BioMol Concepts, Vol. 3 (2012), pp. 183–192 • Copyright © by Walter de Gruyter • Berlin • Boston. DOI 10.1515/bmc-2011-0061

Short Conceptual Overview

Peptide-based rotaxanes and catenanes: an emerging class of supramolecular chemistry systems

Alessandro Moretto*, Marco Crisma, Fernando Formaggio and Claudio Toniolo ICB, Padova Unit, CNR, Department of Chemistry, University of Padova, I-35131 Padova, Italy * Corresponding author e-mail: [email protected]

Abstract [2]Rotaxanes are interlocked molecular systems in which the cavity of a macrocycle (wheel) is threaded by a linear compound (axle). The axle is held in place principally by a sterically demanding moiety (stopper) at each end. If the two ends of a rotaxane are covalently linked together, then an assembly of two intertwined macrocycles (termed [2]catenane) is generated. Although rotaxanes and catenanes have been extensively investigated by several research groups, only a limited amount of studies has been devoted to these systems when a peptide molecule characterizes either the axle or the wheel(s) (or both). The purpose of this short conceptual overview is to summarize recent findings on synthetic and naturally occurring peptide-based [2]rotaxanes, [3]rotaxanes, and [2]catenanes and discuss future prospects for the research in this emerging area of supramolecular chemistry. Keywords: catenanes; molecular shuttles; peptides; rotaxanes; supramolecular chemistry.

Introduction A ‘rotaxane’ is a bead-on-a-string type structure in which a linear molecule threads through at least one cyclic molecule (in this supramolecular system, the number of components, e.g., one linear and one cyclic molecule, is indicated by a number in square brackets, e.g., [2], preceding the word rotaxane) (1–6). A sterically bulky moiety (‘stopper ’) at each end of the linear molecule does not allow the assembly to unravel. A ‘catenane’ is a different, but related, topological system where at least two rings are interlocked (if two macrocycles are involved, then a [2]catenane is generated). The most widely utilized forces to keep the various components of a rotaxane and a catenane in place are intermolecular/intersystem π…π (electron rich)donor…acceptor(electron deficient) interactions and H-bonds, typically of the C = O…H-N type. Interestingly, controlled translocation of a wheel between two

(or more) specific sites (called ‘stations’) in the axle of the supramolecular system gives rise to a ‘molecular shuttle’ or ‘molecular machine’. In addition to the stations, oligomeric systems, which commonly consist of repeating oxyethylene building blocks, are often exploited in the construction of rotaxanes and catenanes. In this overview, we discuss these types of topological systems, but limited to those characterized by oligopeptide units either in the axle or in the wheel(s), or in both components. Most of the known peptide-based rotaxanes and catenanes have been synthetically assembled, but a limited number of them are naturally occurring and bioactive.

Contributions to synthetic, peptide-based rotaxanes by Leigh’s group Synthetic peptido[2]rotaxanes, most of them based on different -Gly-Xxx- (Xxx is commonly Gly, but chiral protein amino acids were used as well) dipeptide stations in the axle, were first reported, their three-dimensional (3D) structure characterized by nuclear magnetic resonance (NMR) and at atomic resolution by X-ray diffraction crystallography, theoretically treated, and extensively developed in several papers by Leigh and coworkers (7–24) (Figures 1A and 1B). Such unhindered diamide stations are able to template efficiently the formation of an aromatic tetramide wheel to afford [2]rotaxanes through a five-component clipping reaction (7). Typically, benzhydryl moieties act as stoppers. A small amount of the [3]rotaxanes was also produced (8). Leigh’s group was able to create a molecular shuttle by inducing an intramolecular translational isomerism, governed by solvent effects, in peptido[2]rotaxanes with two stations (8). A new type of phenomenon, whereby the conformation adopted by one of the mechanically interlocked components is regulated through selectively activated non-covalent interactions with another component (‘co-conformation’), was described (9). An unexpected, enhanced H-bonding induced by photoexcitation of an anthracene-9-carboxamide N-terminal stopper of a peptido[2]rotaxane was unraveled (11). Peptido[2]rotaxanes consisting of an intrinsically achiral, aromatic wheel locked onto various chiral -Gly-LXxx- dipeptide stations exhibit a strong and negative induced circular dichroism (CD) band near 245 nm in solvents of low polarity (13). This effect is lost in polar solvents. A remarkable correlation was found during synthesis between the efficiency of the H-bond-directed assembly of peptido[2]rotaxanes and the symmetry distortion of the aromatic tetramide macrocycle

Brought to you by | The University of Auckland Library Authenticated Download Date | 5/22/15 5:54 AM

184 A. Moretto et al.

A O R2 N

O N R1

O

O O

O Cl

O

Cl

N

O

R1 H 2N

N

O

R2 N

O H

H

H

O

NH2

1 R =H, R =H 2 R1=CH3, R2=H 2

N O

O

B

N

O N

1

H

C NO2 HO H N

O O

O

H N

N H

O

O

H N H

O

N

OH

O S

HO

Figure 1 Selected peptido[2]rotaxanes synthesized and studied by the Leigh’s group. (A) Syntheses and chemical structures of two peptido[2]rotaxanes with a -Gly-Gly- (R1 = R2 = H) or a -Sar-Gly- (R1 = Me; R2 = H) station. (B) X-ray diffraction structure of the peptido[2]rotaxane 2 shown in (A) [adapted from Refs. (9) and (13)]. (C) Met-enkephalin[2]rotaxane with the -Gly2-Gly3- station [adapted from Ref. (24)].

