Molecular Pharmacology of an Insect GABA Receptor A dissertation submitted to the University of Cambridge for the degree of Doctor of Philosophy Ian Vincent McGonigle (BA) King’s College Cambridge 2010 2 Preface The work presented here in this dissertation was carried out in the Department of Biochemistry at the University of Cambridge between October 2007 and September 2010. All of the work was carried out by the author except where otherwise stated in the text. This dissertation has not been submitted, in whole or part, for a degree or diploma at any other university. 3 Acknowledgements There are so many people to whom I owe a great deal of gratitude. To begin with I must thank the Medical Research Council UK for funding my studentship. Without financial support this could never have happened. However it is Sarah Lummis, my PhD supervisor, who has really made this possible. From when I first visited Cambridge to interview until my dissertation was written - some three and a half years later - Sarah has been a tremendous support. More than anybody else I owe her a huge thank you; thank you for everything Sarah. On a daily basis my project was supported and guided by Andy Thompson. Andy is an exceptional scientist and his electrophys expertise, his good company and his enthusiasm have made my PhD a truly enjoyable experience; thanks Andy. In moments of molecular biology crisis Kerry Price has always been there with her enthusiasm, knowledge and advice; thanks Kerry. No better person to be sitting beside! Jamie Ashby, with his knowledge of structural biology and computer modelling has also been a pleasure to work with. In the lab, Mariza and Linda deserve mention and thanks. They were always there to answer my questions about practical things and their knowledge and experience were greatly valued. The many students and visitors who have passed through ‘Skylab’ have enhanced the experience of my PhD. Particular mention goes to Kat, David and Sita, who have all been great colleagues. Nick from computer support also deserves thanks. How many times did I have some nightmare of a problem with my computer that he could fix in only a few minutes? Chris and Tom from the photography team also deserve mention for their help with poster and dissertation printing; thanks guys. With Sarah’s kind support and encouragement I was fortunate enough to have made a trip to Argentina, where in Cecilia Bouzat’s lab I learned single-channel electrophysiology. Cecilia was a fantastic hostess. From the moment she picked me up at the Bahía Blanca bus station, bleary-eyed after an overnight bus ride from Buenos Aires, until the moment she saw me fly off to Iguazú, she was wonderful; thank you so much for welcoming me into your lab and into your home. Muchas gracias. In the lab Jeré was a fantastic teacher and great company; thanks for everything Jeré. To Marie-José and Diego, thank you both for entertaining me at the 4 weekends and showing me such wonderful places as Monte Hermoso and Villa Ventana. I really enjoyed those days with you. I loved Argentina, so thank you all for making this possible. My time in Cambridge has been hugely shaped by my time spent at King’s College. King’s has been my home for the past three years and my experiences there will remain unforgettable. Bert Vaux, the graduate tutor at King’s has been a fantastic support and deserves particular mention; thanks Bert for getting me to Thailand for the Fireball world sailing championships, thank you for all of your help along the way and thanks for your friendship, great company and conversation. I have made so many good friends at King’s. Ross, Luke, Dom and Tom, you guys are heroes of the highest order. To Ruben, Ben, Cornelius, Tom and Dom, playing in the band with you was awesome; I owe you all so much for the enjoyment you have given me. To the rest of my good friends at King’s (you know who you are!), thanks guys. To John, André and Shayne at the bar, you guys have been great. Because of your good company and chat I always felt right at home in King’s. To Sandra, you have been truly amazing! Thank you for introducing me to such wonderful things as Serbian coffee, sleeping till noon and to a generally relaxed attitude to life! You are the best! Hats off to Savino’s, Bella Napoli and Manna Mexico. Without such wonderful lunches and coffee I would never have had the energy to get this dissertation written. Furthermore, late night writing couldn’t have been done without Gardie’s and the ‘Van-of-Death’ providing much needed fuel, so thanks to you guys too. In terms of travel, weekend trips home to Dublin wouldn’t have been possible without Ryanair, so thanks to Michael O’Leary for making Irish flights cheaper than British trains!!! Final words of thanks go to my parents, Al and Ben, for teaching me the importance of education and supporting me all the way. Thank you all….. 5 Contents Page Chapter 1. Introduction 1.1 Neurotransmission 17 1.2 The ligand-gated ion channel superfamily 17 1.3 Cys-loop receptors 18 1.4 GABA 19 1.5 GABAA receptor structure 20 1.6 Receptor activation 24 1.6.1 Gating 24 1.6.2 Interface residues 28 1.6.3 Kinetic schemes 30 1.7 Ligand binding site 32 1.8 Insect GABA receptors 34 1.9 RDL receptors 37 2.0 Thesis Aims 43 Chapter 2. Materials and Methods 2.1 Suppliers and addresses 45 2.2 Subcloning of rdl into pGEMHE 46 2.3 Preparation of mRNA 46 2.4 Oocyte preparation 47 2.5 Electrophysiological recordings 47 2.6 Sequence alignment 49 2.7 Modelling 49 2.8 Ligand docking 50 2.9 Quantum mechanical ab-initio calculations for ligands 51 2.10 Site directed mutagenesis 51 2.10.1 Primer design 51 2.10.2 PCR 52 2.10.3 Preparation of electrocompetent E.Coli 53 2.10.4 Transformation of DH5α cells 53 2.10.5 Plasmid minipreps and restriction digests 54 6 2.11 Assessment of antagonists using TEVC 54 2.12 Mutant cycle analysis 54 2.13 Whole insect bioassays 55 2.14 Culture of HEK293 cells 55 2.15 FlexStation recording 56 2.15.1 HEK293 transfection 56 2.15.2 Recording 56 2.16 Single-channel electrophysiology 56 Chapter 3. Biophysical properties of RDL receptors 3.1 Introduction 59 3.2 Results 61 3.2.1 Expression of RDL receptors in Xenopus oocytes 61 3.2.2 RDL expression rate 64 3.2.3 Ionic selectivity of RDL receptors 66 3.2.4 pH sensitivity of RDL receptors 68 3.3 Discussion 70 3.4 Conclusion 73 Chapter 4. Molecular characterisation of agonists that bind to RDL receptors 4.1 Introduction 75 4.2 Results 77 4.2.1 Functional responses 77 4.2.2 Computational ligand analysis 81 4.2.3 Homology modelling and docking 83 4.3 Discussion 86 4.4 Conclusion 89 Chapter 5. Investigating the GABA binding site of RDL receptors 5.1 Introduction 91 5.2 Results 94 5.2.1 Molecular biology 94 5.2.2 Loop B mutants 95 5.2.3 Loop C mutants 100 7 5.2.4 Loop D mutants 101 5.2.5 Loop A mutants 103 5.2.6 Probing binding site mutations using the gain of function mutant L314Q 105 5.3 Discussion 113 5.4 Conclusion 119 Chapter 6. Characterisation of Ginkgo biloba extracts on RDL receptors 6.1 Introduction 121 6.2 Results 123 6.2.1 Ginkgolide A, ginkgolide B and bilobalide are antagonists of RDL receptors 123 6.2.2 Mutant receptors are resistant to antagonists 125 6.2.3 Mutant cycle analysis 127 6.2.4 Molecular modelling and docking 128 6.2.5 Insect bioassays 131 6.3 Discussion 132 6.4 Conclusion 135 Chapter 7. Single-channel analysis of heteromeric 5-HT3 receptors 7.1 Introduction 137 7.2 Results 140 7.2.1 Single-channel recordings 140 7.2.2 Flexstation analysis of heteromeric 5-HT3 receptors expressed in HEK293 cells 142 7.3 Discussion 145 7.4 Conclusion 147 Chapter 8. Future directions and final remarks 8.1 Future directions 149 8.2 Final remarks 152 References 153 8 List of Tables and Figures Chapter 1. Introduction Page Table 1.1 Pharmacology of insect GABA receptor subunits 36 Fig. 1.1 Chemical structure of GABA (γ-amino butyric acid) 20 Fig. 1.2 ClustalW alignment of a selection of Cys-loop receptors 21 Fig. 1.3 Schematic of a Cys-loop receptor 22 Fig. 1.4 Structure of a Cys-loop receptor 23 Fig. 1.5 Examination of the pore structure of GLIC and ELIC 27 Fig. 1.6 The interface between extracellular and transmembrane domains 29 Fig. 1.7 Kinetic scheme for Cys-loop receptor activation 31 Fig. 1.8 Extracellular domain of a Cys-loop receptor 33 Fig. 1.9 Dendrogram of insect GABA receptor subunits 36 Fig. 1.10 Alternative splicing of rdl 38 Fig. 1.11 Distribution of RDL receptors in the adult house cricket 40 Fig. 1.12 Model of the extracellular domain of RDL 41 Chapter 2. Materials and Methods Table 2.1 Oligonucleotide primers used 52 Fig. 2.1 Standard TEVC setup 48 Chapter 3. Biophysical properties of RDL receptors Table 3.1 Parameters derived from concentration- response curves 63 Fig. 3.1 AflII XbaI excision of rdl from pcDNA3.1 61 Fig. 3.2 Electrophysiological traces from wild-type RDL receptors 62 Fig. 3.3 Concentration-response curves for RDL receptors 63 Fig. 3.4 RDL EC50 current amplitudes over 20 min 64 Fig. 3.5 Maximal RDL GABA currents over 72 h 65 Fig. 3.6 The Goldman-Hodgkin-Katz Equation 66 Fig. 3.7 Voltage-reversal experiments for RDL receptors 67 Fig. 3.8 pH of an aqueous solution of GABA 68 Fig. 3.9 pH changes on the RDL current amplitude 69 Fig. 3.10 Alignment of M2 regions of RDL and other Cys-loop receptors 71 9 Chapter 4. Molecular characterisation of agonists that bind to RDL receptors Table 4.1 Parameters derived from concentration-response curves 80 Table 4.2 Dipole separation distances of GABA analogues 82 Table 4.3 Hydrogen bonding partner residues 84 Fig. 4.1 Chemical structures of GABA analogues 78 Fig. 4.2 Electrophysiological traces and concentration-response curves 79 Fig. 4.3 Docking of GABA and active analogues 85 Chapter 5. Investigating the GABA binding site of RDL receptors Table 5.1 Table 5.1 Parameters derived from concentration- response curves 104 Fig. 5.1 Binding site model 93 Fig. 5.2 Gel screen showing PCR product 94 Fig. 5.3 Analysis of colonies containing mutant and wild-type DNA 94 Fig. 5.4 Concentration-response curves for F206 mutants 96 Fig. 5.5 Concentration-response curves for Y208 mutants 98 Fig. 5.6 Mutant cycle analysis for mutants Y208F, F206Y and double mutant F206Y.Y208F 99 Fig. 5.7 Concentration-response curves for mutant Y254F 101 Fig. 5.8 Concentration-response curves for mutant Y109F 102 Fig. 5.9 Concentration-response curves for mutant F146A 103 Fig. 5.10 Electrophysiological traces for L314Q & F206Y.L314Q mutants 106 Fig. 5.11 F206Y.L314Q open channel proportions 107 Fig. 5.12 Distance between loop B F206 and Loop E S176 108 Fig. 5.13 Change in binding energy for the F206Y mutation 110 Fig. 5.14 Y254F.L314Q open channel proportions 111 Fig. 5.15 Loss in binding energy for Y254F mutation 112 Fig. 5.16 Electrostatic polarity of GABA within the binding site residues 115 Fig. 5.17 Aromatic box 116 Fig. 5.18 Loop A aromatic residues F146 & F147 118 Chapter 6. Characterisation of Ginkgo biloba extracts on RDL receptors Table 6.1 Parameters derived from concentration-response curves 125 Table 6.2 Parameters derived from concentration-inhibition curves 127 10 Fig. 6.1 Structures of pictotoxin and ginkgolides 122 Fig. 6.2 Ginkgolides are antagonists of RDL receptors 123 Fig. 6.3 Recovery following ginkgolide inhibition 124 Fig. 6.4 Concentration-response curves for mutant receptors 125 Fig. 6.5 Ginkgolide inhibition of mutant receptors 126 Fig. 6.6 Mutant cycle analysis 128 Fig. 6.7 Docking simulations 130 Fig. 6.8 Whole insect toxicity bioassays for GA, GB, BB and PTX on N.lugens 131 Chapter 7. Single-channel analysis of heteromeric 5-HT3 receptors Table 7.1 Parameters derived from concentration-response curves 144 Fig. 7.1 5-HT3R alternative transcripts 138 Fig. 7.2 Four intracellular arginine residues located in the M2-M3 intracellular loop confer high conductance to the B subunit 139 Fig. 7.3 Sample traces and amplitude histograms for single-channel 5-HT3 AB and ABr1 receptors expressed in HEK293 cells 141 Fig. 7.4 Current-voltage (IV) relationships for AB and ABr1 channels expressed in HEK293 cells 142 Fig. 7.5 Concentration-response curves from 5-HT3 receptors expressed in HEK293 cells 143 11 Abbreviations 3-APP 3-aminopropylphosphonic acid 4-AB 4-amino-1-butanol 5-AV 5-aminopentanoic acid 5-HT 5-hydroxy-tryptamine aa Amino acid Å angstrom (1×10−10 M) ACh Acetylcholine AChBP Acetylcholine binding protein APS Ammonium persulfate BB Bilobalide cAMP Cyclic adenosine monophosphate cDNA Complementary deoxyribonucleic acid CNS Central nervous system cRNA Complementary ribonucleic acid ddH2O Double distilled water DEPC Diethylpyrocarbonate DMEM Dulbecco’s modified eagle’s medium DNA Deoxyribonucleic acid EBOB 4-n-[3H]propyl-4'-ethynylbicycloorthobenzoate EC50 Concentration of agonist at which half IMAX is achieved ECD Extracellular domain E.coli Escherichia coli EDTA Ethylenediaminetetraacetic acid ELIC Erwinia chrysanthemi ligand gated ion channel EPSP Excitatory post-synaptic potential GA Ginkgolide A GABA γ-amino-butyric-acid GABAAR γ-amino-butyric-acid type A receptor GABABR γ-amino-butyric-acid type B receptor GABACR γ-amino-butyric-acid type C receptor GB Ginkgolide B GC Ginkgolide C 12 GHB γ-hydroxy-butyric-acid GLIC Gloeobacter violaceus ligand gated ion channel GlyR Glycine receptor GRD Glycine-like receptor of Drosophila HEK293 Human Embryonic Kidney 293 cells IMAX Maximal current IPSP Inhibitory post-synaptic potential LB Luria-Bertani LCCH3 Ligand-gated chloride channel homologue 3 LGIC Ligand-gated ion channel M (1-4) Transmembrane spanning domain (1-4) MD Molecular dynamic MOD-1 C.elegans Cys-loop receptor Modulation Of locomotion Defective 1 mRNA Messenger ribonucleic acid nH Hill coefficient NR No functional response detected o/n Overnight PABA Para-amino benzoic acid pcDNA3.1 Mammalian expression vector PEG Polyethyleneglycol pGEMHE Xenopus oocyte expression vector PTX Picrotoxin RDL Resistant to Dieldrin RT Room temperature SEM Standard error of the mean SCAM Substituted cysteine accessibility mutagenesis ssDNA Single-stranded DNA TACA 4-amino-2-butenoic acid TBPS t-butylbicyclophosphorothionate TEVC Two electrode voltage clamping THIP 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol TMD Transmembrane domain UV Ultra-violet WT Wild-type ZAC Zinc-activated channel 13 Amino acids A (Ala) alanine M (Met) methionine C (Cys) cysteine N (Asn) asparagine D (Asp) aspartate P (Pro) proline E (Glu) glutamate Q (Gln) glutamine F (Phe) phenylalanine R (Arg) arginine G (Gly) glycine S (Ser) serine H (His) histidine T (Thr) threonine K (Lys) lysine V (Val) valine I (Ile) isoleucine W (Trp) tryptophan L (Leu) leucine Y (Tyr) tyrosine 14 Summary Cys-loop receptors are ligand-gated ion channels that are involved in fast synaptic neurotransmission in the central and peripheral nervous system. The Cys-loop receptor RDL (‘resistant to dieldrin’) is a GABA-gated chloride channel from Drosophila melanogaster and is a major target site for insecticides. The aim of this dissertation was to characterise RDL receptors with particular focus on the agonist binding site. To assess the potency of a range of GABA analogues on RDL receptors, I expressed receptors in Xenopus oocytes and used voltage-clamp electrophysiology to detect receptor responses. I carried out computational modelling of these analogues to determine the dipole separation distances and atomic charges. Computational calculations and functional experiments revealed that agonists require a charged ammonium and an anionic centre, with the most potent agonists having a dipole separation distance of ~5 Å. I made a homology model of the extracellular domain of RDL and docked the active analogues into the putative binding site. I then conducted mutagenesis studies to test the accuracy of this model. Functional data from mutagenesis studies broadly support the location of GABA within this model. This model may be useful for further structure−activity studies and rational drug design. Natural compounds from the traditional Chinese medicine ‘Ginkgo biloba’ (ginkgolide A, ginkgolide B and bilobalide) have potent insecticidal properties and are similar in structure to picrotoxin. I tested the effect of these compounds on RDL receptor function using voltage-clamp electrophysiology. All compounds were found to inhibit RDL receptor function. I probed the binding site of these compounds using site-directed mutagenesis and electrophysiology. Mutations to the 2'A and 6'T channel-lining (M2) residues greatly reduced the potency of these compounds. I then made a homology model of the transmembrane domain of RDL and docked these compounds into the channel. Compounds docked into the channel pore close to the 2' and 6' channel-lining residues and H-bonding interactions were detected at these locations. 15 Ginkgolides are therefore antagonists of RDL receptors, binding in the channel close to the 2' and 6' residues and this may be the mechanism underlying their potent insecticidal properties. The 5-HT3 receptor is a member of the Cys-loop receptor family and shows homology to RDL receptors. To explore different techniques for studying Cys- loop receptor function I assessed the functionality of two brain derived transcripts of the 5-HT3B subunit (Br1 and Br2) using single-channel electrophysiology and a fluorometric assay. Receptors containing Br1 were found to have a conductance identical to the 5-HT3B subunit whilst Br2 receptors were found not to be expressed. This finding has implications for 5- HT3 brain signalling, in which Br1 may play an important role. In conclusion, work here has described how agonists bind to and activate RDL GABA receptors and I have identified a candidate mechanism for the potent insecticidal properties of Ginkgo biloba extracts. I have also confirmed that 5- HT3 receptor brain transcript Br1 forms functional channels with similar properties to the 5-HT3B subunit. 16 Chapter 1 Introduction ______________________________________________ 17 Introduction 1.1 Neurotransmission Neurons convey information to one another through the use of both electrical and chemical signals. Neurons pass an electrical signal down the length of their axons by means of voltage-gated ion channels which open in response to local changes in membrane potential. The passing of this signal along the axon is known as an action potential. Action potentials can trigger the release of vesicles containing neurotransmitters from the pre-synaptic membrane. Neurotransmitters propagate the signal that was generated as an electrical action potential as a chemical signal by binding to their post-synaptic receptors and eliciting the opening of ion channels. These ligand-gated ion channels can contribute to either excitatory post-synaptic potentials (EPSPs) or inhibitory post-synaptic potentials (IPSPs). Ligand-gated ion channels allow the conversion of the chemical signal transmitted by neurotransmitters to a change of post-synaptic potential on a millisecond timescale, thus facilitating fast synaptic transmission. 1.2 The ligand-gated ion channel superfamily The “ligand-gated ion channel” (LGIC) family of membrane proteins refers to a group of structurally related receptors that consist of a ligand-binding domain, and an intrinsic ion-channel domain. Fast synaptic neurotransmission is mediated by these LGICs, which open in response to the binding of neurotransmitter molecules. The three major classes of LGICs are ATP gated ion channels (P2X receptors), the family of glutamate gated-ion channels (NMDA, AMPA and kainate receptors) and the Cys-loop family of LGICs (Hogg et al., 2005). Intracellular second messenger activated receptors such as NAADP (nicotinic acid adenine dinucleotide phosphate), ryanodine and IP3 activated receptors - which are involved in the release of intracellular calcium stores - are also ligand-gated ion channels in this superfamily (Taylor and Laude, 2002). These types of LGICs are distinct from – and should not be 18 confused with – other ion channels which can be activated by distinct ligand- activated receptors; for example, the GABAB receptors are G-protein coupled receptors which can (indirectly, through a small G-protein) activate potassium channels (See Ong and Kerr (2000) for a review). Whilst GABAB and GABAA receptors are the two major classes of vertebrate GABA receptors, from here onwards we will focus on the Cys-loop receptor family of LGICs, with special focus on GABAA receptors, which are pentameric intrinsic ion channel receptors. 1.3 Cys-loop receptors Members of the Cys-loop receptor family share amino acid sequence homology. Receptors include the nicotinic acetylcholine receptor (nAChR), the γ-amino-butyric-acid type A receptor (GABAAR), the glycine receptor (GlyR) the serotonin type 3 receptor (5-HT3R) (Barnard, 1992) and the more recently discovered zinc-activated channel (ZAC) (Davies et al., 2003; Hogg et al., 2005). Cys-loop receptor subunits are approximately 420-600 amino acids in size with a mass of 50-70 kDa and different members of the family share approximately 30% sequence identity. Cys-loop receptors can be distinguished by their different sensitivity to agonists and their intrinsic ionic selectivity. Both nAChR and 5-HT3 receptors select cations, which usually results in EPSPs favouring the generation of an action potential in the post-synaptic neuron. The GABAAR and GlyR usually mediate post-synaptic inhibition of neurotransmission by generating IPSPs via the influx of anions into the post-synaptic neuron. The direction of ion flux through channels is dependent on the local electrochemical gradient and there are cases where anion channels can in fact be excitatory; in the developing neocortex for example, the activation of GABAA channels results in chloride efflux, leading to the generation of EPSPs (Staley et al., 1995). For the most part, however, GABAA receptor activation results in a hyperpolarisation of the post-synaptic neuron which disfavours the opening of voltage-gated ion channels and the 19 generation of an action potential. Glycine receptors dominate motor neuron inhibition, whilst GABA is the major inhibitory neurotransmitter in the brain. Cys-loop receptors have also been identified in insects. For example, recent studies in the fruit fly (Drosophila melanogaster) (Bocquet et al., 2009) honey bee (Apis mellifera) and red flour beetle (Tribolium castaneum) (Jones and Sattelle, 2007; 2006: Littleton and Ganetzky, 2000) have identified Cys-loop receptors, including nAChRs (Sattelle et al., 2005), anion and cation GABA- gated channels (Buckingham et al., 2005) and glutamate- and histamine-gated anion channels (Gisselmann et al., 2002; Stein et al., 2003; Zheng et al., 2002). Cys-loop receptors have also been identified in other species of animal; in the small crab (Cancer borealis), GABA- and glutamate-gated anion channels have been identified (Duan and Cooke, 2000; Swensen et al., 2000). Additionally, the nematode Caenorhabditis elegans possesses a wide range of Cys-loop receptors, with 90 ligand-gated ion channel genes identified to date (Bargmann, 1998). These include the receptors EXP-1, a GABA-gated cation channel that mediates enteric muscle contraction (Beg and Jorgensen, 2003) and MOD-1, a serotonin-gated anion channel (Ranganathan et al., 2000). A review of the Cys- loop receptors of Caenorhabditis elegans has been published recently by Jones and Sattelle (2008). More recently, Cys-loop receptor orthologues have been identified in the bacteria Gloebacter violaceus and Erwinia chrysanthemi, including a proton-gated anion-channel (GLIC) and a putative anion-channel (ELIC), for which an agonist has yet to be identified (Bocquet et al., 2009: Hilf and Dutzler, 2009; 2008). These findings confirm the prevalence of Cys-loop receptors in prokaryotes. 1.4 GABA GABA has long been known to exist in plants and bacteria, where it serves a metabolic role in the Krebs cycle; it is synthesized by decarboxylation of glutamate by the enzyme glutamic acid decarboxylase. In the 1950s GABA was discovered to be a free amino acid in the brain (Roberts and Frankel, 1950) (Fig. 1.1) and GABA was accepted as a neurotransmitter following the observation that GABA application inhibits Ascaris body muscles by opening 20 chloride channels (Del Castillo et al., 1963). It is now known that these actions of GABA are mediated by a GABA-gated chloride channel, the GABAA receptor (Schofield et al., 1987). In vertebrate neurons the activation of GABAA receptors usually permits the diffusion of chloride ions into the cell, hyperpolarising the membrane and decreasing the excitability of the cell. Figure 1.1 Chemical structure of GABA (γ-amino-butyric acid). GABA is a carboxylic acid with a pKa of 4.03. At typical physiological pH, the amino group is protonated and the carboxylate carries a negative charge (Huxtable et al, 1987). 1.5 GABAA receptor structure GABAA receptors are ionotropic ligand-gated ion channels consisting of five subunits which can be assembled with varying stoichiometries. To date, 19 mammalian GABAAR subunits have been isolated by cDNA cloning; 6α, 4β, 3γ, 1δ, 1ε and 3ρ subunits. For distinction, receptors consisting of only ρ subunits are referred to as GABACRs, although these receptors are technically a sub-class of GABAARs (Barnard et al., 1998). The α1 subunit is the most abundant isoform in the adult mammalian brain (Luque et al., 1994) whilst α2 and α3 subunits are found predominantly in the spinal cord (Persohn et al., 1991). β subunits co-express with α and γ subunits, with the γ subunit conferring benzodiazepine sensitivity. δ subunits and γ subunits co-express with α and β subunits and the predominant isoform in the mammalian brain includes 2α1 and 2β2/2β3 subunits, along with either δ or γ subunits (Somogyi et al., 1996). ρ subunits are localised within retinal bipolar cells, with only low densities found in the brain (Cutting et al., 1992; Cutting et al., 1991; Enz and Cutting, 1999). Each subunit consists of a large extracellular, ligand-binding, domain, a transmembrane domain containing four membrane-crossing α- helices (M1–M4) and a large M3–M4 intracellular loop (See Fig. 1.2). The 21 second transmembrane spanning regions (M2) are arranged centrally, forming the water-filled, channel pore. Loops within the extracellular domain are involved in ligand binding and are named A to F (Fig. 1.2 & Fig. 1.3). Receptors consist of a pseudo-symmetrical arrangement of subunits within the neuronal membrane (Fig. 1.4). 22 Figure 1.3 Schematic of a Cys-loop receptor. Loops within the extracellular domain are labelled by using common nomenclature A-to-F. Four transmembrane spanning regions are labelled M1, M2, M3 and M4. A large intracellular loop extends between transmembrane regions M3 and M4. Subunits form a pentameric arrangement to yield a complete receptor. Taken from Thompson and Lummis, 2007, with copyright permission granted from Expert Opinion on Therapeutic Targets. Acetylcholine binding protein (AChBP), is a soluble acetylcholine scavenging protein which is secreted by glial cells in the fresh water snail Lymnaea stagnalis (Smit et al., 2001). This protein is homologous to the nicotinic acetylcholine receptor and other Cys-loop receptors (See in Fig. 1.2). The crystal structure of this protein at 2.7 Å resolution has thus yielded insight into the structure of the extracellular domain of Cys-loop receptors (Brejc et al., 2001). Additionally, the structure of the extracellular domain of the nAChR α1 subunit has been solved to 1.9 Å (Dellisanti et al., 2007). These advances have led to the prediction that the extracellular domains of homologous Cys-loop receptors would be predominantly β-sheet in structure. Electron micrograph images of the full length AChR from the electric organ of the Torpedo ray at 4 Å resolution have revealed the structure of the full receptor (Miyazawa et al., 2003) showing the transmembrane regions to be α-helical (Fig. 1.4). 23 Figure 1.4 Structure of a Cys-loop receptor. Above: Ribbon structure of the refined structure of the acetylcholine receptor from the electric organ of the Torpedo ray at 4 Å resolution (pdb ID: 2bg9) (Miyazawa et al., 2003). Two of the five subunits are displayed in atomic stick format to enhance perspective. Individual subunits are labelled red, yellow and green for distinction. The extracellular domain is largely β-sheet and the transmembrane and intracellular domains are α-helical. Below: Pore view of the receptor from the extracellular perspective. Five subunits (labelled grey, blue, green, red and yellow for distinction) form a pentameric arrangement to yield a complete receptor. 