Tuberculosis Challenge – Alternate Target

Tuberculosis (TB) is a disease that affects millions of people. We have posted a protein drug target puzzle previously on this topic. In our continued effort to make a dent in this disease, we have also partnered with the Sacchettini lab at Texas A&M University to post another drug target puzzle for TB.

The Sacchettini lab is working in collaboration with other groups on understanding biology and virulence factors of tuberculosis bacteria. The ability of Mycobacterium tuberculosis, which causes TB, to survive inside the host depends on sensing the environment and launching appropriate responses to stimuli. This means that specific protein production levels are strictly controlled and tuned. The machinery and the players of this carefully orchestrated battle against our immune system are poorly characterized. In general, protein production levels could be regulated on multiple levels and by different means. One of the ways involves small non-coding RNA molecules which aid in efficient translation of some mRNAs into proteins and the degradation of others. In pathogenic bacteria specifically, the regulation of production of the proteins required for virulence and intracellular survival has been shown to depend on small RNAs. Reviewed here (Oliva G., Sahr T., Buchrieser C. (2015). Small RNAs, 5’ UTR elements and RNA-binding proteins in intracellular bacteria: impact on metabolism and virulence. FEMS Microbiol. Rev. 39 331–349. :

image (3).png

To carry out their missions, small RNAs require protective chaperon protein – Hfq. Specifically, the protein structure adopts an Sm like fold composed of 6 subunits forming a homo-hexameric ring. Hfq and Sm proteins have been identified in numerous bacteria, yet no known homologs have been annotated in Mycobacterium tuberculosis genome. Through careful examination of secondary structure patterns predictions of the Mycobacterium tuberculosis proteome, Rv3208A has been proposed as a possible Hfq candidate.

If we were able to solve the structure, it would mean that we learn about machinery which has been shown to be important for virulence in other pathogens but is not characterized in Mycobacterium yet. By targeting this RNA chaperon protein, instrumental to any small RNA mediated responses, scientists can prevent Mycobacterium tuberculosis from survival inside human host.

Right now, the protein has been crystallized and diffraction data are available, but none of the models that scientists have created have helped to solve the phase and build the structure. By posting this protein, we are hoping that everyone can come up with a model that will help resolve the structure. As always, we are committed to publishing the work and sharing models created by Foldit players. Lets make a dent in TB!

( Posted by  free_radical 76 2490  |  Tue, 11/29/2016 - 19:29  |  6 comments )
Joined: 01/12/2015
Groups: None
This puzzle will be updated

This puzzle will be updated and reposted. Sorry for the delay.


Joined: 09/24/2012
Groups: Go Science
We are on the starting block

I wonder if you'll give it with ED. Wait and see.

Susume's picture
User offline. Last seen 5 hours 6 min ago. Offline
Joined: 10/02/2011
Questions about contacts to the RNA

This is a set of questions that could probably be answered fairly quickly by the scientist(s) at the Sacchettini lab who are working on the TB Hfq protein (puzzles 1311/1312).

In the Hfq proteins that have already been solved, which sidechains "contact" the RNA in the hub of the hexamer? Are these sidechains well conserved across homologs to the solved Hfq proteins? If there is some variation, what are the AAs that most commonly appear in those positions?

Looking at the TB protein we have been given, and the set of homologs to the TB protein (different set from the homologs to the solved Hfq proteins), are there similar sidechains that are well conserved across the homologs to the TB protein? What positions in the TB protein align to those conserved sidechains in the homologs? My focus is not on all conserved sidechains, but on the ones that match the list above of common RNA-contacting sidechains in the solved Hfq proteins.

If these questions can be answered, it would give us a clue which residues to put in the hub of the hexagon. It also suggests a further puzzle for this protein - give us a snippet of RNA as a ligand to contact, and (optionally) include constraints from the conserved contacting residues to the RNA. This could be done as a monomer + RNA, since the hexamer is too big for many machines. Let us know if we should expect H-bonds between the protein and the RNA, or just contact without bonds.

Ideally, I would want a pair of such puzzles: a monomer + RNA for points, and a hexamer + RNA for no points, where we can load solutions back and forth between them. Then we can check in the hexamer whether our solution works with symmetry, but still work on it competitively in the monomer puzzle which will not crash our machines.

kabubi's picture
User offline. Last seen 20 weeks 6 days ago. Offline
Joined: 11/29/2008
Groups: Void Crushers
me too would like to load

me too would like to load 1312 solution on 1311

bertro's picture
User offline. Last seen 1 day 6 hours ago. Offline
Joined: 05/02/2011
Groups: Beta Folders
I second that

Ideally, I would want a pair of such puzzles: a monomer + RNA for points, and a hexamer + RNA for no points, where we can load solutions back and forth between them. Then we can check in the hexamer whether our solution works with symmetry, but still work on it competitively in the monomer puzzle which will not crash our machines.

I definitely second that.

Joined: 01/12/2015
Groups: None
Hi Susume, Here is a reply

Hi Susume,

Here is a reply from one of the scientists in Jim's lab.

1. It is hard to pin-point the RNA interacting residues from the sequence alignments - sequence homology is too low, so you may end up being severely misled. On top of it, Mtb Hfq has extra strands on N-terminus (predicted to form beta-clasp), but it is not clear how it would fit into classic Hfq doughnut arrangement.
2. From studying solved Hfqs with RNA bound few potentially useful observations could be made:
- RNA interacting spots located at the edges of secondary structure elements and/or in between them
- Most common interactions with RNA include: Q and N - H-bonding with the base; F (or Y) - stacking with the base ring; Y - stacking or bonding with the sugar; H - H-bonding with the sugar O and/or with the phosphate group. Mtb Hfq has one of each for those within the useful positions (I would pair this with the secondary structure predictions):
10 20 30 40 50 60 70
| | | | | | |
3. Yes, there are H-bonds with RNA (see above in (2). Structures were solved with polyU (pdb entry 4Y91)

Get Started: Download
  Windows    OSX    Linux  
(10.7 or later)

Are you new to Foldit? Click here.

Are you a student? Click here.

Are you an educator? Click here.
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, Microsoft, Adobe, RosettaCommons