in the 3D structure of the final product (14). Peptido[2]rotaxanes and [3]rotaxanes were assembled in high yields under thermodynamic control using H-bonding interactions and reversible olefin cross-metathesis (15). A [2]rotaxane was synthesized in which fullerene C60 behaves as a stopper and a photoactive unit (16). A reversible chiroptical switching was induced in a bistable [2]rotaxane containing a chiral dipeptide (-Gly-l-Leu-) first station by photoisomerization of a fumaramide second station (17). A [2]rotaxane was prepared by using a mechanically interlocking auxiliary which opens/closes a gate (18). This practical synthesis is unique in the sense that it does not depend on a strong recognition motif between the axle and the wheel. The final step involves cleavage of the auxiliary to release the rotaxane. [2]Rotaxane-based, light-operated molecular machines were developed using an anthracenyl carboxamide stopper and a bis-acrylamide station, with the latter undergoing cis (Z)/trans (E) equilibrium (19). Some rotaxane-based switches were found to function not only in solution but in polymer films as well (20). The first [2]rotaxanes which utilize synthetic cyclic peptides [i.e., cyclo(l-Pro-Gly)n with n = 4 and 5] as the macrocyclic component were prepared. To this end, nonpeptidic diammonium threads were used (21). Protonation of an amino function in the axle controls the location of a dual binding mode macrocycle in a [2]rotaxane. Only in the neutral (amino) form (wheel)amide…peptide(axle) H-bonds hold the wheel over the dipeptide station (23). An interesting [2]rotaxane-based propeptide was recently reported (24) (Figure 1C). In particular, the synthesis and

properties of this rotaxane in which a macrocycle effectively protects a bioactive pentapeptide (Met-enkephalin) thread from proteolytic attack were described. The enzyme (glycosidase)-catalyzed cleavage of a monosaccharide unit in one of the stoppers of the rotaxane triggers the release of Met-enkephalin through a self-immolative mechanism involving an ortho-NO2,para-hydroxy-benzylurethane moiety. Metenkephalin possesses an N-terminal -Gly-Gly- dipeptide sequence station and a C-terminal -l-Phe-l-Met- dipeptide which is bulky enough to act as a stopper for the aromatic tetramide wheel.

Contributions to synthetic, peptide-based rotaxanes by other research groups In addition to the quantitatively and qualitatively relevant production which emanated from Leigh’s group, other laboratories were also active in the area of synthetic peptido rotaxanes (25–37). In particular, Meillon et al. (25) described the thermoregulated CD properties of a [3]rotaxane in which the β-turn-forming, 7-mer peptide amide positions two dibenzo24-crown-8 ether side chains separated by three intervening residues in such a way that they are nicely oriented to generate a threaded 1:1 complex with a dibenzyl-diammonium host molecule. This pseudorotaxane (no stopper is present in the axle) involves the cooperative action of the two crown ether moieties and its formation is thermodynamically favored. A fluorescence transfer process across a (coumarin 2)

Brought to you by | The University of Auckland Library Authenticated Download Date | 5/22/15 5:54 AM

Peptido rotaxanes and catenanes

donor…acceptor(coumarin 343)-tagged peptido[2]rotaxane was studied (27). The -Gly-Gly- station is close to the coumarin donor, which acts as a stopper, and the coumarin acceptor is located on the wheel. The motion of the interlocked molecule was based on the behavior of Leigh’s rotaxanes (16). Cheetham et al. (28) reported the (α-chymotrypsin) enzymatic synthesis and the diazoaromatic-based photoswitchable enzymatic cleavage of a -Phe-Arg- dipeptide-linked rotaxane in which the α-cyclodextrin (α-CD) wheel is positioned over the diazoaromatic station. An insulin molecule monosubstituted with poly(ethylene glycol) (PEG) forms polypseudorotaxanes with α-CD and γ-CD by inserting one PEG chain in the α-CD cavity and two PEG chains in the γ-CD cavity (30). The results also showed that this type of polypseudorotaxanes is able to work as a sustained drug release system. Hirata et al. (31) prepared a novel type of polypseudorotaxanes composed of a PEG chain as the axle and 24-atom glyco-hexa-β-peptide macrocycles. A notable feature is the self-assembling (stacking) of the wheels into a unidirectionally aligned peptide nanotube endowed with a large dipole moment. The supramolecular structure is stable without stoppers due to the multiple hydrophobic interactions operative between the PEG chain and the nanotube formed by the cyclic hexa-β-peptides. A rotaxane framework was prepared where the wheel is decorated with catalytically peptide dendritic branches (32). The motion of the rotaxane components leads to an effective convergence of the catalytic groups about a rigid substrate (esterase model) binding pocket. A detailed photophysical study was conducted on a peptido[2]rotaxane characterized by a triad of photo/electroactive units (ferrocenoyl groups as substituents in the ring and a fullerene C60/ruthenium porphyrin complex as one of the stoppers) organized along a supramolecular gradient (33). Co-conformational equilibria were investigated between imide-Gly and amide stations of rotaxanes where the main interactions arise from H-bonding with the aromatic tetramide ring (34, 35). Platinum-containing rotaxanes were prepared as cytotoxic agents which specifically target cancer cells (36). Two Arg moieties positioned on the wheel potentially improve association with the DNA backbone or cell membrane phosphate groups. Finally, spin-spin coupling constants between nuclei belonging to the dipeptide axle of a [2]rotaxane and the aromatic tetramide wheel were calculated (37).

Nature’s wheel and axle supramolecular systems Even nature has developed an intriguing set of peptide-based rotaxanes without disulfide bridges which exhibit divergent biological activities. The most widely examined compound of this class is microcin J25, produced by various strains of enterobacteria (38–54). At one end of the 21-mer peptide is a relatively small ring (formed by a covalent bond between the N-terminal α-amino group and the side-chain γ-carboxylic group of Glu8). The other end of the chain loops back and threads through the ring. The C-terminus of the peptide chain is encircled by the N-terminus in a clockwise (righthanded) manner (Figure 2A). Overall, this ‘lariat protoknot’

185

is like a ‘lassoed tail’. The 3D structure is extremely stable to heating and denaturing agents. Two bulky aromatic residues (Phe19 and Tyr20) near the C-terminus sterically lock the tail in place. Indeed, Phe is located on one face of the cycle and Tyr protrudes on the other face. The chain reversal (β-hairpin) involves a dynamically disordered 11–14 segment. C-terminal residues give rise to an antiparallel β-sheet with the dipeptide 6–7 of the ring. Interestingly, when the peptide is cut with thermolysin (at level of the 10–11 peptide bond), the two segments stay together (still threaded), but the antimicrobial activity is greatly reduced. However, the in vitro ability to inhibit RNA polymerase is retained. Microcin J25 (and related compounds, see below) are the only known examples of peptido rotaxanes where non-covalent interactions are effectively permanent. During biosynthesis, two processing enzymes, associated with the bacterial inner membrane, help the initial, 58-mer linear polypeptide chain pre-fold into the correct shape (42, 44–46, 48–50, 52–54). Then, the post-translational modification goes on with one of the two enzymes which cuts off a portion of the chain (the ‘leader ’ peptide of 37 amino acids) and forms the isopeptide bond to encircle the C-terminal tail. The requirement for a minimum of eight residues immediately preceding the enzymatic cleavage site of the linear precursor peptide indicates that these residues are important for its recognition by the maturation machinery. However, it may be concluded that the intact leader peptide likely does not perform a chaperone function.

