1
$\begingroup$

I have two enzymes, with heme cofactors. but their overall structure is very different.

I use Pymol regularly to visualize protein pdb structures.

I want to align both structures, but the structures don't align well due to very low homology. However, their active sites/substrate entrance channels may be more similar than the rest of the structure. so,

  1. I want to align only the substrate channels from both enzymes and compare their structure. or
  2. align both proteins only using the heme as a reference point. So both hemes from two enzymes align and then I can manually do the rest.

I am more interested in how the substrate would be accessible in both enzymes differently.

Thanks in advance.

Edit: Sorry for being a bit unclear. We can pick two proteins like cytochrome P450 BM3 with pdb id 6H1L or 6H1O. The second protein can be a similar peroxygenase UPO, e.g. 7pn4. Thanks.

$\endgroup$
3
  • $\begingroup$ @terdon , do the same rules of SO apply to Bioinformatics , the question seems too wide , at least irfan could have named two pdbs code as inputs, toghrther with the aa defining the channels to have us to try to get a result with $\endgroup$
    – pippo1980
    Commented Sep 27, 2023 at 9:26
  • $\begingroup$ If you feel the question is too broad, then ask for clarifications and/or vote to close as "needs more focus". That said, I don't see why this is too broad: the main question here is "given two pdb structures, how can I align and compare only specific parts of those structures instead?" which seems like it should be answerable by those working with this kind of data. $\endgroup$
    – terdon
    Commented Sep 27, 2023 at 9:33
  • 1
    $\begingroup$ Please add more context to this question; be as specific as possible about the problem you're struggling with. Example proteins, or other input files, are very helpful. $\endgroup$
    – gringer
    Commented Sep 28, 2023 at 3:13

1 Answer 1

1
$\begingroup$

OK, trying to answer question 2, think I need to be sure that both enzymes have same identical heme cofactor and then find a rotation/translation matrix that applied to heme of protein A moved exactly on top of heme of protein B , than I have to move the entire protein backbone of protein A using that matrix, is it reasonable ? Am I getting it wrong ? With PyMOL I tried to use Pair/Pair_fit command with two heme proteins : 2d0t https://www.rcsb.org/structure/2d0t and 4zv8 (https://www.rcsb.org/structure/4ZV8) :

pair hem_2d0t , hem_4zv8, same as pair_fit hem_2d0t , hem_4zv8, where hem_2d0t,hem_4zv8 are the heme of the two proteins [select model 2d0t and resn HEM and set_name sele, hem_2d0t or select hem_2d0t, model 2d0t and resn HEM].

See results, superposed heme:

enter image description here

Superposed Proteins :

enter image description here

As for question 1 I tried to answer using the tools available to me. I picked the two same proteins : 2d0t (https://www.rcsb.org/structure/2d0t and 4zv8 (https://www.rcsb.org/structure/4ZV8). Using parKVFinder (https://github.com/LBC-LNBio/parKVFinder , https://www.softxjournal.com/article/S2352-7110(20)30319-8/fulltext) pymol plugin on both proteins stripped of all their HETATM, I calculated their cavities. As an example , here a png of 2d0t cavities :

enter image description here

Then I extracted the residues surrounding the cavity using :

select model 2d0t_merged near_to 5.0 of resn KAF ; [named sele_2d0t]

select model 4zv8_merged near_to 5.0 of resn KAB ; [named sele_4zv8]

being KAF and KAB cavities described as non_bounded atom in pdb format as output of parKVFinder .
I believe parKVFinder has its own output for the residues surrounding cavities, but wanted to try PyMOL

Example for 2d0t :

enter image description here

Then I used

cealign sele_4zv8 , sele_2d0t (https://pymolwiki.org/index.php/Cealign) to align the residues of the two cavities. (cealign uses just CA of the selections provided).

Results , only cavities and surrounding residues :

enter image description here

the two heme protein cavities are not alignrd even if the HIS (2dt0, red) and CYS (4zv8, yellow) are on the same side :

enter image description here

NOTE:

2dt0 in magenta

4zv8 in cyan

I choose 2 different heme protein (could be more extreme though) , just because the question was open to broad attempts, could be interesting too see for example results on not too distant cytochromes

$\endgroup$
1
  • $\begingroup$ Thank you very much. I will try to follow the process and report back. $\endgroup$
    – Irfan
    Commented Oct 2, 2023 at 16:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.