Báo cáo khoa học: Alternative binding proteins: Biological activity and therapeutic potential of cystine-knot miniproteins Harald Kolmar

M I N I R E V I E W

Alternative binding proteins: Biological activity and therapeutic potential of cystine-knot miniproteins Harald Kolmar

Clemens-Scho¨ pf-Institut fu¨ r Biochemie und Organische Chemie, Technische Universita¨ t Darmstadt, Germany

Keywords conotoxin; cyclotide; cystine-knot; knottin; microprotein; miniprotein; squash inhibitor

successful

Correspondence H. Kolmar, Clemens-Scho¨ pf-Institut fu¨ r Biochemie und Organische Chemie, Technische Universita¨ t Darmstadt, Petersenstraße 22, 64287 Darmstadt, Germany Fax: +49 6151 165399 Tel: +49 61511 64742 E-mail: Kolmar@Biochemie-TUD.de

(Received 16 November 2007, revised 5 March 2008, accepted 31 March 2008)

Cystine-knot miniproteins are members of a large family of small proteins that are defined by a common structural scaffold which is stabilized by three intramolecular disulfide bonds. Cystine-knot miniproteins display a broad spectrum of therapeutically useful natural biological activities and several family members are marketed as therapeutics or are in clinical development. Because of their extraordinary intrinsic chemical and proteo- lytic stability they provide promising scaffolds for the introduction of therapeutically relevant engineering functionalities. Several efforts have been reported to generate miniproteins with novel activities by rational design via functional loop grafting or by directed evolution via screening of scaffold-constrained random libraries. Owing to their small size they are amenable to recombinant as well as to chemical routes of synthesis, which opens up new avenues in optimizing biological activity, specificity and bioavailability by site-specific modification, introduction of non-natural amino acids or chemical conjugation.

doi:10.1111/j.1742-4658.2008.06440.x

Structural features and natural biological activities

manner to result in a molecule that contains the cystine knot, and only for a few miniproteins has the pathway of in vitro and in vivo folding been studied in detail [1–7].

A compact knotted triple-stranded b-sheet structure, and in some cases a short 310 helix, is formed through the formation of the disulfide-bonded scaffold from which loops emerge that frequently carry the pharma- cophoric sequence. A large family of cystine-knot pep- tides (cyclotides), exclusively found in plants, has the unusual feature of a head-to-tail cyclized peptide and the interlocking arrangement of their three disulfide bonds [8,9]. They typically comprise 28 to 37 amino acids and are the largest family of circular proteins currently known. A number of 3D structures of cyclo- tides and open-chain cystine-knot peptides have been determined by NMR and X-ray crystallography. The KNOTTIN database [10] (http://knottin.cbs.cnrs.fr/) and the cyclotide database (http://www.cyclotide.com) provide data on the cystine-knot structural family.

Cystine-knot miniproteins are members of a large fam- ily of small proteins that display a plethora of thera- peutically useful biological activities. Also known as knottins, they are typically around 30 amino acid resi- dues in length and have a common stable tertiary fold that is formed and stabilized by a characteristic pattern of disulfide bonds. Their cystine-knot structure is defined by three intramolecular disulfide bonds, where cysteine I in the sequence is connected to cysteine IV, cysteine II to cysteine V and cysteine III to cyste- ine VI. Through this type of disulfide bond formation, a knot is generated when the disulfide between cyste- ines III and VI crosses the macrocycle formed by the two other disulfides and the interconnecting backbone (Fig. 1). There are 15 ways to form an ensemble of three disulfide pairs from six cysteines. Obviously, folding and oxidation have to occur in a coordinated

Abbreviations AGRP, Agouti-related protein.

