˘ ˇ ˆ - DiVA portal396748/FULLTEXT01.pdf · 2.5 Chemical and biomimetic engineering ..... 28 2.6...

��

��

��

��

��

��

�!"#��$� %�%&#'�

(!!)��*+��*��,(!&)��-.��++/�.00��+1�2��2�21123�4��/+-/�

�� !��"��#$$��$�%$"�&��!��&��&'!��!��( ��&�'!��)*�+!��,��-��!*

��.��+�/*��#$$*�-��&��!�� 0��1�� ,�2�&��*��!��3��!��3��3��*�.�� *�� $�4*�77 ��*� ��*�53�6�47804$0""�08##�0"*

��0�� 2�� ,!��!� �,� ��&�� !�� 2�� &�� !�� !�� !��&�� !�� !��!�� !�� &�� !�� &� �� *� +!�� !�� ,� ��!�� &� �!�� &��,�2� &�� !�� !�� !�� !�� 2�� *

3��!��&��!��&��&��&��2�� ,�� 9�*� ��,��&� �!�� !�� ,�� !�� ,��:��!��&��;��*�+!��,��&�� ;�� 0�� 0�� &&�� 9��<��*�+��&��!��!��!��!��&&�� 2��&� ��=�� &�� !�� !�� &��!,��&�� &��%� �>�� !�� !�� ,��:��&��&&��&��*�+!��,��&��&��<��&��!,��,!��!��!��>��!��!�� &�� ,��!� �0�� ,0� �� !��0��&�� ,!�� !��,��!��!��0��&��!��&&��*

��!��!��&��&&��&��&�� ,�� &� �� (�*�*�� )�� 0!��;��,��0��,��!��*�+!��!��,!��!�!��;��&�� ,��!� �� !,�� !�� &��*� +!��&��!��,��!��&��"��!��&�$�80&��!��&��*� ��!��$��$��6�?��&��!��!��!,��!��!��,��&��*

� ��2��6�?��!��&��!��0��

� �� !"� #�� $ �� #��#�%&'()#�� #��*+(',-.��#�� /

@�+��!��/��.��#$$

5336�$A"$0A$4�53�6�47804$0""�08##�0"��%�%��%��%��0$�"7�4�(!��%BB��*2�*��B��C��D��%�%��%��%��0$�"7�4)

To my mother and father

List of papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Aboye, T. L., Johan, K. R., Gunasekera, S., Bruhn, J. B., El-

Seedi, H., Göransson, U. (2011) Discovery, synthesis, and structural determination of a toxin-like disulfide-rich peptide from the cactus Trichocereus pachanoi. Manuscript.

II Aboye, T. L., Clark, R. J., Craik, D. J., Göransson, U. (2008) Ultra-stable peptide scaffolds for protein engineering: Synthsis and folding of the circular cystine knotted cyclotide cycloviola-cin O2. ChemBioChem, 9(1): 103–113

III Park, S., Gunasekera, S., Aboye, T. L., Göransson, U. (2010) An efficient approach for the total synthesis of cyclotides by microwave assisted Fmoc-SPPS. International Journal of Pep-tide Research and Therapeutics, 16(3): 167-176

IV Aboye, T. L., Clark, R. J., Burman, R., Roig, M. B., Craik, D. J., Göransson, U. (2011) Interlocking disulfides in circular pro-teins. Toward efficient oxidative folding of cyclotides. Antioxi-dants & Redox Signaling, 14(1): 77-86

V Aboye, T. L., Burman, R., Göransson, U. Design, synthesis, structural and biological evaluation of backbone engineered cy-clotides. Manuscript

Reprints were made with permission from the respective publishers.

Contents

1� Introduction ............................................................................................ 13�1.1� Serendipitous discovery of knotted microproteins ......................... 13�1.2� Structure of cystine knotted microproteins ..................................... 14�1.3� Stability and biological activities .................................................... 19�

2� A hybrid of solid and solution phase synthesis of microproteins .......... 20�2.1� Linear chain assembly .................................................................... 20�2.2� Native chemical ligation/cyclization .............................................. 23�2.3� Synthesis of activated precursors for NCL/cyclization .................. 24�2.4� Oxidative folding: locking in the native state ................................. 28�2.5� Chemical and biomimetic engineering ........................................... 28�2.6� Biomolecular engineering ............................................................... 29�

3� Aims of the present study ....................................................................... 31�4� Synthesis and structural determination of a novel disulfide-rich peptide

(Paper I) .................................................................................................. 32�4.1� Discovery of an acyclic disulfide-rich peptide ............................... 32�4.2� Solid phase synthesis and oxidative folding ................................... 33�4.3� Solution NMR structural determination ......................................... 35�

5� Fmoc-based synthesis and oxidative folding of bracelet, cycloviolacin O2 (Papers II and III) ................................................................................... 38�5.1� The bracelet cyclotides ................................................................... 38�5.2� Synthesis of linear chain and protected peptides ............................ 38�5.3� Peptide cyclization using peptidyl α-thioester ............................... 39�5.4� Optimization of oxidative folding conditions ................................. 40�5.5� NMR structural studies and disulfide bond determination ............. 43�5.6� Conclusion ...................................................................................... 44�

6� Folding pathways to lock in the native disulfide bonds (Paper IV) ...... 45 6.1� Oxidative folding of cyclic cystine knot microproteins ................. 45�6.2� Quantification of heterogeneous folding intermediates .................. 45�6.3� The diversity of cyclotide folding pathways .................................. 48�

7� From native to engineered macrocyclic cystine knotted peptides

(Paper V) ................................................................................................. 50

7.1� Backbone engineering of the cyclotide scaffold ............................. 50�7.2� Rational design strategy using backbone spacers ........................... 51�7.3� Synthesis, biological and structural evaluation of mer-cyclotides . 52�7.4� Conclusion ...................................................................................... 58�

8� Concluding Remarks .............................................................................. 59�9� Summary of popular Science ................................................................. 61�10� Acknowledgments ................................................................................ 63�11� References ............................................................................................ 66�

Abbreviations

MeCN acetonitrile

Boc tert-butyloxycarbonyl

CCK cyclic cystine knot

CKM cystine knot microprotein

COSY correlated spectroscopy

CyO2 cycloviolacin O2

Dbz 3,4-diaminobenzoic acid

DCM dichloromethane DIPEA N,N-diisopropylethylamine

DMF N,N-dimethylformamide DMSO dimethylsulfoxide DQF-COSY double quantum filtered correlated spectroscopy

DTT dithiothreitol

E-COSY exclusive correlation spectroscopy

ESI-MS electrospray mass spectrometry

Fmoc 9-fluorenylmethoxycarbonyl

GFK growth factor cystine knot

GSH reduced glutatione

GSSG oxidized glutatione

HATU (2-(7-Aza-1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate)

HBTU O-(1-benzotriazolyl)-1,1,3,3-

tetramethyluroniumhexafluorophosphate

HF hydrogen fluoride

HOBt N-hydroxybenzotriazole ICK inhibitor cystine knot

KB1 kalata B1

MBHA methylbenzhydrylamine MeOH methanol

MPAA 4-mercaptophenylacetic acid

Nbz N-acylbenzimidazolone

NCL native chemical ligation

NEM N-ethylmaleimide

NMR nuclear magnetic resonance

NOESY nuclear Overhauser effect spectroscopy

Pbf 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl

PyBOP benzotriazole-1-yl-oxy-

trispyrrolidinophosphoniumhexafluorophosphate

R fully reduced peptide

RP-HPLC reversed phase high performance liquid chromatography

SPPS solid phase peptide synthesis

t-Bu tert-butyl

TCEP tris(2-carboxyethyl)phosphine

TBTU (2-(1H-Benzotriazole-1-yl)-1,1,3,3-tetramethyluronium tetra-fluoroborate)

TFA trifluoroacetic acid TFE 2,2,2-trifluoroethanol

TIPS triisopropylsilane

TOCSY total correlation spectroscopy

TRIS tris(hydroxymethyl)aminomethane

Trt trityl

2D, 3D two, three-dimensional

1SS, 2SS, 3SS isomers containing One-, two-, or three- disulfide bond

Amino acids residues and polymeric building blocks

13

1 Introduction

1.1 Serendipitous discovery of knotted microproteins Peptides regulate most physiological processes, acting as endocrine signals, neurotransmitters or growth factors. They are used therapeutically in diverse areas including neurology, endocrinology and hematology (Vlieghe, 2010; Edwards, 1999). Due to their advantageous features of high specificity and low toxicity, peptides are considered drugs of the future.

In recent years, naturally occurring cystine-knotted peptides attracted great interest (Sommerhoff, 2010; Gunasekera, 2008b) because of their higher stability, manifested by high resistance to protease degradation, and rigidity (Gunasekera, 2008b; Colgrave and Craik, 2004). Most potential pep-tide drug leads are composed of amino acids in linear chains with open ends that are targets for proteolytic enzymes, leading to short plasma half-life times. This feature, coupled with their poor absorption due to their generally hydrophilic nature, results in poor oral bioavailability, making them unsuita-ble for oral drug therapies (Vagner, 2008; Werle and Bernkop-Schnürch, 2006). Cyclization has been used to reduce susceptibility to degradation of peptide drugs (Clark, 2010) and when combined with oxidative folding, it has been found to stabilize peptides in harsh environments by forming rigid, compact molecules (Clark, 2011; Werle and Bernkop-Schnürch, 2006; Colgrave and Craik, 2004). For example, cyclic cystine-knotted peptides show greater stability than their solely cystine-knotted counterparts (Colgrave and Craik, 2004; Daly and Craik, 2000).

The exceptional properties of such peptides first came to notice through the serendipitous discovery of a bioactive agent in indigenous medicine used in the Congo (Zair) region of Africa (Gran, 1973b). During a Red Cross relief mission in Zaire in the 1960s, the Norwegian physician noted that dur-ing labor, women sipped a decoction made by boiling leaves of the plant Oldenlandia affinis, to accelerate uterine contractions and facilitate child-birth. In the 1970s, the active ingredient was determined to be a peptide, named kalata B1 after the traditional name for the native medicine, 'kalata-kalata' (Gran, 1973c; a). The stability of the peptide during boiling and oral ingestion (and thus resistance to high temperature and gastrointestinal tract enzymes) indicated that it has oral bioavailability, which is unusual for pep-tides and proteins generally. KB1 was also tested in vivo in rats and rabbits(Gran, 2000), confirming that it has similar uterotonic activity to oxy-

14

tocin, the well-known uterine contraction-inducing hormone. Unfortunately, despite its potential utility as therapeutic agent, inherent problems especially high hemolytic activity and unfavorable cardiotoxic effects (Gran, 2000; Gran, 1973b; c), hampered its further drug development. However, the dis-covery of kB1 prompted further studies, and in the 1990s, several other re-lated plant-derived peptides were isolated (Göransson, 1999; Claeson, 1998; Gustafson, 1994b; Witherup, 1994; Schöpke, 1993) and their three dimen-sional (3D) structures solved (Daly, 1999a; Saether, 1995), revealing a simi-lar circular backbone and cysteine content to kB1. As a result, these macro-cyclic peptides considered part of a related family and later named ‘cyclo-tides’ (Craik, 1999). Today cyclotides have been found in members of the Rubiaceae, Vioalaceae, and Cucurbitaceae plant families (Gerlach, 2010; Laura and Craik, 2010; Poth, 2010; Gruber, 2008; Hernandez, 2000).

There are also several acyclic inhibitory cystine-knot peptides, which are remarkably stable compared to non-knotted peptides and have high receptor specificity. Typical examples of these, from Conus peptide families, are ω-conopeptides that inhibit voltage-gated calcium channels (Olivera, 1984), δ-conopeptides that inhibit voltage-gated Na channel inactivation(Shon, 1994), and κ-conopeptides that target voltage-gated K channels (Shon, 1998). The high stability, selective modulation of biomolecular interactions, diverse range of bioactivity and synthetic amenability of cystine-knot microproteins make them attractive choices as scaffolds for peptide-based drug design and biomolecular engineering.

1.2 Structure of cystine knotted microproteins Cystine knot microproteins (CKM) are small disulfide-rich peptides sharing disulfide connectivity pattern of I-IV, II-V, and III-VI associated with β strands (Pallaghy, 1994; McDonald and Hendrickson, 1993). They are divid-ed into two main categories, based on topological features (Craik, 2001; Isaacs, 1995): growth factor cystine knots (GFK), such as nerve growth fac-tor, platet-derived growth factor-BB and transforming growth factor-β2; and inhibitor cystine knots (ICK) (Figure 1A). ICK are often referred to as knot-tins and are commonly associated with a distorted triple-stranded β-sheet structure that is integral to the cystine knot motif. ICK peptides are either acyclic, for example as in ω- and δ-conotoxins, or cyclic, as in cyclotides. The main feature of the general cystine-knot fold is a topological macrocy-cle, formed by two disulfide bonds and their interconnecting peptide seg-ments, which is threaded by a third disulfide bond. The two fold categories are differentiated by the specific disulfide that penetrates the topological macrocycle formed by other two disulfides, which in the ICK motif involves the III-VI disulfide bond, whereas in GFK, the I-IV disulfide bond is in-volved (Figure 1A). Despite having a cystine knot, however, the loops pro-

15

truding from the cystine-knot core of GFK are larger, more flexible than in other knottins, thus GFK has a lower stability.

Cyclotides are ICK or knottins with a cyclic backbone, comprising a large family of cyclic plant-derived microproteins (Ireland, 2010; Craik, 2004). They range in size from 28 to 37 amino acids and are characterized by a head-to-tail cyclized backbone containing six absolutely conserved cysteine residues that are connected intramolecularly in a knotted topology. These arrangement results in six segments or loops in the backbone between cyste-ine residues, which are successively numbered loop 1 to 6, starting at cyste-ine 1 (Figure 1B). Loops 1 and 4 are highly conserved in terms of both the number and types of residues. Most of the other loops are variable in terms of amino acid sequence, composition and size, resulting in a combinatorial library of native cyclotides built around a generalized cyclic cystine knot (Craik, 2006).

The combination of a cyclic backbone and knotted disulfide topology of cyclotides defines a structural motif known as the cyclic cystine knot (CCK) (Craik, 1999). CCK, like other ICK, are formed by three disulfide bonds whereby two disulfide bonds (I-IV and II-V) and their connecting back-bones, loops 1 and 4, form the embedded ring that is penetrated by the third disulfide bond (III-VI) (Figure 1B). This unique feature endows them with a range of valuable biophysical, and chemical properties including exceptional resistance to thermal, chemical, and enzymatic exposure (Colgrave and Craik, 2004). In contrast to most other cyclic peptides of natural origin, such as cyclosporine and bacitracin, which are non-ribosomally synthesized (Walsh, 2004), cyclotides are gene-encoded microproteins that are riboso-mally synthesized from linear precursor proteins (∼100 amino acids) (Burman, 2009; Gillon, 2008; Saska, 2007; Mulvenna, 2005; Dutton, 2004; Jennings, 2001).

16

Figure 1. General features of CKM and cyclotides: A) topology of cystine knots- knottin (ICK) and growth factor cystine knot; B) Loops in structure of cyclic cys-tine-knots, cyclotides C) twist in backbone of cyclotides (upper panels) defining bracelet and Möbius types (bottom panels), from left to right, respectively. The six Cys residues involved in the knots are labeled as I-VI whereas loops or segments between Cys residues as loop 1-6, in order from the N- to-C termini

At present, more than 20 3D structures of cyclotides are available in protein databases (Ireland, 2010). These structures were mainly determined by solu-tion NMR (Göransson, 2009; Mulvenna, 2005; Barry, 2004; Rosengren, 2003; Heitz, 2001; Daly, 1999a; Saether, 1995) but one was solved by X-ray crystallography (Wang, 2009b). Based on their 3D structures, cyclotides were originally categorized into two main subfamilies: Möbius and bracelet (Figure 1C) (Craik, 1999). In Möbius cyclotides, a cis X-Pro peptide bond in loop 5 results in a 180° twist in the cyclic backbone but bracelet subfamily members lack the Pro residue and have a trans-peptide bond at the corre-sponding position (Figure 1C). Hence, the absence and presence of a Pro residue (typically preceded by an aromatic residue) in loop 5 is the main

17

characteristic of bracelet and Möbius subfamilies, respectively. This loop is hydrophobic in Möbius cyclotides. In contrast, bracelet cyclotides have posi-tively charged residues at this position. Interestingly, novel peptides sharing features of both subfamilies are also emerging, usually called hybrids or chimaeras (Daly, 2006). Two macrocyclic peptides, Momordica cochinchinensis trypsin inhibitors I and II (MCoTI-I and MCoTI-II) were discovered in the Cucurbitaceae plant family after the name “cyclotide” was coined, forming a third subfamily referred to as the trypsin-inhibitor cyclo-tides (Hernandez, 2000). Trypsin inhibitor cyclotides are of particular inter-est from a pharmaceutical perspective because of their ability to penetrate cells and thus, potentially, interact with intracellular targets (Jagadish and Camarero, 2010; Greenwood, 2007).

