By engineering proteins called proteases to find new targets with high selectivity, researchers mark a critical advance toward potential new treatments for a wide range of conditions
Redirected enzymes may open doors to novel therapeutics
Botulinum toxin is a powerful microbial poison approved by the FDA to treat a suite of maladies like chronic migraines, uncontrolled blinking, and certain muscle spasms (and also to remove frown lines). A team of researchers from Harvard University and the Broad Institute of MIT and Harvard has now shown that they can engineer the toxin to create a suite of adaptable and super-selective proteases (enzymes that cut proteins to either activate or deactivate them) targeting a vast array of disease-associated proteins. The team's work, reported in Science, may create new opportunities to treat conditions ranging from in neurodegeneration to rampant inflammation.
“In theory, you have a really high ceiling for the number and type of conditions where you could intervene,” said Travis Blum, a postdoctoral researcher in the Harvard Department of Chemistry and Chemical Biology and first author on the Science paper. The study was the culmination of a collaboration with Min Dong, an associate professor at Harvard Medical School, and David Liu, a core institute member of and director of the Merkin Institute for Transformative Technologies in Healthcare at the Broad, the Thomas Dudley Cabot Professor of the Natural Sciences at Harvard University, and a Howard Hughes Medical Institute Investigator.
Together, the team achieved two firsts: They successfully reprogrammed proteases to cut entirely new protein targets and, at the same time, avoid going back to their original targets. They also solved what Blum called a “classical and canonical challenge in biology”: Unlike most large proteins, their botulinum toxin proteases can slip inside neural cells, too, giving them a wider reach that makes them all the more appealing as potential therapeutics.
Now, the team can design bespoke proteases with tailor-made instructions for which protein to cut and how. “Such a capability could make ‘editing the proteome’ feasible,” said Liu, “in ways that complement the recent development of technologies to edit the genome.”
Current gene-editing technologies often target chronic diseases like sickle cell anemia, caused by an underlying genetic error. Correct the error, and the symptoms fade. But some acute illnesses, like neurological damage following a stroke, aren’t caused by a genetic mistake. That’s where protease-based therapies come in: The proteins can help boost the body’s ability to heal something like nerve damage through a short or even one-time treatment.
Scientists have been eager to use proteases to treat disease for decades. Unlike antibodies, which can only attack specific alien substances in the body, proteases can find and attach to any number of proteins, and, once bound, can do more than just destroy their target. They could, for example, reactivate dormant proteins.
“Despite these important features, proteases have not been widely adopted as human therapeutics,” said Liu, “primarily because of the lack of a technology to generate proteases that cleave protein targets of our choosing.”
But Liu has a technological ace in his pocket: PACE (which stands for phage-assisted continuous evolution). A Liu invention, the platform rapidly evolves novel proteins with valuable features. PACE, Liu said, can evolve dozens of generations of proteins a day with minimal human intervention. Using PACE, the team first taught so-called “promiscuous” proteases — those that naturally target a wide swath of proteins — to stop cutting certain targets and become far more selective. When that worked, they moved on to the bigger challenge: Teaching a protease to attach to an entirely new target, one outside its natural wheelhouse.
“At the outset,” said Blum, “we didn't know if it was even feasible to take these proteases and evolve them or teach them to cleave something new because that had never been done before.” (“It was a moonshot to begin with,” said Michael Packer, a previous Liu lab member and an author on the paper). But the proteases outperformed the team’s expectations. With PACE, they evolved four proteases from three families of botulinum toxin; all four had no detected activity on their original targets and cut their new targets with a high level of specificity (ranging from 218- to more than 11,000,000-fold). The proteases also retained their valuable ability to enter cells. “You end up with a powerful tool to do intracellular therapy,” said Blum. “In theory.”
“In theory” because, while this work provides a strong foundation for the rapid generation of many new proteases with new capabilities, far more work needs to be done before such proteases can be used to treat humans. There are other limitations, too: The proteins are not ideal as treatments for chronic diseases because, over time, the body’s immune system will recognize them as alien substances and attack and defuse them. While botulinum toxin lasts longer than most proteins in cells (up to three months as opposed to the typical protein lifecycle of hours or days), the team’s evolved proteins might end up with shorter lifetimes, which could diminish their effectiveness.
Still, since the immune system takes time to identify foreign substances, the proteases could be effective for temporary treatments. And, to side-step the immune response, the team is also looking to evolve other classes of mammalian proteases since the human body is less likely to attack proteins that resemble their own. Because their work on botulinum toxin proteases proved so successful, the team plans to continue to tinker with those, too, which means continuing their fruitful collaboration with Min Dong, who not only has the required permission from the Centers for Disease Control (CDC) to work with botulinum toxin but provides critical perspective on the potential medical applications and targets for the proteases.
“In an ideal world,” said Blum, “we can think about using these toxins to theoretically cleave any protein of interest.” They just have to choose which proteins to go after next.
Support for this study was provided by National Institute of Biomedical Imaging and Bioengineering, National Institute of General Medical Sciences, National Institute of Neurological Disorders and Stroke, the Burroughs Wellcome Fund, the Howard Hughes Medical Institute, and Ipsen Bioinnovation.