FEBS Journal 275 (2008) 2684–2690 ª 2008 The Author Journal compilation ª 2008 FEBS

2684

H. Kolmar

Cystine-knot miniproteins

Fig. 1. Cystine-knot framework and common structural features of knottins. Schematic representation of the 3D structure. Secondary struc- ture elements and termini are indicated. The disulfide bonds forming the cystine-knot architecture are represented by yellow sticks. Loops that are mainly responsible for biological activity are colored in red. (A) Squash inhibitor EETI-II [58]; (B) cyclotide kalata B1 [59], (C) C-termi- nal domain of human AGRP. Note that this cystine-knot peptide contains an additional disulfide bond that is formed by two cysteines flank- ing the C-terminal loop (indicated with the letter C) which is mainly responsible for receptor binding [30].

and protein 5 from spiders [20]. Ziconotide, a x-cono- toxin MVIIa calcium-channel blocker derived from the mollusc Conus magus is an intrathecally administered conotoxin that has already been marketed for the treatment of severe chronic pain [21–23]. Another x-conotoxin from Conus catus is in phase I ⁄ IIa clinical stage [24]. Both conotoxins inhibit presynaptic voltage- gated calcium channels and thus neurotransmission across nerve synapses. With the role of the x-cono- toxins in treating severe chronic pain, this class of biochemicals is also under investigation in the possible development of a drug to treat brain injury secondary to ischemic stroke because there is some evidence that conotoxin-mediated calcium-channel blockade may be of therapeutic value [25].

The only highly conserved signature of members of the knottin family is the space between adjacent cyste- ine residues and the mode of disulfide bond connectiv- ity. The backbone loops, however, are variable with respect to length and amino acid sequence. Thousands of different knottins exist in animals and plants that define a natural combinatorial library of cysteine inter- vening sequences that are built around the central cystine-knot motif and establish the respective repre- sentative with a particular biological activity. Both open chain and circular miniproteins containing the cystine-knot motif are remarkably stable towards extreme pH, chemical and thermal denaturation, and proteolytic attack [11,12]. This is probably a result of the enormous conformational rigidity that is intro- duced by covalent disulfide linkage of the knotted core [13,14].

receptor TRPV1,

excitatory

an

than inhibit,

rather

In contrast with the predominant role of cystine- knot peptides acting as channel inhibitors, the venom from a tarantula that is native to the West Indies has been reported to contain knottins that target the cap- saicin channel expressed by sensory neurons of the pain pathway, and function as TRP agonists. These vanillotoxins acti- vate, excitatory channels on somatosensory neurons to elicit pain and inflammation in mammals [26].

Cystine-knot miniproteins come from phylogeneti- cally diverse sources, including spiders, cone shells, plants and fungi, and display a plethora of pharma- cologic activities. A list of the main sources and types of cystine-knot miniproteins is given in Table 1. Many miniproteins target voltage-gated ion channels [15]. Examples of ion-channel toxins known to adopt this structure are, among others, the conotoxins from mar- ine cone snails (e.g. x-conotoxins [16], j-conotoxin PVIIA [17], d-conotoxin TXVIA [18] and conotoxin GS [19]), as well as x-agatoxins, robustoxin, versutoxin

Squash inhibitors constitute another large group of knottins displaying common structural features and inhibition of serine common biological activity (i.e.

Table 1. Main sources of cystine knot miniproteins.

Source

Type

Representative

Biological activity

Cucurbitaceae Violaceae Cone snails

Squash inhibitors Cyclotides Conotoxins

Protease inhibitor Oxytocic, cytotoxic, insecticidal Ion channel blocker, neurotoxic

Ecballium elaterium trypsin inhibitor II (EETI-II) Kalata B1 x-conotoxin MVIIa

FEBS Journal 275 (2008) 2684–2690 ª 2008 The Author Journal compilation ª 2008 FEBS

2685

H. Kolmar

Cystine-knot miniproteins

proteases). This type of inhibitor is exclusively found in plants and contain an inhibitor loop with canonical conformation between cysteine I and cysteine II that protrudes into the active site of a serine protease and inhibits its proteolytic activity. Most squash inhibitors are linear but two cyclic representatives have also been described [27].

A cystine-knot miniprotein of human origin is the C-terminal domain of human Agouti-related protein (AGRP) [28]. AGRP is an endogenous antagonist of the melanocortin-3 and melanocortin-4 G-protein-cou- pled receptors [29]. The C-terminal domain of AGRP (AGRP 87–132) possesses five disulfide bonds and a well-defined 3D structure that displays full activity like the full-length protein [30]. A minimized 34-residue domain has been shown to fold autonomously into the cystine-knot motif [31].

