cyclization and derivatization of the potassium

CYCLIZATION AND DERIVATIZATION OF THE POTASSIUM CHANNEL ANTAGONIST TERTIAPIN

A THESIS SUBMITTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF HAWAI‘I AT MĀNOA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN MOLECULAR BIOSCIENCES AND BIOENGINEERING AUGUST 2016

by Vinay Keali‘i Menon

Thesis Committee: Jon-Paul Bingham, Chairperson Dulal Borthakur Qing Li

ACKNOWLEDGEMENTS I would like to acknowledge my PI, Dr. Jon-Paul Bingham, for not only giving me the opportunity to pursue a Master’s in his laboratory, but also for guiding and supporting my research every step of the way. My committee members, Dr. Dulal Borthakur and Dr. Qing Li for taking time out of their hectic schedules to evaluate my proposal and defense presentations and provide me with constructive criticism on how to improve, and for lending their knowledge in the revising of this thesis. And finally, my labmates, Parashar Thapa, Mike Espiritu, Ray Zhang, Chino Cabalteja, Zan Halford, Peter Yu, Nick Sinclair, and Chris Sugai, for always helping me work through obstacles in my project, and passing on their knowledge to me when I first joined the laboratory.

i

LIST OF PUBLICATIONS Thapa, P., Zhang, R., Menon, V., Bingham, JP. 2014. Native Chemical Ligation: A Boon to Peptide Chemistry. Molecules, 19, 14461-14483.

LIST OF PRESENTATIONS Vinay K. Menon and Jon-Paul Bingham (2015) “Cyclization and Derivatization of the Potassium Channel Antagonist Tertiapin (TPNQ)” 27th Annual CTAHR and COE Research Symposium, University of Hawai’i, Honolulu, April 10-11 IACUC protocol # 08-547-8, Evaluation of Novel Native and Synthetic Conopeptides on Fish

ii

ABSTRACT The potent potassium channel antagonist tertiapin-Q was used in order to expand the application of data previously published by Clark et al. and also to further examine the usefulness of a newly published peptide backbone cyclization technique. This peptide was also used in order to investigate the possibility of creating a cyclic, fluorescent peptide probe for the ROMK1 channel, which builds upon the thesis work of previous students in the Bingham laboratory. The data published by Clark et al. describes a correlation between N-to-C termini distance and required linker length for a cyclic peptide. To date, this relationship has not been used to cyclize a peptide toxin that did not originate from Conus spp., therefore using it to cyclize the peptide toxin tertiapin-Q represents a unique challenge to these data. After using solid phase peptide synthesis techniques to synthesize 3 length variants of tertiapin-Q, only the variant with the predicted optimal linker length was able to cyclize successfully, thereby indicating that the previously published data does apply to non-Conus toxins and can therefore be reliably used to cyclize toxins from other organisms, which could lead to new and exciting drug leads. The newly published Fmoc-based cyclization technique represents a safer and more accessible alternative to cyclizing peptides than traditional native chemical ligation since it does not require the use of highly corrosive HF or TFMSA, the former of which necessitates the use of specialty Teflon™ coated lab apparatus. Using this technique to successful cyclize tertiapin-Q during this project represented the first non-Conus toxin to be cyclized in this manner, thereby expanding the application of this relatively untested protocol. With this knowledge, future researchers can utilize this technique, in conjunction with the data published by Clark et al., in order to cyclize any given peptide sequence. Lastly, building upon the work of previous students, conjugating a fluorophore to a cyclic derivative of tertiapin-Q would allow it to be used a fluorescent probe which could be used to trace the distribution and investigate the mechanisms of the molecular targets of this peptide, namely ROMK1 and GIRK1/4. Using the Staudinger reaction to conjugate a DyLight® 650 fluorophore to the peptide resulted in poor yields (approximately 3%) of the desired product. This could indicate that the reactive azide moiety is sterically hindered by the surrounding charged/bulky amino acid side-chains, which suggests that the inclusion of an extra carbon chain would negate this effect, allowing the reaction to proceed to completion. If this is coupled with iii

an increase in the ratio of peptide to fluorophore, then the reaction would be almost guaranteed to proceed with little to no trouble.

iv

TABLE OF CONTENTS ACKNOWLEDGEMENTS ............................................................................................................. i LIST OF PUBLICATIONS ............................................................................................................ ii LIST OF PRESENTATIONS ......................................................................................................... ii ABSTRACT ................................................................................................................................... iii LIST OF TABLES ....................................................................................................................... viii LIST OF FIGURES ....................................................................................................................... ix LIST OF ABBREVIATIONS ....................................................................................................... xv CHAPTER 1 .................................................................................................................................. 1 INTRODUCTION .......................................................................................................................... 1 1.1 European Honey Bee (Apis mellifera) .................................................................................. 1 1.1.1 Geographic Distribution and Diet................................................................................... 1 1.1.2 Venom Profile of A. mellifera ........................................................................................ 1 1.1.3 Mechanism and Activity of Apitoxin Components ........................................................ 2 1.2 Tertiapin ................................................................................................................................ 4 1.2.1 Native Structure and Activity ......................................................................................... 4 1.2.2 Stable Synthetic Derivative: Tertiapin-Q ....................................................................... 4 1.3 ROMK1 Channel................................................................................................................... 5 1.3.1 Introduction to Inwardly Rectifying Potassium Channel (IRK) Family ........................ 5 1.3.2 ROMK1 Localization and Importance ........................................................................... 5 1.3.3 ROMK1 Molecular Structure ......................................................................................... 6 1.3.4 ROMK1 Activation and Probes ...................................................................................... 7 1.4 Cyclic Peptides and Cyclization ............................................................................................ 8 1.4.1 Benefits of Cyclization ................................................................................................... 8 1.4.2 Origins of Chemical Cyclization .................................................................................. 11 1.4.3 Cyclic Peptide Linker Sequence ................................................................................... 14 1.5 Fluorophores and Fluorescent Peptides............................................................................... 16 1.5.1 Benefits of Small Organic Fluorophores ...................................................................... 16 1.5.2 Staudinger Ligation via Lys-Azide............................................................................... 17 1.5.3 Thermo Fisher DyLight® Fluorophore ......................................................................... 18 1.6 Project Objectives and Rationale ........................................................................................ 19 PROJECT HYPOTHESES ........................................................................................................... 20 PROJECT GOALS AND OBJECTIVES ..................................................................................... 21 v

CHAPTER 2 ................................................................................................................................ 22 EXPERIMENTAL PROCEDURES ............................................................................................. 22 2.2 Solid Phase Peptide Synthesis of TPN-Q............................................................................ 22 2.2.1 Swelling of Pre-Loaded Chlorotrityl Resin .................................................................. 22 2.2.2 Activation of the Fmoc Amino Acids........................................................................... 23 2.2.3 Amino Acid Coupling .................................................................................................. 23 2.2.4 Removal of the 9-fluorenylmethoxycarbonyl (Fmoc) N-Terminal Protecting Group . 23 2.2.5 Drying Down the Peptide Resin ................................................................................... 24 2.3 Quantification of Amino Acid Coupling............................................................................. 24 2.3.1 Ninhydrin Assay ........................................................................................................... 24 2.3.2 Percent Coupling Calculation ....................................................................................... 24 2.4 Cleavage from Resin to Yield Linear Peptide ..................................................................... 25 2.5 Cleavage from Resin to Yield Cyclic Peptide ..................................................................... 26 2.5.1 Cleavage from the Chlorotrityl Resin ........................................................................... 26 2.5.2 Cyclization Reaction..................................................................................................... 26 2.5.3 Removal of Orthogonal Protecting Groups .................................................................. 26 2.6 Oxidation of Thiols to Form Disulfide Bonds .................................................................... 29 2.7 Purification of Peptide Samples Using RP-HPLC .............................................................. 30 2.7.1 RP-HPLC Columns ...................................................................................................... 30 2.7.2 Sample Preparation ....................................................................................................... 30 2.7.3 Sample Loading ............................................................................................................ 30 2.7.4 RP-HPLC Gradients ..................................................................................................... 30 2.7.5 Sample Collection......................................................................................................... 31 2.8 Electrospray Ionization Mass Spectrometry (ESI-MS) Methods ........................................ 31 2.8.1 Sample Preparation ....................................................................................................... 31 2.8.2 Sample Injection ........................................................................................................... 31 2.9 Whole Animal Bioassay Protocols ...................................................................................... 32 2.9.1 Injection into Green Swordtail Fish (Xiphophorus helleri) .......................................... 32 2.9.2 Extrapolation of the LD50 Value ................................................................................... 32 CHAPTER 3 ................................................................................................................................ 34 RESULTS AND DISCUSSION ................................................................................................... 34 3.1 In silico Distance Measurements of TPN-Q ....................................................................... 34 vi

3.2 Failed Cyclization of cTPN-Q5 and cTPN-Q7 ................................................................... 35 3.3 Successful Cyclization of cTPN-Q6 ................................................................................... 41 3.4 Oxidation and Purification of Cyclic and Linear TPN-Q6 ................................................. 43 3.5 Whole Animal Bioactivity Assays of Linear and Cyclic TPN-Q6 ..................................... 49 3.6 Oxidation and Purification of cTPN-Q6 with a Lysine-Azide Residue .............................. 52 3.7 Purification of Oxidized cTPN-Q6f Dy650 ........................................................................ 55 3.8 Discussion of Overall Results ............................................................................................. 58 CHAPTER 4 ................................................................................................................................ 61 FUTURE DIRECTIONS .............................................................................................................. 61 4.1 Future Work ........................................................................................................................ 61 REFERENCES ............................................................................................................................. 63

vii

LIST OF TABLES Table 1. The sequence and activity of several scorpion toxins, similar to Lq2, that failed to show any significant activity for ROMK1. Cysteine residues are shown in bold face font……8 Table 2. Lists the activity and source (either synthetic or natural) of various cyclic peptides and proteins. The amino acids sequences are shown in Table 3…………….………………….10 Table 3. Shows the sequences of the cyclic proteins and peptides seen in Table 2. Cysteine residues are shown in bold, and backbone cyclization is represented by a line connecting the N and C termini…………………………………………………………………………………16 Table 4. Summarizes the results of in silico distance measurements using the QUARK structure prediction server and three different 3D molecular modeling software in order to determine the termini distance for TPN-Q...................................................................................34 Table 5. Shows the data gathered from injections of cTPN-Q6 Isomer 1 into X. helleri fish. Each dosage was performed in triplicate and then averaged to produce these data. The columns that were used to construct the plot in Figure 19 are indicated by the thick border…..50 Table 6. Shows the data gathered from injections of cTPN-Q6 Isomer 2 into X. helleri fish. Each dosage was performed in triplicate and then averaged to produce these data. The all three dosages were monitored for a total of 18 hours before being terminated…………………51 Table 7. Shows the data gathered from injections of linear TPN-Q6 into X. helleri fish. Each dosage was performed in triplicate and then averaged to produce these data. The experiment was terminated due to ethical considerations after no activity was observed at the highest dosage…………………………………………………………………………………………...51

viii

LIST OF FIGURES Figure 1. Western Honey Bee (A. mellifera)……………………………………………………1 Figure 2. HPLC chromatogram of the main constituents of apitoxin. The peptide apamin is found in the area indicated by the box. Tertiapin is not detectable since it only comprises ~ 0.1% of the dry weight of apitoxin……………………………………………………………2 Figure 3. Solution NMR structure of native TPN. Showing the C-terminal alpha helix and Nterminal beta strand, structural elements that are essential to the maintenance of TNP’s biological activity (Krissinel and Henrick 2007)………………………………………………..4 Figure 4. Shows the side-view of the general molecular structure of IRK channels, the family of channels that includes ROMK1 (Bichet et al. 2003)…………………………………………6 Figure 5. Side-by-side comparison of a) Fmoc-based cyclization and b) NCL cyclization. Solid shapes represent fully protected peptides, while hollow shapes represent deprotected peptides………………………………………………………………………………………….13 Figure 6. Shows the strong positive correlation that exists between termini distance and the number of linker residues necessary for successful cyclization. Each data point represents a peptide that has been successfully cyclized (Clark et al. 2010)…………..……………………..15 Figure 7. Shows the 3 step reaction mechanism for attaching a phosphine-containing fluorophore to the azide moiety of a peptide through the Staudinger reaction (Saxon and Bertozzi 2000)……………………………………………………………………....18 Figure 8. Shows the wide range of wavelengths that DyLight® molecules maximally absorb at………………………………………………………………………………………………....19 Figure 9. Shows the repetitive step-wise process of solid phase peptide synthesis. This process consists of deprotection, during which the N-terminal Fmoc group (A) is removed by piperidine and coupling, which adds another N-terminally protected amino acid to the growing peptide chain. These steps are repeated until the sequence is complete at which

ix

point the side-chain protecting groups (B) and resin are removed. The anchoring step was not done in this project since the resin that was used was pre-loaded with the first amino acid in the sequence. Adapted from Grant 2002.………………………………………………………..28 Figure 10. Shows the 3 reactions necessary for Fmoc-based cyclization. Step 1 cleaves the peptide from the chlorotrityl resin, step 2 selectively joins the N- and C- termini in a peptide bond through a dehydration reaction, and step 3 removes the side-chain protectional groups. Solid shapes represent fully protected peptides, while hollow shapes represent deprotected peptides………………………………………………………………………………………….29 Figure 11. Flow diagram showing the various versions of TPN-Q that were used for this project. cysteine residues (C), all of which are oxidized to form disulfide bonds, and the additional linker sequences are shown in bold face font………………………………………...35 Figure 12. Analytical RP-HPLC/UV chromatogram of reduced cTPN-Q5, after cyclization and side-chain deprotection, extracted at 214nm. The elution time of the peak indicated by the arrow is 38.4 minutes………………………………………………………………………36 Figure 13. ESI-MS spectra of reduced cTPN-Q5 after cyclization and side-chain deprotection. The dominant signal at 558.5 Da corresponds to [M+5H]5+ of linear TPN-Q5, while the minor signals at 554.9 Da and 693.2 Da correspond to the [M+5H]5+ and [M+4H]4+ charge states, respectively, of cyclic TPN-Q5…………………………………………………..37 Figure 14. Shows the ESI-MS of crude oxidized cTPN-Q6. The signals at 565.7 Da, 706.7 Da, and 471.5 Da correspond to the [M+5H]5+, [M+4H]4+, and [M+6H]6+ charge state, respectively, of oxidized cTPN-Q6. Since this was before purification, the signals at 569.0 Da and 711.2 Da most likely correspond to the [M+5H]5+ and [M+4H]4+ charge state, respectively of oxidized linear TPN-Q6…………………………………………………………38 Figure 15. ESI-MS spectra of reduced cTPN-Q7 after cyclization and side-chain deprotection. The dominant signal at 584.1 Da corresponds to [M+5H]5+ of linear TPN-Q7, while the minor signals at 580.8 Da and 725.3 Da correspond to the [M+5H]5+ and [M+4H]4+ charge states, respectively, of cyclic TPN-Q7………………………………………………………………….39 x

Figure 16. Analytical RP-HPLC/UV chromatogram of reduced cTPN-Q7s, after cyclization and side-chain deprotection, extracted at 214nm. The elution time of the peak indicated by the black arrow is 38.5 minutes, while the peak indicated by the grey arrow elutes at 39.7 minutes. The peaks are much less resolved than in Figure 13, indicating that sonication detrimentally affects the cyclization reaction……………………………………………………40 Figure 17. ESI-MS spectra of reduced cTPN-Q7s after cyclization and side-chain deprotection. None of the three dominant signals correspond any of the theoretical m/z of linear or cyclic TPN-Q7………………………………………………………………………40 Figure 18. Shows the analytical RP-HPLC/UV chromatogram of reduced cTPN-Q6 extracted at 214nm. The retention time of the main peak is 39.4 minutes. The small leading shoulder at 38.1 minutes most likely corresponds to the remnants of linear TPN-Q6 that was unable to cyclize. This is corroborated by the appearance of linear masses in Figure 19………………………………………………………………………………………………...42