A

G21 E6 Y20 P7 F19

E8

G1 F10

V11 P16

T15

B

Figure 2 A lasso-tail peptide and a peptido[2]catenane based on natural amino acid sequences. (A) Cartoon showing the lassoed tail peptide microcin J25 [adapted from Refs. (39) and (41)]. (B) The peptido[2]catenane from an analog of the tetramerization domain of the protein p53 [adapted from Ref. (59)].

Brought to you by | The University of Auckland Library Authenticated Download Date | 5/22/15 5:54 AM

186 A. Moretto et al.

Thus far, the total chemical synthesis of this lassoed tail peptide has proven to be unsuccessful. Getting it into the proper conformation before forming the (Gly1) αNH-γCO (Glu8) lactam ring is a great challenge. Moreover, there are two different topologies wherein the tail can be irreversibly threaded through the macrocycle, but obviously, only one of them corresponds to the native structure. Only the synthesis of the ‘lariat’ peptide (unthreaded ring and covalently linked tail) was published. However, the lariat peptide is not a useful precursor for the preparation of the final entangled ‘noose’ microcin J25. A more recent experimental investigation of folding in a chemically synthesized bis-substituted microcin J25 linear analog suggests that the natural peptide does not spontaneously adopt a near-lasso structure (49). In addition, a detailed computational study points to a role for at least part of the leader sequence and/or its maturation enzymes in facilitating the onset of the correct, right-handed topology. Mutagenesis libraries targeted to the tail portion of microcin J25 showed that multiple amino acid substitutions are well tolerated in terms of antimicrobial activity. Conversely, mutagenesis of the ring portion is detrimental. Therefore, the ring region and the β-hairpin loop formed by the tail play distinct roles in the peptide antimicrobial activity. Furthermore, only a small number of residues in either the cycle or the threaded segment of the tail are important for microcin J25 production and inhibition of transcription (47). Finally, the first successful application of the extraordinarily stable microcin J25 lasso scaffold was reported. The integrin-binding Arg-Gly-Asp motif was grafted as epitope onto the tripeptide 12–14 of the β-hairpin motif. Interestingly, the originally antibacterial microcin J25 was converted into a nanomolar integrin inhibitor (51). Microcin J25 is not the only peptide-based rotaxane known to be lacking disulfide bridges (class II lasso peptides). For example, others are the 18- to 20-residues-long lariatins (55–57) and capistruin (58–60), and the shorter 16-mer RES701-1 (61). Their biological activities are diverse, ranging from specific inhibition against mycobacterial growth to endothelin B receptor antagonism. Some of them possess an eight-residue (Gly1) αNH-γCO (Glu8) ring lactam; others have a nine-residue (Gly1) αNH-βCO (Asp9) ring lactam. All of them are characterized by a short antiparallel β-sheet formed between residues of the cycle and the threading tail. In capistruin, the bulky Arg15 residue is responsible for the trapping of the tail, whereas in RES-701-1, this role is played by the C-terminal triplet of aromatic residues.

Synthetic, peptide-based systems of higher complexity: [2]catenanes A higher level of complexity in supramolecular chemistry systems is given by peptide-based catenanes (62–68) where two circular peptide molecules are interlocked (1–6) [the first protein catenane was authenticated by X-ray diffraction analysis in the bacteriophage HK97 capsid (69)]. In 2001, Yan and Dawson (62) described the first synthetic compound with a topologically linked backbone structure, i.e., a peptido[2]

catenane (Figure 2B). To this end, the 31-amino-acid-long tetramerization domain of the tumor suppressor protein p53 with an N-terminal short extension (with a Cys residue at position 1) was synthesized by solid-phase peptide synthesis. Folding and subsequent dimerization to an α-helix coiled coil system is rapid and highly efficient at pH 6. To form the peptide bonds between the two N-termini and the two C-termini of the p53 analog folded dimer, the highly chemoselective reaction termed ‘chemical ligation’ (70) was utilized. It involves the rapid and specific reaction of an N-terminal Cys thiol function with a C-terminal α-thiolester group to form a peptide bond via thiolester exchange followed by an S→N acyl shift. If the N-terminal Cys and C-terminal thiolester groups are positioned on each peptide of the dimer, a peptido[2]catenane may be generated (in addition to the cyclic dimer). The resulting product is thermally very stable (because of reduction in entropy associated to destabilization of the unfolded state) even though only one of the dimer interfaces is topologically linked. It was also found that p53 [2]catenane molecules self-assemble to produce a dimer of catenanes. The synthesis and thermodynamics of a monomeric peptido[2]catenane obtained from a doubly mutated p53 analog with disruption of the hydrophobic dimer…dimer interface were subsequently reported (63). Finally, it was unambiguously shown that it is possible to assemble a topologically linked p53 peptide complex by initially threading the linear peptide through the cyclic peptide (the latter with a slightly different sequence) to form a [2]pseudorotaxane intermediate (64). Subsequent ring closure using chemical ligation cyclizes the linear component and forms a [2]heterocatenane. Threading a peptide through a peptide was shown to be a highly efficient process. A theory for the stabilizing effect of the catenane topology of the peptide p53 analog was developed (65). There are two distinct contributions: backbone cyclization and constrained relative motion of the two rings in the unfolded state associated with interlocking. Interesting developments in the area of formation of peptide-based catenanes from dynamic combinatorial libraries were recently reported (66–68). These supramolecular systems were produced using reversibly binding units which self-assemble around a receptor target (acetylcholine) (66). Specifically, chiral dipeptides, e.g., -l-Pro-l-Phe-, containing a masked aldehyde function and a hydrazide moiety, reversibly combine through hydrazone linkages. Surprisingly, only one [2]catenane product, consisting of two interlocked macrocyclic dipeptide trimers, formed among the possible diastereomers. The acetylcholine trimethylammonium group was found to be the primary template recognition site. Alternatively, the spontaneous (in the absence of a template) formation of octameric [2]catenanes under dynamic selfassembly conditions was reported (67). Mixtures of -d-ProAib- (Aib is α-aminoisobutyric acid) and -l-Pro-l-Xxx- (Xxx is a chiral aromatic amino acid) monomers were assembled. Nevertheless, formation of the octameric [2]catenanes is highly diastereoselective. The major product consists of two tetrameric macrocycles each containing three units of -d-ProAib- and a single unit of -l-Pro-l-Xxx-. Only the diastereomeric combination of the two dipeptides mentioned above did