24 1.6 Receptor activation 1.6.1 Gating Gating refers to the mechanism of transduction which links agonist binding at the extracellular domain to the opening of the channel gate some 50 Å away. To facilitate comparison between the channel-lining residues of the second transmembrane domain (M2) of different Cys-loop receptor channels, a prime number notation is used starting with the highly conserved positively-charged residues at the cytoplasmic end of M2, defined as 0′, increasing to another ring of charged residues at the extracellular end denoted as 20′ (Miller, 1989). The gate is deemed to be located at the “hydrophobic girdle” between M2 residues 9' and 14' (Bali and Akabas, 2007; Miyazawa et al., 2003; Panicker et al., 2002). The free energy derived from the binding of agonist to its binding site in the extracellular domain induces a conformational change which is transmitted to the gate and causes opening of the ion channel. Whilst most of the studies of channel gating have been done on AChRs, I will discuss studies from a range of receptors to illustrate the similarities across receptor types. Such studies can provide clues as to the mechanisms which underlie Cys-loop receptor function in general terms. Linear free-energy relationship analysis of mutant nAChRs has shown that a local conformational change in the binding site is propagated as a conformational wave to the channel (Grosman et al., 2000). An in-silico study of loop C, which forms part of the ligand binding site, in the human α7 nicotinic acetylcholine receptor (nAChR) described initial coupling between ligand binding and channel gating (Cheng et al., 2006); these targeted molecular dynamic (MD) simulations suggest that gating movements of the α7 receptor may involve relatively small structural changes within loop C of the ligand-binding domain. A more recent structural study of AChBP, co- crystallised with agonists and antagonists bound, shows a ‘capping’ motion of loop C closing the binding site cavity, thereby trapping bound agonist molecules. Residues at the apex of loop C moved inwards as much as 7 Å with agonists bound. However, antagonists caused loop C to move outwards as 25 much as 4 Å providing a mechanism underlying their inhibitory effects (Hansen et al., 2005). These findings agree with the previous modelling of Cheng et al for nAChRs, but the magnitude of loop C movement presented by Hansen would suggest that loop C undergoes larger movements during agonist binding than previously suggested. Environment-sensitive fluorescent probes have been used to identify residues involved in the gating mechanism. When the agonist is applied, changes in fluorescence can be monitored and related to conformational changes involved in gating. A recent study of the GlyR by Lynch and Pless (2009) showed that most fluorescently labelled residues in loops C and F yielded fluorescence changes identical in magnitude for glycine and strychnine (a GlyR antagonist) but distinct maximal fluorescence responses for labels on loops D and E. A similar study of GABAARs has shown that homologous residues within loop E of both β and α subunits display a similar fluorescence change during gating (Muroi et al., 2006), suggesting a similar structural transition occurs in both subunits. A follow on study by the same authors showed that residues in loop D (α1L127C and β2L125C) also produce an increase in fluorescence in response to GABA binding (Muroi et al., 2009). This presents the hypothesis of a ‘global transition’ with similar movements in different subunits occurring during channel gating. Agonist binding causes changes at the extracellular domain which are thought to induce a structural transition which causes a widening of the channel pore, formed by the five M2 helices. Indeed, images of the nAChR trapped in the open state showed that the M2 rods were tilted outwards from their resting configuration (Unwin, 1995). The kinks in the α-helices facing the channel pore (M2-region) had rotated to face the side, increasing the pore diameter by about 4 Å. With the solution of the crystal structures of related prokaryotic Cys-loop receptors - ligand gated ion channels from a bacterium (ELIC) (Hilf and Dutzler, 2008) and from a cyanobacterium (GLIC) (Hilf and Dutzler, 2009), a similar channel in a putatively open conformation - the pore structure in putatively both open and closed states has been revealed. Examining the 26 differences in pore structure between these two receptors suggests that the M2 helices relax and tilt around their M2 longitudinal axis with a widening of the channel at the extracellular end (Fig. 1.5). Studies of nAChRs support this hypothetical transition: a fluorophore attached to nicotinic acetylcholine receptor β M2 region (a Cys side chain introduced at the β 19' position in the M2 region) has been used to detect productive binding of agonist in the nAChR. During agonist application, fluorescence increased by approximately 10%, and the emission peak shifted to lower wavelengths, indicating a more hydrophobic environment for the fluorophore (Dahan et al., 2004). A more recent study of the muscle AChR suggests that small structural rearrangements may underlie channel opening. Ionisable amino acids were engineered into M1 and M3 and the effects on channel gating were monitored. The effect of proton transfer was determined using single-channel electrophysiology and it was determined that the M2 helices move by no more than 1.5 Å (Cymes and Grossman, 2008). This prediction is supported by computational approaches which predict that a 1.5 Å widening of the pore is sufficient to increase the channel conductance to values that are close to those observed experimentally (Corry, 2006). In 5-HT3 receptors an essential proline residue has been identified which has been shown to undergo a trans-cis isomerisation during gating which is necessary for channel opening. This study was carried out using unnatural amino-acid mutagenesis where the proline 8* residue was replaced with proline analogues favouring the trans conformer. The trend observed showed that the trans conformer prevented channel gating (Lummis et al., 2005a). This suggests a possible “switching mechanism” for channel gating revolving around the isomerisation of the proline 8* residue. This proline residue is not however present in GABA or glycine receptors, thusly confounding the possibility of a unified model for channel gating for all LGICs. There is some disagreement in the literature regarding the magnitude of the structural transitions underlying receptor activation, however there is a common agreement amongst many studies; a structural rearrangement amongst 27 extracellular binding site loops leads to a structural transition which causes a widening of the channel pore leading to ion flux. While specific residues may take on different roles in different receptors, it seems that this overview is appropriate for many, if not all, of the Cys-loop receptors. Figure 1.5 Examination of the pore structure of GLIC and ELIC, prokaryotic Cys-loop receptors, in putative open and closed states, respectively. Above: Two adjacent M2 regions of the pore with channel-lining residues labeled and side chains in stick representation. A difference in the M2 helical axis is obvious with a tilting away from the pore at the extracellular end of GLIC. Below: View of the channel pore from the extracellular position. M2 residues are in the space filling representation to demonstrate the pore cavities dimensions. A wider pore is observed in GLIC suggesting that the tilting of the helices results in a widening of the channel pore. 28 1.6.2 Interface residues Studies of the regions which bridge the extracellular and transmembrane domains have led to the identification of several critical “interface residues.” Structures from the extracellular domain which constitute this interface include the β1-β2 loop, the Cys-loop, the β8-β9 loop and the end of the β10 loop. The pre-M1 region, the M2-M3 linker and the C-terminal of M4 are also involved in this interface (Fig. 1.6) (Bartos et al., 2009a). Studies of chimeric receptors composed of AChBP tethered to the pore domain of the 5-HT3AR has led to an understanding of some of these coupling regions. Whilst the chimera is expressed at the cell surface, binding of ACh is unable to trigger channel opening, however with the substitution of three regions (β1-2 loop, Cys-loop and β8-9 loop) in AChBP, ACh binds with lower affinity but is capable of triggering channel opening (Bartos et al., 2009a). It has also been demonstrated that the ECD of α7 AChRs can be successfully coupled to the 5- HT3AR, as the ECD of α7 has similar interface residues (Bouzat et al., 2008) (Fig. 1.6). This construct has proven to be a good model for studying the role of interface regions, leading to the hypothesis that activation depends on a “complex network of loops,” where specific interactions between interface residues can not always be successfully substituted with loops from other receptors (Bouzat et al., 2008). 29 Figure 1.6 The interface between extracellular and transmembrane domains. Above: a structure of the Torpedo nAChR with one of its subunits highlighted with the extracellular domain in yellow and the transmembrane and intracellular domains in red. The interface is shown in the dashed square. b View of the structures at the interface. Left: The different segments are coloured as follows: orange (β1-2 loop), ice blue (β8-9 loop), green (Cys-loop), purple (β10-terminal), pink (pre-M1), blue (M2-M3 linker), and cyan (post-M4). Right: Surface representation of the interface loops. c Different views of the interface with key residues labelled. Ile210 in Torpedo nAChR corresponds to Leu210 in the human receptor. Below: Subunit sequences at the receptor interface. Taken from Bartos et al., 2009a. with copyright permission from Mol. Neurobiol. Similar work on GABAA receptors has shown that receptor activation depends on electrostatic interactions between charged residues in the β1-2 loop and the Cys-loop (Asp57 and Asp149) as well as Lys279 in the M2-M3 linker (Kash et al., 2003). An equivalent residue in the 5-HT3AR (Lys81 in the β1-2 loop) lies close to the extracellular end of M2 (26'A and 27'I) and mutational analysis has 30 shown that this residue is involved in channel opening (Reeves et al., 2005). In the GABACR, residue Glu92, situated in the β1-2 loop has been shown to form a salt-bridge with pre-M1 residue Arg258 (Price et al., 2007; Wang et al., 2007). In the same study it was shown that this specific interaction is absent in the 5HT3AR (Price et al., 2007). An interaction between residues in the pre-M1 and the M2-M3 linker has been identified in the human muscle nAChR. It was suggested that agonist binding disrupts a salt-bridge between pre-M1 Arg209 and Glu45 in the β1-2 loop, which in turn triggers a wave of interactions which propagate towards the channel (Lee and Sine, 2005). This hypothesis is supported by a related study which showed that Arg209 and Glu45 move early during the gating process (Purohit and Auerbach, 2007). Recent studies investigating the sequence of movements of residues during gating has led to the construction of a Φ map, which suggests that agonist binding triggers motions which causes a movement of the Cys-loop and β1-2 loop, followed by the M2-M3 linker, several M2 residues and finally the gate at the channel pore (9'-14' region) (Bafna et al., 2008; Chakrapani et al., 2004; Grutter et al., 2005; Zouridakis et al., 2009) The coupling of agonist binding to channel opening some 50 Å away involves several conserved regions, particularly the pre-M1 region, loop C and loop β1- β2. Though specific residues and interactions seem to vary across receptor types, there seems to be an overall degree of similarity across the Cys-loop family, suggesting a conserved mechanism of receptor gating. 1.6.3 Kinetic schemes Kinetic schemes have been posited which fit the observed single-channel behavior of Cys-loop receptors. Most of these schemes have been for the nAChR and GlyR, which are amenable to single-channel analysis (Auerbach, 2010; Lape et al., 2008; Sine and Engel, 2006). Sine describes a “primed” state which precedes channel opening at the nAChR (Mukhtasimova et al., 2009) (Figure 1.7). The channels’ closed to open transition is agonist-independent and 31 is preceded by two primed closed states; the singly primed state has an intermediate duration and triggers brief openings, whereas the doubly primed state has a brief duration and triggers long-lived openings. The priming step is thus a spontaneous transition that occurs after agonist binding. Similarly, Colquhoun describes a “flip” state of the GlyR, “a structural change which takes place while the channel is still shut.” This suggests that there is a structural change which occurs after agonist has bound but before the channel has opened. Both of these schemes suggest that a structural transition occurs which precedes channel opening and that initial binding of a ligand can influence the binding of subsequent ligands to adjacent binding sites. A similar conformational change for the high-conductance isoform of the 5-HT3AR has been reported (Corradi et al., 2009), suggesting that this may be a conserved mechanism across the Cys-loop receptor family. Figure 1.7 Kinetic scheme for Cys-loop receptor activation. Agonist binding, priming and channel gating steps are shown. C, C' and C'' symbolize closed states, whereas O' and O'' symbolize open states. A symbolises an agonist molecule. In the absence of agonist, the C' and C'' states are negligible, indicating that the first step in the activation process generates AC, from which there are three possible paths towards A2C''. Taken from Mukhtasimova et al., 2009, with copyright permission from Nature. 32 A structural study of AChBP described how agonists stabilise loop C in a fully contracted state, whereas peptide inhibitors stabilise loop C in a fully extended conformation (Ulens et al., 2009). This study utilised the strategy of co- crystallisation of AChBP with bound agonists and antagonists. These data, demonstrating a structural rearrangement of loop C following agonist binding, support the theoretical “flip” or “primed” states described by Colquhoun and Sine. Indeed, by providing a structural explanation of such transition states, Ulens has shown how such mechanisms may be part of Cys-loop receptor gating. Gating of the Cys-loop receptors seems to involve small structural changes at the ligand binding site which cause a change in the stability of the closed state of the M2 helices. The opening of the gate is dependant on this structural change. Whilst the basic mechanism for gating is generally understood, all of the intricate molecular determinants have yet to be characterised. There has thus yet to be posited a unified model of gating dynamics for all members of the Cys-loop family. 1.7 Ligand binding site The agonist binding site is located in the extracellular region at the interface between adjacent subunits. There is thus a theoretical maximum of five agonist binding sites per receptor. The binding site is comprised of six discontinuous loops (A-F). The ligand binds between Loops A, B and C, on the principal subunit, and loops D, E and F on the complementary subunit (Akabas, 2004) (Figure 1.8). 33 Figure 1.8 Extracellular domain of a Cys-loop receptor. The ligand-binding site is comprised of residues from loops A, B and C from the principal subunit and D, E and F from the complementary subunit. Functional experiments show that many Cys-loop receptors have a Hill coefficient greater than one - suggesting cooperativity - indicative of more than one agonist binding site per receptor. Single-channel experiments for the 5- HT3AR have led to the hypothesis that three occupied ligand binding sites lead to maximal receptor activation (Corradi et al., 2009). A similar study in the 5- HT3AR showed that receptors with three binding sites at non-consecutive subunit interfaces exhibit maximal mean channel open time and receptors with binding sites at three consecutive or two non-consecutive interfaces exhibit intermediate open time. However, receptors with binding sites at two consecutive or one interface exhibit a brief open time (Rayes et al., 2009). The increased receptor activity associated with the occupation of non-consecutive binding sites suggests that the structural rearrangement inherent in channel opening involves movements within the extracellular domains which affect adjacent binding sites. This hypothesis of ‘structural transitions’ is supported by voltage-clamp fluorometry studies of the GlyR (Pless and Lynch, 2009a; 2009b; 2009c) which show that an agonist specific structural change occurs in the loops of the extracellular domain that is correlated with agonist efficacy. 34 Many residues involved in binding GABA in the GABAA receptor have been identified within the extracellular binding site loops. In particular, hydroxylated residues, aromatic residues and charged residues have been identified as being important in ligand binding. Several aromatic residues which form an “aromatic box,” a hydrophobic surface favourable for ligand stabilisation, have been identified: α1Phe64, β2Tyr62, β2Tyr97 and β2Tyr205. Hydroxylated residues α1Ser68, β2Thr160, β2Thr202, β2Ser204 and β2Ser209 are critical for ligand binding. Charged residues are also involved in GABA binding: α1Arg120, α1Asp183, α1Arg66 and β2Arg207. These residues probably stabilise the carboxylate moiety of the ligand (Amin and Weiss, 1993; Boileau et al., 2002; Newell and Czajkowski, 2003; Wagner et al., 2004; Westh-Hansen et al., 1997; Westh-Hansen et al., 1999). A different repertoire of residues which are involved in GABA binding have been identified in the GABAC receptor: Tyr102, Arg104, Tyr106, Phe138, Val140, Arg158, Tyr198, Phe240, Thr244 and Tyr247 (Amin and Weiss, 1994; Harrison and Lummis, 2006; Lummis et al., 2005b; Zhang et al., 2008). Similarly to the GABAA receptor, these findings suggest an important role for aromatic, hydroxylated and charged residues in the GABAC receptor binding site. A review of GABA binding sites has described these data recently (See Lummis, 2009). In GABAA receptors GABA binds in a partially folded conformation, whilst in GABAC receptors it is in an extended conformation (Jones et al., 1998; Woodward et al., 1993). Although conformations of GABA within the binding sites of different GABA receptors varies somewhat, an overall similarity of location seems to be conserved. 1.8 Insect GABA receptors Insect GABA receptors are one of the major targets of insecticides (Raymond- Delpech et al., 2005) and Fipronil, an antagonist of insect GABA receptors, is one of the most important insect control chemicals to date (Sammelson et al., 2004). Other insecticides such as α-endosulfan and lindane target GABA receptors as a means of inducing a lethal convulsant state within their pestilent targets (Chen et al., 2006). 35 Several transcripts have been identified in D.melanogaster which may form GABA-gated anion channels. Whilst transcripts CG6927, CG7589, CG11340, and CG12344 have not yet been functionally assessed (Buckingham et al., 2005), transcripts RDL, GRD and LCCH3 have been heterologously expressed and assessed functionally. These three oligomeric invertebrate GABA receptor subunits have a similar degree of sequence homology with the characterised vertebrate GABAA receptor subunits as they do with each other (30- 38%)(Hosie and Sattelle, 1996a)(Figure 1.9). These insect GABA receptor subunits form heteromeric receptors with varying properties when expressed heterologously (Gisselmann et al., 2004; Hosie et al., 1997) (Table 1.1). It has been shown that GRD and LCCH3 coassemble to form GABA-gated cation channels when cRNA is co-injected into Xenopus laevis oocytes (Gisselmann et al., 2004). This suggests that RDL may play an important role in the formation of heteromeric GABA receptors. Furthermore, RDL orthologues have been identified in many other insect species; examples include the house fly (Eguchi et al., 2006), red flour beetle (Jones and Sattelle, 2008), honeybee (Dupuis et al., 2010), German cockroach (Kaku and Matsumura, 1994), and mosquito (Du et al., 2005). Furthermore, RDL receptors from the mosquito, cockroach and small brown planthopper have been functionally assessed (Shotkoski et al., 1994; Shotkoski et al., 1996; Eguchi et al., 2006; Narusuye et al., 2007). 36 Figure 1.9 A dendrogram illustrating the relative similarity of known insect GABA receptor subunits and other ligand-gated anion-channels. Vertebrate GABA receptor subunits are marked α,β, etc., GLY refers to GlyR subunits while Glu Cl – and Hc G1 refer to glutamate-gated chloride-channels and a putative GABAR or GlyR subunit from Haemonchus contortus. The RDL subunit is most closely related to vertebrate GABAA ρ subunits. Taken from Hosie et al., 1997, with copyright permission from Trends in Neurosciences. Table 1.1 Pharmacology of insect GABA receptor subunits when co-expressed and expressed alone. Subunit combination Heterologous expression Pharmacology Reference RDL homomer Oocytes, Sf-21; S2 Picrotoxinin-sensitive; Bicuculline-insensitive (Hosie et al., 1996; Millar et al., 1994; Buckingham et al., 1994; Lee et al., 1993; ffrench-Constant et al., 1993a) RDL + LCCH3 Oocytes, Sf-21 Picrotoxinin-insensitive; Bicuculline-sensitive (Zhang et al., 1995b) RDL + GRD Not reported Not reported N/A GRD + LCCH3 Oocytes (cation channel) Picrotoxinin-sensitive; Bicuculline-insensitive (Gisselmann et al., 2004) 37 Native insect GABA receptors are insensitive to the GABAA competitive antagonist bicuculline and sensitive to GABAA receptor agonists 3- aminopropanesulfonic acid (3-APS), isoguvacine, muscimol and the GABAA and Glycine non-competitive antagonist picrotoxin (Buckingham et al., 1994). This pharmacology distinguishes them from both vertebrate GABAC receptors, where 3-APS is an antagonist, and GABAA receptors which are sensitive to bicuculline and show sensitivity to many allosteric modulators (Sieghart, 1995; Hosie and Sattelle, 1996b). Insect GABA receptors are also sensitive to a range of non-competitive antagonists including Fipronil, TBPS and EBOB (Buckingham et al., 1994). The replacement of alanine 302 with serine or glycine in all subunits encoded by the rdl gene renders RDL-containing receptors 100-fold less sensitive to PTX than wild-type (ffrench-Constant et al., 1993a). RDL forms picrotoxin insensitive receptors when expressed with LCCH3 (Zhang et al., 1995b), suggesting that LCCH3 is the predominant subunit in these heteromers. However, when LCCH3 is coexpressed with GRD, picrotoxin sensitivity is restored on the cationic GABA-gated channel (Gisselmann et al., 2004). RDL is the best studied of these subunits as it can be expressed readily in Xenopus oocytes where it forms functional homomers with pharmacology similar to that of native insect neurons (Zhang et al., 1994). As such, RDL receptors are considered to be pharmacologically representative of insect GABA-gated ion channels in general. 1.9 RDL receptors The RDL subunit was originally identified in mutant D.melanogaster strains showing resistance to the insecticide dieldrin, hence the name “RDL” (ffrench- Constant et al., 1991; ffrench-Constant et al., 1993b). Dieldrin blocks the channel of RDL receptors and an Ala to Ser mutation was identified in the pore lining M2 region (Ala302), which conferred resistance. RDL receptor subunits in D. melanogaster can occur as a variety of different splice variants resulting in varying agonist sensitivities (Buckingham et al., 2005; Hosie et al., 2001; Hosie et al., 2001). The regions which are modified lie in exons 3 and 6 (Fig. 38 1.10). These alternative transcripts are named a, b (exon 3), c and d (exon 6). The RDLac variant is considered the canonical isoform and this is the isoform used in this study. Figure 1.10 Regions of rdl (the gene for the RDL subunit) which are alternatively spliced. Exons 3 and 6 undergo alternative splicing resulting in altered agonist sensitivity (Buckingham et al., 2005). Exons 3 and 6 correspond to extracellular loops D and F respectively. Reproduced with permission, from McGonigle et al., 2009, Biochemical Society Transaction, (Pt 6):1404-6. © the Biochemical Society (http://www.biochemsoctrans.org). RDL receptors were first expressed heterologously in Xenopus oocytes by ffrench-Constant et al. (1993a) and then in an insect cell line (S2 cells) by Millar et al. (1994). More recent studies have focussed on the role of alternative splicing in this receptor subunit; indeed, studies on the three cloned splice variants of rdl have revealed differences in agonist potency with GABA EC50s of 9-30, 56-60 and 100-150 µM for the ac, ad and bd splice variants, respectively (Buckingham et al., 2005; Hosie and Sattelle, 1996a), suggesting that the variant residues are involved in receptor activation. Since these spliced regions lie in the extracellular domain, close to the binding site of all Cys-loop receptors, alterations in this region could reduce agonist affinity and/or gating efficacy. Heterologously expressed RDL receptors have similar characteristics to GABA receptors of cultured Drosophila neurons (Zhang et al., 1994), and RDL 39 subunits are widely distributed throughout the adult and embryonic Drosophila CNS (Aronstein and ffrench-Constant, 1995; Harrison et al., 1996; Hosie et al., 1997); thus it is reasonable to assume that these receptors play a major role in inhibitory neurotransmission in the insect CNS. Indeed it has been shown using RNA knock down that RDL receptors are essential to olfactory learning and the associated conditioned stimulus pathway in Drosophila (Liu et al., 2009; Liu and Davis, 2009). A more recent study has shown, using gene knockdown, that RDL has a role in learning in Drosophila and in particular RDL negatively modulates olfactory associative learning (Liu et al., 2007). This highlights the functional importance of RDL in insect neuronal signalling. This study also provided high quality immunocytochemical staining images of RDL distribution throughout the fly brain, showing RDL expression to be distributed widely, with RDL receptors detected on neuronal dendrites and axons but not cell bodies. Nonetheless, RDL receptors were detected widely throughout the antennal lobes and mushroom bodies. It has been shown that RDL subunits can also coassemble with glutamate receptor subunits to form functional GABA-gated anion channels (Ludmerer et al., 2002). Since RDL can form heteromeric receptors with both Glutamate receptors as well as other insect GABA receptor-subunits (Gisselmann et al., 2004) (Table 1.1), it probably plays a role in the insect CNS, potentially generating a diverse set of GABA receptors. Studies of single-channel properties of RDL containing GABA receptors from cultured Drosophila neurons are similar to those of RDL homomers (Zhang et al., 1995b); the single-channel conductance for inward currents was 21 pS for RDL homomers, versus 28 pS for GABA receptors on cultured neurons (Zhang et al., 1994). The distribution of putative RDL-like GABA receptors has been demonstrated in the brain of the adult house cricket Acheta domesticus, using specific antisera (Strambi et al., 1998) (Fig. 1.11). The mapping of RDL subunits using RDL- specific anti-sera confirms the widespread distribution of RDL containing GABA receptors in the insect nervous system, similarly to that shown in Drosophila (Liu et al., 2009). 