The conserved structural features of all cyclotides include: the cyclic cys-tine knot; a β-hairpin involving loops 4, 5 and 6, which may be associated with a third distorted β-strand, forming a triple stranded anti-parallel β-sheet; and a Glu residue in loop 1 that forms an important network of hydrogen bond with loop 3 (Göransson, 2009; Rosengren, 2003). The knotted-cystine core of cyclotides is critical for maintaining the overall fold and has been shown to be important for biological activity and stability (Colgrave and Craik, 2004). Examples of typical sequences and corresponding structures from each of the categories highlighted above (Möbius, bracelet, hybrid, and trypsin inhibitor) are given in Figure 2A and B. Note that although trypsin inhibitor cyclotides have no sequence similarity to other cyclotides subfami-lies, they are generally classified as such on the basis that they contain a CCK motif (Hernandez, 2000).

18

Figure 2. A) Sequences and B) structures of representative members of the cyclotide subfamilies: The bracelet, cycloviolacin O2 (pdb code: 2KNM); hybrid, kalata B8 (pdb code: 2B38); Möbius, kalata B1 (pdb code: 1NB1) and trypsin inhibitor, MCOTI-II (pdb code: 1IB9). The doted lines connecting kalata B8 with Möbius and bracelet indicates, kalata B8 shares properties of both subfamilies. MCOT-II is illus-trated as isolate indicating that it has no sequence homology with these members except the disulfide bonds. Cyclic backbone is shown by broken line connecting N- and C-termini of the sequence.

19

1.3 Stability and biological activities

The covalent disulfide linkage of the knotted core confer conformational rigidity, and appear to be important even in the absence of cyclization since acyclic analogs of kB1 also fold to native-like structure and are highly re-sistant to extreme pH, thermal denaturation and proteolytic attack (Barry, 2003; Daly and Craik, 2000). However, the losses of biological activity of these acyclic permutants have been attributed to the greater conformational flexibility resulting from absence of cyclization. Similarly, the acyclic form of a naturally occurring cyclotide, violacin A, has a similar overall fold and stability to the cyclic counterparts, but with weaker hemolytic activity (Ireland, 2006b).

Cyclotides have a wide range of biological activities, including anti-HIV (Daly, 2004; Gustafson, 1994a), insecticidal (Gruber, 2007; Jennings, 2005; Jennings, 2001), anti-tumor (Svangård, 2004; Lindholm, 2002), antifouling (Göransson, 2004) (cyO2), anti-microbial (Tam, 1999a), hemolytic (Ireland, 2006a; Schoepke, 1993), neurotensin antagonism (Witherup, 1994), trypsin inhibition (Hernandez, 2000), and uterotonic activities (Gran, 2000; Gran, 1973b; c). The potent insecticidal activity of cyclotides kB1 and kalata B2 (kB2) has prompted the belief that cyclotides might act as plant host-defense agents (Gruber, 2007; Jennings, 2005; Jennings, 2001).

20

2 A hybrid of solid and solution phase synthesis of microproteins

2.1 Linear chain assembly Solid phase peptide synthesis (SPPS), which was pioneered by Merrifield (Merrifield, 1964; 1963), involves reaction between an immobilized inter-mediate on solid support and reagents in solution, followed by filtration and washing of the product with suitable solvents. The strategy initially intro-duced by Merrifield exploited Boc/benzyl chemistry and was based on a system of graduated acid lability. In Boc/benzyl-based syntheses, trifluoroa-cetic acid (e.g. 33% trifluoroacetic acid, TFA, in dichloromethane (DCM)) is used for iterative removal of the t-butyloxycarbonyl (Boc) protecting group from the α-amino (Nα) functionality, while side-chain protecting groups and the peptide-resin linkage are simultaneously cleaved using a stronger acid, anhydrous hydrogen fluoride (HF) after completing peptide elongation. Fol-lowing this initial development, a milder method for SPPS that avoids use of HF in the final deprotection step, the fluorenylmethoxycaronyl (Fmoc)/t-butyl approach became popular (Atherton, 1978; Carpino and Han, 1972). This methodology differs from Boc/benzyl chemistry in that the Nα and side-chain protection are based on a system of orthogonality, i.e. each type of protection selectively removed by a mechanism that leaves the other types of protection intact. In Fmoc/t-butyl chemistry, a secondary amine, usually 20-50% piperidine in dimethylformamide (DMF), is used for iterative re-moval of the Fmoc protecting group, while side-chain protecting groups and the peptide-resin linkage are simultaneously cleaved after complete peptide assembly using TFA. The work underlying this thesis was based on manual Fmoc-chemistry and the remaining discussion mainly pertains to this tech-niques.

The success of SPPS is dependent on the choice of solid support, i.e. the resin, for synthesis. Polystyrene based resins, such as 2-chlorotrityl chloride and methylbenzhydrylamine (MBHA) functionalized resins, are the com-monly used solid supports. They are hydrophobic resins that have a matrix of polystyrene cross-linked with 1-2% divinylbenzene (Lee, 2008; Barlos, 1991). However, hybrid resins composed of poly(ethylene glycol) and poly-styrenes-(PEG)-PS resins, such as NovaSyn TGT resin are polar, and have high solvation power (Kates, 1998). In addition, fully hydrophilic resins,

21

such as ChemMatrix, have been developed for particularly hydrophobic or long peptides (Garcia-Martin, 2006).

A linker is a small functional unit between the resin and the enlongating peptide, and serves as a protecting group for the C-terminal groups of the peptide. The linker system must be chosen appropriately, depending on whether the C-terminal of the final peptide is a carboxylate (e.g. chlorotrityl acid, clear acid resin), a carboxamide (e.g. rink amide-MBHA, rink amide Clear resin), or thioester (sulfonamide resins), with consideration given to minimize potential side reactions, including suppression of racemization. The resins and linkers used in the work this thesis is based upon are shown in Figure 3. Some specialized linkers, called ‘safety catch’ linkers, originally introduced by Kenner (Kenner, 1971), are stable to repetitive treatment with both strong acids and bases. Kenner’s ‘safety catch’ linkers are widely used to generate peptides containing a C-terminal thioester for cyclization/native chemical ligation (NCL) reactions (Escher, 2001; Shin, 1999). A related ‘safety catch’ linker, the Dawson Dbz linker (Blanco-Canosa and Dawson, 2008), is also stable under basic conditions and is particularly useful in Fmoc-based thioesterification. Such linkers are transformed into reactive groups via alkylation with trimethylsilyl-diazomethane (TMS-CHN2) or iodoacetonitrile in the case of Kenner’s safety catch linkers (Backes and Ellman, 1999; Shin, 1999; Backes, 1996; Backes and Ellman, 1994; Kenner, 1971) but via acylation and subsequent intramolecular cyclization in the case of Dawson Dbz linker (Blanco-Canosa and Dawson, 2008).

Coupling reagents are used to activate the C-terminal carboxyl of incom-ing amino acids. O-acyluronium/guanidinium and O-acylphosphonium cou-pling reagents enable in situ generation of active esters in the presence of a tertiary nitrogen base (e.g., diisopropylethylamine, DIPEA). These coupling reagents have gained popularity due to their facile handling, short coupling time and minimal loss of configuration during coupling (Han and Kim, 2004). The highly efficient coupling reagents HBTU, TBTU and HATU (see abbreviation and structures in Figure 3) are the most widely used. In the work described in this thesis, HBTU was used either alone or in combination with HOBt for amino acid coupling, whereas PyBOP was used for thioester formation. A tertiary amine, such as DIPEA was used as a base to form the carboxylate ion of the C-terminal carboxyl group.

22

Figure 3. Structures and nomenclature of resins and linkers (left hand box), and coupling reagents (right hand box) used in the work this thesis is based upon. The linker groups are highlighted by the dashed boxes in the resin structures. 2-(1H-Benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU), 2-1H-Benzotriazole-1-yl)-1,1,3,3-tetramethyluronium tetrafluoroborate (TBTU) and 2-(7-Aza-1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HATU), 1-hydroxy benzotriazole (HOBt), benzotriazol-1-yloxytripyrrolidino phos-phonium hexafluorophosphate (PyBOP)

Sequence-dependent side reactions can occur during Fmoc SPPS. For exam-ple, diketopiperazine formation occurs at the C-terminal deprotected dipep-tide stage, and intramolecular cleavage of the resin ester linkage by the free amino function of the penultimate amino acid occurs after deprotection of the Fmoc group, particularly in peptides containing Gly or Pro at the C-terminus (Carpenter, 1994; Fujiwara Y, 1994; Pedroso, 1986). In such cases, the use of hindered trityl-based resins could minimize loss of the dipeptide from the resin (Chan and White, 2000). Cyclization of aspartic acid residues, forming aspartimide, is another side reaction that can occur, during either chain elongation or final TFA cleavage, when Asp (OtBu)-Gly is present in the sequence. In this case, hydrolysis of the aspartimide ring leads to a mix-ture of both α- and β-peptides, its reaction with piperidine, used for Fmoc removal, also results in α- and β-piperidides (Quibell, 1994). Reversible protection of the nitrogen of the Asp-Gly amide bond, using preformed di-peptides such as Fmoc-Asp(OtBu)-(Hmb)Gly-OH or Fmoc-Asp(OtBu)-(Dmb)Gly-OH, can completely block the amide nitrogen from attacking the β-carboxyl group (Mergler, 2003a; Mergler, 2003b). Of these two dipep-tides, Fmoc-Asp(OtBu)-(Dmb)Gly-OH is the most efficient for incorporat-ing Asp-Gly amide protection. For the work reported in this thesis, resins

23

and the corresponding linkers were selected to either generate protected pep-tides or to obtain peptide acids, Carboxyl terminal activated peptides and thioesters directly after cleavage from the resins (see Figure 3 and for further details see Papers II, III, and V).

2.2 Native chemical ligation/cyclization Once the linear peptide chains have been fully assembled, the next step in synthesizing cyclic cystine-knotted microproteins is formation of cyclic pep-tides via native chemical ligation (NCL). The basic principle of NCL is that two peptide fragments, one containing an N-terminal cysteine and the other a C-terminal α-thioester, can be joined together to form a peptide bond chemoselectively at the ligation site (Tiefenbrunn and Dawson, 2010; Dirksen and Dawson, 2008; Dawson, 1994). The reactions proceed efficient-ly without side reaction involving δ-, and ε-amines, phenolates and hydrox-yls, generating a single protein. NCL has been applied to synthesize long peptides or proteins by ligating two or more fragments (Tiefenbrunn and Dawson, 2010; Dawson, 1994). When both the N-terminal cysteine and C-terminal thioester are incorporated in the same peptide sequence, an intramo-lecular NCL reaction can occur, which connects the N to the C temini of the linear molecule, generating a circular peptide, as first reported by Camarero and Muir (Camarero, 1998a; Camarero, 1998b; Camarero and Muir, 1997). Subsequently, in a series of articles, Tam and Lu demonstrated that an NCL strategy could be used to produce cyclic cystine-rich peptides (cyclotides) in a reaction they named the “thia zip reaction” (Tam, 1999b; Tam and Lu, 1998; 1997).

Members of all three cyclotide subfamilies have been chemically synthe-sized using NCL to generate the cyclic backbone (Clark and Craik, 2010; Thongyoo, 2008; Gunasekera, 2006; Thongyoo, 2006; Daly, 1999b; Tam, 1999a). For efficient cyclization, linear precursors were synthesized with a cysteine residue at the N-terminus and a thioester linkers at the C-terminus, as illustrated in Figure 4. It has been reported that the presence of unprotect-ed internal cysteine residues in the assembled peptide chain allows cycliza-tion to proceed through a series of reversible thiol-thiolactone exchanges in the direction of N-terminus, resulting in formation of an Nα-amino thio-lactone linking the N and C termini (Tam, 1999b). Once the final Nα-amino thiolactone bond is formed between the thiol group of the N-terminal cyste-ine and the C-terminal carbonyl group, a spontaneous S,N acyl-migration via a 5-membered ring intermediate results in a head-to-tail cyclized native pep-tide bond (Figure 4) (for further details see Papers II, III, and V).

24

Figure 4. Peptide/protein cyclization via intramolecular thia zip reaction. The final thiolactone generated and the amide bond is shown in rectangle.

In the absence of Cys at the N-terminal, removable acyl transfer auxiliaries, being attached at the N-terminus, have been successfully applied (Offer, 2002). Following ligation, these auxiliaries can be removed under acidic conditions. Xaa-Ala has also been used as a ligation site, exploiting a Cys residue in place of native Ala residue for ligation and desulfurizing the final polypeptide product with Raney Nickel (Pentelute and Kent, 2007) or Pd/Al2O3 (Yan and Dawson, 2001) or radical reaction (Wan and Danishefsky, 2007) to obtain the target sequence.

2.3 Synthesis of activated precursors for NCL/cyclization

NCL/cyclization involves a C-terminal thioester. Hence, synthesis of peptide and protein C-terminal α-thioesters is crucial. The procedure applied to gen-erate thioesters depends on which SPPS strategy is used in the peptide or proteins synthesis. In the case of Boc-SPPS, it is straightforward to synthe-size a thioester using a thioester linker (Camarero and Muir, 1997; Dawson, 1994), whereas extra steps are usually required in Fmoc-SPPS (Mezzato, 2005; Camarero, 2004; Ingenito, 1999; Shin, 1999). Consequently, several

25

methods to afford peptide thioesters using Fmoc SPPS have been developed, including use of cleavage cocktails (Li, 1998), thioesterification of protected peptides (Eggelkraut-Gottanka, 2003; Futaki, 1997) and activa-tion/thioesterification following peptide chain assembly (Camarero, 2004; Shin, 1999). Since thioesters are unstable to repeated deprotection using piperidine, other non-nucleophilic bases have been investigated for the re-moval of the Nα-Fmoc group during peptide elongation, including: 25% 1-methylpyrrolidine/2% hexamethyleneimine/2% (w/v) HOBt in N-methylprolidinone (NMP):dimethylsulfoxide (DMSO) (1:1) (Hojo, 2005; Li, 1998) or 1% 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU)/HOBt mixture (Bu, 2002; Clippingdale, 2000). A popular route for peptide thioester synthesis is to use a modified Kenner's sulfonamide safety-catch linker method and standard Fmoc/t-Bu SPPS followed by sulfonamide alkylation to activate release of the peptide as a peptidyl-thioester from the solid support (Figure 5A) (Mende and Seitz, 2007; Mezzato, 2005; Ingenito, 1999; Shin, 1999).

The drawbacks of this method include epimerization during the long-lasting first residue loading, longer duration of alkylation and thioesterification, low isolated yields, and in the case of post-translationally modified peptides, such as glycopeptides, the activating alkylation step can also result in unde-sirable alkylation of unprotected hydroxyl groups on carbohydrate and me-thionine residues. Consequently, other approaches have been developed, such as synthetic strategies that release fully protected peptide acids (Eggelkraut-Gottanka, 2003) or unprotected activated peptide amides (Blanco-Canosa and Dawson, 2008) from the solid support, prior to conver-sion to thioesters (Figure 5B and C).

26

Figure 5. Different approaches of thioester synthesis. Thioester and activated pre-cursors are encircled.

Mende et al have recently reported a combination of on-resin macrocycliza-tion via an N-terminus cleavable cyclization linker and thiolytic ring-opening at a sulfonamide “safety-catch” linker, providing purified peptidyl thioester after libration from the resin that can directly be used for NCL (Figure 6) (Mende, 2010; Mende and Seitz, 2010; Mende and Seitz, 2007). This approach seems to be an elegant way of generating purified thioester on-resin by automated synthesis for subsequent cyclization/NCL. However, the time-consuming steps in sulfonamide-based thioester generation, i.e. alkylation and thioesterification, remain the same. Thioesters have also been generated biosynthetically by protein-splicing techniques and cyclized in vivo spontaneously or in vitro (Austin, 2009; Camarero, 2007; Kimura, 2006; Camarero and Muir, 1999).

There are six cysteines in cyclotides and theoretically cyclization can be mediated through any one of them (Daly and Craik, 2000). However, to avoid steric interactions, the least hindered available site such as Gly-Cys, Ala-Cys, or Ser-Cys, is usually selected as the point of disconnection to start

27

assembly and subsequent recyclization (Gunasekera, 2008a; Thongyoo, 2006; Daly, 1999b). In the studies underlying this thesis two approaches were used to cyclize peptides: formation of a protected peptide, followed by in-solution thioesterification (Papers II and III); and synthesis of an activat-ed peptidyl-Nbz that was thioesterified in situ (Paper V).

Figure 6. On-resin cyclization and thioesterification, showing an approach to re-move truncated impurity in thioester synthesis (Mende, 2010; Mende and Seitz, 2007). The required final thioester is encircled.