Among cystine-knot peptides, cyclotides are famous for their diverse biological activities. Many cyclotides have insecticidal activities [32]. Reports of African natives who use a decoct of the plant Oldenlandia affi- nis as an oral oxytocic [20,33] led to the discovery of cyclotide kalata B1, the pharmacologically active com- pound that mediates this effect. Moreover, cyclotides with hemolytic [34], cytotoxic [35–37], anti-HIV [38] and antifouling [39] activities have also been described.

Scaffold stability and potential routes to oral application

long-term with regard to proteolytic susceptibility, stability and their potential for oral administration, has been reported [40,41]. The squash miniproteins displayed remarkable plasma stability and over 40% were still present after 24 h of incubation in plasma, whereas no intact insulin used as a reference peptide was detected over the same time period. In contrast to cyclotides, all peptides were affected by enzymatic degradation caused by some gastro-intestinal proteo- lytic enzymes. However, it has also been demon- strated that the substitution of theoretically preferred cleavage sites in the miniprotein backbone can lead from attack by specific proteolytic to resistance enzymes [41]. Importantly for potential oral adminis- tration, knottins were found to be resistant to proteo- lytic enzymes located at the brush border membrane of the intestinal mucosa [40,41]. In addition to the absorption barrier, enzymatic degradation is the most important barrier for orally administered peptides and proteins [42]. Two miniproteins were shown to perme- ate more efficiently through excised rat intestinal mucosa than the reference peptide drugs insulin and bacitracin [12,40,41]. The latter is currently in use as an antibiotic in animal health care and orally admin- istered via the feed. From these data it can be con- the oral administration of knottins, cluded that whether cyclic or linear, seems to be generally possi- ble. However, individual optimization of a particular drug candidate with respect to proteolytic resistance and efficient delivery across the intestinal mucosa, via miniprotein sequence variation or modification as well as via formulation development, will probably be necessary.

Introduction of novel bioactivities by loop grafting

epitope

To extend the broad range of natural biological activi- ties of cystine-knot peptides towards those with thera- peutic relevance, several protein engineering efforts have been reported. They take advantage of the fact that miniprotein loops tolerate the substitution of indi- vidual amino acids as well as the insertion of addi- tional amino acids. This is exemplified by the finding that replacement of the loop between cysteine I and cysteine II of the trypsin inhibitor EETI-II from the squash Ecballium elaterium by sequences resulted in novel functionalities (i.e. antibody binding) [43].

Cyclotides and knottins of the squash family of prote- ase inhibitors have been reported to be extraordinarily stable. Craik and coworkers demonstrated the stability of the prototypic cyclotide kalata B1 to chaotropic agents 6 m guanidine hydrochloride and 8 m urea, to temperatures approaching boiling, to acid, as well as to a range of proteases, conditions under which most proteins readily unfold or are degraded [13]. Cycliza- tion is not a prerequisite for this observed stability because an acyclic permutant of cyclotide kalata B1 displayed comparable stability and structural rigidity [13] as did open-chain variants of the cyclic minipro- tein McoTI-II (L. Chiche, personal communication). However, upon removal of the disulfide bond by reduction, they became significantly more susceptible to chemical or enzymatic breakdown than the oxidized species. From these findings it can be concluded that the cystine knot is more important than the circular backbone in the chemical stability of the cyclic cystine- knot miniproteins [13].

them,

RGD or KGD peptides, or peptide-mimetics, have been developed as inhibitors of platelet aggregation, and some of tirofiban (Aggrastat) such as or eptifibatide (Integrilin), are in clinical use [44].

Recently, a systematic investigation of three differ- ent open-chain knottins of the squash inhibitor fam- together with the AGRP cystine-knot domain, ily,

FEBS Journal 275 (2008) 2684–2690 ª 2008 The Author Journal compilation ª 2008 FEBS

2686

H. Kolmar

Cystine-knot miniproteins

and KGD-containing

the case of serine protease inhibitor miniproteins, the inhibitor loop between cysteine I and cysteine II of the sequence becomes the natural target for loop replace- ment in an effort to generate novel protease inhibitors. For example, an inhibitory peptide (PMTLEYR), derived from the third domain of the turkey ovomu- coid inhibitor and optimized for specific porcine pan- creatic elastase inhibition, was introduced into the trypsin-specific squash inhibitor EETI II from Ecballi- um elaterium. The resulting hybrid showed a specificity and an affinity to porcine pancreatic elastase which was similar to that of the free inhibitory peptide but with significantly higher proteolytic stability [50].

peptide RGD-containing sequences with known activity against the aIIbb3 recep- tor on platelets were successfully grafted onto two cys- tine-knot microproteins, the trypsin inhibitor EETI-II and the melanocortin receptor-binding domain of the human AGRP. The engineered proteins were much more potent in inhibiting fibrinogen binding, aIIbb3 activation and platelet aggregation compared with the that were observed grafted peptides. Differences between the engineered proteins containing the same grafted peptide sequence in different miniprotein scaf- folds indicate the importance of the structural scaffold and the amino acids neighboring the functional peptide sequences [45,46].

