biomedicines Review
Structural Features and Binding Modes of Thioether-Cyclized Peptide Ligands Manuel E. Otero-Ramirez 1 , Toby Passioura 1 1 2
*
and Hiroaki Suga 1,2, *
Department of Chemistry, Graduate School of Science, The University of Tokyo, 7-3-1 Hongo, 113-0033 Tokyo, Bunkyo-ku, Japan;
[email protected] (M.E.O.-R.);
[email protected] (T.P.) JST CREST, The University of Tokyo, 7-3-1 Hongo, 113-0033 Tokyo, Bunkyo-ku, Japan Correspondence:
[email protected]; Tel.: +81-03-5841-8372
Received: 12 November 2018; Accepted: 11 December 2018; Published: 13 December 2018
Abstract: Macrocyclic peptides are an emerging class of bioactive compounds for therapeutic use. In part, this is because they are capable of high potency and excellent target affinity and selectivity. Over the last decade, several biochemical techniques have been developed for the identification of bioactive macrocyclic peptides, allowing for the rapid isolation of high affinity ligands to a target of interest. A common feature of these techniques is a general reliance on thioether formation to effect macrocyclization. Increasingly, the compounds identified using these approaches have been subjected to x-ray crystallographic analysis bound to their respective targets, providing detailed structural information about their conformation and mechanism of target binding. The present review provides an overview of the target bound thioether-closed macrocyclic peptide structures that have been obtained to date. Keywords: macrocyclic; peptide; crystallography
1. Introduction Macrocyclic peptides are an appealing chemical class for drug discovery. The structural constraints imbued by their macrocyclic structure provide both an improved target affinity and biological stability relative to their linear counterparts, and many bioactive macrocyclic peptides’ natural products are known. Some of these are clinically useful (e.g., cyclosporine, vancomycin) and others have formed the basis of successful drug discovery programs (e.g., Octreotide development from somatostatin and Caspofungin development from echinocandins). Taken together, this suggests that macrocyclic peptides are a “privileged” chemical class for the discovery of bioactive compounds. Over the last decade, a number of biochemical techniques have been developed for the identification of bioactive macrocyclic peptides. While the technical detail of these approaches has been reviewed in detail recently elsewhere, a brief description is warranted here. Such techniques rely on biochemical peptide synthesis (i.e., ribosomal translation) to produce partially randomized peptide libraries that are then cyclized post-translationally, either enzymatically or through non-enzymatic chemical reactions, to yield macrocyclic libraries, which can be screened for target affinity. While these techniques employ diverse methodologies, all of them involve the co-localization of each peptide macrocycle and a cognate encoding nucleic acid. This allows for screening of the pooled libraries, which can be recovered by PCR and deconvoluted by DNA sequencing (Figure 1). This, in turn, allows for the screening of very high diversity libraries, ranging from ~106 (comparable to robotic high-throughput screening methods) to in excess of 1012 compounds, depending on the technique used. Consequently, these techniques have become powerful tools for the isolation of macrocyclic peptides with potent activities against targets of interest.
Biomedicines 2018, 6, 116; doi:10.3390/biomedicines6040116
www.mdpi.com/journal/biomedicines
Biomedicines 2018, 6, 116 Biomedicines 2018, 6, x FOR PEER REVIEW
2 of 12 2 of 12
Figure 1. General scheme for affinity selection of macrocyclic peptides from ribosomally synthesized libraries. AADNA DNA library (left) is translated in such a as way as to produce a link between each library (left) is translated in such a way to produce a link between each peptide peptide and its DNA, cognate DNA, and is (often cyclized (oftenreactions through involving reactions involving Cysand residues and and its cognate and is cyclized through Cys residues leading to leading to of formation of thioethers) a DNA–peptide library. This is against pannedaagainst a formation thioethers) to producetoaproduce DNA–peptide library. This is panned surfacesurface-immobilized target protein. A DNA library enriched for sequences encoding peptides that immobilized target protein. A DNA library enriched for sequences encoding peptides that bind to the bind target isand recovered used for iterative rounds oforselection or can be sequenced to targettoisthe recovered can be and usedcan forbe iterative rounds of selection can be sequenced to determine determine the sequences of thewith peptides with a target affinity. the sequences of the peptides a target affinity.
Diverse are known Diverse cyclizing cyclizing linkages linkages are known in in both both natural natural product product macrocyclic macrocyclic peptides peptides and and in in synthetic macrocyclic peptides that are identified using the biochemical screening techniques described synthetic macrocyclic peptides that are identified using the biochemical screening techniques above. These include amide bonds, diverse head-to-side chain, side chain-to-tail, described above. Thesehead-to-tail include head-to-tail amideand bonds, and diverse head-to-side chain, side chainand side chain-to-side chain architectures. Cyclizing moieties involving thioethers are relatively to-tail, and side chain-to-side chain architectures. Cyclizing moieties involving thioethers are common in both natural product cyclic peptides and synthetic cyclic peptides. In the casecase of relatively common in both natural product cyclic peptides and synthetic cyclic peptides. In the natural products, anan entire class of natural products, entire classofofmicrobially-produced microbially-producedantibiotics, antibiotics,the thelantibiotics, lantibiotics, are are defined defined by the presence of thioether containing lanthionine moieties, although other non-lantibiotic thioether by the presence of thioether containing lanthionine moieties, although other non-lantibiotic thioether containing (e.g.,(e.g., thiocoraline) are also known. biochemical screening approaches, containing natural naturalproducts products thiocoraline) are also For known. For biochemical screening the relativelythe nucleophilic thiol of cysteine can residues be used as a reactive for “handle” cyclizing approaches, relatively nucleophilic thiolresidues of cysteine can be used “handle” as a reactive chemistries, and the vast majority of macrocyclic compounds identified through such approaches for cyclizing chemistries, and the vast majority of macrocyclic compounds identified through have such involved thioether linkagesthioether of different types.of different types. approaches have involved linkages The focuses on the features of target-bound thioetherthioether cyclized peptides, The present presentreview review focuses on structural the structural features of target-bound cyclized identified through biochemical screening approaches (to the best of our knowledge, no structures peptides, identified through biochemical screening approaches (to the best of our knowledge, no for thioether peptides bound tobound protein been reported). These are of structures forbridged thioethernatural bridged natural peptides totargets proteinhave targets have been reported). These two general types, (i) bicyclic peptides containing three cysteine residues bridged by reaction with are of two general types, (i) bicyclic peptides containing three cysteine residues bridged