Peptide cyclization has become known for not only expanding upon the properties of linear peptides but also remedying some of their limitations. Peptide cyclization decreases the spatial vibration of the peptide molecule resulting in conformational changes. This structural change also creates an increase in the surface area for contact with the biological target. These factors result in an increase in binding affinity and selectivity to the target. Along with these pharmacodynamic benefits, the pharmacokinetic properties of peptides can be improved. Cyclization rigidifies the structure, thus lowering the energy barrier required for a peptide to adapt to the membrane environment and bind to transport proteins for cellular entry by passive diffusion or active transport.
Linear peptides are vulnerable to rapid degradation by endo- and exoproteases in vivo, with many showing half-lives in plasma of only seconds to minutes. Cyclization overcomes this problem by fixing the peptide into a rigid, pre-organized conformation that protects backbone amide bonds and removes free N- and C-termini, the preferred targets of aminopeptidases and carboxypeptidases. The reduction in conformational entropy also reduces the likelihood that the peptide will adopt the extended conformation required for productive binding to the catalytic cleft of serine or cysteine proteases. Experimental support for these benefits is provided by the finding that cyclization of the angiogenesis-targeting peptide A7R increased its serum stability from < 2 h to >12 h in mouse plasma, which correlated directly with improved tumor accumulation in xenograft models. Equivalent improvements have been reported for cyclized cell-penetrating peptides (cTAT), which maintained full transduction activity after 24 h in 90 % human serum, while the linear parent was undetectable within minutes. Researchers have several cyclization options available including disulfide, lactam, hydrocarbon staple, and triazole linkages to manipulate ring size and rigidity enabling multi-parameter optimization without losing binding affinity. In addition, such bicyclic scaffolds provide another layer of protection, and cyclic peptides incorporating two orthogonal linkages have been found to be >10-fold more stable toward trypsin compared with their monocyclic analogs. Importantly, this gain in protease resistance is achieved without the need for D-amino acids or retro-inverso (RR) motifs, two features which can present significant synthetic challenges and also raise regulatory concerns. For this reason, cyclization is the preferred strategy for converting labile peptide hits into leads with improved proteolytic stability.
For lead optimization to succeed compounds must show stable target binding alongside sufficient membrane permeability and favorable pharmacokinetic properties. Cyclic peptides are known to have advantages over linear analogues in these aspects. In addition to improving target binding stability, conformational constraint also decreases the polar surface area (PSA) by tucking hydrophilic side chains inward and positioning hydrophobic residues outward. The reduced PSA increases passive permeability across biological membranes. Quantitative comparisons in Caco-2 monolayers have demonstrated up to 15-fold improvements in the apparent permeability (Papp) for cyclic CPPs over their linear counterparts. Furthermore, cyclization relieves the entropic cost of membrane insertion; molecular dynamics simulations suggest that cyclic peptides can spontaneously adopt low-energy, membrane-inserted conformations with hydrophobic vectors parallel to the membrane surface. The conformational constraint also decreases nonspecific protein binding and results in cleaner intracellular distribution profiles. In terms of PK properties, cyclic peptides are typically associated with a lower rate of renal clearance because the conformational constraint increases the plasma protein binding affinity, reducing the free-fraction of the compound available for glomerular filtration. For example, in pre-clinical animal models, the cyclic integrin antagonist cilengitide (t1/2 ≈ 4 h) had approximately four-fold longer terminal half-life compared to a linear control peptide of similar size, which enabled twice-daily dosing for cilengitide in Phase II clinical trials. Furthermore, other modifications such as selective N-methylation and the addition of lipophilic ncAAs can be used in a synergistic manner with the cyclic scaffold to tune the clearance pathways and oral bioavailability without affecting target engagement.