Brought to you by | The University of Auckland Library Authenticated Download Date | 5/22/15 5:54 AM

Peptido rotaxanes and catenanes

generate the octameric [2]catenanes. According to the results of an X-ray diffraction analysis, each tetrameric ring is saddle shaped with three type II′ β-turns (71–74) created by the -d-Pro-Aib- dipeptide sequence. Cooperative H-bonding and C-H…π interactions serve to rigidify and stabilize a core of the structure which is efficiently nucleated by the aromatic ring. Finally, a π-acceptor and π-donor disulfido[2]catenane from a dynamic combinatorial library of distinct building blocks each containing two l-Cys residues was synthesized (68). The heterotetramer [2]catenane is formed by two interlocked heterodimeric donor…acceptor moieties. The yield of the water-soluble major component is enhanced by the addition of an electron-rich template, which intercalates between the two electron-deficient units of the catenane. The product adopts the thermodynamically most stable conformation, where the number of favorable interactions between the donor and acceptor is maximized while the repulsive interactions between electron-rich cores are minimized. This represents the first construction of a peptido[2]catenane which contains both donor and acceptor units in the same ring. Air oxidation-induced, interunit Cys…Cys disulfide bond formation and reversible exchanges allow generation of the bis-dimeric interlocked species.

Contributions to synthetic, peptide-based rotaxanes by the authors’ group Based on our long-standing expertise on peptide synthesis and preferred conformations, our group is currently investigating the properties of a new set of symmetrical and nonsymmetrical peptido[2]rotaxanes with an oligopeptide in their axle. Initially, we chose two symmetrical axles, each characterized by either two Aib residues or by two incipient 310-helical (71–74) -(Aib)4- homopeptide ester sequences (75). A fumardiamide-derived central station, two ethoxy linkers, and two 9-fluorenylmethoxycarbonyl (Fmoc) Nα-protection stoppers completed the chemical structures of the [Fmoc(Aib)n-O-(CH2)2-NH]2-FUM (n = 1 and 4; FUM, fumaric unit) axles. The fumardiamide moiety proved to be an excellent template for the formation of the Leigh’s aromatic tetramide macrocyclic wheel. The X-ray diffraction structures of the two peptido[2]rotaxanes are shown in Figure 3. In each structure, the ring is held in place by amide…amide H-bonds, and by phenyl…phenyl edge-to-face and phenyl…olefin π-stacking interactions. In the -(Aib)4- peptido[2]rotaxane, the regularity of the two incipient 310-helices is not disturbed by the presence of the macrocycle. Then, we de novo designed, synthesized, and characterized a reversible molecular device based on a peptido[2] rotaxane (75). We envisaged the non-symmetrical (E)-FUM configured axle shown in Figure 4A. The N-terminal region is formed by a tripeptide with the sequence -d-Leu-(Gly)2-, followed by a 310-helical -(Aib)6- segment. Next, we incorporated an ethoxyamide-FUM unit and an l-Leu residue. We also selected diphenylacetyl and diphenylmethylamino moieties and the aromatic tetramide ring for the stoppers

187

A

B

Figure 3 Two representations each of the two X-ray diffraction structures of the symmetrical [Fmoc-(Aib)n-O-(CH2)2-NH]2-FUM (A), n=1; (B), n=4, [2]rotaxanes. Left: the carbon, oxygen, and nitrogen atoms are depicted in gray, red, and blue, respectively. The intramolecular C = O…H-N H-bonds are shown as black dotted lines. Right: space-filling representations (macrocycle in yellow and axle in green) [adapted from Ref. (72)].

and the wheel. The NMR technique proved to be instrumental in detecting the reversible motion of the wheel from the C-terminal maleidiamide (Z)-MAL station (IIa), generated by photon stimuli from the C-terminal (E)-FUM station (I), to the N-terminal -Gly-Gly- station (IIb) (Figure 4B). The solvent polarity appears to be responsible for the reversibility of the process. By virtue of the size of the inner cavity of the wheel relative to the outer diameter of the peptide helix, a rotation of the wheel concomitant with its translation along the axle seems plausible. We are currently studying the mechanism of the wrapping motion in this molecular machine in detail and extending the scope of this research to a system where both the wheel and the N-terminal stopper are decorated with appropriate probes to follow the relative motion of the two components. Next, we decided to expand this field by synthesizing and studying peptido[2]rotaxane molecular shuttles, part of the axle of which is composed of 2,5-dioxopiperazines (termed also 2,5-diketopiperazines, DKPs), planned as stations for the reversible motion of the aromatic tetrabenzamido wheel (76). DKPs are members of a class of cyclic organic compounds which result from double peptide bond formation between two α-amino acids to afford a bis-lactam (77). We first synthesized the axle of the symmetrical peptido[2]rotaxane using the Boc-Lys-Lys(diphenylacetyl)-OCH3 dipeptide (Boc, tertbutoxycarbonyl), which was subsequently dimerized via reaction of the Lys ε-NH2 functions with fumaric acid chloride. Then, we assembled the tetrabenzamido wheel on top of the fumardiamide station to obtain rotaxane 1 (Figure 5) having two diphenylacetamido stoppers. Acidic removal of the two Boc-Lys α-amino protections furnished 2, which was subsequently cyclized to the DKP-based peptido[2]rotaxanes 3 and 4, using CN- catalysis and heating in CH3CN. Upon irradiation at 256 nm in dimethylsulfoxide solution, the tetrabenzamido

Brought to you by | The University of Auckland Library Authenticated Download Date | 5/22/15 5:54 AM

188 A. Moretto et al.

O

O

A

NH NH

O O

Me NH Me

Me NH

Me NH

O Me NH

Me NH

Me NH

O

O

O Me O Me O Me O Me O Me O

NH

H

NH

NH O

NH

(E)-FUM (l)