40 Figure 1.11 Distribution of RDL receptors in the adult house cricket Acheta domesticus. Diagram summarising, in frontal view, the distribution of RDL-like GABA receptors (right) and of the GABA-like immunoreactivity (left). Except for the mushroom bodies, immunoreactivity in the neuropils has been omitted for clarity. Nine groups (I-X) of GABAergic neurons were indicated as well as neurons immunoreactive to the RDL-GABA receptor subunit antiserum. (AC) Anterior calyx of mushroom body; (PC) posterior calyx of mushroom body; (P) peduncle of mushroom body; (α) α lobe of mushroom body; (β) β lobe of mushroom body; (AL) antennal lobe; (CB) central body; (OL) optic lobe; (PI) pars intercerebralis; (AN) antennal nerve; (AMC) antennal motor center; (DLP) dorsolateral protocerebrum; (VLP) ventrolateral protocerebrum. Taken from Strambi et al., 1998, with copyright permission granted from Cold Spring Harbour Laboratory Press. RDL subunits are homologous to GABAA, glycine, 5-HT3 and nACh receptors (Dougherty, 2008) (See Fig. 1.2). Extrapolation from vertebrate receptor data suggests that all of these proteins are pentameric, consisting of five subunits arranged pseudo-symmetrically around a central ion pore (Miyazawa et al., 2003; Sine and Engel, 2006). RDL however has a larger M3-M4 intracellular loop than its vertebrate orthologues, containing some 86 more residues (See Fig. 1.2). There is as yet no structural information available for RDL receptors, however we have recently published a homology model of the RDL extracellular domain based on AChBP, which shows a range of agonists 41 docked into the agonist binding site and predicts a range of ligand-receptor interactions (McGonigle and Lummis, 2010) (See Fig. 1.12). Figure 1.12. Model of the extracellular domain of RDL. Above RDL extracellular domain dimer. A 3D homology model of the extracellular domain of an RDL GABA receptor based on the crystal structure of AChBP (pdb: 1i9b) at 2.7 Å resolution. Extracellular loops A-F contribute to the agonist binding site. A GABA molecule is docked close to loops A, B and C. Below: Sequence alignment of RDL and related Cys-loop receptors. Aromatic residues that contribute to the binding pocket in the RDL receptor have mostly aromatic residues in the equivalent locations in other Cys-loop receptors. Taken from McGonigle et al., 2010, with copyright permission from ACS Biochemistry. RDL receptors are widely accepted as a good model for the behaviour of insect GABA receptors (Buckingham et al., 2005). They mimic closely the 42 pharmacology of native insect GABA receptors (Buckingham et al., 1994) and RDL receptors are located widely in the insect CNS (Sattelle et al., 2000; Strambi et al., 1998). Insect GABA receptors are also the major site for insecticide action (Buckingham et al., 2005; Casida, 1993; Casida, 2009) and RDL receptors expressed in Xenopus oocytes represent an effective insecticide/drug screening platform. RDL receptors are thus a model system for studying insect GABA receptors and Cys-loop receptors in general. By characterising RDL receptors we will gain a greater insight into the mechanisms underlying agonist binding as well as receptor function. Achieving a clear understanding of the determinants of the agonist binding site in RDL receptors will be a great aid in the rational design of RDL-specific antagonists, potentially paving the path to a new wave of selective insecticides as well as a greater understanding of insect GABA receptors. 43 2.0 Thesis Aims In this thesis I use biophysical, pharmacological, and computational methods as well as site directed mutagenesis to assess the structure and function of RDL receptors. In chapter 3 I determine the biophysical properties of wild-type RDL receptors using electrophysiological methods, assessing pH sensitivity, ionic selectivity and expression rates of receptors. In chapter 4 I report a structure-activity study of the agonist binding site, testing the potency of a range of GABA analogues and performing molecular modelling and docking simulations at the binding site. In chapter 5 I utilise site-directed mutagenesis and functional studies to assess the accuracy of my model of the binding site model. In chapter 6 I assess the potency of Ginkgo biloba extracts on RDL receptor function and use mutagenesis and molecular modelling to assess the nature of the binding interaction. In chapter 7 I assess the functionality of two brain-derived transcripts of the 5-HT3B subunit (Br1 and Br2) using electrophysiological and fluorometric methods. The overall aim of this thesis is to achieve a greater understanding of how Cys- loop receptors function at a molecular level. 44 Chapter 2 Materials & Methods ______________________________________________ 45 Materials 2.1 Suppliers and addresses Ambion UK [Applied Biosystems], Lingley House, 120 Birchwood Boulevard, Warrington, WA3 7QH, UK Bruxton Corporation, 6416 34th Ave, NW Seattle, WA 98107-2607, USA Dow Corning Corporation, Corporate Centre, PO Box 994, MIDLAND MI, 48686-0994, USA Garner Glass, 91711- 4921 Claremont, CA, USA Gibco [Invitrogen Ltd], 3 Fountain Drive, Inchinnan Business Park, Paisley, PA4 9RF, UK GraphPad Software, 2236 Avenida de la Playa, La Jolla, CA 92037, USA Greiner, Brunel Way, Stroudwater Business Park, Stonehouse, GL10 3 SX, UK Harvard Apparatus Ltd, PO BOX 126, Edenbridge, Kent TN8 6WF, UK Labtech, 1 Acorn house, The Broyle, Ringmer, East Sussex, BN8 5NN, UK Molecular Devices UK, Winnersh Triangle, 660-665 Eskdale Road, Wokingham, Berkshire, RG41 5TS, UK Molecular Devices US, 1311 Orleans Drive, Sunnyvale, CA 94089-1136, USA National Instruments, 11500 N Mopac Expwy, Austin, TX 78759-3504, USA NEB (New England Biolabs), 240 County Road, Ipswich, MA 01938-2723, USA Qiagen, QIAGEN House, Fleming Way, Crawley, West Sussex, RH10 9NQ, UK Sigma-Aldrich Company Ltd., The Old Brickyard, New Road, Gillingham, Dorset, SP8 4XT, UK Stratagene [Agilent Technologies, UK], South Queensferry, West Lothian, EH30 9TG, UK Sutter Instrument Company, One Digital Drive, Novato, CA 94949, USA 46 Methods 2.2 Subcloning of rdl into pGEMHE RDL subunit cDNA (GenBank accession number P25123) was gifted by N.Millar. The rdl gene (ac splice variant) was subcloned from pcDNA3.1 (Gibco: Invitrogen) into pGEMHE, the Xenopus oocyte expression vector, using the restriction sites AflII and XbaI; the pGEMHE vector contains 3'- and 5'-untranslated regions of a Xenopus β-globin gene, which has been shown to facilitate very high expression of a number of exogenous proteins in Xenopus oocytes (Liman et al., 1992). Positive clones containing the rdl gene insert were detected by carrying out restriction digests. 2.3 Preparation of mRNA 20 µg wild-type DNA in pGEMHE was linearised by digestion with Nhe1 HF (NEB) in 50 µL volume. 2 µL of proteinase K (1 mg/mL) was added to the reaction mixture and the sample was incubated at 37°C for 1 h in order to degrade any protein in the cDNA sample. DEPC treated water was added to give a final volume of 300 µL. DNA was purified with one phenol/chloroform extraction followed by one chloroform extraction. DNA was precipitated by the addition of 0.1 volumes 3M sodium acetate and 3 volumes 100% ice-cold ethanol. The mixture was left at -20°C for at least 1 h. DNA was pelleted by centrifugation for 15 min at 13,000 g at 4°C. The pellet was washed with 70% ethanol, air dried and resuspended in 10 µL DEPC water. Linearised DNA was diluted to 1 mg/mL and 1 µL was used for an in vitro transcription reaction using T7 mMESSAGE mMACHINE kit (Ambion). cRNA concentration was determined by measuring the A280 of a diluted aliquot of mMESSAGE product (2 µL diluted in 500 µL ddH2O) using a benchtop spectrophotometer, taking the original template cDNA into account. An aliquot of cRNA was then diluted to 50 ng/µL for oocyte injection. 47 2.4 Oocyte preparation Harvested Xenopus laevis oocytes were washed in four changes of ND96 saline (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 5 mM HEPES, pH 7.4), de- folliculated in ND96 saline with 1 mg/mL collagenase for approximately 2 h, washed again in four changes of saline, and then transferred to injection media (ND96 saline containing 2.5 mM Na-pyruvate, 0.66 mM theophyllen and 52 µM of gentamycin). Healthy stage V-VI oocytes were selected and injected with 5 ng mRNA and injected oocytes were stored in ND96 saline at 17ºC in 94 mm cell culture dishes (Greiner). Electrophysiological recordings were taken from oocytes ~24 h post-injection. 2.5 Electrophysiological recordings Oocytes from the African clawed frog Xenopus laevis are a favoured heterologous expression system amongst electrophysiologists working on ion channels. Two-electrode voltage clamping (TEVC) of the oocyte allows any changes in current that result from an ion channel opening event to be recorded digitally and quantified. This technique allows the response following agonist application to be monitored, yielding quantitative results within real time. Conventional TEVC of Xenopus oocytes was performed using standard electrophysiological protocols. A GeneClamp 500B amplifier was connected to a PC running CLAMPEX v6.0.3 software via a DigiDate1200 Series Interface (all Axon Instruments, Molecular Devices UK). Glass microelectrodes were pulled from GC150TF-10 glass capillaries (Harvard apparatus) using a P-87 micropipette puller (Sutter) to a resistance of 1-2 MΩ and back filled with 3 M KCl. 3 M KCl agar bridges connected ground electrodes to the bath (Fig. 2.1). Oocytes were held at a potential of -60 mV unless stated otherwise and perfused continuously with ND96 saline (recording saline) at a rate of 10-15 mL/min. 48 GABA (Sigma) was diluted in recording saline and applied to the bath using a simple row of valve switches (See Fig. 2.1). Doses of GABA were applied successively to achieve a concentration-response curve. Concentration−response data were normalised to maximal current (IMAX) fitted to the four-parameter logistic equation using GraphPad PRISM version 4.03 for Windows (GraphPad USA, http://www.grphpad.com). Significance was calculated using a one-way ANOVA or an unpaired t-test (Prism v3.02). I = IMIN + (IMAX − IMIN)/[1 + 10 log(EC50−[A]) nH] Where IMAX = maximal response plateau IMIN = minimum response plateau [A] is the concentration of agonist, and nH is the Hill coefficient Four-parameter logistic equation Figure 2.1 Standard TEVC setup. Left: TEVC rig with gravity driven saline perfusion and drug valves for smooth saline/drug transitions (above). Oocyte is immobilised by a small piece of copper wire in the recording bath and impaled by borosilicate pipettes (below). Right: TEVC rig schematic 49 Automated TEVC was carried out using the MultichannelSystems© Robocyte R8 electrophysiology platform. This system allows pre-injected oocytes to be loaded individually into a 96-well plate with V-shaped wells (Greiner) for high-throughput automated recording. Supplier’s ready-made measuring heads were used; each head comprises two glass microelectrodes, two silver/silver chloride (Ag/AgCl) reference electrodes, and a perfusion inlet and outlet. Recording protocols can be defined in simple scripts – small text files written in the dedicated Roboocyte scripting language. Based on preselected parameters, the program decides whether a cell is suitable for further recording, and, if not, the Roboocyte simply proceeds to the next oocyte in the well plate to continue the measurements. The Robocyte system can thus operate continuously without any supervision until all tests are performed or all cells within the 96-well plate are tested. For a comprehensive review of the Robocyte system, see Leisgen et al. (2007). 2.6 Sequence alignment The Drosophila melanogaster (GRBDROME) subunit was aligned with the sequence of the AChBP (primary accession number P58154) using FUGUE (Shi et al., 2001). FUGUE is an alignment program which assesses sequence similarity as well as quantifying this in the context of three-dimensional (3D) structure. FUGUE defines the conformation of the aligned protein considering main chain conformation, secondary structure, solvent accessibility and also H- bonding. FUGUE considers the local environment to calculate the likelihood of amino-acid substitutions, insertions and deletions and uses this information to determine the scoring system of the alignment. 2.7 Modelling A 3D model of the extracellular domain of the RDL receptor was created using MODELLER (Sali and Blundell, 1993), based on the AChBP (PDB accession number 1i9b) crystal structure. A dimer was generated by superimposing the modified monomer onto two protomers of the AChBP. 50 A model of the transmembrane regions of RDL was generated using the sequence of GLIC, a prokaryotic pentameric ligand-gated ion channel from Gloebacter violaceus (GLIC) (PDB ID: 3EAM). Homology models of mutant receptors were generated by the same method, by manually mutating the sequence of RDL before aligning using FUGUE. Models were analysed using RAMPAGE (http://raven.bioc.cam.ac.uk/rampage.php)(Lovell et al., 2003). Amino acids found in the unfavoured regions of this plot were visually inspected using ViewerLite (www.accelrys.com), to determine if their orientation will affect the binding site model. 2.8 Ligand docking Binding site ligand structures generated using ChemBio3D Ultra 11.0 were energy minimised using the MM2 force field. Docking of the ligands into the RDL receptor homology model was carried out using GOLD 4.0. The binding site was defined using the αC atoms of the conserved aromatics F206, Y254 and Y109, with a binding site radius of 10 Å. Ten genetic algorithm runs were performed on each docking exercise, giving a total of 10 solutions for each analogue. The structures were analysed using the implemented GoldScore fitness function to identify the most accurate simulation. Special attention was paid to GABA folding and conformation as well as positioning relative to conserved binding site residues and experimental evidence. Hydrogen bonding interactions between ligands and binding site residues were identified using the hydrogen bond monitor function in ViewerLite (www.accelrys.com). For channel-blocking antagonists, the PDB structures for ligands were downloaded from the CCDC (Cambridge Crystallographic Data Centre) database (http://www.ccdc.cam.ac.uk/). Compounds were docked into the homology model of the RDL transmembrane domain using the program GOLD 4.0. Ten docking experiments were performed for each compound. 51 2.9 Quantum mechanical ab-initio calculations for ligands Molecular modelling was carried out with ChemBio3D Ultra 11.0 (CambridgeSoft, UK; http://www.cambridgesoft.com). Ligand structures were generated in the charged zwitterionic state using the ChemDraw program. Ligands structures were energy minimized using the MM2 force field and Mulliken charges (partial atomic charges) were calculated using the GAMESS interface. Dipole distances were calculated using the atomic distance tool in ChemBio3D Ultra 11.0. 2.10 Site directed mutagenesis 2.10.1 Primer design Mutagenic primers containing the mutation and flanking complementary sequence were designed manually. Primers were designed to be 25-45 bases in length with the mutation in middle with at least 10-15 bases either side. GC content was at least 40% and Tm was typically ~80°C. Primers were terminated with one or more C/G at either end. A local restriction site was also disrupted using a silent base substitution where applicable. Where no restriction sites could be removed a restriction site was inserted. This facilitates quick detection of mutant clones. All primers were obtained from Sigma-Aldrich. 52 Table 2.1 Oligonucleotide primers used Mutant Forward primer (5'->3') Reverse primer (5'->3') F206A GAAATCGAAAGTGGCGGTTACAC GTGTAACCGCCACTTTCGATTTC F206Y GAAATCGAAAGTTACGGTTACAC GTGTAACCGTAACTTTCGATTTC Y254F GGCAACTTTTCTCGTTAGCCTGC GCAGGCTAACGAGAAAAGTTGCC Y254A GGCAACGCTTCTCGTTTAGCCT AGGCTAAACGAGAAGCGTTGCC E204D GCCAGCTGTGCCACATTGAAATCGATAGCTTCGG CCGAAGCTATCGATTTCAATGTGGCACAGCTGGC E204A GCCAGCTGTGCCACATTGAAATCGCAAGCTTCGG CCGAAGCTTGCGATTTCAATGTGGCACAGCTGGC S205T GCCAGCTGTGCCACATTGAAATCGAAACCTTCGG CCGAAGGTTTCGATTTCAATGTGGCACAGCTGGC S205A GCCAGCTGTGCCACATTGAAATCGAAGCCTTCGG CCGAAGGCTTCGATTTCAATGTGGCACAGCTGGC Y208S GCCACATTGAAATCGAGAGCTTCGGTTCTACGATGCGAGATATCCG CGGATATCTCGCATCGTAGAACCGAAGCTCTCGATTTCAATGTGGC Y208F GAAAGTTTCGGTTTCACGATGCGA TCGCATCGTGAAACCGAAACTTTC R111A GGACTTCACATTGGATTTTTACTTTGCTCAGTTTTGGACCGATCC GGATCGGTCCAAAACTGAGCAAAGTAAAAATCCAATGTGAAGTCC R111K GGACTTCACATTGGATTTTTACTTTAAGCAGTTTTGGACCGATCC GGATCGGTCCAAAACTGCTTAAAGTAAAAATCCAATGTGAAGTCC Y109F GGACTTCACATTGGATTTTTTCTTTCGTCAGTTTTGGACCGATCC GGATCGGTCCAAAACTGACGAAAGAAAAAATCCAATGTGAAGTCC Y109S CACATTGGATTTTAGCTTTCGTCAGTTTTGG CCAAAACTGACGAAAGCTAAAATCCAATGTG Y109A GGACTTCACATTGGATTTTGCCTTTCGTCAGTTTTGGACCGATCC GGATCGGTCCAAAACTGACGAAAGGCAAAATCCAATGTGAAGTCC Y109R.R111Y GGACTTCACATTGGATTTTCGCTTTTATCAGTTTTGGACCGATCC GGATCGGTCCAAAACTGATAAAAGCGAAAATCCAATGTGAAGTCC S176T CGTAGTAAGACCTTGCTATTGTTC GAACAATAGCAAGGTCTTACTACG S176A CGTAGTAAGACCTCGCTATTGTTC GAACAATAGCGAGGTCTTACTACG A302S GCAACGCCGGCGCGTGTGTCGCTCGGTGTGACAAC GTTGTCACACCGAGCGACACACGCGCCGGCGTTGC T306S TGCAACGCCGGCGCGTGTGGCTCTCGGTGTGAGTACCGTGTTG TGGTCATTGTCAACACGGTACTCACACCGAGAGCCACACGCGC T306V TGCAACGCCGGCGCGTGTGGCTCTCGGTGTGGTAACCGTGTTG TGGTCATTGTCAACACGGTTACCACACCGAGAGCCACACGCGC L314Q GAC AAT GAC CAC TCA GAT GTC GTC AAC AAA TGC GAC AAT GAC CAC TCA GAT GTC GTC AAC AAA TGC R256A CCTAACCACAGGCAACTATTCGGCTTTAGCCTGCG CGCAGGCTAAAGCCGAATAGTTGCCTGTGGTTAGG R256K CCTAACCACAGGCAACTATTCGAAATTAGCCTGCG CGCAGGCTAATTTCGAATAGTTGCCTGTGGTTAGG T306V.A302S GGCGCGTGTGTCGCTCGGTGTGGTAACCGTGTTG CAACACGGTTACCACACCGAGCGACACACGCGCC 2.10.2 PCR Polymerase chain reaction (PCR) was carried out using a high-fidelity Pfu DNA polymerase (Stratagene) and mutagenesis reactions were made up in a 1.5 mL Eppendorf tube as follows: 5 µL (10x) Buffer 1 µL Template DNA (100 ng) 1 µL (25 mM NTPs) (Stratagene) 3 µL Oligo (F) (150 ng) 3 µL Oligo (R) (150 ng) 32 µL dH20 5 µL DMSO* 1 µL Pfu turbo Covered with 50 µL mineral oil *10% DMSO was added as it was found to greatly boost the PCR product PCR was carried out using standard protocol (QuikChange, Stratagene). Annealing temperatures were modified as required for successful incorporation of the mutation. A melting temperature of Tm Oligo - 5ºC was routinely used. This temperature was altered arbitrarily (± 5ºC) where reactions failed to produce a product. Dpn1 digestion (1 µL Dpn1 (NEB)/mutagenesis reaction) at 53 37°C for 2 h was carried out to eliminate methylated template DNA from the PCR products. 1 µL of the reaction was run on a 1% agarose gel for 20 min to visualise any PCR product. Cycling Parameters 95°C for 1 minute 22 Cycles of: 1. 95°C for 1min 2. 60°C for 1min 3. 72°C for 10 min (or 2 min/kb DNA) 72°C for 7min (extension time after cycles) 4°C for infinity 2.10.3 Preparation of electrocompetent E.coli A 3 mL culture of DH5α E.coli was grown over night in LB in an incubator shaker at 37ºC. A 300 mL flask of LB was inoculated with the 3 mL culture and was incubated at 37°C in incubator-shaker until the OD600 was 0.5. The culture was then chilled on ice for 15 min and then spun at 2,500 g (JA-14 rotor) for 15 min at 4ºC. The supernatant was decanted and discarded and cells were rinsed in 300 mL ice cold dH2O. Cells were then re-spun as before and the supernatant discarded again. Cells were then rinsed in 100 mL ice cold dH2O, respun as before and then resuspended in 30 mL ice cold 10% glycerol and transferred to 50 mL Falcon tubes. Cells were then spun at 2,500 g for 15 min at 4ºC. Cells were finally resuspended in 2 mL 10% glycerol. Aliquots of 90 µL were pipetted into 1.5 mL Eppendorfs on dry ice. Samples were stored at -80°C. 2.10.4 Transformation of DH5α cells 45 µL of DH5α cells were electroporated (200 Ω, 1.8 kV, tau value 3 msec) with 1-7 µL mutagenesis reaction and 200 µL LB (at RT) was added immediately afterwards. The cell suspension was then incubated at 37°C for 1 h. Cells were plated out on LB + amp (100 µg/mL) plates and placed in an incubator over night. 54 2.10.5 Plasmid minipreps and restriction digests Colonies from the transformed DH5α plate were selected and grown over night at 37°C in 2 mL LB + ampicillin (100 µg/mL) in an incubator shaker. Plasmid DNA samples were extracted (QIAGEN Miniprep) and 5 µL of DNA was used to set up restriction digest. Samples were then stored at –20°C. Restriction digests were set up in 1.5 mL Eppendorf tubes according to the table: Vector DNA 5 µL 10X Buffer 1 µL Restriction enzyme 1 µL ddH2O 3 µL Digests were incubated for at least 1 hour at 37°C. Loading buffer, 5 µL, was added and samples were run on a 1% agarose gel with ethidium bromide added. Gels were viewed on a UV bed and photographs were taken. 2.11 Assessment of antagonists using TEVC TEVC was performed using the method outlined in section 2.5. Oocytes showing steady GABA-evoked responses were selected before antagonists were applied. All recordings were carried out at EC50 concentrations of GABA. Antagonists were co-applied with GABA at a range of antagonist concentrations and responses were measured yielding IC50 values (the concentration of antagonist (M) at which half-maximal inhibition is achieved). 2.12 Mutant cycle analysis Mutant cycle analysis can be used to investigate interactions between pairs of residues in proteins as well as interactions between residues and molecular groups on ligands (Hidalgo and MacKinnon, 1995). When assessing the effect of molecular groups on ligands the effect on ligand affinity of changing a ligand group or a protein residue, either individually or both together, is measured. If the individual effects are independent and additive, indicating no effective interaction between the two groups, then the coupling coefficient (Ω), defined in eqn. 1 below, will be equal to one. 55 The subscripts WT and MUT represent the wild type and mutant receptor, respectively, and L1 and L2 denote the two ligands being compared. Values of Ω > 2.5 have been shown to identify direct interactions between molecular groups (Schreiber and Fersht, 1995). The interaction energy, ∆∆Gint, between the two substituted groups is given by RTlnΩ, where R is the gas constant (8.314 JK-1M-1) and T is temperature (293 K). Ω values < 1 were normalised by the function Ω-1. 2.13 Whole insect Bioassays Bioassays were carried out by Mirel Puinean and Martin Williamson at Rothamsted Research, Hertfordshire. N.lugens, the brown planthopper was chosen as it is a typical crop pest. Adult brachypterous (short-winged) susceptible females were used to assess responses to compounds. Bioassays consisted of 3 replicates per dose, each one with 10-20 individual insects. Insects were removed directly from rearing cages, lightly anaesthetised with CO2 and dosed on the prothorax using 0.25 µL of acetone as the solvent carrier. Endpoint readings were taken at 48 h. Bioassays for 50 ppm and 1000 ppm were carried out. 2.14 Culture of HEK293 cells Human embryonic kidney (HEK) 293 cells were maintained on 90 mm tissue culture plates at 37 °C and 7% CO2 in a humidified atmosphere. They were cultured in DMEM: F12 (Dulbecco's Modified Eagle Medium/Nutrient Mix F12 (1:1)) with Glutamax I™ containing 10% foetal calf serum and passaged when confluent. 56 2.15 FlexStation recording 2.15.1 HEK293 transfection Cells were stably transfected with pcDNA3.1 (Gibco: Invitrogen) containing the complete coding sequences for the human 5-HT3A receptor subunit (Genbank accession number P46098), as previously described (Hargreaves et al., 1996). For heteromeric receptors, approx 106 HEK293 cells (stably expressing human 5-HT3A receptors) were transfected (electroporation: 1300 V, 30 ms) with 5 µg of human 5-HT3B subunit DNA (Genbank accession number O95264; pcDNA3.1, Invitrogen) using the microporator system (DigitalBio, Labtech), as per manufacturer’s instructions. Cells were then plated (approximately 3 × 104 cells/well) onto black sided, clear bottomed, 96 well plates (Greiner) for FlexStation assays. Cells were incubated 1–2 days before assay. 2.15.2 Recording HEK293 cells in each well were briefly washed once with 100 µL flex buffer (115 mM NaCl, 1 mM KCl, 1 mM CaCl2, 1 mM MgCl2, 10 mM glucose, 10 mM HEPES pH 7.4) at room temperature, and then incubated for 1 h at 37 °C in 100 µL flex buffer containing 1% v/v membrane potential dye (Reagent A-Blue Kit, Labtech). Fluorescence was measured every 2 s for 200 s and, at 20 s, 50 µL of agonist or flex buffer was added to each well. Softmax Pro (Molecular Devices, UK) or Prism (GraphPad, USA) was used for data analysis. The percent change in fluorescence, which was calculated as F (peak fluorescence) minus F0 (baseline fluorescence at 20 s) divided by agonist Fmax (peak fluorescence), and was compared across agonist concentrations yielding data points to be plotted as a concentration-response curve. 2.16 Single-channel electrophysiology HEK293 cells were transfected with subunit cDNA for human 5-HT3A and 5- HT3B subunits (pcDNA3.1, Invitrogen) using calcium phosphate precipitation as previously described (Corradi et al., 2009). For a 35 mm dish of cells, the 57 total amount of cDNA was 5 µg, and the ratio of A to B subunit cDNA was 1:5. Single-channel recordings were obtained from HEK293 cells one or two days after transfection in the cell-attached configuration (Corradi et al., 2009; Hamill et al., 1981) at a membrane potential of -70 mV, -100 mV and -120 mV at 20°C. The bath and pipette solutions contained 142 mM KCl, 5.4 mM NaCl, 0.2 mM CaCl2, and 10 mM HEPES, pH 7.4. Solutions free of magnesium and with low-calcium were used to minimize channel block by divalent cations. Patch pipettes were pulled from 7052 capillary tubes (Garner Glass) and coated with Sylgard (Dow Corning). Pipette resistances ranged from 5 to 7 MΩ. 5-HT was added to the pipette solution at roughly EC50 concentration. Single-channel currents were recorded using an Axopatch 200 B patch-clamp amplifier (Molecular Devices, US), digitized at 5 µs intervals with the PCI- 6111E interface (National Instruments), recorded to the computer hard disk using the program Acquire (Bruxton Corporation), and detected by the half- amplitude threshold criterion using the program TAC 4.0.10 (Bruxton Corporation) at a final bandwidth of 10 kHz (Gaussian digital filter). 58 Chapter 3 Biophysical properties of RDL receptors ______________________________________________ 59 Introduction 3.1 Introduction Oocytes from the African clawed frog, Xenopus laevis, are a favoured heterologous expression system amongst electrophysiologists working on ion channels. Two-electrode voltage clamping (TEVC) of the oocyte allows any changes in current that result from an ion channel opening event to be observed and quantified. This technique allows the response following agonist application to be monitored, yielding quantitative results within real time. Responses for a range of agonist concentrations can be recorded from the same oocyte, allowing concentration-response curves to be generated, yielding an EC50 value for the agonist. Here I determine the EC50 value for GABA at RDL receptors using conventional TEVC as well as the automated Robocyte R8 system. By measuring the amplitude of maximal currents elicited from oocytes expressing RDL receptors over a time course, we can determine the optimum incubation time for conducting electrophysiological experiments. Here I measure the maximal currents over a 72 hour period post-injection. RDL receptors are GABA-activated chloride channels. Allowing the passage of anions into the cell allows the generation of IPSPs, facilitating inhibitory neurotransmission under normal physiological conditions. By reducing the chloride concentration in the extracellular saline, changes in the reversal potential for the receptor can be used to determine the charge selectivity of the channel pore. This strategy has been employed in determining the charge selectivity of the acetylcholine receptor (Galzi et al., 1992). Here I have reduced the chloride concentration in the extracellular saline by 90%, substituting chloride ions with the large isethionate anion, to determine the effect of chloride concentration on channel currents. 60 Changes in the extracellular pH have been shown to affect the maximal currents of GABAA receptors in a subunit dependant manner (Krishek et al., 1996a). Here I investigate the effect of a change of pH on RDL receptors currents. The aims of this chapter were to characterise the biophysical properties of RDL GABA receptors expressed in Xenopus oocytes, specifically determining the GABA EC50 value, the expression rate, the ionic selectivity of the channel pore and the sensitivity of the receptor to changes in pH. 61 Results 3.2 Results In order to test RDL receptor function in Xenopus oocytes, the rdl gene was subcloned from pcDNA3.1 into the oocyte expression vector pGEMHE. The rdl gene was excised from pcDNA3.1 using the restriction enzymes AflII and XbaI (Fig. 3.1) and cloned into pGEMHE using the same restriction sites. Figure 3.1 AflII XbaI excision of rdl from pcDNA3.1. 