Modeling an Enzyme Active Site using Molecular Visualization Freeware

Visualizing biological macromolecules is a critical skill for students and professionals in the biological sciences. In this protocol, we demonstrate how to model the active site of the enzyme glucokinase using four freely available programs for molecular modeling. This tutorial highlights several steps of the protocol for each program, which includes selecting bound ligands and using the ligands to display amino acids and water molecules within five angstroms.

The supporting information of this manuscript contains a dedicated video for each program, which details all steps of the protocol with further explanation. The structure will model PDBID. 3FGU represents the catalytic complex of the enzyme glucokinase.

The enzyme active site is bound to two of its substrates, beta-D-glucose, which has given the identifier BGC and a magnesium ion, MG.Additionally, this substrate analog, a phospho amino phosphoric acid, adenylate ester, ANP is bound to the glucokinase active site. This non-hydrolyzable analog of adenosine triphosphate, ATP prevents the phosphorylation reaction from occurring, which captures the active site complex pre-catalysis. The interface of the UCSF ChimeraX modeling program contains dropdown menus, a toolbar, the structure viewer and a command line.

We'll begin the protocol with step 1.4, selecting residues within five angstroms to define an active site. To select the ligands press control shift and click on any atom or bond in each of the three ligands. Press the up arrow key until all three ligands are highlighted with a green glow.

Define the selection for future use by clicking in the dropdown menu, select define selector, type ligands for the selection name and click okay. Again, using the select menu, select zone. Toggle this to residues and ensure that the top box is checked.

Click okay. Notice the parts of the cartoon within five angstroms of these ligands are highlighted. To display the side chains as sticks and show the active site water molecules use the actions atom bonds menu to show them.

Or toggle them on with these buttons here. To clear the selection, click anywhere in the empty space. The final result of this protocol should be a model with the active site of the protein and the ligands shown as sticks colored in a contrasting way.

Key polar binding interactions are shown with dotted lines and some of the residues making contacts are labeled. The iCn3D interface contains dropdown menus, the structure viewer and a command log. This view shows the select sets and sequence and annotations popup menus, which appear as the user executes commands that require them.

We'll begin the protocol at step 2.4, selecting residues within five angstroms to define an active site. To select the ligands, use the select dropdown menu and click select on 3D, ensure that residue is checked. Holding down the alt button on a PC or the option button on a Mac, click on the first ligand.

Then press control and click on the remaining two ligans to add them to the selection. Save the selection using the dropdown menu, click select, save selection, input a name and click save. The select sets pop-up menu will now appear with the three ligands selected.

Now select the residues within five angstroms of the ligands. Use the dropdown menu select by distance. In the popup menu that appears, change the second item spear with a radius to five angstroms by typing in the block, click display.

And then close the window by clicking the X in the upper right-hand corner. Save the five angstrom active site by clicking the dropdown menu, select save selection and use the keyboard to input a name. Then click save.

Now create a new selection that combines the two sets. These can be combined in the select sets popup menu. On a PC, control click on the two sets or on a Mac command click.

Again, click the dropdown menu, select save selection and type a new name and then click save. To show interactions such as hydrogen bonds use the analysis menu and select interactions. We'll only be interested in hydrogen bonds and salt bridges here.

So we'll uncheck the rest of these. We'll select three ligands. And for the second set, the residues within five angstroms.

Click 3D display interactions and close the window. This shows some of the residues that are interacting but it doesn't show the entire five angstrom active site. To display that will use the select sets menu again.

Click on five angstrom full and then in the dropdown menu, click style side chains, sticks. To apply CPK coloring, click on the color dropdown menu and click atom. The final result of the protocol should be a model that looks like this with the active site of the protein and the ligand shown as sticks colored in a contrasting way.

Important binding interactions are shown with dotted lines and all of the residues within one of the selections that was created during the protocol are labeled. The Jmol interface contains dropdown menus, a toolbar, the structure viewer, a popup menu and the Jmol console containing the command line. We begin the Jmol protocol at step 3.4, selecting residues within five angstroms to define an active site.