by reaction trivalent reagents, and and (ii) monocyclic peptides formed either through with trivalent reagents, (ii) monocyclic peptides formed either throughthe thereaction reactionof ofaa peptide peptide containing through the spontaneous reaction containing two two cysteines cysteines with with aa bivalent bivalent linker linker or or through the spontaneous reaction of of an an N-terminal N-terminal chloroacetyl chloroacetyl group group (ClAc) (ClAc) with with aa downstream downstream cysteine cysteine residue residue (monocyclic (monocyclic thioether thioether containing containing peptides). These two classes will be discussed in turn. peptides). These two classes will be discussed in turn. 2. Case Studies 2. Case Studies 2.1. Bicyclic Thioether Containing Peptides 2.1. Bicyclic Thioether Containing Peptides To the best of our knowledge, only a single protein, the urokinase-type plasminogen activator (uPa), To the best of our knowledge, only a single protein, the urokinase-type plasminogen activator has been crystalized in a complex with thioether cyclized bicyclic peptide ligands, however structures of (uPa), has been crystalized in a complex with thioether cyclized bicyclic peptide ligands, however uPa with several different bicyclic peptide ligands have been described. The first of these to be reported structures of uPa with several different bicyclic peptide ligands have been described. The first of these was the structure of the bicyclic peptide uPa–UK18 in complex with uPa [1]. This study found that to be reported was the structure of the bicyclic peptide uPa–UK18 in complex with uPa [1]. This study uPa–UK18 forms an extended structure with no clear classical secondary structural elements (i.e., no found that uPa–UK18 forms an extended structure with no clear classical secondary structural helical or sheet elements), but which is nonetheless stabilized by eight intramolecular hydrogen bonds elements (i.e., no helical or sheet elements), but which is nonetheless stabilized by eight (Figure 2a). Both of the rings of uPa–UK18 formed contacts with uPa, covering an area of approximately intramolecular hydrogen bonds (Figure 2a). Both of the rings of uPa–UK18 formed contacts with uPa, 700 Å2 and involving 14 intermolecular hydrogen bonds. Subsequent studies showed that the glycine covering an area of approximately 700 Å2 and involving 14 intermolecular hydrogen bonds. Subsequent studies showed that the glycine residue at position 13 of uPa–UK18, which exhibited a
Biomedicines 2018, 6, 116 Biomedicines 2018, 6, x FOR PEER REVIEW
3 of 12 3 of 12
residue position angle 13 of uPa–UK18, which exhibitedcould a positive dihedral angle in-alanine the crystal positiveatdihedral in the crystal structure, be replaced with D or structure, D-serine, could be replaced with D -alanine or D -serine, producing peptides with a slightly increased inhibitory producing peptides with a slightly increased inhibitory activity and a greatly increased serum activity a greatly increased stability relative to the parental uPa–UK18 molecules [2]. stabilityand relative to the parentalserum uPa–UK18 molecules [2].
Figure ofof triple thioether closed bicyclic peptides boundbound to the to active of human Figure 2. 2. Co-crystal Co-crystalstructures structures triple thioether closed bicyclic peptides the site active site of urokinase-type plasminogen activator. The 17 residue 1,3,5-tris(bromomethyl)benzene (TBMB) cyclized human urokinase-type plasminogen activator. The 17 residue 1,3,5-tris(bromomethyl)benzene peptide is shown in (a) (PDB 3QN7), with the3QN7), 13 residue (TBMB) uPa–UK18 cyclized peptide uPa–UK18 is shown in (a) (PDB with1,3,5-triacryloyl-1,3,5-triazinane the 13 residue 1,3,5-triacryloyl00 -(benzene- 1,3,5-triyl)tris(2-bromoacetamide) (TATA) cyclized (TATA) peptide UK811 4MNX) and(PDB N,N0 ,N 1,3,5-triazinane cyclized(PDB peptide UK811 4MNX) and N,N′,N′′-(benzene-1,3,5-triyl)tris(2(TBAB) cyclized peptide (PDB 4MNY) shown in 4MNY) (b,c), respectively. In and each(c), case, the trivalent bromoacetamide) (TBAB)UK903 cyclized peptide UK903 (PDB shown in (b) respectively. In linker moiety is highlighted in magenta, with the first (N-terminal) loop shown in cyan and the second each case, the trivalent linker moiety is highlighted in magenta, with the first (N-terminal) loop shown (C-terminal) loopsecond shown(C-terminal) in orange. loop shown in orange. in cyan and the
Notably, however, the cyclizing mesitylene moiety of uPa–UK18 did not interact with the peptide Notably, however, the cyclizing mesitylene moiety of uPa–UK18 did not interact with the loops, other than through the three covalent linkages (i.e., it did not appear to direct the secondary peptide loops, other than through the three covalent linkages (i.e., it did not appear to direct the structure of the scaffold other than through cyclization). To test whether different linkers could secondary structure of the scaffold other than through cyclization). To test whether different linkers alter the secondary structure of bicyclic uPa inhibitors, Heinis and co-workers identified bicyclic could alter the secondary structure of bicyclic uPa inhibitors, Heinis and co-workers identified peptides based on different trivalent thiol reactive linkers, using an affinity selection approach [3]. bicyclic peptides based on different trivalent thiol reactive linkers, using an affinity selection These studies found that the use of different linkers dramatically altered the sequences of the peptides approach [3]. These studies found that the use of different linkers dramatically altered the sequences identified, and the crystallographic studies demonstrated that these peptides bound to uPa through of the peptides identified, and the crystallographic studies demonstrated that these peptides bound distinct mechanisms (Figure 2b,c). While most of the linkers used did not form strong non-covalent to uPa through distinct mechanisms (Figure 2b,c). While most of the linkers used did not form strong intramolecular interactions, the TBAB linker (which included the most polar moieties of any of the non-covalent intramolecular interactions, the TBAB linker (which included the most polar moieties linkers tested) formed multiple intramolecular hydrogen bonds, suggesting that this linker may be able of any of the linkers tested) formed multiple intramolecular hydrogen bonds, suggesting that this to directly influence the secondary structure of the peptidic loops. Interestingly, however, the exchange linker may be able to directly influence the secondary structure of the peptidic loops. Interestingly, of the linker between the peptides decreased the inhibitory activity of all of the peptides for which this however, the exchange of the linker between the peptides decreased the inhibitory activity of all of was tested, suggesting that even in the absence of observable hydrogen bonding, small differences in the peptides for which this was tested, suggesting that even in the absence of observable hydrogen the chemical structure can lead to dramatically different conformations in the constrained peptides. bonding, small differences in the chemical structure can lead to dramatically different conformations in the constrained peptides. 2.2. Monocyclic Thioether Containing Peptides As a resultThioether of the novel techniques for their synthesis and screening, a number of high affinity 2.2. Monocyclic Containing Peptides thioether monocyclic peptide ligands have been identified in recent years, and for several of these, As a result of the novel techniques for their synthesis and screening, a number of high affinity high resolution x-ray crystal structures have been obtained in complex with a protein target. To the best thioether monocyclic peptide ligands have been identified in recent years, and for several of these, of our knowledge, all of the co-crystal structures of the monocyclic thioether closed peptide ligands high resolution x-ray crystal structures have been obtained in complex with a protein target. To the reported to date have involved peptides synthesized in genetically reprogrammed reactions, so as to best of our knowledge, all of the co-crystal structures of the monocyclic thioether closed peptide include N-terminal chloroacetyl groups that cyclize with downstream Cys resides. Co-crystal structures ligands reported to date have involved peptides synthesized in genetically reprogrammed reactions, so as to include N-terminal chloroacetyl groups that cyclize with downstream Cys resides. Co-crystal