3.2.1 Expression of RDL receptors in Xenopus oocytes Application of GABA to Xenopus oocytes expressing RDL receptors elicited concentration-dependant inward currents (Fig. 3.2). Little desensitisation occurred when GABA was applied successively to an oocyte expressing RDL receptors. RDL receptors recovered quickly following withdrawal of agonist allowing successive applications of agonist to be applied generating a concentration-response course. Plotting current amplitude against a range of GABA concentrations revealed an EC50 of ~20 µM (Fig. 3.3) which is similar to previously published results (10-31 µM) (Belelli et al., 1996; Hosie et al., 2001; Hosie and Sattelle, 1996a; McGurk et al., 1998). The dose response profiles as yielded by the automated Roboocyte R8 system and by conventional 62 TEVC are identical, giving confidence that an accurate EC50 value has been arrived at (Table 3.1). Although the Robocyte system allows accurate and reliable measurements of receptor function, it was felt that conventional TEVC allows more user control and adaptability. For cases where oocyte health varies or where immediate control of drug/wash is required, conventional TEVC affords the user greater control to quickly alter the recording protocol. For these reasons, unless otherwise stated, all further experiments in this dissertation were carried out using conventional TEVC. Figure 3.2 Electrophysiological traces from wild-type RDL receptors expressed in Xenopus laevis oocytes using conventional TEVC. Inward currents are evoked when GABA is applied. A range of doses of GABA were applied yielding data to plot a concentration- response curve. 63 Figure 3.3 Concentration-response curves for RDL receptors expressed in Xenopus oocytes for both the automated electrophysiology system (Robocyte R8) and conventional TEVC. The concentration-response curves for the automated system (Robocyte R8) and conventional TEVC are similar. Table 3.1 Parameters derived from concentration-response curves. Recording system pEC50 ± SEM EC50 (µM) nH ± SEM n Robocyte R8 4.671 ± 0.042 21.3 1.8 ± 0.3 6 TEVC 4.705 ± 0.039 19.7 1.8 ± 0.2 19 Data = mean ± SEM, nH is the Hill coefficient, n indicates number of replicates. Data are not significantly different (p=0.93, Student’s t-test). 64 3.2.2 RDL expression rate To determine the stability of receptor expression and to detect any decay of current amplitude occurs, EC50 currents were measured over a twenty minute period, a time scale representative of a typical experiment on a single oocyte. No decrease in current amplitude was detected over twenty minutes, confirming that receptors express stably (Fig. 3.4). Figure 3.4 EC50 current amplitudes from RDL receptors expressed in Xenopus oocytes over 20 min. RDL EC50 currents do not decay over a 20 min period as demonstrated by sample electrophysiological traces (above) and a graph of mean currents (± SEM) (n≥5 oocytes). RDL receptor maximal currents (evoked by 100 µM GABA) were recorded over a 72 hour period (post injection of cRNA) to determine expression rate in Xenopus oocytes. No response was observed at 6 h. At 60 hours no currents could be recorded and the oocytes health had deteriorated with a high membrane leak (>200 nA). Maximal currents (~7 µA) were observed at 40 h (Fig. 3.5). 65 Figure 3.5 Maximal currents from RDL receptors expressed in Xenopus oocytes over 72 h. A Sample current traces for 100 µM GABA (IMax) at 6 h, 24 h, 30 h and 48 h. B Maximal current amplitudes over 72 h (n≥6 oocytes for each data point). 66 3.2.3 Ionic selectivity of RDL receptors Maximal GABA currents were recorded over a range of voltages, in both standard ND96 saline (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 5 mM HEPES, pH 7.4), as well as a low chloride substitute (6 mM NaCl, 2 mM KCl, 1 mM MgCl2, 5 mM HEPES, 90 mM Na-Isethionate, pH 7.4). Reducing the chloride concentration resulted in an increase in the voltage reversal potential as evinced by a shift to a higher voltage of the IV curve, as characteristic for a chloride selective channel (Fig. 3.6). Published values for the internal ion concentrations for the Xenopus oocyte are: [K+], [Na+] and [Cl- ]; 110, 10 and 38 mM respectively (Costa et al., 1989). Since the internal chloride concentration of the Xenopus oocyte has been determined to be close to 40 mM (Costa et al., 1989), we can use the Goldman-Hodgkin-Katz Equation (Fig. 3.6) to determine the relative permeability of cations and anions through the channel pore. Figure 3.6 The Goldman-Hodgkin-Katz Equation. Vm is the membrane reversal potential. R is the universal gas constant (8.314 J.K-1.mol-1), T is the temperature in Kelvin (°K = °C + 273.15), F is the Faraday's constant (96485 C.mol-1), pK is the membrane permeability for K +, pNa is the relative membrane permeability for Na +, pCl is the relative membrane permeability for Cl-, [K]o is the concentration of K + in the extracellular fluid, [K]i is the concentration of K+ in the intracellular fluid, [Na]o is the concentration of Na + in the extracellular fluid, [Na]i is the concentration of Na+ in the intracellular fluid, [Cl]o is the concentration of Cl - in the extracellular fluid and [Cl]i is the concentration of Cl - in the intracellular fluid. A reversal potential of -22.2 mV was determined for receptors in ND96 saline and +11.2 mV in low chloride (10 mM) saline (Fig. 3.7). Such an increase in the reversal potential is characteristic of a chloride selective channel as shown by the Goldman-Hodgkin-Katz Equation rationale. Taking the combined extracellular concentrations of potassium and sodium together the relative permeability of cations to anions was calculated for both experiments. The 67 relative permeability of cations and anions was determined to be 0.07:1 and 0.26:1 in ND96 and low chloride saline respectively. Figure 3.7 Voltage-reversal experiments for RDL receptors in ND96 and low chloride saline. Above: Sample current traces for RDL receptors (maximal currents [GABA] = 100 µM) in normal ND96 and low chloride (10 mM) saline. Below: IV curves for normal and low chloride saline. Data = mean ± SEM for at least four oocytes. An increase in the reversal potential as shown with the reduced chloride saline is characteristic of a chloride selective channel. Reversals potentials were significantly different (p=0.03, paired t-test) and slopes were not significantly different (p=0.16, paired t-test). 68 3.2.4 pH sensitivity of RDL receptors Since GABA itself is an acid I felt it important to assess the affect of GABA on the pH of a solution and to determine if perturbations in pH have an affect on receptor function. The concentration-pH relationship for GABA in aqueous solution was determined using the program ChemBuddy (www.chembuddy.com) (Fig. 3.8). RDL receptors reach maximal activation at 100 µM GABA. At this concentration of GABA the pH of an unbuffered aqueous solution is 4.2. If high concentrations of GABA were to be reached in the synaptic cleft or the extrasynaptic milieu, local pH perturbations may play a role in modulating receptor function. Solutions of GABA in buffered recording saline (pH 7.4) were tested for changes in pH with the addition of GABA. Concentrations as high as 1 mM GABA effected no change in the pH of the saline. Figure 3.8 pH of an aqueous solution of GABA. Concentration-pH relationship for GABA (pKa 4.03 (Huxtable et al, 1987)) in aqueous solution. Generated using the ChemBuddy program (www.chembuddy.com). The effect of ten and one hundred-fold changes in extracellular pH on RDL receptor function was determined. An increase in the extracellular pH to 8.5 and 9.5 caused a decrease in the EC50 current of approximately 10% and 20% respectively. A decrease in the pH to 6.5 and 5.5 caused no decrease in the EC50 current (Fig. 3.9). 69 pH mean current SEM n 5.500 0.915 0.024 6 6.500 0.909 0.028 7 7.500 1.000 0.000 6 8.500* 0.881 0.038 8 9.500* 0.786 0.031 6 Figure 3.9 The effect of a pH change on the current amplitude of RDL receptors. Above: Sample traces showing the effect of a change in pH during a non-desensitising EC50 response. Changing the pH to 9.5 during the course of the receptor response decreased the current amplitude within ten seconds, whilst a decrease in pH to 5.5 had no effect. Below The effect of a change in pH upon GABA EC50 currents. Data = mean ± SEM for at least six oocytes. Currents at pH 5.5 and 6.5 are not significantly different from pH 7.5 (p>0.05, t-test). * denotes currents at pH 9.5 and 8.5 are significantly different from pH 7.5 (p<0.05, t-test). 70 Discussion 3.3 Discussion In this chapter I have assessed the biophysical properties of RDL receptors expressed in Xenopus oocytes. I have determined the GABA EC50 and I have assessed the rate at which the oocyte expresses RDL receptors. I have demonstrated the anionic selectivity of RDL receptors and I have also assessed the sensitivity of these receptors to changes in extracellular pH. The RDL subunit, an insect GABAA –like receptor, can be expressed readily in the Xenopus oocyte expression system. Maximal current responses were detected ~40 h post-injection. To facilitate experiments it was deemed optimal to inject in the morning and to record responses the following day, i.e. 24 h later, thus utilising the greatest window of opportunity for measuring robust current responses. The MultichannelSystems© Robocyte R8 automated TEVC electrophysiology platform yielded an EC50 value not significantly different to those arrived at using conventional TEVC. The GABA EC50 value is in line with previously published values (Hosie et al., 2001; Hosie and Sattelle, 1996a). This makes the Robocyte R8 a useful and reliable tool for studying ligand-gated ion channels. Cys-loop receptors elicit a cellular response by allowing the passage of cations or anions either into or out of the cell in an agonist dependant manner. The selectivity of the channel to cations/anions determines whether the effect of channel opening will lead to EPSPs or IPSPs, respectively. The ion selectivity filter is located close to the second transmembrane region of the receptor and residues that line the pore have been indentified to be responsible for charge selectivity in these receptors (Galzi et al., 1992; Gunthorpe and Lummis, 2001; Keramidas et al., 2000; Thompson and Lummis, 2003) (Fig. 3.10). 71 SFWINMDAAPARVGLGITTVLTMTTQSSGSRA SFWLNRESVPARTVFGVTTVLTMTTLSISARN GFYLPPNSG-ERVSFKITLLLGYSVFLIIVSD VFLLPADSG-EKISLGITVLLSLTVFMLLVAE SFWLNRNATPARVALGVTTVLTMTTLMSSTNA * : :: : : :* :* :. Figure 3.10 Alignment of M2 regions of RDL and other Cys-loop receptors. The -1' residue (highlighted) has been shown to be responsible for ionic selectivity in the 5-HT3A receptor (Gunthorpe and Lummis, 2001; Thompson and Lummis, 2003), the nAChR (Galzi et al., 1992) and the glycine receptor (Keramidas et al., 2000). RDL, like the glycine and GABAA receptors, has an alanine residue at the -1' position suggesting that the channel is also anion selective. RDL receptors are similar to other anionic selective channels, with neutral residues at the -1' position, and the findings here show that the channel is indeed selective for anions. My results here show that the channel is predominantly anion selective and that at low chloride levels, cations may compete for passage through the channel pore. A mutagenesis study of the β subunit of GABAA receptors has shown that mutating the residues at positions -5' to 0' to those of the α7 nicotinic acetylcholine receptor subunit converts the channel from being anion selective to cation selective (Jensen et al., 2002). This study fits with other work which showed that the ionic selectivity filter lies in the M1-M2 cytoplasmic loop of acetylcholine receptors (Wilson and Karlin, 2001; 1998). Similar studies on the 5-HT3R have highlighted the critical role of the -1' residue (Gunthorpe and Lummis, 2001; Thompson and Lummis, 2003) and proximal mutations (0, -1' and -2' residues) have been shown to convert GlyRs from anionic to cationic selective (Keramidas et al., 2000), with the converse mutations rendering the acetylcholine receptor anion selective (Galzi et al., 1992). Given the critical role of this region in determining the ionic selectivity of Cys-loop receptors, it is unsurprising that M2 ___________________ -11' -4' 0' 6' 12' 20' Glycine α1 GABA α1 5-HT3A AChRα7 RDL 72 RDL receptors are anion selective, given their similarity to their vertebrate orthologues. Changes in external pH are a feature of certain pathological processes such as ischaemia, anoxia and epileptiform activity (Chen and Chesler, 1992; Chesler, 1990; Kraig et al., 1983; Urbanics et al., 1978) and therefore inhibitory neurotransmission may be affected by such biophysical perturbations. Additionally, activation of GABAA receptors may change the external pH following bicarbonate efflux through the integral anion selective channel (Kaila, 1994; Kaila and Voipio, 1987) and this phenomenon could lead to physiological changes in GABA signalling. Indeed α1β1 GABAA receptors EC50 currents are potentiated by lowering the pH of the external medium from 7.4 to 5.4 by (73 ± 10%, mean ± SEM). In contrast, increasing the pH from 7.4 to 9.4 resulted in a reduction (36 ± 11%) in the response to GABA (Wilkins et al., 2005). While an increase in pH has been shown previously to have little or no affect on GABAAR currents (Huang et al., 1999; Krishek et al., 1996a), Wilkins et al. (2005) showed that GABA-activated currents are potentiated at pH 8.4 for both αβ and αβγ subunit-containing receptors, but only at GABA concentrations below the EC40. This same study also identified a critical role of the 24' lysine residue in the M2-M3 extracellular linker in the modulation of GABA currents. A prior study of GABA receptors has shown that this pH modulation is subunit dependent (Krishek et al., 1996a), with a histidine residue at the 17' position in the β subunit being critical to modulation at low pH (Wilkins et al., 2002). The homologous residue in RDL is different (17'S) while the 24' residue is conserved (24'K) so it is possible that the protonation state of this residue could affect channel gating. However, RDL receptors displayed relative resistance to changes in extracellular pH when compared to their vertebrate orthologues, with no potentiation observed at higher or lower pH. In fact current reduction was observed at higher pH (pH 8.5 and 9.5) with no change at lower pH (pH 6.5 and 5.5), distinguishing RDL receptors from GABAA receptors. These results demonstrate that a large change in pH has only a small effect on RDL receptor function and this finding may have wider implications for insect physiology and neural signalling. 73 Conclusion 3.4 Conclusions RDL receptors express stably in Xenopus oocytes with large currents (~ 5 µA) detectable at ~30 h. The Robocyte R8 is an effective automated electrophysiology platform for testing Cys-loop receptor function, post mRNA injection, and it is a reliable accompaniment to standard TEVC, yielding identical EC50 values. RDL receptors are predominantly chloride selective ion channels, highlighting their important role in inhibitory neurotransmission. RDL receptors are resistant to changes in extracellular pH, with only moderate effects at pH 8.5 and pH 9.5. These findings provide a biophysical profile for the functioning of RDL receptors, showing their agonist sensitivity, ionic selectivity and insensitivity to pH modulation. These findings may contain useful information for further studies on the physiological properties of insect GABA receptors. 74 Chapter 4 Molecular characterisation of agonists that bind to RDL receptors ______________________________________________ 75 Introduction 4.1 Introduction Cys-loop receptors, such as GABAA, GABAC (a subclass of GABAA), glycine, 5-HT3, and nicotinic acetylcholine (nACh) receptors, have homologous regions which form their agonist binding sites. These are located in the extracellular domain and are comprised of six discontinuous loops, named A-F (See Fig. 1.12). There are as yet no high resolution structural data of a complete vertebrate Cys-loop receptor, but the lower resolution nACh receptor structure, and homologous structures, such as those from the related acetylcholine binding protein (AChBP), have been useful for creating homology models (Brejc et al., 2001; Miyazawa et al., 2003; Sixma et al., 2003; Unwin et al., 2002). Homology models have been created for many different vertebrate Cys- loop receptors including nACh, 5-HT3, GABA and glycine receptors (Bartos et al., 2009b; Padgett et al., 2007; Pless et al., 2008; Thompson et al., 2008), but invertebrate Cys-loop receptors have been largely ignored. The lack of models for these proteins is surprising, as these proteins are the targets of a number of invertebrate specific ligands, such as insecticides, and a better understanding of critical binding site features could assist the development of novel, more effective compounds (Casida, 1993; 2009). Some information, however, can be extrapolated from vertebrate models, which supports previous mutagenesis and labelling studies in showing that aromatic residues in the binding loops contribute to the formation of an “aromatic box” that is important for ligand binding in all of these receptors (Beene et al., 2002; Beene et al., 2004; Harrison and Lummis, 2006; Padgett et al., 2007; Pless et al., 2007). However, critical residues are not necessarily equivalent for different receptors; even in the closely related GABAA and GABAC receptors, for example, GABA has different orientations in the binding pocket: in the former there is a cation-π interaction between GABA and a tyrosine residue in loop A, while in the latter this interaction is with a tyrosine in loop B (Lummis et al., 2005b; Padgett et al., 2007). Nevertheless, in both these receptors, there is evidence that the 76 carboxylate residue is close to arginine residues in loop D (Harrison and Lummis, 2006; Wagner et al., 2004), suggesting that GABA has broadly similar orientations in both GABA receptor binding sites. There are a range of invertebrate Cys-loop receptors that are activated by GABA, but, as mentioned above, we know little about the molecular details of insect GABA receptor binding sites. The aim of this study was to identify the molecular determinants of agonist binding and potential interactions with binding site residues in RDL receptors, one of the best studied classes of insect GABA receptor. 77 Results 4.2 Results 4.2.1 Functional responses: A range of GABA analogues was tested for activity at RDL receptors (Fig. 4.1). The most potent of these were muscimol, TACA and isoguvacine (Table 4.1; Fig. 4.2). THIP and β-alanine are weaker agonists with EC50 s of 220 and 800 µM respectively, while 5-aminovaleric acid (5-AV) and taurine were very weak agonists with EC50 values of 1.1 mM and > 10 mM respectively. GHB, a GABAB receptor agonist, and 3-APP, a GABAC and GABAB receptor antagonist, had no activating effect at RDL receptors. Glycine, 4-AB, PABA, and tyramine also failed to activate RDL receptors when applied at concentrations up to 10 mM. EC50 values for previously tested analogues (muscimol, β-alanine, TACA and isoguvacine) are close to published values (Hosie and Sattelle, 1996a). None of the compounds tested were antagonists of RDL receptors. 78 Figure 4.1 Chemical structures of GABA analogues examined in this study. Taken from McGonigle and Lummis, 2010, with copyright permission from ACS Biochemistry. 79 Figure 4.2. A. Electrophysiological traces showing maximal currents elicited by agonists tested at RDL receptors; GABA (100µM), TACA (300 µM), Isoguvacine (600 µM), β- alanine (10 mM) and Taurine (10 mM). B. Concentration-response curves showing the relative potencies of GABA and GABA analogues at wild-type RDL receptors expressed in Xenopus oocytes. EC50 values for analogues tested: Muscimol < GABA < TACA < isoguvacine < THIP < 5-AV < β-alanine < taurine. Data represent mean ± SEM, n > 3. Taken from McGonigle and Lummis, 2010, with copyright permission from ACS Biochemistry. 80 Table 4.1 Parameters derived from concentration-response curves. Agonist pEC50 ± SEM EC50 (µM) nH ± SEM n IMax/IMaxGABA Muscimol 5.044 ± 0.04 9.04 2.2 ± 0.6 5 0.87 ± 0.03 GABA 4.705 ± 0.04 19.7 1.8 ± 0.2 19 1.0 ± 0.06 TACA 4.435 ± 0.02 36.7 1.8 ± 0.2 3 1.0 ± 0.02 Isoguvacine 4.188 ± 0.03 64.9 1.9 ± 0.2 5 0.97 ± 0.03 THIP 3.645 ± 0.10 226 2.4 ± 1.0 5 0.57 ± 0.06 β-alanine 3.096 ± 0.03 801 2.0 ± 0.3 10 1.0 ± 0.03 5-AV 2.950 ± 0.15 1120 1.0 ± 0.3 4 0.85 ± 0.1 Taurine < 2.0 > 10,000 - 3 - GHB N/R - - - - 3-APP N/R - - - - Glycine N/R - - - - 4-AB N/R - - - - PABA N/R - - - - Tyramine N/R - - - - Data = mean ± SEM; N/R indicates no response, nH is the Hill coefficient, n indicates number of replicates. IMax/IMaxGABA indicates the maximal response compared to GABA. Taken from McGonigle and Lummis, 2010, with copyright permission from ACS Biochemistry. 81 4.2.2 Computational ligand analysis The atomic distance between the ammonium nitrogen and the carboxylate oxygen, or its equivalent substituent, was calculated for all ligands tested. Comparing these data to the ligand EC50 values (Table 4.2) indicates that the most potent activators have a dipole separation distance of ~5 Å. Glycine has a dipole separation distance of 2.35 Å, and may be too short to activate RDL receptors, while for tyramine the distance was 7.94 Å, thus this molecule may be too long to fit into the binding site cleft. GHB and 3-APP, however, which have dipole separation distances of 5.1 Å and 4.5 Å respectively, are not agonists, while 5-AV, which has a dipole separation distance of 6.5 Å, can activate receptors. These data suggest that other factors also play a role. One of these may be charge distribution; partial atomic charge calculations suggest there is a dependence on the electrostatic potentials of atomic groups at either end of the ligand. Partial atomic charges of + 0.3 at ammonium hydrogen atoms and - 0.5 to - 0.6 on carboxylate oxygen atoms were common amongst the most potent ligands (Table 4.2). However for 3-APP and 4-amino-1-butanol (4-AB), which do not activate receptors, the charges on phosphonic acid and hydroxyl oxygens are -1.2 and - 0.4 respectively. Additionally, 4-AB and GHB are not zwitterions and thus carry no formal charge, and so even if their length is close to optimal for receptor activation, their inactivity could be attributed to the absence of charge at either end of the ligand. The hydroxyl of GHB cannot substitute the ammonium of GABA, confirming the importance of this group for ligand binding at RDL receptors. 82 Table 4.2 Dipole separation distances of GABA analogues. Ligand Potency pEC50 Dipole separation (Å) Carboxylate Oxygens* Ammonium Hydrogens Muscimol 5.04 5.22 -0.26 (Isox-N), -0.47 0.32, 0.32, 0.30 GABA 4.71 4.83 -0.60, -0.54 0.31, 0.31, 0.29 TACA 4.44 4.88 -0.57, -0.53 0.31, 0.31, 0.29 Isoguvacine 4.19 4.86 -0.55, -0.53 0.29, 0.31, 0.29 THIP 3.65 5.14 -0.25 (Isox-N), -0.46 0.29, 0.31, 0.31 β-alanine 3.09 4.37 -0.58, -0.51 0.29, 0.29, 0.29 5-AV 3.06 6.47 -0.61, -0.56 0.31, 0.31, 0.30 Taurine < 2.0 4.46 -1.11, -1.12 0.30, 0.30, 0.28 GHB N/R 5.10 -0.61, -0.61 -0.38 (Hydroxyl-O) † 3-APP N/R 4.50 -1.23, -1.23 (Phos-O) 0.29, 0.31, 0.31 Glycine N/R 2.35 -0.47, -0.57 0.32, 0.32, 0.27 4-AB N/R 6.20 -0.38 (Hydroxyl-O) 0.19, 0.20, 0.24 PABA N/R 6.39 -0.54, -0.54 0.30, 0.30, 0.30 Tyramine N/R 7.94 -0.47 (Hydroxyl-O) 0.31, 0.30, 0.30 Mulliken charges of GABA analogues calculated using the GAMESS interface. Ammonium hydrogens and carboxylate oxygens, or their equivalent substituents for the dipole, are listed. *Ligands without carboxyl groups were assessed by their equivalent groups: Isoxazole nitrogens in THIP and muscimol, phosphate oxygen in 3-APP, and hydroxyl oxygen in 4-AB and tyramine. † GHB does not have an ammonium group. The charge on the hydroxyl oxygen is negative. Taken from McGonigle and Lummis, 2010, with copyright permission from ACS Biochemistry. 83 4.2.3 Homology modelling and docking Homology modelling of the extracellular domain of RDL was based on the crystal structure of AChBP and analogue docking was ranked based on the GOLDscore fitness function, which ranks simulations by comparing interaction energies by consideration of predicted protein-ligand hydrogen bond energy, protein-ligand van der Waals energy, ligand internal van der Waals energy and ligand torsional strain energy, and has been established as an accurate method for scoring ligand-protein docking (Verdonk et al., 2003). The most favored orientation for GABA in the binding site was with the ligand in the cleft between loops B, C, and D (Fig. 4.3). The carboxyl group of GABA was deepest in the cleft, located close to residues Y109 and R111 in loop D. The ammonium was located between aromatic residues contributed by loop B (F206) and loop C (Y254), suggesting a cation-π interaction with one or both of these residues. A hydrogen bond was predicted between the ammonium of GABA and the backbone carbonyl of S205. 5-AV docked in a similar orientation with a hydrogen bond predicted at the same location. β-alanine was found to dock in two almost equally favored orientations. One of these was similar to the orientation of GABA described above, with the carboxyl deep in the cleft and the ammonium sandwiched between F206 and Y254. In this orientation there was a hydrogen bond between the ligand ammonium and the S205 backbone carbonyl oxygen. In the second orientation β–alanine was close to S131 above loop E, and there was a hydrogen bond between the ligand ammonium and the side chain hydroxyl. All other agonists docked with the ammonium moiety oriented close to loop B residues E204 and S205, with hydrogen bonds predicted between these residues and TACA and taurine (Table 4.3). Muscimol and isoguvacine were predicted to hydrogen bond with residues L249 and Y254 respectively. These orientations also position their carboxylate moieties close to loop D residue R111. 84 Table 4.3 Hydrogen bonding partner residues predicted from docking of functional GABA analogues. Analogue H-bonding residues predicted GABA S205 β-alanine S205 5-AV S205 Muscimol L249 Isoguvacine Y254 TACA S205, E204 Taurine S205, E204 THIP None detected Taken from McGonigle and Lummis, 2010, with copyright permission from ACS Biochemistry. 85 Figure 4.3 Docking of GABA and active analogues (β-alanine, TACA, 5-AV, taurine, isoguvacine, muscimol and THIP) into the RDL binding site. Analogues docked with the ammonium group located deep in the cleft formed between loop B and loop C aromatic residues. The carboxylate moiety or equivalent substituent was predicted to be located facing towards R111. Atomic colour scheme: Grey (Carbon); Red (Oxygen); Blue (Nitrogen); White (Hydrogen); Orange (Sulphur). Taken from McGonigle and Lummis, 2010, with copyright permission from ACS Biochemistry. 86 Discussion 4.3 Discussion In the present study I have examined the characteristics of agonist binding in RDL receptors, which are GABA-gated chloride selective Cys-loop receptors found in insect nervous systems. They are important to understand as insecticidal targets, and knowledge of their structure and function could clarify how these receptors function in other species. I have created a homology model of the RDL binding site, and demonstrated that GABA and a range of GABA analogues can dock into this site with a range of potential interactions with binding site residues. I have also examined the characteristics of the agonists and, using functional and modelling data, described the critical features required to activate these receptors. The GABA-gated chloride channels produced when RDL subunits are expressed in Xenopus oocytes have distinct pharmacological characteristics to the vertebrate GABAA receptors: they are activated by GABA with an EC50 of ~20 µM; they exhibit minimal desensitization; they can be inhibited by PTX but not by bicuculline and they can be activated by the GABAA receptor agonist muscimol and the GABAC receptor agonist TACA. Muscimol, which is a full agonist of GABAA receptors and a partial agonist of GABAC receptors, is a full agonist at RDL receptors, while THIP, which is a full agonist of GABAA receptors and an antagonist of GABAC receptors, is a partial agonist at RDL receptors. These characteristics are similar to GABA-gated receptors in cockroach neurons as previously reported (Sattelle et al., 1988; Schnee et al., 1997). Other GABA-gated channels that have been heterologously expressed include those from Musca (MdRdl) and Laodelphax, where GABA-gated anion channels were identified with similar GABA EC50 values; these findings are similar to GABA-gated chloride channels in lobster, cockroach, and crab neurons and are in contrast to those of GABA-gated cation channels that have been identified in Caenorhabditis elegans, small crab, and Drosophila (Beg 87 and Jorgensen, 2003; Duan and Cooke, 2000; Eguchi et al., 2006; Gisselmann et al., 2004; Swensen et al., 2000). Our examination of the GABA agonists suggests that GABA activates RDL receptors in the extended conformation, with agonist optimum length of ~5Å. This is similar to the extended conformation of GABA at GABAC receptors but differs to the partially folded conformation of GABA at GABAA receptors (Verdonk et al., 2003). 