Tutorial 06

Chapter 60. Tutorial 06

60.1. Detecting Competition between a Ligand and Water

The observed X-ray crystallographic structures of enzymes are often used in order to design novel inhibitors. However at a later stage of the research, when the X-ray structure of the enzyme+inhibitor complex has been determined, it sometimes turns out that the ligand does not bind as predicted but has found an alternative binding mode. This can happen because there are positions, which may be called Disfavoured Sites in some receptor clefts from which a specific chemical group seems to be excluded because water would interact better.

In the current release of Grid there is a directive which switches the Programme into Reverse Mode for the detection of such disfavoured places. When Grid is used in this Reverse Mode it computes a balance sheet for the Probe interacting with bulk water on the one hand, and the Probe interacting with its receptor cleft on the other.

In this Tutorial, the active site of the enzyme dihydrofolate reductase (DHFR Brookhaven file 1RG7) will be investigated. The DHFR pdb structure, the inhibitor methotrexate (MTX), the GRID command files and the GRID-kont files are included in the Tutorial directory. Move to the tutorial directory typing:

cd Tutorial06

and inspect the content of the directory

ls -l

Inspect the grid_normal.in command file. This command file forces GRID to produce a dhfr_normal.kont file containing three energy contours for each of the probes reported therein. GRID program can be started by typing Grid<grid_normal.in. However, the computation requires about 20 minutes to be completed. You can save time reading the precomputed GRID-kont files attached in the directory.

Read the file with Gview (if you want to know more about Gview click here)

Gview dhfr_normal.kont

Gview automatically shows the protein structure and the GRID contours for the first N2 probe (N2 is a flat nitrogen with two hydrogens used to mimic the amino groups of the methotrexate). Try to reproduce the content of the Figures reported below.

These Figures show the observed structure of the binding cleft occupied by the inhibitor methotrexate (not used in the calculation). To import the MTX inhibitor, click in the Gview graphic windows in:

File->Open and then select the mtx.pdb from the dialogue).

Change the stile of the MTX representation by clicking on the mtx.pdb in the:

Edit->Selection window, press edit style and change the rendering style to "Sticks" and atom colour to "atom type". Then press Exit and again Exit from Selection window.

To display the atom name press ESC in the Gview graphic window (the actual cursor will change to "arrow" style), then click on methotrexate atoms.

View->Toggle mode->put the arrow in the selected atom and click Press ESC button to exit.

To change the energy levels, select from the Edit menu of programme Gview:

Edit->Field style

Move the values up to -9.0 Kcal and then press Exit.

The attractive yellow regions predict the positions where a -NH2 group will bind the protein residues at -9.0 kcal/mol. As you see the predicted regions correspond to the experimental one of the MTX.

To display the contribution to the Molecular Interaction Field produced by individual protein atoms, select the tab "Field" and click "atom contribution" (atom type color).

The blue regions of NH2-protein attractive interaction will be now mainly coloured in red. This is due to the oxygens of the active site interacting with the NH2 probe. Select the random color by clicking on "atom contribution" (contrast color).

The regions of NH2-protein attractive interaction will be coloured in different colours now, showing that different residues are giving a cooperative interaction with the NH2 groups of methotrexate. For example, the N3 atom of the MTX inhibitor makes strong hydrogen bonds to three atoms of two residues, such as 856 OG1 oxygen of Thr113 and 178 and 179 OD1 OD2 oxygens of Asp27.

The amino substituent N6 of MTX donates a couple of short strong hydrogen bonds to a pair of backbone carbonyl oxygen atoms (34 O of Ile5 and 700 O of Ile94).

Try to visualise the other Molecular Interaction Fields such as those produced by carboxyl probe and the hydrophobic field. You should be able to reproduce these figures below.

Carboxy (O::) probeHydrophobic (DRY) probe

 

Methotrexate in the binding cleft of dihydrofolatereductase. The green regions on the left are Grid contours computed on the unliganded enzyme with a DRY probe. The red regions on the right represents the carboxy probe interaction with the enzyme.

60.1.1. Disfavoured Site

However, there is an important Disfavoured Site on dihydrofolate reductase (DHFR). To show the site, import the file dhfr_reverse.kont that has been produced by the Reverse Mode in GRID. Inspect the grid_reverse.in command file. The directive responsible for the Reverse Mode is the directive LEAU that has been set to 3 in this calculation. See the User Manual in Competition Between Probe and Water for more information.

Read the GRID kont file produced in Reverse Mode with Gview.

Gview dhfr_reverse.kont

Gview automatically shows the protein structure and the GRID contours for the carbonyl oxygen probe. Import the methotrexate structure as before and set the contour levels of the probe to + 4.0 (the disfavouring energy).

The Grid contours surrounding N6 and N3 in the Figure below might well have been generated by an amino Probe. However they were actually generated by a carbonyl oxygen Probe in Reverse Mode, and these contours therefore show that water would compete so well in this part of the binding cleft that any carbonyl oxygen atom of a ligand would be disfavoured.

Now the inhibitor MTX was designed many years ago by starting with the structure of the natural substrate dihydrofolate (DHF), and replacing an oxygen of the substrate by the amino group at N6. The implicit assumption was that the two ligands would bind similarly, and that the amino group N6 of MTX would go to the position in the binding site normally occupied by the DHF oxygen. However, the Grid contours in Figure below show that the carbonyl oxygen of dihydrofolate would be disfavoured by 4-5 Kcal/Mole if DHF bound like MTX, and an aromatic OH Probe in Reverse Mode shows that an oxygen in the tautomeric phenolic form of the substrate would also be disfavoured (try yourself).

Dihydrofolate itself escapes from this predicament by an 180 degree rotation of the purine ring, and with hindsight one can see that the original design strategy for MTX was unsound. In fact the rotated orientation of the DHF ring seems absolutely essential for correct enzyme function, because dihydrofolate reductase could give a hydrogenated product with the wrong stereochemistry if its substrate were able to bind like MTX, and the whole metabolic pathway of which DHFR is a part might then be disrupted.

The disfavoured region round N6 in Figure below is apparently a critically important feature of the enzyme structure, ensuring that the natural substrate makes no mistake but always binds with the correct ring orientation.

Methotrexate in the binding cleft of dihydrofolatereductase. The yellow regions are Grid contours computed on the unliganded enzyme in Reverse Mode at -4 kcal/mol.

You have now completed you sixth Tutorial. Well done!! We look forwards to hearing from you if we can help in any way.

Latest versions

Login

Username

Password

Register | Lost password?