View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Structure 688

structure fold (for Cox12, an oxidase subunit at Selected Reading

fair distance to the CuA center) warrants a direct Abajian, C., Yatsunyk, L.A., Ramirez, B.E., and Rosenzweig, A.C. Cox17 interaction scenario (Arnesano et al., 2005) (2004). J. Biol. Chem. 279, 53584–53592. remains to be seen. A particularly challenging Arnesano, F., Balatri, E., Banci, L., Bertini, I., and Winge, D.R. problem appears the loading of a metal ion into (2005). Structure 13, this issue, 713–722. the CuB center of oxidase (see Figure 1); how is a Carr, H.S., and Winge, D.R. (2003). Acc. Chem. Res. 36, 309–316. copper ion inserted into a site buried by one-third Cobine, P.A., Ojeda, L.D., Rigby, K.M., and Winge, D.R. (2004). J. into the hydrophobic membrane environment? Biol. Chem. 279, 14447–14455. For that, an interesting, even though topologically Glerum, D.M., Shtanko, A., and Tzagoloff, A. (1996). J. Biol. Chem. demanding, clue toward a cotranslational action 271, 14504–14509. of Cox11 on subunit I has been given recently Horng, Y.-C., Cobine, P.A., Maxfield, A.B., Carr, H.S., and Winge, (Khalimonchuk et al., 2005). D.R. (2004). J. Biol. Chem. 279, 35334–35340. Khalimonchuk, O., Ostermann, K., and Rödel, G. (2005). Curr. Genet. 47, 223–233. 10.1007/s00294-005-0569-1. Maxfield, A.B., Heaton, D.N., and Winge, D.R. (2004). J. Biol. Chem. Bernd Ludwig 279, 5072–5080. Molecular Genetics Palumaa, P., Kangur, L., Voronova, A., and Sillard, R. (2004). Bio- chem. J. 382, 307–314. Institute for Biochemistry Puig, S., and Thiele, D.J. (2002). Curr. Opin. Chem. Biol. 6, 171–180. Goethe University Richter, O.-M.H., and Ludwig, B. (2003). Rev. Physiol. Biochem. Marie-Curie-Str. 9 Pharmacol. 147, 47–74. D-60439 Frankfurt/Main Williams, J.C., Sue, C., Banting, G.S., Yang, H., Glerum, D.M., Hen- Germany drickson, W.A., and Schon, E.A. (2005). J. Biol. Chem. 280, 15202– 15211.

Structure, Vol. 13, May, 2005, ©2005 Elsevier Ltd All rights reserved. DOI 10.1016/j.str.2005.04.005

Circular Proteins: tein kalata B1 from the African plant Oldenlandia affinis (Saether et al., 1995)(Figure 1). The cystine-knot struc- Ring around with NOESY ture consists of three disulfide bonds, two of which form a ring structure in the backbone, the third one passing in between these. This crosslinking pattern The secrets of ribosomally synthesized circular pro- gives the protein backbone a high stability and forces teins are slowly revealed by gene sequencing and so- it to adopt a β sheet structure. Such protein cystine- lution NMR studies of novel cyclotides and their pre- knot structures are rather common; for example, they cursors, as demonstrated by Mulvenna et al. (2005) in are also found in certain growth hormones (Vitt this issue of Structure. et al., 2001). It is now known that kalata B1 is a typical example Naturally occurring small circular have long of a large class of cyclic cystine-knot miniproteins, been known to exist in the bacterial world. The antibi- commonly referred to as the cyclotides (Craik et al., otic peptide gramicidin S and the potent immunosup- 2004). Similar proteins have been found in many plant pressive peptide cyclosporin A are two well-known ex- species, and there may well be several thousand dif- amples that are used as drugs. These cyclic bacterial ferent cyclotides in plants (www.cyclotide.com). The peptides often contain uncommon amino acids, and group of cyclic proteins now also extends beyond they are biosynthetically produced by large peptide plants to the rhesus macaque monkeys, where a few synthetases, rather than the standard ribosomal protein years ago a highly unusual theta- antimicrobial synthesis machinery. Cyclic peptides, and by extension peptide was uncovered (Tang et al., 1999). In fact, hu- circular proteins, have therefore remained a bit of an mans have a for a highly similar protein anomaly until the middle of the 1990’s, when the dis- called retrocyclin. The cyclic theta-defensin is distinct covery of several ribosomally synthesized macro-cyclic from the cyclotides as it has a ladder type disulfide proteins was reported. These proteins were all ex- crosslinking pattern (Figure 1). Recently it was shown tracted from plant material, and in addition to having that the bacterial microcin J25 antimicrobial peptide a cyclic backbone, they were also found to contain a contains a looped cyclic backbone structure involving cystine-knot structure. The first cyclic cystine-knot pro- the side chain of a Glu residue and one end of the pro- tein to be fully structurally characterized was the pro- tein (Rosengren et al., 2003)(Figure 1). Clearly, there Previews 689

