MCLELLAND Angela Estelle

An investigation into the antidyskinetic and reinforcing effects of UWA-101 in Sprague Dawley rats

Angela Estelle McLelland, B.Sc. (Hons), M.D.

This thesis is presented for the degree of Doctor of Philosophy of The University of Western Australia

School of Medicine and Pharmacology Pharmacology and Anaesthesiology Unit

2019

Thesis Declaration

I, Angela McLelland, certify that:

This thesis has been substantially accomplished during enrolment in this degree.

This thesis does not contain material which has been submitted for the award of any other degree or diploma in my name, in any university or other tertiary institution.

In the future, no part of this thesis will be used in a submission in my name, for any other degree or diploma in any university or other tertiary institution without the prior approval of The University of Western Australia and where applicable, any partner institution responsible for the joint-award of this degree.

This thesis does not contain any material previously published or written by another person, except where due reference has been made in the text and, where relevant, in the Authorship Declaration that follows.

This thesis does not violate or infringe any copyright, trademark, patent, or other rights whatsoever of any person.

The research involving animal data reported in this thesis was assessed and approved by The University of Western Australia Animal Ethics Committee. Approval RA/3/100/1165, RA/3/100/1182, RA/3/100/1178. The research involving animals reported in this thesis followed The University of Western Australia and national standards for the care and use of laboratory animals.

This thesis contains published work and/or work prepared for publication, some of which has been co-authored.

Signature:

Date: 16/08/19

Abstract

Long-term treatment of people with Parkinson’s disease (PD) with the gold-standard drug, L-3,4-dihydroxyphenylalanine (L-DOPA), often results in debilitating excessive involuntary movements known as L-DOPA induced dyskinesia (LID). There is limited treatment for LID, with only one drug, amantadine, currently licensed for this purpose, and its long-term efficacy is questionable. Research into pharmacological treatment of LID has included the investigation of MDMA-analogue UWA-101, which has affinity for the transporters of dopamine and serotonin, but not noradrenaline.

The first aim of this thesis was to investigate the antidyskinetic and reinforcing effects of increasing doses of UWA-101 in Sprague-Dawley rats in three separate studies.

Two of these studies involved the reserpinised rat model, and a second aim was to determine the potential of this model as an early screening procedure for LID drug discovery, prior to more extensive lesion models. This procedure serves to induce PD-like symptoms such as akinesia in a short time-frame through the administration of reserpine, which depletes monoamine neurotransmitter stores via vesicular monoamine transport 2 (VMAT2) inhibition. In this model, it has been proposed that rat vertical movements are analogous to human dyskinesia, whereas reductions in horizontal activity (akinesia) reflect the loss of dopamine and mimic PD symptoms.

UWA-101 is an analogue of the illicit drug MDMA ('ecstasy'), so the third aim was to determine abuse liability or aversive subjective effects of UWA-101 that may reduce its treatment utility with three separate measurements: the conditioned place preference test, as a

5 measure of the reinforcing or aversive effects of a drug,; the elevated plus maze test, as a measure of anxiogenic effects of a drug; and ultrasonic vocalisations (USV), as a measure of both hedonic (50 kHz) and aversive (20 kHz) effects of a drug.

Using custom-made locomotion detection equipment, UWA-101 (1 and 3 mg/kg) significantly reduced L-DOPA-induced hyperkinesia and behavioural dyskinesia correlates in reserpinised rats. Yohimbine, riluzole, and rimonabant, antidyskinetic compounds used to assess the validity of the reserpinised rat model, also significantly reduced L-DOPA-induced hyperkinesia and behavioural dyskinesia correlates. It can be concluded that the reserpinised rat may serve as an early and relatively quick screening model prior to more extensive lesion models.

Treatment of reserpinised rats with UWA-101 (3 mg/kg) was then preceded by treatment with drugs of known pharmacological affinities to assess the underlying mechanism of the anti-hyperkinetic effect of UWA-101. Fluoxetine (a serotonin transporter

[SERT] inhibitor), WAY-100635 (a 5-HT1A receptor antagonist) and riluzole (a glutamate release inhibitor) all blocked UWA-101 reversal of L-DOPA-induced hyperkinesia and dyskinesia correlates. These observations indicate that UWA-101 reversal of L-DOPA induced hyperkinesia may occur through reversal of SERT function, as does MDMA, and through activation of the 5-HT1A receptor, a process that is also dependent on glutamate release.

UWA-101 (3 mg/kg) did not produce conditioned place preferences or aversions in the CPP test, changes in USVs or anxiolytic effects in the EPM test. In contrast, dexamphetamine (1.5 mg/kg) as a positive control, induced a robust place preference, increased 50 kHz USVs and increased open arm activity in the EPM test. Therefore, UWA-

101 did not appear to have abuse liability or to produce aversive side-effects at this dose.

UWA-101 reduced L-DOPA induced hyperkinesia in line with earlier lesion studies in rats and primates. Similar effects were induced by other drugs shown to have antidyskinetic activities in non-human lesioned animal models, including yohimbine, riluzole and rimonabant, further supporting the utility of the reserpinised model. These findings support the view that multiple neurotransmitters other than dopamine may be involved in the development or expression of L-DOPA induced dyskinetic behaviour. Furthermore, the mechanisms underlying the anti-hyperkinetic effect of UWA-101 seem to involve the serotonergic system.

In conclusion, the reserpinised rat model is useful as an early screen for antidyskinetic compounds, prior to more expensive and time-consuming lesion studies. The findings with

UWA-101 support earlier research in such lesion studies indicating its potential utility as a therapeutic agent for LID without evidence of abuse liability or aversive side-effects, and provides the first evidence that UWA-101, like MDMA, reverses the SERT.

Table of Contents

Abstract……………………………………………………………………………………….5 List of Tables………………………………………………………………………………...15 List of Figures……………………………………………………………………………….17 List of Abbreviations………………………………………………………………………..21 Acknowledgements………………………………………………………………………….23 Authorship declaration: Co-authored publications………………………………..……..25 Statement of candidate contribution………………………………………………………27

Chapter 1 – General Introduction…………………………………………………………29 1. Introduction...... 29 1.1 Development of L-3,4-dihydroxyphenylalanine (L-DOPA)-induced dyskinesia (LID) ...... 31 1.1.1 Overview of the basal ganglia...... 31 1.1.2 Factors involved in LID development...... 36 1.1.2.1 Hypothesis of LID emergence...... 36 1.1.2.2 Delaying the onset of LID...... 38 1.1.2.3 Modifications of L-DOPA in established LID...... 41 1.2 Pharmacological treatment of LID...... 43 1.2.1 Dopamine receptors...... 43 1.2.1.1 D1-like receptors...... 43 1.2.1.2 D2-like receptors...... 45 1.2.2 Serotonin receptors...... 53 1.2.2.1 5-HT1A/b receptors...... 54 1.2.2.2 5-HT2A receptors...... 58 1.2.3 Monoamine transporters...... 60 1.2.4 Glutamate receptors...... 64 1.2.4.1 N-methyl-D-aspartate (NMDA) receptors...... 64 1.2.4.2 Alpha-amine-3-hydroxyl-methyl-4-isoxazleproprionic acid (AMPA) receptors...... 67 1.2.4.3 Metabotropic receptors...... 69 1.2.5 Adrenergic receptors...... 73 1.2.6 Cannabinoid receptors...... 77 1.3 Animal models of LID...... 80 1.3.1 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated primate...... 81 1.3.2 6-hydroxydopamine (6-OHDA) lesion models...... 86 1.3.2.1 6-OHDA rat...... 86 1.3.2.2 6-OHDA mouse...... 91 1.3.3 Reserpinised rat...... 95 1.3.4 Genetically modified mouse...... 99 1.4 Animal models of reward and anxiety...... 104

1.4.1 Conditioned place preference (CPP) ...... 107 1.4.2 Ultrasonic vocalisation (USVs) ...... 110 1.4.2.1 50 kHz USVs...... 112 1.4.2.2 22 and 40 kHz USVs...... 115 1.4.3 Elevated plus maze (EPM)...... 120 1.5 Experimental aims...... 124

Chapter 2 – An inexpensive alternative to commercial locomotion detection systems 2. Preface...... 127 2.1 Abstract...... 127 2.2 Introduction...... 128 2.3 Methods...... 130 2.3.1 Construction...... 130 2.3.1.1 IR emitter/sensor pairs...... 130 2.3.1.2 Housing...... 132 2.3.1.3 Attachment...... 133 2.3.2 Computer and software...... 135 2.3.3 Experimental procedures...... 136 2.3.3.1 Animals...... 136 2.3.3.2 Drugs...... 137 2.3.3.3 Testing...... 137 2.3.4 Statistical analysis...... 139 2.4 Results...... 139 2.4.1 Data collection...... 139 2.4.2 Monetary cost of equipment...... 141 2.4.3 Behavioural and locomotion data...... 142 2.5 Discussion...... 144

Chapter 3 – An inexpensive alternative to commercial prudcts in the recording of ultrasonic vocalisations...... 147 3. Preface...... 147 3.1 Abstract...... 147 3.2 Introduction...... 148 3.3 Methods...... 150 3.3.1 Microphones...... 150 3.3.2 Recording system...... 150 3.3.3 Validation...... 151 3.3.4 Animals...... 151 3.3.5 Apparatus...... 152 3.3.6 Drugs...... 152 3.3.7 Procedure...... 152 3.3.8 USV recording analysis...... 153 3.4 Results...... 153 3.4.1 Artificial ultrasonic emission...... 153

3.4.2 Ultrasonic vocalisations...... 154 3.4.2.1 22 kHz ultrasonic vocalisations...... 154 3.4.2.2 50 kHz ultrasonic vocalisations...... 155 3.4.3 Cost of equipment...... 156 3.5 Discussion...... 157 Chapter 4 – Influence of antidyskinetic compounds on L-DOPA-induced locomotion and behaviour in reserpine-treated rats...... 159 4. Preface...... 159 4.1 Abstract...... 159 4.2 Introduction...... 160 4.3 Methods...... 162 4.3.1 Animals...... 162 4.3.2 Drugs...... 163 4.3.3 Apparatus...... 163 4.3.4 Procedure...... 164 4.3.5 Statistical analysis...... 165 4.4 Results...... 166 4.4.1 Effect of reserpine on healthy rats...... 166 4.4.2 Effect of L-DOPA on reserpinised rats treated with vehicle or antidyskinetic drugs...... 167 4.4.2.1 Time contrasts...... 167 4.4.2.2 Test...... 169 4.4.2.3 Effect of antidyskinetic drugs on L-DOPA-treated reserpinised rats...... 171 4.4.3 Effect of antidyskinetic drugs on reserpinised rats...... 172 4.5 Discussion...... 174

Chapter 5 – Reduction of L-DOPA-induced hyperkinesia by MDMA-analogue, UWA- 101, in the reserpine-treated rats is attenuated by the 5-HT1A receptor, the serotonin transporter, or inhibition of glutamate release...... 181 5. Preface...... 181 5.1 Abstract...... 181 5.2 Introduction...... 182 5.3 Methods...... 185 5.3.1 Animals...... 185 5.3.2 Drugs...... 185 5.3.3 Apparatus...... 186 5.3.4 Procedure...... 186 5.3.5 Statistical analysis...... 187 5.4 Results...... 188 5.4.1 Effect of reserpine on healthy rats...... 188 5.4.2 Effect of L-DOPA on reserpine-treated rats...... 189 5.4.3 Effect of UWA-101 on reserpine-treated rats administered L-DOPA...... 189 5.4.4 Effect of experimental drugs on UWA-10-induced effects in rats...... 189 5.4.4.1 Yohimbine and riluzole...... 189

5.4.4.2 WAY 100635...... 190 5.4.4.3 Fluoxetine...... 192 5.5 Discussion...... 192

Chapter 6 – Rewarding and anxiolytic effects of the MDMA-analogue UWA-101 in healthy Sprague Dawley rats...... 201 6. Preface...... 201 6.1 Abstract...... 201 6.2 Introduction...... 202 6.3 Methods...... 204 6.3.1 Animals...... 204 6.3.2 Drugs...... 205 6.3.3 Conditioned place preference (CPP) ...... 205 6.3.4 Ultrasonic vocalisations (USVs) ...... 206 6.3.5 Elevated plus maze test (EPM) ...... 207 6.3.6 Data analysis...... 208 6.4 Results...... 209 6.4.1 CPP...... 209 6.4.1.1 Preconditioning...... 209 6.4.1.2 Drug-free test...... 209 6.4.1.3 State-dependent test...... 210 6.4.2 USVs...... 210 6.4.2.1 Audio recordings...... 210 6.4.2.2 Behaviour...... 211 6.4.3 EPM...... 212 6.5 Discussion...... 214

Chapter 7 – General Discussion...... 223 7. Summary of experimental aims...... 223 7.1 Significance of findings...... 223 7.1.1 Reserpinised rat model of LID...... 223 7.1.1.1 Custom-made locomotion detection systems...... 223 7.1.1.2 Validation of the reserpinised rat as a model of LID...... 224 7.1.1.3 The mechanism underlying the anti-hyperkinetic effect of UWA- 101...... 227 7.1.2 Reinforcing effects of UWA-101 in reward and anxiety models...... 228 7.1.2.1 Custom-made ultrasonic microphones...... 228 7.1.2.2 The effects of UWA-101 on reward and anxiety...... 229 7.2 Discussion of findings...... 230 7.2.1 Custom-made locomotion detection systems...... 230 7.2.2 The reserpinised rat as a model of LID...... 231 7.2.3 Anti-hyperkinetic mechanism of UWA-101...... 236 7.2.4 Custom-made ultrasonic microphones...... 238 7.2.5 Reinforcing effects of UWA-101...... 239

7.3 Improvements and future directions...... 242 7.3.1 Custom-made locomotion detection system...... 242 7.3.2 Validation of the reserpinised rat model of LID...... 243 7.3.3 Anti-hyperkinetic mechanism of UWA-101...... 244 7.3.4 Custom-made ultrasonic microphones...... 245 7.3.5 Reinforcing effects of UWA-101...... 246 7.3.6 Overall future directions...... 247 7.4 Conclusions...... 247

References...... 249 Appendix I...... 285

List of Tables Table 1.1. Common genetic mouse models based on familial PD gene mutations and/or displaying phenotypes reminiscent of PD pathogenesis or progression.

Table 2.1. A comparison of the prices between IR-based rodent locomotor activity monitoring systems from the current study and commercial suppliers. Three separate quotes were obtained for comparison with a price per unit (activity monitoring for one rat) with and without rodent boxes and for five units (activity monitoring for five rats) with and without rodent enclosures. All prices in USD.

Table 4.1. Mean (SD) and Student’s t test results for all behavioural measurements between saline- and reserpine (4 mg/kg)-treated Sprague-Dawley rats.

Table 4.2. Percentages in L-DOPA-induced horizontal movements (HM), elevation time (ET), elevation bout (EB), repetitive movements while elevated (RME) and non-repetitive movements while elevated (NRME) after exposure to antidyskinetic drugs in reserpine- treated Sprague Dawley rats.

Table 4.3. Ratios in L-DOPA-induced horizontal movements (HM), elevation time (ET), elevation bout (EB), repetitive movements while elevated (RME) and non-repetitive movements while elevated (NRME) after exposure to antidyskinetic drugs in reserpine- treated Sprague Dawley rats.

Table 5.1. Mean (+ SD) of all measurements during habituation session between saline- and reserpine-treated Sprague Dawley rats.

Table 6.1. Time spent in drug- or saline-paired chambers of conditioned place preference boxes during three daily 15-min preconditioning sessions

Table 6.2. Mean (+ SEM) of additional behaviours measured during the EPM test between saline, dexamphetamine and UWA-101 groups. (+)-Amph = dexamphetamine.

List of Figures Figure 1.1. The classic model of the basal ganglia in a A) normal state, B) state of progressive Parkinson’s disease in which the substantia nigra pars compacta is degenerating leading to overcompensation of the input from the subthalamic nucleus (STN) to the internal globus pallidus (GPi) and thus ‘overactivity’ in the indirect pathway, C) late stage of

Parkinson’s Disease in which the substantia nigra pars compacta has degenerated to the degree where L-DOPA treatment is necessary resulting in ‘overactivity’ of the direct pathway and consequently, involuntary movements termed dyskinesia. Glu = Glutamate projections,

GABA = γ-amino-butyric acid (GABA) projections.

Figure 1.2. A recent model of voluntary movements executed by the basal ganglia incorporating additional pathways between neural structures. Str = Striatum, STN =

Subthalamic nucleus, GPe = Globus pallidus pars externa, GPi = Globus pallidus pars interna, SNr, Substantia nigra reticulata, GABA = GABA-ergic projections, Glu =

Glutamatergic projections, DA = Dopaminergic projections. Green lines highlight the projection of the hyperdirect pathway, pink lines highlight the projection of the indirect pathway and blue lines highlight the projection of the direct pathway.

Figure 2. Schematic diagram demonstrating the effect of housing IR LEDS. IR light emitted from LEDs (light grey) in the animal enclosures A) overlapping across three different sensors

(light grey) due to wide emission field and B) targeting aligned sensor (dark grey) after being housed to narrow the IR beam and reduce emission field covered.

Figure 2.2. Mean Locomotor activity + SEM of Sprague-Dawley rats treated with saline (n =

10) or L-DOPA (n = 10) A) Horizontal activity measured as number of beam breaks per 5 minute bin B) Immobile time measured as time (seconds) when the animal remained motionless per 5 minute bin C) Elevation time measured as length (seconds) of beam break in the top level of emitters and sensors only per 5 min bin D) Elevation bout measured as length

(seconds) of upper level beam break per elevation movement per 5 minute bin E) Elevated

17 movement measured as number of beam breaks located in the top level of IR emitters and sensors only per 5 minute bin F) Centre elevated movement measured as the number of beam breaks located in the top level of IR emitters and sensors while located in the centre of the enclosure floor per 5 minute bin G) Repeated movement while elevated as measurement by repetitive horizontal movement during upper beam break and H) Non-repeated movement while elevated as measured as remaining horizontal movement during upper beam beak during testing session in enclosures with transecting unidirectional IR beams using IR emitters and sensors. * Significantly difference compared to saline group (p < 0.05).

Figure 2.3. Mean locomotor activity + SEM of Sprague Dawley rats treated with saline (n =

10) or L-DOPA (n = 10) over total testing session. A) Horizontal activity B) Immobile time

C) Elevation time measured D) Elevation bout E) Elevated movement F) Centre elevated movement G) Repeated movement while elevated H) Non-repeated movement while elevated.

* Significantly difference compared to saline group (p < 0.05).

Figure 3.1. Recording of an ultrasonic tone emitted by speakers from the PC on A)

Panasonic microphone electrets and B) a commercial bat detector after transformation of ultrasonic input by a factor of 10.

Figure 3.2. Recording of a 40 kHz signal from an ultrasonic transducer on a Panasonic microphone.

Figure 3.3. Comparison of 22 kHz USVs between A) custom-made Panasonic microphones and B) a commercial bat detector. Both microphones recorded 22 kHz accurately and clearly, although the bat detector automatically reduces the frequency of ultrasonic USVs by a factor of 10 as indicated by the frequency axis.

Figure 3.4. Recording of 50 kHz A) trill call and B) flat call from a commercial bat detector emitted by a dexamphetamine-treated rat. Internal circuitry of the bat detector translated the input frequency by a factor of 10 during spectrogram analysis.

Figure 4.1. Time effects on horizontal movements in reserpinised male Sprague-Dawley rats treated with A) Saline only B) L-DOPA and saline or UWA-101 C) L-DOPA and saline, yohimbine, riluzole or rimonabant (note, only one group of L-DOPA + saline was used for figures B and C).

Figure 4.2. Effect of L-DOPA and antidyskinetic drugs on A) horizontal movements B) elevation time C) elevation bout D) elevated movements E) EMC F) repetitive movements while elevated G) non-repetitive movements while elevated. *p < 0.05 vs. Res/Veh/Veh group

^p < 0.05 vs. Res/L-DOPA/Veh group.

Figure 4.3. Effect of various antidyskinetic drugs on A) horizontal movements B) elevation time C) elevation bout D) elevated movements E) EMC F) repetitive movements while elevated and G) non-repetitive movements while elevated in reserpinised rats 30 m before administration of L-DOPA.*p < 0.05 vs. Res/Veh group.

Figure 5.1. Effect of various experimental drugs on the reduction of L-DOPA-induced behaviour by UWA-101 (3 mg/kg) in reserpine-treated Sprague Dawley rats. *p < 0.05 vs.

Veh/Veh group, ^p < 0.05 vs. L-DOPA/Veh group, +p < 0.05 vs. L-DOPA/UWA/Veh group.

Figure 6.1. Time spent in the drug-paired chamber of conditioned place preference (CPP) boxes during pre-conditioning testing (baseline), drug-free (DF) testing, and state-dependent

(SD) testing by healthy Sprague Dawley rats.

Figure 6.2. Frequency of ultrasonic vocalisations emitted by dexamphetamine-, UWA-101- or saline-treated Sprague Dawley. (+)-Amph = dexamphetamine

Figure 6.3. Rearing and immobility time during USV testing between saline, dexamphetamine and UWA-101 groups. (+)-Amph = dexamphetamine

Figure 6.4. Comparisons between saline- and drug-treated groups of A) Percentage open arm time (OAT) B) Percentage open arm entries (OAE). (+)-Amph = dexamphetamine

Figure 7.1. Alternating IR emitters (light grey) and sensors (dark grey) along two walls of a locomotion testing box to avoid IR light from emitters overlapping with multiple sensors.

List of Abbreviations

5-HT Serotonin

6-OHDA 6-hydroxydopamine AIMs Abnormal involuntary movements AMPA Αlpha-amino-3-hydroxy-methyl-4-isoxazolepropionic acid

ANOVA Analysis of variance CPP Conditioned place preference

D1, D2, D3, D4, D5 Dopamine receptors, type 1, 2, 3, 4 and 5

DA Dopamine DAT Dopamine transporter df Degrees of freedom EPM Elevated plus maze

GABA Gamma-aminobutyric acid Glu Glutamate kg Kilograms

L-DOPA L-3,4-dihydroxyphenylalanine

LID L-DOPA-induced dyskinesia

MDMA 3,4-Methylenedihydroxymethamphetamine mg Milligrams

MPTP 1-Methyl-4-phenyl-1,2,3,6-tettrehydropyridine

NAT Noradrenaline transporter

NMDA N-methyl-D-aspartate

PD Parkinson’s disease

SD Standard deviation SERT Serotonin transporter

SEM Standard error of the mean

USVs Ultrasonic vocalisations

Acknowledgements

This research was supported by an Australian Governement Research Training

Program (RTP) scholarship.

I am very grateful to my supervisor Mat for his instruction, advice, pizza and wine gatherings and instilling in me a greater depth of understanding of the scientific process.

Also, to my co-supervision Matthew for inventing the UWA-101 compound in the first place and for his unwavering enthusiasm for this project.

I am especially indebted to Carl, who made the study possible by giving up his weekends and weeknights to help build my equipment and write the software code in addition to working full time and studying his own degree. Thank you for your support, patience, and for teaching me how to solder.

Thank you to all the staff at the research facility for caring for the animals and thank you to the animals themselves. Without them, we would have no study.

Further thanks Vivian, Kyran and others who have come and gone over the years for making the lab an enjoyable place to work; and my family for their support and general life help.

Thanks especially to Mum who always believed this thesis would be submitted and who gave me an exceptional start to life.

Authorship declaration: Co-authored publications

This thesis contains work that has been published. Details of the work: McLelland, AE, Winkler, CE, Martin-Iverson MT (2015) A simple and effective method for building inexpensive infrared equipment used to monitor animal locomotion. Journal of Neuroscience Methods: doi: 10.1016/j.jneumeth.2015.01.006 Location in thesis: Chapter 2 Student contribution to work: The thesis author (AM) designed the studies, procured the animals, built the experimental equipment, performed the experiments, collected and processed the data, analysed the data and wrote the manuscripts Co-author signatures and dates: Carl Wink Signature:

Date: 22/0 Mathew Martin-Iverson Signature:

Date: 22/08

Student signature: Date: 16/08/19 I, Mathew Martin-Iverson, certify that the student’s statements regarding their contribution to each of the works listed above are correct. Coordinating supervision signature: Date: 22/08/19

Statement of candidate contribution

This thesis is presented as a collection of papers. For all papers (Chapters 2 – 6) arising from this thesis, the thesis author (AM) designed the studies, procured the animals, built the experimental equipment, performed the experiments, collected and processed the data, analysed the data and wrote the manuscripts.

For all chapters, the principal PhD supervisor (Mathew Martin-Iverson) assisted in designing the studies and provided detailed feedback on the analysis and manuscript drafts.

For all experiments, the co-supervisor (Matthew Piggott) provided the UWA-101 compound being tested, assisted in data interpretation, and provided comments on manuscripts.

Co-author Carl Winkler contributed significantly to the design and building of experimental equipment, designed and wrote the software code for data analysis, and provided contributions to the manuscripts.

Chapter 1 - General Introduction

This thesis is presented as a series of papers investigating the properties of the 3,4 methylenedihydroxymethamphetamine (MDMA) analogue UWA-101 as an antidyskinetic candidate for the treatment of L-3,4-dihydroxyphenylalanine (L-DOPA)-induced dyskinesia

(LID) in Parkinson’s disease (PD) patients. This drug was investigated using the reserpinised rat model of LID assessing L-DOPA-induced hyperkinesia and locomotion, measures previously shown to be behavioural correlates of LID in humans. However, as many prescription drugs are used for non-medical purposes in humans, additional research was conducted to investigate the presence of reinforcing effects by UWA-101 which may compromise its appropriateness for clinical use. Therefore, in addition to the reserpinised rat model of LID, we investigated the effects of UWA-101 in established animal models of reward and anxiety. In particular, measures of conditioned place preference (CPP), ultrasonic vocalisations (USVs) and open arm activity in the elevated plus maze (EPM) were used to investigate the hedonic or anxiogenic effects of UWA-101 in healthy rats.

1 Introduction Parkinson’s Disease (PD) is the second most common neurodegenerative disease, affecting approximately 1-2% of the population over 60 years of age worldwide (Fahn, 2003;

Mayeux, 2003; Tanner & Aston, 2000). PD is characterised pathologically by nigrostriatal dopaminergic pathway degeneration, leading to dopamine deficiency in the striatum and therefore abnormal function of the basal ganglia (Obeso et al., 2000), and the presence of intracellular inclusions known as Lewy bodies. The onset of PD generally occurs after 50 years of age; however, approximately 4% of PD patients are classified as “early-onset” in which clinical symptoms appear earlier (Van Den Eeden, 2003). Symptoms include the development of bradykinesia (slowness of movement), akinesia (loss or impairment of voluntary movements), rigidity (stiffness, resistance, and especially “cogwheel” rigidity

29 which consists of start-and stop movements through the range of motion of a joint similar to a ratchet) in patients’ limbs, hypokinesia (reduced movement amplitude), and resting tremor

(Dauer & Prezedborski, 2003).

One of the most effective treatments for major motor symptoms of PD is a combination of L-3,4-dihydroxyphenylalanine (L-DOPA), the immediate dopamine precursor, and a peripheral aromatic L-amino acid decarboxylase (AADC) inhibitor that inhibits the major enzyme responsible for synthesising dopamine from L-DOPA, as well as serotonin (5- hydroxytryptamine or 5-HT) from its immediate precursor, 5-hydroxytryptophan.

The first and still most common treatment combination for PD is known as Sinemet: a combination of L-DOPA and the peripheral AADC inhibitor carbidopa. After crossing the blood brain barrier, and entering neurons, L-DOPA is decarboxylated by AADC in the residual dopaminergic neurons, and other monoaminergic neurons, thereby restoring depleted striatal dopamine levels and reducing major motor symptoms of PD (Carta et al., 2007; Lloyd et al., 1975; Oertel & Quinn, 1997). During this time, the patient is said to be in an ‘ON’ state. Conversely when the L-DOPA effect wears off and the PD motor symptoms re-emerge, the patient is said to be in ‘OFF’ state.

Although L-DOPA combined with peripheral AADC inhibitors is considered the most effective treatment for alleviating major PD motor symptoms, chronic administration of L-

DOPA can induce dyskinesia – involuntary, purposeless and repetitive movements, particularly in early-onset PD patients (Cenci & Lindgren, 2007; Cotzias et al., 1969). L-

DOPA-induced dyskinesia (LID) occurs in nearly 40% (weighted for N, vs unweighted) of patients within six years of treatment and 87% of patients for over nine years of treatment

(Ahlskog & Muenter, 2001); although, the bulk of these dyskinesias may be clinically significant (for example, they may resolve with a decrease in L-DOPA dose). Interestingly, there was a substantial difference in the time taken to develop dyskinesias in patients who

30 developed parkinsonism before L-DOPA first became available (“pre-L-dopa era” patients) compared to patients who developed parkinsonism subsequent to when L-DOPA first became available (“modern era” patients). The frequency of dyskinesias in “pre-L-DOPA era” patients increased to a median of 50% after 6 months of L-DOPA therapy, while the frequency of dyskinesias by 4-6 years had reached only a median of 38.7% in “modern era” patients (Ahlskog & Muenter, 2001). One explanation for this difference is that “pre-L-

DOPA era” patients would likely have had more advanced Parkinson’s disease at the time of initiating L-DOPA therapy, having developed symptoms prior to L-DOPA being an available therapy with dopamine receptors possibly developing greater sensitivity to fluctuating dopamine levels compared to “modern era” patients. Indeed, within the first 7-12 months of commencing L-DOPA therapy, “pre-L-DOPA era” patients had a median duration of 7.2 years duration of parkinsonism prior to commencing L-DOPA therapy and a LID incidence of 47.5% (unweighted) and 54.8% (weighted) compared to “modern era” patients who had a median duration of 2.0 years duration of parkinsonism prior to commencing L-DOPA therapy and a LID incidence of 7.5% (unweighted) and 7.0% (weighted) (Ahlskog & Muenter, 2001).

However, other researchers have reported reduced incidence of dyskinesias in Parkinson’s disease patients of increasing age (Kumar et al., 2005). Of a cohort comprising 91

Parkinson’s disease patients who developed symptoms from 1976 to 1990, 53% of patients aged 50-59 years developed dyskinesias within 5 years of commencing L-DOPA therapy compared to 14% of patients aged 80-89 years; although, overall 28.5% of the 91 patients had developed dyskinesias within 5 years of L-DOPA treatment. However, these data were reported without statistical analysis, severity and classification of dyskinesias were not described, and duration of parkinsonism prior to commencing L-DOPA therapy was not specified. Any type of motor complication arising from chronic L-DOPA treatment occurs in

50% of patients administered L-DOPA for over five years and almost 100% patients with

31 early-onset PD (Fahn, 2000; Golbe, 1991). Another study reported an incidence of 94% (49 of 52) in patients between 55 to 86 years after 15 years of commencing L-DOPA therapy

(Hely et al., 2005). However, only 12% of the 52 patients demonstrated disabling LID

(UPDRS Grade 3) after 15 years. Furthermore, the mean duration of L-DOPA therapy before the onset of dyskinesias was 5.3 years.

Overall, despite the varying reports of LID incidence, the most commonly reported incidence is between 40 to 50% of patients within 5 – 6 years of initiating L-DOPA therapy, particularly in prospective randomised control trials (Ahlskog & Muenter, 2001; Holloway et al., 2004; Rascol et al., 2000). The variation in these reports may partly be due to different study methods, different selection of patients (Ahlskog & Muenter, 2001) and different patient populations (Schrag & Quinn, 2000; Fabbrini et al., 2007), but may also depend on whether the dyskinesias are troublesome enough, or even noticeable enough, for patients to report them, as patients can be unaware of mild dyskinesias (Schrag & Quinn, 2000; Vitale et al., 2001), or if dyskinesias are not present during clinical consultations.

Of course, it is also important to identify whether dyskinesias impact patient’s quality of life. Indeed, patients have reported that, when at their worst, dyskinesias can be socially embarrassing (Khlebtovsky et al., 2012) and impact their ability to perform activities of daily living (ADLs) such as eating and drinking (Manson et al., 2000; Chapuis et al., 2005).

Furthermore, dyskinesias can significantly impact patients’ quality of life through impaired emotional wellbeing and increased stigma (Montel et al., 2009) with increasing severity of dyskinias associated with increasing depression (Pechevis et al., 2005). However, it is important to note that patients with dyskinesia are more likely to prefer dyskinesias to parkinsonian symptoms such as bradykinesia (Hung et al., 2010). Therefore, patients with

Parkinson’s disease are often left with a difficult choice regarding whether to accept less efficacious therapies with lower rates of dyskinesias or commence L-DOPA, the gold

32 standard option for treatment, with dyskinetic side effects that can impaired emotional and social wellbeing.