Biomedicines 2018, 6, 116 Biomedicines 2018, 6, x FOR PEER REVIEW
4 of 12 4 of 12
structures of suchbound peptides bound to transmembrane enzymes, and other proteins of such peptides to transmembrane transporters,transporters, enzymes, and other proteins (involved in (involved in protein-protein have been reported, these areindiscussed in detail below. protein-protein interactions)interactions) have been reported, and these areand discussed detail below. Transporters 2.2.1. Transporters
In the case of the transmembrane transport proteins, thioether cyclized peptide ligands have been used specifically as co-crystallization ligands in order to facilitate the crystallographic analysis In the the first example example of this, high affinity macrocyclic peptide of these relatively intractable proteins. In identified to to the the Pyrococcus Pyrococcus furiosus furiosus multidrug multidrug and andtoxic toxiccompound compoundextrusion extrusion(PfMATE) (PfMATE) ligands were identified representative member member of of a diverse family of xenobiotic xenobiotic efflux proteins that confer transporter, a representative and neoplastic cells [4,5]. Structurally, the transporter was multidrug resistance resistanceto tomicrobial microbialpathogens pathogens and neoplastic cells [4,5]. Structurally, the transporter foundfound to adopt two different “straight” or “bent”or conformations, according to the spatialtoarrangement was to adopt two different “straight” “bent” conformations, according the spatial of its transmembrane (TM) domains, in particular, TM5, and TM6 (Figure arrangement of its transmembrane (TM) domains, TM1, in particular, TM1, TM5, and3a,b). TM6 (Figure 3a,b).
Figure 3. Crystal structures structures of of apo-Pyrococcus apo-Pyrococcus furiosus furiosus multidrug multidrug and and toxic compound compound extrusion Figure 3. Crystal (PfMATE) in in the the outwards outwards “straight” “straight” (a—PDB (a—PDB 3VVN) 3VVN) and and "bent” "bent” (b—PDB (b—PDB 3VVO) 3VVO) conformations. conformations. (PfMATE) The transmembrane (TM) domains that are most affected by the structural transition are colored colored in in The transmembrane (TM) domains that are most affected by the structural transition are blue (TM1), red (TM5), and orange (TM6). Among these, TM1 shows the most apparent structural blue (TM1), red (TM5), and orange (TM6). Among these, TM1 shows the most apparent structural change, with with the the “bent” “bent” conformation conformation named named after after the change, the bending bending of of its its central central motif. motif.
Four peptide peptide ligands and MaL6), and each of Four ligands to to PfMATE PfMATEwere wereidentified identified(MaD5, (MaD5,MaD3S, MaD3S,MaD8, MaD8, and MaL6), and each these exhibited a strong inhibitors of these exhibited a stronginhibition inhibitionofofPfMATE PfMATEextrusion, extrusion,suggesting suggestingtheir theirpossible possible use use as as inhibitors as well as co-crystalization ligands. Three of these (MaD5, MaD3S, and MaD8) included a single as well as co-crystalization ligands. Three of these (MaD5, MaD3S, and MaD8) included a single DD -Phe residue and exhibited “lariat” structures with relatively small N-terminal macrocyclic regions Phe residue and exhibited “lariat” structures with relatively small N-terminal macrocyclic regions (5–7 residues) residues). ByBy contrast, in the MaL6 peptide (which did (5–7 residues) and andlonger longerC-terminal C-terminaltails tails(9–13 (9–13 residues). contrast, in the MaL6 peptide (which not include any D - residues), all 17 residues were included in the macrocyclic structure. The co-crystal did not include any D- residues), all 17 residues were included in the macrocyclic structure. The costructures of theseofpeptides bound bound to PfMATE showedshowed that MaD5 MaD3S bound bound within within a deep crystal structures these peptides to PfMATE that and MaD5 and MaD3S central cleft pocket of a straight conformation-locked PfMATE, with the macrocyclic region of each a deep central cleft pocket of a straight conformation-locked PfMATE, with the macrocyclic region of peptide occupying a substrate recognition site (the N-lobe cavity). The interaction of the macrocyclic each peptide occupying a substrate recognition site (the N-lobe cavity). The interaction of the domains with PfMATEwith wasPfMATE mainly mediated through hydrophobic the C-terminal macrocyclic domains was mainly mediated throughinteractions, hydrophobicwith interactions, with tails adopting disordered positions that were not completely defined (Figure 4a). In contrast, MaL6 the C-terminal tails adopting disordered positions that were not completely defined (Figure 4a).and In MaD8 were shown to bind to the extracellular opening of the TM domains, with MaD8 binding to contrast, MaL6 and MaD8 were shown to bind to the extracellular opening of the TM domains, witha site deep withintothe channel, MaL6 to and the outward face of to thethe transporter MaD8 binding a site deepand within thebinding channel, MaL6 binding outward (Figure face of 4b). the The distinct binding mode of each peptide suggested distinct mechanisms of inhibition, with the transporter (Figure 4b). The distinct binding mode of each peptide suggested distinct mechanisms of long tails of the lariat, for example, appearing to restrict the motion of the Nand C-terminal lobes inhibition, with the long tails of the lariat, for example, appearing to restrict the motion of the N- and necessary for conformational during changes extrusion. Additionally, peptides were to C-terminal lobes necessary for changes conformational during extrusion.the Additionally, the found peptides reach the transporter intracellularly after penetrating the bacterial membrane, proceeding to inhibit the were found to reach the transporter intracellularly after penetrating the bacterial membrane, transporter’s through spatial blocking through restriction of the dynamic movement. proceeding tofunction inhibit the transporter’s function or through spatial blocking or TM’s through restriction of the Overall, these studies allowed for the identification of cyclic peptides that both inhibited PfMATE and TM’s dynamic movement. Overall, these studies allowed for the identification of cyclic peptides that both inhibited PfMATE and facilitated its crystallization, providing insight into both the modes of macrocyclic peptide binding and inhibition, and the mechanism of the transporter’s activity.
Biomedicines 2018, 6, 116
5 of 12
Biomedicines 6, x FOR PEER REVIEW facilitated2018, its crystallization, providing
5 of 12 insight into both the modes of macrocyclic peptide binding and inhibition, and the mechanism of the transporter’s activity.
