The head-to-tail topology is contrasted with the side-chain cyclization strategy, another common approach for peptide engineering. Head-to-tail cyclization forms an amide bond between the N-terminus and C-terminus, whereas side-chain cyclization connects side chains of amino acids within the peptide. The side chains can be joined by amide, disulfide, or other types of linkages. For example, amide bonds can be formed between side-chain carboxy and amino groups, and disulfide bonds can be formed between cysteine residues. In addition, the effect of ring closure on the peptide structure and function could be different between the two topologies.
Variety of topologies were employed or designed with time which could be considered to be a more or less logical consequence of the variety of limitations of linear peptides and biological targets. Head-to-tail (cap-to-cap) cyclization addressed issues of proteolytic stability because the resulting peptide is no longer a substrate for most exopeptidases (loss of N- or C-terminal recognition sites) and the conformationally constrained peptide is more difficult to cleave for most endoproteases (sterically congested backbone). It also imposes a unique closed loop topology to the peptide, offering conformational preorganization and a reduced entropic penalty on binding, and allowing higher affinity binding to large protein surfaces, which many small molecules can not address. This closed loop topology also offers an increase in rigidity which increases cell membrane permeability (the polar backbone amides are protected in intramolecular hydrogen bonds and can move more freely across lipid bilayer). Side-chain cyclization in the form of disulfide bridges, lactam linkages, or other linking chemistries, on the other hand, is largely the result of natural design for the most part, to stabilize a bioactive conformation while still offering some degree of structural flexibility. Peptides can be designed with specific secondary structures (β-hairpin, α-helical bundle, etc.) through a crosslink that can nucleate the structure and hold it in place without completely restricting the backbone flexibility. The conformational and functional impact of side-chain cyclization is substantial: it can alter the presentation of side-chain chemical groups to create a surface that can mimic protein epitopes or enzyme active sites, for example, while still offering enough conformational flexibility to engage in induced-fit recognition. The choice of topology will determine not only the synthetic strategy and yield, but also the final stability, selectivity, and pharmacokinetic behavior of the macrocycle.
Fig. 1 Cyclic peptides in general.1,5
Head-to-tail cyclization is when an amide bond is formed between the N-terminus and C-terminus of a peptide. Head-to-tail cyclization normally starts with a linear peptide and then is chemically or enzymatically cyclized afterwards. For example, in solid-phase peptide synthesis (SPPS), the linear peptide is bound to a resin. Then the N- and C-termini are deprotected and a bond between them is promoted.
Head-to-tail cyclization is the intramolecular amide bond formation between the N-terminus and C-terminus of a linear peptide precursor. The process of cyclization is not trivial and involves careful considerations of activation chemistry, reaction conditions, and conformational preorganization to favor ring closure. The linear peptide sequence is often assembled on solid phase using Fmoc or Boc chemistry, and the side-chain functional groups are orthogonally protected as needed. The peptide precursor is cleaved from the resin under conditions that do not remove the side-chain protection groups, and the partially protected peptide is often soluble in organic solvents where cyclization is carried out. The C-terminus is then activated with phosphonium or aminium coupling reagents such as HBTU or HATU to form an activated ester that is subject to nucleophilic attack by the N-terminus. High dilution (usually submillimolar concentrations) is often used to maximize the efficiency of intramolecular nucleophilic capture, while minimizing parasitic intermolecular oligomerization, which occurs by the same mechanism and lowers the yield of the desired product. The tetrahedral intermediate formed during the reaction collapses with concomitant loss of the leaving group to form the macrocyclic amide. The use of tertiary amine bases that do not act as nucleophiles can also promote cyclization by deprotonating the N-terminus to form a nucleophilic amide anion. The increased conformational rigidity imparted by head-to-tail cyclization often has a positive impact on the bioactivity and target binding of the peptide. Cyclic peptides formed via head-to-tail cyclization have been reported to have increased metabolic stability and membrane permeability relative to their linear analogues.