1.1 Development of L-3,4-dihydroxyphenylalanine (L-DOPA)-induced dyskinesia (LID)

1.1.1 Overview of the basal ganglia The basal ganglia comprise several main neural structures involved in the motor pathways which are affected during PD progression. These neural structures include substantia nigra, the caudate nucleus and the putamen (fused into a single structure, the striatum, in rodents), the internal and external globus pallidus, and the subthalamic nucleus (DeLong & Wichmann,

2007; Obeso et al., 2008; Obeso et al., 2000). Multiple neural input pathways are received by the basal ganglia including the dopaminergic nigrostriatal pathway, the glutamate corticostriatal and thalamostriatal pathways, the noradrenaline pathway from the locus coeruleus, the serotonergic pathway from the raphe nuclei and the glutamate and acetylcholine pathways from the peduncolopontine nucleus. The main output pathways of the basal ganglia are gamma-amino-butyric acid (GABA)-ergic and project from the internal globus pallidus (GPi) and substantia nigra pars reticulata (SNr) to the thalamus and further to the premotor and motor cortices. These major output pathways have been implicated in the development of LID during progressive degeneration of the dopaminergic neurons in the substantia nigra pars compacta (SNc) in PD (Katz et al., 2005; Robertson, 1992).

Specifically, it has been proposed that reduced output through these pathways leads to disinhibition of the thalamus and over-excitation of the cortex resulting in dyskinetic movement (Fig. 1C).

The striato-thalamic GABA-ergic output circuit in the classic model of the basal ganglia is divided into two separate pathways, both of which originate with axons of GABA-ergic medium spiny neurons (MSNs). These pathways are termed the direct and the indirect pathways and lead to the thalamus via the GPi and SNr (Figure 1.1). The indirect pathway

33 travels from the frontal cortex to the striatum, continuing to the external globus pallidus

(GPe), through to the subthalamic nucleus (STN) and then reaching the thalamus via the

GPi/SNr. However, the direct pathway bypasses the GPe and STN entirely, leading straight from the striatum to the GPi/SNr and through to the thalamus. The MSNs found in the indirect pathway express inhibitory dopamine D2 receptors and co-localise the neurotransmitter, enkephalin, with GABA, whereas the MSNs in direct pathway express stimulatory dopamine D1 receptors and co-localise two neurotransmitters dynorphin and substance P with GABA.

Figure 1.1. The classic model of the basal ganglia in a A) normal state, B) state of progressive Parkinson’s disease in which the substantia nigra pars compacta is degenerating leading to overcompensation of the input from the subthalamic nucleus (STN) to the internal globus pallidus (GPi) and thus ‘overactivity’ in the indirect pathway, C) late stage of Parkinson’s Disease in which the substantia nigra pars compacta has degenerated to the degree where L-DOPA treatment is necessary resulting in ‘overactivity’ of the direct pathway and consequently, involuntary movements termed dyskinesia. Glu = Glutamate projections, GABA = γ-amino-butyric acid (GABA) projections.

The classic model of the basal ganglia suggests that as the SNc degenerates during the progression of PD, the loss of dopaminergic input from the SNc to the striatum leads to underactivity in the direct pathway and overactivity in the indirect pathway (Fig. 1B). This results in over-excitation of the GPi/SNr pathways increasing inhibition of the thalamus and

34 reducing excitation of the cortex resulting in reduced movement. It is thought that administration of exogenous L-DOPA replaces the dopaminergic input from the SNc inducing overactivity in the direct pathway and underactivity in the indirect pathway which results in cortical over-excitation and dyskinesia expression. This was originally thought to be the underlying mechanism of LID priming during L-DOPA administration in patients.

However, this model has important limitations. For example, stereotaxic lesioning of the posteroventral GPi has been shown to reduce dyskinesia in PD (Parkin et al., 2002). This is in contrast with the basal ganglia model in which lesioning of the internal globus pallidus should increase dyskinesia. However, stimulation of the posteroventral GPi has been found to also significantly reduce dyskinesia in patients with advanced PD (Volkmann et al., 1998), suggesting that although the model is useful for making predictions about PD and LID, it is also possibly incomplete with regard to the accepted understanding of the internal globus pallidus.

There is incomplete information regarding the role of the SNr in PD major motor symptoms and LID emergence. L-DOPA and D1 dopamine receptor agonists exert important effects at D1 dopamine receptor in the SNr (Katz et al., 2005; Robertson, 1992) and L-DOPA enhances the synaptic plasticity in SNr neurons of PD patients (Prescott et al., 2009). This indicates that plasticity in the SNr in a PD state might be compromised without dopaminergic input and lead to inflexible output responses. The classic model of the basal ganglia does not adequately incorporate this alteration of SNr plasticity or the firing rate and frequency output from the SNr. This aspect is important to consider as the firing rate of SNr neurons is known to change after L-DOPA treatment (Meissner et al., 2006) and it has even been suggested that

SNr neuronal oscillations in the theta/alpha frequency band may be driven by increased synchronised afferent activity after L-DOPA administration and that such abnormal activity

35 might at least facilitate LID-like abnormal involuntary movements in 6-hydroxydopamine

(OHDA)-lesioned rats (Meissner et al., 2006).

The classic model of the basal ganglia primarily focuses on the direct and indirect pathways to the thalamus in terms of output for movement control. A second, more recent model of the basal ganglia focuses on these pathways in addition to connections between other basal ganglia structures not directly part of the direct and indirect pathways. This pathway is termed the ‘hyperdirect’ cortico-STN-GPi/SNr pathway (Figure 1.2) and incorporates the role of SNr and the additional cortical innervation of the basal ganglia via the sub-thalamic (‘hyperdirect’) pathway (Nambu, 2004; Roberston, 1992). This model of the basal ganglia suggests that three corollary signals are involved in voluntary movement: 1) a signal to activate the hyperdirect pathway resulting in inhibition of all possible ‘motor programs’ (both selected and competing); 2) a corollary signal to activate the direct pathway to stimulate the selected motor program; and 3) the signal to activate the indirect pathway to inhibit all related targets in a manner to ensure precise initiation and termination of the appropriate motor program, and inhibit all irrelevant motor programs (Nambu et al., 2015).

However, there is little evidence to suggest that the hyperdirect pathway plays a role in the development of PD motor symptoms or LID.

Previous research suggests activity in the GPi/SNr during PD or LID is dependent on striatal dopaminergic activity modulating GABA transmission in both the direct and indirect pathways. As there is little, if any, degeneration of glutamatergic activity in the cortico-STN pathway that may lead to altered GPi/SNr neuronal activity, the mechanism by which the hyperdirect pathway would be influenced by altered striatal dopaminergic activity resulting in a change of thalamic inhibition has yet to be determined. Indeed, studies in MPTP-treated primates indicate that while the direct and indirect pathways are altered during the

36 development of parkinsonian symptoms, the hyperdirect pathway seems unaffected by either the administration of MPTP or L-DOPA (Nambu et al., 2000).

Research suggests that the SNr has a substantial role in symptoms of both PD and

LID. Studies in PD patients show oral L-DOPA-induced ON state, combined with high frequency stimulation (HSF), decreases neuronal firing rates while increasing the amplitude of field evoked potentials (fEPs) in the SNr, compared to when patients are in an OFF state

(Prescott et al., 2009). The long-term potentiation (LTP)-like changes and synaptic plasticity in this study suggests the presence of plasticity in the SNr is sensitive to dopaminergic levels.

These neuronal changes in humans are supported by studies in animal models of PD. The antiparkinsonian behaviours and abnormal involuntary movements observed in 6-OHDA- lesioned rats primed and challenged with L-DOPA coincide with a decrease in firing rate in the SNr and an excessive increase in striatal dopamine release associated with the lowest level of dopamine metabolism (Meissner et al., 2006). As the SNR projects GABA-ergic outputs to the thalamus, it is likely that reduced SNr activity after L-DOPA treatment disinhibits thalamic firing, resulting in increased movement, whether antiparkinsonian or dyskinetic.

Figure 1.2. A recent model of voluntary movements executed by the basal ganglia incorporating additional pathways between neural structures. Str = Striatum, STN = Subthalamic nucleus, GPe = Globus pallidus pars externa, GPi = Globus pallidus pars interna, SNr, Substantia nigra reticulata, GABA = GABA-ergic projections, Glu = Glutamatergic projections, DA = Dopaminergic projections. Green lines highlight the projection of the hyperdirect pathway, pink lines highlight the projection of the indirect pathway and blue lines highlight the projection of the direct pathway.

1.1.2 Factors involved in LID development

1.1.2.1 Hypothesis of LID emergence

Although the emergence of dyskinesia during L-DOPA treatment has been thoroughly established, the mechanism behind LID remains elusive. The classic model of the basal ganglia predicts that a dopamine deficiency in the SNc leads to underactivity of the direct pathway and overactivity of the indirect pathway, resulting in a reduction of movement, usually referred to as ‘parkinsonism’ (Goole & Amighi, 2009). However, in the development of dyskinesia, it is suggested that stimulatory D1 and inhibitory D2 dopamine receptors are overstimulated by exogenous L-DOPA, reversing basal ganglia activity and resulting in overactivity of the direct pathway and underactivity of the indirect pathway (Jenner, 2008).

Thus, the model of the basal ganglia predicts that excessive activity of the striatal neurons projecting to the GPi/SNr induced by L-DOPA stimulation disinhibits the thalamus to an abnormal extent, resulting in antiparkinsonian effects and abnormal involuntary movements.

How this overactivity of the direct pathway and underactivity of the indirect pathway occurs to produce dyskinesia is not fully established. One possible explanation is that extracellular dopamine levels over a L-DOPA treatment cycle occur in unregulated, pulsatile oscillations, leading to overactivity in both D1 and D2 receptor mediated activity (Cheshire &

Williams, 2012). Indeed, clinical studies have examined differences between oral and slow- release duodenal L-DOPA administration, but with inconsistent results. While one study reported a reduction in daily time spent with moderate to severe dyskinesia in patients after six months of duodenal L-DOPA administration, no washout period was included between the experimental treatment and patients’ original medication of oral L-DOPA and dopamine agonists and the change in dyskinesia severity was not assessed (Antonini, 2007; Antonini et al., 2007). In contrast, no changes in dyskinesia were found between patients using conventional medication, including oral L-DOPA, apomorphine, catechol-O-methyltrans-

38 ferase (COMT) inhibitors, monoamine oxidase-B (MAO-B) inhibitors, or glutamate N- methyl-D-aspartate (NMDA) receptor inhibitors, and patients administered duodenal L-

DOPA in either video assessment or questionnaires (Nyholm et al., 2005). One double- blinded, randomised control trial reported significant increases in ON-time without dyskinesia or without troublesome dyskinesia in patients administered intestinal L-DOPA and oral placebo compared to patients administered placebo intestinal gel and oral L-DOPA over

12 weeks (Olanow et al., 2014). However, no difference was found in ON-time with troublesome dyskinesia, nor was the severity of dyskinesia assessed. Thus, while sustained L-

DOPA administration does appear to have some impact on time-based dyskinesia measurements, there is little research on any changes in dyskinesia severity during sustained- release L-DOPA treatment. However, longer-acting direct dopamine agonists such as pramipexole produce fewer, and less severe, dyskinesia (Holloway, 2000; Parkinson Study

Group, 2000), unless the patients have previous priming with L-DOPA (Brodsky et al.,

2010), indicating that more sustained-release PD treatments may be beneficial in some circumstances.

Research also suggests that non-dopaminergic neurotransmitter pathways are involved in the severity and frequency of LID experienced by PD patients as multiple neurotransmitter systems are affected during the progression of PD (Halliday et al., 1990; Perry et al., 1991;

Scatton et al., 1983). Increasing evidence suggests that LID emergence is a treatment complication involving neurotransmitters apart from dopamine (Halliday et al., 1990; Perry et al., 1991; Scatton et al., 1983). Such neurotransmitters as glutamate, serotonin, cannabinoids, and noradrenaline have also been studied to determine if drugs that modulate these neurotransmitter systems may assist in treatment of LID (see Section 1.2 for references).

Amantadine, a glutamate NMDA receptor antagonist, is one of the best known antidyskinetic drugs used to alleviate LID. However, its inconsistent effects as a chronic treatment in

39 patients (Blanchet et al., 1998; Crosby et al., 2003; da Silva-Junior et al., 2005; Del Dotto et al., 2011; Luginger et al., 2000; Snow et al., 2000; Thomas et al., 2004; Wolf et al., 2010) highlights the need for further research into dyskinetic therapies as the efficacy of current treatments is not sufficient to warrant standard use. As there is a wide variety of avenues to explore for the treatment of LID, this review aims to identify the most probable mechanism underlying the development of LID and discuss the most promising areas of research regarding pharmacological LID treatment.

1.1.2.2 Delaying the onset of LID It has been established that there are three main factors that affect dyskinesia in PD: the drug administered to treat major motor symptoms of PD (Holloway, 2004; Noyes et al.,

2006; Oertel et al., 2006; Rascol et al., 2000), the mode of drug administration (Manson et al., 2002), and the extent of nigral cell loss. This section will review the effect of initial PD treatment on dyskinesia priming and expression.

Because established LID is difficult to treat, clinicians and researchers have focused on alternatives to L-DOPA treatment in PD patients to prevent dyskinesia from developing initially. These alternatives primarily include monoamine oxidase Type B (MAO-B) inhibitors (Parkinson Study Group, 1993; Olanow et al., 1995), and dopamine receptor agonists. MAO-B inhibitors have been shown to increase the average number of dyskinesia- free ON hours compared to placebo when used as an adjunctive therapy with L-DOPA in PD patients (Waters et al., 2004), while dopamine receptor agonists have been associated with a lower risk of dyskinesia onset (Clarke & Deane, 2001; Manson et al., 2002; Rascol et al.,

2006; Rinne et al., 1998).

MAO-B inhibitors, such as deprenyl and rasagiline, are thought to exert antiparkinsonian effects through reduction in dopamine metabolism by inhibiting monoamine oxidase Type B (O’Carrol et al., 1983), as increased levels of dopamine and reduced

40 deaminated metabolites have been found in post-mortem brain studies of deprenyl-treated patients (Riederer & Youdim, 1986). MAO-B inhibitors, particularly deprenyl, are at least mildly effective in motor symptomatic treatment during early PD and are generally well- tolerated in PD patients with no reports on MAO-B inhibitor-induced dyskinesia (Parkinson

Study Group, 1989; Rascol et al., 2011), thus delaying the need for L-DOPA treatment by up to 12 months (Parkinson Study Group, 1993). In advanced PD patients, deprenyl (Waters et al., 2004) and rasagiline (Rascol et al., 2005) also reduce daily OFF time, and increasing daily ON time without troublesome dyskinesia when used in conjunction with L-DOPA.

However, as the disease progresses, MAO-B inhibitors appear to become less effective in treating motor symptoms and more efficacious dopamine replacement therapies must be introduced (Hauser, 2009; Rascol et al., 2005). This reduction in MAO-B inhibitor treatment efficacy is not well understood, but may be influenced by decreasing dopamine levels as the

SNc continues to degenerate, reducing dopamine receptor stimulation.

There are two main classes in the dopamine receptor family: D1-like and D2-like receptors. The D1-like receptor class includes D1 and D5 receptors, which are thought to potentiate LID development, whereas D2-like receptor class includes D2, D3, and D4 receptors, and are thought to suppress LID development (Darmopil et al., 2009; Grondin et al., 1999). Specifically, D1 receptor sensitivity to ligands, rather than receptor expression, is associated with LID in a macaque model of LID (Aubert et al., 2005). Consistent with this,

D1-receptor antagonists dose-dependently suppress L-DOPA-induced abnormal involuntary movements (AIMs) in 6-OHDA-lesioned rats (Westin et al., 2007). In PD patients, selective

D1 receptor agonists induce dyskinesia in a similar manner to L-DOPA, whereas agonists of

D2 and/or D3 receptors induce little dyskinesia when they are used as a monotherapy in L-

DOPA-naïve patients (Rascol et al., 2006). This is reflected in clinical treatment, where agonists used as a monotherapy or in conjunction with L-DOPA are primarily non-selective,

41 such as apomorphine (Katzenschlager et al., 2005; Manson et al., 2002), or have a high affinity for D2-like receptors (D2, D3, and D4 receptors), such as lisuride (Stocchi et al.,

2002), cabergoline (Bedard et al., 1986; Clarke & Deane, 2001; Gerlach et al., 2003; Rinne et al., 1998), pramipexole (Constantinescu et al., 2007; Parkinson Study Group, 2009; Piercey et al., 1996), and ropinirole (Rascol et al., 2000; Rascol et al., 2006).

It has been suggested that a lower risk of dyskinesia development with dopamine agonists is due to their longer half-lives which prolongs their pharmacological effect, leading to a more tonic and physiological stimulation of dopamine receptors in the basal ganglia

(Chase, 1998; Rascol et al., 2006). However, both apomorphine infusion (Colzi et al., 1998;

Manson et al., 2002) and lisuride infusion (Stocchi et al., 2002) are associated with less severe dyskinesia, but not changes in incidence of dyskinesia, despite the more tonic and physiological stimulation of dopamine receptors through controlled-release infusions. An alternative explanation proposes that the stimulation of D1 receptors (more so than that of D2 receptors) is critical in the priming process, leading to the emergence of dyskinesia (Huot &

Fox, 2013). This explanation is supported by the significant reduction in dyskinesia incidences in patients treated with ropinirole, which has one of the lowest affinities for D1 receptor of the DA receptor agonists, compared to L-DOPA (Rascol et al., 2000; Rascol et al., 2006). Thus, it is possible that excessive D1 receptor stimulation may induce dyskinesia, but that pulsatile stimulation worsens dyskinesia severity.

Because dopamine agonists are associated with fewer, and less severe, dyskinesia, and can treat the major motor symptoms of PD, the need for L-DOPA might be brought into question. One of the most well-known advantages for administering L-DOPA to PD patients is its therapeutic efficacy is much greater than dopamine agonists in treating major motor symptoms (Holloway, 2004; Rascol et al., 2006). As the disease progresses, dopamine agonists become less effective in relieving major motor symptoms, with many patients being

42 supplemented with L-DOPA (Rascol et al., 2000), indicating that dopamine agonists are most effective in the early stages of PD although not as efficacious as L-DOPA (Holloway, 2000;

Holloway et al., 2004; Rascol et al., 2000). Although L-DOPA is also associated with complications such as wearing-off, “on-off” effects, and dyskinesia, the side effects of dopamine agonists are considered more severe as they include somnolence, oedema, hallucinations, and possibly contribute to the development of compulsive disorders in PD patients (Camargos et al., 2009). While dopamine agonists are known to have a reduced incidence of dyskinesia development, their side effects and reduced symptomatic efficacy as the disease progresses supports the continued need for L-DOPA therapy, and thus highlights the necessity for research into the underlying mechanisms of LID development.

A more difficult factor to mitigate in dyskinesia development is the degree of nigral cell loss. In addition to observed parkinsonism in humans exposed to 1-methyl-4-phenyl-

1,2,3,6-tetrahydropyridine (MPTP), was the observation of almost immediate development of dyskinesia after levodopa initiation to treat this MPTP-induced parkinsonism (Langston,

2017). This finding supported the hypothesis that LID is influenced by degree of nigral cell loss in addition to duration of L-DOPA treatment. These findings have also been replicated in rodents, with the time to LID development corrlating positively with completeness of nigral lesions (Winkler et al., 2002) indicating that the greater degree of nigral cell loss increases the likelihood of dyskinesia development.

New research into the neurodegeneration in PD has identified several novel avenues to compensate for the loss of nigral dopaminergic neurons. One approach is to halt the loss of further dopaminergic neurons through inhibition of tumour necrosis factor (TNF), a cytokine produced by activated immune cells and involved in neuroinflammation. Injection of a lentivirus encoding dominant-negative TNF delays further loss of nigral dopaminergic neurons 5 weeks after initial 6-OHDA lesions in rats (Harms et al., 2011). Other studies have

43 looked at compensatory axon growth to re-inervate severely dopamine-depleted striatal areas

(Bez et al., 2016). Another approach is the use of gene therapy to alter the activity of neurotransmitter systems instead of regenerating them. In 6-OHDA-lesioned mice, the blockade of dorsal striatal p11, a cellular scaffold protein that potentiates a variety of neurotransmitter receptors, via administration of adeno-associated virus RNA-encoding vectors improves L-DOPA-induced AIMs (Marongiu et al. 2016). Some gene therapies have even reached clinical trials, 15 subjects with moderately advanced Parkinson’s disease receiving VY-AADC01, an adeno-associated viral vector serotype-2 which encodes the complementary DNA for the enzyme aromatic L-amino acid decarboxylase (Christine et al.,

2019). These subjects demonstrated dose-related improvements in patient-reported ON-time without troublesome dyskinesia and quality of life at 12 months. However, this trial was open-label and included a small sample size and involves a neurosurgical procedure with associated anaesthetic and surgical risks.

These findings support the complex interaction between the dopaminergic agent used, the mode of administration, and degree of disease progression in developing dyskinesias in

Parkinson’s disease. Peripheral administration of L-DOPA in PD animal models results in higher extracellular striatal levels of dopamine compared to healthy animals upon the same treatment (Abercrombie et al., 1990). Furthermore, high individual and total doses of levodopa are associated with a greater risk of developing dyskinesias in PD (Fahn et al.,

2004; Cilia et al., 2014). Therefore, until an efficacious therapy is identified to ultimately halt or reverse nigrostriatal dopaminergic cell loss (and act effectively as a cure for

Parkinson’s disease), research into more easily modifiable factors such as dopaminergic agent and mode of therapy administration is warranted to combat and improve LID in PD patients.

1.1.2.3. Modifications of L-DOPA in established LID There are three main categories of LID:

1) ‘peak-dose’ dyskinesia, the most common form of dyskinesia thought to arise from

L-DOPA’s 90-minute half-life leading to over-stimulation of the dopamine receptors

during L-DOPA’s peak period,

2) ‘biphasic’ dyskinesia which are maximal at the beginning and at the end of the ON

motor response, appearing in the form of dyskinesia-improvement-dyskinesia,

3) ‘off-dystonia’, manifesting in the distal leg or foot and is often painful (Nutt,

2003).

Peak-dose LID commonly responds to dose reduction, but that also results in worsening parkinsonism and increasing “OFF” states (Stocchi, 2006; Thanvi et al., 2007).

One explanation for this worsening parkinsonism is that the therapeutic window for L-DOPA narrows during chronic treatment so that the critical threshold dose that induces dyskinesia is almost equivalent to that required to induce an ‘ON’ state in a patient. Therefore, a reduction in L-DOPA dosage that relieves dyskinesia almost inevitably induces ‘OFF’ states of immobility as the disease and treatment progress (Nutt, 1990), making it difficult to determine the appropriate dose for each patient that both treats major PD motor symptoms and avoids the development of dyskinesia.

Some have proposed that the severity of dyskinesia is increased during pulsatile L-

DOPA therapy (Obeso et al., 2000; Thanvi et al., 2007). However, the rate of LID is not affected by slow-release formulations of L-DOPA (Ahlskog & Muenter, 2001; Block et al.,

1997). More continuous therapy is based on the slow release of L-DOPA to maintain constant plasma concentrations. Duodenal L-DOPA infusions have been particularly effective in reducing dyskinesia severity in PD patients, both in the acute and chronic treatment compared to oral L-DOPA administration (Antonini et al., 2008; Bredberg et al., 1993; Nilsson et al.,

1998; Nisson et al., 2001; Zibetti et al., 2013). Specifically, LID duration decreases 3-8

45 months after duodenal L-DOPA therapy compared to oral L-DOPA and decreases further after 4-7 years (Nisson et al., 2001; Zibetti et al., 2013). Additionally, ‘OFF’ period duration is significantly decreased (Zibetti et al., 2013) and quality of life is significantly increased

(Antonini et al., 2008) compared to oral L-DOPA treatment.

The incidence of dyskinesia in PD patients undergoing sustained-release L-DOPA therapy is more difficult to determine as many studies assess dyskinesia improvement by the

Unified Parkinson’s Disease Rating Scale (UPDRS) section (IV) regarding dyskinesia duration, disability (severity), painfulness and presence of early morning dystonia. This scale does little to measure incidence of dyskinesia specifically and so is not a clear indication of whether sustain-release L-DOPA formulations reduce the incidence of dyskinesia.

Additionally, many patients exposed to alternative routes of L-DOPA administration have already developed LID during oral L-DOPA therapy which would most likely lead to alterations in dyskinesia severity and duration rather than incidence (Zibetti et al., 2013).

However, studies have implicated that immediate- and sustained-release L-DOPA formulations induce dyskinesia in a similar proportion of L-DOPA-naïve patients (Ahlskog &

Muenter, 2001; Block et al., 1997). This suggests that pulsatile L-DOPA therapy may not induce dyskinesia, but may affect dyskinesia severity or duration once they have emerged

(Antonini et al., 2008; Zibetti et al., 2013).

Although duodenal infusions are effective in minimizing L-DOPA side effects, they are also associated with complications. Such complications reported include movement of the catheter tip from the jejunum, sometimes in all cases (Antonini et al., 2008), severe psychosis related to the infusion in one patient in a study of 25 participants, infections occurring in up to 48% of patients (12 cases out of 25 participants), intestinal tube dislocation with migration to the stomach (occurring 34 times), and the percutaneous endoscopic gastronomy being pulled out accidentally (Nilsson et al., 1998; Zibetti et al., 2013). These studies reported

46 participant withdrawal from or discontinuation of duodenal therapy after the study due to such complications, highlighting the need for careful consideration before undergoing this form of L-DOPA therapy.

1.2 Pharmacological treatment of LID 1.2.1 Dopamine receptors 1.2.1.1. D1-like receptors Because L-DOPA is metabolised into dopamine, and is therefore non-selective for dopamine receptor subtypes (although dopamine has highest affinity for D3 receptors, which are most dense in the mesolimbic rather than nigrostriatal pathway), it is difficult to determine which of the five dopamine receptors (D1-5) are primarily involved in LID development. However, as levels of D1 and D2 receptor mRNA are the most abundant in the striatum of the human brain (Meador-Woodruff et al., 1996), particularly the caudate nucleus

(Hall et al., 1994), these have been the most extensively studied dopamine receptors in LID research. Previous research has found that D1 receptor inactivation in mice inhibits the development of LID (Darmopil et al., 2009) and that D1 receptor binding increased in MPTP- lesioned primates after L-DOPA administration (Aubert et al., 2005). However, an increase in

D1 receptor sensitivity in dyskinetic MPTP-treated monkeys, as measured by D1 agonist- induced GTPγS binding, was significantly different from non-dyskinetic but not MPTP- treated monkeys (Aubert et al., 2005). This indicates that, although D1 receptor expression in non-dyskinetic monkeys increases, the sensitivity of these receptors appears to decrease, whereas increased D1 expression and sensitivity seems to occur in dyskinetic monkeys. It is possible that this increase in supersensitivity involves the serotonergic pathway as the 5,7- dihydroxytryptamine-induced destruction of serotonergic fibres was found to attenuate dopamine D1 receptor supersensitivity in 6-OHDA-lesioned rats (Brus et al., 1994).

However, D1 receptor antagonism by SCH 23390 dose-dependently decreased severity and duration of AIMS in 6-OHDA-lesioned rats (Westin et al., 2007). One explanation for this is that D1 receptor antagonism by SCH 23390, but not D2 receptor antagonism by raclopride, significantly reverses L-DOPA-induced increases in phosphorylation of extracellular signal-regulated kinases 1 and 2, ERK1/2 (Westin et al.,

2007). Additionally, the D1 receptor agonist ABT-431 induced dyskinesia at similar frequencies and severities as L-DOPA in a small sample of patients with PD (Rascol et al.,

2001a), indicating that D1 receptor stimulation is the primary contributor to LID expression.

However, it is unknown whether the latency period for initial ABT-431-induced dyskinesia in

L-DOPA-naïve PD patients would be similar to that of L-DOPA.

Very little research has been conducted on the D5-mediated effects of LID, possibly because the D1 subtype has most often been described to be associated with D1-like functions

(Kahn et al., 2000). This understanding was derived from the ﬁnding that D1 receptors are predominantly localized in the primary dopaminergic projection areas, such as striatum and nucleus accumbens (Hall et al., 1994). Additionally, it was not previously possible to identify the distribution of D5 receptors in the brain (Kahn et al., 2000). Now, however, growing evidence indicates that D5 receptors contribute individually in dopaminergic transmission as seen by stronger D5 receptor coupling to G proteins (Kimura et al., 1995), supporting the evidence that D5 receptors have a higher affinity for dopamine compared to D1 receptors

(Grandy et al., 1991; Sunahara et al., 1991; Tiberi et al., 1991). However, studying D5 receptors is problematic as selective D5 receptor-binding ligands are notoriously difficult to develop (Beaulieu & Gainetdinov, 2011). There has also been disagreement regarding D5 receptors involvement in locomotion and their low to moderate distribution in brain areas relevant to dyskinesia development, particularly, the substantia nigra and nucleus accumbens

(Beaulieu & Gainetdinov, 2011; Ciliax et al., 2000; Kahn et al., 2000; Missale et al., 1998).

However, studies have shown that D5 receptors contribute individually to behaviours including increased contralateral rotation in 6-OHDA-lesioned rats after administration of antisense oligodeoxynucleotide for the D5 receptor (Dziewczapolski et al., 1998) and reproductive behaviours in healthy rats after administration of D1-like receptor agonists

(Apostolakis et al., 1996), indicating that D5 receptors may be involved in other behaviours, possibly of a PD nature. Although Dziewczapolski et. al. (1998) study provided insight into the involvement of D5 receptors in locomotion, there was no indication as to how or to what extent D5 receptors might be useful in LID treatments for PD patients; however, future studies may suggest differently.

1.2.1.2. D2-like receptors The previous section focussing on evidence for the role of dopamine D1 receptors in

LID does not discount the possible contributions of D2-like receptors to dyskinesia development. Most of these previously mentioned studies failed to test the behavioural effects of D2-like receptor antagonism in dyskinetic animals. It is important to consider the involvement of D2 receptors in L-DOPA-induced motor side effects as eticlopride (D2 receptor antagonist, 0.1 and 1.0 mg/kg) has been shown to reduce AIMs in 6-OHDA-lesioned rats both alone and in conjunction with SCH 23390 (0.1 and 1.0 mg/kg), a D1 receptor antagonist (Taylor et. al., 2005). It has been shown that D2 receptor binding is not significantly impaired by L-DOPA treatment, regardless of dyskinetic development (Aubert et. al., 2005), and thus it should not be assumed that D2 receptor changes during dyskinesia development are the same as, or even similar to, changes in D1 receptors.

It has also been found that D2-like receptors such as dopamine D3 receptors could be involved in LID. However, the mechanisms underlying this involvement are less clear.

Significant D3 receptor mRNA levels are found in the caudate, putamen and are particularly enriched in the nucleus accumbens (Meador-Woodruff et al., 1996). Antagonism of D3

49 receptors is known to increase rodent locomotion: even after habituation to their environment, rats treated with 100 µmol/kg U99194A (a D3 receptor antagonist) induced a nearly 500% increase in activity compared to saline controls (Waters et al., 1993). Also, co- administration of apomorphine (after reserpine treatment) or dexamphetamine with U99194A synergistically increases locomotion in rats compared to saline or either drug alone (Waters et al., 1993). This further supports the suggestion that stimulation of D3 receptors reduces locomotor activity as blockade of these receptors removes the inhibitory influence D3 receptors exert and allow for a greater expression of dopaminergic activity through D1 and D2 receptor stimulation (Waters et al., 1993). In contrast, the D3 receptor-preferring agonist R-

(+)-7-OH-DPAT decreases locomotion in rats compared to saline controls without affecting dopamine release or synthesis rate, also supporting the suggestion that D3 receptor stimulation inhibits locomotion, increasing its potential as a therapeutic target for LID

(Svensson et al., 1994). It is possible that D3 receptors mediate locomotion through facilitation or inhibition of GABA-ergic MSN firing which could lead to decrease or increase in locomotion respectively. Indeed, it has been indicated that presynaptic D3 receptor stimulation predominately facilitates GABA-ergic neural transmission in MSNs of D2 receptor knock-out mice after exposure to quinpirole (D2 and D3 receptor agonist), although it was suggested that D3 receptors also mediate GABA-ergic MSN inhibition to a smaller degree (Mizuno et al., 2007).

It has been shown that D3 receptor binding is higher in the putamen and GPi of

MPTP-treated dyskinetic monkey brains compared to that of nondyskinetic or normal primate brains obtained from a brain bank (Guigoni et al., 2005). Additionally, partial D3 receptor agonist BP 897 and D3 receptor antagonists, nafadotride and ST 198, effectively reduced LID in MPTP-treated monkeys compared to L-DOPA controls (Bezard et al., 2003). However, the antagonists also decreased antiparkinsonian effects of L-DOPA, suggesting that perhaps only

50 partial agonism normalises D3 receptor function to reduce LID without compromising L-

DOPA-induced beneficial effects (Bezard et al., 2003), or that D3 receptor stimulation increases GABA-ergic neural transmission, partially reducing locomotion or movement

(Minuzo et al., 2007).