Figure 19. ESI-MS spectra of reduced cTPN-Q6 after cyclization and side-chain deprotection. The dominant signal at 566.5 Da corresponds to the [M+5H]5+ charge state. The minor signals at 570.1 Da and 712.0 Da correspond to the [M+5H]5+ and [M+4H]4+ charge states, respectively, of linear TPN-Q6………………………….............................................................43 Figure 20. Shows the analytical RP-HPLC/UV chromatogram of reduced linear TPN-Q6 extracted at 280nm. The retention time of the main peak is 35.2 minutes………………………44 Figure 21. Shows the analytical RP-HPLC/UV chromatogram of oxidized linear TPN-Q6 extracted at 214nm. The retention time of the main peak is 34.6 minutes………………………44 Figure 22. A) Shows the ESI-MS spectra of reduced linear TPN-Q6. The dominant signal at 570.1 Da corresponds to [M+5H]5+ charge state, while the 712.0 Da signal corresponds [M+4H]4+ charge state B) Shows the ESI-MS spectra of oxidized linear TPN-Q6. One of the dominant signal at 569.1 Da corresponds to [M+5H]5+ charge state, while the 711.2 Da signal corresponds [M+4H]4+ charge state……………………………………………………………...45

xi

Figure 23. Shows the analytical RP-HPLC/UV chromatogram of crude oxidized cTPN-Q6 extracted at 214nm. The two black arrows correspond to isomer 1 and 2, while the grey arrow corresponds to the deletion product. The elution time of Isomer 1 is 37.8 minutes and comprises 9.8% of the total area, the elution time of Isomer 2 is 38.4 minutes comprising 14.60%, and the deletion product that elutes at 39.6 minutes makes up 12.78%.........................46 Figure 24. Shows the ESI-MS of crude oxidized cTPN-Q6. The signals at 565.7 Da, 706.7 Da, and 471.5 Da correspond to the [M+5H]5+, [M+4H]4+, and [M+6H]6+ charge state, respectively, of oxidized cTPN-Q6. Since this was before purification, the signals at 569.0 Da and 711.2 Da most likely correspond to the [M+5H]5+ and [M+4H]4+ charge state, respectively of oxidized linear TPN-Q6…………………………………………………………46 Figure 25. Shows the analytical RP-HPLC/UV chromatogram of purified cTPN-Q6 Isomer 1 extracted at 214nm. The retention time is 37.6 minutes…………………………………………47 Figure 26. Shows the analytical RP-HPLC/UV chromatogram of purified cTPN-Q6 Isomer 2 extracted at 214nm. The retention time is 38.3 minutes…………………………………………47 Figure 27. A) Shows the ESI- MS spectra of cTPN-Q6 Isomer 1. The dominant signal at 565.6 Da corresponds to the [M+5H]5+ charge state, while the signal at 706.7 Da corresponds to the [M+4H]4+ charge state. B) Shows the ESI- MS spectra of cTPN-Q6 Isomer 2. The dominant signal at 565.7 Da corresponds to the [M+5H]5+ charge state, while the signal at 706.8 Da corresponds to the [M+4H]4+ charge state…………………………………………….48

Figure 28. Shows the analytical RP-HPLC/UV chromatogram of purified cTPN-Q6 deletion product extracted at 214nm. The retention time is 40.0 minutes………………………………..48 Figure 29. ESI-MS spectra of the cTPN-Q6 deletion product. The dominant signal at 453.8 Da corresponds to a deletion of approximately 18 Da on the [M+6H]6+ charge state of cTPN-Q6…………………………………………………………………………………………49

xii

Figure 30. Shows the plot of the average dosage vs. the average dosage/survival time for the whole animal bioassay of cTPN-Q6 Isomer 1. Each data point represents an average of each dosage done in triplicates. The R-squared value indicates a strong linear relationship in the data, and the y-intercept of best fit line is equivalent to the LD50 value for this isomer………...52 Figure 31. Shows the structure of Fmoc-Lys-Azide that was used during the synthesis of cTPN-Q6f. Importantly, this synthetic amino acid has an azide group (R-N3) attached to the ε-carbon instead of an amino group (R-NH3) as is found in a standard lysine residue………….53 Figure 32. Shows the analytical RP-HPLC/UV chromatogram of reduced cTPN-Q6f extracted at 214nm. This molecule has been cyclized and contains a Lys-Azide within the linker sequence. The retention time is 37.1 minutes…………………………………………….53 Figure 33. Shows the analytical RP-HPLC of oxidized cTPN-Q6f. The retention time of the main peak is 35.1 minutes……………………………………………………………………….54 Figure 34. A) Shows the ESI-MS spectra of reduced cTPN-Q6f. The signals at 484.8 Da and 581.3 Da correspond to the [M+6H]6+ and [M+5H]5+ charge states, respectively. B) Shows the ESI-MS spectra of oxidized cTPN-Q6f. The signals at 580.8 Da, 725.3 Da and 484.0 Da correspond to the [M+5H]5+, [M+4H] 4+, and [M+6H]6+ charge states, respectively……………54 Figure 35. A) Shows the analytical HPLC/UV chromatogram of cTPN-Q6f Dy650 detected by a PDA detector at 214nm. The main peak elutes at 35.1 minutes and the second peak at 39.0 minutes B) Shows the analytical HPLC/Fluorescence chromatogram of cTPN-Q6f Dy650 detected by Scanning Fluoroscence (Ex/Em= 646/674 nm) in tandem with the PDA detector. The elution time is approximately 40.8 minutes, and most likely corresponds to the PDA peak at 39.0 minutes. The lag in retention time is due to the travel distance between the two detectors………………………………………………………………………………………….56 Figure 36. Shows the ESI-MS spectra of cTPN-Q6f Dy650. The signals at 580.9 Da and 726.0 Da correspond to the [M+5H]5+ and [M+4H]4+ charge states, respectively, of unlabeled cTPN-Q6f………………………………………………………………………………………..57

xiii

Figure 37. Shows a flow chart summarizing the peptides produced during this project with their associated expected and observed parent masses and the mass difference between the two.………....................................................................................................................................60

xiv

LIST OF ABBREVIATIONS αα

Amino Acid

AFP

Autofluorescent Protein

AgTx

Agitoxin

BK

Large Conductance Calcium-Activated Potassium Channels

Boc

tert-butyloxycarbonyl

CCD

Cortical Collecting Duct

CD

Circular Dichroism

ChTx

Charybdotoxin

Cl-

Chloride Ion

cTPN-Q5

TPN-Q that has a 5 amino acid linker sequence and has gone through the cyclization reaction

cTPN-Q6

TPN-Q that has a 6 amino acid linker sequence and has gone through the cyclization reaction

cTPN-Q6f

TPN-Q that has a 6 amino acid linker sequence containing a LysAzide residue and has gone through the cyclization reaction

cTPN-Q7

TPN-Q that has a 7 amino acid linker sequence and has gone through the cyclization reaction

cTPN-Q7s

cTPN-Q7 that was sonicated during the cyclization reaction

Da

Daltons

DCM

Dichloromethane

DI

Deionized

DIPEA

N,N-Diisopropylethylamine

DMF

Dimethylformamide

DTT

Dithiothreitol

ED50

Half Maximal Effective Dose

EDT

1,2-ethanedithiol

EK

Equilibrium Potential of Potassium

ESI-MS

Electrospray Ionization Mass Spectrometry

Fmoc

9-Fluorenylmethoxycarbonyl

GIRK

G-coupled protein Inward Rectifying Potassium Channel xv

HATU

N, N, N, N-tetramethyl-O-(7-azabenzotriazol-1-yl)uronium hexafluorophosphate

HCTU

2-(6-chloro-1H-benzotriazole-1-yl)-1,1,3,3-tetramethylaminium hexafluorophosphate

HF

Hydrogen Fluoride

IbTx

Iberiotoxin

IC50

Half Maximal Inhibitory Concentration

IMM

Inner Mitochondrial Membrane

IR

Infrared

IRK

Inward Rectifying Potassium Channel

K+

Potassium Ion

KCN

Potassium Cyanide

KCNJ1

Gene that encodes the ROMK1 protein

Kir1.1a-f

Splice Variants of Potassium Inward-Rectifying Channel, Subfamily J, Member 1 (ROMK1)

LD50

Half Maximal Lethal Dose

LN2

Liquid Nitrogen

m/z

Mass-to-Charge ratio

MCoTI

Momordica cochinchinensis trypsin inhibitor

MCD

Malonyl-CoA Decarboxylase

MeCN

Acetonitrile

mEq

Milliequivalents

Mg2+

Magnesium Ion

MgTx

Margatoxin

mmol

Millimole

MTBE

Methyl tert-Butyl Ether

Na+

Sodium Ion

nAChR

Nicotinic Acetylcholine Receptors

NCL

Native Chemical Ligation

NIR

Near Infrared

NMR

Nuclear Magnetic Resonance xvi

NTx

Noxiustoxin

NV

Ninhydrin Value

PBS

Phosphate-Buffered Saline

PDA

Photodiode Array

Ph

Phenyl

PIP2

Phosphatidylinositol 4,5-bisphosphate

PLA2

Phospholipase A2

POP

Prolyl Oligopeptidase

ROMK

Renal Outer Medullary Inward Rectifying Potassium Channel

RP-HPLC

Reverse Phase High Performance Liquid Chromatography

RTD-1

Rhesus θ-defensin

SFTI

Sunflower Trypsin Inhibitor

SK

Small Conductance Calcium-Activated Potassium Channels

Sol. A

Solvent A (0.1% v/v trifluoroacetic acid aqueous)

Sol. A’

Solvent A’ (0.1% v/v formic acid aqueous)

Sol. B

Solvent B (90% v/v acetonitrile in 0.8% v/v trifluoroacetic acid aqueous)

Sol. B’

Solvent B’ (90% v/v acetonitrile in 0.8% v/v formic acid aqueous)

SPPS

Solid Phase Peptide Synthesis

SV

Substitution Value

TAL

Thick Ascending Limb

TCEP

Tris(2-carboxyethyl)phosphine

TFA

Trifluoroacetic Acid

TIPS

Triisopropylsilane

TPN

Native Tertiapin

TPN-Q

Synthetic stable derivative of native tertiapin: residue M13 is replaced with a Glutamine (Q)

TPN-Q6

TPN-Q with a 6 amino acid linker sequence

TPN-Q7

TPN-Q with a 7 amino acid linker sequence

xvii

CHAPTER 1 INTRODUCTION 1.1 European Honey Bee (Apis mellifera) 1.1.1 Geographic Distribution and Diet The Western Honey Bee (Apis mellifera) is native to Europe, Asia and Africa but was introduced into North America in the 1600s (Figure 1). Currently there are 28 recognized subspecies. Their diet consists of nectar, a sugar-rich fluid produced by flowers in order to attract pollinators, and pollen which is a yellow-green powdery substance representing the male gamete for flowering plants. Pollen often mixed with nectar to form a protein-rich substance called “beebread” that is fed to bee larvae. (Information sheet 4: What bees eat)

Figure 1. Western Honey Bee (A. mellifera) 1.1.2 Venom Profile of A. mellifera All subspecies of A. mellifera produce a complex venom called apitoxin. Apitoxin is produced in glands located in the abdomen of the animal; the active components make up 12% while the other 88% is water . Of this 12% the main peptidic component is melittin, other components include apamin, phospholipase A2, histamine, noradrenaline, dopamine, tertiapin, adolapin, hyaluronidase, and protease inhibitors (White and Meier 1995) (Figure 2).

1

0.00

10.00

Rt (min)

Figure 2. HPLC chromatogram of the main constituents of apitoxin. The peptide apamin is found in the area indicated by the box. Tertiapin is not detectable since it only comprises ~ 0.1% of the dry weight of apitoxin (Rader et al. 1987) 1.1.3 Mechanism and Activity of Apitoxin Components Approximately 52% of the peptides in apitoxin is melittin, a 26mer with no disulfide bonds. Melittin has multiple effects attributed to its interaction with phospholipids. This, in turn, is made possible by its hydrophobic N-terminus and hydrophilic C-terminus allowing melittin to exhibit amphipathic properties. Melittin also interferes with the activity of several ion pumps such as Na+/K+-ATPase and H+/K+-ATPase. Furthermore, it generally increases cell permeability to Na+ ions. Melittin has also been found to increase the permeability of corneal tissue to Cl-, in addition to Na+ (Tossi et al. 2000; Yang and Carrasquer 1997).

2

The ubiquitous enzyme phospholipase A2 (PLA2) comprises 10-12% of apitoxin, and is considered to be one of its most detrimental components. Upon recognizing the Sn2 acyl position of phospholipids, PLA2 catalyzes hydrolysis of the bond resulting in a lysophospholipid and arachidonic acid. Arachidonic acid is then modified by cyclooxygenase enzymes in order to produce eicosanoids, like prostaglandins, which are potent regulators of the inflammatory response. This results in inflammation and pain at the site of injection, due to the disproportionate release of arachidonic acid from the phospholipid membrane (Argiolas and Pisano 1983; Dennis 1994). The effects of PLA2 is exacerbated by the previously mentioned peptide, melittin, which is a known stimulator of PLA2 activity (Clark et al. 1987; Habermann 1972). The peptide adolapin, comprises between 2 and 5 percent of the peptidic component of apitoxin. In stark contrast to PLA2, adolapin seems to function in an anti-inflammatory and analgesic capacity by inhibiting prostaglandin synthesis through inhibition of cyclooxygenase activity (Koburova et al. 1985; Shkenderov and Koburova 1982; Son et al. 2007). The 18mer peptide neurotoxin apamin composes 2-3% of apitoxin (Habermann 1984; Son et al. 2007). Apamin acts as a selective antagonist against SK (small conductance calciumactivated potassium) channels, which are expressed throughout the central nervous system. Once these channels are triggered by an increase in intracellular calcium concentration, the resulting efflux on potassium ions into the extracellular space leads to a similar hyperpolarization as is seen during action potentials (Bond et al. 1999; Faber and Sah 2007). In an attempt to understand the suspected role that SK channels play in the synaptic plasticity that is crucial for learning and memory, apamin has been used as a molecular probe to further explore their electrical properties (Castle et al. 1989; Stackman et al. 2002). Other components of apitoxin, including histamine, noradrenaline, dopamine, hyaluronidase, and protease inhibitors, are involved in the allergic response (Monroe et al. 1997), increasing heart rate (Rang HP 2015; Seeman 2010), and helping to spread inflammation (Chen and Abatangelo 1999), respectively.

3

1.2 Tertiapin 1.2.1 Native Structure and Activity The final component of apitoxin is tertiapin (TPN). This 21 amino acid peptide toxin has the sequence NH2-Ala-Leu-Cys-Asn-Cys-Asn-Arg-Ile-Ile-Ile-Pro-His-Met-Cys-Trp-Lys-LysCys-Gly-Lys-Lys-CONH2. Structural analysis via NMR, circular dichroism (CD), and distance geometry, has revealed that TPN forms an alpha helix between H12-G19 and a beta strand between the residues R7-I9; the rest of the peptide is undifferentiated (Xu and Nelson 1993) (Figure 3). The native form of TPN has been show to specifically binds to various types of inward rectifier potassium channels, including the G-protein coupled inward rectifying potassium channel 1 (GIRK1), GIRK4 (Kd ≈ 8nM), and renal outer medullary potassium channel 1 (ROMK1) (Kd ≈ 2nM) (Jin and Lu 1998). It is theorized that C-terminal α-helix functions as the pharmacophore, occluding the extracellular side of the conduction pore thereby blocking the K+ current in a dose-dependent manner (Jin et al. 1999). This level of specificity for potassium channels is unique to TPN and makes it an invaluable tool for studying the physiological role(s) played by its targets (Jin and Lu 1998). C

N Figure 3. Solution NMR structure of native TPN. Showing the C-terminal alpha helix and Nterminal beta strand, structural elements that are essential to the maintenance of TNP’s biological activity (Krissinel and Henrick 2007) 1.2.2 Stable Synthetic Derivative: Tertiapin-Q When TPN was isolated by Jin and Lu, they noted that the M13 residue could undergo spontaneous oxidation to yield a sulfoxide derivative which detrimentally “altered both the chromatographic behavior and the inhibitory activity of tertiapin”. They found that the oxidized form of TPN not only eluted at a lower organic percentage but also exhibited a 20-fold decrease in its affinity (Jin and Lu 1998). In order to remedy this, the group systematically replaced M13 4

with fourteen different residues and discovered that the glutamine derivative (TPN-Q) exhibited affinity for the ROMK1 channel similar to that of native TPN, while remaining unoxidizable in air (Jin and Lu 1999). Unlike the related voltage-gated K+ channels, inward rectifier K+ channels are poorly understood, partially due to the lack of affinity of the molecular probes used (Imredy et al. 1998; Lu and MacKinnon 1997). The nanomolar affinity of TPN-Q make it a useful probe for elucidating the molecular structure and physiological functions of the ROMK1 channel as well as inward rectifier K+ channels in general.

1.3 ROMK1 Channel 1.3.1 Introduction to Inwardly Rectifying Potassium Channel (IRK) Family Among the plethora of ion channels, potassium channels are the most widely distributed (Littleton and Ganetzky 2000), occurring across most cell types, and controlling a variety of functions (Hille 2001; Kandel et al. 2000). Of these channels, there are four main types: voltagegated potassium channels, leak channels, calcium-activated potassium channels and inward rectifying potassium channels. IRK channels differ from voltage-gated potassium channels in that IRK channels are more permeable to K+ ions during hyperpolarization than during depolarization. At voltages more negative than the equilibrium potential of potassium (EK), less than -80mV, the conductance of IRK channels increases, allowing an influx of K+ into the cell, until resting membrane potential is reestablished. Conversely, these channels are not able to efficiently pass K+ ions out of the cells at voltages that exceed the resting membrane potential. In this manner, IRK channels play a pivotal role in the maintenance of the resting membrane potential, the duration of the action potential and neuronal excitability (Isomoto et al. 1997).