5-AV, which is one CH2 longer than GABA, and β- alanine, which is one CH2 shorter, can also activate the receptors, suggesting that they are both close enough in size to generate essential contacts within the binding site, although the potencies of both these compounds (EC50 ~1 mM) are considerably less than GABA. Glycine, which is two CH2 groups smaller than GABA, is inactive, as is tyramine, which is considerably larger than GABA. Similarly, a correlation between agonist affinity and agonist length has been previously shown in GABAA receptors where affinity is correlated with ligand length, with a decreased binding rate for shorter ligands, suggesting a length-based selectivity mechanism at these receptors (Jones et al., 1998). Replacement of the ligand carboxylate group with a hydroxyl, as in 4-AB, results in an inactive ligand, confirming the requirement for an anionic group at this point of the ligand. Taurine, which is an analogue of β-alanine with the carboxylate replaced by a sulphonate, can also act as an agonist, although its low potency indicates that sulphonate does not substitute very effectively for carboxylate. Replacement of the carboxylate with a phosphoric acid group, however, as in 3-APP, results in an inactive ligand, demonstrating that the type of acidic group is important. 3-APP also had no antagonistic properties at RDL receptors which is in contrast to GABAC receptor data, where replacement of the carboxylic acid group of GABA with a phosphonic acid, phosphinic acid or sulphonic group produces potent antagonists (Woodward et al., 1993). The ammonium group is also critical. GHB, the well known “date rape” drug, is an analogue of GABA with the ammonium group replaced by a hydroxyl. This compound had no activity at RDL receptors. 88 The data therefore support the requirement for a charged dipole for receptor activation, suggesting electrostatic interactions with binding site residues are important. Computational ligand calculations reveal a partial negative charge at the carboxylate moiety and a partial positive charge at the ammonium moiety for all active agonists. Thus, since the most potent ligands have a dipole distance close to 5 Å, and both ends of the ligand are thought to interact with the receptor, we can assume that the agonist binding site lies at the cleft between the binding loops with a distance of ~5 Å from one another. This suggests that loops B, C, and D comprise the part of the binding site with which agonists interact in this receptor, as placement of ligands between the other loops would require a longer ligand. This hypothesis is supported by the model which shows that loops B, C, and D form a clearly defined pocket in which the ligands bind (Fig. 4.3). Docking of ligands identified several residues with hydrogen bonding potential, in particular in loops B and C, which have been shown to have important contact points for ligands in 5-HT3, nACh, glycine and GABAC receptors (Beene et al., 2002; Lummis et al., 2005b; Pless et al., 2008; Thompson et al., 2008). Most of the agonists (GABA, β-alanine, TACA and taurine) could form a hydrogen bond with S205 (loop B), and TACA and taurine could also form H-bonds with E204 (also loop B). Muscimol and isoguvacine could form H- bonds with loop C residues L249 and Y254 respectively. The ammonium group of all the functional agonists docked between loop B and loop C aromatics (F206 and Y254 respectively) suggesting a cation-π interaction. All Cys-loop receptors studied to date have been shown to have a cation-π interaction with an aromatic residue in the binding pocket and the ammonium of their natural ligand, although the location of the interacting residue varies from receptor to receptor. Such interactions have mostly been with loop B aromatics (in nACh, 5-HT3, Gly and GABAC receptors), but interactions with loop C (MOD-1) and loop A (GABAA) aromatics have also been reported (Mu et al., 2003; Padgett et al., 2007). 89 Conclusion 4.4 Conclusions In conclusion, I have identified the charge and dipole requirements for agonist recognition in RDL receptors and have identified residues within loop B, loop C and loop D which could interact with agonists. I have also shown that loop B and/or loop C aromatic residues could contribute to the binding and/or function of the receptor via a cation-pi interaction. The data therefore provide a model of the agonist binding site, which can be used for further structure activity studies and rational drug design. 90 Chapter 5 Investigating the GABA binding site of RDL receptors ______________________________________________ 91 Introduction 5.1 Introduction The aim of this chapter was to characterise putative binding site residues in order to assess the accuracy of my binding site model. I use mutagenesis to assess the effect of altering the side chains of residues identified as potentially important in agonist binding. The effects of these substitutions are assessed by TEVC, monitoring changes in the concentration-response relationship with GABA. ‘Binding’ residues in the homologous GABAA receptor have been identified within the extracellular loops using mutagenesis, single-channel analysis and homology modelling: several aromatic residues form an ‘aromatic box,’ a hydrophobic surface favourable for ligand stabilisation. Hydroxylated residues and charged residues are also involved in GABA binding through direct salt- bridge or hydrogen bonding interactions (Amin and Weiss, 1993; Boileau et al., 2002; Newell and Czajkowski, 2003; Wagner et al., 2004; Westh-Hansen et al., 1997; Westh-Hansen et al., 1999). Furthermore, aromatic residues from loops A, B, C and D have been found to form an aromatic box stabilising ligand binding in other Cys-loop receptors and several studies have in fact identified these regions as the binding site (Bartos et al., 2009b; Lummis, 2009; Lummis et al., 2005b; Melis et al., 2008; Padgett et al., 2007; Pless et al., 2008; Thompson et al., 2008; Wagner et al., 2004; Xiu et al., 2009). Homology modelling and docking experiments on RDL receptors in Chapter 4 led to the identification of several candidate binding residues (See Fig. 5.1). These residues are in similar regions to binding residues in the homologous GABAA receptor. I determine the importance of Loop B residues F206, Y208, E204 and S205, assessing the role of aromaticity, hydroxylation and charge. Loop C residues Y254 and R256 as well as Loop D residues Y109 and adjacent residue R111 are investigated, as well as Loop A aromatic residues F146 and F147. The aromatic residues investigated here are conserved across many of the Cys-loop receptors (See Fig. 1.2) and may well be important in ligand 92 binding. Indeed, in all of the mammalian Cys-loop receptors studied to date, a cation-π interaction with the native ligand has been identified, highlighting the importance of aromatic residues in ligand-binding. Additionally, charged residues may be involved in salt-bridge or electrostatic interactions with the charged ends of GABA and the charged residues investigated here may be involved in such interactions. Such interactions seem likely, given the requirement for a charged dipole as shown in the previous chapter. Thus, here I aim to identify the partner binding residues for GABA in the RDL binding site. 93 Figure 5.1 Binding site model. Above: Ribbon model of the RDL extracellular domain with GABA docked into its binding site. Loops A (Blue), B (Yellow), C (Green), D (Red), E (Black) and F (Purple) are highlighted. Below: Close up view of the putative agonist binding site with GABA docked and important residues labelled. Atomic colour scheme: Red (Oxygen); Blue (Nitrogen); Grey (Carbon); White (Hydrogen). Loops A-E are shown in wire format and coloured as above. 94 Results 5.2 Results 5.2.1 Molecular biology Mutagenesis PCR reactions yielded a DNA product of the expected size (6 kb), confirming the successful amplification of the pGEMHE vector (Fig. 5.2). Digestion of mini-prep DNA from E.coli colonies transfected with PCR product allowed the identification of mutant DNA samples (Fig. 5.3). Figure 5.2 Gel screen showing PCR product Figure 5.3 Analysis of wild-type (WT) and mutant (Mut) DNA as revealed by the mutagenic removal of a reporter restriction site (HindIII). Samples show the removal of a HindIII site in the mutant DNA (mut), confirming the mutation has been incorporated. 95 5.2.2 Loop B mutants Loop B is an important region of the Cys-loop receptor binding site and residues in this loop have been shown to be critical for ligand binding and/or receptor activation; specifically, aromatic residues in loop B have been shown to be critical, forming a cation-π interaction with the native ligand in the 5-HT3, Glycine, GABAC and nACh receptors (Beene et al., 2002; Thompson et al., 2008; Lummis et al., 2005b; Zhong et al., 1998; Pless et al., 2008). Furthermore, charged residues in loop B have also been shown to be critical for receptor function in the 5-HT3 receptor (Thompson et al., 2008). For these reasons, charged and aromatic residues in RDL loop B were investigated. F206 Mutation of F206 to alanine (F206A) resulted in an increase in EC50 for GABA (53 µM), from 20 µM in wild-type receptors, and a reduced Hill coefficient of 0.9 (Fig. 5.4; Table 5.1) (Hill coefficient is 1.8 in wild type receptors). The reduction in the Hill coefficient to 0.9 would suggest that cooperativity within the receptor has been reduced. Mutation of F206 to tyrosine (F206Y) resulted in an EC50 of 0.5 µM (a 38-fold decrease from wild type). There was no significant change in Hill coefficient for F206Y, which would suggest that cooperativity has not been altered. β- alanine was tested on this mutant to assess whether the gain of function was GABA specific or observable for other agonists. β-alanine had an EC50 of 17 µM, which is a 47-fold decrease in EC50, comparable to the change for GABA (Table 5.1). Since β-alanine showed a near parallel shift in EC50 with GABA, it is likely that the biophysical affects of the F206Y mutation affect these two ligands similarly. Such a large gain in agonist sensitivity would suggest that this residue may contribute to ligand binding and/or receptor gating. 96 Figure 5.4 Concentration-response curves for F206 mutations. Above: GABA and β- alanine concentration-response curves for F206A, F206Y and wild type (WT) receptors. Each data point is the mean ± SEM for at least four oocytes. Below: Sample electrophysiological traces showing maximal and approximately EC50 currents from wild type and mutant receptors. 97 Y208 Loop B residue Y208 did not feature prominently in the binding site model, however its proximity to F206 prompted close attention to this region. Y208 was predicted to point away from the GABA binding site, pointing inwards towards the core of the extracellular domain. The effect of mutating this residue and also coupling the Y208F mutation to the F206Y mutation was assessed to identify potential interactions between these residues. Mutation of Y208 to phenylalanine (Y208F) yielded functional receptors with a GABA EC50 of 71 µM (a 3.7-fold increase from wild type) with no significant change in the Hill coefficient. This mutation increased the EC50 for β-alanine to 2.8 mM (a 3.5-fold increase from wild type). Similarly, mutation of Y208 to serine (Y208S) resulted in a GABA EC50 of 47 µM (a 2.5-fold increase from wild type), with no change in Hill coefficient (Fig. 5.5; Table 5.1). The double mutant F206Y.Y208F had a GABA EC50 of 3.6 µM, with a Hill coefficient of 1.1. Double mutant cycle analysis of these two residues shows that they are coupled with interaction energy of 1.7 kJ/mol (Fig. 5.6). 98 Figure 5.5 Concentration-response curves and sample electrophysiological traces for Y208 mutations. Above: GABA concentration-response curves and sample traces for Y208F, Y208S and double mutant 206Y.208F. Below: β-alanine concentration- response curve and sample traces for Y208F. Each data point is the mean ± SEM for at least four oocytes. 99 Figure 5.6 Mutant cycle analysis for mutants Y208F, F206Y and double mutant F206Y.Y208F. A coupling energy of 1.7 kJ/mol suggests that the effects of these mutations on receptor function are coupled. This suggests that these residues are situated close together in an orientation which may contribute to the GABA binding site. Values of Ω > 2.5 have been shown to identify direct interactions between molecular groups (Schreiber et al., 1995). 100 E204 Loop B glutamate (E204) was mutated to alanine (E204A) and aspartate (E204D). Xenopus oocytes injected with 50 ng of cRNA for these mutants failed to give responses to GABA, suggesting that GABA binding or gating has been disrupted. S205T Loop B serine (S205) was mutated to threonine (S205T) and alanine (S205A), to test whether this residue is critical to receptor function. Oocytes injected with 50 ng of cRNA for these mutants failed to give responses to GABA. 5.2.3 Loop C mutants Loop C is an important part of the ligand binding site in Cys-loop receptors; recent studies have suggested that loop C moves in a ‘capping motion’ during agonist binding (Ulens et al., 2009; Hansen et al., 2005) and previous studies have demonstrated a role for aromatic and charged residues in agonist binding in the related 5-HT3, GABAA and GABAC receptors (Padgett et al., 2007; Beene et al., 2007; Harrison and Lummis, 2006). For these reasons, aromatic and charged residues in RDL loop C were investigated to determine their role in agonist binding. Y254 Loop C tyrosine (Y254) was mutated to alanine (Y254A) and phenylalanine (Y254F). Oocytes injected with Y254A cRNA failed to give responses to GABA. Y245F mutant receptors had an increased GABA EC50 of 124 µM (6.5-fold greater than wild-type), with no significant change in Hill coefficient. Testing of β-alanine on this mutant revealed an EC50 of 1800 µM, a 2.2-fold increase compared with wild-type receptors (Fig. 5.7; Table 5.1). 101 Figure 5.7 Concentration-response curves for mutant Y254F. Above: GABA and β- alanine concentration-response curves for Y254F mutant receptors. Each data point is the mean ± SEM for at least four oocytes. Below: Sample electrophysiological traces. R256 Functional responses from mutant receptors R256K and R256A could not be detected following 50 ng RNA injection, suggesting that binding and/or gating has been disrupted. 5.2.4 Loop D mutants Loop D forms part of the agonist binding site in Cys-loop receptors (Akabas, 2004); in the GABAA receptor an aromatic residue (α1Phe65) has been shown to form part of the “aromatic box,” allowing ligand binding, and in the GABAC receptor, a charged residue (R104) was shown to be critical to agonist binding (Harrison and Lummis, 2006). The homologous residues to these in RDL (Y109 and R111) were mutated to determine their role in ligand binding. 102 Y109 Mutation of Loop D tyrosine residue Y109 to alanine (Y109A) failed to give responses following injection of 50 ng cRNA. Similarly, mutation of the same residue to a serine (Y109S) also failed to give responses. However, mutation to phenylalanine (Y109F) resulted in functional receptors with a GABA EC50 of 398 µM (21-fold increase from wild type), but with no change in Hill coefficient (Fig. 5.8; Table 5.1). Figure 5.8 Concentration-response curve and sample traces for mutant Y109F. Each data point is the mean ± SEM for at least four oocytes. R111 Mutation of Loop D arginine residue R111 to alanine (R111A) or lysine (R111K) failed to give responses following injection of 50 ng cRNA. Since R111 is positioned next to Y109 in a β-sheet region it was suspected that these residues may be interacting with each other via a cation-π interaction. This is a likely interaction, as there is on average 1 cation-π interaction for every 77 amino acids in the protein data bank. Also, it is known that Tyr/Trp···Lys/Arg motifs contribute significantly to stabilising protein secondary structure (Dougherty, 2007). As part of an Arg-X-Tyr motif, this residue is likely to be involved in an interaction with Y109. However, responses could not be detected from R111Y.Y109R injected oocytes, suggesting that functional receptors could not be formed. 103 5.2.5 Loop A mutants A loop A aromatic residue (β2Tyr97) has been shown to be important in the human RDL orthologue, the GABAA receptor, where a cation-π interaction between the ammonium of GABA and β2Tyr97 has been detected (Padgett et al., 2007). Although in my homology model the candidate loop A aromatic residue (F147) was predicted to be quite distant from the ammonium of GABA (See Fig. 5.1), it was thought important to investigate this residue in order to validate my model. F146 F146 was assessed to determine if it is involved in agonist binding and/or receptor activation. Mutation of this phenylalanine residue to alanine (F146A) resulted in a GABA EC50 of 90 µM, with no change in Hill coefficient (Fig. 5.9; Table 5.1). Figure 5.9 Concentration-response curve and sample traces for mutant F146A. Each data point is the mean ± SEM for at least four oocytes. 104 F147 F147 is positioned adjacent to F146 in the β-sheet region of loop A. The side chain of F147 was predicted by my homology model to face towards the central core of the protein, away from the binding site cleft. This residue was mutated to alanine but no responses were detected from this mutant. Table 5.1 Parameters derived from concentration-response curves. Mutant pEC50(µM) ± SEM EC50(µM) nH ± SEM n GABA evoked responses WT 4.705 ± 0.039 19.7 1.8 ± 0.2 19 F206A 4.278 ± 0.063* 53 0.9 ± 0.1* 5 F206Y 6.342 ± 0.026* 0.5 1.7 ± 0.2 8 Y208F 4.149 ± 0.032* 71 1.7 ± 0.3 6 Y208S 3.839 ± 0.029* 47 2.2 ± 0.3 5 F206Y. Y208F 5.447 ± 0.078* 3.6 1.1 ± 0.2 6 E204D N/R - - - E204A N/R - - - S205T N/R - - - S205A N/R - - - Y254F 3.910 ± 0.041* 124 2.3 ± 0.5 5 Y254A N/R - - - R256K N/R - - - R256A N/R - - - Y109F 3.400 ± 0.060* 398 1. 7 ± 0.4 4 Y109S N/R - - - Y109A N/R - - - R111K N/R - - - R111A N/R - - - Y109R. R111Y N/R - - - F146A 4.044 ± 0.061* 90 2.0 ± 0.6 4 F147A N/R - - - β-alanine evoked responses WT 3.096 ± 0.030 801 2.0 ± 0.3 10 F206Y 4.770 ± 0.058* 17 2.0 ± 0.4 4 Y208F 2.567 ± 0.035* 2835 2.6 ± 0.5 4 Y254F 2.745 ± 0.050* 1800 2.1 ± 0.4 6 N/R indicates no response, nH is the Hill coefficient, n indicates number of replicates. * denotes significance from WT (p<0.05, t-test). 105 5.2.6 Probing binding site mutations using the gain of function mutant L314Q The RDL mutation L314Q (M2 14' residue) was reported by Hosie et al. to induce spontaneous channel opening with a 200-fold decrease in GABA EC50 (Biochemical Society Summer Meeting, St Andrews, July 2009). The 14' residue lies close to the 16' residue in the adjacent subunit and he proposed that the interaction of these residues could influence closed state stability. Cysteine substitution of 14' and 16' markedly reduced the amplitude of GABA responses. Reduction of disulphide bonds with DTT restored large amplitude GABA responses suggesting these residues are interacting directly. Spontaneously open channels can be detected by applying picrotoxin (PTX) in the absence of agonist. A rise in the baseline current is indicative of open channels being blocked by PTX. I felt that the L314Q mutation would make a useful reporter for changes in the intrinsic open channel probability of the receptor. By coupling a gain/loss of function mutation to the L314Q mutation, changes in the proportion of open channels, as measured by PTX block, can be used to indicate whether the mutation of interest can affect gating in an agonist independent manner. L314Q Cells expressing L314Q mutant receptors displayed a large leak current. GABA evoked responses recovered very slowly when compared to wild type. PTX induced a rise in the current indicative of channel block and channels recovered slowly requiring several agonist applications (Fig. 5.10). F206Y.L314Q F206Y.L314Q receptors displayed a large leak current and GABA evoked very small or no currents (Fig. 5.10). Mean maximal GABA currents were compared to the amplitude of PTX induced block to give an estimation of the proportion of channels that are spontaneously open (Fig. 5.10). The open channel population for F206Y.L314Q was compared to that for L314Q (Fig. 5.11). This ratio was found to be 2.9, indicating that the introduction of the F206Y mutation increased the proportion of open channels by 2.9-fold. 106 Figure 5.10 Electrophysiological traces for L314Q and F206Y.L314Q mutations. GABA and PTX were applied at EC100 and IC100 concentrations (100 µM). 107 Figure 5.11 F206Y.L314Q open channel proportions. Above: Changes in open channel proportions (PO) with F206Y mutation. Above: Relative amplitudes of maximal GABA currents (B) and PTX induced block (A) for L314Q and F206Y.L314Q mutants were calculated and open proportions were plotted as [A/(A+B)] . Data points are mean ± SEM, n≥10. Data are significantly different (p<0.0001, t-test). Below: Open channel proportions (PO) calculation and Gibbs free energy of gain/loss of function. Where A is I[PTX]Max, B is I[GABAMax], R is the universal gas constant (8.314 J.K -1.mol-1), T is the temperature in Kelvin (°K = °C + 273.15), where mut2 is the double mutant and mut1 is the L314Q mutant. 108 If the increase in PTX-induced block is a direct reflection of an increase in the proportion of receptors in the open state then we can use the relative open proportions for the mutant and wild type receptors to calculate the free energy of gating associated with the mutation of interest. This value is an indirect measure of the effects on intrinsic receptor gating, as introduced by the F206Y mutation (Fig. 5.11). This energy should therefore represent an interaction between mutant residue F206Y and another extracellular domain residue, since no agonist is present in this experiment. The change in Gibbs free energy for activation of mutant F206Y receptors is 8.9 kJ/mol (RTln[38]), as calculated using the 38-fold increase in GABA EC50 (EC50Mut/EC50WT). This value is close to the value for a single hydrogen bond. This finding led to the hypothesis that the mutant residue F206Y is interacting with another residue in the ECD in a fashion that favours receptor activation. Since the phenolic oxygen is capable of such an interaction, a hydrogen bond acceptor residue is a likely candidate for interaction. This prompted a further analysis of the RDL ECD homology model, identifying Loop E residue S176, which is predicted to lie close, within 3.1 Å, of F206 (Fig. 5.12). Figure 5.12 Distance between loop B F206 and Loop E S176. F206 is predicted to lie 3.1 Å from S176. With the addition of the phenolic hydroxyl, as in the mutation F206Y, an interaction, such as a hydrogen bond, may occur between S176 and Y206. 109 The F206Y mutation increased the open proportion of receptors by 2.9-fold. This increase in intrinsic gating conferred by the 206Y mutation does not fully account for the 38-fold gain of function, as revealed by EC50 value for GABA. This suggests that though the addition of a hydroxyl group can confer a gain of function, the intrinsic gating component does not fully account for the much larger decrease in agonist (GABA and β-alanine) EC50 values. This finding therefore supports the hypothesis that F206 forms part of the agonist binding site, possibly via a cation-π interaction with the ammonium of GABA. According to the Del Castillo-Katz receptor activation scheme, receptor activation involves distinct steps of agonist binding and receptor gating (K and Ø) (Fig. 5.13) (Del Castillo and Katz., 1957). This basic theory has been since expanded upon to include many more hypothetical states but its basic premise remains widely accepted (Burzomato et al., 2004; Corradi et al., 2009; Miller and Smart, 2010). The open probability of the channel is therefore the product of both of these hypothetical biophysical properties, binding and gating. If the gain of function from the F206Y mutation is the product of both an increase in binding energy as well as an decrease in gating energy, we can use the Del Castillo-Katz equation to calculate the value of these separate components. The change in binding energy for GABA with the F206Y mutation was thus calculated to be 3.4 kJ/mol (Fig. 5.13). 110 Del Castillo-Katz Figure 5.13 Change in binding energy for the F206Y mutation. Above: Del Castillo-Katz equation. Ø is the gating coefficient. K represents the equilibrium constant for ligand binding steps. While in this scheme only two agonist binding events are displayed, a theoretical maximum of five agonists may bind a receptor. This simple scheme shows how binding of an agonist precedes channel opening. Below: Calculation of binding energy using the Del Castillo-Katz rationale and the Gibbs free energy equation where, R is the universal gas constant (8.314 J.K-1.mol-1), T is the temperature in Kelvin (°K = °C + 273.15), Ø represents the gating energy and K represents the change in binding energy for GABA. 111 Y254F.L314Q Y254F.L314Q mutant receptors displayed a large leak current and significant GABA evoked currents could not be detected (Fig. 5.14). Y254F mutant receptors showed an increase in the intrinsic gating pathway with a theoretical open channel proportion of 100%. Figure 5.14 254F.314Q open channel proportions. Above: Relative amplitudes of maximal GABA currents and PTX induced block for 314Q and 254F.314Q mutants. Data points are mean ± SEM, n≥4. Data are significantly different (p<0.0001, t-test). Below: Relative open channel proportions (PO) of receptors. 112 The change in Gibbs free energy for the Y254F mutation is -4.6 kJ/mol, as calculated using the 6.5-fold increase in EC50 for GABA (Fig. 5.15). Using the Del Castillo-Katz rationale to calculate the change in GABA binding energy with this mutation, taking Ø, the gating coefficient to be equal to the open probability, revealed a change in binding energy of -1.6 kJ/mol (Fig. 5.15). Figure 5.15 Loss in binding energy for 254F mutation. Calculation of binding energy using the Del Castilo-Katz rationale and the Gibbs free energy equation where, R is the universal gas constant (8.314 J.K-1.mol-1), T is the temperature in Kelvin (°K = °C + 273.15). Ø represents the gating energy and K represents the binding energy. Y109F.L314Q No GABA currents could be detected from oocytes injected with Y109F.L314Q mutant cRNA. Additionally no PTX induced change in current could be detected and oocytes displayed a low level of leak current, indicating that receptors were not being expressed or that channels could not be opened. This suggests that these two mutations, although individually both resulting functional receptors, when combined have deleterious effects on receptor expression. This finding would suggest that these mutations disrupt receptor folding or trafficking to the cell surface. 