Figure Figure 4. Co-crystal Co-crystal structures structures of of pfMATE transporter transporter (gray) (gray) in complex complex with with the thioether thioether cyclized cyclized peptides MaD3S MaD3S (a—PDB (a—PDB3VVS) 3VVS)and andMaL6 MaL6(b—PDB (b—PDB3WBN). 3WBN). The overall location of the peptide in The overall location of the peptide in the the transporter is shown left, upper view magnified viewon (darkened) on the right. transporter is shown on the on left,the while an while upper an magnified (darkened) the right. Intramolecular Intramolecular polar interactions of the peptides are shown in yellow. The thioether linker polar interactions of the peptides are shown in yellow. The thioether cyclizing linker iscyclizing highlighted in D F— D -phenylalanine. is highlighted in magenta. magenta. DF—D-phenylalanine.
In addition addition to to PfMATE, PfMATE, aa different different high affinity affinity thioether cyclized peptide ligand ligand was was identified identified In against another another xenobiotic xenobiotic transporter, transporter, the the CmABCB1 CmABCB1 protein protein from from the red alga Cyanidioschyzon against Cyanidioschyzon merolae [6]. In this case, the macrocyclic ligand, aCAP, was an 18 amino acid thioether-cyclized merolae [6]. In this case, the macrocyclic ligand, aCAP, was an macrocyclic peptide, likelike the the PfMATE ligands described above, also functioned as a CmABCB1 macrocyclic peptide,which, which, PfMATE ligands described above, also functioned as a inhibitor. The co-crystal comprised inward-open conformation for CmABCB1, CmABCB1 inhibitor. Thestructure co-crystalobtained structure obtained an comprised an inward-open conformation for with aCAP bound to the extracellular surface (Figure 5). The peptide was found to interact with CmABCB1, with aCAP bound to the extracellular surface (Figure 5). The peptide was found toa “gate” the extracellular region of the protein by tightly packed acting as aTMs, clamp that interactinwith a “gate” in the extracellular regionformed of the protein formed by TMs, tightly packed acting restrained conformation. the additional and mutational transport studies, a full scheme of as a clamp their that restrained theirFrom conformation. Frommutational the additional and transport studies, the transport mechanism was elucidated. It was found that the extracellular gate in the upper side of a full scheme of the transport mechanism was elucidated. It was found that the extracellular gate in the complex is maintained by strong interactions between theinteractions TM domains, particularly TM1domains, and TM6. the upper side of the complex is maintained by strong between the TM Upon the binding theTM6. substrate the binding cavity of of thethe transporter, withofa Tyr in TM5 particularly TM1 of and Uponinthe substrateitininteracts the cavity the residue transporter, it interacts with a Tyr residue in TM5 that promotes the movement of the upper domains in opposite directions, consequently disrupting the interactions of the extracellular gate and generating its opening, while also accelerating the ATPase activity. However, upon the binding of aCAP, the
Biomedicines 2018, 6, 116
6 of 12
that promotes of the upper domains in opposite directions, consequently disrupting Biomedicines 2018,the 6, x movement FOR PEER REVIEW 6 ofthe 12 interactions of the extracellular gate and generating its opening, while also accelerating the ATPase interactions between the the TMsbinding in the extracellular are reinforced, preventing the opening of the activity. However, upon of aCAP, the gate interactions between the TMs in the extracellular andreinforced, efficiently preventing inhibiting the activity. gate are thetransport opening of the gate and efficiently inhibiting the transport activity.
Figure structures of the homolog and ABCand multidrug transportertransporter CmABCB1 Figure 5.5.Co-crystal Co-crystal structures of P-glycoprotein the P-glycoprotein homolog ABC multidrug (gray) in complex its selectedwith macrocyclic inhibitor, aCAP (PDB 3WMG). Intramolecular polar CmABCB1 (gray)with in complex its selected macrocyclic inhibitor, aCAP (PDB 3WMG). interactions of the peptide are shown in yellow. The thioether cyclizing linker is highlighted in magenta. Intramolecular polar interactions of the peptide are shown in yellow. The thioether cyclizing linker is DW— D-tryptophan. highlighted in magenta. DW—D-tryptophan.
2.2.2. Enzymes 2.2.2. Enzymes In addition to the use of thioether-cyclized macrocyclic peptides as inhibitors/co-crystallization In addition to the use of thioether-cyclized macrocyclic peptides as inhibitors/co-crystallization ligands of membrane transporters, compounds of this class have been identified as inhibitors of ligands of membrane transporters, compounds of this class have been identified as inhibitors of specific enzymes [7–10], and in several cases, these have been subjected to X-ray crystallography specific enzymes [7–10], and in several cases, these have been subjected to X-ray crystallography as a as a basis for the subsequent structure–activity studies. For example, Kawamura et al. initially basis for the subsequent structure–activity studies. For example, Kawamura et al. initially identified identified several peptides with strong binding and inhibition properties against the JmjC-domain several peptides with strong binding and inhibition properties against the JmjC-domain containing containing lysine demethylases (JmjC-KDMs), a family of2+ Fe2+ and 2-oxoglutarate dependent histone lysine demethylases (JmjC-KDMs), a family of Fe and 2-oxoglutarate dependent histone modification enzymes [11]. In particular, the macrocyclic peptide CP2 was found to exhibit the modification enzymes [11]. In particular, the macrocyclic peptide CP2 was found to exhibit the potent potent inhibition of the KDM4A and KDM4C, and was crystallized bound to KDM4A. This peptide inhibition of the KDM4A and KDM4C, and was crystallized bound to KDM4A. This peptide demonstrated a surprising binding mode, localized not to the catalytic 2OG-binding pocket, but to demonstrated a surprising binding mode, localized not to the catalytic 2OG-binding pocket, but to the histone-binding domain, where it formed a two-turn β-sheet stabilized by, and interacting with the histone-binding domain, where it formed a two-turn β-sheet stabilized by, and interacting with KDM4A through multiple hydrogen bonds. The Arg6 of CP2 was found to bind to the sub-pocket KDM4A through multiple hydrogen bonds. The Arg6 of CP2 was found to bind to the sub-pocket usually occupied by trimethyl Lys in the histone substrate, mimicking its positive charge, and localized usually occupied by trimethyl Lys in the histone substrate, mimicking its positive charge, and near the protein’s Fe2+ cofactor2+(Figure 6). Based on this co-crystal structure, a number of alterations localized near the protein’s Fe cofactor (Figure 6). Based on this co-crystal structure, a number of were made to CP2 (e.g., the N-methylation of specific backbone positions or their substitution with a Dalterations were made to CP2 (e.g., the N-methylation of specific backbone positions or their or fluorinated-amino acids), with the aim of improving the stability and cellular uptake. Most of these substitution with a D- or fluorinated-amino acids), with the aim of improving the stability and cellular modifications were well-tolerated, even when combined, and compounds with an improved activity uptake. Most of these modifications were well-tolerated, even when combined, and compounds with in cell culture assays were obtained, demonstrating the potential for the structure-based design of an improved activity in cell culture assays were obtained, demonstrating the potential for the macrocyclic peptides. structure-based design of macrocyclic peptides.