1.3.2 ROMK1 Localization and Importance The ROMK1 channel, a member of the IRK family, is of particular importance due to its splice variants being localized in the thick ascending limb (TAL) and cortical collecting duct (CCD) of human nephrons where they are involved in the recycling and secretion of potassium, respectively (Ho et al. 1993; Kubo et al. 2005; Yano et al. 1994). Specifically, of the six ROMK1 splice variants that have been characterized (Kir1.1a-f) Kir1.1a and Kir1.1c are localized in the distal renal tubule where they are involved in secretion of K+, and Kir1.1b is involved recycling 5

K+ in the TAL of the Loop of Henle (Boim et al. 1995; Kubo et al. 2005). Mutations in the gene that codes for ROMK1 (KCNJ1) are associated with antenatal Bartter’s syndrome type II. Symptoms include salt wasting, hypokalemia, hypercalciuria, and low blood pressure (Fretzayas et al. 2013; Kubo et al. 2005; Puricelli et al. 2010; Schwalbe et al. 1998).

1.3.3 ROMK1 Molecular Structure Since it is a member of the IRK family, ROMK1 is a heterotetramer with each subunit composed of: 2 transmembrane domains (M1 and M2), a pore loop, and cytoplasmic N- and Ctermini (Bichet et al. 2003) (Figure 4). This is in sharp contrast to voltage-gated potassium channels, which typically have six transmembrane segments (Long et al. 2005).

Figure 4. Shows the side-view of the general molecular structure of IRK channels, the family of channels that includes ROMK1 (Bichet et al. 2003) The cytoplasmic N- and C-termini form the cytoplasmic pore and contain two-thirds of the amino acid sequence for IRK channels. The cytoplasmic pore plays a crucial role in the regulation of the channel by interacting with intracellular regulators (such as ATP) and also establishes the electrical potential difference that is necessary for inward rectification. The pore itself is approximately 30 Å long, 7-15 Å in diameter, and is composed of many polar and charged residues arranged in parallel β-sheets (Bichet et al. 2003; Nishida and MacKinnon 2002). The selectivity filter of the pore helix is responsible for the ability of these channels to allow influx of K+ while simultaneously excluding Na+, without this, these channels would not 6

be able to generate an electrical potential. The selectivity filer itself is formed by a “signature sequence” that is conserved across all known K+ ion channel and consists of either TXGYG or TXGFG. Any mutations within this sequence result in an inability to selectively allow influx of K+ (Bichet et al. 2003; Heginbotham et al. 1994). Furthermore, differences in residues that surround the signature sequence itself accounts for variances in the strength of inward rectification between members of the IRK family (Bichet et al. 2003; Slesinger 2001; Yi et al. 2001).

1.3.4 ROMK1 Activation and Probes Members of the IRK family are activated/inhibited by several different cytoplasmic factors, including: PIP2 (phosphatidylinositol-4,5-bisphosphate), arachidonic acid, Na+ ions, Mg2+ ions, pH, trimeric G-proteins, ATP, phosphorylation, redox reactions, and interactions with PDZ domains. The ROMK1 channel is inhibited by an increase in intracellular [H+], which leads to protonation of several C-terminal histidine residues (Chanchevalap et al. 2000; Qu et al. 2000), via the K80 residue (Jiang et al. 2002). However, the mechanism links this intracellular acidification to movement of the transmembrane domains is still under investigation. It has been shown that ROMK1 has a cysteine residue (C49 and C308) at its N- and C-termini that are only accessible to thiol reducing agents (such as dithiothreitol (DTT) and reduced glutathione) while the channel is closed, thereby indicating that some sort of conformational change occurs after the protonation events (Schulte et al. 1998). Understanding the activation and subsequent opening/closing mechanisms of ion channels often requires the use of selective molecular probes. These allow for potent inhibition of the target channel which in turn allows for elucidation and clarification of the channel’s physiological role. These applications necessitate probes with low Kd values in order to minimize the likelihood of non-specific binding. The small molecule VU590 inhibits the ROMK1 channel with an IC50 of 1.5µM (IC50 is directly related to Kd through the Cheng-Prusoff equation (Newton et al. 2008)), however the fact that VU590 also inhibits Kir7.1, which is also expressed in nephrons (Ookata et al. 2000), makes it less useful as a probe for ROMK1 (Lewis et al. 2009). The successors to VU590, VU591 and BNBI, showed more promise by showing no significant inhibition of Kir7.1 and boasting IC50 values of 240 nM and 8 µM, respectively (Bhave et al. 2011). To date, only one venom peptide, other than TPN-Q, has shown activity towards 7

ROMK1: the scorpion toxin Lq2, which demonstrated a Kd of 0.4 µM, which represents a 200fold decrease when compared to the affinity of TPN-Q. Structurally similar scorpion toxins, charybdotoxin (ChTx), iberiotoxin (IbTx), noxiustoxin (NTx), margatoxin (MgTx), agitoxin 1 (AgTx1) and agitoxin 2 (AgTx2), exhibited no significant activity for the ROMK1 channel (Lu and MacKinnon 1997) (Table 1). Table 1: The sequence and activity of several scorpion toxins, similar to Lq2, that failed to show any significant activity for ROMK1. Cysteine residues are shown in bold face font. Activity of Potential ROMK1 Antagonists Sequence Activity GVPINVKCTGSPQCLKPCKD Shaker Ki: 0.16 nM AGMRFGKCINGKCHCTPK

Toxin

Length

AgTx1

38 αα

AgTx2

38 αα

GVPINVSCTGSPQCIKPCKDA GMRFGKCMNRKCHCTPK

Shaker Ki: 0.64 nM

ChTx

37 αα

pGluFTNVSCTTSKECWSVCQ RLHNTSRGKCMNKKCRCYS

Shaker Ki: 227 nM

IbTx

37 αα

pGluFTDVDCSVSKECWSVCK DLFGVDRGKCMGKKCRCYQ

BK Channel Kd: 1.16 nM

MgTx

39 αα

TIINVKCTSPKQCLPPCKAQF GQSAGAKCMNGKCKCYPH

Kv1.3 IC50: 85 pM

TIINVKCTSPKQCSKPCKELY GSSAGAKCMNGKCKCYNN

SGA K+ Channel Kd: 300 nM Synaptosome ED50: 2.9 nM

NTx

39 αα

Ref (Garcia et al. 1994) (Gao and Garcia 2003; Garcia et al. 1994; Naranjo and Miller 1996) (Gao and Garcia 2003; Goldstein and Miller 1993) (Candia et al. 1992; Gao and Garcia 2003) (Cheong et al. 2011) (Carbone et al. 1982; 1993; Possani et al. 1982; Sitges et al. 1986)

1.4 Cyclic Peptides and Cyclization 1.4.1 Benefits of Cyclization Since the structural elucidation of the cyclotide kalata B1 (Saether et al. 1995), these naturally occurring, stable peptides have been discovered on almost every continent in organisms ranging from bacteria to animals (Trabi and Craik 2002). In addition to their extraordinary stability, these peptides have been shown to possess anti-microbial and anti-HIV properties as 8

well, making them extremely versatile (Gustafson et al. 2004). In addition, according to Craik, cyclization also “[ties] up the loose ends” of peptides which enhances the affinity of these molecules for their target and enables them to retain their activity even after being subjected to extreme pH, boiling or exopeptidase activity (Craik 2006). For example, the peptide Rhesus θdefensin (RTD-1) is an 18 amino acid cyclic peptide produced in Rhesus macaque monkeys (Macaca mulatta). RTD-1 is produced in its native, cyclic form in Rhesus white blood cells and plays an integral role in the monkey’s innate immune response (Tang et al. 1999). However, when a synthetic linear version of this peptide, dubbed retrocyclin, was tested, it was shown to be “essentially inactive” at physiological salt concentrations (Cole et al. 2002; Craik 2006) (Table 2 and 3)

9

Table 2: Lists the activity and source (either synthetic or natural) of various cyclic peptides and proteins. The amino acids sequences are shown in Table 3. Activity and Source of Various Cyclic Proteins and Peptides Peptide/Protein

Source/Synthetic Technique

Activity Comments

(Daly and Craik 2009; Jennings et al. 2001; Thapa et al. 2014a) Cyclic: antibiotic, (Cole et al. 2002; Linear: anti-HIV Tang et al. 1999) (2000; MartinezAntibacterial agent Bueno et al. 1994) Cyclic: antibacterial (Blond et al. Linear: 40-fold weaker 1999) (Luckett et al. Trypsin inhibitor 1999) (Hernandez et al. Trypsin inhibitor 2000) POP enzyme IC50: (Hellinger et al. 25.0 ± 0.3 µM 2015) Improved ligand (Camarero et al. binding 2001) Improved ligand (Camarero et al. binding 1998) nAChR IC50 (native): 1.0 ± 0.2 µM (Clark et al. nAChR IC50 (cyclic): 2005) 1.3 ± 0.4 µM nAChR IC50 (linear): 0.83 ± 0.04 µM (2006b; Clark et nAChR IC50 (cyclic): al. 2010) 0.43 ± 0.08 µM Shows increased (Lovelace et al. resistance to 2006) exoprotease digestion Improved cellular (D'Souza et al. uptake 2014) POP enzyme IC50: (Hellinger et al. 27.8 ± 0.3 µM 2015) Insecticidal and antiHIV agent

Kalata B1

Oldenlandia affinis

RTD-1

Macaca mulatta

Bacteriocin AS-48

Enterococcus faecalis S-48

Microcin J25

Escherichia coli

SFTI-1

Helianthus annuus

MCoTI-I and II

Momordica cochinchinensis

psysol 2

Psychotria solitudinum

SH3

Native Chemical Ligation (NCL)

WW (rsp-5) domain

NCL

α-MII

NCL

α-Vc1.1

NCL

χ-MrIA

NCL

MCoTI-M5

Fmoc-Based

psysol 2

Fmoc-Based

10

Ref.

1.4.2 Origins of Chemical Cyclization Cyclic peptides are not limited to natural products. The discovery of solid phase peptide synthesis (SPPS) by Merrifield in the 1960’s (Merrifield 1963) and the subsequent introduction of native chemical ligation (NCL) by the Kent lab group (Dawson et al. 1994) resulted in a dramatic expansion in the number of peptides that could be cyclized. Using standard SPPS to synthesize peptides that are longer than ~40 amino acids results in a high number of side reactions, intermolecular aggregation, and addition/deletion products that severely limit production of the target material (Chan and White 2000; Dawson and Kent 2000; Grant 2002). Through a three step reaction (thiol-thioester exchange, transthioesterification and an S-to-N acyl shift) NCL ligates 2 smaller peptide fragments together and can be used to overcome the length limitation and create peptides as long as 250 amino acids (Chandrudu et al. 2013; Muir et al. 1997), however a more exciting application of NCL is the creation of cyclic peptides. Traditional NCL necessitates that one peptide fragment contain an N-terminal cysteine and that the other contain a C-terminal thioester moiety, if both conditions are not met then the intermolecular thiol-thioester exchange cannot occur. If, however, the thioester and free cysteine are present on the same peptide then intramolecular NCL can be used to generate a cyclic peptide (Figure 5) (Thapa et al. 2014b). After an initial proof of concept by the Craik lab group (Clark et al. 2006a), this technique has been used to stabilize the previously mentioned kalata B1 through incorporation of a poly R sequence (Gunasekera et al. 2008) and also to inhibit the protease activity of the foot-and-mouth disease virus (Thongyoo et al. 2008). Using appropriately sized linker sequences, NCL has also been used to successfully cyclize disulfide-rich peptide toxins such as the α-conotoxins MII (Clark et al. 2005) and Vc1.1 (Clark et al. 2010), and the χconotoxin MrIA (Lovelace et al. 2006). However, the structural requirements of NCL present somewhat of a challenge. The Cterminal thioester can be achieved through Boc SPPS chemistry, which requires the use of HF for resin cleavage and TFA for Boc removal (Hackeng et al. 1999). These high acid conditions present issues when used in conjunction with acid-labile functionalities, therefore being able to achieve cyclization using the milder Fmoc chemistry would be much more desirable. Assembling peptides on 2-chlorotrityl chloride resin enabled the Craik lab group to cleave the peptide under mild acidic conditions (1% TFA), while simultaneously leaving the side-chain protecting groups intact. After cleavage, head-to-tail cyclization was then carried out using N, N, 11

N, N-tetramethyl-O-(7-azabenzotriazol-1-yl)uronium hexafluorophosphate (HATU) as a leaving group and N,N-diisopropylethylamine (DIPEA) which acts as a base to prevent side reactions (Figure 5) (Cheneval et al. 2014). This approach eliminates the need for both a C-terminal thioester and an N-terminal cysteine and allows for milder conditions during synthesis. This Fmoc-based approach opens the door to cyclize any potential peptide.

12

a)

b)

RS

1) Intramolecular Nucleophilic Attack

1) Resin Removal

COOH

O

H2 N

S

O

H2N

2) Cyclization 2) S-N Acyl Shift O N H

3) Side-Chain Deprotection O N H

Figure 5. Side-by-side comparison of a) Fmoc-based cyclization and b) NCL cyclization. Solid shapes represent fully protected peptides, while hollow shapes represent deprotected peptides 13

1.4.3 Cyclic Peptide Linker Sequence When Clark et al. successfully cyclized the α-conotoxin MII, they noted that the distance between the N- and the C- termini plays a crucial role. In order for the resulting cyclic peptide to retain its native bioactivity, this distance must be spanned by an appropriately sized linker sequence of amino acids otherwise excess strain would cause perturbations in the 3D structure, thereby limiting or destroying the bioactivity (Clark et al. 2005). Equation 1 shows the linear relationship between the distance between the N- and the C- termini and the number of amino acids necessary in the linker. Figure 6 shows this equation translated as a graph. Equation 1: # 𝑜𝑓 𝑟𝑒𝑠𝑖𝑑𝑢𝑒𝑠 𝑖𝑛 𝐿𝑖𝑛𝑘𝑒𝑟 =

𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑏𝑒𝑡𝑤𝑒𝑒𝑛 𝑇𝑒𝑟𝑚𝑖𝑛𝑖 (Å) − 5.5 1.01

Typically, the linker sequence is composed of alanine and glycine residues, since their relatively small sidechains minimize the possibility of side reactions (Clark et al. 2010).

