UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A potentially stable five-helix bundle cavitein and applications of caviteins in ester hydrolysis and… Yang, Hui 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2015_september_yang_hui.pdf [ 3.79MB ]
Metadata
JSON: 24-1.0166361.json
JSON-LD: 24-1.0166361-ld.json
RDF/XML (Pretty): 24-1.0166361-rdf.xml
RDF/JSON: 24-1.0166361-rdf.json
Turtle: 24-1.0166361-turtle.txt
N-Triples: 24-1.0166361-rdf-ntriples.txt
Original Record: 24-1.0166361-source.json
Full Text
24-1.0166361-fulltext.txt
Citation
24-1.0166361.ris

Full Text

A POTENTIALLY STABLE FIVE-HELIX BUNDLE CAVITEIN AND APPLICATIONS OF CAVITEINS IN ESTER HYDROLYSIS AND PROTEIN-PROTEIN INTERACTION by Hui Yang B.Sc., Nanjing University, 2010 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemistry) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July 2015 © Hui Yang, 2015 ii Abstract  This thesis will present the study of a peptide sequence that is favourable for forming a five-helix bundle through the use of template assembled synthetic proteins (TASPs), and applications of our synthetic proteins in rate enhancement of ester hydrolysis and protein-protein interactions. Chapter 1 will introduce proteins, de novo protein design, and TASPs. It will also review past and current research on the applications of the synthetic molecules mimicking biological behavior.  Chapter 2 will focus on several modified de novo designed peptide sequences that are intended to be more favourable in folding into a five-helix bundle than a four-helix bundle. We found that our designed sequence narrowed the free energy gap between the five-helix bundle and the four-helix bundle relative to the systems where the peptide sequence was designed to favour a four-helix bundle.  Chapter 3 will concentrate on investigating the rate enhancement of ester hydrolysis by histidine-containing TASPs. These TASPs increased the rate of ester hydrolysis, and the position of the histidines was found to be relevant to activity.  Chapter 4 will detail the attempts at using a template-assembled synthetic protein to inhibit protein-protein interactions between a Bak peptide and a Bcl-xL protein. Our synthesized protein was found to be a good binding partner towards the Bcl-xL protein and manifested moderately enhanced proteolytic resistance. Several heterocaviteins were also synthesized to study both their inhibitive activities to the Bcl-xL protein, and their proteolytic stability. iii  Chapter 5 will summarize and conclude the work throughout this thesis.                iv Preface  Chapter 2 was carried out in Chemistry Building A, UBC, by myself (Hui Yang) and Professor John C. Sherman. I was responsible for the synthesis and characterization of all the related peptides and caviteins. This work has not been published.  Chapter 3 was carried out in Chemistry Building A, UBC, by myself and Professor John C. Sherman. I was responsible for the synthesis and characterization of all the related peptides and caviteins. Dr. Jon O. Freeman originally designed and synthesized peptide pQ4H and cavitein Q4H. A version of Chapter 3 has been published. Yang, H., Sherman, J. C. Bioorg. Med. Chem. Lett. 2013, 23, 1752-1753. DOI: 10.1016/j.bmcl.2013.01.063. I was responsible for writing the manuscript, which was edited by Professor John Sherman.  Chapter 4 was carried out in Chemistry Building A, UBC, by myself, Dr. Elena Polishchuk, Qi Qian and Professor John C. Sherman. Dr. Elena is the laboratory director at biological services laboratory and Qi Qian is a graduate student at Professor Martin Tanner’s lab. I was responsible for the synthesis and characterization of all the related peptides and caviteins. Dr. Adam Mezo originally designed and synthesized peptide pN1GG and cavitein N1GG. Dr. Polishchuk provided guidance and assistance on overexpression of the Bcl-xL protein and Qi Qian provided guidance and assistance with purification techniques of the Bcl-xL protein. This work has not been published.    v Table of Contents Abstract………………………………………………………………………………………ii Preface………………………………………………………….……………………………iv Table of Contents………………………………………………….…………………………v List of Tables………………………………………………………………...…………….…x List of Figures…………………………………………………………………………….…xi List of Symbols and Abbreviations…..………………………...………………..…….…xvii Acknowledgements………………………………………………………...…………….xix Dedication……………………………………………………………..………..………..…xx Chapter 1 Introduction…………………………………………….……………………1  1.1 General Introduction………………………………………………..…………1  1.2 Basics of Protein Structure……………………………………………………5   1.2.1 Primary Structure………………………………………...……………5   1.2.2 Secondary Structure…………………………………….……………11   1.2.3 Super Secondary Structure………………………..…………………14   1.2.4 Tertiary Structure……………………………………………………17   1.2.5 Quaternary Structure…………………………………...……………19  1.3 Protein Folding………………………………………………………………20  1.4 Factors Contributing to Stabilities of Protein Folding………………………22  1.5 De Novo Protein and Template Assembled Synthetic Proteins……...………23   1.5.1 De Novo Protein Design…………………………………..…………23   1.5.2 Template Assembled Synthetic Proteins…………………….………25  1.6 Applications of Synthetic Biomolecules………………………………….…31 vi   1.6.1 General Applications of Synthetic Biomolecules……………………31   1.6.2 Synthetic Biomolecules Mimicking the Functions of Natural Enzymes    ………………………………………………………………….……34    1.6.2.1 Mimicking the Functions of Peroxidase……………..………34    1.6.2.2 Catalyzing a Retro-aldol Reaction………………………...…36    1.6.2.3 Mimicking the Function of Hydrolase…………….………37   1.6.3 Synthetic Biomolecules in Protein-protein Interactions…….………38    1.6.3.1 Hot Spot-based Design of Synthetic Biomolecules…….……38    1.6.3.2 Protein-protein Interactions in Hippo Signaling Pathway..…40    1.6.3.3 Protein-protein Interactions in Clathrin-mediated Endocytosis    …………………………………………………………………….…41    1.6.3.4 Protein-protein Interactions in Programmed Cell Death….…42  1.7 Thesis Aims and Research Overview..………………………………..….….43 Chapter 2 A Potentially Stable Five-helix Bundle Cavitein………………………….46  2.1 Introduction…………………………………………………….……………46  2.2 Results and Discussion………………………………………………………48   2.2.1 Cavitein Design……………………………………….……..………48   2.2.2 Selection of Cavitand……………………………………..…………50    2.2.2.1 Synthesis of Benzylthiol Cavitand…………………..………51   2.2.3 Peptide Sequence Design and Synthesis……………………….……54   2.2.4 Cavitein Synthesis………………………………………………...…59   2.2.5 Characterization of the Caviteins……………………………………62    2.2.5.1 Far-UV Circular Dichroism Spectra…………………………62    2.2.5.2 GuHCl Denaturation Experiments………………………..…66  2.3 Chapter Summary and Conclusion………………………………………………72  2.4 Experimental………………………………………………………………..……74 vii   2.4.1 Cavitand Synthesis………………………………………..…………74   2.4.2 Peptide Synthesis………………………………………….…………75    2.4.2.1 General Procedures of Peptide Synthesis……………………75    2.4.2.2 Syntheses of Peptides: L, M and S…………………..………76   2.4.3 Cavitein Synthesis……………………………………………..….…78    2.4.3.1 General Procedures of Cavitein Synthesis…………..………78    2.4.3.2 Syntheses of Caviteins………………………………………79   2.4.4 Circular Dichroism Experiments………………….…………………82   2.4.5 GuHCl Denaturation Studies………………………………...………82 Chapter 3 Application of Caviteins in Ester Hydrolysis…………………..…………84  3.1 Introduction………………………………………………….………………84  3.2 Results and Discussion………………………………………………………87   3.2.1 Cavitein Design………………………………………...……………87   3.2.2 Peptide Design and Synthesis…………………….…………………91   3.2.3 Cavitand Synthesis………………………………………..…………92   3.2.4 Cavitein Synthesis…………………………………………...………93   3.2.5 Far-UV Circular Dichroism Spectra…………………………………96   3.2.6 Ester Hydrolysis Study………………………………………………97    3.2.6.1 Initial Rates of Cavitein Q4-H with Different Substrates…...98    3.2.6.2 Initial Rates of Cavitein Q4-H2 with Different Substrates...100   3.2.7 HPLC Results of Post-reaction Caviteins………………….………105  3.3 Chapter Summary and Conclution…………………………………………112  3.4 Experimental………………………………………………..………………113   3.4.1 Cavitand Synthesis…………………………………………………113   3.4.2 Peptide Synthesis……………………………………………..……114 viii    3.4.2.1 General Procedures of Peptide Synthesis…………….……114    3.4.2.2 Syntheses of Peptides: pQ4-H and pQ4-H2………………115   3.4.3 Cavitein Synthesis…………………………………………….……116    3.4.3.1 General Procedures of Cavitein Synthesis…………………116    3.4.3.2 Syntheses of Caviteins Q4-H and Q4-H2………….………117   3.4.4 Circular Dichroism…………………………………………………118   3.4.5 UV-spectrophotometric Assay……………………………..………118   3.4.6 Post-reaction Caviteins Analysis……………………………...……119 Chapter 4 Application of Caviteins in Protein-protein Interactions………………120  4.1 Introduction…………………………………………………..……….……120  4.2 Results and Discussion……………………………………………..………124   4.2.1 Cavitein Design……………………………………………….……124   4.2.2 Peptide Design and Synthesis………………………………………130   4.2.3 Cavitein Synthesis……………………………………….…………132   4.2.4 Overexpression of Bcl-xL Protein……………………………..……142   4.2.5 Far-UV Circular Dichroism Spectra………………………….……145   4.2.6 Fluorescence Polarization Assay…………………………………...149   4.2.7 Trypsin Cleavage Assay……………………………………………154   4.2.8 GuHCl Denaturation Studies……………………………….………159  4.3 Chapter Summary and Conclusion…………………………………………162  4.4 Experimental……………………………………………………..…………163   4.4.1 Cavitand Synthesis…………………………………………………163   4.4.2 Peptide Synthesis…………………………………………...………163    4.4.2.1 General Procedures of Peptide Synthesis……..……………163    4.4.2.2 Synthesis of Free Bak Peptide…………………………...…164 ix    4.4.2.3 Synthesis of Bak Peptide and o-pN1GG Peptide……..……165    4.4.2.4 Synthesis of Flu-Bak Peptide………………………….…166   4.4.3 Cavitein Synthesis……………………………………………….…167    4.4.3.1 General Procedures of Cavitein Synthesis…………………167    4.4.3.2 Syntheses of Caviteins N1GG, o-N1GG and 4B………..…167    4.4.3.3 Syntheses of Heterocaviteins………………………….……168     4.4.3.3.1 Syntheses of Heterocaviteins 1B3N and 1B3O       ………………………………………….…..168     4.4.3.3.2 Synthesis of Heterocavitein 3B1O…………169     4.4.3.3.3 Syntheses of Heterocavitein 2B2O and BOBO       …………………………………………...…170   4.4.4 Overexpression of Bcl-xL Protein…………………………….….…171    4.4.4.1 Plasmid Replication……………………………………...…171    4.4.4.2 Overexpression and Purification of Bcl-xL Protein…...……172   4.4.5 Circular Dichroism…………………………………………………173   4.4.6 Protein Binding Studies……………………………………….……174   4.4.7 Trypsin Cleavage Studies……………………………………..……175   4.4.8 GuHCl Denaturation Studies………………………………….……176 Chapter 5 Thesis Summary and Conclusions……………………………….………178 References…………………………………………………………………………………182      x List of Tables  Table 2.1 Designed peptide sequencescompared with original peptide. ………………54 Table 2.2 Nomenclature for the caviteins and their molecular weights. ………………60 Table 2.3 Percent helicity for all the caviteins………………………………………... 66 Table 2.4 Comparison of ∆G°H2O values between peptide S-based caviteins and caviteins with peptides that are idealized for forming four-helix bundle……72 Table 3.1 Nomenclature for the peptides and caviteins within this chapter…………... 92 Table 3.2 Initial rates of cavitein Q4-H on different substrates………………………100 Table 3.3 Initial rates of cavitein Q4-H2 on different substrates.…………………… 102 Table 3.4 Summary of initial rate……………………………….…………………… 104 Table 3.5 Results of analysis of the post-reaction caviteins. ……………………...… 111 Table 4.1 Designed peptide compared with original peptide. ……………………..…132 Table 4.2 Nomenclature and compositions for the caviteins. ………………………...133 Table 4.3 Molecular weights of caviteins obtained from MALDI-TOF and ESI.....…142 Table 4.4 Percent helicity for all the caviteins. ………………………………………149 Table 4.5 Summary of competitive inhibition experiment results. …………………..154 Table 4.6 Summary of trypsin cleavage assay results. …………………………….…157 Table 4.7 Relative overall efficiency (e) and overall efficiency (e * n) of all the caviteins………………..……………………………………………...……158 Table 4.8 ∆G°H2O values for caviteins 1B3O, 2B2O and BOBO. ……...……………161    xi List of Figures  Figure 1.1 Anfinsen’s famous experiment on protein folding. …………………..………2 Figure 1.2 a) Structure of amino acids, all the natural amino acids are L-amino acids except glycine. b) Primary structure of a peptide. c) Backbone of a protein’s primary structure……………………………………………………………..6 Figure 1.3 Amino acids used within this thesis, along with their respective three and one  letter coding…………………………………………………………...………7 Figure 1.4 Depicts formation of peptide bond (highlighted in red) by two amino acids  (leucine and alanine) with the expulsion of a water molecule………..………8 Figure 1.5 The resonance in the peptide bond makes the carbonyl group and the amide  nitrogen in the same plane, and the steric hindrance from the side chains in  the cis form is shown. ……………………………………………………...…9 Figure 1.6 Phi torsion angle (Φ) and psi torsion angle (Ψ). ………………….…...……10 Figure 1.7  The genetic code. (Phe: phenylalanine, Met: methionine, Pro: proline, Ser:  serine, Thr: threonine, Trp: tryptophan, Tyr: tyrosine) ………………..……11 Figure 1.8 Structure of right-handed alpha helix and the overall dipole moment within the helix (only the backbone atoms are shown). ………………….……...…12 Figure 1.9 Beta sheets with parallel and antiparallel form. The hydrogen bonds are more linear in the antiparallel form than that in the parallel form. ……………..…13 Figure 1.10 Ramachandran diagram, the general case of 81,234 non-Gly, non-Pro,  non-prePro residues. Reproduced with permission from Proteins 2003, 50,  444. Copyright 2003 Wiley-Liss, Inc. ………………………………………14 Figure 1.11 Three examples of the super secondary structures. a) Helix-loop-helix, b)  Beta strand-helix-beta strand, c) Greek key (four adjacent antiparallel beta  strands folded upon itself) …………………………………………….….…15 Figure 1.12 Super secondary structure of a four-helix bundle with two interhelical turns……………………………………………………………………….…16 Figure 1.13 The structure of ara h 2, a five-helix bundle held together by four disulfide bonds. Reproduced with permission from Allergy 2011, 66, 881. Copyright 2011 Clearnce Center, ………………………………….……………...……17 Figure 1.14 The tertiary structures of aplysia limacina myoglobin. ………………..……18 Figure 1.15 The quaternary structure of horse deoxyhaemoglogin. ……………..………19 Figure 1.16 The funnel landscape depicts that the protein reaches a thermodynamically  stable state as it folds down the landscape. There are varous local minima  representing folded intermediates. Reprinted with permission from Polymer  2004, 45, 548. Copyright 2003 Elsevier Science Ltd. ………………………21 Figure 1.17 De novo protein design of a four-helix bundle by Ho and DeGrado. ………24 xii Figure 1.18 Template assisted synthetic protein. ……………………………...…………26 Figure 1.19 Cyclic peptide-based template assisted protein designed by Mutter and co-workers…………………………………………………………………….27 Figure 1.20 An arylthiol cavitand. ………………………………………………….……29 Figure 1.21 First synthesized cavitands by Cram. ………………………………….……30 Figure 1.22 Two types of our frequently used cavitand. On the left is the four-member  ring arylthiol cavitand and the right one is the five-member ring benzylthiol  cavitand. ……………………………………………………….……………31 Figure 1.23 The crystal structures of several proteins. a) Depicts the structure of the 20S  proteasome from yeast in complex with the proteasome activator PA 26. b)  Depicts the structure of the AIM2 pyrin domain. c) Depicts the structure of  transmembrane segment IV of the NHE1 isoform of the Na+/H+ exchanger.  …………………………………………………………………….…………33 Figure 1.24 Computer model of MP3. a) Helices with key residues depicted as sticks. b)  The sequence of the peptide sequence used in MP3. Reprinted with permission from Chem. Eur. J. 2012, 18, 15962. Copyright 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. ……………………………35 Figure 1.25 A) A designed hydrophobic pocket encloses the aromatic portion of the substrate. B) & C) The active site for proton shuffling. Reproduced with permission from Science 2008, 319, 1390. Copyright 2008 American Association for the Advancement of Science. ……………….………...……36 Figure 1.26 The structure of the helix-loop-helix monomer with the six histidine residues as the reactive site. Reprinted with permission from J. Am. Chem. Soc. 1997, 119, 11363. Copyright 1997 American Chemical Society. …..………..……38 Figure 1.27 The region of a PPI interface. The projecting hot spot residue “1” binds  into a pocket in the complementary surface region of the other protein.  Reprinted with permission from J. Am. Chem. Soc. 2013, 135, 6251.  Copyright 2013 American Chemical Society. ………………………………39 Figure 1.28 Formation of the disulfide bond of the cyclic peptide which inhibites PPI between YAP and TEAD. Reprinted with permission from Med.Chem. Lett. 2014, 5, 993. Copyright 2014 American Chemical Society. ………………..40 Figure 1.29 Photoswichable peptide inhibitors with cysteine residues at different positions a) i and i+11 b) i and i+7. Reprinted with permission from  Angew. Chem. Int. Ed. 2013, 52, 7705. Copyright 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. ………………………………...………...………………42 Figure 1.30  Aim of this thesis………………………………………….…………………45 Figure 2.1 Wheel model for a peptide with sequence EELL KKL EEL KKG, where hydrophobic residues are marked by square and hydrophilic residues are marked by circle. ……………………………………………………………49 Figure 2.2 Comparison of the hydrophobic regions between the four-helix bundle (left)   and the five-helix bundle (right). ……………………………………………50 xiii Figure 2.3 Energy minimized structures of [n]cavitands. Reproduced with permission from Chem. Eur. J. 2001, 7, 1639. Copyright 2001 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, Fed. Rep. of Germany. ………………..…51 Figure 2.4 Synthesis of [5]-benzylthiol cavitand from 2-methyl resorcinol and diethoxymethane. ……………………………………………………………53 Figure 2.5 Wheel model for a peptide sequence that is more favorable in forming a four-helix bundle. ……………………………………………...……………55 Figure 2.6 Activation of peptide with 2,2’-dipyridyl disulfide. …………………...……56 Figure 2.7 Wheel model for peptide L, in which the original residues E8 and K6 are replaced by leucine residues to increase the hydrophobic region of the new helix. ……………………………………………………………...…………57 Figure 2.8 Wheel model for peptide sequence M. ……………………………...………58 Figure 2.9 Wheel model for peptide sequence S. ………………………………………59 Figure 2.10 Synthesis of caviteins L4, L5, M4, M5, S4 and S5. …………………..……60 Figure 2.11 Analytical reverse phase HPLC trace of purified caviteins a) L4 and b) L5. Their purity was assessed by an analytical reverse phase column using a H2O:ACN gradient (40% H2O to 20% H2O over 15 minutes). ……..………61 Figure 2.12 Analytical reverse phase HPLC trace of purified caviteins a) M4 and b) M5. Their purity was assessed by an analytical reverse phase column using a H2O:ACN gradient (45% H2O to 30 % H2O over 15 minutes) ……..………61 Figure 2.13 Analytical reverse phase HPLC trace of purified caviteins a) S4 and b) S5. Their purity was assessed by an analytical reverse phase column using a H2O:ACN gradient (57% H2O to 47% H2O over 10 minutes) ………...……62 Figure 2.14 CD spectra of cavitein L4 and cavitein L5. The spectra were acquired at 25 ℃ in 50 mM phosphate buffer at pH 7.0 using a cavitein concentration of 50 µM. …………………………………………..……………………………...64 Figure 2.15 CD spectra of cavitein M4 and cavitein M5. The spectra were acquired at 25   C in 50 mM phosphate buffer at pH 7.0 using a cavitein concentration of 50 µM…………………………………………………………………….64 Figure 2.16 CD spectra of cavitein S4 and cavitein S5. The spectra were acquired at 25 ℃ in 50 mM phosphate buffer at pH 7.0 using a cavitein concentration of 50 µM…………………………………………………………………………65  Figure 2.17 GuHCl dentaturation of cavitein L4 and cavitein L5. Each curve was acquired at pH 7.0 in 50 mM phosphate buffer at 25 °C at a cavitein concentration of 40 μM. ……………………………………………….……………………68 Figure 2.18 Denaturation curves of cavitein L4 and cavitein L5 recorded in elevated temperatures in 8 M GuHCl solution. ………………………………………69 Figure 2.19 GuHCl dentaturation of cavitein M4 and cavitein M5. Each curve was acquired at pH 7.0 in 50 mM phosphate buffer at 25 °C at a cavitein concentration of 40 μM. ……………………………….……………………70 xiv Figure 2.20 GuHCl dentaturation of cavitein S4 and cavitein S5. Each curve was acquired at pH 7.0 in 50 mM phosphate buffer at 25 °C at a cavitein concentration of 40 μM. …………………………………………….………………………71 Figure 3.1 Structure of a carboprotein with imidazoyl groups on the peptide used as a rate enhancer towards ester hydrolysis. ………...…………………...........…85 Figure 3.2 Possible mechanism for the ester hydrolysis reaction by an enzyme. ………86 Figure 3.3 Histidine in its conjugate acid form of imidazole side chain in equilibrium   with its conjugate base. ……………………………………………...………87 Figure 3.4 Wheel model of cavitein Q4-H. ……………………………….……………88 Figure 3.5 Wheel model for cavitein Q4-H2. ……………………………..……………90 Figure 3.6 Positions of the histidine residues in Cavitein Q4-H and Q4-H2……...……91 Figure 3.7 Synthesis of arylthiol cavitand from resorcinol and acetaldehyde. …………93 Figure 3.8 Synthesis of cavitein Q4-H2. …………………………………….…………94 Figure 3.9 Preparatory Reverse phase HPLC purification of cavitein Q4-H2. Cavitein Q4-H2 was purified with a reverse phase preparatory column using a H2O:ACN gradient (65% H2O to 54% H2O over 22 minutes). ……..………95 Figure 3.10 Analytical reverse phase HPLC trace of purified Q4-H2 cavitein. The purity of cavitein Q4-H2 was assessed with an analytical reverse phase column using a H2O:ACN gradient (65% H2O to 54% H2O over 22 minutes). ….…96 Figure 3.11 CD spectra of cavitein Q4-H (red) and cavitein Q4-H2 (black). The spectra   were acquired at 25 ℃ in 50 mM phosphate buffer at pH 7.0 using a cavitein   concentration of 30 µM. ……………………………………….……………97 Figure 3.12 Linear fitting of reaction progress of a) peptide qQ4-H and b) cavitein Q4-H in the first five minutes; reaction progress of c) peptide pQ4-H, d) cavitein Q4-H on p-nitrophenyl acetate (PNPA), p-nitrophenyl butyrate (PNPB) and p-nitrophenyl octanoate (PNPO) (phosphate buffer, pH 7), p-nitrophenolate a b s o r b a n c e  s i g n a l  a t  4 1 0  n m  i n  t h e  f i r s t  t h i r t y minites……………………………………………………………………..99 Figure 3.13 Linear fitting of reaction progress of a) peptide qQ4-H2 and  b) cavitein Q4-H2 in the first five minutes; reaction progress of c) peptide pQ4-H2, d) cavitein Q4-H2 on p-nitrophenyl acetate (PNPA), p-nitrophenyl butyrate (PNPB) and p-nitrophenyl octanoate (PNPO) (phosphate buffer, pH 7), p-nit rophenolate absorbance s ignal  at  410 nm in the fi rst  thi r ty minites……………………………………………………………………101 Figure 3.14 Comparisons between cavitein Q4-H and cavitein Q4-H2 with a) p-nitrophenyl acetate (PNPA), b) p-nitrophenyl butyrate (PNPB) and c) p-nitrophenyl octanoate (PNPO). ……………………………………………103 Figure 3.15 Post-reaction cavitein Q4-H in reaction with PNPA was analyzed by RP-HPLC and shows four different cavitein derivatives after reaction. …..106 Figure 3.16 Post-reaction cavitein Q4-H in reaction with PNPB was analyzed by RP-HPLC and shows five different cavitein derivatives after reaction. …..107 xv Figure 3.17 Post-reaction cavitein Q4-H2 in reaction with PNPA was analyzed by RP-HPLC and shows four different cavitein derivatives after reaction. …..108 Figure 3.18 Post-reaction cavitein Q4-H2 in reaction with PNPB was analyzed by RP-H P LC  an d  s h o w s  a  b r o ad  p e ak  t h a t  c o n t a i n s  f o u r  c av i t e i n derivatives. ……………………………………………………………...…109 Figure 3.19 Post-reaction cavitein Q4-H in reaction with PNPO was analyzed by RP-HPLC and shows no cavitein derivative after reaction. ………………...…110 Figure 3.20 Post-reaction cavitein Q4-H2 in reaction with PNPO was analyzed by RP-HPLC and shows only one cavitein derivative after reaction. ……...…110 Figure 4.1 Binding pocket of Bcl-xL protein bound to the Bak peptide. Reproduced with permission from Science 1997, 275, 985. Copyright 1997 American Association for the Advancement of Science. ………………….…….……121 Figure 4.2 Bak peptide inhibitors of the Bcl-xL/Bak interactions. a) the formation of a hydrogen-bond surrogate alpha helix; b) a benzoylurea-derived alpha helix; c) a terphenyl-based proteomimetics; d) an unnatural oligomer with strong conformational propensities.……………….……………………….123 Figure 4.3 a) Sequence of the Bak peptide, existing as a random coil in aqueous solution; b) wheel model of the Bak peptide, hydrophobic amino acid residues are marked with different color. …………………………………….…………126 Figure 4.4 Wheel model of a hetero Bak cavitein with three o-pN1GG peptide and one   Bak peptide. ……………………………………………….…………….…128 Figure 4.5 Structures of lysine compared with ornithine and homolysine. ………...…129 Figure 4.6 Wheel models of hetero Bak caviteins with various number of o-N1GG peptide. ………………………………………………………………….…130 Figure 4.7 Synthesis of cavitein 4B. ………………………………………….……….134 Figure 4.8 Preparatory Reverse phase HPLC purification of cavitein 4B. Cavitein 4B was purified with a reverse phase preparatory column using a H2O:CAN gradient (66% H2O to 63% H2O over 12 minutes). ……………….…….…134 Figure 4.9 Analytical reverse phase HPLC trace of purified cavitein 4B. The purity of cavitein 4B was assessed with an analytical reverse phase column using a H2O:ACN gradient (78% H2O to 65% H2O over 13 minutes) …….………135 Figure 4.10 Synthesis of partially incorporated caviteins: a) partially incorporated Bak   cavitein, b) partially incorporated o-N1GG cavitein. …………..…………137 Figure 4.11 a) Preparatory Reverse phase HPLC purification of hetero Bak Caviteins was performed with a reverse phase preparatory column using a H2O:ACN gradient (65% H2O to 36% H2O over 30 minutes); b) Preparatory Reverse phase HPLC purification of hetero o-N1GG Caviteins was performed with a reverse phase preparatory column using a H2O:ACN gradient (65% H2O to 36 % H2O over 30 minutes). …..………………………………...…………138 Figure 4.12 Synthesis of heterocavitein 1B3O and 3B1O. ……………………….……139 Figure 4.13 Synthesis of heterocavitein 2B2O and BOBO. ……………………………140 xvi Figure 4.14 Analytical reverse phase HPLC trace of purified heterocaviteins: a) 1B3O, b) 2B2O or BOBO, c) BOBO or 2B2O, d) 3B1O. The purity of these heterocaviteins were assessed with an analytical reverse phase column using a H2O:ACN gradient (74% H2O to 56% H2O over 20 minutes). ……………141 Figure 4.15 Overexpression of Bcl-xL protein. …………………………………………143 Figure 4.16 The sequence of the recombinant Bcl-xL protein. …………………………144 Figure 4.17 SDS-polyacrylamide gel eletrophoresis of Bcl-xL protein. ……...…...……144 Figure 4.18 Analytical reverse phase HPLC trace of purified Bcl-xL protein. The purity of   Bcl-xL protein was assessed with an analytical reverse phase column (Waters   Delta Pak C-18 column (300 × 3.9 mm2, 300 Å, 15 µm)) using a H2O:ACN   gradient (85% H2O to 10% H2O over 75 minutes) ………………...………145 Figure 4.19 CD spectra of cavitein N1GG and cavitein o-N1GG. The spectra were acquired at 25 ℃ in 50 mM phosphate buffer at pH 7.0 using cavitein concentrations of 40 µM. ………………………………………….….……146 Figure 4.20 CD spectra of cavitein 4B, 3B1O and Bak peptide. The spectra were acquired at 25 ℃ in 50 mM phosphate buffer at pH 7.0 using cavitein concentrations of 25 µM and peptide of 100 µM. …………………………………….……...147 Figure 4.21 CD spectra of cavitein 1B3O, 2B2O and BOBO. The spectra were acquired   at 25 ℃ in 50 mM phosphate buffer at pH 7.0 using cavitein concentrations   of 25 µM. ……………………………………………………….…….……148 Figure 4.22 The basic concept of fluorescence polarization. ………………..…….……150 Figure 4.23 Binding of fluorescent peptides to Bcl-xL protein. ……………..…….……151 Figure 4.24 A fluorescence polarization assay shows that cavitein 4B binds Bcl-xL with higher affinity than Bak peptide. The Kd value for the Bak Cavitein was determined by competitive inhibition of fluorescein tagged Bak peptide (15 nM) and Bcl-xL (500 nM) complex. The experimental data was fitted to a sigmoidal dose-response nonlinear regression model. ……………….……152 Figure 4.25 Fluorescence polarization assay of cavitein 4B and all the hetero Bak caviteins binds Bcl-xL protein. The experimental data was fitted to a sigmoidal dose-response nonlinear regression model..………….…………153 Figure 4.26 Proteolytic stabilities of all the caviteins. ………………………………….156 Figure 4.27 GuHCl dentaturation of cavitein 4B and heterocavitein 3B1O. Each curve was acquired at pH 7.0 in 50 mM phosphate buffer at 25 °C at a cavitein concentration of 25 μM. ……………………………………………...……160 Figure 4.28 GuHCl dentaturation of heterocavitein 2B2O, BOBO and 1B3O. Each curve was acquired at pH 7.0 in 50 mM phosphate buffer at 25 °C at a cavitein concentration of 25 μM. …………………………………...………………161 Figure 4.29 HPLC results of trypsin cleavage assay. The extra three peaks in 30 min HPLC for cavitein 4B correspond to multiple cleaved cavitein as there are four peptide in it. (confirmed by MALDI-TOF).