Ethanol-oxytocin Interactions at Homomeric Glycine Receptors

A thesis submitted to fulfil requirements for the degree of Master of Philosophy

-

Dominique Taylor Faculty of Pharmacy The University of Sydney 2018

Statement of Authorship & Originality

This is to certify that to the best of my knowledge, the content of this thesis is my own work. This thesis has not been submitted for any degree or other purposes.

I certify that the intellectual content of this thesis is the product of my own work and that all the assistance received in preparing this thesis and sources have been acknowledged.

Dominique Taylor

Acknowledgements I would like to acknowledge my supervisor Dr Mary Collins (Chebib) for her expert advice throughout my project. Thanks to my co-supervisors Dr Nathan Absalom for all of his assistance in practical and theoretical aspects of the research; and Dr Michael Bowen for all his assistance.

I would also like to thank Dr Steven Devenish for his expertise on glycine receptors, and for generous assistance in supply and manufacture of glycine receptors. I’m also grateful to all the other members of the lab; (including but not limited to) Tim Bakas, Dr Petra van Nieuwenhuijzen, Dr Vivian Liao, Dinesh Indurthi, Taima Quddah, Dr Philip Ahring, Leonny Hartiadi, for their company and frequent guidance in standard lab activities.

A special mention to my husband Ben, along with Henry, Charlie, Tilly, Lulu, Jack and Scruff Johansson, who differentially supported, distracted, confused, and amused me throughout this endeavour. Justin and Stacey Taylor also get a shout out for bestowing me with some of their mad critical thinking skills.

[ii] Contents Statement of Authorship & Originality ...... ii Acknowledgements ...... ii List of Figures ...... v List of Tables ...... v List of Abbreviations ...... vi Abstract ...... vii Chapter 1: All About Alcohol’s Actions ...... 1

1.1 GABAA Receptors Mediate Many of Alcohol’s Effects ...... 3

1.1.1 GABAA Receptor Functions and Properties ...... 4 1.1.2 Characteristics of δ Subunit-Containing Receptors ...... 4 1.1.3 Why are δ-GABAA Receptors a Likely Target? ...... 6 1.2 A Review of Glycine Receptors ...... 7 1.2.1 Glycine Receptor Subunits Influence Receptor Characteristics ...... 10 1.3 Glycine Receptors Mediate Ethanol Effects and Behaviours ...... 12 1.3.1 Ethanol Modulates Glycine Receptors in Reward Pathways ...... 13 1.3.2 Zinc and Ethanol Synergistically Modulate Glycine Receptors ...... 14 1.3.3 Mechanism of Ethanol Action at Glycine Receptors ...... 15 1.4 Oxytocin-ethanol Interactions at Glycine Receptors ...... 16 1.4.1 Oxytocin and Alcohol ...... 18 1.4.2 Oxytocin-ethanol Interactions at Ligand-gated Ion Channels ...... 19 1.5 Research Question & Aims ...... 21 1.5.1 Significance ...... 21 Chapter 2: Experimental Procedures ...... 22 2.1 Molecular Biology ...... 22 2.1.1 Materials ...... 22 2.1.2 Transformation & Growth of DNA Plasmids ...... 22 2.1.3 Linearisation and Transcription of DNA ...... 23 2.1.4 Site-Directed Mutagenesis ...... 23 2.2 Electrophysiology ...... 24 2.2.1 Materials ...... 24 2.2.2 Xenopus laevis Oocyte Preparation ...... 24 2.2.3 Expression of Recombinant Glycine Receptors in Xenopus Oocytes ...... 25 2.2.4 Electrophysiological Recording from Recombinant Receptors ...... 26 2.2.5 Data and Statistical Analysis ...... 27 Chapter 3: Results ...... 28

[iii] 3.1 Expression of Homomeric Glycine Receptors in Xenopus Oocytes ...... 28 3.2 Effects of Ethanol on Wild Type α1 and α2 Glycine Receptors ...... 31 3.2.1 Zinc-induced Potentiation of α1 Glycine Receptors ...... 33 3.3 Oxytocin-glycine Interactions at α1 and α2 GlyRs ...... 34 3.4 Ethanol-oxytocin Interactions at Wild Type α1 GlyRs...... 35 3.5 Ethanol-oxytocin Interactions at α2 GlyRs ...... 37 3.6 Ethanol-oxytocin Interactions at α1δL2 Glycine Receptors ...... 39 Chapter 4: Discussion and Conclusion ...... 41 4.1 Ethanol-induced Potentiation of Glycine Receptors ...... 41 4.1.1 Zinc Modulation of α1 GlyRs ...... 43 4.2 Interactions of Oxytocin and Ethanol ...... 43 4.2.1 Vasopressin Does Not Modulate Ethanol’s Effects ...... 45 4.2.2 Oxytocin May Interact with Zinc at α1 GlyRs ...... 46 4.3 Oxytocin-ethanol Interactions at α1δL2 GlyRs ...... 47 4.4 Conclusion and Further Considerations ...... 48 4.4.1 Further Considerations and Future Directions ...... 49 4.4.2 Limitations of These Experiments ...... 50 References ...... 52

[iv] List of Figures

Figure 1. Ethanol binding sites at αβδ/γ GABAA receptors ...... 6 Figure 2. Schematic diagram of the ECD and subunit composition of glycine receptors ...... 7 Figure 3. Structural and binding properties of an α1 homomeric GlyR ...... 9 Figure 4. Ethanol binding pocket at the α1 subunit of the glycine receptor...... 16 Figure 5. The Molecular Structure of Nonapeptides Oxytocin and Arginine Vasopressin ...... 18 Figure 6. Microinjection of mRNA into Xenopus laevis oocyte...... 25 Figure 7. Two-electrode voltage clamp schematic diagram ...... 26 Figure 8. C-R curves for glycine at α1 and α2 WT GlyRs ...... 28 Figure 9. C-R curve comparison of α1dL2 mutant and α1 GlyRs ...... 30 Figure 10. Ethanol induced potentiation of wild type α1 and α2 receptors ...... 32 Figure 11. Zinc modulates activity at α1 GlyRs ...... 33 Figure 12. Effects of oxytocin on glycine-gated current at WT GlyRs ...... 34 Figure 13. Summary of oxytocin-ethanol interactions at α1 GlyRs ...... 35 Figure 14. Concentration dependent effects of OT on EtOH activity at WT α1 GlyRs ...... 36 Figure 15. Effects of OT on EtOH-induced potentiation of α2 GlyRs ...... 37 Figure 16. Summary of oxytocin-ethanol and vasopressin interactions at α2 GlyRs ...... 38 Figure 17. Ethanol induced potentiation of α1δL2 GlyRs expressed in Xenopus oocytes...... 39 Figure 18. Summary of EtOH-OT interactions at ethanol-sensitive α1δL2 mutant GlyRs ...... 40

List of Tables

Table 1. Summary of Plasmid Properties ...... 22 Table 2. Sense Oligonucleotide Primer Sequences ...... 23 Table 3. QuikChange II PCR Cycle Conditions ...... 24 Table 4. Summary of nonlinear regression results for α1 and α2, and α1δL2 GlyR...... 29

[v] List of Abbreviations

Abbreviation Term AVP vasopressin; arginine vasopressin BZ benzodiazepine CA3 Cornus Ammonis area 3 (hippocampus) cDNA complementary DNA CR concentration-response CSF cerebrospinal fluid DRG dorsal root ganglion (spinal cord) ECD extracellular domain (of Cys-loop receptor) EtOH ethanol; ethyl alcohol GABA gamma-aminobutyric acid; γ-aminobutyric acid

GABAA ionotropic γ-aminobutyric acid receptor subtype A Gly glycine GlyR glycine receptor GlyT glycine transporter HC hippocampus ICD intracellular domain (of Cys-loop receptor) LGIC ligand-gated ion channel LORR loss of righting reflex NAcc nucleus accumbens nAChR nicotinic acetylcholine receptor NDMAR N-methyl-D-aspartate receptor NMDA N-methyl-D-aspartate OT oxytocin OXTR oxytocin receptor SD standard deviation SEM standard error of the mean TMD transmembrane domain (of Cys-loop receptor) VTA ventral tegmental area WT wild type ZnT zinc transporter

[vi] Abstract

Alcohol is one of the most widely used drugs, yet its targets in the brain have not been reliably established, and effective treatments for alcohol addiction are lacking. Various ligand-gated ion channels (LGICs) are established targets of ethanol in the central nervous system, including

NMDA receptors and the Cys-loop γ-aminobutyric acid type A (GABAA), 5-HT3, nicotinic acetylcholine (nACh), and glycine (Gly) receptors (Harris et al 2008, Spanagel 2009).

Historically, the role of GABAA receptors in alcohol’s sedative, anxiolytic, and depressant effects has been supposed. More recently, extrasynaptic δ subunit-containing GABAA receptors have come to light as targets of behaviourally relevant concentrations of alcohol in the brain (Olsen et al 2007).

Glycine receptors (GlyRs) have also been implicated in ethanol’s (EtOH) effects on motor coordination, reward pathways, sensory processing and perception. Furthermore, glycine receptors may be involved in regulating alcohol consumption (Perkins et al 2010, Soderpalm et al 2017). Ethanol may interact with the receptor via residues in transmembrane M2 and M3 regions, in addition to residues in Loop 2 of the extracellular domain (Burgos et al 2015, Crawford et al 2008, Horani et al 2015). Loop 2 residues of the extracellular domain may mediate signal transduction in the channel of both GABAA and glycine receptors (Cederholm et al 2010), and may be important in ethanol’s actions at these receptors (Crawford et al 2007).

Furthermore, Loop 2 of the GABAA receptor δ subunit may be necessary for its high sensitivity to ethanol, and an ethanol ultra-sensitive mutant GlyR may be created through insertion of this sequence into the α subunit (Perkins et al 2009).

Recent behavioural and electrophysiological experiments have demonstrated that the neuropeptide oxytocin (OT) attenuates ethanol-induced motor impairment in rats, and prevents ethanol-induced potentiation of GABA-gated currents via selective and specific interaction at δ-GABAA receptors (Bowen et al 2015). Although oxytocin’s binding site remains unclear, its interactions at these receptors presents a useful tool for investigation of the receptor conformations necessary for both alcohol and oxytocin’s actions. The actions of oxytocin at glycine receptors have not yet been characterised, although the functional and structural similarities between GABAA and Gly receptors - particularly as inhibitory LGIC targets of ethanol - suggested that a similar oxytocin-ethanol interaction could be possible at glycine receptors.

Overall, this project aimed to investigate the interactive effects of ethanol and oxytocin at recombinant human homomeric α1 and α2 GlyRs. The role of Loop 2 in oxytocin’s effects and interactions with ethanol at ethanol-sensitive mutant α1δL2 GlyRs was also a focus of this [vii] investigation. To achieve this, human wild type α1, α2, and mutant α1δL2 homomers were expressed recombinantly in Xenopus laevis oocytes. Two-electrode voltage clamp electrophysiology was then used to evaluate the effects of EtOH and OT at these receptors.

Presently, ethanol was found to potentiate α1 and α2 GlyRs at pharmacologically relevant concentrations, with the threshold concentration for potentiation greatly reduced (from 30 mM EtOH to as low as 1 mM) in the α1δL2 mutant receptor. At all of these receptors, oxytocin significantly attenuated ethanol-induced potentiation of glycine-gated current when co- applied at low micromolar concentrations. Importantly, neither ethanol nor oxytocin activated the receptors independently, and oxytocin did not modulate glycine currents in the absence of ethanol. The structurally similar neuropeptide arginine vasopressin (AVP) did not modulate ethanol potentiation when tested in place of OT at α2 receptors, suggesting structural moieties unique to the oxytocin molecule may allow modulation of ethanol’s actions at these receptors.

In addition to their significantly increased sensitivity to ethanol, Loop 2-mutated GlyRs were more sensitive to glycine, although the general function of the receptor appeared unchanged. Importantly, the modulatory effects of oxytocin upon ethanol was preserved across wild type α1 and mutant α1δL2 receptors, indicating that Loop 2 is unlikely to be involved in mediating oxytocin’s interactions with the glycine receptor. However, the binding site for oxytocin at glycine and GABAA receptors has not been identified, necessitating further investigation into this novel action of oxytocin in the brain.

[viii] Chapter 1: All About Alcohol’s Actions

Alcohol use is prevalent in many modern societies, and is often used recreationally to facilitate social lubrication and relaxation. Despite this, alcohol misuse gives rise to a range of toxic and systemic side effects, and is often a participant in poly-drug use (Williams & Parker 2001). In 2016, 77% of Australians over 14 years had consumed alcohol in the past 12 months, with 36% of those over 12 years having engaged in significant “binge drinking” within the past year (Claydon et al 2016). While chronic alcohol abuse is associated with a host of health problems, the major risks from such acute intoxication involve accidental harm to oneself and others, with these risks increasing with the amount of alcohol ingested. The combination of alcohol- induced motor impairment, behavioural effects, and loss of inhibition contributes to issues such as road accidents, risk-taking behaviour, and alcohol-related violence. Knowledge of how alcohol acts in the brain to produce these effects may be important for understanding the link between its use and these behaviours. Though multiple targets of alcohol in the central nervous system have been identified, there is little agreement upon the principal mechanisms through which physiologically relevant concentrations of alcohol generate intoxication.

Commonly consumed for recreation, sedation, and anxiolysis, ethyl alcohol is the drug of choice for many (Claydon et al 2016, Williams & Parker 2001), however the mechanisms through which it exerts these effects have remained elusive. Following typical oral ingestion, ethanol is absorbed via passive diffusion; readily crossing biological membranes to become distributed throughout the body. Ethanol is highly water soluble, with high membrane permeability so readily crosses the blood brain barrier, allowing it to interact with a range of potential neural substrates. At sufficiently high concentrations, alcohol may affect a range of biological targets, including glutamate (NMDA), GABA, glycine, serotonin (5-HT3) and nicotinic acetylcholine (nACh) receptors, L-type calcium channels, and G protein-coupled inwardly rectifying potassium (GIRK) channels (Spanagel 2009). Additionally, ethanol undergoes primarily hepatic metabolism by alcohol dehydrogenase to form the carcinogenic metabolite acetaldehyde, which then yields acetate and a hydrogen atom via acetaldehyde dehydrogenase.

In Australia and much of the European Union, the legal driving limit on blood alcohol concentration is 0.05% wt/vol, corresponding to a blood concentration of around 11 mM (or 50 mg/dL). Below this concentration, performance in skilled tasks may be impaired, accompanied by talkativeness and a mild sense of euphoria (Vonghia et al 2008). However, beyond this limit, motor and reflex impairment is typically apparent, while altered perception, impaired judgement, and sedation may also occur. By comparison, concentrations above 30

[1] mM typically produce signs of acute toxicity, including vomiting and substantial loss of control; while lethal concentrations range between 50 and 110 mM (Gable 2004, Olsen et al 2007). Because ethanol readily crosses the blood brain barrier, concentrations found in the brain are likely to resemble circulating levels. The effects of acute alcohol intoxication are loosely correlated with blood alcohol concentrations, though may be influenced by other factors including individual tolerance, period of consumption, and body weight (Vonghia et al 2008). Recreational, intoxicating self-administration of alcohol may therefore generate brain concentrations between 3 and 30 mM, although the lower, less acutely toxic levels may be more relevant to typical moderate consumption.

Interestingly, a number of reports suggest that some individuals are able to tolerate and survive extremely high doses of alcohol. In the USA, Johnson et al (1982) reported the survival of a young woman following hospital admission (for abdominal pain) with a serum alcohol level of 1.5% (BAC 10-15% lower than this). In this case, the patient presented as ambulant, alert, and responsive - though slightly confused. Over a two year period, Urso et al (1981) also reported over 70 individuals who had presented to a hospital emergency room (primarily due to injuries) who had recently consumed alcohol. Despite appearing sober (ambulant, responsive, able to perform simple tasks), these individuals were found to have blood alcohol concentrations between 120 and 540 mg/dL. As such, blood alcohol concentration is not necessarily an accurate predictor of the behavioural and physiological effects which accompany intoxication - particularly for those with frequent binge use or alcohol dependence. Conversely, in alcohol-naïve or intolerant individuals, the threshold for severe intoxication and lethality may be much lower than expected.

Although sufficient levels of alcohol interact with a wide range of targets in the brain, members of the Cys-loop family of ligand-gated ion channels (LGICs) appear to be modulated by doses of alcohol within the range typically consumed by humans (reviewed in Howard et al 2014). These receptors may share a common region for allosteric or intersubunit interactions with ethanol, due to their relatively conserved structure and subunit arrangement. The cation- conducting nACh and 5-HT3 receptors may be potentiated by ethanol, though are inhibited by long-chain alcohols (Howard et al 2014, Zuo et al 2004). The anion-conducting GABAA and Gly Cys-loop receptors are also some of the best illustrated targets of ethanol in the CNS.

Potentiation of inhibitory GABAA receptors is thought to be largely responsible for its sedative and hypnotic effects, both through enhancement of inhibition and the suppression of excitatory signals. Together with GABAA receptors, glycine receptors present another established mediator of alcohol in the central nervous system. Glycine receptors likely contribute to some of alcohol’s effects upon reward pathways, sensory processing, motor and

[2] respiratory control, and may be involved in regulating alcohol consumption (Perkins et al 2010, Soderpalm et al 2017). This chapter will explore the pivotal role of inhibitory Cys-loop receptors in generating the behavioural and cognitive effects of ethanol. While the next section focuses on GABAA receptors, the role of glycine receptors as ethanol targets is discussed in detail in Section 1.2.

1.1 GABAA Receptors Mediate Many of Alcohol’s Effects

GABAA receptors have long been posited as one of ethanol’s main targets in the brain, and are also targets for barbiturates and benzodiazepines such as diazepam, which generate sedative, hypnotic, anxiolytic, and anticonvulsant effects. Due to the pharmacological similarities in behavioural effects of these drugs, alcohol has historically been presumed to act at GABAA receptors. Involvement of the GABAergic system in mediating alcohol’s effects is also suggested by the neuroadaptations that arise from sustained chronic alcohol use. Some of these neuroadaptive changes include downregulation of GABAA receptors, which may facilitate symptoms of alcohol withdrawal syndrome, including anxiety, seizure, and agitation. However, the more common synaptic GABAA receptors do not appear responsive to physiologically relevant doses of alcohol (Wallner et al 2006b). Effects on the majority of GABAA receptors are only observed beyond 60 mM – far above (though possible) levels resulting from typical self- administration and recreational use (Olsen et al 2007). At such high concentrations, ethanol is likely to act non-specifically at a variety of other targets in addition to GABAA receptors, thereby limiting the validity of in vitro responses to these levels (Spanagel 2009).

Extrasynaptic delta subunit-containing GABAA receptors (δ GABAA receptors) have since come to light as a major target of more relevant doses of alcohol, such as the low millimolar concentrations arising from consumption of half a glass of wine (Wallner et al 2006b). So far, alcohol’s potentiating effects at these receptors have been demonstrated in a range of cell types including neurons, HEK cells, and Xenopus laevis oocytes (Olsen et al 2007), while metabolomic experiments have suggested similarities in the metabolic footprints of low- dose alcohol and δ-GABAA receptor specific compounds (Rae et al 2014). However, a small number of studies have failed to find any effects of low dose alcohol at δ GABAA receptors (Borghese et al 2006); a number of explanations for this have been suggested, such as difficulties incorporating the delta subunit, variable conditions in recombinant systems, and variations in receptor stoichiometry (Hurley et al 2009, You & Dunn 2007).

[3] 1.1.1 GABAA Receptor Functions and Properties

As the main inhibitory neurotransmitter in the brain, y-aminobutyric acid (GABA) plays a pivotal role in the function of the central nervous system. In the brain, GABA acts at two main classes of receptors; ionotropic GABAA, and G-protein coupled GABAB receptors. GABAA receptors are Cys-loop ligand-gated chloride channels, and are widely expressed throughout the brain. Binding of GABA to these receptors induces a conformational change, allowing the influx of chloride anions through the channel pore - this serves to hyperpolarise the postsynaptic cell membrane and dampen incoming postsynaptic excitatory potentials. As functional heteropentamers, most GABAA receptors comprise two α, two β, and one γ subunit, with θ, δ, or ε subunits replacing the γ subunit in different regions of the brain. Subunit composition determines receptor properties and function, distribution within the cellular membrane, and anatomical distribution throughout the brain (Korpi et al 2002, Sieghart & Sperk 2002, Wisden et al 1992). While the more common γ subunit-containing receptors are generally located within synapses, receptors containing the δ subunit are located extrasynaptically (Herd et al 2008, Wallner et al 2003).

