In the evolving landscape of drug discovery, cyclic peptides represent an exciting opportunity. These compounds offer a unique balance of size and specificity that positions them to bridge the gap between small molecule drugs and larger biologics like antibodies.
However, most cyclic peptides demonstrate low oral bioavailability: they are digested in the stomach before they can enter the bloodstream, or they’re not absorbed into the bloodstream by the gastrointestinal tract and can have little therapeutic effect (1). Biologics face a similar challenge and are administered intravenously rather than with a more convenient pill form.
To address the challenge of low oral bioavailability of cyclic peptides, a team from the Ecole Polytechnique Fédérale de Lausanne in Switzerland developed a “one-pot” method to synthesize a diverse library of cyclic peptides, which they then screened for stability, activity and permeability (1). Their method, which was published December 2023 in Nature Chemical Biology, streamlined the process of identifying and optimizing cyclic peptides and marked a substantial improvement from their earlier studies, where the developed cyclic peptides exhibited almost no oral bioavailability (%F). Using this new method, the team successfully developed a cyclic peptide with 18%F oral bioavailability in rats.
This blog covers the details of this study as well as a brief background on cyclic peptides.
Why Cyclic Peptides?
Cyclic peptides are bigger than most small organic molecule drugs and have the opportunity, therefore, to engage in more specific and selective interactions with protein targets. Studies also show that cyclic peptides can bind targets with greater affinity than their linear analogs, are less prone to degradation by proteases and show greater membrane permeability (2). While these features are certainly useful for traditional protein targets with well-defined binding pockets, in theory, cyclic peptides could do a better job of binding relatively featureless protein targets or disrupting protein: protein interactions (3).
Compared to antibodies, cyclic peptides can be less expensive to produce and may have less negative immunogenic responses (2,3).
Despite their promise for drug discovery, cyclic peptides face a number of challenges (3). One is that few cyclic peptides have good oral bioavailability. Peptides are large compared to small molecule organic compounds, often have multiple hydrogen bond donors and are relatively polar (1). These three factors can contribute to poor oral bioavailability, or the proportion of a drug that enters the blood stream relative to the amount that was ingested.
Finding strategies to improve the oral bioavailability of cyclic peptides could be one more way to ease development and further adoption of this class of drug compounds.
Synthesizing a Cyclic Peptide Library
To begin their study, the researchers first considered how they would design and prepare cyclic peptides for downstream screening (1).
Though several cyclic peptide library preparation and screening approaches have been reported, the researchers argue that some established methods, such as DNA-encoded libraries, are cumbersome to prepare. Previously, they developed a library prep where the formation of a disulfide bond yielded the cyclic peptides. They were able to quickly prepare a 20,000-member library using their method but encountered challenges: the disulfide bonds were not stable, and the cyclic peptides had no oral bioavailability.
In this follow up study, the researchers instead relied on forming cyclic peptides via a more stable thioether linkage. Specifically, they reacted short peptides containing a thiol group at each end with a biselectrophile linker. After showing that the thioether cyclized peptides were metabolically stable, they developed a method to cyclization and sequentially acylate the cyclic peptides in 96-well plates. (Acylation can make cyclic peptides less polar).
Using this method, they prepared the cyclic peptides in a single well with high efficiency for each chemical step. A straightforward precipitation provided purified peptides for downstream screening. The researchers used this strategy to prepare an 8,448-member library of cyclic peptides where each consisted of two amino acids, a linker, and a group introduced via the acylation reaction.
Screening and Optimizing
Having prepared the library, the researchers then used it to screen for activity against thrombin. In a fluorogenic thrombin activity screen, 73 of the 8,448 cyclic peptides inhibited thrombin by more than 50%. A follow up screen with the 20 most potent cyclic peptides prepared in the same way yielded similar inhibition activities, pointing to the consistency of the library prep method. They also screened for proteolytic stability, permeability and metabolic stability of the 20 cyclic peptides.
Overall, no one compound from the original library met their activity, stability and permeability requirements. So, the researchers built a new 168-member library based on the cyclic peptide with the highest activity from the first library. Five peptides demonstrated sufficient activity and permeability, but their metabolic stability was too low.
They then modified the peptide prep for the most active peptides identified from the second library. Instead of cyclizing the peptide with a thioether bond, they used ring closing metathesis reaction to yield cyclic peptides with a hydrocarbon linker. The other peptide components remained the same. In one peptide, the hydrocarbon linker led to three-fold higher metabolic stability compared to the same peptide with a thioether linker. However, oral bioavailability was rather low, however (only 4.5%F).
To potentially improve the oral bioavailability of their top hydrocarbon linked cyclic peptide, they carried out another library preparation where they only varied the peptide’s acyl group. One peptide from this library demonstrated high potency, permeability and stability along with twofold higher oral bioavailability in rats than the parent compound.
Key Takeaways
By relying on the formation of a thioether bond, the researchers were able to develop a library that can be prepared quickly and adapted to form a range of cyclic peptide structures. The thioether cyclized peptides can also be optimized relatively easily for improved oral bioavailability. As interest in cyclic peptides continues to grow, this and other convenient library prep and screening methods will expedite both discovery and development.
References
- Merz, M.L. et al. (2023) “De novo development of small cyclic peptides that are orally bioavailable.” Nat. Chem. Bio. doi: 10.1038/s41589-023-01496-y
- Ji, X., Nielsen, A.L., Heinis, C. (2023) “Cyclic Peptides for Drug Development.” Angew. Chem. Int. Ed. 63, e2023308251. doi: 10.1002/anie.202308251
- Li, X., Craven, T.W., Levine, P.M. (2022) “Cyclic Peptide Screening Methods for Preclinical Drug Discovery.” J. Med. Chem. 65, 11913. doi: 10.1021/acs.jmedchem.2c01077
Related Posts
Latest posts by Jordan Nutting (see all)
- Silencing the Immunogenicity of AAV Vectors - April 4, 2024
- Discovering Cyclic Peptides with a “One-Pot” Synthesis and Screening Method - February 29, 2024
- How do Self-Amplifying RNA Vaccines Work? - February 6, 2024
