Cyclic Peptide Drug Design: Recent Advances and Future Directions

Cyclic peptides have become one of the most active areas in drug discovery. Their ring structure gives them advantages that linear peptides cannot match, including better stability, stronger target binding, and in some cases the ability to cross cell membranes.

This article covers the latest advances in cyclic peptide drug design and what they mean for the future of peptide therapeutics.

What Makes Cyclic Peptides Special

Linear peptides are chains of amino acids with free ends. They flex freely in solution and are quickly broken down by enzymes in the body. This limits their usefulness as drugs.

Cyclic peptides connect their ends to form a ring. This ring structure restricts the molecule's flexibility and shields the peptide bonds from enzymes. The result is a molecule that lasts longer in the body and binds more tightly to its target.

The cyclic structure also reduces the number of conformations the peptide can adopt. This pre-organization means the molecule does not have to pay an entropic penalty when it binds to its target, leading to stronger binding.

Types of Cyclic Peptides

Type	Cyclization Method	Examples
Head-to-tail	Peptide bond between N and C terminus	Cyclosporine A
Side chain to side chain	Disulfide, lactam, or thioether bridge	Octreotide
Head-to-side chain	Bond from N-terminus to side chain	Various research compounds
Side chain to tail	Bond from side chain to C-terminus	Various research compounds
Stapled	Hydrocarbon cross-link between side chains	ALRN-6924 (clinical)
Bicyclic	Two independent ring closures	Phage-derived bicyclics

Cyclosporine A, one of the most successful cyclic peptide drugs, generates over $1 billion in annual sales. It was first discovered in a soil fungus in 1971 and remains a cornerstone of organ transplant medicine.

Recent Advances in Synthesis

The last few years have brought major improvements in how cyclic peptides are made.

Improved Macrocyclization Chemistry

Traditional head-to-tail cyclization is difficult for large peptides because the two ends are far apart in solution. New catalytic methods have improved cyclization yields significantly.

Thiol-ene and thiol-yne click chemistry now allow rapid ring closure under mild conditions. These methods tolerate a wide range of functional groups and produce clean products.

Native chemical ligation has been adapted for cyclization, allowing the formation of native peptide bonds in the ring closure step. This avoids the need for non-natural linkers.

On-Resin Cyclization

Performing cyclization while the peptide is still attached to the solid-phase synthesis resin simplifies the process. The pseudo-dilution effect of the resin reduces dimerization and oligomerization.

New acid-labile linkers allow on-resin cyclization followed by mild cleavage. This approach works well for lactam bridges and disulfide bond formation.

Enzymatic Cyclization

Nature's own tools are now being used for peptide cyclization. Enzymes like sortase A, butelase 1, and PatG catalyze head-to-tail cyclization with high efficiency and selectivity.

Butelase 1 is particularly impressive. It can cyclize peptides with as few as five amino acids and works at near-quantitative conversion in minutes.

Method	Advantages	Limitations
Solution-phase chemical	Flexible, well-established	Low concentration needed, competing reactions
On-resin chemical	Pseudo-dilution effect, integrated workflow	Limited to SPPS-compatible chemistry
Enzymatic (sortase A)	Mild conditions, specific	Requires recognition sequence
Enzymatic (butelase 1)	Very fast, minimal recognition sequence	Limited commercial availability
Thiol-ene click	Fast, mild, bioorthogonal	Introduces non-natural linkage

Computational Design Tools

Computers are playing an increasingly important role in cyclic peptide drug design.

Structure Prediction

Predicting the three-dimensional structure of a cyclic peptide is crucial for understanding how it will interact with its target. Machine learning methods, building on the success of AlphaFold for proteins, are being adapted for peptide structure prediction.

Rosetta's generalized kinematic closure (GenKIC) protocol can sample cyclic peptide conformations and predict the most stable structures. This helps researchers design molecules with the right shape for their target.

Virtual Screening

Large virtual libraries of cyclic peptides can now be screened computationally against protein targets. Molecular docking methods score how well each peptide fits the binding site.

Recent advances in scoring functions have improved the accuracy of these predictions. Researchers can screen millions of virtual compounds before synthesizing the best candidates.

Machine Learning for Optimization

Machine learning models trained on experimental data can predict properties like binding affinity, cell permeability, and metabolic stability. These models guide the optimization of lead compounds and reduce the number of synthesis and testing cycles needed.

Oral Bioavailability: The Holy Grail

Most peptide drugs must be injected because they cannot survive the digestive system. Making cyclic peptides that work when taken by mouth is one of the biggest challenges in the field.

The Problem

Oral peptide drugs face three major barriers:

1. Enzymatic degradation: Stomach acid and digestive enzymes break down peptides
2. Poor membrane permeability: Peptides are too large and polar to cross the intestinal wall
3. First-pass metabolism: The liver clears peptides before they reach the bloodstream

Progress on Permeability

Cyclic peptides have an inherent advantage for oral delivery. Their constrained structure can form intramolecular hydrogen bonds that shield the polar backbone from the lipid membrane.

N-methylation of backbone amides further improves permeability by reducing the number of hydrogen bond donors. Cyclosporine A, which has seven N-methylated residues, is one of the few orally bioavailable cyclic peptides.