Biomedicines 2018, 6, 116 Biomedicines 2018, 6, x FOR PEER REVIEW
7 of 12 7 of 12
6. Co-crystal Co-crystalstructure structureofofJmjC-domain JmjC-domainLysine Lysine Demethylase KDM4A (gray) in complex Figure 6. Demethylase KDM4A (gray) in complex withwith the the thioether cyclized peptide CP2 (PDB 5LY1), a selected macrocyclic peptide inhibitor (the thioether cyclized peptide CP2 (PDB 5LY1), a selected macrocyclic peptide inhibitor (the inhibitor’s inhibitor’sis sequence is shown above the structure). Intramolecular polar interactions inasCP2 are sequence shown above the structure). Intramolecular polar interactions in CP2 are shown yellow 2+ (replacing 2+ (replacing Fe2+ in KDM4A crystals) as aNiteal sphere, and Niin shown as yellow Zn2+ ionand dashed lines, Zn2+dashed ion as alines, teal sphere, Fe2+ KDM4A crystals) as a green sphere. as a thioether green sphere. The thioether cyclizing linker is highlighted magenta. DY—D-tyrosine. The cyclizing linker is highlighted in magenta. DY—Din -tyrosine.
As a second example of the X-ray crystallographic analysis of a thioether-cyclic peptide targeting an enzyme, enzyme, Yamagata Yamagata et et al. al. analyzed the structure of the S2iL5 macrocyclic peptide bound to its + + target, NAD -dependent deacetylase Sirtuin 2 (SIRT2) [12,13].[12,13]. S2iL5 was originally isolated target, the thehuman human NAD -dependent deacetylase Sirtuin 2 (SIRT2) S2iL5 was originally from a peptide around around a trifluoroacetyl Lys residue (KTfa ), designed to act as isolated from a library peptideconstructed library constructed a trifluoroacetyl Lys residue (KTfa), designed to aact mechanism-based “warhead” targeted to the SIRT2 active site. The co-crystal structure of S2iL5 as a mechanism-based “warhead” targeted to the SIRT2 active site. The co-crystal structure of and that S2iL5 a remarkable structure, in whichinthe cyclicthe peptide S2iL5SIRT2 and demonstrated SIRT2 demonstrated thatassumed S2iL5 assumed a remarkable structure, which cyclic scaffold was stabilized by a central water molecule, effectively coordinated by Arg8, Arg9, and Asn11 peptide scaffold was stabilized by a central water molecule, effectively coordinated by Arg8, Arg9, (Figure 7a). (Figure This structure allowed for allowed the peptide to a groove presenting the warhead and Asn11 7a). This structure for to thebind peptide to bindoftoSIRT2, a groove of SIRT2, presenting directly into the activeinto site.the Surprisingly, only did the and engage in a β-sheet-like the warhead directly active site. not Surprisingly, notpeptide only did theprotein peptide and protein engage in mode of interaction facilitated the of the SIRT2 to a more but the analysis of but the a β-sheet-like modethat of interaction thattransition facilitated transition ofclosed SIRT2 state, to a more closed state, unbound SIRT2 showed thatSIRT2 the SIRT2-specific insertion region (residues went through a the analysis of the unbound showed that the SIRT2-specific insertion 289–304) region (residues 289–304) drastic structural change structural from a fullchange α-helix from to a loop conformation bindingupon (Figure 7b,c). went through a drastic a full α-helix to aupon looppeptide conformation peptide The subsequent on both the peptide and the is binding (Figure mutational 7b,c). The studies subsequent mutational studies on protein both therevealed peptidethat andthis theregion protein remarkably flexible and may an important role in the substrate with in each revealed that this region is play remarkably flexible and may play anrecognition, important role theinteraction substrate with the peptide contributing synergistically to the overall kinetics of the conformational change. recognition, with each interaction with the peptide contributing synergistically to the overall kinetics Furthermore, these studies indicated that this large change was the macrocyclic of the conformational change. Furthermore, these structural studies indicated thatfacilitated this largeby structural change skeleton of the peptide, its structural and numerous withplasticity the target. was facilitated by the through macrocyclic skeleton plasticity of the peptide, throughinteractions its structural and numerous interactions with the target.
Biomedicines2018, 2018,6,6,116 x FOR PEER REVIEW Biomedicines
88 of of 12 12
Figure Figure7.7.Co-crystal Co-crystalstructure structureof ofthe theSirtuin Sirtuin22(SIRT2) (SIRT2)first firstunit unitin incomplex complexwith withthe theS2iL5 S2iL5macrocyclic macrocyclic peptide inhibitor. The (a—PDB 4L3O) sequence of the inhibitor and its binding site peptide inhibitor. The (a—PDB 4L3O) sequence of the inhibitor and its binding siteon onthe thegroove groove between the small and large domains of the enzyme (gray) are shown. The peptide maintains a between the small and large domains of the enzyme (gray) are shown. The peptide maintains a highly highly rigid structure with several hydrogen bonding interactions (yellow dashed lines), including rigid structure with several hydrogen bonding interactions (yellow dashed lines), including ones ones mediated by a water molecule (red sphere).TfaThe KTfa warhead is highlighted in light green. mediated by a water molecule (red sphere). The K warhead is highlighted in light green. Below, the Below, the significant structural changes in the SIRT2-specficic region (red) upon binding of the peptide significant structural changes in the SIRT2-specficic region (red) upon binding of the peptide between between the small (blue) and large (green) domains are apparent when comparing the free enzyme the small (blue) and large (green) domains are apparent when comparing the free enzyme (b—PDB (b—PDB 1J8F) and its bound form (c).2+The Zn2+ cofactor in the small domain is shown as a gray sphere, 1J8F) and its bound form (c). The Zn cofactor in the small domain is shown as a gray sphere, and the and the thioether cyclizing linker is highlighted in magenta. thioether cyclizing linker is highlighted in magenta.
