Subtleties of neuronal receptor behaviour described in new research
Recent work from departmental researchers in collaboration with a group at McGill University in Canada has provided new insights into a critical aspect of neuronal receptor behaviour.
Crystal structure of the wild-type ligand binding domain dimer of kainite receptor GluK2. Two L-glutamate molecules and the anion and cation binding pockets are shown (Click to enlarge)
The work on kainate receptors, a family of glutamate receptors, was carried out by Departmental Lecturer Phil Biggin, postdoc Maria Musgaard, and colleagues in Canada. Published in Nature Structural and Molecular Biology, the findings help to build a picture of how the finely tuned response of these receptors is regulated (1).
Glutamate receptors are found on the membrane of neuronal cells. They control all the fast neurotransmission in the brain and are important for key functions such as memory and learning. The kainate receptor family is one of three principal families of glutamate receptors and is found across different regions of the brain.
Binding of the neurotransmitter glutamate, which is released from one neuron, to kainate receptors on another, allows sodium ions to pass through a channel in the receptor and into the cell, causing depolarisation of the membrane. This generates an electrical current that can be propagated.
To ensure that the ion channel only remains open for a short period of time, the receptor goes into a refractory state known as desensitisation after depolarisation of the membrane. In this state, it is unable to respond to glutamate and must go back to the resting state to do so. All ligand-responsive ion channels exhibit desensitisation but little is known about how this step is controlled.
Researchers have characterised the structure of parts of the kainate receptor in its resting, activated and desensitised states using X-ray crystallography to understand how it responds to glutamate. The overall topology of a similar glutamate receptor is known and a working model for glutamate receptors generated in which the ligand-binding domain (LBD) is arranged as a pair of dimers attached to a transmembrane channel.
In the model, the receptor’s channel is closed at the membrane in its resting state, but opens up when glutamate binds. In the desensitised state, the channel closes but glutamate remains bound.
Attempts to understand desensitisation better have focused on the structural changes that take place in the LBD dimer interface during this shift. Kainate receptors uniquely have binding sites for two cations (sodium) and one anion (chloride) in this interface, and previous work has shown that the type of cation bound can control the speed at which the receptor enters the desensitised state.
Dr Biggin and colleagues wanted to pursue this observation further by investigating the role of the cation in receptor desensitisation. ‘We wanted to make receptor mutants that locked the interface and prevented desensitisation’, he explains. ‘The obvious step was to engineer a disulphide bond across the interface as previously suggested or to do something that would interfere with the sodium binding site.’
Using the expertise of the Canadian collaborators, they made both these mutants – with either double cysteine mutations to generate a disulphide bond or a tethered cation occupying the position of sodium – and tested how these behaved.
Schematic showing one of the kainate receptor dimers (blue and orange) and the proposed mechanism for activation (centre), deactivation and desensitisation. The proposed sequence of glutamate and ion binding during the transitions between these states is shown (Click to enlarge)
Initial experiments looking at lots of receptor molecules all at once in the cells appeared to show that both mutants behaved the same - they could be activated by glutamate but could not be desensitised. But when the researchers looked at the level of a single receptor, they saw a striking difference between the two mutants, explains Dr Biggin.
‘For the tethered cation mutant, the channel opens and current goes through. There is sustained activation and the receptor cannot be desensitised. But with the disulphide bond mutant, the channel behaviour is strange. It just flickers – it is generally closed with the occasional spike corresponding to the open state.’
The difference came as a surprise. ‘We weren’t expecting to see this dramatic difference between the two mutants because we thought they were going to be operating in basically the same way – you are locking the interface and expect to see the same effect.’
The crystal structures, which indicated only minor structural difference between the mutants, provided few clues as to what was happening. So the Oxford researchers turned to molecular simulation to explore the basis for the functional differences observed.
‘When you run the simulations of the crystal structures, the tethered cation mutant behaves as expected – the tethered cation remains bound pretty much the whole time,’ says Dr Musgaard. ‘But if you simulate the double cysteine mutant, the dimer interface relaxes quickly into a more open structure and at the same time, there is a closure of regions which corresponds to a closed channel.’
The simulations provided a structural explanation for the mutants’ different behaviours and highlighted the role of ions in controlling transitions between receptor states. The tethered cation mutant with its continuously occupied ion pocket in the interface is held in an open state. The double cysteine mutant, in which the ions are seen to fall off and the interface relaxes leading to a closed channel, is preferentially closed but there is just enough flexibility that it can open occasionally.
To pin down the role of the ions more precisely, the group carried out further simulations and experimental work, looking at the behaviour of the receptor in the absence of any ions and mutants in which the cation binding site is disturbed but with no structural changes to the interface.
Together, their results demonstrate the importance of the cation binding pocket in the interface in controlling transitions between activated and desensitised states of kainate receptors. Cation binding is necessary for activation of the receptor, but for desensitisation to proceed the cation must fall off.
The key role played by cation binding surprised the researchers who expected that they might find that something in the ligand when bound would have an effect that would eventually lead to desensitisation. Without the molecular simulations, the role of the cation would not have been uncovered.
How might ionic concentration influence signalling of kainate receptors in the brain itself? Dr Biggin says that the question is an intriguing one but points out that it is a big step to relate these findings to what is happening at the level of neurons talking to each other. ‘We need the next level up to knit this all together – someone who works on whole tissue samples to take it to a meaningful physiological context.’