Chaperone Folds “Mirror Image” Protein

Proteins made exclusively of D-amino acids—rather than naturally occurring L-amino acids—are attractive as potential therapeutics because they resist enzymatic degradation and reduce immune responses in the body. As the methods for synthesizing these “mirror image” proteins improve, scientists are able to make increasingly larger versions. But they then face the challenge that these bulky proteins might not be able to fold properly without help.

As proteins “get larger and larger, it’s more likely that they depend on a chaperone for efficient folding,” says Michael S. Kay, a biochemist at the University of Utah School of Medicine who focuses on d-peptide therapeutics. After painstakingly synthesizing one of these large molecules, he adds, “we don’t want to be left holding the bag at the end with a beautiful protein that can’t fold and therefore we can’t use.”

So the question arises whether naturally occurring chaperone proteins can fold mirror-image proteins as well as their l-amino acid counterparts. Kay hoped that GroEL/ES, which is a chaperone protein complex involved in the folding of hundreds of different proteins, would be enough of a generalist to fold mirror-image proteins too.

Kay and his colleagues found that GroEL/ES is indeed “ambidextrous.” They used recombinant GroEL/ES to successfully fold a synthetic mirror-image version of 4-hydroxy-tetrahydrodipicolinate synthase (DapA), the smallest protein they could find that depends on the chaperone and has a functional assay that doesn’t require chiral reagents (Proc. Natl. Acad. Sci. USA 2014, DOI: 10.1073/pnas.1410900111). DapA is also the longest protein chemists have synthesized so far.

“We purposely picked a protein that we knew was a worst-case scenario, a protein that really needs chaperone assistance to fold in a cell—without it, it doesn’t fold at all—and used that to determine whether we could get away with using recombinant chaperones to help fold large d-proteins,” Kay says.

The success with DapA gives Kay hope that other d-proteins can also be folded by conventional chaperones. The way to establish GroEL/ES’s generality will be to synthesize a mirror-image version of the chaperone and see whether it too can fold a diverse panel of L- and D-proteins, Kay says. The problem is that such a large cellular machine is beyond the reach of current protein synthesis methods. Although GroEL’s symmetry helps, each subunit has more than 500 amino acids.

Kay’s lab specializes in d-peptide drug design, particularly for antiviral agents. Researchers in the lab discover potential therapeutics by synthesizing mirror-image targets and using them in a phage-display assay. Until now, they’ve been limited to target proteins with about 50 to 100 amino acids.

In the current study, the team pushed the limits of chemical peptide synthesis and native chemical ligation to make both natural and mirror-image versions of DapA, which has 312 amino acids. They used a peptide hydrazide ligation technique developed by Lei Liu’s group at Tsinghua University, in Beijing. The method allowed Kay and coworkers to use an efficient convergent synthesis strategy with fewer steps than would have been necessary with a traditional ligation method.

“This is a fantastic study showing that chaperones are not guided by chirality but rather by general chemical features of amino acid side chains,” says Sachdev Sidhu, a professor at the University of Toronto and cofounder of the D-protein therapeutic company Reflexion Pharmaceuticals. “On the practical end, the work opens up a lot of new opportunities to use natural chaperones to fold synthesized D-proteins, and this should enable both new theoretical studies and potential therapeutic development.”

Kay’s long-term goal is to make an entire cell out of D-proteins and other right-handed molecules. As an intermediate goal, he wants to develop a D-ribosome that could make d-proteins via in vitro translation, eliminating the need for complex chemical protein syntheses.

“What’s nice about the ribosome is that it’s defined; there’s only one subunit that’s larger than 300 residues” in Escherichia coli, Kay says. “We’re getting close to the ability to synthesize each of the individual subunits, which could then be assembled on mirror-image ribosomal RNAs. A D-ribosome would allow us to efficiently synthesize other components of the cell.”