B

E/Z isomerization

(m)

(llb)

(lla)

Reversible interconversion

Figure 4 A reversible molecular device based on a peptido[2]rotaxane. (A) Chemical structure of the non-symmetrical axle (E)-FUM [adapted from Ref. (72)]. (B) The peptido[2]rotaxane molecular shuttle showing reversible interconversion of the wheel between the C- and N-terminal stations. Computer-generated structures of the (E)-FUM (I), (Z)-MAL (IIa), and (Z)-MAL (IIb) peptido[2]rotaxanes. In I and IIa, the wheel is on the C-terminal station, whereas in IIb, it is on the N-terminal station [adapted from Ref. (72)]. The proposed mechanism (rotation/translation of the wheel along the peptide axle), m, is also schematically depicted.

wheel travels along the axle, eventually reaching one of the two DKP stations and providing peptido[2]rotaxane 5. Both types of stations can interact effectively with the wheel through multiple hydrogen bonding.

Very recently, two peptido[2]rotaxanes were synthesized using an axle based on an N,N-disubstituted DKP as a template to which bulky tritylphenyl stoppers were attached through ‘click’ chemistry (78). An NMR study provided

Rotaxane synthesis

1 bis-Boc deprotection

DKP formation 3

2

DKP formation Photoswitch and shuttling 4

5

Figure 5 The peptido[2]rotaxanes 1 and 2 and their related DKP-based peptido[2]rotaxanes 3–5 synthesized in Padova [adapted from Ref. (73)]. Brought to you by | The University of Auckland Library Authenticated Download Date | 5/22/15 5:54 AM

Peptido rotaxanes and catenanes

evidence for two different H-bonding motifs, in one of which the triazole C-H groups form C-H…O = C H-bonds with the tetralactam wheel carbonyl oxygen atoms. This motif could be controlled and switched reversibly by competitive binding of anions (e.g., Cl-) which block one side of the wheel and thus cause large changes in the relative orientation of the axle and the wheel. The crystal structure of a related pseudorotaxane clearly confirmed the DKP to be an excellent station for the synthesis of rotaxanes.

Conclusions This overview summarizes the results thus far reported in the literature on peptide-based rotaxanes and catenanes. Among the former compounds, almost exclusively, the simplest peptido[2]rotaxanes were prepared (or extracted and sequenced) and their properties were investigated. In all of these supramolecular systems, peptides are principally incorporated in the axle (often as stations, but as spacers as well). However, examples of cyclic peptides as wheels are also known (21). It is worth mentioning here that even nature developed peptido[2]rotaxane systems with specific bioactivities (38–61). In particular, the peptide-based rotaxane microcin J25 is one of the few natural molecules which inhibit RNA polymerase. This property makes it a perfect target for antibacterial therapy. Synthesizing analogs which mimic its characteristics could afford novel, potentially very useful, bioactive compounds. Moreover, lasso peptides combine unique properties relevant for many diverse applications in medicinal chemistry as robust scaffolds. On the other hand, polypeptide-based catenanes represent excellent simple models to get information on the complexity of protein folding and dynamics phenomena, as well as intriguing lead compounds for the development of new materials. Considering the rather limited number of peptide-based rotaxanes, and particularly catenanes, available, it is quite clear that there is ample space for further experimental and theoretical investigations in these interesting and emerging areas. Peptide science is a classical interdisciplinary subject, which is expected to heavily contribute to narrow the gap and strengthen the ties between synthetic organic chemistry on one side and applications to biology and nanoscience on the other side. In this framework, peptide unique modular properties and huge chemical diversity of coded and non-coded amino acid building blocks offer enormous advantages to creative supramolecular chemists involved in the developments and applications of peptide-based rotaxanes and catenanes. Other advantages related to the introduction of peptides into rotaxanes and catenanes include: (i) Use of a helical backbone which does afford a good yield of the supramolecular entity, because this preorganized system does not need to fold prior to H-bonding with the corresponding template sites. (ii) Production of molecular shuttles endowed with rigid axles, which is a very rare opportunity without being forced to result in highly insoluble systems. Moreover, a rigid axle will permit to hold appropriate probes far apart on a non-folding template. (iii) Tailoring of their solubility properties to different

189

media, suitable also to the development of biocompatible systems. (iv) Opportunity to investigate how the conformational properties of two folded elements allosterically communicate structural information.