1.1.2 Characteristics of δ Subunit-Containing Receptors

δ subunit-containing GABAA receptors are found extrasynaptically, where low ambient concentrations necessitate their high sensitivity and affinity to GABA (Herd et al 2008, Wallner et al 2003). The δ subunit generally assembles with two α4 or α6, and two β3 subunits (Sur et al 1999), and confers distinct properties to functional receptors. α4β3δ and α6β2/3δ receptors appear to have the highest affinity to GABA, however the efficacy of GABA at these receptors is three-fold lower than at α4β3 and α6β2/3 receptors (Zheleznova et al 2009). δ-GABAA receptors are also distinguishable from synaptic receptors due to their benzodiazepine insensitivity. Consistent with their role in tonic activity, these receptors have an EC50 for GABA in the low micromolar range (EC50 ≈1.4 μM for α4β3δ), and have a low level of desensitisation upon prolonged exposure to GABA (Mody et al 2007, You & Dunn 2007). In addition to the presence of the delta subunit, other subunits contribute to the unique properties of these receptors. For instance, α4βδ receptors are highly plastic, and appear to be rapidly upregulated in response to certain hormones, in addition to being sensitive to <10 mM alcohol (Carver et al 2014, Sundstrom-Poromaa et al 2002). Receptors incorporating β3 subunits are also more sensitive to alcohol than those containing β2 (Wallner et al 2003).

Activity at extrasynaptic receptors contributes to diffuse and long-lasting tonic inhibition (Belelli et al 2009), which modulates excitability of the postsynaptic membrane. Tonic inhibition is the persistent activation of GABAA receptors located outside of the synapse, by

[4] low and extracellular concentrations of GABA. Extrasynaptic δ-GABAA receptors are therefore able to modulate excitability of the postsynaptic membrane. Although they comprise around

10% of all GABAA receptors in the brain, their persistent inhibitory activity is essential for accurate signal transduction in the brain (Mody et al 2007). In contrast, phasic inhibition occurs within the synapse, where presynaptic release of GABA activates receptors on the postsynaptic membrane, facilitating intermittent hyperpolarisation and inhibition of the cell.

GABAA receptors predominantly containing γ subunits are largely responsible for generation of these phasic inhibitory currents. Tonic inhibition may also supplement this phasic inhibition, and facilitate the function of inhibitory synapses (Brickley et al 1996). The anatomical distribution of these tonically active δ GABAA receptors in the brain reflects many of the effects seen with low to moderate intoxication – including the effects on areas of substantial δ expression such as the cerebellum, dentate gyrus, and thalamus brain (Laurie et al 1992, Wisden et al 1992). Here, the link between subunit distribution and in vivo effects may reflect ethanol-induced enhancement of tonic inhibition.

In the cerebellum, dense perisynaptic expression of α6βδ receptors occurs in cerebellar granule cells (Hamann et al 2002). These cells are the most abundant neurons in the brain, while around 90% of tonic inhibition of these cells arises from αβδ activity, suggesting alterations in tonic inhibition of these cells could have important effects. Granule cells receive excitatory input from the cerebral cortex via mossy fibres regarding motor planning and initiation, in addition to GABAergic feedback inhibition from Golgi cells. They also send excitatory signals to Purkinje cells, the main motor output neurons of the cerebellum. Tonic inhibition via α6βδ receptors is important for moderating granule cell excitability and activation (Duguid et al 2012). By enhancing tonic inhibition, ethanol may decrease granule cell excitability - dampening cortical inputs, and preventing output to Purkinje cells - thereby interfering with motor input and output. Given the role of the cerebellum in motor control, these α6βδ receptors probably contribute to ethanol-induced ataxia and impairment seen in vivo. Rats with a single nucleotide polymorphism in the α6 subunit gene also demonstrated heightened ethanol-induced impairment on cerebellum dependent tasks, accompanied by increased in tonic inhibition at these receptors, further supporting the role of enhanced inhibitory tone in contributing to these effects (Hanchar et al 2005).

The thalamus is a midbrain structure involved in sensory relay, and critical to the control of sleep and wakefulness. α4βδ receptors are heavily expressed in thalamic relay nuclei, which are essential for the generation of sleep waves, therefore enhancement of tonic inhibition on these cells by alcohol may generate sedation, and contribute to alcohol-induced sleep disruption. As part of the hippocampus, the dentate gyrus is involved in memory formation

[5] and retrieval. Dentate gyrus granule cells (DGGCs) express α4β3δ and possibly α4βδ receptors, which participate in tonic inhibition of these cells (Herd et al 2008, Nusser & Mody 2002). As the principal excitatory cells of the hippocampus, DGGCs relay information from the entorhinal cortex to inhibitory interneurons of the CA3 region and elsewhere in the hippocampus. Ethanol-induced potentiation of tonic inhibition may interfere with these circuits, thereby contributing to memory impairments and even the fragmentary blackouts seen with sufficient doses of alcohol in humans (Wetherill & Fromme 2011).

1.1.3 Why are δ-GABAA Receptors a Likely Target?

In addition to correlations between the anatomical distribution of αβδ receptors and known effects of alcohol intoxication in humans, several lines of evidence also suggest that these extrasynaptic receptors are targets of behaviourally and physiologically relevant doses of alcohol. A number of groups have demonstrated low dose ethanol-induced potentiation of

GABA-gated current at recombinant δ-GABAA receptors, in human embryonic kidney (HEK) cells (Wallner et al 2006b) and Xenopus laevis oocytes (Bowen et al 2015, Wallner et al 2014), in addition to assays and recording from mammalian brain slices (Glykys et al 2007, Hanchar et al 2005, Wei et al 2004). Such data indicate that concentrations of ethanol as low as 3 mM potentiate GABA current at recombinant δ-GABAA receptors in Xenopus oocytes, corresponding to the doses at which subjective feelings of intoxication arise (Olsen et al 2007). Investigation by Perkins et al (2009) has suggested Loop 2 is responsible for the high ethanol sensitivity of these receptors.

Figure 1. Ethanol binding sites at αβδ/γ GABAA receptors. GABA, THIP and muscimol binding sites are indicated at the α-β interface, with ethanol (EtOH) and benzodiazepine (BZ) residues or binding pockets shown (according to legend symbols) at subunit interfaces. Modified from Wallner et al (2014).

[6] Further support for the role of these receptors in mediating alcohol’s effects comes from Ro15-4513; a behavioural antagonist of alcohol, able to attenuate alcohol-induced ataxia and motor impairment, sedation, and cognitive impairment. Experiments by Wallner et al (2006a) demonstrated that 100 nM Ro15-4513 antagonises (3-30 mM) ethanol-induced potentiation of GABA-activated currents at recombinant αβ3δ receptors expressed in Xenopus oocytes, and attenuates alcohol’s enhancing effects on tonic inhibition in the cerebellum (Santhakumar et al 2007). Taken together, this seems to suggest that these receptors (which are sensitive to both compounds) are important mediators of the effects of behaviourally relevant alcohol doses.

The binding site of ethanol at GABAARs has not been reliably established, though may involve a unique benzodiazepine binding site of the ethanol antagonist Ro15-4513, or a homologous binding site (Wallner et al 2014). Based on their investigation of Ro15-4513, Wallner et al. have proposed these binding sites of alcohol, shown in Figure 1. These allosteric sites are thought to occur at pockets present at the α-β and δ-β subunit interfaces.

1.2 A Review of Glycine Receptors

Glycine is the smallest amino acid, with a single amino acid comprising its side chain. As a non- essential amino acid, glycine may be synthesised from serine in a reversible reaction catalysed by a hydroxymethyltransferase enzyme. Although it primarily functions as a protein precursor, glycine (Gly) is also one of the main inhibitory neurotransmitters in the brain and spinal cord. Ionotropic glycine receptors are responsible for this inhibitory activity, is also a necessary co- agonist for the activation of excitatory NMDA receptors by glutamate.

Figure 2. Schematic diagram of the extracellular domain and subunit composition of α homomeric & αβ heteromeric glycine receptors. Glycine binding sites are shown at α-α and α-β subunit interfaces as orange diamonds. Binding properties of the β-β interface have not been established. The stoichiometry of heteromeric GlyRs has been debated, though both 2α:3β (Betz & Laube 2006) and 3α:2β (Durisic et al 2012) are illustrated here.

Glycine receptors are chloride conducting members of the Cys-loop family of LGICs, which also includes GABAA, 5-HT3, and nicotinic acetylcholine receptors. Glycine receptors (GlyRs) form homomeric and heteromeric pentamers from α and β subunits, with five subunits arranged

[7] about a central, membrane spanning pore (Figure 2). The stoichiometry of the heteropentamers has not been confirmed, with both 3α:2β and 2α:3β reported (Durisic et al 2012, Grudzinska et al 2005). There are four known variants of the α subunit (Lynch 2004). Subunit variants α1-3 form both α homomers and αβ heteromers. Conversely, the α4 subunit variant may form αβ heteromers; in humans α4 is encoded by a pseudogene (GLRA4) and thus is not expressed (Simon et al 2004).

As shown in Figures 2 and 3, agonist (Gly) binding sites may occur at both α-α and α-β interfaces (of the extracellular domain). In α homomers, five glycine binding sites occur at the α-α interface; while the channel may be opened by occupation of at least three of these (Beato et al 2002). These sites are also activated by amino acids agonists β-alanine and taurine (Lynch 2009). A subset of Gly sites also bind the competitive antagonist strychnine (though all homomeric sites could potentially do so). Based on the known structure of other Cys-loop LGICs, homologous models of the glycine receptor have been developed. The extracellular domain of the α1 subunit has been modelled after the crystal structure of acetylcholine binding protein, which appears in Figure 3a (Absalom et al 2003, Betz & Laube 2006). Betz & Laube also present a homology model of the α1 transmembrane domain based on the nicotinic acetylcholine receptor - shown in Figure 3b. More recently, homology models of α1 GlyRs have been produced using the prokaryotic Gloeobacter ligand-gated ion channel (GLIC), and have been used to examine potential ethanol binding residues in the transmembrane region (Howard et al 2011, Sauguet et al 2013).

As illustrated in Figure 3b, each subunit comprises an extracellular domain (ECD) containing an agonist binding site, in addition to the disulphide bridge forming the Cys-loop. Zinc may bind to the ECD to effect positive (low [Zn2+]) or negative (high [Zn2+]) modulation of the receptor (Laube et al 1995, McCracken et al 2013a). Macrocyclic lactones like ivermectin may also bind to a hydrophobic site here (Estrada-Mondragon & Lynch 2015). Coupled to the ECD are four membrane spanning α helices (M1-4) forming the transmembrane domain (TMD), with M2 helices from adjacent subunits forming the walls of the channel pore. Binding sites for volatile anaesthetics and ethanol may occur in the upper segment of the TMD. Channel blockers are thought to bind deeper into the TMD pore. The intracellular domain houses sites for clustering proteins, phosphorylation, and may play a role in Gβγ modulation of the receptor (Estrada- Mondragon & Lynch 2015, San Martin et al 2012). When an agonist binds at the ECD, it rotates relative to the transmembrane region, such that the upper TMD slopes outward and opens the channel (Bode & Lynch 2014). This membrane spanning pore allows the influx of chloride ions, contributing to membrane hyperpolarisation. Though generally selective for Cl-, the channel is also permeable to other anions including bromide, iodide, and nitrate (Lynch 2004). In

[8] neurons, the summation of these glycinergic inhibitory currents hinders cell depolarisation, thereby preventing further propagation of excitatory signals.

(a) (b)

Figure 3. Structural and binding properties of an α1 homomeric GlyR. (a) Extracellular domains of a homomeric α1 glycine receptor, with glycine shown bound at subunit interfaces. (b) View of the interface of two adjacent α1 subunits, the extracellular (ECD), transmembrane (TMD), and intracellular (ICD) domains, with binding sites as indicated. Adapted from Betz and Laube (2006)

In the brain, glycine and GABAA receptors are often co-localised to the same cells, where they both mediate inhibitory chloride currents at the postsynaptic terminal. Because of this, the alkaloid strychnine (selective for GlyRs) is used to differentiate between Gly and GABAA receptor mediated inhibition. The coexistence of GABAA and glycine receptors in the rostral brain may give rise to cross-inhibition of GABAergic signalling by GlyRs, thereby forming more complex inhibitory circuits. Neurons and neighbouring astrocytes express glycine transporters (GlyT1 and GlyT2), which remove excess Gly from glycinergic synapses to maintain signal acuity. Transporters found outside the active synapse may also regulate glycine available for tonic activation of GlyRs, with GlyTs at glutamatergic synapses regulating NMDA receptor (NMDAR) function.

GlyR mediated inhibition is an important component of neural function, as evident in their wide expression throughout the central nervous system, integration in diverse processes, and implication in disease states. The importance of α1 GlyR mediated inhibition is highlighted in

[9] hyperekplexia, where the acoustic startle reflex is highly exaggerated due to mutations in the GLRA1 gene encoding the α1 subunit. Alterations in the function or expression of α3 subunits in the spinal cord may also contribute to neuropathic pain (Harvey et al 2004, Imlach et al 2016), presenting a possible therapeutic target. Presynaptic GlyRs may also be viable targets for glycinergic dysfunction, and could be targeted with cannabinoid modulation in the treatment of hyperekplexia (Xiong et al 2014). Importantly, GlyRs are a major target of ethanol’s effects in the central nervous system (Perkins et al 2010, Soderpalm et al 2017) - their modulation may assist the development of treatment for alcohol abuse, and more selective ethanol alternatives.

1.2.1 Glycine Receptor Subunits Influence Receptor Characteristics

The α subunit variants share more than 90% sequence homology, while the β subunit shares around 47% homology with the α1 subunit (Lynch 2004). Further α subunit heterogeneity arises from alternative splicing and RNA editing, which produces splice variants of α1-3 subunits (Eichler et al 2009, Kuhse et al 1991, Malosio et al 1991a, Miller et al 2004). Because of this disparity, the subunit composition of glycine receptors influences their functional properties to varying degrees. Furthermore, their differential distribution throughout the CNS reflects their specific characteristics and function. α1 subunit mRNA is abundantly expressed in the spinal cord, and to a lesser extent in the retina, superior and inferior colliculi, and brainstem, where they likely form αβ receptors (Lynch 2009, Malosio et al 1991b). Here, they mediate fast synaptic inhibition, and are important for the relay of sensory and nociceptive information.

By comparison, α2 homomers predominate in the developing CNS, but are largely replaced by α1β within weeks after birth. This switch has been directly observed in the spinal cord (Lynch 2009), although α2 may predominate in much of the forebrain even after this, suggesting that the phenomenon may follow a regional pattern (Jonsson et al 2012, McCracken et al 2017). In the developing CNS, α2 GlyRs may play a role in cortical neurogenesis (Avila et al 2013b), the production of progenitor cells (Avila et al 2014), and differentiation of spinal interneurons (McDearmid et al 2006). Recent evidence also suggests they replace α3 containing GlyRs in spinal nociceptive neurons following nerve injury (Imlach et al 2016). In the healthy adult brain, α2 mRNA may also be found in higher brain regions like the midbrain’s ventral tegmental area (VTA), part of the dopaminergic reward pathways. In the forebrain, α2 mRNA has also been detected in the amygdala and in reward areas like the nucleus accumbens (Blednov et al 2015). Lower levels of expression occur in the cerebellum, hippocampus, and olfactory bulb (Malosio et al 1991b). α2 may also be found in cortical lamina VI (Malosio et al

[10]

1991b) - this is best developed in the motor cortex, and typically projects to the thalamus, suggesting a role of α2 in supraspinal motor control.

Like α1, α3 subunits are expressed in the adult spinal cord, where they probably form heteropentamers with the β subunit. They are generally localised to nociceptive neurons, where they may be replaced by α2 GlyRs in neuropathic pain (Imlach et al 2016). Because of their differing functional properties (reduced activation) (Mangin et al 2003, Wang et al 2006), replacement by α2 subunits reduces Gly-mediated inhibitory currents in the spinal cord, thereby impairing glycinergic modulation of pain sensation. Additionally, α3 GlyRs are modulated by inflammatory compounds such as prostaglandins, and their phosphorylation may also play a role in pain sensation (Harvey et al 2004). In rodents α3 mRNA has been detected in the cochlea (Tziridis et al 2017), and hippocampus, where it is colocalised with β subunits at glutamatergic terminals (Eichler et al 2009, Malosio et al 1991b).

The β subunit is widely distributed, with mRNA detected at high levels throughout the brain and spinal cord (Malosio et al 1991b). Incorporation of the β subunit in αβ GlyRs influences their synaptic localisation, sensitivity to compounds, and ligand binding properties (Grudzinska et al 2005). Importantly, the M2 segment of the β subunit confers picrotoxin resistance to heteromeric GlyRs (Pribilla et al 1992), allowing some distinction between α and αβ GlyRs. The neuronal assembly protein gephyrin is responsible for anchoring the GlyR to the postsynaptic membrane, and is thought to interact exclusively with the β subunit’s intracellular domain (Meier et al 2000, Meyer et al 1995). As such, αβ GlyRs are clustered on the postsynaptic membrane, where they mediate fast synaptic inhibition. In contrast, α subunits lack this interaction, and likely occur at presynaptic and extrasynaptic sites (Lynch 2009, Muller et al 2008). However, experiments by Takagi et al (1992) demonstrated an interaction between the α2 subunit and gephyrin, suggesting that α2 homomers could also cluster postsynaptically. Despite this, they may not effectively mediate synaptic glycinergic transmission, due to their slow activation in response to agonist binding (Mangin et al 2003). Furthermore, Patrizio et al (2017) observed differences in clustering of α1 and α3 containing αβ GlyRs with the same stoichiometry. This suggests that the α subunit could exert some influence upon membrane localisation, perhaps through interaction with the gephyrin binding motifs of the β subunit.

Presynaptic glycine receptors may modulate vesicular release of neurotransmitters from the presynaptic terminal. They have been shown to reduce GABA release in the VTA (Ye et al 2004), hinting at a role in the reward system. In parts of the mid- and forebrain, presynaptic GlyRs may also generate a weak depolarisation of the presynaptic membrane, triggering the release of neurotransmitters (particularly glutamate) into the synapse (Lee et al 2009, Turecek

[11]

& Trussell 2001). These may be activated by spillover glycine from nearby synapses, and reflect a non-traditional mechanism of Gly-mediated neurotransmission.

Extrasynaptic glycine receptors are thought to participate in tonic inhibition, a phenomenon well studied in subpopulations of GABAAR. McCracken et al (2017) demonstrated tonic inhibition in areas of the mouse forebrain, mediated by strychnine and picrotoxin sensitive current - indicating extrasynaptic homomeric GlyRs were responsible. These currents were observed in the prefrontal cortex, hippocampus, and striatum. In the CA1 area of the hippocampus, GlyR mediated tonic currents were also detected by Zhang et al (2008). Through tonic inhibition, extrasynaptic GlyRs may regulate neuronal excitability by increasing the cell’s activation threshold or by dampening incoming excitatory currents.

1.3 Glycine Receptors Mediate Ethanol Effects and Behaviours

Glycine receptors are a well-established target of ethanol’s actions in the central nervous system (Burgos et al 2015). Ethanol acts as a positive allosteric modulator of glycine receptors, with potential binding at sites in the transmembrane (Harris et al 2008, Sauguet et al 2013) or extracellular domains (Crawford et al 2007, Perkins et al 2008). In addition to direct association with the receptor, alcohol may indirectly modulate GlyRs through downstream signalling and protein kinase activation (Burgos et al 2015, Yevenes et al 2008). The majority of literature regarding alcohol’s actions at GlyRs focuses on its activity at homomeric α1 receptors - they are easily expressed in recombinant systems, and are well represented in the caudal brain and spinal cord (Lynch 2009). However, the expression of α2 and α3 subunits in higher brain regions targeted by ethanol has necessitated further investigation into the effects of alcohol at these receptors (Jonsson et al 2012, McCracken et al 2013a).