The Jmol console is the best way to select the residues within five angstroms. Type this command to select residues within five angstroms of the three ligands. 193 atoms are selected but these do not represent the full amino acid residues.

To select those, use the typed command, select within(group, selected)and press enter. Notice additional selection halos appear. To show these residues as sticks, right click to bring up the popup menu, hover over style, scheme and then click sticks.

Notice there are still some empty halos here. These are the water molecules in the active site. To select only the water molecules, we can re-execute this command and then modify it.

Click within the console then use the up arrow keys to find that command and click enter to re-execute it. To display the water molecules as atoms, we want to remove the selection of the ligand and the protein. We'll type two commands to do that.

Our ligans are considered hetero groups but water is also considered one. So within this command, we need to define that we're not removing the water. Press enter and now only the water molecules are selected.

Click on the dropdown display menu, hover over atom and click 20%of van der Waals radius. The green magnesium ion is still shown as sticks. More commonly ions are shown as spheres.

Click in the Jmol console and then type select MG and then space fill 50%The ligands inside chains are colored identically. To distinguish them from each other, it's useful to recolor the ligands. In the console, I'll execute a multiline command, which I've copied-pasted from a cheat sheet that I elaborate on in the Jmol supplemental video.

The result of this protocol should be a model that looks like this with the active site of the protein shown as sticks. And the ligands shown us sticks in a softer color scheme. Yellow lines indicate the binding interactions and individual residues are labeled as desired.

The PyMOL interface contains dropdown menus, the structure viewer, the names object panel and the mouse controls menu. The main command line is also labeled in this figure. We begin the primal protocol at step 4.4, selecting residues within five angstroms to define an active site.

To select the ligands click on each one of them. A new selection pops up, which can be renamed by clicking the A button. Using the keyboard, delete the letters sele and type ligans in place of them.

Press enter. We can use this selection to define the area around it. Begin by clicking the A button and select the menu item duplicate.

In this new selection, sel01, click A and select the rename menu item. Using the keyboard, delete the existing letters and type active. This is still selecting our three ligands so we'll have to modify it again, using the A actions button.

Click the button, select modify and expand the selection by five angstroms. Now we've captured the ligands and the residues within five angstroms. The S on the S button stands for show.

Click on this to display the protein in different ways. We'll show this licorice as sticks. Click in the empty space to clear the selection.

This has captured the amino acids within five angstroms but not the water. We can again, duplicate the selection and now modify it to select only the water. In the actions A button, duplicate.

Selection 2 appears. Let's rename this to active water. This time we'll use the A button to modify, around to select the atoms around our selection.

Atoms within four angstroms. This has selected the water but also some atoms of the side chains. To modify this further, use the A button to modify and restrict to solvent.

Now we can see just the water molecules are lit up. Again in the A button, we can apply a preset. Select preset, ball and stick and now the water molecules are shown as spheres.

The final result of the protocol is a model that looks like this with the active site in ligands shown as sticks colored in a contrasting way. Yellow dash lines show polar binding interactions and individual residues are labeled using the selections created in the names object panel. Errors in the execution of the protocol can lead to suboptimal results.

For example, the entire protein being displayed as sticks. To troubleshoot, the user will first need to hide the stick representation for the entire structure. And then redisplay the stick representation for only the object called active using the S button.

Here the models generated using each program are shown side by side. Although there are differences in the display of the enzyme, the same key features and interactions can be seen in each of the four models. A user interested in mastering one of the programs containing a command line want to learn to apply and modify typed commands.

As I alluded to in the Jmol protocol, a useful tool is a command cheat sheet. A plain text file containing frequently used codes to reference. To become an advanced user, it is useful to understand which parts of the protocol can be adapted and modified.

With practice these protocols can be applied to model any enzyme active site of interest.