How to elucidate binding differences of natural and artificial ligand unsing decomposition analysis

[Paul Schrank]

Hey guys,

Sorry for the dumb question. Because I am not so well versed in MMPBSA/MMGBSA calculation i wanted to ask, if you might have some ideas (maybe speaking from experience), how to most promisingly elucidate binding differences between natural and artificial ligands for an enzyme. In this particular case I wanted to see, how the ligands would differ in there catalytic convertibale conformation (which i tried replicate by constraining the position reactive group of the substrate Gln/Asn side chain in respect to the active site residues Cys69, Asp260). 

The glutamine dipeptide in this case would be the natural ligand and the asparagine dipeptide the artificial one. Beforehand I identified (through some QM studies) that the difference most preasumably lies in the nucleophilic attack of the Cys69 to the reactive group and the concerted protonation of the amide side chain by the Asp260 (for which the needed conformation was possible to be adapted by the natural, but not the artificial ligand).

Through my constrained MD I tried to replicate the this "catalytic" conformation for both ligands and through an MMPBSA/MMGBSA analysis (+Decomposition), which residues most preasumably play a role in the correct adaption of the conformation (by finding residues, energetically speaking, who positively and negatively influence the adaptation of the conformation).

But it seems, that through the parameters I used for calculation (mostly default, I provided in the documents) for MMPBSA/MMGBSA prove to be rather unsuccessful, as the natural and artifical ligand don't seem to differ much. Do you maybe have any idea what the issue could be, or is my approach just not correct? 

Sorry for the rather unrelated question, I just wanted to know if my trail using this approach with MMPBSA makes sense or not :(

Here is the link with the result files:

https://gigamove.rwth-aachen.de/de/download/a66f3e8be7045050c2ada86637d6e48b

Thank you in advance, but I could also understand if you can't or don't have the time to answer this, as this question is rather unrelated to the general code. But maybe you could give some inside on how to choose the correct calculation parameters for the question on hand?   

Best regards

Paul Schrank

[marioe911116]

Couple of observations and suggestions:

-I see the N-terminal region is kinda "disordered" and it seems to produce a lot of "noise" in the simulation. The movement of this N-terminal seems to have a direct effect onto loops interacting with the substrate and this could influence the results. Is this N-terminal region solved in the starting structure by any experimental method or is it a based on predictions?

-Based on the diferences in the ligands, I am actually quite happy with the results taking into account the limitations of the method. ~One kcal/mol difference between Gln and Asn means almost an order of magnitude difference in the binding due to an additional carbon in the Gln. This method doesn't consider charge transfer or other QM processes, which could have an important role in this case.

-Finally, there is QM/MMGBSA available in gmx_MMPBSA (https://valdes-tresanco-ms.github.io/gmx_MMPBSA/dev/examples/QM_MMGBSA). Maybe you could try with that and see if it is a better fit for your problem....

hope this helps!

[Paul Schrank]

Firstly thank you very much for your quick response :) 

Maybe I go through your points one by one:

-Yes you assumed correctly the N-terminal is modeled and not experimentally solved. The aspartic acid in the so called FRAPD tag was published to yield an improvement in catalytic activity for the natural enzyme, so I didn’t want to neglect the effect during my MD run. But you are right maybe I can consider redoing the MD without the N-terminal extension

-oh ok I didn’t know the small difference in Binding energy has such an impact on the conversion rate. I guess you hint to the PBSA calculations correct? In this case I just saw that the Cysteine had a large electrostatic repulsion to both ligand, from which I though it could falsely the results. Is there maybe another ipb or igb method which is more fitting to these constrained MD runs? 

-Yeah I also figured to try to analyze using the QM method but sadly it is not compatible with the decomposition analysis, correct? Because I would like to identify hot spots for mutation from my results, so I think that the decomposition analysis will be mandatory :(

Thank you for your kind suggestions! 

Best regards 

Paul

[marioe911116]

You can perform alanine scanning, and mutate those residues closer to the ligand and see the impact on the calculated binding free energy...

check this paper for more details

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0137787

[Paul Schrank]

Firstly thank you very much for your quick response :) 

Maybe I go through your points one by one:

-Yes you assumed correctly the N-terminal is modeled and not experimentally solved. The aspartic acid in the so called FRAPD tag was published to yield an improvement in catalytic activity for the natural enzyme, so I didn’t want to neglect the effect during my MD run. But you are right maybe I can consider redoing the MD without the N-terminal extension

-oh ok I didn’t know the small difference in Binding energy has such an impact on the conversion rate. I guess you hint to the PBSA calculations correct? In this case I just saw that the Cysteine had a large electrostatic repulsion to both ligand, from which I though it could falsely the results. Is there maybe another ipb or igb method which is more fitting to these constrained MD runs? 

-Yeah I also figured to try to analyze using the QM method but sadly it is not compatible with the decomposition analysis, correct? Because I would like to identify hot spots for mutation from my results, so I think that the decomposition analysis will be mandatory :(

Thank you for your kind suggestions! 

Best regards 

Paul


Gesendet von Yahoo Mail für iPhone

--
You received this message because you are subscribed to a topic in the Google Groups "gmx_MMPBSA" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/gmx_mmpbsa/fbpo4oD9aKg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to gmx_mmpbsa+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gmx_mmpbsa/130c3f57-e229-4c69-87e3-1d6d6d73a49bn%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.