28

2.4 Oxidative folding: locking in the native state Following cyclization, the final and generally rate-limiting step in the syn-thesis of cyclic cystine-knot microproteins, particularly in bracelet cyclo-tides, is oxidative folding, which refers to the combination of native disulfide bond formation and conformational folding, resulting in the native three-dimensional folded protein (Cemazar, 2008; Craik and Daly, 2005). The formation of disulfide bonds is crucial for the folding, stability, and biologi-cal activity of cystine knot microproteins. Due to the interdependence be-tween conformational folding and disulfide formation, and the plethora of disulfide connectivities that may occur at different stages of the oxidative folding process, generation of a single natively folded peptide is often chal-lenging. In principle, the number of potential pathways and transient inter-mediate species that may occur in the oxidative folding of disulfide-rich proteins increases as the number of cysteine residues increases, however, only one species is expected to have the fully oxidized native peptide fold. The number of ways (sp) of forming p disulfide bonds from n cysteine resi-dues is given by the formula (Sela and Lifson, 1959),

sp =n!

p! n − 2p( )!2p

For example, numerous intermediate disulfide species (15 one-disulfide spe-cies (1SS), 45 two-disulfide species (2SS) and 15 three-disulfide species (3SS)) are theoretically possible on the oxidative folding pathways of a cy-clotide with six Cys residues. Among these, only one product will have the fully oxidized native fold, whereas the remaining 74-disulfide connectivities are intermediates. However, due to unfavorable energy state, some of these disulfide bonds may not be observed and most disulfide-rich proteins fold with remarkable efficiency and specificity for a particular disulfide connec-tivity under suitable conditions. Under favorable conditions, native cyclo-tides, particularly Möbius and trypsin inhibitor cyclotides, can be generated in a one-pot cyclization and oxidative folding(Cemazar, 2006; Kimura, 2006; Daly, 2003a).

2.5 Chemical and biomimetic engineering Cyclic cystine-knot and related acyclic microproteins have attracted the at-tention, due to their remarkable stability, bioactivities and readily accessible synthetic methodology (Garcia and Camarero, 2010; Jagadish and Camarero, 2010; Sommerhoff, 2010). In cyclotides, a cystine-knot core is incorporated with six backbone loops that are variable in size and sequence, offering the

29

possibility of synthetically substituting them with new, otherwise suscepti-ble, bioactive sequences that are stabilized by the knots. To demonstrate such an approach, cyclotides have been grafted with foreign bioactive epti-topes that have anticancer (Gunasekera, 2008b) properties, stabilizing the labile non-natural bioactive small peptides. In this study a poly-Arg hex-apeptide with antiangiogenic properties was grafted onto the kB1 framework in loops 2, 3, 5 and 6; the analog with the poly-Arg sequence grafted into loop 3 showed the highest receptor inhibitory activity, compared to the un-grafted linear poly-Arg epitope and the other grafted analogs. Similarly, Thongyoo et al showed that the trypsin inhibitor site located in loop 1 of the MCoTI cyclotides could be engineered in its specificity for proteases other than trypsin (Thongyoo, 2008). Despite success in grafting Möbius and tryp-sin inhibitory cyclotides, engineering of bracelets, the largest family of cy-clotides, is challenging due to their intractable-oxidative folding properties.

2.6 Biomolecular engineering Cyclotides have also been successfully synthesized and engineered using a biomolecular approach, in which modified inteins and a Met residue, are fused in-frame to the C- and N- termini of a target cyclotide, respectively (Austin, 2009; Camarero, 2007; Kimura, 2006). Peptide-bond cleavage of the Met residue and modified inteins results in an N-terminal cysteine and a C-terminal thioester, respectively, on the same linear cyclotide, which can then be intramolecularly condensed to generate the circular backbone micro-proteins (Kimura, 2006). With this strategy, linear kB1 mutant library have been expressed in Escherichia coli (E. coli) with C-terminal modified intein fusion protein, generating a C-terminal thioester and an N-terminal Met resi-due that liberated an N-terminal Cys residue when finally unmasked. Precur-sor proteins were isolated and then cyclized/oxidatively folded in vitro in a one-pot reaction. Camarero and his colleagues have also synthesized native MCoT-II cyclotides, as well as libraries of multiple MCoTI-I mutants, fully inside the living cells of E. coli using an intein-mediated cyclization and spontaneous oxidative folding (see Figure 7) (Austin, 2009; Camarero, 2007). These approaches are very attractive for generating combinatorial libraries of mutant cyclotides and can serve as complementary route to solid phase peptide synthesis.

30

Figure 7. Biosynthetic approach to produce cyclotides inside living E. coli cells. Backbone cyclization of the linear cyclotide precursor is mediated by a modified protein-splicing unit, which subsequently folds spontaneously in the living bacterial cells.

31

3 Aims of the present study

The work presented in this thesis was part of a research project concerning plant proteins focusing on cyclotides at the Division of Pharmacognosy, Department of Medicinal Chemistry, Uppsala University. The long-term aims of the project are to understand the biological functions of this unique protein family and to explore their possible use in biotechnological, pharma-ceutical and agricultural applications. Specific objectives of this study were • to synthesize cystine knotted peptide and evaluate their structure and

bioactivity

• to devise methods for synthesis of native bracelet cyclotides including native ligation/cyclization and oxidative folding

• to determine and compare the oxidative folding pathways for the differ-

ent cyclotide subfamilies (by mapping subsets of folding species)

• to devise synthetic and rational design strategy for backbone engineering of cyclotides including peptidyl-thioester synthesis

32

4 Synthesis and structural determination of a novel disulfide-rich peptide (Paper I)

4.1 Discovery of an acyclic disulfide-rich peptide Tripatide is a novel disulfide rich peptide that was initially isolated from the cactus, Trichocereus pachanoi, a member of the plant family Cactaceae. Some species of this family are known for their hallucigenic constituent, mescaline (Ogunbodede, 2010; Helmlin, 1992; Poisson, 1960). The arial parts of the plant were extracted, fractionated and analyzed by MS in a search for polypeptides following protocol described by Claeson, et al (Claeson, 1998). The MS analysis showed that the crude fractions contained polypeptides, among which tripatide of major abundance, which lead to its purification and characterization.

MS analysis of the native peptide showed that it had a molecular weight of 3602.1. Alkylation of the native peptide with iodoacetamide did not lead to any apparent change in mass. However, when native peptide was reduced with dithiothreitol for 2 h followed by iodoacetamide alkylation, S-carbamidomylated peptide mass was increased by 348 Da revealing the di-sulfide rich nature of the peptide, i.e. six cysteine residues. Quantitative ami-no acid analysis was then performed (at the Amino Acid Analysis Center, Department of Biochemistry and Organic Chemistry, Uppsala University) by acid hydrolysis over 24 h at 110°C with 6 N HCl containing 2 mg/mL phe-nol. The hydrolysates were analyzed with automated amino acid analyzer using ninhydrin detection. The results showed that the peptide is composed of 1xAla, 3xArg, 4xAsn/Asp, 1xSer, 2xGln/Glu, 3xPro, 6xGly, 6xCys, 4xVal, 1xIle, 3xLeu, and 1xPhe.

Following determination of its amino acid composition, the sequence of S-carbamidomethylated peptide was determined by automated Edman deg-radation using protein sequencer, which revealed the following 28 residues sequence: CVLIGQRCDNDRGPRCCSGQGNCVPLPF. This sequence was further confirmed by MS/MS analysis of tryptic cleavage fragments of the S-carbamidomethylated peptide following a previously reported protocol (Broussalis, 2001). The remaining still ambiguous sequence of seven C-terminal residues was confirmed by MS/MS analysis of chymotryptic cleav-age fragment of S-carbamidomethylated peptide as: LGGVCAV.

33

The average mass of fully sequenced peptide (MW 3608.2 Da) differs from that of native peptide (3602.1 Da) by 6 Da, showing that all Cys resi-dues are involved in disulfide bond linkage. The native peptide was named tripatide based on the name of the plant species it was isolated from, Tricho-cereus pachanoi. A sequence similarity search in known protein databases revealed that tripatide is a unique peptide with very low sequence similarity (∼40%) to agelenin (Inui, 1992) and related inhibitor cysteine knot peptides (Gracy, 2008) such as agatoxins. Due to their inherent stability, relatively small size that makes them readily accessible to chemical synthesis and highly specific bioactivities, inhibitor cystine knot peptides have potential applications in peptide based drug design and biomolecular engineering.

4.2 Solid phase synthesis and oxidative folding Tripatide was isolated with very low yields, only sufficient to determine its sequence. Thus in order to obtain sufficient amounts for further analysis, it was synthesized using Fmoc-SPPS. The first residue attachment to 2-chlorotrityl chloride resin was conducted in DCM in the presence of DIPEA. The rest of the amino acid residues were coupled in DMF using HBTU as activating agent and DIPEA as base via stepwise activation and coupling of Fmoc-protected amindo acids followed by Fmoc deprotection cycle until the peptide was fully assembled, as illustrated in Figure 8. The coupling effi-ciency was confirmed by a ninhydrin test at each step (Sarin, 1981). The final peptidyl-resin was treated with a cleavage mixture to release the pep-tide from the resin. The resulting peptide was confirmed by ESI-MS (ob-served. 3607.7 Da, av.; calculated. 3608.2 Da, av.), partially purified by RP-HPLC and freeze dried.

34

Figure 8. Fmoc-SPPS of tripatide with sequence: CVLIGQRCDNDRGPRCCSGQGNCVPLPFLGGVCAV. Fmoc group was depro-tected with 50% Pip/DMF (2 ×1 min). Nα-Fmoc-amino acid (1 mmol, 4 equiv.) was suspended in 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophos-phate (HBTU; 0.5M, 2 mL, 1 mmol, 4 equiv.) and activated by addition of DIPEA (180 mL, 1 mmol, 4 equiv.). Fully assembled peptide was cleaved and deprotected with TFA/TIPS/water (95/2.5/2.5%, 2 h). Reaction scale was 0.25 mmol.

The reduced peptide was then subjected to oxidative folding in buffer sys-tems of 0.1 M NH4HCO3, 2 mM/GSH/ 2 mM GSSG (standard buffer) con-taining either 30% MeOH (aq.) or 30% DMSO (aq.) or only standard buffer over 1-48 h. The oxidative folding reaction was monitored by RP-HPLC and ESI-MS by removing aliquots and quenching with 2% formic acid at those time intervals. The native peptide was accumulated as a single peak after 8 h with the yield of about 70% in the DMSO containing buffer. Native peptide was eluted with longer retention time than reduced counterpart similar to cyclotides (Daly, 1999b) indicating an overall hydrophobic fold. The folding pattern and efficiency was similar in all the buffer systems that indicate tripatide was readily amenable for in vitro folding in ordinary buffer sys-tems. The identity of synthetic peptide was confirmed by co-injecting isolat-ed and synthetic tripatide, which was found to co-elute. Typical RP-HPLC traces illustrating the folding in 30% DMSO (aq.) are shown in Figure 9.

35

Figure 9. RP-HPLC traces illustrating the oxidative folding profile of tripatide in 30% aq. DMSO. R is duced and N is oxidatively folded tripatide.

4.3 Solution NMR structural determination Given the unique sequence of tripatide and its potential application in pep-tide based drug design and biomolecular engineering, determination of its solution structure and comparing structural similarity with known polypep-tides was clearly warranted. Therefore, it was subjected to detailed structural analysis using solution NMR techniques. 2D 1H NMR analysis including DQF-COSY, TOCSY and NOESY showed that narrow line width and well dispersed amide region (10.1-7.4 ppm) indicating a highly ordered structure. Resonance assignments were achieved using standard 2D assignment strate-

36

gies (Wüthrich, 1986). The tripatide comprise three proline residues for which Pro25 and Pro27 were confirmed to be in trans conformation while Pro14 in cis conformation based on the presence of Hαi-1-Hδi (trans) or Hαi-

1-Hαi (cis) NOEs between preceding residue (i-1) and proline (i), respective-ly. Secondary Hα shifts, i.e. deviations of the observed Hα shifts relative to shifts observed in random coil peptides are indicative of secondary structure. Figure 10A shows the secondary shifts for tripatide.

Figure 10. A) Hα chemical deviation of tripatide from random coil shift showing secondary structural elements and B) ribbon representation of tripatide structure.

3D structural fold of tripatide comprises a triple stranded β-sheet comprising residues 7-8, 22-26 and 30-34 (Figure 10B). The strands are connected by a series of turns including well defined β-turns comprising residues 3-6, 10-13 and 17-20, and a less defined turn comprising residues 26-29. The elements of secondary structure are stabilized by an extensive network of hydrogen bonds, and it is clear that these interactions together with the disulfide bonds are the major contributors to the stability of the overall fold.

In an attempt to determine disulfide connectivity by using the NMR struc-tural data, the derived set of structural restraints were used to calculate struc-tures using all fifteen possible disulfide connectivities. All combinations without the 1-17 bond are highly unfavorable. In contrast all three variants with 1-17 disulfide bond: 1-17, 8-23, 16-33; 1-17, 8-33, 16-23 and 1-17, 8-16, 23-33 all agrees very well with the available NMR data and only show minor violations to experimental restraints. Furthermore, when compared not only do families with different disulfide bond connectivities superimpose well within the families but if separate families were overlayed the different variants are close to identical. In fact even the projection of the side chains of the six cysteines (χ1angle) are identical between two models, with differ-ences in χ2 angles sufficient to allow different connectivities. Thus strategies for detemining disulfide connectivities based on projection of side chains, as previously done for cyclotide (Rosengren, 2003) was not possible. Moreo-ver, disulfide bond determination by chemical means was also not successful

37

due to very low yields of partially reduced species and absence of positively charged residues for enzymatic cleavage where most cysteines are localized.

A sequence alignment of tripatide with selected related ICK peptides that have more sequence similarity with tripatide, is shown in Figure 11A. Sub-jecting the determined structure to a structure similarity search using the DALI server showed that tripatide is closely related to ICK peptides also in terms of secondadry and tertiary structures. The most similar structures were spider toxins, in particular agelenin, which structurally compared in Figure 11B.

Figure 11. A) Comparison of sequences of selected ICK peptides with tripatide, B) overlay structures of tripatide (blue) and agelenin (grey) (pdb code: 2E2S).

Tripatide contains multiple Gly/Cys residues, constituting 34% of the pep-tide composition. This is interesting, since Gly/Cys rich peptides, such as avesin A, hevein and related Cysteine/Gly-rich domain of chitin-binding proteins have been found to have chitin binding properties with potential antimicrobial activity (Li and Claeson, 2003; Broekaert, 1992; Broekaert, 1990). Despite this potential feature, preliminary chitin binding test of tripatide was unsuccessful. However, inhibitor cystine knots such as age-lenin, which showed relatively closer sequence similarity to tripatide, exhib-ited insecticidal and ion channel binding activity showing the potential of this peptide in peptide based drug design and biomolecular engineering that further needs to be confirmed.

38

5 Fmoc-based synthesis and oxidative folding of bracelet, cycloviolacin O2 (Papers II and III)

5.1 The bracelet cyclotides Cyclotides are mainly categorized into bracelet and Möbius subfamilies. Bracelet constitutes the major part of cyclotides and highly intractable to currently available synthetic and oxidative folding strategies, which is of course, a major concern in the context of this thesis and the studies it is based upon. Previously two strategies were used to obtain native bracelet cyclotides, random oxidation strategy and a two step regioselective protec-tion approach (Tam, 1999a; Tam and Lu, 1998). However, both approaches have not been resulted in high yields. Unlike tripatide, cyclotides require additional cyclization steps. To develop a suitable cyclization and oxidative folding strategy, a series of steps were optimized using the model bracelet, cycloviolacin O2 (cyO2) including selection of a suitable linker and resin for linear chain coupling and subsequent protected peptide generation.

5.2 Synthesis of linear chain and protected peptides The approach adapted for the linear chain assembly of cyO2 was identical to that applied for the synthesis of tripatide. However, in the synthesis of cyO2, a linear activated peptide (peptide thioester) must be constructed to form cyclic peptide. For this purpose, the sequence of cyO2 was assembled, using a similar procedure to that applied in the tripatide synthesis, on highly acid labile linker (4-carboxytrityl linker) that was used to release protected pep-tides for thioesterification.

Initial attempts to assemble linear peptide on 2-chlorotrityl chloride resin were unsuccessful because of incomplete coupling starting from the fifth or sixth residue addition. Consequently, chlorotrityl resin was replaced with Gly preloaded NovaSyn TGT resin, which has the same 4-carboxytrityl link-er. The chemical properties of this resin are similar to those of 2-chlorotrityl resin but it offers additional advantage of the highly polar PEG-polystyrene base matrix with better solvating power. Since it provides suitable targets for both thioester introduction and native ligation, the Gly16-Cys17 site selected

39

for the planned native ligation/cyclization and hence starting point for linear chain assembly. The linear peptide was synthesized by manual Fmoc-SPPS using HBTU/DIPEA activation protocol. The temporary Nα-Fmoc was re-moved by 2x1 min treatments with Pip/DMF (50% v/v). Whenever neces-sary, as judged by ninhydrin tests, residues were recoupled. Once the full linear chain peptide assembled, protected peptides were cleaved from the resin with optimized mild cleavage protocol of CH3COOH/TFE/DCM (1:1:8, 3h) (Figure 12). The resulting peptide was repeatedly washed with n-hexane, concentrated in vacuo and directly used for thioester formation.