Of potential therapeutic relevance is the design of cyclic miniproteins that inhibit human mast cell tryp- tase. Human b-II tryptase is a tetrameric serine prote- ase that resides almost exclusively within mast cells and presumably plays an important role in allergic asthma [51]. Based on a McoTI-II squash inhibitor knottin scaffold, a cyclic miniprotein was produced by combined recombinant and chemical synthesis and identified as the most potent proteinaceous inhibitor of human mast cell tryptase known to date, which is able to bind to all four active sites of the tryptase tetramer simultaneously, thereby blocking the access of natural substrates [52].

Introduction of novel bioactivities by directed evolution

Another example for successful loop grafting is the design of miniproteins that act as thrombopoietin agonists. Thrombopoietin is the primary regulator of platelet production. A peptide sequence of 14 residues in length that has been shown to act as a high-affinity thrombopoietin antagonist [47] was introduced into squash (EETI-II) and human (AGRP) knottin scaf- folds [48]. These variants were shown to block antago- nistically thrombopoietin-mediated receptor activation. Dimerization of the c-Mpl receptor by thrombopoietin is necessary and sufficient for receptor activation. Covalent chemical linkage of c-Mpl-binding minipro- teins yielded dimers that act as potent bivalent c-Mpl receptor agonists with a half maximal effective concen- tration (EC50) in the picomolar range, almost as active as natural thrombopoietin with regard to stimulation of megakaryocyte colony formation from human bone marrow mononuclear cells, and elicited doubling of platelet counts in mice when administered in vivo [48]. This suggests that designed dimeric cystine-knot mini- proteins may have considerable potential for the future development of small and stable receptor agonists.

Successful functional redesign has also been reported for cyclotides. Craik and coworkers exploited the cycl- otide framework by synthesizing and structurally char- acterizing two grafted analogues of the cyclotide kalata B1. The modified peptides have polar or charged residues substituted for residues that form part of a surface-exposed hydrophobic patch that plays a significant role in the folding and biological activity of kalata B1. Both analogues retain the native cyclotide fold, but lack the undesired hemolytic activity of their parent molecule [49].

Several approaches have been described for using the stable autonomous folding unit of the knottin scaffold to identify novel binding ligands using constrained ran- dom libraries. Sequence requirements of the GPNG b-turn of the E. elaterium trypsin inhibitor II, with respect to correct folding and maintenance of trypsin inhibitory activity, were explored by combinatorial library screening [7]. The same scaffold was used to identify peptide sequences involved in macromolecular recognition via mRNA display. A peptide library of EETI-II was generated, where the six residues in the first loop were randomized. The constrained library was screened against the natural target of wild-type EETI-II, bovine trypsin, and this permitted the identi- fication of a minimal consensus trypsin-binding motif [53]. An N-terminal shortened 23-residue version of EETI-II, which lacks the first disulfide bond but retains the cystine-stabilized b-sheet motif, was used as a scaffold for conformationally constrained insertion of a randomized sequence on a b-turn. A phage dis- play library of more than 108 variants comprising 10 randomized amino acids was successfully screened, among others, against malarial antigen AMA-1, the

Because the knottin scaffold retains its rigid struc- ture upon loop modification, structure-based modeling of target protein binding is feasible, retains its rigid structure-based structure upon loop modification, modelling and optimization of miniprotein binding to the target protein under consideration is feasible. In

FEBS Journal 275 (2008) 2684–2690 ª 2008 The Author Journal compilation ª 2008 FEBS

2687

H. Kolmar

Cystine-knot miniproteins

mitochondrial membrane protein Tom70 and the HIV viral protein Nef [54]. These examples indicate that screening of constrained random peptide libraries which are embedded into a knottin scaffold may, in the near future, provide a range of novel compounds with possible pharmaceutical applications.