Pramipexole is the preferential D2-like receptor agonist that is used in the clinic with highest affinity for D3 receptor (63-fold greater than for D2 receptors, but only 5-fold greater potency based on EC50,). It effectively reduces LID in patients with PD without affecting antiparkinsonian effects of the patients’ original medication (Utsumi et. al., 2013). In contrast, quinpirole (2.5 mg/kg), another D2-like receptor agonist with highest affinity for the

D3 receptor (35-fold higher affinity than for the D2 receptor), induced limb, axial, locomotor and grooming AIMs as well as contralateral rotations in 6-OHDA-lesioned rats (Drake et. al.,

2013). Additional experimentation with D2 and D3 receptor antagonists administered in conjunction with pramipexole might identify the mechanism for this increase in AIMs. In particular, (+)-4-propyl-9-hydroxynapthoxazine has been shown in human brain imaging studies to have a very high preference for D3 receptors over D2 receptors, with very little non- specific binding (Graff-Guerrero et al., 2010) and may be a candidate as an antidyskinetic drug.

Although there are clinical studies being conducted using D4 receptor antagonists for

LID treatment, the antidyskinetic effect of D4 receptors does not seem to be as thoroughly investigated as that of some other dopamine receptors. Although moderate levels of D4 receptors have been found in the striatum and even lower levels in the nucleus accumbens of primates, rodent studies have demonstrated D4 receptor involvement in LID (Rivera et al.,

2002). The dopamine D4 receptor antagonist, L-745,870, significantly decreases L-DOPA- induced ON time with overall dyskinesia and disabling dyskinesia in MPTP-lesioned macaques without affecting good quality ON-time or Parkinsonian severity compared to L-

DOPA administered alone (Huot et al., 2012a). Although results regarding D4 receptor antagonism as a potential target for dyskinesia therapy have so far been promising, this area has been less thoroughly studied than other antidyskinetic possibilities. It is therefore difficult to comment on the likelihood of D4 antagonists being a clinically useful dyskinesia therapy.

Overall, D1 and D2 receptors were thought to be the most influential and promising targets for LID of the five dopamine receptor subtypes (Jenner et al., 2008). However, research shows that D3, D4 and D5 receptors contribute individually to AIMs in both rodents and primates. It is also likely that these dopamine receptors influence each other’s activity and therefore effect on locomotion. For example, agonists of D3 receptors are known to inhibit locomotion (Waters et al., 1993) primarily by acting on postsynaptic D3 receptors

(Kling-Petersen et al., 1995; Waters et al., 1994; Waters et al., 1993), and possibly by increasing GABA-ergic release from MSNs, inhibiting D1 and D2 receptor activity (see above). This reasoning is based on the evidence that antagonism of both D1 and D2 receptors inhibit AIMs in animal models of LID (Westin et al., 2007; Taylor et al., 2005). Additionally,

D3 receptors might also influence the activity of D4 receptors by reducing their effect on locomotion as D4 receptor antagonism has shown to reduce AIMs (Huot et al., 2012b) and due to their proximity in the striatum (Rivera et al., 2002; Meador-Woodruff et. al., 1996).

Activation of the usual dopamine D2 receptor by agonists (e.g., dopamine released by L-

DOPA, or direct D2-preferring agonists) should exacerbate dyskinesias. However, D3 preferential agonists and antagonists both inhibit dyskinesias (Jenner et al., 2008). It is likely that the effects of D3 antagonists (i.e., they have not always been replicated) are mediated by interactions with D2 receptors. D3 agonists have no effect on their own when expressed in cell membranes in vitro, but greatly increase the efficacy of D2 receptors when they are co- expressed (Maggio et al., 2003). That is, D2 and D3 receptors form heterodimers that produce a “sensitized” D2 receptor and these heterodimers are potently acted on by the agonist,

52 pramipexole. It is likely that these heterodimers act to inhibit dopamine release via presynaptic autoreceptors, since D3 preferential agonists are ineffective at reducing DA release in D2-knock out mice (Zapata & Shippenberg, 2005). However, there is also evidence that D3 receptor stimulation is responsible for dyskinesias (Visanji et al., 2009). These data are not as paradoxical as they appear. It is likely that postsynaptic D2/D3 heterodimers are responsible for the development of sensitization to D2 receptors after L-DOPA treatment, and blockade of these by antagonists can therefore improve dyskinesia, whereas agonists may act preferentially on presynaptic, DA release inhibiting D2/D3 heterodimers.

1.2.2 Serotonin (5-HT) receptors Serotonin receptors have been found to greatly influence the expression and severity of LID in animal models and humans (Carta & Bjorklund, 2018). Similar to dopaminergic neurons requiring AADC to convert L-DOPA into dopamine, serotonergic neurons require

AADC to convert the 5-hydroxytryptophan into serotonin. This AADC in serotonergic neurons gives them the capacity to covert L-DOPA into dopamine (Arai et. al., 1994;

Yamada et. al., 2007; Arai et. al., 1995; Maeda et. al., 2005). Rat neurotoxin models of PD have shown that up to 80% of all L-DOPA-derived dopamine is released from serotoninergic axon terminals (Tanaka et. al., 1999; Cenci & Lindgren, 2007; Kannari et. al., 2001; Lopez et. al., 2001) while other studies have demonstrated some of the highest regional uptake of

SERT-binding radioligands, and therefore concentration of SERT, occurring in the substantia nigra of rats (Kretzschmar et al., 2003). Serotonergic neurons also only have feedback mechanisms for regulating serotonin release via 5-HT1 presynaptic autoreceptors, and do not have feedback mechanisms to regulate dopamine release, which is regulated in dopaminergic neurons via D2 and D3 presynaptic autoreceptors, as well as multi-neuron feedback loops.

Therefore, because serotonergic neurons lack the feedback mechanisms for regulating dopamine release, as their terminal vesicles become filled with dopamine, and dopamine is

53 released instead of serotonin, the serotonergic neurons lose their normal serotonergic feedback mechanisms, resulting in greater rates of release of neurotransmitter than normal

(Waldmeier, 1985). Thus, dopamine is released in at non-physiological manner, creating excessive fluctuations in extracellular striatal dopamine which induce dyskinesia in animal models (Parkinson Study Group, 2000; Rascol et. al., 2000; Constantinescu et. al., 2007). The serotonergic receptors and transporters that influence LID in animal models and human will be further discussed here.

1.2.2.1 5-HT1A/1B receptors Serotonin 5-HT1A and 5-HT1B receptors are G-protein-coupled receptors that are thought to influence LID expression. Specifically, stimulation of 5-HT1A receptors alone has been shown to reduce dyskinesia-like behaviour in both 6-OHDA-lesioned rats and MPTP- treated primates (Carta et al., 2007). This is possibly due to unregulated release of exogenous

L-DOPA-derived dopamine from 5-HT neurons, since L-DOPA-induced dyskinesia-like behaviour was significantly reduced after serotonergic lesioning by 5,7-dihydroxytryptamine

(5,7-DHT) in 6-OHDA-lesioned rats (Carta et al., 2007) and 5-HT1A receptor agonism by sarizotan in MPTP-treated primates (Bibbiani et al., 2001). These effects in primates were then blocked by administration of WAY100635, indicating that stimulation of 5-HT1A receptors was responsible for these motor effects (Bibbiani et al., 2001). As stimulation of 5-

HT1A autoreceptors inhibits serotonin release from neurons (Davidson & Stamform, 1995), it is possible that unregulated exogenous-L-DOPA-derived dopamine release from serotonergic neurons may also be inhibited by 5-HT1A release (Cheshire & Williams, 2012).

Stimulation of 5-HT1A receptors can also improve the efficacy of L-DOPA while reducing LID expression in 6-OHDA-lesioned rats as seen with the partial 5-HT1A agonist, buspirone (Eskow et. al., 2007). As buspirone is a partial agonist, the reduction of LID expression may occur through reduced dopamine release by serotonin neurons when 5-HT1A

54 receptors are stimulated and when dopamine release is initially high, while L-DOPA efficacy is improved by blocking 5-HT1A receptors when dopamine release by serotonin neurons is low and serotonin release is greater. Other researchers have also found that 5-HT1A agonism by +8-OH-DPAT, a 5-HT1A receptor agonist, in the dorsal raphe nucleus of 6-OHDA- lesioned rats can reduce LID expression, and that positive correlations exist between percentage of intact 5-HT innervation of the rostral raphe nucleus in 6-OHDA-lesioned rats and axial, limbic and orolingual AIMS (Eskow et. al., 2009). The raphe nucleus is not the only structure to play an important serotonergic role in LID expression: it has been shown that 5-HT1A stimulation by +8-OH-DPAT in the primary motor cortex reduced L-DOPA- induced c-fos and diminished the onset of L-DOPA-induced AIMs, an effect which was also reversed by administration of WAY100635 (Ostock et. al., 2011). This evidence indicates that anatomically selective 5-HT1A receptor agonists might be useful in combating LID expression.

Other studies have provided further evidence that dopamine can be released as a ‘false neurotransmitter’ from serotonergic terminals. For example, 6-OHDA-lesioned rats displaying L-DOPA-induced AIMs have higher dopamine levels than non-dyskinetic rats also treated with L-DOPA and that this increase in dopamine efflux is blunted by 5-HT1A and 5-

HT1B agonist administration (Lindgren et. al. 2010). It was also demonstrated exogenous L-

DOPA-derived extracellular dopamine levels were significantly attenuated after +8-OH-

DPAT administration in 6-OHDA-lesioned, an effect which was reversed after WAY100635 pre-treatment (Kannari et. al., 2001). Two highly selective ‘biased’ 5-HT1A receptor agonists,

F13714 and F15599, blunted peak L-DOPA-induced increases in dopamine microdialysate levels in dernervated, but not intact, striatum of rats (Newman-Tancredi et al., 2018).

However, this attenuation of extracellular dopamine levels was not found after 5-HT1B receptor agonism by CGS-12066 (Kannari et al., 2001), indicating that the 5-HT1A receptor is

55 primarily responsible for the increase in L-DOPA-derived dopamine levels in the absence of dopaminergic neurons.

In contrast, 5-HT1B receptor mRNA levels are increased in the lesioned striatum of unilaterally 6-OHDA-lesioned rats and 5-HT1B expression is markedly further elevated after administration repeated L-DOPA administration (Zhang et. al., 2008; Padovan-Neto et al.,

2019). It is possible that, while 5-HT1B receptors do not induce antidyskinetic effects in animal models when stimulated alone, that they play a more important when stimulated in conjunction with 5-HT1A receptors. Indeed, 5-HT1A and 5-HT1B receptor agonists, when administered in combination, have been found to attenuate L-DOPA-induced AIMs in 6-

OHDA-lesioned rats (Muñoz et. al., 2008; Muñoz et. al., 2009). This attenuation of LID in rodents and primates seems to be dose-dependent, but still primarily dependent on 5-HT1A stimulation, as LID scores between +8-OH-DPAT + CP-94253, a 5-HT1B receptor agonist

(0.5 mg/kg and 2.5 mg/kg, respectively) pre-treatment and +8-OH-DPAT + CP-94253 (1.0 mg/kg and 2.5 mg/kg) pre-treatment did not appear to differ substantially, however this was not statistically verified. Additionally, the +8-OH-DPAT + CP-94253 (1.0 mg/kg and 2.5 mg/kg) pre-treatment combination significantly increased PD score compared to L-DOPA administered alone in MPTP-treated macaques, indicating that 5-HT1 receptor agonist combinations can reduce the efficacy of L-DOPA. It seems likely that both the decrease in

LID scores and increase in PD scores were primarily mediated by 5-HT1A receptor stimulation, as the +8-OH-DPAT + CP-94253 (0.5 mg/kg and 2.5 mg/kg, respectively) combination reduced LID score without affecting the PD score while +8-OH-DPAT + CP-

94253 (1.0 mg/kg and 2.5 mg/kg) increased and decreased PD and LID scores, respectively

(Munoz et al., 2008). Therefore, because only the dose of the 5-HT1A receptor agonist was altered, as was the efficacy of L-DOPA consequently, these observations suggest that, while a combination of 5-HT1 agonists seems to act synergistically to induce changes to L-DOPA-

56 induced effects, the stimulation of the 5-HT1A receptor seems to be the more important of the two receptors.

Although evidence for the efficacy of 5-HT1 receptor agonists in animal studies has been promising, there have been inconsistencies regarding antidyskinetic effects of 5-HT1 receptor agonists in human trials. Significant differences were found in Part IV (Dyskinesia) scores of the UPDRS, but not the diary-based data, between sarizotan- and placebo-treated

PD patients during a double-blind, placebo-controlled trial (Goetz et. al., 2007; Bara-Jimenez et. al., 2005). However, other studies have indicated that sarizotan failed to produce antidyskinetic effects altogether in Phase III trials when compared to placebo (Muller et al.,

2006; Rascol et al., 2006; Goetz et al., 2008). Because of this, sarizotan-based research has been essentially abandoned, and research into other 5-HT1A agonists with antidyskinetic efficacy is proving difficult. For example, tandospirone reduces dyskinesia in approximately

50% of patients, but can also reduce antiparkinsonian effects of L-DOPA (Kannari et. al.,

2002).

It is not fully understood why these compounds exhibit inconsistent antidyskinetic effects in PD patients. It is possible that co-stimulation of 5-HT1A and 5-HT1B receptors may exert a stronger antidyskinetic effect in patients. Unfortunately, there has been little clinical research in the antidyskinetic effects of 5-HT1B agonists, although, if their effects in animal models of LID are any indication, reduced dyskinesia via 5-HT1B stimulation would be amplified with 5-HT1A co-stimulation. This could be another avenue for LID in PD patients, with effective 5-HT1B agonists administered in combination with 5-HT1A agonists to determine if existing antidyskinetic effects may be amplified as seen in primates (Muñoz et. al., 2008) and reduce effects on L-DOPA-induced antiparkinsonism.

1.2.2.2 5-HT2A receptors It has been shown that serotonin 5-HT2 receptors influence motor behaviours. In particular, 5-HT2A receptors are thought to underlie increased motor behaviours in 6-OHDA- lesioned rats after intrastriatal 5-HT2A/C agonism by [(_)-1-(4-iodo-2,5-dimethoxyphenyl)-2- aminopropane] (DOI) compared to sham-operated rats (Bishop et. al., 2004). This effect was blocked by intrastriatal infusion of both the 5-HT2 antagonist ketanserin and the 5-HT2A preferring antagonist M100907, rather than the 5-HT2C preferring antagonist RS1-222, suggesting that 5-HT2A receptors are responsible for the effects of DOI. Like 5-HT1A receptors, 5-HT2A receptors are found both pre- and post-synaptically: stimulation of the presynaptic 5-HT2A receptors are thought to increase glutamate release (Aghajanian &

Marek, 1999), while stimulation of postsynaptic 5-HT2A receptors increases NMDA- mediated neurotransmission (Neuman & Rahman, 1996). It is, therefore, possible that increased motor behaviour in rats after 5-HT2 receptor stimulation is mediated by increases in glutamatergic neurotransmission.

Additionally, researchers have found that 5-HT2A receptor binding is increased in the striatum of dyskinetic MPTP-treated monkeys chronically administered L-DOPA (Huot et. al., 2012a). Moreover, 5-HT2A receptors are found in moderate density in the SNc

(Pompeiano et al., 1994) which, when stimulated, increase dopamine outflow in the striatum

(Lucas & Spampinato, 2000; Pehek et al., 2006). Such evidence indicates that antagonism of

5-HT2A receptors, particularly in the corticostriatal pathway, would reduce corticostriatal glutamate transmission and, in doing so, possibly LID. Because 5-HT2A receptors mediate dopamine outflow, 5-HT2A receptor antagonism would most likely reduce the release of exogenous L-DOPA-derived dopamine, which would reduce dyskinesia, but may also impair antiparkinsonian effects of L-DOPA.

Research into possible antidyskinetic effects of 5-HT2A receptor antagonists has identified compounds that reduce LID. A 5-HT2 receptor antagonist, ritanserin, has also

58 shown antidyskinetic effects in PD patients treated with L-DOPA although, like clozapine and quetiapine, was also shown to exacerbate PD motor symptoms (de Noordhout and

Delwaide, 1986; Meco et. al., 1988). These studies also used a single-blind rather than double-blind placebo procedure, making the need for cautious interpretation of their results necessary. Pimavanserin, a 5-HT2A/C receptor antagonist reduced LID in both primates

(Vanover et al., 2008) and PD patients (Roberts, 2006; Mills et al., 2008) without compromising the antiparkinsonian effects of L-DOPA, however it is possible that 5-HT2C receptors also play a role in this reduction. One study found that the R enantiomer of the

MDMA molecule, has a relatively high antagonistic affinity for 5-HT2A receptors (in the low micromolar range but no affinity for 5-HT1A receptors or monoamine transporters, Huoet et al., 2011a). They also reported reduced LID in primates with R-MDMA (Huot et. al., 2011a).

This evidence supports the involvement of 5-HT2A, but MDMA is thought to be cytotoxic and certainly has potential for recreational abuse and misuse (Henry et al., 1992; Burgess et al., 2000). Although these effects have not been demonstrated for the R-enantiomer specifically, clinical trials to determine this 5-HT2A-mediated antidyskinetic effect of R-

MDMA in humans is potentially difficult.

Despite the research into the influence of 5-HT2A receptors on LID frequency and severity, little is known about how these compounds induce antidyskinetic effects at a cellular level. It has been suggested that 5-HT2A receptors mediate nigrostriatal dopamine release

(Huot et al., 2012) and are involved in glutamate activity (Riahi et. al., 2011) in MPTP- treated dyskinetic primates. It has been shown that depletion of serotonin with the serotonin neurotoxin 5,7-dihydroxytryptamine potentiates dopamine effects of dexamphetamine (Carter

& Pycock, 1979; Martin-Iverson et al., 1983), 5-HT2A/B receptor antagonists block the effects of dopamine receptor agonists (Carter & Pycock, 1978) and activation of 5-HT2A receptors in the striatum increases the effects of a dopamine antagonist (Lucas & Spampinato, 2000).

Therefore, it seems that serotonin acting on 5-HT2 receptors block dopamine’s effects. It is likely that increasing 5-HT2 receptor activity may ameliorate dyskinesias by reducing postsynaptic dopamine effects, consistent with a sensitised dopamine response after long- term L-DOPA contributing to dyskinesia.

Although there is evidence to support the hypothesis that the serotonergic system may be involved in expression of LID, it cannot adequately account for all aspects of dyskinesia development. Firstly, dyskinesia can also develop after treatment with dopamine receptor agonists, although their mechanism involves direct stimulation of the post-synaptic dopamine

D2 receptors. This is shown through the suppression of L-DOPA-induced AIMs, but not apomorphine-induced AIMs, after the removal of serotonergic innervation to the striatum in

6-OHDA-lesioned rats (Munoz et. al., 2009). However, the dyskinetic effects are less with direct agonists than with L-DOPA (Scholz et. al., 2008), but are increased if there has been prior exposure to L-DOPA, suggesting a “priming” effect that may be related to a sensitization phenomenon (Scholz, et al, 2008), which depends upon intermitted stimulation of receptors (Martin-Iverson et. al., 1987; Martin-Iverson et. al., 1988a; Martin-Iverson et. al.,

1988b). Thus, despite increasing evidence of serotonergic involvement in LID development, it is unlikely to be the only mechanism or treatment avenue.

1.2.3 Monoamine transporters

More globally, the monoamine receptors described above can be influenced and manipulated through indirect mechanisms, primarily involving the monoamine transporters such as the dopamine and serotonin transporters (DAT and SERT, respectively) using monoamine reuptake inhibitors.

Monoamine reuptake inhibitors such as methylphenidate, tesofensine, and BTS-74398 have been trialled both preclinically and clinically exert antiparkinsonian and/or

60 antidyskinetic effects with or without the presence of of L-DOPA. However, these compounds have demonstrated mixed results both preclinically and clinically. Oral methylphenidate, while producing no improvement of parkinsonian symptoms alone, improved tapping and walking speeds when coadministered with L-DOPA infusions in 17 patients with Parkinson’s disease (Nutt et al., 2004). There was also an increase in duration of dyskinesias experienced, although not in severity of dyskinetic movements. However, larger and more global trials have investigated methylphenidate for gait impairment in Parkinson’s disease patients with mixed results. A randomised, controlled, multicentre study found that methylphenidate improved gait hypokinesia and freezing in 81 patients with advanced

Parkinson’s disease ((Moreau et al., 2012). Contrastly, a double-blind, placebo-controlled study of 27 subjects (only 17 of whom completed the trial) found that methylphenidate did not improve gait and worsened motor function, somnolence and quality of life (Espay et al.

2011). Therefore, there is “insufficient evidence” to support the use of methylphenidate as a standard option for clinical care in Parkinson’s disease patients with gait problems (Fox et al.,

2018).

Tesofensine, serotonin–noradrenaline–dopamine reuptake inhibitor, from the phenyltropane family of drugs, was initially considered obesity. However, it has been investigated in advanced Parkinson’s disease patients to reduce motor fluctuations, also with mixed results. In a randomised, placebo controlled trial of 254 Parkinson’s disease patients

(184 of whom completed the trial), tesofensine (0.25-0.5 mg) modestly improved motor fluctuations on UPDRS subscales II and III and OFF time after 14 weeks of administration, but induced adverse drug reactions at higher doses, namely gastrointestinal and neuropsychiatric (Rascol et al., 2008). This study was also unable to demonstrate a dose- response curve with the doses used (0.125 mg, 0.25 mg, 0.5 mg, 1 mg p.o.). However, in a 4- week, randomised, double-blind placebo-controlled study, 1.5 mg oral tesofensine three times

61 a week did not improve UPDRS subscale III score in seven patients with advanced PD compared to 2 placebo patients. It also did not increase anti-parkinsonian benefit when combined with L-DOPA or worsen dyskinesia severity (Bara-Jimenez et al., 2004).

Regarding adverse effects, a meta-analysis of two studies involving tesofensine (0.25 mg, 0.5 mg 1 mg p.o.) administration in either Alzheimer’s or Parkinson’s disease patients demonstrated significant weight loss compared to the placebo group (Astrup et al., 2008).

Although not investigated to treat dyskinesias, BTS-74398, another non-selective

DAT/NAT/SERT inhibitor, significantly reverses parkinsonism in MPTP-lesioned marmosets without inducing dyskinesias when administered as a monotherapy (Hansard et al., 2002).

Furthermore, when combined with low dose L-DOPA, BTS-74398 does not exacerbate the severity of dyskinesias exhibited by MPTP-lesioned primates; however, it does not enhance antiparkinsonian active of L-DOPA either (Hansard et al., 2004). It is perhaps because of these mixed pre-clinical results that BTS-74398 has not yet been investigated in clinical trials.

One of the earliest promotions of MDMA improving LID in Parkinson’s disease patients occurred with a BBC article reporting reduced dyskinesias in a patient with early- onset Parkinson’s disease after recreational use of MDMA (BBC, 2001). In nonhuman primates and rodents, MDMA has also demonstrated antidyskinetic effects (Bishop et al.,

2006; Crespi et al., 1997; Johnston et al., 2012; Nash & Brodkin, 1991); however, the mechanism underlying these effects is still uncertain. Racemic MDMA and its S-entantiomer are non-selective triple monoamine reuptake inhibitors, while th R-enantiomer primarily binds to the SERT, by reversing the transport gradient and promoting increased monoamine release (Han & Gu, 2016; Rothman et al., 2001; Verrico et al., 2007; Battalgia et al., 1988;

Huot et al., 2010). However, MDMA also has weak affinity for several other receptors, including 5-HT2A and D1 and D2 receptors (Battaglai et al., 1988). Therefore, any potential

62 antidyskinetic effects are not necessarily solely attributed to monoamine reuptake inhibition mechansims. Furthermore, several studies have demonstrated toxicity to serotonergic neurons

(Xie et al., 2006; Ricaurte et al., 1988; Wilson et al., 1989; O’Hearn et al., 1988), highlighting the need for further research into promoting therapeutic properties of MDMA while reducing its toxicity and potential for misuse and abuse.

Analogues of MDMA have been produced and researched, with the analogue referred to as UWA-101 demonstrating significant potential as an antidyskinetic compound in MPTP- nonhuman primates and 6-OHDA-lesioned or reserpine-treated rats administered L-DOPA

(Huot et al., 2011; Johnston et al., 2012). Pharmacological differences between UWA-101 and its parent compound include UWA-101 demonstrating a 10-fold higher affinity than

MDMA for DAT (Ki = 1.3 + 0.9 M and 11.6 + 2.9 M, respectively), while retaining a high affinity for SERT (Ki = 0.5 + 0.1 M and 1.8 + 0.8 M, respectively). Furthermore, UWA-

101 demonstrated a >5-fold decrease in affinity for 5-HT2A receptor (>50 versus 11.4 + 3.3

M, respectively), and >3-fold decreased in affinity for the noradrenaline tranporter (NAT;

>50 versus 18.7 + 9.2 M, respectively) compared to MDMA (Johnston et al., 2012).

In particular, the R-enantiomer of UWA-101 referred to as UWA-121 has been shown to be responsible for the antidyskinetic effects observed in racemic UWA-101 administration, possibly through dual inhibition of the DAT and SERT (Huot et al., 2001, 2014).

Furthermore, as a MDMA analogue, investigations into UWA-101’s potential for toxicity and misuse have suggested a lack of psychoactivity and cytotoxicity in MPTP-nonhuman primates as demonstrated by preserving pre-pulse inhibition in rodents (which is reduced after exposure to psychoactive compounds) and inducing no cytotoxicity in human serotonergic cells (Johnston et al., 2012). While encouraging, studies into UWA-101 have not investigated the underlying antidyskinetic mechanism in vivo or have assessed properties that may predispose it to misuse and/or abuse. Therefore, further information is required to

63 improve the pharmacological profile of the racemic UWA-101 and even its R- and S- enantiomers before it can be considered for research in humans.

1.2.4 Glutamate receptors As dopamine-modulated GABA-ergic neurons in the basal ganglia are primarily driven by glutamate, it is conceivable that the changes in glutamatergic action would impact the firing rate of GABA-ergic MSNs and possibly influence the severity of dyskinesia in PD patients. Indeed, augmented tyrosine phosphorylation of NR2B subunits has been found to contribute to the enhancement of striatal glutamate NMDA receptor sensitivity and therefore contributes to the plastic alterations in dopaminergic responses in L-DOPA-treated parkinsonian rats (Oh et. al., 1998). In extension, drugs affecting glutamate receptors can influence the severity and duration of LID. The α-amino-3-hydroxy-5-methyl-4- isoxazolepropionic acid (AMPA) receptor positive allosteric modulator CX516 (100 mg/kg) induces dyskinetic movements in MPTP-treated primates when administered alone and produces dose-dependent increases in dyskinesia when co-administered with medium- or high-dose L-DOPA (Konitsiotis et. al., 2000). Additionally, glutamate receptor antagonists can reduce LID and LID-like behaviour in both humans and animal models (Konitsiotis et. al., 2000; Kobylecki et. al., 2010; Snow et al., 2000; Blanchet et. al., 1999). These results suggest that excessive glutamate signalling may correlate with LID severity. Research suggests that there are four main mechanisms by which the glutamate system influences LID:

NMDA receptor antagonism, AMPA receptor antagonism, metabolic glutamate subtype 5

(mGlu5) receptor antagonism, and glutamate release inhibition.

1.2.4.1. N-methyl-D-aspartate (NMDA) receptors The tetramic structure of ionotropic NMDA receptors comprise at least one NR1 and one NR2 (NR2A-D) subunit (Dingledine et al., 1999; Laube et al., 1998) and primarily linked

64 to calcium channels. Glutamate binds to the NR2 subunit, whereas the positive allosteric modulator glycine binds to the NR1 subunit (Paoletti & Neyton, 2007). Unlike other glutamate receptors, such as AMPA, metabotropic, or kainite receptors, glycine binding to

NMDA receptors is required to activate these receptors (Kleckner & Dingledine, 1998). The high affinity for glutamate binding in NMDA receptors (NR1/NR2A EC50= 2.89 + 0.12 µM,

Anson et al., 1998, NR2B EC50=1.5 + 0.6 µM, Laube et al., 1997) indicates these receptors would be a prime target of glutamate-dependent LID treatment.

Arguably, the most well-known research into the influence of NMDA receptors on

LID is that concerning amantadine, a NMDA receptor antagonist and one of the only drugs used clinically to reduce LID in PD patients. Encouraging findings in randomised, double- blind, placebo-controlled Phase III clinical trials (the EASE-LID and EASE-LID 3 trials) have been reported, with amantadine extended-release capsules (ADS-5102) taken once daily before bedtime significantly reducing total UDRS scores, OFF time, and on time with or without troublesome dyskinesia compared to placebo within 2 weeks of initiating treatment and was generally well tolerated (Elmer et al., 2018; Pahwa et al., 2017; Oertel et al., 2017).

Regarding safety and tolerability, adverse drug reactions were assessed for up to 25 weeks in the EASE LID study (Elmer et al., 2018; Pahwa et al., 2017) and up to 13 weeks in the EASE

LID 3 study (Elmer et al., 2018; Oertel et al., 2017). Pooled together (Elmer et al., 2018), most common ADRs (>10% of pooled patient data) reported by patients administered ADS-

5102 included hallucinations (visual or auditory), dizziness, dry mouth, peripheral oedema, constipation, falls, and orthostatic hypotension. Pooled data between the EASE LID and

EASE LID 3 studies (Elmer et al., 2018) demonstrated that ADS-5102-induced ADRs resulted in premature discontinuation of treatment by 20 patients (20%) compared to 8 patients (8.2%) of placebo-controlled patients. Amantadine significantly reduces choreic and dystonic LID in MPTP-treated primates during high and low dose L-DOPA treatment,

65 although amantadine also significantly reduced the duration and intensity of motor stimulation in low dose L-DOPA treatment (Blanchet et. al., 1999). Amantadine also reduces

LID severity by up to 50% as measured through clinical and subjective observations in acute double-blind, placebo-controlled studies of PD patients without affecting motor function or antiparkinsonian effects of L-DOPA (Luginger et. al., 2000; Crosby et. al., 2003; Del Dotto et. al., 2001; da Silva-Junior et. al., 2005; Snow et. al., 2000). Furthermore, amantadine produces long-term antidyskinetic effects in L-DOPA-treated PD patients administered amantadine for over a year (Wolf et. al., 2010).

However, other studies have discovered inconsistencies in amantadine’s efficacy for reducing the severity or frequency of LID. In MPTP-treated primates, amantadine was able to reduce LID significantly at 1.25 mg/kg when combined with a low dose of L-DOPA, but a dose of 2.5 mg/kg amantadine significantly reduced antiparkinsonian effects of L-DOPA as well as LID when combined with a low dose of L-DOPA and only mildly reduce LID when combined with a high dose of L-DOPA (Blanchet et. al., 1999). Amantadine’s efficacy has also been inconsistent in L-DOPA-treated PD patients, when LID scores returned to baseline after an average of 4.9 months of amantadine administration (Thomas et. al., 2004). These inconsistencies demonstrate important limitations in amantadine’s efficacy in reducing LID in PD patients.

Despite these limitations of amantadine’s therapeutic effects on LID, there is still strong evidence that NMDA receptors are involved in LID induction. Other NMDA receptor antagonists also reduce LID-like behaviour in animal models. For example, CI-1041 a novel

NMDA receptor antagonist selective for the NR1A/NR2B subtype completely prevented the induction of LID in MPTP-treated primates without any significant effect on the antiparkinsonian benefits of L-DOPA (Tahar et. al., 2004). There is some discrepancy in these results however, as maximum dyskinetic scores and minimum parkinsonian scores were

66 used instead of average scores for both measures. Researchers utilising the MPTP-primate model generally use average scores as a method of determining differences between groups

(van der Stelt et. al., 2005; Gomez-Ramirez et. al., 2006; Huot et. al., 2011). Therefore, such results should be interpreted with caution.

Additionally, Co 101244, a NMDA receptor antagonist selective for the NR2B receptor subtype, was found to have antidyskinetic effects in MPTP-treated primates

(Blanchet et. al., 1999). In contrast, MDL 100,453, selective for the NR2A receptor subtype, increased antiparkinsonian and dyskinetic effects of L-DOPA (Blanchet et al., 1999). It is therefore possible that NMDA receptor antagonists selective for the NR2B subtype have greater antidyskinetic efficacy in vivo. This is also supported by the similarities in behavioural effects between LY 235959 and Co 101244, which both have high affinities for the NR2B receptor subtype, in animal models of dyskinesia, while MDL 100,543 has a lower affinity for the NR2B subtype and has less antidyskinetic efficacy in LID animal models

(Blanchet et. al., 1999; Papa and Chase, 1996).