…………………………..176 xvii List of Symbols and Abbreviations  [θ]M.R.E. mean residue ellipticity Å angstroms ACN acetonitrile AUC analytical ultracentrifugation CCA α-cyano-4-hydroxycinnamic acid CD circular dichroism Bcl-xL B-cell lymphoma-extra large BH Bcl-2 homology DCM dichloromethane DHB 2,5-dihydroxybenzoic acid DIPEA diisopropylethylamine DMA N, N-dimethylacetamide DMF N, N-dimethylformamide DMSO dimethylsulfoxide DOSY diffusion-ordered nuclear magnetic resonance spectroscopy ESI-MS electrospray ionization mass spectrometry EtOH ethanol FMOC 9H-fluoren-9-ylmethoxycarbonyl GuHCl guanidine hydrochloride h hour (s) HBTU  1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate HOBt 1-hydroxybenzotriazole HPLC high performance liquid chromatography IPTG isopropyl 1-thio-β-D-galactopyranoside m/z mass-to-charge ratio MALDI-TOF matrix-assisted laser desorption/ionization time of flight MeOH methanol min minute (s) mL milliliter mM milimolar mP millipolarization xviii MS mass spectrometry MW molecular weight NBS N-bromosuccinimide NMP N-methylpyrrolidone NMR nuclear magnetic resonance PDB protein data bank Ph phenyl PNPA p-nitrophenyl acetate PNPB p-nitrophenyl butyrate PNPO p-nitrophenyl octanoate PPI protein-protein interaction RP-HPLC reverse phase high performance liquid chromatography rpm revolutions per minute rt room temperature SDS-PAGE sodium dodecyl sulfate-polyacrylamide gel electrophoresis SPPS solid phase peptide synthesis TASP template assembled synthetic protein TFA trifluoroacetic acid Trt trityl UV ultraviolet           xix Acknowledgements  Thank you to my supervisor, Dr. John Sherman, for his guidance throughout my graduate studies at UBC. He has taught me a great deal about chemistry and scientific thinking in general.  Thank you to Dr. Elena Polishchuk and Qi Qian for their guidance and assistance in my protein-protein interaction project.   Thank you to all the previous members in Sherman group and all the faculties in Chemistry Department, UBC, for their direct and indirect assistance.  Thank you to my parents and my wife for their unwavering love and support.           xx     To Mikayla Ng1 Chapter 1 Introduction  1.1 General Introduction  Proteins are a major building block of cells. They are vital for the existence of every living species. Most proteins exist in a compact, autonomously organized (well-folded) and well-defined three dimensional form.1 Scientists are drawn to explain how proteins fold into their unique structures and what factors contribute to this folding problem.  The phenomenon of protein folding has been the focus of protein structure study since the 1960s, starting with a famous experiment from Chris Anfinsen and colleagues (Figure 1.1). They used a well-folded ribonuclease A, containing four disulfide bonds, and observed a complete denaturation when it was incubated in the presence of β-mercaptoethanol (a reducing agent) and 8 M urea. Oxidation under denaturing conditions gave rise to “scrambled” species in which four disulfide bonds were randomly paired. However, when urea was removed from this system, the original native structure of ribonuclease was recovered by treatment with a trace amount of the reducing agent β-mercaptoethanol. This shows that the folding of ribonuclease A is fully reversible without any external cofactors,2 which can also be interpreted as following: the complex three-dimensional structure of protein molecules depends exclusively on the amino acid sequence. 2     Unveiling the relationship between the amino acid sequence and protein folding has been one of the greatest challenges of molecular biology. Although a thorough knowledge of how proteins fold remains elusive, theoretical and experimental studies of protein folding have come a long way since Anfinsen’s discovery.3   When talking about the protein folding problem, we inevitably have to mention the “Levinthal Paradox”. Let us estimate the time required for a protein to fold, where each bond connecting two neighboring amino acids can have three possible states. In a polypeptide chain of 101 amino acids, there could be 3100 (around 5.15 × 1047) different possibilities for Figure 1.1 Anfinsen’s famous experiment on protein folding.  3 the sequence to fold and only one of them corresponds to the native state. It would take 7.36×1037 years to find the native state even if the protein is sampling the new configuration at a rate of 7 billion possibilities (world population) per second. In reality, however, proteins fold in a timescale of seconds or less. To explain this paradox, Levinthal pointed out that proteins must have some predetermined folding pathways rather than just folding by a random search (vide infra).4  Later, many studies, stimulated by the search for indications of predetermined folding pathways, gave rise to a number of models illustrating the folding process, such as the nucleation/growth model, 5 the framework model,6 the hydrophobic collapse model,7 the jigsaw puzzle model,8 and others. However, these models of protein folding are limited in their application. Thus, none of these are qualified to be a general model to explain the entire protein-folding problem.  The creation of synthetic de novo (from scratch) proteins offers a way to predict and control protein folding. The design of proteins through this method has progressed significantly within the last two decades.9, 10 William DeGrado, a pioneer of de novo protein design with his early work on a minimalist approach founded on first principles, made a great contribution in the de novo protein design field.11 Later on, template assembled synthetic proteins (TASPs) were introduced by Mutter and Vulleumies. In this system, peptides are covalently linked to a rigid scaffold which provides a simple way to explore deeper rules that control the protein folding behavior.12, 13 Our research group, and consequently the second chapter of the thesis, focuses primarily on the study of predicting and controlling the folding of a protein. 4  Overwhelmed by the challenges of the protein-folding problem, scientists asked themselves: why should we care about protein-folding? The reason why we have invested so much effort into it is because proteins are molecules with extraordinary functional diversity which have the potential to influence our daily life profoundly in the future. For example, enzymes are proteins that function as catalysts as they significantly speed up chemical reactions which would otherwise be extremely slow. One such example is the breakdown of proteins by the enzyme pepsin, which is a digestive enzyme found in the stomach.14 Some proteins, like actin and collagens, build up our body structure. Actin is involved in muscle contraction and movement, while collagens provide support for connective tissues.15, 16 What is more, our immune system relies on specialized proteins to identify antigens and defend against bacteria and viruses. These specialized proteins are called antibodies. Finally, hormonal proteins coordinate certain bodily activities: insulin helps regulate blood-sugar concentration.17 Therefore, life on earth is deeply dependent on proteins, and it cannot be sustained without them.   My research in the third and fourth chapter of this thesis concentrates mainly on the applications of our synthetic proteins in mimicking a catalyst and a protein-protein interaction.     5 1.2 Basics of Protein Structure  1.2.1 Primary Structure  Protein are polymers of 20 different amino acids joined together by peptide bonds. There are four levels of protein structure in nature. Primary structure is the linear sequence of amino acids in the polypeptide chain (Figure 1.2). As we can see in Figure 1.2, the backbone (the regularly repeating part) of protein primary structure consists of amide bonds and alpha carbons. The chemical groups that are bonded to the alpha carbons of the backbone are the side chains of amino acids. It is the side chains that give rise to different tendencies for amino acids to participate in interactions with each other and with water.  6   Of the 20 common amino acids, there are amino acids with hydrophobic side chains like alanine and leucine. They have the tendency to repel water and pack against each other. Amino acids like glutamic acid and lysine are hydrophilic. Their side chains are able to form hydrogen bonds or salt bridges. These differences in amino acids significantly determine their contributions to protein stability and function. The amino acids used within this thesis, along with their respective three and one letter coding, are depicted in Figure 1.3. Figure 1.2 a) Structure of amino acids, all the natural amino acids are L-amino acids except glycine. b) Primary structure of a peptide. c) Backbone of a protein’s primary structure. 7    Amino acids are connected by peptide bonds (more generally, amide bonds), which are chemical bonds formed when a carboxylic acid condenses with an amino group with the expulsion of a water molecule (Figure 1.4).   Figure 1.3 Amino acids used within this thesis, along with their respective three and one   letter coding 8     A peptide bond is quite stable because of resonance. One consequence of resonance is that it increases the polarity of the peptide bond, which contributes to the behaviour of protein folding (Figure 1.5). Also, because of resonance, the three atoms (the carbonyl oxygen, the carbonyl carbon and the amide nitrogen) that make up the peptide bond are coplanar, and the free rotation of this bond is hindered. This reduces the number of possible conformations that a polypeptide chain can adopt. There are theoretically two conformations, trans and cis, but only the trans conformation is favourable because of steric interactions between two consecutive side chains.   Figure 1.4 Depicts formation of peptide bond (highlighted in red) by two amino acids    (leucine and alanine) with the expulsion of a water molecule 9   Free rotation is allowed in the other two bonds: the N–Cα and Cα–C bonds (where Cα is the carbon atom to which the side chain is attached) (Figure 1.6). The angle between the N–Cα bond and the adjacent peptide bond is known as the phi torsion angle (Φ), while the psi torsion angle (Ψ) is between the C–Cα bond and the adjacent peptide. Figure 1.5 The resonance in the peptide bond makes the carbonyl group and the amide   nitrogen in the same plane, and the steric hindrance from the side chains in    the cis form is shown. 10   G. N. Ramachandran and his co-workers, in 1963, introduced the Ramachandran diagram (detailed in section 1.2.2) to visualize the phi and psi torsion angles of amino acids in protein structure.18 Because of steric restriction from Cα and its side chains, Φ and Ψ torsion angles have their favoured and permitted regions corresponding to each amino acid.19  The genetic code holds the information needed to make proteins. It tells us that all protein-based amino acids are coded by a codon consisting of three consecutive nucleotides: the basic unit of DNA. The genetic information is transcribed into messenger RNA (mRNA) from DNA and then it is translated directly into the amino acids forming the proteins. There are four different nucleotides: adenine (A), guanine (G), thymidine (T) (replaced by uridine (U) in mRNA) and cytosine (C). Figure 1.7 shows the correspondence between the 64 possible three-base codons in messenger RNA and the 20 naturally amino acids. Ψ Φ Cα Cα Cα Figure 1.6 Phi torsion angle (Φ) and psi torsion angle (Ψ). 11   1.2.2 Secondary Structure  Next is the secondary structure, and the polypeptide chain at this level takes a highly regular form as opposed to random coils. The hydrogen bonds between N-H and C=O groups of the amino acids in the polypeptide backbone are mainly responsible for inducing the protein secondary structure.   Three general types of secondary structures are observed in a folded polypeptide chain. The most commonly found form is the alpha helix (Figure 1.8), which has a right-handed coiled conformation, though a left-handed alpha helix also can be formed.20 (Steric Figure 1.7  The genetic code. (Phe: phenylalanine, Met: methionine, Pro: proline, Ser:    serine, Thr: threonine, Trp: tryptophan, Tyr: tyrosine) 12 constraints favor the right-handed conformation because all protein forming amino acids except glycine have the L-configuration.) In an alpha helix, hydrogen bonds form between the carbonyl oxygen atom of each residue (n) and the amide nitrogen four residues (n+4) further along in the sequence. There are 3.6 residues per turn in an alpha helix, corresponding to a rotation of 100° per residue. For each peptide bond there is a dipole moment which comes from the amide resonance structure and the higher electronegativity of oxygen over carbon. The overall net dipole of an alpha helix is significant, because the dipole moments are aligned along the helical axis in the same direction.  Figure 1.8 Structures of right-handed alpha helix and the overall dipole moment within the helix (only the backbone atoms are shown). 13  The beta sheet, which is formed by backbone hydrogen bonding between segments of extended polypeptide chains, is the second most common form of secondary structure (Figure 1.9). Usually, more than two strands are involved in forming a beta sheet, and the hydrogen bonds between the strands link them side by side. There can be three possible arrangements of beta sheets: parallel, antiparallel and mixed sheets.   Finally, the beta turn, alternatively a hairpin turn, is a tight turn that reverses the direction of the polypeptide chain. It is stabilized by one or more backbone hydrogen bonds. This structure makes it possible for the compact folding of polypeptide chain.21  From the Ramachandran diagram (mentioned in section 1.2.1), we are able to see clearly that different secondary structures have their own preferred phi and psi regions. The observed values of phi and psi torsion angles from a predetermined protein structure should fall in the allowed regions (to reach the minimum of steric constraints) in the Ramachandran diagram (Figure 1.10). Figure 1.9 Beta sheets with parallel and antiparallel form. The hydrogen bonds are more linear in the antiparallel form than that in the parallel form. parallel antiparallel    14   1.2.3 Super Secondary Structure   Before moving on to the tertiary structure of protein, we first introduce the super secondary structure of proteins. Though Anfinsen pointed out that all the information of protein folding is present in its primary structure, it still remains challenging for scientists to predict how tertiary structure is formed directly from the amino acid sequence. The super secondary structure, which is simply a combination of a few secondary structure elements, is Figure 1.10 Ramachandran diagram, the general case of 81234 non-Gly, non-Pro, non-prePro residues. Reproduced with permission from Proteins 2003, 50, 444. Copyright 2003 Wiley-Liss, Inc. 15 introduced, for it is far easier to investigate how secondary structures fold into super secondary structure than into tertiary structure. With a deep understanding of the super secondary structure formation, the veil of protein tertiary structure will be gradually uncovered.  The simplest form of a super secondary structure consists of two alpha helices joined together by a loop (a peptide chain segment with no regular conformations), and it can bind with a DNA molecule22 or calcium cation.23 More complicated forms include greek keys (four adjacent antiparallel beta strands folded upon themselves), helix bundles, helix-sheets, etc. (Figure 1.11).    A helix bundle is a small protein form composed of several nearly parallel or antiparallel alpha helices (Figure 1.12). Each helix crosses the next at an angle of a) b) c) Figure 1.11 Three examples of the super secondary structure. a) Helix-loop-helix, b) Beta   strand-helix-beta strand, c) Greek key (four adjacent antiparallel beta strands   folded upon itself) 16 approximately -20°, so that the entire view of the bundle has a left-handed twist. The alpha helix involved in the helix bundle is always found to be amphiphilic. The hydrophobic side chains are buried inside the central cone, interacting with each other to stabilize the whole bundle. In contrast, the hydrophilic side chains are on the exterior (surface) of the bundle, interacting with the surrounding aqueous environment.      The helix bundle has commonly been found in nature. For example, the influenza virus uses a protein, which is a long three-helix bundle, to fuse into cells.25 Myohemerythrin, a four-helix bundle, is responsible for oxygen storage and it is found in most of the marine worms.26 And ara h 2, a five-helix bundle, is responsible for peanut allergen recognition; its five helices are held together by four disulfide bonds (Figure 1.13).27 Figure 1.12 Super secondary structure of a four-helix bundle with two interhelical turns.24 17   1.2.4 Tertiary Structure  Now we come to the tertiary structure of a protein. It is a protein structure in a globular form. It can be composed of alpha helices, beta sheets, or both, as well as of loops and links that have no regular conformation (Figure 1.14). The tertiary structure is stabilized by a large number of weak interactions.  Figure 1.13 The structure of ara h 2, a five-helix bundle held together by four disulfide bonds. Reproduced with permission from Allergy 2011, 66, 881. Copyright 2011 Clearnce Center, Inc.  18   Comparing the protein super secondary structure with the tertiary structure, we can see the difference between these two structures. The tertiary structure of a protein describes not only the space positions of its secondary structures, but also regions without secondary structure, and the secondary structures within it do not have to form a super secondary structure. The super secondary structure is a combination of several secondary structures and it does not include regions with no secondary structure; also, this folded pattern can be commonly found in many proteins. For instance, there are eight alpha helices in the protein of the globin family. These helices are not packed like helix bundles except the last two, which take a helix-turn-helix form.28  Figure 1.14 The tertiary structures of aplysia limacina myoglobin.29 19 1.2.5 Quaternary Structure  Finally, we come to the top of the protein structure hierarchy, quaternary structure. It describes how polypeptide chains with defined tertiary structure associate (noncovalently) with each other to form stable complexes, or oligomers. The force that enables proteins to associate specifically with each other comes from their irregular surface. It is not exclusively the weak bonds between the surfaces that do the work; hydrogen bonds and positive and negative charge interactions are also responsible for stabilizing quaternary structure. Figure 1.15 shows the protein structure of hemoglobin, an iron-containing oxygen-transport protein assembled by four globular protein units that exists in red blood cells.  Figure 1.15 The quaternary structure of horse deoxyhaemoglogin.30 20 1.3 Protein Folding  As we discussed above, globular proteins have four levels of structure. If we put a linear polypeptide chain (a globular protein, not a fibrous protein) at physiological temperature in an aqueous solution, it will not exist as it was: in a linear state! Instead, it will usually autonomously organize into a compact, well-defined three-dimensional structure or so-called native state (an operative and functional state of a properly folded protein) in less than a few seconds. (Although proteins without a fixed or ordered three-demensional structure (intrinsically disordered proteins, IDP) have been found and they are enriched in many signalling and regulatory functions.31) According to Anfinsen’s point of view, all the information that determines how the protein folds are encoded in its primary structure.  In most cases, a well-folded polypeptide chain is globular at physiological temperature in an aqueous solution. The internal core of the globule is mostly packed by amino acids with hydrophobic side chains, and these side chains are held together by van der Waals forces. The charged and the polar side chains are on the surface where they can interact with surrounding water molecules. In this way, the protein will fold into its native state, which is energetically stable.  When talking about the native state of a protein, we inevitably have to mention the folding funnel hypothesis, which assumes that the free energy of a protein achieves a minimum when the protein reaches its native state (Figure 1.16). 32, 33 The free energy landscape represents the free energy of a protein from its unfolded structure to its native structure. It illustrates that a protein does not have to follow a particular path to reach its native state, but instead, it may adopt various folding pathways that lead it towards the native 21 state.4, 34, 35   Several biophysical techniques have been used to explore protein folding, such as rapid mixing methods, 36 high-pressure NMR spectroscopy, 37 chemically induced nuclear polarization38 and others. Also, effort has been put into computational prediction of protein folding. In this method, scientists are able to predict how amino acid sequences fold into a three-dimensional structure computationally depending on a few assumptions. However, this method is still not accurate enough due to the uncertainty of these assumptions.39, 40   Figure 1.16 The funnel landscape depicts that the protein reaches a thermodynamically    stable state as it folds down the landscape. There are varous local minima    representing folded intermediates. Reprinted with permission from Polymer    2004, 45, 548. Copyright 2003 Elsevier Science Ltd.  22 1.4 Factors Contributing to Stability of Protein Folding  There are three major factors that greatly influence the protein folding process and make a protein reach its global energy minimum. These factors are hydrogen bonds, electrostatic interactions, and hydrophobic interactions.  Hydrogen bonds in proteins are formed when a hydrogen atom has a significant partial positive charge due to the high electronegativity of the atom (in this case, oxygen or nitrogen) to which the hydrogen atom covalently binds, and is attracted to a nearby partially negative charged atom (oxygen or nitrogen). The contribution towards protein stability for each intramolecular hydrogen bond is estimated to be 1.5 ± 1.0 kcal/mol.41  In a hydrogen bond, the donor atom is the one that is covalently attached to hydrogen while the acceptor atom is the non-bonded atom. Both the donor and the acceptor atoms have partial charges. When two fully charged atoms are involved, a salt bridge forms instead of a hydrogen bond. These are the electrostatic interactions in the protein folding process.  Last but not least, we look at the hydrophobic interactions. These are the major driving force in protein folding.42 Just like oil droplets dispersed in water, the hydrophobic side chains on a polypeptide will tend to clump together, expelling water as they cannot form hydrogen bonds with water molecules. For this reason, it causes the gain in solvent entropy, which outstrips the loss of entropy in the hydrophobic zone of the protein. This is why it is one of the major driving forces in the protein folding process.  When these three factors facilitate the protein to fold into an orderly, well defined form, the whole structure experiences a considerable conformational entropy decrease. This 23 is the chief destabilizing contribution that influences protein folding. Thus, we can see that there is a compensatory effect in the total energy balance of the protein folding process.  1.5 De Novo Protein and Template Assembled Synthetic Proteins  1.5.1 De Novo Protein Design  Synthetic de novo proteins are a class of molecules that mimic the native properties of natural proteins and provide us with an approach to explore the process of protein folding. The main goal in this area is to design artificial proteins in an attempt to unveil the protein-folding problem, which is to unlock the “hidden code” in the protein primary structure for forming tertiary structure. A further purpose of this strategy is to create novel protein molecules with functional activities for applications in a wide range of areas.43-48  Before designing a protein, it is a must to have a deep understanding of the properties of all the twenty amino acids, because their physical and chemical properties directly influence the way a protein folds. After decades of statistical analyses of numerous proteins with known conformation, scientists now have a better understanding of the role amino acids play in protein secondary structure.49-52  Some early de novo designed proteins were very successful in mimicking small super secondary structure. One case is the four-helix bundle designed by Ho and DeGrado (Figure 1.17). They first designed and chemically synthesized a stable helical tetramer in which 24 peptides were linked by loops, and then they demonstrated how hydrophobic interactions act in forming this four-helix bundle.53 Their peptide sequence was designed based on the minimalist approach11 in which the only non-polar residues are leucine and the only polar residues are glutamic acid and lysine. Thus, the resulting peptide is amphiphilic for it has both hydrophilic and hydrophobic side chains oriented in a certain order. Four identical peptides form a four-helix bundle where the hydrophobic side chains are buried within the internal core of the bundle while the hydrophilic side chains are on the surface of the bundle in contact with the aqueous environment.     Later Hecht et al. reported the de novo designed protein Felix, a nonrepetitive antiparallel four-helix bundle. In the 1990s, structures like helical hairpins54 and three-helix Figure 1.17 De novo protein design of a four-helix bundle by Ho and DeGrado. 25 bundles55 were designed in the manner of this method. What is more, an antiparallel beta sheet that has three or four strands was made possible through de novo protein design.56, 57  In 1994, Chemielewski and co-workers, in their work on disulfide cross-linked peptides, unexpectedly obtained a five-helix bundle.58 Their peptide sequence was designed based upon the criteria of DeGrado’s53. They used homocysteine (HCys) to cross-link the peptides. The helicity of this unexpected five-helix bundle was found to be concentration-independent, whereas the wild type peptide lost its helicity at low concentration. Also, their five-helix bundle is much more thermodynamically stable than its peptide counterpart. Our lab has turned to de novo designed peptide sequences that are idealized for a five-helix bundle, and this will be detailed in Chapter 2 of this thesis.  1.5.2 Template Assembled Synthetic Proteins  What we have seen in DeGrado’s de novo designed protein, a four-helix bundle, is that the four alpha helices are linked together by loops. There is another method that can hold together secondary structural units: grafting them onto a template molecule which is accessible by synthetic chemistry. A molecule that is qualified to be the template needs to facilitate intramolecular interactions between the fixed secondary structural units. Also the type, spatial arrangement, and number of the units to be fixed can be predetermined through this template molecule.59 Mutter and Vuilleumier introduced this approach in which the peptides were covalently linked to a rigid scaffold, resulting in a branched structure (Figure 1.18). 59,60 In this way, the unfavourable loss of entropy from folding is minimized. This simple system was termed as Template Assembled Synthetic Protein (TASP). 61 26    Mutter and his co-workers assembled four unordered peptide strands on cyclic and acyclic peptide-based templates to form four-helix bundles mimicking native-like properties of proteins.59, 60,62-64 The lysine residues, which have an amino group in their side chains, are covalently attached to the carboxyl termini of the unordered peptides. The numbers of the lysine residues (usually 4) and their spatial position predetermine that this peptide-based template is designed to form a four-helix bundle (Figure 1.19).   Figure 1.18 Template assisted synthetic protein. 27   Also, by taking advantage of this lysine-containing peptide-based template, a synthetic protein that mimics the pore-forming structure of a calcium channel was successfully designed and synthesized by Montal and co-workers. 65 They assembled four identical 22-residue peptides, which comes from the S3 segment of the fourth internal repeat of the DHP-sensitive calcium channel, to a 9-residue peptide-based template and they formed a four-helix bundle with single-channel conductance property.   W. Haehnel used the concept of template-assembled synthetic protein in a combinatorial synthesis of de novo proteins. 66 Their synthetic protein was orthogonally Figure 1.19 Cyclic peptide-based template assisted protein designed by Mutter and    co-workders. 