Ethanol has been shown to potentiate α1 GlyRs expressed in Xenopus laevis oocytes, at concentrations above 10 mM (Crawford et al 2007, Davies et al 2004). Additionally, α2 homomers respond to millimolar concentrations of ethanol when expressed in Xenopus oocytes, though are less sensitive than α1 to low (<50 mM) concentrations of ethanol (Mascia et al 1996). Comparable concentrations of ethanol also facilitate glycine-gated currents in cultured spinal neurons from rat and mouse brain (Aguayo et al 1996, Engblom & Akerman 1991). Cortical neuron excitability may also be influenced by ethanol-induced potentiation of glycine receptors, as observed in slices from rat brain (Badanich et al 2013).

Several groups have implicated supraspinal glycine receptors in ethanol reward behaviours (Blednov et al 2015, Li et al 2012, Molander et al 2005, Ye et al 2004). GlyRs may also contribute to other behavioural and sensorimotor effects of alcohol. For instance, ethanol- induced loss of righting reflex (LORR) reflects the central depression and motor impairment

[12] generated by intoxication. The righting reflex involves complex sensorimotor pathways, and its loss typically occurs with comparable doses of alcohol in humans and other mammals. Both Williams et al (1995) and Ye et al (2009) observed glycine dependent enhancement of the ethanol-induced LORR in rats, and that this was dose-dependently inhibited by strychnine. Moreover, transgenic expression of mutant EtOH-insensitive α1 subunits in mice reduces their sensitivity to EtOH-induced motor impairment and LORR (Findlay et al 2002). Hence, alcohol- induced LORR may involve GlyRs, possibly in the cochlea (Tziridis et al 2017), motor and frontal cortices, spinal cord, and reflex pathways.

1.3.1 Ethanol Modulates Glycine Receptors in Reward Pathways

Behavioural studies using α2 and α3 subunit knockout mice by Blednov et al (2015) described reduced voluntary intake of ethanol in α2 knockouts. Given their distribution in reward areas of the brain, this alludes to a role for α2-3 containing GlyRs in regulating ethanol consumption and its rewarding effects. The nucleus accumbens (NAcc) is an important component of the brain’s reward system, and contains ethanol-sensitive glycine receptors (Förstera et al 2017). Here, strychnine-sensitive GlyRs may regulate ethanol-induced dopamine release in rats, thereby contributing to alcohol’s reinforcing effects (Molander et al 2005).

The ventral tegmental area (VTA) is involved in the brain’s reward system, with dopaminergic projections to the NAcc forming part of the mesolimbic pathway. Additional projections from the VTA to the prefrontal cortex form the mesocortical dopamine pathway, which is involved in motivation and executive function. In the VTA, the firing of dopaminergic neurons is regulated by inhibitory GABAergic inputs, which may in turn be inhibited by presynaptic GlyRs. In rats, microinjection of glycine directly into the VTA has been shown to increase the firing of dopaminergic neurons in this area through its actions at presynaptic GlyRs (Ye et al 2004). Li et al (2012) also demonstrated reduced voluntary ethanol consumption in rats following direct delivery of glycine (5-10 µM) into the VTA. As such, GlyR modulation in the VTA appears important for modulating cross-inhibition of GABAAR and consequent disinhibition of dopaminergic afferents in these pathways.

Further evidence for the role of glycinergic activity in alcohol’s behavioural effects comes from Org25935, a synthetic compound which selectively inhibits glycine transporter 1 (GlyT1). Neighbouring astrocytes express GlyT1 at both glycinergic and glutamatergic (NMDAR) terminals throughout the brain. The transporter catalyses the uptake of extracellular glycine from in and around the synapse following its vesicular release, thereby maintaining low extracellular glycine and allowing precise temporospatial transmission of Gly and NMDA receptors. Through blockade of GlyT1, Org25935 effectively increases extracellular

[13] concentrations of glycine; hence more glycine is available to activate GlyRs and to co-agonise NMDARs. This would increase inhibitory signalling via GlyRs, and excitatory signalling via NMDARs; differentially so in areas where each receptor is expressed. Vengeliene et al (2010) explored the effects of Org25935 in rats chronically exposed to ethanol, finding that Org25935 “persistently” reduced relapse-like consumption (voluntary increased use following a period of imposed abstinence). Together, this supports a role for glycinergic and NMDA signalling in alcohol related reward. Org25935 has also been shown to decrease ethanol preference and consumption in male rats (Molander et al 2007), however this was not reflected in human trials, where Org25935 showed no benefit compared to placebo treatment in preventing relapse in detoxified alcohol-dependent individuals (de Bejczy et al 2014).

While there is substantial evidence for the role of ethanol in modulating GlyRs throughout the brain, the endogenous agonist of GlyR in higher brain regions is less clear – here, taurine and glycine may be differentially involved in GlyR agonism (Avila et al 2013a). While glycine is released by astrocytes in the cortex, for example (Shibasaki et al 2017), it is not clear whether its release occurs within glycinergic synapses, and therefore whether glycine is an important endogenous agonist at ethanol targets in the higher brain.

1.3.2 Zinc and Ethanol Synergistically Modulate Glycine Receptors

Zinc is a positive allosteric modulator of GlyRs at low (<10 µM) concentrations, while concentrations above 100 µM decrease GlyR activity (Laube et al 1995). In the brain, zinc is the second most abundant transition metal (other than iron), with around 90% of this stored in metal-protein complexes such as zinc-finger proteins (involved in gene regulation). In addition, many neurons also contain free ionic zinc. Histochemical methods have demonstrated a relative abundance of “chelatable” zinc in the cerebral cortex and limbic areas, particularly the hippocampal formation (Assaf & Chung 1984, Huang 1997). In these areas, much of this zinc is contained within synaptic vesicles - during strong stimulation, synaptic concentrations of zinc can reach upwards of 1 µM (McCracken et al 2013a). In the hippocampus, mossy fibres project from granule cells in the dentate gyrus to the CA3 region, and may co-release zinc with glutamate (Assaf & Chung 1984). Interestingly, presynaptic glycine receptors are also found in these cells (Kubota et al 2010), and may be potentiated by low ambient levels of zinc. Thus in stimulated synapses zinc may modulate presynaptic GlyRs, perhaps regulating further neurotransmitter release.

Importantly, zinc exerts synergistic effects with alcohol at α1-3 GlyRs, further amplifying ethanol-induced potentiation (McCracken et al 2010b, McCracken et al 2013b). Zinc may also be necessary for ethanol sensitivity of the α1 GlyR - mutation of the zinc-binding residue D80

[14] in the extracellular domain of the α1 subunit prevents zinc binding and reduces EtOH- potentiation of recombinant GlyRs expressed in Xenopus oocytes (McCracken et al 2013a). Knock-in mice with this mutation also display reduced ethanol preference and voluntary intake, further highlighting the role of zinc in modulating GlyRs in the context of ethanol’s behavioural effects (McCracken et al 2013a).

1.3.3 Mechanism of Ethanol Action at Glycine Receptors

The identification of ethanol’s binding sites is complicated by the physicochemical properties of the drug. Without co-crystallisation at the glycine receptor, molecular modelling is the best alternative for understanding ethanol’s sites of action. The best illustrated mechanism of ethanol at glycine receptors appears to involve the interaction of ethanol with an intrasubunit pocket, shared with other n-alcohols and anaesthetics (Betz & Laube 2006, Burgos et al 2016). Several studies have demonstrated the importance of residues in the transmembrane M2 (S267) and M3 (A288) regions for ethanol’s actions at α1 GlyRs (Horani et al 2015, Mihic et al 1997, Ye et al 1998). Additionally, M1 and M4 regions may also be involved (Burgos et al 2016, McCracken et al 2010a). Together, these residues may form an “action pocket”, which could contain multiple ethanol targets that are increasingly occupied with greater ethanol concentrations (Burgos et al 2016, Perkins et al 2010).

Recently, Sauguet et al (2013) developed a model of ethanol binding at LGICs using an ethanol- sensitised mutant of the prokaryotic GLIC receptor. By mutating F238A (F14’) in the M2 helix, which lines the pore, the GLIC was made to resemble ethanol’s mammalian targets including GABA, Gly (Q14’), and nACh receptors. Co-crystallisation of the ethanol-sensitised GLIC, and site-directed mutagenesis were used to map the ethanol binding pocket. When superposed onto the α1 GlyR model generated by Borghese et al (2012), residues corresponded with those previously identified as important for EtOH actions, further supporting TMD EtOH interactions.

In addition to this TMD interaction, ethanol may also interact with the extracellular domain of the α subunit (Crawford et al 2007, Perkins et al 2008). Loop 2 of the GlyR ECD may be involved in transduction of receptor activation and gating (Cederholm et al 2010), thus interactions of ethanol in this region could potentially influence these processes to enhance receptor function. Mutations of position 52 in extracellular Loop 2 alter ethanol sensitivity, while preserving receptor EC50 (Crawford et al 2008, Crawford et al 2007). Point mutation of α1’s A52 residue to a serine also generates similarly reduced ethanol sensitivity as α2 GlyRs, which has a branched threonine at this position (Mascia et al 1996). Moreover, Crawford et al (2007) have suggested that positions 52 (ECD) and 267 (TMD M2) are parts of one interconnected pocket shown in Figure 4, allowing ethanol to act at a cavity involving both domains.

[15]

Figure 4. Ethanol binding pocket at the α1 subunit of the glycine receptor. (a) Overall α1 subunit structure illustrating the extracellular, transmembrane, and intracellular domains. (b) Magnified view of the EtOH binding pocket extending from residues A52 in Loop 2 of the ECD through to S267 in the TMD. Space-filling molecules are shown for A52 and S267 residues. An interconnected pocket within which ethanol targets may reside is highlighted in red in (a) and (b). From Crawford et al (2007).

Furthermore, an ethanol- “ultra-sensitive” α1 GlyR may be created by replacing Loop 2 of the

α1 GlyR subunit with that of the GABAAR δ subunit (Naito et al 2014, Perkins et al 2009). In these α1δL2 mutant GlyRs, the α Loop 2 sequence at position 50 (SIAETTMDYR) was replaced with the homologous sequence at position 43 of the δ subunit (HISEANMEYT). As a result, the threshold for ethanol-induced potentiation was dramatically reduced from 30 mM in wild type α1 GlyRs to 1 mM in α1δL2 - concentrations that may occur in the brain following minimal alcohol consumption. Importantly, general function of the α1 receptor appeared intact with the exception of a leftward shift in glycine EC50 in the α1δL2 mutant (Perkins et al 2009).

Substitution with the γ GABAAR sequence does not generate EtOH sensitivity (Perkins et al 2009), though Naito et al (2014) have demonstrated increased EtOH sensitivity of several other Loop 2 mutants. As such, this evidence not only supports the role of extracellular loop 2 in ethanol sensitivity of the GlyR, but also reinforces the importance of the δ subunit in conferring ethanol sensitivity at GABAA receptors.

1.4 Oxytocin-ethanol Interactions at Glycine Receptors

Oxytocin is a hormone and neuropeptide comprising nine amino acids, with a role in a variety of important reproductive and neural processes. Following its production in the paraventricular and supraoptic nuclei of the hypothalamus, oxytocin is secreted into the bloodstream by the posterior pituitary gland. From here it is distributed to target areas such as

[16] the uterine and mammary tissues. Of particular importance are some of the hormone’s reproductive functions - oxytocin stimulates uterine contractions during parturition, and initiates lactation through the milk ejection reflex. Although male and female oxytocin production appears comparable and oxytocin receptors are present in the human male reproductive system, the physiological functions of the hormone in males are poorly understood. As a centrally-acting neuropeptide, oxytocin is vital for mammalian pair bonding, and regulation of social behaviour (Lee et al 2016, Neumann & Landgraf 2012), and normal maternal behaviour (Ross & Young 2009). Medically, exogenous oxytocin has traditionally been used to induce labour in pregnant women and to prevent postpartum haemorrhage (Elbourne et al 2001). More recently, oxytocin has come to light as a focus of mental health research, particularly with regards to anxiety and addiction (Bowen & Neumann 2017b, MacFadyen et al 2016, Pedersen et al 2013, Sarnyai & Kovács 2014).

The principal endogenous target of oxytocin is the oxytocin receptor (OXTR); a G-protein coupled receptor, coupled to Gq. The Gq protein activates phospholipase C, effecting increased cytosolic calcium and triggering an intracellular cascade. Oxytocin receptors are expressed in mammary myoepithelium, and in the uterine myometrium and endometrium in late pregnancy. Oxytocin receptors are also present in the central nervous system, where their distribution and abundance is influenced by gender, together with environmental and peripheral factors (Evans et al 2014). Importantly, the distribution of oxytocin receptors in the brain also varies significantly among mammals (Boccia et al 2013, Young et al 1996). In rats, oxytocin binding sites are well represented in the olfactory and limbic systems, and the basal ganglia; lower levels occur in the thalamus, hypothalamus, and some cortical areas (Gimpl & Fahrenholz 2001). In humans, immunostaining has also demonstrated the presence of oxytocin receptors primarily in the hypothalamus and limbic system, together with the thalamus, brainstem, and several cortical regions (Boccia et al 2013). Both similarities and differences have been noted between rat and human patterns of OXTR expression (Boccia et al 2013). As such, species-specific mechanisms of oxytocin must be considered when examining potentially OXTR-mediated effects.

Oxytocin is released into the periphery by the posterior pituitary, while cerebrospinal fluid (CSF) is thought to facilitate its distribution throughout the brain. Because oxytocin does not readily cross the blood-brain barrier (though may do so in small amounts), peripheral oxytocin levels are generally independent of those in the CSF (Neumann & Landgraf 2012, Neumann et al 2013). Several studies demonstrate a rise in CSF OT concentrations following intranasal administration of oxytocin (Mitchell et al 2015, Neumann et al 2013, Pedersen et al 2013, Striepens et al 2013), however the mechanism through which this occurs remains unclear.

[17]

Once in the CSF and brain’s extracellular fluid, oxytocin is available to act at a range of potential central targets.

oxytocin vasopressin

P8 P3

Figure 5. The Molecular Structure of Nonapeptides Oxytocin and Arginine Vasopressin. Skeletal structures are shown for oxytocin and vasopressin. Vasopressin differs from oxytocin by two amino acid as shown; isoleucine is replaced by phenylalanine at position 3, and leucine by arginine at position 8. On vasopressin, these differences are loosely indicated by the perforated circles.

Drugs of abuse interact with the endogenous oxytocin system; and reciprocally, oxytocin has been observed to modulate the rewarding properties of these drugs (Lee et al 2016, McGregor & Bowen 2012). Typically, addictive substances activate the mesolimbic dopamine reward pathway, which connects the VTA to the NAcc in the striatum; and the amygdala and hippocampus of the limbic system. In particular, dopaminergic innervation of the nucleus accumbens is thought to be crucial for drug based reward. Additionally, dopaminergic activation of the hippocampus and amygdala modulate memory formation and emotion respectively. Oxytocin interacts with these mesolimbic dopaminergic neurons to mediate the rewarding and reinforcing effects of addictive substances. In humans, the specific mechanisms through which this occur are unknown, particularly given that binding sites for oxytocin may not be detected in the human VTA or NAcc (Boccia et al 2013, Gimpl & Fahrenholz 2001).

1.4.1 Oxytocin and Alcohol

Emerging research suggests the oxytocin system may be a useful target in the treatment of substance addictions, including alcohol use disorder (Bowen & Neumann 2017a, Lee et al 2016). While oxytocin receptors themselves are unlikely to be functional targets of ethanol, they may be influenced by ethanol’s activity in the brain, such as through convergent pathways with GABAA and glycine receptors. Interestingly, oxytocin and alcohol may generate the same behavioural effects in humans - both decrease amygdalic fear responses, are dose- dependently anxiolytic, while also (perhaps surprisingly) increasing aggressive behaviour in a small subset of predisposed individuals, and promoting in-group behaviours (Mitchell et al 2015). Furthermore, intranasally administered oxytocin was shown to reduce the symptoms of

[18] alcohol withdrawal and craving in alcohol-dependent humans (Pedersen et al 2013).

Behavioural studies have consistently demonstrated the ability of exogenous oxytocin to modulate alcohol related reward. Intraperitoneally administered oxytocin has also been shown to reduce voluntary ethanol consumption in rats (MacFadyen et al 2016). Peters et al (2017) also demonstrated that centrally-administered exogenous oxytocin effected a reduction in voluntary ethanol consumption in rats who had chronically consumed alcohol. Additionally, Peters et al observed that ethanol-induced dopamine release in the nucleus accumbens was attenuated by oxytocin in both ethanol-naïve and repeatedly exposed rats. This reduction in dopamine release could explain the reduced drive towards ethanol consumption, due to oxytocin’s interference in the mesolimbic reward pathway. While present in rats, OT binding targets may not occur in the human NAcc (Boccia et al 2013, Gimpl & Fahrenholz 2001), thus if OT interferes with dopamine release in the human NAcc, it may do so through interaction with targets innervating this area rather than direct action in the NAcc.

1.4.2 Oxytocin-ethanol Interactions at Ligand-gated Ion Channels

Indirect interactions of oxytocin with glycine and GABAA receptors has been observed (Hussy et al 2001, Koksma et al 2003). For example, OT may modulate the expression of GABAAR subunits to contribute to neuroprotection against ischaemia and oxidative stress (Kaneko et al

2016). This may involve downstream actions through the OXTR, though a direct GABAAR gated or phosphorylation mechanism has not been investigated. Moreover, direct interactions of large neuropeptides with ligand-gated ion channels are not well explored.

Oxytocin Interacts with EtOH at GABAA Receptors Recent behavioural experiments have demonstrated that oxytocin attenuates ethanol-induced motor impairment in rats, possibly via δ GABAARs without involvement of the oxytocin receptor, while electrophysiological experiments demonstrate that oxytocin prevents ethanol- induced enhancement of GABA-gated currents in δ receptors (Bowen et al 2015). Behavioural experiments by Bowen et al looked at the effects of oxytocin on ethanol-induced motor impairment in rats, finding that intraperitoneal delivery of 1.5 g/kg EtOH generated significant motor impairment and sedation at 5 and 35 minutes following application. Intracerebroventricular co-administration of oxytocin however, abolished ethanol-induced sedation and ataxia across these timepoints.

In electrophysiological experiments using Xenopus oocytes, moderate alcohol concentrations (30 mM) significantly enhanced GABA-activated current at α4β1δ and α4β3δ receptors – this effect was also completely abolished by 10 μM oxytocin. The delta subunit appeared necessary

[19] to facilitate sensitivity to alcohol, and for oxytocin’s influence on ethanol-induced effects. The lack of effect by oxytocin on GABA-activated current when ethanol was replaced by THIP (a δ selective agonist) suggests that these effects are specific to ethanol actions, and alcohol- induced conformational changes may be required for oxytocin to act at these receptors. Moreover, these effects were not reproduced by vasopressin, a functionally similar neuropeptide differing from OT by only two amino acids (Figure 5), suggesting that the interaction necessitated the specific pattern of structural moieties present in the OT molecule.

Because of the similarities between Ro15-4513 and oxytocin, in terms of behavioural and functional ethanol antagonism, alcohol could bind at the proposed BZ site described by Wallner et al (Figure 1). However, due to structural differences, differing hydrophilicity, and because oxytocin doesn’t act as an inverse agonist, oxytocin could bind at a distinct site, or some overlapping site (Bowen et al 2015). The selective and specific interactions of oxytocin and ethanol provide a useful tool for further investigation of how the two compounds bind individually and together at this receptor, and may assist in clarifying alcohol’s mode of action at δ-containing receptors.

Oxytocin Interactions at Glycine Receptors Several indirect interactions between oxytocin and glycine receptors are cited in the literature. For example, presynaptic glycine receptors may indirectly modulate the release of vasopressin - and to a lesser extent oxytocin, from the posterior pituitary (Hussy et al 2001). Additionally, immunohistochemical staining has demonstrated the presence of glycine receptors on both vasopressin and oxytocin terminals (Hussy et al 2001). As such, their activity could have implications for neuropeptide hormone signalling within the CNS. Both oxytocin and glycine may modulate ethanol-induced reward through a similar or convergent mesolimbic pathway. Molander et al (2005) demonstrated that perfusion of Gly into the rat NAcc increases dopamine output, while Gly prevented EtOH-induced accumbal dopamine release and correspondingly reduced voluntary intake and preference in a subpopulation of these rats. Similarly, oxytocin impairs ethanol-induced accumbal dopamine release in rats, and reduces their voluntary alcohol use (Peters et al 2017).