Head-to-tail cyclization is a different approach that has its own synthetic and functional advantages. The most prominent is that the reaction manifold is relatively simple: a single, well-defined cyclic product forms, mimicking natural ribosomal macrocycles in topology. This not only gives rise to structural elements that are inherently proteolytically resistant and conformationally rigid, but also has synthetic advantages. For example, because N- and C-termini are the only functional groups undergoing reaction in head-to-tail cyclization, there is no positional isomerism as can occur in side-chain cyclization where multiple nucleophilic side chains can be acylated. These properties make purification and characterization easier, and structural elements like cyclotides and bacteriocins are notable for their stability in biological matrices because they do not have free termini and are generally sterically inaccessible to proteases, affording long circulations and persistent bioactivity. Conformational rigidity also helps target selectivity because the pharmacophore array of a cyclic peptide can present a preorganized interaction surface to the target, but not be as easily reconfigured on off-target surfaces. Finally, on-resin cyclization can take advantage of pseudo-dilution effects because the peptide is bound to an insoluble support. This effectively reduces oligomerization without needing extremely low solution concentrations, and increases reaction rates as well as atom economy. A linear peptide bound to resin is also easier to workup, because the cyclic product is trapped on resin and can easily be washed free of excess reagents before cleavage and deprotection. A few issues have also been seen with head-to-tail cyclization. Reaction optimization can be more difficult as one must exert fine control on the cyclization reaction itself to suppress dimerization and oligomerization. Synthesis can also be more complex, with the need for additional reaction conditions and purification steps. The rigidity in conformation can also be a problem if other conformations of the peptide are needed for activity.
Side-chain cyclization of peptides is a form of peptide macrocyclization that is structurally orthogonal to head-to-tail topology in that the covalent bridge in the macrocycle forms between side chains of the peptide's amino acid residues and not the peptide backbone termini. A common implementation of side-chain cyclization is the use of lactam bridges which are formed between the ε-amino group of a lysine and the carboxylate side chain of aspartic or glutamic acid, or the formation of disulfide bridges between the thiol groups of cysteine side chains. Side-chain cyclization is synthetically orthogonal to head-to-tail closure of a peptide's termini because the backbone termini remain free and functional for further chemical modifications or interactions. Side-chain cyclization can be useful for macrocyclizing sequences with sterically encumbered termini, and sequences that would have unfavorable ring size and thus high ring strain upon head-to-tail closure. Additionally, side-chain cyclization can be used to introduce conformational constraints into specific regions along the peptide chain, providing more targeted structural modulation of a peptide's secondary structure and pharmacophore. Side-chain cyclization can be more synthetically challenging than head-to-tail because it requires the use of orthogonal protecting groups: those on the side chains to be cyclized are deprotected, and those on other side chains (and of course the peptide termini) remain protected. As the two ends of a head-to-tail macrocycle are structurally equivalent, only a single product is possible, however side-chain cyclization between two possible side chains can result in multiple regioisomeric products, so sequence design and protecting group strategy must be used to impart chemoselectivity. Macrocycles incorporating side-chain cyclization often have an intermediate level of structural rigidity: more rigid than unmodified linear peptides but sometimes less conformationally constrained than head-to-tail macrocycles. This combination of characteristics has been especially useful in the design of cell penetrating peptides, and peptides that must bind to and adapt to conformationally dynamic surfaces.
