional activity of cyclic cholecystokinin analogs. Eur. J. Med. Chem. 1998, 33, 171-180.

(164) Ahmed, S. I.; Wibowo, F.; Gembitsky, D. S.; Bozso, Z.; Murphy, R. F. et al. Importance of the C-terminal phenylalanine of gastrin for binding to the human CCK(2) receptor. J. Peptide Res. 2001, 58, 332-337. 211 (165) Kopin, A. S.; McBride, E. W.; Shcaffer, K.; Beinbom, M. CCK receptor polymorphisms: an illustrationof emerging themes in pharmacogenomics. Trends Pharm. Sci. 2000, 21, 346-353.

(166) DiMaio, J.; Schiller, P. W. A cycUc enkephalin with high in vitro opiate activity. Proc. Natl. Acad. Sci. U. S. A. 1980, 77, 7126-7126.

(167) Dinh, T. D.; Van Vranken, D. L. Control of ditryptophan cross-linking: dihydrotryptophan as a tryptophan precursor in peptide synthesis. J. Peptide Res., 53, 465-474.

(168) Hruby, V. J.; Kao, L.-F.; Montgomery Pettitt, B.; Karplus, M. The conformational properties of the delta opioid peptide cyclic [D-Pen^,D-Pen^] enkephalin in aqueous solution determined by NMR and energy minimization calculations. J. Am. Chem. Soc. 1988, 3351 - 3359.

(169) Mosberg, H. I.; Sobczyk-Kojiro, K.; Subramanian, P.; Crippen, G. M.; Ramalingam, K. et al. Combined use of stereospecific deuteration, NMR, distance geometry, and energy minimization for the conformational analysis of the highly delta-opioid receptor selective peptide [D-Pen^,D-Pen^] enkephalin. J. Am. Chem. Soc. 1990, 112, 822-829.

(170) Pellegrini, M.; Mierke, D. F. Molecular complex of cholecystokinin-8 and N- terminus of the cholecystokinin A receptor by NMR spectroscopy. Biochemistry 1999, 38, 14775-14783.

(171) Albrizio, S.; Carotenuto, A.; Fattorusso, C.; Moroder, L.; Picone, D. et al. Environmental mimic of receptor interaction: conformational analysis of CCK-15 in solution. J. Med. Chem. 2002, 45, 762-769.

(172) Hruby, V. J.; Kao, L.-F.; Himing, L. D.; Burks, T. F.; Pierce Chem. Co.: Rockford, IL, 1985; 487-490.

(173) Mosberg, H. I.; Schiller, P. W. IH N.M.R. investigation of conformational features of cyclic enkephalinamide analogs. Int. J. Pept. Protein Res. 1984, 23, 462-466.

(174) Qian, X.; Shenderovich, M. D.; Kover, K. E.; Davis, P.; Horvath, R. et al. Probing the stereochemical requirements for receptor recognition of opioid agonists through topographic modifications in position 1. J. Am. Chem. Soc. 1996, 118, 7280-7289. 212 (175) Nikiforovich, G. V.; Prakash, O. M.; Gehrig, C. A.; Hruby, V. J. Solution conformations of the peptide backbone for DPDPE and its beta-MePhe4- substituted analogs. Int. J. Pept. Protein Res. 1993, 41, 347-361.

(176) Kover, K. E.; Jiao, D.; Fang, S.; Hruby, V. J. Conformational properties of the unnatural amino acid beta-methylphenylalanine in a linear octapeptide system; correlations of 13C-NMR chemical shifts with the side-chain stereochemistry of these amino acid residues. J. Org. Chem. 1994, 59, 991 - 998.

(177) Pachler, K. G. R. Nuclear magnetic resonance (N.M.R.) study of some .alpha.- amino acids. I. Coupling constants in alkaline and acidic medium. Spectrochim. Acta 1963, 19, 2085-2092.

(178) Pachler, K. G. R. Nuclear magnetic resonance study of some .alpha.-amino acids. II. Rotational isomerism. Spectrochim. Acts 1964, 20, 581-587.

(179) Cung, M. T.; Marraud, M. Conformational dependence of the vicinal proton coupling constant for the C.alpha.-C.beta, bond in peptides. Biopolymers 1982, 21, 953-967.

(180) Kover, K. E.; Jiao, D.; Fang, S.; Hruby, V. J. Conformational analysis of four beta-methlphenylalanine stereoisomers in a bioactive peptide by z-filtered relay NMR spectroscopy. Magnetic Resonance in Chemistry 1993, 31, 1072-1076.

(181) De Leeuw, F. A.; Altona, C. Component vicinal coupling constants for calculating side-chain conformations in amino acids. Int. J. Pept. Protein Res. 1982, 20, 120- 125.

(182) Foumie-Zaluski, M. C.; Durieux, C.; Lux, B.; Belleney, J.; Pham, P. et al. Conformational analysis of cholecystokinin fragments CCK4, CCK5, and CCK6 by IH-NMR spectroscopy and fluorescence-transfer measurements. Biopolymers. 1985, 24, 1663-1681.

(183) Keire, D. A.; Solomon, T. E.; Reeve, J. R., Jr. NMR evidence for different conformations of the bioactive region of rat CCK-8 and CCK-58. Biochem. Biophys. Res. Commun. 2002, 293, 1014-1020.

(184) 1D6G. Brookhaven Protein Data Bank.

(185) Bax, A.; Davis, D. G. MLEV-17-based two-dimensional homonuclear magnetization transfer spectroscopy. J. Magn. Reson. 1985, 65, 355-360. 213 (186) Jeener, J.; Meyer, B. H.; Bachman, P.; Ernst, R. R. Investigation of exchange processes by two-dimensional NMR spectroscopy. J. Chem. Phys. 1979, 71, 4546-4553. 214 APPENDIX

Kaiser test

A few samples resin beads were placed in a small test tube and washed with methanol several times. One drop of each of the solutions were added: (A) 5% ninhydrin in ethanol, (B) 80% hquefied phenol in ethanol, and (C) 2% 0.001 M aq. Potassium cyanide in pyridine. The mixture was heated 5 minutes. Free amine is indicated by dark blue resins and solution.

TNBS test

A few samples resin beads were placed in a small test tube and washed with methanol several times. One drop of 10% DIPEA in DMF was added. One drop of a solution of 1% 2, 4, 6-trinitrobenzenesulfonic acid (TNBS) in DMF was added. The mixture was allowed to stand for 5 minutes. The beads were washed with methanol. A positive free amine is indicated by red beads.

Chloranil test

A few samples resin beads were placed in a small test tube and washed with methanol several times. One drop of 2% acetaldehyde in DMF was added. One drop of

2% /7-chloranil in DMF was added. The mixture was allowed to stand for 5 minutes.