Cyclized polypeptides have been identified in nature and the cyclisation is often an integral property of the molecule's biological activity. There are four broad classes of cyclisation, including side chain-to-side chain, head-to-tail (also known as backbone cyclisation), head-to-side chain and side chain-to-tail. Side chain-to-side chain cyclisation occurs when a bond is formed between side chain functionalities of different amino acid residues. The most widely studied example of side chain-to-side chain cyclisation is that of intramolecular disulphide bond formation between the thiol functionalities of two cysteine residues, a type of cyclic structure often found in peptides and proteins, including insulin and antibodies. It is estimated that up to 50% of cysteine residues in polypeptides are present in the form of a disulphide bond. Head-to-tail terminus cyclisation is another common form of cyclisation. As the first residue in a chain of amino acids has an amino functionality (i.e. N-terminus), and the last residue has a carboxylate functionality (i.e. C-terminus), polypeptides are generally directional in structure. Cyclisation can be achieved by joining the N- and C-termini of the peptide via an amide bond. Head-to-tail peptide cyclisation has been observed in both microorganisms and plants, such as kalata B1 from Oldenlandia affinis and bacteriocin AS-48 produced by the bacterium Enterococcus faecalis..
Fig.1 Diverse peptide cyclization methods.1,2
Direct amide bond formation: The amide bond between carboxylic acid and amine groups can be formed by direct condensation in the presence of high temperatures or microwave irradiation. These harsh reaction conditions, however, are not compatible with most peptides and proteins. As a result, reactions under milder conditions have been developed for polypeptide ligation. Typically, this is achieved by first converting the C-terminal –OH into a better leaving group, e.g. an acyl halide, acyl azide, anhydride or activated ester, using a coupling reagent. Nucleophilic (e.g. Lys, Ser, Thr) and carboxylate (e.g. Asp, Glu) amino acid side chains therefore have to be protected to avoid side reactions, and as a result, this approach is more suited for peptides that can be synthesised in a fully protected form. In addition, additives are often used in combination with coupling reagents to suppress racemisation at the ligation site and improve the rate of the reaction. For example, in the total synthesis of a 13 residue depsicyclic peptide antibiotic, texiobactin, the cyclisation step was carried out successfully using a combination of coupling reagents (HOAt/OxymaPure/HATU) with a tertiary amine base DIEA.
Native chemical ligation: Native chemical ligation (NCL) enables two unprotected peptide fragments with C-terminal thioester and N-terminal cysteine residues respectively to be joined together. Both fragments can be obtained either chemically (synthesized on solid-phase) or recombinantly (produced by cells). In addition, the reaction can take place in aqueous solution at neutral pH, and is compatible with the presence of chaotropic reagents (such as guanidine hydrochloride) and reducing agents. Mechanistically, the initial step involves a nucleophilic attack by the thiol group of the N-terminal cysteine on the carbonyl carbon of the thioester group. This results in rapid and reversible transthioesterification, followed by an S-to-N acyl shift that generates the desired peptide bond. The reaction is both regio- and chemoselective, meaning that neither internal cysteine residues nor any other nucleophilic amino acid side chains interfere.
Ligations relying on a C-terminal aldehyde: Serine and threonine residues can also be used for the ligation of unprotected peptide fragments, and thus for cyclisation. Thus, for example, a C-terminal glycolaldehyde ester reacts chemoselectively with an N-terminal serine or threonine (or cysteine) residue to form an oxazolidine intermediate, which upon rearrangement furnishes a peptide bond in the form of a pseudoproline structure. However, the reaction is slow and an unnatural functionality is left at the ligation site. A modified approach using a C-terminal salicylaldehyde was developed to overcome these limitations. After oxazolidine formation from the chemoselective reaction of the salicylaldehyde with the N-terminal Ser/Thr, an N,O-benzylidene acetal amide intermediate is generated upon O–N acyl shift. Using TFA, the acyl group can then be removed, leaving a native peptide bond at the ligation site. This Ser/Thr ligation approach has been successfully applied to the synthesis of a number of cyclic peptide natural products, including daptomysin, cyclomontanin B, mahafacyclin B, etc.
