Protein Design Critique: Cubane FeS Binder

A few weeks ago, we challenged Foldit players to design a protein that could bind an iron-sulfur (FeS) cluster, in Puzzle 1688: Cubane FeS Binder Design. A cubane type [4Fe-4S] iron-sulfur cluster is a "cube" made out of alternating iron and sulfur atoms, and is bound by carefully-placed cysteine residues in a protein. Iron-sulfur metallo-proteins are responsible for electron transfer in light-harvesting, cellular respiration, many other processes. We'd like to design an iron-sulfur protein so that we can better understand electron transfer in proteins. By changing the environment around the iron-sulfur cluster, we could tune the electron transfer properties of the protein, which could open the door to metabolic engineering and new chemistry!

We asked Dr. Anindya Roy, the Baker Lab’s expert on redox proteins, to take a look at Foldit players’ designs from Puzzle 1688. Below are some comments from Anindya, which we hope players will take into account for the Round 2 puzzle, which is online now!

Fold diversity

We were very excited about the structural diversity of designs by Foldit players, who developed a variety of different protein folds! Many natural redox proteins adopt a ferredoxin fold, with a secondary structure pattern of (β-α-β)2, and we were worried that Foldit players might also favor the same ferredoxin fold. We were happy to see lots of helical bundles and other α/β folds with different secondary structure patterns, because these folds might have properties that are not possible with the ferredoxins typically found in nature. We encourage players to keep exploring helical bundles and other folds!

Room for improvement

In these initial designs, the two main areas for improvement are excessive loops and incomplete burial of the FeS cluster.

The cubane FeS cluster should be buried inside the protein core as much as possible. If the FeS cluster is to be used to catalyze a chemical reaction, then we want the active site to be protected from the water surrounding the protein. The top-scoring design by toshiue and Wilm, shown below, does a good job of burying the FeS cluster. The frozen FeS-binding loop is highlighted in blue and purple, with helices packing nicely against the cluster on three sides, shielding it from water.

If we zoom in on the FeS cluster, we can see some other nice features of this design. We like to see large, aromatic residues packed near the FeS cluster, like the TRP residue at the left of this protein. Players should try to design aromatic PHE, TRP, and TYR residues around the FeS cluster. Also, because the FeS cluster is negatively charged, it can be stabilized with complementary positively-charged residues, like the LYS residue shown beneath the cluster here. Players should also design positively charged LYS and ARG residues near the FeS cluster.

Unfortunately, we’re afraid this design has too many residues in loops, and not enough secondary structure. We can see in the first image that the frozen loop has been extended to make an even longer loop, which is unlikely to fold as intended. In order for these protein designs to fold up with high stability, we want to minimize the amount of loops in the structure. The more residues in helices and sheets, the better!

Below is a design by Galaxie and grogar7 that has a much smaller proportion of loop residues. The FeS-binding loop is flanked closely by long, stable helices on either side, and all of the other helices are connected by minimal loops. This design would have a much better chance of folding up into a stable structure.

However, in this design the FeS cluster is not completely buried, and will be exposed to the water surrounding the protein. This means we have less control over the electron transfer properties of the FeS cluster, which makes it harder to design an enzyme that can catalyze chemical reactions. One way to improve this design would be to extend the helices on either side of the FeS cluster in order to bury it away from the surrounding solvent.

This design also features lots of positively charged LYS and ARG residues at the binding site, which help to stabilize the negatively charged FeS cluster. Keep in mind that these charged residues have polar atoms that like to make hydrogen bonds. On the protein surface, they can make hydrogen bonds with the surrounding water; but if they’re buried away from solvent then they need to make hydrogen bonds within the protein!

In summary, we encourage Foldit players to design more helical bundles and other folds, with a focus on minimizing loop residues and burying the FeS cluster in the protein core! Large aromatic residues (PHE, TRP, TYR) and positively charged residues (LYS, ARG) help to stabilize the FeS binding site! Play Puzzle 1701: Cubane FeS Binder Design: Round 2 now!

( Posted by  bkoep 172 3398  |  Fri, 07/19/2019 - 22:57  |  9 comments )
spvincent's picture
User offline. Last seen 35 min 13 sec ago. Offline
Joined: 12/07/2007
Groups: Contenders
Are Histidines ok?

Are Histidines ok?

aroy10's picture
User offline. Last seen 47 weeks 1 day ago. Offline
Joined: 06/20/2019
Groups: None
Yes Histidines are ok, as

Yes Histidines are ok, as long as its not super buried in the core of the protein, then it becomes energetically unfavorable unless its completely satisfied through H-bonding.

spvincent's picture
User offline. Last seen 35 min 13 sec ago. Offline
Joined: 12/07/2007
Groups: Contenders
Also, there don't seem to be

Also, there don't seem to be any bondable atoms showing on the FeS cube.

aroy10's picture
User offline. Last seen 47 weeks 1 day ago. Offline
Joined: 06/20/2019
Groups: None
The Iron in the cube is

The Iron in the cube is already bonded with 4 Cys, ( one is missing in the puzzle, we will put in later on). The bondable atom are S in the cube, which are already H-bonded from the loop given to fold around. We encourage to put mainly hydrophobic residues to protect it from bulk water in solution. If possible ARG-Lys residues can be designed around the cube as long as they are not completely buried.

georg137's picture
User offline. Last seen 1 hour 32 min ago. Offline
Joined: 08/07/2010
Groups: Contenders
About this critique

The feedback and given information tie wet lab objectives to foldit design considerations. This is extremely helpful to have these insights beforehand, and the foldit community will produce more successful design solutions.

It is rare to see this level of information given prior to a design and is a positive step away from "We are revisiting old Foldit puzzles so we can see how useful the recent additions to the game have been".

aroy10's picture
User offline. Last seen 47 weeks 1 day ago. Offline
Joined: 06/20/2019
Groups: None
Thank you, We are trying to

Thank you, We are trying to make sure all the designs that come out of foldit is in sync with the wet lab objectives.

Joined: 06/06/2013
Groups: Gargleblasters
Thanks for the feedback!

It helps us non-scientists figure out what might make a good fold

Joined: 07/24/2019
Groups: None
Simple question about our life.

Hi, a have one question.
What all people doing here?
Why we dont living with real life?

jeff101's picture
User offline. Last seen 3 hours 7 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
Some simple answers?
explains the goals of Foldit.

Download links:
  Windows    OSX    Linux  
(10.12 or later)

Are you new to Foldit? Click here.

Are you a student? Click here.

Are you an educator? Click here.
Other Games: Mozak
Only search
Recommend Foldit
User login
Top New Users

Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, RosettaCommons