The resins were washed with methanol. The presence of a secondary amine is indicated by a dark blue stain on the resin beads. 215

In vitro isolated tissue bioassays for opioid agonist

The in vitro tissue bioassays were performed by Peg Davis in Dr. Porreca's laboratory in the Department of Pharmacology, University of Arizona. IC50 values represent the mean of no less than four tissues. IC50 estimates, relative potency estimates, and their associated standard errors were determined by fitting the data to the Hill equation by a computerized nonlinear least-square method. The results were not corrected for the actual peptide content.

In the guinea pig isolated ileum (GPI) bioassay, male Hartley guinea pigs under anesthesia were sacrificed by decapitation and a non-terminal portion of the ileum was removed. The longitudinal muscle with myenteric plexus (LMMP) were carefully separated fi-om the circular muscle and were cut into strips. The tissues were tied to a gold chain with suture silk and mounted between platinum wire electrodes in 20 mL baths at a tension of 1 g containing 37°C oxygenated (95% O2, 5% CO2) Kreb's buffer

(118 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl2, 1.19 mM KH2PO4, 1.18 mM MgS04, 25 mM NaHCOa, and 11.48 mM glucose), and allowed to equilibrate for 15 min. The tissues were stimulated electrically (0.1 Hz, 0.4 msec duration) at supramaximal voltage.

Following an equilibration, the compound was added to the baths in 15 - 60 |j,L aliquots until maximum inhibition was observed. Percent inhibition was calculated by using the average contraction height for 1 min. preceding the addition of the compound divided by the contraction height 3 min. after exposure to the dose of the compound. Response to an 216 IC50 dose of PL-017 (10 nM) were measured to determine tissue integrity before compound testing begins.

In the mouse isolated vas deferens (MVD) assay, male ICR mice under ether anesthesia were sacrificed by cervical dislocation and the vasa deferentia was removed.

The tissues were tied to a gold chain with suture silk and mounted between platinum wire electrodes in 20 ml organ baths at a tension of 0.5 g and bathed in oxygenated (95% O2,

5% CO2) magnesium free Kreb's buffer at 37°C. They were stimulated electrically (0.1

Hz, single pulses, 2.0 msec duration) at supramaximal voltage. Following an equilibrium period, compounds were added to the bath cumulatively in volumes of 14 - 16 |j,L until maximum inhibition is reached. Response to an IC50 dose of DPDPE (10 nM) were measured to determine tissue integrity before compound testing begins.

CCK studies in the guinea pig isolated ileum

The in vitro tissue bioassays were performed by Peg Davis in Dr. Porreca's laboratory in the Department of Pharmacology, University of Arizona. Male Hartley guinea pigs under ether anesthesia were killed by decapitation and a non-terminal portion of the ileum removed. The longitudinal muscle with myenteric plexus (LMMP) was carefully separated from the circular muscle and cut into strips as described previously

(Porreca and Burks, 1983) . These tissues were tied to gold chains with suture silk and mounted between platinum wire electrodes in 20 ml organ baths at a tension of 1 g and 217 bathed in oxygenated (95:5 02'C02) Kreb's bicarbonate buffer at 31\ C. Tissues were stimulated electrically (0.1 Hz, 0.4 msec duration) at supramaximal voltage to stabilize base-line force and tissue health. Response to an IC50 dose of PL-017 (100 nM) was measured to determine tissue integrity before analog testing began. Following an equilibration period, tissues were challenged with KCl (67 mM) to determine initial maximal muscle contractility.

An initial non-cumulative CCK-8 dose-response curve was constructed using concentrations from 1 to 100 nM. The test compound was tested added to the bath in concentrations from 1 to 1000 nM. If no agonist activity was observed, 3 minutes later a dose of CCK-8 was added to determine the test compound's antagonist activity until a complete CCK-8 dose-response curve had again been reconstructed using a dose of antagonist which would seem to cause a three fold shift rightward. Tissues were again challenged with KCl to determine tissue changes during the assay. After thorough washing, electrical stimulation was again applied, tissue resiliency tested with 100 nM

PL-017, and an opioid dose-response curve was constructed with the test compound.

For CCK studies, contraction height was calculated as a percentage of the maximal KCl contraction and calculated as an A50 . For opioid studies, percentage inhibition was calculated using the average tissue contraction height for 1 min preceding 218 the addition of the agonist divided by the contraction height 3 min after exposure to the dose of agonist. IC50 values represent the mean of not less than 3 tissues. A50, IC50 and

Emax estimates were determined by computerized non-linear least-squares analysis

(MINSQ, Micromath

[^^SJGTP-y-S binding assays

The assays were performed by Dr. Shou-Wu Ma at the laboratories of Dr.

Josephine Lai at the Department of Pharmacology, University of Arizona. Cells expressing hDOR (or rMOR for mu-receptor studies) were incubated with increasing concentrations of the test compounds in the presence of 0.1 nM [ S]GTP-y-S (1000 -

1500 Ci/mmol, NEN, Boston, MA) in assay buffer (total volume of 1 mL, duplicate samples) as a measure of agonist-mediated G protein activation. After incubation (90 min, 30°C), the reaction was terminated by rapid filtration under vacuum through

Whatman GF/B glass fiber filters, followed by four washes with ice-cold 25 mM Tris/

120 mM NaCl, pH 7.4. Filters were pretreated with assay buffer prior to filtration to reduce nonspecific binding. Bound reactivity was measure by liquid scintillation spectrophotometry after an overnight extraction with EcoLite (ICN, Biomedicals, Costa

Mesa, CA) scintillation cocktail.

Phosphoinoside (PI) hydrolysis assays

The assays were performed by Dr. Shou-Wu Ma at the laboratories of Dr.

Josephine Lai at the Department of Pharmacology, University of Arizona. Transfected 219 cells were incubated with 0.5 mL IMDM with 0.2 |j.M [^H]myoinositol (final concentration) for 20-22 hrs at 37°C in the presence of humidified air (95% O2/ 5% CO2).

After the removal of media, the cells were incubated further for 1 hr at 37°C.

After removal of the media solution, 0.5 mL of IMDM was added and then LiCl stock solution to make a final concentration of 10 mM. Following a 10 min equilibration time, the test compound was added and the cells were incubated for 1 hr at at 37°C in the presence of humidified air (95% O2/ 5% CO2). The cells were placed in ice and the test compound and media were removed. The reactions were terminated with the addition of

0.5 mL cold methanol. The cells were scraped and transferred to a chloroform/water mixture (1 mL/0.5 mL). The sample was centrifuged at 2100 rpm for 10-15 min at 4°C.

