In de novo protein design, scientists make new proteins “from scratch, from first principles,” said David Baker, a University of Washington biologist and one of the collaborators on the research. (The phrase “de novo” is borrowed from Latin, meaning “from the beginning.”)
The new protein, called LOCKR (for Latching Orthogonal Cage Key pRoteins), behaves like a switch. LOCKR’s potential applications are “endless,” said biologist Scott Boyken, who worked on the studies as a postdoctoral researcher in Baker’s lab.
Almost all of the interesting things that happen in your body happen with the help of proteins, which in humans add up to about 20,000 types. Sliding proteins contract muscle cells. Some proteins detect specific chemicals, like the receptor tuned to whiffs of mustard and wasabi. Other proteins digest starch into sugar.
For centuries, people have bent natural proteins to their needs. Rennet, the enzyme proteins that line calf stomachs, can be used to curdle milk to make cheese, for instance. California Institute of Technology chemical engineer Frances Arnold won a Nobel Prize last year for her work using a technique called directed evolution to create more powerful versions of existing proteins.
But borrowing from nature comes with a messy history, said study author Hana El-Samad, a biochemistry professor at the University of California at San Francisco, because proteins have been remixed by evolution for billions of years. “These molecules don’t do only one function,” El-Samad said. “They do 10.” For biologists who rely on existing proteins, it can feel like designing a washing machine that, if moved to the wrong corner of a house, starts to act like a blender.
LOCKR, on the other hand, “doesn’t do anything else except act as a switch,” El-Samad said.
Baker and his colleagues designed LOCKR from the ground up, starting with the building blocks of proteins, amino acids. Simply strung together, amino acids are as powerful as limp noodles. These chains must twist into three-dimensional structures to become functional. But predicting the ways amino acid chains will fold has taken decades. Much of Baker’s work relies on computer simulation; his lab also developed a video game, Foldit, which turns the science of folding proteins into a crowdsourced puzzle.
Over the past five years, scientists have been able to build protein structures from unnatural amino acid chains. These proteins properly fold up, but, in many cases, that’s all they do. They’re inert. “Kind of like rocks,” Baker said.
Not LOCKR. Three biologists working in Baker’s lab — Boyken, Wally Novak and Robert Langan — designed the prototype LOCKR protein in 2016. The protein is a cage of six helical sub-structures. Five of the helixes are tightly bound together. The sixth sits looser, like a tooth ready to be yanked out.
“When you add the key it basically binds to the site where the sixth helix, which we call the latch, is bound. That binding displaces the latch,” Baker said. In other words, in the presence of the key molecule, that sixth helix pops open, exposing an active site. What exists at that site can be tailored to fit a specific need.
“It’s magnificent what they’re doing, without a doubt,” said chemist Michael Hecht, a synthetic biologist at Princeton University. “What’s especially exciting here,” he added, is that the LOCKR protein works within living cells.
The study authors conducted an “impressive set of demonstrations,” Khalil said, such as using LOCKR to control cell death. They also built a circuit of LOCKR proteins that resembled how an engineered delivery cell might secrete a drug.
This is not the first designer protein that can switch a cell between a living and dead state, Hecht said. But LOCKR is more broadly applicable than those predecessor molecules. “It is much more plug and play than what other people have done,” he said.
There are patents pending for LOCKR’s commercial use, but its creators have made the system freely available to academics. Any scientist with an Internet connection can access the blueprint for LOCKR: the DNA code used to build the string of the amino acids that builds the protein. “Not a single physical piece of DNA was exchanged between our labs,” El-Samad said. “The only thing that was exchanged between our labs through this collaboration was a file that had the DNA sequence.”
Biologists can print the code for LOCKR onto a custom strand of DNA and insert this into a cell. El-Samad envisioned an injection of cells, programmed with LOCKR circuits, as a method to control brain inflammation after a traumatic injury, for instance.
“You will start seeing uses for LOCKR in mammals and explorations for therapeutic uses — well, actually, it’s already begun,” Baker said. Companies are in the early, experimental stages of exploring LOCKR as a cancer treatment. In theory, it would be similar to CAR T-cell therapy, which reprograms a patient’s white blood cells to attack cancer cells.
Synthetic biologists predict that LOCKR and other designer proteins would not interact with anything other than their targets. This remains to be proved. “The first clinical trials on de novo designed proteins are just starting,” Baker said. “So we’ll know in a couple of years.”
Immune systems, vigilant for signs of invading bacteria and viruses, are attuned to foreign proteins. Researchers are working on strategies to prevent designer proteins from triggering immune responses.
They hope to use LOCKR for precise applications, El-Samad said. “It’s a cell that is engineered that goes and does its job and then goes away.”