1.2.4.2 Alpha-amino-3-hydroxy-methyl-4-isoxazolepropionic acid (AMPA) receptor While NMDA receptors are the glutamate receptors primarily linked to calcium channels, there are a minority of AMPA receptors in the CNS lacking the GluR2 subunit that confers impermeability to Ca2+. As mentioned previously, an AMPA receptor positive allosteric modulator has been shown to induce dyskinesia in MPTP-treated monkeys, and to increase LIDs, although which subtype of AMPA receptor is responsible is unknown. Pertinent to this observation, is that a calcium-permeable AMPA receptor antagonism by IEM 1460 is effective in reducing LID in animal models (Kobylecki et. al., 2010), suggesting that there may be a role for

AMPA receptors without the GluR2 subunit in LID. Other researchers have found that co- administration of AMPA and NMDA receptor antagonists reduces L-DOPA-induced motor complications in both 6-OHDA-lesioned rats and MPTP-treated primates (Bibbiani et. al., 2005),

67 demonstrating that smaller doses of drugs that block more than one receptor can be more effective in reducing LID than each drug administered alone.

However, Bibbiani et al. (2005) measured only the number of rotations after drug administration in the 6-OHDA-lesioned rats, a measure primarily of postsynaptic dopamine receptor supersensitivity known as a wearing-off response, which has previously been argued to be separate to the dyskinetic measures of fore- and hindlimb (Metz & Whishaw, 2002). In addition to the rotations, this study might have employed measurements to assess the abilities of the limbs or measure the Abnormal Involuntary Movements commonly induced by high dose L-

DOPA administration (Dekundy et. al., 2007; Andersson, et al., 1999; Bishop et. al., 2006). The drugs between the rodent and primate experiments were not similar, with only GYKI-47261 being used in both species. Not only this, but in the primate experiment, only LID severity was measured, whereas LID frequency is also considered an important and reliable measurement.

While these findings implicate glutamate receptors involvement in altered L-DOPA-induced motor responses, they provided limited information on how glutamate receptor antagonists influence the duration and types of dyskinesia seen in either rats or primates.

Other avenues of LID research have also focused on glutamate release inhibition. It is incompletely known how glutamate release inhibiting drugs reduce LID severity and frequency and there is argument as to whether these drugs demonstrate this effect consistently. One such drug under intense study is riluzole. The mechanism of riluzole is not yet confirmed, but there is evidence to suggest that inactivation of voltage-dependent sodium channels, activation of G- protein-dependent signal transduction pathways, and non-competitive blockage of NMDA receptors may be involved in the riluzole-induced inhibition of glutamic acid release (Doble et al., 1996). However, evidence from other studies in rat cerebellar granule cells suggests that riluzole does not mediate glutamate release through sodium channels directly, as riluzole-induced inhibition of intracellular calcium evoked by NMDA or glutamic acid was still evident after

68 sodium channels were blocked by tetrodotoxin (Hubert et al., 1994). Instead, studies have proposed that diminution of glutamate release could be attributed to a riluzole-induced reduction in cytosolic calcium mediated by P/Q-type calcium channels in cerebrocortical nerve terminals of rats (Wang et al., 2004). Riluzole reduces axial, limb and orolingual AIMs by up to 33% at 4 mg/kg in 6-OHDA-lesioned rats (Dekundy et. al., 2007). Other studies have also found riluzole effective in reducing waking hours spent with LID by 24% overall and 30% for severe dyskinesia in six PD patients administered 50 mg/day riluzole during an open-label, six-week study (Merims et. al., 1999). In contrast, riluzole (100-200 mg/day) had no effect on LID severity in a double- blind, placebo-controlled study of 15 PD patients over 3 weeks (Bara-Jimenez, et. al., 2006).

Despite the inconsistent efficacy of riluzole, there are a number of differences between these studies that should be addressed:

1) the study by Merims et. al. (1999) was designed to measure a reduction in LID frequency, whereas the study by Bara-Jimenez et. al. (2006) was designed to measure the reduction of LID severity, particularly at an amplitude similar to that of amantadine,

2) the participants in the Merims et. al. (1999) study were administered 50 mg/day riluzole for 4 weeks, two weeks longer than the participants administered up to 200 mg/day riluzole in the Bara-Jimenez et. al. (2006) study.

Therefore, the results of these studies are not directly comparable, however, they do highlight the need for further research into the efficacy of riluzole in reducing LID in PD patients.

1.2.4.3 Metabotropic receptors In addition to ionotropic glutamate receptors, there are eight different types of metabotropic glutamate receptors (mGluR): mGlu1-8 (Foord et. al., 2005) also classified as

G-protein coupled receptors. Research has focused on the type 5 receptor (mGlu5) as a possible target for therapeutic antidyskinetic effects in LID. These effects are thought to be

69 possible as mGlu5 receptors are abundant within the striatum (Testa et. al., 1995) and mGlu5 receptor binding levels are increased in the striatum, particularly the putamen, of MPTP- treated primates treated with either saline or L-DOPA compared to controls as well as in post-mortem studies of PD patients also suffering from LID compared to nondyskinetic PD patients (Ouattara et. al., 2011; Samadi et. al., 2008). Additionally, MPTP-treated primates administered L-DOPA combined with CI-1041 (an NMDA antagonist) or cabergoline (a dopamine D2 antagonist) had mGlu5 receptor binding similar to controls and were less likely to develop dyskinesia (Ouattara et. al., 2011). This indicates that MPTP-induced striatal dopamine loss is related to the upregulation of mGlu5 receptor binding and that NMDA or

D2 receptor antagonism influences mGlu5 receptor binding as well as the likelihood of developing dyskinesia in MPTP-treated primates. Similar findings have also be reported in 6-

OHDA-lesioned rats, where an increase in mGluR5 uptake was noted in contralateral motor and somatosensory cortices following administration of L-DOPA and was correlated positively with increased AIM scores (Crabbe et al., 2018). It is possible, from these results, that striatal dopamine loss increases the density of mGlu5 which are then overactive during

L-DOPA administration and that this overactivity is linked to the stimulation of both NMDA and dopamine D2 receptors.

Not only is mGlu5 receptor binding increased after striatal loss of dopamine and L-

DOPA administration, but pharmacological studies have confirmed the influence of mGlu5 receptors in the development and expression of LID. Studies in animal models of LID have shown that mGlu5 receptor antagonism reduces L-DOPA-induced AIMs and reduces mGlu5 mRNA levels. In the 6-OHDA-lesioned rat, for example, the mGlu5 receptor negative allosteric modulator (NAM) 3-[(2-methyl-1,3-thiazol-4-yl)ethynyl] pyridine (MTEP) significantly reduced the development of AIMs without interfering with the antiparkinsonian effects of L-DOPA (Mela et. al., 2007; Rylander et. al., 2009). Additionally, MTEP reduced

70 the L-DOPA-induced upregulation of prodynorphin mRNA levels, indicating that MTEP normalizes the response of striatal neurons to L-DOPA in the ‘direct pathway’ (Mela et. al.,

2007).

While these results are encouraging, research into alternative mGlu5 receptor antagonists is needed. This is primarily indicated by one study finding that the peak antiparkinsonian benefit of L-DOPA (25-40 mg/kg) combined with MTEP (18 and 36 mg/kg) was significantly reduced compared to L-DOPA administered with vehicle to MPTP-treated primates (Johnston et. al., 2010), suggesting that MTEP might have a narrow therapeutic window or simply that its effects are not well translated between species. However, other mGlu5 receptor antagonists have shown promising antidyskinetic effects in MPTP-treated primates including ADX-48,621 (Hill et. al., 2010), fenobam (Rylander et. al., 2010) and

MPEP (another NAM, Morin et. al., 2010). The mGlu5 receptor antagonist AFQ056 was found to reduce LID scores during 1 hour peak periods and also reduced the maximum dyskinesia score for MPTP-treated primates, although it did not have any significant effect on the dyskinesia score for the total observation period (Gregoire et. al., 2011). This drug did not, promisingly, induce any significant effects on the antiparkinsonian effects of L-DOPA.

Indeed, AFQ056 attenuated developed LID in PD patients, although this effect was more prominent in patients with moderate to severe dyskinesia rather than only severe dyskinesia

(Berg et. al., 2011). However, other researchers have noted improvements in dyskinesia duration rather than severity as indicated through modified Abnormal Involuntary Movement

Scale (mAIMS) and the Unified Parkinson’s disease Rating Scale part IV (UPDRS-IV), and at 200 mg of AFQ056, dyskinesias were improved without worsening PD motor symptoms

(Stocchi et al., 2013). Although the effects for AFQ056 both in animal models of LID and clinical trials of PD patient appear to be limited, these results do provide additional support for involvement of mGlu5 receptors in LID development and expression in PD patients.

Other mGlu5 receptor NAMs have also made it to clinical trials, including mavoglurant and dipraglurant. A meta-analysis of 9 clinical trials (8 trialling mavoglurant and 1 trialling dipraglurant) demonstrated that mGlu5 receptor NAMs were more effective in reducing mAIMS compared to placebo in LID patients, but did not change UPDRS-IV scores (Wang et al., 2018). Although relatively well tolerated, patient groups administered the mGlu5R

NAMs reported a higher incidence of dizzinesss, visual hallucinations and fatigue compared to placebo patients.

Although it is unknown whether the increase in mGlu5 receptor levels occurs primarily on the neurons in the ‘direct pathway’, it could explain the increased in glutamatergic activity in the direct pathway that influences LID development and severity

(Samadi et. al., 2008; Luginger et. al., 2000; Crosby et. al., 2003; Kobylecki et. al., 2010).

However, research into this area is still needed to determine whether any unexpected effects occur after chronic mGlu5 receptor antagonist administration in dyskinetic PD patients, such as tolerance or adverse effects and whether other antagonists might have greater antidyskinetic efficacy or larger therapeutic windows than drugs such as AFQ056.

Considering the influence of the glutamatergic system on dopaminergic activity and dopamine release, it seems logical that pharmacologically manipulating glutamate in the basal ganglia would influence the expression and severity of LID. Based on the results from previous studies, glutamatergic antidyskinetic therapies appear to be the most promising non- dopaminergic avenue of research for LID. In particular, NMDA receptor antagonists have been shown to be most effective, although it could be suggested that a combination of glutamate receptor compounds might exert an additive or synergistic effect. This would be especially beneficial in terms of reduced side effects due to the lower doses needed.

However, the inconsistent efficacy of amantadine and riluzole in clinical trials and the reduction in antiparkinsonian effects after mGlu5 receptor antagonist administration

72 demonstrates that, while the glutamatergic system is one of the most promising avenue for antidyskinetic research, other avenues of research should also be considered to assist in the understanding of the mechanism underlying LID and also to increase the range of therapeutics available so that treatment can be as individualistic as possible.

1.2.5 Adrenergic receptors Adrenergic receptors are classified as G protein-coupled receptors that are bound by adrenaline and noradrenaline. Adrenergic receptors are divided into alpha or beta receptors and can be further identified as specific subtypes: the alpha receptor subtypes are α1A, α1B,

α1D, α2A, α2B, and α2C, while the beta adrenergic receptor subtypes include β1, β2, and β3.

Although less well researched than other LID-mediating receptors, adrenergic receptors have also been shown to influence LID emergence and severity in humans and animal models. For example, moderate lesioning of the noradrenaline system in the striatal and hippocampal regions of Sprague-Dawley rats can reduce the development and expression of AIMs and rotation, without influencing L-DOPA efficacy in the forepaw adjusting step (FAS) test

(Barnum et. al., 2012).

There is still controversy as to the depth of involvement of some adrenergic receptors in LID in humans. Although post-mortem studies have found that, compared to non-PD controls, α1 and β1 receptor densities are increased and α2 receptor densities are decreased in

PD patients during disease progression, these changes were found in the frontal cortex which is less involved in LID emergence than structures such as the basal ganglia (Cash et. al.,

1984). Additionally, the density of β1 receptors was not significantly different in the basal ganglia in a small sample of PD patients compared to controls (Waeber et. al., 1991; Hara et al., 2010). Moreover, α2-adrenergic receptors are present on presynaptic noradrenergic terminals and post-synaptic GABA-ergic medium sized spiny neurons in the striatum (Savola

73 et. al., 2003). These studies present a greater argument for the involvement of α-adrenergic receptors over β receptors.

As α-adrenergic receptor densities are higher in the striatum compared to β receptor densities, much research has been focused on the influence of α-adrenergic receptors in LID expression, although this does not discount the therapeutic potential for β adrenergic receptors as a target for LID treatment. It is thought that the stimulation of α2A and α2C receptors facilitate activation of direct striatal pathways, increasing the risk for LID development (Gottwald & Aminoff, 2011). The α2-adrenergic receptor antagonist yohimbine and α2-adrenergic receptor agonist clonidine have been shown to reduce LID severity in

MPTP-treated primates (Gomez-Mancilla et. al., 1991) and 6-OHDA-lesioned rats (Dekundy et. al., 2007). Additionally, fipamezole (90 mg), another α2 adrenergic receptor antagonist was found to reduce LID without exacerbating parkinsonian symptoms in 115 patients during a phase 2B double-blind, placebo-controlled study (LeWitt et. al., 2012). Although fipamezole displayed antidyskinetic effects, this was only seen in the American (United

States, US) subgroup, but not the Indian subgroup. The authors could not explain this difference, possibly indicating that fipamezole might have antidyskinetic effects in only certain populations. Additionally, there have been no further clinical trials of this drug since

2011, indicating that the company did not consider it worthwhile pursuing for Parkinson’s disease, after three studies that completed between 2008 and 2011 (NIH, ClinicalTrials.gov).

Fipamezole is also a SERT inhibitor and histamine H3 receptor agonist, both mechanisms of which have antidyskinetic actions (Durif et. al., 1995; Gomez-Ramirez et. al., 2006).

Antidyskinetic actions of histamine receptor agonists seem to be limited, as co-administration of L-DOPA with immepip or imetit (H3 receptor agonists) significantly reduces chorea, but not dystonia in MPTP-treated primates and immepip administered as a monotherapy

74 significantly increased cumulated parkinsonian disability, a rating scale combining measures of range of movement, bradykinesia, posture and alertness (Gomez-Ramirez et. al., 2006).

Other adrenergic compounds have also shown efficacy in mediating LID, including β receptor antagonists. Both idazoxan and propranolol, α2 and β adrenergic receptor antagonists, respectively, reduced L-DOPA-induced AIMs in 6-OHDA-lesioned rats, although only propranolol was able to reduce AIMs at both doses of L-DOPA (4 and 12 mg/kg; Barnum et. al., 2012). Barnum et al. (2012) also reported that LIDs were reduced by moderate depletion of noradrenaline, indicating that increases in noradrenaline interacting at

α2 and/or β adrenergic receptors can increase LIDs. Idazoxan (10 mg/kg) has also been effective in reducing LID in MPTP-treated primates, as well as enhancing the antiparkinsonian activity of L-DOPA (Henry et. al., 1999). Although there was no evidence of enhanced antiparkinsonian response to L-DOPA, single oral doses of idazoxan (20mg) improved the severity of LID in 18 patients without deteriorating the antiparkinsonian response to L-DOPA (Rascol et. al., 2001b). One limitation in this study was that, although it was placebo-controlled, there was no mention of a double-blind procedure to reduce experimental bias. Nonetheless, the antidyskinetic effects of idazoxan are evident in both pre- clinical and clinical studies. However, despite all the preclinical and early clinical evidence from late 1990s to 2011, no  receptor antagonists have been developed successfully for the treatment of L-DOPA-induced dyskinesias, suggesting that they have limited usefulness.

Despite the clear evidence that adrenergic receptors can exert influence over LID expression and severity, few studies have focused on the actual mechanism behind this effect.

One possible explanation for the reduction in LID after α-adrenergic receptor antagonist administration is that these compounds reduce striatal dopamine and L-DOPA levels. Indeed, this has been found in unilaterally 6-OHDA-lesioned rats in which idazoxan (9 mg/kg) reduced extracellular dopamine and L-DOPA levels in both the lesioned and intact striatum

75 as measured by the area under the curve (Buck et. al., 2010). However, there were no significant differences found between vehicle and idazoxan (3 and 9 mg/kg) during the time course analysis, but significant differences were found in the time course of dyskinesia animals treated with idazoxan compared to controls. These discrepancies indicate that, while

α-adrenergic receptor antagonists may reduce the extracellular levels of dopamine and L-

DOPA, further research is required to clarify the relationship of these effects to LID.

As previously mentioned, although β adrenergic receptors are less prominent in the basal ganglia of humans, β antagonists have been useful in reducing LID. Antagonists such as propranolol (Dekundy et. al., 2007; Barnum et. al., 2012; Lindenbach et. al., 2011) have shown great efficacy in reducing LID in animal models. There is question of whether propranolol affects the antiparkinsonian action of L-DOPA, however, as propranolol (20 mg/kg) reduced rotarod performance in 6-OHDA-lesioned rats after L-DOPA administration

(Dekundy et. al., 2007). In contrast, other studies in rats have found that propranolol (20 mg/kg) can increase antiparkinsonian response to L-DOPA as measured through the FAS test and distance travelled (cm) as well as reduce vertical activity (Lindenbach et. al., 2011) which has been suggested to correlate with dyskinesia-like behaviours in rats (Johnston et. al., 2005).

Although there are discrepancies concerning the effect of β-adrenergic receptor antagonists on the antiparkinsonian actions of L-DOPA, to our knowledge there have been no reports of reduced antiparkinsonism after α-adrenergic receptor antagonist administration.

Indeed, clinical studies for these compounds have reported few adverse side effects while reporting significant decreases in LID, although only in specific populations of patients.

However, there is no recently completed or currently recruiting clinical trials testing effects of adrenergic antagonists for LID on the USA clinical trials registry. Therefore, while these

76 compounds induce antidyskinetic effects in animals, it appears less likely that they will be a large focus for clinical research into antidyskinetic compounds.

1.2.6 Cannabinoid receptors Cannabinoid receptors are G protein-coupled receptors and have also been implicated as a possible therapeutic target for LID. These receptors are primarily presynaptic commonly found on glutamatergic (Gerdeman & Lovinger, 2001; Katona et al., 2006) and GABA-ergic

(Szabo et al., 1998; Katona et al., 2006) terminals and inhibit neurotransmitter release

(Gerdeman & Lovinger, 2001; Szabo et al., 1998). The two main subtypes of cannabinoid receptors are CB1 and CB2 receptors, CB1 receptors appear to mediate the majority of cannabinoid effects in the central nervous system (Mackie, 2005). CB1 receptor signalling seems to play a large role in dopamine-mediated mechanisms within the basal ganglia (Di

Marzo et. al., 2000); indeed, CB1 receptors are most highly concentrated in the basal ganglia, particularly the SNr, which is part of the output pathway of the basal ganglia, and the striatum of rats, rhesus monkeys and humans, but not on the dopamine neurons themselves in the SNc or on their terminals (Herkenham et. al., 1990; Macki, 2005). All the cells in the striatum that are positive for CB1 receptor mRNA are both types of GABA-ergic projection neurons, with few intrinsic interneurons expressing CB1 receptor mRNA, and no GABA-ergic interneurons expressing CB1 receptor mRNA (Hohmann & Herkenham, 2000). Since CB1 receptors are primarily on terminals, any receptors expressed by this mRNA would not be in the striatum, but in the output pathways (primarily globus pallidus and SNr). There are high densities of

CB1 receptors in the striatum, highest in the lateral striatum, and many of these are on corticostriatal glutamate terminals (Van Waes et al., 2012), consistent with the finding that activation of CB1 receptors in the striatum primarily reduces glutamate release (Gubellini et. al., 2002; Gerdeman & Lovinger, 2001). In line with this, it has also been reported that CB1 receptors are important in long-term depression (LTD) in corticostriatal synapses (Gerdeman

77 et al., 2002). There are also CB1 receptors on GABA-ergic terminals (Kofalvi et al., 2005), presumably from cells extrinsic to the striatum as there no CB1 receptor mRNA is expressed in GABA-ergic interneurons of the striatum (Hohmann & Herkenham, 2000). There is no direct effect of CB1 receptor agonists on dopamine release or GABA transport (Kofalvi et al.,

2005).

As the indirect pathway of the basal ganglia is hypothesised to be overactive during progression of PD, it is possible that this overactivity is due to excess glutamate release from corticostriatal neurons, leading to excess GABA-ergic release from striatal neurons projecting to the GPe resulting in inhibition of external nuclei and reduced movement. As CB1 receptor stimulation decreases the release of glutamate, it is possible that CB1 receptor agonists selective for striatal CB1 receptors could decrease glutamate signalling in corticostriatal neurons leading to disinhibition of structures such as the thalamus, thus exerting antiparkinsonian effects. Additionally, as CB1 receptor inhibition decreases GABA release, CB1 receptor antagonists selective for CB1 receptors located in the GPe would most likely increase extracellular GABA levels in the striatum to also inhibit GABA-ergic projection neurons nuclei, thus also exerting anti-parkinsonian effects. Therefore, CB1 receptor agonists have potential as LID therapeutics and antagonists have potential as targets for antiparkinsonian therapy.

As previously discussed, LID is hypothesised to emerge through overactivity of the direct pathway in the basal ganglia. This would most likely occur through increased glutamatergic transmission in the striatum, leading to increased GABA-ergic activity in the

GPi thereby disinhibiting structures such as the thalamus, leading to dyskinesia. Therefore, it is suggested that CB1 receptor agonists would also assist in LID therapy by reducing the glutamate release in the corticostriatal neurons and decrease striatal activity.

Although, selectivity of CB1 receptors by agonists (in the striatum) or antagonists (in the GPi) would theoretically be needed to reduce the expression and severity of LID, clinical and non-clinical studies have identified possible candidates for LID reduction in conjunction with dopamine replacement therapy, without affecting the antiparkinsonian actions of L-

DOPA. The CB1 agonist WIN 55,212-2 dose-dependently (0.5 and 1.0 mg/kg) attenuates L-

DOPA-induced axial, limb and oral AIMs in 6-OHDA-lesioned rats (Morgese et. al., 2007).

Indeed, the CB1 receptor agonist, nabilone, effectively reduces LID in both MPTP-treated primates (Fox et. al., 2002) and PD patients (Sieradzan et. al., 2001) at 0.1 mg/kg and 0.3 mg/kg, respectively. Although the clinical trial was designed as a double-blind, placebo- controlled pilot study, the sample size of seven participants might not have enough statistical power to accurately reflect the effects of nabilone. Additionally, this was an acute study, in which participants were given two challenges of L-DOPA (one session in conjunction with nabilone and the other with placebo), and thus more information is needed regarding side effects, drug interactions, and the possibility of wearing-off effects.

Both the agonist, WIN 55,212-2 (1.0 m/kg) and the CB1 receptor antagonist, rimonabant (SR141716; 1.0 mg/kg), reduce L-DOPA-induced vertical activity without affecting L-DOPA-induced horizontal activity in reserpinised rats, measurements that are proposed to represent a correlation with dyskinetic and antiparkinsonian activity, respectively

(Segovia et. al., 2003). This is intriguing considering these drugs have opposing effects on the

CB1 receptor. The authors speculated that the differing ability of each drug to modulate activity of the direct and indirect striatal pathways may underlie this finding. This hypothesis was not well supported by the additional findings that WIN 55,212 (1 mg/kg) reduced horizontal activity and mobile time in rats treated with either the dopamine D1 receptor agonist CI-APB or the dopamine D2 receptor agonist quinipirole, but SRI 41716 had no significant effect on activity induced by either dopaminergic compound. These results do

79 little to help identify the mechanism in which SR 41716 reduced L-DOPA-induced activity in rats and future research would be needed to investigate this further. However, WIN 55,212-2

(3.0 mg/kg) seems to compromise L-DOPA’s antiparkinsonian effects as indicated by a significant reduction in L-DOPA-induced horizontal activity, although rimonabant seems to effectively reduces LID in MPTP-treated marmosets and reserpine-treated rats without compromising the antiparkinsonian effects of L-DOPA (van der Stelt et. al., 2005).

Other CB1 receptor antagonists seem to have less consistent antidyskinetic effects in animal models of LID. For example, the selective CB1 antagonist 1-[7-(2-chlorophenyl)-8-(4- chlorophenyl)-2-methylpyrazolo[1,5-a]-[1,3,5]triazin-4-yl]-3-ethylaminoazetidine-3- carboxylic acid amide benzenesulfonate (CE; 0.03-1 mg/kg) had no effect on LID in MPTP- treated rhesus monkeys, but did increase response duration to L-DOPA (suboptimal doses:

50-100 mg) by 30% in co-administration trials (Cao et. al., 2007). These results demonstrate that the cannabinoid CB1 receptor could be a potential target for pharmacological treatment of LID with agonists, but perhaps compromise the therapeutic effects of L-DOPA, while the antagonists may have benefit for PD, without worsening dyskinesia.

1.3 Animal models of LID Although studies over the two decades have made significant progress in possible mechanisms for LID, further research is needed as no new treatments have eventuated from this knowledge (Utsumi et al., 2013; Del Dotto et al., 2001; Da Silva-Junior et al., 2005).

Studies that have employed animal models of LID have provided insight into the neurotransmitter systems that influence the expression and severity of LID and have been useful in demonstrating the efficacy of antidyskinetic compounds when administered with L-

DOPA (Johnston et al., 2005; Dekundy et al., 2007). Animal models of LID have also been proposed in the last decade or so to provide a screening process for potential antidyskinetic

80 compounds before further pre-clinical or clinical studies are conducted. However, there is some question about the translation of the face validity of these models in identifying mechanisms underlying LID, and therefore possible therapeutic avenues, from animal models to the predictive validity of clinical improvement in PD patients suffering from LID (Lane &

Dunnett, 2008).

There are currently several lesion or genetic-based animal models of LID, in combination with repeated L-DOPA treatments. Lesion animal models of PD and LID are primarily based on neurotoxins either destroying the neurons in the dopaminergic pathway to the striatum, or creating a dopamine deficiency through inadequate monoamine packaging in the neurons. Genetic animal models of PD and LID are founded on the identification of genes known to contribute to PD, however as only about 3-5% of sporadic cases and about 30% of familial cases of PD in humans can be explained by genetics (Klein & Westenberger, 2012), these models are limited in their ability to accurately reflect the features of these conditions in humans. Nonetheless, both lesion and genetic animal models of LID have been invaluable for obtaining new knowledge for possible LID treatments. Lesion models more commonly use both rodents and primates as rodent models are excellent for screening potentially antidyskinetic compounds while primates are able to display dyskinetic symptoms highly representative of those found in humans. Similarly, genetic models of PD and LID more commonly utilise rodents as they provide a pragmatic and representative form of genetic manipulation.

1.3.1 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated primate One of the most common and robust models of LID is the MPTP-treated primate model. This neurotoxin was discovered when young drug-dependent adults developed an idiopathic parkinsonian syndrome after intravenous self-administration of a “synthetic heroin” known as 1-methyl-4-phenyl-4-propionoxypiperidine (MPPP; Langston et al., 1983;

Davis et al., 1979). While the MPPP was not thought to induce this parkinsonian syndrome, researchers concluded that ill-made batches of MPPP also contained MPTP, compound resulting from the dehydration of 4-hydroxy-4-phenyl-N-methylpiperidine (HPMP), a compound involved in the chemical reaction process, and MPPP (Davis et al., 1979). These drug-dependent adults displayed biochemical, neuropathological and clinical characteristics of which most were almost identical to the cardinal symptoms of human PD, with the exception of Lewy bodies (Langston et al., 1983). Due to these similar symptoms, MPTP is considered one of the most important and frequently used parkinsonian toxins administered in animal models as it causes direct toxicity of dopaminergic neurons, specifically, and it induces symptoms in humans that are closely reminiscent of those seen in PD patients

(Langston et al., 1983; Davis et al., 1979).

MPTP crosses the blood-brain barrier in primates, and crosses with much reduced concentration in rodents (Riachi et al., 1991; Andersen et al., 1994), where it is converted into

1-methyl-4-phenyl-2,3-dihydropyridium (MPDP) in non-dopaminergic cells (particularly serotonergic neurons) by monoamine oxidase B (MAO-B). It then spontaneously oxidizes to form 1-methyl-4-phenylpyridinium (MPP+; Vila & Przedborski, 2003). MPP+ is then released into the extracellular space (mechanism unknown) and, due to its high affinity for the dopamine transporter (DAT), is taken up into the dopaminergic cells through carrier transport systems on the plasma membrane. Researchers therefore used this compound to create a new model of PD in primates, and later, LID after chronic subcutaneous administration of L-

DOPA. Primates are more sensitive to the effect of MPTP compared to other animals

(Schmidt & Ferger, 2001), and as they develop dyskinesia closely resembling those seen in humans (Gomez-Ramirez et al., 2006; Iderberg et al., 2012), the MPTP-treated primate model is an excellent tool for studying LID.

The MPTP-lesioned primate model encompasses all the major behavioural attributes of human PD, including abnormal posture, akinesia, rigidity and postural tremor (Iravani et al., 2003; Cao et al., 2007) as well as non-motor symptoms such as cognitive impairment

(Schneider & Kovelowski, 1990; Schneider, 1990) and sleep/wake disorders (Almirall et al.,

1999). The duration of MPTP administration can range anywhere between a few weeks

(Luquin et al., 1992) and several months (Schneider & Kovelowski, 1990; Schneider, 1990).

After moderate to severe parkinsonism has been firmly established, the primates undergo L-

DOPA administration until dyskinesia emerge, which generally takes one to two months

(Iravani et al., 2003; Gomez-Ramirez et al., 2006).

Motor activity can sometimes be measured through infrared photo sensors attached to the cages; however, behaviour is generally recorded on video and each animal is given a parkinsonian score and a dyskinesia score, although this also requires inter-observation reliability assessment. Parkinsonian scoring systems are used to assess motor disability in the form of alertness, checking movements, posture, balance, motility, reactions to stimuli and vocalization (Iravani et al., 2003; Savola et al., 2003) as well as bradykinesia and disability

(Savola et al., 2003; Huot et al., 2011) most commonly with an overall maximum parkinsonian disability score of 36. Dyskinesia scoring is generally rated from 0 = absent, 1 = mild, fleeting, rare throughout the observation period; 2 = moderate, not interfering with normal activity, more prominent throughout the observation period; 3 = marked, frequent and, at times, continuous dyskinesia interfering with normal activity; 4 = severe, virtually continuous, essentially replacing normal activity (Savola et al., 2003; Henry et al., 1999). The two most common primates used in this MPTP model are marmosets and macaques (rhesus and cynomolgus; Savola et al., 2003; Huot et al., 2011; Iravani et al., 2003), however some research has been conducted on the MPTP-treated squirrel monkey (Boyce et al., 1990; Quik et al., 2007).

Marmosets are often utilized in the MPTP-treated primate model of LID due to their small size and convenient housing requirements as well as their display of robust L-DOPA- induced dyskinetic-like behaviours. The MPTP-treated marmoset model has high face validity for modelling LID as animals display both choreic and dystonic dyskinetic behaviours. In addition, previous research has shown that certain antidyskinetic compounds can selectivity reduce chorea, but not dystonia, in these animals (Gomez-Ramirez et al.,

2006), suggesting that these different dyskinetic behaviours are not all modulated by the same neural pathways.

The MPTP-treated marmoset model has been also been shown to have some predictive validity in identifying antidyskinetic compounds for LID therapy; fipamezole displayed significant antidyskinetic effects after L-DOPA was repeatedly administered to

MPTP-treated marmosets (Savola et al., 2003) and was later found to significantly reduce dyskinesia in PD patients (Lewitt et al., 2012). This predictive validity, however, is limited as compounds such as idazoxan, while effective in reducing dyskinesia in MPTP-lesioned primates (Henry et al., 1999), had inconsistent effects in clinical trials in terms of antidyskinetic efficacy and tolerability (Rascol et al., 2001b). Other drugs such as 3,4,- methylenedioxymethamphetamine (MDMA) and its analogues have also been reported to reduce dyskinesia (Iravani et al., 2003; Johnston et al., 2012) as well as extend L-DOPA ON time (Huot et al., 2011). MDMA has been shown to reduce dyskinesia PD patient case studies

(BBC, 2001), although due to cytotoxicity and psychoactive effects, it is difficult to rigorously test these effects in ethically-approved trials.

Although MPTP-treated macaque primates (rhesus and cynomolgus monkeys) are larger in size and are less convenient to house, the face validity of this model is higher than that of MPTP-treated marmosets. In particular, macaques present choreic movements

(constant writing and jerking motions), dystonic movements, and ballistic movements (large-

84 amplitude flinging and flailing), although ballistic movements tend to occur less often

(Bezard et al., 2003). These movements are exemplary in their similarity to dyskinesia seen in humans, and can be measured using the Dyskinesia Disability Scale (Bezard et al., 2003).