14

Figure 6. Shows the strong positive correlation that exists between termini distance and the number of linker residues necessary for successful cyclization. Each data point represents a peptide that has been successfully cyclized (Clark et al. 2010)

15

Table 3: Shows the sequences of the cyclic proteins and peptides seen in Table 2. Cysteine residues are shown in bold, and backbone cyclization is represented by a line connecting the Nand C- termini

Protein/Peptide Kalata B1 RTD-1 Bacteriocin AS-48 Microcin J25

Sequence of Various Cyclic Proteins and Peptides Sequence GLPVCGETCVGGTCNTPGCTCSWPVCTRN GFCRCLCRRGVCRCIVTR MAKEFGIPAAVATVLNVVEAGGWVTTIVSILTAVGSGGLSLLAA AGRESIKAYLKKEIKKKGKRAVIAW VGIGTPISFYGGGAGHVREYF GRCTKSIPPICFPD

SFTI-1 MCoTI-I MCoTI-II psysol 2 SH3 WW (rsp-5) domain

SGSDGGVCPKILQRCRRDSDCPGACICRGNGYCG SGSDGGVCPKILKKCRRDSDCPGACICRGNGYCG GLPICGESCVGGTCNTPGCTCWPVCTRN AEYVRALFDFNGNDEEDLPFKKGDILRIRDKPEEQWWNAEDSEG KRGMIPVPYVEKYG CFEIPDDVPLPAGWEMAKTSSGQRYFLNHIDQTTTWQDPRKAM LSQG

α-MII

GCCSNPVCHLEHSNLC

α-Vc1.1

GCCSDPRCNYDHPEIC

χ-MrIA

NGVCCGYKLCHOC

MCoTI-M5

SGSKGGVCPRILRRCRRDSDCPGACICRGNGYCG

1.5 Fluorophores and Fluorescent Peptides 1.5.1 Benefits of Small Organic Fluorophores The elucidation of cellular mechanisms has been a consistent goal over the last few decades. The pursuit of this has led to the development of many tools for teasing out the intricate 16

relationships between molecules and their targets. One of the most useful of these techniques has been the application of fluorescent molecules as a means to visualize the distribution, targets and/or structure of proteins. The review by Giepmans et al. expounds on how the majority of fluorescent probes that have been used in the past ten years to study protein structure and function have been autofluorescent proteins (AFPs) (Giepmans et al. 2006), such the ubiquitous GFP. Even though AFPs have been used for a wide range of applications, such as creating photoactivatable proteins and sensing membrane potentials (Shaner et al. 2005), these molecules are not without their fair share of limitations. The first of these is size; even on the low end, AFPs are >200 residues in length. This large size creates problems when distance between the AFP itself and activity is of importance. Secondly, the range of wavelengths produced by AFPs is relatively limited, most notably no AFPs exist that emit in the near infrared (NIR) range. And lastly, AFPs have as of yet been unable to serve as probes for certain biomolecules and enzymatic activity (Johnsson and Johnsson 2007). Small organic fluorophores, on the other hand, have the distinct advantage of being able “to use chemistry to dictate the properties and position of the fluorescent dye”. These molecules can also take advantage of the chemical reactions of a living cell in order to localize the dye to a specific sub-cellular region (Wysocki and Lavis 2011). One method of chemically attaching a small organic fluorophore is via the Staudinger reaction. 1.5.2 Staudinger Ligation via Lys-Azide The classical Staudinger ligation technique of reacting a phosphine (R3P) containing compound with an azide moiety (R-N3) (1992; Staudinger and Meyer 1919), can be modified to selectively conjugate a small organic fluorophore to a peptide sequence even in the presence of water (Saxon and Bertozzi 2000) (Figure 7). In the case of this modified reaction, a lysine residue modified with an ε-azide group must be incorporated in order to form the key aza-ylide intermediate. This reaction results in a stable amide linkage between the peptide and the phosphine (Yanagisawa et al. 2008). This labeling reaction is superior to other forms of bioconjugation/fluorescent labeling due to the high selectivity of the phosphine for the azide moiety. In a properly constructed peptide, there is no other azide group present other than the intended target, this results in a highly selective reaction that is nearly bereft of side-products. Although due to the large size of the fluorophore itself, this reaction can take several days, 3-5 on average. 17

1.5.3 Thermo Fisher DyLight® Fluorophore The Thermo Fisher DyLight® fluorophore is a family of phosphine-containing molecules that can be conjugated to peptides via the Staudinger reaction. Furthermore, DyLight® fluorophores exhibit higher intensity and photostability when compared to comparable fluorophore molecules, such as LI-COR and CyDye molecules. DyLight® also retains its fluorescence over a broad pH range (pH 4-9), and its high degree of water solubility allows it to achieve high ratios of dye to peptide without precipitation. Finally, DyLight® molecules have absorption spectra that cover the entire visible spectra as well as some in the NIR and infrared (IR) wavelengths (Figure 8) giving these molecules a high degree of flexibility. Combined, these features of DyLight® make it highly favorable for its use in peptide bioconjugation, an area which this thesis explores.

-N2

-OCH3

Figure 7. Shows the 3 step reaction mechanism for attaching a phosphine-containing fluorophore to the azide moiety of a peptide through the Staudinger reaction (Saxon and Bertozzi 2000)

18

Figure 8. Shows the wide range of wavelengths that DyLight® molecules maximally absorb at.

1.6 Project Objectives and Rationale Part of the goal of this project is to further test the validity of the data presented by Clark et al. that shows an approximately linear relationship between linker length and termini distance (refer to Figure 6). Also, since all of the data used by Clark et al. was collected from Conus spp., this project represents a challenge to these data by testing whether this relationship holds for other venomous organisms, in this case A. mellifera. Additionally, this project seeks to fill gaps in the knowledge about the molecular targets of TPN-Q, most notably the ROMK1 channel. On its own, TPN-Q is a useful molecular probe in studying the relatively mysterious functions of the ROMK1 channel due to its uniquely high level of specificity. However, since cyclic peptides have been shown to be more resistant to enzymatic degradation than their linear counterparts, creating a cyclic, fluorescent derivative of TPN-Q, would open the door for an orally available fluorescent probe for the ROMK1 channel, thereby allowing researchers to determine distribution and localization of this channel. This probe could also aid in the elucidation of the molecular mechanism that results in a change of configuration of ROMK1 after an increase intracellular [H+]. This work also aims to expand upon research conducted by two previous University of Hawaii graduate students who successfully used the Staudinger ligation to conjugate the same fluorophore to two linear toxins (hongotoxin and iberiotoxin) and a smaller cyclic conotoxin (Tx2081). ROMK1 is also medically relevant due to its localization in the ascending Loop of Henle of nephrons, where mutations have been linked to subtypes of Bartter syndrome which can lead 19

to renal failure, and in the inner-mitochondrial membrane (IMM) of cardiac cells, where it has been shown to be involved in preventing hypoxia-induced brain damage as a result of strokes and other ischemic attacks. Cyclic TPN-Q could be used as a scaffold for delivering therapeutics to these key areas, and due to the high specificity of TPN-Q, very few side effects would arise. Finally, a homolog of the ROMK1 channel has been discovered in the Fall Armyworm (Spodoptera frugiperda), a known agricultural pest species. Infestations of these worms on corn, alfalfa, soybean, and cotton crops has been estimated to cause $300-500 million worth of damage annually. Cyclic TPN-Q could potentially be used as a phyla-specific pesticide for this worm that could be sprayed on infected plants. Since TPN-Q is a peptide it would readily decompose in the soil leading to fewer environmental impacts than traditional pesticides.

PROJECT HYPOTHESES 

In order for TPN-Q to be successfully cyclized an appropriately sized linker sequence must be placed between the N- and C-termini. Since the distance between the termini remains relatively constant, we expect only one length variant will be able to cyclize successfully.



Since the three-dimensional conformation of the peptide is recovered through cyclization, we would only expect bioactivity to be demonstrated in the cyclic TPN-Q and not in linear TPN-Q



Adding a lysine-azide residue within the non-critical linker sequence will enable the conjugation of a phosphine-containing fluorophore molecule to TPN-Q via the Staudinger reaction

20

PROJECT GOALS AND OBJECTIVES 

Goal 1: Determine the effect of linker length on the bioactivity of cyclic TPN-Q o Objective 1: Determine the theoretical linker length through in silico distance measurements o Objective 2: Synthesize, oxidize and purify 3 length variants of linear and cyclic TPN-Q



Goal 2: Assess bioactivity of length variants through whole animal LD50 assays o Objective 1: Determine LD50 value through injections into fish



Goal 3: Produce a cyclic, fluorophore derivative of TPN-Q o Objective 1: Re-synthesize TPN-Q, while incorporating a Lys-Azide within the linker. Bioactivity data will be used in order to choose appropriate linker length o Objective 2: Cyclize, oxidize and purify Lys-Azide TPN-Q o Objective 3: Conjugate a phosphine fluorophore to the Lys-Azide TPN-Q through the Staudinger reaction. o Objective 4: Purify fluorescently labeled TPN-Q

21

CHAPTER 2 EXPERIMENTAL PROCEDURES 2.1 Determination of Predicted Optimal Linker Length Since successful cyclization of a peptide requires the inclusion of a linker sequence (refer to Section 1.4.3), the first step before attempting to cyclize TPN-Q was to determine the distance between the N- and C- termini of TPN-Q. This was done using in silico protein structure prediction and measurement software. First, the sequence of TPN-Q was fed into the QUARK ab initio Protein Structure Prediction server, provided through the Zhang Lab webpage at the University of Michigan (Xu and Zhang 2012). Next, each of the ten PDB files generated was opened using three different open-source molecular modeling software: Jmol, RasMol and Avogadro. Using the distance measuring tools provided in each of these programs, the distances between the C-terminal carbon atom and the N-terminal nitrogen atom were measured and recorded, in either Angstroms (Å) or nanometers (nm). These values were then averaged across each of the ten files for all three programs utilized. The averages for each of the programs were then averaged with each other in order to produce the final value that was considered to be the Nto-C distance for TPN-Q (Table 4). This final value was then substituted into Equation 1 in order to determine the optimal linker length that would be required to cyclize TPN-Q. This length was used to inform the synthesis of TPN-Q. Due to inaccuracies in the distance measurements and equations, two length variants were also synthesized: one variant with one less residue in the linker (-1) and one variant with one extra residue in the linker (+1). 2.2 Solid Phase Peptide Synthesis of TPN-Q 2.2.1 Swelling of Pre-Loaded Chlorotrityl Resin 2-Chlorotrityl chloride resin pre-loaded with H-Lys(Boc) was weighed out to contain 1 milliequivalents (mEq) (SV= 0.724 mEq/g) and placed inside of a glass reaction vessel. Approximately 20 mL of dimethylformamide (DMF) was added and the reaction vessel was allowed to shake for 30 minutes. Subsequently, the resin was wrapped in aluminum foil left to swell for at least 12 hours. 22

2.2.2 Activation of the Fmoc Amino Acids Four mmol of the required amino acid was weighed out and placed into a scintillation vial. In a separate scintillation vial, 4 mmol of 2-(6-chloro-1H-benzotriazole-1-yl)-1,1,3,3tetramethylaminium hexafluorophosphate (HCTU) was weighed out. The HCTU was dissolved in 8 mL of DMF, and shaken by hand to mix. This 4-fold excess of activated amino acid to resin creates a supersaturated environment that aids in driving the coupling reaction to completion. The HCTU-DMF solution was then transferred to the scintillation vial containing the amino acid. Finally, 800 µL of N,N-diisopropylethylamine (DIPEA) was added in order to act as a base for any protons released during the activation reaction (Grant 2002). 2.2.3 Amino Acid Coupling The activated amino acid solution was transferred to the reaction vessel and allowed to shake for 30-60 minutes (depending on the amino acid being added). After shaking, a small amount (3-5 mg) of resin was removed from the reaction vessel and used to quantify the coupling reaction using the Ninhydrin assay (Sarin et al. 1981). Once the threshold coupling percentage (≥99.5%) was attained, the process was repeated with the next amino acid in the sequence. 2.2.4 Removal of the 9-fluorenylmethoxycarbonyl (Fmoc) N-Terminal Protecting Group Before adding another amino acid, the resin was flow-washed 3 times with DMF to remove any of the previous amino acid. Then, approximately 5 mL of 50% v/v piperidine:DMF solution is added to the reaction vessel and allowed to shake for 1 minute. The solution is then drained and repeated. This step ensures that the Fmoc group attached to the N-terminal of the previous amino is removed, thereby allowing the next amino acid to attach to the growing chain. After deprotection, the resin is flow-washed 3 more times with DMF in order to remove any leftover piperidine. This procedure of washing, activating, coupling and deprotection is repeated until the desired peptide sequence is completed (Figure 9).

23

2.2.5 Drying Down the Peptide Resin After the final amino is added, deprotected and washed with DMF, the resin was washed with dichloromethane (DCM) to remove the DMF and placed in an inert N2 environment until dry. The dried resin was then weighed and stored in a scintillation vial at -20 oC until cleavage. 2.3 Quantification of Amino Acid Coupling 2.3.1 Ninhydrin Assay After the requisite coupling time but before removal of the N-terminal Fmoc group, a small aliquot of resin (3-5 mg) was extracted from the reaction vessel using a Pasteur pipette. This aliquot is placed in a fritted glass thimble and flow washed with 50% v/v methanol in DCM. The dried resin was then transferred to a test tube and weighed. The following reagents were added to the tube containing the resin and a negative control tube: 2 drops of 76% w/v phenol in ethanol, 4 drops of 0.2 mM KCN in pyridine and 2 drops of 0.28 mM Ninhydrin in ethanol. Both tubes were then covered with Parafilm® and placed in a sand bath regulated to 90-100 oC for precisely 5 minutes. Both tubes were then diluted with 3 mL of 60% v/v ethanol in water and centrifuged for approximately 30-60 seconds in order to pellet the resin particles. The supernatant of each tube was then decanted into separate 1 cm quartz cuvettes and placed in a spectrophotometer. The absorbance of the resin was then measured at 570 nm (Sarin et al. 1981). 2.3.2 Percent Coupling Calculation The absorbance of the sample at 570 nm (Abs570) was then multiplied by 200 and divided by the resin mass in milligrams in order to calculate the Ninhydrin Value (NV) (Equation 2). This value represents the amount of free α-amino groups that did not successfully react during the coupling phase. Equation 2: 𝑁𝑉 =

(𝐴𝑏𝑠570 ) × 200 𝑟𝑒𝑠𝑖𝑛 𝑚𝑎𝑠𝑠 (𝑚𝑔)

This NV is then substituted into Equation 3 along with the substitution value (SV) of the previous amino acid in order to calculate the percent coupling of the current amino acid (Kaiser et al. 1970). If the threshold coupling percentage of 99.5% was reached, then the coupling was 24

considered successful and the synthesis would continue. However, if the coupling percentage was less than 99.5% two options were available depending on how far away the calculated percentage was from the threshold value. These options included washing out the previous amino acid solution and adding a fresh solution of the same amino acid (so-called “double-coupling”), or adding more DIPEA (~200 µL) and letting the reaction vessel shake for another 10-15 minutes. Equation 3: % Coupling = 1 − (

NV ) × 100 SV

2.4 Cleavage from Resin to Yield Linear Peptide In order to cleave the synthesized peptide from the chlorotrityl resin and remove the orthogonal protecting groups a highly acidic cocktail called Reagent K was used in the ratio of 10 mL of Reagent K to 300 mg of resin. The cocktail was comprised of the following substituents: 82.5% v/v TFA, 5% v/v DI water, 5% v/v phenol, 5% v/v thioansiole and 2.5% v/v 1,2-ethanedithiol (EDT). EDT can also be substituted with triisopropylsilane (TIPS), which serves the same function and does not have an offensive odor. However, it was discovered that the yields for this peptide were significantly better if EDT was used. This Reagent K solution, along with the peptide-resin, was placed in an Erlenmeyer flask with a magnetic stir bar and allowed to stir for 2.5 hours at room temperature. Subsequently, the mixture of Reagent K and resin was poured into a glass Buchner funnel containing a Whatman® filter (No.4). The eluates, which contained the peptide, were then pulled through the filter under vacuum and collected in a pear flask, thereby separating it from the resin particles. The Erlenmeyer flask was then washed with 2-3 mL of 100% TFA, which was also poured through the filter. The volume of the pear flask was then distributed evenly across the necessary number of 50 mL Falcon® tubes along with approximately 40 mL of methyl tert-butyl ether (MTBE) chilled in liquid N2 (LN2). The tubes were then centrifuged for 7 minutes at 3000 rpm in order to form the initial peptide pellet, at which point the supernatant was decanted off, the pellet was re-suspended in chilled MTBE and centrifuged in the same manner in order to ensure optimal extraction of the peptidic material. This final peptide pellet in each Falcon® tube was then dissolved in 10-15 mL of 25% (v/v) acetic acid in DI water and transferred to an appropriately sized round-bottom flask. The 25

volume of the flask was then shelled using LN2 and freeze-dried overnight. The lyophilized material was then weighed and stored in a scintillation vial at -20 oC until required. 2.5 Cleavage from Resin to Yield Cyclic Peptide 2.5.1 Cleavage from the Chlorotrityl Resin Between 600 and 800 mg of dried peptide resin was placed in a fritted reaction vessel and was washed 10 times with 5 mL of 1% v/v TFA in DCM for 5 minutes. The eluates containing the peptide were collected in a round-bottom flask after each wash. The contents of the roundbottom were then diluted with 150 mL of 50% v/v MeCN, 0.05% v/v TFA in DI water. The majority of the DCM and TFA were removed under vacuum using a rotary evaporator and the remaining contents were shelled using LN2 and freeze-dried overnight. The resulting powder was then weighed and stored at -20 oC until needed. This procedure cleaved the linear peptide from the chlorotrityl resin while leaving the protectional groups attached. 2.5.2 Cyclization Reaction Between 400 and 500 mg of the crude, protected peptide was placed in an Erlenmeyer flask containing 2 mM of DMF, 5 mM HCTU and 10 mM DIPEA and was allowed to stir for 3 hours at room temperature. The mixture was then diluted with twice its volume of 50% v/v MeCN, 0.05% v/v TFA in DI water, transferred to a round-bottom flask, shelled using LN2 and freeze-dried (Cheneval et al. 2014). Due to the low volatility of DMF, the mixture frequently melted during the lyophilization process. The mixture was re-shelled as needed until a dried material remained. The material was then weighed and stored at -20 oC until needed. This procedure joined the N- and the C- termini of the peptide in a native peptide bond, yielding a cyclic peptide with protectional groups intact. 2.5.3 Removal of Orthogonal Protecting Groups Side-chain deprotection of the cyclic peptide was carried out in the same fashion as linear peptides (see 2.3 Cleavage from Resin to Yield Linear Peptide). Between 300 and 400 mg of cyclic protected material was subjected to Reagent K cleavage and then subsequently extracted using chilled MTBE. The resulting peptide pellets were dissolved in 25% v/v acetic acid in DI

26

water and freeze-dried overnight. The resulting lyophilized material was then weighed and stored at -20 oC until needed (Figure 10).

