Aims
The Integral Membrane Proteins group at the SGC aims to solve structures in human membrane proteins, understand the function and biological role of these proteins and to provide tools such as protein purification protocols, proteins for antibody generation and small molecule probes.
A pipeline to produce human membrane protein structures
In the past two years we have developed a clone-to-structure pipeline for the production of human membrane proteins at the SGC. We selected 186 human membrane proteins from a variety of different families, with the aim of developing methods that are generally applicable to human IMPs.
31 Ion channels
9 Gap junctions
14 ABC transporters
52 Solute carrier transporter
26 Enzymes
54 G-protein coupled receptor
All our targets are expresed using the Baculovirus/insect cell exressiion system, 
with a series of N- and C-terminal truncation constructs cloned into two expression vectors with TEV cleavable tags. The vectors have either a 6-Histidine tag on the N-terminus or a 10 Histidine and a Flag tag on the C-terminus. For each protein we make a series of N- and C- terminal truncation constructs, usually between 6 and 20 per protein, based on secondary structure, transmembrane helix and disorder prediction, sequence alignments and the literature. These constructs are cloned into a ligation independant cloning adapted pFastBac vector and bacmids are generated for use as expression vectors in sf9 insect cells. The cloning and screening work is now performed by the SGC's Biotechnology group lead by Dr Nicola Burgess-Brown.
To identify useful constructs we use a small scale purification screening system. This involves extraction of membrane proteins with the detergent DDM, followed by immobilized metal affinity chromatography and gel electrophoresis to identify which constructs are capable of giving soluble protein in the detergent DDM. Successful constructs are then purified again in a series of detergents and any constructs with a reasonable amount of protein produced are then purified on a larger scale and tested on a gel filtration system with in line light scattering and refractive index monitoring, to find conditions where the protein gives a symmetrical gel filtration peak, indicating that the protein is monodisperse. Proteins are then scaled up to the 1 to 10 L volumes of insect cells and purified for crystallisation studies using high throughput nanoscale crystallisation systems. Crystals are screened and optimized using the microfocus beam at the Diamond Light Source Ltd.
From our initial pipeline of 186 proteins we were able to prepare 24 proteins in sufficient quantity to set up crystallisation trials and three have given crystals (as shown in the figure). These crystals come from proteins from the solute carrier, ABC transporter and enzymes families, suggesting that our methods are generally applicable. For one of these targets we have obtained a crystal structure, with data to 2.85A.
Structure of a human mitochondrial ABC transporter
ABCB10 is a mitochondrial transporter which moves its substrate from the mitochondrial matrix into the inter membrane space, across the inner membrane. In common with other ABC transporters it has two nucleotide binding domains (NBDs) which bind and hydrolyse ATP, and two transmembrane domains (TMDs) which form a binding site for the substrate. ABCB10 is a homodimer, having two identical chains with one NBD and one TMD on each chain. In the figure the two blue sections make up one chain, light blue for the NBD and dark blue for the transmembrane section. The second chain is coloured red and pink. We have solved the structure with the ATP analogue AMPPCP and magnesium bound to the NBD, and the resolution of the structure is sufficient to show lipids and detergent molecules associated with the transmembrane helices. Unexpectedly the structure has an 'open inwards' conformation, rather than an 'open outwards' conformation observed for other ABC transporter/nucleotide complexes. For more information on the structure of ABCB10, please follow this link....