28 assembled by small libraries of peptide building blocks. The template molecule is a cyclic decapeptide, in which there are four cys residues. The side chains of these four cys residues are protected by different protecting groups (Acm, StBu and Mmt), thus making it possible to construct a hydrophilic protein in a controlled way, which include: successively cleaving of the protecting groups on the template and coupling of amphipathic helices in a predefined orientation.  Aromatic ring-templates were also used in TASP design. Fairlie and coworkers reported a series of aromatic templates based on benzene, benzanilide and a cyclic octapeptide to synthesize four-helix bundles.67 They found that the size, shape and directionality of the rigid templates are extremely important to their capacity of enhancing the helicity of an amphiphilic peptide. In addition to lysine-containing peptide-based and aromatic ring-based templates, carbohydrates 68 and porphyrin rings 69 were also used to mimic the properties of particular natural molecules and showed good results.  The cavitand, named by Cram in 1982, represents a rigid organic molecule used as a scaffold in TASPs design. Our group has reported TASPs based on cavitand templates that show significant native-like structures in various helical bundle systems.70-73 Figure 1.20 shows different views of a [4]-cavitand (4 indicates the number of resorcinol units). We can see that the rigidity of its bowl-like macrocyclic structure results from the methylene units which bridge together the phenolic groups on the resorcinarene. The upper and lower rims of the cavitand can be modified for different functions and behaviours. The group R’ on the lower rim or feet, is usually modified to control the solubility of the TASPs.74, 75 The group R 29 on the upper rim is often functionalized as a nucleophilic group, such as a thiol, which is responsible for grafting the peptide onto the template.   Figure 1.21 shows the first cavitand synthesized by Cram and his co-workers. Cavitand 2a was synthesized in two steps: first, the acid-catalyzed condensation of resorcinol and acetaldehyde afforded the resorcin[4]arene 1; second, bridging of the eight hydroxyl groups with bromochloromethane in the presence of potassium carbonate afforded the methylene-bridged cavitand 2.76 Figure 1.20 An arylthiol cavitand. 30    Our lab focused mainly on the four-helix bundle, but larger TASPs like five- and six-helix bundles were also explored with expanded cavitand systems.77 Both benzylthiol and arylthiol cavitands are frequently used as templates in our research. We used the benzylthiol cavitand to synthesize the five member-ring cavitein, because only the [5]-benzyl cavitand is known (Figure 1.22); that is, the [5]-arylthiol cavitand is not known.  Figure 1.21 First synthesized cavitands by Cram. 31    The term “cavitein” is the combination of cavitand and protein, and is used to describe our TASPs, because peptides are grafted onto cavitand templates to form caviteins. Therefore, the cavitein is a subgroup of the whole TASP family.  1.6 Applications of Synthetic Biomolecules  1.6.1 General Applications of Synthetic Biomolecules  As mentioned earlier in this chapter, proteins have a wide range of functions within living organisms. One major type of biochemical function of proteins is catalysis. Catalysis requires specific intermolecular binding between the catalytic molecule and the substrate. The catalytic molecules have specific chemical reactivity, like proteasomes which can only degrade oxidized and ubiquitinated proteins (Figure 1.23a). 78,79 Binding is the other chief function that proteins have. Proteins may bind to other proteins in the case of a transporter or Figure 1.22 Two types of our frequently used cavitand. On the left is the four-member    ring arylthiol cavitand and the right one is the five-member ring benzylthiol   cavitand. 32 a receptor, or to some macromolecules such as DNA in the case of protein AIM2 (also known as “absent in melanoma 2”) which contributes to the defence against bacterial and viral DNA (Figure 1.23b). 80, 81 Thirdly, proteins sometimes operate as molecular switches, such as the Na+/H+ exchanger-1 (Figure 1.23c). 82, 83 Na+/H+ exchanger-1 is a ubiquitous plasma membrane protein that is regulated via its lipid-interacting domain. It functions as a molecular switch and requires conformational changes that depend on a delicate balance between structural stability and flexibility. Finally, some fibrous proteins can serve as structural components of cells and organisms. These structural proteins create strong materials such as silk, a protein polymer that is of interest for biomaterials and scaffolds for tissue engineering.84 33   Though nature offers us many multi-functional proteins, there is still a broad development space for scientists to improve the functional activities of natural proteins into a more efficient and functional state. Natural proteins are large molecules. However, scientists try to simplify their studies by exploring only a small part (“the active region”) of a protein that is responsible for its biological function. Therefore, it seems wiser to cut the “noisy” Figure 1.23 The crystal structures of several proteins. a) Depicts the structure of the 20S   proteasome from yeast in complex with the proteasome activator PA 26. b)    Depicts the structure of the AIM2 pyrin domain. c) Depicts the structure of    transmembrane segment IV of the NHE1 isoform of the Na+/H+ exchanger. 34 parts off and keep the functional residues. The synthetic biomolecule is thus a strategy that allows us to focus on the key part of a protein and add artificial modifications to create novel molecules with far better functions.85, 86  A newly synthetic integrin binding peptide created by Nortcliffe et al. is such an example.87 They artificially decorated the RGD (aspartic acid-glycine-arginine) peptide with different nitric oxide releasing groups and found that the binding ability of the furoxan-decorated RGD peptide to the integrin is much stronger than that of the wild type RGD peptide. They also found that it has anticancer properties. Synthetic biomolecules with functions such as ethanol biosensors88, 89 and antibacterial90 also have been reported. The following section reviews several recent studies on the applications of a few synthetic biomolecules as enzyme mimics and protein-protein interactions modulators.  1.6.2 Synthetic Biomolecules Mimicking the Functions of Natural Enzymes  1.6.2.1 Mimicking the Function of Peroxidase  Many synthetic supramolecules were designed to mimic the functions of natural enzymes with high selectivity in catalyzing biological transformations. 91,92 Synthetic enzymes are more stable and structurally simpler than their natural counterparts, like some de novo designed heme proteins with high peroxidase activities reported by Pavone et al. 93 Peroxidase is a type of enzyme that is widespread in many living organisms and it plays an important role in cell redox metabolism. Naturally existing peroxidases are large molecules 35 that contain more than 300 amino acid residues and are unstable. Pavone et al. designed a new artificial metalloenzyme, named MiniPeroxidase 3, which has a four-helix bundle structure.93 In the structure of MiniPeroxidase 3, the two peptide chains are covalently linked to a deuteroporphyrin molecule forming a helix-loop-helix/heme/helix-loop-helix sandwich arrangement. The stability of this artificial enzyme comes not only from its highly helical structure but also from the ligation between the histidine residue and the heme. What is more significant is that the size of this artificial molecule is 5 times smaller than that of the natural peroxidase (Figure 1.24).   Figure 1.24 Computer model of MP3. a) Helices with key residues depicted as sticks. b)    The sequence of the peptide sequence used in MP3. Reprinted with     permission from Chem. Eur. J. 2012, 18, 15962. Copyright 2012     WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. 36 1.6.2.2 Catalyzing a Retro-aldol Reaction  De novo designed proteins can also catalyze a retro-aldol reaction by mimicking an aldolase. In the case of fructose-bisphosphate adolase, the mimic catalyzes a reversible reaction that splits an aldol (fructose 1, 6-bisphosphate) into DHAP (dihydroxyacetone phosphate) and GAP (glyceraldehyde 3-phophate). Baker et al.94 used computational methods to select protein molecules with enzyme activities from a large number of possibilities. After constructing and characterizing the selected molecules, they found that these proteins displayed high retro-aldolase activity, with rate enhancement up to four orders of magnitude over the uncatalyzed reaction. Their most active designed models are on a jelly-roll scaffold or a TIM (triose phosphate isomerase)-barrel fold, in which there are structures such as imine-forming lysines, hydrogen-bonding residues, and a hydrophobic pocket, all of which are important to the enzymes’ activities of catalyzing the retro-aldol reaction (Figure 1.25).   Figure 1.25 A) A designed hydrophobic pocket encloses the aromatic portion of the substrate. B) & C) The active site for proton shuffling. Reproduced with permission from Science 2008, 319, 1390. Copyright 2008 American Association for the Advancement of Science. 37 1.6.2.3 Mimicking the Function of Hydrolase  The process of hydrolysis is ubiquitous and is of extreme importance to organisms. The hydrolysis of ATP (adenosine triphosphate) is involved in energy metabolism and storage. The hydrolysis of polysaccharides is the only way for polysaccharides to be absorbed by cells. Other substances such as esters, peptides and DNAs also can be hydrolyzed in specific conditions. One early study by Baltzer and co-workers was on the de novo protein design of a peptide that can catalyze hydrolysis and transesterification reactions of esters.95 Their polypeptide had 42 residues and was designed to fold into a hairpin helix-loop-helix structure (Figure 1.26). Two of these hairpin structures further folded into a dimer yielding a four-helix bundle. The hydrolytic ability comes from its eight histidine residues located in the reactive site of the folded polypeptide itself. These histidine residues can be grouped into two for they have two different pKa values ranging from 5 to 7. From this unique structure, they found that the four-helix bundle has hydrolysis activity 1140 times higher than that of the 4-methylimidazole catalyzed reaction. After a deep investigation of this dimer, they suggested that the reaction goes through a mechanism where the ester oxygens are bonded by protonated histidines which flank the unprotonated nucleophilic histidines in the transition state. Inspired by this work, our lab explored one of our histidine-containing TASPs to investigate its hydrolytic activity which will be detailed in chapter 3 of this thesis. 38   1.6.3 Synthetic Biomolecules in Protein-protein Interactions  Protein-protein interaction (PPI) is a process in which two or more proteins bind together to carry out biological functions. It occurs in almost every aspect of cell functioning such as signal transduction (the activity of potassium cation channel regulated by PPIs),96 regulation of gene expression (the expression of topoisomerase IIβ is regulated by Sp1 protein),97 and muscle contraction (PPI between cortactin and profilin-1)98. 1.6.3.1 Hot Spot-based Design of Synthetic Biomolecules  Protein-protein interactions mediate a large number of important regulatory pathways. Application of synthetic biomolecules are widely used in this field. These synthetic Figure 1.26 The structure of the helix-loop-helix monomer with the six histidine residues as the reactive site. Reprinted with permission from J. Am. Chem. Soc. 1997, 119, 11363. Copyright 1997 American Chemical Society. 39 molecules act as inhibitors that disrupt the original interactions between two proteins. The synthetic molecules are small molecules compared with proteins, because they contain only the active regions of proteins. Therefore, it is vital to detect the most active key region of the target protein. Hot spot-based design of synthetic biomolecules is one such strategy that helps us design molecules that mimic protein activities.   Wells and co-workers named a small number of residues of a protein a “hot spot” because these residues contribute to the majority of the binding free energy.99 Usually, these hot spots are in clusters and are in contact with each other to create hot regions.100 It is these hot regions that dominate the interactions between two proteins; only 10% of the residues in the binding site are involved (Figure 1.27). 101 Figure 1.27 illustrates hot spots and hot regions during PPI.  Figure 1.27 The region of a PPI interface. The projecting hot spot residue “1” binds into a pocket in the complementary surface region of the other protein. Reprinted with permission from J. Am. Chem. Soc. 2013, 135, 6251.  Copyright 2013 American Chemical Society. 40 1.6.3.2 Protein-protein Interactions in Hippo Signaling Pathway  The Hippo signaling pathway, which controls organ sizes in animals through the regulation of cell proliferation and apoptosis, is worth noting.102, 103 The PPI between proteins YAP (Yes-associated protein) and TEAD (Transcriptional enhancer factor domain) mediates the oncogenic function of YAP. A disruption to their interactions by an inhibitor may offer help towards treatment of YAP-related cancers. Hu and et al. designed and synthesized a cyclic peptide truncated from YAP that acted as an effective inhibitor of the YAP-TEAD interaction.104 They designed a truncated peptide from residues 87 to 96 of YAP, as that is the “hot region” of the YAP-TEAP interaction. In this way, they avoided using the original YAP protein, which is a large molecule. Modifications like L91A and D93A have been made to increase the hydrophobicity of the peptide to make it fit better in the hydrophobic pocket on the surface of TEAD protein. A disulfide bridge was made for peptide conformational constraint, which also led to its stronger binding affinity towards TEAD than the wild type YAP protein (Figure 1.28).  Figure 1.28 Formation of the disulfide bond of the cyclic peptide which inhibites PPI between YAP and TEAD. Reprinted with permission from Med. Chem. Lett. 2014, 5, 993. Copyright 2014 American Chemical Society. 41 1.6.3.3 Protein-protein Interactions in Clathrin-mediated Endocytosis  The clathrin-mediated endocytosis is a process that occurs in all eukaryotic cells. It involves the taking in of matter by a living cell through invagination of its membrane to form a vacuole with the help of a clathrin protein.105 The binding of clathrin to the membrane is mediated by the AP2 (activating protein 2) complex.105 In order to regulate endocytosis, Gorostiza and co-workers developed several photoswichable inhibitors of PPI between clathrin and AP2 based on the structure of the β-arrestin C-terminal peptide which can bind to the AP2.106 Their designing strategy was to photocontrol the helicity of the peptide because the more helical the peptide is, the more stable it binds to the AP2.107 The photoisomerizable crosslinker BSBCA (3,3’-bis(sulfonato)-4,4’-bis (chloroacetamido)azobenzene), which was responsible for reversibly changing the stability of the helix in different wavelengths of UV light, was covalently linked to the peptide through pairs of cysteines. One of their molecules, TL-1, induced a helical structure more readily at long wavelength UV light than at short wavelength, resulting in inhibiting the PPI between clathrin and AP2 (Figure 1.29). 42   1.4.3.4 Protein-protein Interactions in Programmed Cell Death   Protein-protein interactions also play an important role in regulating programmed cell death. Cells will commit suicide by activating an intracellular death program when they are no longer needed.108 One example of PPI in programmed cell death is between the Nef and CXCR4 proteins, both of which are responsible for HIV induced immunodeficiency. The Nef protein, an HIV encoded protein, is involved in uninfected CD4+ T cell apoptosis induced by HIV-1.109 When the Nef protein binds with the CXCR4 surface receptor, apoptosis of CD4+ T cells occurs, leading to immunodeficiency.110 Zhang and co-workers designed and synthesized a 15-residue peptide derived from the N-terminus of the viral Figure 1.29 Photoswichable peptide inhibitors with cysteine residues at different positions a) i and i+11 b) i and i+7. Reprinted with permission from Angew. Chem. Int. Ed. 2013, 52, 7705. Copyright 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. 43 macrophage inflammatory protein II (vMIP II) and it significantly blocked the Nef-induced apoptosis through tight binding to CXCR4.111  Our lab has now used a template assembled synthetic protein (TASP) to investigate protein-protein interactions in programmed cell death. We developed a Bak peptide-containing cavitein with an increased affinity to the Bcl-xL protein relative to the free Bak peptide, and this will be detailed in Chapter 4 of this thesis.  1.7 Thesis Aims and Research Overview  The first goal for this thesis is to design a peptide sequence that is favourable for forming a pentameric helical bundle. Since our group’s first work on using a cavitand as a template in de novo designed proteins in 1995,70 numerous helix bundles have been designed and synthesized, 70,71,112-114 in which there are three-helix bundles, four-helix bundles, five-helix bundles, and six-helix bundles. All the designed peptide sequences are idealized for a four-helix bundle. Since stable five-helix bundles are present in nature, 115-117 it means that it is not impossible to design a helix peptide that is more favourable for forming a pentameric bundle. If our purpose is fulfilled, it will further prove the maturity of our de novo protein designing strategy.  The second goal is to put our previously synthesized caviteins into application, to see if any of them are able to mimic an enzyme catalyzing one type of reaction. Though our lab has done years of research on the TASPs, no applications in the related chemistry or biology field have been pursued directly. Thus, it is of great interest and potential to go deep into this 44 area. Esterase mimicking is our first trial.  The third goal is to utilize our TASP strategy to design caviteins that may have application in PPIs. This is our second trial of exploring applications of our caviteins in the biological field.  The following chapters focus on my research which consists of two parts: Chapter 2 is on designing a peptide sequence which has the pentamer forming tendency, and chapter 3 and chapter 4 are on discovering the potential applications our TASP strategy (Figure 1.30). Chapter 5 will summarize and conclude the work throughout this thesis. 45 Figure 1.30  Aim of this thesis     46 Chapter 2 A Potentially Stable Five-helix Bundle     Cavitein  2.1 Introduction  The biological functions of proteins are determined by their structures. As we discussed in chapter 1, proteins are polymers of amino acids and have nearly infinite potential structures. How they fold, and into what structures they fold are the central questions of the protein folding problem. One approach toward reconciling the protein folding problem is to reduce its complicated structure by creating a simpler system. It is the tertiary structure that is difficult to predict. Instead of focusing on the tertiary structure, we are more capable of creating a super secondary structure. Alpha-helixes and beta-sheets are the two main secondary structural units involved in assembling a super secondary structure. It is relatively easy to create a secondary structural unit, but it is difficult to organize them. Thus, our group uses a template to organize secondary structural units. One common pattern of super secondary structure of proteins is the alpha-helix bundle, and this one is what we have mimicked.  A cavitand is a rigid bowl-shaped organic macrocyclic molecule that is used as a template to assist in the formation of an alpha-helical peptide bundle. One main advantage of using a cavitand as the template in designing de novo (from scratch) proteins is that it acts as a scaffold and drastically reduces the number of available conformations of the final protein. 47  As Isaac Newton once said: if I have seen further, it is by standing upon the shoulders of giants. The previous generations in our lab offered us precious experience in this area: Dr. Gibb and Dr. Mezo were the first ones working on the cavitand-based de novo protein (a four-helix bundle). They used an aryl-thiol cavitand with methyl groups as feet as the template, and only two atoms (-SCH2-) were used as the linkage between the peptide and the cavitand.70 Later, the linker’s significant influence towards the native-like property of the four-helix bundle was further studied by Dr. Mezo and Dr. Seo. It was found that methylene (at the α-possition of the resorcinol) variants were more characteristic of molten globules, whereas glycine (linked to arylthiol cavitand) variants aided in producing native-like structure.72 They optimized the number of glycine residues in the linker to two in order to obtain a native-like structure.113 Dr. Seo also synthesized a series of helical bundles with different ring sizes (four-, five- six-helix bundle) and demonstrated that their originally designed peptide sequence was indeed superior for a tetrameric bundle in terms of thermodynamic stability and structural specificity compared with larger ring size bundles.112 More recently, Dr. Freeman has successfully generated a crystal structure of a four-helix bundle cavitein dimer.118 Later, he investigated the monomer-dimer equilibria of the TASP systems through conformationally constrained designs such as mutations of the peptide sequence, the addition of a metal ion (nickel), and the incorporation of a disulfide bond.114  All our early studies were focused on the four-helix bundle and our strategy of designing a peptide sequence and mimicking its native-like property is sophisticated. However, little has been done in designing a peptide sequence that is idealized for a five-helix bundle. Since the five-helix bundles are present in nature (though they are less common than the four-helix bundles) such as residues 6-85 of the λ repressor protein117 and residues 48 1815-1973 of the talin which is a rod-shaped protein, 119 we are confident that it is not impossible for us to design a peptide sequence that is more favourable for forming a pentamer than a tetramer.  2.2 Results and Discussion  2.2.1 Cavitein Design   Any alpha-helical peptide with the tendency to form a helical bundle must be amphiphilic, because the helical bundle cannot be formed without either the hydrophobic side chains or the hydrophilic side chains of the residues. The hydrophobic effect which is one of the main driving forces that stabilizes the entire system, comes from the inner part of the helical bundle where all the hydrophobic residues interact with each other.   For example, one of our helical forming peptide sequences (EELL KKL EELL KKG) shows highly stable native-like structure as a four-helix bundle. It is difficult to distinguish the hydrophobic region and hydrophilic region from its linear written form, but it will be much clearer from its wheel model. (Figure 2.1)  Because an alpha helix has 3.6 residues per turn, which corresponds to a rotation of 100° per residue, we put the peptide sequence in a wheel model. In the wheel model, it is easy to see that the hydrophobic residues are not separated. Rather, they are positioned in a specific region as are the hydrophilic residues. 49    In Figure 2.1, the five leucine residues comprise the hydrophobic region of the peptide and it is this region where the hydrophobic effect occurs. We can see the whole hydrophobic region of a four-helix bundle on the left part of Figure 2.2: the four grey circles are the view of the peptides from the amino-terminal end, and the squared part is the hydrophobic region of the four-helix bundle. On the right is a five-helix bundle, whose hydrophobic region is marked by a pentagon. Comparing these models, we noticed that the hydrophobic region per peptide of the pentamer is larger than that of the tetramer because the angle θ2 is obviously wider than θ1. This stimulated us to increase the ratio of the hydrophobic to hydrophilic residues in our originally designed peptide sequence, which Figure 2.1 Wheel model for a peptide with sequence EELL KKL EEL KKG, where    hydrophobic residues are marked by square and hydrophilic residues are    marked by circle. 50 favors a four-helix bundle. In this way, we hope that the newly designed peptide sequence will be idealized for a five-helix bundle.   2.2.2 Selection of Cavitand  Among all the cavitands studied in our group, only benzyl rimmed cavitands have been synthesized in expanded ring sizes.77 Figure 2.3 displays models of four different sizes of benzyl cavitands (from 4 to 7) and we can see that for ring sizes larger than five, they become flattened and not as “circular” as the four- or five-member rings. The methyl groups on the aromatic ring can be functionalized to methylthiol groups for forming the linkage with the peptides.112 The [4]-benzylthiol cavitand has been shown to produce caviteins with native-like structures by our previous group members, and lower stability has been shown in Figure 2.2 Comparison of the hydrophobic regions between the four-helix bundle (left)    and the five-helix bundle (right). 51 caviteins with larger ring-size benzylthiol cavitands using the same peptide sequence idealized for the four-helix bundle.112    2.2.2.1 Synthesis of Benzylthiol Cavitand  The syntheses of the [4]-benzylthiol cavitand and [5]-benzylthiol cavitand have been previously reported by Dr. Naumann et al. What follows is a brief description of the Figure 2.3 Energy minimized structures of [n]cavitands. Reproduced with permission from Chem. Eur. J. 2001, 7, 1639. Copyright 2001 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, Fed. Rep. of Germany. 52 synthesis of [5]-benzylthiol cavitand with a few modifications (Figure 2.4).77 Unlike making a [4]-cavitand, which is thermodynamically controlled, the first step in synthesizing the [5]-benzylthiol cavitand is kinetically controlled and requires only 30 minutes to yield the resorcinarene mixture 3n. Bridging of the hydroxyl groups with bromochloromethane in the presence of potassium carbonate after refluxing for 10 hours afforded the methylene-bridged benzyl cavitand mixture 4n. [5]-benzyl cavitand (5) was separated from mixture 4n. (yield: 4.3%) Next, bromination of [5]-benzyl cavitand (5) with recrystallized N-bromosuccinimide (NBS) afforded [5]-benzylbromo cavitand 6. Compound 6 was then reacted with thiourea in DMF solution. After steps of workup and purification, the final [5]-benzylthiol cavitand 7 was afforded with a yield of 12%. 53      Figure 2.4 Synthesis of [5]-benzylthiol cavitand from 2-methyl resorcinol and     diethoxymethane. 54 2.2.3 Peptide Sequence Design and Synthesis   The peptide sequences designed in this study (Table 2.1) are based on our previous group member Dr. Seo’s sequence, CGGG-EE LLK KLE ELL KKG, which was idealized for a four-helix bundle and had high stability and native-like properties (Figure 2.5). 112 This peptide sequence was originally designed based on DeGrado’s minimalist approach, 11 in which the only non-polar residues are Leu and the only polar residues are Glu and Lys, and these two kinds of amino acids are distributed 3-4 residues apart in the sequence.  Table 2.1 Designed peptide sequences compared with original peptide. Peptide Peptide Sequence n* m** Original S-pyridyl-CGGG-EELLKKLEELLKKG-NH2 0.63 4 L S-pyridyl-CGG-AEELLKLLLELLKKG-NH2 1.7 3 M S-pyridyl-CGG-ALKLLEELLKLG-NH2 1.5 2 S S-pyridyl-CGG-ALKLLLEG-NH2 2 1 * n: the ratio of non-polar residues to polar residues. ** m: the number of salt bridges.  55    Later, research on the linker of the original sequence showed that a two-glycine-residue linker is optimal in obtaining a native-like structure of a four-helix bundle.113 Thus, the linkers of our designed peptides all have two Gly. Following the studies of the linker, an Ala was added to the sequence, for the Ala right after the linker residues helps to place the linker at the hydrophobic region of the helix. It also increases the peptide’s tendency to form a helix.113 The C-terminus of the sequence was amidated so that the repulsion between charges on the helices would be minimized and the macrodipole effect would be reduced.70 It was also capped with a Gly residue to stabilize the last turn of the helix. Figure 2.5 Wheel model for a peptide sequence that is more favorable in forming a    four-helix bundle. 56  At the beginning of each peptide sequence, there is a Cys residue. It is responsible for forming a disulfide bond with the template cavitand. Because peptides with Cys residues tend to dimerize during the process of linking to the template molecule, they need to be activated by reacting with 2,2’-dipyridyl disulfide (DPDS) before linking to the cavitand (Figure 2.6).   