Every two weeks, Douglas Johns, a clinical director at Merck, gives his mother-in-law a subcutaneous shot of a monoclonal antibody to treat her high cholesterol. As someone who doesn’t feel comfortable administering the shot herself, she once asked Johns why researchers couldn’t develop a pill to achieve the same effect. At the time, Johns couldn’t disclose to his mother-in-law that he was actually part of a team at Merck developing a pill called MK-0616 to provide the same level of treatment against high cholesterol.
Today, MK-0616 is in a Phase 3 clinical trial, after Phase 2b data released last year demonstrated that the drug is safe and effective at reducing high cholesterol (1). MK-0616 works the same way as monoclonal antibodies: It blocks the proprotein convertase subtilisin/kexin type 9 (PCSK9) protein from interacting with low-density lipoprotein (LDL) receptors. In the liver, PCSK9 degrades receptors that reduce the amount of LDL (bad cholesterol). With PCSK9 activity blocked, LDL receptors can clear out LDL in the blood. “I believe that an oral treatment should help doctors introduce this really, really powerful mechanism earlier in a person’s, what we call their patient journey. So [instead of] having to wait until you've gone a huge distance with a bunch of different drugs, this might be something that they could take earlier on and with easier access,” said Johns.
Christie Ballantyne, a cardiologist and researcher at Baylor College of Medicine who led the Phase 2b trial, said that he was pleasantly surprised by the efficacy of MK-0616 after discussions he had with experts a decade ago. “They said, ‘listen, because of the nature of this target and the interaction … we'll never be able to make an oral molecule for PCSK9,’” said Ballantyne.
The new oral drug that made it possible is no ordinary molecule. MK-0616 belongs to a class of drugs known as macrocyclic peptides. Based on their sizes, these peptides find themselves somewhere in between small molecule drugs (which make up more than 90 percent of marketed drugs) and large biologic drugs (like antibodies) made from living cells or organisms. Their size makes them ideally suited for targeting bigger protein-protein interactions that happen within cells, which small molecules often struggle against. While small molecules are often available in a format suitable for oral consumption and are adept at getting into cells, they are too small to selectively bind to many shapes of proteins, resulting in off-target effects. On the other hand, large biologics can selectively bind to the protein of interest, but they’re too large to get into the cell, and they must be injected rather than administered orally. Macrocyclic peptides are neither too big nor too small, earning them the Goldilocks designation.
“There is evidence from peptides like ours as well as others that if you can operate in that [middle] space, you can do the things that biologics do with the delivery approaches that are maybe exemplified by small molecules, and you completely have the best of both worlds,” said Johns.
Put a ring on it
Regular peptides don’t operate in the perfect Goldilocks middle zone naturally. Macrocyclic peptides have specific features that allow them to claim that title. The cyclization process turns them into a ring shape, while the macro part refers to their size. They usually contain between four and 25 amino acids in their rings. This contrasts with regular linear peptides that exist as a string of amino acids all in a line. Linear peptides are susceptible to quick degradation since enzymes recognize their ends and chew them up. Macrocyclic peptides are more resistant to degradation with their ends tied up into a ring.
Instead of being a floppy piece of string, you make it into a donut.
– Jan Kihlberg, Uppsala University
Researchers have found that macrocyclizing peptides also forces them into a more rigid conformation that helps them bind much more strongly to targets than the usual flexibile peptides (2). “Instead of being a floppy piece of string, you make it into a donut,” said Jan Kihlberg, a medicinal chemist at Uppsala University. Plus, he added that macrocyclic peptides can cover more area as a ring or a disc than as a straight line, allowing them to bind and block bigger, flatter targets. “If you want to block a tunnel, you have to be the size of a tunnel,” he said.
A rigid, donut-like macrocyclic peptide proved to be an effective molecule for Johns’ team to use because they needed it to block a big, flat interface on PCSK9. “[Targeting PCSK9 is] great for antibodies because they're huge,” said Johns. “They just sit there, and they prevent stuff from coming together. And we're able to accomplish the same thing … and be big enough and rigid enough to keep that protein-protein interaction from happening.”
To find a macrocyclic peptide that could target PCSK9, Johns and his team collaborated with small startup companies that had platforms available to screen vast numbers of macrocyclic peptides. They ran a few screens for PCSK9 just as test cases, and the results led them on a journey to develop the MK-0616 compound as the first oral drug to target PCSK9 for high cholesterol. “It was really an interesting tour de force of clinical chemistry in terms of evolving the molecule to become a better agent for human beings, meeting more of the clinical criteria,” said Ballantyne.