Lys-Glu and Asp-Lys lactam bridges are very common. They are formed by side-chain cyclization using the nucleophilic ε-amino group of lysine to react with the electrophilic carboxylate side chains of glutamic acid or aspartic acid to form a constrained amide bond. This is typically performed by solid phase synthesis of the linear precursor with orthogonal protection groups, such as an acid labile group for the lysine side chain (e.g. ivDde or Mmt) and a base labile protecting group (e.g. allyl) for the Glu/Asp side chain, then chemoselectively deprotecting the side chains to form the desired linkage. This is performed by deprotecting the lysine side chain with mild acid, then coupling it with a reagent like HATU or PyBOP to make the carboxylate side chain of the partner acid residue sufficiently electrophilic to be nucleophilically attacked by the deprotonated lysine amine, forming a lactam-bridged macrocycle. The resulting seven or eight-membered ring (seven if Asp and eight if Glu is used) puts a kink in the peptide backbone which is then able to nucleate β-turns (closer spacings) or α-helical segments (larger spacings). Stapled peptides often rely on this form of cyclization, as the i to i+4 lactam bridge (Lys at position i and Glu at i+4, for example) not only locks the peptide into a helical conformation, but forms an additional hydrogen bond which more closely mimics the native backbone hydrogen bonding pattern of an α-helix. The size of the bridge (ring length) can be crucial for proper folding: shorter linkers introduce more ring strain, while longer spacers do not constrain the peptide backbone enough to induce proper folding. Other challenges with this approach can include oligomerization (due to competing amide formation), and the need for orthogonal deprotection that reduces the overall synthetic yield. Disulfide bridges between cysteine residues are also a common type of side-chain cyclization. These can be formed through a variety of functional group chemistries, but are most often made by oxidation of the cysteine thiols.
Side-chain cyclization has additional architectural implications as well. There is a trade-off between target selectivity and conformational flexibility. This has a direct impact on both drug activity and drug properties. As described above, side-chain cyclization connects two residues in the peptide backbone via a lactam or disulfide bridge. In so doing it preorganizes the peptide backbone in a low energy, bound-like conformation and reduces the entropic cost of binding. This is particularly useful for protein–protein interaction inhibitors which bind to large flat surfaces and can use side-chain cyclization to bring together distant binding surfaces and design ligands with high specificities, in some cases similar to that of antibodies. An additional benefit is the decreased conformational entropy of the peptide in the unbound state. Lead optimization must only consider a smaller number of possible bioactive conformers. The trade-off to increased selectivity is decreased conformational flexibility. In some cases, too much restriction can be disadvantageous. Active sites of enzymes are known to change conformation in an induced-fit manner when binding to substrates. An inflexible cyclic peptide may not be able to mimic the conformational dynamics of the substrate, and although it may bind with high affinity it may not be a good catalyst inhibitor. Cell-penetrating peptides must have some conformational plasticity to traverse the lipid bilayer. In some cases an increase in structural rigidification through crosslinking may decrease cell permeability. Furthermore, retention of free termini increases the polar surface area of the molecule, which may also affect cell permeability. On the other hand, the additional termini could be used for further PEGylation or for attachment of targeting moieties. The crosslink length and chemistry can also be varied to either increase or decrease flexibility. For example, a shorter lactam crosslink will place more of a conformational restriction than a longer length, or a disulfide bond.
Backbone–side-chain hybrid rings involve the use of more than one mode of cyclization in the same peptide. In this case, head-to-tail backbone cyclization is joined by at least one side-chain crosslink to yield bicyclic or polycyclic topologies. The use of these arrangements is a more advanced approach to peptide cyclization, in that it often allows greater topological diversity than can be obtained from any one single cyclization method alone.
Backbone–side-chain hybrid rings are typically designed and synthesized based on the logic that no single cyclization chemistry can be a panacea for all challenges, and that combining the useful aspects of two or more cyclization strategies can have a synergistic effect when appropriately arranged on a peptide backbone. Such motifs are distinguished from ordinary monocyclic peptides by containing at least one backbone macrocycle connecting the N- and C-termini and one or more side-chain crosslinks linking particular residues in the loop, and are usually prepared through a synthetic sequence in which the peptide is first assembled as a linear precursor on solid support with mutually orthogonal protection groups on side-chain functionalities that can form the intended crosslinks. The resulting fully protected linear peptide is cleaved from resin, then cyclized at the backbone under conditions of high-dilution to avoid intermolecular reactions, often by native chemical ligation or head-to-tail amide bond formation, to give a protected monocyclic peptide. This is then opened up under orthogonal conditions that selectively deprotect only the two residues forming the intended side-chain crosslink, which is then cyclized intramolecularly under standard conditions for the chosen chemistry to give the hybrid bicyclic (or polycyclic) product. The properties of the backbone and side-chain cyclic motifs are generally complementary in that the former enforces global tertiary structure and resistance to exoproteolysis, whereas the latter imparts local secondary structure elements and confers additional protection against endoproteolysis. The range of different chemistries available for backbone and side-chain cyclization is quite broad, including amide bonds, disulfide bonds, and thioether and hydrocarbon staples, which can be selected to tune the properties of the hybrid ring as desired, for example by adjusting hydrophobicity, redox susceptibility, and backbone dynamics. Computational design approaches can be used to predict the best positions for such crosslinks to stabilize the desired tertiary structure while minimizing steric clash, which is an important consideration since backbone and side-chain constraints may be in tension with each other.