For the purpose of increasing the ratio of non-polar residues to polar residues, a few mutations in the peptide sequence have been made. On the interface between the hydrophobic region and hydrophilic region of the original peptide sequence, there is a Lys residue and Glu residue. Both of them are mutated to non-polar Leu residues as shown in Figure 2.7, so that the hydrophobic region is increased and the number of salt bridges falls from 4 to 3. We name this first mutated peptide sequence L for it is the Longest among the sequences we have designed. Figure 2.6 Activation of peptide with 2,2’-dipyridyl disulfide. 57    The designed peptide L (shown in Figure 2.7) has around 3.6 turns in its helical structure (3.6 residues per turn in an alpha helix). The next designed peptide sequence M (Middle) is shortened by around one turn (it is 2.8 turns in total) after taking away one salt bridge (a Lys and a Glu) and one Leu residue from peptide L. The ratio for the non-polar to polar residues in peptide M is similar to L (L: 1.7; M: 1.5) (Figure 2.8). Figure 2.7 Wheel model for peptide L, in which the original residues E8 and K6 are    replaced by leucine residues to increase the hydrophobic region of the new    helix. 58   As for peptide S (Short), it is further shortened by taking away one salt bridge and two Leu residues from peptide M; it has 1.7 turns in all (Figure 2.9). Therefore only one salt bridge is left in this sequence, and the non-polar/polar ratio increases from 1.5 (in M) to 2. Figure 2.8 Wheel model for peptide sequence M. 59   2.2.4 Cavitein Synthesis  The synthesis of the cavitein was achieved by reacting the benzylthiol cavitand with an excess of activated peptide (6 equivalents for [4]-benzylthiol cavitand and 8 equivalents for [5]-benzylthiol cavitand) in the presence of diisopropylethylamine (DIPEA) in dimethyl formamide (DMF) solution at room temperature for 6 hours (Figure 2.10), after which the resulting cavitein was purified by reverse phase HPLC.   Figure 2.9 Wheel model for peptide sequence S. 60   Nomenclature for the caviteins is shown in Table 2.2. For example: L5 denotes a five-helix bundle cavitein made from peptide L, and M4 denotes a four-helix bundle cavitein made from peptide M. The molecular weights of these caviteins were confirmed by ESI and MALDI mass spectroscopy. Figure 2.11, Figure 2.12 and Figure 2.13 show HPLC purification of these caviteins and the assessment of purity by analytical HPLC. Table 2.2     Nomenclature for the caviteins and their molecular weights. Peptides  Cavitands [4]-benzylthiol cavitand [5]-benzylthiol cavitand L L4: 8589.0 L5: 10734.3 M M4: 7107.2 M5: 8884.1 S S4: 5172.9 S5: 6464.4   Figure 2.10 Synthesis of caviteins L4, L5, M4, M5, S4 and S5. 61     Figure 2.11 Analytical reverse phase HPLC traces of purified caviteins a) L4 and b) L5.    Their purity was assessed by an analytical reverse phase column using a    H2O:ACN gradient (40% H2O to 20% H2O over 15 minutes). Figure 2.12 Analytical reverse phase HPLC traces of purified caviteins a) M4 and b) M5.   Their purity was assessed by an analytical reverse phase column using a    H2O:ACN gradient (45% H2O to 30% H2O over 15 minutes) 62   2.2.5 Characterization of the Caviteins 2.2.5.1 Far-UV Circular Dichroism Spectra   Circular dichroism (CD) signals come from the differential absorption of left and right circularly polarized light, and this technology is widely used in assessing and quantifying secondary structural elements that may be present within a protein system. Different types of secondary structures display their unique characteristic spectra. As for an alpha helix, it exhibits two distinctive negative bands near 222 nm and 208 nm and one positive band near 195 nm. If it is a beta sheet, a single minimum band at 218 nm and a maximum band at 200 nm will be observed. When the band reaches its negative minimum at 200 nm, it most likely exists as a random coil. The wavelength falls between 190 nm to 240 nm, which is the far-UV region. Absorption in this region is dominated mainly by the Figure 2.13 Analytical reverse phase HPLC traces of purified caviteins a) S4 and b) S5.    Their purity was assessed by an analytical reverse phase column using a    H2O:ACN gradient (57% H2O to 47% H2O over 10 minutes) 63 peptide bond. For example, the positive maximum band at 195 nm and the negative minima band at 208 nm are due to π → π* amide transitions whereas the negative band at 222 nm is due to the n → π* transition which comes from the amide hydrogen bonded helix.  The secondary structure of caviteins L4 and L5, (Figure 2.14) M4 and M5, (Figure 2.15) S4 and S5 (Figure 2.16) were investigated by far-UV circular dichroism. All these caviteins were found to be alpha helical, because their spectra all exhibit the three characteristc bands for an alpha helix: two negative bands near 222 nm and 208 nm respectively and one positive maximum band near 195 nm.   The CD experiments were conducted at both low and high concentrations (5 µM and 50 µM respectively) for the caviteins, and the spectra in both conditions overlapped. Therefore, only high concentration spectra are shown below. The concentration independent CD spectra suggest that the caviteins in the solution exist as a monomeric species instead of being aggregated.72  Helix-helix interactions may exist if [θ]222/[θ]208 is above 1.120 In our cases, all the [θ]222/[θ]208 ratios for the caviteins are below 1; however, it still does not mean that there are no helix-helix interactions in the caviteins. The absorption not only comes from the peptide part, but an indused signal can come from the cavitand chromophore as well.121  64  Figure 2.14 CD spectra of cavitein L4 and cavitein L5. The spectra were acquired at 25 ℃    in 50 mM phosphate buffer at pH 7.0 using a cavitein concentration of 50 µM. Figure 2.15 CD spectra of cavitein M4 and cavitein M5. The spectra were acquired at 25 ℃    in 50 mM phosphate buffer at pH 7.0 using a cavitein concentration of 50 µM. 65   Table 2.3 lists the percent helicities for all the involved caviteins. We noticed that the percent helicity values of the four-helix bundles and five-helix bundles are similar for the same peptide-based caviteins. The percent helicity of peptide M-based cavitein is lower than that of peptide L-based cavitein, and peptide S-based cavitein has the lowest percent helicity. The length of the peptide sequence is the main factor that influences the percent helicity in this situation (peptide L has 3.6 helical turns in its sequence, peptide M has 2.8 turns and peptide S has 1.7 turns).122  The percentage of the alpha-helicity was calculated using the following equation: Helicity (%) = 100 * ([θ]222 - 3000) / -39000 Figure 2.16 CD spectra of cavitein S4 and cavitein S5. The spectra were acquired at 25 ℃    in 50 mM phosphate buffer at pH 7.0 using a cavitein concentration of 50 µM. 66 where the reference values of [θ]222 = 0 and [θ]222 = -36000 (deg cm2 dmol-1) correspond to 0% and 100 % helicity respectively.123 However, any aromatic structures (in our case the cavitand chromophore) are able to influence the absorption at [θ]222.121 Thus, interpretation concerning percent helicity is not an accurate way to describe this situation, but its trend is more relevant than the absolute values. Table 2.3     Percent helicity for all the caviteins. Cavitein [θ]222 (deg cm2 dmol-1) Helicity (%) L4 -15803 48 L5 -16582 50 M4 -12906 41 M5 -13754 43 S4 -11065 36 S5 -10166 33   2.2.5.2 GuHCl Denaturation Experiments  Denaturation experiments using GuHCl were performed to investigate the thermodynamic conformational stabilities of the caviteins. Since the CD signal at 222 nm results from the n → π* transition in the amide hydrogen bonded helix, the loss of this signal indicates the degree of denaturation.  We use ∆G°, the standard free energy of unfolding of the protein, to indicate the stability of the native state. The denaturation of the protein is a two-state unfolding process: 67 N (native)  ⇌   D (denatured)  By incrementally adding the denaturant to force the equilibrium rightwards, the free energy of the native state in the absence of denaturant, ∆G°H2O, can be obtained from the free energy of the native folded protein via extrapolation.124  The GuHCl denaturation curves of caviteins L4 and L5 are shown in Figure 2.17. The experiments were conducted at both high and low concentrations of cavitein(40 µM and 4 µM respectively) and the denaturation curves overlapped, which suggest that the caviteins denature as monomers. We can see that these two caviteins are extremely stable in the presence of GuHCl. Both of them are slightly denatured in 6 M GuHCl solution. The denaturation curves for caviteins L4 and L5 cannot be completed as they are not fully denatured even in 8 M GuHCl solution. However, according to the trend of these two curves, we speculate that cavitein L5 would be more thermodynamically stable than L4, because the curve of L4 declines faster than that of L5 in the 6 M to 8 M (GuHCl concentration) region. Therefore, to reach the same denaturation percentage, cavitein L5 requires a higher concentration of GuHCl than that of cavitein L4, which normally leads L5 to have a lower negative value of ∆G°H2O. 68   Can caviteins L4 and L5 be further denatured? The GuHCl denaturation experiments at elevated temperatures were performed to answer this question (Figure 2.18). We measured the CD absorptions at 222 nm of caviteins L4 and L5 solutions in the presence of 8 M GuHCl from 20 ℃ to 95 ℃, and found that both of the caviteins denatured further as the temperature rises, but did not fully denature even in 8 M GuHCl at 95℃. Figure 2.17 GuHCl dentaturation of cavitein L4 and cavitein L5. Each curve was    acquired at pH 7.0 in 50 mM phosphate buffer at 25 °C at a cavitein    concentration of 40 μM. 69   GuHCl denaturation experiments were also performed for the shortened caviteins M4 and M5 under the same conditions (Figure 2.19). Because of peptide M’s shortened helical structure (3.6 turns in peptide L; 2.8 turns in peptide M), both of the peptide M-based caviteins can be denatured to a further degree than peptide L-based caviteins. However, they are still not completely denatured in 8 M GuHCl solution. Nevertheless, according to the trend of these two curves, it is obvious that cavitein M5 is more thermodynamically stable than cavitein M4. As we can see from the incomplete diagram, cavitein M4 almost reaches complete denaturation while cavitein M5 is still half way folded in 8 M GuHCl solution. Figure 2.18 Denaturation curves of cavitein L4 and cavitein L5 recorded in elevated    temperatures in 8 M GuHCl solution. 70   For the shortest peptide-based caviteins S4 and S5, GuHCl denaturation experiments were also performed under the same conditions (Figure 2.20). Because peptide S only contains 1.7 helical turns, S4 and S5 are less stable than the L and M series. The denaturation curves are complete for them and the free energies of the native state in the absence of denaturant, ∆G°H2O, were determined and are shown in Table 2.4, along with ∆G°H2O of caviteins with peptides that are idealized for forming four-helix bundles.  Figure 2.19 GuHCl dentaturation of cavitein M4 and cavitein M5. Each curve was    acquired at pH 7.0 in 50 mM phosphate buffer at 25 °C at a cavitein    concentration of 40 μM. 71   ∆G°H2O of cavitein S4 is -3.8 kcal/mol (-0.95 kcal/mol per helix) and ∆G°H2O of caivtein S5 is -3.4 kcal/mol (-0.68 kcal/mol per helix). Compared with the original peptide sequence in different ring-sized systems studied by Dr. Seo,112 where ∆G°H2O of the four-helix bundle is -9.4 kcal/mol (-2.4 kcal/mol per helix) and ∆G°H2O of the five-helix bundle is -5.4 kcal/mol (-1.1 kcal/mol per helix), the energy gap (either the whole cavitein or per helix) between our four-helix bundle and five-helix bundle is much smaller than Dr. Seo’s.   Figure 2.20 GuHCl dentaturation of cavitein L4 and cavitein L5. Each curve was    acquired at pH 7.0 in 50 mM phosphate buffer at 25 °C at a cavitein    concentration of 40 μM. 72 Table 2.4 Comparison of ∆G°H2O values between peptide S-based caviteins and caviteins ffffffffffffffffffwith peptides that are idealized for forming four-helix bundle. Cavitein ∆G°H2O (kcal mol-1) Per helix ∆G°H2O (kcal mol-1) S4 -3.8 ± 0.2 -0.95 ± 0.05 S5 -3.4 ± 0.2 -0.68 ± 0.04 four-helix bundle* -9.4 ± 0.7 -2.4 ± 0.2 five-helix bundle* -5.4 ± 0.6 -1.1 ± 0.2 *four-helix bundle and five-helix bundle are caviteins with peptides that are idealized for forming four-helix bundle synthesized by Dr. Seo.112   The shortened peptide sequence does lower the thermodynamic stability of the caviteins which makes it possible for us to quantify and compare their stability; however, ∆G°H2O values of the peptide S-based caviteins do not suggest that peptide S is favourable in forming a five-helix bundle even though the energy gap between them is narrowed.  2.3 Chapter Summary and Conclusion  In this chapter, we designed three peptide sequences that were designed to be more favorable in forming a five-helix bundle than a four-helix bundle. The first designed peptide L-based caviteins, in which the peptide L contains 18 amino acids and has 3.6 helical turns, were found to have extremely high thermodynamic stabilities even at high temperature. The second designed peptide M-based caviteins, in which the peptide M contains 15 amino acids and has 2.8 helical turns, were also not fully denatured even by 8 M GuHCl. However, what is important is that both caviteins L5 and M5 appear to show higher thermodynamic 73 stabilities than their four-helix bundle counterparts, though we are not able to quantify the free energies of their native states due to their incomplete denaturation curves.  As for the peptide S-based caviteins, in which the peptide S contains 11 amino acids and has 1.7 helical turns, they were fully denatured by GuHCl and their ∆G°H2O values were determined. Peptide S is slightly more favourable in forming a four-helix bundle than a five-helix bundle, but the free energy gap between its five-helix bundle and four-helix bundle is narrowed compared to the systems where the peptide sequence was designed to favour a four-helix bundle.  Thus, we conclude that our strategy of designing a peptide that is more favourable in a five-helix bundle than a four-helix bundle by increasing the hydrophobic region in the helix is reasonable and partially successful. What disappoints us is that we cannot determine the free energies of peptide L-based and peptide M-based caviteins, though these two five-helix bundles appear to have higher thermodynamic stabilities than those of the their four-helix counterparts. However, TASPs using a cavitand as the template do provide us with a meaningful way to study the protein folding problem, and there is no other way to force a single peptide sequence into both a four- and a five-helix bundle and quantify their free energy difference.     74 2.4 Experimental  2.4.1 Cavitand Synthesis  All the chemicals that are involved in the synthesis of the benzylthiol cavitand were reagent grade and purchased from Sigma-Aldrich. N-bromosuccinimide was recrystallized in water and was dried in reduced pressure in the presence of P2O5. All the solid reagents and products were dried in reduced pressure with P2O5 before use. Degassed water, acetic acid solution and DMF (dimethylformamide) were prepared through freeze-pump-thaw.   The synthesis of benzylthiol cavitand from 2-methyl resorcinol and diethoxymethane contains four steps. The detailed procedures have been previously reported by Dr. Naumann (Synthesis of the [5]-methyl rimmed cavitand) and Dr. Seo (Synthesis of [5]-benzylthiol cavitand)77,112 1D 1H NMR spectra of all the products were observed on Bruker AVANCE 300 MHz spectrometers at 298 K. The molecular weights were confirmed by Waters LC-MS equipped with ESCI ion source. We made a few modifications in the recrystallization steps of obtaining [5]-benzyl rimmed cavitand: The volume of each solvent that precipitates different ring sizes of cavitands is optimized based upon Dr. Naumann’s procedures.77 The first portion of ethyl acetate (150 mL) was added to the solid to dissolve mainly the [4]-benzyl cavitand after the mixture was purified by column chromatography. After filtration, chloroform (80 mL) was added to the filtered solid which mainly contains [5]-benzyl cavitand and [6]-benzyl cavitand to dissolve most of the [6]-benzyl cavitand. After a second filtration, the solution was 75 condensed to 40 mL and then it was stored in the fridge overnight. After a third filtration, the filtrate at this stage was reduced to 15 mL and then 50 mL of ethyl acetate was added. [5]-benzyl cavitand was precipitated after the solution was refrigerated for 30 minutes. 2.4.2 Peptide Synthesis 2.4.2.1 General Procedures of Peptide Synthesis  All the peptide syntheses were performed on a CS Bio (Menlo Park, CA) 136XT peptide synthesizer. Solvents used in the synthesizer like N-methylpyrrolidone (NMP) and DMF, and all the amino acids were purchased from Advanced Chemtech (Louisville, KY). Other chemicals involved were from Sigma-Aldrich. Rink’s amide resin was used to afford C-terminal amides.125 Peptide was synthesized on a 0.25 mmol scale. Standard Fmoc protected amino acids were used during the synthesis with acid labile protecting groups for alanine, cysteine, leucine, glycine, lysine and glutamic acid side chains. The following amino acids were used: Fmoc-L-Ala-OH, Fmoc-L-Cys(Trt)-OH, Fmoc-L-Leu-OH, Fmoc-L-Gly-OH, Fmoc-L-Lys(Boc)-OH and Fmoc-L-Glu(OtBu)-OH.   Below is the standard coupling cycle for the Fmoc strategy protocols: First, piperidine (20% in NMP) deprotects the amino groups by removing the Fmoc groups. The second step involves the activation of the amino acid in the presence of diisopropylethylamine (DIPEA) plus 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate (HBTU) and 1-hydroxybenzotriazole (HOBT). After the activation of amino acid, it was then coupled to the peptide chain.  76  After the synthesis was done in the synthesizer, the resin was washed with DCM. The N-terminus of the peptide (the C-terminus was linked to the resin) was acetylated with acetic anhydride before it was cleaved from the resin. The cleavage solution contained a mixture of TFA/H2O/ethane dithiol in a 90:5:5 ratio. The cleavage process lasted 2 hours. Afterwards, the peptide was purified by reversed-phase HPLC using a linear gradient of acetonitrile (0.05% TFA) and water (0.1% TFA) on a Waters C-18 Delta Pak column (300 × 19 mm2, 300 Å, 15 µm) at a flow rate of 10 mL min-1. The fraction signal was monitored by a UV detector at 229 nm to detect the amide chromophore. The fraction solution from HPLC purification was frozen and the solvent was removed by lyophilization. Analytical reversed-phase HPLC (Waters Delta Pak C-18 column (300 × 3.9 mm2, 300 Å, 15 µm)) was used to assure the purity of the peptide by the observation of a single peak. The identity of the peptide was confirmed by mass spectrometry (ESI-MS) with a Bruker Esquire to analyze the molecular weight of the peptide using electrospray ionizaiton as the ion source.  2.4.2.2 Syntheses of Peptides: L, M and S  After the free N-terminus of the peptide linked resin (~0.25 mmol) was collected from the synthesizer and washed by methanol, it was reacted with acetic anhydride (6 mL) in 2 mL of dried NMP solvent for 2 hours at room temperature. After filtration and washing by DCM, the peptide linked resin was treated with 10 mL of ice cold TFA/H2O/ethane dithiol (90% TFA, 5% H2O) solution for approximately 2 hours. After the peptide was cleaved from the resin and dissolved in this mixed solution, the resin was filtered off and the filtrate was transferred to a round bottom flask. Its volume was then reduced to approximately 5 mL in 77 vacuo. This concentrated cleavage solution was slowly pipetted into ice cold diethyl ether to precipitate the peptide. The solid peptide was separated from the ether by centrifugation (3000 rpm) for 5 minutes. The steps of washing with diethyl ether and centrifugation were repeated for another two times to remove TFA and small molecular weight impurities. The remaining peptide pellet was dissolved in water with a little acetic acid to speed up the dissolving process. RP-HPLC was performed after the peptide solution was filtered by a 0.45 µm pore disposable filter. The resulting peptide was a white cotton-like solid after lyophilization (L: 155 mg, yield: 27%; M: 134 mg, yield: 33%; S: 144 mg, yield: 51%). MS (L peptide before activation): m/z 1969.8 MS (M peptide before activation): m/z 1598.6 MS (S peptide before activation): m/z 1115.1  To activate the peptide with 2,2’-dipyridyl disulfide, the peptide (25 µmol) was dissolved in 3 mL methanol and 2,2’-dipyridyl disulfide (100 µmol) was added. The reaction mixture was stirred at room temperature for 1 hour under the argon protection. The methanol was then reduced to approximately 1 mL in reduced pressure and the remaining solution was pipetted into ice cold diethyl ether to precipitate the peptide. The solid peptide was separated from the ether by centrifugation (3000 rpm) for 5 minutes. The steps of washing with diethyl ether and centrifugation were repeated for another two times to remove small molecular weight impurities. The remaining peptide pellet was then dissolved in water with a little acetic acid to speed up the dissolving process. RP-HPLC was performed after the peptide solution was filtered by a 0.45 µm pore disposable filter. The resulting peptide was a white 78 cotton-like solid after lyophilization (L: 34 mg, yield: 65%; M: 25 mg, yield: 58%; S: 24 mg, yield: 78%). MS (L peptide after activation): m/z 2078.8 MS (M peptide after activation): m/z 1708.6 MS (S peptide after activation): m/z 1224.1 2.4.3 Cavitein Synthesis 2.4.3.1 General Procedures of Cavitein Synthesis  All the reagents used for the syntheses of the caviteins were reagent grade and purchased from Sigma-Aldrich. The solvent for the reaction was HPLC grade N, N-dimethylformamide (DMF). It was freeze-pump-thawed under argon atmosphere and dried by 4 Å molecular sieves. The crude caviteins were purified by reversed-phase HPLC and their purities were assessed by analytical RP-HPLC. Both MALDI-TOF (Matrix Assisted Laser Desorption Ionization) and ESI were used to confirm the identities of the caviteins. Molecular weights determined by MALDI were acquired on a Bruker Biflex IV. The matrix used was 2,5-dihydroxybenzoic acid (DHB).     79 2.4.3.2 Syntheses of Caviteins Cavitein L4  In 1 mL of dried and degassed DMF solvent, [4]-benzylthiol cavitand (1.4 µmol) and purified activated peptide (6 equiv, 8.4 µmol) were dissolved. DIPEA (40 µmol, 7 µL) was added to the solution. The solution was bubbled by argon gas at room temperature overnight. The reaction solution was condensed in vacuo to a volume of around 200 µL. The solution was then diluted with H2O and was filtered through a 0.45 µm pore disposable filter. The crude cavitein was purified by RP-HPLC. A white solid was obtained after lyophilization. (7.1 mg, yield: 58%) MS (cavitein L4): m/z 8589.0  Cavitein L5  In 1 mL of dried and degassed DMF solvent, [4]-benzylthiol cavitand (1.4 µmol) and purified activated peptide (6 equiv, 8.4 µmol) were dissolved. DIPEA (40 µmol, 7 µL) was added to the solution. The solution was bubbled by argon gas at room temperature overnight. The reaction solution was condensed in vacuo to a volume of around 200 µL. The solution was then diluted with H2O and was filtered through a 0.45 µm pore disposable filter. The crude cavitein was purified by RP-HPLC. A white solid was obtained after lyophilization. (5.9 mg, yield: 40 %) MS (cavitein L5): m/z 10734.3 80 Cavitein M4  In 1 mL of dried and degassed DMF solvent, [4]-benzylthiol cavitand (1.4 µmol) and purified activated peptide (6 equiv, 8.4 µmol) were dissolved. DIPEA (40 µmol, 7 µL) was added to the solution. The solution was bubbled by argon gas at room temperature overnight. The reaction solution was condensed in vacuo to a volume of around 200 µL. The solution was then diluted with H2O and was filtered through a 0.45 µm pore disposable filter. The crude cavitein was purified by RP-HPLC. A white solid was obtained after lyophilization. (4.3 mg, yield: 42%) MS (cavitein M4): m/z 7107.2  Cavitein M5  In 1 mL dried of and degassed DMF solvent, [4]-benzylthiol cavitand (1.4 µmol) and purified activated peptide (6 equiv, 8.4 µmol) were dissolved. DIPEA (40 µmol, 7 µL) was added to the solution. The solution was bubbled by argon gas at room temperature overnight. The reaction solution was condensed in vacuo to a volume of around 200 µL. The solution was then diluted with H2O and was filtered through a 0.45 µm pore disposable filter. The crude cavitein was purified by RP-HPLC. A white solid was obtained after lyophilization. (3.2 mg, yield: 26%) MS (cavitein M5): m/z 8884.1  81 Cavitein S4  In 1 mL of dried and degassed DMF solvent, [4]-benzylthiol cavitand (1.4 µmol) and purified activated peptide (6 equiv, 8.4 µmol) were dissolved. DIPEA (40 µmol, 7 µL) was added to the solution. The solution was bubbled by argon gas at room temperature overnight. The reaction solution was condensed in vacuo to a volume of around 200 µL. The solution was then diluted with H2O and was filtered through a 0.45 µm pore disposable filter. The crude cavitein was purified by RP-HPLC. A white solid was obtained after lyophilization. (2.9 mg, yield: 40%) MS (cavitein S4): m/z 5172.9  Cavitein S5  In 1 mL of dried and degassed DMF solvent, [4]-benzylthiol cavitand (1.4 µmol) and purified activated peptide (6 equiv, 8.4 µmol) were dissolved. DIPEA (40 µmol, 7 µL) was added to the solution. The solution was bubbled by argon gas at room temperature overnight. The reaction solution was condensed in vacuo to a volume of around 200 µL. The solution was then diluted with H2O and was filtered through a 0.45 µm pore disposable filter. The crude cavitein was purified by RP-HPLC. A white solid was obtained after lyophilization. (2.1 mg, yield: 23%) MS (cavitein S5): m/z 6464.4  82 2.4.4 Circular Dichroism Experiments  The CD spectra were acquired on a JASCO J-815 CD spectrophotometer. The experiments were conducted at 25 ℃ with a quartz cell of 1 mm path length. Both high and low concentrations of cavitein were assayed (5 µM and 50 µM in 50 mM phosphate buffer at pH 7.0). For each test, the solution was filtered by a 0.45 µm pore disposable filter. Each spectrum was the result of averaging of three scans with solvent baseline correction. Mean residue ellipticity was calculated by the following equation: (θobs: measured ellipticity in millidegrees, C: concentration of the cavitein in M, n: number of residues in the cavitein, l: path length in centimeters.) [θ]M.R.E. = θobs / (10Cnl) 2.4.5 GuHCl Denaturation Studies  Guanidine HCl denaturation studies on caivteins L4, L5, M4, M5, S4 and S5 were performed in a 50 mM potassium phosphate buffer (pH 7.0) in the presence of GuHCl with its concentration varying from 0 to 8.0 M. The CD signals were observed at [θ]222. The experiments were conducted at 25 ℃ with a quartz cell of 1 mm path length. Both low and high concentrations of caviteins were prepared for the experiments (4 µM and 40 µM respectively) and they were filtered by a 0.45 µm pore disposable filter before testing. Each data point was the result of the average of three scans with solvent baseline correction. The denaturation curves were analyzed by a non-linear least-squares method. ∆G°H2O was determined by the following equation:  83   y: fraction folded ([θ]222);  x: concentration of GuHCl;  F: the least-squares analysis of pre-transitional [θ]222 intercept;  U: the least-squares analysis of post-transitional [θ]222 intercept;  G: ∆G°H2O;  m: ∆G°/[GuHCl];  R: the universal gas constant;  T: temperature;  a: a constant determined by least-squares analysis.  GuHCl denaturation experiments in elevated temperatures were performed in cavitein concentration of 50 µM. CD signals at 222 nm of caviteins L4 and L5 solutions in the presence of 8 M GuHCl were measured from 20 ℃ to 95 ℃. The solutions were filtered by a 0.45 µm pore disposable filter before testing. Each data point was the result of the average of three scans with solvent baseline correction.   84 Chapter 3 Application of Caviteins in Ester Hydrolysis  3.1 Introduction  Our lab has achieved great success in the past two decades in our research on template-assembled de novo designed proteins, and mimicking the native-like properties of natural proteins. However, not much has been done with regard to application of our synthetic proteins. In the summary chapter of our former group member Dr. Freeman’s thesis, he proposed that the potential applications of the cavitein can be found in the HIV-1 Gag polyprotein where there is a six helical peptide region126,127 and the slac1 protein which exists as a series of parallel five-helix bundles.128  We, as our first step to bringing the structurally well-defined caviteins into a practical use, start with the application of the caviteins in ester hydrolysis, for it seems less challenging than the ideas proposed by Dr. Freeman.  