Conclusions and outlook

thermal and biological stability,

Over the years, a careful cost–benefit analysis of the process of synthesis of biopharmaceuticals has become increasingly more relevant. Because of their small size, of only a few dozen amino acids, and in contrast to other larger scaffold proteins with potential therapeutic relevance, knottins are amenable to recombinant syn- thesis and to chemical synthesis [56]. The latter also permits the introduction of non-natural amino acids [57], as well as the defined chemical dimerization or coupling of linker molecules for improvement of bio- logical activities. With the first natural knottin already successfully marketed as an analgesic, it can be expected that over the next few years therapeutic can- didates will follow that have been isolated from the vast spectrum of natural resources or generated by peptide grafting, rational design or random library screening.

References

1 Cemazar M, Daly NL, Haggblad S, Lo KP,

Yulyaningsih E & Craik DJ (2006) Knots in rings. The circular knotted protein Momordica cochinchinensis trypsin inhibitor-II folds via a stable two-disulfide inter- mediate. J Biol Chem 281, 8224–8232.

2 Cemazar M, Gruber CW & Craik DJ (2008) Oxidative folding of cyclic cystine knot proteins. Antioxid Redox Signal 10, 103–111.

found that

3 Craik DJ & Daly NL (2005) Oxidative folding of the cystine knot motif in cyclotide proteins. Protein Pept Lett 12, 147–152. 4 Daly NL, Love S, Alewood PF & Craik DJ (1999)

In conclusion, the cystine-knot framework, owing to its inherent is an attractive scaffold for the development of peptide- based pharmaceuticals. One of the major advantages of this scaffold lies in the fact that the cystine-knot core is a reproducible structural unit that can be func- tionalized by decoration with bioactive loop residues. This results in a conformationally constrained surface exposition of one or more particular sequences and establishes the miniprotein scaffold with a desired bio- logical activity. Alternatively, the natural bioactivity of a given knottin can be used as a starting point for improvement of the pharmacologic properties. Besides, rational protein design miniproteins with predefined binding characteristics can also be obtained by combi- natorial library screening. As for most protein-based compounds, primary targets for pharmaceutical inter- vention are located extracellularly for the purpose of direct accessibility. Interestingly, recent data indicate that knottins may also be useful biomolecules for interference with intracellular target molecules. Craik and coworkers the cyclic miniprotein McoTI-II, a member of the trypsin inhibitor family of cystine-knot peptides, is readily internalized by macro- line with low cyto- phage and a breast cancer cell toxicity. Further studies will be required to determine whether this scaffold has the potential to transport grafted bioactivities into target cells [55].

Chemical synthesis and folding pathways of large cyclic polypeptides: studies of the cystine knot polypeptide kalata B1. Biochemistry 38, 10606–10614. 5 Price-Carter M, Gray WR & Goldenberg DP (1996)

Folding of omega-conotoxins. 2. Influence of precursor sequences and protein disulfide isomerase. Biochemistry 35, 15547–15557. 6 Price-Carter M, Gray WR & Goldenberg DP (1996)

Folding of omega-conotoxins. 1. Efficient disulfide-cou- pled folding of mature sequences in vitro. Biochemistry 35, 15537–15546. 7 Wentzel A, Christmann A, Kratzner R & Kolmar H

(1999) Sequence requirements of the GPNG beta-turn of the Ecballium elaterium trypsin inhibitor II explored by combinatorial library screening. J Biol Chem 274, 21037–21043.

8 Craik DJ, Cemazar M & Daly NL (2007) The chemistry and biology of cyclotides. Curr Opin Drug Discov Devel 10, 176–184.

The main advantage of using knottins for pharma- ceutical applications lies in the fact that high affinities (up to the picomolar range) and selectivities can be obtained. Moreover, many, but not all, knottins are stable against proteolytic attack and display high plasma stabilities in comparison to other proteins and peptides. However, as a result of their small size, knot- tins are amenable to fast renal clearance and their half-life in the human body is in the range of only a few hours (H. Kolmar, unpublished data). Introduc- tion of poly(ethylene glycol) chains into proteins (pegy- lation) is a well-established method of decreasing the renal clearance of the conjugated proteins but has not yet been applied to knottins. Scaffold immunogenicity is also an issue that has not been addressed syste- matically.

