skip to primary navigationskip to content

Research Interest

Biomolecular imaging using atomic force microscopy (AFM)

We study the structure and function of proteins at the single-molecule level using atomic force microscopy (AFM).


AFM imaging. The sample is mounted on mica, and the tip is scanned across it in a raster pattern. A laser beam is bounced off the back of the cantilever and is reflected by a mirror onto a split photodiode. When the tip encounters an object on the surface of the mica, it becomes deflected, causing the position of the laser beam on the photodiode to change. The scanner moves in the Z-dimension to restore the beam to its original position. The XYZ movements of the scanner are analysed by a computer in order to generate the image. Imaging can be carried out either in air or under fluid.








Current projects

The subunit arrangement in ionotropic receptors

Ionotropic receptors are involved in fast synaptic transmission. Most of these receptors are composed of at least two types of subunit. Although the subunit stoichiometry is sometimes known, the arrangement of subunits around the central ion channel is often undetermined. We produce receptors containing epitope-tagged subunits and incubate them with anti-epitope antibodies. AFM imaging of the receptor-antibody complexes enables us to deduce the structures of the receptors. For instance, we have shown that the subunit arrangement within the α4β3δ form of the GABAA receptor is αβαδβ, counter-clockwise when viewed from the extracellular face of the membrane. GABAA receptors containing δ-subunits, although a minor component of the total GABAAreceptor population, have interesting properties, such as an extrasynaptic location and a high sensitivity to GABA. They are also the target for several general anaesthetics, and have been associated with conditions such as epilepsy and pre-menstrual syndrome. They are therefore attractive targets for drug development. Mike talks about this work here (

Ionotropic glutamate receptors are widely distributed in the central nervous system and play a major role in excitatory synaptic transmission. All three ionotropic glutamate subfamilies (i.e. AMPA-type, kainate-type and NMDA-type) assemble as tetramers of four homologous subunits. AFM imaging of receptor-antibody complexes showed that while the GluA1/GluA2 AMPA receptor assembles with an alternating (i.e. 1/2/1/2) subunit arrangement, the GluN1/GluN2A NMDA receptor adopts an adjacent (i.e. 1/1/2/2) arrangement. Hence, the two types of ionotropic glutamate receptor appear to be built in different ways from their constituent subunits. We are currently looking at the assembly of kainate receptors, and at the association of various accessory proteins (e.g. stargazin, cornichons, etc.) with the ionotropic glutamate receptors.





Subunit arrangement in ionotropic glutamate receptors. Double decoration of AMPA and NMDA receptors by subunit specific antibodies. As illustrated in the diagrams, anti-GluA1 antibodies decorate AMPA receptors at ~180°, whereas anti-GluN2A antibodies decorate NMDA receptors at ~90°. Scale bar, 20 nm; height scale, 0-3 nm.









Activation-induced structural changes in ionotropic receptors

We are using used AFM imaging to visualize activation-induced structural changes in ionotropic receptors reconstituted into a lipid bilayer. For instance, in the absence of agonist, the extracellular domain of the GluN1/GluN2A NMDA receptor protrudes from the bilayer surface by 8.6 nm. In the presence of the co-agonists glycine and glutamate, the height of the extracellular region falls to 7.3 nm. Fast-scan AFM imaging, combined with UV photolysis of caged glutamate, permits the detection of a rapid reduction in the height of individual NMDA receptors. The reduction in height does not occur in the absence of glycine or in the presence of the NMDA receptor antagonist D-AP5, indicating that the observed structural change is caused by receptor activation. These results represent the first demonstration of an activation-induced effect on the structure of the NMDA receptor at the single-molecule level. A change in receptor size following activation could have important functional implications, in particular by affecting interactions between the NMDA receptor and its extracellular synaptic partners. We are currently studying the behaviour of the GluN1/GluN3A excitatory glycine receptor.

Fast-scan imaging of NMDA receptor height shifts. Frames were captured at 1 Hz. A single receptor was monitored in buffer containing 100 μM glycine and 100 μM caged-glutamate. UV irradiation occurred at 107 s. A fitted Gaussian curve is overlaid on each histogram.


The interaction of the sigma-1 receptor with ionotropic receptors and ion channels

The sigma-1 receptor is widely expressed in the central nervous system, where it has a neuroprotective role in ischaemia and stroke, and an involvement in schizophrenia. The sigma-1 receptor interacts functionally with a variety of ion channels, including the NMDA receptor. We showed that the sigma-1 receptor binds directly and specifically to GluN1 subunits within GluN1/GluN2A NMDA receptor heterotetramers. This interaction likely accounts for at least some of the observed modulatory effects of sigma-1 receptor ligands on the NMDA receptor.