The process of ester hydrolysis, splitting an ester into an acid and an alcohol by a molecule of water, has been mimicked by synthetic biomolecules. Tofteng and co-workers designed a TASP, named carboprotein (Figure 3.1), which contains a carbohydrate template that helps to assemble an imidazole functionalized four-helix bundle, to explore its influence in a hydrolytic reaction.129 Their use of the tetraaminooxyacetyl functionalized monosaccharide as the carbohydrate template helps to lower the folding entropic cost. Their peptide sequence which mainly consists of Leu, Glu and Lys was designed based on the minimalist approach.11 Four identical peptides which were linked to the carbohydrate 85 template by oxime bonds folded into a four-helix bundle. The imidazoyl groups on the peptides not only were able to chelate to the metal ion (Cu (II)), but also made the carboprotein a rate enhancer in the reaction of ester hydrolysis.129    One mechanism (Figure 3.2) for ester hydrolysis catalyzed by an enzyme takes place in the following way, which is an example of nucleophilic catalysis: The nitrogen (the one without hydrogen) in the imidazole ring is nucleophilic so that it can attack the carbonyl carbon which is electrophilic. After this process, they form a tetrahedral intermediate (not shown). The break down of this tetrahedron intermediate forms one molecule of alcohol and one molecule of an acylimidazole. This acylated imidazole group is readily hydrolyzed by water via forming another tetrahedral intermediate; finally, the histidine residue turns into its original form and one molecule of carboxylic acid is formed.130  Figure 3.1 Structure of a carboprotein with imidazoyl groups on the peptide used as a rate enhancer towards ester hydrolysis. 86    The reason Tofeng’s carboprotein enhances the reaction rate of ester hydrolysis is because their TASP is functionalized with imidazole groups, which comprise the side chain of histidine (Figure 3.1). Histidine is an amino acid that is often found in enzyme active sites participating in enzyme catalyzed hydrolylic reactions. The imidazole side chain and its conjugate acid form are of utmost importance to the hydrolytic reaction. The conjugate acid form is responsible for transferring a proton in the formation and break down of the tetrahedral intermediate. In the conjugate base form, histidine is nucleophilic and also serves as a general base. These properties make histidine a versatile residue among all the amino acids (Figure 3.3).  Figure 3.2 Possible mechanism for the ester hydrolysis reaction by an enzyme. 87    3.2 Results and Discussion  3.2.1 Cavitein Design  Looking back at all the caviteins that have been synthesized in our group, we found one cavitein that contains histidine residues. We sought to use this cavitein as a rate enhancer in the process of ester hydrolysis (the measured rate indicates the rate of formation of p-nitrophenol; this may involve hydrolysis, acylation, or both; we will refer to the effect of caviteins and peptides on this process as rate enhancement. This point will be detailed in section 3.2.7). This histidine-containing cavitein was synthesized by Dr. Freeman and it was named Q4-H for two reasons. (1) Two pairs of salt bridges (Glu and Lys) from the original peptide sequence that mimics native-like properties of a natural protein (EELL KKLE ELLKK) were mutated to Gln residues (EQLL KQLE EQLKQ). The four Gln containing Figure 3.3 Histidine in its conjugate acid form of imidazole side chain in equilibrium with   its conjugate base. 88 cavitein (Q4) was used to study the crystal structure of a cavitein, and it was found that it exists as an asymmetric cavitein dimer in the solid state and as a monomer in solution.118 (2) For Q4-H, four histidine residues are centrally positioned in the hydrophobic region of the helix near the C-terminus. The peptide sequence for pQ4-H is GG EQLL KQLE QLLK QHG. (Figure 3.4)     The peptide sequence of pQ4-H was originally designed to form a four-helix bundle cavitein monomer in the presence of a metal ion in solution. The four histidine residues in  Figure 3.4 Wheel model of cavitein Q4-H. 89 the central hydrophobic core of the cavitein were found to chelate a nickel ion, which conformationally constrained this TASP system to exist predominantly as a monomeric species at pH 7.0 aqueous solution (60μM, 80μM, 100μM).114  Based on the peptide sequence of pQ4-H, we designed a new peptide, pQ4-H2, to investigate how the position of the histidine residues in the sequence contributes to the rate enhancement. The peptide sequence of pQ4-H2 is GG EQLH KQLE QLLK QLG.   As shown in Figure 3.4 and Figure 3.5, the histidine residues in the pQ4-H were moved downward (with respect to the cavitand) from the C-terminus to the N-terminus to make peptide pQ4-H2. As shown in Figure 3.4 above, the histidine residues of Q4-H were positioned in the hydrophobic core, which may decrease the hydrophobic effect for forming a helical bundle. In cavitein Q4-H2, one leucine residue, which was originally positioned at the interface between the hydrophobic and hydrophilic regions, was switched in position with the histidine residue. (Figure 3.5) Thus the histidine is moved out a little bit from the inner core with respect to cavitein Q4-H. 90    Figure 3.6 gives us a clearer view of the positions of the histidine residues in both Q4-H and Q4-H2 caviteins. On the left is the Q4-H cavitein, in which the histidine residues are located near C-terminus far away from the template molecule. The one on the right is the Q4-H2 cavitein, in which the histidine residues are moved towards the N-terminus (close to the template molecule) and they are at the interface between the hydrophobic region and hydrophilic region of the peptide.   Figure 3.5 Wheel model for cavitein Q4-H2. 91   3.2.2 Peptide Design and Synthesis  Both of the pQ4-H and pQ4-H2 peptides have two glycines at the N-terminus which provide the peptides with an appropriate linker to the cavitand, and the glycine at the C-terminal provides a C-cap. The peptides were made via solid phase peptide synthesis using a Rink resin, which produces a C-terminal amide. The amidated terminus helps minimize the macrodipole effect and reduces charge-charge repulsion at the C-terminus of the helix bundle.   Also, the N-terminus of the sequence was activated with a chloroacetyl functional group in order to link to the cavitand template. The chloroacetyl is incorporated before the  Figure 3.6 Positions of the histidine residues in Cavitein Q4-H and Q4-H2. 92 peptide is cleaved from the resin so that the chloroacetyl chloride will not influence any side chains on the peptide. Nomenclature for the peptides and caviteins is shown in Table 3.1. Table 3.1     Nomenclature for the peptides and caviteins within this chapter.  Peptide Sequence  pQ4-H peptide ClCH2CO-GG EQLL KQLE QLLK QHG-NH2 Q4-H cavitein pQ4-H2 peptide ClCH2CO-GG EQLH KQLE QLLK QLG-NH2 Q4-H2 cavitein   3.2.3 Cavitand Synthesis  The template cavitand for both Q4-H and Q4-H2 caviteins is a methyl footed [4]-arylthiol cavitand. This type of cavitand is usually used for an N-acetyl linked cavitein (unlike the benzylthiol cavitand in the previous chapter which is usually used for a disulfide linked caviteins). Because the synthesis of this cavitand has been previously reported, the following only summarizes its synthesis.  Resorcinol and acetaldehyde were added into the solution of methanol and hydrochloric acid. (Figure 3.7) After stirring up to five days, the crude resorcinarene (yield: 69%) was obtained. It was then dissolved in 2-butanone and reacted with recrystallized N-bromosuccinimide to obtain the tetrabromoresorcinarene (yield: 70%). Bridging of the phenolic groups with bromochloromethane in the presence of potassium carbonate after refluxing overnight afforded the methylene-bridged arylbromo cavitand (yield: 51%). After a halogen-lithium exchange reaction, the anion was quenched by sulfur. Finally, the reaction 93 mixture was acidified by hydrochloric acid, and the arylthiol cavitand (11) as the final product was produced with a yield of 44%.74   3.2.4 Cavitein Synthesis  The peptide was attached to the arylthiol cavitand template through the chloroacetylated N-terminus of the peptide to form the final cavitein. The syntheses of the caviteins were achieved by reacting the arylthiol cavitand with an excess of activated peptide (6 equivalents of the cavitand) in the presence of diisopropylethylamine (DIPEA) in dimethylformamide (DMF) solution at room temperature overnight, (Figure 3.8) and then the  Figure 3.7 Synthesis of arylthiol cavitand from resorcinol and acetaldehyde. 94 product was purified by reverse phase HPLC. The molecular weights of these caviteins were confirmed by ESI and MALDI mass spectroscopy. Figure 3.9 and Figure 3.10 show the HPLC purification of Q4-H2 cavitein and the assessment of purity by analytical HPLC.    Figure 3.8 Synthesis of cavitein Q4-H2. 95  Figure 3.9 Preparatory Reverse phase HPLC purification of cavitein Q4-H2. Cavitein    Q4-H2 was purified with a reverse phase preparatory column using a    H2O:ACN gradient (65% H2O to 54% H2O over 22 minutes). 96   3.2.5 Far-UV Circular Dichroism Spectra  The secondary structure of cavitein Q4-H2 was investigated by circular dichroism (CD) experiments. Figure 3.11 shows the CD results compared with cavitein Q4-H which was previously synthesized by Dr. Freeman. It is shown that cavitein Q4-H2 has an alpha helix structure since it displays three characteristic bands of an alpha helix: two distinctive negative bands near 222 nm and 208 nm and one positive maximum band near 195 nm. Figure 3.10 Analytical reverse phase HPLC trace of purified Q4-H2 cavitein. The purity of   cavitein Q4-H2 was assessed with an analytical reverse phase column using a   H2O:ACN gradient (65% H2O to 54% H2O over 22 minutes) 97 Helix-helix interactions may exist if [θ]222/[θ]208 is above 1.120 However, this should be taken lightly due to the contributing absorption from the cavitand chromophore.   3.2.6 Ester Hydrolysis Study  A UV-spectrophotometric assay was used to investigate the hydrolytic activities of these two caviteins.129 The substrates were a series of esters with different lengths of carbon chain: p-nitrophenyl acetate (PNPA), p-nitrophenyl butyrate (PNPB) and p-nitrophenyl Figure 3.11 CD spectra of cavitein Q4-H (red) and cavitein Q4-H2 (black). The spectra    were acquired at 25 ℃  in 50 mM phosphate buffer at pH 7.0 using a cavitein   concentration of 30 µM. 98 octanoate (PNPO). p-Nitrophenol is the product of this ester hydrolysis reaction, and it shows a good UV absorbance signal at 410 nm.  3.2.6.1 Initial Rates of Cavitein Q4-H with Different Substrates  The initial rates of cavitein Q4-H with different substrates directly reflects its reaction activity towards esters. The experiments were performed in pH 7.0 phosphate buffer at room temperature. In comparison, the concentration of the peptide was made four times larger than that of the cavitein.  Figure 3.12 depicts the curve of the reaction progress over time and its linear fitting for the first five minutes. On the top (Figure 3.12 a) and c)) is the pQ4-H peptide with the three esters and on the bottom (Figure 3.12 b) and d)) is the Q4-H cavitein. It is apparent that both the pQ4-H peptide and Q4-H cavitein are capable of enhancing the reaction rate, especially for the short chain ester PNPA. The initial rates measured for long chain ester PNPO with both the peptide and cavitein are lower than shorter chain substrates PNPA and PNPB.      99   As we can see in Figure 3.12c and Figure 3.12d, the reaction rate gradually slows down over time, and the line is curved instead of straight, as is the most obvious in the case of cavitein Q4-H with PNPB. The explanation for this will be discussed later. Table 3.2 lists the initial rates in the first five minutes from the data in Figure 3.12a and Figure 3.12c. The initial rate measured here means the increment of the absorbance of UV light at 410 nm divided by the reaction time.129 The absorbance comes from the reaction product p-nitrophenol so that the faster the absorbance goes up, the faster the rate of reaction. The rate Figure 3.12 Linear fitting of reaction progress of a) peptide qQ4-H and b) cavitein Q4-H in the first five minutes; reaction progress of c) peptide pQ4-H, d) cavitein Q4-H on p-nitrophenyl acetate (PNPA),  p-nitrophenyl butyrate (PNPB) and p-nitrophenyl octanoate (PNPO) (phosphate buffer, pH 7), p-nitrophenolate absorbance signal at 410 nm in the first thirty minutes. 100 enhancement of cavitein Q4-H over background with substrate PNPB is significant, about 26 fold in the first five minutes. The initial reaction rate with PNPA is enhanced, but it is only around 4-5 fold over background, as there is a significant rate of PNPA’s hydrolysis in buffer alone.  Table 3.2     Initial rates of cavitein Q4-H on different substrates.a  Rate/∆mAbsU410nm / min System PNPA PNPB PNPO Q4-H cavitein 0.1 mM 79.5 ± 0.06 62.1 ± 0.02 12.6 ± 0.04 pQ4-H peptide 0.4 mM 38.0 ± 0.01 12.1 ± 0.01 14.5 ± 0.05 Unenhanced 17.4 ± 0.03 2.4 ± 0.03 1.1 ± 0.006 a. The reaction was conducted in phosphate buffer, pH 7, room temperature.  3.2.6.2 Initial Rate of Cavitein Q4-H2 with Different Substrates  The reaction activity of the newly designed cavitein Q4-H2 was investigated with the same three ester substrates (Figure 3.13a-d). It is shown in Figure 3.13c that the initial rate of peptide pQ4-H2 with PNPO is higher than that with PNPB in the first several minutes, but surprisingly, the relative rates reverse over time. In Figure 3.13d, cavitein Q4-H2 exhibits a moderate reaction rate with long chain ester PNPO, and the rate with PNPB exceeds that of PNPA during first ten minutes. These results illustrate that both peptide pQ4-H2 and cavitein Q4-H2, especially cavitein Q4-H2, are more capable of enhancing the rate of hydrolysis reaction with longer chain substrates. Also, we observed that the enhanced 101 reaction rates did not last long, as the slopes of the curves decrease over time. This hindrance in the reaction will be discussed in detail in section 3.2.7.     Figure 3.13 Linear fitting of reaction progress of a) peptide qQ4-H2 and b) cavitein Q4-H2 in the first five minutes; reaction progress of c) peptide pQ4-H2, d) cavitein Q4-H2 on p-nitrophenyl acetate (PNPA), p-nitrophenyl butyrate (PNPB) and p-nitrophenyl octanoate (PNPO) (phosphate buffer, pH 7), p-nitrophenolate absorbance signal at 410 nm in the first thirty minutes. 102 Table 3.3     Initial rates of cavitein Q4-H2 on different substrates.a  Rate/∆mAbsU410nm / min System PNPA PNPB PNPO Q4-H2 cavitein 0.1 mM 73.9 ± 0.2 99.7 ± 0.1 51.4 ± 0.07 pQ4-H2 peptide 0.4 mM 34.9 ± 0.005 7.2 ± 0.007 16.1 ± 0.05 Unenhanced 17.4 ± 0.03 2.4 ± 0.03 1.1 ± 0.006 a. The reaction was conducted in phosphate buffer, pH 7, room temperature.   The initial rates for cavitein Q4-H2 are shown in Table 3.3. The rate enhancement of cavitein Q4-H2 over background with substrate PNPB is the most outstanding, about 42 fold. Also noteworthy is that cavitein Q4-H2 outperforms cavitein Q4-H with the larger esters.  Figure 3.14 is a comparison between these two caviteins with respect to their reaction activity towards ester substrates with different chain lengths. As shown in Figure 3.14a, the initial rates for short chain substrate PNPA are similar for the two caviteins. Figure 3.14b shows that the rate enhanced by cavitein Q4-H2 with PNPB is approximately 1.6 times faster than that of cavitein Q4-H. For the longest chain substrate, PNPO, in Figure 3.14c, cavitein Q4-H2 enhances the reaction with a rate 4.1-fold higher than that of Q4-H.  103  Figure 3.14 Comparisons between cavitein Q4-H and cavitein Q4-H2 with a) p-nitrophenyl   acetate (PNPA), b) p-nitrophenyl butyrate (PNPB) and c) p-nitrophenyl    octanoate (PNPO). 104  Consideration of the design of caviteins Q4-H and Q4-H2 will help us understand the differences in these reaction rate enhancements. We know that the only difference between these two caviteins is the positions of histidine residues in their peptide sequences. In the helix bundle of Q4-H, the histidine residues are located near the C-terminus, away from the template molecule. Thus these four histidines are at the top of the bundle (if we consider the cavitand template as the bottom), and they are close to each other in the center of the inner core. Histidine residues in cavitein Q4-H2 are near to the template and at the interface between the hydrophobic region and hydrophilic region of the helix bundle. These histidine residues are at the bottom of the helix bundle close to the template, and they are separated from each other at the interface of the bundle.  Since the peptide is free of a template, the influence of the positions of the histidine residues to the hydrolysis rates is minor, which can be seen from the similar initial rates of peptide pQ4-H and peptide pQ4-H2 with the three substrates. (Table 3.4) Table 3.4     Summary of initial rates.  Rate/∆mAbsU410nm / min System PNPA PNPB PNPO Q4-H cavitein 0.1 mM 79.5 ± 0.06 62.1 ± 0.02 12.6 ± 0.04 Q4-H2 cavitein 0.1 mM 73.9 ± 0.2 99.7 ± 0.1 51.4 ± 0.07 pQ4-H peptide 0.4 mM 38.0 ± 0.01 12.1 ± 0.01 14.5 ± 0.05 pQ4-H2 peptide 0.4 mM 34.9 ± 0.005 7.2 ± 0.007 16.1 ± 0.05 Unenhanced 17.4 ± 0.03 2.4 ± 0.03 1.1 ± 0.006   105  As for the caviteins, histidine residues on both Q4-H and Q4-H2 caviteins can easily reach the short chain substrate: in Q4-H, the histidines are only covered by one glycine residue whereas the histidines in Q4-H2 are at the interface of the bundle where it is open for them to meet the substrate. As for the longer chain substrates, binding between the caviteins and substrate may contribute to the rate enhancement. In the cases of cavitein Q4-H2, its reaction rates with longer chain substrates PNPB and PNPO increases a lot compared either to the background or its corresponding peptide.  3.2.7 HPLC Results of Post-reaction Caviteins  In the UV-spectrophotometric assay of the caviteins towards ester substrates, we noticed that the reaction rates of the hydrolysis process slowed down as the reactions progressed. Over the course of the experiment, the caviteins were oversaturated by the ester substrates. So what happened to the caviteins? It is possible that the cavitein was consumed, and therefore its hydrolytic activity was hindered. HPLC analysis of the post-reaction caviteins and mass spectrometry were performed to examine this situation.  Figure 3.15 shows the separated post-reaction cavitein Q4-H after hydrolyzing PNPA. After the reaction, the solution was stored in the fridge overnight and then prepared for HPLC. The fractions were separated using a reverse phase analytical C-18 column. Each peak marked with a capital letter stands for one kind of “reacted cavitein” which was collected separately, and its molecular weight was examined by mass spectrometry. It was found that after reaction with PNPA, cavitein Q4-H turns into four different molecules 106 covalently bound to various numbers of acetyl groups (peak A has three acetyl groups on the Q4-H, peak B: four, peak C: five, peak D: six).  Figure 3.15 Post-reaction cavitein Q4-H in reaction with PNPA was analyzed by RP-HPLC   and shows four different cavitein derivatives after reaction. 107 Figure 3.16 Post-reaction cavitein Q4-H in reaction with PNPB was analyzed by RP-HPLC   and shows five different cavitein derivatives after reaction.  The same situation occurred for substrate PNPB with cavitein Q4-H (Figure 3.16), and for all three substrates with cavitein Q4-H2. (Figure 3.17, Figure 3.18, and Figure 3.20) Only the original Q4H cavitein signal was detected from the reaction between cavitein Q4-H and PNPO. (Figure 3.19) Table 3.5 summarizes the numbers of small groups attached to post-reaction caviteins. 108  Figure 3.17 Post-reaction cavitein Q4-H2 in reaction with PNPA was analyzed by    RP-HPLC and shows four different cavitein derivatives after reaction. 109  Figure 3.18 Post-reaction cavitein Q4-H2 in reaction with PNPB was analyzed by    RP-HPLC and shows a broad peak that contains four cavitein derivatives. 110  Figure 3.19 Post-reaction cavitein Q4-H in reaction with PNPO was analyzed by    RP-HPLC and shows no cavitein derivative after reaction.  111  Table 3.5     Results of analysis of the post-reaction caviteins. Systems Amount of observed attached groups acetyl butyryl octanoyl Q4-H cavitein 3,4,5,6 0,1,2,3,4 0 Q4-H2 cavitein 4,5,6,7 0,1,2,3 0,1   The data from the HPLC analysis and mass spectrometry show that the reactive groups in the caviteins are blocked by reacting irreversibly with the substrates, which may explain why the p-nitrophenol-forming process slows down over time. Aside from the Figure 3.20 Post-reaction cavitein Q4-H2 in reaction with PNPO was analyzed by    RP-HPLC and shows only one cavitein derivative after reaction.  112 histidine residues, there are two lysine residues in each peptide strand. Literature on reactions of both the lysine and imidazole with p-nitrophenyl acetate show that the lysine residue is readily acetylated while the imidazole is regenerated by the spontaneous hydrolysis of the acylimidazole.166-169 In some cases, the number of attached acetyl groups on both caviteins is more than four. This means that the lysine residues in the cavitein did participate in the reaction because even if the histidine residues were irreversibly acylated, the rest of the acetyl groups must be located on the lysine residues as each cavitein only contains four histidine residues.   As for the reaction between PNPO and cavitein Q4-H, no peak was observed for a new cavitein by HPLC from the reaction between PNPO and cavitein Q4-H after 16 hours. Thus, it is evident that hydrolysis occurs, because a cavitein adduct would be detected by HPLC if it formed, since 0.6 equivalents of p-nitrophenol per cavitein was formed after 1 hour. We used the UV data to calculate the amount of p-nitrophenol generated after 30 minutes. In all cases, less than 200% of p-nitrophenol was formed relative to the amount of original cavitein.   3.3 Chapter Summary and Conclusion  Synthetic supramolecules have been designed to mimic the functions of natural enzymes, which are highly selective in catalyzing biological transformations.91 In this chapter, we accomplished our goal of putting a previously synthesized cavitein into application for the first time, mimicking the process of ester hydrolysis. 113   Cavitein Q4-H exhibits good activity in ester hydrolysis with its histidine residues in the hydrophobic core of the helix bundle. Cavitein Q4-H2, which is a newly designed cavitein based on cavitein Q4-H, with its four histidine residues moved inward from the C-terminus toward the N-terminus positioning on the juncture between the hydrophobic region and hydrophilic region of the helix bundle, also shows rate enhancement, particularly with the substrate that has a long carbon chain. We showed that the positioning of the histidine residues does affect caviteins’ activity in these reactions via altering the peptide sequence.   We found that lysine residues can also participate in the process, as some of the post-reaction caviteins are adducts of a number of acyl groups. The reaction between cavitein Q4-H and long chain substrate PNPO is pure hydrolysis, as no acylated adducts were observed. In all other cases, we can only conclude that acylation is also involved in the reaction, but it is not known to what extent it is. However, the HPLC experiments suggest that the extent of acylation for the short chain substrate PNPA is higher than the longer chain substrate PNPB.  3.4 Experimental  3.4.1 Cavitand Synthesis  All the chemicals that are involved in the synthesis of the methyl-footed arylthiol cavitand were reagent grade and purchased from Sigma-Aldrich. N-bromosuccinimide was recrystallized in water and was dried in vacuo in the presence of P2O5. All the solid reagents 114 and products were dried in vacuo with P2O5 before use. Degassed water, HCl solution and DMF (dimethylformamide) were prepared through freeze-pump-thaw. Tetrahydrofuran (THF) was dried by freshly cut sodium and distilled under nitrogen atmosphere.  The synthesis of methyl-footed arylthiol cavitand from resorcinol and acetaldehyde includes four steps. The detailed procedures have been previously reported by several former group members.70 The work-up steps of the arylthiol cavitand in the last step were based on Dr. Freeman’s modification, in which the tri-substituted arylthiol cavitand was not removed before mixture was used to make the cavitein.  1D 1H NMR spectra were observed on a Bruker AVANCE 300 MHz spectrometer at 298 K. Molecular weights of the products were confirmed by Waters LC-MS equipped with ESCI ion source.  3.4.2 Peptide Synthesis  3.4.2.1 General Procedures of Peptide Synthesis  All the peptide syntheses were performed on a CS Bio (Menlo Park, CA) 136XT peptide synthesizer. Solvents used in the synthesizer, N-methylpyrrolidone (NMP) and DMF, and all the amino acids were purchased from Advanced Chemtech (Louisville, KY). Other chemicals involved were from Sigma-Aldrich. Rink’s amide resin was used to afford C-terminal amides.125 The peptides were synthesized on a 0.25 mmol scale. Standard Fmoc 115 protected amino acids were used during the synthesis with acid labile protecting groups for leucine, glycine, lysine, glutamine, glutamic acid and histidine side chains. The following amino acids were used: Fmoc-L-Leu-OH, Fmoc-L-Gly-OH, Fmoc-L-Lys(Boc)-OH, Fmoc-L-Gln(Trt)-OH, Fmoc-L-Glu(OtBu)-OH and Fmoc-L-His(Fmoc)-OH.   Before peptides were cleaved from the resin, they were activated with chloroacetyl chloride. After the resin was washed with DCM, it was treated with 95% of TFA/H2O for 2 hours to cleave the peptide from the resin. Afterwards, the peptide was purified by reversed-phase HPLC using a linear gradient of acetonitrile (0.05% TFA) and water (0.1% TFA) on a Waters C-18 Delta Pak column (300 × 19 mm2, 300 Å, 15 µm) at a flow rate of 10 mL min-1. The fraction signal was monitored by a UV detector at 229 nm to detect the amide chromophore.The solution with fraction from HPLC purification was frozen and the solvent was removed through lyophilization. Analytical reversed-phase HPLC (Waters Delta Pak C-18 column (300 × 3.9 mm2, 300 Å, 15 µm)) was used to assure the purity of the peptide by the observation of a single peak. The identity of the peptide was confirmed by mass spectrometry (ESI-MS) with a Bruker Esquire using electrospray ionizaiton as the ion source. 3.4.2.2 Syntheses of Peptides pQ4-H and pQ4-H2  The synthesis of peptide pQ4-H has been reported by Dr. Freeman previously which is similar to the synthesis of peptide pQ4-H2 in the following procedure.   After the peptide linked resin (~0.25 mmol) was collected from the synthesizer and washed by methanol, it was reacted with chloroacetyl chloride (100 µL, 1.26 mmol) and DIPEA (150 µL, 0.86 mmol) in dried DMF (5 mL) for 1 hours at room temperature. Later, a second portion of chloroacetyl chloride (50 µL, 0.63 mmol) was added to the reaction system 116 for another hour. After it was filtered and washed by DCM, the peptide linked resin was treated with 10 mL of ice cold TFA/H2O (95% TFA) solution for approximately 2 hours. The peptide was cleaved from the resin and dissolved in the above solution. The resin was filtered off and the filtrate was transferred to a round bottom flask. Its volume was then reduced in vacuo to approximately 5 mL. This concentrated cleavage solution was pipetted into ice cold diethyl ether to precipitate the peptide. The solid peptide was separated from the ether by centrifugation (3000 rpm) for 5 minutes. Washing with diethyl ether and centrifugation were repeated for another two times to completely removed TFA and small molecular weight impurities. The remaining peptide pellet was dissolved in water with a little acetic acid to speed up the dissolving process. RP-HPLC was performed after the peptide solution was filtered by a 0.45 µm pore disposable filter. The final peptide was a white cotton-like solid after lyophilization (~120 mg, yield: ~ 35%). MS (pQ4-H peptide): m/z 1996.3 MS (pQ4-H2 peptide): m/z 1996.3 3.4.3 Cavitein Synthesis  3.4.3.1 General Procedures of Cavitein Synthesis  All the reagents used for the synthesis of the caviteins were reagent grade and purchased from Sigma-Aldrich. The solvent for the reaction was HPLC grade N, N-dimethylformamide (DMF). It was freeze-pump-thawed under inert gas argon and dried by 4 Å molecular sieves. The crude caviteins were purified by reversed-phase HPLC using a 117 Phenomenex Jupiter C-4 column (250 × 20 mm2, 300 Å, 15 µm) with an acetonitrile (0.05% TFA) and water (0.1% TFA) solvent system at flow rate 10 mL/min. Their purities were assessed by analytical RP-HPLC. MALDI-TOF (Matrix Assisted Laser Desorption Ionization) and ESI were both used to confirm the identities of the caviteins. Molecular weights determined by MALDI were acquired on a Bruker Autoflex. The matrix used was 2,5-dihydroxybenzoic acid (DHB).  