Crystal structure of a monomeric retroviral protease solved by protein folding game players

Firas Khatib1, Frank DiMaio1, Foldit Contenders Group, Foldit Void Crushers Group, Seth Cooper2, Maciej Kazmierczyk3, Miroslaw Gilski3,4, Szymon Krzywda3, Helena Zabranska5, Iva Pichova5, James Thompson1, Zoran Popović2, Mariusz Jaskolski3,4 & David Baker1,6

Following the failure of a wide range of attempts to solve the crystal structure of M-PMV retroviral protease by molecular replacement, we challenged players of the protein folding game Foldit to produce accurate models of the protein. Remarkably, Foldit players were able to generate models of sufficient quality for successful molecular replacement and subsequent structure determination. The refined structure provides new insights for the design of antiretroviral drugs.

Foldit is a multiplayer online game that enlists players worldwide to solve difficult protein-structure prediction problems. Foldit players leverage human three-dimensional problem-solving skills to inter- act with protein structures using direct manipulation tools and algo- rithms from the Rosetta structure prediction methodology1. Players collaborate with teammates while competing with other players to obtain the highest-scoring (lowest-energy) models. In proof-of- concept tests, Foldit players—most of whom have little or no back- ground in biochemistry—were able to solve protein structure refine- ment problems in which backbone rearrangement was necessary to correctly bury hydrophobic residues2. Here we report Foldit player successes in real-world modeling problems with more complex devia- tions from native structures, leading to the solution of a long-standing protein crystal structure problem.

Many real-world protein modeling problems are amenable to com- parative modeling starting from the structures of homologous pro- teins. To make use of homology modeling techniques in Foldit, we introduced a new capability called the Alignment Tool, which allows players to manually move alignments and thread their sequence onto the structures of known homologs (Supplementary Fig. 1). Players are able to combine different regions from multiple templates into a single hybrid structure (partial threading) and load in previously saved solutions as templates to hybridize with their current models.

Our aim was for Foldit players to use these new tools to solve real- world problems; the Critical Assessment of Techniques for Protein Structure Prediction (CASP) experiment was an ideal venue in which to test this. CASP is a biennial experiment in protein structure predic- tion methods in which the amino acid sequences of structures that are close to being experimentally determined—referred to as CASP targets—are posted to allow groups from around the world to predict the native structure (http://predictioncenter.org/casp9/). Each group taking part in CASP is allowed to submit five different predictions for each sequence. Foldit participated as an independent group during CASP9 and made predictions for the targets with fewer than 165 resi- dues that the CASP organizers did not indicate as oligomeric. For targets with homologs of known structure—the Template-Based Modeling category—Foldit players were given different alignments to templates predicted by the HHpred server3 via the new Alignment Tool. Despite these new additions to the game, the performance of Foldit players over all CASP9 Template-Based Modeling targets was not as good as those of the best-performing methods, which made better use of information from homologous structures; extensive energy minimiza- tion used by Foldit players tended to perturb peripheral portions of the chain away from the conformations present in homologs.

Continues...

To disable M-PMV’s protease, we need to know exactly what it looks like. Like real scissors, the proteases come in two halves that need to lock together in order to work. If we knew where the halves joined together, we could create drugs that prevented them from uniting. But until now, scientists have only been able to discern the structure of the two halves together. They have spent more than ten years trying to solve structure of a single isolated half, without any success.

The Foldit players had no such problems. They came up with several answers, one of which was almost close to perfect. In a few days, Khatib had refined their solution to deduce the protein’s final structure, and he has already spotted features that could make attractive targets for new drugs.

“This is the first instance that we are aware of in which online gamers solved a longstanding scientific problem,” writes Khatib. “These results indi*cate the potential for integrating video games into the real-world scientific process: the ingenuity of game players is a formidable force that, if properly directed, can be used to solve a wide range of scientific problems.”

This work was supported by the Center for Game Science at the University of Washington, US Defense Advanced Research Projects Agency (DARPA)

AlohaChris wrote:More importantly did the high scorers get credit in the paper? Will they share in the profits of any future drug?