Recent research has identified a set of design rules for orally bioavailable cyclic peptides:

Parameter	Target Range
Molecular weight	Less than 1,000 Da
Number of hydrogen bond donors	Less than 5
Number of N-methylated amides	2 to 4
Polar surface area	Less than 200 square angstroms
LogP	2 to 6
Number of rotatable bonds	Minimized by cyclization

Formulation Approaches

Beyond molecular design, formulation strategies can enhance oral bioavailability. Permeation enhancers like sodium caprate open tight junctions between intestinal cells temporarily, allowing peptides to pass through.

Nanoparticle encapsulation protects peptides from enzymatic degradation and promotes absorption. Lipid-based formulations improve solubility and transport across membranes.

Dr. Patrick Lokey, Professor of Pharmaceutical Chemistry put it plainly: "We are getting closer to the oral cyclic peptide dream every year. The combination of better molecular design and smarter formulation is going to crack this problem within the next decade."

Clinical Applications

Several cyclic peptide drugs have reached the market, and many more are in clinical trials.

Approved Cyclic Peptide Drugs

Drug	Indication	Type	Year Approved
Cyclosporine A	Organ transplant rejection	Fungal natural product	1983
Daptomycin	Bacterial infections	Lipocyclic peptide	2003
Romidepsin	T-cell lymphoma	Depsipeptide	2009
Pasireotide	Cushing's disease	Somatostatin analog	2012
Bremelanotide	Hypoactive sexual desire	Melanocortin agonist	2019

Promising Clinical Candidates

Several cyclic peptides are currently in Phase II or Phase III clinical trials for conditions including cancer, metabolic disease, and autoimmune disorders.

Stapled peptides that target protein-protein interactions are particularly promising. These molecules can disrupt interactions that were previously difficult to address with small molecules.

Bicycle Therapeutics is developing bicyclic peptides that combine the target specificity of antibodies with the tissue penetration of small molecules. Their lead compound is in Phase II trials for solid tumors.

Library Screening Approaches

Finding the right cyclic peptide for a given target often starts with screening large libraries.

Phage Display

Phage display is a powerful method for discovering cyclic peptide binders. Billions of random peptide sequences displayed on bacteriophage surfaces are screened against a target protein.

After multiple rounds of selection, the surviving phage encode peptides with high affinity for the target. Cyclization is typically achieved through disulfide bonds between flanking cysteine residues.

mRNA Display

mRNA display allows screening of even larger libraries (over 10 trillion sequences). The peptide is physically linked to its encoding mRNA, enabling selection and amplification.

This platform has been adapted for cyclic peptide discovery using non-natural amino acids and chemical cyclization after translation.

DNA-Encoded Libraries

Chemical synthesis on DNA templates creates massive libraries where each compound is tagged with a unique DNA barcode. After selection, sequencing reveals the structures of the best binders.

This approach is particularly useful for exploring chemical space that goes beyond natural amino acid building blocks.

Future Directions

The field of cyclic peptide drug design is moving quickly. Several trends will shape the next decade.

AI-Driven Design

Artificial intelligence will increasingly guide the design of cyclic peptides. Generative models can propose novel sequences optimized for multiple properties simultaneously, including binding affinity, selectivity, stability, and permeability.

Bicyclic and Polycyclic Peptides

Adding more rings increases the structural complexity and rigidity of the molecule. Bicyclic peptides, constrained by two independent bridges, offer even greater stability and selectivity than monocyclic analogs.

Cell-Penetrating Cyclic Peptides

Engineering cyclic peptides that efficiently enter cells would open up intracellular targets. Research on arginine-rich cyclic sequences and charge-hiding strategies is making progress.

Hybrid Molecules

Cyclic peptides combined with small molecule pharmacophores, antibodies, or nanoparticles create hybrid drugs with unique properties. These molecules blur the boundaries between traditional drug classes.

FAQ

What is the advantage of cyclic peptides over linear peptides as drugs?

Cyclic peptides are more resistant to enzymatic degradation, have better binding affinity due to pre-organized structure, and in some cases show improved cell permeability. These properties make them more drug-like than linear peptides while retaining the ability to target large protein surfaces.

How are cyclic peptides different from small molecule drugs?

Small molecule drugs are typically less than 500 Da and bind in well-defined pockets on proteins. Cyclic peptides are larger (500 to 2,000 Da) and can cover larger surface areas, making them suitable for targeting protein-protein interactions and other flat binding surfaces that small molecules cannot effectively engage.

Can cyclic peptides be taken orally?

Most current cyclic peptide drugs require injection. However, significant progress is being made in designing orally bioavailable cyclic peptides. Strategies include N-methylation, lipophilicity optimization, and advanced formulation approaches. A few cyclic peptides like cyclosporine A are already given orally.

How long does it take to develop a cyclic peptide drug?

From initial discovery to market approval, a cyclic peptide drug typically takes 10 to 15 years, similar to other drug classes. However, advances in computational design and library screening are shortening the early discovery phase from years to months.

What are the biggest challenges in cyclic peptide drug development?

The top challenges are achieving oral bioavailability, scaling up manufacturing to commercial quantities, and developing analytical methods for quality control. Cost of goods for complex cyclic peptides can also be higher than for small molecule drugs.