FEBS Journal 275 (2008) 2684–2690 ª 2008 The Author Journal compilation ª 2008 FEBS

2688

9 Craik DJ, Cemazar M, Wang CK & Daly NL (2006) The cyclotide family of circular miniproteins: nature’s combi- natorial peptide template. Biopolymers 84, 250–266.

H. Kolmar

Cystine-knot miniproteins

10 Gelly JC, Gracy J, Kaas Q, Le-Nguyen D, Heitz A &

24 Craik DJ & Adams DJ (2007) Chemical modification of conotoxins to improve stability and activity. ACS Chem Biol 2, 457–468. Chiche L (2004) The KNOTTIN website and database: a new information system dedicated to the knottin scaf- fold. Nucleic Acids Res 32, D156–D159.

11 Ireland DC, Colgrave ML & Craik DJ (2006) A novel suite of cyclotides from Viola odorata: sequence varia- tion and the implications for structure, function and stability. Biochem J 400, 1–12. 25 Perez-Pinzon MA, Yenari MA, Sun GH, Kunis DM & Steinberg GK (1997) SNX-111, a novel, presynaptic N-type calcium channel antagonist, is neuroprotective against focal cerebral ischemia in rabbits. J Neurol Sci 153, 25–31. 12 Werle M, Kafedjiiski K, Kolmar H & Bernkop-

26 Siemens J, Zhou S, Piskorowski R, Nikai T, Lumpkin EA, Basbaum AI, King D & Julius D (2006) Spider toxins activate the capsaicin receptor to produce inflam- matory pain. Nature 444, 208–212. Schnurch A (2007) Evaluation and improvement of the properties of the novel cystine-knot microprotein McoEeTI for oral administration. Int J Pharm 332, 72– 79. 13 Colgrave ML & Craik DJ (2004) Thermal, chemical,

27 Chiche L, Heitz A, Gelly JC, Gracy J, Chau PT, Ha PT, Hernandez JF & Le-Nguyen D (2004) Squash inhibitors: from structural motifs to macrocyclic knot- tins. Curr Protein Pept Sci 5, 341–349. 28 Bolin KA, Anderson DJ, Trulson JA, Thompson DA, and enzymatic stability of the cyclotide kalata B1: the importance of the cyclic cystine knot. Biochemistry 43, 5965–5975. 14 Kratzner R, Debreczeni JE, Pape T, Schneider TR,

Wilken J, Kent SB, Gantz I & Millhauser GL (1999) NMR structure of a minimized human agouti related protein prepared by total chemical synthesis. FEBS Lett 451, 125–131. 29 Wilczynski AM, Joseph CG & Haskell-Luevano C Wentzel A, Kolmar H, Sheldrick GM & Uson I (2005) Structure of Ecballium elaterium trypsin inhibitor II (EETI-II): a rigid molecular scaffold. Acta Crystallogr D Biol Crystallogr 61, 1255–1262. 15 Norton RS & Pallaghy PK (1998) The cystine knot

structure of ion channel toxins and related polypeptides. Toxicon 36, 1573–1583. 16 Pallaghy PK & Norton RS (1999) Refined solution (2005) Current trends in the structure-activity relation- ship studies of the endogenous agouti-related protein (AGRP) melanocortin receptor antagonist. Med Res Rev 25, 545–556.

structure of omega-conotoxin GVIA: implications for calcium channel binding. J Pept Res 53, 343–351. 17 Scanlon MJ, Naranjo D, Thomas L, Alewood PF,

30 McNulty JC, Thompson DA, Bolin KA, Wilken J, Barsh GS & Millhauser GL (2001) High-resolution NMR structure of the chemically-synthesized melanocortin receptor binding domain AGRP(87-132) of the agouti- related protein. Biochemistry 40, 15520–15527.

Lewis RJ & Craik DJ (1997) Solution structure and proposed binding mechanism of a novel potassium channel toxin kappa-conotoxin PVIIA. Structure 5, 1585–1597.

