I am trying to identify the binding pocket of a porphyrin like ligand which is a sodium copper salt. Forums and literature tell me that the counter ions of the ligand should be removed and the ligand prepared using standard techniques. I want to know if the metal-free ligand sufficiently represents the metal-bound ligand, and will positive or negative results obtained from docking of metal free ligands correlate with results obtained from docking metal-bound ligands.
Finally, I would like to know if I can perform docking with a metal-bound ligand and how I can prepare a such a ligand structure for docking.
Hi Avaneesh,
I agree with Haresh, we cant do docking for metal complexes... In case of preparing your ligand molecules from servers, it wont consider metal ions. But as per my knowledge, I think we can do docking for metal complexes also (not sure) using gold and glide. We you are accessible for those software, try once.....
Good luck.:-)
Hi,
I'm not sure to get it right... the copper is part of your ligand (like the iron is part of the porphyrin), but the sodium ion is not (your ligand is negatively charged, and sodium is only there to compensate for your ligand charge). If this is correct, then you can remove the sodium, but you will have to keep the copper.... if your docking software can handle it.
How to prepare your ligand will depend on the software you use, but my first trial would be to use the software with default settings and carefully check in the log files whether copper was recognized properly.Chances are that if you software is able to recognize metals, it will tell you so.. otherwise you might get an error message....
good luck !
Yes Laurent,
You got it right, the copper is a part of the ligand, the sodium just neutralizes it which in solution would not really be associated with the ligand.
So with this in mind, i generated and minimized the ligand without the copper ion, and when i do dock it with my target protein, no binding sites are observed. This as far as i understand only means that the ligand without copper does not interact with any particular pocket, and tells us nothing of the copper-bound ligand.
I then decided to add the copper by placing it roughly where its supposed to be (in the minimized pdb file, i found the mid point of the 4 copper-interacting nitrogen atoms and placed it there). However when i tried using autodock to perform a docking, it is generating bonds between two of the nitrogens and the copper ion. This again obviously does not reflect the real ligand, which brings me back to my question of how i can generate a metal-ion bound ligand that reflects the real thing. and if adding it would actually make a difference in the docking results.
A typical drug molecule might be the hydrochloride salt of an amine, in which case you dock without the chloride ion. Or you might have the sodium salt of a carboxylate, in which case you dock without the sodium ion. In the above two examples, the counterions are detached from the ligand in aqueous solution. In your case, you need the metal ion because it is part of the ligand. It binds more strongly and stays attached to the ligand. Standard docking procedures will not work because they typically are not parametrized for a transition metal. Most likely, your metal ion is 4-coordinated by the ligand and when it binds to the protein, it becomes 5- or 6- coordinated by amino acids such as histidine, which become the axial coordinating ligands of the metal ion. This ability to make axial bonds with the protein will have a big effect on the binding energy and binding constant.
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology