Interactive Protein Tutorial
Sequence and structure in kinase domains
Kinase domains occur in a wide range of proteins and catalyse the transfer of phosphate, usually from ATP, to proteins. Amongst these proteins a basic architecture is strongly conserved even though there may be significant variation in the primary amino acid sequence. The similarities conserve the basic function and the difference confers specificity of activation, inhibition, localisation and target substrate. To reveal the similarities we will first study one protein kinase in detail, the catalytic subunit of cAMP-dependent protein kinase (PKA). You can find more information on this molecule at the PDB site under code 1CDK.
You must use your lecture notes and your assigned essential reading to get the most from this tutorial. Click on the the buttons below to view important features of this and other kinase domain structures.
BACKBONE
You can use this button to simplify the wireframe model of the PKA catalytic subunit by showing only the alpha-carbons of the polypetide backbone. You should be able to identify some helical regions. You can stop and re-start the rotation as you like throughout this tutorial. Remember that your mouse can be used to rotate the structure manually and to zoom in and out (plus shift button). The right hand button controls other features but use these only when you have passed through this tutorial at least once.SECONDARY STRUCTURE
This is shown more clearly by a ribbon diagram. The computer calculates where regions of secondary structure occur and draws them as ribbons. The α-helical region is now clearly defined and there are also regions of β-structure. Stop the rotation and tilt the model using your mouse until it looks like the images from your lectures (try this example).NON-PEPTIDE REGIONS
With this button you can examine those elements of the structure that are not composed of amino acids. All of these hetero-atoms are shown in spacefill for contrast. A group representing the N-terminal myristate is shown in uniform grey. Note that this group is tucked into a deep crevice between helices and loops. The non-hydrolysable analogue of ATP (AppNHp) is the larger molecule coloured in "cpk" format and which fits into a deep cleft in the secondary structure. The smaller hetero-group, coloured in cpk, is a phosphate group. You should note that this phosphate sits in a loop near the cleft occupied by ATP. You will also be able to see which parts of the ATP molecule are phosphates because they are coloured the same as the other phosphate group..You can switch the hetero-atoms off now to avoid confusion. They will re-appear later.
The linker
In this next section we will examine the linker region, a key feature of protein kinases. In PKA this is a long N-terminal alpha helix of 21 amino acid residues. You should be able to identify this secondary structural feature in the primary sequence. This linker may also be referred to the as the A-helix.
Show the linker helix alone
You will see that this helix lies along the surface of the protein. You will also see that it reaches almost from the "top" of the molecule to the "bottom"
Here you can simplify the structure for clarity
By highlighting the other secondary structural elements and the hetero groups you should still be able to figure out which helix is the linker and be ready to identify it later.
In these next two sections we will divide the kinase into two lobes, one small and one large.
The small lobe
Here you can display the linker and the small lobe.
By highlighting the secondary structure you will be able to see that the small lobe contains a five stranded, anti-parallel beta-sheet. Some of the beta strands are linked by loops and turns whilst others are bridged by a larger unit which contains two helices, B and C, within it. Remembering that the linker or A helix is N-terminal take some time now to work out which helix is B and which is A. This will be important later.
You can add the ATP-like molecule now. Note that this lies very close to two strands of the beta sheet where it makes specific contacts with amino acid side chains. This is the glycine-rich loop. You will meet this loop again later. Some of the structure in this region and the contacts are so specific that they can be used to regulate kinase activity, and because they are highly conserved, to identify likely protein kinases from amino acid sequences.
By filling in much of the remainder of the protein you can see that the beta-sheet of the small lobe defines the "top" half of the deep cleft into which ATP fits.
The large lobe
By subtracting the small lobe we are left with the large lobe which obviously forms the other, "bottom" half of the ATP binding site.
As you can see this lobe is principally composed of alpha-helices with a small, two stranded beta-sheet. Once again the beta-sheet is in close proximity to the ATP molecule but particularly with the purine ring and the ribose residue.
The core of this large lobe is formed from a very common protein structural element, a four helix bundle (see Branden and Tooze)
You can see that this densly packed structure forms a compact core for the large lobe.
The whole kinase
Now that you have identified parts of PKA take some time to check that you can distinguish them amongst elements of the complete strucuture.
We can now revert to the backbone model of the large lobe and clearly identify the four helix bundle again.