Linear chain assembly and protected peptide formation was also per-formed using manual microwave peptide synthesizer (CEM Corp, Matthews, NC), with significantly reduced reaction time for coupling (from 32 min to about 8 min) and protected peptide formation (from 3 h to 45 min), together with increased coupling and deprotection efficiency. The protocol was then applied successfully to two other members of cyclotides: Möbius, kB1 and trypsin inhibitor, MCoTI-II (Paper III).

Figure 12. Synthetic strategy for synthesizing protected peptide, peptidyl-thioester and cyclic peptide with sequence CSCKSKVCYRNGIPCGESCVWIPCISSAIG. Peptide thioester was efficiently cyclized under neutral pH (pH 7.4), denaturing (6 M guanidine) and reductive (6 equiv. DTT) conditions.

5.3 Peptide cyclization using peptidyl α-thioester Formation of peptidyl α-thioester is a key step in native ligation/cyclization. To introduce thioester at the C-terminus of protected peptides, previously published method (Eggelkraut-Gottanka, 2003) was adapted, with modifica-tion, i.e. PyBOP/4-acetamidothiophenol/DIPEA (5/15/500 μL, 3 h) in DCM. A high equivalent of DIPEA was used since C-terminal residue, Gly is not prone to epimerization and high equivalent of DIPEA increases the rate of thioester formation (Paramonov, 2005). The standard TFA-scavenger cleav-age protocol was used to remove side-protecting groups on protected pep-tidyl α-thioester. Following thioesterification, peptide α-thioester was effi-

40

ciently cyclized to generate the reduced backbone cyclized single peptide as illustrated in Figure 12.

Similarly, taking advantage of microwave irradiation, the duration of thi-oesterification of cyO2 in more polar aprotic solvent, DMF, was optimized to be as PyBOP/acetamidothiophenol/DIPEA (5/5/10, 5 min) reducing both the reaction time and thioesterifying reagent equivalents (Paper III). The microwave based thioesterification was successfully applied in kB1 and MCOTI-II synthesis, in which the resulting crude peptides were cyclized and oxidatively folded in ‘one pot’ reaction in 0.25 M Tris-HCl containing 50% iPrOH, 8/2 mM GSH/GSSG.

5.4 Optimization of oxidative folding conditions Once the reduced cyclic peptides have been obtained, the last and in many cases, the rate-limiting step is oxidative folding. This is a combination of two mutually dependent systems: oxidative folding to native conformation and connecting cysteines in only one of the 15 possible combinatorial forms, i.e. the native conformation (for example cyclotides have native disulfide connectivity of I-IV, II-V, III-VI). Although this is topologically complex, formation of such native folding is straightforward for Möbius and trypsin inhibitory cyclotides, even under one pot cyclization and oxidative folding conditions (Kimura, 2006; Thongyoo, 2006; Daly, 1999b). However, it is a bottleneck when dealing with bracelet cyclotides. In order to circumvent this problem, several external factors that reportedly enhance oxidative folding of peptides and proteins were explored, including salt concentration and temperature (Kubo, 1996), pH, metal ions, co-solvents, (DeLa Cruz, 2003) and duration of reaction but without significant improvement.

In this screening, folding at low temperature in the presence of co-solvent of 40% MeOH containing 0.4 mM EDTA, and GSH/GSSG (2/1 mM) result-ed in small yield of native peptide, raising possibilities that prompted more detailed investigation (Figure 13). Non-ionic detergents, such as Brij 35, Tween 40, and Tween 60, which reportedly improve folding of hydrophobic peptides (DeLa Cruz, 2003) by increasing surface exposure of hydrophobic patches have also shown to improve the native yields of this peptide. Alt-hough all these detergents proved to improve the folding, we chose to con-tinue with Brij 35 since it quantitatively favors the last RP-HPLC eluting hydrophobic peaks (major misfolded intermediate (MI) and native (N)), and because other detergents usually solidify at the temperature and concentra-tion used. Following further optimization of oxidative folding by varying redox systems, duration of reaction and temperature, the best identified choice of folding conditions was GSH/cytamine, 48 h of reaction time and 3°C (Figure 13).

41

Figure 13. Effect of redox agents and temperature on folding in Tris-HCl (0.1M, pH 8.5), EDTA (1 mM), MeOH (40%) and Brij 35 (6%). A) Folding in the GSH/cystamine redox system over 24 h gave higher yields than in the GSH/GSSG system. Concentrations of 2/2 mM GSH/cystamine favored a higher total yield of hydrophobic peptides (N, MI). B) Higher total yields were achieved at lower tem-peratures; hence 2/2 m MGSH/cystamine at 3°C was used in subsequent experi-ments.

However, under all experimental conditions tested up to this point, misfold-ed products (particularly MI) with the same mass as native peptide but sig-nificantly differing in RP-HPLC retention time were predominant. Surfaces of cyO2 are composed of larger surface exposed hydrophobic patch and smaller localized surface exposed hydrophilic patch. In order to favor the contact of both of these surfaces, we have replaced 40% MeOH with 20% DMSO. DMSO reportedly aids oxidative folding of peptide with basic and hydrophobic residues (Tam, 1991) and this was found to be the case with cyO2 as its presence improved yields. Based on this, concentration of DMSO, influence of various additives and duration of reaction were then re-optimized to maximize the yield of native peptide under these conditions (Figure 14). In that process, 35% DMSO, 2 mM GSH/2 mM cystamine and 6% Brij 35 over 48 h of reaction times were found to be optimum resulting in 40% N.

42

Figure 14. Folding in DMSO with Tris-HCl (0.1M, pH 8.5) and EDTA (1 mM) at 3°C. A) Effects of DMSO with and without different reagents: the presence of all folding additives (DMSO/Brij 35/GSH/cystamine) gave the highest yield after 24 h B) The DMSO concentration was then optimized to 35%. * Indicates condition that was further optimized to give yields with > 50%.

The reaction environment under such conditions, i.e. presence of oxidizing agent (DMSO) at higher pH, (Tam, 1991; Friedman and Koenig, 1971) re-sulted in plateau whereby MI and N reached equilibrium and no further in-crement of N could be observed. This might result from complete oxidation of reshuffling agent, GSH, (Friedman and Koenig, 1971) that might have converted to GSSG thereby trapping reshuffling of non-native 3SS species. Accordingly, we added an extra amount of GSH/cystamine after 24 h under controlled conditions that resulted in improved yield (52% N).

43

5.5 NMR structural studies and disulfide bond determination

NMR spectral assignments for native cyO2, synthetic cyO2 and the MI were carried out by 2D NMR techniques (TOCSY and NOESY) at 290 K as de-scribed before (Rosengren, 2003; Wüthrich, 1986). The chemical shifts were referenced to water peaks at 4.85 ppm. A comparison of the αH secondary shifts of the three peptides, shown in Figure 15, reveals that native and syn-thetic cyO2 have similar values, demonstrating they both have the same 3D structure but MI differs significantly. Attempts to determine 3D structure of MI were unsuccessful due to absence of interresidue NOEs under a range of different experimental conditions. Hence, chemical means were used to de-termine the disulfide bridges. Partial reduction and alkylation of MI with NEM resulted in three major partially reduced products (one 2SS, and two 1SS) which was then completely reduced with DTT and subsequently alkyl-ated with iodoacetamide in analogy with a previously published protocol (Göransson and Craik, 2003). Cleavage of each product with trypsin and MS/MS sequencing revealed the disulfide connectivity of MI as Cys1-Cys5, Cys10-Cys17 and Cys19-Cys24, showing that the intermediate has com-pletely non-native disulfide bond connectivity with significantly different 3D-structure from native peptide (Figure 15).

Figure 15. A) NMR secondary shift comparison. Isolated (red) and synthetic (green) cyO2 have similar values, which differ significantly from those for the MI (blue), B) disulfide connectivity pattern of MI (upper panel) and native cyO2 (N) (lower pan-el).

44

5.6 Conclusion The occurrence and stability of oxidative folding intermediates are influ-enced by the solvent accessibility, and both the reactivity and proximity of thiol and disulfide groups (Arolas, 2006; Cemazar, 2003; Bulaj, 1998; Friedman and Koenig, 1971). The major misfolded product MI accumulated to a significant level and resistant to reshuffling, probably since it has buried three non-native disulfide bonds and/or a conformationally biased non-native fold. Hence, rearrangements of such species must be preceded by conforma-tional unfolding step that allows disulfide bonds to be reshuffled to form native bridges with native fold. DMSO due to its both hydrophilic and hy-drophobic surface stabilizing and non-ionic detergents, brij35 due its hydro-phobic surface interacting nature found to assist this process.

In contrast, members of other subfamilies of cyclotides, Möbius, kB1 and trypsin inhibitor, MCoTI-II, are characterized by a major two-disulfide bond (2SS) intermediate having native like 3D fold and disulfide connectivity. However, for kB1 it was rather a kinetic trap that needs disulfide bond re-shuffling to a direct precursor (Daly, 2003a; Daly, 2003b) while in MCoTI-II it was a direct precursor (Cemazar, 2006). Scrambled 3SS intermediates have also been observed in oxidative folding pathways of other cystine knot microproteins such as hirudin, (Chang, 2000) tic anticoagulant peptide, (Chang, 1996) amaranthus α-amylase, (Cemazar, 2004; Cemazar, 2003) and potato carboxypeptidase inhibitor (Chang, 1994).

45

6 Folding pathways to lock in the native disulfide bonds (Paper IV)

6.1 Oxidative folding of cyclic cystine knot microproteins

Oxidative folding involves the combination of two interdependent processes, i.e. mutual conformational folding and disulfide bond formation to lock in the single native state (Arolas and Ventura, 2010; Chang, 2010; 2007). The-se disulfide bonds can stabilize the native structure significantly, both by lowering the entropy of the unfolded state and by forming stabilizing interac-tions in the native fold.

Having established the first folding condition for the bracelet cyclotides, it appeared necessary to study the heterogeneity of the folding intermediates accumulating in the folding process of cyclotides. Currently, the oxidative folding pathways of cyclotides are only partly known. Previous studies showed the presence of two disulfide intermediates with native-like fold and disulfide connectivities on the oxidative folding pathways of Möbius, kB1 and trypsin inhibitor, MCoTI-II cyclotides (Cemazar, 2006; Daly, 2003a; Daly, 2003b). The intermediate identified was on-pathway for MCOTI-II but not for kalata B1. In this work, the influence of folding conditions on the heterogeneity and kinetics of folding intermediates that characterized the folding pattern of each subfamily was valued and compared.

6.2 Quantification of heterogeneous folding intermediates

In this study, intermediates accumulated in the oxidative refolding processes were trapped in a time-course fashion by covalent chemical modification of unreacted Cys thiols under low pH condition to quench the folding process and simultaneously alkylate free thiols. Intermediates could be in fully re-duced (R), one-disulfide (1SS), two-disulfide (2SS) or three-disulfide (3SS) form. The alkylation reaction was performed with N-ethylmaleimide (NEM) at pH 3. The increases in mass of NEM trapped intermediates would be n × 125 mg/mmol, where n is the number of free thiols. The alkylated species were then easily separated as a group of 1SS, 2SS, 3SS, N and R, thus assist-

46

ing in tracking and determining the heterogeneity and kinetics of folding. Without using NEM alkylation, differentiation between these species is usu-ally difficult due to their close or overlapping [M+2H]2+ or [M+3H]3+ form in MS analysis, together with isotopic contribution.

Cyclotides containing three disulfide bonds can potentially adopt 75 dif-ferent disulfide isomers (15 × 1SS, 45 × 2SS, 15 × 3SS). All these 75 iso-mers could potentially occur during oxidative folding though most of them are not observed due to unfavorable energetics. Oxidative folding pathway is, therefore, characterized by the heterogeneity, kinetics and structures of these isomers that accumulate along the process of oxidative folding.

For this study, native cyclotides from both subfamilies, Möbius (kB1 and kB2) and bracelet (cyO2), as well as a hybrid cyclotide (kB8) were fully reduced with dithiothreitol (DTT) and their oxidative refolding was initiated under four different folding conditions. The folding conditions were initially optimized using kB1 as a model peptide using NH4HCO3 buffer systmem with or without iPrOH in varying concentrations of redox agents. Four opti-mized buffers suitable for these studies were then selected, i.e. 1) 0.1 M NH4HCO3 containing 2mM GSH/GSSG (standard buffer), 2) 0.1 M NH4HCO3 containing 2mM GSH/GSSG, 50% iPrOH, 3) 0.1 M NH4HCO3, 20% DMSO and 4) modified cyO2 buffer (0.1 Tris-HCl, 1 mM EDTA, 6% Brij35, 35% DMSO, 2 mM cystamine, 2 mM GSH at room temp).

The intermediates accumulated during oxidative refolding processes were quenched and labeled by NEM alkylation of free thiols at various time inter-vals (0.03, 0.5, 1, 2, 4, 6, 24, and 72 h). The resulting samples were analyzed by LC-MS and quantified using area under the curve (AUC) for each species (R, 1SS, 2SS, 3SS and N) obtained by integration of LC-MS signal intensi-ties of selected ions. The time vs. % AUC plot of each species (R, 1SS, 2SS, 3SS, N and R) showed the kinetics, and extent of accumulation of each spe-cies in the folding pathways under specific folding conditions. Typical oxi-dative refolding intermediates accumulated in cyO2 buffer are shown in Figures 16.

The analysis showed that the presence of iPrOH favored formation of na-tive-like species such as 2SS or 3SS, while the presence of increased concen-tration of GSSG favored the formation of species with higher oxidation states such as 3SS, some of which are non-native in the absence of iPrOH. kB1 and kB2 were also efficiently folded in cyO2 buffer reaching about 80% N, although small percentages of 2SS and non-native 3SS accumulated (Fig-ures 16 and 17). Native fold of kB8 in the same buffer was reached equilib-rium with non-native 3SS at about 60% native fold without significantly accumulating 2SS. On the other hand, native cyO2 seems to reach equilibri-um with non-native 3SS in the ratio of 3:7 (Native: non-native).

47

Figure 16. Folding kinetics in cyO2 buffer of: A) kB1, B) kB2, C) kB8 and D) cyO2. N and M in the graph represent native and non-native folded 3SS species, respectively.

Analysis of the oxidative folding pattern of these peptides showed that Mö-bius (kB1 and kB2) have relatively similar folding pattern in most buffer systems while the bracelet cyclotide, cyO2, has different folding profile with maximal native fold in cyO2 buffer. The hybrid peptide, kB8, seems to ex-hibit folding pattern that is intermediate between those of the two subfami-lies, but with slightly higher similarity to the Möbius pattern (Figures 16 and 17).

48

Figure 17. Folding efficiency as judged by native percentage yield under four dif-ferent folding conditions

6.3 The diversity of cyclotide folding pathways Although all cyclotides are of comparable size and have the same number of disulfides, their folding pathways, kinetics and efficiency of native fold for-mation significantly vary in vitro (Figures 16 and 17). As previously report-ed in various studies of disulfide rich peptides, this could be due to differ-ences in degrees and properties of surface exposed residues (Wang, 2009a; Daly, 2003a; Daly, 2003b; Chang, 2000; Creighton, 1997). In Möbius cyclo-tides (kB1 and kB2) surface exposed hydrophobic patches are localized in loops 2, 5 and 6 whereas in bracelets (cyO2) that surface is localized in loops 2 and 3, but bracelets have also hydrophilic patches in loop 5. However, the hybrid (kB8) lacks such large hydrophobic patches and is generally charac-terized by more hydrophilic surfaces. These variations in surface properties likely influence their folding properties.

Therefore, combining the previous oxidative folding studies (Cemazar, 2006; Daly, 2003a; Daly, 2003b) with the results obtained in the work pre-sented in Papers II and IV, the following major pathways can be proposed, in accordance with previously described folding models (Chang, 2010): 1) a mixed folding pathway characterized by heterogeneous intermediates (1SS, 2SS, and 3SS) (Paper IV), but selected 2SS intermediate (Daly, 2003a; Daly, 2003b) adopting native disulfide bonds and native-like structure (the nucleation-condensation model, exemplified by Möbius, such as kB1 and kB2), 2) very few 2SS intermediates (Cemazar, 2006) having native disul-fide bonds and native-like structures (the framework model, exemplified by trypsin inhibitor, MCOT-II), and 3) homogeneous 3SS intermediates (Pa-

49

pers II and IV) comprising primarily non-native structures (the hydrophobic collapse model, exemplified by bracelet, cyO2 ). Details are illustrated in Figure 18. Observations of such specific intermediates were, however, found to be dependent on folding conditions employed.