-5 0.9 mL aliquots of the supernatant containing the water soluble [ H]-inositol phosphates was diluted with 2 mL of H2O, which was purified into anion exchange columns (CAGl-

X8, 100-200 mesh Bio-Rad laboratory). Following the loading of the sample, the

-3 columns were washed with 5 mL water to remove [ H]inositol precursor, followed by 5

•3 mL of 5 mM sodium tetraborate/ 60 mM sodium formate. The [ H] inositol phosphate was eluted with 2 mL 0.2 M ammonium formate and 0.1 M formic acid mixture into scintillation vials. 9 mL of Aquamix was added to each vial and radioactivity was measured by Uquid scintillation spectrometry. measured for reactivity. 220 Appendix. Rf values from thin layer chromatography of linear peptides

TLC' No. Sequence A B C 1 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NHi 0.86 0.77 0.26

2 Tyr-D-Ala-Gly-Trp-MeNle-Asp-Phe-NH2 0.79 0.61 0.10

3 Tyr-D-Nle-Gly-Trp-MeNle-Asp-Phe-NH2 0.86 0.77 0.25

4 T yr-D-V al-Gly-T rp-MeNle-Asp-Phe-NH2 0.81 0.76 0.26

5 Tyr-D-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 0.76 0.56 0.15

6 Tyr-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 0.58 0.27 0.05

7 T yr-Gly-Gly-Trp-MeNle-Asp-Phe -NH2 0.81 0.66 0.15

8 Asp-T yr-Nle-Gly-T rp-Nle-Asp-Phe-NH2 0.72 0.64 0.24

9 Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 0.83 0.67 0.14

10 Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2 0.84 0.69 0.17

11 T yr-D-Ala-Gly-T rp-Nle-Asp-Phe-NH2 0.76 0.55 0.09

12 T yr-D-Nle-Gly-T rp-Nle-Asp-Phe-NH2 0.84 0.69 0.23

13 Tyr-D-Phe-Gly-D-Trp-MeNle-Asp-Phe-NH2 0.84 0.69 0.10

14 Tyr-D-Ala-Gly-D-Trp-MeNle-Asp-Phe-NH2 0.76 0.54 0.10

15 T yr-D-V al-Gly-D-Trp-MeNle-Asp-Phe-NH2 0.82 0.73 0.31

16 Tyr-D-Nle-Gly-D-Trp-MeNle-Asp-Phe-NH2 0.84 0.75 0.24

17 Tyr-D-Ala-Gly-D-Trp-Nle-Asp-Phe-NH2 0.80 0.64 0.12

18 Tyr-D-Phe-Gly-D-Trp-Nle-Asp-Phe-NH2 0.85 0.74 0.10

19 T yr-D-Pro-Gly-D-Trp-Nle-Asp-Phe-NH2 0.71 0.48 0.07

20 Tyr-D-Nle-Gly-D-Trp-Nle-Asp-Phe-NH2 0.86 0.73 0.10 221 Rf values from thin layer chromatography of linear peptides (continued)

No. Sequence TLC" A B C 23 Tyr-D-Phe-Gly-D-Nal(2')-Nle-Asp-Phe-NH2 0.86 0.76 0.19

24 Tyr-D-Phe-Gly-Nal(l')-Nle-Asp-Phe-NH2 0.88 0.79 0.34

25 Tyr-D-Phe-Gly-Trp(5'Phe)-Nle-Asp-Phe-NH2 0.69 0.67 0.18

26 Tyr-D-Phe-Gly-D-Trp(5'Phe)-Nle-Asp-Phe-NH2 0.8 0.78 0.26

27 Tyr-D-Phe-Gly-Dht-NMeNleAsp-Phe-NH2 0.8 0.78 0.18

28 Tyr-D-Hph-Gly-Trp-NMeNleAsp-Phe-NH2 0.76 0.68 0.22

29 Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2 0 0 0

30 Tyr-c[D-Cys-Gly-Trp-D-Cys]-Asp-Phe-NH2 0.70 0.43 0

31 Tyr-c[D-Pen-Gly-Trp-Cys]-Asp-Phe-NH2 0.67 0.37 0.04

32 Tyr-c[D-Cys-Gly-Trp-Pen]-Asp-Phe-NH2 0.75 0.66 0.04

33 Tyr-c[D-Cys-Gly-D-Trp-Cys]-Asp-Phe-NH2 0.7 0.44 0.04

34 Tyr-c[D-Cys-Gly-D-Trp-Pen]-Asp-Phe-NH2 0.74 0.46 0.04

35 Tyr-c[Lys-Gly-Trp-Glu]-Asp-Phe-NH2 0.56 0.14 0

36 Tyr-c[D-Lys-Gly-Trp-Glu]-Asp-Phe-NH2 0.65 0.25 0

37 Tyr-c[D-Lys-Gly-Trp-D-Glu]-Asp-Phe-NH2 0.68 0.32 0

38 Tyr-c[D-Glu-Gly-Trp-Ly s] - Asp-Phe -NH2 0.62 0.24 0

39 Tyr-c[D-Lys-Gly-D-Trp-Glu]-Asp-Phe-NH2 0.61 0.24 0

40 Tyr-c[D-Om-Gly-Trp-Glu] - Asp-Phe-NHa 0.63 0.23 0

41 Tyr-c[D-0m-Gly-Trp-Asp]-Asp-Phe-NH2 0.62 0.24 0

42 Tyr-c[D-Dap-Gly-Trp-Glu]- Asp-Phe -NH2 0.66 0.28 0 Ill Rf values from thin layer chromatography of linear peptides (continued)

No. Sequence TLC A B C 43 Tyr-c[D-Dap-Gly-Trp-Asp]-Asp-Phe-NH2 0.69 0.29 0

"Rf values on thin-layer chromatograms of silica gel were observed in the following solvent systems: (A) 1-butanol/water/acetic acid (4:1:1) (B) ethyl acetate/1-butanol/water/acetic acid (5:3:1:1) (C) chloroform/methanol/water (7:1:2). 223 Physiochemical properties of CCK/opioid peptide analogues.

HPLC No. Sequence Retention time" k' 1 T yr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NHa 20.4 3.96

2 Tyr-D-Ala-Gly-Trp-MeNle-Asp-Phe-NH2 18.1 3.41

3 T yr-D-Nle-Gly-T rp-MeNle-Asp-Phe-NH2 19.7 3.80

4 T yr-D-V al-Gly-Trp-MeNle-Asp-Phe-NH2 18.9 3.61

5 Tyr-D-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 18.4 3.48

6 T yr-Pro-Gly-Trp-MeNle-Asp-Phe-NH2 17.7 3.32

7 T yr-Gly-Gly-Trp-MeNle-Asp-Phe-NH2 17.8 3.34

8 Asp-T yr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 17.9 3.36

9 Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 17.9 3.36

10 Tyr-D-Phe-Gly-Tip-Nle-Asp-Phe-NH2 19.1 3.65

11 T yr-D-Ala-Gly-Trp-Nle-Asp-Phe -NH2 16.1 2.93

12 T yr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 18.4 3.48

13 T yr-D-Phe-Gly-D-Trp-MeNle-Asp-Phe-NH2 20.1 3.90 224

Physiochemical properties of linear CCK/opioid peptide analogues.

