To create novel protein-protein interactions, Magnet Bio uses a presenter protein (light pink) that binds a TrueGlue small molecule (dark blue) that together bind to the target disease protein (light blue).
Credit: Bruce Hua
Advances in screening and protein targeting are reviving interest in molecular glues as a strategy to tackle previously undruggable diseases.
When a protein binds to another protein, a whole host of biological processes become possible. Targeting these protein-protein interactions — once considered undruggable — has become a cornerstone of drug discovery and development these days. Currently, molecular glues that link two proteins together, with the purpose of making a disease-causing protein easier to degrade, are one of the most sought-after techniques to target protein-protein interactions. By 2030, the molecular glue market is projected to reach $5.6 billion.
As CEO of Magnet Biomedicine, Brian Safina leads the company in their pursuit to discover new molecular glues.
Credit: Magnet Biomedicine
One of the leading biotech companies in this space, Magnet Biomedicine, announced a collaboration and license agreement with Eli Lilly in February worth $40 million upfront with a potential value over $1.25 billion. Seeking to discover, develop, and commercialize molecular glues in oncology, the partnership is founded on Magnet’s TrueGlue™ discovery platform. The system uses cutting-edge screening technologies and chemical libraries to select proteins and identify small molecule TrueGlues to bind them together.
“The concept of identifying small molecules that can bring together any two proteins is a fascinating one. I think it’s truly where a startup can make an impact in a very new type of science,” said Brian Safina, CEO and President of Magnet.
DDN talked with Safina about the benefits to molecular glues, how Magnet’s platform is unique, and their first TrueGlues under development.
If molecular glues were discovered over 30 years ago, why has it taken so long to bring them to this stage?
Great question. My first answer is that I think no one has had the guts to try, but it’s also a very difficult challenge. The screening technologies have come a long way, and we now have DNA-encoded libraries where you’re able to sample millions of compounds, as opposed to good old-fashioned high throughput screening, where you might have 50,000 compounds, or up to a million. But that requires a lot of infrastructure. We’ve also had improvements in our mechanistic understanding of the biology in selecting the two proteins, the target and the presenter protein, where the presenter protein is the one that essentially does the gluing once the TrueGlue is attached. We’ve also had advances in proteomics, improving our understanding of the function of the different proteins that are intracellular and extracellular. Plus, I think small molecule degraders put glues on the map and said, “Hey, this is actually possible.”
All of these things came together to show that the time is now to take a step back and think wholeheartedly about how we can discover small molecules that can induce proximity in a cooperative way.
What are the main advantages of molecular glues compared to other drug modalities?
The biggest advantage is going after the hard-to-drug targets. With small molecules being invariably small, they don’t have a large interface to bind to proteins that don’t have defined pockets, like kinases and a variety of different enzymes. So, when you then bind to another protein, and it’s the entire complex that is now binding to this hard-to-drug protein, you have more surface area. That’s represented by our pipeline in the immunology space where we are going after cytokines that are currently dominated by biologics, because biologics can be engineered to be very specific to a small portion of another target.
The second piece is that in the world of oncology, if you bind to a target and it is dependent upon binding to a protein that is uniquely restricted to that cell type, then you can get selectivity within that cell type, which allows you to avoid the toxicities.
How is the TrueGlue™ discovery platform different from platforms at other companies?
Molecular glues are a new branch on the tree. I think they are going to gain a lot more speed.
– Brian Safina, Magnet Biomedicine
Most platforms revolve around degradation. Our screening is derived from selecting two presenters and then utilizing libraries to screen for glues. I think that’s unique. There are other types of screening methodologies that look for proteins that are already expressed in cells and try to find ones that are matchmakers. Our approach is to select the target because it’s a good target and it’s involved in the disease, and then depending upon the properties we want to impart in our drug, we then select the second presenter whether it’s restricted to that cell type or it’s expressed extracellularly. That’s the real difference. It’s a very rational approach to selecting the two partners, as opposed to going and finding two partners where you could also find a glue.
What is the importance of discovering novel protein-protein interactions at Magnet?
You don’t want to be restricted to just known protein-protein interactions. You want to be more creative and have more wiggle room. Let’s put it this way, if we don’t restrict it to proteins that are known to naturally bind to each other, it allows us to have much more of an empirical screening methodology that we embark upon. This is represented in our pipeline right now where all of our current protein pairs, which are made up of the target protein and a second protein that remains confidential, have no business interacting as far as we know. But we’re finding small molecules that can co-opt this protein-protein interaction.
What do your first TrueGlues target?
Our work on the suppression of tumorigenicity 2 (ST2) protein is our lead program at the moment. This is a receptor for interleukin-33 (IL-33) and it’s a very interesting target. Both IL-33 and ST2 have very interesting biology in the world of chronic obstructive pulmonary disease (COPD) and asthma, where there is significant unmet need still. There are antibodies that are advancing in this area, but we have now identified a very specific protein, which remains confidential, that is expressed endogenously in the body, where we can now take advantage of that protein to form a glue with the ST2 receptor. We’ve also done this for IL-17, tumor necrosis factor alpha (TNF-α), IL-6 receptor and interferon beta (IFN-β). It’s really a pretty interesting proof of concept. And now we’re making these more potent and advancing them into more advanced studies in vivo over the next three to six months.
Have there any been results so far that you found surprising?
Science is full of unexpected results, that’s for sure. We’ve had unexpected results throughout the whole process in trying to figure out which are the right pairs and what’s the right screening approach. In the early days, we certainly boiled the oceans in all the different ways we could approach the problem. And with good experimentation emerges the answer, and then once you identify what’s working, you can then work to improve upon that. So, we’re always improving the technology, both in the screening and we’re improving the way we construct our chemical libraries. We’re also expanding the types of presenter proteins. So, there have been lots of surprises that did not align with what we would have predicted. Of course, this whole process has resulted in the know-how that makes the company special.
These are really novel types of drugs that we’re working on, and to me, molecular glues are a new branch on the tree. I think they are going to gain a lot more speed. Being a part of something new and innovative like this is really why we all go into science.
This interview has been condensed and edited for clarity.