Figure 18. A) Schematic representation of the energy landscape of cyclic cystine knot microprotein oxidative folding. The energy of the peptide species is displayed as a function of their disulfide connectivity and conformation. MI (red) is the major misfolded intermediate, in addition to other 3SS intermediate in the oxidative refold-ing pathway of cycloviolacin O2, whereas 2SS (light blue) is the major oxidative folding intermediate in folding pathways of Möbius and MCoTI-II. All intermedi-ates eventually converge to native fold (N, green). B) The right hand panel shows detailed folding pathways for each subfamily: Möbius, MCoTI-II, bracelet, (left to right). Folding pathways of Möbius cyclotides (left line) seem to be heterogeneous (yielding 1SS, 2SS and 3SS intermediates) while that of cyO2 (right line) is homog-enous (3SS) in the cyO2 buffer system. The folding of both cyclotides is character-ized by off-pathway intermediates (indicated as *), unlike MCoTI-II (middle line) that has a limited direct precursor on-pathway, indicated as [2SS]. [[ 2SS]] in cyO2 pathways indicates the non-observed but likely occurring 2SS intermediates. Adapted from Cemazar et al with modifications (Cemazar, 2004).

50

7 From native to engineered macrocyclic cystine knotted peptides (Paper V)

7.1 Backbone engineering of the cyclotide scaffold In contrast to most known proteins, cyclotides have a unique surface ex-posed patch of hydrophobic residues since the peptide core is occupied by tightly packed cystine knot, leaving no room for internalization of hydro-phobic residues (Jagadish and Camarero, 2010; Göransson, 2009; Craik, 2006; Daly, 1999b). Cyclotides have wide ranges of bioactivities, together with exceptional stability (Colgrave and Craik, 2004). Bracelet cyclotides, such as cyO2, circulin A, and B are particularly potent in bioactivity tests, such as cytotoxicity and/or anti-HIV activity assays, indicating their poten-tial utility as therapeutics (Burman, 2010; Gunasekera, 2008b; Ireland, 2008; Craik, 2004; Gustafson, 2000; Hallock, 2000; Gustafson, 1994b). However, their in vitro hemolytic activity and the narrow therapeutic index highlight the challenges that needs to be addressed before these peptides can be used as drug lead candidates (Burman, 2010; Gustafson, 2000; Hallock, 2000; Gustafson, 1994b).

As noted in Chapter 5 and 6, bracelet cyclotides have been largely intrac-table for chemical synthesis that mimics the native peptide fold (Gunasekera, 2008a; Tam and Lu, 1997). Consequently, biomolecular engineering and use of this cyclotide subfamily as peptide based drug scaffolds have been hin-dered. Hence, two main challenges need to be addressed: 1) the oxidative folding to obtain peptide with improved yields, built on previously devel-oped methods and 2) removal of the undesired cytotoxicity/hemolytic activi-ty to generate an inert molecular scaffold.

Bracelet cyclotides have several non-conserved residues dispersed in dif-ferent loops as illustrated in Figure 19. The variation in size and composition of the loops might reflect the variation in their structural importance for spe-cific types of biological activities, such as cytotoxicity, antimicrobial, or anti-HIV activity (Gruber, 2008; Lindholm, 2002; Göransson, 1999; Tam, 1999a; Gustafson, 1994b). By systematically investigating the influence of non-conserved residues on the folding and cytotoxicity of the resulting engi-neered-cyclotides we aimed to generate a cyclotide with lower cytotoxic activity than the wild-type, which could potentially be used as a stable inert scaffold for protein engineering. In this work, the polymer-peptide hybridi-zation approach previously used for conotoxin-SIIIA (Green, 2007) was adapted to address such challenges in bracelets.

51

Figure 19. The generalized backbone framework of bracelet cyclotides, showing relatively conserved and relatively variable loci. Backbone spacers represent partly non-conserved loops.

7.2 Rational design strategy using backbone spacers Peptidomimetics has proved to be effective approach for designing analogs of small peptides, such as opiod peptides (Hruby and Agnes, 1999) and so-matostatins (14 amino acids) (Janecka, 2001). However, modification of disulfide rich peptides by the molecular size reduction methods commonly used in peptidomimetics often results in reduction or losses of both bioactivi-ty, and the stability. This is mainly due to removal of the interlocking cystine knots and the essential residues, as observed with ω-conotoxin GVIA (Price-Carter, 1998). An alternative approach has been successfully applied in en-gineering SIIIA-conotoxin (Green, 2007), analogs of neuropeptide Y (Langer, 2001), antibiotic tyrocidine A (Trauger, 2001), and glucagon like peptide-1 (Doyle, 2001), resulting in improved bioactivities and pharmaco-kinetic properties. In this approach, non-essential residues were systemati-cally replaced with inert, flexible backbone spacers improving the peptides’ bioactivity and stability.

In the work reported in Paper V, such replacements were used to explore the impact of non-conserved residues and their replacement on bioactivity and oxidative folding. It is desirable for a drug scaffold to be biologically inert. Therefore, this approach might be useful for transforming cyclotides into inert scaffolds to utilize them as peptide based molecular template. Hence, analogs of bracelet cyclotides, named mer-cyclotides, were designed on the basis of backbone hybridization with inert backbone spacers. The backbone spacers were flexible polymers including aminohexanoic acid (Hex, U), and PEG spacers: 5-amino-3-oxapentanoic acid (Pen, J), and 8-amino-3,6-dioxaoctanoic acid (Oct, B). These building blocks were used to replace non-conserved residues. The non-conserved residues targeted for replacement were selected by comparing related sequences of bracelet cyclo-tides and using cyO2 as a base template, as illustrated in Figure 20. The six

52

designed analogs shown in Figure 20 B and D were then synthesized using Fmoc-SPPS.

Figure 20. Rational design of bracelet cyclotide analogs: A) Sequence comparison of selected bracelet cyclotides, B) Designed analogs based on similarity analysis, C) Building blocks used in analog design, D) 3D structure of cyO2 showing site of modification (pdb of cyO2 is 2KNM). BP is bracelet modified with PEG at loops 3(Oct-Pen) and 5(Pen); BA is bracelet modified with PEG and 6-aminohexanoic acid at loops 3(Oct-Hex) and 5(Pen); BPD is the same as BP but in loop 6, Asn is replaced with Asp; BP26 is bracelet modified with PEG at loop 6 (Oct) and Ile in loop 2 is also replaced with Gly; BP26J is the bracelet modified with PEG at loop 6(Pen) and also Ile at loop 2 is replaced with G, and BP3 is bracelet modified with PEG only at loop 3(Oct-Pen). Residues selected for replacement are highlighted in different colors. cyA, cyO1, cyO3, and cyO4 in sequence comparison are abbrevia-tions for cyclopsychotride A, cycloviolacin O1, O2, O3 and O4, respectively.

7.3 Synthesis, biological and structural evaluation of mer-cyclotides

In the previously described cyclization of cyO2, thioesterification of protect-ed peptides was used, followed by cyclization in solution. For the synthesis of the mer-cyclotides, an approach with fewer steps that included one-pot thioesterification and cyclization was adapted. All the designed hybrids were assembled on rink amide MBHA resin following Fmoc-SPPS, as previously reported (Harpaz, 2010; Pentelute, 2010; Tiefenbrunn, 2010; Blanco-Canosa and Dawson, 2008) with some important modifications. In Blanco-Canosa and Dawson approach, the removable carboxyl terminal activating linker was assembled using mono Fmoc-3,4-diaminobenzoic acid (Fmoc-Dbz),

53

which need to be synthesized since this compound is not commercially available. Hence, this approach significantly prolongs the overall synthesis of the peptides. In the report presented in Paper V, we showed a convenient method of loading the linker (Dbz) using commercially available diFmoc-3,4-diaminobenzoic acid (diFmoc-Dbz).

In the synthesis of mer-cyclotides, Dbz was coupled as a first residue, which was then used for carboxyl terminal activation and subsequent solu-tion phase in situ thioesterification and cyclization of peptides, as illustrated in Figure 21. Coupling of Gly and polymer-type-amino acids were per-formed at two equivalents (0.5 mmol) relative to reaction scale (0.25 mmol), while all other residues (including diFmoc-Dbz) were assembled following the established procedure (Blanco-Canosa and Dawson, 2008). The fully assembled hybrid analogs were acylated on the resin using 4-nitrophenylchloroformate over 55 min and then activated with DIPEA over 20 min, resulting in activated N-acylbenzimidazolone (Nbz). The activated mer-cyclotide-Nbz was cleaved from the resin, thioesterified in situ (without thioester isolation) and natively ligated to form reduced cyclic peptide at neutral pH. Typical LC-MS traces of in situ thiolysis and cyclization prod-ucts of crude mer-cyclotide-Nbz are shown in Figure 22. All the analogs were efficiently activated and cyclized, with ∼10% RP-HPLC purified over-all yields of cyclized mer-cyclotides, based on resin loading.

54

Figure 21. Fmoc-SPPS of mer-cyclotides and subsequent acylation, activation and cyclization. The shaded region on two Fmoc groups shows the newly employed Dbz coupling strategy. The rest elliptically shaded region highlights the activated groups.

The resulting reduced cyclic peptides were oxidatively folded in ordinary buffer systems containing such as 0.1 M NH4HCO3/2 mM GSH/2 mM GSSG (standard buffer) with improved yields relative to template (native cyO2). Analogs BP, BPD and BA, with replaced loops 3 and loop 5 were folded with two major isomers in the ratio of ∼3 e: 2 l, hereafter referred to as early(e) and late(l) eluting, respectively (folding of BP is shown in Figure 23A). Evaluation of the cytotoxic activity of selected analogs, both isomers of BP and BA, indicated that they were devoid of the potent cytotoxicity associated with native cyO2 (0.2 μM) (Lindholm, 2002), at a concentration tested: BP (BPe at 38.5 (192-fold relative to native cyO2), BPl at 25.6 μM (128-fold)) and BA (BAe at 15.3 μM (76-fold), BAl at 7.3 μM (36-fold)).

55

Figure 22. LC-MS traces of cyclized crude A) BP26J, and B) BP26 with MS part on the right hand side (base peak ions shown). * Indicates the desired cyclic product.

To further explore the possibility of generating the single native-like ana-logs, an approach recently reported by Gunasekera et al was adapted (Gunasekera, 2008a). In this report the replacement of Ile in loop 2 with Gly, together with insertion of an extra Thr residue in loop 6 preceding Tyr resi-due of bracelet cyclotide cyO1, resulted in efficient folding. In the study described in Paper V, cyO2 was used as a framework, whereby Ile in loop 2 was replaced with Gly and (Thr)-Tyr-Arg sequence was replaced with Oct (analog BP26) or only (Thr)-Tyr with Pen (analog BP26J), where (Thr) indi-cates the extra residue inserted as reported by Gunasekera et al (Figure 20C and D). Both analogs were efficiently folded to yield one major product un-der standard buffer system NH4HCO3, 2 mM GSH/GSSG) (BP26J) or under standard buffer containing 50% iPr-OH (BP26) but with retention time in RP-HPLC earlier than their respective reduced counterparts (Figure 23B). Evaluation of the cytotoxicity of the analog BP26 at a maximum concentra-tion of 50 μM (250-fold) indicated that it has no cytotoxic activity as com-pared to native cyO2.

In order to have further insights into the influences of loop 3 and 5 re-placements, another analog BP3 with only loop 3 replaced with PEG (Oct-Pen) was designed (Figure 20). This analog was found to fold under standard buffers with or without 50% iPr-OH, similar to BP, BA and BPD, but accu-mulating one major product that elutes late in RP-HPLC (Figure 23C).

56

Figure 23. Oxidative folding product of: A) BP, B) BP26, and C) BP3, for each analog the left panel shows RP-HPLC traces for reduced (R) whereas right panel shows traces of corresponding folded form (f). Thiols were removed before folding.

Further 1D 1H NMR analysis of selected analogs, BP (both isomers), BP26 and BP26J showed that amide region of BP26, BP26J and early eluting iso-mers of BP (BPe) exhibited spectra with several overlapping peaks, whereas late eluting isomers of BP (BPl) exhibited well resolved 1D 1H NMR (Figure 24A). These findings indicated that the late eluting isomers of BP, BPD and BA, which showed the same folding pattern, have a well-folded structure. Furthermore, 2D 1H NMR analyses comprising TOCSY and NOESY at 298K on both isomers of BP showed that unlike early eluting form (BPe), the late eluting (BPl) isomer of BP had a well dispersed narrow line NMR peaks in the amide region. The Hα chemical shift deviation from random coil chemical shift for BPl and native cyO2 is shown in Figure 24B. Conse-quently, BP3 (which resulted in one major late eluting product) also likely have well-structured native-like fold although this need to be confirmed by further analysis.

57

Figure 24. A) 1D 1HNMR analysis of BP26, BP26J and both variants of BP, B) Hα chemical shift deviation of BP (late eluting variant, BPl) from random coil in com-parison with native cyO2. All the NMR analysis was performed on Varian NMR (800 MHz) at the Swedish NMR Centre. 2D 1H NMR comprising TOCSY with mixing time of 80 ms and NOESY with mixing time of 300 ms at 298k were per-formed on both isomers of BP. Spectra were acquired with 4096 data points in f2 and 512 increments in f1 dimension. Water suppression was achieved using a modi-fied WATERGATE sequence (Piotto, 1992). Chemical shifts were referenced to water signal at 4.76 ppm. Note: dark bar represents BPl (with upper sequence) while light grey bar represents cyO2 (with lower sequence).

The absence of cytotoxic activity with BP26 might indicate that either Tyr residue in loop 6 or the native conformational folding might be essential for bioactivity since this analog has no well-structured peaks in the amide region in 1D 1H NMR analysis. In addition, previous studies have shown that modi-

58

fication of both Lys and an Arg residues in cyO2 has slight influence on cytotoxicity of the native peptide (Herrmann, 2006).

The loss of activity of BP and BA, despite having well folded structures, might indicate that residues in loop 3 and probably two residues, KS in loop 5 is essential to preserve strong native cytotoxicity (Göransson, 2009; Herrmann, 2006). Residues in loop 3 (I14-Ser15) were found to form stabi-lizing network of hydrogen bonding with Glu6 residue in loop 1 contributing to strong cytotoxicity (Göransson, 2009; Herrmann, 2006). Therefore, loop 3 could be a target for engineering out the undesired cytotoxic/hemolytic ac-tivity. In contrast, to exploit such native bioactivity, it is clearly essential to preserve loop 3.

7.4 Conclusion Reducing the cytotoxicity of native cyclotides by hybridization with inert polymers to generate mer-cyclotides together with amenability of the result-ing mer-cyclotides to synthetic approach is a promising method in exploiting diverse family of cyclotides. Furthermore, the approach of practically easy direct coupling of commercially available diFmoc-Dbz linker in Fmoc-SPPS and the relatively short final processing steps, i.e. acylation/activation as well as in situ thiolysis and cyclization make these methods an attractive system of combinations that can be used to reduce unwanted bioactivity for utilization of bracelets as inert scaffolds.

59

8 Concluding Remarks

This thesis describes attempts to engineer the ultra-stable cystine knot framework of microproteins and associated design, chemical synthesis and structural studies. The work included synthesis and oxidative folding of acy-clic and macrocyclic cystine-knotted peptides with further biological and structural evaluation. A novel acyclic cystine knot microprotein with 35 amino acids, named tripatide, was discovered. In the first phase of the stud-ies, Fmoc-SPPS of tripatide, followed by oxidative folding, showed that this peptide readily amenable to standard peptide synthesis. Solution NMR anal-ysis of this peptide, which has a unique sequence, showed that it is structur-ally related to inhibitor cystine knot peptides.

The work reported in this thesis mainly included the synthesis and engi-neering of bracelet cyclotides. For this purpose, methods for synthesizing the model peptide, cycloviolacin O2, including protected peptide synthe-sis/thioesterification and cyclization, were initially developed. The reduced cyclic peptide was then oxidatively folded under optimized conditions, in-cluding the presence of co-solvent, DMSO and non-ionic detergent, Brij35 at 3°C, affording improved yield of more than 50% of the native peptide.

Following that study, the heterogeneity and efficiency of oxidative fold-ing pathways of Möbius, hybrid and bracelet cyclotides were evaluated in four different buffers. The results showed that Möbius cyclotides have heter-ogeneous folding pathways, accumulating one-, two-, and three- disulfide bond intermediates while bracelets have homogenous three disulfide bond intermediates. However, the major intermediates accumulating in the folding pathways vary depend on the reaction conditions.