113 Discussion 5.3 Discussion In this chapter I have characterised the agonist binding site of RDL receptors. I have used mutagenesis to determine which residues are critical to ligand binding. I have focused on residues in loops A, B, C and D, regions which are known to contain binding residues in several other Cys-loop receptors. Loop B residue F206 was found to be sensitive to mutation, with an alanine substitution disrupting cooperativity, as evinced by a reduced Hill coefficient, and with a tyrosine mutation greatly increasing agonist sensitivity. This residue is equivalent in location to conserved aromatic residues in 5-HT3, GABAA, GABAC and nACh receptors, where the equivalent residues have been shown to be key binding residues and specifically forming a cation-π interaction with the native ligand in 5-HT3 and GABAC receptors (Beene et al., 2002; Lummis et al., 2005b; Padgett et al., 2007; Thompson et al., 2008; Zhong et al., 1998). Further analysis of F206, by incorporating the L314Q reporter mutation, which causes spontaneous channel opening, revealed that this residue plays a role in both ligand binding as well as channel gating in RDL receptors. The change in binding energy for GABA associated with the F206Y mutation was found to be 3.4 kJ/mol. This value is greater than the difference in cation-π forming ability of tyrosine and phenylalanine (Dougherty, 1996; Mecozzi et al., 1996), suggesting that a direct interaction between the phenolic oxygen of mutant residue F206Y and GABA may occur. This finding fits well with work on the homologous residue in other Cys-loop receptors and suggests that RDL receptors bind GABA in a similar fashion to the binding of ligands in vertebrate Cys-loop receptors. Loop B residue Y208 was found not to be a binding residue, however it is energetically coupled to F206 and plays a role in receptor gating. These results suggest that this region of loop B is involved in receptor activation and is sensitive to mutation. This finding fits well with a recent structure-function study of loop B in the related 5-HT3R, which showed that many residues in 114 loop B are essential for receptor function via both structural stabilisation of loop B and direct ligand binding (Thompson et al., 2008). The position of Y208 as predicted by my homology model, facing into the hydrophobic core of the extracellular domain, is consistent with such a structural stabilisation. The Loop B residue E204 was found to be critical to receptor function. Currents could not be recorded from mutants E204D or E204A. Since neither of these mutants responded, it seems likely that this residue is important for receptor activation or receptor folding. E204 is situated close to F206, which is important in receptor activation. It thus seems probable that this region of Loop B is sensitive to mutations and those residues likely to be involved in structural roles, such as charged residues, cannot be mutated without deleterious effects. Mutation of the homologous residue in the 5-HT3R (a threonine) to either an alanine or a serine resulted in non-functional receptors, although receptors expressed and formed a binding site and radioligand binding studies (using an antagonist) showed a six-fold increase in binding affinity for the alanine mutation and no change for the serine mutant when compared to wild type receptors (Thompson et al., 2008). This suggests a critical role for this residue in receptor activation. My results are therefore similar to those of Thompson et al. (2008), since no functional receptors could be detected. Docking simulations from the previous chapter show the ammonium of GABA to be positioned 6.7 Å from E204 (Fig. 5.16). Although the distance here exceeds the predicted length of a salt-bridge of ~4 Å (Kumar and Nussinov, 2002) this finding should leave open the possibility for such an interaction. 115 Figure 5.16 Electrostatic polarity of GABA within the binding site. GABA docked with the carboxylate group 3.7 Å from Loop D residue R111 and with the ligand ammonium 6.7 Å from Loop B residue E204. Loop B residue S205 is one of very few residues conserved across the Cys-loop receptor family (See Figure 1.2). It is thus unsurprising that mutations made to this residue are not tolerated. This residue is located next to F206, which has proven to be a key determinant of agonist sensitivity. It is probable that mutations at position 205 disrupt the structure of Loop B, since even the conservative mutant S205T failed to respond. Mutation of the homologous residue in the 5-HT3 receptor (also a serine) to both alanine and threonine resulted in a two and four-fold decrease in binding affinity respectively, with only the threonine mutant showing functionality, with a three-fold increase in EC50. Further investigations into this residue should utilise unnatural amino acids which can monitor very subtle changes in side chain chemistry without the deleterious effects observed here. Loop C residue Y254 was found to be important in ligand binding. Mutation of Y254 to alanine (Y254A) resulted in non-functional receptors. This is unsurprising since Y254 is a conserved residue across many Cys-loop receptors, suggesting an important functional or structural role. The Y254F mutation however, resulted in an increase in GABA EC50, suggesting that subtle changes at this location can affect receptor function. Additionally, this result supports the hypothesis that the tyrosine phenolic hydroxyl group is involved in ligand binding and/or receptor activation. Analysis with the 116 reporter mutation L314Q revealed that the energy of binding between GABA and Y254 is less than a single hydrogen bond, suggesting that a cation-π interaction may occur here. Furthermore, docking simulations in the previous chapter, placed the ammonium of GABA between F206 and Y254 (Fig. 5.17), suggesting that Y254 may well form a cation-π interaction with the ammonium of GABA. This would be similar to the interaction observed in the C.elegans 5- HT-gated channel MOD-1 (Mu et al., 2003). Proximal Loop C residue R256 was also found to be critical to receptor activation and/or formation. Since this residue was not prominent in docking simulations carried out in the previous chapter, I feel that this residue is more likely to be involved in receptor structure rather than ligand binding. However, in a study of the homologous residue in the GABAC receptor, R249 was substituted with alanine, glutamate, or aspartate with relatively small (4–30-fold) increases in GABA EC50 (Harrison and Lummis, 2006). This would suggest that this residue may play different roles in different receptors, with it perhaps being more important in RDL. Figure 5.17 Aromatic box residues. Above: Wire diagram of the binding site with aromatic box residues Y109, F206 and Y254 in space filling representation, with GABA docked. Below: Close up view of GABA docked with the ammonium group sandwiched between the faces of aromatic residues F206 and Y254. 117 Mutation of Loop D residue Y109 to alanine resulted in non-functional receptors and the conservative phenylalanine mutation resulted in a large rise in GABA EC50. This residue is thus critical to GABA sensitivity. Indeed, the homologous residue in the GABAA receptor (α1Phe65) has been shown to be involved in ligand binding by increasing the “general hydrophobicity of the region,” showing a requirement for aromaticity at this position (Padgett et al., 2007). However another study of the same residue in GABAA receptors showed that mutation to leucine increased the GABA EC50 from 6 to 1260 µM, with the IC50 values of bicuculline and SR95531 (competitive antagonists) increasing by similar amounts (Sigel et al., 1992). This would suggest an important role in ligand binding. The position of this residue in the ECD homology model supports the hypothesis that this residue forms part of the aromatic box in this receptor, providing a hydrophobic surface for water exclusion and ligand stabilisation close to the predicted GABA binding location (Fig. 5.17). The adjacent residue R111 was found to be a key binding residue in RDL receptors with R111K and R111A mutants failing to respond. R111 is predicted to lie 3.7 Å from the carboxylate of GABA, based on docking simulations (Fig. 5.16). A salt bridge interaction is likely to occur if the simulation is accurate. Furthermore, it has been shown in the GABAC receptor that the homologous residue (R104) is critical to receptor activation: when R104 is substituted with alanine or glutamate an increase in GABA EC50 >10,000-fold was observed. In the same study, molecular modelling indicated a role of this residue in binding GABA (Harrison and Lummis, 2006). Loop A aromatic F146 was found not to be a binding residue in RDL receptors. Referring to the model predicted that the ammonium of GABA is 9.7 Å from the aromatic side chain of F146 (Fig. 5.18). The model is consistent with the observed experimental evidence in this chapter and suggests that this residue is not close enough to interact directly with GABA. Conversely however, this residue is of central importance to GABA binding in the GABAA receptor, where the equivalent aromatic residue (β2Tyr97), is involved in a cation-π interaction with the ammonium of GABA (Padgett et al., 2007). This would 118 suggest that GABA binds in a different position in RDL than in GABAA receptors. Adjacent residue F147 is predicted to face into the hydrophobic core of the extracellular domain (Fig. 5.18). Alanine substitution (mutation F147A) resulted in nonfunctional receptors. It is possible that the alanine substitution alters the tertiary structure of the β-sheet backbone of Loop A resulting in misfolded protein. The hydrophobic nature of this residue may be important in maintaining receptor structure or correct receptor folding. Figure 5.18 Loop A aromatic residues F146 & F147. GABA is positioned in its predicted binding orientation with its distance from F146 labeled. 119 Conclusion 5.4 Conclusions RDL receptors are sensitive to mutations in Loop B, Loop C and Loop D aromatic residues (F206, Y254 and Y109 respectively), suggesting that these residues form the aromatic box involved in stabilising ligand binding. Furthermore, Loop B aromatic residues F206 and Y208 are energetically coupled, suggesting an interaction during receptor activation. However, Loop A aromatic residues are unlikely to be involved in GABA binding. Loop B glutamate and serine residues E204 and S205 may play important structural roles as well as being directly involved in ligand binding. In particular, the negatively charged side chain of E204 is likely to be involved in a salt-bridge interaction with the positive ammonium of GABA. A Loop D arginine residue (R111) is also critical to RDL receptor function and modelling and docking experiments predict a direct role for this residue in GABA binding. The mutation F206Y causes a large gain of function and this residue is predicted to form a cation-π interaction with the ammonium of GABA. Additionally, Loop C aromatic Y254 is sensitive to mutations and is predicted to interact with GABA, also via a cation-π interaction. Molecular modelling supports this hypothesis and this should be further tested using unnatural amino acid mutagenesis. 120 Chapter 6 Characterisation of Ginkgo biloba extracts on RDL receptors ______________________________________________ 121 Introduction 6.1 Introduction The Ginkgo biloba tree has been used in Chinese medicine for over 2,500 years (Drieu et al., 2000) and is amongst the most used herbal medicines today. Ginkgo biloba extracts contain flavonoids (22-24%) and terpene trilactones (6- 8%) (DeFeudis and Drieu, 2000). Ginkgo biloba extracts are also potent insecticides, e.g. bilobalide has an LD50 of 0.26 ng/insect when tested on planthoppers, while ginkgolide A and ginkgolide B had values of 64 and 16 ng/insect respectively (Ahn, 1997). These terpene trilactone compounds have been shown to be non-competitive antagonists of vertebrate glycine and GABAA receptors with IC50 values in the low micromolar range (Ivic et al., 2003; Huang et al., 2004; Kondratskaya et al., 2004). Ginkgolides and bilobalide are therefore pharmacologically similar to picrotoxin (PTX) a non- competitive antagonist of GABAA and glycine receptors (Ivic et al., 2003; Li and Slaughter, 2007). PTX has been well characterised as binding close to the 2' channel-lining residue (Chen et al., 2006; Das and Dillon, 2005; Olsen, 2006). Similarly, molecular modelling of the glycine receptor pore has yielded insight into the binding parameters of ginkgolide B, implicating a role for the 6' and 2' channel-lining residues (Hawthorne et al., 2006; Heads et al., 2008). The aim of this study was to assess the potency of ginkgolide A, ginkgolide B, bilobalide and picrotoxin (Fig. 6.1) on RDL receptor function and to determine the role of the 2' and 6' channel-lining residues. 122 Figure 6.1 Structures of picrotoxin (picrotoxinin) and ginkgolides. Ginkgolide A and ginkgolide B differ only by a single atom, the R1 group being a hydrogen atom and hydroxyl group respectively. Ginkgolides, bilobalide and picrotoxinin are all caged compounds. 123 Results 6.2 Results 6.2.1 Ginkgolides are antagonists of RDL receptors Co-application of ginkgolides and bilobalide with GABA resulted in diminished currents (Fig. 6.2), while pre-application (30 µM for 30 s) had no effect on subsequent GABA responses. Concentration-inhibition curves were prepared (Fig. 6.2) and IC50 values were 0.95 µM, 0.69 µM, 0.25 µM and 1.1 µM for GA, GB, BB and PTX respectively (Table 6.2). Following ginkgolide inhibition (10 µM antagonist), receptors recovered slowly, with the effects of inhibition not fully reversed after twenty minutes of washout and four GABA applications (EC50) (Fig. 6.3). Fig. 6.2 Ginkgolides are antagonists of RDL receptors. Concentration-inhibition curves and sample electrophysiological traces showing the effects of ginkgolide A (GA), ginkgolide B (GB), bilobalide (BB) and picrotoxin (PTX) on GABA (EC50) evoked currents. GABA evoked responses are diminished in the presence of all compounds. 124 Figure 6.3 RDL receptor recovery following ginkgolide inhibition. Above: Recovery time of wild-type RDL receptors following GA, GB, BB and PTX inhibition. PTX, BB, GB and GA inhibition (10 µM antagonist) was not fully reversed following four applications of GABA [EC50] and twenty minutes of saline washout. Each data point is mean ± SEM from at least three oocytes. Below: Electrophysiological traces showing recovery from inhibition. 125 6.2.2 Mutant receptors are resistant to antagonists Mutant receptors A2'S, T6'V and double mutant A2'S T6'V formed functional receptors with EC50 values of 23 µM, 75 µM and 21 µM respectively (Fig. 6.4) (Table 6.1). Both T6'V and A2'S mutant receptors showed decreased sensitivity to all compounds: IC50 values for T6'V were > 100 µM, 260 µM, > 1 mM and 1.1 mM for GA, GB, BB and PTX respectively; IC50 values for A2'S were 2.8 µM, 10.1 µM, 160 µM and 260 µM for GA, GB, BB and PTX respectively (Table 6.2; Fig. 6.5). Double mutant A2'S T6'V was resistant to all compounds, with no effects observable with 100 µM antagonist. A2'S and T6'V mutant receptors also displayed accelerated recovery from antagonists when compared to wild type. Following antagonist withdrawal, A2'S and T6'V mutant receptors recovered fully with a subsequent GABA application. Figure 6.4 Concentration-response curves for mutants used in this study. Each data point is the mean ± SEM of at least three different oocytes Table 6.1 Parameters derived from concentration-response curves. Mutant EC50 ± SEM EC50 (µM) nH ± SEM n WT 4.71 ± 0.04 19.7 1.8 ± 0.2 19 A2'S 4.64 ± 0.06 22.7 1.9 ± 0.4 4 T6'V 4.13 ± 0.06* 74.6 1.4 ± 0.3 9 A2'S.T6'V 4.69 ± 0.06 20.6 2.1 ± 0.5 5 nH is the Hill coefficient, n indicates number of replicates. * denotes significance from WT (p<0.05, t-test). 126 Figure 6.5 Ginkgolide inhibition of mutant RDL receptors. Above: Concentration- inhibition curves prepared from mutant receptor responses. Antagonist was co-applied with GABA (EC50 concentration) and the inhibited response was normalised to response in the absence of antagonist. Each data point is the mean ± SEM of at least three different oocytes. Below: Electrophysiological traces showing the effects of 1 µM ginkgolide A (GA), ginkgolide B (GB), bilobalide (BB) and picrotoxin (PTX) on GABA (EC50) evoked currents. A2'S and T6'V mutant receptors show resistance to these compounds. 127 Table 6.2 Parameters derived from concentration-inhibition curves. Antagonist pIC50 ± SEM IC50 (µM) nH ± SEM n Wild-type receptors PTX 5.976 ± 0.165 1.1 0.5 ± 0.1 3 GA 6.024 ± 0.371 0.95 0.5 ± 0.2 3 GB 6.118 ± 0.222 0.69 0.7 ± 0.3 4 BB 6.610 ± 0.167 0.25 0.7 ± 0.2 4 A2'S receptors PTX 3.593 ± 0.082* 260 1.1 ± 0.2 4 GA 5.555 ± 0.052 2.80 1.3 ± 0.2 3 GB 4.997 ± 0.111* 10.1 0.7 ± 0.2 4 BB 3.800 ± 0.787* 160 0.5 ± 0.2 3 T6'V receptors PTX 2.961 ± 0.185* 1090 0.9 ± 0.2 5 GA ≤4 - - 3 GB 3.588 ± 0.362* 260 1.0 ± 0.7 4 BB ≤3 - - 4 nH is the Hill coefficient, n = number of replicates. * denotes significance from wild-type (p<0.05, unpaired t-test). 6.2.3 Mutant cycle analysis Mutant cycle analysis was carried out using IC50 values arrived at from concentration-inhibition curves. By assessing the effect of changing a single chemical group on the ligand both on wild-type and mutant receptors we can determine how coupled these two atomic groups are by calculating a coupling coefficient. Coupling coefficients suggest a positive interaction between the R1 group of ginkgolide B and both the 2' and 6' residues. Interaction energies (∆∆G.int = -RTlnΩ) of 3.9 and 4.8 kJ.mol-1 for 2' and 6' residues respectively suggest a non-covalent interaction between ligand and receptor at this position (Fig. 6.6). This prediction at 6'T is in line with the expected energy of a single hydrogen bond (5-30 kJmol-1). 128 Figure 6.6 Mutant cycle analysis. For the 6' threonine residue the coupling energy (∆∆G.int = -RTlnΩ) was 4.8 kJmol-1. For the 2' alanine it was 3.9 kJmol-1. 6.2.4 Molecular modelling and docking Ginkgolide A and ginkgolide B docked into the channel close to the 6'T residue (Fig. 6.7). Both compounds docked towards two of the M2 helical bundles. Hydrogen bonding with the carbonyl oxygen of the 6'T residue was detected for Ginkgolide A. Ginkgolide B docked into the same location in a similar orientation but the same H-bonding interaction with this residue was not detected. Picrotoxinin and bilobalide also docked within 5 Å of the 6' and 2' channel-lining residues of contiguous subunits; however no specific interactions were detected. In A2'S receptors, both ginkgolide A and ginkgolide B showed hydrogen bonding interactions with the 2'S residue whilst hydrogen 129 bonding at the 6'T residue was absent. Furthermore, both ginkgolide compounds docked in the same orientation at 2'S receptors, orientations distinct from that observed at wild type receptors, with both compounds inverted and lower in the channel. In T6'V receptors no hydrogen bonding interactions were observed for any compounds. 130 Figure 6.7 Homology models of the RDL transmembrane domain and docking simulations. A homology model of the full RDL pentamer based on Gloebacter violaceus (GLIC) (PDB: 3EAM). Two subunits were removed for the purpose of this figure, revealing the binding orientation within the pore. M2 transmembrane regions are α-helical (red, grey and green; denoting separate subunits). Above: GA, GB, BB and PTX docked into WT, A2'S and T6'V mutant receptors. Below: Close up view of GA and GB binding orientations in WT and A2'S mutant receptors, where hydrogen bonds were detected. 131 6.2.5 Insect bioassays Bioassays were carried out by Mirel Puinean and Martin Williamson at Rothamsted Research, Hertfordshire. After 48 h with an application of 1000 ppm compound, two out of four compounds (BB and GA) produced 100% mortality. The difference between these compounds and GB and PTX could be observed after 1 h when most of the individuals dosed with BB and GA, although alive, had their hind legs paralyzed while the insects dosed with GB and PTX, although affected, could still move all their legs. However, after 48 h at this high dose 100% of the individuals were killed (p<0.05, Student’s t-test) (Fig. 6.8). A further bioassay using 50 ppm solutions of the tested compounds produced a better resolution between their toxicity. BB and GA emerged as the most toxic with 100% mortality (p<0.05, Student’s t-test). GB and PTX were significantly less potent yielding 60% and 34% mortality respectively, and significantly different to control treated insects (p>0.05, Student’s t-test). Figure 6.8 Whole insect toxicity bioassays for GA, GB, BB and PTX on N.lugens. Each experiment consisted of ten insects being dosed with compound. Data is mean ± SEM for three experiments. 132 Discussion 6.3 Discussion I have shown that ginkgolide A, ginkgolide B and bilobalide are antagonists of RDL receptors with IC50 values in the sub-micromolar range. Reversal from inhibition is slow, requiring >20 min to wash out. This distinguishes the ginkgolides from picrotoxin, which washes out within 10 min. This behaviour of ginkgolides differs from that reported at glycine receptors, where recovery is observed following one minute of washout (Ivic et al., 2003). An investigation into the role of the 6'T and 2'A channel-lining residues using mutagenesis studies revealed that these residues are critical to the ginkgolide sensitivity of RDL receptors, suggesting that ginkgolides bind in the channel close to these residues. Modelling of the transmembrane domain of homomeric RDL receptors reveals that the 6'T and 2'A residues are amongst the channel-lining M2 residues which face into the channel pore. Docking of ginkgolides into the channel identified a hydrogen bond interaction between ginkgolides A and B and the hydroxyl oxygen of the 6'T residue. This binding location is supported by mutant cycle analysis which suggested an interaction between ginkgolides and the 6'T. This binding site is similar to that reported for ginkgolide binding in the glycine receptor pore (Hawthorne et al., 2006; Heads et al., 2008). Bilobalide also docked close to the 6'T but no specific interactions were observed. Docking of these compounds into the A2'S mutant receptor revealed hydrogen bonding between ginkgolides and the 2'S hydroxyl, while hydrogen bonding with the 6'T residue was absent. T6'V mutant receptors showed no hydrogen bonding interactions, suggesting that this mutation disrupts binding and that this residue may be the principal attachment point for ginkgolides. Since the 6'T residue and the mutant A2'S residues have hydroxyl groups on their side chains and both of these residues showed H-bonding interactions in the docking simulations, it may be that interactions with these side chains is a determinant of ginkgolide binding. Indeed, the A2'S “resistance to dieldrin” mutation 133 caused ginkgolides to dock lower in the channel, in what may be a lower affinity binding site. Such an altered binding position may underlie the decreased ginkgolide sensitivity conferred by this mutation. Picrotoxin inhibits both cation-selective (nACh and 5-HT3 receptors) and anion-selective (GABAA, GABAC, GlyR, and GluCl receptors) receptor channels (Das et al., 2003; Erkkila et al., 2004; Etter et al., 1999; Schmieden et al., 1989; Zhang et al., 1995a). Picrotoxin is an equimolecular complex of picrotoxinin and picrotin (Curtis and Johnston, 1974) and picrotin is inactive in inhibiting GABAARs and GABACRs, indicating that the inhibitory effect of picrotoxin is related to the picrotoxinin component (Curtis et al., 1974; Qian et al., 2005). Conversely picrotoxinin and picrotin are equally potent in inhibiting α1 homomeric GlyR activation (Lynch et al., 1995). RDL receptors are also inhibited by picrotoxin (Hosie et al., 1995), however the blocking effect at RDL receptors is mediated solely by picrotoxinin as RDL receptors are resistant to picrotin (Shirai et al., 1995). This finding shows RDL receptors to be more GABAA-like than GlyR-like in terms of picrotoxin sensitivity. So far, ginkgolides have only been shown to inhibit anion selective receptors (Ivic et al., 2003; Huang et al., 2004; Kondratskaya et al., 2004) and more recently Jensen et al. (2010) described how ginkgolide X, a synthetic ginkgolide derivative, is selective for glycine receptors, and suggested a distinct binding mode for ginkgolides at anionic receptors. The ginkgolides and picrotoxin block RDL receptors with similar IC50 values to those at vertebrate GABAA receptors (Huang et al., 2004; Krishek et al., 1996b). What is surprising, however, is that picrotoxin is a convulsant but the ginkgolides are not, and instead have neuro-protective, anxiolytic and other beneficial properties (Ahlemeyer and Krieglstein, 2003; Mdzinarishvili et al., 2007; Zhu et al., 2004). The reasons for this are still unclear, but may include differences in bioavailability as well as differences in insect and human GABA receptor structure. Other properties of these compounds include symptomatic relief from Alzheimer disease and a reduction in the effects of dementia (DeFeudis and Drieu, 2000). Furthermore, a study on human volunteers, using 134 a computer-analysed electroencephalogram, showed that Ginkgo biloba extract (EGb) has effects similar to those of cognitive enhancers as well as tacrine (an anticholinesterase), a marketed antidementia drug currently available in the United States (Itil et al., 1996). A more established property of Ginkgo biloba extract is its potent insecticidal activity. For many decades leaves from the Ginkgo biloba tree have been used in books to prevent insect activity, and more recent studies to determine potency have shown excellent activity on a range of insects (Sun et al., 2006; Ahn, 1997). One potential mode of action is via insect GABA receptors, as many insecticidal compounds are known to exert their effects through these proteins. My data showing moderate potency of ginkgolides at RDL receptors, and perhaps more importantly, slow reversibility, indicates that these receptors could indeed be the insecticidal target of these compounds. This would be a mechanism of action that fits well with reported potencies (Sun et al., 2006; Ahn, 1997), with bilobalide being the most potent insecticide and also the most potent RDL receptor antagonist. Insecticides which act on GABA receptors such as hexachlorocyclohexanes, polychloroboranes and chlorinated cyclodienes (such as lindane, toxaphene and endosulfan respectively) were used widely as pesticides before their GABAergic mechanism of action was revealed (Casida, 1993). Fipronil is still one of the most common insecticides (used mainly for pest control), with 119,000 lbs being used in California in 2006 (http://cdpr.ca.gov/). Identifying compounds which selectively block insect GABA receptors could facilitate novel insecticide design. The environmental and economic benefits of developing bio-organic insecticides which poses little toxicological threat to humans are great. Selective toxicity is the ultimate goal of the pesticide industry, whilst ever progressively, environmentalism is in the sights of modern industry. For these reasons, I deem Ginkgo biloba extracts to be an underexploited bio-organic alternative to current pesticide strategies. This study provides chemical insight into their action which will facilitate further development of these “green” insecticides. 135 Conclusion 6.4 Conclusions Ginkgolide A, ginkgolide B and bilobalide are antagonists of RDL GABA receptors and recovery from inhibition is slow. These compounds block the channel of RDL receptors, binding close to the 2' and 6' residues. Interaction energies between ginkgolides and their binding residues are in the region of a single hydrogen bond and the expected interaction includes a hydrogen bond with the 6'T hydroxyl side chain. I hypothesise that the blocking action of ginkgolides at RDL receptors is the mechanism underlying their potent insecticidal activity. 