In a third example, the co-crystal structure of a thioether-cyclized macrocycle inhibitor (Ce-2d) In a third example, the co-crystal structure of a thioether-cyclized macrocycle inhibitor (Ce-2d) of the Caenorhabditis elegans co-factor independent phosphoglycerate mutase enzyme (iPGM) was of the Caenorhabditis elegans co-factor independent phosphoglycerate mutase enzyme (iPGM) was determined [14]. Similar to the MaD5, MaD3S, and MaD8 peptides described above, Ce-2d exhibited a determined [14]. Similar to the MaD5, MaD3S, and MaD8 peptides described above, Ce-2d exhibited a “lariat” structure with a D-Tyr initiated, thioether-closed macrocycle of eight residues, followed by
Biomedicines 2018, 6, 116
Biomedicines 2018, 6, x FOR PEER REVIEW
9 of 12
9 of 12
“lariat” structure with a D-Tyr initiated, thioether-closed macrocycle of eight residues, followed by a linear “tail” of three aminoamino acids. The crystal of Ce-2d with the C. elegans iPGM a linear “tail” of three acids. The structure crystal structure of Ce-2d with the C. demonstrated elegans iPGM that the macrocycle of Ce-2d was bound between phosphatase domains of iPGM demonstrated that the macrocycle of Ce-2d wasthe bound between and the transferase phosphatase and transferase in a dominantly polar cavity, leaving the four amino acid tail free to form a small α-helical structure in domains of iPGM in a dominantly polar cavity, leaving the four amino acid tail free to form a small 2+ and Mn2+ ion cofactors 2+ 2+ contact with the solvent, with Tyr11 in close proximity to the Zn (Figure 8). α-helical structure in contact with the solvent, with Tyr11 in close proximity to the Zn and Mn ion The superposition with theof2-phosphoglycerate showed that Ce-2dsubstrate did not cofactors (Figure of 8).this Thestructure superposition this structure withsubstrate the 2-phosphoglycerate interact directly with either the substrate or the active site, evidencing an allosteric effect. Studies of showed that Ce-2d did not interact directly with either the substrate or the active site, evidencing an the truncated variants of of Ce-2d demonstrated that of theCe-2d terminal Tyr 11 was that critical activity, probably allosteric effect. Studies the truncated variants demonstrated thefor terminal Tyr 11 was because of the hydrogen bonding of the carboxamide of this residue with iPGM Glu87. These studies critical for activity, probably because of the hydrogen bonding of the carboxamide of this residue also a more potent analogue Ce-2d, Ce-2a, which exhibited the sub-nanomolar inhibition withidentified iPGM Glu87. These studies alsoofidentified a more potent analogue of Ce-2d, Ce-2a, which of iPGM, but for which the co-crystal not be determined. Ce-2a includes a longer exhibited the sub-nanomolar inhibitionstructure of iPGM,could but for which the co-crystal structure could not be (seven residue) “tail”includes than Ce-2d and a (seven terminal Cys residue. of this analog suggested an determined. Ce-2a a longer residue) “tail” The thanmodeling Ce-2d and a terminal Cys residue. 2+ co-factor of iPGM, explaining the considerably 2+ interaction of this terminal Cys with the Zn stronger The modeling of this analog suggested an interaction of this terminal Cys with the Zn co-factor of inhibitory activity of Ce-2a comparedstronger with Ce-2d. Overall,activity these thioether-closed macrocycles were iPGM, explaining the considerably inhibitory of Ce-2a compared with Ce-2d. found to induce allosteric inhibition, stabilizing a locked-open structure of iPGM, and, in the case Overall, these thioether-closed macrocycles were found to induce allosteric inhibition, stabilizingofa 2+ co-factor. Ce-2a, sequestering its of ZniPGM, locked-open structure and, in the case of Ce-2a, sequestering its Zn2+ co-factor.
Figure 8. 8. Co-crystal Co-crystal structure structure of of C. C. elegans elegans iPGM iPGM (gray) (gray) in in complex complex with with the the peptide peptide inhibitor, inhibitor, Ce-2d Ce-2d Figure (PDB 5KGN). The left panel shows the positioning of the tightly folded peptide in the groove between (PDB 5KGN). The left panel shows the positioning of the tightly folded peptide in the groove between the phosphatase phosphatase and and transferase transferase domains domains of of the the protein. protein. On the right, right, aa magnified magnified view view shows shows the the the multiple Tyr Tyr residues residues in in the thepeptide, peptide, as as well wellas asthe theintramolecular intramolecular polar polar interactions interactions (yellow (yellow dashed dashed multiple cofactor lines) and and the the presence presenceof of the the N-terminal N-terminalTyr11 Tyr11amide’s amide’soxygen oxygenin inclose closeproximity proximitytotothe theZn Zn2+2+ cofactor lines) 2+ (blue sphere) cofactor is shown as a blue sphere, and the thioether (purple (purplesphere). sphere).The TheMn Mn2+ thioether cyclizing cyclizing moiety moiety is is highlighted highlighted in in magenta. magenta.
Finally, Finally,the thethioether thioethercyclized cyclizedpeptide peptideligands ligandshave havealso also been been co-crystalized co-crystalized bound bound to to the the human human pancreatic digestion and implicated as as playing a role in pancreaticamylase amylase(HPA), (HPA),an anenzyme enzymeinvolved involvedininstarch starch digestion and implicated playing a role type-2 diabetes [15]. These peptides exhibited conserved RFGYAY and ( D Y)PYSCWXRH motifs, and a in type-2 diabetes [15]. These peptides exhibited conserved RFGYAY and (DY)PYSCWXRH motifs, lariat and were competitive inhibitors of HPA, with with inhibition constants in the and a architecture, lariat architecture, and potent were potent competitive inhibitors of HPA, inhibition constants in single digit nanomolar range. The co-crystal structure of one of these peptides, a nonapeptide termed the single digit nanomolar range. The co-crystal structure of one of these peptides, a nonapeptide piHA-Dm, showed that it occupied catalytic of the protein, consistent its competitive termed piHA-Dm, showed that it the occupied thesite catalytic site of the protein,with consistent with its mechanism. The tail region lariatofassumed highly ordered α-helical the competitive mechanism. The of tailthe region the lariata assumed a highly orderedstructure, α-helical while structure, five-membered cycle was localized and tightly compacted within the catalytic site and bound directly while the five-membered cycle was localized and tightly compacted within the catalytic site and (or through water(or molecules) several key catalytic amino acids the protein (Figure bound directly through to water molecules) to several key in catalytic amino acids9).inFurthermore, the protein even the9). initiating D -Tyr was to formDimportant interactions with the protein, throughwith boththe its (Figure Furthermore, evenshown the initiating -Tyr was shown to form important interactions side chain and amide moieties. The D -stereochemistry of the initiating Tyr was found to allow a parallel protein, through both its side chain and amide moieties. The D-stereochemistry of the initiating Tyr arrangement forming a tripeptide motif with Pro2 that made crucial with the was found to with allowTyr3, a parallel arrangement with Tyr3, forming a tripeptide motifinteractions with Pro2 that made
crucial interactions with the protein that were strikingly similar to the known inhibitors of HPA. On the basis of these findings, the authors hypothesized that the (DY)PY and YAY motifs may be responsible for the inhibitory activity observed in all of the peptides identified, and that similar two-
Biomedicines 2018, 6, 116
10 of 12
protein that were similar to the known inhibitors of HPA. On the basis of these findings, Biomedicines 2018, 6, x strikingly FOR PEER REVIEW 10 ofthe 12 authors hypothesized that the (DY)PY and YAY motifs may be responsible for the inhibitory activity phenolic moiety may have the potential for similar the development of inhibitors targeting observed in all ofmotifs the peptides identified, and that two-phenolic moiety motifs mayamylases have the more generally. potential for the development of inhibitors targeting amylases more generally.
