(Note: These tutorials are meant to provide illustrative examples of how to use the AMBER software suite to carry out simulations that can be run on a simple workstation in a reasonable period of time. They do not necessarily provide the optimal choice of parameters or methods for the particular application area.)
Copyright Ross Walker 2006

TUTORIAL 3

Examining the pKa's of Asp and Glu Residues

By Ross Walker

In this tutorial we will be using the WW domain from the formin-binding protein (FBP). The pdb for this (PDB ID: 1e0l) is again an ensemble of NMR structures:

1e0l.pdb

The aim of this tutorial is to show how one can use Amber's GB model and minimisation routines to estimate the pKa shifts for glutamic acid and aspartic acid residues in proteins. We will examine two different structures from an NMR ensemble and calculate the pKa shift in each case. The results should show that the shift in pKa is very much dependent on the protein conformation.

In this tutorial we will only require the first two structures from the 1e0l ensemble so I have stripped these out and cleaned them up for you here:

FBP1.pdb, FBP2.pdb

This protein does not contain any histidine residues so we do not need to worry about their protonation states here. It also does not contain any cysteine residues so we do not need to worry about having to manually add any disulphide bridges in xleap. Feel free to visualise these structures in VMD if you want to see the differences between them.

Note: In this tutorial we will be doing everything by hand and as such we can only look at a handful of structures. In reality we would script this entire process to make it automatic. We could then look at a huge ensemble of structures and generate average pKa shifts.

This protein contains five glutamic acid residues (7, 10, 17, 31, 35) and one aspartic acid residue (15). Since it will take a while to do all of them we shall just do 2 of the glutamic acid residues and the aspartic acid residue.

The shift in pKa between solution and protein can be calculated as follows:

DpKa = -(DG(X- + H+ -> XH)protein - DG(X- + H+ -> XH)solution ) / kBT*ln(10)

(KBT = 0.597 KCal/mol [at 300K])

Hence in order to perform the above calculation we will need, for each residue, the protein structure with and without the proton and the residue on its own in solution with and without the proton.

The table of energies we will need to complete is therefore as follows:

Structure 1 PROTEIN
NO H
(A)		 PROTEIN
GLU 7 H
(C)		 PROTEIN
GLU 10 H
(E)		 PROTEIN
ASP 15 H
(G)
Structure 2 PROTEIN
NO H
(B)		 PROTEIN
GLU 7 H
(D)		 PROTEIN
GLU 10 H
(F)		 PROTEIN
ASP 15 H
(H)
    SOLUTION
GLU NO H
(I)		 SOLUTION
GLU H
(K)		  
    SOLUTION
ASP NO H
(J)		 SOLUTION
ASP H
(L)		  

So, let's create the input pdb files we will require using xleap.

CLICK HERE TO GO TO SECTION 2

(Note: These tutorials are meant to provide illustrative examples of how to use the AMBER software suite to carry out simulations that can be run on a simple workstation in a reasonable period of time. They do not necessarily provide the optimal choice of parameters or methods for the particular application area.)
Copyright Ross Walker 2006