31 Jackson PJ, McNulty JC, Yang YK, Thompson DA, Chai B, Gantz I, Barsh GS & Millhauser GL (2002) Design, pharmacology, and NMR structure of a mini- mized cystine knot with agouti-related protein activity. Biochemistry 41, 7565–7572. 18 Kohno T, Kim JI, Kobayashi K, Kodera Y, Maeda T & Sato K (1995) Three-dimensional structure in solu- tion of the calcium channel blocker omega-conotoxin MVIIA. Biochemistry 34, 10256–10265. 19 Hill JM, Alewood PF & Craik DJ (1997) Solution 32 Gruber CW, Cemazar M, Anderson MA & Craik DJ (2007) Insecticidal plant cyclotides and related cystine knot toxins. Toxicon 49, 561–575. 33 Grain L (1973) Isolation of oxytocic peptides from structure of the sodium channel antagonist conotoxin GS: a new molecular caliper for probing sodium chan- nel geometry. Structure 5, 571–583.

Oldenlandia affinis by solvent extraction of tetrapheny- lborate complexes and chromatography on sephadex LH-20. Lloydia 36, 207–208.

20 Pallaghy PK, Nielsen KJ, Craik DJ & Norton RS (1994) A common structural motif incorporating a cystine knot and a triple-stranded beta-sheet in toxic and inhibitory polypeptides. Protein Sci 3, 1833–1839. 21 Schroeder CI, Doering CJ, Zamponi GW & Lewis RJ 34 Schopke T, Agha MI, Wray V & Hiller K (1994) Trit- erpenoid saponins from Bellium bellidioides. Phyto- chemistry 36, 449–453.

(2006) N-type calcium channel blockers: novel therapeutics for the treatment of pain. Med Chem 2, 535–543. 22 Doggrell SA (2004) Intrathecal ziconotide for refractory 35 Lindholm P, Goransson U, Johansson S, Claeson P, Gullbo J, Larsson R, Bohlin L & Backlund A (2002) Cyclotides: a novel type of cytotoxic agents. Mol Cancer Ther 1, 365–369. pain. Expert Opin Investig Drugs 13, 875–877.

FEBS Journal 275 (2008) 2684–2690 ª 2008 The Author Journal compilation ª 2008 FEBS

2689

36 Lindholm P, Gullbo J, Claeson P, Goransson U, Johansson S, Backlund A, Larsson R & Bohlin L (2002) Selective cytotoxicity evaluation in anticancer drug screening of fractionated plant extracts. J Biomol Screen 7, 333–340. 23 Klotz U (2006) Ziconotide – a novel neuron-specific cal- cium channel blocker for the intrathecal treatment of severe chronic pain – a short review. Int J Clin Pharma- col Ther 44, 478–483.

H. Kolmar

Cystine-knot miniproteins

37 Svangard E, Goransson U, Hocaoglu Z, Gullbo J, Larsson R, Claeson P & Bohlin L (2004) Cytotoxic cyclotides from Viola tricolor. J Nat Prod 67, 144–147. 38 Gustafson KR, McKee TC & Bokesch HR (2004) Anti- 49 Clark RJ, Daly NL & Craik DJ (2006) Structural plas- ticity of the cyclic-cystine-knot framework: implications for biological activity and drug design. Biochem J 394, 85–93.

HIV cyclotides. Curr Protein Pept Sci 5, 331–340. 39 Goransson U, Sjogren M, Svangard E, Claeson P &

Bohlin L (2004) Reversible antifouling effect of the cycl- otide cycloviolacin O2 against barnacles. J Nat Prod 67, 1287–1290. 50 Hilpert K, Wessner H, Schneider-Mergener J, Welfle K, Misselwitz R, Welfle H, Hocke AC, Hippenstiel S & Hohne W (2003) Design and characterization of a hybrid miniprotein that specifically inhibits porcine pan- creatic elastase. J Biol Chem 278, 24986–24993. 40 Werle M, Kolmar H, Albrecht R & Bernkop-Schnurch

51 Sommerhoff CP & Schaschke N (2007) Mast cell tryp- tase beta as a target in allergic inflammation: an evolv- ing story. Curr Pharm Des 13, 313–332.

A (2007) Characterisation of the barrier caused by lumi- nally secreted gastro-intestinal proteolytic enzymes for two novel cystine-knot microproteins. Amino Acids doi:10.1007/s00726-007-0569-1.