136 Chapter 7 Single-channel analysis of heteromeric 5-HT3 receptors ______________________________________________ 137 Introduction 7.1 Introduction 5-HT3 receptors are 5-hydroxytryptamine-gated Cys-loop receptors that are largely sodium selective. Receptor activation results in a rapidly activating and then desensitising inward current resulting in depolarisation of the cell. Thus far, five 5-HT3 receptor genes have been identified (subunits A-E), and functional receptors containing each of these have been expressed (Davies et al., 1999; Niesler et al., 2003; Niesler et al., 2007). Only A subunits can form functional homomeric receptors, but other subunits can combine with A subunits to form heteromeric complexes. Isoforms of the human A-subunit have also been described, and recently two novel transcriptional variants of the B subunit (Brain-1 and Brain-2) have also been cloned (Bruss et al., 2000a; Bruss et al., 2000b; Tzvetkov et al., 2007). RNA for these B variants has been quantified using real-time RT-PCR showing that the B subunit is abundantly expressed in the human brain as well as in the colon and small intestine (Fig. 7.1). Tzvetkov et al. (2007) reported that in the brain less than 1% of the 5- HT3B subunit RNA coded for the conventional B-subunit, while the remaining B-subunit RNA was accounted for by approximately 75% Brain-2 and 24% Brain-1. Protein sequences have been predicted from the alternative transcripts, and compared to the canonical 5-HT3B subunit (Tzvetkov et al., 2007) (Fig. 7.1). The Cys-loop, four transmembrane domains (designated TM1 to TM4) and the HA-stretch are common to all three 5-HT3B isoforms. The HA-stretch is responsible for the higher conductance of 5-HT3B-containing heteromeric receptors (Kelley et al., 2003) (Fig. 7.2). The N-terminal localisation signal and the β1–β2 loop structure mediating channel gating (Reeves et al., 2005) are missing in the Brain-2 isoform. The six amino acids of the N-terminus of the Brain-1 isoform differ from the canonical form. 138 Figure 7.1 5-HT3R alternative transcripts. Above: Relative abundance of B, brain-1 (Br1) and brain-2 (Br2) transcripts in brain and gut. Below: Br1 and Br2 secondary structure. The part of the localization signal that differs between the intestinal and Brain-1 form is marked by hatching. Potential glycosylation sites are shown in bold. Figure taken from Tzvetkov et al., 2007, with copyright permission from Gene. Unpublished data from our lab suggests that Br1 and Br2 can each co-assemble with the A subunit, forming functional receptors in Xenopus oocytes with distinct agonist sensitivities and desensitisation kinetics. When A and Br1 cRNA is injected; and when A and Br2 cRNA is injected, receptor responses 139 show distinct EC50 values (with EC50 AABr2>ABr1>AB), suggesting that both Br1 and Br2 subunits can be incorporated into receptors with the A subunit. A previous study has reported that A-homomers have a conductance of 0.4 pS whilst A-B heteromers have a much higher conductance of 16 pS (Davies et al., 1999), highlighting the functional significance of the B subunit. This difference is due to the presence of four intracellular arginine residues, which if mutated in the A subunit to the B subunit equivalent residues, increases the conductance close to that of A-B heteromers (Davies et al., 1999) (Fig. 7.2). Figure 7.2 Four intracellular arginine mutations located in the M2-M3 intracellular loop confer high conductance to the B subunit. Taken from Davies et al., 1999 with copyright permission from Nature. Since the conductance of A homomers is too low to detect using single-channel electrophysiology, identifying A-B heteromers is facilitated by the fact that only A-B heteromers have a conductance high enough to be detected. The aim of this study was to determine whether Br1 and Br2 subunits form functional receptors in HEK293 cells and to determine the conductance of these different receptor channels. 140 Results 7.2 Results 7.2.1 Single-channel recordings Channels detected from whole-cell attached patches of HEK293 cells expressing human 5-HT3AB receptors opened in bursts interrupted by brief sojourns in the closed state. Bursts were typically 20-200 ms in duration although brief single openings were also observed (Fig. 7.3). Several hundred events were recorded and current voltage relationships were plotted using the mean amplitude for channels at each voltage. Conductance was calculated using the Nernst equation and was found to be 30 ± 1.2 pS (Fig. 7.4). ABr1 channels were measured using the same method. Channel events were similar to AB receptors with typical bursts of ~100 ms and with similar kinetic behaviour (i.e. brief open-closed time and burst frequency) (Fig. 7.3). Conductance was 33 ± 1.1 pS (Fig. 7.4). No channels could be detected from ABr2 transfected cells. Transfection parameters were varied, including varying the ratio of A to Br2 DNA, varying the total DNA concentration used and also using different DNA samples. Nonetheless, no Br2 currents could be detected. Whole-cell macroscopic currents of cells confirmed the expression of 5-HT3 receptors (data not shown) but no single-channel events could be detected. 141 Figure 7.3 Sample traces and amplitude histograms for single-channel 5-HT3 AB and ABr1 receptors expressed in HEK293 cells. Each histogram represents data from at least three patched cells. 142 Figure 7.4 Current-voltage (IV) relationships for AB and ABr1 channels expressed in HEK293 cells. Receptors show identical conductance. Data is mean ± SEM for at least three cells. Data are not significantly different (p>0.05, unpaired t-test). 7.2.2 Flexstation analysis of heteromeric 5-HT3 receptors expressed in HEK293 cells HEK293 cells stably expressing 5-HT3A receptors were transfected with B, Br1 and Br2 receptor subunit DNA using the microporator system (DigitalBio). Concentration-response data were prepared for each receptor type (A, AB, ABr1 and ABr2) and relative fluorescence was plotted, yielding concentration- response curves (Fig. 7.5). Data for each receptor type was collected from at least five separate transfections. Maximum fluorescence was decreased for heteromeric receptors compared to control (A homomeric) receptors (Fig. 7.5). EC50 values were 0.36 µM, 0.26 µM, 0.53 µM and 0.46 µM for A, AB, ABr1 and ABr2 receptors respectively, not significantly different (p>0.05, t-test) and Hill coefficients were 0.49, 1, 1 and 0.9 for A, AB, ABr1 and ABr2 receptors respectively (Figure 7.5; Table 7.1). 143 Figure 7.5 Concentration-response curves for homomeric and heteromeric 5-HT3 receptors expressed in HEK293 cells stably expressing 5-HT3A homomers. Above: Concentration-response curves. Middle: Sample traces from Flexstation (0.33 µM and 100 µM 5-HT). Below: Maximum relative fluorescence (RFU) for 5-HT-evoked responses in homomeric A (control) and heteromeric AB, ABr1 and ABr2 receptors. 144 Table 7.1 Parameters derived from concentration-response curves. Subunit Composition pEC50 (µM) mean ± SEM EC50 (µM) nH mean ± SEM RFU Max mean ± SEM n A 6.447 ± 0.17 0.36 0.5 ± 0.1 151.5 ± 9.6 15 AB 6.580 ± 0.07 0.26 1.0 ± 0.1* 125.1 ± 6.8* 15 ABr1 6.280 ± 0.06 0.53 1.0 ± 0.1* 110.9 ± 4.5* 15 ABr2 6.336 ± 0.07 0.46 0.9 ± 0.1* 110.5 ± 5.1* 15 nH is the Hill coefficient, n indicates number of replicates. pEC50 values are not significantly different. * denotes significance from A homomers (where p<0.05, unpaired t-test). 145 Discussion 7.3 Discussion 5-HT3 receptor antagonists are currently used for the treatment of post- operative, radiotherapy- and chemotherapy-induced nausea and vomiting as well as irritable bowel syndrome. It is expected that 5-HT3 antagonists will have wider therapeutic applications in the future, thus it is important to understand the consequences of subunit composition on the pharmacology and physiology of these receptors (Thompson and Lummis, 2007; Thompson et al., 2007). With the higher conductance conferred by the addition of the B subunit and with its abundant expression levels (Tzvetkov et al., 2007), B subunits are critical to brain 5-HT3 signalling. Single-channel currents from AB heteromers were previously reported to be 16 pS (Davies et al., 1999; Kelley et al., 2003). My findings here of 30 pS are higher than these finding and may be due to different experimental conditions. More importantly, however, ABr1 channels showed similar behaviour and identical conductance to AB channels, confirming that they can be expressed in HEK293 cells and form functional channels with similar properties to AB receptors. This is an important finding, confirming the functionality of this abundantly expressed brain 5-HT3 subunit. Since the Br2 is missing such a large amount of sequence, it is unsurprising that functional channels were not observed with single-channel electrophysiology. This finding would suggest that either Br2 subunits can be expressed in oocytes but not HEK293 cells or that our previous findings are in fact an experimental artefact. Attempts to further validate these findings using the Flexstation yielded ambiguous results. The EC50 value arrived at for A homomers (0.36 µM) is close to a previously published value of 0.2 µM (Price and Lummis, 2005), where the same fluorometric assay was used. Other studies using electrophysiological methods have published EC50 values of 2.1 µM and 1.4 µM for the same receptors expressed in HEK293 cells and Xenopus oocytes respectively (Reeves et al., 2001; Spier et al., 2000). Whilst there was found to be no change in EC50 for stable A cells transfected with the different B 146 subunits, there was a significant increase in Hill coefficient and a decrease in the maximum response (RFU-Max). This is in contrast to that reported when these receptors are expressed in Xenopus oocytes where the opposite is observed - a decrease in Hill coefficient from 2.2 to 1.1 when the B subunit is incorporated into receptors (Lochner and Lummis, 2010). Additionally, since the B subunit has a higher conductance than the A subunit (Kelley et al., 2003), an increase in RFU-Max may be expected for heteromeric receptors, although we cannot be sure if this is the case. Therefore this observed decrease in RFU- Max may be either an experimental artefact or a consequence of decreased open probability in heteromeric channels. Of course, transfections may not have been successful, but the significantly different Hill coefficients would suggest otherwise. As our preliminary work on Br2 receptors (in oocytes) yielded an A-like concentration-response profile (i.e. similar Hill coefficient and EC50), it seems likely that this is indeed an experimental artefact and that the Br2 subunit does not form functional channels. Furthermore, since this subunit is missing extracellular loops D and A as well as the β1-β2 loop, which is essential for gating (Reeves et al., 2005), it is unlikely to be involved in the formation of a functional 5-HT binding site. Despite this finding, the high abundance of Br2 transcripts detected in the brain warrant further investigation of the role of Br2. Perhaps this transcript plays an intracellular role in regulating receptor transcription and/or surface expression. Such an explanation could account for our previous results in oocytes, where a change in EC50 was observed for ABr2 receptors. Br2 cRNA may have altered the formation and/or regulation of homomeric A receptors. On the other hand, expressed Br2 subunits may have been incorporated into receptors with a stoichiometry too low to increase channel conductance (perhaps one Br2 subunit per receptor), but enough to affect the concentration-response profile. Further studies could determine if Br2 subunits reach the cell surface using immuno-staining or with conjugated fluorescent tags. 147 Conclusion 7.4 Conclusions The 5-HT3 Br1 subunit can form functional heteromeric channels with the A subunit and these heteromeric channels have a conductance of 30 pS, identical to AB receptors. The Br1 subunit is therefore an important component in brain 5-HT3 signalling. The Br2 subunit does not form functional channels with the A subunit in HEK293 cells and is unlikely to form functional 5-HT3 receptors. However, the high abundance of the Br2 transcript suggests that it may play an intracellular role in receptor regulation and/or surface expression and further experiments on Br2 should focus on these aspects. 148 Chapter 8 Future directions and final remarks ______________________________________________ 149 Future directions The aim of this thesis was to investigate the structure and function of Cys-loop receptors and to generate a greater understanding of these receptors in general. I began by assessing the biophysical properties of RDL receptors. I followed on from this by assessing the efficacy of a range of agonists on RDL receptors, thereby identifying the critical determinants of agonist binding. Following on from this work, I identified several residues in loops B, C and D that are involved in ligand binding. I have also identified the binding location of ginkgolides and bilobalide in the channel pore. Finally, during a visit to Cecilia Bouzat’s lab in Bahía Blanca, Argentina, I made single-channel recordings of heteromeric 5-HT3 receptors, confirming the functionality of brain variant subunit Br1. The work on biophysical properties of RDL receptors has confirmed that these receptors express quickly (within 24 h) in Xenopus oocytes and that they are chloride selective channels. Further work on this property could include mutagenesis of the -1' residue (alanine) to a glutamate - the equivalent residue in the cation-selective nAChR and 5-HT3R - to determine if the ion selectivity filter is in the same location in RDL receptors. RDL receptors showed resistance to changes in pH. This insensitivity to changes in pH distinguishes these receptors from their vertebrate orthologues, the GABAA receptors, and this difference may be due to the absence of charged residues involved in receptor gating. Future work could involve mutagenesis of charged residues in the pre-M1 and β1-β2 loop regions, since these regions contain many key gating residues (Bartos et al., 2009a). This may lead to a greater general understanding of pH sensitivity in Cys-loop receptors. In Chapter 4 I identified the critical features of agonists that bind to and activate RDL receptors; a charged amine and an anionic centre are required for agonists to bind. Additionally, there is a size requirement for agonists, with the 150 most potent agonists being ~5 Å in length. This work provides a set of structure activity relationship (SAR) data which could be used to generate potential antagonists. Further work could include the generation of a large set of GABA analogues, which may lead to the identification of an RDL specific antagonist. This may be useful for insecticide development as well as a tool for further studies in invertebrate neuroscience. In Chapter 5 I identified several residues that are involved in ligand binding in RDL receptors. Loop B residues E204 and F206 and loop C residue Y254 seem the most likely candidates for interactions with the ammonium end of GABA. Loop D residues R111 and Y109 seem the most likely candidates for interactions with the carboxylate of GABA. Homology modelling and docking experiments support these binding interactions. Further studies could involve unnatural amino acid mutagenesis of these residues: fluorination of aromatic residues would determine whether there is a role for π electron density on the face of aromatic rings in ligand binding. Charged residues should be substituted with unnatural amino acids with similar side chains structure but with varying charge. These studies would clarify beyond doubt the role of these candidate ‘binding’ residues. In Chapter 6 I identified Ginkgo biloba extracts, ginkgolides and bilobalide, as antagonists of RDL receptors with IC50 values similar to picrotoxin. Ginkgolides wash out more slowly than picrotoxin, making these compounds more potent blockers of RDL channels. Mutagenesis, modelling and mutant cycle analysis identified the 2' and 6' M2 channel-lining residues as being part of the ginkgolide binding site. Whole insect bioassays confirmed the insecticidal potency of these compounds, leading to the hypothesis that these compounds exert their insecticidal properties through RDL-like receptors. Further work could involve screening these compounds on a range of insect species, identifying susceptible and resistant species. Resistant species should be genetically catalogued with special attention to Cys-loop receptor structure, and particularly, the M2 2' and 6' residues. 151 In Chapter 7 I confirmed that the brain 5-HT3R B subunit variant Br1 co- expresses with the A subunit, forming functional heteromeric receptor channels with a conductance similar to heteromeric AB receptors. Br2 containing channels however could not be detected in this study, leading to the hypothesis that this transcript (Br2) may play a role in other cellular processes such as receptor trafficking and/or regulation of expression. It is also quite possible that this transcript is not translated at all. Further studies on these subunits could involve mutagenesis studies and molecular modelling, as has led to the further understanding of the arrangement of A and B subunits in heteromeric 5-HT3 receptors (Lochner and Lummis, 2010). Immunocytochemistry studies, using antibodies or fluorescent tags, could complement such work to determine whether Br2 subunits reach the cell surface. 152 Final remarks In this thesis I have identified the critical determinants of RDL receptor function: pH sensitivity, ionic selectivity, agonist requirements and binding site residues. I have also identified natural compounds (ginkgolides) which block RDL receptors and which are potent insecticides. Together these findings provide potential for the development of RDL-targeting insecticides as well as pharmacological probes which may be useful in further studies on insect GABA receptors. I have also confirmed that 5-HT3 receptor brain subunit Br1 forms functional receptor channels. Overall this work has lead to a greater understanding of Cys-loop receptor function and the findings presented here may benefit further biomedical and neuroscience research. 153 References Ahlemeyer, B, Krieglstein, J (2003) Neuroprotective effects of Ginkgo biloba extract. Cell Mol Life Sci 60(9): 1779-1792. Ahn, YJ (1997) Potent Insecticidal Activity of Ginkgo biloba Derived Trilactone Terpenes Against Nilaparvata lugens. ACS Symposium series 658: 90-105. Akabas, MH (2004) GABAA receptor structure-function studies: a reexamination in light of new acetylcholine receptor structures. Int Rev Neurobiol 62: 1-43. Amin, J, Weiss, DS (1993) GABAA receptor needs two homologous domains of the β-subunit for activation by GABA but not by pentobarbital. Nature 366(6455): 565-569. Amin, J, Weiss, DS (1994) Homomeric ρ1 GABA channels: activation properties and domains. Receptors Channels 2(3): 227-236. Aronstein, K, ffrench-Constant, R (1995) Immunocytochemistry of a novel GABA receptor subunit Rdl in Drosophila melanogaster. Invert Neurosci 1(1): 25-31. Auerbach, A (2010) The gating isomerization of neuromuscular acetylcholine receptors. J Physiol 588(4): 573-586. Bafna, PA, Purohit, PG, Auerbach, A (2008) Gating at the mouth of the acetylcholine receptor channel: energetic consequences of mutations in the αM2-cap. PLoS One 3(6): e2515. Bali, M, Akabas, MH (2007) The location of a closed channel gate in the GABAA receptor channel. J Gen Physiol 129(2): 145-159. Bargmann, CI (1998) Neurobiology of the Caenorhabditis elegans genome. Science 282(5396): 2028-2033. Barnard, EA (1992) Receptor classes and the transmitter-gated ion channels. Trends Biochem Sci 17(10): 368-374. Barnard, EA, Skolnick, P, Olsen, RW, Mohler, H, Sieghart, W, Biggio, G, Braestrup, C, Bateson, AN, Langer, SZ (1998) International Union of Pharmacology. XV. Subtypes of γ-aminobutyric acidA receptors: classification on the basis of subunit structure and receptor function. Pharmacol Rev 50(2): 291-313. 154 Bartos, M, Corradi, J, Bouzat, C (2009a) Structural Basis of Activation of Cys- Loop Receptors: the Extracellular-Transmembrane Interface as a Coupling Region. Mol Neurobiol 40(3): 236-252 Bartos, M, Price, KL, Lummis, SC, Bouzat, C (2009b) Glutamine 57 at the complementary binding site face is a key determinant of morantel selectivity for α7 nicotinic receptors. J Biol Chem 284(32): 21478- 21487. Beene, DL, Brandt, GS, Zhong, W, Zacharias, NM, Lester, HA, Dougherty, DA (2002) Cation-π interactions in ligand recognition by serotonergic (5-HT3A) and nicotinic acetylcholine receptors: the anomalous binding properties of nicotine. Biochemistry 41(32): 10262-10269. Beene, DL, Price, KL, Lester, HA, Dougherty, DA, Lummis, SC (2004) Tyrosine residues that control binding and gating in the 5-HT3 receptor revealed by unnatural amino acid mutagenesis. J Neurosci 24(41): 9097- 9104. Beg, AA, Jorgensen, EM (2003) EXP-1 is an excitatory GABA-gated cation channel. Nat Neurosci 6(11): 1145-1152. Belelli, D, Callachan, H, Hill-Venning, C, Peters, JA, Lambert, JJ (1996) Interaction of positive allosteric modulators with human and Drosophila recombinant GABA receptors expressed in Xenopus laevis oocytes. Br J Pharmacol 118(3): 563-576. Bocquet, N, Nury, H, Baaden, M, Le Poupon, C, Changeux, JP, Delarue, M, Corringer, PJ (2009) X-ray structure of a pentameric ligand-gated ion channel in an apparently open conformation. Nature 457(7225): 111- 114. Boileau, AJ, Newell, JG, Czajkowski, C (2002) GABAA receptor β2Tyr97 and Leu99 line the GABA-binding site. Insights into mechanisms of agonist and antagonist actions. J Biol Chem 277(4): 2931-2937. Bouzat, C, Bartos, M, Corradi, J, Sine, SM (2008) The interface between extracellular and transmembrane domains of homomeric Cys-loop receptors governs open-channel lifetime and rate of desensitization. J Neurosci 28(31): 7808-7819. Brejc, K, van Dijk, WJ, Klaassen, RV, Schuurmans, M, van Der Oost, J, Smit, AB, Sixma, TK (2001) Crystal structure of an ACh-binding protein reveals the ligand-binding domain of nicotinic receptors. Nature 411(6835): 269-276. 155 Bruss, M, Barann, M, Hayer-Zillgen, M, Eucker, T, Gothert, M, Bonisch, H (2000a) Modified 5-HT3A receptor function by co-expression of alternatively spliced human 5-HT3A receptor isoforms. Naunyn Schmiedebergs Arch Pharmacol 362(4-5): 392-401. Bruss, M, Eucker, T, Gothert, M, Bonisch, H (2000b) Exon-intron organization of the human 5-HT3A receptor gene. Neuropharmacology 39(2): 308- 315. Buckingham, SD, Biggin, PC, Sattelle, BM, Brown, LA, Sattelle, DB (2005) Insect GABA receptors: splicing, editing, and targeting by antiparasitics and insecticides. Mol Pharmacol 68(4): 942-951. Buckingham, SD, Hosie, AM, Roush, RL, Sattelle, DB (1994) Actions of agonists and convulsant antagonists on a Drosophila melanogaster GABA receptor (Rdl) homo-oligomer expressed in Xenopus oocytes. Neurosci Lett 181(1-2): 137-140. Burzomato, V, Beato, M, Groot-Kormelink, PJ, Colquhoun, D, Sivilotti, LG (2004) Single-channel behavior of heteromeric α1β glycine receptors: an attempt to detect a conformational change before the channel opens. J Neurosci 24(48): 10924-10940. Casida, JE (1993) Insecticide action at the GABA-gated chloride channel: recognition, progress, and prospects. Arch Insect Biochem Physiol 22(1- 2): 13-23. Casida, JE (2009) Pest toxicology: the primary mechanisms of pesticide action. Chem Res Toxicol 22(4): 609-619. Chakrapani, S, Bailey, TD, Auerbach, A (2004) Gating dynamics of the acetylcholine receptor extracellular domain. J Gen Physiol 123(4): 341- 356. Chen, JC, Chesler, M (1992) Modulation of extracellular pH by glutamate and GABA in rat hippocampal slices. J Neurophysiol 67(1): 29-36. Chen, L, Durkin, KA, Casida, JE (2006) Structural model for γ-aminobutyric acid receptor noncompetitive antagonist binding: widely diverse structures fit the same site. Proc Natl Acad Sci U S A 103(13): 5185- 5190. Cheng, X, Wang, H, Grant, B, Sine, SM, McCammon, JA (2006) Targeted molecular dynamics study of C-loop closure and channel gating in nicotinic receptors. PLoS Comput Biol 2(9): e134. Chesler, M (1990) The regulation and modulation of pH in the nervous system. Prog Neurobiol 34(5): 401-427. 156 Corradi, J, Gumilar, F, Bouzat, C (2009) Single-channel kinetic analysis for activation and desensitization of homomeric 5-HT3A receptors. Biophys J 97(5): 1335-1345. Corry, B (2006) An energy-efficient gating mechanism in the acetylcholine receptor channel suggested by molecular and Brownian dynamics. Biophys J 90(3): 799-810. Costa, PF, Emilio, MG, Fernandes, PL, Ferreira, HG, Ferreira, KG (1989) Determination of ionic permeability coefficients of the plasma membrane of Xenopus laevis oocytes under voltage clamp. J Physiol 413: 199-211. Curtis, DR, Johnston, GA (1974) Amino acid transmitters in the mammalian central nervous system. Ergeb Physiol 69(0): 97-188. Cutting, GR, Curristin, S, Zoghbi, H, O'Hara, B, Seldin, MF, Uhl, GR (1992) Identification of a putative γ-aminobutyric acid (GABA) receptor subunit ρ2 cDNA and colocalization of the genes encoding ρ2 (GABRR2) and ρ1 (GABRR1) to human chromosome 6q14-q21 and mouse chromosome 4. Genomics 12(4): 801-806. Cutting, GR, Lu, L, O'Hara, BF, Kasch, LM, Montrose-Rafizadeh, C, Donovan, DM, Shimada, S, Antonarakis, SE, Guggino, WB, Uhl, GR, et al. (1991) Cloning of the γ-aminobutyric acid (GABA) ρ1 cDNA: a GABA receptor subunit highly expressed in the retina. Proc Natl Acad Sci U S A 88(7): 2673-2677. Cymes, GD, Grosman, C (2008) Pore-opening mechanism of the nicotinic acetylcholine receptor evinced by proton transfer. Nat Struct Mol Biol 15(4): 389-396. Dahan, DS, Dibas, MI, Petersson, EJ, Auyeung, VC, Chanda, B, Bezanilla, F, Dougherty, DA, Lester, HA (2004) A fluorophore attached to nicotinic acetylcholine receptor β M2 detects productive binding of agonist to the αδ site. Proc Natl Acad Sci U S A 101(27): 10195-10200. Das, P, Bell-Horner, CL, Machu, TK, Dillon, GH (2003) The GABAA receptor antagonist picrotoxin inhibits 5-hydroxytryptamine type 3A receptors. Neuropharmacology 44(4): 431-438. Das, P, Dillon, GH (2005) Molecular determinants of picrotoxin inhibition of 5-hydroxytryptamine type 3 receptors. J Pharmacol Exp Ther 314(1): 320-328. 157 Davies, PA, Pistis, M, Hanna, MC, Peters, JA, Lambert, JJ, Hales, TG, Kirkness, EF (1999) The 5-HT3B subunit is a major determinant of serotonin-receptor function. Nature 397(6717): 359-363. Davies, PA, Wang, W, Hales, TG, Kirkness, EF (2003) A novel class of ligand- gated ion channel is activated by Zn2+. J Biol Chem 278(2): 712-717. DeFeudis, FV, Drieu, K (2000) Ginkgo biloba extract (EGb 761) and CNS functions: basic studies and clinical applications. Curr Drug Targets 1(1): 25-58. Del Castillo, J, Katz, B (1957) Interaction at end-plate receptors between different choline derivatives. Proc R Soc Lond B Biol Sci 146(924): 369- 381. Del Castillo, J, Morales, T., and Sanchez, V. (1963) Action of piperazine on the neurotransmitter system of Ascaris lumbricoides. Nature 200: 706-707. Dellisanti, CD, Yao, Y, Stroud, JC, Wang, ZZ, Chen, L (2007) Crystal structure of the extracellular domain of nAChR α1 bound to α- bungarotoxin at 1.94 Å resolution. Nat Neurosci 10(8): 953-962. Dougherty, DA (1996) Cation-π interactions in chemistry and biology: a new view of benzene, Phe, Tyr, and Trp. Science 271(5246): 163-168. Dougherty, DA (2007) Cation-π interactions involving aromatic amino acids. J Nutr 137(6 Suppl 1): 1504S-1508S; discussion 1516S-1517S. Dougherty, DA (2008) Cys-loop neuroreceptors: structure to the rescue? Chem Rev 108(5): 1642-1653. Drieu, K, Vranckx, R, Benassayad, C, Haourigi, M, Hassid, J, Yoa, RG, Rapin, JR, Nunez, EA (2000) Effect of the extract of Ginkgo biloba (EGb 761) on the circulating and cellular profiles of polyunsaturated fatty acids: correlation with the anti-oxidant properties of the extract. Prostaglandins Leukot Essent Fatty Acids 63(5): 293-300. Du, W, Awolola, TS, Howell, P, Koekemoer, LL, Brooke, BD, Benedict, MQ, Coetzee, M, Zheng, L (2005) Independent mutations in the Rdl locus confer dieldrin resistance to Anopheles gambiae and An. arabiensis. Insect Mol Biol 14(2):179-183. Duan, S, Cooke, IM (2000) Glutamate and GABA activate different receptors and Cl(-) conductances in crab peptide-secretory neurons. J Neurophysiol 83(1): 31-37. Dupuis, JP, Bazelot, M, Barbara, GS, Paute, S, Gauthier, M, Raymond- Delpech, V. (2010) Homomeric RDL and heteromeric RDL/LCCH3 158 GABA receptors in the honeybee antennal lobes: two candidates for inhibitory transmission in olfactory processing. J Neurophysiol 103(1): 458-468. Eguchi, Y, Ihara, M, Ochi, E, Shibata, Y, Matsuda, K, Fushiki, S, Sugama, H, Hamasaki, Y, Niwa, H, Wada, M, Ozoe, F, Ozoe, Y (2006) Functional characterization of Musca glutamate- and GABA-gated chloride channels expressed independently and coexpressed in Xenopus oocytes. Insect Mol Biol 15(6): 773-783. Enz, R, Cutting, GR (1999) GABAC receptor ρ subunits are heterogeneously expressed in the human CNS and form homo- and heterooligomers with distinct physical properties. Eur J Neurosci 11(1): 41-50. Erkkila, BE, Weiss, DS, Wotring, VE (2004) Picrotoxin-mediated antagonism of α3β4 and α7 acetylcholine receptors. Neuroreport 15(12): 1969-1973. Etter, A, Cully, DF, Liu, KK, Reiss, B, Vassilatis, DK, Schaeffer, JM, Arena, JP (1999) Picrotoxin blockade of invertebrate glutamate-gated chloride channels: subunit dependence and evidence for binding within the pore. J Neurochem 72(1): 318-326. ffrench-Constant, RH, Mortlock, DP, Shaffer, CD, MacIntyre, RJ, Roush, RT (1991) Molecular cloning and transformation of cyclodiene resistance in Drosophila: an invertebrate γ-aminobutyric acid subtype A receptor locus. Proc Natl Acad Sci U S A 88(16): 7209-7213. ffrench-Constant, RH, Rocheleau, TA, Steichen, JC, Chalmers, AE (1993a) A point mutation in a Drosophila GABA receptor confers insecticide resistance. Nature 363(6428): 449-451. ffrench-Constant, RH, Steichen, JC, Rocheleau, TA, Aronstein, K, Roush, RT (1993b) A single-amino acid substitution in a γ-aminobutyric acid subtype A receptor locus is associated with cyclodiene insecticide resistance in Drosophila populations. Proc Natl Acad Sci U S A 90(5): 1957-1961. Galzi, JL, Devillers-Thiery, A, Hussy, N, Bertrand, S, Changeux, JP, Bertrand, D (1992) Mutations in the channel domain of a neuronal nicotinic receptor convert ion selectivity from cationic to anionic. Nature 359(6395): 500-505. Gisselmann, G, Plonka, J, Pusch, H, Hatt, H (2004) Drosophila melanogaster GRD and LCCH3 subunits form heteromultimeric GABA-gated cation channels. Br J Pharmacol 142(3): 409-413. 