In collaboration with Dr. Olivier Soriani (University of Nice), we have investigated the interaction between the sigma-1 receptor and the Nav1.5 voltage-gated Na+ channel, which has been implicated in promoting the invasiveness of cancer cells. AFM imaging demonstrates that the sigma-1 receptor binds to Nav1.5 with four-fold symmetry. Hence, each set of six transmembrane regions in Nav1.5 likely constitutes a sigma-1 receptor binding site. Interestingly, two known sigma-1 receptor ligands, haloperidol and (+)-pentazocine disrupt the sigma-1 receptor/Nav1.5 interaction both in vitro and in living cells. At the moment, we are looking at the interaction between the sigma-1 receptor and the hERG voltage-gated K+ channel, and trying to elucidate the mechanisms by which drugs perturb the interactions between the sigma-1 receptor and its ion channel targets.

The sigma-1 receptor decorates Nav1.5 with four-fold symmetry. Images of isolated Nav1.5 particles decorated by one, two, three or four Sig1R particles. Images are 75 nm on each side; height range, 2 nm.



The structure of proteins involved in autosomal dominant polycystic kidney disease (ADPKD)

ADPKD is one of the commonest inherited human disorders, with a population prevalence of over 1:1,000 in all ethnic groups. There are over 50,000 affected and at-risk individuals in the UK, and up to 12.5 million worldwide. The disorder is a leading cause of end-stage renal failure, and accounts for ~6% of patients on renal replacement programmes in the UK. ADPKD is characterised by the progressive loss of normal renal parenchyma secondary to the development of multiple fluid-filled cysts derived from renal tubular epithelial cells. It is caused by mutations in two genes, PKD1 and PKD2, whose protein products, polycystin-1 and polycystin-2 (or TRPP2) form a mechanosensory Ca2+-permeable ion channel complex. This complex transduces extracellular mechanical stimuli via the renal primary cilium and regulates multiple intracellular Ca2+-sensitive signalling pathways.

Polycystin-2 (TRPP2) interacts with other members of the transient receptor potential family, including TRPC1 and TRPV4. We have used AFM imaging to show that both the TRPP2/TRPC1 heteromer and the TRPP2/TRPV4 heteromer have a fixed subunit stoichiometry of 2:2, and an alternating subunit arrangement. Isolated polycystin-1 appears as two unequally-sized ‘blobs’, connected by a ‘string’. The larger blob represents the C-terminal region of the protein, suggesting that the smaller blob at the N-terminus might project from the membrane of kidney tubule cells, sensing fluid flow in the tubule. Co-immunoprecipitation experiments indicate that polycystin-1 interacts with TRPP2. In the presence of TRPP2 the volume of the larger polycystin-1 blob is reduced and the string length is increased, indicating an ‘unravelling’ effect on polycystin-1. Interestingly, this effect requires the expression of active TRPP2, since channel-dead TRPP2 had no effect. We are currently investigating the interactions between polycystin-1 and TRPP2 with fibrocystin/polyductin (FCP), which is also present on the primary cilium (see below).


Structure of polycystin-1. AFM images of isolated polycystin-1 molecules. A schematic illustration of the domain structure of the molecule is shown at the right. Scale bar, 25 nm; height range, 5 nm.









Investigation of the mechanism underlying the interaction of urinary exosomes with the primary cilium

Autosomal recessive polycystic kidney disease (ARPKD), the most common cause of hereditary childhood PKD, is caused by mutations in the gene PKHD1, which encodes the protein fibrocystin/polyductin (FCP). PC1, PC2 and FCP are believed to be present on the primary cilium. Here, the PC1/PC2 complex senses tubular fluid flow. The function of FCP is less clear, but it too complexes with PC2. PC1, PC2 and FCP are also expressed in exosomes, small vesicles present in urine. Exosomes are derived from intraluminal vesicles of multi-vesicular bodies (MVBs), and are released into the renal tubule when the MVBs fuse with the apical membrane of the tubular epithelial cells. They can mediate transfer of protein expression between cells, and they are known to play a crucial role in left-right axis determination in the embryonic node. Exosomes interact specifically with primary cilia. Significantly, exosomes from ARPKD mice decorate cilia much more extensively than exosomes from WT mice, suggesting that the binding is normally followed by fusion or endocytosis, a step that might be inhibited when FCP is mutated, thus accounting for the increased decoration. In a project funded by Kidney Research UK, and in collaboration with Prof. Fiona Karet (University of Cambridge), we are currently studying the mechanisms underlying the interaction of urinary exosomes with the primary cilium.

