LEGOs for biochemists

Figuring out how proteins fold—that is, the way that amino-acid sequences determine the unique structures and functions of protein molecules, which then act as “biology’s workhorses”—remains one of the biggest open questions in biology today. One common approach analyzes numerous proteins with similar structures and functions—a protein family—to try to tease out the fundamental interactions responsible for a given property. Now a Caltech team of chemical engineers, chemists, and biochemists has created a huge family of proteins that, even though they have very different sequences, all fold the same way.

Grad student Christopher Otey and his colleagues analyzed three natural protein structures and pinpointed locations at which they could be broken apart and reassembled, like LEGO pieces. The proteins were then broken into eight pieces each and reassembled into all possible eight-piece combinations, creating 38, or 6,561, sequences. Nearly half of these constructs were able to fold themselves to constitute an artificial protein family. Says Otey, “In this single experiment, we’ve been able to make about 3,000 new proteins.”

The viable proteins have an average of about 72 sequence changes relative to any known protein. “We can use the new proteins and new sequence information to learn about the original proteins,” Otey adds. “For example, we can determine which combinations of amino acids contribute to specific protein properties.”

The original proteins belong to a family called the cytochrome P450s, which play critical roles in drug metabolism, hormone synthesis, and the biodegradation of many chemicals. The researchers broke these roughly 460-amino-acid proteins into LEGO blocks of about 60 to 70 amino acids each. It has taken researchers 40 years to collect 4,500 natural P450 sequences, but the Caltech team required only a few months to create their new P450s.

“During evolution, nature conserves protein structure. We do the same thing by shuffling natural proteins with the help of computational tools. By changing protein sequences, we can generate new functions,” Otey says. “One of our goals is to be able to create new and possibly useful proteins for pharmaceuticals, to do chemical syntheses, or to be used in sensors or other biotechnology applications.”

The paper appeared in the April 10 issue of the Public Library of Science Biology. The other authors include Frances Arnold, Caltech’s Dickinson Professor of Chemical Engineering and Biochemistry; biochemistry postdoc Marco Landwehr; Jeffrey Endelman, PhD ’05 in bioengineering; chemistry grad student Jesse Bloom; and postdoc Kaori Hiraga, now at the New York State Department of Health. —RT