Crystal structure structureof ofHPA HPA(gray) (gray)inincomplex complex with macrocyclic peptide inhibitor, piHAFigure 9. Crystal with thethe macrocyclic peptide inhibitor, piHA-Dm Dm (PDB 5KEZ). The left-hand the position of macrocyclic peptide binding in the site active (PDB 5KEZ). The left-hand panelpanel showsshows the position of macrocyclic peptide binding in the active of site enzyme. of the enzyme. On the right, a magnified view the intramolecular hydrogen bonding the On the right, a magnified view shows the shows intramolecular hydrogen bonding interactions interactions (yellow dashed within the peptide, those including with two water molecules (red (yellow dashed lines) withinlines) the peptide, including withthose two water molecules (red spheres). DY-P-Y motif has been identified as playing a crucial and role isinhighlighted inhibition,inand is spheres). The The DY-P-Y motif has been identified as playing a crucial role in inhibition, green. highlighted green. The thioether cyclizinginlinker is highlighted in magenta. The thioetherincyclizing linker is highlighted magenta.
2.2.3. Other Targets Targets 2.2.3. Other In In contrast contrast to to traditional traditional small small molecules, molecules, cyclic cyclic and and bicyclic bicyclic peptide peptide ligands ligands can can be be identified identified against essentially any protein target, and are not limited to targets with appropriate binding against essentially any protein target, and are not limited to targets with appropriate binding pockets, pockets, as transporters and enzymes, making amenable targetingofof protein–protein protein–protein such as such transporters and enzymes, making themthem amenable to to thethe targeting interactions. However, to the best of our knowledge, only a single co-crystal structure thioether interactions. However, to the best of our knowledge, only a single co-crystal structure for for aa thioether containing inhibitor (not (not including containing peptide peptide macrocycle macrocycle protein–protein protein–protein interaction interaction inhibitor including the the KDM4 KDM4 and and SIRT2 inhibitors described above, which, while technically involved in protein–protein interactions, SIRT2 inhibitors described above, which, while technically involved in protein–protein interactions, involve In this this study, study, Matsunaga involve enzymatic enzymatic processes) processes) has has been been reported. reported. In Matsunaga and and co-workers co-workers identified identified aa high affinity (K = 3.5 nM) thioether-closed cyclic peptide ligand to Plexin B1 (termed PB1m6), D high affinity (KD = 3.5 nM) thioether-closed cyclic peptide ligand to Plexin B1 (termed PB1m6), which was also a potent inhibitor of the interaction of Plexin B1 with Semaphorin 4D; an interaction which was also a potent inhibitor of the interaction of Plexin B1 with Semaphorin 4D; an interaction that regulates osteoblast osteoblastdifferentiation differentiationand andisisa possible a possible target osteoporosis Remarkably, that regulates target forfor osteoporosis [16].[16]. Remarkably, the the co-crystal structure of Plexin B1 with PB1m6 demonstrated that the peptide ligand bound at co-crystal structure of Plexin B1 with PB1m6 demonstrated that the peptide ligand bound at a site asignificantly site significantly distant from the Semaphorin interacting region and was an allosteric inhibitor distant from the Semaphorin interacting region and was an allosteric inhibitor (Figure (Figure 10). Interestingly, the PB1m6 bound PB1m6 a short of section of anti-parallel stabilized 10). Interestingly, the bound formed formed a short section anti-parallel β-sheet,β-sheet, stabilized by four by four backbone amide hydrogen bonds and Arg–Trp cation–pi interactions at the two turns, backbone amide hydrogen bonds and Arg–Trp cation–pi interactions at the two turns, demonstrating the demonstrating the capacity of relatively short thioether cyclized peptides to form recognizable capacity of relatively short thioether cyclized peptides to form recognizable secondary structure motifs. secondary structure motifs.
Biomedicines 2018, 6, 116 Biomedicines 2018, 6, x FOR PEER REVIEW
11 of 12 11 of 12
Figure 10. Co-crystal structure of the thioether closed cyclic peptide PB1m6 bound to human Plexin B1 Figure 10. Co-crystal structure of the thioether closed cyclic peptide PB1m6 bound to human Plexin (PDB 5B4W). Hydrogen bonding between the backbone amides (dashed yellow lines) stabilizes the B1 (PDB 5B4W). Hydrogen bonding between the backbone amides (dashed yellow lines) stabilizes antiparallel β-sheet like structure of PB1m6. The sequence of the peptide (left-to-right, N-to-C) is also the antiparallel β-sheet like structure of PB1m6. The sequence of the peptide (left-to-right, N-to-C) is shown. The thioether linker moiety is highlighted in magenta. also shown. The thioether linker moiety is highlighted in magenta.
3. Conclusions 3. Conclusions The very high affinity and selectivity of thioether-closed macrocyclic peptides, as well as the The very high affinity ofusing thioether-closed macrocyclic as wellligands as the relative ease with which theyand can selectivity be identified modern techniques, makepeptides, them intriguing relative easeapplications. with which As they can be identified using modern can techniques, makediverse them intriguing for diverse described above, such compounds form highly structures ligands for diverse applications. As described above, such compounds can form highly because of their ability to adopt different spatial conformations, and sections of recognizable diverse protein structures because of their ability to adopt spatial conformations, and sections of secondary structure (α-helices and β-sheets) candifferent be observed. Unlike smaller molecules, the bound recognizable protein secondary (α-helices and β-sheets) can be observed. Unlike smaller structures of macrocyclic peptidestructure ligands appear to always involve intramolecular interactions in the molecules, bound structures of macrocyclic peptide ligands and appear always involve macrocycle,the which presumably stabilize their binding conformations allowtothem to adopt the intramolecular interactions in theto macrocycle, presumably stabilize theirpaucity binding requisite conformations for binding highly diversewhich “pockets”. At present, the relative of conformations and allow adopt the requisiteare conformations for binding highly diverse structural information (onlythem a fewtoco-crystal structures currently available) makes to it difficult to draw “pockets”. At present, the relative paucity of structural information (only a few co-crystal structures further general conclusions about the binding modes of thioether-closed macrocyclic peptides to their are currently available) makes it difficult to draw conclusions about is theincreasing binding targets. However, the utility of such compounds, andfurther the factgeneral that their rate of discovery modes thioether-closed macrocyclic to their targets. will However, theinutility offuture, such year by of year, leads us to believe that manypeptides more co-crystal structures be solved the near compounds, and theinsights fact thatinto their of discovery increasing year by year, leads us to believe allowing for further therate structural biologyisof these particularly interesting compounds. that many more co-crystal structures will be solved in the near future, allowing for further insights Funding: This work biology was supported by Joint ANR-JSTinteresting grant (ANR-14-JITC-2014-003 and JST-SICORP), JST CREST into the structural of these particularly compounds. (JPMJCR12L2), and AMED (JP18am0301001 and JP18am0101090) to H.S. Funding: This workThe wasauthors supported bytoJoint ANR-JST grant (ANR-14-JITC-2014-003 JST-SICORP), Acknowledgments: would like to show their sincere gratitude to past and and current members of JST the CREST (JPMJCR12L2), and AMED (JP18am0301001 and JP18am0101090) to H.S. Bioorganic Chemistry laboratory at the University of Tokyo, for their valuable effort, support and contribution.