As with humans, titration of L-DOPA (Hely et al., 2005); Rezak, 2007) and antidyskinetic drugs (Thomas et al., 2004) is often required in PD and LID primate studies, indicating that some individual subjects in either macaque or human studies may be more susceptible to dyskinesia development (Bezard et al., 2003).

MPTP administration is relatively simple compared to neurotoxins used in other models of LID. Macaques can be treated with MPTP subcutaneously (Blanchet et al., 1999;

Tahar et al., 2004; Johnston et al., 2010; Riahi et al., 2011) or intravenously (Cao et al., 2007;

Luquin et al., 1992; Munoz et al., 2008; Schneider & Kovelowski, 1990) to maximize the effect of the drug by avoiding peripheral metabolism. Additionally, macaques can be administered L-DOPA orally using human medications such as Sinemet (L-

DOPA/Carbidopa; Blanchet et al., 1999; Cao et al., 2007) and Prolopa (L-DOPA/

Benserazide; Tahar et al., 2004; Riahi et al., 2011), simulating the route of administration most commonly used in humans (Stocchi et al., 2002; Stocchi, 2006; Peaston & Bianchine,

1970) and thus increasing the model’s face validity of LID.

Despite the effectiveness of the MPTP-treated primate model, there has been some comment of its face validity. One of the main drawbacks to this model is that, physiologically, the death of the dopaminergic cells is much quicker compared to PD in humans as the effects of MPTP can be seen in a few months rather than years. However, this does have the advantage of lessening the cost of the overall study in terms of time taken to complete the experiment, care of the animals, and drug costs. Despite the more acute degeneration of dopaminergic neurons in MPTP-treated primates, this model of LID has an advantage over others in that the physiological similarities in neural structures as well as the

85 observational similarities in dyskinetic movements after chronic L-DOPA administration between primates and humans are greater than any other animal model currently employed in

LID research.

1.3.2 6-hydroxydopamine (6-OHDA) lesion models 1.3.2.1 6-OHDA-lesioned rat Although MPTP is an excellent neurotoxin for animal models of PD and LID, this compound induces milder loss of nigral tyrosine hydroxylase-immunostained cells and striatal dopamine depletion in rats than 6-OHDA (Ferro et al., 2005). Although 6-OHDA- and

MPTP-lesioned rats express similar deficits in memory tasks, and few exploratory and motor alterations in open fields and catalepsy tests, locomotor activity tends to increase in MPTP- lesioned rats and decreased in 6-OHDA-lesioned rats (Ferro et al., 2005). Thus, the MPTP- treated rat is generally considered a model for early stage PD and the 6-OHDA-treated rat are used as models of late stage PD and therefore is more commonly used (Munoz et al., 2008;

Meissner, et al., 2006; Barnum et al., 2012; Kobylecki et al., 2010).

6-OHDA is a hydroxylated analogue of dopamine and a potent neurotoxin that induces degeneration of monoamine neurons, including dopaminergic neurons (Ungerstedt,

1968). Because 6-OHDA destroys sympathetic nerve terminals in the peripheral nervous system when administered systemically (Porter et al., 1963), it is commonly injected intracranially into the substantia nigra, medial forebrain bundle, or striatum of rats, and taken up by dopamine neurons (Ungerstedt, 1976; Dekundy et al., 2007; Munoz et al., 2008).

Furthermore, intracranial administration avoids the induction of bilateral effects that lead to significant impairments of the animal including aphagia and adipisia often necessitating artificial feeding to reduce severe loss of body weight (Arnt, 1985, Ungerstedt, 1971). The extent of the lesion can be measured through apomorphine-induced rotations (Ungerstedt

1976; Ungerstedt & Arbuthnott, 1970) or the actual lesion size can be measured by

86 determining the percentage of dopamine depletion in post-mortem tissue (Castaneda et al.,

1990). Previously, researchers measured the antiparkinsonian effects of chronically administered L-DOPA through rotations, both direction and number rotations (Ungerstedt

1976; Ungerstedt & Arbuthnott, 1970). However, there was criticism regarding the use of this method to measure LID in animal models. Indeed, the correlation between amphetamine- induced rotations and LID in 6-OHDA-lesioned rats was poor (Tronci et al., 2012) and while increases in rotation correlated to bromocriptine- or L-DOPA-induced AIMs, the absolute number of rotations showed no statistically significant correlation with AIMs (Lundblad et al., 2002).

Therefore, a more representative measurement of LID was needed in this model in which behaviours were induced by L-DOPA, affected the side of the body contralateral to the lesion, and were repetitive, purposeless and not ascribable to normal behaviour patterns as seen in humans (Cenci et al., 1998). Rather than measuring rotation, researchers began to analyse abnormal involuntary movements (AIMs) developed in 6-OHDA-lesioned rats after chronic L-DOPA treatment that had striking similarities to dyskinetic movements in L-

DOPA-treated PD patients and a scale was developed to quantify these movements in 6-

OHDA-lesioned rats (Cenci et al., 1998; Cenci & Lundblad, 2007). The AIMs are divided into four subtypes: locomotive (increased locomotion with contralateral side bias), axial dystonia (contralateral twisting, posturing of neck and upper body), orolingual dyskinesia

(stereotyped jaw movements and contralateral tongue protrusions), and forelimb dyskinesia

(repetitive, rhythmic jerks or dystonic posturing of the contralateral forelimb and/or grabbing movement of the contralateral paw) and are rated on a severity scale from 0 to 4 (Cenci et al.,

1998; Cenci & Lundblad, 2007). Compared to analysis of L-DOPA-induced rotations, this rating scale has greater accuracy in measuring and predicting the dyskinetic effects of L-

DOPA and the antidyskinetic efficacy of other compounds in 6-OHDA-lesioned rats

(Dekundy et al., 2007; Andersson et al., 1999).

Since this implementation of the AIMs scale as a measurement of LID-like behaviour in rodents, much research has been conducted to validate this model and identify compounds which reduce AIMs (Andersson et al., 1999; Eskow et al., 2007; Delfino et al., 2004;

Morgese et al., 2007). Indeed many compounds found to reduce LID in humans have also been used to validate the 6-OHDA-lesioned rat model of LID. For example, amantadine, clozapine and yohimbine all reduce AIMs in 6-OHDA-lesioned rats and reduce LID in patients with PD (Dekundy et al., 2007). Therefore, the 6-OHDA-lesioned rat model could be used as a screening test for identifying compounds that reduce the adverse effects of L-DOPA in animals and indicate the antidyskinetic compounds that might also be effective in humans.

Most studies focus on the effects of drugs in rats with unilateral lesions, also known as the ‘hemiparkinson’ model (Dekundy et al., 2007; Delfino et al., 2004; Morgese et al.,

2007), although bilateral lesions are also possible (Castaneda et al., 1990). Unilateral lesions are generally preferred to bilateral lesions for a number of reasons: 1) the expense of a unilateral lesion is generally lower than that of bilateral lesion surgeries, primarily because the quantity of 6-OHDA required is less with a unilateral surgery, 2) bilateral 6-OHDA lesions can induce severe motor reduction, this usually results in the animals requiring intensive care which can also be expensive as well as time consuming, 3) unilateral lesions also provide a within-subject control in the form of the non-lesioned side of the basal ganglia, thus behavioural effects of the lesions or proceeding drugs administered can be compared contra- or ipsilateral to the lesion (Cenci et al., 2002). However, bilateral lesions are sometimes advantageous as the lesioning of both sides of the brain induces essential elements of parkinsonian motor syndromes (Castaneda et al., 1990; Cenci et al., 2002).

The AIMs rating scale in 6-OHDA-lesioned rats has improved the accuracy of measuring the efficacy of potentially antidyskinetic compounds. There are supportive consistencies between antidyskinetic effects of compounds in the 6-OHDA-lesioned rat and

PD patients (Dekundy et al., 2007) and researchers have developed a greater understanding of neural and molecular mechanisms underlying LID because of this advancement (Kobylecki et al., 2010; Munoz et al., 2008). However, this model is not without its drawbacks: primarily, the 6-OHDA-lesioned rat model is representative only of the late stages of PD after severe degeneration of dopaminergic neurons has occurred. This is because of the acute degeneration >90% of the dopaminergic neurons in the SNpc of the rats, rather than progression cell death as is characterized by PD. Therefore, caution is required when interpreting the results of 6-OHDA-lesioned rat studies as these outcomes may not apply to all stages of PD.

Another test used in this model is the forepaw adjusting step (FAS) test which quantifies forelimb akinesia, a consequence of the lesion representative of parkinsonian symptoms. Rats with >80% unilateral dopamine depletion perform poorly on the test with the forepaw contralateral to the lesion (Chang et. al., 1999; Bishop et al., 2009), while L-DOPA reduces this deficit. Therefore, the test indicates whether any additional compounds administered to the rats compromises the antiparkinsonian effects of L-DOPA (Eskow et. al.,

2007). Generally the score for percent intact stepping is derived by summing the total steps with the forepaw contralateral to the lesion, dividing by the number of steps with the forepaw ipsilateral to the lesion, and multiplying this number by 100. Lower percent intact scores indicate greater forelimb akinesia. However, this test does require separate acclimation periods before administration of test compounds to reduce potential practice effects.

Other complementary scoring systems used in 6-OHDA rat models include food handling analysis, gait symmetry analysis, tactile stimulation analysis (Proft et al., 2011), and

“dyskinesia amplitude” (Winkler et al., 2002; Rylander et al., 2010; Breger et al., 2013).

Food handling analysis involves measuring the time taken for rats to eat small pieces of uncooked pasta as this task allows abnormalities of posture, limb and mouth movements to be observed after a 6-OHDA lesion (Allred et al., 2008; Proft et al., 2011; Whishaw et al.,

1997). Time measurements begin with the first contact of the snout with the pasts until the entire piece is consumed and both paws return to the ground (Proft et al., 2011). A rating scale from 0 (eating time < 60 seconds) to 3 (eating time >120 seconds) is used to categorise the food handling with an average of three tests being the end result. Gait symmetry, allowing the rat to roam freely on a table top rather than remain in a transparent cylinder, can be paired with spontaneous turning preference (as described by Cenci et al., 2002 and Dekundy et al.,

2007) as a predictor of neostriatal dopamine depletion (Metz et al., 2005; Fornaguera &

Schwarting, 1999; Proft et al., 2011). Lastly, the tactile stimulation test measures the responsiveness of a rat to light touch after a 6-OHDA lesion (Proft et al., 2011) as rats with a unilateral 6-OHDA lesion are known to display biased responses to tactile stimulation

(Schallert & Hall 1988). Dyskinesia ampltitude is definied as the degree of deviation of a dyskinetic body part from its natural resting position (Cenci & Crossman, 2018). Amplitude scores are used in axial and forelimb AIMs, range from a score of 1 (“consistent lateral deviation of head and neck at ~30o angle” or “tiny oscillations of the paw and distal forelimb around a fixed position”, respectively) to a score of 4 (“torsion of head , neck and trunk at

>90o angle, causing rat to lose balance” or “vigorous limb and shoulder movements of maximum amplitude, may have a ballistic character”, respectively; Winkler et al., 2002;

Rylander et al., 2010; Breger et al., 2013) Using a collection of tests for detecting parkinsonism or dyskinetic behaviours increases the robustness and strength of the study and differentiate what deficits are improved with the therapy being investigated.

Despite the sufficient face, predictive, and construct validities of the 6-OHDA- lesioned rat as a model of LID, there are some major drawbacks to consider. Of course it is difficult for any model to be truly representative of LID, however the degeneration of dopaminergic neurons in the 6-OHDA-lesioned rat is acute compared to the more progressive emergence of Parkinsonism in the MPTP-treated primate. While the chronic development of

L-DOPA-induced AIMs is representative of LID, the time and expense required for the AIMs to appear can impact heavily on a laboratory’s resources. Because of these limitations, it may be more efficient to utilize a more acute and less expensive animal model as a screening process for novel antidyskinetic compounds rather than a model such as the 6-OHDA- lesioned rat which requires more resources and time.

1.3.2.2 6-OHDA-lesioned mouse One reason why rats are generally preferred over mice for models of LID is the practical size of the rat which is convenient for surgical manipulation and behavioural analysis. However, the vast range of genetically engineered mouse strains is an attractive option for studying genetic components of LID after neurotoxin administration. Mouse models of PD and LID have been developed based on two neurotoxins: 6-OHDA or MPTP.

The MPTP-treated mouse model is generally less preferential because of its limitations: the effects of L-DOPA after MPTP administration are highly dependent on strain (Jackson-Lewis

& Przedborski, 2007), motor activity can return to baseline even with continued MPTP treatment and high doses (200 mg/kg) of L-DOPA are often required (Nicholas, 2007).

Behavioural analysis in the 6-OHDA-lesioned mouse model is less sensitive to these variables and is therefore more favourable as an animal model of PD.

The 6-OHDA lesion procedure in mice is almost identical to that used in rats, although one of the first studies of 6-OHDA-lesioned mice noted a lesion success rate of only

14% as many of the mice died prematurely (Lundblad et al., 2004). Despite this poorer

91 outcome, the 6-OHDA-lesioned mice showed significant increases in dyskinetic behaviour after L-DOPA treatment compared to sham-operated or saline-treated mice (Lundblad et al.,

2004). Additionally, sensitization of L-DOPA-induced AIMs were significantly greater throughout each dosage phase and increased with dose escalation as well compared to controls. The 6-OHDA-lesioned mouse model also demonstrates sensitivity in predicting the development of LID, with rotarod, spontaneous rotations, apomorphine-induced rotations and locomotor activities significantly correlating with LID development compared to other activities such as cylinder, corridor, balance beam and psycho-stimulant behaviour tests

(Smith et al., 2012).

Further validation of this model was seen when lesioned C57BL/6 mice treated with

L-DOPA for three weeks developed orolingual, axial, and limb AIMs, but not when groups were treated with D2/D3 agonist ropinirole or adenosine A2A antagonist KW-6002 at doses that improved forelimb akinesia (Lundblad et al., 2005). This suggests that the 6-OHDA- lesioned mouse model mimics the effects of specific antiparkinsonian agents in humans as both ropinirole and KW-6002 are antiparkinsonian agents do not induced dyskinesia in PD patients. Other compounds that exert antidyskinetic effects in rat and primate models of LID such as amantadine, buspirone, or riluzole also alleviate L-DOPA-induced AIMs in lesioned mice (Lundblad et al., 2005).

Not only are there pharmacological consistencies between these models, but molecular markers of LID or PD found in other animal models are also present in the 6-

OHDA-lesioned mouse model. This suggests that not only can inter-species AIMs be treated in a similar fashion, but that the underlying processes that induce these AIMs are also the same. For example, increased striatal FosB expression and increased phosphorylation of ERK

1/2 as well as upregulation of striatal prodynorphin mRNA is found in both 6-OHDA- lesioned mice (Pavon et al., 2005) and rats (Andersson et al., 1999) displaying AIMs induced

92 by chronically L-DOPA treatment. Additionally, striatal pre-proenkephalin levels are higher in both 6-OHDA-lesioned mice (Pavon et al., 2005) and MPTP-treated primates (Quik et al.,

2002) compared to animal controls with no nigrostriatal damage. These results suggest that not only does the 6-OHDA-lesioned mouse model have behavioural and pharmacological similarities to other validated animal models of LID, but there are also significant overlapping neurochemical tendencies between these models after lesioning or L-DOPA treatment.

One major drawback in the 6-OHDA-lesioned mouse model is the high rate of mortality. In particular, these unexpected deaths often occur soon after the surgeries and before behavioural testing can be performed, leading to ethical concerns about animal welfare. Mortality rates of 82% and 30% have been reported in mice with 6-OHDA lesions in the medial forebrain bundle or striatum, respectively (Lundblad et al., 2004). Intense post- operative care such as separating meagre and healthy animals into different cages to prevent competition for food, replacing ordinary food with food pellets soaked in sugar/water solution, administering sterile physiological glucose-saline solution (s.c.) every day, and in extreme cases hand feeding the mice due to ipsilateral twisting of the head which can make eating difficult can reduce mortality rates, but the justification can be contentious (Cenci &

Lundblad, 2007). While post-procedural care is needed with other animal models, particularly in bilateral 6-OHDA-lesioned rats which often have to be tube-fed (Castaneda et al., 1990), the unilateral 6-OHDA-lesioned rat and the MPTP-treated primate generally require less intensive post-procedural care than the unilateral 6-OHDA-lesioned mouse (Joyce & Iversen,

1984; Benazzouz et al., 1992; Kordower et al., 2000). It is not fully understood why 6-

OHDA-lesioned mice have such a high post-operative mortality rate compared to 6-OHDA- lesioned rats, which allows for few alterations to the surgery procedure, excepting a change in injection location, to avoid the necessity for intensive post-operative care. Therefore, unless

93 the outcomes of a study are genetically dependent, rats are generally preferred over mice in 6-

OHDA lesion models.

Previous studies have suggested that the location of the 6-OHDA lesion can impact on the induction and severity of neurochemical and behavioural effects of L-DOPA treatment in rodents. In particular, significant differences in behaviour have been found in mice with lesions in the medial forebrain bundle (MFB) or striatum (Francardo et al., 2011). 6-OHDA lesions in the MFB lead to maximally severe L-DOPA-induced AIMs in all mice and high numbers of L-DOPA-induced contralateral rotations, indicating substantial supersensitivity, compared to mice with striatal lesions in which only a third exhibited L-DOPA-induced

AIMs of reduced severity and fewer contralateral rotations (Francardo et al., 2011). These results suggest that mice with striatal lesions are more prone to experiencing spontaneous dopaminergic neuronal recovery than MFB-lesioned mice (Francardo et al., 2011).

Although research into strain differences within the 6-OHDA-lesioned mouse model has been sparse, strain-dependent behavioural differences after L-DOPA administration have been identified. Specifically, FVB/C57BL/6 mice with 6-OHDA lesions in the MFB showed significantly less net contraversive rotations after 4.5 mg/kg L-DOPA administration and exhibited higher L-DOPA-induced AIMs scores compared to FVB mice (Thiele et al., 2011).

This indicates L-DOPA has lesser antiparkinsonian effects and greater dyskinetic effects in lesioned FVB/C57BL/6 mice than FVB mice and demonstrates that results must be interpreted in the context with the strain selected for the 6-OHDA-lesioned mouse model.

However, this is the only known study comparing mouse strains in the 6-OHDA-lesion model as many studies prefer to use the pure C57BL/6 mice to develop and expand on mouse models (Cenci & Lundblad, 2007; Lundblad et al., 2004; Francardo et al., 2011).

The relative ease of genetic manipulation in the mouse makes the 6-OHDA-lesioned mouse model an attractive tool to study the different genetic components in PD and LID.

However, the mortality rate, selected strain, and the location of the lesion must be considered when developing the model or testing potential therapies as these factors substantially influence antiparkinsonian symptoms and dyskinetic effects of L-DOPA. Thus interpretation of study outcomes without considering these factors could influence overall conclusions of the efficacy of such potential therapies and possibly lead to false alarms or missed opportunities.

1.3.3 Reserpinised rat For decades it has been known that reserpine depletes monoamines in the central nervous system through irreversible inhibition of the vesicular monoamine transporter 2

(VMAT2; Scherman, 1986). In particular, the blockage of dopamine vesicular uptake results in the accumulation of neurotoxic dopamine oxidation byproducts (Caudle et al., 2008) and can reduce dopamine in the substantia nigra pars compacta and striatum by 85% and >95%, respectively, approximately 2 hours after reserpine administration (Heeringa & Abercrombie,

1995). These byproducts can occur 1) when synaptic dopamine forms dopamine-quinones by reacting with molecular oxygen, depleting the antioxidant glutathione and generating reactive oxygen species (ROS) and 2) when enzymatic metabolic breakdown of dopamine (via monoamine oxidase) increases the formation of ROS (Tsang & Chung, 2009). It is when the production of ROS exceeds the ability of the antioxidant system to eliminate them that oxidative damage occurs (Dröge, 2002).

Oxidative stress-induced neuronal damage can lead to alterations in both motor (Faria et al., 2005; Teixeira et al., 2009) and cognitive skills (Chen et al., 2010) in rodents; impairments which are commonly found in PD patients (Zibetti et al., 2013). Evidence of oxidative stress-induced damage is also found in brain tissue of PD patients (Beal, 2002),

MPTP-treated (Obata, 2002) and 6-OHDA-lesioned (Riobó et al., 2002) animals. These neurological similarities suggest that the reserpinised rat is a representative model of PD, at

95 least at a cellular level (Bilska et al., 2007; Spina & Cohen, 1989). However, it is important to note that reserpine does not induce degeneration of striatal neurons as is a defining characteristic of PD, and so has contentious construct validity as a model of PD in this regard.

However, other aspects of the reserpine rat model, particularly reserpine-induced behaviour, has been informative in PD research. More specifically, reserpine treatment elicits a transient syndrome characterized by rigidity, akinesia and postural flexion (Colpaert, 1987) which have substantial similarities to the major motor symptoms of PD. These symptoms are found in acute and chronic studies of reserpine treatment (Segovia et al., 2003; Salamone &

Baskin, 1996; Fernandes et al., 2012) and demonstrates the versatility of the reserpine rat model of PD for studies focused on different timeframes and stages of dopamine depletion.

Furthermore, the reserpine rat model has demonstrated overlapping pharmacological effects with PD in humans. Indeed, compounds that reverse major motor symptoms in PD patients also reverse reserpine-induced behavioural effects in rodents. L-DOPA (in conjunction with a peripheral AADC inhibitor), apomorphine, bromocriptine are all commonly used to treat PD symptoms in patients and are effective in antagonizing hind limb rigidity in reserpinised rats (Coplaert, 1987; Goldstein et al., 1975). Other compounds which are effective in relieving these reserpine-induced symptoms include amantadine (NMDA receptor antagonist), pargylne (MAO inhibitor), and pergolide (DA receptor antagonist)

(Colpaert, 1987). This demonstrates that symptoms elicited from reserpine treatment can be reversed by dopaminergic compounds, but also by a range of non-dopaminergic drugs that utilize pathways previously researched in PD (Parkes et al., 1970; Cooper et al., 1992).

Although reserpine-induced akinesia in animals has been a widespread model of PD, other research has emerged to suggest that reserpine might also be useful for studying LID in animal models. This proposition is based on the selective hyperactivity behaviours exhibited in reserpinised rats after L-DOPA administration which is reduced by administration of

96 compounds thought to be antidyskinetic (Johnston et al., 2005; Segovia et al., 2003; Johnston et al., 2012; Lee et al., 2006). Indeed, amantadine, one of the few compounds used to treat

LID in patients (da Silva-Junior et al., 2005), and 3,4-methylenedioxy-N-methylamphetamine

(MDMA) which reduces LID in MPTP-treated primates (Blanchet et al., 1998; Johnston et al., 2005) both reduce L-DOPA-induced hyperactivity in reserpinised rats (Johnston et al.,

2005).

In particular, antidyskinetic compounds seem to selectively reduce L-DOPA-induced vertical activity, while exerting no significant effect on L-DOPA-induced horizontal activity

(Johnston et al., 2005; Segovia et al., 2003; Johnston et al., 2012). Moreover, haloperidol, which reduces both LID and antiparkinsonian effects of L-DOPA in PD patients (Klawans &

Weiner, 1974), reduces both L-DOPA-induced vertical and horizontal activity in reserpinised rats (Johnston et al., 2005).

This has prompted the suggestion that L-DOPA-induced vertical activity is modulated by pharmacology that overlaps with LID in primates (both human and non-human) and may therefore be an animal correlate of LID in PD patients. Furthermore, L-DOPA-induced horizontal activity may be the animal correlate of L-DOPA-induced antiparkinsonism in humans due to a similar effects induced by the same compounds (Johnston et al., 2005). It is predicted that increases in horizontal movement (movement on all four paws) indicates a decrease in PD-like major motor symptoms after L-DOPA administration and that an increase in vertical movement (rearing, movement on two hind paws) is indicative of an increase in

LID-like behaviour (Segovia et al., 2003; Johnston et al., 2005; Johnston et al., 2012).

The reserpine rat model has also provided evidence for the involvement of other neurotransmitter pathways which may modulate LID, such as the cannabinoid pathway.

Specifically, the CB1 receptor antagonist rimonabant (SR-141716) among other cannabinoid compounds reduce vertical, but not horizontal, L-DOPA-induced activity in reserpinised rats

(Segovia et al., 2003). These results suggest that rimonabant reduces behavioural correlates of LID, without impairing L-DOPA’s antiparkinsonian effects. Further evidence of this was found in MPTP-treated primates which displayed reduced LID without impairing antiparkinsonian effects of L-DOPA after administration of rimonabant (van der Stelt et al.,

2005).

The similar behavioural effects of MDMA-analogues, UWA-101 in particular, between MPTP-treated primates and reserpinised rats have also been demonstrated. As with rimonabant, MDMA and UWA-101 were most effective in reducing LID in primates and vertical activity in rats without impairing L-DOPA’s antiparkinsonian effects in primates or reducing horizontal activity in rats, respectively (Johnston et al., 2012). Thus, the effects of both MDMA analogues and cannabinoid compounds support the correlation between LID- like behaviour and vertical activity, and antiparkinsonian effects and horizontal activity across different species models of LID. This is also evident through the testing of compounds in MPTP-treated primates and reserpinised rats which have displayed antidyskinetic effects in humans.

In addition to the predictive validity, the L-DOPA-treated reserpinised rat model demonstrates substantial face validity, not just through dopamine depletion, but also in cellular responses to exogenous L-DOPA administration. After administration of L-DOPA

(100 mg/kg, i.p.), extracellular dopamine levels in the brains of reserpinised rats were increased to about 200% of normal values 30 and 100 minutes after the injection of L-DOPA before gradually decreasing reserpine levels 6 hours after L-DOPA injection (Ahlenius &

Engel, 1971). This is consistent with effects of L-DOPA in humans as it is well established that exogenous L-DOPA is metabolised into dopamine in the brain, particularly the striatum, of PD patients (Lloyd et al. 1975). It has also been found that acute dopamine depletion by reserpine can increase dopamine receptor sensitivity 6-12 hours after reserpine injection

98 without an increase in receptor density (Trugman & James, 1992). The authors suggested that this receptor supersensitivity reflected the OFF state during L-DOPA treatment seen in PD patients and that this response may increase the risk of dyskinesia during ON states. This was also consistent with a study of dopamine receptor density in primate brains: although D1 receptor density did not change, binding sensitivity was linearly correlated to presence of dyskinesia (Aubert et al., 2005).

A more progressive rat model of PD has also been proposed in which reserpine is administered at a lower dose over a number of weeks (Missale et al., 1989; Fernandes et al.,

2012). Although this model has not yet been tested as a reliable model of LID, chronic administration of both reserpine and L-DOPA could serve as an intermediate level between the acute reserpine rat model and the 6-OHDA-lesioned rat model of LID. In particular, this potential new model may serve as a screen test for possible antidyskinetic compounds to determine long-term effects, as opposed to the acute effects found in the current reserpine rat model, which may then be more thoroughly tested in 6-OHDA-lesioned rats or MPTP-treated primates. As 6-OHDA-lesioned rat and MPTP-treated primate models can be costly in both time and money, the utilization of a model that is both inexpensive and acute as a screening test for antidyskinetic compounds would be advantageous as a starting point for research into potential drug leads.

1.3.4 Genetically modified mouse Unfortunately, there is little research focused on developing a genetic model of LID.

Although several human genes have been associated with increased risk of either PD or LID

(Sharma et al., 2010), only genetic models of PD have been established (Crabtree & Zhang,

2012). Studies have found that the presence of both the 13 and 14 allele of the D2 receptor encoding gene lowers the risk of developing LID. In contrast, alleles that results in low catechol-O-methyltransferase (COMT) enzyme activity are overexpressed in PD patients that

99 exhibit dyskinesia compared to controls (Watanabe et al., 2003), while patients with dyskinesia carry the 40-bp VNTR allele of the DAT more frequently than nonafflicated patients (Kaiser et al., 2003), and are at a higher risk of developing dyskinesia earlier during treatment with dopaminergic compounds if they have the Met allele of the brain derived neurotrophic factor (Foltynie et al., 2008). Although these studies have provided valuable information concerning the genetic risk factors and influences of PD and LID, majority of PD cases, and follow-on LID side effects, are still considered idiopathic.

One promising model of PD is the Pitx3-deficient mouse model, which could also have the potential to serve as a model of LID (Ding et al., 2011). Pitx3 is a critical transcription factor for the survival of midbrain dopaminergic neurons (Nunes et al., 2003) and Pitx3-deficiency in mice leads to significant but selective loss of DA neurons in the substantia nigra and defects in the nigrostriatal pathway (Hwang et al., 2004, Smidt, et al.,

2004). It has also been found that Pitx3-expressing neurons in the substantia nigra are preferentially destroyed after MPTP treatment (Luk et al., 2013), indicating that neurons destroyed in MPTP-treated animals resulting in PD-like symptoms are also involved in the locomotor deficits exhibited in Pitx3-deficient mice. Additionally, these mice display accompanying locomotor deficits, similar to those seen in PD, which are ameliorated after L-

DOPA administration (Hwang et al., 2005).

Studies of Pitx3-deficiency in mice have found significant neuronal changes reminiscent of other dopamine-deficient animal models. Indeed, these overlapping changes are most readily found in neuropeptide and receptor expression. For example, D1 receptor, dynorphin, and substance P mRNA levels are decreased in the caudate putamen of Pitx3- deficient mice (Smits et al 2005). This, paired with the unaltered D2 receptor and enkephalin mRNA levels, are reminiscent of changes found in DA -/- mice (Zhou & Palmiter, 1995) and partially DA-lesioned (>10% of remaining striatal DA neurons) adult rats (Nisenbaum et al.,

100

1996). Additionally, the changes seen in adult animals with nearly complete DA-lesions are similar to those of partially-lesioned animals, with the exception of increased enkephalin and

D2 receptor levels (Young et al., 1986; Voorn et al., 1987; Gerfen et al., 1990, 1991; Li et al.,

1990). These results suggest that dopamine denervation, via a range of methods, reliably induces similar alterations in neuropeptide and dopamine receptor expression in genetic and lesion-induced rodent models of PD, providing support for the use of the Pitx3-deficient mouse model.

Although Pitx3-deficient mice display similar neuronal responses to dopamine depletion compared to other animals, there are discrepancies in behavioural effects across models and species. The most common behavioural features of parkinsonian animals include rigidity, slow or little movement, and reduced co-ordination. In contrast to animal exposed to

6-OHDA, MPTP, or reserpine, Pixt3-deficient mice exhibit few impairments in overall locomotor activity (Hwang et al., 2005). However, specific measures of parkinsonian behaviour such as traverse time or number of steps on a challenging beam as well as downwards orientation time during the pole test show that Pitx3-deficient mice have significantly greater parkinsonism compared to wildtype mice which is, importantly, reversed by L-DOPA (Hwang et al., 2005). Furthermore, exposure to L-DOPA significant increase activity and performance in Pixt3-deficient mice compared to L-DOPA-treated wildtype mice, indicating DA receptor supersensitivity in Pitx3-deficient mice similar to that found in other neurotoxin-induced models of PD (Graham et al., 1990; Heikkila et al., 1981; Hefti et al., 1980).

As mentioned previously, this mouse model has potential as a research tool for LID therapies (Ding et al., 2011). Indeed, after sufficient nigrostriatal degeneration of dopaminergic neurons has been achieved in Pixt3-deficient mice (Pixt3ak/ak), reproducible dyskinesia have been observed after repeated L-DOPA administration (Ding et al., 2011).

101

While this is supportive of the Pitx3-deficient mouse model, there are some differences in dyskinesia measurements that should be noted. Unlike most LID studies that measure severity and frequency of dyskinesia through AIMs, dyskinesia in Pixt3ak/ak mice were measured through abnormal sliding movements with both front paws (and over time, a hind paw as well) against a cylinder wall 15 minutes after L-DOPA injection for 2 minutes (Ding et al., 2011). The sum of abnormal movements in the two front paws with a hind paw then provided an indicative score of total dyskinesia. In contrast, dyskinesia exhibited by Pixt3ak/+ mice given unilateral 6-OHDA lesions in the same study were measured through limbic and axial AIMs and were recorded for three minutes every 20 minutes during a 2-hour period.

This is an important distinction as the authors used the LID findings from the Pixt3ak/+ 6-

OHDA lesioned mice as a confirmation for the LID findings obtained from the Pixt3ak/ak mice. The authors reported that non-lesioned Pixt3ak/+ mice showed no two or three paw dyskinesia when treated with L-DOPA; however, there was no mention of whether three paw dyskinetic behaviour was observed in Pixt3ak/+ 6-OHDA-lesioned mice which would have been a stronger confirmation of the Pixt3ak/ak findings rather than analysing only limbic and axial AIMs. Thus, although a genetic mouse model for LID has been proposed, studies are still required to ensure these outcomes can be repeated and that this model has sufficient face, predictive and construct validity to be used as such a model.