HPLC Retenti No. Sequence on time" k' 14 Tyr-D-Ala-Gly-D-Trp-MeNle-Asp-Phe-NH2 18.0 3.39

15 Tyr-D-Val-Gly-D-Trp-MeNle-Asp-Phe-NH2 18.3 3.47

16 Tyr-D-Nle-Gly-D-Trp-MeNle-Asp-Phe-NH2 18.4 3.50

17 Tyr-D-Ala-Gly-D-Trp-Nle-Asp-Phe-NH2 16.7 3.06

18 Tyr-D-Phe-Gly-D-Trp-Nle-Asp-Phe-NH2 18.9 3.62

19 Tyr-D-Pro-Gly-D-Trp-Nle-Asp-Phe-NH2 16.8 3.11

20 Tyr-D-Nle-Gly-D-Trp-Nle-Asp-Phe-NH2 18.7 3.56

22 Tyr-D-Phe-Gly-Nal(2')-NleAsp-Phe-NH2 20.7 4.04

23 Tyr-D-Phe-Gly-D-Nal(2')-Nle-Asp-Phe-NH2 21.2 4.17

24 Tyr-D-Phe-Gly-Nal(l')-Nle-Asp-Phe-NH2 20.3 3.95

25 T yr-D-Phe-Gly-Trp(5' Phe)-Nle-Asp-Phe-NH2 22.1 4.39

26 Tyr-D-Phe-Gly-D-Trp(5'Phe)-Nle-Asp-Phe-NH2 22.6 4.51

27 Tyr-D-Phe-Gly-Dht-NMeNleAsp-Phe-NH2 15.5 2.79 225

Physiochemical properties of linear CCK/opioid peptide analogues (continued).

HPLC No. Sequence Retention time^ k' 28 Tyr-D-Hph-Gly-Trp-NMeNleAsp-Phe-NH2 19.9 3.85

29 T)T-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2 13.7 2.34

30 Tyr-c[D-Cys-Gly-Trp-D-Cys]-Asp-Phe-NH2 13.9 2.39

31 T yr-c[D-Pen-Gly-Trp-Cys] -Asp-Phe-NH2 14.5 2.53

32 Tyr-c[D-Cys-Gly-Trp-Pen]-Asp-Phe-NH2 13.8 2.36

33 Tyr-c[D-Cys-Gly-D-Trp-Cys]-Asp-Phe-NH2 13.3 2.24

34 Tyr-c[D-Cys-Gly-D-Trp-Pen]-Asp-Phe-NH2 13.4 2.26

35 Tyr-c[Lys-Gly-Trp-Glu]-Asp-Phe-NH2 11.8 1.90

36 Tyr-c[D-Lys-Gly-Trp-Glu]-Asp-Phe-NH2 11.9 1.91

37 Tyr-c[D-Lys-Gly-Trp-D-Glu]-Asp-Phe-NH2 11.8 1.87

38 T yr-c[D-Glu-Gly-Trp-Lys]- Asp-Phe-NH2 12.6 2.08

39 Tyr-c[D-Lys-Gly-D-Trp-Glu]-Asp-Phe-NH2 12.4 1.1

40 Tyr-c[D-0m-Gly-Trp-Glu]-Asp-Phe-NH2 11.8 1.8 226

Physiochemical properties of linear CCK/opioid peptide analogues.

HPLC No. Sequence Retention time^ k' 41 Tyr-c[D-0m-Gly-Trp-Asp]-Asp-Phe-NH2 n/d n/d

42 Tyr-c[D-Dap-Gly-Trp-Glu]-Asp-Phe-NH2 13.3 2.23

43 Tyr-c[D-Dap-Gly-Trp-Asp]-Asp-Phe-NH2 12.4 2.02

''HPLC k' = [(peptide retention time - solvent retention time)/ solvent retention time] in a solvent system of 10% ACN in 0.1% TFA and a gradient 10 to 90% ACN over 40 min. An analytical CI8 column was used with a flow rate of 1 mL/min for compounds 1-34. 227 Amino acid analysis of linear CCK/opioid peptide analogues

No. Tyr Gly Trp Asx Phe Ala Pro Val

1 0.84(1) 0.94(1) 1.0(1) 1.62(2)* — —

2 0.98(1) 1.0(1) - 1.1(1) 1(1) 1.0(1)* ~

3 0.98(1) 1.1(1) - 1.1(1) 1.0(1) 1.0(1) ~

4 0.67(1) 0.84(1) 1.0 0.83(1) ~ 0.6(1)*

5 0.98(1) 1.06(1) 1.26(1) 1.0(1) ~ 1.0(1)*

6 0.96(1) 0.92(1) 1.0(1) 1.0(1) ~ ~

7 0.89(1) 1.86(2) 1.1(1) 1.0(1) ~ ~

8 0.86(1) 1.9(1) - 1.92(2) 1.0(1) ~ --

9 0.87(1) 0.89(1) 1.1(10 1.0(1) — --

10 0.85(1) 0.92(1) 1.0(1) 1.7(2)* ~ ~

11 1.0(1) 1.0(1) - 1.2(1) 1.0(1) 1.0(1)* —

12 0.93(1) 1.1(1) - 1.1(1) 1.0(1) ~ ~

13 0.86(1) 0.95(1) 1.0(1) 1.63(2) ~ --

14 0.98(1) 1.0(1) - 1.1(1) 1.0(1) 1.0(1) --

15 0.69(1) 0.83(1) 1.0(1) 0.91(1) — 0.7(1)* 16 0.78(1) 0.87(1) 1.0(1) 1.0(1)

17 0.78(1) 0.82(1) 1.0(1) 0.75(1) 0.76(1)* ~

18 0.71(1) 0.82(1) 1.0(1) 1.6(2)* ~ ~

19 0.87(1) 1.0(1) - 1.2(1) 1.0(1) — 0.99(1)*

20 0.88(1) 1.1(1) - 1.2(1) 1.0(1) — ~ Nle shows but was not ran against a standard. MeNle does not elute. Trp is degraded during hydrolysis. Asterisk (*) indicates the D-isomer. Asx or Phe were used as standards for integrations. 228

Amino acid analysis (continued)

No. Tyr Gly Trp Asx Phe 22 0.82(1) 0.86(1) 1.0(1) 1.99(2)* 23 0.96(1) 1.2(1) - 1.0(1) 2.65(2)

24 0.73(1) 0.93(1) 1.0(1) 1.86(2) 25 0.79(1) 0.84(1) 1.0(1) 1.76(2) 26 0.74(1) 0.87(1) 1.0(1) 1.72(2) 27 0.89(1) 0.85(1) 1.0(1) 1.8(1) 28 0.82(1) 0.81(2) 1.0(1) 0.85(1)

Nle shows but was not ran against a standard. MeNle does not elute. Trp is degraded during acid hydrolysis. Asterisk (*) indicates the D-isomer. Asx was used to compare integrations. 229 Amino acid analysis (continued)