3.4.3.2 Syntheses of Caviteins Q4-H and Q4-H2  The synthesis of cavitein Q4-H has been reported by Dr. Freeman previously which is similar to the synthesis of cavitein Q4-H2 in the following procedure.  In 1 mL of dried and degassed DMF solvent, arylthiol cavitand (1.4 µmol) and purified activated peptide (6 equiv, 8.4 µmol) were dissolved. DIPEA (40 µmol, 7 µL) was added to the solution. The solution was bubbled by argon gas at room temperature overnight. DMF was condensed in vacuo to a volume of around 200 µL. The solution was then diluted with H2O and was filtered with a 0.45 µm pore disposable filter. The crude cavitein was purified by RP-HPLC. A white cotton-like solid was obtained after lyophilization. (6 mg, yield: ~50%) MS (Q4-H cavitein): m/z 8557.5 MS (Q4-H2 cavitein): m/z 8557.5  118 3.4.4 Circular Dichroism  The CD spectra were acquired on a JASCO J-815 CD spectrophotometer. The experiments were conducted at 25 ℃ with a quartz cell of 1 mm path length. The concentration of cavitein was 30 µM in 50 mM phosphate buffer at pH 7.0 and it was filtered by a 0.45 µm pore disposable filter before testing. Each spectrum was the result of the average of three scans with solvent baseline correction. Mean residue ellipticity was calculated by the following equation: (θobs: measured ellipticity in millidegrees, C: concentration of the cavitein in M, n: number of residues in the cavitein, l: path length in centimeters.) [θ]M.R.E. = θobs / (10Cnl)   3.4.5 UV-spectrophotometric Assay  The initial rate was obtained from a UV spectrophotometric assay. p-nitrophenolate was formed during hydrolysis reaction and its absorbance signal was observed at 410 nm at room temperature. All the substrates used in experiments (p-nitrophenyl acetate, p-nitrophenyl butyrate, p-nitrophenyl octanoate and paraoxon) were purchased from Sigma-aldrich. The UV data was collected on a Varian Cary 4000 UV/Vis absorption spectrophotometer. The stock solutions of p-nitrophenyl acetate (100 mM), p-nitrophenyl butyrate (100 mM), p-nitrophenyl octanoate (100 mM) and paraoxon (100 mM) were prepared and wrapped in aluminium foil at 4 °C. The stock solutions were then diluted with 119 phosphate buffer (0.5 M, pH 7) 100 times before use. Both the free peptide (0.4 mM) and cavitein (0.1 mM) were dissolved in these diluted solutions.   3.4.6 Post-reaction Caviteins Analysis  After the UV-spectrophotometric assay, solutions of all the six reactions (cavitein   Q4-H with PNPA, PNPB, PNPO; cavitein Q4-H2 with PNPA, PNPB, PNPO) were saved and stored in the fridge overnight. The post-reaction caviteins were analyzed by reverse phase HPLC with H2O : ACN gradient (65% ACN to 45% ACN over 20 minutes). Each fraction from the HPLC band was collected in separate vial. Their identities were examined by MALDI-TOF and ESI. Molecular weights determined by MALDI were acquired on a Bruker Autoflex. The matrix used was 2,5-dihydroxybenzoic acid (DHB). For the HPLC spectrum of cavitein Q4-H2 with PNPB, it showed a broad peak instead of separated peaks. Therefore, the whole fraction of this broad peak was collected and examined by MALDI-TOF and ESI, and four different MS signals were still identifiable.       120 Chapter 4 Application of Caviteins in Protein-protein    Interactions  4.1 Introduction  Programmed cell death (also named apoptosis), which means cells commit suicide by activating an intracellular death program when they are no longer needed,108 occurs in multicellular organisms.131 Many fatal diseases, such as neurodegenerative disorders and autoimmunity, are related to dysfunction of this cell death process.132   The Bcl-2 family of proteins plays an important role in regulating programmed cell death.133 There are two groups of proteins in the Bcl-2 family: proteins Bcl-xL and Bcl-2 inhibit programmed cell death while others such as Bax, Bak, Bad and Bid promote apoptosis.134,135 The Bcl-2 homology (BH) domains found in all members of the Bcl-2 protein family are involved in protein-protein interactions (PPI) between these two groups,136,137 by forming homo- and heterodimers that antagonize their opposite activities.138   Research groups have focused on the Bcl-xL protein, an anti-apoptotic protein, and Bak protein, a pro-apoptotic protein, because these two proteins are widely studied and structurally elucidated.139 The binding site of the Bcl-xL protein consists of BH1, BH2 and BH3 (BH for Bcl-2 homology) domains, all of which form a hydrophobic cleft. The binding site of the Bak protein also has a BH3 domain.140 It has been demonstrated that the BH3 domain from the Bak protein not only is essential for interacting with the Bcl-xL protein, but 121 it is required for its pro-apoptotic activity as well.141 Figure 4.1 depicts the binding pocket of Bcl-xL protein bound to a helical peptide (named Bak peptide) which is derived from the BH3 domain of Bak protein.    The minimal region of the Bak protein required for PPI activity is a 16-residue peptide (GQVG RQLA IIGD DINR) derived from the BH3 domain (residue 72-87 of the Bak protein).142 As mentioned in Chapter 1, these 16 residues are called the hot spot that contributes to the majority of the binding free energy between the Bak and the Bcl-xL Figure 4.1 Binding pocket of Bcl-xL protein bound to the Bak peptide. Reproduced with permission from Science 1997, 275, 985. Copyright 1997 American Association for the Advancement of Science. 122 proteins. Fesik and co-workers mutated some of the hot spot residues (val74, Arg76, Leu78, Ile80, Ile81, Gly82, Asp83, Asp84, Ile85) to alanine and found that their presence significantly influences the Bak peptide’s binding affinity towards the Bcl-xL protein.140 This truncated active Bak peptide exists as a random coil in aqueous solution. However, it will fold into an alpha helix when interacting with the Bcl-xL protein.140  Because this truncated Bak peptide is relatively small (16 residues compared to a large protein molecule), it prompted us to design molecules based on this Bak peptide sequence that can mimic its behavior with the Bcl-xL protein.  Arora and co-workers introduced a highly alpha helical peptide derived from a hydrogen-bond surrogate which displayed much stronger inhibition activity than that of the wild type Bak peptide in the PPI between the Bak peptide and the Bcl-xL protein (Figure 4.2a).143 Their macrocyclic peptide was highly helical due to the replacement of a hydrogen bond with a covalent carbon-carbon bond derived from a ring-closing metathesis reaction. They pointed out that the uniqueness of their approach was that all the side chains on the peptide were not disturbed for molecular recognition, which greatly increased its inhibition ability.   Besides their work, other potent small-molecule and Bak peptide-based inhibitors of the Bcl-xL/Bak interactions have been developed,144-155 such as a benzoylurea-derived alpha-helix (Figure 4.2b), a terphenyl-based proteomimetics (Figure 4.2c), and an unnatural oligomer with strong conformational (helix-forming) propensities (Figure 4.2d).143-146 123    Here we incorporated four identical Bak peptides onto a template molecule, a cavitand, making a Bak cavitein, to explore its inhibition ability. Compared to all the other systems described above, the Bak peptide in our cavitein system is the original Bak peptide Figure 4.2 Bak peptide inhibitors of the Bcl-xL/Bak interactions. a) the formation of a hydrogen-bond surrogate alpha helix;143 b) a benzoylurea-derived alpha helix;144 c) a terphenyl-based proteomimetics;145 d) an unnatural oligomer with strong conformational propensities.146 124 itself, without any artificial modifications, whereas the Bak peptide in their systems either has extra covalent bonds or is decorated with small functional molecules. The increase in helicity of their modified Bak peptide is a big improvement; but in our system, helicity is driven by bundling, which competes with binding to the Bcl-xL protein. The advantage of their artificiality is prtease resistance, and we use bundling for this. What is more, in our cavitein system, we can even make hetero Bak caviteins, in which our previously made highly helical peptides are incorporated, so that we can further explore the highly helical peptides’ influence on the activities of the Bak peptide in the same cavitein.   4.2 Results and Discussion  4.2.1 Cavitein Design  This truncated 16-residue Bak peptide from BH3 domain of Bak protein was chosen in our cavitein design. The Bak peptide is alpha helical when bound to the Bcl-xL protein in its hydrophobic cleft and is a random coil in the aqueous solution.140 We used the arylthiol cavitand (same as that in Chapter 3) as a template, on which four identical Bak peptides were incorporated to make a Bak cavitein.  The cavitand reduces the number of available conformations of the synthetic proteins and potentially provides protection to protease via helical bundling.11,13,156 Moreover, much 125 of the arduous synthetic tasks and artificiality of other Bak derivatives are avoided when using TASP strategy.  In the wheel model of the Bak peptide shown in Figure 4.3, it is noticeable that there is a four-residue hydrophobic region (val74, Leu78, Ile81, Ile85) within the Bak peptide sequence when it is put in a helical form. When protein-protein interactions between the Bcl-xL protein and the Bak peptide occur, it is this hydrophobic region that points into the hydrophobic cleft of the Bcl-xL protein when PPI occurs. Moreover, in its hydrophilic region, there are three charged side chains (Arg76, Asp83, Asp84) which are close to the oppositely charged residues in the Bcl-xL protein (Glu129, Arg139, Arg100).140 The separation of the hydrophobic region and the hydrophilic region also offers the possibility for our designed Bak cavitein to form a four-helix bundle when the four Bak peptides are incorporated on the cavitand template. 126    Cavitein N1GG/Ar/Me is one of the TASPs made by our previous group member Dr. Adam Mezo.72 Its peptide (N1) sequence is EELL KKLE ELLK KG and it has two glycine residues at the N-terminus as part of the linker (pN1GG). The pN1GG peptides are incorporated into the arylthiol cavitand with methyl feet. This cavitein showed highly stable native-like structure, and its oligomeric structure was explored by Dr. Jon Freeman. We Figure 4.3 a) Sequence of the Bak peptide, existing as a random coil in aqueous solution;   b) wheel model of the Bak peptide, hydrophobic amino acid residues are    marked with a different color. 127 abbreviate this peptide and cavitein as pN1GG peptide and N1GG cavitein in this chapter for ease of discussion.  Because of N1GG’s high helicity and stable native-like structure (∆G°H2O = -11.2 ± 0.1 kcal mol-1), 157 a series of heterocaviteins with both pN1GG peptides and Bak peptides were designed in order to increase the helicity and proteolytic stability of our Bak peptide inhibitor.  Figure 4.4 displays the first hetero Bak cavitein that we designed. We incorporated three pN1GG peptides with only one Bak peptide on the arylthiol rimmed cavitand. It was named cavitein 1B3N (B for Bak peptide, N for pN1GG peptide, with numbers for the amount of each peptide. The same rule is used for all the nomenclature of caviteins within this chapter). The hydrophobic regions in the pN1GG peptides and Bak peptides should interact with each other forming a helical bundle, and this bundling in the centre core should drive the Bak peptide to be in a more protected helical form.  128    The multitude of lysine residues in the pN1GG peptide make the cavitein extremely vulnerable to trypsin, which is a protease, because trypsin will cleave a peptide if there is any lysine or arginine residue in the peptide chain (not followed by a proline). Therefore, mutations of the lysine residues were pursued. There are two common candidates that can replace lysine, homolysine and ornithine, for their structural similarity. (Figure 4.5) However, it is reported that homolysine esters can still be hydrolyzed by trypsin, though at a low rate.158 Thus, we replaced all four lysine residues (Lys5, Lys6, Lys12, Lys13) in the pN1GG peptide with ornithine because peptides with ornithines have shown good resistance to Figure 4.4 Wheel model of a hetero Bak cavitein with three o-pN1GG peptides and one   Bak peptide. 129 hydrolysis by trypsin.159 The ornithine-replaced peptide sequence is therefore named as o-pN1GG.   Hetero Bak caviteins with two o-pN1GG peptides have two peptide arrangements. In the first one, two o-pN1GG and two Bak peptides are neighboring each other (cavitein 2B2O); while in the other, these two different peptides are interlaced in a more symmetric way (cavitein BOBO). (Figure 4.6a) The hetero Bak cavitein with only one o-pN1GG peptide has also been designed for comparison of properties with the others (cavitein 3B1O). (Figure 4.6b) With the design of the Bak cavitein and a series of hetero Bak caviteins as inhibitors, the behavior of their inhibition in the PPI between the Bcl-xL protein and the Bak peptide and the influence of the o-pN1GG peptide on the whole system will be explored.    Figure 4.5 Structures of lysine compared with ornithine and homolysine. 130  4.2.2 Peptide Design and Synthesis  Both the Bak and o-pN1GG peptides have two glycines at the N-terminus which provide an appropriate linker for peptides to the arylthiol rimmed cavitand, and the C-terminal glycine provides a C-cap. Also, the N-terminus was activated with a chloroacetyl Figure 4.6 Wheel models of hetero Bak caviteins with various number of o-N1GG peptide. 131 functional group in order to link the peptides to the cavitand template. The activation reaction is performed before the peptide is cleaved from the resin so that the chloroacetyl chloride will not influence any side chains on the peptide.   A fluorescein-tagged Bak peptide was used as the molecule that binds to the Bcl-xL protein so that a fluorescent polarization signal can be monitored.160 In its peptide sequence, 5-carboxyfluorescein was linked to the N-terminus of the Bak peptide right before the first Gly residue in the sequence. Because 5-carboxyfluorescein also has a carboxyl group, it can be coupled to the peptide in the same way as an amino acid: First, the Fmoc group on the terminal glycine was deprotected by piperidine; second, 5-carboxyfluorecein was activated in the presence of DIPEA (diisopropylethylamine) and HBTU (1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate); and finally, 5-carboxyfluorecein was coupled onto the peptide just as a molecule of amino acid.  The free Bak peptide was synthesized with its N-terminus acetylated to decrease the macrodipole effect.161 All the four peptides from above were made via solid phase peptide synthesis using a Rink resin, which produces a C-terminal amide. The amidated terminus helps minimize the macrodipole effect and reduces charge-charge repulsion at the C-terminus of the helix bundle. Nomenclature and sequences as well as the molecular weights of the above peptides are shown in Table 4.1.   132 Table 4.1 Designed peptide compared with original peptide. Peptide Peptide Sequence Molecular Weight* free-Bak CH3CO-GQVG RQLA IIGD DINR-NH2 1765.1 Bak ClCH2CO-GGQVG RQLA IIGD DINR-NH2 1858.2 flu-Bak (5’-Fluorescein)-GQVG RQLA IIGD DINR-NH2 2083.5 o-N1GG ClCH2CO-GG-EELLOrnOrnLEELLOrnOrnG-NH2 1803.0 * n: The molecule weights of these peptides were obtained from MALDI-TOF.  4.2.3 Cavitein Synthesis  The peptides were attached to an arylthiol cavitand template (its synthesis was detailed in Chapter 3) through a chloroacetylated N-terminus to form the final caviteins. The nomenclature and the peptide compositions for all caviteins within this chapter are summarized in Table 4.2.        133 Table 4.2     Nomenclature and compositions for the caviteins.  Peptides Number of Bak Peptides Number of o-pN1GG Peptides Number of pN1GG Peptides Caviteins  4B 4 0 0 3B1O 3 1 0 2B2O 2 2 0 BOBO 2 2 0 1B3O 1 3 0 1B3N 1 0 3 N1GG 0 0 4 o-N1GG 0 4 0   The synthesis of Bak cavitein 4B was achieved by reacting the arylthiol cavitand with an excess of activated Bak peptide (6 equivalents to the cavitand) in the presence of DIPEA in NMP (N-methylpyrrolidone) solution at room temperature overnight. (Figure 4.7) NMP was used as the reaction solvent instead of DMF because the Bak peptide has a better solubility in NMP. The crude cavitein was then purified by reverse phase HPLC. The mass of cavitein 4B was confirmed by ESI and MALDI mass spectroscopy. Figure 4.8 and Figure 4.9 show the HPLC purification and the assessment of purity by analytical HPLC. 134  Figure 4.7 Synthesis of cavitein 4B. Figure 4.8 Preparatory Reverse phase HPLC purification of cavitein 4B. Cavitein 4B was   purified with a reverse phase preparatory column using a H2O:ACN gradient   (66% H2O to 63% H2O over 12 minutes). 135   Different types of heterocaviteins have been prepared by our group previously.162 One method of synthesizing heterocaviteins is by a one-pot reaction. It is an easy, time-efficient way to obtain heterocaviteins. However, HPLC purification requires tremendous effort, especially in the case of Bak heterocaviteins, if we need to separate caviteins BOBO and 2B2O, because these two caviteins are difficult to separate by HPLC. Thus, the step-wise method was adopted in synthesizing the Bak heterocaviteins. Figure 4.9 Analytical reverse phase HPLC trace of purified cavitein 4B. The purity of    cavitein 4B was assessed with an analytical reverse phase column using a    H2O:ACN gradient (78% H2O to 65% H2O over 13 minutes) 136  The first step is to synthesize four partially incorporated caviteins. The cavitand template was dissolved in the degassed DMF or NMP in the presence of DIPEA, and a peptide solution was added drop-wise for approximately 40 minutes. Afterwards, the reaction was immediately quenched and purified by HPLC.  Figure 4.10a shows that when the final target cavitein is 1B3O, which means the partially incorporated cavitein 12 should be the dominant product from the reaction, we optimized the ratio of arylthiol cavitand to Bak peptide to 3:1. With the excess amount of the cavitand in the reaction, we got partial cavitein 12 as the major product, two peptide-linked caviteins 13 and 14 as the moderate yield products and the trisubstituted cavitein 15 as the minor product. If our final target cavitein is 3B1O, we only need to use o-pN1GG peptide instead of Bak peptide in the first step. (FIgure 4.10b) The product with two peptides substituted from either of these two reactions can be used to synthesize caviteins 2B2O and BOBO. This first step reaction only lasts for 40 minutes, and 20% acetic acid solution was used to quench the reaction to prevent dimerization between the unreacted thiol groups. Lyophilization was performed under an argon atmosphere to minimize oxidization of the unlinked thiol groups. The molecular weights of these partially incorporated caviteins 12, 13, 14, 15, 16, 17, 18 and 19 were confirmed by ESI and MALDI mass spectroscopy. Figure 4.11 shows the corresponding HPLC purifications. As shown in the HPLC spectra, caviteins 13 and 14 (Figure 4.11a) are separable by HPLC, but they are not distinguishable at this stage. Same with caviteins 17 and 18 (Figure 4.11b).   137  Figure 4.10 Synthesis of partially incorporated caviteins: a) partially incorporated Bak    cavitein, b) partially incorporated o-N1GG cavitein. 138  Figure 4.11 a) Preparatory Reverse phase HPLC purification of hetero Bak Caviteins was   performed with a reverse phase preparatory column using a H2O:ACN gradient   (65% H2O to 36% H2O over 30 minutes); b) Preparatory Reverse phase HPLC   purification of hetero o-N1GG Caviteins was performed with a reverse phase   preparatory column using a H2O:ACN gradient (65% H2O to 36% H2O over 30   minutes). 139  The major products 12 and 16, from the first reactions, were used to synthesize caviteins 1B3O and 3B1O. These two caviteins were obtained by reacting the partially incorporated cavitein 12 or 16 with an excess of activated peptide o-pN1GG or Bak (1:5 ratio) in the presence of DIPEA in DMF or NMP solution at room temperature for 5 hours (Figure 4.12). The crude caviteins then were purified by reverse phase HPLC.    The partially incorporated caviteins 13, 14, 17 and 18 from the first step reactions were used to synthesize caviteins 2B2O and BOBO. It was achieved by reacting the partially Figure 4.12 Synthesis of heterocavitein 1B3O and 3B1O. 140 incorporated cavitein 13, 14, 17 or 18 with an excess of activated peptide o-pN1GG or Bak (1:4 ratio) in the presence of DIPEA in DMF or NMP solution at room temperature for 5 hours (Figure 4.13). The crude caviteins were then purified by reverse phase HPLC. As mentioned above, 13 and 14 were not distinguished so far but they were separated by HPLC, which made caviteins BOBO and 2B2O not distinguishable, though they were synthesized separately. It is the same situation for 17 and 18.   The molecular weights of these caviteins were confirmed by ESI and MALDI mass spectroscopy (Table 4.3). Figure 4.14a-d shows the purity assessment of caviteins 1B3O, 2B2O, BOBO and 3B1O by analytical HPLC. Figure 4.13 Synthesis of heterocavitein 2B2O and BOBO. 141      Figure 4.14 Analytical reverse phase HPLC trace of purified heterocaviteins: a) 1B3O, b)   2B2O or BOBO, c) BOBO or 2B2O, d) 3B1O. The purity of these     heterocaviteins were assessed with an analytical reverse phase column using a   H2O:ACN gradient (74% H2O to 56% H2O over 20 minutes). 142 Table 4.3 Molecular weights of caviteins obtained from MALDI-TOF and ESI. Cavitein MW Cavitein MW 4B 8009.5 1B3N 8017.5 1B3O 7843.2 N1GG 8015.1 2B2O/BOBO 7897.9 o-N1GG 7794.7 3B1O 7951.6     4.2.4 Overexpression of Bcl-xL Protein  The gene on the pET29b plasmid, which is responsible for expressing the Bcl-xL protein, was synthesized and cloned by GenScript USA Inc. Figure 4.15 shows the process of overexpression of the Bcl-xL protein from plasmid replication to protein overexpression. The pET29b plasmid was first transferred into Escherichia coli DH5α cells for plasmid replication. Because the transformed cells were kanamycin resistant, the replicated plasmid can be extracted from the surviving DH5α cells. E. coli, BL21(DE3) strain was used for protein overexpression. The replicated plasmid was transferred into BL21 (DE3) cells and the transferred cells were kanamycin resistant and were used to overexpress the Bcl-xL protein in the presence of IPTG (Isopropyl β-D-1-thiogalactopyranoside) at 37°C. 143    The Bcl-xL protein, from the first residue, methionine (Met, M), to the 209th residue, arginine, was expressed in E. coli using a pET29b plasmid. In its primary structure sequence, there are five extra amino acids (EGDIH-) at the N-terminus of this recombinant protein due to cloning, and a polyhistidine tag (-LEHHHHHH) at the C-terminus to facilitate purification by a nickel IDA column142 (Figure 4.16). The identity of the purified recombinant protein was confirmed by SDS-polyacrylamide gel eletrophoresis (PAGE) (Figure 4.17) and its purity was assessed by analytical HPLC. (Figure 4.18) Figure 4.15 Overexpression of Bcl-xL protein. 144    Figure 4.16 The sequence of the recombinant Bcl-xL protein. Figure 4.17 SDS-polyacrylamide gel electrophoresis of Bcl-xL protein. 145  4.2.5 Far-UV Circular Dichroism Spectra  Figure 4.19 shows the far-UV spectra of our previously synthesized cavitein N1GG and cavitein o-N1GG which shares the same peptide sequence as pN1GG except all the lysine residues were replaced by ornithine. As we can see from the spectra, the o-N1GG cavitein has similar secondary structure to the N1GG cavitein. Because of the similarity in their secondary structures, it is reasonable for us to replace the pN1GG peptide with the o-pN1GG peptides. Therefore, not only will the o-pN1GG peptides function in the same way as the pN1GG peptide, but also the o-pN1GG peptides should be resistant to trypsin. Figure 4.18 Analytical reverse phase HPLC trace of purified Bcl-xL protein. The purity of   Bcl-xL protein was assessed with an analytical reverse phase column (Waters   Delta Pak C-18 column (300 × 3.9 mm2, 300 Å, 15 µm)) using a H2O:ACN    gradient (85% H2O to 10% H2O over 75 minutes) 146   The secondary structures of caviteins 4B and 3B1O were investigated by far-UV circular dichroism (CD) along with the Bak peptide (Figure 4.20). It is shown that the free Bak peptide exists as a random coil in aqueous solution because the spectrum of a random coil always reaches its negative minimum at 200 nm. As for cavitein 4B, which contains four Bak peptides, it still largely exists as a random coil, but its negative minimum moved a bit towards longer wavelength and the signal at 222 nm is stronger than that of Bak peptide alone. This change may come from the helical bundling in cavitein 4B though at a low level, or may simply arise from the absorption of the cavitand chromophore. Compared with the band of cavitein 3B1O, which contains three Bak peptides plus one o-pN1GG peptide, it is Figure 4.19 CD spectra of cavitein N1GG and cavitein o-N1GG. The spectra     were acquired at 25 ℃  in 50 mM phosphate buffer at pH 7.0 using cavitein    concentration of 40 µM. 147 obvious that the cavitein 3B1O is somewhat helical for its spectrum displays two negative minimum bands at both 222 nm and 208 nm and one positive maximum band at 195 nm.   Figure 4.21 depicts the far-UV CD spectra of the rest of the three hetero-caviteins: 2B2O, BOBO and 1B3O. All of them display alpha helical structure because they all have three characteristic bands: distinctive negative bands near 222 nm and 208 nm and one positive maximum band near 195 nm. The spectrum of cavitein BOBO is similar to that of 2B2O whereas the spectrum of 1B3O is more distinctive because its [θ]222/[θ]208 is above 1, which may indicate the existence of helix-helix interaction.120 However, this should be taken Figure 4.20 CD spectra of cavitein 4B, 3B1O and Bak peptide. The spectra were acquired at 25 ℃  in 50 mM phosphate buffer at pH 7.0 using cavitein concentration of 25 µM and peptide of 100 µM. 148 lightly due to the contributing absorption from cavitand chromophore.121 From the far-UV CD spectra, it is found that the more o-pN1GG peptides present in the cavitein, the more helical structure it has, and the peptide o-pN1GG may have the function of inducing the helical character of the Bak peptide strands.   Table 4.4 lists the percent helicities for all the involved caviteins within this chapter. It is apparent that the percent helicity rises with the number of o-pN1GG peptides increases. However, any aromatic structures (in our case the cavitand chromophore) are able to influence the absorption at [θ]222.121 Thus, it should be taken lightly concerning the interpretation of percent helicity. Figure 4.21 CD spectra of cavitein 1B3O, 2B2O and BOBO. The spectra were acquired at   25 ℃  in 50 mM phosphate buffer at pH 7.0 using cavitein concentrations of 25   µM. 149 Table 4.4     Percent helicity for all the caviteins. Cavitein [θ]222 (deg cm2 dmol-1) Helicity (%) 4B 5312 22 3B1O 8536 30 2B2O/BOBO 13694 42 BOBO/2B2O 13174 41 1B3O 14718 46 o-N1GG 18017 54 N1GG 21083 62   4.2.6 Fluorescence Polarization Assay  Fluorescence polarization is a technique widely used in assays concerning PPI because it has a low limit of detection in the sub-nanomolar range and allows real-time measurements for kinetic assays. The following are some basics of this technology (Figure 4.22).  The initial depolarized light first goes through an excitation filter and turns into polarized light. When a small peptide tagged with a fluorescent functional group (in our case, the flu-Bak peptide) is excited by this polarized light at the excitation wavelength of the fluorophore, the fast tumbling of this fluorescent peptide will cause the emitted light to be nearly depolarized. When the fluorescein-tagged peptide is bound to a large molecule, such as a protein (in our case, the Bcl-xL protein), the resulting complex will tumble much more slowly and the emitted light will still be polarized. Thus, the observation of the fluorescence 150 polarization signal reflects the interaction between the fluorescent-tagged peptide and the large molecule. Its value is proportional to the fraction of the bound peptide. Figure 4.22 describes the basic concept of fluorescence polarization.    