At present, investigation of any direct interaction of oxytocin with glycine receptors has not been described. However, a number of (smaller) heptapeptide modulators have been characterised at αβ GlyRs (Cornelison et al 2016). It was found that several heptapeptides were able to potentiate receptor function, and a number of motifs necessary for this were identified. Although none of these heptapeptide motifs appear to overlap with oxytocin’s structure, this demonstrates that sizeable peptide molecules may be able to modulate the glycine receptor.

[20]

Both GABAA and Gly receptors are chloride-conducting members of the Cys-loop family. Both are major targets of ethanol’s actions in the CNS, and both contribute to ethanol-induced reward and motor impairment, together with its effects upon memory, physiological arousal, and mood. Due to their colocalisation on neurons in regions of the brain and involvement in convergent neural pathways, glycinergic and GABAergic signals may differentially modulate a number of ethanol’s central effects. As such, because of their structural and functional similarities as inhibitory LGIC targets of ethanol, it was hypothesised that oxytocin could interact with glycine receptors in a manner similar to that observed by Bowen et al at GABAA receptors. This hypothesis forms the basis of the work presented in this thesis.

1.5 Research Question & Aims

This project will explore the action of ethanol at recombinant α-homomeric glycine receptors expressed in Xenopus oocytes, and to examine any potential interactions of oxytocin with ethanol at these receptors. The results of this project aim to work towards a better understanding of the following:

(1) Does ethanol consistently modulate the activity of glycine receptors at concentrations relevant to human alcohol consumption?

(2) How are alcohol’s actions modulated by the presence of oxytocin?

(3) What might be the mechanism through which an oxytocin-ethanol interaction occurs?

1.5.1 Significance

Building on existing knowledge within the literature, this project will provide information allowing better understanding of alcohol’s interaction with α1 and α2 homomeric glycine receptors. Moreover, direct effects of oxytocin at the glycine receptor can be examined electrophysiologically for the first time. If oxytocin’s anti-alcohol effects first observed at δ-

GABAA receptors (Bowen et al 2015) can be replicated at α GlyRs, this would hint at an additional mechanism through which oxytocin modulates the effects of ethanol. Furthermore, such effects could potentially extend to how oxytocin may modulate ethanol’s effects on native glycine receptors in the brain. Further down the track, this could contribute to development of more targeted, subunit-selective, and less toxic alternatives and treatments relative to alcohol misuse and abuse. In a broader context, increased knowledge of how alcohol works in the brain may assist in understanding alcohol-related violence, misuse and abuse – knowledge of what’s happening at a molecular and cellular level in the brain allows for a more integrative approach to addressing these issues.

[21]

Chapter 2: Experimental Procedures

2.1 Molecular Biology

2.1.1 Materials

Restriction endonucleases Hpa1 and Nhe1, and CutSmart buffer were purchased from New England Biolabs, USA. The mMessage mMachine T7 kit was purchased from Ambion, USA. Human a1 and a2A glycine receptor mRNA were generated in-house by Dr Steven Devenish.

Glycine receptor subunits α1 and α2 were encoded by GLRA1 and GLRA2 inserts subcloned into the pGEMHE vector. GABAA receptor subunits α4, β3 and δ were encoded by GABRA4, GABRB3 and GABRD inserts, and were subcloned into the pCDNA1 (α4 and δ) or pGEMHE (β3) plasmids (Table 1). Oligonucleotide primers were purchased from Invitrogen, USA.

Table 1. Summary of Plasmid Properties

Insert Accession Number Polymerase Restriction Vectors (NCBI) Promoter Enzymes

GLRA1 NG_011764 T7 NheI pGEMHE

GLRA2 NG_016459 T7 NheI pGEMHE GABRA4 NM_000809 T7 Hpa1 pcDNA1 GABRB3 NC_000015 T7 NheI pGEMHE GABRD NC_000001 T7 NheI pCDNA1

2.1.2 Transformation & Growth of DNA Plasmids

All plasmids were transformed into XL1-Blue competent E. coli (Stratagene, USA). Competent cells were briefly thawed on ice, then 50 µl competent cells were added to 2 µl of 100-500 ng/µl DNA. Cells were left on ice for 10 minutes before heat-shocking the bacteria for 2 minutes at 42° C, then were immediately returned to ice for a further 10 minutes.

To first allow for the expression of ampicillin-resistant genes, 500 µl of SOC media was added and cells were incubated for 1 h at 37ᴼ C. The cells were then plated on Luria Broth (LB)-agar plates containing 100 µg/ml ampicillin and incubated overnight at 37ᴼC.

Using a sterile pipette tip, selected colonies were isolated and transferred to centrifuge tubes containing LB media. To again encourage selection of bacteria expressing ampicillin-resistant genes, they were first incubated for 1 h at 37° C in a shaking incubator, before ampicillin was added to a final concentration of 100 µg/ml, and the tubes were returned to the shaking incubator overnight.

[22]

Plasmid DNA was purified with a Miniprep kit (QIAGEN) in accordance with the kit instructions. Bacteria was pelleted in a centrifuge and the supernatant was removed, with Buffer P1 used to resuspend the pellet. Cells were lysed with Buffer P2, which was terminated by the addition of Buffer N3. The mixture was centrifuged in a QIAprep column and tube washed with ethanol several times, and the collected purified DNA was eluted in 20 µl elution (EB) buffer.

2.1.3 Linearisation and Transcription of DNA

Vectors containing sub-cloned human α1 and α2A GlyRs (pGEMHE); α4, β3 and δ GABAAR cDNA were used to amplify template DNA. Plasmids containing α4 and δ subunit cDNA were linearised using Hpa1, while α1, α2 and β3 were linearised with Nhe1 restriction endonuclease (New England Biolabs). The linearisation reaction was assembled (20 µl DNA, 5 µl CutSmart buffer, 1-2 µl enzyme, H2O added to 50 µl total), and incubated at 37° C for the appropriate period. A nucleotide removal kit (QIAGEN) was then used to remove unincorporated nucleotides. PNI Buffer was added to the sample, which was spun in a column and collection tube. Flow through was discarded, with Buffer PE added and the former step repeated. DNA was eluted with EB Buffer.

To generate mRNA from the linearised cDNA, a capped transcription reaction was performed using the T7 mMessage mMachine kit (Ambion, USA), and RNA was purified using DNAse and lithium chloride precipitation. mRNA quality was determined with 0.5% agarose denaturing gel electrophoresis, and mRNA concentration was determined using a NanoDrop® ND-1000 UV-vis spectrophotomer. RNA was stored at -80 °C prior to injection into Xenopus oocytes.

2.1.4 Site-Directed Mutagenesis

Several amino acid substitutions into the Loop 2 sequence of the α1 GlyR subunit were made in order to increase sensitivity of the receptor to ethanol (Perkins et al 2009). Mutant α1 glycine receptor subunits were made with the Quikchange II site directed mutagenesis kit. Due to the number of amino acid substitutions required, mutagenesis was carried out in two separate reactions; the first creating four substitutions (S50H:A52S:T54A:T55N) and the second creating the additional D57E and R59T substitutions.

Table 2. Sense Oligonucleotide Primer Sequences

α1 Loop 2 Mutation Sense Oligonucleotide Primer (5’-3’)

GlyRd1F cagctttggtcccatttctgaggcaaccatggact

GlyRd2F tgctgagacaaacatggaatatacggtcaacatct

[23]

The QuikChange II Site-Directed Mutagenesis Kit (Stratagene CA, USA) was used to incorporate amino acid substitutions with sense and antisense oligonucleotide primers containing the substitutions (Table 2). A PCR cycle was performed according to the manufacturer’s instructions with the cycle conditions shown in Table 3.

Table 3. QuikChange II PCR Cycle Conditions

Segment Number of Cycles Temperature (° C) Time

1 1 95 30 s 16 95 30 s 2 55 1 min 68 6 min

Following the mutagenesis reaction, the resultant DNA was treated with DpnI to remove the original DNA vector and transformed into XL1-Blue competent E. coli (Stratagene, USA). Positive colonies were grown and DNA isolated as explained previously. Mutant DNA samples were sequenced by the Australian Genome Research Facility (Sydney, Australia).

2.2 Electrophysiology

2.2.1 Materials

Gentamycin, glycine, HEPES (hemisodium salt), sodium pyruvate, tetracycline and theophylline were purchased from Sigma-Aldrich, Australia. Oxytocin and arginine vasopressin were supplied by Dr Michael Bowen.

2.2.2 Xenopus laevis Oocyte Preparation

All procedures involving use of Xenopus laevis frogs and oocytes were approved by the Animal Ethics Committee of the University of Sydney (Reference number: 2013/5915).

Frogs were housed with compatible triplets in a 12 h light/dark cycle, in tanks with a 15 cm deep reticulated aquatic system. The aquatic solution was maintained at 18 to 22°C, with persistent filtering by 50 µM carbon filter, and regular monitoring of pH, nitrate, and hardness levels. Frogs were fed Wardley’s cichlid floating pellets (Hartz Mountain Corporation NJ, USA) twice weekly, and supplemented with Wardley’s reptile sticks with fortified calcium weekly. Tanks were cleaned twice-weekly the day following feeding to control (reduce) nitrate levels.

Mature female Xenopus laevis frogs were anaesthetised with 0.17% tricaine (buffered with 0.06% sodium bicarbonate) for 15 minutes. Following confirmation of the loss of righting

[24] reflex, frogs were transferred to ice for duration of the surgery. A small 1-2 cm abdominal incision was made through the skin and muscle layers with sterile blades. Ovarian lobes were removed with forceps and kept in oocyte releasing 2 buffer (OR2 – 82.5 mM NaCl, 2 mM KCl, 1 mM MgCl2, 5 mM HEPES hemisodium; pH 7.4). Skin and muscle layers were sutured separately, and frogs were left to recover for 6 months before being reselected for surgery. After five recoverable surgeries, each frog underwent a terminal surgery using a lethal dose of tricaine (0.5%). Ovarian lobes were separated into small sections and the oocytes were defolliculated with 2 mg/mL Collagenase A (Boehringer Mannheim, IN, USA) in OR2 buffer for 1 hour. Released oocytes were rinsed with ND96 storage solution (96 mM NaCl, 2 mM KCl, 1 mM

MgCl2, 1.8 mM CaCl2, 5 mM HEPES hemisodium, 2.5 mM pyruvate, 0.5 mM theophylline; pH 7.4). Apparently healthy stage V-VI oocytes were isolated and kept in ND96 storage solution at 18° C prior to microinjection.

Following collagenase digestion to remove extracellular connective tissue, isolated stage V-VI oocytes were sorted to obtain a healthy, uniformly sized selection of cells. Oocytes were microinjected with 0.5-5 ng of mRNA. Oocytes were then maintained at 18 °C in ND96 buffer (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, 5 mM HEPES, pH 7.4), supplemented with 0.5 mM theophylline, 5 mmol/L sodium pyruvate, 50 μg gentamycin and tetracycline. Theophylline supplementation was used to prevent oocyte maturation (O'Connor & Smith 1976), while antibiotics gentamycin and tetracycline were used to prevent microbial contamination. Oocytes were incubated under these conditions for 4-6 days prior to electrophysiological experiments to attain optimal receptor expression.

2.2.3 Expression of Recombinant Glycine Receptors in Xenopus Oocytes

Following collagenase digestion to remove extracellular connective tissue, isolated stage V-VI oocytes were sorted to obtain a healthy, uniformly sized selection of cells. Oocytes were microinjected with 0.5-5 ng of mRNA, and maintained at 18 °C in ND96 buffer (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, 5 mM HEPES, pH 7.4), supplemented with 0.5 mM theophylline, 5 mmol/L sodium pyruvate, 50 μg gentamycin and tetracycline.

Figure 6. Microinjection of mRNA into Xenopus laevis oocyte.

[25]

Theophylline supplementation was used to prevent oocyte maturation (O'Connor et al., 1976), while antibiotics gentamycin and tetracycline were used to prevent microbial contamination. Oocytes were incubated under these conditions for 2-4 days prior to electrophysiological experiments to attain optimal receptor expression.

2.2.4 Electrophysiological Recording from Recombinant Receptors

To investigate receptor activity upon exposure to the compounds of interest, whole cell currents were measured using a two-electrode voltage clamp, with a Geneclamp 500B or Warner OC-725 amplifier filtered through a Warner LPF-8 filter (providing a filtering frequency of 14 Hz). Analogue data was digitised through a Powerlab/200 analogue to digital converter, in combination with LabChart version 3.5 (AD Instruments, Australia), with an acquisition rate of 1 KHz/s. Individual oocytes were transferred to the recording apparatus chamber and immersed in ND96 buffer solution. Glass recording microelectrodes were made using a micropipette puller (Narishige, Japan), and filled with 3 M KCl, with a resistance between 0.2 and 1 MΩ. Oocyte membranes were pierced with electrodes, and membrane potential was clamped at -60 mV, with continual superfusion of the chamber with ND96 buffer (Figure 7).

Figure 7. Two-electrode voltage clamp schematic diagram

After the establishment of a stable baseline current, compounds were applied via bath perfusion. To prevent receptor desensitisation and contamination of compound responses, oocytes were washed (4-8 min) between each successive application.

[26]

2.2.5 Data and Statistical Analysis

Data analysis and statistical testing were conducted in GraphPad Prism. The amplitude of each glycine response current (I) was expressed relative to the maximal current (Imax), and normalised using the following equation:

푁표푟푚푎푙푖푠푒푑 푅푒푠푝표푛푠푒 = 퐼푚푎푥

Normalised concentration-response curves were generated according to the Hill equation:

[A]푛퐻 퐼 = 푚푎푥 푛퐻 푛퐻 ([A] + EC50 )

Where [A] denotes ligand concentration, nH is the Hill coefficient, and EC50 is the drug concentration generating a half-maximal response.

Glycine-gated currents in the presence of ethanol and oxytocin were normalised and analysed similarly, when applied individually and in combination with Gly. Statistical analyses were conducted in GraphPad Prism (GraphPad Software, La Jolla, California, USA) using a combination of one-sample and paired sample t tests, and ANOVA as appropriate. Post-hoc analyses used were Dunnett’s test (for comparison with control) and Sidak’s test (for comparing means in data with one another).

[27]

Chapter 3: Results

3.1 Expression of Homomeric Glycine Receptors in Xenopus Oocytes

Concentration-response relationships were established for homomeric α1 and α2 glycine receptor subtypes (Figure 8). Glycine was applied to Xenopus oocytes via bath perfusion, at concentrations from 1 μM to 2 mM (α1), and 20 μM to 3 mM (α2). Between 5 and 8 minutes wash time were allowed between successive applications to minimise the risk of receptor desensitisation, and to allow removal of residual compounds from the bath solution housing the oocytes. For results hereafter, statistical analysis was performed using either an unpaired t-test or one-way ANOVA (post-hoc tests listed in 2.2.5). Throughout this chapter, results are reported as mean ±SEM to 3 significant figures or 2 decimal places unless stated otherwise.

Figure 8. Concentration-response curves for glycine at α1 and α2 WT GlyRs expressed in Xenopus laevis oocytes. Using two-electrode voltage clamp electrophysiology, glycine doses were applied at concentrations from 1 μM (α1) and 20 μM (α2) up to 3 mM. Log-transformed glycine concentrations (M) are represented on the x-axis. Mean I/Imax values were calculated as response relative to maximum Gly dose (1 mM for a1, 2 mM for a2), where Imax is equal to 100, and are represented on the y-axis. Values were derived from n≥5 oocytes for each Gly dose. Where visible, error bars represent SEM (symbols may eclipse smaller SEM error bars). Nonlinear regression was used to generate curves of best-fit using GraphPad Prism.

Oocytes expressing homomeric α1 GlyRs displayed negligible responses to 1-3 μM glycine, though current was generated at Gly concentrations above this (Figure 8). In line with existing literature (Naito et al 2014), α1 receptors were more sensitive to glycine than α2, with EC50 values of 66.42 μM and 263.4 μM respectively. Because nonlinear regression analyses were conducted on log-transformed EC50 values, logEC50 values were used in to test whether the

[28] potency of glycine at these subtypes differed significantly, with the use of a t-test. Glycine was significantly more potent at α1 than α2 receptors (p <0.0001, t=19.24, df=33.53; two-tailed t test with Welch’s correction for unequal variance). Compared to selected literature values

(Naito et al 2015, Perkins et al 2009), the present experimental EC50 was lower for α1 and higher for α2 homomers respectively. This may reflect inherent variability present with the use of Xenopus laevis oocytes as a recombinant expression system, or may arise from further potentiation by trace amounts of zinc present in the buffer in the case of α1.

Table 4. Summary of nonlinear regression results for WT α1 and α2, and mutant α1δL2 GlyRs.

Receptor Imax Hill Slope Log EC50 Experimental EC50 Literature EC50 nA ± SEM μM ± SEM μM μM ± SEM

α1 GlyR 2488 2.45 ± 0.2627 -4.18 ± 0.0255 66.42 92.3 6 ± 8.61

α1δL2 GlyR 3044 1.29 ± 0.1199 -4.53 ± 0.0345 29.57 14.615 ± 42; 22.9 ± 4.11

α2 GlyR 4170 2.17 ± 0.0019 -3.58 ± 0.0230 263.4 115.0 6 ± 21.01

Experimental Imax, Hill slope, log EC50 and EC50 were determined from n≥5 oocytes. Experimental values were calculated using nonlinear regression in GraphPad Prism – significance was tested using unpaired t tests, with

Welch’s correction used where unequal standard deviation was known. Imax values are mean values calculated directly from experimental data. Literature values were taken from (1) Naito et al (2014) and (2) Perkins et al (2009).

The observed Imax value refers to the maximum current generated by the application of maximal concentrations of agonist (glycine). Maximum current reflects both the level of receptor expression on the oocyte cell surface, and the efficacy of glycine at these receptors.

Some variation in Imax was seen across receptor subtypes (Table 4), with the rank order α1<α1δL2<α2. Glycine efficacy was significantly greater at α2 than α1 (p=0.0014, t=3.801, df=17.05; two-tailed t test with Welch’s correction). As the same amount of α1 and α2 RNA was injected, these variations may reflect either subunit-specific changes in the efficacy of glycine, or differential levels of expression on the oocyte cell membrane. Additionally, the Imax of individual cells also varied between experiments using the same GlyR construct, possibly due to altered expression levels (which may change over time post-injection).

Mutant α1δL2 GlyRs - in which a Loop 2 sequence from the δ GABAA receptor subunit replaces that of α1 GlyR - were also expressed in Xenopus oocytes. Due to their increased ethanol sensitivity, this construct was used to investigate the interaction of oxytocin with lower

[29] concentrations of ethanol (Section 3.6). As above, the mutant’s Imax was not significantly different from that of either wild type receptor, even though twice the volume of α1δL2 RNA was injected (compared with WT RNA). The concentration-response relationship of glycine at these receptors was also established and compared with that of the wild type α1 receptor (Figure 9).

Figure 9. Concentration-response curve comparison of α1dL2 mutant and α1 GlyRs. Using two-electrode voltage clamp electrophysiology, glycine doses were applied at concentrations from 100 nM to 10 mM (α1δL2). The CR curve for the α1 GlyR is superimposed here to allow comparison of wild type and mutant receptors. Log-transformed glycine concentrations (M) are represented on the x-axis. I/Imax values were calculated as response relative to maximum Gly dose (1 mM), and are represented on the y-axis. Where visible, error bars represent SEM. Nonlinear regression was used to generate curves of best-fit using GraphPad Prism.

Maximum current (Imax) at α1 was not significantly increased by introduction of the δL2 mutation (Table 4). The glycine EC50 was more than halved from 66.42 μM (α1) to 29.57 μM (α1δL2), with the α1δL2 concentration-response curve shifted to the left (Figure 9). Comparing the logEC50 values revealed a significant increase in Gly potency at α1δL2 (p<0.0001, t=46.47, df=42.61; two-tailed t test with Welch’s correction). Since the EC50 is determined by both the binding affinity of an agonist and the intrinsic gating properties of a receptor, this supports a role for Loop 2 in one or more of these functions. This shift in potency may be reconciled with the involvement of loop 2 in transducing agonist activation to effect channel gating (Absalom et al 2003, Cederholm et al 2010). Increased potency of the α1δL2 mutant relative to wild type α1 is also reported in the literature (Perkins et al 2009).