In their Phase 1 clinical trial of MK-0616 in humans, Johns said that he expected huge variability and possibly no detection of the drug at all in some people (3). Instead, most patients showed measurable and consistent levels of the drug. “That's when I got goosebumps,” said Johns. “We reduced free PCSK9 to levels that were basically to zero, which is what the antibodies do. And then we knew we had something.”
Based on the clinical trial data so far, other macrocyclic peptide researchers are excited about MK-0616’s potential and what it means for future macrocyclic peptide drugs. “What it does show is the incredible potency that you can get with these larger compounds against undruggable targets that have previously been impossible to inhibit with small molecules,” said Scott Lokey, a chemist at the University of California, Santa Cruz who was not involved with developing MK-0616.
As a result of MK-0616’s early success, Johns said that Merck is enthusiastic about continuing to study macrocyclic peptides for other drug targets, too, although he did not disclose which diseases they plan to treat. “We are committed to this platform,” Johns said. “It's a super exciting opportunity.”
Chameleon-like molecules
Merck’s high cholesterol drug sailed through a few challenges that other macrocyclic peptides won’t overcome so easily. For one, making macrocyclic peptides as orally bioavailable drugs makes it more difficult for them to be absorbed. Johns noted that the oral bioavailability of MK-0616 lies between just one and two percent, but because it’s so potent at inhibiting PCSK9, they were able to stick with a normal dose. If another target circulates in the blood at a much higher level, or if the macrocyclic peptide isn’t as potent in binding to it, low oral bioavailability of a drug will be a problem.
We are committed to this platform ... It's a super exciting opportunity.
– Douglas Johns, Merck
Another common challenge that MK-0616 skirted is the limited ability of macrocyclic peptides to sneak into cells. MK-0616’s target, PCKS9, circulates in the blood, and thus MK-0616 didn’t need to get past the cell membrane. But many protein-protein interactions relevant to disease take place inside cells that macrocyclic peptides have a hard time reaching. “Cyclization isn't enough to make them permeable,” said Lokey. The cell permeability problem is due to the water molecules on their backbones that must be removed in order to go through a fatty lipid cell membrane. As the saying goes, oil and water don’t mix.
Earlier in his career, Lokey noticed that some compounds found in nature act on intracellular targets, meaning that they must be getting into cells somehow. To find out how, he began studying natural cyclic peptides like cyclosporine, an immunosuppressive drug produced by the Tolypocladium inflatum fungi (4). Cyclosporine can pass through cell membranes because it changes its conformation and internally hides its backbone; once it passes through the cell membrane, it flips back again. In 2006, Lokey and his colleagues used simple model systems to provide some of the first evidence that the cell permeability of cyclosporine results from its ability to hide its polar backbone (5).
This surprising behavior from natural products like cyclosporine is called chameleonicity. It’s a property that scientists hope to find in more macrocyclic peptides to use as therapeutics.
David Baker and Gaurav Bhardwaj, computational biologists at the Institute for Protein Design at the University of Washington, design new peptides from scratch using computational methods. “What our research has shown is how to design completely new macrocycles. We can design their structures very accurately. That means we come up with a 3D model of a peptide, and then we can design a peptide — that means specify its sequence — such that it actually folds up to that structure,” said Baker.
In a recent study, they designed novel macrocyclic peptides that achieved cell permeability using similar chamelonicity strategies as cyclosporine by changing their structures before and after entering the cell (6). The team also identified principles that allowed the macrocyclic peptides to be orally bioavailable at rates up to 40 percent in mice. “But we're not as good yet at getting this kind of high-affinity binding [to the target]. So, the challenge is getting all these properties at the same time,” said Baker.
Making new rules
More than 25 years ago, Christopher Lipinski and his colleagues at Pfizer came up with the “rule of 5” to guide chemists in choosing drug molecules that will be successful when taken orally (7). The rules, which include guidelines like selecting ones with a molecular weight under 500 Da, have profoundly influenced the drug discovery field ever since.
Macrocyclic peptides and other larger compounds don’t match up to the rules. “We call it the ‘beyond rule of 5,’ affectionately called the ‘bRo5,’” said Johns.