As the Pixt3-deficient mouse is a possible model of LID and a model of PD, it is possible that other genetic mouse models of PD could also serve as models of LID. There are several established genetic mouse models of PD primarily based on a number of gene modifications including alpha-synuclein dysfunction, proteasome dysfunction, dopamine metabolism dysfunction, mitochondrial dysfunction, synphilin-1 dysfunction, LLRK2 dysfunction and nigrostriatal pathway dysfunction (Table 1.1). As most of these models are known to respond behaviourally to L-DOPA, it is possible that chronic or high dose L-DOPA

102 administration might induce AIM-like behaviours in these mice which would enable researchers to study the effects of genetic factors on the emergence and severity of LID.

Additionally, other genes such as those associated with risk of LID development (D2 receptor genes, Met allele of BDNF gene, TaqDI*1 and TaqIA*2) also provide avenues for LID research in genetic mouse models.

Table 1.1. Common genetic mouse models based on familial PD gene mutations and/or displaying phenotypes reminiscent of PD pathogenesis or progression.

Model Mutation/Gene Behavioural and Neuronal Effects Responsive to L-DOPA

Alpha- A53T, A30P, Bradykinesia, rigidity, postural Yes synuclein E46K instability, decreased range of motion, tremor, protein aggregation, loss of TH+ neurons, lysosome pathology, gliosis, mitochondrial dysfunction, inclusion formation

Proteasome park 2, psmc 1 Decreased coordination, decreased startle Yes response, decreased motor activity, protein aggregation, loss of TH+ neurons, gliosis, abnormal mitochondria, altered DA metabolism

DA drd 2, th, Bradykinesia, decreased locomotion, Yes – Metabolism moab, vmat2 abnormal gait/ posture, protein rescues aggregation, loss of TH+ neurons, perinatal decreased mitochondria activity, gliosis, lethality altered DA levels

Mitochondrial park 6, park 7, Weight loss, decreased locomotion, Yes tfam decreased grip strength, gait impairment, tremor, rigidity, abnormal mitochondria, altered DA metabolism, increased ROS, inclusion formation

Synphilin-1 sncaip, R621C Decreased motor performance and skill, Unknown decreased learning, decreased step length, inclusion formation, altered DA levels

103

LLRK2 park 8 Mixed reports: hyperactivity, increased Yes motor function, decreased locomotor activity, protein aggregation, axonal degeneration, altered DA metabolism, systemic synucleinopathy

1.4 Animal models of reward and anxiety Over the years, concern about the growing misuse and abuse of medicines by patients and the general population has been increasing. Misuse of a substance, such as prescription drugs, is defined as use that deviates from the original prescription instructions, or inventing/exaggerating symptoms to obtain a prescription for that drug (Fischer & Boggs,

2010). Substance abuse is defined as repeated substance use resulting in significant impairment or distress through substance-related legal issues, persistent social problems, use in physically dangerous situations, or failure to fulfil major role obligations (American

Psychiatric Association, 2000). Although substance abuse is widely recognised as primarily associated with use of illicit drugs, research is increasingly focused on the misuse and abuse of prescription drugs.

The misuse and abuse of prescription drugs has been climbing over recent years and remain near peak levels, presumably because of the decline in illegal drugs since 1990

(Johnston et. al., 2011). Indeed, 2.6% of the population in the US engage in nonmedical use of prescription drugs (NMUPD), a figure which is now second only to marijuana use in persons aged 12 and older according to the National Survey on Drug Use and Health

(SAMHSA, 2013). Additionally, 20.5% of the Australian population reported nonmedical use of a prescription drug at least one in their lifetime, while 7.8% admitted to nonmedical use of prescription drugs within the last 12 months based on results of the National Drug Strategy

Household Survey (AIHW, 2005). Another concerning trend is the 111% increase in the estimated number of hospital Emergency Department (ED) visits for nonmedical use of

104 opioids during 2004-2008 (from 144,600-305,900 visits), while the number of ED visits involving nonmedical use of benzodiazepines increased 89% during 2004-2008 (from

143,500-271,700 visits; Cai et al., 2010). A wide range of prescription drugs are commonly used for nonmedical reasons and are thus categorized as pain relievers, tranquilizers, stimulants, or sedatives (SAMHSA, 2009). These results highlight the strong prevalence of

NMUPD in different populations, and thus emphasises the need for additional research and therapeutic drugs with fewer reinforcing effects.

Initiation of NMUPD can occur for a number of reasons. Among student groups, motivations for using a range of prescription drugs non-medically commonly include experimenting to experience rewarding/reinforcing effects, sleep prevention, exam preparation, improvement or maintenance of overall academic performance, coping with stress, or to satisfy their curiosity (Betancourt et al., 2013; Rozenbroek & Rothstein, 2011).

Additionally, ‘Adderall’, an amphetamine-dextroamphetamine combination agent used to treat Attention Deficit Hyperactive Disorder (ADHD), is one of the most misused prescription drugs among college students who most often take it to experiment with drugs, lose weight or improve concentration (Teter et. al., 2006). This is of particular concern as drugs that act as dopamine agonists, such as cocaine and amphetamines, have a high abuse liability and encouraged the epidemic growth of psychostimulant addiction over the past years (Pulvirenti & Koob, 1994).

One possible reason for such widespread misuse is the large range of legitimate uses of these drugs which encourages the opinion that the health risk of consuming these drugs is lower (Johnston et. al., 2011; Compton & Volkow, 2006). These drugs are also legal and publicly advertised to the consumer both through media and healthcare professionals, adding to the illusion that prescription drugs are safer to use than illicit drugs, as seen when Boyd et. al. (2006) surveyed adolescents about their motivation for abusing prescribed substances.

105

Therefore, advice by healthcare professionals as well as general social approval of these drugs might also influence their increasing prevalence of abuse. While some nonmedical users of prescription drugs do so without much significant aftermath, others, particularly those using prescription drugs for their reinforcing effects, can develop physical and psychological dependence on these substances. Such dependence often results in addiction and serious health consequences.

As discussed by Tcheremissine (2008), the most recent view of addiction proposes that almost all addictive drugs act on one common dopamine site in the mesolimbic reward system, either directly or indirectly. Illicit drugs such as amphetamine and cocaine increase dopaminergic activity in the brain, particularly in the mesolimbic reward system, and are thus prime examples of stimulating the reward system and inducing rewarding effects. Other drugs which may affect the reward system indirectly are thought to induce different subjective effects such as decreases in anxiety. Indeed, benzodiazepines which are classified as prescription drugs are commonly misused and abused for their sedation and anxiolytic effects (Caplan et. al., 2007). The underlying mechanism for anxiolytic effects is not yet fully determined, although neurotransmitters such as histamine (Williams & Miller, 2003), serotonin (Wood & Toth, 2001), and glutamate (Cosford et al., 2003) are thought to be involved. Both the direct and indirect pathways are known to lead to drug-seeking behaviour if not actual addiction.

Because the prevalence of prescription drug misuse and abuse in a number of populations continues to increase over time, it is important during drug discovery and pre- clinical research to test for misuse or abuse potential and psychoactivity in novel compounds in an attempt to prevent NMUPD in patients and other consumers. Rewarding or reinforcing effects induced by the drug can be tested in animal models including the conditioned place

106 preference (CPP) test, the elevated plus maze (EPM) test, and the analysis of drug-induced ultrasonic vocalisations (USV).

1.4.1 Conditioned place preference (CPP) CPP is a procedure that uses classical conditioning to associate primary motivational effects of a drug, the unconditioned stimulus (UCS), with an initially neutral environment by consistently pairing them together during conditioning (Mucha et al., 1982). Because of this, throughout the conditioning period, the neutral environment develops secondary motivational properties such that the contextual stimuli acquire the properties of a conditioned stimulus

(CS) (Mucha et al., 1982). Hence, the reinforcing properties of the drug are associated with the paired environment or CS which is approached by the animal in preference to a second neutral environment (Mucha et al., 1982). The counterpart of CPP is Conditioned Place

Avoidance (CPA), where aversive properties of a drug act as a UCS, and the contextual stimuli become a compound CS for an avoidance response (Mucha et al., 1982). This model enables the study of drugs which directly and indirectly affect the reward circuits of the brain, and thus can incorporate analysis of a variety of drugs (Mucha et al., 1982).

However, such a paradigm has its limitations - for instance it has been found some animals that have learnt a response while in a specific (drugged) state can only reproduce this response when in that same drugged state as opposed to a drug-free state (Overton, 1978).

This is known as state-dependency and is an example of compound stimulus conditioning in which one component of a compound stimulus (i.e. the subjective cue produced by the drug) is conditioned in preference to another due to overshadowing (salience) or blocking (better or earlier prediction). Although less common, a few studies have explicitly examined state- dependency effects on CPP testing; however, most studies that mention state-dependency do so in passing. Spyraki et al. (1985) reported that picrotoxin-induced CPA was state-dependent in that significant aversion was only seen in the drug-free state rather than the drugged state,

107 while Oberling et al. (1993) demonstrated that lithium chloride only induces CPA when animals are in a drugged state. This effect can be seen in other species too: mice pre-treated with amphetamine (10 mg/kg) rather than saline, or amphetamine (2 mg/kg) show increased preference for novel environment compared to a drug-paired environment when administered amphetamine (2 mg/kg) rather than saline (Laviola & Adriani, 1998). However, not all CPP or CPA effects are state-dependent, which can suggest pathways for the development of such learned behaviour (Elliot, 1988).

The influence of state-dependency on the outcome of an experiment is important to consider, as in most cases the animals are conditioned in a drugged state and tested for CPP in a drug-free state (Fudala et al., 1985; Mueller & Stewart, 2000; Horan et al., 1997). This issue can be easily accounted for by incorporating an additional test of CPP after administration of the conditioning drug(s) (Tzschentke, 2007). This allows experimenters to detect any state-dependency if the CPP only occurs in the drugged test or both drugged and non-drugged tests and make further inferences about the mechanism or subjective effects of the tested drug.

Additionally, a phenomenon known as ‘time-state’ dependency can develop in certain strains of rats. This occurs when CPP is inhibited because the time of conditioning and the time of testing do not match. This time-state dependency was tested by Cain et. al. (2004) using Long Evans and Wistar rats (Rattus norvegicus) to determine any differences in food- induced CPP development if time of testing was mismatched to the conditioning schedule

(i.e. a different point in the circadian rhythm). It was shown that a difference in time of testing compared to conditioning had no effect on the development of CPP in Long Evans rats, but prevented Wistar rats from exhibiting CPP produced when conditioning and test times matched. This is resolved by conducting the test at the same time of day as the conditioning sessions.

108

The validity of the CPP test in accurately measuring reward has been established through a number of studies. Drugs such as cocaine and amphetamine are widely misused and abused in humans (SAMHSA, 2013) induce consistent place preference in rodents conditioned in CPP studies (Nazarian et al., 2004; Cervo & Samanin, 1995; Hiroi & White,

1991; Carr & White, 1983). Additionally, drugs that do not act on the dopamine system directly, but are still widely consumed in humans, such as ethanol, also induce strong place preference in the CPP test in rodents (Bozarth, 1990). Thus, the induction of a place preference in animals by drugs known to be abused by humans provides evidence for the predictive validity of this paradigm in identifying drugs that induce rewarding effects.

However, CPP must also provide evidence of reward in commonly abused prescription drugs for the test to be useful in research focused on the abuse liability of prescription drugs. Animal studies testing the effects of prescription drugs commonly misused and abused in humans have provided considerable evidence regarding the predictive validity of the CPP paradigm. For example, diazepam, a prescription benzodiazepine that induces reinforcing effects in nondrug-abusing individuals and is chosen 100% of the time in moderate drinkers over placebo (De Wit et al., 1989), also induces significant place preference in rodents at 1 mg/kg (Nomikos & Spyraki, 1988). Methylphenidate, a psychostimulant commonly prescribed for attention deficit/hyperactivity disorder (ADHD) and known for its abusive potential in humans (Kollins et al., 2001), also induces strong conditioned place preference in rats (Martin-Iverson et al., 1985; Nomikos & Spyraki, 1988;

Spyraki et al., 1985). Lastly, morphine’s rewarding effects (Preston et al., 1991; Lamb et al.,

1991) and naloxone’s lack of reinforcing effects in humans (Jasinski et a., 1967; Orman &

Keating, 2009) is also reflected in rats as morphine and naloxone induce a conditioned place preference and conditioned place aversion, respectively (Mucha & Iversen, 1984).

109

Not only can the CPP model replicate in rodents the rewarding effects seen in humans, but CPP can help to identify the mechanisms of these drug-induced rewarding effects. This is an important endeavour in order to develop effect treatments for drug abuse.

For example, ethanol-induced CPP in rats is selectively blocked by orexin-2 receptor antagonists (Shoblock et al., 2011) and morphine-induced CPP in rats is blocked by NMDA- receptor antagonists (Tzschentke & Schmidt, 2003). Studies of this nature identify promising directions to explore when researching ways to reduce abuse potential of drugs, particularly medications. Indeed, the NMDA receptor antagonist memantine successfully attenuates the expression of morphine physical dependence in humans and thus may be a useful treatment for general opioid abuse (Bisaga et al, 2001; Herman et al., 1995). Whether memantine also modulates the analgesic properties of morphine is yet unknown; however, this research has highlighted the usefulness of the CPP paradigm as a tool for identifying rewarding drugs and possible mechanisms to treat drug abuse.

1.4.2 Ultrasonic vocalisations (USVs) It is widely considered that many mammals including primates and rodents express affective states in a range of environments. Studying the range of factors that influence the affective states of animals has contributed significantly to the understanding of emotion which has implication in fields of research such as psychology, pharmacology, and neurology. As rodents, and other such animals, are unable to provide semantic reports on affective states, behaviour must be used as an indicator of affective state. For an animal behaviour to be considered to reflect an affective state, it must meet specific criteria. Firstly, as humans commonly communicate emotion through facial/vocal displays, communication of affective states in laboratory animals should also be expressed as facial or vocal displays and be associated with predictive changes in approach/avoidance behaviours. Secondly, the same or similar, categories of appetitive stimuli that elicit positive affective states in humans

110 should also increase facial/vocal displays associated with positive affect in laboratory animals, while aversive stimuli should reduce them, as seen in humans. Finally, the identified neurobiological mechanisms of animal affective states should be consistent with those of human affective states.

One aspect of rodent behaviour that meets these three criteria is the emission of ultrasonic vocalisations (USVs). These vocalisations are described as ultrasound as they occur at a frequency greater than the upper limit of the human hearing range (about 20 kHz).

USVs are emitted by a range of animals including bats (Flaquer et al., 2007), rodents (Ahrens et al., 2009; Berger et al., 2013), whales (Samarra et al., 2010), porpoises (Wang et al., 2005), and fish (Mitamura et al., 2005). Anderson (1954) was one of the first researchers to document ultrasonic vocalisations in rodents in the context of their known ability to emit

“pure tones between 20-30 kcy/sec-sounds” (20-30 kHz) that were not audible. He demonstrated evidence of their ultrasonic emissions through a “sonic amplifier” covering a range of 0 to 80 kHz. However, ultrasonic sounds had been documented in bats over a decade prior (Pierce & Griffin, 1938). Although USVs are now thought to represent affect in rats,

Anderson (1954) proposed that these USVs served for communication between individual rats and may have used them for orientation similar to echolocation in bats.

Previous research has reported the emission of different types of USVs in several strains of rats including Lister (Commissaris et al., 2000), Long-Evans (Wright et al., 2012b;

Vivian & Miczek et al., 1993), Sprague-Dawley (Hamed et al., 2012), and Wistar (Borta et al., 2006; Wohr et al., 2009; Commissaris et al., 2000). These USVs have been divided into three broad categories: 40 kHz isolation calls emitted by rat pups, 22 kHz calls by adult rats during exposure to aversive stimuli, and 50 kHz calls often emitted by adult rats when exposed to appetitive stimuli. Previous research has shown that 50 kHz USVs are most often elicited by appetitive stimuli such as food, sex, and play; the same categories of appetitive

111 stimuli that elicit positive affective states in humans. Furthermore, adult 22 kHz USVs are most often produced in response to aversive stimuli such as social defeat and the presence of a predator, stimuli that are considered aversive to humans and therefore elicit a negative affective state. Infant isolation calls (40 kHz) may also represent a negative affective state as they share similar characteristics with adult 22 kHz USVs in aversive and dangerous situations and are best elected separation from the infants’ mother. The neurobiological mechanisms underlying these USVs are also similar to those underlying affective states in humans. Although mice also emit USVs, previous studies suggest that these USVs do not indicate negative or positive affect, unlike rats (for review, see Portfors, 2007) and therefore only USVs emitted by rats will be discussed here.

1.4.2.1 50 kHz USVs

Adult 50 kHz USVs are thought to reflect a positive affective state in rats, based on their elicitation by appetitive stimuli. Although categorized as ’50 kHz’ USVs, these calls encompass a broad frequency range (30-90 kHz) lasting between 30-50 ms and contain a number of subtypes including flat (constant frequency) and up to 14 types of frequency- modulated (FM) calls. The mean peak frequency of flat calls are around 55 kHz, while FM calls such as step calls and trills have a mean peak frequency of 35 and 70 kHz, respectively

(Burgdorf et al., 2008). Appetitive stimuli shown to induce 50 kHz USVs include anticipation of food, rewarding social interaction such as play with a littermate or mating, exposure to an oestrous female, ‘tickling’ by a researcher, treatment with drugs of abuse such as psychostimulants, and even fresh rodent bedding (Natusch & Schwarting, 2010).

However, as 50 kHz USVs are also emitted in aversive contexts such as aggression, drug-withdrawal, CO2 exposure, and pain. Deeper analysis into these calls show that a greater proportion of these USVs emitted in less appetitive contexts are flat calls. FM calls such as

112

‘trills’, on the other hand, are positively correlated to positive affect and appetitive contexts.

Indeed, flat 50 kHz USVs occur significantly more often during aggression than FM 50 kHz

USVs which are emitted at a significantly higher rate during play and mating (Burgdorf et al.,

2008). Furthermore, rats emit 50 kHz USVs, most of which were flat, in a novel clean cage or home cage when separated from their cage mates (Wohr et al., 2008). These findings suggest that flat calls may be involved in social-exploration/contact signalling, while FM calls are primarily associated positive affect. This may explain the emission of flat 50 kHz USVs in aversive contexts such as aggression or drug-withdrawal as these USVs might serve to function as social co-ordination or initiating/maintaining contact, respectively.

As subtypes of 50 kHz USVs such as FM calls are strongly associated with appetitive stimuli and positive affect, this behaviour may be used as another model of reward in addition to others such as place preference and electrical brain stimulation (ESB). Indeed, conditioned cues associated with drugs which induce reinforcing effects in humans also increase emission of 50 kHz USVs, often without influencing 22 kHz call rate (Knutson et al., 1999).

Furthermore, repeated amphetamine (1 mg/kg) administration also increases FM 50 kHz

USVs compared to saline without increasing the rate of flat 50 kHz USVs (Ahrens et al.,

2009; Wright et al., 2010), while repeated cocaine administration increases 50-kHz USVs in rats dose-dependently, although many studies do not specify differences in call subtypes

(Williams & Undieh, 2010; Browning et al., 2011; Ma et al., 2010). ESB also induces 50 kHz

USVs, particularly in areas involved in intense self-stimulation behaviour such as the prefrontal cortex, nucleus accumbens, ventral palladium, lateral pre-optic areas, lateral hypothalamus, ventral tegmental area (VTA) and raphe (Burgdorf et al., 2007). Furthermore, lesioning of either the VTA or medial forebrain bundle, or pharmacological antagonism of

D1/D2 receptors reduces FM 50 kHz USVs emitted during heterospecific play (‘tickling’) compared to rats with sham lesions or vehicle treatment, without moderating 20 kHz USVs or

113 flat 50 kHz (Burgdorf et al., 2007). Thus, it appears that although analysis of combined 50 kHz USVs certainly provides insightful findings in relation to reward, analysing specific subtypes of 50 kHz USVs provides more information into the possible subjective effects of particular environments or drugs and provides a possible explanation for the emission of 50 kHz USVs in more aversive contexts.

Not only have brain areas been identified in influencing USV emission, but neuroplasticity mechanisms as well. As compounds that induce addictive behaviours also generally increase the rate of 50 kHz USVs, it is likely that these behaviours share a commonality. Indeed, the expression of brain-derived neurotrophic factor (BDNF) protein, a crucial mediator of neuroplasticity, is significantly increased after cocaine administration in rats, in addition to dose-dependent cocaine-induced 50 kHz USVs (Williams & Undieh,

2010). These effects were also reversed with D1 receptor antagonism by SCH23390 (0.1 mg/kg), suggesting that 50 kHz USVs are augmented by a D1 receptor-dependent neurotrophin signalling pathway in cocaine-treated rats (Williams & Undieh, 2010).

Despite these findings, there are still inconsistencies in scientific literature of rat

USVs. Although the emission of 50 kHz USVs in aversive contexts may be explained by the proportion of flat calls compared to FM calls, reinforcing drugs have inconsistent effects on

50 kHz vocalization. For example, morphine administered at doses shown to be rewarding in rats during CPP tests fail to promote 50 kHz USVs of any kind in rats during conditioning in

CPP, after repeated administration, in paired or single testing sessions (Wright et al., 2012a), or even during appetitive social interaction with other rats (Vivian & Miczek et al., 1993).

Moreover, morphine does not influence the emission of 50 kHz USVs during tickling by researchers (Panksepp & Burgdorf, 2000). However, increases in 50 kHz USVs do occur in anticipation of morphine administration such as the morphine-paired chamber during CPP

114 testing even without a clear place preference compared to controls, but not as a result of its effects (Hamed et al., 2012; Knutson et al., 1999; Burgdorf et al., 2001).

Inhibition of 50 kHz USVs may be a result of morphine’s affinity for both κ- and µ- opioid receptors (KOR and MOR, respectively; Mignat et al., 1995). It is possible that the emission of 50 kHz USVs is more sensitive to the combined stimulation of KOR and MOR than other reward tests such as CPP. Indeed, the MOR agonist [D-Ala(2),N-Me-

Phe(4),Gly(5)-ol]-enkephalin (DAMGO) produces CPP (Bals-Kubik et al., 1993) and increases the rat of 50 kHz USVs when injected into the VTA (Burgdorf et al., 2007).

Contrastly, KOR selective agonists (U50,488H) produce conditioned place aversion when injected into the VTA (Bals-Kubik et al., 1993), indicating that KOR stimulation may be responsible for morphine’s lack of effect on 50 kHz USVs. However, to our knowledge, no studies of the effect of KOR stimulation on 50 kHz USVs have been conducted. Nonetheless, these studies highlight the necessity for caution when interpreting such findings as they appear to be drug-dependent, regardless of rewarding effects demonstrated in other tests.

1.4.2.2 22 and 40 kHz USVs The negative affective state represented by adult 22 kHz and infant 40 kHz USVs may reflect anxiety and/or depressive states displayed in humans. The vocalizations classified as

22 kHz have frequencies between 18 and 32 kHz often lasting between 100 and 2000 ms and have a sound pressure level of around 40-85 dB (Brudzynski et al., 1993, Brudzynski &

Ociepa, 1992), whereas the USVs classified as 40 kHz in infant rats tend to be frequency modulated calls between 30 and 55 kHz, although 40 kHz is most commonly the frequency with peak energy and emitted at approximately 55-80 dB (Wohr & Schwarting, 2008;

Kraebel et al., 2001). Stimuli known to induce 20 kHz USVs, often paired with freezing behaviour (tense, motionless crouching), are generally aversive and include exposure to predators (Brudzynski & Ociepa, 1992; Blanchard et al., 1991), inescapable pain such as foot

115 shocks (Borta et al., 2006), aversive air-puffs (Pohorecky, 2008), post-ejaculation refractory periods (Barfield & Geyer, 1972), and social defeat (Vivian & Micsek, 1993). On the other hand, 40 kHz USVs in infant are also known as isolation calls as rat pups between 4 to 16 days of age emit these USVs when separated from their mother. Previous studies have demonstrated reliability and predictive validity of infant 40 kHz in rats, as they predict fearful behaviours in adulthood.

Brudzynski et al., (1993) reported that singly-housed rats would vocalize more in a novel environment compared to rats housed in pairs. This is interesting because another study showed that rats exposed to a predator would emit USVs when conspecifics were present but not if the exposed rats were alone. This study concluded that 22 kHz calls functioned as alarm cries to warn conspecifics of predators. However, the finding that singly-housed rats vocalise more than rats housed in pairs suggests that 22 kHz have other functions in addition to communicating danger to other rats. Indeed, studies using air puffs or CPA conditioning or other aversive stimuli induce 22 kHz calls when rats are tested alone, and researchers have reported that animals tested in pairs emit fewer overall and 22 kHz calls compared to rats tested individually. Thus, 22 kHz calls could be a by-product of fear and/or anxiety in aversive situations in addition to serving as an alarm when predators are nearby. Not only this, but experimenters who have induced 22 kHz call emission through touching

(Brudzynski et al., 1993) have done so by touching parts of the animal that are generally out of the animal’s range of vision. For example, touching the dorsal neck area and the rostral portion of the spine are the reactive regions and produce the most 22 kHz calls, while touching the sides or front of the body, regions which can be better observed by the animals, are less effective in inducing vocalisation (Brudzynski & Ociepa, 1992).

Although aversive stimuli can induce 22 kHz calls, not all fear/anxiety tests do so. It appears that rats emit 22 kHz calls when exposed to an immediate threat (such as a cat) or

116 when they cannot escape the aversive stimuli they are exposed to. Hot-plates and inescapable foot-shocks induce 22 kHz calls while the EPM, an anxiety test, induces no USVs in drug naïve rats regardless of the percentage of open arm time or entries recorded (Borta et al.,

2006). Thus, it is possible that the ability of the rats to freely move about the maze and escape the open space of the unprotected arms is responsible for the lack of USVs emitted during this test.

Not only should the type of anxiety test be considered carefully, but the strain and sex of rats as well. Call differences have been found between strains or sexes exposed to the same testing procedure. Sprague Dawley rats tend to emit 22 kHz calls for longer durations than

Long Evans rats, and although there was no main effect of sex, Long Evans females vocalized more often than Long Evans males (Graham et al., 2009). Additionally, 22 kHz calls between Lewis and Wistar rats differ significantly when treated with CP 55940 (10, 25, or 50 µg/kg) during the test phase of foot-shock conditioning (Arnold et al., 2010). CP 55940 dose-dependently increased the number of 22 kHz calls during testing compared to vehicle in

Wistar rats, while CPP 55940 had no effect on conditions USVs in Lewis rats (Arnold et al.,

2010). Thus, caution is needed when interpreting USVs as strain and sex can significantly impact on results.

In addition to their correlation with exposure to aversive stimuli, 22 kHz USVs also overlap neurologically with anxiety-related behaviours. Electrical or chemical stimulation of the dorsolateral periaqueductal grey, which is involved in pain, fear, and anxiety, produces 22 kHz calls, flight responses, and increased defensive reactions in rats (Depaulis et al., 1992;

Yajima et al., 1980). This suggests that the modulation of aversive vocalization overlaps that of anxiety-like reactions as both are influenced by the dorsal periaqueductal grey matter.

Such findings also lend support to the hypothesis that 22 kHz calls are a vocal representative of anxiety and/or fear.

117

In additional to brain regions modulating 22 kHz call emissions, multiple neurotransmitter systems are known to influence these USVs through pharmacological manipulation. It must be noted that most of the following studies have examined the acute, rather than chronic, effects of drug administration on 22 kHz USVs. The main hypothesis for these studies is that if 22 kHz calls are an indication of fear and/or anxiety, then they should be modulated by anxiolytic and anxiogenic compounds.

Indeed, the serotonergic system has been implicated in modulating the emission of 22 kHz USVs. Selective serotonin reuptake inhibitors (SSRI’s), which are commonly used to treat depression and anxiety in humans, significantly reduce foot shock-induced 22 kHz calls compared to saline. However, this effect is not consistent across all SSRI’s as fluoxetine exerts only a partial inhibitory effect and citalopram “produces a biphasic dose-response curve”. These differences may occur due to the different pharmacological profile of each drug as fluoxetine also acts as a 5-HT2 receptor antagonist, which may be responsible for fluoxetine’s limited influence over USVs. 5-HT1A receptor agonists inhibit PAG neuronal activity which may be an underlying mechanism of their reduction of 22 kHz calls in rats.

Other studies have reported that both increased and decreased activity of serotonin receptors can reduce 22 kHz calls in rats. This suggests that the location of these receptors is an important factor as these effects may also be mediated by more than one serotonergic pathway.

Other antidepressants such as tricyclic antidepressants have limited influence over reducing 22 kHz calls, although the tricyclic antidepressant clomipramine (20 mg/kg) does reduce the period required for extinction of fear-induced USVs compared to vehicle- and diazepam-treated rats (Kikusui et al., 2001). Additionally, typical antipsychotics such as haloperidol have little effect on USVs, although the atypical antipsychotic clozapine in known to reduce 22 kHz calls. In contrast to haloperidol, D1 receptor antagonists increase 22

118 kHz calls, suggesting that they are anxiogenic compounds. Although known for its dopamine receptor antagonistic activity, it is possible that clozapine modulates 22 kHz calls through its antagonism for 5-HT2A receptors. However, this needs expanding upon as fluoxetine is thought to have limited effect on USVs due to its 5-HT2 receptor antagonism, maybe that’s because of 5-HTB/C. Other dopaminergic compounds that reduce USVs are D2 and D2/3 receptor agonists which inhibit foot shock-induced USVs. Additionally, the high potency of these compounds suggests that they are agonising pre-synaptic receptors, although if working post-synaptically, they may have abuse potential.

Noradrenergic systems have been implicated in the mediation of 22 kHz calls in rats.

Specifically, α2 adrenoceptor agonists increase the number of 22 kHz calls in intruder paradigms while antagonists such as yohimbine decrease the number of 22 kHz calls induced by inescapable foot shocks. However, α2 adrenoceptor antagonists also display affinities for serotonin receptors and so may be reducing 22 kHz in this way.

Other neurotransmitter systems known to modulate 22 kHz calls include opioids, histamine, glutamate, and GABA. Opioid receptor agonists such as morphine, glutamate

NMDA receptor antagonists such as MK801, and H1 receptor antagonists tend to reduce aversive USVs. In contrast, GABA receptor antagonists and opioid receptor antagonists such as naloxone are firmly established to increase the number of 22 kHz calls in rats.

One important feature to note is that chronic studies have found significant changes in the number of 22 kHz calls in adult rats when tested repeatedly. Specifically, diazepam reduced foot shock-induced 22 kHz USVs in repeated tests over three weeks. However, the vehicle control group also showed a reduction in 22 kHz USVs during foot shock testing over the same period. It is difficult to determine the underlying cause of this reduction as so few chronic studies focused on USVs have been conducted. Nonetheless, it is possible that the animals became desensitized to the tests, as rats show reduced foot-shock-induced

119 hypolocomotion and freezing behaviour by the third testing session, suggesting the development of stress adaptation (Ohi et al., 1989). Thus, the reduction of stress induced-22 kHz USVs along with other behaviours associated with stress and anxiety further indicates that 22 kHz USVs are a model of panic-like anxiety.

Generally, compounds that are anxiolytic in humans tend to reduce 22 kHz calls in rats, although mention of increased 50 kHz calls is needed as well. In contrast, compounds that are anxiogenic in humans tend to increase 22 kHz calls in rats. These results provide strong evidence to suggest that the quantity of specific drug-induced USVs emitted by rodents is a predictive and valid indicator of rodent affect and is therefore a potentially useful tool determining the potential for misuse and abuse of drugs in pre-clinical research.

1.4.3 Elevated plus maze (EPM) The Elevated Plus Maze (EPM) test is a valid method of measuring anxiety in rodents. The EPM test involves conflict between the incentive to explore the environment and the natural aversion to heights and open spaces in rodents as they prefer to spend time in the closed, rather than open, arms of the maze (Montgomery, 1958). The test reliably measures spontaneous locomotive activity and behaviour which can be modified by stimuli that presumably increase or decrease anxiety of the animal when placed in that environment

(Pellow et. al., 1985). Thus, reduced anxiety of a rat by an anxiolytic stimulus will be reflected by increased percentage of open arm entries and percentage of time spent in the open arms of the maze. Conversely, an anxiogenic stimulus will increase the time spent in the closed arms as a result of increased fear and anxiety. However, the rodent’s activity can be influenced by its incentive to forage for food: if the animal is not hungry, its desire to explore might be reduced and so might remain in the closed arms. This could be interpreted as a measure of increased anxiety when it is in fact reflecting the rodent’s dietary needs.