Acknowledgments: like to show their sincere gratitude to past and current members of Conflicts of Interest:The Theauthors authorswould declaretono conflict of interest. the Bioorganic Chemistry laboratory at the University of Tokyo, for their valuable effort, support and contribution. Conflicts of Interest: The authors declare no conflict of interest.
Biomedicines 2018, 6, 116
12 of 12
References 1. 2.
3. 4.
5.
6.
7. 8. 9.
10.
11.
12.
13.
14.
15. 16.
Angelini, A.; Cendron, L.; Chen, S.; Touati, J.; Winter, G.; Zanotti, G.; Heinis, C. Bicyclic peptide inhibitor reveals large contact interface with a protease target. ACS Chem. Biol. 2012, 7, 817–821. [CrossRef] Chen, S.; Gfeller, D.; Buth, S.A.; Michielin, O.; Leiman, P.G.; Heinis, C. Improving binding affinity and stability of peptide ligands by substituting glycines with D-amino acids. Chembiochem 2013, 14, 1316–1322. [CrossRef] [PubMed] Chen, S.; Bertoldo, D.; Angelini, A.; Pojer, F.; Heinis, C. Peptide ligands stabilized by small molecules. Angew. Chem. Int. Ed. Engl. 2014, 53, 1602–1606. [CrossRef] [PubMed] Tanaka, Y.; Hipolito, C.J.; Maturana, A.D.; Ito, K.; Kuroda, T.; Higuchi, T.; Katoh, T.; Kato, H.E.; Hattori, M.; Kumazaki, K.; et al. Structural basis for the drug extrusion mechanism by a MATE multidrug transporter. Nature 2013, 496, 247–251. [CrossRef] [PubMed] Hipolito, C.J.; Tanaka, Y.; Katoh, T.; Nureki, O.; Suga, H. A macrocyclic peptide that serves as a cocrystallization ligand and inhibits the function of a MATE family transporter. Molecules 2013, 18, 10514–10530. [CrossRef] [PubMed] Kodan, A.; Yamaguchi, T.; Nakatsu, T.; Sakiyama, K.; Hipolito, C.J.; Fujioka, A.; Hirokane, R.; Ikeguchi, K.; Watanabe, B.; Hiratake, J.; et al. Structural basis for gating mechanisms of a eukaryotic P-glycoprotein homolog. Proc. Natl. Acad. Sci. USA 2014, 111, 4049–4054. [CrossRef] [PubMed] Hayashi, Y.; Morimoto, J.; Suga, H. In vitro selection of anti-Akt2 thioether-macrocyclic peptides leading to isoform-selective inhibitors. ACS Chem. Biol. 2012, 7, 607–613. [CrossRef] [PubMed] Ito, K.; Sakai, K.; Suzuki, Y.; Ozawa, N.; Hatta, T.; Natsume, T.; Matsumoto, K.; Suga, H. Artificial human Met agonists based on macrocycle scaffolds. Nat. Commun. 2015, 6, 6373. [CrossRef] [PubMed] Kawakami, T.; Ishizawa, T.; Fujino, T.; Reid, P.C.; Suga, H.; Murakami, H. In vitro selection of multiple libraries created by genetic code reprogramming to discover macrocyclic peptides that antagonize VEGFR2 activity in living cells. ACS Chem. Biol. 2013, 8, 1205–1214. [CrossRef] [PubMed] Yamagishi, Y.; Shoji, I.; Miyagawa, S.; Kawakami, T.; Katoh, T.; Goto, Y.; Suga, H. Natural product-like macrocyclic N-methyl-peptide inhibitors against a ubiquitin ligase uncovered from a ribosome-expressed de novo library. Chem. Biol. 2011, 18, 1562–1570. [CrossRef] [PubMed] Kawamura, A.; Munzel, M.; Kojima, T.; Yapp, C.; Bhushan, B.; Goto, Y.; Tumber, A.; Katoh, T.; King, O.N.; Passioura, T.; et al. Highly selective inhibition of histone demethylases by de novo macrocyclic peptides. Nat. Commun. 2017, 8, 14773. [CrossRef] [PubMed] Yamagata, K.; Goto, Y.; Nishimasu, H.; Morimoto, J.; Ishitani, R.; Dohmae, N.; Takeda, N.; Nagai, R.; Komuro, I.; Suga, H.; et al. Structural basis for potent inhibition of SIRT2 deacetylase by a macrocyclic peptide inducing dynamic structural change. Structure 2014, 22, 345–352. [CrossRef] [PubMed] Morimoto, J.; Hayashi, Y.; Suga, H. Discovery of macrocyclic peptides armed with a mechanism-based warhead: isoform-selective inhibition of human deacetylase SIRT2. Angew. Chem. Int. Ed. Engl. 2012, 51, 3423–3427. [CrossRef] [PubMed] Yu, H.; Dranchak, P.; Li, Z.; MacArthur, R.; Munson, M.S.; Mehzabeen, N.; Baird, N.J.; Battalie, K.P.; Ross, D.; Lovell, S.; et al. Macrocycle peptides delineate locked-open inhibition mechanism for microorganism phosphoglycerate mutases. Nat. Commun. 2017, 8, 14932. [CrossRef] [PubMed] Jongkees, S.A.K.; Caner, S.; Tysoe, C.; Brayer, G.D.; Withers, S.G.; Suga, H. Rapid Discovery of Potent and Selective Glycosidase-Inhibiting De Novo Peptides. Cell. Chem. Biol. 2017, 24, 381–390. [CrossRef] [PubMed] Matsunaga, Y.; Bashiruddin, N.K.; Kitago, Y.; Takagi, J.; Suga, H. Allosteric Inhibition of a Semaphorin 4D Receptor Plexin B1 by a High-Affinity Macrocyclic Peptide. Cell. Chem. Biol. 2016, 23, 1341–1350. [CrossRef] [PubMed] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