Now spacefill the whole kinase domain and see if you can spot the groove or cleft between the two lobes.
You can toggle to a simpler model with this button. You need to identify this cleft because it contains the entry for the substrate peptide and the ATP and also serves as the route out for the phosphopeptide and ADP products. You can toggle back to the previous view to demonstrate this. For clarity the ATP analogue will be left in situ whilst you do this.
Phosphorylation of Threonine 197
In the next sections of the tutorial we will pay close attention to the key role of this phosphate and other residues that make up the active site of the kinase. The loop that has been phosphorylated is called the P-loop in PKA and the activation loop in other kinases. This modification is a key feature of protein kinase regulation because it is used to co-ordinate and to orientate other key amino acid side chains in the protein. You can reveal these with this button, You may want to use your mouse to zoom in.
The active site
With this button you can view the secondary structure in the model and replace the ATP-like molecule. You will also see the phosphate group that we mentioned earlier (phosphorylated threonine at 197, pT197). This time the whole amino acid is included and you can see that it forms part of a loop at the edge of the groove through which substrates enter and products leave.
You can subtract a lot of the structure with this button. You will now be able to see two amino acids that bond to this phosphate. One is a histidine (H87) from the C-helix of the small lobe. You can toggle back to the second part of the section on the small subunit to identify this helix more clearly. The second amino acid is an arginine (R165) from another loop called the catalytic loop. The residues in the catalytic loop must be correctly positioned for efficient kinase activity and an essential contribution to this alignment is provided by the arginine to phosphate interaction. Also note the reverse-turn (beta 1-beta 2) structure from the larger beta-sheet in the small subunit. This is the glycine rich loop that co-ordinates phosphates the beta phosphate of ADP or ATP.
Also note that a small "beta-sheet", in conjunction with a similar structural motif from the large subunit forms a small "sandwich" between which ADP or ATP will fit. The purine ring is "sandwiched" by hydrophobic interactions with side chains from both the small and large lobe. The loop connecting this large subunit beta-structure to the phosphorylated threonine (T197) is the magnesium-binding loop. This loop co-ordinates a magnesium ion which is in turn co-ordinated to the phosphates of the nucleotide. Consequently, the polyphosphate tail of the ATP is positioned correctly for efficient catalysis.
The magnesium ion (blue) can be revealed with this button. You will also have revealed an additional strand of the small subunit beta-sheet along with the B-helix. This strand contains a key lysine (K72) residue that binds to the alpha and beta phosphates of ATP and to a glutamate residue (E91) in the C-helix. Make sure that you can identify these two residues. To digress a little from PKA, other kinases e.g. cyclin dependent kinase 2, this local structure is formed only in the enzyme is activated by binding to cyclin. Cyclin forces the correct rotation of the C-helix, called the PSTAIRE helix in cdk2, to put the glutamate equivalent to E91 (the E in PSTAIRE) in the correct position. Therefore in the absence of cyclin the catalytic site is poorly "constructed" and the enzyme is an inefficient catalyst. You should also note that cyclin causes the activation loop to be exposed and phosphorylated and this change also contributes to the restructuring of the active site for efficient catalysis (see above).
The peptide substrate
We will now examine a substrate peptide similar to one to which the gamma-phosphate of ATP would be transferred in nature. However, in this model the peptide is a pseudo-substrate and it does not contain an hydroxyl-containing amino acid to accept the phosphate, this has been swapped for an alanine residue.
Add the pseudo-substrate into the model.
We can now highlight this alanine residue and see its proximity to the terminal phosphate of the ATP. We can now see clearly that during catalysis the phosphor-transfer reaction would only require the bond on one side of the gamma phosphate to break and re-form on the other side to the phospho-acceptor residue (here replaced by dysfunctional alanine). This merely requires the efficient movement of electrons since they are extremely close together for efficient transfer. Now that we have seen fine details of the active site we can fill in the rest of the kinase molecule to see this detail in context.
In conclusion
Use these buttons to add and subtract more structural detail and see how much of the active site and the rest of the molecule you can still recognise. Obscuring parts of the structure have not been space filled so that you can view the interior of the protein around the active site.
Button 1
Button 2
Button 3
Button 4
Button 5
Rendering Options: (if Jmol isn't working, try JSmol!)