Figure 1. Solution Structures of Some Cyclic Proteins Discovered in Recent Years Clockwise from the top left: tricyclon A (the subject of the paper by Mulvenna et al. [2005] in this issue), the prototypic cyclotide kalata B1, the theta-defensin RTD-1 from macaque monkeys, microcin J25 (a bacteriocin secreted by some enterobacteria); and SFTI-1 (a trypsin inhibitor from sunflower seeds). All structures were determined by standard homonuclear NOESY NMR techniques, using proteins that were not isotope labeled. In all cases the cyclic nature of the protein was confirmed by observing the nOe’s (nuclear Overhauser effects) between the linking amino acids. (Figure provided courtesy of J. Mulvenna and D. Craik.) could be many more different types of circular proteins bonds are reduced. The knot rather than the end-to- than we currently think. end peptide bond appears to have the major effect on The notion that circular proteins would exist at all is the stability of the protein. not immediately obvious. Genome sequencing efforts Finally it is worthwhile to consider potential applica- have strongly reinforced the idea that all genes code tions of this unique class of circular proteins. Kalata B1 for proteins as a linear array of amino acids. Therefore, and some related proteins are known to play a role in a dedicated biosynthetic mechanism must exist to plant host defense, where they have anti-insecticidal achieve the end-to-end cyclization reaction. One can activity (Craik et al., 2004). Thus, improving insect resis- think of nonenzymatic reactions similar to intein protein tance of certain crops could be an area of future appli- splicing or special enzymes tailor-made to catalyze this cation. Even though many of the cyclotides are some- reaction, as is the case for the bacterial cyclic peptides. what hemolytic in vitro, they appear to have no negative In this issue of Structure, David Craik and his col- effects on . This property, taken together with leagues from the University of Queensland in Brisbane, their phenomenal stability, may allow their use as stabi- Australia, report on the structure of the novel circular lizers of peptides with clinical potential. Many linear plant protein tricyclon A from Viola tricolor (Mulvenna peptides are known that have high potency and high et al., 2005). The structure resembles that of other plant selectivity and great potential for clinical use. However, cyclotides, but reveals a more extensive β sheet struc- the pharmaceutical industry has been reluctant to con- ture (Figure 1). Moreover, these investigators also sider such peptides as potential drugs, because they cloned and analyzed the sequence of the 22 kDa pre- are often rapidly degraded under in vivo conditions. cursor protein that is posttranslationally processed to Consequently there is considerable interest in stabiliz- give the cyclic protein. These data provide important ing their backbone. Some researchers have done this insights into the recognition sequences for the pro- by cyclizing peptides through chemical synthesis, lead- cessing of the linear array of amino acids, and such ing to improved stability and activity in several cases work will eventually allow researchers to uncover the (e.g., Nguyen et al., 2005). Several groups have made protein cyclization mechanism. the linear peptide backbone indigestible to proteases One may wonder what the biological function of end- by utilizing unnatural D- or β-amino acids (Hunter et al., to-end cyclization of the cyclotides is. The first thought 2005; Steer et al., 2002). This has created peptides that that comes to mind is that it removes the flexible N- are more protease resistant, but it has not always led and C-terminal ends of the protein, which would be to an improved activity. Others have successfully stabi- substrates for exoproteases, thereby increasing the lized the peptide backbone and increased activity by stability of the protein. Likewise, the cystine-knot ties adding extra thioether linkages at specific places along down all the flexible regions in the remainder of the pro- the backbone, as is done naturally in the lantibiotics, a tein and thereby also makes the cyclotides resistant to class of produced by Lactoba- all endoproteases. Indeed, these proteins can be boiled cilli (Rew et al., 2002). By analogy, the cyclotides pro- or treated with a broad range of proteases or denatur- vide a unique nontoxic, rock-solid scaffold on which ing agents and survive unharmed, unless the disulfide such pharmaceutical peptides can potentially be Structure 690

