Barreling the way to designer proteins

From the Stoddard lab, Basic Sciences Division and the Baker lab, UW

Proteins are the undisputed workhorses of the cell; their functions often depend on the ability to fold into specific three-dimensional structures and to bind specifically to small molecules. The ability to design and make proteins from scratch that will bind target molecules of one’s choosing could greatly increase biotechnological applications that are currently limited to backbone geometries of known protein structures. Despite recent advancements in the design of protein folds from scratch and redesign of existing native scaffolds to bind small molecules, some outstanding hurdles remain. One unsolved challenge is the de novo design of b proteins, which are comprised of b-strands and sheets that tend to generate protein aggregates if not perfectly aligned. In nature, many b-barrel, proteins containing a pleated b-sheet rolled into a cylindrical structure, bind small ligand molecules. However, successful de novo design of b-barrels has not been reported. Another unsolved challenge is the design of proteins customized to bind any small molecule of interest. The staggering number of possible positions and spatial orientations of a small molecule within a protein cavity complicate this objective. Furthermore, precise orchestration of the backbone and side chain geometry without sacrificing protein conformation and function is required. Therefore, accurate and efficient computational approaches are required to narrow down optimal solutions.

A collaboration between investigators in the David Baker laboratory at University of Washington and Barry Stoddard’s lab in the Basic Sciences Division at Fred Hutch has reported a new study that overcome many of these hurdles. In a recent issue of the journal Nature, the authors describe a computational method for designing a novel, engineered protein fold that was then customized to bind a small molecule of interest. This led to the creation of a designed b-barrel protein that binds a small molecule in a way that leads to the generation of a fluorescent signal. Dr. Stoddard explained how this was a great collaborative effort: “Our laboratory participated in this project by characterizing the biophysical behavior and structure of the designed protein. These experiments are crucial for the continued development and application of protein engineering, both by validating the behavior of engineered protein constructs, and by providing details of their form that is used to improve computational engineering approaches.”

The authors started by devising general principles for designing stably folded β-barrel proteins. They discovered that structural irregularities, known as glycine kinks and β-bulges, must be introduced into β-barrels to alleviate molecular strain and to maintain the continuous pattern of hydrogen bonds needed to form the cylindrical structure. With this approach, they built computational models of hundreds of β-barrel ‘backbones’ and identified amino-acid sequences that stabilize each backbone. Four designs that were predicted to be most stable were synthesized, one of which folded into a monomeric β-barrel. Strikingly, the computationally predicted model was very similar to the experimentally determined structure, with high stability. The authors then used a novel computational approach to model a large number of possible positions of a small molecule in space and how it docks into a binding site. They designed a β-barrel protein that binds DFHBI, a small molecule that fluoresces upon binding. Using this approach, the authors successfully generated proteins that bound DFHBI with moderate affinities. To improve the binding affinity, X-ray crystal structures of protein variants were used to guide additional rounds of computational design, where every amino-acid residue was systematically mutated to find changes that improved binding affinity. The resultant designer proteins bound DFHBI with greater affinity and enhanced its fluorescence in vitro and in vivo.


Structures of a naturally evolved green fluorescent protein (top panels) and the comparable de novo engineered ‘mFAP' fluorescent protein (bottom panels). The engineered construct is less than half the size of GFP, is extremely thermostable and soluble, and fluoresces only when it binds its chromophore in trans. Figure provided by Dr. Stoddard

Dr. Stoddard explained the significance of this work: “This study represents, in several ways, the current state-of-the-art for protein engineering. In it, investigators in the Baker lab at the UW successfully combined the de novo design of a folded protein scaffold with a novel algorithm to search and query protein sequence and protein conformations in order to imbue the designed protein with a biological function.  Either of these accomplishments are extremely noteworthy on their own, but when combined they provide a powerful new approach for the creation of new proteins with important biological and biochemical properties.  In this case, a fringe benefit of this study was the creation of an entirely novel fluorescent reporter that differs significantly from commonly used reagents such as GFP.” Indeed, such designer fluorescence activating proteins can be used to monitor gene expression, track molecules in cells, or function as biosensors.

When asked about next steps, Dr. Stoddard revealed: “My own laboratory’s next steps in this project are two-fold: first to continue to work with computational biologists to create novel protein functions with increasing complexity, and also to utilize this newly created reagent in our own work developing novel cell imaging reagents.”

Dou J, Vorobieva AA, Sheffler W, Doyle LA, Park H, Bick MJ, Mao B, Foight GW, Lee MY, Gagnon LA, Carter L, Sankaran B, Ovchinnikov S, Marcos E, Huang P-S, Vaughan JC, Stoddard BL, Baker D. 2018. De novo design of a fluorescence-activating beta-barrel. Nature. Sep;561(7724):485-491

Funding was provided by the National Institutes of Health, Howard Hughes Medical Institute, Fulbright Commission for Belgium and Luxembourg, Marie Curie International Outgoing Fellowship, Washington Research Foundation, and Open Philanthropy.

Research reported in the publication is a collaboration between Cancer Consortium members Barry Stoddard (Fred Hutch) and David Baker (UW)