27

Anchoring

Repetitive Cycle

Nα-deprotection -A

Coupling -H2O

1) Nα-deprotection 2) Resin Cleavage 3) Side-Chain Deprotection

Figure 9. Shows the repetitive step-wise process of solid phase peptide synthesis. This process consists of deprotection, during which the N-terminal Fmoc group (A) is removed by piperidine and coupling, which adds another N-terminally protected amino acid to the growing peptide chain. These steps are repeated until the sequence is complete at which point the side-chain protecting groups (B) and resin are removed. The anchoring step was not done in this project since the resin that was used was pre-loaded with the first amino acid in the sequence. Adapted from (Grant 2002).

28

Step 1 1% TFA in DCM

2mM DMF 5mM HCTU 10mM DIPEA

Step 2

Step 3 82.5% TFA 5% DI Water 5% Phenol 5% Thioanisole 2.5% EDT/TIPS Figure 10. Shows the 3 reactions necessary for Fmoc-based cyclization. Step 1 cleaves the peptide from the chlorotrityl resin, step 2 selectively joins the N- and C- termini in a peptide bond through a dehydration reaction, and step 3 removes the side-chain protectional groups. Solid shapes represent fully protected peptide, while hollow shapes represent deprotected peptide. 2.6 Oxidation of Thiols to Form Disulfide Bonds Oxidation of cysteines to form disulfide bridges was carried out using NH4HCO3 buffer in DI water. 10 mL of 100 mM of this buffer was prepared fresh per 1 mg of peptide. The pH was adjusted to approximately 7.8-8.0 using concentrated NaOH and HCl, typically 20-30 mg of peptide was oxidized at once. The peptide was then placed in an Erlenmeyer flask, dissolved in the buffer. Perforated Parafilm® was then placed over the mouth of the flask in order to prevent contaminants from falling but also allow adequate air flow for oxidation. The flask was then placed in a cold room that was temperature controlled to approximately 4 oC and allowed to stir on a magnetic stir plate. After 5 days of oxidation, the flask was removed from the cold room and the oxidation was quenched using 100% acetic acid until the pH dropped to 4. Finally, the 29

contents of the flask were transferred to a round-bottom flask and shelled using LN2 and freezedried overnight. The lyophilized peptide was then weighed and stored and -20 oC until needed. 2.7 Purification of Peptide Samples Using RP-HPLC 2.7.1 RP-HPLC Columns The column selected for purification depended on the mass of peptide that needed to be purified. Peptide samples ranging in mass from 20-50 mg were purified using a large semipreparatory column (Vydac® C18, 10 µm, 22 mm i.d. x 250 mm), 5-20 mg of material was purified on a semi-preparatory column (Vydac® C18, 5 µm, 10 mm i.d. x 250 mm), 1-5 mg was separated using an analytical column (Vydac® C18, 5 µm, 4.6 mm i.d. x 250 mm) and finally a capillary column (Phenomenex® C18, 5 µm, 1.00 mm i.d. x 250 mm) was used in order to assess the purity of a sample and required only µg quantities of material. 2.7.2 Sample Preparation Due to high number of charged amino acids present, TPN-Q was easily soluble in relatively small amounts of Sol. A (0.1% v/v TFA in water). Before injection onto RP-HPLC/UV systems peptide samples were dissolved in Sol. A up to a concentration of 2 mg/mL. The solution was then filtered using 0.45 µm nylon filters and injected using experimental syringes after extra care was taken to remove any excess air present. 2.7.3 Sample Loading After flushing the system with a high ratio of organic solvent (80% v/v Sol. B in Sol. A), and then re-equilibrating the system to 5% v/v Sol. B in Sol. A, the sample to be purified was injected onto a Waters® 600E/2695 and pumped onto the attached column. After injection, the system was kept at 5% v/v Sol. B in Sol. A for approximately 10-20 minutes in order to allow any salts in the sample to elute off the system (i.e. desalting). Once the system reestablished baseline absorbance readings, the solvent gradient was initiated. 2.7.4 RP-HPLC Gradients From 5-65 min. of the purification run, the initial solvent conditions (5% v/v Sol. B in Sol. A) underwent a 1% increase in Sol. B per minute and a 1% decrease in Sol. A per minute. 30

Therefore, at t = 65min., the system was at 65% v/v Sol. B in Sol. A. Over the next 5 minutes (65-70 min.) Sol. B increased by 3% v/v per min while Sol. A decreased in a similar fashion. The subsequent 5 minutes (70-75 min.) of the gradient was 80% v/v Sol. B in Sol. A in order to flush the column of any remaining contaminants. Finally, the last 5 minutes of the run (75-80 min.) reset the flow back to the initial conditions through a 15% v/v increase in Sol. A and similar decrease in Sol. B per minute. 2.7.5 Sample Collection Depending on the flow rate being using for purification, the desired eluates were collected in either 20 mL scintillation vials or 2 mL Eppendorf tubes. The desired peaks were determined using a Waters® Photodiode Array (PDA) 996 detector and visualized simultaneously at 214 nm, corresponding to the absorbance of the peptide backbone, and 280 nm corresponding to the absorbance of aromatic side chains. After collection, the samples were frozen in LN2 and lyophilized overnight and then subsequently stored at -20 oC until required. 2.8 Electrospray Ionization Mass Spectrometry (ESI-MS) Methods 2.8.1 Sample Preparation Lyophilized peptides were reconstituted in 50 µL of 50% v/v Sol. A’ in Sol. B’. Sol. A’ was composed of 0.1% v/v formic acid in DI water, while Sol. B’ was made up of 90% v/v MeCN in 0.8% v/v formic acid in DI water. The average concentration of samples was approximately 1-10 µM. 2.8.2 Sample Injection Less than 5 µL of the above sample preparation was manually injected onto a PE Sciex API 3000 LC/MS/MS using a Hamilton® syringe. The injection line containing the sample joined with the main flow line which flowed at 50 µL per minute (50% Sol. A’ in Sol. B’) into the turbo spray ionizer. Data acquisition of the mass to charge ratio was monitored using Analyst® 1.4.1 software.

31

2.9 Whole Animal Bioassay Protocols 2.9.1 Injection into Green Swordtail Fish (Xiphophorus helleri) A 1.15 mM stock of each peptide was created by dissolving the appropriate amount of lyophilized peptide in PBS solution. Injections of three different dosages into X. helleri fish, ranging in weight from 200-400 mg, were carried out in triplicate for each peptide being assayed. Due to a limitation of peptidic material, these different dosages were achieved through varying the volume injected. These volumes were 2 µL, 4 µL, and 8 µL. A total of ten animals were sacrificed for each peptide, including one negative control animal that was only injected with PBS. The time of initial injection was recorded and the experiment was monitored until all the specimens had died or a total of 24 hours had elapsed, at which point the experiment was terminated. After the experiment, all specimens that had been injected were euthanized using a diluted solution of tricaine mesylate before being stored at -20 oC prior to disposal. Euthanized specimens were then autoclaved in order to prep them for proper disposal. Animals that did not receive any injections were returned to their storage tanks (protocol # 08-547-8, Evaluation of novel native and synthetic conopeptides on Fish; PI: Bingham). 2.9.2 Extrapolation of the LD50 Value The dosage, in µg toxin/g fish, of each injection was calculated and then averaged. This was then plotted against the average of dosage/survival time, measured in µg toxin/g fish/min. This was done for each of the three dosages for each peptide. An equation was then generated using a linear line of best fit, which was in turn used to calculated the y-intercept of the data. This value was taken to be the LD50 value of the peptide (Meier and Theakston 1986). 2.10 Fluorescent Labelling through the Staudinger Reaction 51.36 nmols of pure, oxidized cTPN-Q6f was dissolved in 485 µL of 1X PBS in order to give an approximately 0.1mM peptide solution. Subsequently, 5.14 µL of a pre-prepared 10mM stock of DyLight® 650 in DMF was added to the peptide solution in order to have a 1:1 mole ratio of DyLight® to peptide. The reaction was allowed to stir in a scintillation vial with a MicroFlea stir bar for 5 and 8 days at which point, various concentrations of the reaction mixture

32

was diluted in Sol. A and purified on an analytical RP-HPLC system using a PDA detector, in series with a Scanning Fluorescence detector.

33

CHAPTER 3 RESULTS AND DISCUSSION 3.1 In silico Distance Measurements of TPN-Q Table 4. Summarizes the results of in silico distance measurements using the QUARK structure prediction server and three different 3D molecular modeling software in order to determine the termini distance for TPN-Q Server Model # Software

QUARK 1

2

3

4

5

6

7

8

9

10

Average

Jmol (nm)

1.136

0.928

0.973

0.999

0.962

1.395

1.434

0.897

1.456

1.084

1.126

RasMol (Å)

11.36

9.28

9.73

9.99

9.62

13.95

14.33

8.97

14.55

10.84

11.262

Avogadro (Å)

11.36

9.28

9.727

9.991

9.623

13.947

14.335

8.968

14.555

10.842

11.263

Average (Å)

11.36

9.28

9.729

9.990

9.621

13.949

14.335

8.969

14.555

10.841

11.263

Each of the ten 3D structures generated by the QUARK server was opened in Jmol, RasMol and Avogadro. Then, using the appropriate function, these molecular modeling programs were used to calculate the distance between the N- and C- termini. These measurements were then averaged within each software and finally between software in order to generate the final measurement of 11.263 Å. This value was then substituted into Equation 1, in order to generate the predicted optimal number of residues required for the linker sequence, which is approximately 5.2 residues. This value was rounded up, since a linker of 5 amino acids would have been too short, in order to give a predicted optimal linker length of 6 amino acids.

34

TPN-Q ALCNCNRIIIPHQCWKKCGKK +5 Amino Acids

+6 Amino Acids

cTPN-Q5 AGAGAALCNCNRIIIPHQCWKKCGKK

+7 Amino Acids cTPN-Q7 AGAGAGAALCNCNRIIIPHQCWKKCGKK

cTPN-Q6 GAGAGAALCNCNRIIIPHQCWKKCGKK Linear TPN-Q6 GAGAGAALCNCNRIIIPHQCWKKCGKK Figure 11. Flow diagram showing the various versions of TPN-Q that were used for this project. cysteine residues (C), all of which are oxidized to form disulfide bonds, and the additional linker sequences are shown in bold face font. 3.2 Failed Cyclization of cTPN-Q5 and cTPN-Q7 Two of the length variants that were synthesized via SPPS correspond to +1 and -1 of the optimal linker length predicted by Clark et al. 2010 (refer to Figure 6). These peptides are referred to as cTPN-Q7 and cTPN-Q5, respectively. After synthesis (refer to section 2.1) and cleavage (refer to section 2.4), the putative cyclic peptides were assessed for purity via analytical RP-HPLC (Figures 12 and 14) and characterized via ESI-MS (Figures 13 and 15). The theoretical m/z for the [M+5H]5+ and the [M+4H]4+ charge states of cTPN-Q5 are 554.27 Da and 692.59 Da, respectively. While these masses appear in Figure 13, the dominant mass is 558.5 Da which corresponds to the [M+5H]5+ charge state of linear TPN-Q5. Since cyclization occurs via a dehydration reaction, if the peptide had been successfully cyclized a signal corresponding to a loss of 18 Da would have been observed. Taking the observed charge state signal (558.5 Da) and back-calculating to the parent mass yields a mass of 2784.37 Da. Comparing this observed mass to the expected parent mass for cTPN-Q5 of 2766.37 Da represents an addition of exactly 18 Da.

35

From these data it is clear the cyclization of TPN-Q5 was unsuccessful, yielding a linear peptide instead of a cyclic one. This is presumably due the fact that TPN-Q5 contains a linker sequence that is one amino acid shorter than the predicted optimal length. Thus, the sequence was unable to bridge the gap between the N- and C- termini.

Rt: 38.4 minutes 80

A214

% MeCN 5

Figure 12. Analytical RP-HPLC/UV chromatogram of reduced cTPN-Q5, after cyclization and side-chain deprotection, extracted at 214nm. The elution time of the peak indicated by the arrow is 38.4 minutes

36

Relative Intensity

558.5 [M+5H]5+

693.2 [M+4H]4+ 554.9 [M+5H]5+

m/z, Da Figure 13. ESI-MS spectra of reduced cTPN-Q5 after cyclization and side-chain deprotection. The dominant signal at 558.5 Da corresponds to [M+5H]5+ of linear TPN-Q5, while the minor signals at 554.9 Da and 693.2 Da correspond to the [M+5H]5+ and [M+4H]4+ charge states, respectively, of cyclic TPN-Q5 For cTPN-Q7, the theoretical m/z for the [M+5H]5+ and [M+4H]4+ charges states are 579.90 Da and 724.62 Da, respectively. As with cTPN-Q5, the cyclization might have been marginally successful since the masses 580.8 and 725.3 do appear in the ESI-MS spectrum of cTPN-Q7 (Figure 15), but the spectrum is dominated by the mass 584.1 Da, which once again corresponds to the [M+5H]5+ charge state of linear TPN-Q7. This failed cyclization can once again be attributed to the length of the linker. Since cTPN-Q7 contained a linker that was longer than the predicted optimal value, the resulting N-terminus of the molecule was too flexible making it difficult for it to react with the C-terminus. This is further compounded by the fact that there are two adjacent lysine residues protected by Boc groups present on the C-terminal. These

37

relatively large amino acids with bulky side-chain protecting groups would cause further steric hindrance for the dehydration reaction that is necessary for cyclization.

Rt: 38.2 minutes 80

A214

% MeCN 5

Figure 14. Analytical RP-HPLC/UV chromatogram of reduced cTPN-Q7, after cyclization and side-chain deprotection, extracted at 214nm. The elution time of the peak indicated by the arrow is 38.2 minutes

38

Relative Intensity

584.1 [M+5H]5+

580.8 [M+5H]5+

725.3 [M+4H]4+

m/z, Da Figure 15. ESI-MS spectra of reduced cTPN-Q7 after cyclization and side-chain deprotection. The dominant signal at 584.1 Da corresponds to [M+5H]5+ of linear TPN-Q7, while the minor signals at 580.8 Da and 725.3 Da correspond to the [M+5H]5+ and [M+4H]4+ charge states, respectively, of cyclic TPN-Q7 After this unsuccessful attempt, cyclization of cTPN-Q7 was reattempted with the inclusion of sonication for the entire 3 hours of the reaction, in lieu of a traditional stir plate. It was thought that the sonication would provide additional thermal energy in order to aid the cyclization reaction. This attempt was also unsuccessful and appeared to produce worse results as the ESI-MS spectrum showed new masses that did not correspond to any of the theoretical masses of cTPN-Q7 or linear TPN-Q7 (Figure 17). Furthermore, the RP-HPLC chromatogram shown in Figure 15 shows peaks that are much less resolved when compared to Figure 14. This sample was called cTPN-Q7s.

39

Rt: 38.5 minutes 80

Rt: 39.7 minutes

A214

% MeCN 5

Figure 16. Analytical RP-HPLC chromatogram of reduced cTPN-Q7s, after cyclization and side-chain deprotection, extracted at 214nm. The elution time of the peak indicated by the black arrow is 38.5 minutes, while the peak indicated by the grey arrow elutes at 39.7 minutes. The peaks are much less resolved than in Figure 13, indicating that sonication detrimentally affects the cyclization reaction. 453.7

491.0

Relative Intensity

471.3

m/z, Da Figure 17. ESI-MS spectra of reduced cTPN-Q7s after cyclization and side-chain deprotection. None of the three dominant signals correspond any of the theoretical m/z of linear or cyclic TPNQ7 40

Part of the goal of this project was to investigate the differences between the linear and cyclic forms of TPN with an added linker sequence. Therefore, since both cTPN-Q5 and cTPNQ7 failed to cyclize, oxidizing and purifying their respective linear forms would have been meaningless as there would be no point of comparison. 3.3 Successful Cyclization of cTPN-Q6 The final length variant includes the predicted optimal linker of six amino acids, cTPN-Q6. After being subjected to identical synthesis and cleavage protocols as cTPN-Q5 and cTPN-Q7, cTPN-Q6 was analyzed via analytical RP-HPLC (Figure 18) and characterized using ESI-MS (Figure 19) but with very different results. The theoretical m/z for the [M+5H]5+ charge state of cTPN-Q6 is 565.68 Da. Figure 19 clearly shows a dominant mass of 566.5 Da. Since the accuracy of the machine is ±1 Da, it is clear that this peak corresponds to the target mass. The two smaller signals on the spectrum, 570.1 Da and 712.0 Da, correspond to the [M+5H]5+ and [M+4H]4+ charge states of the linear form of TPN-Q6 which have a theoretical m/z of 569.28 Da and 711.35 Da, respectively. These peaks imply that the cyclization reaction did not go all the way to completion since some of the linear form is still present. This also explains the small leading shoulder present in the RP-HPLC chromatogram (Figure 18). With the successful cyclization of this variant, oxidation and purification of both the linear and cyclic forms were carried out as explained in Section 2.5 and 2.6, respectively.