AFM imaging of the primary cilium. Images of isolated primary cilia that have flattened out on the mica substrate. Scale bar, 500 nm; height scale, 20 nm.


The structure and behaviour of synaptotagmin

Synaptotagmin 1 is the major Ca2+ sensor for membrane fusion during neurotransmitter release. The cytoplasmic domain of synaptotagmin consists of two C2 domains, C2A and C2B. On binding Ca2+, the tips of the two C2 domains rapidly and synchronously penetrate lipid bilayers. In collaboration with Dr. Ed Chapman (University of Wisconsin-Madison), we have used AFM to investigate the forces of interaction between synaptotagmin and lipid bilayers using single-molecule force spectroscopy. C2AB was attached to an atomic force microscope cantilever via a polyethylene glycol linker. With wild type C2AB, the force profile for a bilayer containing phosphatidylserine (PS) had both Ca2+-dependent and Ca2+-independent components. No force was detected when the bilayer lacked PS, even in the presence of Ca2+. The binding characteristics of C2A and C2B indicate that the two C2 domains co-operate in binding synaptotagmin to the bilayer, and that the relatively weak Ca2+-independent force depends only on C2A.

Force of interaction between synaptotagmin and a lipid bilayer. C2AB was attached to an AFM cantilever through a flexible linker. Force curves are shown for C2AB interacting with a PS-containing bilayer in the presence and absence of Ca2+. The force (height of the red triangle) is considerably smaller in the absence of Ca2+.

Free C2AB associates with bilayers in the form of aggregates of varying stoichiometries, and aggregate size increases with increasing PS content. Repeated scanning of bilayers has revealed that C2AB progressively dissociates from the bilayer, leaving behind residual indentations in the bilayer. C2AB binding to bilayers and the formation of indentations is significantly compromised by mutations that interfere with Ca2+ binding to syt or reduce the positive charge on the surface of C2B. We believe that bilayer perturbation by syt might be significant with respect to its ability to promote membrane fusion.

AFM image of synaptotagmin (white particles) in association with a lipid bilayer (orange surface). Between successive images one synaptotagmin particle dissociates, leaving behind an indentation in the bilayer (arrow).

AFM imaging also reveals that C2A and C2B physically interact with one another, and that this interaction is critical to the ability of synaptotagmin to drive neurotransmitter release. For example, the substitution of the flexible linker between the two C2 domains by a rigid 18-proline rod visibly separates the two C2 domains and at the same time almost abolishes the activity of the protein in neurons.

Structure of synaptotagmin. Images of wildtype C2AB and a mutant in which the flexible linker between the two C2 domains has been replaced by a rigid 18-proline rod. Scale bar, 20 nm.




We are currently imaging various other mutants that affect the behaviour of synaptotagmin, to shed light on the relationship between structure and function.

Structure and behaviour of seipin

Disruption of the gene BSCL2 in humans, encoding the protein seipin, causes congenital generalized lipodystrophy with severe insulin resistance and dyslipidaemia. In collaboration with Dr. Justin Rochford (University of Aberdeen), we have used AFM to define the oligomeric structure of seipin and to determine whether disease-causing mutations affect its capacity to oligomerize. We have shown that wild-type human seipin forms oligomers of 12 subunits in a circular configuration but that the L91P and A212P mutants of seipin do not. We suggest that this inability to assemble correctly underlies the pathology caused by these mutations.

Seipin forms dodecamers. Images showing antibody-decorated seipin. Images are 100 nm on each side; height scale, 0-5 nm. Angles between adjacent antibodies are multiples of 30°, indicating a dodecameric assembly state.




We are currently investigating the interactions between seipin and the proteins AGPAT2 and lipin 1, which operate sequentially in the intracellular lipogenesis pathway.


Current lab members

Fahim Kadir

Fahim Kadir
Zhenyu (Cathy) HePia Jeggle
Zhenyu (Cathy) He
Pia Jeggle
Dilshan BalasuriyaMesbah Talukder
Dilshan Balasuriya Mesbah Talukder