Building on previous studies of cycloviolacin O2 synthesis and oxidative folding, the internal factors contributing to inefficient folding and potential toxicity of bracelet cyclotides were evaluated via bracelet-polymer hybridi-zation. Interestingly, hybrid analogs designed by replacing non-conserved residues in loop 3 (ISSAI) and 5 (KS) were oxidatively folded, resulting in two isomers, neither of which exhibited cytotoxicity at the maximum con-centration tested for respective products. Furthermore, 1D and 2D 1HNMR analysis showed that in each case one of the two isomers, which eluted late in RP-HPLC, were well structured.

In the future am eager to see cyclotides, particularly bracelet cyclotides, with improved bioactivity profiles, as these peptides found to be very attrac-tive with diverse sequences and potent in vitro bioactivities that are also

60

associated with undesired bioactivities such as hemolytic and cytotoxic ac-tivities. With the currently available techniques the solution to oxidative folding and removal of undesired cytotoxic activity doe not seem to be straightforward. Some deviation such as systematic hybridization of these peptides with inert polymers might be required. However, with the ongoing improvements in techniques for the solid phase synthesis of cyclic peptides, and emergency of biomolecular engineering, an alternative approaches for solving these problems may (hopefully) soon be met

61

9 Summary of popular Science

Peptides and proteins are polymers composed of amino acids linked in chains by amide bonds that are found in all known organisms, in which they provide structural support and participate in diverse functions including as hormones, neuropeptides, and antibiotics. Humans have been using these peptides as therapeutics exploiting low toxicity nature of the peptides since they are already in many functions of the human body and they also readily metabolized to amino acid building blocks. Such peptides have been isolated in the laboratory from various plants and animals.

Most peptides and proteins have both an amino end and a carboxyl end. Consequently, most of them are readily degraded to small peptide fragments by enzymes in the gastrointestinal tracts. This generally leads to loss of their natural bioactivities. However, cyclic peptides, in which both an amino end and a carboxyl end are connected, are much more resistant to enzymatic degradation.

This thesis focuses on a class of peptides called cyclotides. The discovery of cyclotides was started in the 1970's when Lorents Gran, a Norwegian physician, reported that African women in Zaire used a herb to accelerate childbirth. The local native Zairian used to call this plant ‘kalata-kalata’ and the scientific name was Oldenlandia affinis. Lorents Gran took this plant back to Norway and isolated a substance, which was found to promote uter-ine contractions. This was the first discovery of cyclotide. Since then, nu-merous cyclotides have been isolated from Oldenlandia affinis and various other plants. The cyclotides can also produced by chemical synthesis and molecular biology approaches.

Cyclotides comprise a class of small proteins with special properties that distinguishes them from other peptides and proteins. As the name implies, cyclotides are cyclic and thus their amino acid chain lack the amino end and the carboxyl end. This circular chain of amino acids is further tied together with three disulfide bonds. The cyclization and further tying together with the extra disulfide bonds of the amino acid chain make them compact and rigid. These properies make them extremely stable to heat, chemical and enzymatic digestion. Such characteristics are very unusual for proteins and peptides that have both ends open and hence are readily digested by intesti-nal enzymes.

These properties attracted the attention of many researchers engaged in peptide-based drug discovery. In addition to isolation from plants, several of

62

these peptides were synthesized in the laboratory for further studies. Merri-field (1963) pioneered a method for the synthesizing peptides, which was called solid phase peptide synthesis. In the solid phase peptide synthesis, the reactions involved are performed using a solid support on which the first residue is attached. The subsequent amino acids are assembled stepwise, by mixing a solution of appropriate amino acids with the solid support and washing the product attached to the solid support with a suitable solvent.

In the reactions used to produce cyclotides, the peptides need to be as-sembled on a solid support to generate, ultimately, a peptide with a C-terminal thioester and N-terminal cysteine. Following complete assembly of such a peptide, cyclization can be performed in aqueous solution at neural pH. With this approach they can be efficiently cyclized. Once cyclic peptides have been obtained, the next step towards producing cyclotide is to fold them to a compact and rigid state with characteristics disulfide bonds. This has been found to be challenging for certain group of cyclotides.

Cyclotides have broad potential applications. They have diverse natural biological activities including antimicrobial, anti-HIV, enzyme inhibitory and insecticidal activities. The stability of cyclotides also offer a great poten-tial that can be utilized to stabilize non-natural bioactive peptides that are not stable on their own. As a result of these potential applications, interests in utilization of these peptides is remarkably increasing with time.

63

10 Acknowledgments

First and foremost I praise the only Almighty God for his miraculous help in undertaking my work and covering this lengthy journey peacefully. This work was carried at the Division of Pharmacognosy, Department of Medicinal Chemistry, Faculty of Pharmacy, Uppsala University by the fund-ing from the Swedish International Development Cooperation Agency (SIDA), the Department for Research cooperation (SAREC). I would especially like to thank the following people: My supervisor, Dr. Ulf Göransson for your support, guidance and teaching throughout my research work. I am very happy for letting me continue with my area of research interest and assisting me in doing so. Thank you for all the help you gave me throughout my stay in Sweden. My assistant supervisor Prof. Lars Bohlin thank you for assisting me to start up my PhD studies at your division and for creating excellent network be-tween Addis Ababa University and Uppsala University. Special thanks to Prof. Tsige Gebre-mariam, Dr. Kaleab Asres and Dr. Nigussu Mekonnen for your extremely valuable effort to design and start up the Sida/SAREC project. Thank you all staff members at International Science Programme: Hussein Aminaey, Assoc. Prof. Peter Sundin, Dr. Linnea Sjöblom, Malin Åkerblom, Pravina Gajjar and others. I would like to express my deepest gratitude to you all. All the former and present members of cyclotide group: Per Claeson for setting up the peptide research Lab in the division, Erika svangård for cre-ating good cyclotide impression, Anders Herrmann for teaching me HPLC and MS, Sunithi Gunasekera for teaching me NMR and evaluating my thesis, Mariamawit Yeshak and Habtamu Gabisa and their children for all the great time we had together, SungKyu Park for sharing office room with me, and evaluating my thesis, Robert Burman for helping in processing my thesis and evaluating it, Stefan Svahn,and Adam Strömstedt for becoming our cyclotide group and expanding antimicrobial lab, Aida, Rehan, Sohaib,

64

and Eman for joining our research group, Walid Madian for all your great help, in this connection I am privileged to appreciate all your hard work and commitment. All other members of the research group in the division: Anders Backlund, Ulrika Melin, Sonny Larsson, Martin Sjögren, Erik Hedner (for all your funs and jokes), Catarina Ekenäs and Josefin Rosén, Cecilia Alsmark, Christina Wedén and Elisabet Vikeved, Jenny Felth (for your nice sup-port during my application for defense), it has been a great pleasure to meet you all. Hesham El-Seedi and his family for taking me with you for journey to Mariehamn by boat and also showing me the direction in Uppsala. Working staff at Swedish NMR Centre thank you for helping with NMR work. Kerstin Ståhlberg, Maj Blad, Gunilla Eriksson, Anna-Maria Andersson, and Eva Strömberg for keeping thing in such a wonderful order. The undergraduate students who involved in the project: Jonas, and Lisa Albrecht, it was great to work with you. All the people at the Division of Analytical Pharmaceutical Chemistry; Curt, Douglas, Per, Ylva, Olle, Henrik, Matilda, Annica, Marika, Anita, Jacob, Victoria, Alex and Axel, it has been my pleasure meeting you all. To my collaborator in Australia, David Craik for your wonderful cyclotide research and making the scientific community aware of the cyclotide im-portance, Richard Clark and Johan Rosengren for all the help with NMR and the knowledge I got from such help, Björn Hellman for collaborating with us. Special thanks to my Swedish families of Ethiopian origin Dr. Ararso Etana and his wife Ayelech Etefa and their wonderful son Yohannes. It is beyond my imagination to describe all the help and guidance you gave my wife and me not only during hard times but also through all our stay in Swe-den. Thank you Sheke Abdela and his wife, Zainab; Haji Roba and his wife, Halo; Halima; Abe and his wife; Abaa Biyaa and his family; Fatu-ma, Hedata Gudata; Zekarias and his family, thank you all for the great time we had together. Colleagues and friends from Addis ababa University: Nigus Dessalew, Workalemahu Mikre, Dr. Ariaya Hymete, Dr. Ephrem Engidawork and Henok Baye, work hard and expand your field.

65

Our friends Girma Badhadha, and his wife Hiwot Amanu, and their daughter Soliyana. Thank you for all the great time we had together, keep on going soon you will be a doctor too. My friend Jemal Dema, and his wife, Makida and their daughter Atnun. We had a great time together throughout home country to Sweden. Former Cyslo member friends: Jema Haji, Dereje Beyene, Mengistu Urge, Husein Komicha, Diriba Muleleta and all the rest, thank you all for such amazing Saturday get together. My Ethiopian friends and Families My close friends, Girma Eshetu and his wife, Mesu, and their wonderful children. Thank you for all the help. I would like to express my deepest gratitude to my brother Zenebe, and his wife Kebu and their wonderful children. You have always been there and supported me throughout my studies, Also My brothers Zewdu and his fam-ily, Shiferaw and his family, my Sisters and their families, Mulunesh, Taji, (Aberash) Tadelech, (Bekelech) Ketema. Thank you for all your great help. My family, Daba Melka and Tujare Beyecha for welcoming me into your wonderful family. Haimanot and Kumsa and their children, Genet and Gemechu and their son; Asheber, Melkamu, Gelana, Etabez, Rawda, Zenu and her husband Asfaw, and their son Yohannes it is great to have you all as family and friends. The last but not the least comes for my lovely wife, Argane (Takkiti). I do not have words to describe all the things you have done for me. I believe a few of the living human being become lucky and successful in such a chal-lenging and hard times, you and me have passed through. You have unwea-ryingly waited all these times, now, you should be proud for seeing the suc-cess of your work too.

66

11 References

Arolas, J.L., Aviles, F.X., Chang, J.-Y., and Ventura, S. (2006). Folding of small disulfide-rich proteins: clarifying the puzzle. Trends Biochem Sci 31, 292-301.

Arolas, J.L., and Ventura, S. (2010). Protease Inhibitors as Models for the Study of Oxidative Folding. antioxid redox sign 14, 97-112.

Atherton, E., Fox, H., Harkiss, D., and Sheppard, R.C. (1978). Application of polyamide resins to polypeptide synthesis: an improved synthesis of [small beta]-endorphin using fluorenylmethoxycarbonylamino-acids. J Chem Soc, Chem Commun, 539-540.

Austin, J., Wang, W., Puttamadappa, S., Shekhtman, A., and Camarero, J.A. (2009). Biosynthesis and Biological Screening of a Genetically Encoded Library Based on the Cyclotide MCoTI-I. Chembiochem 10, 2663-2670.

Backes, B.J., and Ellman, J.A. (1994). Carbon-Carbon Bond-Forming Methods on Solid Support. Utilization of Kenner's "Safety-Catch" Linker. J Am Chem Soc 116, 11171-11172.

Backes, B.J., and Ellman, J.A. (1999). An Alkanesulfonamide “Safety-Catch” Linker for Solid-Phase Synthesis. J Org Chem 64 2322-2330

Backes, B.J., Virgilio, A.A., and Ellman, J.A. (1996). Activation Method to Prepare a Highly Reactive Acylsulfonamide ‚ 'Safety-Catch' Linker for Solid-Phase Synthesis. J Am Chem Soc 118, 3055-3056.

Barlos, K., Gatos, D., and Schäfer, W. (1991). Synthesis of Prothymosin α (ProTα)—a Protein Consisting of 109 Amino Acid Residues. Angew Chem, Int Ed Engl 30, 590-593.

Barry, D.G., Daly, N.L., Bokesch, H.R., Gustafson, K.R., and Craik, D.J. (2004). Solution structure of the cyclotide palicourein: implications for the development of a pharmaceutical framework. Structure 12, 85-94.

Barry, D.G., Daly, N.L., Clark, R.J., Sando, L., and Craik, D.J. (2003). Linearization of a naturally occurring circular protein maintains structure but eliminates hemolytic activity. Biochemistry 42, 6688-6695.

Blanco-Canosa, J., and Dawson, P. (2008). An Efficient Fmoc-SPPS Approach for the Generation of Thioester Peptide Precursors for Use in Native Chemical Ligation. Angew Chem, Int Ed Engl 47, 6851–6855.

Broekaert, I., Lee, H.I., Kush, A., Chua, N.H., and Raikhel, N. (1990). Wound-induced accumulation of mRNA containing a hevein sequence in laticifers of rubber tree (Hevea brasiliensis). Proc Natl Acad Sci U S A 87, 7633–7637.

Broekaert, W.F., Marien, W., Terras, F.R.G., De Bolle, M.F.C., Proost, P., Van Damme, J., Dillen, L., Claeys, M., and Rees, S.B. (1992).

67

Antimicrobial peptides from Amaranthus caudatus seeds with sequence homology to the cysteine/glycine-rich domain of chitin-binding proteins. Biochemistry 31, 4308-4314.

Broussalis, A.M., Göransson, U., Coussio, J.D., Ferraro, G., Martino, V., and Claeson, P. (2001). First cyclotide from Hybanthus (Violaceae). Phytochemistry 58, 47-51.

Bu, X., Xie, G., Law, C.W., and Guo, Z. (2002). An improved deblocking agent for direct Fmoc solid-phase synthesis of peptide thioesters. Tetrahedron Lett 43, 2419-2422.

Bulaj, G., Kortemme, T., and Goldenberg, D.P. (1998). Ionization-reactivity relationships for cysteine thiols in polypeptides. Biochemistry 37, 8965-8972.

Burman, R., Gruber, C.W., Rizzardi, K., Herrmann, A., Craik, D.J., Gupta, M.P., and Goransson, U. (2009). Cyclotide proteins and precursors from the genus Gloeospermum: Filling a blank spot in the cyclotide map of Violaceae. Phytochemistry.

Burman, R., Svedlund, E., Felth, J., Hassan, S., Herrmann, A., Clark, R.J., Craik, D.J., Bohlin, L., Claeson, P., Göransson, U., et al. (2010). Evaluation of toxicity and antitumor activity of cycloviolacin O2 in mice. Pept Sci 94, 626-634.

Camarero, J.A., Cotton, G.J., Adeva, A., and Muir, T.W. (1998a). Chemical ligation of unprotected peptides directly from a solid support. J Pept Res 51, 303-316.

Camarero, J.A., Hackel, B.J., de Yoreo, J.J., and Mitchell, A.R. (2004). Fmoc-Based Synthesis of Peptide α-Thioesters Using an Aryl Hydrazine Support. J Org Chem 69, 4145-4151.

Camarero, J.A., Kimura, R.H., Woo, Y.H., Shekhtman, A., and Cantor, J. (2007). Biosynthesis of a fully functional cyclotide inside living bacterial cells. Chembiochem 8, 1363-1366.

Camarero, J.A., and Muir, T.W. (1997). Chemoselective backbone cyclization of unprotected peptides. Chem Commun, 1369-1370.

Camarero, J.A., and Muir, T.W. (1999). Biosynthesis of a Head-to-Tail Cyclized Protein with Improved Biological Activity. J Am Chem Soc 121, 5597-5598.

Camarero, J.A., Pavel, J., and Muir, T.W. (1998b). Chemical Synthesis of a Circular Protein Domain: Evidence for Folding-Assisted Cyclization. Angew Chem, Int Ed Engl 37, 347-349.

Carpenter, K.A., Weltrowska, G., Wilkes, B.C., Schmidt, R., and Schiller, P.W. (1994). Spontaneous Diketopiperazine Formation via End-to-End Cyclization of a Nonactivated Linear Tripeptide: An Unusual Chemical Reaction. J Am Chem Soc 116, 8450–8458.

Carpino, L.A., and Han, G.Y. (1972). 9-Fluorenylmethoxycarbonyl amino-protecting group. J Org Chem 37, 3404-3409.

Cemazar, M., Daly, N.L., Häggblad, S., Lo, K.P., Yulyaningsih, E., and Craik, D.J. (2006). Knots in rings. The circular knotted protein Momordica cochinchinensis trypsin inhibitor-II folds via a stable two-disulfide intermediate. J Biol Chem 281, 8224-8232.

68

Cemazar, M., Gruber, C.W., and Craik, D.J. (2008). Oxidative folding of cyclic cystine knot proteins. antioxid redox sign 10, 103-111.

Cemazar, M., Zahariev, S., Lopez, J.J., Carugo, O., Jones, J.A., Hore, P.J., and Pongor, S. (2003). Oxidative folding intermediates with nonnative disulfide bridges between adjacent cysteine residues. Proc Natl Acad Sci U S A 100, 5754-5759.

Cemazar, M., Zahariev, S., Pongor, S., and Hore, P.J. (2004). Oxidative folding of Amaranthus alpha-amylase inhibitor: disulfide bond formation and conformational folding. J Biol Chem 279, 16697-16705.

Chan, W.C., and White, P.D. (2000). Basic procedures. In Fmoc Solid Phase Peptide Synthesis: A Practical Approach (Practical Approach Series), W.C. Chan, and P.D. White, eds. (Oxford University Press), pp. 41-75.