So far, it’s been a challenge to determine a set of common rules that allow macrocyclic peptides to be orally bioavailable and pass through the cell membrane to activate intracellular targets. Researchers at Chugai Pharmaceutical spent the last several years developing a new technological platform to elucidate these new rules and develop library technologies to identify macrocyclic peptides with the right properties (8). Atsushi Ohta, a chemical biologist at Chugai Pharmaceutical who worked on the project, said that the results were beyond their expectations.
The team showed that their platform identifies macrocyclic peptides that are orally bioavailable and cell permeable. They discovered LUNA18, an oral macrocyclic peptide that blocks the intracellular protein-protein interaction between the Kirsten rat sarcoma viral oncogene homolog (KRAS) and son of sevenless 1 (SOS1) proteins involved in promoting cancer growth. KRAS mutations have been notoriously difficult drug targets in the past (9).
To make LUNA18 a better compound for this protein-protein interaction, they finetuned the side chains in its ring-like structure. There was no need for LUNA18 to undergo conformational chameleonic changes like cyclosporine to reach the intracellular target. Because LUNA18 successfully gets into cells, Kihlberg said that its journey from the intestine into the bloodstream and then into the target cell is a much lengthier and more complex journey than modulating an extracellular target with an injectable compound. “This is an impressive piece of work,” he said. LUNA18 is now under study in a Phase 1 clinical trial.
Other researchers focus on improving the nature of the macrocyclization ring-making process to add more functionalities to the drug. David Spring, a chemist at the University of Cambridge, is pioneering new stapling methods to attach additional handles to a macrocyclic peptide that could make it more cell permeable, for example. “It gives you advances in optimization and extra positions to add functionality without disturbing important interactions that the peptide makes,” he said.
Spring is also developing new technologies for identifying macrocyclic peptides to use as antibiotics. In January of this year, researchers at Roche showed that zosurabalpin, a macrocyclic peptide compound, can treat mice with carbapenem-resistant Acinetobacter baumannii (CRAB) infections (10). It would be the first new antibiotic with antibacterial activity against this pathogen in more than 50 years. Spring hopes to identify many more macrocyclic peptides like this one with antibacterial activity. “In the future, what we're trying to do is make huge libraries of billions of cyclic peptides, and screen them in cell assays directly. There's no technology that can do that,” Spring said.
Working it all out
Macrocyclic peptide drugs could also offer advantages over other drug modalities even in cases where successful treatments already exist. Last year, the FDA approved zilucoplan, a macrocyclic peptide from the biopharmaceutical company UCB that treats people with myasthenia gravis (MG), a rare autoimmune condition that leads to muscle weakness (11). Zilucoplan works by blocking the extracellular complement five protein, which is erroneously activated in MG, leading to the destruction of the neuromuscular junction tissue.
These recent successes from pharma have really galvanized people, and I think they’ve turned a lot of skeptics into believers.
– Scott Lokey, University of California, Santa Cruz
Two monoclonal antibody treatments were available before zilucoplan, but these have several drawbacks for patients. They require visits to an infusion center every few weeks, whereas patients can administer zilucoplan themselves once daily with a subcutaneous injection. James Howard, a neurologist at the University of North Carolina at Chapel Hill who was involved in the Phase 3 trial said that zilucoplan also allows patients to be treated with another class of drugs called neonatal fragment crystallizable receptor (FcRn) inhibitors that can’t be combined with monoclonal antibodies. “I only work in this one disease, but I envision it being applied to other disorders. So, I think the field is open,” said Howard.
Before macrocyclic peptides are widely applicable to a variety of diseases, however, researchers must continue to find unique ways for them to get through cell membranes. “We’re finding that membrane permeability can be achieved in these larger molecules in a lot of unexpected ways,” said Lokey. He and his colleagues recently showed that a macrocyclic peptide crossed cell barriers by shifting into a novel saddle-shaped fold (12). “The chemical space that allows for permeability is vast and pretty largely unexplored,” he said.
It’s not only about getting into a cell. After that, the macrocyclic peptide must also bind to the intracellular target with enough potency to treat the disease of interest. “You can frequently find something that binds to the target, but it's not permeable. Or it's permeable, but does not bind to the target. When you try to optimize for one, you lose the other property,” said Bhardwaj. “The next step is how do we combine [to] do them together?”
Ultimately, research on macrocyclic peptides has generated optimism for the future. “The field can explode and just start generating cool molecules and leads,” said Lokey. “These recent successes from pharma have really galvanized people, and I think they've turned a lot of skeptics into believers.”
This article was updated on May 8, 2024 to correct the injection type. The monoclonal antibody injection is subcutaneous, not intravenous.