Fluorescence polarization experiments were conducted at 4 ℃, excitation wavelength at 485 nm was used, and the signals of emission at 535 nm were monitored. To determine the dissociation constant KD1 for the Bcl-xL protein with flu-Bak peptide, we used a 15 nM solution of flu-Bak pepitde in PBS buffer and added increasing amounts of Bcl-xL protein to obtain a saturation binding curve (Figure 4.23). KD1 (253 nM, which is close to 264 nM in the literature143) for the Bcl-xL protein with flu-Bak peptide was calculated from the following equation and the IC50 value was obtained from the resulting curve.  Figure 4.22 The basic concept of fluorescence polarization. 151  where KD1 stands for the dissociation constant of fluoresceine probe; RT stands for the total concentration of Bcl-xL protein; LTF stands for the total concentration of flu-Bak peptide; and FBP stands for the fraction of bound flu-Bak peptide.    A decrease in the fluorescence polarization signal was observed upon competitive inhibition of the flu-Bak-Bcl-xL complex by the Bak peptide or cavitein 4B. The KD values for the free Bak peptide (147 ± 16 nM, close to 154 nM in literature)143 and cavitein 4B  (51 ± 8 nM) were obtained from the fluorescence polarization assay. Figure 4.24 presents the inhibition curves from the competitive experiments. The data show that our synthetic protein, KD1 = (RT * (1 - FBP) + LTF * FBP2) / FBP - LTF  Figure 4.23 Binding of fluorescent peptides to Bcl-xL protein.  152 cavitein 4B, has three-fold greater affinity towards Bcl-xL protein than the Bak peptide itself. The following equation was used to calculate the dissociation constant KD2:  where KD1 stands for the dissociation constant of fluoresceine probe; KD2 stands for the dissociation constant of cavitein; RT stands for the total concentration of Bcl-xL protein; LTF stands for the total concentration of flu-Bak peptide; FBP stands for the fraction of bound flu-Bak peptide; and LT stands for the total concentration of cavitein.  KD2 = KD1 * FBP * {LT / [LTF * FBP2- (KD1 + LTF + RT) * FBP + RT] - 1 / (1- FBP)} Figure 4.24 A fluorescence polarization assay shows that cavitein 4B binds Bcl-xL with    higher affinity than Bak peptide. The Kd value for the Bak Cavitein was    determined by competitive inhibition of fluorescein tagged Bak peptide (15 nM)   and Bcl-xL (500 nM) complex. The experimental data was fitted to a sigmoidal   dose-response nonlinear regression model.  153  Hetero Bak cavitein 3B1O shows a slightly weaker binding affinity towards Bcl-xL protein than the Bak peptide counterpart. For heterocaviteins with two o-pN1GG peptides, BOBO and 2B2O, their binding affinities towards Bcl-xL protein are similar to each other, but weaker than 3B1O. Furthermore, heterocaviteins with three o-pN1GG peptides (1B3O) or three pN1GG peptides (1B3N) exhibit no association with the Bcl-xL protein. Figure 4.25 presents the inhibition curves of all these caviteins from the competitive inhibition experiments and Table 4.5 summarizes their IC50 values along with their KD values obtained from the experiment.  Figure 4.25 Fluorescence polarization assay of cavitein 4B and all the hetero Bak caviteins   binds Bcl-xL protein. The experimental data was fitted to a sigmoidal    dose-response nonlinear regression model. 154 Table 4.5 Summary of competitive inhibition experiment results. Inhibitor IC50 (nM) KD2 (nM) Bak Peptide 558 ± 25 147 ± 16 Cavitein 4B 408 ± 22 51 ± 8 Cavitein 3B1O 623 ± 37 181 ± 19 Cavitein 2B2O or BOBO 935 ± 73 394 ± 52 Cavitein BOBO or 2B2O 961 ± 72 375 ± 46 Cavitein 1B3O No activity Cavitein 1B3N No activity    From the competitive inhibition between the Bak peptide and the cavitein inhibitor, we found that cavitein 4B shows an improved binding affinity compared with the Bak peptide. However, the binding affinity of the hetero Bak caviteins decreases as the number of o-pN1GG peptide increases. With three o-pN1GG peptides or pN1GG peptide, the heterocavitein totally lost its ability of binding towards the Bcl-xL protein. This is not surprising, as the increased bundling would be expected to make the Bak strand less available for binding to the Bcl-xL protein.  4.2.7 Trypsin Cleavage Assay  The proteolytic cleavage assays were performed to test the stabilities of our Bak peptide containing caviteins in the presence of trypsin, which is a protease that hydrolyzes proteins and is found in the digestive system.163 Trypsin cleaves the amide bond selectively 155 so that only at the carboxyl side of lysine or arginine residues in the peptide chain (not followed by a proline) can the peptide be cleaved by trypsin.  To test if our cavitand-based template is able to protect the incorporated Bak peptides from proteolytic cleavage, we assayed cavitein 4B and the Bak peptide in the presence of trypsin, which is expected to cleave the peptide at the arginine residue (Arg 76).143 The experiment was conducted at 4 °C with 0.5 mM Bak peptide or cavitein in phosphate buffer saline in the presence of 1 ng/μL of trypsin. The amount of the remaining cavitein was calculated by the area of the band from analytical HPLC. Comparison of the initial reaction rate of cleavage by trypsin indicated that cavitein 4B (0.0083 mM/min) is proteolyzed roughly 2.5-fold slower than the free Bak peptide (0.019 mM/min, 0.018 mM/min from literature143) (Figure 4.26). As expected, the rigidity of our cavitand template affords proteolytic stability to the Bak peptides on the cavitand.  Our previously synthesized cavitein N1GG (cavitein with four pN1GG peptide) is easily cleaved by trypsin into various fractions due to the multiple lysine residues on its peptide chain. However, cavitein o-N1GG (all the lysine residues in the N1GG cavitein are replaced by ornithine) is fully resistant towards trypsin (data not shown).  The results in Figure 4.26 depict the initial reaction rates of all the heterocaviteins in the presence of trypsin. Heterocavitein 3B1O, which has one o-pN1GG peptide, is proteolyzed approximately 6-fold slower than the free Bak peptide. Caviteins BOBO and 2B2O have similar cleavage rates, which are around 14-fold slower than the free Bak peptide. Furthermore, the heterocavitein with three o-pN1GG peptides (1B3O) is fully protected from being cleaved by trypsin. Table 4.6 summarizes their initial reaction rates from the 156 cleavage assay. All these data indicate that the o-pN1GG peptide in the cavitein plays a important role in stabilizing the Bak peptide from proteolytic cleavage. This stabilization may come from the induced helical structure of the Bak peptide by o-pN1GG peptides and their helical bundling.      Figure 4.26 Proteolytic stabilities of all the caviteins. 157 Table 4.6 Summary of trypsin cleavage assay results. Substrate Initial Rate (mM/min) Bak Peptide 0.019 Cavitein 4B 0.0083 Cavitein 3B1O 0.0031 Cavitein 2B2O or BOBO 0.0013 Cavitein BOBO or 2B2O 0.0014 Cavitein 1B3O Cannot be cleaved    Table 4.7 lists the relative overall efficiency of Bak peptide, caviteins 4B, 3B1O, 2B2O and BOBO. The relative overall efficiency (e) is the product of [trypsin stability (a) relative to Bak peptide] and [inhibition ability (b) relative to Bak peptide] divided by the number of Bak peptides (n): e = a * b / n. So both the inhibition ability and proteolytic stability as well as the numbers of active peptides are considered in obtaining this value. As we can see, all the designed caviteins show improvement compared with the Bak peptide. Cavitein 4B, which shows a good binding affinity but only moderate resistance towards protease, is 1.7-fold more efficient than Bak peptide. Cavitein BOBO and 2B2O, though their inhibition abilities are poor, have strong resistance towards trypsin, and thus have higher efficiency than that of cavitein 4B. The second row of Table 4.7 is the overall efficiency, which takes the cavitein as a whole entity, because it is very unlikely that more than one Bak strand can be involved in inhibitory, we treat each cavitein as a single inhibitor. Cavitein 4B, as a whole, exhibits 6.8-fold overall efficiency and all other heterocaviteins shows around 5 times higher efficiency than the free Bak peptide. 158 Table 4.7 Relative overall efficiency (e) and overall efficiency (e * n) of all the caviteins.  Bak Peptide 4B 3B1O 2B2O (BOBO) BOBO (2B2O) e 1 1.7 1.6 2.7 2.6 e * n 1 6.8 4.8 5.4 5.2    We used the relative overall efficiency to compare all our synthesized caviteins with the Bak peptide and found that each Bak peptide in the cavitein 4B is more efficient than Bak peptide itself when both inhibition ability and proteolytic stability are considered. Meanwhile, the o-pN1GG peptide does have the function of protecting the Bak peptide from being hydrolyzed by trypsin, and further increases the overall efficiency of the whole cavitein system.   As for the stoichiometry of the Bcl-xL - cavitein system, we assume that the ratio of Bcl-xL protein to bound cavitein is most likely 1:1 because the cavitein is a small molecule compared with the Bcl-xL protein, and it might be sterically unfavorable to have two or more Bcl-xL proteins bind with one cavitein. Experiments to determine the stoichiometry have been considered: Analytical ultracentrifugation (AUC) is challenging because there is a complex mixture of entities in solution with different molecular weights; Gel permeation chromatography (GPC) is great for normal polymers but impractical for biomolecules which are easily denatured; DOSY NMR also works theoretically, but the data interpretation of such a complex mixture of species makes it impractical. Mass spectrometry has been carried out using MALDI-TOF, but only the signals of the Bcl-xL protein and the cavitein were detected. This probably results from the high energy of the laser that disrupts any complexes. 159 There is also no observable signal of the complex from ESI for the mass is out of the detectable range for the complexes.  4.2.8 GuHCl Denaturation Studies  The GuHCl denaturation curves of caviteins 4B and 3B1O are shown in Figure 4.27. We can see that the thermodynamic stabilities of these two caviteins are very weak. They start to denature significantly in the presence of a low concentration of GuHCl. Figure 4.28 exhibits the GuHCl denaturation curves of cavitein 2B2O, BOBO and 1B3O. The ∆G°H2O of cavitein 1B3O is calculated to be approximately 3 kcal/mol higher than those of the other two counterparts. (Table 4.8) The studies of the denaturation are consistent with the results from the proteolytic cleavage assay in which cavitein 1B3O shows full resistance towards trypsin. Because of its relatively strong thermodynamic stability, it becomes the most resistant one when treated with trypsin. In the same way, 2B2O and BOBO, which have relatively moderate thermodynamic stability, show their moderate standing when they are treated with trypsin. 160    Figure 4.27 GuHCl dentaturation of cavitein 4B and heterocavitein 3B1O. Each curve was   acquired at pH 7.0 in 50 mM phosphate buffer at 25 °C at a cavitein    concentration of 25 μM. 161  Table 4.8 ∆G°H2O values for caviteins 1B3O, 2B2O and BOBO. Cavitein ∆G°H2O (kcal mol-1) BOBO or 2B2O -2.36 ± 0.1 2B2O or BOBO -1.79 ± 0.3 1B3O -5.03 ± 0.5    Figure 4.28 GuHCl dentaturation of heterocavitein 2B2O, BOBO and 1B3O. Each curve   was acquired at pH 7.0 in 50 mM phosphate buffer at 25 °C at a cavitein    concentration of 25 μM. 162 4.3 Chapter Summary and Conclusion  In this project, we incorporated Bak peptides onto the cavitand template instead of the peptides which are solely designed based on the minimalist approach as we normally do in our lab. Moreover, we used our synthesis strategy to make a series of heterocaviteins to further study the inhibitory activity to the Bcl-xL/Bak interaction, as well as their proteolytic stability.   Bak cavitein 4B was synthesized where the Bak peptides exist mainly as random coils. Only when o-pN1GG peptide was incorporated did the CD signal show characteristic bands of an alpha helix. From the competitive inhibition assay, cavitein 4B shows a 3-fold higher affinity in binding with Bcl-xL protein and a 2.5-fold stronger resistance to proteolysis than the wild type Bak peptide. Hetero Bak caviteins, which have 1 to 3 o-pN1GG peptides incorporated, show lower inhibitory abilities whereas their proteolytic stabilities are largely improved. The GuHCl denaturation experiments further confirm that the presence of o-pN1GG peptides stabilize the entire cavitein system. As for caviteins BOBO and 2B2O, they are obtained and assayed separately, but we did not distinguish them as their similar activity did not warrant it.  The cavitein system provides a platform onto which, in principle, any potentially helical peptide can be incorporated. Functions such as inhibition of protein-protein interactions can then be explored. Also, challenging synthesis and artificiality are minimized, while proteolytic stability is enhanced.  163 4.4 Experimental  4.4.1 Cavitand Synthesis  The synthesis of arylthiol cavitand with methyl feet is the same as that mentioned in Chapter 3. Detailed description can be found in section 3.4.1. 4.4.2 Peptide Synthesis 4.4.2.1 General Procedures of Peptide Synthesis  All the peptide syntheses were performed on a C S Bio (Menlo Park, CA) 136XT peptide synthesizer. Solvents used in the synthesizer like N-methylpyrrolidone (NMP) and DMF, and all the amino acids were purchased from Advanced Chemtech (Louisville, KY). Other chemicals involved were from Sigma-Aldrich. Rink’s amide resin was used to afford C-terminal amides.125 The peptides were synthesized on a 0.25 mmol scale. Standard Fmoc protected amino acids were used during the synthesis with acid labile protecting groups for alanine, arginine, asparagine, aspartic acid, glycine, glutamine, isoleucine, leucine and valine side chains. The following amino acids were used: Fmoc-L-Ala-OH, Fmoc-L-Arg(Pbf)-OH, Fmoc-L- Asn(Trt)-OH, Fmoc-L-Asp(OtBu)-OH, Fmoc-L-Gln(Trt)-OH, Fmoc-Gly-OH, Fmoc-Ile-OH, Fmoc-L- Leu-OH and Fmoc-L-Val-OH.  Before the peptide was cleaved from the resin, it was activated with chloroacetyl chloride. After the resin was washed with DCM, it was treated with 95% of TFA/H2O to cleave the peptide from the resin for 2 hours. Afterwards, the peptide was purified by reverse 164 phase HPLC using a linear gradient of acetonitrile (0.05% TFA) and water (0.1% TFA) on a Waters C-18 Delta Pak column (300 × 19 mm2, 300 Å, 15 µm) at a flow rate of 10 mL min-1. The fraction signal was monitored by a UV detector at 229 nm. The fraction solution from HPLC purification was frozen and the solvent was removed by lyophilization. Analytical reversed-phase HPLC (Waters Delta Pak C-18 column (300 × 3.9 mm2, 300 Å, 15 µm)) was used to assure the purity of the peptide by the observation of a single peak. The identity of the peptide was confirmed by mass spectrometry (ESI-MS) with a Bruker Esquire using electrospray ionization as the ion source. 4.4.2.2 Synthesis of Free Bak Peptide  The free N-terminus peptide linked resin (~0.25 mmol peptide) was acetylated by acetic anhydride (6 mL) in NMP (N-methyl-2-pyrrolidinone, 2 mL) for 2 hours at room temperature. After the peptide linked resin was filtered and washed with DCM, it was treated with 10 mL of ice cold TFA/H2O (95% TFA) solution for approximately 2 hours. After the peptide was cleaved from the resin and dissolved in the above solution, the resin was filtered off and the filtrate was transferred to a round bottom flask. Its volume was condensed in reduced pressure to approximately 5 mL. This concentrated cleavage solution was pipetted into ice cold diethyl ether to precipitate the peptide. The solid peptide was separated from the ether after centrifugation (3000 rpm) for 5 minutes. Steps of washing with diethyl ether and centrifugation were repeated for another two times to completely remove TFA and small molecular weight impurities. The remaining peptide pellet was dissolved in water with a little acetic acid to speed up the dissolving process. RP-HPLC was performed after the peptide solution was filtered by a 0.45 µm pore disposable filter. The purity was assessed by 165 the observation of a single peak from analytical reversed-phase HPLC (95% pure). After lyophilization, the peptide was obtained as a white solid (150 mg).  MS (free Bak peptide): m/z 1765.1  4.4.2.3 Synthesis of Bak Peptide and o-pN1GG Peptide  After the free N-terminus peptide linked resin (~0.25 mmol) was collected from the synthesizer and washed by methanol, it was reacted with chloroacetyl chloride (100 µL, 1.26 mmol) and DIPEA (150 µL, 0.86 mmol) in dried DMF (5 mL) for 1 hour at room temperature. Later, a second portion of chloroacetyl chloride (50 µL, 0.63 mmol) was added to the reaction system for another hour. After it was filtered and was washed with DCM, the peptide linked resin was treated with 10 mL of ice cold TFA/H2O (95% TFA) solution for approximately 2 hours. After the peptide was cleaved from the resin and dissolved in the above solution, the resin was filtered off and the filtrate was transferred to a round bottom flask. Its volume was condensed in reduced pressure to approximately 5 mL. This concentrated cleavage solution was pipetted into ice cold diethyl ether to precipitate the peptide. The solid peptide was separated from the ether after centrifugation (3000 rpm) for 5 minutes. Washing with diethyl ether and centrifugation were repeated twice for complete removal of TFA and small molecular weight impurities. The remaining peptide pellet was dissolved in water with a little acetic acid to speed up the dissolving process. RP-HPLC was performed after the peptide solution was filtered by a 0.45 µm pore disposable filter. Its purity was assessed by the observation of a single peak from analytical reverse phase HPLC (95% pure). After lyophilization, the peptide was obtained as a white solid (180 mg). 166 MS (Bak peptide): m/z 1856.1 MS (o-pN1GG peptide): m/z 1803.0 4.4.2.4 Synthesis of Flu-Bak Peptide  The 5-carboxylfluorecein was linked to the solid phase peptide linked resin just like an amino acid because it also has a carboxyl group on it. The peptide was cleaved from the resin by a 1.5-h treatment of 95% TFA : 5% H2O solution. After the peptide was cleaved from the resin, the resin was filtered off and the cleavage solution was transferred to a round bottom flask. Its volume was condensed in reduced pressure to approximately 5 mL. This concentrated cleavage solution was pipetted into ice cold diethyl ether to precipitate the peptide. The solid peptide was separated from the ether after centrifugation (3000 rpm) for 5 minutes. Washing with diethyl ether and centrifugation were repeated twice to completely remove TFA and small molecular weight impurities. The remaining peptide pellet was dissolved in water with a little acetic acid to speed up the dissolving process. RP-HPLC was performed after the peptide solution was filtered by a 0.45 µm pore disposable filter. Its purity was assessed by the observation of a single peak by analytical reverse phase HPLC (95% pure). After lyophilization, the peptide was obtained as a white solid (90 mg). MS (flu-Bak peptide): m/z 2083.5    167 4.4.3 Cavitein Synthesis 4.4.3.1 General Procedures of Cavitein Synthesis  All the reagents used for the syntheses of the caviteins were reagent grade and purchased from Sigma-Aldrich. The solvent for the reaction was HPLC grade N, N-dimethylformamide (DMF) and reagent grade NMP. They were freeze-pump-thawed under argon and dried by 4 Å molecular sieves. The crude caviteins were purified by reversed-phase HPLC using a Phenomenex Jupiter C-4 column (250 × 20 mm2, 300 Å, 15 µm) with an acetonitrile (0.05% TFA) and water (0.1% TFA) solvent system at flow rate 10 mL/min. Their purities were assessed by analytical RP-HPLC. MALDI-TOF (Matrix Assisted Laser Desorption Ionization) and ESI were both used to confirm the identities of the caviteins. Molecular weights determined by MALDI were acquired on a Bruker Autoflex. The matrix used was 2,5-dihydroxybenzoic acid (DHB). 4.4.3.2 Syntheses of Caviteins N1GG, o-N1GG and 4B  The synthesis of cavitein N1GG has been reported by Dr. Freeman previously. It is similar to the syntheses of caviteins o-N1GG and 4B in the following procedures.  In 1 mL of dried and degassed NMP solvent, arylthiol cavitand (1.4 µmol) and purified activated peptide (6 equiv, 8.4 µmol) were dissolved. DIPEA (28 µmol, 5 µL) was added to the solution. The solution was bubbled by argon gas at room temperature for 4 hours. The reaction solution was condensed in vacuo to a volume of around 200 µL. The solution was then diluted with H2O was filtered with a 0.45 µm pore disposable filter. The 168 crude cavitein was purified by RP-HPLC. A white cotton-like solid was obtained after lyophilization. (6 mg, yield: ~50%) MS (N1GG cavitein): m/z 8014.2 MS (o-N1GG cavitein): m/z 7791.2 MS (4B cavitein): m/z 8007.0 4.4.3.3 Syntheses of Heterocaviteins  The process of syntheses of heterocaviteins includes two parts. The first one is to synthesize the partially incorporated caviteins 12, 13, 14, 16, 17 and 18, and the second is to make the whole caviteins (1B3O from 12; 3B1O from 16; BOBO and 2B2O are from 13 and 14 or 17 and 18). 4.4.3.3.1 Synthesis of Heterocaviteins 1B3N and 1B3O  In 0.5 mL of dried and degassed NMP solvent, arylthiol cavitand (3 equiv, 7.0 µmol) was dissolved in the presence of DIPEA (56 µmol, 10 µL). Purified activated Bak peptide (1 equiv, 2.3 µmol) was predissolved in 0.5 mL of NMP. The peptide solution was then added drop-wise into the cavitand solution in a 40-minute period under argon atmosphere. Afterwards, acetic acid solution (20% in water) was added to quench the reaction and to prevent the dimerization between the unreacted thiol groups. The solution was then filtered with a 0.45 µm pore disposable filter. The crude partially incorporated caviteins mixture of 12, 13 and 14 was separated by RP-HPLC and each fraction was collected. Lyophilization was performed under argon atmosphere to prevent oxidization of the unlinked thiol groups. 169 Caviteins 12, 13 and 14 were obtained as white solids after lyophilization. (12: 2.5 mg, yield: ~43%; 13 (or 14): 1.5 mg; 14 (or 13): 0.5 mg) MS (12 cavitein): m/z 2544.3 MS (13 cavitein): m/z 4366.4 MS (14 cavitein): m/z 4366.4  For the synthesis of the whole cavitein, cavitein 12 (2.5 mg, 1 µmol) was dissolved in 0.7 mL of dried and degassed DMF solvent in the presence of DIPEA (28 µmol, 5 µL). Purified activated pN1GG or o-pN1GG peptide (5 equiv, 5 µmol) was predissolved in 0.3 mL of DMF and was slowly added into the reaction vial. The reaction was bubbled by argon gas at room temperature for 4 hours. DMF was condensed in vacuo to a volume of around 200 µL. The solution was then diluted with H2O and was filtered with a 0.45 µm pore disposable filter. The crude cavitein was purified by RP-HPLC. A white solid was obtained after lyophilization. (1B3N:4.5 mg, yield: ~57%; 1B3O: 5.5 mg, yield: ~70%) MS (1B3N cavitein): m/z 8013.5 MS (1B3O cavitein): m/z 7843.2 4.4.3.3.2 Synthesis of Heterocavitein 3B1O  Synthesis of heterocavitein 3B1O is similar to that of 1B3O described in the above section. In the beginning, it was the o-pN1GG peptide that reacted with the excessive amount of arylthiol cavitand (1:3). Caviteins 16, 17 and 18 were obtained as white solid after lyophilization. (16: 3 mg, yield: ~47%; 17 (or 18): 2 mg; 18 (or 17): 1 mg) 170 MS (16 cavitein): m/z 2490.4 MS (17 cavitein): m/z 4259.3 MS (18 cavitein): m/z 4259.3  The whole cavitein was made in NMP solution and cavitein 3B1O was obtained as a white solid after lyophilization. (3B1O: 4.5 mg, yield: ~58%) MS (3B1O cavitein): m/z 7951.6  4.4.3.3.3 Syntheses of Heterocaviteins 2B2O and BOBO  Partially incorporated caviteins 17 and 18 were used to synthesize heterocaviteins 2B2O and BOBO.  Caviteins 17 (4 mg, 1 µmol) and 18 (2 mg, 0.5 µmol) were dissolved in 0.7 mL of dried and degassed NMP solvent in the presence of DIPEA (28 µmol, 5 µL) separately. Purified activated Bak peptide (4 equiv) was predissolved in 0.3 mL of NMP solvent (× 2) and was slowly added into the reaction vial. The reactions were bubbled by argon gas at room temperature for 4 hours. The reaction solutions were condensed in vacuo to a volume of around 200 µL. Both of the solutions were then diluted with H2O were filtered with a 0.45 µm pore disposable filter. The crude caviteins were purified by RP-HPLC. White solids were obtained after lyophilization. (2B2O or BOBO :4 mg, yield: ~51%; BOBO or 2B2O: 2.5 mg, yield: ~64%) MS (2B2O or BOBO cavitein): m/z 7897.9 171 MS (BOBO or 2B2O cavitein): m/z 7897.9  4.4.4 Overexpression of Bcl-xL Protein 4.4.4.1 Plasmid Replication  The gene on the pET29b plasmid which is responsible for expressing the Bcl-xL protein was synthesized and cloned by GenScript USA Inc. Piscataway, NJ, USA. The following description is the procedures of the transformation of the genes into competent cells for gene replication and the preparation of the cells for the protein overexpression.  A single DH5α colony was picked from a plate that had been incubated for 16 hours at 37 ℃ and transferred into 25 mL of LB (Luria-Bertani) medium. The culture was incubated for 6 hours at 37 ℃ with a shaking rate of 300 rpm. 10 mL of this starting culture was added into 250 mL of LB medium and was incubated overnight at 22 ℃ with moderate shaking at 225 rpm. When its OD600 reached 0.55, the culture was cooled in ice bath for 10 minutes. The cells were harvested by centrifugation at 3,900 rpm for 10 minutes at 4 ℃. The cell pellet was resuspended gently in 80 mL of ice-cold transformation buffer (MnCl2・4H2O: 55 mM, CaCl2・2H2O: 15 mM, KCl: 250 mM, PIPES (piperazine-1,2-bis[2-ethanesulfonic acid]): 10 mM) and then centrifuged at 3,900 rpm for 10 minutes at 4 ℃. Later 20 mL of ice-cold transformation buffer and 1.5 mL of DMSO were added to resuspend the pellet, and then the solution was stored in ice for 10 minutes. The pET29b plasmid (1 ng/µL) was added into the tube that had the competent DH5α cells, and then the combined mixture was stored on ice for another 30 minutes. The tube was then transferred to a 42 ℃ water bath for exactly 172 90 seconds. After that, it was cooled down in an ice-bath for 2 minutes. 800 µL of SOC medium (2% tryptone, 0.5% yeast extract, NaCl: 10 mM, KCl: 2.5 mM, MgCl2: 10 mM, MgSO4: 10 mM, and glucose: 20 mM) was added and the mixture was incubated at 37 ℃ with a shaking rate of 225 rpm for 45 minutes. After that, 200 µL of the culture was transferred onto an agar LB medium with 50 µg/mL kanamycin and incubated at 37 ℃ for 12 hours. The collection and purification of the replicated plasmid were performed following QIAGEN plasmid purification protocol.  4.4.4.2 Overexpression and Purification of Bcl-xL Protein  Plasmid pET29b was transformed into E. coli BL21 (DE3), and the cells were incubated overnight in 5 mL of Luria-Bertani (LB) medium containing 50 µg/mL kanamycin at 37 ℃ with shaking at 225 rpm. The overnight cultures were poured into 500 mL of LB medium containing 50 µg/mL kanamycin, and the combined mixture were incubated at 37 ℃ with shaking at 225 rpm until the OD550 reached mid-log growth (~ 0.5). IPTG (isopropyl 1-thio-β-D-galactopyranoside) was added (final concentration: 1 mM) to induce protein overexpression and the cultures continued to incubate at 22 ℃ with shaking at 185 rpm for another 2 hours. The cells were then harvested at 5,000 rpm for 40 min at 4 ℃.  The pellets were then resuspended in 15 mL of lysis buffer (50 mM phosphate buffer at pH 7.4, 150 mM NaCl, 25 mM imidazole, 10% glycerol and 1% tween) and lysed three times at 20,000 psi using an ice-cooled French Pressure Cell. The cell lysate was centrifuged at 12,000 rpm for 1 hour and then was filtered with a 0.45 µm pore disposable filter before affinity chromatography. A 10 mL column containing Chelating Sepharose was first washed 173 with 1 column volume of H2O, 3 column volumes of EDTA (50 mM) and 2 column volumes of H2O. Then, it was charged with 2 column volumes of 100 mM NiSO4, followed by 2 column volumes of H2O, and then 3 column volumes of running buffer (50 mM phosphate buffer, 150 mM NaCl, 25 mM imidazole, 0.1% tween, pH 7.4). The filtered cell lysate was then loaded into the column and 5 column volumes of previous used running buffer were passed through; later 5 column volumes of a higher concentration of imidazole running buffer (50 mM phosphate buffer, 150 mM NaCl, 125 mM imidazol, 0.1% tween, pH 7.4) were passed through. Finally, running buffer with 250 mM imidazole (50 mM phosphate buffer, 150 mM NaCl, 250 mM imidazol, 0.1% tween, pH 7.4) was passed through and was collected in tubes (5 mL each). The purified protein was concentrated by passing it through an Amicon Ultra-4 (Millipore, 10,000 MWCO) membrane filter at 5,000 rpm and it was exchanged into the assay buffer (20 mM phosphate buffer, 50 mM NaCl, 1 mM EDTA, 0.05% pluronic F-68, pH 7.4) by spinning five times through this filter, each time with 1.5 mL of the assay buffer. The concentrated solution with purified Bcl-xL protein was snap-frozen in liquid nitrogen and stored at -80 ℃.  