Dual-cyclic peptides are a type of hybrid topology in which backbone cyclization is used in combination with a second crosslink, to generate bicyclic structures. These topologies have found applications in which their enhanced stability and expanded functional capacity are superior to monocyclic analogs. A common example is the construction of bicyclic antimicrobial peptides via head-to-tail cyclization in combination with a side-chain disulfide to form a topology similar to naturally occurring θ-defensins, which are circular peptides found in primates that feature both a backbone macrocycle and two disulfide bonds. In these molecules, the resistance to exopeptidases and stable presentation of cationic residues from the backbone ring is combined with the ability to stabilize a β-hairpin that is important for pore formation due to the internal disulfide bond, to yield a highly potent and broad-spectrum antimicrobial peptide with low hemolytic activity.
Stapled analogs of cyclic peptides are an extension of the concept of peptide stapling, which was originally developed for linear peptides to stabilize helical secondary structures, to macrocyclic structures, in which a hydrocarbon or lactam staple is installed across α-helical segments in addition to a backbone macrocycle. For instance, a cyclic peptide with an intended target of protein–protein interactions may be modified with an all-hydrocarbon staple across the i and i+4 residues that is located within the macrocyclic ring, to form a constrained helical bundle that displays superior proteolytic stability and improved cell permeability. The staple can be installed through ring-closing metathesis of olefinic amino acids that are included during SPPS, followed by backbone cyclization using an orthogonal chemistry, to yield the stapled cyclic peptide which benefits from both global protection from backbone closure and local stabilization from the staple, with the two effects adding to form a synergistic increase in oral bioavailability and target affinity.
In general, head-to-tail cyclization offers the most stable, most protease-resistant macromolecules that resemble natural ribosomal products. These molecules have the potential to be the most metabolically stable and orally available, but head-to-tail cyclization usually involves low yields, the potential for epimerization, and difficulties in local conformation tuning without affecting the global ring constraint. Side-chain cyclization is easier to incorporate from a synthetic perspective and allows cyclization of sequences where the termini may not be accessible or when backbone flexibility is desired. The resulting macrocycles maintain an intermediate level of rigidity and therefore an induced-fit binding is still possible, but there are additional new vulnerabilities such as potential cleavage of the lactam bridges in acidic conditions or of the disulfide bond in reducing environments. Side-chain cyclization often involves the use of complex orthogonal protection schemes, and as such may result in higher synthetic steps. Hybrid backbone–side-chain topology offers the rigidity and stability from the backbone closure with additional local structural reinforcement from the side-chain crosslinks. Hybrid macrocycles are the most protease-resistant and can have the greatest target selectivity and affinities. The challenges that come with hybrids are the higher synthetic complexity and the overall lower yields due to multiple sequential reactions. The competing conformational constraints can also lead to strain on the macrocycle and a balance of these effects must be achieved. Functionally, head-to-tail macrocycles have the best potential for long circulation times and a broad proteolytic resistance. Side-chain cyclization is the best suited for situations when some flexibility and reactivity is desired and hybrid topologies are usually used for the most difficult targets where high affinity and system stability are required.