Disulphide bonds: The creation of disulphide bonds stands out as the primary chemical technique for inducing cyclisation. Up to 30% of eukaryotic proteins contain at least one disulphide bond, which can stabilise the three-dimensional structure of the protein and can also be used to regulate protein function. The incorporation of cysteine residues into amino acid chains is straightforward by either chemical or recombinant means, and disulphide bonds are formed spontaneously on exposure to air. This principle has been used to generate cyclic proteins by introduction of two cysteine residues.18 For proteins containing a number of cysteine residues there are multiple possible patterns of disulphide bonds. For example, 6 cysteine residues can form 3 intramolecular disulphide bonds in 15 (5 × 3 × 1) possible ways. Although in cells this problem is solved by the production of enzymes that catalyse the formation of the correct disulphide bond pattern, in a reaction vessel it is difficult to control and usually results in a mixture of products.
Subtiligase variants: Subtiligase is a double mutant of the serine protease subtilisin BPN' from Bacillus amyloliquefaciens. It has a range of substrate sequences that it can recognise. Furthermore, protein engineering has enabled variants to be generated that have a broader substrate scope, or which can function in a Ca2+-independent manner. The combination of efficiency and broad substrate scope make subtiligase variants an appealing tool to achieve traceless ligation to backbone cyclised peptidyl products. However, these desirable properties require the use of C-terminal (thio)ester substrates by subtiligase variants. While recombinant production of a protein with a C-terminal thioester is possible by an intein-mediated strategy, this is sequence-dependent and typically has long operation steps that result in a low yield. Therefore, ester and thioester substrates are often created by chemical synthesis, restricting potential applications.
Sortases: Sortases are a family of cysteine transpeptidases that are widely distributed in bacterial species, especially Gram-positive bacteria. Of those sortases which have been characterised, sortase A from Staphylococcus aureus (SrtA) has been the most commonly used for modification of peptides and proteins. SrtA specifically recognises the amino acid sequence LPXTG, where X can be any amino acid, and catalyses the cleavage of the amide bond between Thr and Gly residues followed by formation of a thioester intermediate at the C-terminus of Thr and subsequent nucleophilic attack of a peptide containing a N-terminal Gly. With respect to peptide cyclisation, the length and concentration of the peptide has been shown to play a role in determining SrtA preference for backbone cyclisation or oligomerisation. It has been reported that a minimum peptide substrate length of 19 residues (including the LPXTG motif) is necessary for cyclisation to be favoured over intermolecular reactions (i.e. formation of dimers and trimers, in linear or cyclic forms). Increased peptide concentration (>1 mM) unsurprisingly led to increased intermolecular di- and trimerisation. SrtA has also been used for cyclisation of larger recombinant proteins, including several cytokines, green fluorescence protein (GFP) and ubiquitin C-terminal hydrolase L3.
Asparaginyl endopeptidases: Asparaginyl endopeptidases (AEPs) are a family of cysteine proteases that hydrolyze peptide bonds C-terminal to an Asx residue (i.e. an Asn or Asp residue). AEPs are widely distributed in the plant kingdom, with the majority of them possessing transpeptidation ability, resulting in the production of naturally occurring cyclic peptides. Butelase 1 from Clitoria ternatea and asparaginyl endopeptidase 1 from Oldenlandia affinis (OaAEP1) are the two most well-studied examples which have been applied to peptide and protein cyclisation.
Transglutaminase: Transglutaminases are a family of enzymes, widely distributed in microorganisms, plants and animals, which catalyse an acyl transfer reaction between the carboxyamide group of glutamine residues and primary amines (including the ε-amino group of lysine residues). NH3 is released as a by-product. The crosslinking amide bond, termed an isopeptide bond, is chemically and proteolytically stable. A calcium-dependent microbial transglutaminase from Streptomyces mobaraensis was used to cyclize a range of peptide sequences between 11 and 23 amino acids. The enzyme was found to have a broad substrate specificity, but the substrate required an Ala and Leu dipeptide sequence at the N-terminal side of the glutamine residue. In general, isopeptide bond formation will occur provided both substrate lysine and glutamine residues are accessible to the enzyme. Sequence of the amino acids flanking the glutamine residue may also influence the reaction yield.