References 1. Amabilino DB, Stoddart JF. Interlocked and intertwined structures and superstructures. Chem Rev 1995; 95: 2725–828. 2. Vögtle F, Dünnwald T, Schmidt T. Catenanes and rotaxanes of the amide type. Acc Chem Res 1996; 29: 451–60. 3. Balzani V, Gómez-López M, Stoddart JF. Molecular machines. Acc Chem Res 1998; 31: 405–14. 4. Balzani V, Credi A, Venturi M. Light powered molecular machines. Chem Soc Rev 2009; 38: 1542–50. 5. Mullen KM, Beer PD. Sulfated anion templation of macrocycles, capsules, interpenetrated and interlocked structures. Chem Soc Rev 2009; 38: 1701–13. 6. Forgan RS, Sauvage J-P, Stoddart JF. Chemical topology: complex molecular knots, links and entanglements. Chem Rev 2011; 111: 5434–64. 7. Leigh DA, Murphy A, Smart JP, Slawin AMZ. Glycylglycine rotaxanes. The hydrogen bond directed assembly of synthetic peptide rotaxanes. Angew Chem Int Ed 1997; 36: 728–32. 8. Lane AS, Leigh DA, Murphy A. Peptide-based molecular shuttles. J Am Chem Soc 1997; 119: 11092–3. 9. Clegg W, Gimenez-Saiz C, Leigh DA, Murphy A, Slawin AMZ, Teat SJ. “Smart” rotaxanes: shape memory and control in tertiary amide peptido[2]rotaxanes. J Am Chem Soc 1999; 121: 4124–9. 10. Leigh DA, Troisi A, Zerbetto F. Reducing molecular shuttling to a single dimension. Angew Chem Int Ed 2000; 39: 350–3. 11. Wurpel GWH, Brouwer AM, van Stokkum IHM, Farran A, Leigh DA. Enhanced hydrogen bonding by optical excitation: unexpected subnanosecond photoinduced dynamics in a peptidebased [2]rotaxane. J Am Chem Soc 2001; 123: 11327–8. 12. Biscarini F, Cavallini M, Leigh DA, Leon S, Teat SJ, Wong JKY, Zerbetto F. The effect of mechanical interlocking on crystal packing: predictions and testing. J Am Chem Soc 2002; 124: 225–33. 13. Asakawa M, Brancato G, Fanti M, Leigh DA, Shimizu T, Slawin AMZ, Wong JKY, Zerbetto F, Zhang S. Switching “on” and “off” the expression of chirality in peptide rotaxanes. J Am Chem Soc 2002; 124: 2939–50. 14. Brancato G, Coutrot F, Leigh DA, Murphy A, Wong JKY, Zerbetto F. From reactants to products via simple hydrogenbonding networks: information transmission in chemical reactions. Proc Natl Acad Sci USA 2002; 99: 4967–71. 15. Hannam JS, Kidd TJ, Leigh DA, Wilson AJ. “Magic rod” rotaxanes: the hydrogen bond-directed synthesis of molecular shuttles under thermodynamic control. Org Lett 2003; 5: 1907–10. 16. Da Ros T, Guldi DM, Farran Morales A, Leigh DA, Prato M, Turco R. Hydrogen bond-assembled fullerene molecular shuttle. Org Lett 2003; 5: 689–91. 17. Bottari G, Leigh DA, Pérez EM. Chiroptical switching in a bistable molecular shuttle. J Am Chem Soc 2003; 125: 13360–1. 18. Hannam JS, Lacy SM, Leigh DA, Gimenez-Saiz C, Slawin AMZ, Stitchell SG. Controlled submolecular translation motion in synthesis: a mechanically interlocked auxiliary. Angew Chem Int Ed 2004; 43: 3260–4. 19. Pérez EM, Dryden DTF, Leigh DA, Teobaldi G, Zerbetto F. A generic basis for some simple light-operated mechanical molecular machines. J Am Chem Soc 2004; 126: 12210–1.

Brought to you by | The University of Auckland Library Authenticated Download Date | 5/22/15 5:54 AM

190 A. Moretto et al.

20. Leigh DA, Morales MAF, Pérez EM, Wong JKY, Gimenez-Saiz C, Slawin AMZ, Carmichael AJ, Haddleton DM, Brouwer AM, Buma WJ, Wurpel GWH, Leon S, Zerbetto F. Patterning through controlled submolecular motion: rotaxane-based switches and logic gates that function in solution and polymer films. Angew Chem Int Ed 2005; 44: 3062–7. 21. Aucagne V, Leigh DA, Lock JS, Thomson AR. Rotaxanes of cyclic peptides. J Am Chem Soc 2006; 128: 1784–5. 22. Marlin DS, Gonzáles Cabrera D, Leigh DA, Slawin AMZ. Complexation-induced translational isomerism: shuttling through stepwise competitive binding. Angew Chem Int Ed 2006; 45: 77–83. 23. Leigh DA, Thomson AR. Switchable dual binding mode molecular shuttle. Org Lett 2006; 8: 5377–9. 24. Fernandes A, Viterisi A, Coutrot F, Potok S, Leigh DA, Aucagne V, Papot S. Rotaxane-based propeptides: protection and enzymatic release of a bioactive pentapeptide. Angew Chem Int Ed 2009; 48: 6443–7. 25. Meillon J-C, Voyer N, Biron E, Sanschagrin F, Stoddart JF. Thermoregulated optical properties of peptidic pseudorotaxanes. Angew Chem Int Ed 2000; 39: 143–5. 26. Yui N, Ooya T, Kawashima T, Saito Y, Tamai I, Sai Y, Tsuji A. Inhibitory effect of supramolecular polyrotaxane-dipeptide conjugate on digested peptide uptake via intestinal human peptide transporter. Bioconjugate Chem 2002; 13: 582–7. 27. Onagi H, Rebek J Jr. Fluorescence resonance energy transfer across a mechanical bond of a rotaxane. Chem Commun 2005; 4604–6. 28. Cheetham AG, Hutchings MG, Claridge TDW, Anderson HL. Enzymatic synthesis and photoswitchable enzymatic cleavage of a peptide-linked rotaxane. Angew Chem Int Ed 2006; 45: 1596–9. 29. Krinas CS, Demetropoulos IN. A systematic algorithm capable to yield open molecular knots. Application to alkanes, polyethylene oxides and peptides. Chem Phys Lett 2007; 433: 422–6. 30. Higashi T, Hirayama F, Arima H, Uekama K. Polypseudorotaxanes of pegylated insulin with cyclodextrins: application to sustained release. Bioorg Med Chem Lett 2007; 17: 1871–4. 31. Hirata T, Fujimura F, Kimura S. A novel pseudorotaxane composed of cyclic β-peptide as bead component. Chem Commun 2007; 1023–5. 32. Salhi A, Andreea RS. Catalytic properties of dendritic peptides in a host [2]rotaxane system. Biopolymers (Pept Sci) 2007; 88: 640-0 (505). 33. Mateo-Alonso A, Ehli C, Guldi DM, Prato M. Charge transfer reactions along a supramolecular redox gradient. J Am Chem Soc 2008; 130: 14938–9. 34. Günbas DD, Zalewski L, Brouwer AM. Energy landscape of a hydrogen-bonded non-degenerate molecular shuttle. Chem Commun 2010; 46: 2061–3. 35. Günbas DD, Zalewski L, Brouwer AM. Solvatochromic rotaxane molecular shuttles. Chem Commun 2011; 47: 4977–9. 36. Wang X, Smithrud DB. Pt-rotaxanes as cytotoxic agents. Bioorg Med Chem Lett 2011; 21: 6080–3. 37. Pikulska A, Kauch M, Pecul M. Theoretical prediction of the spin-spin coupling constants between an axis and macrocycle of a rotaxane. J Phys Chem A 2011; 115: 10795–800. 38. Bayro MJ, Mukhopadhyay J, Swapna GVT, Huang JY, Ma L-C, Sineva E, Dawson PE, Montelione GT, Ebright RH. Structure of antibacterial peptide microcin J25: a 21-residue lariat protoknot. J Am Chem Soc 2003; 125: 12382–3. 39. Rosengren KJ, Clark RJ, Daly NL, Gransson U, Jones A, Craik DJ. Microcin J25 has a threaded sidechain-to-backbone ring

