Prioritizing kinase:inhibitor complexes

[jchodera]

I've finally had a chance to start looking through kinase:inhibitor complexes that @steven-albanese deposited here.

The criteria I am using to prioritize these files are:

1. Anything where the inhibitors populate a mixture of protonation states in solution
2. Anything where the titratable inhibitor groups are buried or near charged/titratable residues in the binding site

From critieria (1), it looks like Epik's predictions suggest the following inhibitors have multiple protonation states populated at pH 7.4 in solution:

• Alectinib - two protonation states that differ by 0.2 kT
• Axitinib - two protonation states that differ by 1.5 kT
• Cabozantinib - two protonation states that differ by 0.5 kT
• Crizotinib - two protonation states that differ by 0.5 kT
• Dabrafenib - two protonation states that differ by 1 kT
• Dasatinib - two protonation states that differ by 1 kT; three within 1.7 kT
• Imatinib - two protonation states that differ by 0.4 kT
• Ponatinib - two protonation states that differ by 0.4 kT
• Regorafenib - two protonation states that differ by 0.5 kT
• Vemurafenib - two protonation states that differ by 0.8 kT

These kinase inhibitors would all be great to look at.

Of the PDB files of kinase:inhibitor complexes that @steven-albanese has collected, this list has

• Alectinib-ALK
• Axitinib-VEGFR1
• Crizotinib-ALK
• Crizotinib-MET
• Dabrafenib-BRAF
• Dasatinib-BCR-ABL
• Imatinib-BCR-ABL
• Regorafenib-VEGFR1

Some of the inhibitors above aren't represented here because their crystal structures are to closely related mammalian kinases (just a few amino acids different) that I am working on adding now.

[jchodera]

Looking at Alectinib-ALK, the morpholine ring of Alectinib is titratable on the nitrogen, and this is somewhat water-exposed by equidistant between GLU and ARG residues.

[jchodera]

In Axitinib-VEGFR1, a titratable proton in the indazole ring appears to be hydrogen bonding to the carbonyl oxygen of GLU102.

[jchodera]

In Crizotinib-4ANQ, there is a titratable proton on the piperidine ring that is next to GLU105, but in solution, the penalty for putting the proton there is about 5 kT.

[jchodera]

Dabrafenib-BRAF has the titratable proton of the sulfonamide linker group sterically clashing with the nitrogen of ASP146, which should wipe out the protonated population.

[jchodera]

Dasatinib-BCR-ABL doesn't have much going on except for two ASP residues in the vicinity of the two titratable nitrogens.

[jchodera]

Regorafenib-VEGFR1 has two titratable protons on the purine group with a GLU, ASP, LYS, and a backbone carbonyl all very close.

[jchodera]

There appears to be some problem with retrieving files from the PDB right now, so I don't think I can generate the Ponatinib structures at the moment. Hopefully, this is enough to go on for now, though!

[jchodera]

I've just added the Ponatinib structures too. Check this out: There's a GLU, an ASP, and a HIS in the binding site pretty close to the titratable piperazine ring. Here's an image from 3OXZ:

[jchodera]

Any luck here? I think @bas-rustenburg is just about ready to start running constant-pH MD simulations.

[mrgunner]

We should have some results tomorrow.

On May 15, 2016, at 11:43 AM, John Chodera notifications@github.com wrote:

Any luck here? I think @bas-rustenburg is just about ready to start running constant-pH MD simulations.

— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub

[jchodera]

Just checking in on this!

[jchodera]

Any chance you were able to generate anything useful even without the CONECT record problem being solved?

[SalahBioPhysics]

Yes, the CONNECT section was useful only so we can animate the work. But that's fine we are able to carry on the work.