No. Tyr Gly Trp Asx Phe 29 0.488(1) 0.91(1) 0.93(1) 1.0(1) 30 0.49(1) 0.92(1) 1.0(1) 1.0(1) 31 0.67(1) 0.94(1) 1.1(1) 1.0(1) 32 0.48(1) 1.0(1) - 0.7(1) 1.0(1) 33 0.35(1) 0.9(1) - 0.54(1) 1.0(1) 34 0.58(1) 1.1(1) - 1.0(1) 1.0(1)

Cys and D-Cys showed in chromatogram but it was not tested against a standard and it was partially degraded by acid hydrolysis. Trp is degraded during acid hydrolysis. Low Tyr content might have been affected by the Cys degradation. 230 Amino acid analysis of cyclic lactam analogues of CCK/opioid peptide

No. Tyr Gly Trp Asx Phe Lys Glx

35 0.75(1) 0.97(1) 1.0(1) 1.0(1) 0.86(1) 0.99(1)

36 0.74(1) 0.92(1) 1.0(1) 1.0(1) 0.93(1)* 0.77(1)

37 0.72(1) 0.94(1) 1.2(1) 1.0(1) 0.73(1)* 1.0(1)*

38 0.91(1) 1.0(1) - 1.1(1) 1.0(1) 0.99(1) 0.9(1)*

39 0.86(1) 1.0(1) - 1.1(1) 1.0(1) 0.87(1)* 0.94(1)

40 n/d

41 n/d

42 0.93(1) 1.1(1) - 1.1(1) 1.0(1) — 1.1(1)

43 0.86(1) 0.94(1) 2.0(2) 1.0(1) ~ ~

Trp is degraded during hydrolysis. Asterisk (*) indicates the D-isomer. Phe was used to compare integrations, n/d, not yet determined. High Resolution Mass Spectrometry for CCK/opioid peptides

Molecular HR-MS No. Sequence Formula Calcd Obsd 1 Tyr-D-Phe-Gly-Trp-MeNle-Asp-Phe-NHa CsiHeiNgOioNa 982.4434 982.4449

2 Tyr-D-Ala-Gly-Trp-MeNle-Asp-Phe-NHi C45H58N9O10 884.4301 884.4330

3 T yr-D-Nle-Gly-Trp-MeNle-Asp-Phe-NHs C48H63N90ioNa 948.4590 948.4601

4 Tyr-D-Val-Gly-Trp-MeNle-Asp-Phe-NH2 C47H6iN90ioNa 934.4434 934.4433

5 Tyr-D-Pro-Gly-Trp-MeNle-Asp-Phe-NHi C47H59N90ioNa 932.4277 932.4230

6 T yr-Pro-Gly-Trp-MeNle-Asp-Phe-NHa C47H60N9O10 910.4458 910.4438

7 Tyr-Gly-Gly-Trp-MeNle-Asp-Phe-NH2 C44H55N90ioNa 892.3964 892.3963

8 Asp-Tyr-Nle-Gly-Trp-Nle-Asp-Phe-NH2 CsiHeeNioOiaNa 1027.4889 1027.4590

9 Tyr-Nle-Gly- Trp-Nle-Asp -Phe-NH2 C47H6iN90ioNa 934.4434 934.4423

10 Tyr-D-Phe-Gly-Trp -Nle-Asp-Plie-NH2 C50H60N9O10 946.4458 946.4410

11 Tyr-D-Ala-Gly-Trp-Nle-Asp-Phe-NH2 C44H56N9O10 870.4145 870.4205

12 Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2 C47H62N9O10 912.4615 912.4656

13 Tyr-D-Phe-Gly-D-Trp-MeNle-Asp-Phe-NH2 C51H62N9O10 960.4615 960.4621 High Resolution Mass Spectrometry for CCK/opioid peptides (continued)

Molecular HR-MS No. Sequence Formula Calcd Obsd

15 Tyr-D-Val-Gly-D-Trp-MeNle-Asp-Phe-NHi C45H58N9O10 912.4658 912.4610

16 Tyr-D-Nle-Gly-D-Trp-MeNle-Asp-Phe-NH2 C48H63N9O10 926.4776 926.4750

17 Tyr-D-Ala-Gly-D-Trp-Nle-Asp-Phe-NH2 C44H56N9O10 870.4150 870.4125

18 T yr-D-Phe-Gly-D-T rp-Nle-Asp-Phe-NH2 C50H60N9O10 946.4458 946.4465

19 T yr-D-Pro-Gly-D-T rp-Nle-Asp-Phe-NHi C46H58N9O10 896.4302 896.4293

20 Tyr-D-Nle-Gly-D-Trp-Nle-Asp-Phe-NHi C47H62N9O10 912.4615 912.4634

22 Tyr-D-Phe-Gly-Nal(2')-NleAsp-Phe-NH2 C52H61N8O10 957.4506 957.4487

23 Tyr-D-Phe-Gly-D-Nal(2')-Nle-Asp-Phe-NH2 C52H6oN80ioNa 979.4325 979.4315

24 Tyr-D-Phe-Gly-Nal(l')-Nle-Asp-Phe-NH2 C52H6oN80ioNa 979.4325 979.4323

25 T yr-D-Phe-Gly-T rp(5' Phe)-Nle-Asp-Phe-NH2 C56H64N9O10 1022.4771 1022.4760

26 Tyr-D-Phe-Gly-D-Trp(5'Phe)-Nle-Asp-Phe-NH2 C56H64N9O10 1022.4771 1022.4799

toU) to High Resolution Mass Spectrometry for CCK/opioid peptides (continued)

Molecular HR-MS No. Sequence Formula Calcd Obsd 27 Tyr-D-Phe-Gly-Dht-NMeNleAsp-Phe-NHa C51H64N9O10 962.4771 962.4811

28 T yr-D-Hph-Gly-Trp-NMeNleAsp-Phe-NH2 C51H61N9O10 960.4615 960.4322

29 Tyr-c[D-Cys-Gly-Trp-Cys] - Asp-Phe-NHi C4iH47N90ioS2Na 912.2784 912.2830

30 Tyr-c[D-Cys-Gly-Trp-D-Cys] -Asp-Phe -NH2 C4iH47N90ioS2Na 912.2784 912.2834

31 T yr-c[D-Pen-Gly-Trp-Cys] - Asp-Phe -NH2 C43H52N9O10S2 918.3278 918.3216

32 T yr-c[D-Cys-Gly-Trp-Pen] - Asp-Phe-NHa C43H5iN90ioS2Na 940.3097 940.3103

33 Tyr-c[D-Cys-Gly-D-Trp-Cys] - Asp-Phe-NH2 C4iH47N90ioS2Na 912.2784 912.2769

34 Tyr-c[D-Cys-Gly-D-Trp-Pen] - Asp-Phe-NH2 C43H5iN90ioS2Na 940.3097 940.3072

35 Tyr-c[Lys-Gly-Trp-Glu] - Asp-Phe-NH2 C46H56NioOiiNa 947.4023 947.3983

36 Tyr-c[D-Lys-Gly-Trp-Glu]-Asp-Phe-NH2 C46H57N10O11 925.4203 925.4226

37 Tyr-c[D-Lys-Gly-Trp-D-Glu]-Asp-Phe-NH2 C46H57N10O11 925.4203 925.4241

38 Tyr-c[D-Glu-Gly-Trp-Lys]-Asp-Phe-NH2 C46H57N10O11 925.4203 925.4185

toU) U) High Resolution Mass Spectrometry for CCK/opioid peptides (continued)