structure and not a head-to-tail cyclized backbone. J Am Chem Soc 2003; 125: 12464–74. 40. Wilson K-A, Kalkum M, Ottesen J, Yuzenkova J, Chait BT, Landick R, Muir T, Severinov K, Darst SA. Structure of microcin J25, a peptide inhibitor of bacterial RNA polymerase, is a lassoed tail. J Am Chem Soc 2003; 125: 12475–83. 41. Rosengren KJ, Blond A, Afonso C, Tabet J-C, Rebuffat S, Craik DJ. Structure of thermolysis cleaved microcin J25: extreme stability of a two-chain antimicrobial peptide devoid of covalent links. Biochemistry 2004; 43: 4696–702. 42. Destoumieux-Garzon D, Duquesne S, Peduzzi J, Goulard C, Desmadril M, Letellier L, Rebuffat S, Boulanger P. The iron-siderophore transported FhuA is the receptor for the antimicrobial peptide microcin J25: role of microcin Val(11)-Pro(16) β-hairpin region in the recognition mechanism. Biochem J 2005; 389: 869–76. 43. Craik DJ, Rosengren KJ, Sando L, Simonsen SS. Threaded ring and complex topologies in antimicrobial peptides: nature’s engineering templates. In: Blondelle SE, editor. Understanding biology using peptides. Proceedings of the 19th American Peptide Symposium. New York, NY: Springer, 2006: 243–7. 44. Duquesne S, Destoumieux-Garzon D, Zirah S, Goulard C, Peduzzi J, Rebuffat S. Two enzymes catalyze the maturation of a lasso peptide in Escherichia coli. Chem Biol 2007; 14: 793–803. 45. Bierbaum G, Jansen A. Tying the knot: making of lasso peptides. Chem Biol 2007; 14: 734–5. 46. Clarke DJ, Campopiano DJ. Maturation of McjA precursor peptide into active microcin MccJ25. Org Biomol Chem 2007; 5: 2564–6. 47. Pavlova O, Mukhopadyay J, Sineva E, Ebright RH, Severinov K. Systematic structure-activity analysis of microcin J25. J Biol Chem 2008; 283: 25589–95. 48. Cheung WL, Pan SJ, Link AJ. Much of the microcin J25 leader peptide is dispensable. J Am Chem Soc 2010; 132: 2514–5. 49. Ferguson AL, Zhang S, Dikiy I, Panagiotopoulos A, Debenedetti PG, Link AJ. An experimental and computational investigation of spontaneous lasso formation in microcin J25. Biophys J 2010; 99: 3056–65. 50. Pan SJ, Link AJ. Sequence diversity in the lasso peptide framework: discovery of functional microcin J25 variants with multiple amino acid substitutions. J Am Chem Soc 2011; 133: 5016–23. 51. Knappe TA, Manzenrieder F, Mas-Moruno C, Linne U, Sasso F, Kessler H, Xie X, Marahiel MA. Introducing lasso peptides as molecular scaffolds for drug design: engineering of an integrin antagonist. Angew Chem Int Ed 2011; 50: 8714–7. 52. Yan K-P, Zirah S, Li Y, Goulard C, Ducasse R, Blond A, Peduzzi J, Rebuffat S. Molecular knots as templates for protein engineering: the story of lasso peptides. In: Lebl M, Meldal M, Jensen KJ, Hoeg-Jensen T, editors. Peptides 2010: tales of peptides. Proceedings of the 31st European Peptide Symposium. San Diego, CA: Prompt Scientific Publishing, 2010: 34–5. 53. Yan K-P, Zirah S, Li Y, Goulard C, Rebuffat S. Towards lasso peptide engineering: insight into the maturation mechanism of microcin J25. In: Lebl M, Meldal M, Jensen KJ, Hoeg-Jensen T, editors. Peptides 2010: tales of peptides. Proceedings of the 31st European Peptide Symposium. San Diego, CA: Prompt Scientific Publishing, 2010: 396–7. 54. Katoh T, Goto Y, Reza MS, Suga H. Ribosomal synthesis of backbone macrocyclic peptides. Chem Commun 2011; 47: 9946–58. 55. Iwatsuki M, Tomoda H, Uchida R, Gouda H, Hirono S, Omura S. Lariatins, antimycobacterial peptides produced by Rhodococcus sp. K01-B0171, have a lasso structure. J Am Chem Soc 2006; 128: 7486–91. 56. Iwatsuki M, Uchida R, Takakusagi Y, Matsumoto A, Jiang C-L, Takahashi Y, Arai M, Kobayashi S, Matsumoto M, Inokoshi J,

Brought to you by | The University of Auckland Library Authenticated Download Date | 5/22/15 5:54 AM

Peptido rotaxanes and catenanes

191

Tomoda H, Omura S. Lariatins, novel anti-mycobacterial peptides with a lasso structure, produced by Rhodococcus jostii K01-B0171. J Antibiot 2007; 60: 357–63. 57. Iwatsuki M, Koizumi Y, Gouda H, Hirono S, Tomoda H, Omura S. Lys17 in the “lasso” peptide lariatin A is responsible for anti-mycobacterial activity. Bioorg Med Chem Lett 2009; 19: 2888–90. 58. Knappe TA, Linne U, Zirah S, Rebuffat S, Xie X, Marahiel MA. Isolation and structural characterization of capistruin, a lasso peptide predicted from the genome sequence of Burkholderia thailandensis E264. J Am Chem Soc 2008; 130: 11446–54. 59. Knappe TA, Linne U, Robbel L, Marahiel MA. Insights into the biosynthesis of the lasso peptide capistruin. Chem Biol 2009; 16: 1290–8. 60. Kuznedelov K, Semenova E, Knappe TA, Mukhamedyarov D, Srivastava A, Chatterjee S, Ebright RH, Marahiel MA, Severinov K. The antibacterial threaded-lasso peptide capistatin inhibits bacterial RNA polymerase. J Mol Biol 2011; 412: 842–8. 61. Katahira R, Shibata K, Yamasaki M, Matsuda Y, Yoshida M. Solution structure of endothelin B receptor selective antagonist RES-701-1 determined by 1H NMR spectroscopy. Bioorg Med Chem 1995; 3: 1273–80. 62. Yan LZ, Dawson PE. Design and synthesis of a protein catenane. Angew Chem Int Ed 2001; 40: 3625–7. 63. Blankenship JW, Dawson PE. Thermodynamics of a designed protein catenane. J Mol Biol 2003; 327: 537–48. 64. Blankenship JW, Dawson PE. Threading a peptide through a peptide: protein loops, rotaxanes, and knots. Protein Sci 2007; 16: 1249–56. 65. Zhou H-X. Effect of catenation on protein folding stability. J Am Chem Soc 2003; 125: 9280–1. 66. Lam RTS, Belenguer A, Roberts SL, Naumann C, Jarrosson T, Otto S, Sanders JKM. Amplification of acetylcholine-binding catenanes from dynamic combinatorial libraries. Science 2005; 308: 667–9. 67. Chung M-K, White PS, Lee SJ, Gagné MR. Synthesis of interlocked 56-membered rings by dynamic self-templating. Angew Chem Int Ed 2009; 48: 8683–6.