41

Rt: 38.1 minutes 80

Rt: 39.4 minutes

A214

% MeCN 5

Figure 18. Shows the analytical RP-HPLC chromatogram of reduced cTPN-Q6 extracted at 214nm. The retention time of the main peak is 39.4 minutes. The small leading shoulder at 38.1 most likely corresponds to the remnants of linear TPN-Q6 that was unable to cyclize. This is corroborated by the appearance of linear masses in Figure 19

42

Relative Intensity

566.5 [M+5H]5+

570.1 [M+5H]5+

712.0 [M+4H]4+

m/z, Da Figure 19. ESI-MS spectra of reduced cTPN-Q6 after cyclization and side-chain deprotection. The dominant signal at 566.5 Da corresponds to the [M+5H]5+ charge state. The minor signals at 570.1 Da and 712.0 Da correspond to the [M+5H]5+ and [M+4H]4+ charge states, respectively, of linear TPN-Q6. The successful cyclization of cTPN-Q6 also expands the findings reported by Clark et al. 2010 (refer to Figure 6). Since both cTPN-Q5 and cTPN-Q7 failed to cyclize, we can assume that the relationship between termini distance and linker length can not only be applied to toxins from Conus spp., but to other peptide toxins as well. With this information, future studies can more reliably apply this relationship when attempting to cyclize peptide toxins. 3.4 Oxidation and Purification of Cyclic and Linear TPN-Q6 Oxidation and purification of reduced linear TPN-Q6 (Figure 20 and 22A) appeared to yield only one isomer as evidenced by the chromatogram shown in Figure 21. The 569.1 Da and 711.2 Da peaks in the ESI-MS spectra (Figure 22B) corresponds to the theoretical m/z of the [M+5H]5+ and [M+4H]4+ charge states, respectively, thereby confirming the successful oxidation of the material.

43

Rt: 35.2 minutes

80

A280

% MeCN 5

Figure 20. Shows the analytical RP-HPLC chromatogram of reduced linear TPN-Q6 extracted at 280nm. The retention time of the main peak is 35.2 minutes.

Rt: 34.6 minutes 80

A214

% MeCN 5

Figure 21. Shows the analytical RP-HPLC chromatogram of oxidized linear TPN-Q6 extracted at 214nm. The retention time of the main peak is 34.6 minutes.

44

A)

B)

Relative Intensity

570.1 [M+5H]5+

569.1 [M+5H]5+

711.2 [M+4H]4+

712.0 [M+4H]4+

m/z, Da

m/z, Da

Figure 22. A) Shows the ESI-MS spectra of reduced linear TPN-Q6. The dominant signal at 570.1 Da corresponds to [M+5H]5+ charge state, while the 712.0 Da signal corresponds [M+4H]4+ charge state B) Shows the ESI-MS spectra of oxidized linear TPN-Q6. One of the dominant signal at 569.1 Da corresponds to [M+5H]5+ charge state, while the 711.2 Da signal corresponds [M+4H]4+ charge state. However, the oxidation of cTPN-Q6 seemed to yield multiple isomers (Figure 23). Since this peptide contains four cysteine residues, two disulfide bonds can be formed in three possible isomeric forms. Therefore, since this oxidation was not done selectively, we would expect to find three peaks, corresponding to the three possible isomers. However, upon purifying the oxidized material it was discovered that only two of three main peaks correspond to isomers, while the third appears to be an 18 Da deletion product of the [M+6H]6+ charge state (Figure 28 and 29). This deletion usually results from the loss of water from one of the amino acids in the sequence. These two isomers were purified (Figure 25-27) and subsequently called Isomer 1 and Isomer 2, based on their relative elution time. The deletion product was initially called Isomer 3, since it has the latest retention time.

45

Rt: 37.8 minutes 80

Rt: 38.4 minutes

A214

% MeCN

Rt: 39.6 minutes

5

Figure 23. Shows the analytical RP-HPLC chromatogram of crude oxidized cTPN-Q6 extracted at 214nm. The two black arrows correspond to isomer 1 and 2, while the grey arrow corresponds to the deletion product. The elution time of Isomer 1 is 37.8 minutes and comprises 9.8% of the total area, the elution time of Isomer 2 is 38.4 minutes comprising 14.60%, and the deletion product that elutes at 39.6 minutes makes up 12.78%.

Relative Intensity

565.7 [M+5H]5+

711.2 [M+4H]4+ 569.0 [M+5H]5+ 471.5 [M+6H]6+

706.7 [M+4H]4+

m/z, Da Figure 24. Shows the ESI-MS of crude oxidized cTPN-Q6. The signals at 565.7 Da, 706.7 Da, and 471.5 Da correspond to the [M+5H]5+, [M+4H]4+, and [M+6H]6+ charge state, respectively, of oxidized cTPN-Q6. Since this was before purification, the signals at 569.0 Da and 711.2 Da most likely correspond to the [M+5H]5+ and [M+4H]4+ charge state, respectively of oxidized linear TPN-Q6

46

Rt: 37.6 minutes

80

A214

% MeCN 5

Figure 25. Shows the analytical RP-HPLC chromatogram of purified cTPN-Q6 Isomer 1 extracted at 214nm. The retention time is 37.6 minutes.

80

Rt: 38.3 minutes

A214

% MeCN 5

Figure 26. Shows the analytical RP-HPLC chromatogram of purified cTPN-Q6 Isomer 2 extracted at 214nm. The retention time is 38.3 minutes.

47

A) 565.6 [M+5H]5+

565.7 [M+5H]5+

Relative Intensity

B)

706.8 [M+4H]4+

706.7 [M+4H]4+

m/z, Da

m/z, Da

Figure 27. A) Shows the ESI-MS spectra of cTPN-Q6 Isomer 1. The dominant signal at 565.6 Da corresponds to the [M+5H]5+ charge state, while the signal at 706.7 Da corresponds to the [M+4H]4+ charge state. B) Shows the ESI-MS spectra of cTPN-Q6 Isomer 2. The dominant signal at 565.7 Da corresponds to the [M+5H]5+ charge state, while the signal at 706.8 Da corresponds to the [M+4H]4+ charge state.

80

Rt: 40.0 minutes

A214

% MeCN 5

Figure 28. Shows the analytical RP-HPLC chromatogram of purified cTPN-Q6 deletion product extracted at 214nm. The retention time is 40.0 minutes

48

Relative Intensity

453.8 [M+6H]6+ -18

m/z, Da Figure 29. ESI-MS spectra of the cTPN-Q6 deletion product. The dominant signal at 453.8 Da corresponds to a deletion of approximately 18 Da on the [M+6H]6+ charge state of cTPN-Q6. The appearance of an extra isomer in cTPN-Q6 when compared to linear TPN-Q6 can be attributed to the 6 amino acid linker sequence present in both peptides. In cTPN-Q6, both ends of this linker sequence is held in a stable peptide bond, thereby preventing it from interacting with the solutions. In linear TPN-Q6, on the other hand, while the C-terminal side of the linker sequence is attached to the rest of the peptide, the N-terminal side is allowed to freely associate with the rest of the oxidation buffer. This fact could confer added stability to linear TPN-Q6 allowing it to fold more easily and only form one disulfide isomer. It is also possible that the other disulfide isomers of linear TPN-Q6 are less stable than their cyclic counterparts causing them to breakdown quickly resulting in negligible amounts being present at the end of the oxidation reaction. 3.5 Whole Animal Bioactivity Assays of Linear and Cyclic TPN-Q6 The results of injections of cTPN-Q6 Isomer 1, Isomer 2 and linear TPN-Q6 into X. helleri fish are shown in Tables 5, 6, and 7, respectively. Isomer 1 not only demonstrated bioactivity at all of the dosages attempted, but acted in as little as 7 minutes (Table 5). After creating a plot of the data (Figure 30), the LD50 value for Isomer 1 was extrapolated by solving

49

for the y-intercept of the best fit line of the data. This value worked out to be 32.515 µg/g fish, or 11.532 nmols/g fish. cTPN-Q6 Isomer 2 did not demonstrate any significant activity at any of the dosages even after monitoring for 18 hours (Table 6). However, Isomer 2 did cause death in one fish at Dose #3 and one at Dose #2 (data not shown). This could indicate low levels of activity, but due to the small sample size it is difficult to know. It is also possible that the sample was contaminated with small amounts of Isomer 1 leading to false positives. Linear TPN-Q6, on the other hand, was unable to demonstrate any level of activity through this assay (Table 7). Since this peptide did not show any activity after 60 minutes at the highest dosage, the experiment was terminated due to ethical considerations. 60 minutes was designated as the cutoff time since that was the longest amount of time that it took for Isomer 1 to show activity. Even though the fish that were injected with this peptide appeared to alter their swimming pattern shortly after being injected, this behavior did not persist and by time the experiment was terminated all of the injected fish seemed to have fully recovered. Therefore, it can be assumed that this behavior was caused by the physical trauma of the injection itself. Table 5. Shows the data gathered from injections of cTPN-Q6 Isomer 1 into X. helleri fish. Each dosage was performed in triplicate and then averaged to produce these data. The columns that were used to construct the plot in Figure 30 are indicated by the thick border. Dose # (n=3)

Amount Injected (µg)

Average Fish Weight (g)

Average Dosage (µg/g fish)

Average Survival Time (min)

Average Dosage/ Survival Time (µg/g fish/min)

1

6.5

0.343

19.1

58

0.330

2

13.0

0.200

66.5

46

33.326

3

26.0

0.231

115.6

7

18.017

50

Table 6. Shows the data gathered from injections of cTPN-Q6 Isomer 2 into X. helleri fish. Each dosage was performed in triplicate and then averaged to produce these data. The all three dosages were monitored for a total of 18 hours before being terminated. Dose # (n=3)

Amount Injected (µg)

Average Fish Weight (g)

Average Dosage (µg/g fish)

1

6.5

0.337

20.0

2

13.0

0.427

31.6

3

26.0

0.297

88.0

Average Survival Time (min)

Average Dosage/ Survival Time (µg/g fish/min)

No significant effect after 18 hours

Table 7. Shows the data gathered from injections of linear TPN-Q6 into X. helleri fish. Each dosage was performed in triplicate and then averaged to produce these data. The experiment was terminated due to ethical considerations after no activity was observed at the highest dosage. Dose # (n=3)

Amount Injected (µg)

Average Fish Weight (g)

Average Dosage (µg/g fish)

Average Survival Time (min)

Average Dosage/ Survival Time (µg/g fish/min)

1

N/A

N/A

N/A

N/A

N/A

2

N/A

N/A

N/A

N/A

N/A

3

26.0

0.24

109.8

No observable activity after 1 hour

Comparing results from the injections of cyclic TPN-Q6 (Tables 5 and 6) to the injections of linear TPN-Q6 (Table 7) another trend becomes evident. Since both of the cyclic isomer showed some level of activity and the linear form showed none at all, we can conclude that through the process of cyclizing the peptide backbone of TPN-Q, we were able to restore the bioactivity of this peptide. This implies that the cyclic versions of TPN-Q were most likely able to adopt a 3-dimensional structure that allowed them to be bioactive, while the linear form failed to do this.

51

LD50 Plot of cTPN-Q6 Isomer 1 140 y = 4.7791x + 32.515 R² = 0.8792

Average Dosage (µg/g fish)

120 100 80 60 40 20 0 0

2

4

6

8

10

12

14

16

18

20

Average Dosage/Survival Time (µg/g fish/min)

Figure 30. Shows the plot of the average dosage vs. the average dosage/survival time for the whole animal bioassay of cTPN-Q6 Isomer 1. Each data point represents an average of each dosage done in triplicates. The R-squared value indicates a strong linear relationship in the data, and the y-intercept of best fit line is equivalent to the LD50 value for this isomer. 3.6 Oxidation and Purification of cTPN-Q6 with a Lysine-Azide Residue Before the DyLight® 650 fluorophore could be conjugated to cyclic TPN-Q via the Staudinger reaction, a lysine residue modified with an azide moiety had to be inserted within the linker sequence (refer to section 1.5.2). This was accomplished by using a synthetic Fmoc-LysAzide (Figure 31) reagent while re-synthesizing cTPN-Q6 in order to replace the alanine at position 4. This position was chosen for the reason that it is as far away from the crucial Cterminal alpha-helix as possible, while still being near the center of linker sequence. This modified version of cTPN-Q6 was called cTPN-Q6f. This peptide was then cyclized, oxidized and purified in the same manner as the described in Section 2 (Figure 32-34). The success of the synthesis was determined by the appearance of an extra charge state in the ESI-MS spectra. As a result of the azide, cTPN-Q6f can theoretically acquire one more positive charge than cTPN-Q6. This shifted the dominant charge state from [M+5H]5+ to [M+6H]6+. While the [M+6H]6+ charge state does appear in the ESI-MS spectra of cTPN-Q6 (Figure 24), the signal is weak and only slightly above the background noise of the sample, whereas the [M+6H]6+ charge state clearly dominates the spectra of cTPN-Q6f (Figure 34A). Since cTPN-Q6f was directly synthesized via 52

SPPS to include a Lys-azide residue, we would expect the resulting peptide to consist almost exclusively of this material, thereby leading to a clear dominance of the [M+6H]6+ charge state. While the distribution of ions may have shifted back in favor of the [M+5H]5+ charge state after the oxidation of cTPN-Q6f (Figure 34B), the strength of the [M+6H]6+ charge state signal gives us confidence that the azide remained intact through the oxidation. However, it is unclear why the dominant ion in the spectra reverted back after oxidation of the disulfide bonds.

Figure 31. Shows the structure of Fmoc-Lys-Azide that was used during the synthesis of cTPNQ6f. Importantly, this synthetic amino acid has an azide group (R-N3) attached to the ε-carbon instead of an amino group (R-NH3) as is found in a standard lysine residue.

80

Rt: 37.1 minutes

A214

% MeCN 5

Figure 32. Shows the analytical RP-HPLC/UV chromatogram of reduced cTPN-Q6f extracted at 214nm. This molecule has been cyclized and contains a Lys-Azide within the linker sequence. The retention time is 37.1 minutes. 53

80

Rt: 35.1 minutes

A214

% MeCN 5

Figure 33. Shows the analytical RP-HPLC/UV of oxidized cTPN-Q6f. The retention time of the main peak is 35.1 minutes.

Relative Intensity

A)

484.8 [M+6H]6+

B)

580.8 [M+5H]5+

725.3 [M+4H]4+

484.0 [M+6H]6+

581.3 [M+5H]5+

m/z, Da m/z, Da Figure 34. A) Shows the ESI-MS spectra of reduced cTPN-Q6f. The signals at 484.8 Da and 581.3 Da correspond to the [M+6H]6+ and [M+5H]5+ charge states, respectively. B) Shows the ESI-MS spectra of oxidized cTPN-Q6f. The signals at 580.8 Da, 725.3 Da and 484.0 Da correspond to the [M+5H]5+, [M+4H]4+, and [M+6H]6+ charge states, respectively.

54

3.7 Purification of Oxidized cTPN-Q6f Dy650 Once cTPN-Q6f was oxidized and purified, fluorescent labeling with DyLight® 650 was attempted using the Staudinger reaction (refer to Section 2.10). The sample of cTPN-Q6f that went through the Staudinger reaction was called cTPN-Q6f Dy650. The analytical HPLC chromatogram in Figure 35A show PDA peaks at 35.1 minutes and at 39.0 minutes, while Figure 35B shows a fluorescence peak at approximately 40.8 minutes. The PDA peak at 35.1 minutes exactly matches the retention time of oxidized cTPN-Q6f (Figure 33), therefore this peak is most likely comprised of unlabeled peptidic material. This is further corroborated by the lack of a corresponding fluorescence peak in Figure 35B. Since the fluorescence peak in Figure 35B occurs at 40.8 minutes, it most likely corresponds to the minor PDA peak at 39.0 minutes, indicating that the conjugation reaction did not go to completion, leaving the majority of the peptide in its original, unlabeled state. This is also evidenced by the ESI-MS spectra shown in Figure 36, which is shows the [M+4H]4+ and [M+5H]5+ charge states of unlabeled cTPN-Q6f. The difference in retention time between the fluorescence peak and the PDA peak is due to the travel distance between the two detectors, this distance is not great enough to result in a 5-minute delay between signals. In light of these data, it is clear that the protocol for the Staudinger reaction needs to be modified in order to have a chance of successfully label this larger, cyclic peptide. In order to get better yields of labeled peptide the reaction ratio should be increased from a 1:1 ratio of peptide to fluorophore to 2:1. This excess of peptidic material in the reaction mixture would push the reaction further to completion by ensuring that all of the fluorophore molecules react.