One of the most impactful strategies to enhance proteolytic stability while simultaneously increasing conformational diversity is through the incorporation of mirror-image or non-canonical building blocks. D-amino acids are not substrates for canonical proteases, and the replacement of just 2 L-residues with their D-enantiomers can increase serum half-life from minutes to hours without affecting receptor binding. Systematic scans of the peptide identify "hot-spot" positions (generally glycines or solvent-exposed residues) where D-substitution will have the greatest effect on protease evasion while having the least impact on the pharmacophore. In addition to stereochemical inversion, we incorporate non-natural backbones (N-methylated amides, α,α-dialkyl glycines, β-amino acids) to introduce local rigidity and occlude endoprotease cleavage sites. To increase the lipophilicity of membrane-permeable cyclic peptides, FPs and cyclohexyl alans are used to retain α-helix mimicry while improving Caco-2 permeability (Papp) 5- to 20-fold relative to native sequences. The installation of clickable handles (azido-homoalanine, propargyl-glycine) during SPPS allows for late-stage diversification by CuAAC or SPAAC, which can rapidly explore SAR without re-synthesis of the entire scaffold. Notably, all ncAAs are available as Fmoc-protected building blocks that couple under standard SPPS conditions, without the need for custom reagents.
The translation of strong in vitro performance to effective in vivo results involves attaching stealth or membrane-interactive components to the cyclic framework. Stealth PEGylation (covalent conjugation of polyethylene glycol chains) produces a hydrated volume element that sterically shields from the kidneys due to an increased hydrodynamic radius. Site-selective PEGylation strategies have been developed, with one C-terminal cysteine being targeted with maleimide-PEG reagents under mild and thiol-selective conditions to produce monopegylated products with polydispersity indices (PDIs)<1.05. Linear 20- and 40-kDa PEGs have been shown to increase plasma half-life by 10- to 50-fold in mouse PK studies and branched 2 × 20 kDa architectures further reduce mononuclear phagocyte system (MPS) uptake. For oral administration, lipidation provides orthogonal advantages. We tether C12–C18 fatty acids to stable amide or oxime linkers to the N-terminus of the cyclic peptide or to a strategically engineered lysine side chain. The lipid tail embeds into chylomicron membranes to facilitate lymphatic uptake and prevent first-pass metabolism in the liver. In a recent drug discovery program against an intracellular PPI target, lipidation of a compound that had <2 % oral bioavailability improved it to 18 % and increased t½ in rats from 0.4 h to 3.2 h, without loss of target affinity. As both PEG and lipid moieties are attached post-cyclization, we can conduct rapid "mix-and-match" SAR studies, synthesizing libraries of differentially PEGylated or lipidated analogs in a single week.
Peptide Modification Services at Creative Peptides
The process preferred by regulatory bodies for confirming the ≥95 % purity of a lead cyclic peptide is reversed-phase high-performance liquid chromatography followed by mass spectrometry. In our laboratory, every batch is initially evaluated on a UPLC-MS system that delivers ultra-high-resolution chromatography and sub-ppm mass accuracy. The chromatographic separation is achieved on a C18 column (100 × 2.1 mm, 1.7 µm) at 60 °C using a shallow acetonitrile gradient buffered with 0.1 % formic acid; these conditions separate important epimers, oxidation products, and truncated sequences that differ by as little as 0.02 min in retention time. Absorbance is monitored at 214 nm (peptide bond) and 280 nm (Trp/Tyr), while concurrent electrospray ionization (ESI) in positive-ion mode provides exact-mass confirmation of the expected [M+H]+ and [M+2H]2+ ions. Any impurity peak that comprises >0.5 % of the total UV area is automatically highlighted and further characterized by HR-MS/MS fragmentation, permitting unequivocal assignment of sequence errors, des-amido variants, or side-chain modifications. For peptides >30 residues in size or those with multiple disulfide bonds, we add size-exclusion chromatography (SEC) to the workflow in order to quantify aggregates and fragments >5 kDa that are undetectable by RP-HPLC.
