When Cameron Pye joined biochemist Scott Lokey’s lab at the University of California, Santa Cruz as a graduate student in 2011, he learned about “undruggable” molecules. These cellular compounds couldn’t be targeted with conventional drug discovery approaches that identify a molecule’s inhibtor.1 Around this time, he also learnt about large, ring-shaped molecules that could provide a solution to these undruggable targets: macrocyclic peptides. These compounds target surfaces of protein-protein interactions, offering a different approach to drug mechanisms.
Lokey and his team had been leading research into macrocyclic peptides with several industry collaborators, offering Pye a front-row seat to understand their potential and “crucially what the bottlenecks for adoption were,” said Pye. “[We understood] what were the most critical problems that needed to be solved at a basic science level that would enable this modality.”
Cameron Pye is the cofounder of Unnatural Products, Inc., where his work focuses on synthetic macrocyclic peptides.
Unnatural Products, Inc.
With this vision, Pye joined forces with his fellow lab member Joshua Schwochert to cofound Unnatural Products, Inc., a startup to help overcome the hurdles that impede the adoption and scaling of macrocyclic peptide therapeutics. Today, Pye and his team use a proprietary platform combining machine learning and massively parallel chemistry to design, synthesize, and screen macrocyclic peptides with therapeutic properties for several conditions.
Macrocyclic Peptides Rewrite the Rules for Oral Drugs
Macrocyclic peptides are large molecules with molecular weights between 500 to 2,000 Daltons, which defy a rule of thumb in drug discovery. More than 25 years ago, based on evidence that most orally active drugs are small molecules, researchers formulated guidelines recommending that orally active drugs have a molecular weight under 500 Daltons.2
Despite this, many naturally occurring, promising macrocyclic drugs do not abide by this rule. “Cyclosporin A is one of the poster children in the space,” said Pye. This orally administered immunosuppressant isolated from a fungus weighs 1,202 Daltons.3
“Nature has…evolved these larger, more complex molecules, and in this case these peptidic macrocyclic ones that do really interesting biology. This [serves] as [a] signpost that we can break the traditional rules of small molecule drug discovery,” said Pye.
Like cyclosporin A, researchers have turned to other cyclic peptides, including synthetic ones, to explore treatments for a range of conditions including cancers and autoimmune diseases. Macrocyclic peptides combine high target affinity offered by monoclonal antibodies with improved metabolic stability, making them easy to administer like small molecules through pills.4
“The zeitgeist for the field was sparked by nature's evolved molecules,” said Pye. “The question was, ‘How do we do what nature did?’”
Automation and Industry Collaboration Could Help Bring Macrocyclic Peptides to Therapy
Macrocyclic peptides are biochemically complex molecules, making it challenging to design them. “[We need] a unique set of tools to really effectively navigate it,” explained Pye. To this end, Pye and his colleagues train computer models to accurately predict structure-function relationships and design potent, cell-permeable, and orally available drugs that will act on the intended target.
Synthesizing these large molecules comes with its own challenges. “We brought in automation engineers to help us develop custom robotics that let us make essentially any molecule [that] chemists can think of and quickly turn that into chemical matter that we can test,” Pye explained.
Pye and his colleagues then use a number of approaches such as affinity selection mass spectrometry to screen these to identify promising leads. “It's a ‘many threads make a yarn’ kind of platform approach and that's because no one tool or technology can solve the entirety of all these problems,” said Pye.
In this manner, Pye and his colleagues have established a number of industry collaborations to develop and test macrocyclic peptide therapeutics for rare diseases, cancers, cardiometabolic conditions, and autoimmune disorders.
“There [are] some really exciting proof points out there now that we're able to use this broader modality class to go after some of these really challenging targets,” said Pye. He added that antibodies can act on certain targets, showing that they are druggable, but antibody therapeutic adoption remains low in part because they require injections.
“In some ways, [there] is this greater unmet need…We know it translates. We know [it] is valuable for patients and yet we have kind of the wrong tool for the job,” said Pye. “We're really seeing this opportunity to take biology that we now know works and provide the right tool for the job. And I think macrocycles are that right tool for the job for that.”
- Russo A, et al. Targeting "undruggable" proteins: Design of synthetic cyclopeptides. Curr Med Chem. 2016;23(8):748-762.
- Lipinski CA, et al. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev. 2001;46(1-3):3-26.
- Tedesco D, Haragsim L. Cyclosporine: A review. J Transplant. 2012;2012:230386.
- Garcia Jimenez D, et al. Macrocycles in drug discovery ─ Learning from the past for the future. J Med Chem. 2023;66(8):5377-5396.