55

A) Rt: 35.1 minutes 80

A214

% MeCN

Rt: 39.0 minutes 5

B)

Rt: 40.8 minutes 80

Em674

% MeCN 5

Figure 35. A) Shows the analytical HPLC/UV chromatogram of cTPN-Q6f Dy650 detected by a PDA detector at 214nm. The main peak elutes at 35.1 minutes and the second peak at 39.0 minutes B) Shows the analytical HPLC/Fluorescence chromatogram of cTPN-Q6f Dy650 detected by Scanning Fluoroscence (Ex/Em= 646/674 nm) in tandem with the PDA detector. The elution time is approximately 40.8 minutes, and most likely corresponds to the PDA peak at 39.0 minutes. The lag in retention time is due to the travel distance between the two detectors.

56

Relative Intensity

580.9 [M+5H]5+ 726.0 [M+4H]4+

m/z, Da Figure 36. Shows the ESI-MS spectra of cTPN-Q6f Dy650. The signals at 580.9 Da and 726.0 Da correspond to the [M+5H]5+ and [M+4H]4+ charge states, respectively, of unlabeled cTPNQ6f The earlier eluting peak in Figure 35A (at 35.1 minutes) has the exact same retention time as oxidized cTPN-Q6f (Figure 33), before going through the Staudinger reaction. This indicates that major peak in Figure 35A corresponds to unlabeled peptidic material. This hypothesis is further corroborated by the appearance of signals at 580.9 Da and 726.0 Da in the ESI-MS spectra of this material (Figure 36), which correspond to the [M+5H]5+ and [M+4H]4+ charge states, respectively, of oxidized cTPN-Q6f. Lastly, the second major peak in Figure 35A at 39.0 minutes has a corresponding fluorescence peak in Figure 35B at 40.8 minutes. The reason for the delay in the retention time is due to the travel distance between the PDA and fluorescence detectors. This is not so great so as to require 5 full minutes of travel time, thereby making it highly unlikely that this fluorescence peak correspond to the PDA peak at 35.1 minutes. A control run containing the fluorophore alone also showed a retention time of 40-45 minutes (data not shown).

57

3.8 Discussion of Overall Results The failed cyclizations of cTPN-Q5 and cTPN-Q7 along with the successful cyclization of cTPN-Q6 clearly validate the results published by Clark et al. (Figure 6), by showing that this approximately linear relationship between termini distance and linker length does in fact hold for peptide toxins outside of Conus spp. cTPN-Q5 likely had a linker that was too short to span the gap between the N- and C-termini leading the termini being unable to react with each and resulting in a failed cyclization reaction. Once again, the most likely explanation for the failed cyclization of cTPN-Q7 is the length of its linker. Since this linker was one amino acid longer than predicted resulting in an N-terminus that was too flexible to react effectively with a Cterminus that was already extremely hindered by bulky/charged amino acids, such as lysine and tryptophan. cTPN-Q6, which contained the predicted optimal linker length was able to perfectly balance these two factors (length and flexibility) in order to successfully join its N- and Ctermini in a stable peptide bond, producing a backbone cyclized molecule. This expanded application will allow future researchers to more reliably apply these data to the cyclization of other peptide toxins, which could in turn result in the development of new and exciting stable drug leads for previously inaccessible molecular targets. The in vivo bioassays performed on the cyclic isomers of TPN-Q6 indicate that there is a preferred disulfide conformation for this peptide. This is indicated by the ability of Isomer 1 to exhibit activity in fish at all three administered dosages, whereas Isomer 2 was only able to demonstrated minimal activity, which could have been the result of the sample being contaminated with small amounts of Isomer 1. This difference in disulfide connectivity would ultimately lead to differences in the overall 3D structure of the molecule and therefore differences in bioactivity. Since this oxidation was carried out in a non-selective manner, the exact connectivity of the bioactive isomer is not known. Determining this connectivity and comparing it to that of native TPN-Q will be an integral first piece of information in order to establish whether cyclic peptides retain the same disulfide connectivity of their linear counterparts. Furthermore, the fact that linear TPN-Q6 did not show any effect while the both of the cyclic isomers were able to speaks to the ability of cyclization to restore the threedimensional structure, and therefore the bioactivity, of this peptide. The linear version had a 6 amino acids linker sequence on the N-terminus that was not attached in a peptide bond, resulting

58

in it being free to interact with the solution, leading to perturbations in the 3D structure and diminished bioactivity. The non-selective oxidation of cTPN-Q6 appeared to yield 2 isomeric forms. Since this peptide has 4 cysteines, we would expect it to be able to form 2 disulfide bonds in 3 different configurations: globular, beaded and ribbon, with each configuration resulting in a distinct HPLC peak. However, instead we only see 2 peaks that correspond to isomers. This could indicate that the third configuration is highly unstable, and therefore thermodynamically unfavorable, causing it to be very short-lived throughout the oxidation process and leaving an undetectable amount left over by the end of the reaction. Another possibility if that the two isomers observed from oxidation are topological in nature, instead of structural. This would mean that while their disulfide connectivity is the same, the way in which the bonds are arranged through space is different, resulting in differences in the overall bioactivity of the peptide. Finally, the unsuccessful labelling of cTPN-Q6f with the DyLight® 650 fluorophore indicates that the azide active site was sterically hindered by the surrounded bulky amino acids. This likely made the conjugation reaction highly unfavorable, leading to poor overall yields of the desired products. This problem can be remedied by the inclusion of an additional carbon chain (LC), which will function to extend the reactive azide group farther away from the surface of the peptide thereby reducing, or negating, the steric effects of the other amino acid side-chains and making the labeling reaction much more favorable. This technique has already been successfully demonstrated by a previous lab member to add the same DyLight® 650 fluorophore to iberiotoxin and also by Bingham et al. 2006 wherein iberiotoxin could only be successfully biotinylated through the use of an LC group.

59

cTPN-Q5 AGAGAALCNCNRIIIPHQCWKKCGKK

TPN-Q ALCNCNRIIIPHQCWKKCGKK

Observed Parent Mass: 2784.37 Da Expected Parent Mass: 2766.37 Da δ Mass: +18 Da cTPN-Q7 AGAGAGAALCNCNRIIIPHQCWKKCGKK

TPN-Q ALCNCNRIIIPHQCWKKCGKK

Observed Parent Mass: 2912.5 Da Expected Parent Mass: 2894.5 Da δ Mass: +18 Da TPN-Q ALCNCNRIIIPHQCWKKCGKK

cTPN-Q6 GAGAGAALCNCNRIIIPHQCWKKCGKK

Linear TPN-Q6 GAGAGAALCNCNRIIIPHQCWKKCGKK Expected Parent Mass: 2837.42 Da Observed Parent Mass: 2837.42 Da δ Mass: +0 Da

Expected Parent Mass: 2823.42 Da Observed Parent Mass: 2823.42 Da δ Mass: +0 Da

N3 cTPN-Q6f GAGKGAALCNCNRIIIPHQCWKKCGKK Expected Parent Mass: 2900.5 Da Observed Parent Mass: 2900.5 Da δ Mass: +0 Da Dy650 N cTPN-Q6f Dy650 GAGKGAALCNCNRIIIPHQCWKKCGKK Expected Parent Mass: 4242.89 Da Observed Parent Mass: 2900.5 Da δ Mass: 1342.39 Da Figure 37. Shows a flow chart summarizing the peptides produced during this project with their associated expected and observed parent masses and the mass difference between the two.

60

CHAPTER 4 FUTURE DIRECTIONS 4.1 Future Work Now that a cyclic version of TPN-Q has been successfully produced and has been shown to have bioactivity, the next steps would be to determine the disulfide connectivity of the two putative isomers that were produced during oxidation. This could be done through reduction/alkylation protocols in which the reduced cyclic peptide is completely oxidized in a strong oxidizing environment in order to ensure disulfide bond formation. These disulfide bonds are then carefully broken one at a time utilizing different concentrations of a thiol reducing agent, such as tris(2-carboxyethyl)phosphine (TCEP) or DTT. Then, an alkylating agent is introduced after the breaking of each disulfide in order to prevent them from reforming. At each stage of this process, small aliquots are analyzed via HPLC and ESI-MS and compared to fully reduced and fully oxidized samples in order to monitor the progress of the reaction. Even though the disulfide connectivity of native TPN-Q is known, it is unknown whether this native connectivity will be bioactive after the molecule has been cyclized. Since cyclization invariably distorts the 3-dimensional structure of the peptide, this could have led to a different connectivity, other than the native one, giving rise to the bioactivity that was observed. Furthermore, since only one of the isomers was able to demonstrate bioactivity on fish, it would be interesting to see if that molecule has adopted the native connectivity. It is also possible, as has been demonstrated by other members of the Bingham laboratory, that the isomers formed by TPN-Q are topological as opposed to structural. This would mean that their disulfide connectivities would be the same, but the way in which the bonds are formed through space would be different, potentially leading to differences in activity. Furthermore, a fluorescent cyclic probe of TPN-Q could be used in staining assays in order to assess the distribution and localization of the molecular targets of TPN-Q, namely ROMK1 and GIRK1/4. This information could potentially lead to new druggable targets for the treatment of medically relevant diseases, such as Bartter’s syndrome. Samples of linear TPN-Q6, cTPN-Q6 Isomer 1 and cTPN-Q6 Isomer 2 have been sent to the Vanderbilt University School of Medicine in order for collaborators there to perform electrophysiology assays using the ROMK1 and GIRK1/4 channels. Here again we would expect 61

to see similar activity as we did in the whole animal bioassays on X. helleri fish, namely that, for the reason stated previously, Isomer 1 will be bioactive while Isomer 2 and the linear form will not. We would expect to see similar levels of activity to the unmodified peptide (TPN-Q), since the very nature of backbone cyclization allows the peptide to adopt its native 3D conformation, thereby allowing it to have comparable levels of activity while demonstrating increased resistance to heat, pH and exopeptidases. Showing this increased resistance through in vitro enzymatic degradation assays, would also be a strong sign that this cyclic peptide could be used an orally bioavailable drug.

62

REFERENCES Argiolas A, Pisano JJ. 1983. Facilitation of phospholipase a2 activity by mastoparans, a new class of mast-cell degranulating peptides from wasp venom. Journal of Biological Chemistry. 258(22):3697-3702. Bhave G, Chauder BA, Liu W, Dawson ES, Kadakia R, Nguyen TT, Lewis LM, Meiler J, Weaver CD, Satlin LM et al. 2011. Development of a selective small-molecule inhibitor of kir1.1, the renal outer medullary potassium channel. Molecular Pharmacology. 79(1):42-50. Bichet D, Haass FA, Jan LY. 2003. Merging functional studies with structures of inward-rectifier k+ channels. Nature Reviews Neuroscience. 4(12):957-967. Bingham JP, Bian S, Tan ZY, Takacs Z, Moczydlowski E. 2006. Synthesis of a biotin derivative of iberiotoxin: Binding interactions with streptavidin and the bkca2+-activated k+ channel expressed in a human cell line. Bioconjugate Chemistry. 17(3):689-699. Blond A, Peduzzi J, Goulard C, Chiuchiolo MJ, Barthelemy M, Prigent Y, Salomon RA, Farias RN, Moreno F, Rebuffat S. 1999. The cyclic structure of microcin j25, a 21-residue peptide antibiotic from escherichia coli. European Journal of Biochemistry. 259(3):747755. Boim MA, Ho K, Shuck ME, Bienkowski MJ, Block JH, Slightom JL, Yang YH, Brenner BM, Hebert SC. 1995. Romk inwardly rectifying atp-sensitive k+ channel .2. Cloning and distribution of alternative forms. American Journal of Physiology-Renal Fluid and Electrolyte Physiology. 268(6):F1132-F1140. Bond CT, Maylie J, Adelman JP. 1999. Small-conductance calcium-activated potassium channels. In: Rudy B, Seeburg P, editors. Molecular and functional diversity of ion channels and receptors. p. 370-378. Camarero JA, Fushman D, Sato S, Giriat I, Cowburn D, Raleigh DP, Muir TW. 2001. Rescuing a destabilized protein fold through backbone cyclization. Journal of Molecular Biology. 308(5):1045-1062. Camarero JA, Pavel J, Muir TW. 1998. Chemical synthesis of a circular protein domain: Evidence for folding-assisted cyclization. Angewandte Chemie-International Edition. 37(3):347-349.

63

Candia S, Garcia ML, Latorre R. 1992. Mode of action of iberiotoxin, a potent blocker of the large conductance ca(2+)-activated k+ channel. Biophysical Journal. 63(2):583-590. Carbone E, Wanke E, Prestipino G, Possani LD, Maelicke A. 1982. Selective blockage of voltage-dependent k+ channels by a novel scorpion toxin. Nature. 296(5852):90-91. Castle NA, Haylett DG, Jenkinson DH. 1989. Toxins in the characterization of potassium channels. Trends in Neurosciences. 12(2):59-65. Chan W, White P. 2000. Fmoc solid phase peptide synthesis a practical approach. Oxford University Press: Oxford. Chanchevalap S, Yang ZJ, Cui NG, Qu ZQ, Zhu GY, Liu CX, Giwa LR, Abdulkadir L, Jiang C. 2000. Involvement of histidine residues in proton sensing of romk1 channel. Journal of Biological Chemistry. 275(11):7811-7817. Chandrudu S, Simerska P, Toth I. 2013. Chemical methods for peptide and protein production. Molecules. 18(4):4373-4388. Chen WYJ, Abatangelo G. 1999. Functions of hyaluronan in wound repair. Wound Repair and Regeneration. 7(2):79-89. Cheneval O, Schroeder CI, Durek T, Walsh P, Huang YH, Liras S, Price DA, Craik DJ. 2014. Fmoc-based synthesis of disulfide-rich cyclic peptides. Journal of Organic Chemistry. 79(12):5538-5544. Cheong A, Li J, Sukumar P, Kumar B, Zeng F, Riches K, Munsch C, Wood IC, Porter KE, Beech DJ. 2011. Potent suppression of vascular smooth muscle cell migration and human neointimal hyperplasia by k(v)1.3 channel blockers. Cardiovascular Research. 89(2):282289. Clark MA, Conway TM, Shorr RGL, Crooke ST. 1987. Identification and isolation of a mammalian protein which is antigenically and functionally related to the phospholipasea2 stimulatory peptide melittin. Journal of Biological Chemistry. 262(9):4402-4406. Clark RJ, Daly NL, Craik DJ. 2006a. Structural plasticity of the cyclic-cystine-knot framework: Implications for biological activity and drug design. Biochemical Journal. 394:85-93. Clark RJ, Fischer H, Dempster L, Daly NL, Rosengren KJ, Nevin ST, Meunier FA, Adams DJ, Craik DJ. 2005. Engineering stable peptide toxins by means of backbone cyclization: Stabilization of the alpha-conotoxin mii. Proceedings of the National Academy of Sciences of the United States of America. 102(39):13767-13772. 64

Clark RJ, Fischer H, Nevin ST, Adams DJ, Craik DJ. 2006b. The synthesis, structural characterization, and receptor specificity of the alpha-conotoxin vc1.1. Journal of Biological Chemistry. 281(32):23254-23263. Clark RJ, Jensen J, Nevin ST, Callaghan BP, Adams DJ, Craik DJ. 2010. The engineering of an orally active conotoxin for the treatment of neuropathic pain. Angewandte ChemieInternational Edition. 49(37):6545-6548. Cole AM, Hong T, Boo LM, Nguyen T, Zhao CQ, Bristol G, Zack JA, Waring AJ, Yang OO, Lehrer RI. 2002. Retrocyclin: A primate peptide that protects cells from infection by tand m-tropic strains of hiv-1. Proceedings of the National Academy of Sciences of the United States of America. 99(4):1813-1818. Craik DJ. 2006. Chemistry - seamless proteins tie up their loose ends. Science (New York, NY). 311(5767):1563-1564. D'Souza C, Henriques ST, Wang CK, Craik DJ. 2014. Structural parameters modulating the cellular uptake of disulfide-rich cyclic cell-penetrating peptides: Mcoti-ii and sfti-1. European Journal of Medicinal Chemistry. 88:10-18. Daly NL, Craik DJ. 2009. Design and therapeutic applications of cyclotides. Future Medicinal Chemistry. 1(9):1613-1622. Dawson PE, Kent SB. 2000. Synthesis of native proteins by chemical ligation*. Annual review of biochemistry. 69(1):923-960. Dawson PE, Muir TW, Clarklewis I, Kent SBH. 1994. Synthesis of proteins by native chemical ligation. Science (New York, NY). 266(5186):776-779. Dennis EA. 1994. Diversity of group types, regulation, and function of phospholipase a2. Journal of Biological Chemistry. 269(18):13057-13060. Faber ESL, Sah P. 2007. Functions of sk channels in central neurons. Clinical and Experimental Pharmacology and Physiology. 34(10):1077-1083. Fretzayas A, Gole E, Attilakos A, Daskalaki A, Nicolaidou P, Papadopoulou A. 2013. Expanding the spectrum of genetic mutations in antenatal bartter syndrome type ii. Pediatrics international : official journal of the Japan Pediatric Society. 55(3):371-373. Gao YD, Garcia ML. 2003. Interaction of agitoxin2, charybdotoxin, and iberiotoxin with potassium channels: Selectivity between voltage-gated and maxi-k channels. ProteinsStructure Function and Genetics. 52(2):146-154. 65

