Examples of structure-based functional comparison of enzyme active centers

Question

[natasha.sernova]

Dear all, I’m trying to compare 4 orthologous proteins-enzymes and the corresponding pdb-files for them. Species are rather far from each other, so the sequences are far from identity. Never the less their functions look close – I hope at least their active centers have high structural identity, if it is a correct term in this case.
I would like to learn how to compare them correctly from both points of view - structure and function together. Could you please give me a good example of structure-based functional comparison? So far I failed to find any good article as an example. I hope the article will describe all the tools the authors used to accomplish the task. Thank you very much! Natalia

Answer 1

[Mensur Dlakic]

Russ Altman's group has a publicly available set of programs that extract structural features from PDB files, and can use those to find new structures with similar properties.

https://www.ncbi.nlm.nih.gov/pubmed/18831785

You should be able to find programs that compare protein active sites by Googling, as that is a fairly active area of research. I will share with you my experience with that subject, though my memory is not perfect as this was more than 10 years ago and the computer that had all the relevant info has since been retired. I know that I used programs from Uppsala Software Factory, and I heartily recommend all of them to anyone who is interested in protein structure. Pretty sure that for this purpose I used SPASM.

Anyway, here are two looks of the proteins I know to be functionally related, but they do not align structurally all that well. First, a side view:

Active site view (from the top, with regard to previous image):

These two structures have 64.4% of residues aligned, with RMSD 3.08 Å. While it is not easy to argue about their relationships from the global alignment of the two structures, it is fairly obvious that they are related when one looks just at the alignment of their active sites that shows catalytic residues and ignores the rest. This part is done by USF programs and the result is shown below.

[natasha.sernova]

Dear Professor Dlakic, thank you very much indeed for your answer! By the way, do you like PYMOL? There are so many other tools now like CHIMERA, etc, but my colleagues make me use PYMOL to solve this problem, at least, to make pictures. I'll check your links tomorrow. Thousand thanks for your suggestion - I really need some professional guidance! Sincerely yours, Natalia Sernova

[Mensur Dlakic]

I like PyMol mostly because of its protein displaying aesthetics, and to some degree because I have solved enough visualization problems with PyMol that it would be too much work to rediscover all of them with other programs. Back in the day when I worked more with nucleic acids, Chimera was my choice. Chimera is an excellent program that has been developed for a long time and is still actively maintained. I don't think there is a bad choice between these two programs.

[Joe]

Personally I find Chimera to be much more user friendly, and the fact that its 100% free without having to sort out academic licenses makes it a natural choice for me. Their user help forum is also excellent and they'll help you out very quickly with even the most trivial questions.

There is a new version ChimeraX which is being developed currently which is excellent for handling larger structures like biomacromolecular complexes.

[natasha.sernova]

Thank you very much, but I don't have any free choice. I have to use PyMol. I'll try Chimera next time.

Answer 2

[Joe]

For structural, I'd look at tools on the Zhang lab website (the guys behind ITASSER). TMAlign is particularly good for structural comparisons, though this tells you more about their global similarity than just an active site.

For local similarities, good ol' fashioned RMSD is tried and tested. You could calculate atomic distance differences for just residues known to participate in the active site etc.

[natasha.sernova]

Many thanks, it's a nice group.

Since active site is labelled, it's one more chance.

Active Site Of Enzymes