Molecular HR-MS No. Sequence Formula Calcd Obsd 40 Tyr-c[D-0m-Gly-Trp-Glu]-Asp-Phe-NH2 C45H55N10O11 911.4047 911.4082

41 T yr-c[D-Om-Gly-Trp-Asp] - Asp-Phe-NHi C44H53N10O11 897.3890 897.3848

42 T yr-c[D-Dap-Gly-Tip-Glu] - Asp-Phe-NH2 C43H5iNioOn 883.3734 883.3776

43 Tyr-c[D-Dap-Gly-Trp-Asp] - Asp-Phe-NHi C42H48N10O11 869.3756 869.3774

^High-resolution mass spectra were determined by MALDI-TO-MS or FAB-MS methods. The calculated mass was determined from [M+H] or [M+Na], as notated by an asterisk (*).

K) w RSA203: H-Tyr-D-AU-Gly-Trp-NMeNle-Asp-Phe-NH2

_AJui u Jil

•I I ""I I I 'I I I I I ""I I I"" I I I I I I I I' 8.6 8.5 8.4 8.3 8.2 8.1 8.0 7.9 ppm 5.0 4.9 4.8 4.7 4.6 4.5 4.4 4.3 4.2 4.1 ppii

JUL J,iL '^1 jij'

• I ' • ' • I • ' • ' I ' • • ' 1 • ' ' ' I • • • ' I • • ' • 1 ' • I ' ' ' ' I • ' • • I • • ' ' I ' • • ' I ' ' ' • I ' ' ' ' I ' • ' • I • • ' • I • • • ' I ' ' • • I • ' ' • I • • • • I ' ' K) 10.5 10,0 9.5 9.0 8.5 8.0 7.5 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2,0 1.5 1.0 ppm U) RSA204: H-Tyr-D-Nle-Gly-Trp-NMeNle-Asp-Phe-NH2

4.5 4.4 4.3 4.2 4.1 4.0 ppt 1

ppm

I • I I I I I • I I • • ' ' I ' ' • ' I ' ' ' • I ' • • ' I • • • • I ' • ' • I • • • ' I • ' ' ' I—• • • I • • • • I ' ' • ' I • • • ' I • ' • • I ' ' 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm toU) On Til ujdd 01 SI 01 SZ 0£ Se Ot' OS SS 09 S'Q 07, SL 08 S'S 06 56 001 SOI • I • • • • I • • • • I • • • • ' • • • • I • • • • I • • • • I • • • • ' • • • • I • • • I • • • • I • • I • • I • ' • I • I • . • I . . . • I • . . I . . . • I • • • • I . I

r-\

uidd 8X o'8 rs

uidd ge 0> O'S

CHN-sMd-ds:01ZVS'H RS A211: H-Tyr-Gly-Gly-Tip-NMeNle-Asp-Phe-NH2

'"I" I I'" ""I"' 4.': 4.2 4.1 4.0 3.9 3.8 ppm

9.2 9.0 8.8 8.6 8.4 8.2 8.0 7.8 ppm

JIm

' I'''' I' I - • I • • • I • I • I • • • • I • • • • I • • • • I • • I • - • • I • • • • I • • • • I • • • • I • I • • I. • •. I.... I.... I.... I.... I. 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm

U)K)

RSA218: H-Tyr-Pro-Gly-Trp-NMeNle-Asp-Phe-NH2

5.0 4.7 4.6 4.5

npm

I• • • ' • • I• • '• • I •• I• ••• I• • • I • I• I I • I• • • I'• '' I • '•' I ' ' • 'I ''' • I' • '• I• • •'I•• • 'I• '' ' I ' 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm

to 4^ RSA500; H-Asp-Tyr-D-Phe-Gly-Trp-Nle-Asp-Phe-NH2

i»H'ii»iTim|niiminiriiri'>'ifrpiiiminiiwniif|iiirmt»|ininiii|iiiwmniiiiimi[iinmmi

4.6 4.5 4.4 4.3 4.2 4.1 4.0 3.9 3.8 3.7 ppm

•'-n- —T- • 1 • • • • I • • • 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 ppm

to (O4^ CO

uidd o-I 5 IT • • O'Z- t - •SZ I -IO e••••»•••• &•£ OP I O 1S S'SI 0'9 S'9 O'LI •••I S'L •• -0'8 I - - •S*8 I 0*6 9'6 0*01- I SOT'

oidd LL rL 08 18 S'8 9'8

(udd 0>

3HK-3i1d-'lsv-aW-dil-'

L ""I I I I I'"' '"I I"" • "I 'I"" "T" ""I""""'! I I"" "T" 8.8 8.7 8.6 8.5 8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 ppm 4.7 4.6 4.5 4.4 4.3 i.2 4.1 3.9 3.8 ppm

A> iuUULJ'

I --n- • 1 • 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4,5 4,0 3.5 3.0 2.5 2.0 1.5 ppm to 4^ RSA503: H-Tyr-D-Ala-GIy-Ttp-Nie-Asp-Phe-NH2

I I • I • I • • • • I • • • • I ' ' • • I ' • • • I • ' • • I • • • • I • 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0

N) RSA504: H-Tyr-D-Nle-Gly-Trp-Nle-Asp-Phe-NH2

4.5 4.4 4.3 4.2 4.1 3.9 3.8 3.7 ppm

8.5 8.4 8.3 8,2 8.1 8.0 7.9 7.8 7.7 ppm

I ' I • • • I I ' • I • • • I • • I • I • • • I • • I • I • I • • I • F - I • • ' ' I ' ' ' ' I ' ' ' ' 1 ' ' ' ' I 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7,0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 K) -1^ RSA601: H-Tyr-D-Phe-Gly-D-Trp-NMeNle-Asp-Phe-NH2

I I r I I I I I I I I I "T" —I—'—\—^—\—^—I—^—\—^—I—•—r 8.6 8,5 8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 7.6 ppm 5.0 4.8 4.6 4.4 4.2 4.0 ppm

JU u J I • • I•• • • I• • • I - •I••'•I••'•I'••'I••• • I• I I • I • • • 1 •• • I • • I • 1 • • • I • I • • •• I ' ' ' ' I ' ' ' ' I ' 11.0 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm K) RSA602: H-Tyr-D-Ala-Gly-D-Trp-NMeNle-Asp-Phe-NH2

ppm

• I I I • • I • • • • I • • • • I • • I • • I • • ' ' I ' • ' ' I • • ' ' I ' ' ' • I ' ' ' • I • • ' ' I ' ' ' ' I • • ' ' I ' ' • ' I ' • • ' I ' • • ' I • • ' ' I ' ' ' ' I ' ' • ' I • 11.0 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 I.O ppm K)