120

Therefore, caution is needed when interpreting results from the EPM test in terms of differing levels of anxiety and fear if the animals are put on restricted diets.

Despite this cautionary measure, the EPM model of anxiety in rodents has been thoroughly tested and found to have substantial face and construct validity. Studies have demonstrated that the EPM test involves behavioural, physiological, and pharmacological measures of anxiety. Many of the behaviours associated with anxiety in rodents such as freezing, immobility, and risk assessment occur significantly more when animals are exposed to the open arms rather than closed arms of the maze (Pellow et al., 1985; Rodgers & Dalvi,

1997). Furthermore, plasma corticosterone concentrations are significantly elevated in animals confined to open arms compared to animals confined to closed arms (Pellow et al.,

1985), while drugs shown to reduce anxiety in humans, such as diazepam (Charney et al.,

1983), also increase the proportion of open arm entries and time during the EPM test in rodents (Nomikos et al., 1988). Thus, the proportion of open arm entries and time in the EPM is strongly associated with anxiety-like behaviour in rodents.

It was previously thought that rodents avoided the open arms of the EPM due to the aversion of heights (as the EPM is commonly situated at a height of approximately 50 cm) and novelty. However, it has been identified that open space, or more specifically, an absence of thigmotactic cues (physical contact or proximity to the edge or wall of an environment), primarily influences the decrease in open arm exploration rodents rather than height or novelty (Treit et al., 1993; Fernandes & File, 1996). This conclusion was drawn from systematic studies demonstrating that the anxiolytic actions of diazepam were significantly attenuated when rats re-tested in the EPM, suggesting that re-exposure increases or maintains anxiety-like behaviour, and that no significant difference in open arm exploration was found when the maze height was altered both symmetrically and asymmetrically (Treit et al., 1993).

Contrastly, the presence of ledges in the maze decreased anxiety-like behaviours in untreated

121 rats and those administered anxiolytic drugs (Treit et al., 1993; Fernandes & File, 1996), supporting the hypothesis that thigmotactic cues primarily influence open arm exploration.

Although open arm exploration is regarded as a valid measure of anxiety-like behaviour in rodents, inconsistent reports on pharmacological influence in the EPM have arisen internationally. Compounds such as adrenergic, serotonergic, CKK and NMDA receptor ligands have produced much more variable results then those seen with γ-amino- butyric-acid (GABA)-related drugs based on open arm entries and time (Handly & McBlane,

1993; Dawson & Tricklebank, 1995) which are often attributed to inter-laboratory differences. However, it is also likely that additional measures of ethological behaviour and locomotive activity, which are not often reported, assist in creating a more detailed profile of anxiogenic/anxiolytic substances. Indeed, closed arm entries have a significant factor loading for locomotive activity. Thereby, if closed arm entries were increased due to the stimulating effect of a drug, the proportion of open arm entries and time is likely to decrease as closed arm time and entries are in the denominator of the equations for calculating percentage of open arm entries and time, suggesting an anxiogenic effect when this conclusion might be inaccurate. Thus, factors such as risk assessment, for which total sniffing has a high factor loading, exploration, for which total head-dips has a high factor loading, and anxiety, for which percentage open arm entries/time and percentage protected stretch-attend postures have high factor loading, can all be used to accurately quantify animal anxiety in the EPM

(Rodgers & Dalvi, 1997).

Although drugs can modulate behaviour of the EPM, this does not necessarily indicate rewarding effects. Certainly, drugs that decrease anxiety and therefore induce anxiolytic-like effects in rodents can be rewarding: benzodiazepines such as diazepam produce CPP (Spyraki et al., 1985) and induce anxiolytic-like effects in the EPM test (Violle et al., 2009) in rodents. However, antipsychotics such as quetiapine induce anxiolytic-like

122 effects at high doses in the EPM without producing CPP in rats (McLelland et al., 2014) while amphetamine induces CPP in rodents (Spyraki et al., 1982; McLelland et al., 2014) and rewarding effects in humans (Drevets et al., 2001) and can increase anxiety-like effects in rodents during EPM testing (Biala et al., 2009) and humans (Uhlenhuth et al., 1981; Johanson

& Uhlenhuth, 1980; Kirkpatrick et al., 2013).

It is possible that some anxiolytic drugs, such as quetiapine, are not rewarding due to their antagonistic properties at dopamine receptors, while other anxiolytic drugs induce rewarding effects because they act on major brain systems implicated in drug reward such as dopamine, opioid, or GABA. Alternatively, the anxiolytic actions of the drugs could indirectly affect the reward system; it has been shown that the extended amygdala, including the bed nucleus of the stria terminalis (BNST), which modulates fear and anxiety projects to the ventral tegmental area (VTA; Jennings et al., 2013), an area which has been implicated in reward and aversion (Cohen et al., 2012; Fields et al., 2007; Lammel et al., 2012; Stamatakis

& Stuber, 2012; Tan et al., 2012; Tye et al., 2012; van Zessen et al., 2012). Thus, anxiolytic drugs such as benzodiazepines which induce anxiolytic-like behaviour in rats through the amygdala (Zangrossi & Graeff, 1994) may also indirectly modulate the reward system in this way.

Although the EPM test is designed to measure anxiety-like behaviour rather than reward, the test can indicate which prescription drugs may have higher abuse potential.

Because anxiolytic rather than anxiogenic drugs tend to induce rewarding effects (Ator &

Griffiths, 2003) and are commonly self-administered in both humans and primates (File et al.,

1986), identifying a drug as anxiolytic can help determine its potential for abuse and misuse in humans. Indeed, the 2012 SAMHSA study of 2.9 million initiates of illicit drugs over the age of 12 found that 17.0% used pain relievers, 4.1% used tranquilizers and 3.6% used stimulants (SAMHSA, 2013). In fact, the only drug with higher usage among initiates was

123 marijuana. Furthermore, drug overdoses and hospital care events are more commonly due to opioid analgesic and benzodiazepine abuse compared to abuse with stimulant prescription drugs (Cai et al., 2010). Thus, the EPM test is a useful model of anxiety in rodents that, while not directly indicating dopamine-dependent abuse potential of different drugs, can be used to build a profile of reinforcing effects of medicinal drug candidates.

1.5 Experimental aims The overall aim of this thesis was to investigate the anti-hyperkinetic and reinforcing effects of UWA-101 at multiple doses in Sprague Dawley rats. Therefore, as an overall hypothesis for this thesis, we anticipate that UWA-101 will act as an effective anti- hyperkinetic drug at lower (1 and 3 mg/kg) rather than higher (10 mg/kg) doses through a mechanism involving the serotonin pathway without inducing reinforcing effects.

Specifically, for each test, we hypothesis that:

● UWA-101 will most effectively reduce L-DOPA-induced hyperkinesia at lower doses (1 and 3 mg/kg) in the reserpinised rat ● The reduction in L-DOPA-induced hyperkinesia by UWA-101 (3 mg/kg) will be blocked by the administration of WAY100635 or fluoxetine ● UWA-101 (3 mg/kg) will not induce a CPP in drug-free or state-dependent tests ● Administration of UWA-101 (3 mg/kg) to healthy rats will not induced a greater number of 50 kHz USVs as a measure of reward compared to saline-treated rats ● UWA-101 (3 mg/kg) will not increase the percentage of open arm time or percentage of open arm entries in the EPM compared to saline-treated controls.

To investigate these hypotheses:

• We first aimed to design and construct in-house testing boxes that recorded locomotive and behavioural data of the rats. • Furthermore, we also aimed to construct in-house microphones to record ultrasonic vocalisations as an indicator of affect in rats.

124

• Using the in-house testing boxes, we aimed to investigate the reduction of L- DOPA-induced hyperkinesia with UWA-101 at increasing doses (1, 3, and 10 mg/kg) in comparison to other drugs known to demonstrate antidyskinetic or reduced hyperkinesia effects in animal models or humans. • Again using in-house testing boxes, we aimed to investigate the mechanism underlying the reduction of L-DOPA-induced hyperkinesia by UWA-101 through administering serotonergic compounds in conjuction with UWA-101 to assess any synergistic, additive or blocking effects. • Lastly, we aimed to investigate the induction of reinforcing or anxiolytic effects by UWA-101 on Sprague Dawley rats through three separate models: Conditioned Place Preference testing to demonstrate reinforcing/aversive effects, Elevated Plus Maze testing to demonstrate anxiolytic/anxiogenic effects and Ultrasonsic Vocalisations to demonstrate positive/negative affects.

125

126

Chapter 2 - A simple and effective method for building inexpensive infrared equipment used to monitor animal locomotion

2 Preface

The aim of this chapter was to design and validate a set of locomotion enclosures inbuilt with infrared (IR) sensors. The custom-designed equipment was required to collect and standardise locomotion data in rats on which future chapters of this thesis would be based. This also provided the opportunity to demonstrate a reserpine-induced hypokinesia effect and a L-DOPA-induced hyperkinesia effect in rats to further validate the reserpinised rat model of Parkinson’s disease and collect baseline data for future comparison. Should these drug-induced effect occur, further analysis could then be conducted to assess the impact of time on these effects.

Publication: McLelland, AE, Winkler, CE, Martin-Iverson MT (2015) A simple and effective method for building inexpensive infrared equipment used to monitor animal locomotion.

Journal of Neuroscience Methods: doi: 10.1016/j.jneumeth.2015.01.006

2.1 Abstract

Infrared (IR) technology is a flexible and effective way of measuring animal locomotion. However, the cost of most commercial IR equipment can limit their availability.

We have designed an inexpensive and effective replacement for commercial IR sensors that can be attached to enclosures to monitor animal locomotion. IR components were soldered to circuits connected to a single microcontroller. These IR components were housed inexpensively using plastic tubing and cork discs to further focus and extend detection of the

IR beam. A standard personal computer recorded data from circuit boards connected to an inexpensive interface. This system may be used in a range of lighting conditions without requiring readjustment or recalibration. Validation of our equipment design was done with

127 male Sprague Dawley rats treated with reserpine 22 hours prior to administration of saline or

L-DOPA (125 mg/kg). Data was collected in eight different measures: horizontal activity, immobile time, elevated activity, centre elevated activity, elevation time, elevation bout, and repeated and non-repeated movement while elevated. L-DOPA increased horizontal movement and all elevated activity excepting elevated movement and centre elevated movement, demonstrating selective drug effects. The total cost of our complete IR system

(US$517. 45) was substantially less than the least expensive quote (US$19,666.90) obtained for a commercial IR system. We have successfully designed and constructed a flexible and inexpensive IR system to monitor at least eight measures of rodent locomotion at a significantly lesser cost than quoted by commercial suppliers.

2.2 Introduction

The first infrared (IR) detector was made in 1800 by the discoverer of Pluto and IR light, William Herschel (Rogalski, 2012), and has been used to measure motor activity in rodents and other animals at least since 1970 (Fibiger et al., 1971). Despite the rapid rate of technological improvements, IR photocells remain a major method of measuring motor activity in rodents. The only real competitor to assess locomotion and rearing are camera based systems that are expensive and require equally expensive software. Camera systems do provide additional information in terms of path analysis compared to infrared technology, but these measurements are less often used in pharmacological research than quantified locomotion as this additional information may not be worth the enormous increase in both expense and time needed for data analysis.

128 hydroxydopamine-lesion rat or MPTP-treated primate models of Parkinson’s disease, the conditioned place preference test of reward, and exploratory/investigatory behaviour. The behaviours and locomotion recorded in these tests and models include total distance travelled

(Zakharova et al., 2009), spatial patterns of locomotion (Geyer et al., 1987), immobility/freezing (Gresack et al., 2010), hole pokes (Riley et al., 1979), time spent in specific areas/compartments of the apparatus (Dietz et al., 2007), drinking and eating behaviour (Jahng and Houpt, 2001) and rearing (Gresack et al., 2010), all of which can be recorded using IR technology. Analysis of these behaviours and patterns of locomotion have been used to understand basis of many human behaviours such as drug use (McLelland et al.,

2014), anxiety disorders (Pellow et al., 1985), neurological diseases (Johnston et al., 2005), and learning and memory (Cassel et al., 1998). Thus, access to inexpensive and effective IR devices is extremely valuable in animal research to develop insight into animal and human behaviour and allows many laboratories a wider selection of tests and models to study in a more affordable manner.

One of the greatest advantages to using IR technology is that, unlike mechanical switches, the activation of IR sensors is undetectable to the animal and therefore has minimal influence over its behaviour. Additionally, IR devices are compact, lightweight, and can be used in almost all laboratory light conditions. No newer technology has been able to compete with IR in both usefulness and price, especially as IR emitters and detectors have become extremely inexpensive, as compared to 30-40 years ago. Furthermore, the equipment needed to interface between the photobeam detectors and a computer or other data-logging device has improved considerably and become inexpensive and easily available online. However, costs of pre-made equipment from research equipment companies remains relatively high, often prohibitively so, especially considering how inexpensive the component parts have become, and how simple software is to write at present.

129

Some of the least expensive IR devices can range between US$50 and US$100 from research equipment supply companies (for example, one pair of IR emitters and detectors from a research equipment company that is widely used have a median price of around

US$75 at the time of writing, so that IR emitters and detectors 10 per box and 5 boxes would cost US$3,750), just for the IR devices, and excluding delivery. However, one IR emitter and detector pair can be bought online for US$0.33, at the time of writing, so that 40 can be bought from electronic supply stores for US$13.21, substantially less than from even the least expensive research equipment supply company. This expense does not include an interface to transmit IR beam break data to a computer for further processing (four channel IR controller from one company at the time of writing is approximately US$230, or another US$2,875), and does not include the cost of software, cable, or circuit boards that are generally required to accurately measure animal locomotion. Therefore, our goal was to construct a set of five enclosures as an inexpensive alternative to commercial IR devices and a computer interface that was compatible with most modern computers to measure the activity of five rats simultaneously.

2.3 Methods

Although employing IR technology can vary in complexity, our purposes required measures of both horizontal motor activity (locomotion) and elevated activity (elevated posture and behaviour). Thus, similar to commercial equipment, we created unidirectional IR beams. Our microcontroller detects the break in IR beams when an animal enters the beam and prevents the IR light from reaching the sensor. For these enclosures, the voltage in the IR sensor circuit changes when the beam is broken or unbroken. This voltage is converted to a digital signal, transmitted to a connected computer and recorded. We can thus measure number of beam breaks as well as the duration at which a beam remains broken.

130

2.3.1 Construction

2.3.1.1 IR emitter/sensor pairs

(approximately 30 cm) without operating at maximum capacity. In order to create a detectable IR beam, IR emitters were paired with IR sensors, creating a unidirectional IR beam that was sensitive to interruption.

Although only a 30 cm distance was required, the IR emitters were able to transmit up to 80 cm by supplying 100 mA forward current and increasing the resistance in the IR receiver circuit up to 600,000 ohms. However, transmission over this distance created a substantially larger beam width which was likely to interfere with nearby IR sensors.

We built a network of IR beams to measure horizontal and elevated activity of rats in

20 cm x 30 cm x 25 cm (w x l x h) Poly(methyl methacrylate)1 (PMMA) enclosures

(US$17.81 each). We purchased IR emitters (Model TSAL 7200) from element14

(au.element14.com, US$0.08 each) and sensors (Model Lite-On LTR-301) from Mouser

Electronics (au.mouser.com, US$0.25 each), classified as IR LEDs and NPN phototransistors, respectively. A 5 V DC power supply rated at 4 A ($22.16) was used; however this circuit could be constructed to accommodate a range of power supplies.

Accessories for this circuit to monitor five enclosures included resistors (US$4.68), 8core cable (US$56.13), circuit boards (US$18.71), enclosing boxes (US$5.61), and connector wires (US$18.71). Extras such as glue, power connector plugs, ferrules, solder, and croc clips totalled US$80.52.

1 Commonly known by the trade name Plexiglas®, Perspex®, or Lucite®

131

In order to calculate the requirements of the power supply, it was necessary to verify how much power the IR emitters needed to form a reliable IR beam. The circuit voltage output of the IR sensor determined the strength of the IR beam: the stronger the beam, the lower the voltage is. As the microcontroller discerns a voltage below 1.2 V as a digital zero

(0), it was decided that a voltage change of roughly 0.5 V would be considered reliable in order to rule out any false beam breaks. As we were using 10 IR emitters per enclosure and 5 enclosures, we used the equation:

75 mW x 50 IR emitters = 3.75 A; Equation 1

2.3.1.2 Housing

Shielding is another factor to consider when constructing IR LEDs and sensors.

Although the IR light emitted from an IR LED is invisible to humans (and many other animals), it emits light over a wide field. Therefore, distances larger than 2-5 cm between the

LED and sensor require shielding and housing of the LED (and sometimes the sensor to avoid LED emission overlap, see Figures 2.1A-B) to focus the light towards the sensor more effectively. One solution is to add a lens to the housing unit of the LED to further focus the

IR light towards the sensor (Batson and Turner, 1986). However, this is optional as the equipment can still be effective without the lens (Wilson, 2004, 1996; Wilson et al., 2000;

Wilson et al., 1992).

Each IR emitter and sensor was placed inside a 60 mm length PVC tube with a 16mm and 20 mm diameter opening obtained from a local hardware store (US$0.18 each, 40 PVC tubes totalling US$7.20 per enclosure). The 16 mm end of each tube was attached to the outside wall of the enclosure to ensure uniformity and focus of the IR light towards to the sensor. Before the emitters were soldered to wire, they were mounted onto the middle of individual cork discs (US$3.05 for a pack of four) by pushing the pins gently through the cork and securing the main component with glue. The sensors were mounted by gluing the

132 head of each sensor to the middle of the cork discs. The cork was mounted onto the 20 mm end of the plumbing tube with glue, although when mounting the sensors, grooves were made in the edge of the tubing to accommodate the vertical pins. The IR LEDs and sensors were then soldered to wire so that the voltage of the sensor could be measured while the LED and sensor were aligned on the wall of the enclosure. This was to ensure LED and sensor pairs were properly aligned with no IR overlap from the neighbouring emitters. The shielding of these emitters and sensors also helped to mitigate any effects of the widening beam width on locomotion data recording by narrowing the width of the beam that could be detected by the sensor to create a more uniform signal across the length of the cage.

A) B)

Figure 2.1. Schematic diagram demonstrating the effect of housing IR LEDS. IR light emitted from LEDs (light grey) in the animal enclosures A) overlapping across three different sensors (light grey) due to wide emission field and B) targeting aligned sensor (dark grey) after being housed to narrow the IR beam and reduce emission field covered.

2.3.1.3 Attachment

Although versions of IR equipment include frames with IR sensors and emitters attached in which an animal enclosure can be placed (for example, AM1053 activity monitors from Linton Instrumentation, UK), we attached the housed emitters and sensors to the enclosure for increased mobility and decreased storage. This does have the drawback of

133 decreased flexibility regarding enclosure replacement and altered the placement of the sensors and emitters. However, this was not a disadvantage for our particular study, but should be considered for individual experiments requiring custom-made equipment. This design also has the advantage of avoiding the issue with using home cages where scratches and other discolourations and dirt on the walls can impede the signal by allowing holes to be drilled where the emitters and sensors are located to avoid the impediment from the walls.

The housed sensors and emitters were attached to the outside walls of the enclosures using glue after holes were drilled into the walls to allow for maximum IR beam detection by the sensors. Drilling holes into enclosure walls is optional although as PMMA is transparent to IR light. However, as mentioned above, enclosure walls are susceptible to markings, scratches that come with wear and tear and the mess that animals make. The emitters and sensors were permanently attached to the walls so that the IR beams would transect the enclosure on the lower and higher levels. The lower level consisted of seven pairs of emitters and sensors attached 3 cm above the floor of the enclosure (three pairs along the width and four pairs along the length of the enclosures) at 4 cm intervals, while the upper level consisted of three pairs of emitters and sensors attached to the width of the enclosure at 4 cm intervals, 8 cm above the floor of the enclosures. These measurements were chosen to ensure that movement by the animal would break at least one IR beam in both the X and Y axes of the enclosure. The arrangement of the lower level pairs was designed to record coordinates of the rat’s movement to determine whether the rat was in the peripheral or central areas in the enclosure and record exploratory movement measured as beam breaks. The arrangement of the upper level pairs was designed to record rearing behaviour rather than coordinates of the rat’s location. The area covered by the upper level enclosures was tested and found to cover the vast majority of the enclosure area horizontally. The areas that were not reached by the upper level IR pairs were found to be substantially smaller than the size of a 225-325 gram

134 rat and thus a beam break would be registered even if the animal was rearing in the uncovered areas. Of course, more IR emitters and detectors can be added if desired.

The wires were connected to a circuit board placed in a container resting on the lid of the enclosures. This ensured uniformity throughout the enclosure, minimal tangling of the wires and allowed for easy enclosure cleaning.

2.3.2 Computer and software

The output voltage from the IR sensors were fed to the inputs of the Fez Panda II microcontroller (one microcontroller used, purchased for US$53.57 at the time of writing).

The microcontroller firmware (freely available from https://github.com/cewinkler/Loco.Firmware ) was programmed to translate the inputs to sensors on enclosures. While turned on, the microcontroller was programmed to read the sensors at a rate of 2 Hz and translate the readings into 8-bit binary strings which indicate what state each beam in the enclosure is in. The first four bits indicate the X-axis beams; the fifth to seventh bits indicate the Y-axis beams; and the last bit indicates if any of the upper level beams are broken. The readings are transmitted through the serial port of the microcontroller, which can be received by a PC using a standard universal serial bus (USB) connection.

Software was written using the C# programming language to receive and record the microcontroller data (freely available from https://github.com/cewinkler/Loco.Collection ).

The software was required to listen for the microcontroller serial data; determine the source enclosure; and save the data accordingly. The software saved the enclosure data to files based on information provided by the user. The software also gave a visual representation of beam states for each enclosure.

135

To ascertain locomotor measures from the data, additional software was written using the C# programming language (freely available from https://github.com/cewinkler/Loco.Measures ). The locomotor measure extraction software was designed to read the saved raw data files from the receiver software and calculate the locomotor measures. This process generated comma separated files (CSVs) for each locomotor measure for importing in to R for statistical analysis.

2.3.3 Experimental procedures

To validate the accuracy of our custom-made enclosures in measuring animal behaviour and locomotion, we studied the effect of the dopamine precursor L-3,4- dihydroxyphenyl-alanine (L-DOPA) or saline on rodent behaviour.

2.3.3.1 Animals

Experimental procedures were performed with approval from the University of

Western Australia Animal Ethics Committee and in accordance with the Australian Code of

Practice of Use of Animals for Scientific Purposes and relevant Western Australia State law.

A total of 30 drug and test naïve male Sprague Dawley rats, originally weighing 225-275 grams upon arrival from the Animal Resource Centre of Western Australia, were housed in pairs, in clear polypropylene 47 x 37 x 20 (l x w x h) boxes with woodchip bedding. The animal housing room was temperature controlled and maintained at approximately 22 + 2OC under a 12 hour light-dark cycle (lights on at 6:00 a.m.). Rats had free access to food and water except during testing. Each rat was handled for 2 min daily for 5 days prior to behavioural testing. After reserpine administration, rats were given free access to soft feed and Necta gel packs to minimize health effects of akinesia.

136

2.3.3.2 Drugs

Reserpine (Sigma-Aldrich) was freshly dissolved in 30-40% captisol solution

(captisol dissolved in distilled water) before each administration. L-3,4- dihydroxyphenylalanine methyl ester hydrochloride (Sigma Aldrich) and benserazide were prepared using saline. The pH of all solutions was approximately 6.5-7.0. Saline was used as a vehicle treatment. All drugs were administered at 1ml/kg intraperitoneally.

2.3.3.3 Testing

All rats were administered reserpine (4 mg/kg, freshly dissolved in 30% captisol solution, i.p.) 22 hours before testing. All rats were randomly allocated to received L-DOPA

+ benserazide (125/31.5 mg/kg, n = 10, herein referred to as L-DOPA) or saline (1 mg/kg, n

= 20). Rats were habituated to the testing boxes for an hour before being injected (i.p.) with

L-DOPA (125 mg/kg) or saline (1mg/kg) and tested immediately afterwards for 80 minutes.

All testing was conducted between 10:00 am and 5:00 pm. This testing regime was designed for a separate study currently being prepared for publication.

Our custom-made enclosures were programmed to record five primary measures: horizontal movement (HM) was determined by counting the number of beam breaks on the lower level of IR pairs in the testing time, immobile time (IMT) was determined by counting the number of seconds in which no beams states changed. Alternatively, horizontal movement may be translated into distance travelled by multiplying the number of beams broken on the x or y axes by the distance between the IR devices on each separate axis. For example, the distance between IR devices along an enclosure wall was 4 and 3 cm for the X and Y axes, respectively. The relevant equation used to translate horizontal movement to distance travelled would be the following:

137

Distance travelled = [(distance between IR devices)*(beam breaks on x axis)] + [(distance

between IR devices)*(beam breaks on y axis)]; Equation 2

However, this equation could only be effectively implemented using equipment with more IR devices to reduce the error of assuming the entire length of a square has been travelled.

Elevated movement (EM) was defined by beam breaks in the upper level IR pairs, elevation time (ET) was determined by measuring how long an upper level beam was broken for, centre elevated movement (CEM) was measured similarly to EM with the addition of analysing the coordinates of the lower level beam breaks that occurred simultaneously with the upper level beam breaks. Thereby, we could determine the location of the rat on the floor of the enclosure during rearing and whether the rat was in the centre or periphery of the enclosure floor in addition to its overall movement. Repeated movement while elevated

(RME) was incremented when one lower level beam was consecutively broken at least twice at the same instance that the upper levels beams were broken. Comparatively, non-repeated movement while elevated (NRME) was calculated after data collection using the following equation:

(Total movement while elevated) – (Repeated movement while elevated); Equation 3

Although not recorded by the equipment itself, a new measurement termed ‘elevation bout’ was also included in the analysis. Elevation bout (EB) is determined by the equation:

(Elevation time)/(Elevated movement +1); Equation 4

An animal was considered to be in an elevated position when grooming or rearing.

Therefore, measurements such as elevation bout, elevation time, and elevated movement represent the summation of grooming and rearing behaviour.

2.3.4 Statistical data analysis

138

A mixed two-way Treatment x Time ANOVA was used for comparison of locomotor measures between of groups. Mean locomotion totals between groups were analysed with a two-tailed Student’s t-test. Statistical significance was defined as p < 0.05.

2.4 Results

2.4.1 Data collection

For the purpose of our study, the enclosures, interface, and software program were designed to detect and record HM, IMT, ET, EB, EM, CEM, RME, and NRME either collated in 5 minute bins (Figure 2.2A-H) or as a combined total for the 80 minute testing session (Figure 2.3A-H). Although the data was collated in 5 minute bins, our software allows data collation in any bin length.

139

Horizontal Movement Immobile Time

Elevation Time Elevation Bout

Elevated Movement Centre Elevated Movement

Repeated movement Non-repeated movement while elevated while elevated

 L-DOPA group  Saline group Figure 2.2. Mean Locomotor activity + SEM of Sprague-Dawley rats treated with saline (n = 20) or L-DOPA (n = 10) A) Horizontal activity measured as number of beam breaks per 5 minute bin B) Immobile time measured as time (seconds) when the animal remained motionless per 5 minute bin C) Elevation time measured as length (seconds) of beam break in the top level of emitters and sensors only per 5 min bin D) Elevation bout measured as length (seconds) of upper level beam break per elevation movement per 5 minute bin E) Elevated movement measured as number of beam breaks located in the top level of IR emitters and sensors only per 5 minute bin F) Centre elevated movement measured as the number of beam breaks located in the top level of IR emitters and sensors while located in the centre of the enclosure floor per 5 minute bin G) Repeated movement while elevated as measurement by repetitive horizontal movement during upper beam break and H) Non-repeated movement while elevated as measured as remaining horizontal movement during upper beam beak during testing session in enclosures with transecting unidirectional IR beams using IR emitters and sensors.

140

Figure 2.3. Mean locomotor activity + SEM of Sprague Dawley rats treated with saline (n = 20) or L-DOPA (n = 10) over total testing session. A) Horizontal activity B) Immobile time C) Elevation time measured D) Elevation bout E) Elevated movement F) Centre elevated movement G) Repeated movement while elevated H) Non-repeated movement while elevated. * Significantly different compared to saline group (p < 0.05).

2.4.2 Monetary Cost of Equipment

The cost of our custom-made enclosures includes the Plexiglas® enclosures, IR devices, the Fez Panda microcontroller, power supply, housing units, resistors, circuit boards, terminal blocks, 8 core cable, solder, glue, ferrule, power connector plugs, and wires. Table

2.1. shows the price comparison between the IR system built for the current report and commercial IR systems at time of writing.

141

Equipment for current Quote 1 Quote 2 Quote 3 study (with enclosures/ (with enclosures/ (with (with enclosures/ without without enclosures/ without enclosures) enclosures) enclosures) without enclosures) Per $156.16/ $5,595.70/ $5,095.35/ $5,764.98/ unit $138.35 - $5,077.54 $5,507.68

For 5 $477.89/ $19,666.90/ $22,148.75/ $28,824.90/ units $388.84 - $22,059.70 $27,538.41

The total price for the IR system in the current study amounted to US$477.89 to measure the activity of five rats simultaneously. This amount is substantially less than quoted prices from commercial suppliers at the time of writing, with the least expensive commercial system being priced at US$19,666.90 for five locomotor enclosures in addition to set up materials purchased from commercial suppliers.

2.4.3 Behavioural and locomotor data

Initial analysis of the data revealed violations of homogeneity of variance and thus the data was transformed to square roots, which produced homogeneity of variance. The 80 minute testing session was divided into two 40 minute periods for analysis of drug effects over time.

In rats treated with saline or L-DOPA (125 mg/kg), the two-way mixed ANOVAs for

HM and IMT revealed a significant effect of treatment (F (1, 28) = 28.77, p < 0.001 and F (1,

28) = 24.60, p < 0.001, respectively), but not of period (F (1, 28) = 1.31, p = 0.26 and F (1,

28) = 2.73, p = 0.11, respectively) and a significant treatment x period interaction (F (1, 28) =

28.86, p < 0.001 and F (1, 28) = 30.88, p < 0.001, respectively). Pairwise comparisons

142 revealed that L-DOPA increased HM and decreased IMT in the second, but not the first period, compared to saline (p < 0.05).

Similarly, a two-way mixed ANOVA revealed significant treatment (F (1, 28) =

11.89, p < 0.01), but not period (F (1, 28) = 0.68, p = 0.42) effects and a significant treatment x period interaction (F (1, 28) = 21.73, p < 0.001) for ET. Pairwise comparisons showed that

L-DOPA increased ET in both periods compared to saline (p < 0.05).

Significant effects of treatment (F (1, 28) = 9.88, p < 0.01) and period (F (1, 28) =

7.26, p < 0.05) were seen in the ANOVA for EB as well as a significant interaction (F (1, 28)

= 17.99, p < 0.001). Pairwise comparisons showed that L-DOPA increased EB in both periods compared to saline (p < 0.05).

While there was no significant treatment effect on EM (F (1, 28) = 1.15, p = 0.29), a period effect was seen (F (1, 28) = 4.27, p < 0.05), but without a significant interaction (F (1,

28) = 0.002, p = 0.96) in the two-way mixed ANOVA. However, the mixed ANOVA for

CEM showed no significant effects for treatment (F (1, 28) = 1.53, p = 0.22) or period (F (1,

28) = 2.09, p = 0.15) and no significant interaction (F (1, 28) = 0.49, p = 0.49). Pairwise comparisons showed no significant differences between L-DOPA and saline over the two periods for EM (p > 0.05).

The ANOVAs for RME and NRME both revealed similar results as significant treatment effects were seen for both behaviours (F (1, 28) = 38.03, p < 0.001 and F (1, 28) =

11.02, p < 0.01, respectively), a non-significant effect of period (F (1, 28) = 4.00, p = 0.055 and F (1, 28) = 1.08, p = 0.31, respectively) and a significant interaction for RME (F (1, 28) =

143 saline. Overall, L-DOPA-induced locomotion and behavioural activity was significantly increased compared to saline-treated rats, and also increased significantly over time.

Independent student’s t tests for summated data over the entire testing session revealed that L-DOPA significantly increase HM, ET, EB, RME, and NRME (p < 0.05) and significantly decreased IMT (p < 0.05) compared to saline, while no differences were found between groups for EM and CEM (p > 0.05).

2.5 Discussion

The results of the current study demonstrated significant differences in select locomotion and behavioural measures of reserpine-treated rats administered L-DOPA or saline and tested in enclosures lined with custom-made IR devices and recorded using a custom-made data analysis program compatible with a standard PC. Furthermore, analysis of the data collected by our equipment also revealed significant differences in all measures over time. These differences suggest that our custom-made equipment and software detect and record selective drug-induced summated and time-based changes in rodent locomotion and behaviour. Furthermore, the increases in activity induced by L-DOPA compared to saline seen in the current study are in agreement with those reported previously in reserpine-treated rats (Johnston et al., 2005; Segovia et al., 2003). Therefore, our inexpensive design of an IR system is an effective alternative to commercial IR systems.

