Interactive Protein Tutorial
PH Domain
PH domains occur in a wide range of proteins that are found in eukaryotic cells. Many of these proteins are implicated in signalling mechanisms. (PH domains take their name from pleckstrin, a protein isolated from platelets, that is a major substrate for protein kinase C). The ball and stick structure on the right, obtained from solution NMR data, represents the pleckstrin homology (PH) domain of β-adrenergic receptor kinase (βARK1, also known as G-protein coupled receptor kinase 2 or GRK2). Click on the the buttons below to view important features of this and other PH domain structures. Here are some questions that you should try to answer.- What forms of secondary structure does this protein possess?
- How are the regions of secondary structure arranged to provide a stable fold?
- How much does the 3D structure depend on sequence?
- To which ligands do PH domains bind?
- Do all PH domains bind this type of ligand?
BACKBONE
Press the button to simplify this wireframe model to show the backbone structure. Notice that an α-helical region can now be seen.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 and some stretches that are recognised as turns.The 7 β strands are arranged in an antiparallel, up-and-down fashion forming a barrel-like structure. A C-terminal α helix "caps" the top of the barrel. These are the minimal elements of a PH domain structure. The connecting loops are variable and they may tolerate small or very extensive insertions.
STABILISATION
The fold is stabilised by the long α helix that is packed tightly to the β structure. Unlike many other β barrel structures there is no space within for a hydrophobic ligand.SEQUENCE SIMILARITY
How similar are the sequences of PH domains? Very few residues are in fact conserved. PH domains have mostly been identified by computer analysis of primary sequences that predict secondary structure.Multiple sequence alignment of PH domains in the proteins: dynamin, phospholipase Cδ1, phospholipase Cδ2, pleckstrin1, β-spectrin, bruton's tyrosine kinase.
STRUCTURAL SIMILARITY
How similar are the 3D structures of different PH domains? They are remarkably similar. Similarities of 3D structure of the PH domain of βARK and the PH domain of PLCδ1Variable regions
Although PH domains share a common structure, some regions are variable. For example inserts may be tolerated in loop regions. In some PH domains the loops on the edge opposite the stabilising a helix can form a binding site for polyphosphoinositide ligands.
INSERTS
Phospholipase Cγ possesses a PH domain with two SH2 and one SH3 domains inserted into a loop between two β strands. Viewed as a linear sequence of amino acids you might conclude that the PH domain has been "split" but in 3D the domain retains its compact form because the inserts are accommodated in loops between β strands.Large insert in β3-β4 loop.
LIGANDS
Some PH domains bind the headgroup of the polyphosphoinositide PI-4,5-P2 (phosphatidylinositol-4,5-bisphosphate). They can also bind IP3 (inositol trisphosphate). This can enable protein-lipid binding.polyphosphoinositides
These amino acid residues do not necessarily interact directly with the ligand. Some are linked via hydrogen bonding through water molecules.
Protein-protein binding
You should note that binding to inositol lipids is not a ubiquitous feature of PH domains - others bind G protein βγ subunits and some have no known ligands.
Now make your own explorations. Restart with a ball and stick model (press the button opposite) and use the built-in JMol menus to learn more about the structural properties of this PH domain.