Chang, J. (2007). Diversity of Folding Pathways and Folding Models of Disulfide Proteins. antioxid redox sign 10, 171-178.

Chang, J. (2010). Distinct Folding Pathways of Two Homologous Disulfide Proteins: Bovine Pancreatic Trypsin Inhibitor and Tick Anticoagulant Peptide. antioxid redox sign 14, 127-135.

Chang, J.Y. (1996). The disulfide folding pathway of tick anticoagulant peptide (TAP), a Kunitz-type inhibitor structurally homologous to BPTI. Biochemistry 35, 11702-11709.

Chang, J.Y., Canals, F., Schindler, P., Querol, E., and Aviles, F.X. (1994). The disulfide folding pathway of potato carboxypeptidase inhibitor. J Biol Chem 269, 22087-22094.

Chang, J.Y., Li, L., and Bulychev, A. (2000). The underlying mechanism for the diversity of disulfide folding pathways. J Biol Chem 275, 8287-8289.

Claeson, P., Göransson, U., Johansson, S., Luijendijk, T., and Bohlin, L. (1998). Fractionation protocol for the isolation of polypeptides from plant biomass. J Nat Prod 61, 77-81.

Clark, R.J., Akcan, M., Kaas, Q., Daly, N.L., and Craik, D.J. (2011). Cyclization of conotoxins to improve their biopharmaceutical properties. Toxicon In Press, Corrected Proof, doi:10.1016/j.physletb.2003.1010.1071.

Clark, R.J., and Craik, D.J. (2010). Invited reviewnative chemical ligation applied to the synthesis and bioengineering of circular peptides and proteins. Pept Sci 94, 414-422.

Clark, R.J., Jensen, J., Nevin, S.T., Callaghan, B.P., Adams, D.J., and Craik, D.J. (2010). The Engineering of an Orally Active Conotoxin for the Treatment of Neuropathic Pain. Angew Chem, Int Ed Engl 49, 6545-6548.

Clippingdale, A.B., Barrow, C.J., and Wade, J.D. (2000). Peptide thioester preparation by Fmoc solid phase peptide synthesis for use in native chemical ligation. J Pept Sci 6, 225-234.

Colgrave, M.L., and Craik, D.J. (2004). Thermal, chemical, and enzymatic stability of the cyclotide kalata B1: the importance of the cyclic cystine knot. Biochemistry 43, 5965-5975.

69

Craik, D.J., Cemazar, M., Wang, C.K., and Daly, N.L. (2006). The cyclotide family of circular miniproteins: Nature's combinatorial peptide template. Biopolymers 84, 250-266.

Craik, D.J., and Daly, N.L. (2005). Oxidative folding of the cystine knot motif in cyclotide proteins. Protein Pept Lett 12, 147-152.

Craik, D.J., Daly, N.L., Bond, T., and Waine, C. (1999). Plant cyclotides: A unique family of cyclic and knotted proteins that defines the cyclic cystine knot structural motif. J Mol Biol 294, 1327-1336.

Craik, D.J., Daly, N.L., Mulvenna, J., Plan, M.R., and Trabi, M. (2004). Discovery, structure and biological activities of the cyclotides. Curr Protein Pept Sci 5, 297-315.

Craik, D.J., Daly, N.L., and Waine, C. (2001). The cystine knot motif in toxins and implications for drug design. Toxicon 39, 43-60.

Creighton, T.E. (1997). Protein folding coupled to disulphide bond formation. Biol Chem 378, 731-744.

Daly, N.L., Clark, R.J., and Craik, D.J. (2003a). Disulfide folding pathways of cystine knot proteins. Tying the knot within the circular backbone of the cyclotides. J Biol Chem 278, 6314-6322.

Daly, N.L., Clark, R.J., Göransson, U., and Craik, D.J. (2003b). Diversity in the disulfide folding pathways of cystine knot peptides. Lett Pept Sci 10, 523-531.

Daly, N.L., Clark, R.J., Plan, M.R., and Craik, D.J. (2006). Kalata B8, a novel antiviral circular protein, exhibits conformational flexibility in the cystine knot motif. Biochem J 393, 619-626.

Daly, N.L., and Craik, D.J. (2000). Acyclic permutants of naturally occurring cyclic proteins - Characterization of cystine knot and beta-sheet formation in the macrocyclic polypeptide kalata B1. J Biol Chem 275, 19068-19075.

Daly, N.L., Gustafson, K.R., and Craik, D.J. (2004). The role of the cyclic peptide backbone in the anti-HIV activity of the cyclotide kalata B1. FEBS Lett 574, 69-72.

Daly, N.L., Koltay, A., Gustafson, K.R., Boyd, M.R., Casas-Finet, J.R., and Craik, D.J. (1999a). Solution structure by NMR of circulin A: A macrocyclic knotted peptide having anti-HIV activity. J Mol Biol 285, 333-345.

Daly, N.L., Love, S., Alewood, P.F., and Craik, D.J. (1999b). Chemical synthesis and folding pathways of large cyclic polypeptides: studies of the cystine knot polypeptide kalata B1. Biochemistry 38, 10606-10614.

Dawson, P.E., Muir, T.W., Clark-Lewis, I., and Kent, S.B. (1994). Synthesis of proteins by native chemical ligation. Science 266, 776-779.

DeLa Cruz, R., Whitby, F.G., Buczek, O., and Bulaj, G. (2003). Detergent-assisted oxidative folding of delta-conotoxins. J Pept Res 61, 202-212.

Dirksen, A., and Dawson, P.E. (2008). Expanding the scope of chemoselective peptide ligations in chemical biology. Curr Opin Chem Biol 12, 760-766.

70

Doyle, M.E. (2001). Insertion of an N-Terminal 6-Aminohexanoic Acid after the 7 Amino Acid Position of Glucagon-Like Peptide-1 Produces a Long-Acting Hypoglycemic Agent. Endocrinology 142, 4462-4462.

Dutton, J.L., Renda, R.F., Waine, C., Clark, R.J., Daly, N.L., Jennings, C.V., Anderson, M.A., and Craik, D.J. (2004). Conserved Structural and Sequence Elements Implicated in the Processing of Gene-encoded Circular Proteins. J Biol Chem 279, 46858-46867.

Edwards, C.M.B., Cohen, M.A., and Bloom, S.R. (1999). Peptides as drugs. QJM 92, 1-4.

Eggelkraut-Gottanka, R., Klose, A., Beck-Sickinger, A.G., and Beyermann, M. (2003). Peptide αthioester formation using standard Fmoc-chemistry Tetrahedron Lett 44, 3551-3554.

Escher, S.E., Klüver, E., and Adermann, K. (2001). Fmoc-based synthesis of the human CC chemokine CCL14/HCC-1 by SPPS and native chemical ligation. Lett Pept Sci 8, 349-357.

Friedman, M., and Koenig, N.H. (1971). Effect of Dimethyl Sulfoxide on Chemical and Physical Properties of Wool. Text Res J 41, 605-609.

Fujiwara Y, A.K., Kiso Y (1994). Racemization-free synthesis of C-terminal cysteine-peptide using 2-chlorotrityl resin. Chem Pharm Bull (Tokyo) 42, 724-726.

Futaki, S., Sogawa, K., Maruyama, J., Asahara, T., Niwa, M., and Hojo, H. (1997). Preparation of Peptide Thioesters using Fmoc-Solid-Phase Peptide Synthesis and its Application to the Construction of a Template-Assembled Synthetic Protein (TASP). Tetrahedron Lett 38, 6237-6240.

Garcia, A.E., and Camarero, J.A. (2010). Biological Activities of Natural and Engineered Cyclotides, a Novel Molecular Scaffold for Peptide-Based Therapeutics Curr Mol Pharmacol 3, 153-163.

Garcia-Martin, F., Quintanar-Audelo, M., Garcia-Ramos, Y., Cruz, L.J., Gravel, C., Furic, R., Cote, S., Tulla-Puche, J., and Albericio, F. (2006). ChemMatrix, a Poly(ethylene glycol)-Based Support for the Solid-Phase Synthesis of Complex Peptides. J Comb Chem 8, 213-220.

Gerlach, S.L., Burman, R., Bohlin, L., Mondal, D., and Göransson, U. (2010). Isolation, Characterization, and Bioactivity of Cyclotides from the Micronesian Plant Psychotria leptothyrsa. Journal of Natural Products 73, 1207-1213.

Gillon, A.D., Saska, I., Jennings, C.V., Guarino, R.F., Craik, D.J., and Anderson, M.A. (2008). Biosynthesis of circular proteins in plants. Plant J 53, 505-515.

Göransson, U., and Craik, D.J. (2003). Disulfide mapping of the cyclotide kalata B1: chemical proof of the cyclic cystine knot motif. J Biol Chem 278, 48188-48196.

Göransson, U., Herrmann, A., Burman, R., Haugaard-Jonsson, L.M., and Rosengren, K.J. (2009). The conserved glu in the cyclotide cycloviolacin O2 has a key structural role. Chembiochem 10, 2354-2360.

71

Göransson, U., Luijendijk, T., Johansson, S., Bohlin, L., and Claeson, P. (1999). Seven novel macrocyclic polypeptides from Viola arvensis. J Nat Prod 62, 283-286.

Göransson, U., Sjögren, M., Svangård, E., Claeson, P., and Bohlin, L. (2004). Reversible antifouling effect of the cyclotide cycloviolacin O2 against barnacles. J Nat Prod 67, 1287-1290.

Gracy, J., Le-Nguyen, D., Gelly, J., Kaas, Q., Heitz, A., and Chiche, L. (2008). KNOTTIN: the knottin or inhibitor cystine knot scaffold in 2007. Nucleic Acids Res 36, D314-D319.

Gran, L. (1973a). Isolation of oxytocic peptides from Oldenlandia affinis by solvent extraction of tetraphenylborate complexes and chromatography on Sephadex LH-20. Lloydia 36, 207-208.

Gran, L. (1973b). On the effect of a polypeptide isolated from "Kalata-kalata" (Oldenlandia affinis DC) on the oestrogen dominated uterus. Acta Pharmacol Toxicol 33, 400-408.

Gran, L. (1973c). Oxytocic principles of Oldenlandia affinis. Lloydia 36, 174-178.

Gran, L., Sandberg, F., and Sletten, K. (2000). Oldenlandia affinis (R&S) DC - A plant containing uteroactive peptides used in African traditional medicine. J Ethnopharmacol 70, 197-203.

Green, B.R., Catlin, P., Zhang, M., Fiedler, B., Bayudan, W., Morrison, A., Norton, R., Smith, B., Yoshikami, D., Olivera, B.M., et al. (2007). Conotoxins Containing Nonnatural Backbone Spacers: Cladistic-Based Design, Chemical Synthesis, and Improved Analgesic Activity. Chemistry & Biology 14, 399-407.

Greenwood, K.P., Daly, N.L., Brown, D.L., Stow, J.L., and Craik, D.J. (2007). The cyclic cystine knot miniprotein MCoTI-II is internalized into cells by macropinocytosis. Int J Biochem Cell Biol 39, 2252-2264.

Gruber, C.W., Cemazar, M., Anderson, M.A., and Craik, D.J. (2007). Insecticidal plant cyclotides and related cystine knot toxins. Toxicon 49, 561-575.

Gruber, C.W., Elliott, A.G., Ireland, D.C., Delprete, P.G., Dessein, S., Goransson, U., Trabi, M., Wang, C.K., Kinghorn, A.B., Robbrecht, E., et al. (2008). Distribution and evolution of circular miniproteins in flowering plants. Plant Cell 20, 2471-2483.

Gunasekera, S., Daly, N.L., Anderson, M.A., and Craik, D.J. (2006). Chemical synthesis and biosynthesis of the cyclotide family of circular proteins. IUBMB Life 58, 515-524.

Gunasekera, S., Daly, N.L., Clark, R.J., and Craik, D.J. (2008a). Dissecting the Oxidative Folding of Circular Cystine Knot Miniproteins. antioxid redox sign 11, 971-980.

Gunasekera, S., Foley, F.M., Clark, R.J., Sando, L., Fabri, L.J., Craik, D.J., and Daly, N.L. (2008b). Engineering stabilized vascular endothelial growth factor-A antagonists: synthesis, structural characterization, and bioactivity of grafted analogues of cyclotides. J Med Chem 51, 7697-7704.

Gustafson, K.R., Sowder, R.C., Henderson, L.E., Parsons, I.C., Kashman, Y., Cardellina, J.H., McMahon, J.B., Buckheit, R.W., Pannell, L.K., and

72

Boyd, M.R. (1994a). Circulin-a and Circulin-B - Novel Hiv-Inhibitory Macrocyclic Peptides From the Tropical Tree Chassalia-Parvifolia. J Am Chem Soc 116, 9337-9338.

Gustafson, K.R., Sowder, R.C., II, Henderson, L.E., Parsons, I.C., Kashman, Y., Cardellina, J.H., II, McMahon, J.B., Buckheit, R.W., Pannell, L.K., and Boyd, M.R. (1994b). Circulins A and B: Novel HIV-inhibitory macrocyclic peptides from the tropical tree Chassalia parvifolia. J Am Chem Soc 116, 9337-9338.

Gustafson, K.R., Walton, L.K., Sowder, R.C., Johnson, D.G., Pannell, L.K., Cardellina, J.H., and Boyd, M.R. (2000). New circulin macrocyclic polypeptides from Chassalia parvifolia. J Nat Prod 63, 176-178.

Hallock, Y.F., Sowder, R.C., Pannell, L.K., Hughes, C.B., Johnson, D.G., Gulakowski, R., Cardellina, J.H., and Boyd, M.R. (2000). Cycloviolins A-D, anti-HIV macrocyclic peptides from Leonia cymosa. J Org Chem 65, 124-128.

Han, S.-Y., and Kim, Y.-A. (2004). Recent development of peptide coupling reagents in organic synthesis. Tetrahedron 60, 2447-2467.

Harpaz, Z., Siman, P., Kumar, K.S.A., and Brik, A. (2010). Protein Synthesis Assisted by Native Chemical Ligation at Leucine. Chembiochem 11, 1232-1235.

Heitz, A., Hernandez, J.F., Gagnon, J., Hong, T.T., Pham, T.T., Nguyen, T.M., Le-Nguyen, D., and Chiche, L. (2001). Solution structure of the squash trypsin inhibitor MCoTI-II. A new family for cyclic knottins. Biochemistry 40, 7973-7983.

Helmlin, H.-J., Bourquin, D., and Brenneisen, R. (1992). Determination of phenylethylamines in hallucinogenic cactus species by high-performance liquid chromatography with photodiode-array detection. J Chromatogr, A 623, 381-385.

Hernandez, J.F., Gagnon, J., Chiche, L., Nguyen, T.M., Andrieu, J.P., Heitz, A., Hong, T.T., Pham, T.T.C., and Nguyen, D.L. (2000). Squash trypsin inhibitors from Momordica cochinchinensis exhibit an atypical macrocyclic structure. Biochemistry 39, 5722-5730.

Herrmann, A., Svangård, E., Claeson, P., Gullbo, J., Bohlin, L., and Göransson, U. (2006). Key role of glutamic acid for the cytotoxic activity of the cyclotide cycloviolacin O2. Cell Mol Life Sci 63, 235-245.

Hojo, H.M., Y.; Nakahara, Y.; Ito, E.; Suzuki, Y.; Suzuki, M.; Suzuki, A.; Nakahara, Y. (2005). Chemical Synthesis of 23 kDa Glycoprotein by Repetitive Segment Condensation: A Synthesis of MUC2 Basal Motif Carrying Multiple O-GalNAc Moieties. J Am Chem Soc 127, 13720-13725.

Hruby, V.J., and Agnes, R.S. (1999). Conformation-activity relationships of opioid peptides with selective activities at opioid receptors. Pept Sci 51, 391-410.

Ingenito, R., Bianchi, E., Fattori, D., and Pessi, A. (1999). Solid Phase Synthesis of Peptide C-Terminal Thioesters by Fmoc/t-Bu Chemistry. J Am Chem Soc 121, 11369-11374

73

Inui, T., Hagiwara, K., Nakajima, K., Kimura, T., Nakajima, T., and Sakakibara, S. (1992). Synthesis and disulfide structure determination of agelenin: identification of the carboxy-terminus as an amide form. Pept Res 5, 140-144.

Ireland, D.C., Clark, R.J., Daly, N.L., and Craik, D.J. (2010). Isolation, Sequencing, and Structure‚àíActivity Relationships of Cyclotides. J Nat Prod 73, 1610-1622.

Ireland, D.C., Colgrave, M.L., and Craik, D.J. (2006a). A novel suite of cyclotides from Viola odorata: sequence variation and the implications for structure, function and stability. Biochem J 400, 1-12.

Ireland, D.C., Colgrave, M.L., Nguyencong, P., Daly, N.L., and Craik, D.J. (2006b). Discovery and characterization of a linear cyclotide from Viola odorata: implications for the processing of circular proteins. J Mol Biol 357, 1522-1535.