4.4.5 Circular Dichroism  The CD spectra were acquired on a JASCO J-815 CD spectrophotometer. The experiments were conducted at 25 ℃ with a quartz cell of 1 mm path length. The concentration of cavitein was 25 µM in 50 mM phosphate buffer at pH 7.0 and it was filtered by a 0.45 µm pore disposable filter. Each spectrum was the result of the average of three scans with solvent baseline correction. Mean residue ellipticity was calculated by the 174 following equation: (θobs: measured ellipticity in millidegrees, C: concentration of the cavitein in M, n: number of residues in the cavitein, l: path length in centimeters.) [θ]M.R.E. = θobs / (10Cnl) 4.4.6 Protein Binding Studies  Fluorescence polarization experiments were conducted at room temperature in a 96-well plate.160 Polarization measurements were recorded on a DTX 880 Multimode Detector (Beckmen). Excitation at 485 nm was used for the fluorescein-containing peptide and the emission at 535 nm was monitored. All samples were prepared in 4% DMSO and 0.1% Pluronic F-68 (Sigma- Aldrich) assay buffer. The binding affinity values were determined by fitting the experimental data to a sigmoidal dose-response nonlinear regression model on OriginPro 8.5. To determine the binding affinity of the flu-Bak peptide, we used a 15 nM solution of flu-Bak in PBS buffer at 4 °C and added an increasing concentration from 0 nM to 2.5 μM of the Bcl-xL protein. A saturation binding curve was afforded and the IC50 value obtained to calculate the binding affinity for the fl-Bak and Bcl-xL. As for the competitive inhibition, an increasing concentration (from 10 nM to 100 nM) of the free Bak peptide and all the caviteins were added into a solution of 15 nM flu-Bak peptide and 500 nM Bcl- xL. Fluorescence polarization assays were performed after the above solutions were incubated at 4 °C for 1 hour. The following equations were used to determine the KD1 and KD2 values: 164, 165  175  where KD1 stands for the dissociation constant of fluoresceine probe; KD2 stands for the dissociation constant of cavitein; RT stands for the total concentration of Bcl-xL protein; LTF stands for the total concentration of flu-Bak peptide; FBP stands for the fraction of bound flu-Bak peptide; and LT stands for the total concentration of cavitein.  4.4.7 Trypsin Cleavage Studies143  1 ng/μL of trypsin and 0.5 mM of Bak peptide or cavitein in phosphate buffer saline was incubated at 4 °C. At different time intervals, 50 μL of the above solution was quenched with 50 μL of 2% aqueous TFA solution, and then the combined solution was injected into analytical reversed-phase column (Waters Delta Pak C-18 column (300 × 3.9 mm2, 300 Å, 15 µm)) to analyze the change in the area of the peptide/cavitein peak compared to the area of external control.  Figure 4.29 is the analytical HPLC results of trypsin cleavage assay of Bak peptide and cavitein 4B. We only measured the initial rates in the first ten minutes because multiple peaks of cleaved cavitein 4B fractions were observable in the UV spectrum. MS confirmed that these fractions are the cavitein fragments.  KD1 = (RT * (1 - FBP) + LTF * FBP2) / FBP - LTF   KD2 = KD1 * FBP * {LT / [LTF * FBP2- (KD1 + LTF + RT) * FBP + RT] - 1 / (1- FBP)} 176  4.4.8 GuHCl Denaturation Studies  Guanidine HCl denaturation studies on caviteins B4, 3B1O, 2B2O, BOBO and 1B3O were performed in a 50 mM potassium phosphate buffer (pH 7.0) in the presence of GuHCl with its concentration varying from 0 to 8.0 M. The CD signals were observed at [θ]222. The experiments were conducted at 25 ℃ with a quartz cell of 1 mm path length. The concentration of cavitein was 25 µM and it was filtered by a 0.45 µm pore disposable filter. Each data point was the result of the average of three scans with solvent baseline correction. ∆G°H2O was determined by the following equation:  Figure 4.29 HPLC results of trypsin cleavage assay. The extra three peaks in 30 min HPLC   for cavitein 4B correspond to multiple cleaved cavitein as there are four peptide   in it. (confirmed by MALDI-TOF)  177   y: fraction folded ([θ]222);  x: concentration of GuHCl;  F: the least-squares analysis of pre-transitional [θ]222 intercept;  U: the least-squares analysis of post-transitional [θ]222 intercept;  G: ∆G°H2O;  m: ∆G°/[GuHCl];  R: the universal gas constant;  T: temperature;  a: a constant determined by least-squares analysis.           178 Chapter 5 Thesis Summary and Conclusions     In this thesis, we demonstrated the maturity of our strategy in de novo protein design. Not only did we find peptide sequences that are favourable for forming a pentameric helical bundle, but most importantly, the design of TASPs is found to be effective in applications either as a reaction rate enhancer or in modulation of protein-protein interactions.  In Chapter 2, we designed three peptide sequences with different lengths (L, M and S, and six caviteins based on these peptides: L4 and L5; M4 and M5; S4 and S5). The longer peptide sequence-based caviteins display higher helicity and higher thermodynamic stability than caviteins with shorter peptide sequences. Both caviteins L5 and M5, according to the trend of their denaturation curves, may have higher thermodynamic stability than their four-helix bundle counterparts, but their free energies of native state cannot be quantified since we found their incomplete denaturation even in the presence of 8 M GuHCl. Variable temperature circular dichroism experiments were performed for caviteins L4 and L5 in 8 M GuHCl, which shows that they were still being denatured as the temperature increased. As for peptide S-based caviteins, though they were fully denatured by GuHCl and their ∆G°H2O values were determined, caviteins S4 and S5 both exhibit lower thermodynamic stability than their counterparts with longer peptide sequences. Despite their weakness in the presence of a denaturant, the free energy gap between the five-helix bundle and the four-helix bundle is narrowed compared with the peptide sequence that had been designed for a four-helix bundle. Thus, we can conclude that, by enlarging the hydrophobic region in the helical core, 179 our strategy in designing TASPs with a peptide sequence that intends to favour for a five-helix bundle does provide us with a meaningful way to study the protein folding problem.  In Chapter 3, we successfully put our TASPs (Q4-H and Q4-H2) into application for the first time, mimicking the process of ester hydrolysis.  Both cavitein Q4-H and cavitein Q4-H2 are great hydrolysis rate enhancers. Cavitein Q4-H2 was made by altering the positions of histidine residues in cavitein Q4-H, and we found that the positions of the histidine residues are involved in influencing the rate of the hydrolysis reaction. In the helix bundle of Q4-H, the histidine residues are located near the C-terminus, away from the template molecule, and these four histidines are close to each other in the center of the inner core. Histidines in Q4-H2 are near to the template molecule and located at the interface between the hydrophobic region and hydrophilic region of the helix bundle. These histidine residues are separated from each other at the interface of the bundle. Due to these structural differences, cavitein Q4-H2 shows greater reactivity than Q4-H in ester hydrolysis, particularly for the substrate with a long carbon chain. Meanwhile, we found that for short chain substrates, lysine residues also participated in the process of hydrolysis, and the post-reaction caviteins are adducts of a number of acyl groups from the reaction.   Chapter 4 presents our second attempt to put our TASP system into application. It was also found to be effective in protein-protein interactions. Cavitein 4B was designed, in which we concentrated four Bak peptides into one cavitand molecule. It shows a 3-fold higher affinity in binding with a Bcl-xL protein and a 2.5-fold stronger resistance to proteolysis than the wild type Bak peptide. From its CD signal, Bak peptide strands in cavitein 4B still exist as random coils. When an o-pN1GG peptide strand, which was based upon a highly helical peptide sequence (pN1GG) where all the lysines were replaced by 180 ornithines, was incorporated, cavitein 4B’s CD signal shows characteristic bands of an alpha helix. The hetero Bak caviteins (3B1O, 2B2O/BOBO and 1B3O), which have 1 to 3 o-pN1GG peptides incorporated respectively, show lower inhibitory ability whereas their proteolytic stability is largely improved. This stabilization may come from the induced helical structure of the Bak peptide strands by the o-pN1GG peptide strand and their helical bundling. GuHCl denaturation experiments confirmed that the presence of o-pN1GG peptides helps stabilize the entire cavitein system: The heterocaviteins with more o-pN1GG peptides show higher thermodynamic stability. Because of the similar properties found for caviteins BOBO and 2B2O, we did not attempt to distinguish which is which. The cavitein system provides us with a platform onto which, in principle, any potentially helical peptide can be incorporated, and functions such an inhibition of protein-protein interactions can be exploited. Also, challenging synthesis and artificiality are minimized, while proteolytic stability is enhanced. In this project, we incorporated Bak peptides, which are from the BH3 domain of a natural molecule, Bak protein, onto a cavitand template instead of peptides that were designed artificially based on the minimalist approach as we normally do in our lab. Moreover, we used our synthetic strategy to synthesize a series of heterocaviteins to explore this project in depth.  The first project of exploring a potentially five-helix bundle forming peptide sequence is our group’s extension of basic research in the protein folding problem. We have had a systematic study of our four-helix bundle TASP system, and research on the five-helix bundle directed us towards a more extensive understanding of our strategy in protein-molecule maneuvering. As A. Kornberg described basic research as arising out of curiosity, 181 our curiosity led us to explore this peptide sequence and we proved that the enlargement of the peptide’s hydrophobic region does help its tendency to form a five-helix bundle.  In contrast with basic research, we also did applied research in our cavitein system. Although our histidine-containing caviteins used in ester hydrolysis reactions turned out to be acylated, we still believe that our cavitein system provides the enzyme mimic field with a new platform to design novel artificial enzymes for the following reasons: 1. The cavitand template predefines the positions and the number of the potentially active peptide molecules and helps them bundle each other to induce a helical form; 2. The peptide sequence can be easily modified and its functions optimized.  In the studies of Bcl-xL/Bak interaction, more advantages of our cavitein system are manifested. First, the inhibitory activity of our designed caviteins in Bcl-xL/Bak interaction indicates that our cavitein system has the potential to be widely adopted as an inhibitor in medical applications. Any active helical region in a disease-related protein can be truncated and linked to the cavitand template to form a potentially active inhibitor or antigen. Second, the heterocavitein design strategy offers us a way to improve stability of the target active peptide to proteases. Thirdly, our cavitein system keeps the original peptide sequence itself, without incorporating artificiality to the peptides. Aside from its linkage to the cavitand, it is all “natural”.    182 References  (1) Kauzmann, W. Adv. Prot. Chem. 1959, 14, 1. (2) Sela, M.; White, F. H.; Anfinsen, C. B. Science 1957, 125, 691-692. (3) Schuler, B.; Clarke, J. Nature 2013, 502, 632-633. (4) Levinthal, C. J. J. Chem. Phys. 1968, 65, 44-45. (5) Wetlaufer, D. B. Proc. Natl. Acad. Sci. U. S. A. 1973, 70, 697-701. (6) Baldwin, R. L. Trends Biochem. Sci. 1989, 14, 291-294. (7) Dill, K. A. Biochemistry 1985, 24, 1501-1509. (8) Harrison, S.C.; Durbin, R. Proc. Natl. Acad. Sci. U. S. A. 1985, 82, 4028-4030. (9) DeGrado, W. F.; Summa, C. M.; Pavone, V.; Nastri, F.; Lombardi, A. Annu. Rev. Biochem. 1999, 68, 779-819. (10) Moffet, D. A.; Hecht, M. H. Chem. Rev. 2001, 101, 3191-3203. (11) DeGrado, W. F.; Wasserman, Z. R.; Lear, J. D. Science 1989, 243, 622-628. (12) Mutter, M.; Altmann, E.; Altmann, K. H.; Hersperger, R.; Koziej, P.; Nebel, K.; Tuchscherer, G.; Vuilleumier, S.; Gremlich, H. U.; Müller, K. Helv. Chim. Acta. 1988, 71, 835-847.  (13) Mutter, M.; Vuilleumier, S. Angew. Chem. Int. Ed. 1989, 28, 535-554. (14) Akeson, W. R.; Stahmann, M. A. J. Nutr. 1964, 83, 257-261. (15) Spudich, J. A.;Watt, S. J. Biol. Chem. 1971, 246, 4866-4871. (16) Akeson, W. H.; Amiel, D; Mechanic, G. L.; Woo, S. L.; Harwood, F. L.; Hamer, M. L. Connective Tissue Research 1977, 5, 15-19. (17) Pfeiffer, E. F.; Thum, C.; Clemens, A. H. Horm. Metab. Res. 1974, 6, 339-342. (18) Ramachandran, G.N.; Ramakrishnan, C.; Sasisekharan, V. Journal of Molecular Biology 1963, 7, 95-99. (19) Lovell, S. C.; Davis, I. W.; Arendall, W. B.; De Bakker, P. I. W.; Word, J. M.; Prisant, M. G.; Richardson, J. S.; Richardson, D. C. Proteins, 2003, 50, 437-450. 183 (20) Mortishire-Smith, R. J.; Drake, A. F.; Nutkins, J. C.; Williams, D. H. FEBS Lett. 1991, 278, 244-246. (21) McCallister, E. L.; Alm, E.; Baker, D. Nature Structural Biology, 2000, 7, 669-673. (22) Graeff, M.; Wenkel, S. Biomolecular Concepts 2012, 3, 71-78. (23) Di Sole, F.; Vadnagara, K.; Moe, O. W.; Babich, V. Am. J. Physiol. Renal. Physiol. 2012, 303, 165-179. (24) PDB ID: 1MFT; Lahr, S. J.,  Engel, D. E.,  Stayrook, S. E.,  Maglio, O.,  North, B.,  Geremia, S.,  Lombardi, A.,  DeGrado, W. F. J. Mol. Biol. 2005, 346, 1441-1454. (25) Wilson, I. A.; Skehel, J. J.; Wiley, D. C. Nature 1981, 289, 366-373. (26) Klippenstein, G. L.; Cote, J. L.; Ludlam, S. E. Biochemistry, 1976, 15, 1128-1136. (27) Mueller, G. A.; Gosavi, R. A.; Pomes, A.; Wunschmann, S.; Moon, A. F.; London, R. E.; Pedersen, L. Allergy 2011, 66, 878-885. (28) Bolton, W.; Perutz, M. F. Nature 1970, 228, 551. (29) PDB ID: 1MBA, Bolognesi, M.; Onesti, S.; Gatti, G.; Coda, A.; Ascenzi, P.; Brunori, M. J. Mol. Biol. 1989, 205, 529-544. (30) PDB ID: 2DHB, Bolton, W.; Perutz, M. F. Nature 1970, 228, 551-552. (31) Babu, M.; van der Lee, R.; Sanchez de Groot, N.; Gsponer, J. Current Opinion in Structural Biology 2011, 21, 1-9. (32) a). Onuchic, J. N.; Nymeyer, H.; Garcia, A. E.; Chahine, J.; Socci, N. D. Adv. Protein Chem. 2000, 53, 87-152.  b). Bryngelson, J. D.; Wolynes, P. G. Proc. Natl. Acad. Sci. U. S. A. 1987, 84, 7524-7528. (33) Shea, J. E.; Brooks, C. L. Annu. Rev. Phys. Chem. 2001, 52, 499-535. (34) Socci, N. D.; Onuchic, J. N.; Wolynes, P. G. Proteins 1998, 32, 136–58.  (35) Onuchic, J. N.; Luthey-Schulten, Z.; Wolynes, P. G. Annu. Rev. Phys. Chem. 1997, 48, 545-600. (36) Roder, H.; Shastry, M. R. Curr. Opin. Struct. Biol. 1999, 9, 620-626. (37) Akasaka, K.; Yamada, H. Methods Enzymol. 2001, 338, 134-158. (38) Mok, K. H.; Nagashima, T.; Day, I. J.; Jones, J. A.; Jones, C. J. V.; Dobson, C. M.; Hore, P. J. J. Am. Chem. Soc. 2003, 125, 12484-12492. 184 (39) Garbuzynskiy, S. O.; Ivankov, D. N.; Bogatyreva, N. S.; Finkelstein, A. V. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 147–150.  (40) Pierri, C. L.; De Grassi, A.; Turi, A. Proteins 2008, 73, 351–361. (41) Pace, C. N.; Shirley, B. A.; McNutt, M.; Gajiwala, K. FASEB J. 1996, 10, 75-83. (42) Dill, K. A. Biochemistry 1990, 29, 7133-7155. (43) Berwick, M. R.; Lewis, D. J.; Jones, A. W.; Parslow, R. A.; Dafforn, T. R.; Cooper, H. J.; Wilkie, J.; Pikramenou, Z.; Britton, M. M.; Peacock, A. F. A. J. Am. Chem. Soc. 2014, 136, 1166-1169.   (44) Bellows-Peterson, M. L.; Fung, H. K.; Floudas, C. A.; Kieslich, C. A.; Zhang, L.; Morikis, D.; Wareham, K. J.; Monk, P. N.; Hawksworth, O. A.; Woodruff, T. M. J. Med. Chem. 2012, 55, 4159–4168. (45) King, N. P.; Sheffler, W.; Sawaya, M. R.; Vollmar, B. S.; Sumida, J. P.; André, I.; Gonen, T.; Yeates, T. O.; Baker, D. Science 2012, 336, 1171-1174. (46) Khare, S. D.; Kipnis, Y.; Greisen, P. J.; Takeuchi, R.; Ashani, Y.; Goldsmith, M.; Song, Y.; Gallaher, J. L.; Silman, I.; Leader, H.; Sussman, J. L.; Stoddard, B. L.; Tawfik, D. S.; Baker, D. Nat. Chem. Biol. 2012, 8, 294-300. (47) Khoury, G. A.; Fazelinia, H.; Chin, J. W.; Pantazes, R. J.; Cirino, P. C.; Maranas, C. D. Protein Sci. 2009, 18, 2125-2138. (48) Craik, D. J.; Fairlie, D. P.; Liras, S.; Price, D. Chem. Biol. Drug Des. 2013, 81, 136-147. (49) Chou, P. Y.; Fasman, G. D. Biochemistry 1974, 13, 222-245. (50) Kumar, S.; Bansal, M. Proteins 1998, 31, 460.  (51) Fonseca, N. A.; Camacho, R.; Magalhaes, A. L. Proteins 2008, 70, 188. (52) Goliaei, B.; Minuchehr, Z. FEBS Lett. 2003, 537, 121. (53) Ho, S. P.; DeGrado, W. F. J. Am. Chem. Soc. 1987, 109, 6751-6758. (54) Fezoui, Y.; Weaver, D. L.; Osterhout, J. J. Protein Sci. 1995, 4, 286-295. (55) Bryson, J. W.; Desjarlais, J. M.; Handel, T. M.; DeGrado, W. F. Protein Sci. 1998, 7, 1404-1414. (56) Kortemme, T.; Ramirez-Alvarado, M.; Serrano, L. Science 1998, 281, 253. (57) Carulla, N.; Woodward, C.; Barany, G. Protein Sci. 2002, 11, 1539-1551. (58) Lutgring, R.; Chmeilewski, J. J. Am. Chem. Soc. 1994, 116, 6451-6452. 185 (59) Mutter, M.; Vuilleumier, S. Angew. Chem. Int. Ed. Engl. 1989, 28, 535-554 (60) Mutter, M. Trends Biochem. Sci. 1988, 13, 260-265. (61) Mutter, M.; Tuchscherer, G.; Macromol. Chem. Rapid Commun. 1988, 9, 437-443. (62) Mutter, M.; Altmann, E.; Altmann, K. H.; Hersperger, R.; Koziej, P .; Nebel, K.; Tuchscherer, G. G.; Vuilleumier, S. Helv. Chim. Acta 1988, 71, 835.  (63) Floegel, R.; Mutter, M. Biopolymers 1992, 32, 1283.  (64) Mutter, M. Angew. Chem., Int. Ed. Engl. 1985, 24, 639. (65) Grove, A.; Tomich, J. M.; Montal, M. Proc. Natl. Acad. Sci. U. S. A. 1991, 88, 6418-6422. (66) Haehnel, W. Molecular Diversity 2004, 8, 219-229. (67) Wong, A. K.; Jacobsen, M. P.; Winzor, D. J.; Fairlie, D. P. J. Am. Chem. Soc. 1998, 120, 3836-3841. (68) a). Khew, S. T.; Tong, Y. W. Biochemistry 2008, 47, 585-596. b). Brask, J.; Jensen, K. J. Bioorg. Med. Chem. Lett. 2001, 11, 697-700. (69) Sasaki, T.; Kaiser, E. T. J. Am. Chem. Soc. 1989, 111, 380-381. (70) Gibb, B. C.; Mezo, A. R.; Sherman, J. C. Tetrahedron Lett. 1995, 36, 7587-7590. (71) Causton, A. S.; Sherman, J. C. Bioorg. Med. Chem. 1999, 7, 23-27. (72) Mezo, A. R.; Sherman, J. C. J. Am. Chem. Soc. 1999, 121, 8983-8994. (73) Singh, Y.; Dolphin, G. T.; Razkin, J.; Dumy, P. ChemBioChem 2006, 7, 1298-1314. (74) Gibb, B. C.; Mezo, A. R.; Causton, A. S.; Fraser, J. R.; Tsai, F. C. S.; Sherman, J. C. Tetrahedron 1995, 51, 8719-8732. (75) Mezo, A. R. Study of Cavitand-Based de Novo Four-Helix Bundle Proteins, PhD Thesis, University of British Columbia, Vancouver, 1999. (76) Cram, D. J.; Karbach, S.; Kim, H. E.; Knobler, C. B.; Maverick, E. F.; Ericson, J. L.; Helgseon, R. C. J. Am. Chem. Soc. 1988, 110, 2229-2237. (77) Naumann, C.; Román, E.; Peinador, C.; Ren, T.; Patrick, B. O.; Kaifer, A. E.; Sherman, J. C. Chem. Eur. J. 2001, 7, 1637-1645. (78) Derouiche, F.; Bôle-Feysot, C.; Naïmi, D.; Coëffier, M. Biochemical and Biophysical Research Communications 2014, 452, 740-745. 186 (79) PDB ID: 1FNT, Whitby, F. G.;  Masters, E. I.;  Kramer, L.;  Knowlton, J. R.;  Yao, Y.;  Wang, C. C.;  Hill, C. P. Nature 2000, 408, 115-120. (80) Connolly, D. J.; Bowie, A. G. Biochemical Pharmacology, 2014, 92, 405-414. (81) PDB ID: 3VD8, Jin, T.;  Perry, A.;  Smith, P.;  Jiang, J.;  Xiao, T. S.  J. Biol. Chem. 2013, 288, 13225-13235. (82) Shimada-Shimizu, N.; Hisamitsu, T.; Nakamura, T. Y.; Hirayama, N.; Wakabayashi, S. Mol. Pharmacol. 2014, 85, 18-28. (83) PDB ID: 1Y4E, Slepkov, E. R.;  Rainey, J. K.;  Li, X.;  Liu, Y.;  Cheng, F. J.;  Lindhout, D. A.;  Sykes, B. D.;  Fliegel, L. J.Biol.Chem. 2005, 280, 17863-17872. (84) Preda, R. C.; Leisk, G.; Omenetto, F.; Kaplan, D. L. Methods in molecular biology 2013, 996, 19-41. (85) Guo, W.; Wisniewski, J. A.; Ji, H. Bioorg. Med. Chem. Lett. 2014, 24, 2546-2554.  (86) Moreira, I. S.; Fernandes, P. A.; Ramos, M. J. Proteins 2007, 68, 803-812. (87) Nortcliffe, A.; Fleming, I. N.; Botting, N. P.; O’Hagan, D. Tetrahedron 2014, 70, 8343-8347. (88) Kesik, M.; Akbulut, H.; Söylemez, S.; Cevher, Ş. C.; Hızalan, G.; Arslan Udum, Y.; Endo, T.; Yamada, S.; Çırpan, A.; Yağcı, Y.; Toppare, L. Polym. Chem. 2014, 5, 6295-6306. (89) Panda, J. J.; Chauhan, V. S. Polym. Chem. 2014, 5, 4431-4449. (90) Nicolas, P. FEBS J. 2009, 276, 6483-6496. (91) Meeuwissen, J; Reek, J. N. H. Nature Chemistry 2010, 2, 615-621. (92) Motherwell, W. B.; Bingham, M. J.; Six, Y. Tetrahedron 2001, 57, 4663-4686. (93) Faiella, M.; Maglio, O.; Nastri, F.; Lombardi, A.; Lista, L.; Hagen, W. R.; Pavone, V. Chem. Eur. J. 2012, 18, 15960-15971. (94) Jiang, L.; Althoff, E. A.; Clemente, F. R.; Doyle, L.; Röthlisberger, D.; Zanghellini, A.; Gallaher, J. L.; Betker, J. L.; Tanaka, F.; Barbas III, C. F.; Hilvert, D.; Honk, K. N.; Stoddard, B. L.; Baker, D. Science 2008, 319, 1387-1391. (95) Broo, K. S.; Brive, L.; Ahlberg, P.; Baltzer, L. J. Am. Chem. Soc. 1997, 119, 11362-11372. (96) Bernstein, J. D.; Okamoto, Y.; Kim, M.; Shikano, S. FEBS Open Bio. 2013, 3, 196-203. (97) Guo, H.; Cao, C.; Chi, X.; Zhao, J.; Liu, X.; Zhou, N.; Han, S.; Yan, Y.; Wang, Y.; Xu, Y. J. Neurosci. Res. 2014, 92, 1374-1383. 187 (98) Wang, R.; Cleary, R. A.; Wang, T.; Li, J.; Tang, D. D. J. Biol. Chem. 2014, 289, 14157-14169. (99) Clackson, T.; Wells, J. A. Science 1995, 267, 383-386. (100) Keskin, O.; Ma, B.; Nussinov, R. J. Mol. Biol. 2005, 345, 1281-1294. (101) Carbonell, P.; Nussinov, R.; del Sol, A. Proteomics 2009, 9, 1744-1753. (102) Liu, A. M.; Xue, M. Z.; Chen, J.; Poon, R. T.; Luk, J. M. Expert Opin. Ther. Targets 2010, 14, 855-874. (103) Zhao, B.; Lei, Q.; Guan, K.L. Genes Dev. 2010, 24, 862-874. (104) Zhang, Z.; Lin, Z.; Zhou, Z.; Shen, H. C.; Yan, S. F.; Mayweg, A. V.; Xu, Z.; Qin, N.; Wong, J. C.; Zhang, Z.; Rong, Y.; Fry, D. C.; Hu, T. Med. Chem. Lett. 2014, 5, 993-998. (105) McMahon, H. T.; Boucrot, E. Nat. Rev. Mol. Cell Biol. 2011, 12, 517-533. (106) Nevola, L.; Martín-Ouirós, A.; Eckelt, K.; Camarero, N.; Tosi, S.; Llobet, E.; Gorostiza, P. Angew. Chem. Int. Ed. 2013, 52, 7704-7708. (107) Edeling, M. A.; Mishra, S. K.; Keyel, P. A.; Steinhauser, A. L.; Collins, B. M.; Roth, R.; Heuser, J. E.; Owen, D. J.; Traub, L. M. Dev. Cell 2006, 10, 329-342. (108) Alberts, B.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.; Walter, P. In Molecular Biology of the Cell; Garland Science: New York, NY, 2002; , pp 73. (109) Rasola, A.; Gramaglia, D.; Boccaccio, C.; Comoglio, P. M. J. Immunol. 2001, 166, 81-88. (110) Huang, M. B.; Jin, L. L.; James, C. O.; Khan, M.; Powell, M. D.; Bond, V. C. J. Virol. 2004, 78, 11084-11096. (111) An, G.; Li, X.; S, H.; Liu, S.; Zhou, J.; Ding, Q.; Wei, P.; Zhang, G. Trop. J. Pharm. Res. 2014, 13, 683-688. (112) Seo, E. S.; Sherman, J. C. Peptide Science 2007, 88, 774-779. (113) Seo, E. S.; Scott, W. R. P.; Straus, S. K.; Sherman, J. C. Chem. Eur. J. 2007, 13, 3596-3605. (114) Freeman, J. O.; Sherman, J. C. Chem. Eur. J. 2011, 17, 14120-14128. (115) Lau, D. K. W.; Okon, M.; McIntosh, L. P. Protein Sci. 2012, 21, 1716-1725. (116) Yogesha, S. D.; Rangarajan, E. S.; Vonrhein, C.; Bricogne, G.; Izard, T. Protein Sci. 2012, 21, 583-588. 188 (117) Russo, S.; Schweitzer, J.; Polen, T.; Bott, M.; Pohl, E. J. Biol. Chem. 2009, 284, 5208-5216. (118) Freeman, J. O. ; Lee, W. C. ; Murphy, M. E. P. ; Sherman, J. C. J. Am. Chem. Soc. 2009, 131, 7421-7429. (119) Goult, B. T.; Gingras, A. R.; Bate, N.; Barsukov, I. L.; Critchley, D. R.; Roberts, G. C. K.  FEBS Lett. 2010, 584, 2237-2241. (120) Kohn, W. D.; Kay, C. M.; Hodges, R. S. Protein Sci. 1995, 4, 237-250. (121) Chakrabartty, A.; Kortemme, T.; Padmanabhan, S.; Baldwin, R. L. Biochemistry 1993, 32, 5560-5565. (122) Scholtz, J. M.; Qian, H.; York, E. J.; Stewart, J. M.; Baldwin, R. Biopolymers 1991, 31, 1463-1470. (123) Chen, Y. H.; Yang, J. T.; Chau, K. H. Biochemistry 1974, 13, 3350-3359. (124) Shirley, B. A. In Protein Stability and Folding: Theory and Practice; Methods in Molecular Biology; 1995; Vol. 40, pp 177. (125) Rink, H. Tetrahedron Lett. 1987, 28, 3787-3788. (126) Ganser-Pornillos, B. K.; Yeager, M.; Sundquist, W. Curr. Opin. Struct. Biol. 2008, 18, 203-217. (127) Wright, E. R.; Schooler, J. B.; Ding, H. J.; Kieffer, C.; Fillmore, C.; Sundquist, W.; Jensen, G. J. EMBO J. 2007, 1-9. (128) Thomine, S.; Barbier-Brygoo, H. Nature 2010, 467, 1058-1059. (129) Tofteng, A. P.; Hansen, T. H.; Brask, J.; Nielsen, J.; Thulstrup, P. W.; Jensen, K. J. Org. Biomol. Chem. 2007, 5, 2225-2233. (130) Bender, M. L.; Turnquest, B. W. J. Am. Chem. Soc. 1957, 79, 1652-1655. (131) Green, D. R.; Reed, J. C. Science 1998, 281, 1309-1312. (132) Thompson, C. B. Science 1995, 267, 1456. (133) Reed, J. C. J. Cell Biol. 1994, 124, 1. (134) Adams, J. M.; Cory, S. Science 1998, 281, 1322-1325.  (135) Minn, A. J.; Swain, R. E.; Ma, A.; Thompson, C. B. Adv. Immunol. 1998, 70, 245-279. (136) Sato, T.; Hanada, M.; Bodrug, S.;Irie, S.; Iwama, N.; Boise, L. H.; Thompson, C. B.; Golemis, E.; Fong, L.; Wang, H. G.; Reed, J. C. Proc. Natl. Acad. Sci. 1994, 91, 9238-9242.  189 (137) Zhang, H.; Saeed, B.; Ng, S. C. Biochem. Biophys. Res. Commun. 1995, 208, 950-956. (138) Oltvai, Z. N.; Korsmeyer, S. J. Cell 1994, 79, 189. (139) Fesik, S. W. Cell, 2000, 103, 273-282. (140) Sattler, M.; Liang, H.; Nettesheim, D.; Meadows, R. P.; Harlan, J. E.; Eberstadt, M.; Yoon, H. S.; Shuker, S. B.; Chang, B. S.; Minn, A. J.; Thompson, C. B.; Fesik, S. W. Science 1997, 275, 983-986. (141) Chittenden, T.; Flemington, C.; Houghton, A. B.; Ebb, R. G.; Gallo, G. J.; Elangovan, B.; Chinnadurai, G.; Lutz, R. J. EMBO J. 1995, 14, 5589- 5596. (142) Muchmore, S. W.; Sattler, M.; Liang, H.; Meadows, R. P.; Harlan, J. E.; Yoon, H. S.; Nettesheim, D.; Chang, B. S.; Thompson, C. B.; Wong, S. L.; Ng, S. C.; Fesik, S. W. Nature 1996, 381, 335-341. (143) Wang, D.; Liao, W.; Arora, P. S. Angew. Chem. Int. Ed. 2005, 44, 6525-6529. (144) Rodriguez, J. M.; Ross, N. T.; Katt, W. P.; Dhar, D.; Lee, G.; Hamilton, A. D.; ChemMedChem 2009, 4, 649-656.  (145) Yin, H.; Lee, G.; Sedey, K. A.; Kutzki, O.; Park, H. S.; Orner, B. P.; Ernst, J. T.; Wang, H.; Sebti, S. M.; Hamilton, A. D. J. Am. Chem. Soc. 2005, 127, 10191-10196.   (146) Lee, E. F.; Sadowsky, J. D.; Smith, B. J.; Czabotar, P. E.; Peterson- Kaufman, K. J.; Colman, P. M.; Gellman, S. H.; Fairlie, W. D. Angew. Chem. 2009, 121, 4382-4386.  (147) Guarracino, D. A.; Bullock, B. N.; Arora, P. S. Biopolymers 2011, 95, 1-7.  (148) Walensky, L. D.; Kung, A. L.; Escher, I.; Malia, T. J.; Barbuto, S.; Wright, R. D.; Wagner, G.; Verdine, G. L.; Korsmeyer, S. J. Science 2004, 305, 1466-1470.  (149) Baek, S.; Kutchukian, P. S.; Verdine, G. L.; Huber, R.; Holak, T. A.; Lee, K. W.; Popowicz, G. M. J. Am. Chem. Soc. 2011, 134, 103-106.  (150) Wang, J. L.; Liu, D.; Zhang, Z. J.; Shan, S.; Han, X.; Srinivasula, S. M.; Croce, C. M.; Alnemri, E. S.; Huang, Z. Proc. Natl. Acad. Sci. 2000, 97, 7124-7129.  (151) Ko, E.; Liu, J.; Perez, L. M.; Lu, G.; Schaefer, A.; Burgess, K. J. Am. Chem. Soc. 2011, 133, 462-477.  (152) Kazi, A.; Sun, J.; Doi, K.; Sung, S. S.; Takahashi, Y.; Yin, H.; Rodriguez, J. M.; Becerril, J.; Berndt, N.; Hamilton, A. D.; Wang, H. G.; Sebti, S. M. J. Biol. Chem. 2011, 286, 9382-9392.  (153) Harker, E. A.; Daniels, D. S.; Guarracino, D. A.; Schepartz, A. Bioorg. Med. Chem. 2009, 17, 2038-2046.  190 (154) Sadowski, J. D.; Fairlie, W. D.; Hadley, E. B.; Lee, H. S.; Umezawa, N.; Nikolovska-Coleska, Z.; Wang, S.; Huang, D. C. S.; Tomita, Y.; Gellman, S. H. J. Am. Chem. Soc. 2007, 129, 139-154.  (155) Chin, J. W.; Schepartz, A. Angew. Chem. Int. Ed. 2001, 40, 3806-3809. (156) Mutter, M.; Vuilleumier, S. Angew. Chem. 1989, 101, 551-571. W. F. DeGrado, Z. R. Wasserman, J. D. Lear, Science 1989, 243, 622- 628. (157) Freeman, J. O.; Wallhorn, D.; Sherman, J. C. Peptide Sci. 2007, 88, 725-732. (158) Seely, J. H.; Benoiton, L. Can. J. Biochem. 1968, 46, 387-389. (159) Witkowska, E.; Orlowska, A.; Sagan, B.; Smoluch, M.; Izdebski, J. J. Peptide Sci. 2001, 7, 168-172. (160) Zhang, H.; Nimmer, P.; Rosenberg, S. H.; Ng, S.; Joseph, M. Analytical Biochemistry 2002, 307, 70-75. (161) Hol, W. G. J.; van Duijnen, P. T.; Berendsen, H. J. C. Nature 1978, 273, 443-446. (162) Huttunen-Hennelly, H. E. K.; Sherman, J. C.  Organic & Biomolecular Chemistry 2007, 5, 3637-3650. (163) Rawlings, N. D.; Barrett, A. J. Meth. Enzymol. 1994, 244, 19–61.  (164) Roehrl, H. A. M.; Wang, J. Y.; Wagner, G. Biochemistry 2004, 43, 16056-16066. (165) Roehrl, H. A. M.; Wang, J. Y.; Wagner, G. Biochemistry 2004, 43, 16067-16075. (166) Jabusch, J. R.; Deutsch, H. F. Archives of Biochemistry and Biophysics 1985, 238, 170-177. (167) Koltun, W. L.; Ng, L.; Gurd, F. R. N. J. Biol. Chem. 1963, 238, 1367-1372. (168) Kristol, D. S.; Krautheim, P.; Stanley, S.; Parker, R. C. Bioorganic Chemistry 1975, 4, 299-304. (169) Jencks, W. P.; Carriuolo, J. Biol. Chem. 1959, 234, 1272-1279.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0166361/manifest

Comment

Related Items