52 Avrutina O, Schmoldt HU, Gabrijelcic-Geiger D, Went- zel A, Frauendorf H, Sommerhoff CP, Diederichsen U & Kolmar H (2008) Head-to-tail cyclized cystine-knot peptides by a combined recombinant and chemical route of synthesis. Chembiochem 9, 33–37. 41 Werle M, Schmitz T, Huang HL, Wentzel A, Kolmar H & Bernkop-Schnurch A (2006) The potential of cystine- knot microproteins as novel pharmacophoric scaffolds in oral peptide drug delivery. J Drug Target 14, 137–146. 42 Bernkop-Schnurch A (1998) The use of inhibitory

53 Baggio R, Burgstaller P, Hale SP, Putney AR, Lane M, Lipovsek D, Wright MC, Roberts RW, Liu R, Szostak JW et al. (2002) Identification of epitope-like consensus motifs using mRNA display. J Mol Recognit 15, 126– 134. agents to overcome the enzymatic barrier to perorally administered therapeutic peptides and proteins. J Con- trol Release 52, 1–16. 54 Souriau C, Chiche L, Irving R & Hudson P (2005) New

binding specificities derived from Min-23, a small cystine-stabilized peptidic scaffold. Biochemistry 44, 7143–7155. 55 Greenwood KP, Daly NL, Brown DL, Stow JL & 43 Christmann A, Walter K, Wentzel A, Kratzner R & Kolmar H (1999) The cystine knot of a squash-type protease inhibitor as a structural scaffold for Escherichi- a coli cell surface display of conformationally con- strained peptides. Protein Eng 12, 797–806.

Craik DJ (2007) The cyclic cystine knot miniprotein MCoTI-II is internalized into cells by macropinocytosis. Int J Biochem Cell Biol 39, 2252–2264. 44 Lincoff AM, Califf RM & Topol EJ (2000) Platelet gly- coprotein IIb ⁄ IIIa receptor blockade in coronary artery disease. J Am Coll Cardiol 35, 1103–1115. 56 Gunasekera S, Daly NL, Anderson MA & Craik DJ

(2006) Chemical synthesis and biosynthesis of the cyclo- tide family of circular proteins. IUBMB Life 58, 515– 524. 57 Avrutina O, Schmoldt HU, Gabrijelcic-Geiger D, Le 45 Reiss S, Sieber M, Oberle V, Wentzel A, Spangenberg P, Claus R, Kolmar H & Losche W (2006) Inhibition of platelet aggregation by grafting RGD and KGD sequences on the structural scaffold of small disulfide- rich proteins. Platelets 17, 153–157. 46 Wentzel A, Christmann A, Adams T & Kolmar H

Nguyen D, Sommerhoff CP, Diederichsen U & Kolmar H (2005) Trypsin inhibition by macrocyclic and open- chain variants of the squash inhibitor MCoTI-II. Biol Chem 386, 1301–1306. (2001) Display of passenger proteins on the surface of Escherichia coli K-12 by the enterohemorrhagic E. coli intimin EaeA. J Bacteriol 183, 7273–7284.

58 Favel A, Mattras H, Coletti-Previero MA, Zwilling R, Robinson EA & Castro B (1989) Protease inhibitors from Ecballium elaterium seeds. Int J Pept Protein Res 33, 202–208.

47 Cwirla SE, Balasubramanian P, Duffin DJ, Wagstrom CR, Gates CM, Singer SC, Davis AM, Tansik RL, Mattheakis LC, Boytos CM et al. (1997) Peptide agonist of the thrombopoietin receptor as potent as the natural cytokine. Science 276, 1696–1699. 48 Krause S, Schmoldt HU, Wentzel A, Ballmaier M,

59 Saether O, Craik DJ, Campbell ID, Sletten K, Juul J & Norman DG (1995) Elucidation of the primary and three-dimensional structure of the uterotonic polypep- tide kalata B1. Biochemistry 34, 4147–4158.

FEBS Journal 275 (2008) 2684–2690 ª 2008 The Author Journal compilation ª 2008 FEBS

2690

Friedrich K & Kolmar H (2007) Grafting of thrombo- poietin-mimetic peptides into cystine knot miniproteins yields high-affinity thrombopoietin antagonists and agonists. Febs J 274, 86–95.