68. An-Yeung HY, Pantos GD, Sanders JKM. Dynamic combinatorial synthesis of a catenane based on donor-acceptor interactions in water. Proc Natl Acad Sci USA 2009; 106: 10466–70. 69. Wikoff WR, Liljas L, Duda RL, Tsuruta H, Hendrix RW, Johnson JE. Topologically linked protein rings in the bacteriophage HK97 capsid. Science 2000; 289: 2129–33. 70. Dawson PE, Churchill MJ, Ghadiri RM, Kent SB. Synthesis of proteins by chemical ligation. Science 1994; 266: 776–9. 71. Karle IL, Balaram P. Structural characteristics of α-helical peptide molecules containing Aib residues. Biochemistry 1990; 29: 6747–56. 72. Benedetti E, Toniolo C. The polypeptide 310-helix. Trends Biochem Sci 1991; 16: 350–3. 73. Benedetti E, Di Blasio B, Pavone V, Pedone C, Toniolo C, Crisma M. Characterization at atomic resolution of peptide helical structures. Biopolymers 1992; 32: 453–6. 74. Toniolo C, Crisma M, Formaggio F, Peggion C. Control of peptide conformation by the Thorpe-Ingold effect (Cα-tetrasubstitution). Biopolymers (Pept Sci) 2001; 60: 396–419. 75. Moretto A, Menegazzo I, Crisma M, Shotton EJ, Nowell H, Mammi S, Toniolo C. A rigid helical peptide axle for a [2]rotaxane molecular machine. Angew Chem Int Ed 2009; 48: 8986–9. 76. Moretto A, Longo F, Formaggio F, Toniolo C. A 2,5-dioxopiperazine-based molecular shuttle. In: Lebl M, editor. Peptides: building bridges. Proceedings of the 22nd American Peptide Symposium. San Diego, CA: American Peptide Society, 2011: 114–5. 77. Resurreição ASM, Delatouche R, Gennari C, Piarulli U. Bifunctional 2,5-diketopiperazines as rigid three-dimensional scaffolds in receptors and peptidomimetics. Eur J Org Chem 2011; 217–28. 78. Dzyuba EV, Kaufmann L, Löw NL, Meyer AK, Winkler HDF, Rissanen K, Schalley CA. CH…O hydrogen bonds in “clicked” diketopiperazine-based amide rotaxanes. Org Lett 2011; 13: 4838–41.

Alessandro Moretto was born in Padova in 1971. He studied chemistry at the University of Padova and in the same University he completed his PhD in 2002 working under the supervision of Prof. Toniolo. He then started a postdoctoral work at the Scripps Institute Research Laboratories in San Diego, CA, with Prof. J.W. Kelly followed by another three years of a joined post-doc with the DSM Life Science Company (The Netherlands). Since 2006 he is Assistant Professor of Organic Chemistry at the University of Padova, Italy. He is co-author of more than 70 publications. His current research focuses on the design and investigation of peptide-based selfassembling systems, molecular machines and nanoparticles.

Marco Crisma earned his degree in chemistry at the University of Padova, Italy in 1980, with a thesis on the synthesis and conformational characterization of homopeptides from α-aminoisobutyric acid, supervised by C. Toniolo. After postdoctoral research with G.D. Fasman at Brandeis University, MA, USA (1981–1982), and work in industry in the laboratory of the Italian branch of Quaker Chemical Company, Inc., in 1985 he came back to the Toniolo’s group as researcher of the National Research Council of Italy (CNR). Since 2001 he is Senior Researcher at the Padova Unit of the Institute of Biomolecular Chemistry of CNR. His research interests in the field of peptides based on non-coded amino acids are currently mainly focused on their conformational analysis in the crystal state by X-ray diffraction. Dr. Crisma has co-authored over 330 publications.

Received November 21, 2011; accepted January 2, 2012

Brought to you by | The University of Auckland Library Authenticated Download Date | 5/22/15 5:54 AM

192 A. Moretto et al.

Fernando Formaggio, Professor of Organic Chemistry at the University of Padova, Italy, has a long-standing experience in the synthesis, structural characterization and applications of peptides rich in sterically hindered amino acids. His expertise was built in the group of Claudio Toniolo (University of Padova) with major contributions from a 20-month stay in Arno F. Spatola’s laboratory (Louisville, KY, USA; SPPS of peptide surrogates) and a 5-week stage at DSM Research (The Netherlands; chemo-enzymatic synthesis of alpha-amino acids). He is co-author of more than 300 publications.

Claudio Toniolo is Professor of Organic Chemistry at the University of Padova, Italy, since 1980. He was Visiting Professor in many Universities, including the Indian Institute of Sciences Bangalore, India, the National University of Somalia Mogadishu, Somalia, the Osaka University, Osaka, Japan, and the University of California, San Diego, CA. He is author of about 900 publications and his H-index is 61. He received awards from the Italian Chemical Society and the European Peptide Society. He is currently member of the Editorial Board (or the Editorial Advisory Board) of Chem. Eur. J., ChemBioChem, Biopolymers (Pept. Sci.), Chem. Biol. Drug Design, and Chem. Biodivers.

Brought to you by | The University of Auckland Library Authenticated Download Date | 5/22/15 5:54 AM