[30]

3.2 Effects of Ethanol on Wild Type α1 and α2 Glycine Receptors

Much of the literature on the actions of alcohol at GlyRs focuses on its activity at homomeric α1 receptors, due to their wide distribution in the CNS (Lynch 2009), and perhaps ease of expression in recombinant systems. However, α2 subunit-containing GlyRs are also expressed in regions of the adult brain relevant to the intoxicating effects of alcohol - including the frontal cortices and nucleus accumbens (Jonsson et al 2012, McCracken et al 2013b). To date, there is comparatively little research into the effects of ethanol at these receptors. Investigating the effects of ethanol at both of these subunits may therefore provide a more detailed understanding of its actions in the brain.

While physiologically and behaviourally relevant concentrations range up to 30 mM (for most humans) (Olsen et al 2007), concentrations of up to 100 mM may be considered pharmacologically relevant. When testing ethanol concentrations much greater than 100 mM at recombinant ion channels, it is possible that the osmolarity of the bath perfusion may be disrupted. Because the operation of the GlyR chloride channel is dependent upon the ion concentration gradient, significant disruption of the osmolarity may affect reliable data collection. Because of these factors, the present experiments were limited to ethanol concentrations equal to or below 100 mM.

When using a Gly concentration of approximately EC50, the likelihood of receptor desensitisation increases, while the magnitude of ethanol-induced potentiation decreases (Bowen et al 2015, Mascia et al 1996, Naito et al 2014). Therefore to allow observation of potentiation, submaximal concentrations of glycine (approximately EC5) were co-applied with ethanol. The use of lower concentrations of glycine may also be relevant to α homomers (hypothetically) expressed in vivo – due to the absence of the gephyrin-binding β subunit, these would likely occur extrasynaptically, with exposure to lower glycine concentrations.

To test the modulation of homomeric α1 and α2 GlyRs, the glycine EC5 concentration was established for each subtype. After at least 2 stable consecutive applications of EC5 Gly on oocytes, ethanol in combination with EC5 Gly was perfused into the bath solution. Gly-gated currents elicited by this combination were calculated as a percentage of the EC5-induced current (Figure 10). At α1 GlyRs, ethanol was co-applied with 20 μM Gly (~EC5) in the presence of 100 nM ZnCl2. For both α1 and α2 GlyRs, separate one-way ANOVA with Sidak’s test for multiple comparisons were used to test for significance between EtOH concentrations.

In α1-containing oocytes, Gly-gated current was not observably modulated by 10 mM EtOH, and although 30 mM EtOH potentiated EC5 Gly by 127.7% ±6.24, this was not statistically significant. However, 100 mM EtOH did significantly potentiate glycine by 164.6% ±8.14

[31]

(p<0.0001) (Figure 10i). When tested at α2-containing oocytes, 30 mM EtOH did not significantly potentiate 80 μM glycine (117.5% ±3.20), however potentiation induced by higher ethanol concentrations was significant. Gly was potentiated 143.6% ±2.33 by 60 mM EtOH (p<0.001), and 237.6% ±29.9 by 100 mM EtOH (p<0.001) (Figure 10ii).

Figure 10. Ethanol induced potentiation of wild type α1 and α2 receptors. Using two-electrode voltage clamp electrophysiology, ethanol doses were co-applied with ~EC5 glycine – for α1 GlyR, 100 nM ZnCl2 was present in solution. Representative traces are shown for each receptor below graph.

Ethanol modulation is represented as a percent of EC5 glycine-induced current, as displayed on the y- axis. The perforated line at y=100 reflects glycine-induced current (100%), with error bars displaying the SEM. **p<0.01, ***p<0.001, ****p<0.0001.

As evident in Figure 10, potentiation increased in a concentration-dependent manner for both α1 and α2 GlyRs. Positive modulation elicited by 100 mM EtOH was significantly greater than with 30 mM (α1) or 60 mM (α2) EtOH (p<0.01). Ethanol was also applied in the absence of glycine, though did not generate any current in α1, α2, or α1δL2 oocytes. Though slight potentiation was seen at both α1 and α2 GlyRs with 30 mM EtOH, significant EtOH modulation only occurred at higher concentrations. The magnitude of potentiation by 100 mM EtOH was greater at α2 than α1 GlyRs (143% and 156%), although the variability of α2 was quite large (SEM = 29.9). Conversely, 30 mM EtOH generated greater potentiation at α1 (127.7%) than α2 GlyRs (117.5%). Similar differences were reported by Mascia et al (1996), however the present results may be limited by even slight differences in the concentration of Gly co-applied.

[32]

3.2.1 Zinc-induced Potentiation of α1 Glycine Receptors

Zinc occurs naturally throughout the brain in chelatable form (Huang 1997), including within the synapse (Assaf & Chung 1984). At low concentrations, zinc acts as a positive allosteric modulator of α1 GlyRs, and may be important for ethanol-induced potentiation at these receptors (Laube et al 1995, McCracken et al 2010b). For this reason, zinc was used to enhance the potentiation of glycine current by ethanol at these receptors. To allow comparison across experiments involving α1 GlyRs, 100 nM ZnCl2 was present in solution when examining the effects of Gly, EtOH, and OT, either individually or in combination with one another.

*

Figure 11. Zinc modulates activity at α1 GlyRs. (i) Relative response of α1 GlyRs to 20 μM Gly was magnified by the presence of 100 nM ZnCl2. Responses are displayed as mean ±SEM, and are expressed as percent of Imax response. (ii) Zinc modulation of ethanol-induced potentiation of glycine current at α1

GlyRs. 10, 30, and 100 mM ethanol was applied with 20 μM glycine, with and without 100 nM ZnCl2. Responses are expressed as a percent of the response to control (20 μM Gly with and without zinc), and are shown as mean ±SEM. *p<0.05.

The EC5 concentration of glycine at α1 GlyRs was established as 20 μM from n<5 cells.

Responses to 20 μM Gly in the presence or absence of 100 nM ZnCl2 were calculated as relative to the current generated by Imax of a cell (equal to 100%). Significance testing was performed with two-tailed t-tests with Welch’s correction for unequal variances. The relative response to 20 μM Gly (5.55% ±1.75) was significantly greater in the presence of 100 nM ZnCl2 (17.68% ±4.13; p<0.05, t=2.70, df=5.39) (Figure 11).

To test zinc’s modulatory effects on ethanol, EtOH-induced Gly potentiation was compared in the absence and presence of 100 nM ZnCl2. Modulation was expressed as a percent of 20 μM Gly-induced current (with or without zinc) for each respective condition (Figure 11ii). 100 nM

ZnCl2 significantly enhanced potentiation by 30 mM ethanol (p<0.05, t=2.77, df=6.35). The presence of zinc did not significantly modulate other concentrations of ethanol.

[33]

In the literature, ethanol potentiation is enhanced by zinc at α1 GlyRs (McCracken et al 2010b). There is also evidence that zinc enhancement of ethanol’s actions occurs at α2 and α3 GlyRs (McCracken et al 2013b). In the present data, there was no statistically significant enhancement, though zinc enhanced the mean response to 30 and 100 mM EtOH by 17.6% and 21.5% respectively. Due to natural contamination, zinc is likely present at low levels in the

ND96 solution and may also be present on equipment. The addition of 100 nM ZnCl2 might create a more consistent ambient concentration of zinc, though ultimately any naturally occurring ionic zinc in the environment would likely be free to interact at the receptor (Cornelison et al 2017).

3.3 Oxytocin-glycine Interactions at α1 and α2 GlyRs

As reported by Bowen et al (2015), oxytocin does not modulate GABA-gated activity, nor does it generate current when applied individually to GABAA receptors. In the present experiments, oxytocin performed similarly, and did not generate a response when applied in the absence of glycine. To examine any potential modulation of Gly-activity at α1 and α2 receptors, oocytes were first pre-treated with the appropriate dose of oxytocin for 90 seconds, then exposed to oxytocin in combination with EC5 Gly (with 100 nM ZnCl2 added with α1 GlyRs). Statistical analyses for each receptor were performed using a one-way ANOVA with Dunnett’s test for multiple comparisons to the control (EC5 Gly).

Figure 12. Effects of oxytocin on glycine-gated current at WT GlyRs. Oocytes were pretreated with each concentration of OT for 90 seconds, followed by application of OT with 20 μM Gly (α2) and 100 nM ZnCl2 (α1). Wash out periods of at least 8 minutes were implemented following each application of oxytocin to allow removal from the bath environment. Responses elicited by application of OT with Gly were compared to the respective Gly control for each subtype (equal to 100%), and were expressed as a mean percent ±SEM of that control.

[34]

The application of oxytocin yielded no statistically significant modulation of glycine-gated current at concentrations ranging from 100 nM to 10 mM at either receptor (Figure 12). In α1 expressing oocytes, 3 and 10 μM slightly reduced glycine gated current to 88.3% ±3.25% and 89.7% ±6.81% of control respectively, though neither did so in a concentration dependent manner and thus was not significant. When tested at α2 expressing oocytes, 10 μM oxytocin appeared to positively modulate glycine at these receptors by 121.4% ±12.8, though this was not statistically significant (Figure 12). As such, it is unclear whether these are experimental artefacts or possibly due to concentration dependent effects at each subunit.

3.4 Ethanol-oxytocin Interactions at Wild Type α1 GlyRs

Oxytocin has previously been shown to attenuate the modulatory effects of ethanol at δ-

GABAA receptors when expressed in Xenopus laevis oocytes (Bowen et al 2015). Besides this, there is no literature describing direct interactions of oxytocin at ligand-gated ion channels, including glycine receptors thus far. Due to the similarities between glycine and GABAA receptors as behaviourally relevant, inhibitory LGIC targets of ethanol in the brain, it was hypothesised that oxytocin may exert a similar anti-alcohol effect at α1 GlyRs. To test this hypothesis, α1 homomers expressed in oocytes were pretreated with oxytocin and exposed to 100 mM ethanol, in combination with submaximal glycine (Figure 13).

(i) (ii) **** *

Figure 13. Summary of oxytocin-ethanol interactions at α1 GlyRs. (i) Representative trace illustrating responses to Gly, EtOH, and OT in an α1-expressing oocyte. (ii) Comparison of modulatory activity of ethanol and oxytocin at wild type α1 GlyRs. Mean responses are expressed as a percentage of the response to 20 μM Gly (equal to 100%), with error bars indicating SEM. Statistical analysis was performed with a one way ANOVA with Sidak’s test for multiple comparisons. *p<0.05, ****p<0.0001.

[35]

As illustrated in Figure 13, 100 mM EtOH potentiated Gly current by an average of 164.56% ±8.14% (p<0.0001). With the addition of 10 μM oxytocin, potentiation by 100 mM EtOH was significantly reduced by 48.1% to 116.51% ±8.78% of Gly control (p<0.05). This combination of ethanol, glycine, and oxytocin did not significantly differ from the control response to 20 μM glycine. Additionally, responses generated following co-application of 20 μM Gly and 10 μM OT were 89.68% of Gly control, though this was not a significant decrease.

Figure 14. Concentration dependent effects of oxytocin on EtOH activity at WT α1 GlyRs. Ethanol was coapplied with Gly and zinc at a concentration of 100 mM. Oocytes were pretreated with each concentration of OT, followed by application of OT with 20 μM Gly and 100 nM ZnCl2. All responses are expressed as mean percent relative to the response generated by 20 μM Gly (with zinc), with error bars indicating SEM. Data gathered from n=3- 23 oocytes. Significance testing was performed using a one way ANOVA with Sidak’s test for multiple comparisons. + indicates significant difference to 20 μM Gly & 100 ++, nM ZnCl2, * denotes significance compared with 100 mM EtOH, 20 μM Gly, 100 nM ZnCl2. **p<0.01, +++,***p<0.001, ++++,****p<0.0001.

Further concentrations of oxytocin were also tested at α1 GlyRs to establish concentration dependent effects. While several concentrations of oxytocin appear to reduce ethanol potentiation of glycine (Figure 14), a significant reduction was only observed with 10 μM oxytocin. 3 μM, 300 nM, and 30 nM OT reduced potentiation from 164.6% to 139.1% ±7.47%; 125.7% ±12.47%, and 135.5 ±10.34 of Gly control respectively. While not significantly less than EtOH-induced potentiation, these responses were also not significantly greater than control. With the addition of 100 nM OT, Gly potentiation by 100 mM EtOH was slightly reduced to 150.5% ±19.15 – remaining significantly greater than control (p<0.05).

Unexpectedly, 1 μM OT appeared to exacerbate EtOH-induced potentiation (not significant) up to 180.5% ±21.0%, which was significantly increased relative to the 20 μM Gly response (p<0.0001). Similarly, 10 nM OT significantly increased the response to 179.4% ±40.73% of Gly control (p<0.05). Because the standard error of these values was so high, the larger variability

[36] in this data may be difficult to interpret, though could reflect a multiphasic concentration- response relationship for oxytocin’s actions at α1 receptors.

3.5 Ethanol-oxytocin Interactions at α2 GlyRs

The α2 subunit may be found throughout many higher brain regions, where they probably exist as α2β heteromers, or potentially as α2 homomers (Takagi et al 1992). The subunit’s pattern of expression in the brain suggests α2 may mediate some of alcohol’s effects in these areas (McCracken et al 2013b). As with α1 GlyRs, direct interactions of oxytocin with the receptor have not yet been explored. Similarly to the α1 GlyR, it was hypothesised that oxytocin may modulate ethanol’s effects at these receptors. To test this hypothesis, α2- expressing oocytes were pretreated and exposed to varying concentrations of oxytocin, in combination with 60 mM ethanol and 80 μM Gly (EC5).

Figure 15. Concentration-dependent effects of oxytocin on ethanol-induced potentiation at α2 GlyRs. Ethanol potentiation of glycine was tested with 60 mM ethanol, coapplied with 80 μM Gly. Oocytes were pretreated with each concentration of OT for 90 seconds, followed by application of OT with 80 μM Gly and 60 mM EtOH. All responses are expressed as mean percent relative to the response generated by 80 μM Gly, with error bars indicating SEM. Data represents n=3-15 oocytes. Significance testing was performed using a one way ANOVA with Sidak’s test for multiple comparisons. +indicates significant difference to 80 μM Gly, *denotes significance compared with 60 mM EtOH, 80 μM Gly. ++,**p<0.01, +++,***p<0.001, ++++,****p<0.0001.

In homomeric α2A glycine receptors, 60 mM concentrations of ethanol significantly potentiated 80 µM Gly (~EC5) by an average of 143.56% ±2.33% (p<0.0001) (Figure 15). The addition of 10 μM OT significantly abolished this potentiation by 41.5% to 102.0% ±6.51% (p<0.0001), with a mean response not visibly or significantly different from 80 μM Gly control.

[37]

Co-application with 3 μM OT also significantly reduced Gly potentiation by 60 mM EtOH to 113.3% ±2.20% (p<0.0001). This response was also not significantly different to Gly control, though unlike the higher 10 μM OT concentration, complete attenuation of EtOH-induced potentiation was not observed. Oxytocin concentrations from 100 nM to 1 μM also partially impaired ethanol-induced potentiation in a statistically significant manner (Figure 15). 10 and 30 nM oxytocin also generated some negative modulation of EtOH potentiation, though not significantly. Responses in the presence of these OT concentrations (below 3 μM) maintained significantly greater than the response to 80 μM Gly alone (Figure 15).

In addition to oxytocin-ethanol interactions, the effect of arginine vasopressin (AVP) was also examined at α2 GlyRs. Oxytocin and vasopressin are structurally similar large nonapeptides, differing by only two amino acids. Both compounds occur in the brain and are thought to be involved in social behaviours (Heinrichs & Domes 2008). Based on the absence of anti-alcohol effects previously observed by Bowen et al (2015) at GABAA receptors, it was hypothesised that vasopressin would not modulate the effects of ethanol at α2 GlyRs. To test this, AVP was applied in place of equimolar oxytocin at α2 GlyRs.

(i) (ii)

Figure 16. Summary of oxytocin-ethanol and vasopressin interactions at α2 GlyRs. (i) Representative trace illustrating responses to Gly, EtOH, OT, and AVP in an α2-expressing oocyte. (ii) Comparison of modulatory activity of ethanol, oxytocin, and arginine vasopressin (AVP) at α2 GlyRs. Oocytes were exposed to each treatment, with 90 second pretreatment preceding those involving co-application with OT and AVP. Mean responses are expressed as a percentage of the response to 80 μM Gly (equal to 100%), with error bars indicating SEM. Statistical analysis was performed with a one way ANOVA with Sidak’s test for multiple comparisons. ****p<0.0001.

As previously discussed, 80 μM Gly was significantly potentiated by 60 mM EtOH (143.56% ±2.33%, p<0.0001), an effect which was abolished in the presence of 10 μM OT (p<0.0001; Figure 15 and Figure 16). When applied with 80 μM Gly, 10 μM OT did not appear to modulate

[38]

Gly-gated current (Figure 16Figure 15), and did not generate a response when applied individually. Taken together, the data suggest that the modulatory effect of oxytocin is only observable in the presence of ethanol at α2 GlyRs. Additionally, 10 μM AVP did not demonstrate the attenuation of EtOH-induced potentiation seen with equimolar OT, with potentiation remaining significantly greater than 80 μM Gly-control (147.36% ±3.25%, p<0.0001). Like oxytocin, vasopressin did not generate current when applied independently, and did not modulate 80 μM Gly.

3.6 Ethanol-oxytocin Interactions at α1δL2 Glycine Receptors

Perkins et al (2009) and Naito et al (2014) previously demonstrated the importance of Loop 2 sequences in regulating ethanol sensitivity of both γ GABAA and glycine receptors. They showed that replacement of the α1 Loop 2 sequence with the corresponding sequence of the

GABAA δ subunit conferred increased ethanol sensitivity. In the present experiments, these mutants were used to examine the differential effects of oxytocin and lower concentrations of ethanol. This attempted to validate the potentiation of WT GlyRs seen with higher EtOH concentrations by removing the potential for osmotic effects, in addition to enabling the use of more behaviourally and physiologically relevant concentrations of ethanol (Olsen et al 2007). Furthermore, use of the α1δL2 mutant allowed investigation of the role of Loop 2 in oxytocin and ethanol modulation of these receptors.

Figure 17. Ethanol induced potentiation of α1δL2 GlyRs expressed in Xenopus oocytes. Using two- electrode voltage clamp electrophysiology, ethanol doses were co-applied with ~EC5 glycine (6 μM).

Ethanol modulation is represented as a percent of EC5 glycine-induced current, as displayed on the y-axis. The perforated line at y=100% reflects glycine-induced current, with error bars displaying the SEM. Statistical analysis was performed with a one way ANOVA with Sidak’s test for multiple comparisons. *p<0.05, ***p<0.001.

[39]

Just as the mutant α1δL2 GlyR demonstrated significantly increased sensitivity to glycine (Section 3.1 Expression of Homomeric Glycine Receptors in Xenopus Oocytes), it also showed greater sensitivity to ethanol compared to α1 WT. These receptors responded to ethanol concentrations as low as 1 mM, generating a response 122.66% ±22.76% of Gly-control (not significant). 3 mM EtOH significantly potentiated 6 μM Gly by 160.38% ±10.68% (p<0.05), while

10 mM EtOH elicited 171% potentiation of EC5 Gly current (p<0.001) (Figure 17).

Given the significant magnitude of α1δL2 potentiation by 10 mM EtOH, this concentration was used to further investigate oxytocin interactions at the receptor. In both α1 and α2 wild type

GlyRs, 10 μM oxytocin significantly attenuated potentiation of EC5 glycine by ethanol. Due to the similarities between subunits and the conservation of oxytocin’s apparent effects across these and δ GABAA receptors, this concentration of oxytocin was also used to replicate interaction experiments at α1δL2 mutants.

