Programmable proteins use logic to improve targeted drug delivery

Targeted drug delivery is a powerful and promising area of ​​medicine. Therapy that identifies the exact areas of the body where they are needed (and where they are not) can reduce drug dosage and avoid potentially harmful off-target effects in other parts of the body. For example, targeted immunotherapy can identify cancerous tissues and activate immune cells to fight disease only in those tissues.

The hardest thing to make a therapy truly “smart” is when the drug can move freely throughout the body and decide which areas to treat.

Researchers at the University of Washington have taken a significant step toward this goal by developing proteins with autonomous decision-making capabilities. In a study confirming the effectiveness of the principles, published in Nature Chemical Biologyresearchers demonstrated this by adding smart tail structures to therapeutic proteinsthey could control protein localization based on the presence of certain environmental cues.

These protein tails fold into preprogrammed shapes that determine how they respond to different combinations of signals. In addition, the experiment showed that smart protein tails can be attached to a carrier material for delivery into living cells.

Achievements in synthetic biology also allowed researchers to produce these proteins cheaply and in a matter of days rather than months.

“We've been thinking about these concepts for some time, but have struggled to find ways to scale up and automate production,” said senior author Cole DeForest, a UW professor of chemical engineering and bioengineering. “Now we've finally figured out how to build these systems faster, at larger scale, and with vastly increased logical complexity. We are excited about how this will lead to more sophisticated and scalable treatments for diseases.”

The concept of programmable biomaterials is not new. Scientists have developed a variety of strategies to make systems respond to specific signals, such as pH levels or the presence of certain enzymes, that are associated with a specific disease or area of ​​the body. But it is rare to find a single signal (biomarker) unique to one site, so material that focuses on just one biomarker may affect multiple unintended sites in addition to the target.

One solution to this problem is to find a combination of biomarkers. There may be many areas of the body with certain enzyme or pH levels, but there will likely be fewer areas with both of these factors. Theoretically, the more biomarkers a material can identify, the more accurate the targeting drug delivery May be.

In 2018, DeForest's lab created a new class of materials which responded to multiple biomarkers using Boolean logic, a concept traditionally used in computer programming.

“We realized that we could program the way therapeutics were released simply based on how they bound to the carrier material,” DeForest said. “For example, if we connect a therapeutic payload to a material through two degradable groups connected in series, that is, one after the other, it will be released if either group degrades, acting as an OR gate.

“When the decomposed groups were instead connected in parallel, that is, each in a different half of the cycle, both groups had to be decomposed to release the load, acting as an AND gate. Interestingly, by combining these basic gates, we could easily create advanced logic circuits.”

It was a big step forward, but it couldn't be scaled up—the team created these large, complex, logic-responsive materials by hand using traditional organic chemistry.

But over the next few years, the corresponding field of synthetic biology developed by leaps and bounds.

“The field has developed exciting new protein-based tools that may allow researchers to form persistent connections between proteins,” said co-author Murial Ross, a UW bioengineering graduate student. “This opened the door to new protein structures that were previously inaccessible, making more complex logic operations possible.”

It also became practical to use living cells as factories for the production of these complex proteins, allowing scientists to design customized DNA blueprints for new proteins, insert the DNA into bacteria or other host cells, and then assemble proteins with the desired structure directly from the cells.

With these new tools, DeForest and his team have simplified and improved many parts of the process at once. They designed and produced proteins with tails that spontaneously fold into more customized shapes, creating complex “circuits” that can respond to five different biomarkers. These new proteins can attach to a variety of carriers—hydrogels, tiny beads, or living cells—for delivery into a cell or, theoretically, a disease site. The team even loaded three different proteins into one carrier, each programmed to deliver its own unique cargo based on a different set of environmental cues.

“We were so excited about the results,” DeForest said. “Using the old process, synthesizing just a few milligrams of each of these materials would take months. Now it takes us a couple of weeks to go from construction design to product. It was a complete game changer for us.”

“The sky's the limit. You can provide delayed and independent delivery of many different components in one course of treatment,” Ross said. “And I think we could create much larger logic circuits that the protein could respond to. We're at the point now where the technology is ahead of what we were seriously considering from an application perspective, and it's a great place to be.”

The researchers will now continue to search for new biomarkers that the proteins can target. They also hope to begin collaborating with other labs at UW and beyond to create and implement real-world treatments.

The team also describes other uses for this technology. The same tools can produce therapy within a single cell and target it to specific areas, a kind of microcosm of how this process works in the body. DeForest also envisions diagnostic tools, such as blood tests, that can, say, change a certain color when a complex set of signals is present in a blood sample.

DeForest believes the first practical application will likely be to treat cancer, but with further research the possibilities seem limitless.

“The dream is to be able to pick any random location inside the body—down to individual cells—and program a material to act there,” he said. “It's not an easy task, but with these technologies we are getting closer. With the right combination of biomarkers, these materials will become more and more accurate.”