159 Gisselmann, G, Pusch, H, Hovemann, BT, Hatt, H (2002) Two cDNAs coding for histamine-gated ion channels in D. melanogaster. Nat Neurosci 5(1): 11-12. Grosman, C, Zhou, M, Auerbach, A (2000) Mapping the conformational wave of acetylcholine receptor channel gating. Nature 403(6771): 773-776. Grutter, T, de Carvalho, LP, Dufresne, V, Taly, A, Edelstein, SJ, Changeux, JP (2005) Molecular tuning of fast gating in pentameric ligand-gated ion channels. Proc Natl Acad Sci U S A 102(50): 18207-18212. Gunthorpe, MJ, Lummis, SC (2001) Conversion of the ion selectivity of the 5- HT3A receptor from cationic to anionic reveals a conserved feature of the ligand-gated ion channel superfamily. J Biol Chem 276(24): 10977- 10983. Hamill, OP, Marty, A, Neher, E, Sakmann, B, Sigworth, FJ (1981) Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflugers Arch 391(2): 85-100. Hansen, SB, Sulzenbacher, G, Huxford, T, Marchot, P, Taylor, P, Bourne, Y (2005) Structures of Aplysia AChBP complexes with nicotinic agonists and antagonists reveal distinctive binding interfaces and conformations. EMBO J 24(20): 3635-3646. Hargreaves, AC, Gunthorpe, MJ, Taylor, CW, Lummis, SC (1996) Direct inhibition of 5-HT3 receptors by antagonists of L-type Ca2 + channels. Mol Pharmacol 50(5): 1284-1294. Harrison, JB, Chen, HH, Sattelle, E, Barker, PJ, Huskisson, NS, Rauh, JJ, Bai, D, Sattelle, DB (1996) Immunocytochemical mapping of a C-terminus anti-peptide antibody to the GABA receptor subunit, RDL in the nervous system in Drosophila melanogaster. Cell Tissue Res 284(2): 269-278. Harrison, NJ, Lummis, SC (2006) Locating the carboxylate group of GABA in the homomeric ρ GABAA receptor ligand-binding pocket. J Biol Chem 281(34): 24455-24461. Hawthorne, R, Cromer, BA, Ng, HL, Parker, MW, Lynch, JW (2006) Molecular determinants of ginkgolide binding in the glycine receptor pore. J Neurochem 98(2): 395-407. Heads, JA, Hawthorne, RL, Lynagh, T, Lynch, JW (2008) Structure-activity analysis of ginkgolide binding in the glycine receptor pore. J Neurochem 105(4): 1418-1427. 160 Hidalgo, P, MacKinnon, R (1995) Revealing the architecture of a K+ channel pore through mutant cycles with a peptide inhibitor. Science 268(5208): 307-310. Hilf, RJ, Dutzler, R (2009) Structure of a potentially open state of a proton- activated pentameric ligand-gated ion channel. Nature 457(7225): 115- 118. Hilf, RJ, Dutzler, R (2008) X-ray structure of a prokaryotic pentameric ligand- gated ion channel. Nature 452(7185): 375-379. Hogg, RC, Buisson, B, Bertrand, D (2005) Allosteric modulation of ligand- gated ion channels. Biochem Pharmacol 70(9): 1267-1276. Hosie, AM, Aronstein, K, Sattelle, DB, ffrench-Constant, RH (1997) Molecular biology of insect neuronal GABA receptors. Trends Neurosci 20(12): 578-583. Hosie, AM, Baylis, HA, Buckingham, SD, Sattelle, DB (1995) Actions of the insecticide fipronil, on dieldrin-sensitive and- resistant GABA receptors of Drosophila melanogaster. Br J Pharmacol 115(6): 909-912. Hosie, AM, Buckingham, SD, Presnail, JK, Sattelle, DB (2001) Alternative splicing of a Drosophila GABA receptor subunit gene identifies determinants of agonist potency. Neuroscience 102(3): 709-714. Hosie, AM, Ozoe, Y, Koike, K, Ohmoto, T, Nikaido, T, Sattelle, DB (1996) Actions of picrodendrin antagonists on dieldrin-sensitive and -resistant Drosophila GABA receptors. Br J Pharmacol 119(8): 1569-1576. Hosie, AM, Sattelle, DB (1996a) Agonist pharmacology of two Drosophila GABA receptor splice variants. Br J Pharmacol 119(8): 1577-1585. Hosie, AM, Sattelle, DB (1996b) Allosteric modulation of an expressed homo- oligomeric GABA-gated chloride channel of Drosophila melanogaster. Br J Pharmacol 117(6): 1229-1237. Huang, RQ, Dillon, GH (1999) Effect of extracellular pH on GABA-activated current in rat recombinant receptors and thin hypothalamic slices. J Neurophysiol 82(3): 1233-1243. Huang, SH, Duke, RK, Chebib, M, Sasaki, K, Wada, K, Johnston, GA (2004) Ginkgolides, diterpene trilactones of Ginkgo biloba, as antagonists at recombinant α1β2γ2 GABAA receptors. Eur J Pharmacol 494(2-3): 131-138. Huxtable, R, Franconi, F, Giotti, A (1987) The biology of taurine. Methods and mechanisms. Adv Exp Med Biol 217: 1-404. 161 Itil, TM, Eralp, E, Tsambis, E, Itil, KZ, Stein, U (1996) Central Nervous System Effects of Ginkgo Biloba, a Plant Extract. Am J Ther 3(1): 63- 73. Ivic, L, Sands, TT, Fishkin, N, Nakanishi, K, Kriegstein, AR, Stromgaard, K (2003) Terpene trilactones from Ginkgo biloba are antagonists of cortical glycine and GABAA receptors. J Biol Chem 278(49): 49279- 49285. Jensen, AA, Bergmann, ML, Sander, T, Balle, T (2010) Ginkgolide X is a potent antagonist of anionic Cys-loop receptors with a unique selectivity profile at glycine receptors. J Biol Chem 285(13): 10141-10153. Jensen, ML, Timmermann, DB, Johansen, TH, Schousboe, A, Varming, T, Ahring, PK (2002) The β subunit determines the ion selectivity of the GABAA receptor. J Biol Chem 277(44): 41438-41447. Jones, AK, Sattelle, DB (2008) The Cys-loop ligand-gated ion channel gene superfamily of the nematode, Caenorhabditis elegans. Invert Neurosci 8(1): 41-47. Jones, AK, Sattelle, DB (2007) The Cys-loop ligand-gated ion channel gene superfamily of the red flour beetle, Tribolium castaneum. BMC Genomics 8: 327. Jones, AK, Sattelle, DB (2006) The Cys-loop ligand-gated ion channel superfamily of the honeybee, Apis mellifera. Invert Neurosci 6(3): 123- 132. Jones, MV, Sahara, Y, Dzubay, JA, Westbrook, GL (1998) Defining affinity with the GABAA receptor. J Neurosci 18(21): 8590-8604. Kaku, K, Matsumura, F (1994) Identification of the site of mutation within the M2 region of the GABA receptor of the cyclodiene-resistant German cockroach. Comp Biochem Physiol C Pharmacol Toxicol Endocrinol 108(3): 367-376. Kaila, K (1994) Ionic basis of GABAA receptor channel function in the nervous system. Prog Neurobiol 42(4): 489-537. Kaila, K, Voipio, J (1987) Postsynaptic fall in intracellular pH induced by GABA-activated bicarbonate conductance. Nature 330(6144): 163-165. Kash, TL, Jenkins, A, Kelley, JC, Trudell, JR, Harrison, NL (2003) Coupling of agonist binding to channel gating in the GABAA receptor. Nature 421(6920): 272-275. 162 Kelley, SP, Dunlop, JI, Kirkness, EF, Lambert, JJ, Peters, JA (2003) A cytoplasmic region determines single-channel conductance in 5-HT3 receptors. Nature 424(6946): 321-324. Keramidas, A, Moorhouse, AJ, French, CR, Schofield, PR, Barry, PH (2000) M2 pore mutations convert the glycine receptor channel from being anion- to cation-selective. Biophys J 79(1): 247-259. Kondratskaya, EL, Fisyunov, AI, Chatterjee, SS, Krishtal, OA (2004) Ginkgolide B preferentially blocks chloride channels formed by heteromeric glycine receptors in hippocampal pyramidal neurons of rat. Brain Res Bull 63(4): 309-314. Kraig, RP, Ferreira-Filho, CR, Nicholson, C (1983) Alkaline and acid transients in cerebellar microenvironment. J Neurophysiol 49(3): 831- 850. Krishek, BJ, Amato, A, Connolly, CN, Moss, SJ, Smart, TG (1996a) Proton sensitivity of the GABAA receptor is associated with the receptor subunit composition. J Physiol 492 ( Pt 2): 431-443. Krishek, BJ, Moss, SJ, Smart, TG (1996b) A functional comparison of the antagonists bicuculline and picrotoxin at recombinant GABAA receptors. Neuropharmacology 35(9-10): 1289-1298. Kumar, S, Nussinov, R (2002) Close-range electrostatic interactions in proteins. Chembiochem 3(7): 604-617. Lape, R, Colquhoun, D, Sivilotti, LG (2008) On the nature of partial agonism in the nicotinic receptor superfamily. Nature 454(7205): 722-727. Lee, WY, Sine, SM (2005) Principal pathway coupling agonist binding to channel gating in nicotinic receptors. Nature 438(7065): 243-247. Lee, HJ, Rocheleau, T, Zhang, HG, Jackson, MB, ffrench-Constant, RH (1993) Expression of a Drosophila GABA receptor in a baculovirus insect cell system. Functional expression of insecticide susceptible and resistant GABA receptors from the cyclodiene resistance gene Rdl. FEBS Lett 335(3): 315-318. Leisgen, C, Kuester, M, Methfessel, C (2007) The roboocyte: automated electrophysiology based on Xenopus oocytes. Methods Mol Biol 403: 87-109. Li, P, Slaughter, M (2007) Glycine receptor subunit composition alters the action of GABA antagonists. Vis Neurosci 24(4): 513-521. 163 Liman, ER, Tytgat J, and Hess P. (1992) Subunit stoichiometry of a mammalian K+ channel determined by construction of multimeric cDNAs. Neuron 9(5): 861-871. Littleton, JT, Ganetzky, B (2000) Ion channels and synaptic organization: analysis of the Drosophila genome. Neuron 26(1): 35-43. Liu, X, Buchanan, ME, Han, KA, Davis, RL (2009) The GABAA receptor RDL suppresses the conditioned stimulus pathway for olfactory learning. J Neurosci 29(5): 1573-1579. Liu, X, Davis, RL (2009) The GABAergic anterior paired lateral neuron suppresses and is suppressed by olfactory learning. Nat Neurosci 12(1): 53-59. Liu, X, Krause, WC, Davis, RL (2007) GABAA Receptor RDL Inhibits Drosophila Olfactory Associative Learning. Neuron 56(6): 1090-1102. Lochner, M, Lummis, SC (2010) Agonists and antagonists bind to an A-A interface in the heteromeric 5-HT3AB receptor. Biophys J 98(8): 1494- 1502. Lovell, SC, Davis, IW, Arendall, WB, 3rd, de Bakker, PI, Word, JM, Prisant, MG, Richardson, JS, Richardson, DC (2003) Structure validation by Cα geometry: φ, ψ and Cβ deviation. Proteins 50(3): 437-450. Ludmerer, SW, Warren, VA, Williams, BS, Zheng, Y, Hunt, DC, Ayer, MB, Wallace, MA, Chaudhary, AG, Egan, MA, Meinke, PT, Dean, DC, Garcia, ML, Cully, DF, Smith, MM (2002) Ivermectin and nodulisporic acid receptors in Drosophila melanogaster contain both γ-aminobutyric acid-gated Rdl and glutamate-gated GluCl α chloride channel subunits. Biochemistry 41(20): 6548-6560. Lummis, SC (2009) Locating GABA in GABA receptor binding sites. Biochem Soc Trans 37(Pt 6): 1343-1346. Lummis, SC, Beene, DL, Lee, LW, Lester, HA, Broadhurst, RW, Dougherty, DA (2005a) Cis-trans isomerization at a proline opens the pore of a neurotransmitter-gated ion channel. Nature 438(7065): 248-252. Lummis, SC, D, LB, Harrison, NJ, Lester, HA, Dougherty, DA (2005b) A cation-π binding interaction with a tyrosine in the binding site of the GABAC receptor. Chem Biol 12(9): 993-997. Luque, JM, Malherbe, P, Richards, JG (1994) Localization of GABAA receptor subunit mRNAs in the rat locus coeruleus. Brain Res Mol Brain Res 24(1-4): 219-226. 164 Lynch, JW, Rajendra, S, Barry, PH, Schofield, PR (1995) Mutations affecting the glycine receptor agonist transduction mechanism convert the competitive antagonist, picrotoxin, into an allosteric potentiator. J Biol Chem 270(23): 13799-13806. McGonigle, I, Lummis, SC (2010) Molecular Characterization of Agonists That Bind to an Insect GABA Receptor. Biochemistry 49(13): 2897- 2902 McGonigle, I, Lummis, SC (2009) RDL receptors. Biochem Soc Trans 37(6): 1404-1406. McGurk, KA, Pistis, M, Belelli, D, Hope, AG, Lambert, JJ (1998) The effect of a transmembrane amino acid on etomidate sensitivity of an invertebrate GABA receptor. Br J Pharmacol 124(1): 13-20. Mdzinarishvili, A, Kiewert, C, Kumar, V, Hillert, M, Klein, J (2007) Bilobalide prevents ischemia-induced edema formation in vitro and in vivo. Neuroscience 144(1): 217-222. Mecozzi, S, West, AP, Jr., Dougherty, DA (1996) Cation-π interactions in aromatics of biological and medicinal interest: electrostatic potential surfaces as a useful qualitative guide. Proc Natl Acad Sci U S A 93(20): 10566-10571. Melis, C, Lummis, SC, Molteni, C (2008) Molecular dynamics simulations of GABA binding to the GABAC receptor: the role of Arg104. Biophys J 95(9): 4115-4123. Millar, NS, Buckingham, SD, Sattelle, DB (1994) Stable expression of a functional homo-oligomeric Drosophila GABA receptor in a Drosophila cell line. Proc Biol Sci 258(1353): 307-314. Miller, C (1989) Genetic manipulation of ion channels: a new approach to structure and mechanism. Neuron 2(3): 1195-1205. Miller, PS, Smart, TG (2010) Binding, activation and modulation of Cys-loop receptors. Trends Pharmacol Sci 31(4): 161-174. Miyazawa, A, Fujiyoshi, Y, Unwin, N (2003) Structure and gating mechanism of the acetylcholine receptor pore. Nature 423(6943): 949-955. Mu, TW, Lester, HA, Dougherty, DA (2003) Different binding orientations for the same agonist at homologous receptors: a lock and key or a simple wedge? J Am Chem Soc 125(23): 6850-6851. 165 Mukhtasimova, N, Lee, WY, Wang, HL, Sine, SM (2009) Detection and trapping of intermediate states priming nicotinic receptor channel opening. Nature 459(7245): 451-454. Muroi, Y, Czajkowski, C, Jackson, MB (2006) Local and global ligand- induced changes in the structure of the GABAA receptor. Biochemistry 45(23): 7013-7022. Muroi, Y, Theusch, CM, Czajkowski, C, Jackson, MB (2009) Distinct structural changes in the GABAA receptor elicited by pentobarbital and GABA. Biophys J 96(2): 499-509. Narusuye, K, Nakao, T, Abe, R, Nagatomi, Y, Hirase, K, Ozoe, Y (2007) Molecular cloning of a GABA receptor subunit from Laodelphax striatella (Fallén) and patch clamp analysis of the homo-oligomeric receptors expressed in a Drosophila cell line. Insect Mol Biol 16(6): 723-733. Newell, JG, Czajkowski, C (2003) The GABAA receptor α1 subunit Pro174- Asp191 segment is involved in GABA binding and channel gating. J Biol Chem 278(15): 13166-13172. Niesler, B, Frank, B, Kapeller, J, Rappold, GA (2003) Cloning, physical mapping and expression analysis of the human 5-HT3 serotonin receptor-like genes HTR3C, HTR3D and HTR3E. Gene 310: 101-111. Niesler, B, Walstab, J, Combrink, S, Moller, D, Kapeller, J, Rietdorf, J, Bonisch, H, Gothert, M, Rappold, G, Bruss, M (2007) Characterization of the novel human serotonin receptor subunits 5-HT3C,5-HT3D, and 5- HT3E. Mol Pharmacol 72(1): 8-17. Olsen, RW (2006) Picrotoxin-like channel blockers of GABAA receptors. Proc Natl Acad Sci U S A 103(16): 6081-6082. Ong, J, Kerr, DI (2000) Recent advances in GABAB receptors: from pharmacology to molecular biology. Acta Pharmacol Sin 21(2): 111- 123. Padgett, CL, Hanek, AP, Lester, HA, Dougherty, DA, Lummis, SC (2007) Unnatural amino acid mutagenesis of the GABAA receptor binding site residues reveals a novel cation-π interaction between GABA and β 2Tyr97. J Neurosci 27(4): 886-892. Panicker, S, Cruz, H, Arrabit, C, Slesinger, PA (2002) Evidence for a centrally located gate in the pore of a serotonin-gated ion channel. J Neurosci 22(5): 1629-1639. 166 Persohn, E, Malherbe, P, Richards, JG (1991) In situ hybridization histochemistry reveals a diversity of GABAA receptor subunit mRNAs in neurons of the rat spinal cord and dorsal root ganglia. Neuroscience 42(2): 497-507. Pless, SA, Dibas, MI, Lester, HA, Lynch, JW (2007) Conformational variability of the glycine receptor M2 domain in response to activation by different agonists. J Biol Chem 282(49): 36057-36067. Pless, SA, Lynch, JW (2009a) Distinct conformational changes in activated agonist-bound and agonist-free glycine receptor subunits. J Neurochem 108(6): 1585-1594. Pless, SA, Lynch, JW (2009b) Ligand-specific conformational changes in the α1 glycine receptor ligand-binding domain. J Biol Chem 284(23): 15847-15856. Pless, SA, Lynch, JW (2009c) Magnitude of a conformational change in the glycine receptor β1-β2 loop is correlated with agonist efficacy. J Biol Chem 284(40): 27370-27376. Pless, SA, Millen, KS, Hanek, AP, Lynch, JW, Lester, HA, Lummis, SC, Dougherty, DA (2008) A cation-π interaction in the binding site of the glycine receptor is mediated by a phenylalanine residue. J Neurosci 28(43): 10937-10942. Price, KL, Lummis, SC (2005) FlexStation examination of 5-HT3 receptor function using Ca2+ - and membrane potential-sensitive dyes: advantages and potential problems. J Neurosci Methods 149(2): 172- 177. Price, KL, Millen, KS, Lummis, SC (2007) Transducing agonist binding to channel gating involves different interactions in 5-HT3 and GABAC receptors. J Biol Chem 282(35): 25623-25630. Purohit, P, Auerbach, A (2007) Acetylcholine receptor gating at extracellular transmembrane domain interface: the "pre-M1" linker. J Gen Physiol 130(6): 559-568. Qian, H, Pan, Y, Zhu, Y, Khalili, P (2005) Picrotoxin accelerates relaxation of GABAC receptors. Mol Pharmacol 67(2): 470-479. Ranganathan, R, Cannon, SC, Horvitz, HR (2000) MOD-1 is a serotonin-gated chloride channel that modulates locomotory behaviour in C. elegans. Nature 408(6811): 470-475. 167 Rayes, D, De Rosa, MJ, Sine, SM, Bouzat, C (2009) Number and locations of agonist binding sites required to activate homomeric Cys-loop receptors. J Neurosci 29(18): 6022-6032. Raymond-Delpech, V, Matsuda, K, Sattelle, BM, Rauh, JJ, Sattelle, DB (2005) Ion channels: molecular targets of neuroactive insecticides. Invert Neurosci 5(3-4): 119-133. Reeves, DC, Goren, EN, Akabas, MH, Lummis, SC (2001) Structural and electrostatic properties of the 5-HT3 receptor pore revealed by substituted cysteine accessibility mutagenesis. J Biol Chem 276(45): 42035-42042. Reeves, DC, Jansen, M, Bali, M, Lemster, T, Akabas, MH (2005) A role for the β1-β2 loop in the gating of 5-HT3 receptors. J Neurosci 25(41): 9358-9366. Roberts, E, Frankel, S (1950) γ-Aminobutyric acid in brain: its formation from glutamic acid. J Biol Chem 187(1): 55-63. Sali, A, Blundell, TL (1993) Comparative protein modelling by satisfaction of spatial restraints. J Mol Biol 234(3): 779-815. Sammelson, RE, Caboni, P, Durkin, KA, Casida, JE (2004) GABA receptor antagonists and insecticides: common structural features of 4-alkyl-1- phenylpyrazoles and 4-alkyl-1-phenyltrioxabicyclooctanes. Bioorg Med Chem 12(12): 3345-3355. Sattelle, DB, Harrison, JB, Chen, HH, Bai, D, Takeda, M (2000) Immunocytochemical localization of putative γ-aminobutyric acid receptor subunits in the head ganglia of Periplaneta americana using an anti-RDL C-terminal antibody. Neurosci Lett 289(3): 197-200. Sattelle, DB, Jones, AK, Sattelle, BM, Matsuda, K, Reenan, R, Biggin, PC (2005) Edit, cut and paste in the nicotinic acetylcholine receptor gene family of Drosophila melanogaster. Bioessays 27(4): 366-376. Sattelle, DB, Pinnock, RD, Wafford, KA, David, JA (1988) GABA receptors on the cell-body membrane of an identified insect motor neuron. Proc R Soc Lond B Biol Sci 232(1269): 443-456. Schmieden, V, Grenningloh, G, Schofield, PR, Betz, H (1989) Functional expression in Xenopus oocytes of the strychnine binding 48 kd subunit of the glycine receptor. EMBO J 8(3): 695-700. Schnee, M, Rauh, J, Buckingham, SD, Sattelle, DB (1997) Pharmacology of skeletal muscle GABA-gated chloride channels in the cockroach Periplaneta americana. J Exp Biol 200(Pt 23): 2947-2955. 168 Schofield, PR, Darlison, MG, Fujita, N, Burt, DR, Stephenson, FA, Rodriguez, H, Rhee, LM, Ramachandran, J, Reale, V, Glencorse, TA, et al. (1987) Sequence and functional expression of the GABAA receptor shows a ligand-gated receptor super-family. Nature 328(6127): 221-227. Schreiber, G, Fersht, AR (1995) Energetics of protein-protein interactions: analysis of the barnase-barstar interface by single mutations and double mutant cycles. J Mol Biol 248(2): 478-486. Shi, J, Blundell, TL, Mizuguchi, K (2001) FUGUE: sequence-structure homology recognition using environment-specific substitution tables and structure-dependent gap penalties. J Mol Biol 310(1): 243-257. Shirai, Y, Hosie, AM, Buckingham, SD, Holyoke, CW, Jr., Baylis, HA, Sattelle, DB (1995) Actions of picrotoxinin analogues on an expressed, homo-oligomeric GABA receptor of Drosophila melanogaster. Neurosci Lett 189(1): 1-4. Shotkoski, F, Lee, HJ, Zhang, HG, Jackson, MB, ffrench-Constant, RH (1994) Functional expression of insecticide-resistant GABA receptors from the mosquito Aedes aegypti. Insect Mol Biol 3(4): 283-287. Shotkoski, F, Zhang, HG, Jackson, MB, ffrench-Constant, RH (1996) Stable expression of insect GABA receptors in insect cell lines. Promoters for efficient expression of Drosophila and mosquito Rdl GABA receptors in stably transformed mosquito cell lines. FEBS Lett 380(3): 257-262. Sieghart, W (1995) Structure and pharmacology of γ-aminobutyric acid A receptor subtypes. Pharmacol Rev 47(2): 181-234. Sigel, E, Baur, R, Kellenberger, S, Malherbe, P (1992) Point mutations affecting antagonist affinity and agonist dependent gating of GABAA receptor channels. EMBO J 11(6): 2017-2023. Sine, SM, Engel, AG (2006) Recent advances in Cys-loop receptor structure and function. Nature 440(7083): 448-455. Sixma, TK, Smit, AB (2003) Acetylcholine binding protein (AChBP): a secreted glial protein that provides a high-resolution model for the extracellular domain of pentameric ligand-gated ion channels. Annu Rev Biophys Biomol Struct 32: 311-334. Smit, AB, Syed, NI, Schaap, D, van Minnen, J, Klumperman, J, Kits, KS, Lodder, H, van der Schors, RC, van Elk, R, Sorgedrager, B, Brejc, K, Sixma, TK, Geraerts, WP (2001) A glia-derived acetylcholine-binding protein that modulates synaptic transmission. Nature 411(6835): 261- 268. 169 Somogyi, P, Fritschy, JM, Benke, D, Roberts, JD, Sieghart, W (1996) The γ2 subunit of the GABAA receptor is concentrated in synaptic junctions containing the α1 and β2/3 subunits in hippocampus, cerebellum and globus pallidus. Neuropharmacology 35(9-10): 1425-1444. Spier, AD, Lummis, SC (2000) The role of tryptophan residues in the 5-HT3 receptor ligand binding domain. J Biol Chem 275(8): 5620-5625. Staley, KJ, Soldo, BL, Proctor, WR (1995) Ionic mechanisms of neuronal excitation by inhibitory GABAA receptors. Science 269(5226): 977-981. Stein, LD, Bao, Z, Blasiar, D, Blumenthal, T, Brent, MR, Chen, N, Chinwalla, A, Clarke, L, Clee, C, Coghlan, A, Coulson, A, D'Eustachio, P, Fitch, DH, Fulton, LA, Fulton, RE, Griffiths-Jones, S, Harris, TW, Hillier, LW, Kamath, R, Kuwabara, PE, Mardis, ER, Marra, MA, Miner, TL, Minx, P, Mullikin, JC, Plumb, RW, Rogers, J, Schein, JE, Sohrmann, M, Spieth, J, Stajich, JE, Wei, C, Willey, D, Wilson, RK, Durbin, R, Waterston, RH (2003) The genome sequence of Caenorhabditis briggsae: a platform for comparative genomics. PLoS Biol 1(2): E45. Strambi, C, Cayre, M, Sattelle, DB, Augier, R, Charpin, P, Strambi, A (1998) Immunocytochemical mapping of an RDL-like GABA receptor subunit and of GABA in brain structures related to learning and memory in the cricket Acheta domesticus. Learn Mem 5(1-2): 78-89. Sun, L, Dong, H, Guo, C, Qian, J, Sun, J, Ma, L, Zhu, C (2006) Larvicidal activity of extracts of Ginkgo biloba exocarp for three different strains of Culex pipiens pallens. J Med Entomol 43(2): 258-261. Swensen, AM, Golowasch, J, Christie, AE, Coleman, MJ, Nusbaum, MP, Marder, E (2000) GABA and responses to GABA in the stomatogastric ganglion of the crab Cancer borealis. J Exp Biol 203(Pt 14): 2075-2092. Taylor, CW, Laude, AJ (2002) IP3 receptors and their regulation by calmodulin and cytosolic Ca2+. Cell Calcium 32(5-6): 321-334. Thompson, AJ, Lochner, M, Lummis, SC (2007) The antimalarial drugs quinine, chloroquine and mefloquine are antagonists at 5-HT3 receptors. Br J Pharmacol 151(5): 666-677. Thompson, AJ, Lochner, M, Lummis, SC (2008) Loop B is a major structural component of the 5-HT3 receptor. Biophys J 95(12): 5728-5736. Thompson, AJ, Lummis, SC (2007) The 5-HT3 receptor as a therapeutic target. Expert Opin Ther Targets 11(4): 527-540. 170 Thompson, AJ, Lummis, SC (2003) A single ring of charged amino acids at one end of the pore can control ion selectivity in the 5-HT3 receptor. Br J Pharmacol 140(2): 359-365. Tzvetkov, MV, Meineke, C, Oetjen, E, Hirsch-Ernst, K, Brockmoller, J (2007) Tissue-specific alternative promoters of the serotonin receptor gene HTR3B in human brain and intestine. Gene 386(1-2): 52-62. Ulens, C, Akdemir, A, Jongejan, A, van Elk, R, Bertrand, S, Perrakis, A, Leurs, R, Smit, AB, Sixma, TK, Bertrand, D, de Esch, IJ (2009) Use of acetylcholine binding protein in the search for novel α7 nicotinic receptor ligands. In silico docking, pharmacological screening, and X- ray analysis. J Med Chem 52(8): 2372-2383. Unwin, N (1995) Acetylcholine receptor channel imaged in the open state. Nature 373(6509): 37-43. Unwin, N, Miyazawa, A, Li, J, Fujiyoshi, Y (2002) Activation of the nicotinic acetylcholine receptor involves a switch in conformation of the α subunits. J Mol Biol 319(5): 1165-1176. Urbanics, R, Leniger-Follert, E, Lubbers, DW (1978) Time course of changes of extracellular H+ and K+ activities during and after direct electrical stimulation of the brain cortex. Pflugers Arch 378(1): 47-53. Verdonk, ML, Cole, JC, Hartshorn, MJ, Murray, CW, Taylor, RD (2003) Improved protein-ligand docking using GOLD. Proteins 52(4): 609- 623. Wagner, DA, Czajkowski, C, Jones, MV (2004) An arginine involved in GABA binding and unbinding but not gating of the GABAA receptor. J Neurosci 24(11): 2733-2741. Wang, J, Lester, HA, Dougherty, DA (2007) Establishing an ion pair interaction in the homomeric ρ1 γ-aminobutyric acid type A receptor that contributes to the gating pathway. J Biol Chem 282(36): 26210- 26216. Westh-Hansen, SE, Rasmussen, PB, Hastrup, S, Nabekura, J, Noguchi, K, Akaike, N, Witt, MR, Nielsen, M (1997) Decreased agonist sensitivity of human GABAA receptors by an amino acid variant, isoleucine to valine, in the α1 subunit. Eur J Pharmacol 329(2-3): 253-257. Westh-Hansen, SE, Witt, MR, Dekermendjian, K, Liljefors, T, Rasmussen, PB, Nielsen, M (1999) Arginine residue 120 of the human GABAA receptor α1, subunit is essential for GABA binding and chloride ion current gating. Neuroreport 10(11): 2417-2421. 171 Wilkins, ME, Hosie, AM, Smart, TG (2002) Identification of a β subunit TM2 residue mediating proton modulation of GABA type A receptors. J Neurosci 22(13): 5328-5333. Wilkins, ME, Hosie, AM, Smart, TG (2005) Proton modulation of recombinant GABAA receptors: influence of GABA concentration and the β subunit TM2-TM3 domain. J Physiol 567(Pt 2): 365-377. Wilson, G, Karlin, A (2001) Acetylcholine receptor channel structure in the resting, open, and desensitized states probed with the substituted- cysteine-accessibility method. Proc Natl Acad Sci U S A 98(3): 1241- 1248. Wilson, GG, Karlin, A (1998) The location of the gate in the acetylcholine receptor channel. Neuron 20(6): 1269-1281. Woodward, RM, Polenzani, L, Miledi, R (1993) Characterization of bicuculline/baclofen-insensitive (ρ-like) γ-aminobutyric acid receptors expressed in Xenopus oocytes. II. Pharmacology of γ-aminobutyric acid A and γ-aminobutyric acid B receptor agonists and antagonists. Mol Pharmacol 43(4): 609-625. Xiu, X, Puskar, NL, Shanata, JA, Lester, HA, Dougherty, DA (2009) Nicotine binding to brain receptors requires a strong cation-π interaction. Nature 458(7237): 534-537. Zhang, D, Pan, ZH, Zhang, X, Brideau, AD, Lipton, SA (1995a) Cloning of a γ-aminobutyric acid type C receptor subunit in rat retina with a methionine residue critical for picrotoxinin channel block. Proc Natl Acad Sci U S A 92(25): 11756-11760. Zhang, HG, ffrench-Constant, RH, Jackson, MB (1994) A unique amino acid of the Drosophila GABA receptor with influence on drug sensitivity by two mechanisms. J Physiol 479 ( Pt 1): 65-75. Zhang, HG, Lee, HJ, Rocheleau, T, ffrench-Constant, RH, Jackson, MB (1995b) Subunit composition determines picrotoxin and bicuculline sensitivity of Drosophila γ-aminobutyric acid receptors. Mol Pharmacol 48(5): 835-840. Zhang, J, Xue, F, Chang, Y (2008) Structural determinants for antagonist pharmacology that distinguish the ρ1 GABAC receptor from GABAA receptors. Mol Pharmacol 74(4): 941-951. Zheng, Y, Hirschberg, B, Yuan, J, Wang, AP, Hunt, DC, Ludmerer, SW, Schmatz, DM, Cully, DF (2002) Identification of two novel Drosophila melanogaster histamine-gated chloride channel subunits expressed in the eye. J Biol Chem 277(3): 2000-2005. 172 Zhong, W, Gallivan, JP, Zhang, Y, Li, L, Lester, HA, Dougherty, DA (1998) From ab initio quantum mechanics to molecular neurobiology: a cation- π binding site in the nicotinic receptor. Proc Natl Acad Sci U S A 95(21): 12088-12093. Zhu, XZ, Li, XY, Liu, J (2004) Recent pharmacological studies on natural products in China. Eur J Pharmacol 500(1-3): 221-230. Zouridakis, M, Zisimopoulou, P, Poulas, K, Tzartos, SJ (2009) Recent advances in understanding the structure of nicotinic acetylcholine receptors. IUBMB Life 61(4): 407-423.