(i) (ii)

Figure 18. Summary of ethanol-oxytocin interactions at ethanol-sensitive α1δL2 mutant GlyRs. (i) Representative trace illustrating responses to Gly, EtOH, and OT in an α1δL2-expressing oocyte. (ii) Comparison of the modulatory activity of ethanol and oxytocin at α1δL2 homomeric GlyRs. Oocytes were exposed to each treatment, with 90 second pretreatment preceding co-application with OT. Mean responses are expressed as a percentage of the response to 6 μM Gly (equal to 100%), with error bars indicating SEM. Statistical analysis was performed with a one way ANOVA with Sidak’s test for multiple comparisons. *p<0.05, ***p<0.001.

The potentiation of glycine-gated current by 10 mM EtOH (171%) was significantly decreased to 108.86% ±8.38% of control by the addition of 10 μM oxytocin (p<0.05) – similarly to results observed with WT α1 and α2 GlyRs. Again, responses to ethanol in the presence of oxytocin were not significantly increased relative to Gly-control (Figure 18). Independent application of oxytocin generated no response, and did not significantly potentiate EC5 (6μM) glycine. Overall, the pattern of OT-EtOH interactions in the α1δL2 mutant closely resembled those observed with α1 and α2 GlyRs. A notable difference however, was that these interactions occurred with a much smaller EtOH concentration – 10 mM as opposed to 60-100 mM.

[40]

Chapter 4: Discussion and Conclusion

Ethanol is one of the most widely used and abused substances worldwide. Despite its prevalence, mechanisms through which it exerts its intoxicating effects are not fully understood. As a highly non-specific drug, alcohol has a range of concentration-dependent targets, which include the inhibitory GABAA and glycine receptors. Based on molecular modelling of the prokaryotic GLIC, a potential ethanol binding site has been proposed for α1 glycine receptors at a homologous inter-subunit cavity (Sauguet et al 2013). Further evidence from behavioural, transgenic, and electrophysiological experiments support the role of glycine receptors in the alcohol response (Blednov et al 2015, Harris et al 2008, Mihic 1999, Soderpalm et al 2017). Additionally, oxytocin has been shown to counteract ethanol-induced potentiation of δ-GABAA receptors (Bowen et al 2015), and to influence alcohol use behaviours in murine and human experiments (Mitchell et al 2015, Pedersen et al 2013, Peters et al 2017). The present thesis examined the hypothesis that oxytocin interacts with the glycine receptor to alter its response to ethanol. The expression of human recombinant α1, α2 and α1δL2 GlyRs in Xenopus laevis oocytes enabled investigation of these interactions at a molecular level.

Alcohol was tested at α1 and α2 homomeric glycine receptors, where it generated significant potentiation at concentrations greater than 30 mM. Although it did not modulate glycine in the absence of EtOH, oxytocin attenuated this potentiation at concentrations ranging from 100 nM to 10 µM at α2 GlyRs, and at a concentration of 10 µM at α1 GlyRs. Additionally, neither ethanol nor oxytocin elicited responses at these receptors when glycine was not present. Investigation was extended to the actions of ethanol and oxytocin at mutant α1 glycine receptors. Replacing Loop 2 of the α1 GlyR subunit with that of the δ GABAAR subunit reduced the threshold for ethanol sensitivity, and appeared to increase the potency of glycine. However, this α1δL2 mutation did not appear to alter overall receptor function, nor reduce the threshold for oxytocin’s effects.

4.1 Ethanol-induced Potentiation of Glycine Receptors

As a small amphiphilic molecule, ethanol freely crosses the blood-brain barrier, with blood alcohol concentrations likely reflecting those found in the CNS. In line with this assumption, ethanol concentrations relevant to typical human consumption range between approximately 3 and 30 mM (Olsen et al 2007, Wallner et al 2003). As discussed, ethanol significantly potentiated glycine-gated current at α1 and α2 homomers, in what appeared to be a concentration-dependent manner.

[41]

At α1 homomers, 100 mM ethanol significantly potentiated glycine current, with slight (not significant) potentiation also seen with 30 mM ethanol. For α2 GlyRs, enhancement of glycine current was evident with ethanol concentrations of 60 mM and above. In humans, these concentrations would likely result in severe intoxication, with a typical lethal blood concentration of around 80 mM (Gable 2004). When expressed in this recombinant system, the present data suggest that wild type α1 GlyRs may be more relevant to severe rather than moderate alcohol intoxication. However, testing a sufficient number of cells with additional EtOH concentrations between 30 and 100 mM may elucidate a cut off point for where significant potentiation of α1 receptors occurs - possibly below this lethal threshold. Identifying this cut off point would allow a better understanding of the involvement of α1 containing GlyRs in ethanol intoxication. Ethanol-induced enhancement of glycine was observed below this broad threshold at α2 receptors, suggesting they may be involved in the actions of moderately high doses of ethanol. In humans, a 60 mM concentration of ethanol would likely induce severe intoxication, loss of consciousness, and would stimulate a range of other non-specific targets (Wallner & Olsen 2008).

The present results are similar to those reported by Crawford et al (2007), who demonstrated concentration dependent potentiation of WT α1 between 25 and 100 mM EtOH; and to Davies et al (2004), where 10 to 200 mM EtOH potentiated α1 and α2 GlyRs. This group also reported that ethanol did not produce current in the absence of glycine. Similar experiments on human α homomers expressed in Xenopus laevis oocytes by Mascia et al (1996) demonstrated potentiation of glycine (EC2) by much lower concentrations of ethanol. They demonstrated a threshold for significant potentiation of α1 and α2 GlyRs from ethanol doses of 5 mM and 10 mM respectively. Unlike these reports, the present results do not demonstrate significant potentiation at lower millimolar concentrations.

Variability is often encountered between laboratories, particularly when using Xenopus laevis oocytes as a recombinant system - where variability may arise between individual donor frogs, and between seasons (Weber 1999). The above groups utilised similar methods to those used here, with minor differences such as differences in oocyte buffer and the antibiotics used - which may influence results. Furthermore, these groups used an EC2 concentration of glycine, however the present experiments used an approximate EC5 concentration to allow observation of glycine currents, including in those with apparently lower receptor expression. As discussed, ethanol potentiation is greater with decreasing concentrations of glycine, hence this difference may contribute to disparate ethanol sensitivity.

Recombinantly expressed α1 homomers are reportedly more sensitive to ethanol than α2 (Mascia et al 1996), particularly at ethanol concentrations below 50 mM. Above 50 mM

[42] ethanol, α2 and α1 receptors did not exhibit differences in ethanol sensitivity. By comparison, significant potentiation was only seen above these concentrations in the present data. Overall, the present experiments do not effectively address subunit sensitivities, as ethanol concentrations are not precisely matched between receptor subtypes, with the presence of zinc at α1 GlyRs confounding this further. Importantly, the degree of potentiation exerted by ethanol is likely dependent upon the concentration of glycine present (Mascia et al 1996, Naito et al 2014), and receptor open probability. Even slight differences in these factors may influence the magnitude of ethanol potentiation observed. Receptor open probability was not quantified in these experiments, though would allow more accurate comparison of receptor subtypes in future experiments, possibly with the use of picrotoxin or an open channel blocker.

4.1.1 Zinc Modulation of α1 GlyRs

When testing the effects of Gly, EtOH and OT at wild type α1 GlyRs, zinc was co-applied with these compounds. Here, zinc enhanced responses to submaximal glycine at α1 GlyRs, reflecting the role of low levels of zinc as positive allosteric modulators of α homomeric GlyRs (Laube et al 1995). As a naturally occurring compound in the brain and within synapses (Assaf & Chung 1984), the presence of zinc ions may assist in generating a more native environment for recombinant GlyRs in oocytes. However, vast differences between mammalian GlyRs expressed in vivo and recombinantly still remain. Zinc also significantly enhanced potentiation by 30 mM but not 100 mM ethanol, while 30 mM ethanol did not significantly potentiate Gly current. This may have occurred because 30 mM was a submaximal concentration of ethanol, allowing further potentiation by zinc. Where ethanol is present, zinc may exert these effects through the interaction of both compounds with the receptor that increases receptor affinity for glycine (McCracken et al 2010b). Additionally, the magnitude of error seen with potentiation by 100 mM ethanol appeared to decrease in the presence of zinc. It is not clear whether this is due to zinc itself or is an incidental finding.

4.2 Interactions of Oxytocin and Ethanol

Bowen et al (2015) demonstrated a selective and specific effect of oxytocin at δ GABAA receptors. Aside from this, investigation of any direct interactions of oxytocin with Cys-loop receptors is absent from the literature. Xenopus laevis oocytes do not endogenously express the oxytocin receptor, thus their involvement in oxytocin’s apparent effects are improbable in these experiments (Kimura et al 1992). The present experiments showed that significant observable effects of oxytocin are restricted to the presence of both glycine and ethanol. In the absence of ethanol, oxytocin did not appear to modulate recombinant α1, α1δL2 or α2

[43] receptors. This suggests that the presence of ethanol could have direct or steric influences on the oxytocin binding site, either allowing entry of oxytocin to an otherwise shielded cavity, or by exposing residues necessary for docking at the receptor. Such conformational changes induced by ethanol may also extend to other n-alcohols. Co-crystallisation of ethanol and glycine at the glycine receptor, and the development of molecular models would allow further investigation of these changes.

When co-applied with ethanol, oxytocin mitigated ethanol-induced potentiation of these receptors - appearing to act as a use-dependent negative allosteric modulator of ethanol. At α1 receptors, reductions in ethanol-induced potentiation were observed above 3 µM oxytocin. At α1δL2 receptors, 10 µM oxytocin almost completely abolished enhancement by 10 mM ethanol. When tested at α2 receptors, oxytocin reduced ethanol’s effects at concentrations of 100 nM upwards dose-dependently. At a concentration of 10 µM, oxytocin completely abolished the response to 60 mM ethanol. Complete abolishment may have been observed as the magnitude of ethanol potentiation at α2 (143.56%) was less than that of α1 and α1δL2 (164.56% and 171% respectively). This supports an interactive effect of both oxytocin and ethanol, as the degree of ethanol-induced potentiation influenced the extent to which oxytocin was able to counteract this. It is also important to note that these micromolar concentrations of oxytocin are unlikely to occur in the CNS - either endogenously or following intranasal delivery. In humans, baseline cerebrospinal fluid concentrations (CSF) of oxytocin are generally much lower, with an average baseline concentration of 16.3 pg/ml reported by Kagerbauer et al (2013). Another study reported an increase in CSF oxytocin levels from a baseline of approximately 18 pg/ml up to 30 pg/ml following intranasal delivery (Striepens et al 2013). Because of this, it may be assumed that the present effects of oxytocin reflect a pharmacologically relevant interaction, though may not occur significantly in vivo under normal conditions.

Overall, the effects of oxytocin (10 µM) remained constant between α1 and α1δL2 receptors, suggesting oxytocin does not bind at a Loop 2 site at the α1 subunit. However, these experiments have not illustrated a specific binding site for oxytocin at homomeric glycine receptors. One means of investigating this would be to test the effects of oxytocin under reducing conditions, which would disrupt the cysteine disulphide bridges in the molecule (using a reducing agent such as 2-mercaptoethanol). Disrupting oxytocin’s tertiary structure in this way would allow observation of whether the conformation or the presence of disulphide bonds is necessary for oxytocin-ethanol interactions to occur. Further investigation of oxytocin binding could also involve examining oxytocin’s larger degradation products - or fragments of the compound - for any modulatory effects on glycine or ethanol at α1 and α2 GlyRs.

[44]

Because ethanol is a positive modulator of α GlyRs, it may be worthwhile investigating the effects of other n-alcohols and their interactions with oxytocin. As n-alcohols occupy different volumes in the alcohol binding pocket, they may exert steric influences upon residues important for oxytocin binding. Additionally, because volatile anaesthetics are purported to act at the same site as alcohol, examining possible interactions between oxytocin and these compounds may also aid understanding of oxytocin’s association with these receptors.

In addition to demonstrating oxytocin-ethanol interactions in recombinant GABAA receptors in Xenopus oocytes, Bowen et al (2015) observed attenuation of behavioural and psychomotor effects in rats. Rats exposed to intoxicating doses of ethanol (~ 30 mM), who had also received oxytocin showed significant reductions in ataxia and sedation, though oxytocin did not decrease impairment from much larger doses of ethanol. Due to the importance of glycine receptors in motor coordination and relay, oxytocin’s attenuation of ethanol-induced potentiation at native glycine receptors may have contributed to these improvements in motor activity - particularly if in vivo GlyR populations were indeed enhanced by the relevant ethanol concentrations.

An additional consideration is that oxytocin receptors may potentially be present in residual follicular cells (Miledi & Woodward 1989) not removed in the defolliculation process. Given that oxytocin is an agonist at oxytocin receptors, the lack of response to oxytocin when applied alone did not indicate their presence in these experiments. Future experiments could further determine their absence or presence with the use of an oxytocin receptor antagonist.

4.2.1 Vasopressin Does Not Modulate Ethanol’s Effects

Bowen et al (2015) tested AVP at αβδ GABAA receptors, finding that it did not modulate the effects of alcohol on these receptors. Presently, arginine vasopressin (AVP) did not mimic the effects of oxytocin when tested at α2 GlyRs. Together, these investigations suggest that despite their structural and physiological similarities, oxytocin may possess unique structural features which allow it to interact with these LGICs. OT and AVP are both nonapeptides, differing by only two amino acids - AVP replaces OT’s isoleucine and leucine with phenylalanine and arginine at positions 3 and 8 respectively (Figure 5). Residues at position 3 form part of the cyclic moiety - both isoleucine and phenylalanine are hydrophobic, though AVP’s phenylalanine is larger. At position 8, oxytocin’s lipophilic leucine is replaced by a larger, positively charged arginine. Because AVP did not influence α2 GlyRs, this suggests that the binding site for oxytocin houses a smaller hydrophobic pocket and favours a smaller lipophilic residue over a (larger) positively charged residue.

[45]

The effects of vasopressin at α1 and α1δL2 receptors were not examined in these experiments.

Because AVP does not modulate δ-GABAA receptors (Bowen et al 2015), and was not observed to modulate α2 GlyRs in the present experiments, it was hypothesised that no influence would be observed on other α GlyRs. The binding site of oxytocin is as yet unidentified at these receptors, suggesting that AVP should also be tested at α1 and α1δL2 GlyRs, as different allosteric interactions between subunit residues could possibly influence association with AVP. As discussed, the α1δL2 mutation may not target areas involved binding oxytocin, suggesting that like OT, AVP is unlikely to interact with the Loop 2 region. However, this warrants further investigation to resolve whether the incurred change in the Loop 2 sequence influences response to either neuropeptide.

4.2.2 Oxytocin May Interact with Zinc at α1 GlyRs

As previously discussed, zinc was co-applied with ethanol and oxytocin at α1 receptors. This is of importance when examining effects of oxytocin at α1 GlyRs as zinc may form a conglomerate with oxytocin in solution. In this complex, the zinc ion occupies an interior site of the oxytocin peptide, which could be necessary for effective association of the peptide with the oxytocin receptor. This causes a conformational change of the peptide, freeing the N- terminus and generating a hydrophobic site on the exterior surface (Liu et al 2005). It is not clear which conditions must be present in solution for these complexes to occur, though they may do so naturally, with significant formation under acidic conditions (Wyttenbach et al 2008). Though the above study suggests that the zinc-oxytocin (Zn-OT) complex increases affinity for the OT receptor, it may interfere or facilitate binding at GlyRs and other ligand- gated ion channels. The effects of formation of this complex could be twofold - by decreasing the availability of zinc ions to interact and positively modulate the receptor, and by inducing a conformational change in the OT molecule itself - which could influence its association with the receptor.

The presence of Zn-OT complexes may have contributed to the biphasic concentration response trend observed for oxytocin’s effects on ethanol potentiation at α1 GlyRs. Between 30 and 300 nM, negative modulation of ethanol potentiated glycine current was observed at a similar magnitude to higher (3-10 µM) concentrations of oxytocin. The same pattern was not observed at α2 homomers, where oxytocin’s concentration dependent effects upon ethanol- induced enhancement were observed in a typical, monophasic fashion; and less variation was seen within the data. While it is possible that oxytocin exerts multiphasic effects at α1 receptors, it could also be the case that a reduction in available zinc contributed to this. Oxytocin’s interactive effects were measured relative to currents induced by a combination of

[46] ethanol, glycine and zinc. Zinc and ethanol appear to have synergistic rather than additive effects upon α1 GlyRs (McCracken et al 2010b). As such, decreased availability of zinc ions would be expected to lower ethanol potentiation in the presence of oxytocin, resulting in apparent inhibition of the ethanol response by oxytocin. This may also have contributed to the reduced variability of oxytocin responses at α2 GlyRs, where zinc was not present. It is not clear whether Zn-OT complexes form preferentially at lower concentrations of oxytocin, or if they otherwise override the effects of ethanol at lower oxytocin concentrations only. To better understand this, investigation into the concentration dependence of complex formation is necessary.

Additionally, zinc ions may be beneficial for stabilisation of aqueous oxytocin, and for preventing dimerisation between oxytocin molecules (Avanti et al 2011). Stabilisation may also occur with divalent ions like Mg2+, which were also present in ND96 buffer used throughout oocyte experiments (in the form of 1 mM MgCl2). Thus, increased oxytocin stability from zinc may have also contributed to the differences in concentration related effects between α1 and α2 GlyRs, though magnesium could have had similar effects at both subtypes. As zinc contamination may occur and other GlyR subunits may be potentiated by zinc, the effects of this complex could also have some impact on α2 receptors - though the effects of this complex would likely be more pronounced with higher levels of zinc as tested at α1. One way to further examine the effects of the Zn-OT complex would be to apply ethanol and oxytocin in the presence of the zinc chelator tricine, to reduce environmental zinc and prevent complex formation. Additionally, testing oxytocin’s concentration dependent interactions at both α1 and α2 receptors in the absence of zinc, using the same effective concentrations of ethanol and glycine would allow thorough comparison across receptor subtypes.

4.3 Oxytocin-ethanol Interactions at α1δL2 GlyRs

δ subunit-containing GABAA receptors demonstrate high sensitivity to low levels of ethanol (Olsen et al 2007, Wallner et al 2003). The distribution pattern of these receptors reflects their role in tonic inhibition and the behavioural effects of alcohol consumption. Previously, Loop 2 was found to be an important regulator of ethanol sensitivity in δ-GABAA, and that insertion of the δ Loop 2 sequence in γ-GABAA and Gly receptors conferred increased EtOH sensitivity (Naito et al 2014, Perkins et al 2009).

In the present investigation, significant ethanol-induced enhancement of wild type α1 and α2 glycine receptors was only observed at relatively high concentrations. To investigate oxytocin- ethanol interactions at lower concentrations of ethanol, the ethanol-sensitive α1δL2 mutant receptor was utilised. Use of this mutant also allowed examination of the role of Loop 2 in the

[47] effects of ethanol and oxytocin at both glycine and GABAA receptors expressed in Xenopus laevis oocytes. As anticipated, the introduction of the δL2 mutation greatly increased sensitivity to ethanol. Sensitivity to glycine was also increased, while the response to oxytocin was unchanged. Despite a decrease in glycine EC50 and EC5 relative to α1, the α1δL2 mutant did not differ significantly in terms of Hill slope or Imax. These findings mirror those of Crawford et al (2007) and Perkins et al (2009), who also established that other Loop 2 mutations of α1

GlyRs (such as substituting the GABAA receptor γ subunit sequence or mutating position 52) did not influence EC50, Hill slope or Imax values. This group also found no effect of the α1δL2 mutation upon the response to picrotoxin and strychnine. Also in keeping with the findings of Perkins et al (2009), sensitivity to (presumably) allosteric modulation by oxytocin did not appear to change. Overall, these findings suggest that the increased ethanol sensitivity of α1δL2 mutants does not arise from changes in receptor conformation which alter basic functioning of the α1 GlyR. However, the precise mechanism through which increased ethanol sensitivity arises has not been established. Further investigation - using other allosteric modulators in place of ethanol – may provide some insight into whether the Loop 2 mutation is specific to ethanol or may be extended to other allosteric modulators.

As previously discussed, oxytocin’s effects upon glycine and ethanol did not differ between wild type α1 and α1δL2 receptors. At both subtypes, low micromolar concentrations of oxytocin significantly attenuated ethanol-induced enhancement of Gly current, without modulating glycine current itself. As with other GlyR subtypes tested, α1δL2 homomers were not directly activated by oxytocin. Based on these findings, it appears that Loop 2 of the α1 GlyR is not involved in oxytocin’s modulatory effects on ethanol and glycine at the receptor. The present data also reflects that of Bowen et al (2015), who found similar effects and oxytocin-ethanol interactions at δ GABAA receptors. Considering the structural and functional similarities of these LGICs, Loop 2 may also not be important for oxytocin-ethanol interactions at δ subunit-containing GABAA receptors.