| Cyclization Type | Stability | Structural Flexibility | Application |
| Head-to-tail | High | Medium | Drug discovery |
| Side-chain | Moderate | High | Constrained motifs |
| Backbone | Very high | Low | Protein mimetics |
In the design of peptides for backbone cyclization and mixed topologies, the choice of amino acid residues and their positions in the sequence is important for favorable ring closure. Computational tools are often employed to predict the folding and stability of designed peptides. Common challenges in the synthesis of backbone-cyclized peptides include side product formation, incomplete cyclization, and structural instability. Optimization of reaction conditions, purification methods, and the use of appropriate coupling reagents and catalysts are essential for efficient cyclization. Solid-phase peptide synthesis (SPPS) and enzymatic approaches have been developed for the synthesis of backbone-cyclized peptides, offering improved yields and accessibility.
This class of small molecules includes a variety of bioactive agents, such as PPI, enzyme, and structural inhibitors. PPI inhibitors are designed to disrupt or modulate the interactions between proteins, which can be difficult to achieve due to the large and flat interfaces involved. Cyclic peptides have been used as PPI inhibitors by mimicking the interaction surfaces of the target proteins, resulting in high binding affinity and selectivity. For example, stapled peptides, which use hydrocarbon cross-links to stabilize α-helices, have been developed as PPI inhibitors against the p53-MDM2 interaction, which is implicated in cancer. These inhibitors can penetrate cells and have improved proteolytic stability. Enzyme inhibitors derived from cyclic peptides have also been developed. These inhibitors can mimic the transition state or substrate of the enzyme, leading to high specificity. For example, cyclic peptides have been developed as inhibitors of proteases, which are enzymes involved in many pathological processes. Structural mimetics aim to replicate specific structural motifs of proteins, such as turns or helices, to confer desired biological activities. These mimetics can be designed to stabilize specific conformations, which can increase their stability and bioavailability. In summary, cyclic peptides and macrocycles are versatile platforms for the design of inhibitors with enhanced stability and functionality.
Fig. 2 Current representative mainstream strategies for peptidomimetic PPI inhibitors design and optimization.2,5
Cyclization topology is a key parameter that can be used to influence peptide performance in many areas. A number of different topologies have been applied and there are some key features to consider with each. For example, head-to-tail cyclization can provide high proteolytic stability and preorganization while creating a fully macrocyclic peptide. This type of topology can make peptides good candidates for systemic administration or oral delivery. On the other hand, it can introduce some challenges in terms of conformational restriction potentially reducing binding adaptability. Side-chain cyclization allows for more targeted control of the peptide conformation, for example stabilizing α-helices or β-hairpins while maintaining a flexible backbone. It may also have some weaknesses depending on the linkers chosen, such as the redox sensitivity of disulfides or the hydrolytic instability of lactam. Furthermore, they can require more complex orthogonal protection methods to achieve selective modification. Hybrid topologies that combine both backbone and side-chain constraints can provide a synergistic effect, allowing for both global stability and local structural definition. This can lead to the formation of bicyclic frameworks that provide antibody-like specificity while maintaining drug-like properties. One of the challenges with these types of peptides is that they often require more complex synthesis and may not be as easily scalable. A key functional consequence of the topology is the effect on mode of engagement with the protein surface. This can impact the mechanisms of membrane penetration, metabolic stability, immunogenicity and more. For example, a given topology may be effective as a PPI inhibitor, but less suitable as an enzyme inhibitor if conformational adaptability is compromised. As a result, no one topology can be considered universally superior and the choice of topology should be carefully considered and matched to the target based on empirical or computational data. It will become more and more common to select the appropriate topology for the application through careful computational modeling and empirical hit-to-lead screening, and new synthetic strategies are likely to emerge with advances in computational tools to accurately predict optimal crosslinking patterns and with the re-engineering of the enzymatic machinery for peptide cyclization from synthetic biology approaches.
Choosing between head-to-tail and side-chain cyclization requires deep knowledge of structural constraints, functional goals, and sequence compatibility. Our engineering team helps you select and execute the best strategy to achieve superior stability, conformation, and bioactivity.
We provide:
From early design exploration to cGMP-ready development, our cyclization expertise ensures reliable progress and high-quality results.
References