Although chromatography establishes purity, unambiguous evidence for the intended macrocycle topology and secondary structure is provided by high-field nuclear magnetic resonance (NMR) and circular dichroism (CD) spectroscopy. For peptides of up to 9 residues, complete 2D-NMR characterization is standard: 1H-1H COSY and TOCSY spectra determine scalar connectivities, while NOESY or ROESY experiments give distance restraints that map out the spatial proximity of key protons throughout the ring. 13C and 15N heteronuclear single-quantum coherence (HSQC) spectra are acquired on a 600 MHz spectrometer with cryoprobe, providing atomic-level resolution of side-chain conformations and verifying that cyclization has taken place between the correct residues. Larger or more flexible macrocycles are characterized by CD spectroscopy. Far-UV spectra (190–260 nm) obtained at 25 °C in phosphate buffer immediately provide information about secondary-structure content: negative maxima at 208 and 222 nm reveal α-helicity, a single minimum at 217 nm indicates β-sheet character, and a broad negative band near 200 nm is diagnostic for disordered coil. Thermal denaturation experiments (20–90 °C) determine melting temperatures (Tm) and show that the cyclic scaffold conveys increased conformational stability. Where appropriate, CD spectra are measured in the presence of 30 % trifluoroethanol to simulate membrane environments and confirm that the designed helical propensity is preserved under amphipathic conditions.
Peptide drug discovery happens at the speed of ideas, and our platform is designed accordingly. Projects start with a virtual consult where medicinal chemists, structural biologists, and process engineers collectively author a "design brief." Within 24 h we deliver a project-specific synthetic route that outlines recommended cyclization chemistries, protecting-group strategies, and orthogonal handles for late-stage diversification. Automated microwave-enhanced SPPS then constructs the first 1- mg scale batch in as little as 48 h, while a parallel microfluidic flow-cyclization module streams out 10–50 mg variants for same-week SAR screens. As soon as analytical QC (UPLC-MS, HR-MS/MS, and chiral HPLC) verifies ≥95 % purity, the data package is uploaded to a secure client portal; chemists can therefore run binding or cell-penetration assays before the calendar page turns. If hits need rapid iteration, our "click-and-ship" service taps a library of pre-activated ncAA building blocks and an AI-assisted sequence optimizer that forecasts potency, stability, and permeability effects in silico. This closed-loop cycle—design, synthesize, purify, assay—can be repeated weekly, enabling hundreds of analogues to be assessed in the same timeframe that classical CROs devote to a single analogue. Finally, every synthetic route is captured as a digital twin; if a particular motif advances, the route can be handed off to our scale-up team without re-optimization, saving weeks of downstream development time.
After a lead peptide results from iterative cycles of SAR, our in-house scale-up engine allows for a seamless transition from mg to kg scale. Route-scouting chemists quickly assess the pros and cons of three orthogonal production platforms—solid-phase, solution-phase fragment condensation, or a hybrid recombinant expression strategy—weighing factors of yield, COG, and regulatory risk. For larger macrocycles with multiple disulfide or thioether linkages, we leverage a proprietary oxidative folding methodology in precisely defined redox buffers that can routinely achieve >70 % isolated yield at gram scale without compromising the native disulfide topology. Process analytical technology (PAT) is integrated at every stage: in-line Raman spectroscopy is used to monitor deprotection efficiency, and UPLC-MS analysis is used to trigger on-the-fly diversion of off-spec material such that each batch easily exceeds ICH Q7 standards. Our GMP-capable suite is under a quality-by-design (QbD) program; critical process parameters (CPPs) such as coupling equivalents, base strength, and cyclization temperature are defined well before scale-up to ensure<2 % RSD in batch-to-batch variation.
Peptide Synthesis Services at Creative Peptides
References