4.4 Conclusion and Further Considerations

This research supported earlier reports of ethanol-induced potentiation of α homomeric glycine receptors. As anticipated, oxytocin appeared to mitigate ethanol-induced enhancement of these receptors, with low micromolar concentrations abolishing much of ethanol’s effects. Additionally, oxytocin did not independently modulate glycine receptors, while its modulatory effects were not reproduced by arginine vasopressin at α2 receptors, supporting the specific and selective oxytocin-ethanol interactions first described by Bowen et al (2015) at δ GABAA receptors. By replacing the α1 GlyR Loop 2 sequence with that of the δ

[48]

GABAAR subunit, sensitivity to glycine and ethanol was increased, with the effects of oxytocin appearing unchanged. Because of the relative similarities in ethanol and oxytocin responses between WT and mutant α1 GlyRs, the potentiation of glycine current generated by high millimolar ethanol concentrations at WT α1 may reflect a pharmacological action - rather than being a function of disturbed osmolarity. The threshold for ethanol was reduced such that low millimolar (10 mM) concentrations significantly enhanced α1δL2 function, in agreement with earlier work by Perkins et al (2009) and Naito et al (2014). This indicated that the δ Loop 2 sequence conferred increased ethanol sensitivity to α1 GlyRs expressed in Xenopus oocytes, highlighting the importance of this region in the ethanol response of both glycine and GABAA receptors. Furthermore, since the Loop 2 mutation increased sensitivity to ethanol but not oxytocin, it seems likely that the oxytocin binding site is not found in this region at α1 glycine receptors. Due to the conserved structures and similarities between Gly and GABAA receptors, this may also extend to GABAA receptors, where oxytocin may also interact with a non-Loop 2 site. However the present experiments did not identify the binding site of oxytocin, therefore further investigation is required to elucidate its mechanism of ethanol modulation at the glycine receptor.

4.4.1 Further Considerations and Future Directions

Some difference in absolute maximum current (Imax) was observed between α1 and α2 GlyRs, despite microinjection of the same amount of mRNA and incubation for approximately the same amount of time (2-4 days). To examine whether these changes in Imax were due to altered surface expression levels, it may be useful to compare receptor protein content using cell- surface biotinylation and immunoblotting methods.

The present experiments have not identified a binding site for oxytocin at α homomeric glycine receptors. As oxytocin is already approved for use in humans, and clinical trials are currently underway, there is little cause for in vivo experiments using non-human animals. Additionally, such methods would likely be highly confounded, as oxytocin has multiple apparent actions on ethanol targets - which would also be difficult to isolate pharmacologically. As such, the best course of action would likely involve further recombinant expression of GlyRs, together with pharmacological manipulation (such as with other allosteric modulators), and further mutagenesis studies. Ideally, smaller fragments or large degradation products of oxytocin would also be tested, and may elucidate regions of the compound or receptor important for binding. Co-crystallisation of ethanol and glycine at the glycine receptor may allow further insight into ethanol-induced conformational changes that may allow oxytocin to interact with the receptor, though achieving this may be technically difficult. In silico methods would also

[49] contribute to our understanding of oxytocin’s association with glycine receptors. A model of oxytocin binding could be developed, with ethanol occupying the site proposed by Sauguet et al (2013).

Research into the effective delivery of exogenous oxytocin and its analogues into the CNS is ongoing, geared towards increased central bioavailability and greater efficacy in treating psychiatric disorders. One example is the lipidated OT analogue lipo-oxytocin-1, which generated similar behavioural effects to OT in mice, though appeared longer lasting (Mizuno et al 2015). It would be interesting to examine whether oxytocin analogues generate similar interactions with ethanol at α glycine receptors and δ GABAA receptors. A possibility of extending these experiments to clinical investigation of approved OT-related compounds in the future would also be educative, particularly in terms of whether they are able to attenuate any of ethanol’s Gly or GABAA receptor-mediated effects.

4.4.2 Limitations of These Experiments

Here, human homomeric α glycine receptors were expressed recombinantly in Xenopus laevis oocytes. While Xenopus oocytes present a valuable tool for investigation the function of ion channels like glycine receptors, there are also several inherent limitations of this method. One such limitation is that compounds are generally less potent at receptors expressed in oocytes than those expressed natively or in mammalian cells (such as HEK cells), though the mechanism through which this occurs is not entirely clear. This could be due in part to remnants of the highly invaginated membrane, which may impair access of compounds to the receptors (Goldin 2006). Additionally, recombinant receptors may be irregularly distributed or clustered on the membrane, and may bind to both the yolk and surface membrane - thereby also contributing to decreased potency (Papke & Smith-Maxwell 2009). Additionally, there can be considerable variability between research groups, and even between individual donor frogs, while seasonal variation in oocyte quality and results can also occur (Weber 1999).

Because post-translational modifications of subunits may occur in mammalian systems, and relevant accessory proteins may not be present, ethanol sensitivity of α GlyRs can differ significantly when expressed in vivo rather than in Xenopus oocytes. Differences in receptor function could also be influenced by the presence of the β subunit in heteromeric GlyRs, which are found throughout the brain and spinal cord and are likely involved in the ethanol response in vivo. Though more difficult to express in oocytes, heteromeric αβ GlyRs would ideally be examined for ethanol and oxytocin effects, allowing a better understanding of how native GlyRs might respond in the CNS.

[50]

The brain is a complex system, with intricate temporal and spatial dynamics, which are vital to our understanding of its function. In the context of alcohol consumption, alcohol’s subjective effects involve changes from the molecular to regional level. Because of its differential effects on both excitatory and inhibitory targets, ethanol influences neural pathways and processes in a highly complex manner. In contrast, the present work looks at the effects and interactions of oxytocin and ethanol with a single target at the receptor level - a somewhat reductionist approach. To better understand the validity of the present data, it is important to examine the role of glycine receptors in these pathways and processes, particularly given their role as a major inhibitory neurotransmitter receptor throughout the central nervous system.

[51]

References

Absalom NL, Lewis TM, Kaplan W, Pierce KD, Schofield PR. 2003. Role of Charged Residues in Coupling Ligand Binding and Channel Activation in the Extracellular Domain of the Glycine Receptor. Journal of Biological Chemistry 278: 50151-57

Aguayo LG, Tapia JC, Pancetti FC. 1996. Potentiation of the glycine-activated Cl- current by ethanol in cultured mouse spinal neurons. Journal of Pharmacology and Experimental Therapeutics 279: 1116-22

Assaf SY, Chung S-H. 1984. Release of endogenous Zn2+ from brain tissue during activity. Nature 308: 734-36

Avanti C, Amorij J-P, Setyaningsih D, Hawe A, Jiskoot W, et al. 2011. A New Strategy to Stabilize Oxytocin in Aqueous Solutions: I. The Effects of Divalent Metal Ions and Citrate Buffer. The AAPS Journal 13: 284-90

Avila A, Nguyen L, Rigo J-M. 2013a. Glycine receptors and brain development. Frontiers in Cellular Neuroscience 7: 184

Avila A, Vidal Pía M, Dear T N, Harvey Robert J, Rigo J-M, Nguyen L. 2013b. Glycine Receptor α2 Subunit Activation Promotes Cortical Interneuron Migration. Cell Reports 4: 738-50

Avila A, Vidal PM, Tielens S, Morelli G, Laguesse S, et al. 2014. Glycine receptors control the generation of projection neurons in the developing cerebral cortex. Cell Death And Differentiation 21: 1696

Badanich KA, Mulholland PJ, Beckley JT, Trantham-Davidson H, Woodward JJ. 2013. Ethanol Reduces Neuronal Excitability of Lateral Orbitofrontal Cortex Neurons Via a Glycine Receptor Dependent Mechanism. Neuropsychopharmacology 38: 1176-88

Beato M, Groot-Kormelink PJ, Colquhoun D, Sivilotti LG. 2002. Openings of the Rat Recombinant α1 Homomeric Glycine Receptor as a Function of the Number of Agonist Molecules Bound. The Journal of General Physiology 119: 443-66

Belelli D, Harrison NL, Maguire J, Macdonald RL, Walker MC, Cope DW. 2009. Extrasynaptic GABAA receptors: form, pharmacology, and function. J Neurosci 29: 12757-63

Betz H, Laube B. 2006. Glycine receptors: recent insights into their structural organization and functional diversity. Journal of Neurochemistry 97: 1600-10

Blednov YA, Benavidez JM, Black M, Leiter CR, Osterndorff-Kahanek E, Harris RA. 2015. Glycine receptors containing alpha2 or alpha3 subunits regulate specific ethanol-mediated behaviors. J Pharmacol Exp Ther 353: 181-91

Boccia ML, Petrusz P, Suzuki K, Marson L, Pedersen CA. 2013. Immunohistochemical localization of oxytocin receptors in human brain. Neuroscience 253: 155-64

Bode A, Lynch JW. 2014. The impact of human hyperekplexia mutations on glycine receptor structure and function. Molecular Brain 7: 2

Borghese CM, Blednov YA, Quan Y, Iyer SV, Xiong W, et al. 2012. Characterization of Two Mutations, M287L and Q266I, in the α1 Glycine Receptor Subunit That Modify Sensitivity to Alcohols. The Journal of Pharmacology and Experimental Therapeutics 340: 304-16

[52]

Borghese CM, Stórustovu Sí, Ebert B, Herd MB, Belelli D, et al. 2006. The δ Subunit of γ- Aminobutyric Acid Type A Receptors Does Not Confer Sensitivity to Low Concentrations of Ethanol. Journal of Pharmacology and Experimental Therapeutics 316: 1360-68

Bowen MT, Neumann ID. 2017a. The Multidimensional Therapeutic Potential of Targeting the Brain Oxytocin System for the Treatment of Substance Use Disorders. Current topics in behavioral neurosciences

Bowen MT, Neumann ID. 2017b. Rebalancing the Addicted Brain: Oxytocin Interference with the Neural Substrates of Addiction. Trends in Neurosciences 40: 691-708

Bowen MT, Peters ST, Absalom N, Chebib M, Neumann ID, McGregor IS. 2015. Oxytocin prevents ethanol actions at δ subunit-containing GABAA receptors and attenuates ethanol- induced motor impairment in rats. Proceedings of the National Academy of Sciences 112: 3104- 09

Brickley SG, Cull-Candy SG, Farrant M. 1996. Development of a tonic form of synaptic inhibition in rat cerebellar granule cells resulting from persistent activation of GABAA receptors. The Journal of Physiology 497: 753-59

Burgos CF, Muñoz B, Guzman L, Aguayo LG. 2015. Ethanol effects on glycinergic transmission: From molecular pharmacology to behavior responses. Pharmacological Research 101: 18-29

Burgos CF, Yévenes GE, Aguayo LG. 2016. Structure and Pharmacologic Modulation of Inhibitory Glycine Receptors. Molecular Pharmacology 90: 318-25

Carver CM, Wu X, Gangisetty O, Reddy DS. 2014. Perimenstrual-Like Hormonal Regulation of Extrasynaptic δ-Containing GABAA Receptors Mediating Tonic Inhibition and Neurosteroid Sensitivity. The Journal of Neuroscience 34: 14181-97

Cederholm JME, Absalom NL, Sugiharto S, Griffith R, Schofield PR, Lewis TM. 2010. Conformational changes in extracellular loop 2 associated with signal transduction in the glycine receptor. Journal of Neurochemistry 115: 1245-55

Claydon C, Webber K, Sweeney J. 2016. National Drug Strategy Household Survey 2016. Canberra, Australia

Cornelison GL, Daszkowski AW, Pflanz NC, Mihic SJ. 2017. Interactions between Zinc and Allosteric Modulators of the Glycine Receptor. Journal of Pharmacology and Experimental Therapeutics 361: 1-8

Cornelison GL, Pflanz NC, Tipps ME, Mihic SJ. 2016. Identification and characterization of heptapeptide modulators of the glycine receptor. European Journal of Pharmacology 780: 252- 59

Crawford DK, Perkins DI, Trudell JR, Bertaccini EJ, Davies DL, Alkana RL. 2008. Roles for Loop 2 Residues of α1 Glycine Receptors in Agonist Activation. Journal of Biological Chemistry 283: 27698-706

Crawford DK, Trudell JR, Bertaccini EJ, Li K, Davies DL, Alkana RL. 2007. Evidence that ethanol acts on a target in Loop 2 of the extracellular domain of alpha1 glycine receptors. J Neurochem 102: 2097-109

Davies DL, Crawford DK, Trudell JR, Mihic SJ, Alkana RL. 2004. Multiple sites of ethanol action in α1 and α2 glycine receptors suggested by sensitivity to pressure antagonism. Journal of Neurochemistry 89: 1175-85

[53] de Bejczy A, Nations KR, Szegedi A, Schoemaker J, Ruwe F, Soderpalm B. 2014. Efficacy and safety of the glycine transporter-1 inhibitor org 25935 for the prevention of relapse in alcohol- dependent patients: a randomized, double-blind, placebo-controlled trial. Alcoholism, clinical and experimental research 38: 2427-35

Duguid I, Branco T, London M, Chadderton P, Häusser M. 2012. Tonic Inhibition Enhances Fidelity of Sensory Information Transmission in the Cerebellar Cortex. The Journal of Neuroscience 32: 11132-43

Durisic N, Godin AG, Wever CM, Heyes CD, Lakadamyali M, Dent JA. 2012. Stoichiometry of the Human Glycine Receptor Revealed by Direct Subunit counting. The Journal of neuroscience : the official journal of the Society for Neuroscience 32: 12915-20

Eichler SA, Förstera B, Smolinsky B, Jüttner R, Lehmann T-N, et al. 2009. Splice-specific roles of glycine receptor α3 in the hippocampus. European Journal of Neuroscience 30: 1077-91

Elbourne DR, Prendiville WJ, Carroli G, Wood J, McDonald S. 2001. Prophylactic use of oxytocin in the third stage of labour. The Cochrane database of systematic reviews: Cd001808

Engblom AC, Akerman KE. 1991. Effect of ethanol on gamma-aminobutyric acid and glycine receptor-coupled Cl- fluxes in rat brain synaptoneurosomes. J Neurochem 57: 384-90

Estrada-Mondragon A, Lynch JW. 2015. Functional characterization of ivermectin binding sites in α1β2γ2L GABA(A) receptors. Frontiers in Molecular Neuroscience 8: 55

Evans SL, Monte OD, Noble P, Averbeck BB. 2014. Intranasal oxytocin effects on social cognition: a critique. Brain research 0: 69-77

Findlay GS, Wick MJ, Mascia MP, Wallace D, Miller GW, et al. 2002. Transgenic Expression of a Mutant Glycine Receptor Decreases Alcohol Sensitivity of Mice. Journal of Pharmacology and Experimental Therapeutics 300: 526-34

Förstera B, Muñoz B, Lobo MK, Chandra R, Lovinger DM, Aguayo LG. 2017. Presence of ethanol- sensitive glycine receptors in medium spiny neurons in the mouse nucleus accumbens. The Journal of Physiology 595: 5285-300

Gable RS. 2004. Comparison of acute lethal toxicity of commonly abused psychoactive substances. Addiction (Abingdon, England) 99: 686-96

Gimpl G, Fahrenholz F. 2001. The Oxytocin Receptor System: Structure, Function, and Regulation. Physiological Reviews 81: 629-83

Glykys J, Peng Z, Chandra D, Homanics GE, Houser CR, Mody I. 2007. A new naturally occurring GABAA receptor subunit partnership with high sensitivity to ethanol. Nat Neurosci 10: 40-48

Goldin AL. 2006. Expression of Ion Channels in Xenopus Oocytes In Expression and Analysis of Recombinant Ion Channels, pp. 1-25: Wiley-VCH Verlag GmbH & Co. KGaA

Grudzinska J, Schemm R, Haeger S, Nicke A, Schmalzing G, et al. 2005. The beta subunit determines the ligand binding properties of synaptic glycine receptors. Neuron 45

Hamann M, Rossi DJ, Attwell D. 2002. Tonic and Spillover Inhibition of Granule Cells Control Information Flow through Cerebellar Cortex. Neuron 33: 625-33

Hanchar HJ, Dodson PD, Olsen RW, Otis TS, Wallner M. 2005. Alcohol-induced motor impairment caused by increased extrasynaptic GABAA receptor activity. Nat Neurosci 8: 339-45

[54]

Harris RA, Trudell JR, Mihic SJ. 2008. Ethanol's molecular targets. Science signaling 1: re7

Harvey RJ, Depner UB, Wassle H, Ahmadi S, Heindl C, et al. 2004. GlyR alpha3: an essential target for spinal PGE2-mediated inflammatory pain sensitization. Science 304

Heinrichs M, Domes G. 2008. Neuropeptides and social behaviour: effects of oxytocin and vasopressin in humans. Progress in brain research 170: 337-50

Herd MB, Haythornthwaite AR, Rosahl TW, Wafford KA, Homanics GE, et al. 2008. The expression of GABAAβ subunit isoforms in synaptic and extrasynaptic receptor populations of mouse dentate gyrus granule cells. The Journal of Physiology 586: 989-1004

Horani S, Stater EP, Corringer PJ, Trudell JR, Harris RA, Howard RJ. 2015. Ethanol Modulation is Quantitatively Determined by the Transmembrane Domain of Human alpha1 Glycine Receptors. Alcoholism, clinical and experimental research 39: 962-8

Howard RJ, Murail S, Ondricek KE, Corringer P-J, Lindahl E, et al. 2011. Structural basis for alcohol modulation of a pentameric ligand-gated ion channel. Proceedings of the National Academy of Sciences of the United States of America 108: 12149-54

Howard RJ, Trudell JR, Harris RA. 2014. Seeking structural specificity: direct modulation of pentameric ligand-gated ion channels by alcohols and general anesthetics. Pharmacological reviews 66: 396-412

Huang EP. 1997. Metal ions and synaptic transmission: Think zinc. Proceedings of the National Academy of Sciences 94: 13386-87

Hurley JH, Ballard CJ, Edenberg HJ. 2009. Altering the Relative Abundance of GABAA Receptor Subunits Changes GABA- and Ethanol-Responses in Xenopus Oocytes. Alcoholism: Clinical and Experimental Research 33: 1089-96

Hussy N, Brès V, Rochette M, Duvoid A, Alonso G, et al. 2001. Osmoregulation of Vasopressin Secretion via Activation of Neurohypophysial Nerve Terminals Glycine Receptors by Glial Taurine. The Journal of Neuroscience 21: 7110-16

Imlach WL, Bhola RF, Mohammadi SA, Christie MJ. 2016. Glycinergic dysfunction in a subpopulation of dorsal horn interneurons in a rat model of neuropathic pain. Scientific Reports 6: 37104

Johnson R, Noll E, Rodney WM. 1982. SURVIVAL AFTER A SERUM ETHANOL CONCENTRATION OF 11/2%. The Lancet 320: 1394

Jonsson S, Morud J, Pickering C, Adermark L, Ericson M, Soderpalm B. 2012. Changes in glycine receptor subunit expression in forebrain regions of the Wistar rat over development. Brain research 1446: 12-21

Kagerbauer SM, Martin J, Schuster T, Blobner M, Kochs EF, Landgraf R. 2013. Plasma Oxytocin and Vasopressin do not Predict Neuropeptide Concentrations in Human Cerebrospinal Fluid. Journal of Neuroendocrinology 25: 668-73

Kaneko Y, Pappas C, Tajiri N, Borlongan CV. 2016. Oxytocin modulates GABAAR subunits to confer neuroprotection in stroke in vitro. Scientific Reports 6: 35659

Kimura T, Tanizawa O, Mori K, Brownstein MJ, Okayama H. 1992. Structure and expression of a human oxytocin receptor. Nature 356: 526-9

[55]

Koksma J-J, van Kesteren RE, Rosahl TW, Zwart R, Smit AB, et al. 2003. Oxytocin Regulates Neurosteroid Modulation of GABA_AReceptors in Supraoptic Nucleus around Parturition. The Journal of Neuroscience 23: 788-97

