Characterisation of protein-ligand interactions of glycine binding protein (GBP) as a model of the heteromeric glycine receptor

Pentameric ligand gated ion channels (pLGICs) have an essential role in mammals, insects and some prokaryotes. These membrane receptors convert chemical signals of neurotransmitters into electrical signals of an ion flow, mainly at synapses. Due to this function, the channels are involved in a breadth of biological systems including metabolism, sensory processing and cognition. All the members of the family share a similar structure. A receptor is made up of five subunits. Each subunit has three domains, the extracellular (ECD), transmembrane (TMD) and intracellular (ICD). Neurotransmitters bind in the orthosteric pocket formed at the interface of two ECDs. Ligand binding induces conformational changes that allow ions to flow through a pore generated by the TMD. The ICD is the site of cell- receptor communication. This work focuses on the glycine receptor (GlyR), which has a role, among others, in chronic pain and whose malfunction leads to diseases such as epilepsy and hyperekplexia. In adults, the primary functional oligomer is a heteromeric arrangement of α1β, with other subtypes dispersed throughout the central nervous system (CNS). Current research has produced models of homomeric receptors, however, no structural data are available for the physiologically relevant heteromeric form. An objective of this work was to characterise a new model for the ECD of the heteromeric GlyR, called glycine binding protein (GBP). GBP is based on Aplysia californica acetylcholine binding protein (AcAChBP), a surrogate for the nicotinic acetylcholine receptor (nAChR). Other objectives were the production of another surrogate of a GlyR heteromer utilising models of the homomeric form. Also, the homologous histamine receptor (HisR) from Apis mellifera was investigated to understand how histamine binds and relate this binding to potentially mislabelled receptors in mammals. More detailed descriptions and experimental sections can be found in Chapter 1 and 2. In Chapter 3, the biophysical techniques surface plasmon resonance (SPR), isothermal titration calorimetry (ITC) and tryptophan fluorescence (WF) were assessed utilising AcAChBP and three of its ligands. Comparison of computational docking and thermodynamic parameters led to insights into ligand binding in AcAChBP. Data showed good corroboration between the techniques and highlighted some limitations when considering small molecule interactions. The archetypal ligands strychnine of GlyR and N-methylbicuculline of the GABAA receptor (GABAAR) were utilised to investigate binding in GBP. Structural, computational and biophysical data were used for comparisons. Chapter 4 also presents a structure of the GBP:N-methylbicuculline complex at 2.4 Å and a preliminary model of the AcAChBP:N-methylbicuculline complex at 2.9 Å. These structures emphasise he promiscuous binding properties of N-methylbicuculline and give the first insight into its interaction with GlyR and nAChR. Data from the study of N-methylbicuculline suggest that it is promiscuous due to the flexibility of the orthosteric binding site of pLGIC members as well as the number of interactions the ligand can form, therefore a favourable orientation can always be adopted. Comparisons of structural sequence and binding data for both ligands suggested that GBP may be a good surrogate of the GlyR heteromer. Utilising GBP and AcAChBP as targets, two ligand screens were carried out; an NMR screen of 435 fluorinated compounds and a computational docking screen of 2,000 small molecules from Maybridge. The screens were conducted to discover new ligands of the proteins. Investigation of these ligands could further understanding of protein-ligand interactions with GBP and AcAChBP, as well as the physiological receptors GlyR and nAChR. Similarities in chemical structure and binding poses of compounds that were shown to bind the target proteins were investigated. Validation of binding for selected small molecules was attempted. Two compounds were successfully validated by WF with KD values of 237-393 µM. Exploiting published structures, the possibility to extend the small molecule scaffolds into empty pockets of the two proteins was examined. Extensions of three compounds were carried out suggesting potential group additions that may improve binding interactions. Prior to the initiation of this study, the crystal structures of homomeric GlyR α1 and α3 were reported. It was then considered possible to engineer the sequence to present a heteromeric binding site using a homomeric assembly. Utilising the homomeric channels, more sequence and structural conservation to the GlyR heteromer could be achieved. Therefore, a new surrogate was designed by substituting five residues of loop C of GlyR α3. Loop C acts as a flexible loop that closes upon ligand binding leading to conformational changes, therefore is an important feature for the binding mechanism. Comparison of sequence and structural models of potential HisR constructs was also conducted, which led to the identification of AmHisR. Attempts were made to produce the proteins, however, problems were encountered. Chapter 6 details the approaches attempted to solubilise these membrane proteins. Due to there being no structural studies of HisR, binding models were constructed, and key residues were highlighted. Potentially, the same approach as for GBP could be employed using these models to produce an AcAChBP-based surrogate for HisR.