Garcia ML, Garciacalvo M, Hidalgo P, Lee A, Mackinnon R. 1994. Purification and characterization of 3 inhibitors of voltage-dependent k+ channels from leiurus quinquestriatus var. Hebraeus venom. Biochemistry. 33(22):6834-6839. Garciacalvo M, Leonard RJ, Novick J, Stevens SP, Schmalhofer W, Kaczorowski GJ, Garcia ML. 1993. Purification, characterization, and biosynthesis of margatoxin, a component of centruroides margaritatus venom that selectively inhibits voltage-dependent potassium channels. Journal of Biological Chemistry. 268(25):18866-18874. Giepmans BNG, Adams SR, Ellisman MH, Tsien RY. 2006. Review - the fluorescent toolbox for assessing protein location and function. Science (New York, NY). 312(5771):217224. Goldstein SAN, Miller C. 1993. Mechanism of charybdotoxin block of a voltage-gated k+ channel. Biophysical Journal. 65(4):1613-1619. Gololobov YG, Kasukhin LF. 1992. Rencent advances in the staudinger reaction. Tetrahedron. 48(8):1353-1406. Gonzalez C, Langdon GM, Bruix M, Galvez A, Valdivia E, Maqueda M, Rico M. 2000. Bacteriocin as-48, a microbial cyclic polypeptide structurally and functionally related to mammalian nk-lysin. Proceedings of the National Academy of Sciences of the United States of America. 97(21):11221-11226. Grant G. 2002. Synthetic peptides: A user’s guide. Oxford University Press, Oxford. Gunasekera S, Foley FM, Clark RJ, Sando L, Fabri LJ, Craik DJ, Daly NL. 2008. Engineering stabilized vascular endothelial growth factor-a antagonists: Synthesis, structural characterization, and bioactivity of grafted analogues of cyclotides. Journal of Medicinal Chemistry. 51(24):7697-7704. Gustafson KR, McKee TC, Bokesch HR. 2004. Anti-hiv cyclotides. Current Protein & Peptide Science. 5(5):331-340. Habermann E. 1972. Bee and wasp venoms. Science (New York, NY). 177(4046):314-322. Habermann E. 1984. Apamin. Pharmacology & Therapeutics. 25(2):255-270. Hackeng TM, Griffin JH, Dawson PE. 1999. Protein synthesis by native chemical ligation: Expanded scope by using straightforward methodology. Proceedings of the National Academy of Sciences of the United States of America. 96(18):10068-10073.

66

Heginbotham L, Lu Z, Abramson T, Mackinnon R. 1994. Mutations in the k+ channel signature sequence. Biophysical Journal. 66(4):1061-1067. Hellinger R, Koehbach J, Puigpinós A, Clark RJ, Tarragó T, Giralt E, Gruber CW. 2015. Inhibition of human prolyl oligopeptidase activity by the cyclotide psysol 2 isolated from psychotria solitudinum. Journal of Natural Products. 78(5):1073-1082. Hernandez JF, Gagnon J, Chiche L, Nguyen TM, Andrieu JP, Heitz A, Hong TT, Pham TTC, Nguyen DL. 2000. Squash trypsin inhibitors from momordica cochinchinensis exhibit an atypical macrocyclic structure. Biochemistry. 39(19):5722-5730. Hille B. 2001. Ion channels of excitable membranes. Sinauer Sunderland, MA. Ho K, Nichols CG, Lederer WJ, Lytton J, Vassilev PM, Kanazirska MV, Hebert SC. 1993. Cloning and expression of an inwardly rectifying atp-regulated potassium channel. Nature. 362(6415):31-38. Imredy JP, Chen CF, MacKinnon R. 1998. A snake toxin inhibitor of inward rectifier potassium channel romk1. Biochemistry. 37(42):14867-14874. Information sheet 4: What bees eat. [accessed 2016 February 13]. http://ag.arizona.edu/pubs/insects/ahb/inf4.html. Isomoto S, Kondo C, Kurachi Y. 1997. Inwardly rectifying potassium channels: Their molecular heterogeneity and function. Japanese Journal of Physiology. 47(1):11-39. Jennings C, West J, Waine C, Craik D, Anderson M. 2001. Biosynthesis and insecticidal properties of plant cyclotides: The cyclic knotted proteins from oldenlandia affinis. Proceedings of the National Academy of Sciences of the United States of America. 98(19):10614-10619. Jiang C, Qu ZQ, Xu HX. 2002. Gating of inward rectifier k+ channels by proton-mediated interactions of intracellular protein domains. Trends in Cardiovascular Medicine. 12(1):513. Jin WL, Klem AM, Lewis JH, Lu Z. 1999. Mechanisms of inward-rectifier k+ channel inhibition by tertiapin-q. Biochemistry. 38(43):14294-14301. Jin WL, Lu Z. 1998. A novel high-affinity inhibitor for inward-rectifier k(+) channels. Biochemistry. 37(38):13291-13299. Jin WL, Lu Z. 1999. Synthesis of a stable form of tertiapin: A high-affinity inhibitor for inwardrectifier k+ channels. Biochemistry. 38(43):14286-14293. 67

Johnsson N, Johnsson K. 2007. Chemical tools for biomolecular imaging. Acs Chemical Biology. 2(1):31-38. Kaiser E, Colescott RL, Bossinger CD, Cook PI. 1970. Color test for detection of free terminal amino groups in the solid-phase synthesis of peptides. Anal Biochem. 34(2):595-598. Kandel ER, Schwartz JH, Jessell TM. 2000. Principles of neural science. McGraw-hill New York. Koburova KL, Michailova SG, Shkenderov SV. 1985. Further investigation on the antiinflammatory properties of adolapin--bee venom polypeptide. Acta physiologica et pharmacologica Bulgarica. 11(2):50-55. Krissinel E, Henrick K. 2007. Inference of macromolecular assemblies from crystalline state. Journal of Molecular Biology. 372(3):774-797. Kubo Y, Adelman JP, Clapham DE, Jan LY, Karschin A, Kurachi Y, Lazdunski M, Nichols CG, Seino S, Vandenberg CA. 2005. International union of pharmacology. Liv. Nomenclature and molecular relationships of inwardly rectifying potassium channels. Pharmacological Reviews. 57(4):509-526. Leandro LF, Mendes CA, Casemiro LA, Vinholis AHC, Cunha WR, De Almeida R, Martins CHG. 2015. Antimicrobial activity of apitoxin, melittin and phospholipase a(2) of honey bee (apis mellifera) venom against oral pathogens. Anais Da Academia Brasileira De Ciencias. 87(1):147-155. Lewis LM, Bhave G, Chauder BA, Banerjee S, Lornsen KA, Redha R, Fallen K, Lindsley CW, Weaver CD, Denton JS. 2009. High-throughput screening reveals a small-molecule inhibitor of the renal outer medullary potassium channel and kir7.1. Molecular Pharmacology. 76(5):1094-1103. Littleton JT, Ganetzky B. 2000. Ion channels and synaptic organization: Analysis of the drosophila genome. Neuron. 26(1):35-43. Long SB, Campbell EB, MacKinnon R. 2005. Crystal structure of a mammalian voltagedependent shaker family k+ channel. Science (New York, NY). 309(5736):897-903. Lovelace ES, Armishaw CJ, Colgrave ML, Wahlstrom ME, Alewood PF, Daly NL, Craik DJ. 2006. Cyclic mria: A stable and potent cyclic conotoxin with a novel topological fold that targets the norepinephrine transporter. Journal of Medicinal Chemistry. 49(22):65616568. 68

Lu Z, MacKinnon R. 1997. Purification, characterization, and synthesis of an inward-rectifier k+ channel inhibitor from scorpion venom. Biochemistry. 36(23):6936-6940. Luckett S, Garcia RS, Barker JJ, Konarev AV, Shewry PR, Clarke AR, Brady RL. 1999. Highresolution structure of a potent, cyclic proteinase inhibitor from sunflower seeds. Journal of Molecular Biology. 290(2):525-533. Martinez-Bueno M, Maqueda M, Galvez A, Samyn B, Vanbeeumen J, Coyette J, Valdivia E. 1994. Determination of the gene sequence and the molecular-structure of the enterococcal peptide antibiotic as-48. Journal of Bacteriology. 176(20):6334-6339. Meier J, Theakston RDG. 1986. Approximate ld50 determinations of snake-venoms using 8 to 10 experimental-animals. Toxicon. 24(4):395-401. Merrifield RB. 1963. Solid phase peptide synthesis. I. The synthesis of a tetrapeptide. Journal of the American Chemical Society. 85(14):2149-2154. Monroe EW, Daly AF, Shalhoub RF. 1997. Appraisal of the validity of histamine-induced wheal and flare to predict the clinical efficacy of antihistamines. Journal of Allergy and Clinical Immunology. 99(2):S798-S806. Muir TW, Dawson PE, Kent SB. 1997. Protein synthesis by chemical ligation of unprotected peptides in aqueous solution. Methods in enzymology. 289:266-298. Naranjo D, Miller C. 1996. A strongly interacting pair of residues on the contact surface of charybdotoxin and a shaker k+ channel. Neuron. 16(1):123-130. Newton P, Harrison P, Clulow S. 2008. A novel method for determination of the affinity of protein: Protein interactions in homogeneous assays. Journal of Biomolecular Screening. 13(7):674-682. Nishida M, MacKinnon R. 2002. Structural basis of inward rectification: Cytoplasmic pore of the g protein-gated inward rectifier girk1 at 1.8 angstrom resolution. Cell. 111(7):957-965. Ookata K, Tojo A, Suzuki Y, Nakamura N, Kimura K, Wilcox CS, Hirose S. 2000. Localization of inward rectifier potassium channel kir7.1 in the basolateral membrane of distal nephron and collecting duct. Journal of the American Society of Nephrology. 11(11):1987-1994. Possani LD, Martin BM, Svendsen IB. 1982. The primary structure of noxiustoxin - a k+ channel blocking peptide, purified from the venom of the scorpion centruroides noxius. Carlsberg Research Communications. 47(5):285-289. 69

Puricelli E, Bettinelli A, Borsa N, Sironi F, Mattiello C, Tammaro F, Tedeschi S, Bianchetti MG. 2010. Long-term follow-up of patients with bartter syndrome type i and ii. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association. 25(9):2976-2981. Qu ZQ, Yang ZJ, Cui NR, Zhu GY, Liu CX, Xu HX, Chanchevalap S, Shen WZ, Wu JP, Li YJ et al. 2000. Gating of inward rectifier k+ channels by proton-mediated interactions of nand c-terminal domains. Journal of Biological Chemistry. 275(41):31573-31580. Rader K, Wildfeuer A, Wintersberger F, Bossinger P, Mucke HW. 1987. Characterization of bee venom and its main components by high-performance liquid-chromatography. Journal of Chromatography. 408:341-348. Rang HP RJ, Flower R, Henderson G. 2015. Noradrenergic transmission. Rang & dale's pharmacology. 8th ed. p. 177–196. Saether O, Craik DJ, Campbell ID, Sletten K, Juul J, Norman DG. 1995. Elucidation of the primary and 3-dimensional structure of the uterotonic polypeptide kalata b1. Biochemistry. 34(13):4147-4158. Sarin VK, Kent SBH, Tam JP, Merrifield RB. 1981. Quantitative monitoring of solid-phase peptide-synthesis by the ninhydrin reaction. Analytical Biochemistry. 117(1):147-157. Saxon E, Bertozzi CR. 2000. Cell surface engineering by a modified staudinger reaction. Science (New York, NY). 287(5460):2007-2010. Schulte U, Hahn H, Wiesinger H, Ruppersberg JP, Fakler B. 1998. Ph-dependent gating of romk (kir1.1) channels involves conformational changes in both n and c termini. Journal of Biological Chemistry. 273(51):34575-34579. Schwalbe RA, Bianchi L, Accili EA, Brown AM. 1998. Functional consequences of romk mutants linked to antenatal bartter's syndrome and implications for treatment. Human Molecular Genetics. 7(6):975-980. Seeman P. 2010. Historical overview: Introduction to the dopamine receptors. In: Neve AK, editor. The dopamine receptors. Totowa, NJ: Humana Press. p. 1-21. Shaner NC, Steinbach PA, Tsien RY. 2005. A guide to choosing fluorescent proteins. Nature Methods. 2(12):905-909. Shkenderov S, Koburova K. 1982. Adolapin - a newly isolated analgetic and anti-inflammatory polypeptide from bee venom. Toxicon. 20(1):317-321. 70

Sitges M, Possani LD, Bayon A. 1986. Noxiustoxin, a short-chain toxin from the mexican scorpion centruroides noxius, induces transmitter release by blocking k+ permeability. Journal of Neuroscience. 6(6):1570-1574. Slesinger PA. 2001. Ion selectivity filter regulates local anesthetic inhibition of g-protein-gated inwardly rectifying k+ channels. Biophysical Journal. 80(2):707-718. Son DJ, Lee JW, Lee YH, Song HS, Lee CK, Hong JT. 2007. Therapeutic application of antiarthritis, pain-releasing, and anti-cancer effects of bee venom and its constituent compounds. Pharmacology & Therapeutics. 115(2):246-270. Stackman RW, Hammond RS, Linardatos E, Gerlach A, Maylie J, Adelman JP, Tzounopoulos T. 2002. Small conductance ca2+-activated k+ channels modulate synaptic plasticity and memory encoding. Journal of Neuroscience. 22(23):10163-10171. Staudinger H, Meyer J. 1919. Über neue organische phosphorverbindungen iii. Phosphinmethylenderivate und phosphinimine. Helvetica Chimica Acta. 2(1):635-646. Tang YQ, Yuan J, Osapay G, Osapay K, Tran D, Miller CJ, Ouellette AJ, Selsted ME. 1999. A cyclic antimicrobial peptide produced in primate leukocytes by the ligation of two truncated alpha-defensins. Science (New York, NY). 286(5439):498-502. Thapa P, Espiritu MJ, Cabalteja C, Bingham JP. 2014a. The emergence of cyclic peptides: The potential of bioengineered peptide drugs. International Journal of Peptide Research and Therapeutics. 20(4):545-551. Thapa P, Zhang RY, Menon V, Bingham JP. 2014b. Native chemical ligation: A boon to peptide chemistry. Molecules. 19(9):14461-14483. Thongyoo P, Roque-Rosell N, Leatherbarrow RJ, Tate EW. 2008. Chemical and biomimetic total syntheses of natural and engineered mcoti cyclotides. Organic & Biomolecular Chemistry. 6(8):1462-1470. Tossi A, Sandri L, Giangaspero A. 2000. Amphipathic, alpha-helical antimicrobial peptides. Biopolymers. 55(1):4-30. Trabi M, Craik DJ. 2002. Circular proteins - no end in sight. Trends in Biochemical Sciences. 27(3):132-138. White J, Meier J. 1995. Handbook of clinical toxicology of animal venoms and poisons. CRC Press.

71

Wysocki LM, Lavis LD. 2011. Advances in the chemistry of small molecule fluorescent probes. Current Opinion in Chemical Biology. 15(6):752-759. Xu D, Zhang Y. 2012. Ab initio protein structure assembly using continuous structure fragments and optimized knowledge-based force field. Proteins-Structure Function and Bioinformatics. 80(7):1715-1735. Xu XB, Nelson JW. 1993. Solution structure of tertiapin determined using nuclear-magneticresonance and distance geometry. Proteins-Structure Function and Genetics. 17(2):124137. Yanagisawa T, Ishii R, Fukunaga R, Kobayashi T, Sakamoto K, Yokoyama S. 2008. Multistep engineering of pyrrolysyl-trna synthetase to genetically encode n(epsilon)-(oazidobenzyloxycarbonyl) lysine for site-specific protein modification. Chemistry & Biology. 15(11):1187-1197. Yang S, Carrasquer G. 1997. Effect of melittin on ion transport across cell membranes. Acta Pharmacologica Sinica. 18(1):3-5. Yano H, Philipson LH, Kugler JL, Tokuyama Y, Davis EM, Lebeau MM, Nelson DJ, Bell GI, Takeda J. 1994. Alternative splicing of human inwardly rectifying k+ channel romk1 messenger-rna. Molecular Pharmacology. 45(5):854-860. Yi BA, Lin YF, Jan YN, Jan LY. 2001. Yeast screen for constitutively active mutant g proteinactivated potassium channels. Neuron. 29(3):657-667.

72