grafted, thereby enhancing their “survivability” in vivo. Selected Reading It might even be possible to graft two peptides simulta- neously, one with a specific targeting sequence and the Craik, D.J., Daly, N.L., Mulvenna, J., Plan, M.R., and Trabi, M. second with a designed pharmaceutical activity. Struc- (2004). Curr. Protein Pept. Sci. 5, 297–315. tural studies such as those reported by Mulvenna et al. Hunter, H.N., Jing, W., Schibli, D.J., Trinh, T., Park, I.Y., Kim, S.C., (2005) in this issue of Structure are important because and Vogel, H.J. (2005). Biochim. Biophys. Acta 1668, 175–189. they reveal the optimal scaffold structures and graft Mulvenna, J.P., Sando, L., and Craik, D.J. (2005). Structure 13, this sites to use for such applications. The fact that a sta- issue, 691–701. ble, highly disulfide crosslinked peptide derived from a Nguyen, L.T., Schibli, D.J., and Vogel, H.J. (2005). J. Pept. Sci., 11, marine toxin has recently been successfully brought to in press. the market by Elan for the treatment of chronic pain (www.prialt.com) suggests that such applications of the Rew, Y., Malkmus, S., Svensson, C., Yaksh, T.L., Chung, N.N., Schil- ler, P.W., Cassel, J.A., DeHaven, R.N., Taulane, J.P., and Goodman, cyclotides are not far-fetched at all. Not only can suit- M. (2002). J. Med. Chem. 45, 3746–3754. ably grafted cyclotides be used as pharmaceuticals, but end-to-end cyclization and stabilization of other Rosengren, K.J., Clark, R.J., Daly, N.L., Goransson, U., Jones, A., bioactive proteins and peptides may become feasible and Craik, D.J. (2003). J. Am. Chem. Soc. 125, 12464–12474. as well. Thus, gramicidin and cyclosporin may be only Saether, O., Craik, D.J., Campbell, I.D., Sletten, K., Juul, J., and the first examples of the successful pharmaceutical use Norman, D.G. (1995). Biochemistry 34, 4147–4158. of cyclic peptides and proteins. Steer, D.L., Lew, R.A., Perlmutter, P., Smith, A.I., and Aguilar, M.I. (2002). Curr. Med. Chem. 9, 811–822. Hans J. Vogel and David I. Chan Tang, Y.Q., Yuan, J., Osapay, G., Osapay, K., Tran, D., Miller, C.J., Structural Biology Research Group Ouellette, A.J., and Selsted, M.E. (1999). Science 286, 498–502. Department of Biological Sciences Vitt, U.A., Hsu, S.Y., and Hsueh, A.J. (2001). Mol. Endocrinol. 15, University of Calgary 681–694. Calgary, Alberta T2N 1N4 Canada