Doctors have long sought the ability to deliver drugs and molecules exactly where, and when, the body needs them. However, solutions that can accomplish all of that have been hard to come by. Cracking this problem could bring new precision to therapeutics across a wide range of injuries and reduce unwanted off-target effects. A series of publications from the Marian Hettiaratchi Lab at the Phil and Penny Knight Campus for Accelerating Scientific Impact work toward engineering highly targeted molecules and precisely controlling their release, with hopes of one day helping muscle, bone, and spinal cord injuries heal better and faster.
When a tissue is injured, cells release regenerative cues called growth factors that kickstart regeneration. The body produces a whole cascade of different growth factors, which work together in a specific sequence to rebuild tissue.
Current treatments that use growth factors are limited because they typically release all molecules at once. This blunt approach doesn't match the complex timing of regeneration and may be preventing optimal healing.
Marian Hettiaratchi, Lary Simpson Professor and associate professor in the Department of Bioengineering, recognized that finding a way to control how quickly growth factors become active, rather than all at once, could give clinicians much finer control over healing, with the potential to improve outcomes across injuries.
The Hettiaratchi Lab's approach to precision molecule release centers on engineered proteins called affibodies. Affibodies function like the antibodies naturally found in our immune systems, but they're nearly ten times smaller. Unlike natural antibodies, which evolved to target foreign invaders like viruses, affibodies can be designed entirely in the lab to bind almost any target, including growth factors.
Years ago, Hettiaratchi envisioned a new way to use these engineered affibodies. She imagined engineering affibodies specific to particular growth factors, then adding further modifications so each affibody would release its target at a different rate. The affibody binds and holds the growth factor, but by tuning how tightly an affibody grips its target, the lab could control how quickly or slowly that growth factor is released.
This was an exciting idea, but to make it work, she needed someone to engineer and test a lot of different affibodies to find ones with varying release rates.
A visual representation of the size of antibody (left), and smaller affibody (right). Graphics were made using Alphafold.
Enter graduate student Justin Svendsen, a biochemist by training, but a protein engineer at heart - one who loves to code and build computational predictions. Combining these interests, Svendsen used computational models to design affibodies specific to different growth factors, starting with those that control blood vessel growth, since this process is critical to healing in nearly every kind of injury.
In the first study, published in the journal Biomacromolecules, Svendsen describes designing these affibodies on the computer, then testing them in the lab to confirm they bound only to their intended growth factor. With specific affibodies in hand, Svendsen began to tinker, using computational modeling to predict how tiny genetic mutations would alter an affibody's release rate before testing the predictions at the bench. This approach allowed the team "to test hundreds of variations on a computer before we ever touched a test tube," Svendsen said.
The result was a library of affibodies spanning a range of release rates - a full suite targeting blood vessel growth factors, each carrying subtle mutations that dialed binding affinity up or down. A single mutation could shift an affibody's release timeline dramatically, from minutes to days, and Svendsen found the team could control release for up to seven days.
"We're essentially creating biological timers," Svendsen says. "By introducing a single mutation, we can program an affibody to release its target growth factor in minutes, hours, or even days."
This first study served as a proof of principle. Computational design could reliably produce affibodies with different release rates; however, Svendsen had only tested one growth factor at a time and wanted to see how the approach would scale to multiple factors simultaneously. He thought that if each growth factor had a corresponding affibody engineered for a distinct release rate, one growth factor could reach its fully active state sooner and another later, giving the team the staggered control they were after.
Here, an affibody is shown in green, highlighting the protein's structure. A genetic sequence is coded into amino acids, which group together to form a 3-dimensional structure like this one.
Using computational models, researchers in the Hettiaratchi lab can create genetic mutations and predict how they might affect the release rate of the affibody. In the lab, they then make these genetic changes, altering the amino acids of the protein, shown here in yellow. In this example, the mutation may increase the release rate.
Because this work starts with computational models, researchers can explore many genetic changes at once. Shown here in yellow, a different mutation may instead decrease the release rate.
Through this approach, researchers can build libraries of affibodies specific to different growth factors (shown here as different colors), each with different mutations that impact release rate (shown in yellow). This gives researchers the ability to release some growth factors faster than others, providing greater precision for supporting regeneration.
In a second publication, in the Journal of Controlled Release, Svendsen and co-author Chandler Asnes moved from releasing a single growth factor to coordinating several in sequence. Using different engineered affibodies, they programmed the release rate of three growth factors involved in blood vessel regeneration: VEGF (vascular endothelial growth factor), FGF-2 (fibroblast growth factor), and PDGF (platelet-derived growth factor).
Through simply swapping in different affibodies, each tuned to release its target at a different rate, the team could deliver the three growth factors in any order they chose: VEGF first, then FGF-2, then PDGF, or any other sequence.
Because Svendsen and Asnes had control over each growth factor's release rate, they tinkered and tried different combinations. Through that process, they learned something new about growth factors and regeneration. When the team released all three growth factors at the same rate, blood vessel healing was actually worse than with the carefully timed, staggered release. One likely explanation is that these growth factors don't always work in the same direction, depending on timing, they can either promote new blood vessel growth or cause existing vessels to shrink. The sequence in which they arrive may matter as much as the growth factors themselves.
"Previous approaches just delivered all the proteins into the site at once. Through controlled, staggered release, we've actually learned more about the order in which these growth factors act, and how that timing contributes to regeneration, which we couldn't have teased apart with previous approaches" Hettiaratchi explains.
The findings offer a glimpse of what precision medicine could look like in the future, in which multiple growth factors are delivered exactly where, and when, the body needs them after an injury.
The Hettiaratchi Lab is also looking beyond blood vessel repair, exploring whether the same programmable affibodies could help accelerate bone healing, muscle repair, and even spinal cord regeneration. Because the design process is computational, the team can develop and test new candidate molecules far faster than traditional lab-only methods would allow.
Affibody-based therapeutics are still an emerging field in human health, but Hettiaratchi sees the potential: "It could revolutionize how we approach injuries that require multiple therapeutic interventions delivered with the perfect timing."
"We're excited about applying this across tissue and cell types," says Svendsen. "We think the applications could stretch from sports injuries to far more serious medical conditions."
This work was supported by the National Institutes of Health (NIH), the Wu Tsai Human Performance Alliance, the Oregon Health and Science University Medical Research Foundation New Investigator Grant, the University of Oregon Summer Program for Undergraduate Research (SPUR), the Knight Campus Undergraduate Scholars (KCUS) Program, and University of Oregon Vice President for Research and Innovation (VPRI) Undergraduate Fellowship.