00 RSA603: H-Tyr-D-Val-Gly-D-Trp-NMeNle-Asp-Phe-NH2

I • • • 1 • • I • • I • • • • I • I • I • • • I • • ' ' 1 ' ' ' ' I' ' ' ' I ' I ' • • • I • ' • ' I • • • • I • ' • ' I • ' • • I • • • • I • • ' • I ' • • • I • • • • I • 11.0 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm

K) VO RSA604; H-Tyr-D-Nle-Gly-D-Trp-NMeNle-Asp-Phe-NH2 251

S a

q

U

oo\ RSA622: H-Tyr-D-Phe-Qly-D-Trp-Nle-Asp-Phe-NI

ppm

8,8 8.7 8.6 8.5 8.4 8.3 8.2 8.1 8.0

LJ

r T T 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 l.O ppm

tsJ KJ RSA623: H-Tyr-D-Pro-Gly-D-Tri>-Nte-Asp-Phe-NH2 RSA624: H-Tyr-D-Nie-Gly-D-Ttp-Nle-Asp-Phe-NH2

4.6 4.5 4.4 4.3 4.2 4.1 4.0 3.9 3.8 3,7 3.6 ppm

8.6 8.5 8.4 8.3 8.2 8.1 8.0 7.9 l.i 7.7 ppm

I • • • • I • • • I • I • ' ' ' I • ' ' ' I ' • • ' I • • • ' I • ' • ' I • ' ' • I ' • ' • I ' • ' • I ' ' • ' I ' • • ' I ' • ' ' I ' ' ' ' I ' ' ' ' I ' ' • ' I ' • • • I ' • ' • I • • • ' I • • ' • I 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm RSA801: H-Tyr-D-Phe-Gly-Nal(2')-Nle-Asp-Phe-NH2

4.2 4.1 4.0 3.9 3# 3.7 ppm

7.7 ppm

MJl I • •'' I''' • I'''' I •''' I'' •' I • • • • I •' • • I'' • • I''' • I'' •' I'''' I •' • • I • •'' I •' • • I •' •' I'' • • 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 ppm

to RSA802; H-Tyr-D-Phe-Gly-D-Nal(2')-Nle-A^-Phe-NH2

4.8 4.7 4.6 4.5 4.4 4.3 4.2 4.1 4.0 3.9 ppm

8.8 8.7 8.6 8.5 8.4 8.3

I • • I • • • • I • • • I • I ' • ' ' I • ' • • I • • • ' I • • • ' I • ' ' • I ' • • • I ' ' ' • I • ' ' ' I • • • • I • • • ' I ' ' ' ' I • • • ' I • ' • • I 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5

K> RSA803: H-Tyr-D-Phe-Gly-Nal(r )-Nle-Asp-Phe-NH2 aaJUU jvAhJ^ .1, I I I I 4.7 4.6 4.5 4.4 4.3 4.2 4.0 3.91 3.8 3.7 3.6 ppm

8.8 8.7 8.6 8.5 8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 7.6 7.5 ppm

JIU

I I • • I • • I • • • I • • • • I • • I • • • • I ' • • ' I ' • ' • I • • • • I • • • • I • • ' • I • • • • I • • • • I • ' • • I ' ' • ' I 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 i

(O LTl -J RSA804: H-Tyr-D-Phe-Gly-Dht-NMeNle-Asp-Phe-NH2

4.90 4.85 4.80 4.75 4.70 4.65 4.60 4.55 4.50 4.45 4.40 4.35 ppm

8.8 8.7 8.6 8.5 8.4 8,3 8.2 8.0 7.' 7.8 ppm

I • •I I...... I... I.... I.... I.... 11... I.... I.... I.... I.... I.... I.... II.... I I,I....I 11.0 10.5 10.0 9.5 9.0 8.5 8,0 7.5 7.0 6.5 6.0 5,5 5.0 4.5 4.0 3.5 3.0 2,5 2,0 1,5

K)

00 ! A806: H-Tyr-D-Phe-Gly-D-Tip(5' Phe)-Nle-Asp-Phe-Nl

iinpfiin'iiiHimim|n'nrmi'p 8.8 8.7 8.6 8.5 8.4 8.3 8.: 8.1 8.0 7.9 7.8 7.7

"T— • I • 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 260 RS A102c: Tyr-c[DCys-Gly-Trp-Cys]-Asp-Phe-NH2

J L /U iil —T—^—I——I—'—I—^—I—^—I—•—I—^—I—•—I—^—I—'—I 9.4 9.2 9.0 8.8 8.6 8.4 8.2 8,0 7.8 ppm

LJIiJliJ

• I••••I • II••••('•••I••'•I'''' I • . I • • • • I ' • ' • I • •"T' •n • 11.0 10.5 10.0 9.5 9.0 8.5 8.0 7.5 6.5 6.0 5.5 5.0 4.5 RSA104C; H-Tyr-c[DCys-Gly-Ttp-Pen]-Asp-Phe-NH2

8.6 8.4 8.2 8.0 7.8 ppm

I • • 1 •• • • I • • I • • • • I • I • • 1 ••I • • I • • • I•I I I • • I I • • I • I • I • • I • • I • I ••• 1 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm

K) a\ to RSA103C:H-Tyr-c[D-Pen-Gly-Trp-Cys]-Asp-Phe-NH2

8.8 8.6 8.4 8.2 8.0 7.8 7.6 ppm 4.60 4.55 4.50 4.45 4.40 4,35 4.30 j jm

/ULLLJ I • • I •• • • I • • I •' • I' •'' I •'' • I'' •' I • •' • I' • •'I•'''I'''•I''•'I''•'I''•'I•• • • I'' • •I'•' 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5,0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 ppm

U) RSA121C: H-Tyr-c[DCys-Gly-DTrp-Cys]-Asp-Phe~NH2

5.3 5.2 5.1 5.0 4.9 4.8 4.7 4.6 4.5 4.4 4.3 4.2 4.1 4.0 3.9 ppm

8.8 8.6 8.4 8.2 8.0 7.8 ppm

10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 ppm

to 4^

RSA405: H-Tyr-c|Lys-Gly-Trp-Glu]-Asp-Phe-NH2

4.6 4.5 4.4 4.3 4.2 4.1 3.9 3. i 3.7 3.6 ppm

LU • ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' I ' ' ' ' I ' ' • II.... I .... II....III , 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 ppm

K) o^On RSA402: H-Tyr-c[D-Lys-Gly-Trp-Glu]-Asp-Phe-NH2

4.6 4.5 4.4 4.3 4.2 4.0 ppm

8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 ppm

I I '' I • • • 'I • • • 'I • '• • )'• • • I'' • • I' ' ' ' I • • '• I' • • •I • '• • I I.... I .... I .... I .... I .... I ...... I .... I 10.5 10.0 9.5 9.0 8.5 8.0 7,5 7.0 6.5 6,0 5,5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 pi