Korpi ER, Grunder G, Luddens H. 2002. Drug interactions at GABA(A) receptors. Progress in neurobiology 67: 113-59

Kubota H, Alle H, Betz H, Geiger JR. 2010. Presynaptic glycine receptors on hippocampal mossy fibers. Biochem Biophys Res Commun 393: 587-91

Kuhse J, Kuryatov A, Maulet Y, Malosio ML, Schmieden V, Betz H. 1991. Alternative splicing generates two isoforms of the alpha 2 subunit of the inhibitory glycine receptor. FEBS letters 283: 73-7

Laube B, Kuhse J, Rundstrom N, Kirsch J, Schmieden V, Betz H. 1995. Modulation by zinc ions of native rat and recombinant human inhibitory glycine receptors. J Physiol 483

Laurie DJ, Seeburg PH, Wisden W. 1992. The distribution of 13 GABAA receptor subunit mRNAs in the rat brain. II. Olfactory bulb and cerebellum. J Neurosci 12: 1063-76

Lee E-A, Cho J-H, Choi I-S, Nakamura M, Park H-M, et al. 2009. Presynaptic glycine receptors facilitate spontaneous glutamate release onto hilar neurons in the rat hippocampus. Journal of Neurochemistry 109: 275-86

Lee MR, Rohn MCH, Tanda G, Leggio L. 2016. Targeting the Oxytocin System to Treat Addictive Disorders: Rationale and Progress to Date. CNS drugs 30: 109-23

Li J, Nie H, Bian W, Dave V, Janak PH, Ye J-H. 2012. Microinjection of Glycine into the Ventral Tegmental Area Selectively Decreases Ethanol Consumption. The Journal of Pharmacology and Experimental Therapeutics 341: 196-204

Liu D, Seuthe AB, Ehrler OT, Zhang X, Wyttenbach T, et al. 2005. Oxytocin-Receptor Binding: Why Divalent Metals Are Essential. Journal of the American Chemical Society 127: 2024-25

Lynch JW. 2004. Molecular Structure and Function of the Glycine Receptor Chloride Channel. Physiological Reviews 84: 1051-95

Lynch JW. 2009. Native glycine receptor subtypes and their physiological roles. Neuropharmacology 56: 303-09

MacFadyen K, Loveless R, DeLucca B, Wardley K, Deogan S, et al. 2016. Peripheral oxytocin administration reduces ethanol consumption in rats. Pharmacology Biochemistry and Behavior 140: 27-32

Malosio ML, Grenningloh G, Kuhse J, Schmieden V, Schmitt B, et al. 1991a. Alternative splicing generates two variants of the alpha 1 subunit of the inhibitory glycine receptor. Journal of Biological Chemistry 266: 2048-53

Malosio ML, Marquèze-Pouey B, Kuhse J, Betz H. 1991b. Widespread expression of glycine receptor subunit mRNAs in the adult and developing rat brain. The EMBO Journal 10: 2401-09

Mangin JM, Baloul M, Prado de Carvalho L, Rogister B, Rigo JM, Legendre P. 2003. Kinetic properties of the α(2) homo-oligomeric glycine receptor impairs a proper synaptic functioning. The Journal of Physiology 553: 369-86

[56]

Mascia MP, Mihic SJ, Valenzuela CF, Schofield PR, Harris RA. 1996. A single amino acid determines differences in ethanol actions on strychnine-sensitive glycine receptors. Molecular Pharmacology 50: 402-06

McCracken LM, Blednov YA, Trudell JR, Benavidez JM, Betz H, Harris RA. 2013a. Mutation of a Zinc-Binding Residue in the Glycine Receptor α1 Subunit Changes Ethanol Sensitivity In Vitro and Alcohol Consumption In Vivo. The Journal of Pharmacology and Experimental Therapeutics 344: 489-500

McCracken LM, Lowes DC, Salling MC, Carreau-Vollmer C, Odean NN, et al. 2017. Glycine receptor α3 and α2 subunits mediate tonic and exogenous agonist-induced currents in forebrain. Proceedings of the National Academy of Sciences 114: E7179-E86

McCracken LM, McCracken ML, Gong DH, Trudell JR, Harris RA. 2010a. Linking of Glycine Receptor Transmembrane Segments Three and Four Allows Assignment of Intrasubunit-Facing Residues. ACS Chemical Neuroscience 1: 482-94

McCracken LM, Trudell JR, Goldstein BE, Harris RA, Mihic SJ. 2010b. Zinc Enhances Ethanol Modulation of the α1 Glycine Receptor. Neuropharmacology 58: 676-81

McCracken LM, Trudell JR, McCracken ML, Harris RA. 2013b. Zinc-Dependent Modulation of α2- and α3-Glycine Receptor Subunits by Ethanol. Alcoholism, clinical and experimental research 37: 2002-10

McDearmid JR, Liao M, Drapeau P. 2006. Glycine receptors regulate interneuron differentiation during spinal network development. Proceedings of the National Academy of Sciences 103: 9679-84

McGregor IS, Bowen MT. 2012. Breaking the loop: Oxytocin as a potential treatment for drug addiction. Hormones and Behavior 61: 331-39

Meier J, Meunier-Durmort C, Forest C, Triller A, Vannier C. 2000. Formation of glycine receptor clusters and their accumulation at synapses. J Cell Sci 113 ( Pt 15): 2783-95

Meyer G, Kirsch J, Betz H, Langosch D. 1995. Identification of a gephyrin binding motif on the glycine receptor beta subunit. Neuron 15

Mihic SJ. 1999. Acute effects of ethanol on GABAA and glycine receptor function. Neurochemistry International 35: 115-23

Mihic SJ, Ye Q, Wick MJ, Koltchine VV, Krasowski MD, et al. 1997. Sites of alcohol and volatile anaesthetic action on GABA(A) and glycine receptors. Nature 389: 385-9

Miledi R, Woodward RM. 1989. Membrane currents elicited by prostaglandins, atrial natriuretic factor and oxytocin in follicle-enclosed Xenopus oocytes. The Journal of Physiology 416: 623-43

Miller PS, Harvey RJ, Smart TG. 2004. Differential agonist sensitivity of glycine receptor α2 subunit splice variants. British Journal of Pharmacology 143: 19-26

Mitchell IJ, Gillespie SM, Abu-Akel A. 2015. Similar effects of intranasal oxytocin administration and acute alcohol consumption on socio-cognitions, emotions and behaviour: Implications for the mechanisms of action. Neuroscience & Biobehavioral Reviews 55: 98-106

Mizuno A, Cherepanov SM, Kikuchi Y, Fakhrul AAKM, Akther S, et al. 2015. Lipo-oxytocin-1, a Novel Oxytocin Analog Conjugated with Two Palmitoyl Groups, Has Long-Lasting Effects on Anxiety-Related Behavior and Social Avoidance in CD157 Knockout Mice. Brain Sciences 5: 3-13

[57]

Mody I, Glykys J, Wei W. 2007. A new meaning for “Gin & Tonic”: tonic inhibition as the target for ethanol action in the brain. Alcohol 41: 145-53

Molander A, Lido HH, Lof E, Ericson M, Soderpalm B. 2007. The glycine reuptake inhibitor Org 25935 decreases ethanol intake and preference in male wistar rats. Alcohol Alcohol 42: 11-8

Molander A, Lof E, Stomberg R, Ericson M, Soderpalm B. 2005. Involvement of accumbal glycine receptors in the regulation of voluntary ethanol intake in the rat. Alcoholism, clinical and experimental research 29: 38-45

Muller E, Le-Corronc H, Legendre P. 2008. Extrasynaptic and Postsynaptic Receptors in Glycinergic and GABAergic Neurotransmission: A Division of Labor? Frontiers in Molecular Neuroscience 1: 3

Naito A, Muchhala KH, Asatryan L, Trudell JR, Homanics GE, et al. 2014. Glycine and GABA_A Ultra-Sensitive Ethanol Receptors as Novel Tools for Alcohol and Brain Research. Molecular Pharmacology 86: 635-46

Naito A, Muchhala KH, Trang J, Asatryan L, Trudell JR, et al. 2015. Manipulations of extracellular Loop 2 in α1 GlyR ultra-sensitive ethanol receptors (USERs) enhance receptor sensitivity to isoflurane, ethanol, and lidocaine, but not propofol. Neuroscience 297: 68-77

Neumann ID, Landgraf R. 2012. Balance of brain oxytocin and vasopressin: implications for anxiety, depression, and social behaviors. Trends in Neurosciences 35: 649-59

Neumann ID, Maloumby R, Beiderbeck DI, Lukas M, Landgraf R. 2013. Increased brain and plasma oxytocin after nasal and peripheral administration in rats and mice. Psychoneuroendocrinology 38: 1985-93

Nusser Z, Mody I. 2002. Selective modulation of tonic and phasic inhibitions in dentate gyrus granule cells. Journal of neurophysiology 87: 2624-8

O'Connor CM, Smith LD. 1976. Inhibition of oocyte maturation by theophylline: possible mechanism of action. Dev Biol 52: 318-22

Olsen RW, Hanchar HJ, Meera P, Wallner M. 2007. GABA(A) Receptor Subtypes: the ‘One Glass of Wine’ Receptors. Alcohol (Fayetteville, N.Y.) 41: 201-09

Papke RL, Smith-Maxwell C. 2009. High-throughput electrophysiology with Xenopus oocytes. Combinatorial chemistry & high throughput screening 12: 38-50

Patrizio A, Renner M, Pizzarelli R, Triller A, Specht CG. 2017. Alpha subunit-dependent glycine receptor clustering and regulation of synaptic receptor numbers. Sci Rep 7: 10899

Pedersen CA, Smedley KL, Leserman J, Jarskog LF, Rau SW, et al. 2013. Intranasal Oxytocin Blocks Alcohol Withdrawal in Human Subjects. Alcoholism: Clinical and Experimental Research 37: 484-89

Perkins DI, Trudell JR, Crawford DK, Alkana RL, Davies DL. 2008. Targets for ethanol action and antagonism in loop 2 of the extracellular domain of glycine receptors. J Neurochem 106: 1337- 49

Perkins DI, Trudell JR, Crawford DK, Alkana RL, Davies DL. 2010. Molecular targets and mechanisms for ethanol action in glycine receptors. Pharmacol Ther 127: 53-65

[58]

Perkins DI, Trudell JR, Crawford DK, Asatryan L, Alkana RL, Davies DL. 2009. Loop 2 Structure in Glycine and GABA(A) Receptors Plays a Key Role in Determining Ethanol Sensitivity. The Journal of biological chemistry 284: 27304-14

Peters ST, Bowen MT, Bohrer K, McGregor IS, Neumann ID. 2017. Oxytocin inhibits ethanol consumption and ethanol-induced dopamine release in the nucleus accumbens. Addiction biology 22: 702-11

Pribilla I, Takagi T, Langosch D, Bormann J, Betz H. 1992. The atypical M2 segment of the beta subunit confers picrotoxinin resistance to inhibitory glycine receptor channels. The EMBO Journal 11: 4305-11

Rae CD, Davidson JE, Maher AD, Rowlands BD, Kashem MA, et al. 2014. Ethanol, not detectably metabolized in brain, significantly reduces brain metabolism, probably via action at specific GABA(A) receptors and has measureable metabolic effects at very low concentrations. J Neurochem 129: 304-14

Ross HE, Young LJ. 2009. Oxytocin and the neural mechanisms regulating social cognition and affiliative behavior. Frontiers in Neuroendocrinology 30: 534-47

San Martin L, Cerda F, Jimenez V, Fuentealba J, Munoz B, et al. 2012. Inhibition of the ethanol- induced potentiation of alpha1 glycine receptor by a small peptide that interferes with Gbetagamma binding. The Journal of biological chemistry 287: 40713-21

Santhakumar V, Wallner M, Otis TS. 2007. Ethanol acts directly on extrasynaptic subtypes of GABAA receptors to increase tonic inhibition. Alcohol 41: 211-21

Sarnyai Z, Kovács GL. 2014. Oxytocin in learning and addiction: From early discoveries to the present. Pharmacology Biochemistry and Behavior 119: 3-9

Sauguet L, Howard RJ, Malherbe L, Lee US, Corringer P-J, et al. 2013. Structural basis for potentiation by alcohols and anaesthetics in a ligand-gated ion channel. Nature communications 4: 1697-97

Shibasaki K, Hosoi N, Kaneko R, Tominaga M, Yamada K. 2017. Glycine release from astrocytes via functional reversal of GlyT1. Journal of Neurochemistry 140: 395-403

Sieghart W, Sperk G. 2002. Subunit composition, distribution and function of GABA(A) receptor subtypes. Current topics in medicinal chemistry 2: 795-816

Simon J, Wakimoto H, Fujita N, Lalande M, Barnard EA. 2004. Analysis of the Set of GABAA Receptor Genes in the Human Genome. Journal of Biological Chemistry 279: 41422-35

Soderpalm B, Lido HH, Ericson M. 2017. The glycine receptor - a functionally important primary brain target of ethanol. Alcoholism, clinical and experimental research

Spanagel R. 2009. Alcoholism: a systems approach from molecular physiology to addictive behavior. Physiol Rev 89: 649-705

Striepens N, Kendrick KM, Hanking V, Landgraf R, Wüllner U, et al. 2013. Elevated cerebrospinal fluid and blood concentrations of oxytocin following its intranasal administration in humans. Scientific Reports 3: 3440

Sundstrom-Poromaa I, Smith DH, Gong QH, Sabado TN, Li X, et al. 2002. Hormonally regulated α(4)β(2)δ GABA(A) receptors are a target for alcohol. Nature neuroscience 5: 721-22

[59]

Sur C, Farrar SJ, Kerby J, Whiting PJ, Atack JR, McKernan RM. 1999. Preferential coassembly of alpha4 and delta subunits of the gamma-aminobutyric acidA receptor in rat thalamus. Mol Pharmacol 56: 110-5

Takagi T, Pribilla I, Kirsch J, Betz H. 1992. Coexpression of the receptor-associated protein gephyrin changes the ligand binding affinities of α2 glycine receptors. FEBS letters 303: 178-80

Turecek R, Trussell LO. 2001. Presynaptic glycine receptors enhance transmitter release at a mammalian central synapse. Nature 411: 587-90

Tziridis K, Buerbank S, Eulenburg V, Dlugaiczyk J, Schulze H. 2017. Deficit in acoustic signal-in- noise detection in glycine receptor α3 subunit knockout mice. European Journal of Neuroscience 45: 581-86

Urso T, Gavaler JS, Van Thiel DH. 1981. Blood ethanol levels in sober alcohol users seen in an emergency room. Life Sciences 28: 1053-56

Vengeliene V, Leonardi-Essmann F, Sommer WH, Marston HM, Spanagel R. 2010. Glycine Transporter-1 Blockade Leads to Persistently Reduced Relapse-like Alcohol Drinking in Rats. Biological Psychiatry 68: 704-11

Vonghia L, Leggio L, Ferrulli A, Bertini M, Gasbarrini G, Addolorato G. 2008. Acute alcohol intoxication. European Journal of Internal Medicine 19: 561-67

Wallner M, Hanchar HJ, Olsen RW. 2003. Ethanol enhances α(4)β(3)δ and α(6)β(3)δ γ- aminobutyric acid type A receptors at low concentrations known to affect humans. Proceedings of the National Academy of Sciences of the United States of America 100: 15218-23

Wallner M, Hanchar HJ, Olsen RW. 2006a. Low-dose alcohol actions on α4β3δ GABA(A) receptors are reversed by the behavioral alcohol antagonist Ro15-4513. Proceedings of the National Academy of Sciences of the United States of America 103: 8540-45

Wallner M, Hanchar HJ, Olsen RW. 2006b. Low dose acute alcohol effects on GABA(A) receptor subtypes. Pharmacology & therapeutics 112: 513-28

Wallner M, Hanchar HJ, Olsen RW. 2014. Alcohol Selectivity of β3-Containing GABAA Receptors: Evidence for a Unique Extracellular Alcohol/Imidazobenzodiazepine Ro15-4513 Binding Site at the α+β- Subunit Interface in αβ3δ GABAA Receptors. Neurochemical Research 39: 1118-26

Wallner M, Olsen RW. 2008. Physiology and pharmacology of alcohol: the imidazobenzodiazepine alcohol antagonist site on subtypes of GABA(A) receptors as an opportunity for drug development? British Journal of Pharmacology 154: 288-98

Wang D-S, Mangin J-M, Moonen G, Rigo J-M, Legendre P. 2006. Mechanisms for Picrotoxin Block of 2 Homomeric Glycine Receptors. 3841-55 pp.

Weber W-M. 1999. Ion currents of Xenopus laevis oocytes: state of the art. Biochimica et Biophysica Acta (BBA) - Biomembranes 1421: 213-33

Wei W, Faria LC, Mody I. 2004. Low Ethanol Concentrations Selectively Augment the Tonic Inhibition Mediated by δ Subunit-Containing GABAA Receptors in Hippocampal Neurons. The Journal of Neuroscience 24: 8379-82

Wetherill RR, Fromme K. 2011. Acute alcohol effects on narrative recall and contextual memory: an examination of fragmentary blackouts. Addictive behaviors 36: 886-9

[60]

Williams KL, Ferko AP, Barbieri EJ, DiGregorio GJ. 1995. Glycine enhances the central depressant properties of ethanol in mice. Pharmacology Biochemistry and Behavior 50: 199-205

Williams L, Parker H. 2001. Alcohol, cannabis, ecstasy and cocaine: drugs of reasoned choice amongst young adult recreational drug users in England. International Journal of Drug Policy 12: 397-413

Wisden W, Laurie DJ, Monyer H, Seeburg PH. 1992. The distribution of 13 GABAA receptor subunit mRNAs in the rat brain. I. Telencephalon, diencephalon, mesencephalon. J Neurosci 12: 1040-62

Wyttenbach T, Liu D, Bowers MT. 2008. Interactions of the Hormone Oxytocin with Divalent Metal Ions. Journal of the American Chemical Society 130: 5993-6000

Xiong W, Chen S-R, He L, Cheng K, Zhao Y-L, et al. 2014. Presynaptic glycine receptors as a potential therapeutic target for hyperekplexia disease. Nature neuroscience 17: 232-39

Ye J-H, Wang F, Krnjević K, Wang W, Xiong Z-G, Zhang J. 2004. Presynaptic Glycine Receptors on GABAergic Terminals Facilitate Discharge of Dopaminergic Neurons in Ventral Tegmental Area. The Journal of Neuroscience 24: 8961-74

Ye JH, Sokol KA, Bhavsar U. 2009. Glycine Receptors Contribute to Hypnosis Induced by Ethanol. Alcoholism: Clinical and Experimental Research 33: 1069-74

Ye Q, Koltchine VV, Mihic SJ, Mascia MP, Wick MJ, et al. 1998. Enhancement of Glycine Receptor Function by Ethanol Is Inversely Correlated with Molecular Volume at Position α267. Journal of Biological Chemistry 273: 3314-19

Yevenes GE, Moraga-Cid G, Peoples RW, Schmalzing G, Aguayo LG. 2008. A selective Gβγ-linked intracellular mechanism for modulation of a ligand-gated ion channel by ethanol. Proceedings of the National Academy of Sciences 105: 20523-28

You H, Dunn SMJ. 2007. Identification of a domain in the δ subunit (S238-V264) of the α4β3δ GABAA receptor that confers high agonist sensitivity. Journal of Neurochemistry 103: 1092-101

Young LJ, Huot B, Nilsen R, Wang Z, Insel TR. 1996. Species differences in central oxytocin receptor gene expression: comparative analysis of promoter sequences. J Neuroendocrinol 8: 777-83

Zhang LH, Gong N, Fei D, Xu L, Xu TL. 2008. Glycine uptake regulates hippocampal network activity via glycine receptor-mediated tonic inhibition. Neuropsychopharmacology 33: 701-11

Zheleznova NN, Sedelnikova A, Weiss DS. 2009. Function and modulation of δ-containing GABA(A) receptors. Psychoneuroendocrinology 34S1: S67-S73

Zuo Y, Nagata K, Yeh JZ, Narahashi T. 2004. Single-Channel Analyses of Ethanol Modulation of Neuronal Nicotinic Acetylcholine Receptors. Alcoholism: Clinical and Experimental Research 28: 688-96

[61]