Ireland, D.C., Wang, C.K., Wilson, J.A., Gustafson, K.R., and Craik, D.J. (2008). Cyclotides as natural anti-HIV agents. Biopolymers 90, 51-60.

Isaacs, N.W. (1995). Cystine Knots. Curr Opin Struct Biol 5, 391-395. Jagadish, K., and Camarero, J.A. (2010). Cyclotides, a promising molecular

scaffold for peptide-based therapeutics. Pept Sci 94, 611-616. Janecka, A., Zubrzycka, M., and Janecki, T. (2001). Somatostatin analogs. J

Pept Res 58, 91-107. Jennings, C., West, J., Waine, C., Craik, D.J., and Anderson, M. (2001).

Biosynthesis and insecticidal properties of plant cyclotides: The cyclic knotted proteins from Oldenlandia affinis. Proc Natl Acad Sci U S A 98, 10614-10619.

Jennings, C.V., Rosengren, K.J., Daly, N.L., Plan, M., Stevens, J., Scanlon, M.J., Waine, C., Norman, D.G., Anderson, M.A., and Craik, D.J. (2005). Isolation, solution structure, and insecticidal activity of kalata B2, a circular protein with a twist: do Mobius strips exist in nature? Biochemistry 44, 851-860.

Kates, S.A., McGuinness, B.E., Blackburn, C., Griffin, G.W., Sole, N.A., Barany, G., and Albericio, E. (1998). ‘‘High-load’’ polyethylene glycol-polystyrene (PEG-PS) graft supports for solid-phase synthesis. Biopolymers 47, 365-380.

Kenner, G.W., McDermott, J.R., and Sheppard, R.C. (1971). Safety catch principle in solid phase peptide synthesis. J Chem Soc D 12, 636-637.

Kimura, R.H., Tran, A.T., and Camarero, J.A. (2006). Biosynthesis of the Cyclotide Kalata B1 by Using Protein Splicing. Angew Chem, Int Ed Engl 45, 973-976.

Kubo, S., Chino, N., Kimura, T., and Sakakibara, S. (1996). Oxidative folding of ω -conotoxin MVIIC: Effects of temperature and salt. Biopolymers 38, 733-744.

Langer, M., La Bella, R., Garcia-Garayoa, E., and Beck-Sickinger, A.G. (2001). 99mTc-Labeled Neuropeptide Y Analogues as Potential Tumor Imaging Agents. Bioconjugate Chem 12, 1028-1034.

Laura, C., and Craik, D.J. (2010). Naturally occurring circular proteins: distribution, biosynthesis and evolution. Org Biomol Chem 8, 5035-5047.

74

Lee, T.-K., Choi, J.-H., Byun, J.-W., and Lee, Y.-S. (2008). Preparation of MBHA resin by benzotriazole-mediated amidoalkylation. Tetrahedron Lett 49, 5380-5382.

Li, S.S., and Claeson, P. (2003). Cys/Gly-rich proteins with a putative single chitin-binding domain from oat (Avena sativa) seeds. Phytochemistry 63, 249-255.

Li, X., Kawakami, T., and Aimoto, S. (1998). Direct preparation of peptide thioesters using an Fmoc solid-phase method. Tetrahedron Lett 39, 8669-8672.

Lindholm, P., Göransson, U., Johansson, S., Claeson, P., Gullbo, J., Larsson, R., Bohlin, L., and Backlund, A. (2002). Cyclotides: A novel type of cytotoxic agents. Mol Cancer Ther 1, 365-369.

McDonald, N.Q., and Hendrickson, W.A. (1993). A structural superfamily of growth factors containing a cystine knot motif. Cell 73, 421-424.

Mende, F., Beisswenger, M., and Seitz, O. (2010). Automated Fmoc-Based Solid-Phase Synthesis of Peptide Thioesters with Self-Purification Effect and Application in the Construction of Immobilized SH3 Domains. J Am Chem Soc 132, 11110-11118

Mende, F., and Seitz, O. (2007). Solid-Phase Synthesis of Peptide Thioesters with Self-Purification. Angew Chem, Int Ed Engl 46, 4577–4580.

Mende, F., and Seitz, O. (2010). 9-Fluorenylmethoxycarbonyl-Based Solid-Phase Synthesis of Peptide α-Thioesters. Angew Chem Int Ed Engl, n/a-n/a.

Mergler, M., Dick, F., Sax, B., Stähelin, C., and Vorherr, T. (2003a). The Aspartimide Problem in Fmoc-based SPPS. Part II. J Pept Sci 9, 518–526.

Mergler, M., Dick, F., Sax, B., Weiler, P., and Vorherr, T. (2003b). The Aspartimide Problem in Fmoc-based SPPS. Part I. J Pept Sci 9, 36–46.

Merrifield, R.B. (1963). Solid phase peptide synthesis I: The synthesis of a tetrapeptide. . J Am Chem Soc 85, 2149-2153.

Merrifield, R.B. (1964). Merrifield, R.B. 1964. Solid phase peptide synthesis II: The synthesis of bradykinin. . J Am Chem Soc 86, 304-305.

Mezzato, S., Schaffrath, M., and Unverzagt, C. (2005). An orthogonal double-linker resin facilitates the efficient solid-phase synthesis of complex-type N-glycopeptide thioesters suitable for native chemical ligation. Angew Chem, Int Ed Engl 44, 1650-1654.

Mulvenna, J.P., Sando, L., and Craik, D.J. (2005). Processing of a 22 kDa precursor protein to produce the circular protein tricyclon A. Structure 13, 691-701.

Offer, J., Boddy, C.N.C., and Dawson, P.E. (2002). Extending Synthetic Access to Proteins with a Removable Acyl Transfer Auxiliary. J Am Chem Soc 124, 4642-4646.

Ogunbodede, O., McCombs, D., Trout, K., Daley, P., and Terry, M. (2010). New mescaline concentrations from 14 taxa/cultivars of Echinopsis spp. (Cactaceae) ("San Pedro") and their relevance to shamanic practice. J Ethnopharmacol 131, 356-362.

75

Olivera, B.M., McIntosh, J.M., Curz, L.J., Luque, F.A., and Gray, W.R. (1984). Purification and sequence of a presynaptic peptide toxin from Conus geographus venom. Biochemistry 23, 5087-5090.

Pallaghy, P.K., Nielsen, K.J., Craik, D.J., and Norton, R.S. (1994). A common structural motif incorporating a cystine knot and a triple-stranded beta-sheet in toxic and inhibitory polypeptides. Prot Sci 3, 1833-1839.

Paramonov, S.E., Gauba, V., and Hartgerink, J.D. (2005). Synthesis of collagen-like peptide polymers by native chemcial ligation. Macromolecules 38, 7555-7561.

Pedroso, E., Grandas, A., de las Heras, X., Eritja, R., and Giralt, E. (1986). Diketopiperazine formation in solid phase peptide synthesis using p-alkoxybenzyl ester resins and Fmoc-amino acids. Tetrahedron Lett 27, 743-746.

Pentelute, B.L., Barker, A.P., Janowiak, B.E., Kent, S.B.H., and Collier, R.J. (2010). A Semisynthesis Platform for Investigating Structure‚àíFunction Relationships in the N-Terminal Domain of the Anthrax Lethal Factor. ACS Chem Biol 5, 359-364.

Pentelute, B.L., and Kent, S.B.H. (2007). Selective Desulfurization of Cysteine in the Presence of Cys(Acm) in Polypeptides Obtained by Native Chemical Ligation. Org Lett 9, 687-690.

Piotto, M., Saudek, V., and Sklenar, V. (1992). Gradient-tailored excitation for single-quantum NMR spectroscopy of aqueous solutions. J Biomol NMR 2, 661-665.

Poisson, J. (1960). Presence of mescaline in a Peruvian Cactaceae. Ann Pharm Fr 18, 764-765.

Poth, A.G., Colgrave, M.L., Philip, R., Kerenga, B., Daly, N.L., Anderson, M.A., and Craik, D.J. (2010). Discovery of Cyclotides in the Fabaceae Plant Family Provides New Insights into the Cyclization, Evolution, and Distribution of Circular Proteins. ACS Chemical Biology, null-null.

Price-Carter, M., Hull, M.S., and Goldenberg, D.P. (1998). Roles of individual disulfide bonds in the stability and folding of an omega-conotoxin. Biochemistry 37, 9851-9861.

Quibell, M., Owen, D., Packmanb, L.C., and Johnson, T. (1994). Suppression of Piperidine-mediated Side Product Formation for Asp( OBut)-containing Peptides by the Use of N-(2-hydroxy-4-methoxybenzyl) (Hmb) Backbone Amide Protection. J Chem Soc, Chem COmmun, 2343-2344.

Rosengren, K.J., Daly, N.L., Plan, M.R., Waine, C., and Craik, D.J. (2003). Twists, knots, and rings in proteins. Structural definition of the cyclotide framework. J Biol Chem 278, 8606-8616.

Saether, O., Craik, D.J., Campbell, I.D., Sletten, K., Juul, J., and Norman, D.G. (1995). Elucidation of the primary and three-dimensional structure of the uterotonic polypeptide kalata B1. Biochemistry 34, 4147-4158.

Sarin, V.K., Kent, S.B., Tam, J.P., and Merrifield, R.B. (1981). Quantitative monitoring of solid-phase peptide synthesis by the ninhydrin reaction. Anal Biochem 117, 147-157.

76

Saska, I., Gillon, A.D., Hatsugai, N., Dietzgen, R.G., Hara-Nishimura, I., Anderson, M.A., and Craik, D.J. (2007). An asparaginyl endopeptidase mediates in vivo protein backbone cyclization. J Biol Chem 282, 29721-29728.

Schoepke, T., Hasan Agha, M.I., Kraft, R., Otto, A., and Hiller, K. (1993). Compounds with hemolytic activity from Viola tricolor and V. arvensis. Sci Pharm 61, 145-153.

Schöpke, T., Hasan Agha, M.I., Kraft, R., Otto, A., and Hiller, K. (1993). Hämolytisch aktive komponenten aus Viola tricolor L. und Viola arvensis Murray. Sci Pharm 61, 145-153.

Sela, M., and Lifson, S. (1959). On the reformation of disulfide bridges in proteins. Biochim Biophys Acta 36, 471-478.

Shin, Y., Winans, K.A., Backes, B.J., Kent, S.B.H., Ellman, J.A., and Bertozz, C.R. (1999). Fmoc-Based Synthesis of Peptide-αThioesters:� Application to the Total Chemical Synthesis of a Glycoprotein by Native Chemical Ligation. J Am Chem Soc 121, 11684-11689

Shon, K.-J., Stocker, M., Terlau, H., Stühmer, W., Jacobsen, R., Walker, C., Grilley, M., Watkins, M., Hillyard, D.R., Gray, W.R., et al. (1998). κ-Conotoxin Pviia Is a Peptide Inhibiting theShaker K+ Channel. J Biol Chem 273, 33-38.

Shon, K.J., Hasson, A., Spira, M.E., Cruz, L.J., Gray, W.R., and Olivera, B.M. (1994). Delta-conotoxin GmVIA, a novel peptide from the venom of Conus gloriamaris. Biochemistry 33, 11420-11425.

Sommerhoff, C.P., Avrutina, O., Schmoldt, H., Gabrijelcic-Geiger, D., Diederichsen, U., and Kolmar, H. (2010). Engineered Cystine Knot Miniproteins as Potent Inhibitors of Human Mast Cell Tryptase [beta]. J Mol Biol 395, 167-175.

Svangård, E., Göransson, U., Hocaoglu, Z., Gullbo, J., Larsson, R., Claeson, P., and Bohlin, L. (2004). Cytotoxic cyclotides from Viola tricolor. J Nat Prod 67, 144-147.

Tam, J.P., and Lu, Y.A. (1997). Synthesis of large cyclic cystine-knot peptide by orthogonal coupling strategy using unprotected peptide precursor. Tetrahed Lett 38, 5599-5602.

Tam, J.P., and Lu, Y.A. (1998). A biomimetic strategy in the synthesis and fragmentation of cyclic protein. Prot Sci 7, 1583-1592.

Tam, J.P., Lu, Y.A., Yang, J.L., and Chiu, K.W. (1999a). An unusual structural motif of antimicrobial peptides containing end-to-end macrocycle and cystine-knot disulfides. Proc Natl Acad Sci U S A 96, 8913-8918.

Tam, J.P., Lu, Y.A., and Yu, Q.T. (1999b). Thia zip reaction for synthesis of large cyclic peptides: Mechanisms and applications. J Am Chem Soc 121, 4316-4324.

Tam, J.P., Wu, C.R., Liu, W., and Zhang, J.W. (1991). Disulfide bond formation in peptides by dimethyl-sulfoxide - Scope and applications J Am Chem Soc 113, 6657-6662.

77

Thongyoo, P., Roque-Rosell, N., Leatherbarrow, R.J., and Tate, E.W. (2008). Chemical and biomimetic total syntheses of natural and engineered MCoTI cyclotides. Org Biomol Chem 6, 1462-1470.

Thongyoo, P., Tate, E.W., and Leatherbarrow, R.J. (2006). Total synthesis of the macrocyclic cysteine knot microprotein MCoTI-II. Chem Commun, 2848-2850.

Tiefenbrunn, T.K., Blanco-Canosa, J., and Dawson, P.E. (2010). Alternative chemistries for the synthesis of thrombospondin-1 type 1 repeats. Pept Sci 94, 405-413.

Tiefenbrunn, T.K., and Dawson, P.E. (2010). Chemoselective ligation techniques: Modern applications of time-honored chemistry. Pept Sci 94, 95-106.

Trauger, J.W., Kohli, R.M., and Walsh, C.T. (2001). Cyclization of Backbone-Substituted Peptides Catalyzed by the Thioesterase Domain from the Tyrocidine Nonribosomal Peptide Synthetase‚Ä. Biochemistry 40, 7092-7098.

Vagner, J., Qu, H., and Hruby, V.J. (2008). Peptidomimetics, a synthetic tool of drug discovery. Curr Opin Chem Biol 12, 292-296.

Vlieghe, P., Lisowski, V., Martinez, J., and Khrestchatisky, M. (2010). Synthetic therapeutic peptides: science and market. Drug Discovery Today 15, 40-56.

Walsh, C.T. (2004). Polyketide and Nonribosomal Peptide Antibiotics: Modularity and Versatility. Science 303, 1805-1810.

Wan, Q., and Danishefsky, S.J. (2007). Free-Radical-Based, Specific Desulfurization of Cysteine: A Powerful Advance in the Synthesis of Polypeptides and Glycopolypeptides. Angew Chem Int Ed Engl 46, 9248-9252.

Wang, C.K., Colgrave, M.L., Ireland, D.C., Kaas, Q., and Craik, D.J. (2009a). Despite a conserved cystine knot motif, different cyclotides have different membrane binding modes. Biophys J 97, 1471-1481.

Wang, C.K., Hu, S.H., Martin, J.L., Sjogren, T., Hajdu, J., Bohlin, L., Claeson, P., Goransson, U., Rosengren, K.J., Tang, J., et al. (2009b). Combined X-ray and NMR analysis of the stability of the cyclotide cystine knot fold that underpins its insecticidal activity and potential use as a drug scaffold. J Biol Chem 284, 10672-10683.

Werle, M., and Bernkop-Schnürch, A. (2006). Strategies to improve plasma half life time of peptide and protein drugs. Amino Acids 30, 351-367.

Witherup, K.M., Bogusky, M.J., Anderson, P.S., Ramjit, H., Ransom, R.W., Wood, T., and Sardana, M. (1994). Cyclopsychotride A, a biologically active, 31-residue cyclic peptide isolated from Psychotria longipes. J Nat Prod 57, 1619-1625.

Wüthrich, K. (1986). NMR of Proteins and Nucleic Acids (New York, John Wiley & Sons, Ltd.).

Yan, L.Z., and Dawson, P.E. (2001). Synthesis of Peptides and Proteins without Cysteine Residues by Native Chemical Ligation Combined with Desulfurization. J Am Chem Soc 123, 526-533.

��4��

�3��2��5��1 ��6��

��3�� 3��1 ��6��7�� 4��7��1�1� ��1��1��8�� 3��9��3�� 7�� 1�� 3��1�3�� 1��5�� 4�1�� 5��1 ��6��8�:6��;��1��7��7��1� ��31�3�� <��4��1�� 5��1 ��6��=8>

5��1��2��1� ��8118�1�2��2�21123�4�� ! "

��

��

��

˘ ˇ ˆ - DiVA portal396748/FULLTEXT01.pdf · 2.5 Chemical and biomimetic engineering ..... 28 2.6...

Documents

Transcript of ˘ ˇ ˆ - DiVA portal396748/FULLTEXT01.pdf · 2.5 Chemical and biomimetic engineering ..... 28 2.6...