RSA406: H-Try-c[DLys-Gly-Trp-D-Glu]-Asp-Phe-NH2

4.6 4.5 4.4 4.3 4.2 4.1 4.0 3.9 ppm

8.5 8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 7.6 ppm

I • • I• • • • I•• • -I '' ''I' ' ' ' I •• •'I• '••I '• • •I • ' ' •I '' • 'I • '' ' I • '• 'I ' • • 'I ' • • • I• • • • I' ' • •I • ' ''I•'• • I'' • •I• • • 'I 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2,5 2.0 1.5

K) VOOn RSA407: H-Tyr-c[D-Glu-Gly-Trp-Lys]-Asp-Phe-NH2

4.0 3.9 3.8 ppm

I I I [HII I I [ |i I I |i I I I 8.8 8.7 8.6 8.5 8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 7.6 7.5 7.4 7.3 ppm

I • • I • • I • • I • I • • I - • • • I • • • • I • • . • I • . • . I • • • • I . I• • I • • • I• • I• • • I • • • I•• • • I' ' ' 'I ' ' ' • I' • ' 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 ppm

O-4 RSA408: H-Tyr-c[D-Orn-Giy-Trp-Asp]-Asp-Phe-NH2

5.0 4.8 4.6 4.4 4.2 4.0 3.8 ppm

8.9 8.8 8.7 8.6 8.5 8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 7.6 7.5 ppm

r • I • • I • • • • I • • • • I• •• • I• • •• 1 • • • • I' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' • 'I ' ' ' ' I '' ' ' I ' • '' I • '' ' I ' ' ' • I '' • 'I '' ' ' I • • • •I ' • '• I' 11.0 10.5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 ppm

K> --J RSA409: H-Tyr-c[D-Dap-Gly-Trp-Glu]-Asp-Phe-NH2

4.5 4.4 4.3 4.2 4.1 4.0 ppm

ppm

T 10,5 10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 ppm

to to 273

L 6 & d 6 tx

(No O)

q O m* *o

p

wo •*1^

o

ffi

a. o «A 00

5 d RSAlOlc: Tyr-c[DCys-Gly-Trp-Cys]-Asp-Phe-NH2T=7Q0l<:

I —T —I 1 9.0 8.8 8.6 8.4 8.2 8.0 7.8 7.6 ppm

jy

• I ' • • ' I • • • • I • • • • I • • • • I• • • • I• '• • I• • • 'I ' 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 RSAl01: H-Tyr-c[DCys-Gly-Trp-Cys]-Asp-Phe-NH2 T=293K

aJ. ""1 1 1"" I I [•••••••..| I I .| I |. I |. "I"' -T^T— 8.9 8.8 8.7 8.6 8.5 8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 7.6 ppm 4.8 4.7 4.6 4.5 4.4 4.3 ppm

i)LJmJ' il ii

'I • • '' I • • '• I• ' • ' I• • • • I• • • • I' • • • I• • • 'I ' • • • I' • • I• • • I • I • 1 • • • I • • I• • I• '• • I• 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm Richard Agnes sample: H-Tyr-ctDCys-Gly-Trp-Cys)-Asp-Phe-NH2 y.btnlVl, O.bmL UMSU IH 600MHz lemp=298K June 5, 2002

"T" -r- -T" -7- —I— 8.8 8.6 8.4 8.0 7.8 7.6 ppm

i I .!•

I• • • • I' ' ' ' I • '' • I• • • • I• • • • I' • • • I• • • • I 'T-' 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4,0 3.5 3.0 2.5 2.0 1.5 ppm to 0\ RSAlOl: Tyr-c[DCys-Gly-Trp~Cys]-Asp-Phe-NH2 T=303K

•i 1 1 1 1 1 1 1 1--" 4.8 4.7 4.6 4.5 4.4 4.3 4.2 ppm

JiJ Vjm/

I I I I I I I I I I I I I I 8.8 8.7 8.6 8.5 8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 7.6 ppm

—T' —T"' T-pn. —r-' r' T"' —T— 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 ppm to

-J 278

e a£1. s &

rr _ p

. ^ 3

• vdo

'sd

JSu

& < I.I Ti 9^ U B ot

f- c • 00 ' OS VD£ki 1 TOCSY experiment for RSAIOIC (H-Tyr-c[D-Cys-Gly-Trp-Cys]-Asp-Phe-NH2) N) O ppm

i '•••1 1 1" 7.9 7.8 7.3 7.2 7.1 7.0 ppm

Region of TOCSY spectra for RSAlOlc (H-Tyr-cp-Cys-Gly-Trp-CysJ-Asp-Phe-NHi) NOESY experiment spectra for RSAlOlc (H-T>T-c[D-Cys-Gly.Trp-Cys]-Asp-Phe-NH,) Temperature dependence of amide proton chemical shifts of RSAlOlc.

Temperature dependence of D-Cys NH

8.63 E a 8.62 Q. 8.61 8.6 £ Series 1 (0 8.59 • Linear (Series1) "5o 8.58 8.57 Eo y = -0.0029X + 9.454 £ 8.56 o 8.55 = 0.9989 285 290 295 300 305 310 315 Temperature (K)

Temperature dependence of Gly NH

8.8 EQ. Q. 8.75

8.7 • Series 1 £m Linear (Series 1) ure 8.65 Ea> 8.6 £ y = -0.0075X + 10.947 O 8.55 R2 = 0.9684 285 290 295 300 305 310 315 Temperature (K) 283 Temperature dependence of amide proton chemical shifts of RSAlOlc (continued).

Temperature dependence of Trp NH

8.82 Ea. 8.8 a. 8.78 8.76 Series 1 8.74 • Linear (Seriesi) leu 8.72 8.7 1a> £ 8.68 y = -0.0056X + 10.398 o 8.66 = 0.9348 285 290 295 300 305 310 315 Temperature (K)

Temperature dependence of Cys NH

7.52 E a. 7.52 a. 7.52 • Seriesi £u 7.52 « 7.52 • Linear (Seriesi) u 7.52 E 0) 7.52 y = 1E-14X + 7.52 7.52 = 7E-13 285 290 295 300 305 310 315 Temperature (K) 284 Temperature dependence of amide proton chemical shifts of RSAlOlc (continued).

Tempertaure dependence of Asp NH

8.44 E a Q. 8.42 8.4 r Series1 (0 8.38 re • Linear (Series 1) o 8.36 E 0) 8.34 y = -0.0045X + 9.7374 £ O 8.32 R2 = 0.9995 285 290 295 300 305 310 315 Temperature (K)

Temperature dependence of Phe^ NH

8.05 E a. a. 8 • Series 1 v> 7.95 « Linear (Series 1) o E 7.9 d) y = -0.0071X + 10.083 £ o 7.85 R2 = 0.9981 285 290 295 300 305 310 315 Temperature (K)