Although our specific enclosure design may not be suited for all studies, the IR devices, computer interface, and data analysis program are able to accommodate specific requirements of experiment designs in a number of different research fields including, but not limited to, pharmacology, psychology, and pathology. To highlight the inexpensive construction of our equipment, we obtained three separate quotes for complete IR-based animal locomotor monitoring systems, including five enclosures to monitor five individual

144 rats simultaneously, from commercial research equipment suppliers. Table 2.1. shows the price comparison between the IR system built for the current report and commercial IR systems at time of writing.

As mentioned previously, the attachment of the IR devices to the testing enclosure does decrease the flexibility of the size or species of animal that can be tested as the placement of the devices cannot be altered to accommodate these variables. However, other laboratories may overcome this potential disadvantage by attaching the IR devices to a frame surrounding the testing enclosure or by using an enclosure more suited to the animal to be studied.

The use of IR technology in animal research has important advantages as it is undetectable, unlike mechanical switches, inexpensive, unlike video cameras, and relieves the researcher of observing the animals at the time of testing, as bias is likely to influence study outcomes when observers are not ‘blind’ to experimental treatment of the animal. However, because of the often prohibitive expense of commercial IR systems and equipment, many laboratories may find it difficult to incorporate this versatile technology into their research.

145 of rodent behaviours with no faults or breakages in the equipment or in the software program and with significant differences found in behaviours after drug administration. It is important to note, however, that this design of apparatus provides information regarding level of activity and movement rather than more qualifying features of movement such as gait and fine motor movements.

Compared to commercial suppliers of IR-based rodent locomotor monitoring systems, the cost of constructing these systems is substantially less. As the data obtained from these enclosures also show significant statistical differences in animal locomotion, we have demonstrated that it is possible to construct inexpensive IR-based locomotor monitoring enclosures able to detect and record different drug-induced behaviour in rodents. IR technology can also be employed in a range of research fields and thus can contribute to further investigation of animal behaviour in a range of animal models and experiments.

146

Chapter 3 - An inexpensive alternative to commercial products in the recording of rodent ultrasonic vocalizations 3 Preface

In addition to antidyskinetic effects, this thesis also investigated the reinforcing effects of UWA-101 on Sprague Dawley rats using three tests including ultrasonic vocalisations. This aim of this chapter was to design and validate the construction of microphones to record ultrasonic vocalisations in rats on which future chapters of this thesis would be based. Adequate recording of ultrasonic vocalisations would indicate that these microphone capsules could serve as inexpensive replacements of commercial bat detectors and allow for easy use due to their flexible design.

3.1 Abstract

Ultrasonic technology is a flexible and effective way of measuring ultrasonic vocalizations of animals. However, the cost of most commercial ultrasonic recording equipment can be limiting. In this study, we describe the design and construction of a flexible and inexpensive recording system to detect and record ultrasonic vocalisations (USVs) at a reduced cost than that quoted by commercial suppliers. Panasonic electret microphone capsules were soldered to cables and stereo jacks, and plugged into sound cards connected to

Raspberry Pi mini-computers. Using a wireless router, a personal computer was used to remotely control the Raspberry Pis and commence recording using Python software. Initial validation of this novel ultrasonic recording system was achieved using artificial ultrasonic tone emitters. Additional validation involved recording USVs emitted by male Sprague

Dawley rats treated with saline (1 mg/kg) or dexamphetamine (1.5 mg/kg). Using a commercial bat detector with a recording capacity of 20-120 kHz as a recording reference, the Panasonic microphones reliably recorded ultrasonic frequencies from an artificial source

147 up to 40 kHz. The microphones also reliably recorded 22 kHz USVs emitted from saline- treated rats, but did not detect 50 kHz USVs emitted from dexamphetamine-treated rats. The total cost of our complete recording system with the capacity to record the 22 kHz USVs of four rats simultaneously (US$327.72) was substantially less than a commercially purchased equivalent (up to US$5703.16).

3.2 Introduction

Ultrasound is a sound pressure wave that has a frequency greater than the upper limit of the human hearing range (about 20 kHz). Technology capable of detecting or recording these ultrasonic frequencies has been used in a range of research and industrial fields for decades (Dunn & Fry, 1959; Hueter & Bolt, 1951) and applied to electronic appliances (Du et al., 2011), civil engineering (Ciancio & Helinski, 2010), and other areas of science (Banahan et al., 2014; Wohr & Schwarting, 2008). Using this technology, researchers have developed a greater understanding of a range of phenomena in fields such as physics, chemistry and biology. Ultrasonic emitters and receivers have also been useful in the study of animals such as bats (Flaquer et al., 2007), rodents (Ahrens et al., 2009; Berger et al., 2013), whales

(Samarra et al., 2010), porpoises (Wang et al., 2005), and fish (Mitamura et al., 2005).

In rodents, ultrasonic research has recently focused on the hypothesis that ultrasonic vocalisations (USVs) are indicative of the affective states of these animals (Burgdorf et al.,

2007; Hamed et al., 2012; Mallo et al., 2007). Specifically, USVs in rats are thought to represent positive or negative affect as determined by the frequency range and shape of such vocalizations. USVs representative of positive affect are known as 50 kHz USVs as they often occur within the frequency range of 40-60 kHz. These USVs are generally short (<300 ms) and can occur as flat USVs, upward/downward ramps, trills, flat-trill combinations, and step up/step down USVs (Wright et al., 2010). USVs representative of negative affect are

148 known as 22 kHz USVs as they occur mostly in the 18-30 kHz frequency range, particularly around 22 kHz. These vocalizations tend to be longer (300-1000 ms) and often occur as a singular call rather than a collection of shorter USVs as found with 50 kHz USVs (Wright et al., 2010). USVs emitted by rat pups are known as 40 kHz USVs and can occur between 30-

50 kHz and are often parabolic or hyperbolic in shape (Brudzynski et al., 1999; Hashimoto et al., 2004). USVs are categorized as representing positive or negative affect by analysing the contexts in which they are emitted. For example, 50 kHz USVs are often emitted during play, sexual intercourse, or drug self-administration (Berger et al., 2013; Burgdorf et al., 2008;

Burgdorf et al., 2011; Mallo et al., 2007; Wright et al., 2010). Adults emitting 22 kHz USVs do so when a perceived threat is present such as the release of cat odour in fear conditioning, aggression due to the presence of an intruder, or during electric shock testing, drug- withdrawal or pre-pulse inhibition testing (Berger et al., 2013; Borta et al., 2006; Graham et al., 2009; Kim et al., 2010; Litvin et al., 2007; Tunstall et al., 2009). Rat pups most commonly emit 40 kHz USVs when separated from their mother (Insel et al., 1986; Naito &

Tonoue, 1987).

Investigations into these USVs have increased our understanding of the emotional state of rats in a range of affect models. In particular, 22 kHz USVs have been useful in studies of stress and anxiety as they are associated with negative affect in adult rats. The analysis of 22 kHz USVs in adult rats has provided insight into anxious emotional states and their relation to substance use such as the link between psychological stress and increased ethanol consumption (Pohorecky, 2008). Studies have also used 22 kHz USVs in fear conditioning (Graham et al., 2009), drug withdrawal (Berger et al., 2013), as a measure of social interaction and fear transmission (Kim et al., 2010), and in the study of neural substrates involved in fear and anxiety (Allen et al., 2007).

149

While USVs are attracting considerable interest in research fields such as pharmacology and psychology, the cost of the equipment required to record and analyse these

USVs can be a considerable obstacle. Commercial ultrasonic microphones, most commonly sold as bat call detectors usually recording frequencies between 15-125 kHz, and the corresponding software used to analyse the recordings generated by these microphones can cost thousands of dollars. Well-known suppliers price ultrasonic microphones between

US$420 – $1800 each, with software costing an additional US$195 – $3100. These costs can impact significantly on smaller laboratories and limit research opportunities. Therefore, we describe the construction and testing of an inexpensive and flexible alternative to commercial ultrasonic microphones to record 22 kHz USVs in adult rats than can accommodate simultaneous recordings from multiple microphones.

3.3 Methods

3.3.1 Microphones

The microphones used to record the 22 kHz USVs were Panasonic WM-61A electret condenser microphone capsules purchased online for US$4.64 each (or US$18.56 for four microphones, excluding US$0.46 postage fee). These microphones were soldered to 50-ohm

2 core black twin audio shielded cable (US$0.87 per metre) and attached to a 3.5 mm stereo jack plug (US$1.15 each).

3.3.2 Recording system

The recording system was based on four Raspberry Pi (Model B) mini-computers

(US$38.33 each) connected to a TP-LINK router (US$45.80) via ethernet cables for wireless remote control via a personal computer (PC). Each Pi had one sound card (US$2.75 each) connected via a universal serial bus (USB) hub (US$2.75 each) and a SD memory card

150

(US$7.34 each), and was powered by a USB plugpack adaptor (US$9.17 each). Python software was used for recording commands.

3.3.3 Validation

Initial testing of the Panasonic microphone input frequency range was conducted using common speakers on a PC with open-source tone generation software. The capability of these microphones to record ultrasonic sounds was compared to that of a commercial bat detector (US$104.85 each) to determine their effectiveness in accurately recording ultrasonic frequencies up to 30 kHz, as 22 kHz USVs often fall within the range of 18-30 kHz. The

Panasonic microphones were then used to record frequencies from a 40-kHz transducer.

These recordings were analysed to determine the accuracy of detection by these microphones.

After these tests, the microphones were used to record the 22 kHz USVs of adult rats.

3.3.4 Animals

Experimental procedures were performed with approval from the University of

Western Australia Animal Ethics Committee and in accordance with the Australian Code for

Care and Use of Animals for Scientific Purposes, and relevant Western Australia State law.

Twenty male Sprague Dawley rats, weighing 180-220 grams upon arrival from the Animal

Resource Centre of Western Australia, were housed in pairs, in clear polypropylene 47 x 37 x

20 (l x w x h) boxes with woodchip bedding. The animal housing room was temperature controlled and maintained at approximately 22 + 2OC under a 12-hour light-dark cycle (lights on at 6:00 a.m.). Rats had free access to food and water except during testing. Each rat was handled for 2 min daily for 5 days prior to behavioural testing.

151

3.3.5 Apparatus

Rats were tested in polypropylene 47 x 37 x 20 (l x w x h) boxes identical to the housing boxes. One bat detector and one Panasonic microphone were attached to the top of a box wall to record USVs. The bat detector was used as a recording reference for the

Panasonic microphone.

3.3.6 Drugs

(+)-Amphetamine sulphate (referred to as dexamphetamine) was dissolved in saline and administered at 1.5 mg/kg (dose base on previous literature: Ahrens et al., 2013; Simola et al., 2012). Saline was administered at 1 ml/kg. All injections were intraperitoneal. The data reported here served as the baseline component to a pharmacological experiment to be reported elsewhere, hence the injections were administered.

3.3.7 Procedure

All testing took place between 9:00 a.m. and 3 p.m. On the test day, the rats were weighed and given a saline or dexamphetamine injection immediately prior to being placed into testing boxes. These testing boxes were lined with bedding from each individual rat’s housing box and were cleaned with ethanol after each recording session. Testing boxes contained bedding as previous studies have reported significant difference in USV calling rates in rats tested in boxes with and without bedding (Natusch & Schwarting, 2010). Rats were habituated to testing boxes (one rat per box) for 20 min before USVs were recorded for a 10-min period.

152

3.3.8 USV recording analysis

Acoustic data were recorded with a sampling rate of 140,000 Hz as a wav file. For acoustic analysis, recordings were transferred to Audacity® (free download from http://audacity.sourceforge.net/) and transformed to high resolution spectrograms. After noise removal, USVs were counted by an experimenter blinded to the treatment of each group.

3.4 Results

3.4.1 Artificial ultrasonic emission

Both the Panasonic microphones and the commercial bat detector accurately and clearly recorded the ultrasonic frequencies emitted by speakers connected to the PC. Figures

3.1A -B show the comparison between to two devices during these recordings. The Panasonic microphones also reliably recorded output from a 40-kHz transducer as shown in Figure 3.2.

Figure 3.1. Recording of an ultrasonic tone emitted by speakers from the PC on A) Panasonic microphone electrets and B) a commercial bat detector after transformation of ultrasonic input by a factor of 10.

153

Figure 3.2. Recording of a 40 kHz signal from an ultrasonic transducer on a Panasonic microphone

3.4.2 Ultrasonic vocalisations

The recordings by the Panasonic microphone was compared to the recordings by the commercial bat detector based on the quality of USVs in different frequency ranges.

Although the bat detector reduced the overall frequency of the USVs by a factor of 10 to transform the USVs to an audible format, the shape and length of the USVs remained unchanged and were therefore a useful comparison to the Panasonic microphones.

3.4.2.1 22 kHz ultrasonic vocalisations

As 22 kHz USVs are known to occur between 18-30 kHz, we used this range to classify USVs recorded by our microphones. Only one saline-treated rat emitted USVs within the range of 18-30 kHz on both microphones (Figures 3.3A-B). These USVs were consistent with the shape and length of 22 kHz USVs as described in previous studies. These USVs

154 were also the same shape and length in both the commercial bat detector and the Panasonic microphone recordings. No 22 kHz USVs were emitted by the dexamphetamine-treated rats.

A) B)

3.4.2.2 50 kHz ultrasonic vocalisations

50 kHz USVs occur between 40-60 kHz, and thus this range was used to identify 50 kHz USVs in addition to the known shapes and length of 50 kHz USVs in spectrograms of previous studies. Dexamphetamine, but not saline-, treated rats were found to emit 50 kHz

USVs using the commercial bat detector recordings, but not the Panasonic microphone recordings (Figure 3.4).

155

A) B) Figure 3.4. Recording of 50 kHz A) trill call and B) flat call from a commercial bat detector emitted by a dexamphetamine-treated rat. Internal circuitry of the bat detector translated the input frequency by a factor of 10 during spectrogram analysis.

3.4.3 Cost of equipment

The cost of our custom-made microphones includes the Panasonic WM-61A electret

microphone capsules, the shielded cable, stereo jack plugs, Raspberry Pi mini-computers, TP-

LINK router, sound cards, USB hubs, SD memory cards, and plug packs. Table 3.1 shows the

price comparison between these microphones and accessories and commercial USV

recording systems.

The total price for the USV recording system used to record the USVs of four rats

simultaneously in the current study amounted to US$327.72. This amount is substantially less

than the prices quoted by commercial suppliers at the time of writing, with the least

expensive commercial systems being priced between US$833-$1004 for four simultaneously

recording ultrasonic microphones remotely controlled by a PC.

156

Table 3.1. Quotes from six different commercial suppliers for ultrasonic microphones and corresponding software to record ultrasonic vocalizations of four rats simultaneously. All prices are in USD and correct as of 12/11/2014. Product Quote 1 Quote 2 Quote 3 Quote 4 Quote 5 Quote 6 Per Microphone 695 999 139.08 543.23 504.27 104.85 4 Microphones 2780 3996 556.32 2172.92 2017.08 419.4 Software 195 399 - 1552.08 575.38 - Accessories 309.16 309.16 309.16 309.16 309.16 309.16 Total 3979.16 5703.16 1004.56 4577.39 3405.89 833.41

Not only did we find a substantial difference in the cost of the equipment, but the small size

of the Panasonic microphones increased their versatility compared to (0.5 x 0.5 x 0.5 cm)

than commercial ultrasonic detectors. This allowed us to place the microphones in the testing

cage more discreetly than the bat detector which was encased in a 3 x 4 x 15 cm plastic

covering.

3.5 Discussion

The results of the current study demonstrate that microphones designed for recording

sonic frequencies can be used to effectively detect and record artificial and biological

ultrasonic frequencies. Indeed, the custom-made Panasonic WM-61A microphones used in

this study clearly, accurately and simultaneously recorded 22 kHz USVs emitted by saline-

treated rats compared to a commercial bat detector. These simultaneous recordings were

made possible by the connection of Raspberry Pi mini-computers to a PC via a wireless

router. This inexpensive ultrasonic recording system represents an effective alternative to

commercial bat detectors or ultrasonic microphones and associated software and hardware.

The small size and portability of the components of this recording system and the total cost of

US$327.72 may enable research in laboratories previously limited by the size, portability or

cost of the equivalent commercial products. Indeed, our four custom-made Panasonic

157 microphones cost $40, which was less than 4% of the average cost of four commercial microphones.

As USVs are thought to indicate affective states in rats, much research has focused on anxiety in rats through tests such as fear conditioning and intruder aggression with 22 kHz

USVs being analysed in addition to physical behaviour (Graham et al., 2009; Kim et al.,

2010). As the Panasonic WM-61A microphones have sufficient sensitivity to detect and record frequencies up to 40 kHz, their employment in research into 22 kHz USVs by adult rats or 40 kHz USVs by rat pups would provide an inexpensive and effective alternative to commercial ultrasonic recording systems. As these Panasonic microphones do not detect 50 kHz USVs, research into the positive affective state of rats using these microphones is limited. As the recording system is also versatile, other electret microphone capsules could be used to replace the Panasonic microphones to detect and record frequencies above 40 kHz such as those seen in reward or anti-anxiety tests.

The use of ultrasonic recording technology in animal research has important advantages as it is undetectable, inexpensive, and relieves the researcher of observing the animals at the time of testing which may influence behaviour. We have designed an ultrasonic recording system to detect rodent USVs using inexpensive Panasonic microphones as an effective alternative to expensive ultrasonic recording equipment from commercial suppliers. The results from this study demonstrate inexpensive microphones can detect and record USVs in adult rats and can be employed in a range of research fields.

158

Chapter 4 - Influence of antidyskinetic compounds on L-DOPA- induced locomotion and behaviour in reserpinised rats 4 Preface In Chapter 2, we demonstrated that the custom-made infrared locomotion equipment successfully detected locomotive behaviour of the rats tested and that reserpine and L-DOPA significantly lowered and increased the level of activity and locomotion, respectively. This chapter aims to expand on these findings by observing the effect of UWA-101 on locomotion in the L-DOPA-treated reserpinised rat model. This chapter describes the dose-dependent effect of UWA-101 on the locomotion of reserpinised rats in the presence or absence of L-

DOPA as compared to previously studied antidyskinetic compounds.

Publication: McLelland, AE, Winkler, CE, Piggott MJ, Martin-Iverson, MT. Influence of antidyskinetic compounds on L-DOPA-induced locomotion and behaviour in reserpinised rat. Publication in process.

4.1 Abstract

The L-DOPA-treated reserpinised rat model is a potentially useful and predictive model of L-DOPA-induced dyskinesias in Parkinson’s disease in screening for antidyskinetic drugs. L-DOPA-induced horizontal and elevated activity correlate with antiparkinsonian and dyskinetic effects, respectively. We aimed to validate the L-DOPA-treated reserpinised rat model by testing drugs previously shown to exert antidyskinetic effects in animal models of

L-DOPA-induced dyskinesias or Parkinson’s disease patients. L-DOPA-treated reserpinised rats were administered yohimbine (10 mg/kg), riluzole (3 mg/kg), rimonabant (3 mg/kg), or

UWA-101 (1, 3, and 10 mg/kg) before locomotive monitoring. Yohimbine, riluzole, rimonabant, and UWA-101 (1 and 3 mg/kg) significantly attenuated L-DOPA (125 mg/kg)- induced time spent in an elevated position and repetitive movements while elevated. L-

DOPA-induced elevation time, non-repetitive movements while elevated and horizontal movements were all reduced by UWA-101 (1 and 3 mg/kg), riluzole, and yohimbine, while

159 only central elevated movements were reduced by riluzole. The highest dose of UWA-101

(10 mg/kg) exerted no significant effects on L-DOPA-induced rodent behaviour.

Interestingly, the lower doses of UWA-101 (3 and 10 mg/kg) were the most effective doses in reversing reserpine-induced behaviour in the absence of L-DOPA compared to the other antidyskinetic compounds. These results support the association between antidyskinetic effects and a reduction in L-DOPA-induced elevated activity in the reserpinised rats.

However, the reduction in horizontal activity caused by most of the drugs suggests a decreased antiparkinsonian effect, supporting the view that antidyskinetic drugs balance a reduction of dyskinesias with a reduction in L-DOPA efficacy, thus requiring dose-titration.

4.2 Introduction

In Parkinson’s disease (PD), L-3,4 dihydroxyphenylalanine (L-DOPA) remains one of the most effective treatments for major motor symptoms. Most PD patients require chronic L-

DOPA therapy, although this frequently leads to motor complications such as dyskinesias

(Nadjar et al., 2009; Rascol et al., 2003; Rezak, 2007). These L-DOPA-induced dyskinesias

(LID), defined as involuntary and repetitive movements, are thought to emerge because of excessive dopaminergic activity, particularly in the striatum (Buck & Ferger, 2008; Carta et al., 2006; Meissner et al., 2006).

The severity and frequency of LID are influenced by glutamate release inhibition, cannabinoid CB1 receptor antagonism, α2 adrenergic receptor antagonism, and serotonin transporter (SERT) or dopamine transporter (DAT) inhibition. SERT inhibitors are effective in reducing dyskinesia-like behaviour in rodents, possibly through indirect stimulation of presynaptic 5-HT1A autoreceptors, inhibiting further monoaminergic release (Bishop et al., 2012;

Bishop et al., 2006; Dekundy et al., 2007; Yamato et al., 2003). Antidyskinetic effects have also been observed with the monoamine transporter reverser 3,4-methylenedioxy-N-

160 methylamphetamine (MDMA) (Bishop et al., 2006; Crespi et al., 1997; Johnston et al., 2012;

Nash & Brodkin, 1991) and its analogue UWA-101 (Johnston et al., 2012) in 6-OHDA- lesioned or reserpinised rats and MPTP-lesioned primates administered L-DOPA. In both cases, the R-enantiomers have been shown to be responsible for the antidyskinetic effects originally observed in the racemates (Huot et al., 2011, 2014). These antidyskinetic effects were proposed to occur through dual DAT/SERT inhibition as R-MDMA and UWA 121 (the

R-enantiomer of UWA-101) shows reasonable affinity for both proteins and UWA-101 inhibited both serotonin and dopamine reuptake in a functional assay (Johnston et al., 2012).

Because some pharmacological animal models, such as the 6-OHDA-lesioned rat model, require costly personnel, time and resources to assess antidyskinetic compounds, researchers have investigated other models to screen potentially antidyskinetic compounds more quickly and efficiently before advancing on to lesion models. One such model is the acute reserpinised rat model of PD. Reserpine irreversibly inhibits both peripheral (VMAT1) and central nervous system (VMAT2) vesicular monoamine transporter isoforms, with the latter leading to a loss of presynaptic vesicular storage capacity and depletion of noradrenaline, serotonin, and dopamine in the brain (Erickson et al., 1996; Kirshner, 1962; Lohr et al., 2014; Weihe et al., 1994). The dopamine depletion leads to parkinsonian-like akinetic states and hind limb rigidity in animals, the reversal of which indicates potential antiparkinsonian efficacy of new agents (Fernandes et al., 2012; Finberg & Youdim, 2002; Salamone & Baskin, 1996). High dose L-DOPA in reserpinised rats can also induce specific behavioural correlates of LID in humans and other animal models (Johnston et al., 2005; Segovia et al., 2003). The reserpinised rat model of LID is based on correlations between vertical activity (rearing) and dyskinesia and between horizontal activity and parkinsonism (Johnston et al., 2005; Johnston et al., 2012; Segovia et al., 2003).

Because the rats are given a L-DOPA challenge 18-24 hours after reserpine administration, the time taken to develop abnormal behaviour is markedly reduced as well as decreased personnel

161 and resource costs compared to the 6-OHDA-lesioned rat model, making this model a quick and efficient alternative for preliminary antidyskinetic drug screening.

Although vertical activity in the L-DOPA-treated reserpinised rat model is described as a correlate of dyskinesia-like behaviour, this activity has previously been based on rearing alone

(Johnston et al., 2005; Johnston et al., 2012; Segovia et al., 2003). Studies have described a range of dyskinetic behaviours including chorea and dystonia in humans and orolingual, axial, locomotive, and limbic abnormal involuntary movements in animals (Cenci et al., 1998;

Dekundy et al., 2007; Gomez-Ramirez et al., 2006; Johnston et al., 2012; Kobylecki et al., 2010;

Nutt, 2003). It is therefore of interest to incorporate measurements of abnormal L-DOPA-induced elevated posture and behaviour in addition to rearing to more accurately represent the complex side-effects associated with L-DOPA treatment in this model.

Thus, the aim of this study was to validate and refine upon the reserpinised rat model by incorporating additional measurements of abnormal L-DOPA-induced behaviour after treatment with drugs exhibiting antidyskinetic efficacy in the 6-OHDA-lesioned rat model,

MPTP-lesioned primate model, and/or clinical trials of LID.

4.3 Methods

4.3.1 Animals

Experimental procedures were performed with approval from the University of

Western Australia Animal Ethics Committee and in accordance with the Australian Code of

Practice of Use of Animals for Scientific Purposes and relevant Western Australian State law.

A total of 100 male Sprague Dawley rats, originally weighing 225-275 g upon arrival from the Animal Resource Centre of Western Australia, were housed in pairs, in clear polypropylene boxes (47 x 37 x 20 cm high). The animal housing room was temperature- controlled and maintained at approximately 22 + 2°C under a 12 h light-dark cycle (lights on

162 at 6:00 a.m.). Rats had free access to food and water except during testing. Each rat was handled for 2 min daily for 5 days prior to behavioural testing. After reserpine administration, rats were given free access to soft feed and Necta gel packs to minimize health effects of akinesia.

4.3.2 Drugs

All drugs were administered as solutions, prepared immediately before use, and injected intraperitoneally. Reserpine (Sigma-Aldrich) and rimonabant were dissolved in 30-

40% w/v solution of Captisol® (Ligand Technology, La Jolla, CA, USA) solution with distilled water. Saline solutions of L-3,4-dihydroxyphenylalanine methyl ester hydrochloride

(Sigma Aldrich) with benserazide (Sapphire, a peripheral aromatic amino decarboxylase inhibitor to prevent the premature metabolism of L-DOPA to dopamine) and UWA-101

(synthesized by as described previously, Gandy et al., 2010) were used, while solutions of yohimbine (Sapphire) and riluzole (Sapphire) were prepared using distilled water and 5% dimethyl sulfoxide-saline solution, respectively. The pH of all solutions was approximately

6.5-7.0 using Litmus paper. Saline was used as a vehicle control treatment. All drug and saline solutions were administered at 1 mL/kg.

4.3.3 Apparatus

The five testing boxes used in this experiment were custom-made (McLelland et al.,

2015). Briefly, each acrylic box 32cm x 18cm x 25cm (l x w x h) had two rows of infrared

(IR) sensors and emitters glued on the outside of two opposing wall. Another row of IR emitters and sensors were attached along the outside of the other pair of opposing wall. The lower level emitters and sensors were attached to form a grid of IR beams throughout the

163 testing box and the upper level emitters and sensors were attached to form a parallel pattern of IR beams along the length of the testing box.

The emitters and sensors were wired to one custom-made circuit board per box. All five circuit boards were then connected to a 72 mHz Fez Panda II (obtained from Australian

Robotics, Canberra) connected to a standard desktop computer via USB port. Custom-made software written with Visual C# Express (freely available from https://github.com/cewinkler/Loco.Firmware) was used to collect and extract data. The boxes were placed in lidless polyethylene crates 45 x 25 x 36 cm (l x w x h) to avoid interference of the IR beams across the five boxes.

4.3.4 Procedure

Rats were injected with either reserpine (4 mg/kg) or saline (n = 10) between

2:00 pm and 6:00 pm. After 20 h, rats were injected with saline then habituated to the testing boxes for 30 min (the habituation session). Rat were then injected with either saline (n = 20),

UWA-101 (1, 3, or 10 mg/kg, n = 10 for each dose), rimonabant (3 mg/kg, n = 10), riluzole

(3 mg/kg, n = 10), or yohimbine (10 mg/kg n = 10) and observed for 30 min (the pre-test session). The 85-minute test session commenced immediately after injection of L-DOPA

(125 mg/kg) + benserazide (31.5 mg/kg) or saline (n = 10). All rats were randomly allocated to a specific drug group (n = 10-20). All rats were tested between 10:00 am and 5:00 pm. All testing occurred over one day, and rats were only tested once. Beam breaks detected by IR sensors on the bottom level were used as a measure of horizontal movements (number of beam breaks to indicate locomotion). Elevated movements were measured through upper level beam breaks and elevated movements in the centre of the test box (EMC) were counted when an upper level beam break occurred while peripheral beams on the lower level remained unbroken. Repetitive movements while elevated were counted when one lower

164 level beam was consecutively broken at least twice within a 3s time frame (1 sample/s sampling rate) while an upper level beam was broken. Non-repetitive movements while elevated were calculated after data collection using the following equation: (Total horizontal movements while elevated) – (Repetitive movements while elevated). Elevation time was measured as the length of time an upper beam was broken. Mean elevation bout duration, a subset measure of elevation time, was calculated after data collection using the equation:

Elevation Time/(Elevated Movements +1).

4.3.5 Statistical analysis

Data were analyzed using the ‘R studio’ statistical program version 0.98.1028. Results from rats treated only with saline or reserpine were analyzed using a between-subjects

Student’s t-test. Linear regression was performed to analyse effect of drug over time. Data from all reserpinised rats were compared using one-way, between-subjects ANOVA, using the ez package version 4.2-2 for R. Planned pairwise comparisons for each measurement between all reserpinised groups were then made and corrected using Bonferroni’s tests.

Corrections were made for multiple comparisons used with pooled variance unless Levene’s test for homogeneity showed significantly different variance. Alpha was set to p= 0.05.

165

4.4 Results

Initial analysis of the data revealed violations of assumptions of homogeneity of

variance. To avoid these violations, the data was transformed using a square root formula,

which resulted in homogeneity of variance within an acceptable range.

4.4.1 Effect of reserpine on healthy rats

Locomotor activity during the habituation session for rats treated with reserpine or

saline is presented in Table 4.1. Student’s t-tests revealed reserpine that horizontal

movements, elevation time, elevation bout, repetitive and non-repetitive movements while

elevated were all significantly lower in the reserpinised group compared to saline-treated rats.

Differences between groups in elevated movements or elevated movement in the centre did

not reach significance (p > 0.05).

Table 4.1. Mean (SD) and Student’s t test results for all behavioural measurements between saline- and reserpine (4 mg/kg)-treated Sprague-Dawley rats. Measurement Mean (SD) Significance Saline (n=10) Reserpine (n=20) Horizontal Movements 11.46 (1.53) 9.41 (2.29) p < 0.001

Elevation Time (s) 14.09 (2.85) 10.68 (2.66) p < 0.001

Elevation Bout 5.22 (4.6) 3.15 (2.59) p < 0.01

Repetitive Movements While 3.93 (0.58) 3.5 (0.70) P < 0.05 Elevated Non-Repetitive Movements 9.34 (1.87) 6.87 (2.21) p < 0.001 While Elevated

Elevated Movements 3.56 (1.82) 3.48 (1.29) NS

Elevated Movements in Centre 2.78 (1.59) 2.37 (1.13) NS

166

4.4.2 Effect of L-DOPA on reserpinised rats treated with vehicle or antidyskinetic drugs

4.4.2.1 Time contrasts

A linear regression was performed to analyze the effect of L-DOPA over time. L-

DOPA significantly increased horizontal movements over time with significant linear and cubic effects (p < 0.001, R2 = 0.33, F (20, 169) = 4.16, p < 0.001). A significant quadratic effect was found in elevated movements (p < 0.05, R2 = 0.14, F (20, 169) = 1.40, p = 0.12), but no significant effects were found in the regression for EMC (p > 0.05, R2 = 0.13, F (20.

169) = 1.29, p = 0.19). The linear regression for elevated time revealed significant linear, quadratic and cubic effects (p < 0.05, R2 = 0.21, F (20,169) = 2.29, p < 0.01) with increased elevated time across the testing period, while significant linear and quadratic effects were seen in elevated bout (p < 0.05, R2 = 0.19, F (20, 169) = 1.98, p < 0.05), again increasing over time. The linear regression for repetitive movements while elevated revealed significant linear and cubic effects (p < 0.05, R2 = 0.29, F (20, 169) = 4.83, p < 0.001) increasing over time, while non-repetitive movements while elevated showed significant linear and cubic effects (p < 0.05, R2 = 0.16, F (20. 169) = 1.6, p = 0.057), also increasing over time. These results demonstrate that L-DOPA induced a significant increase in horizontal movement, elevated time, elevated bout, and repeated and non-repeated movements over time between

25 and 85 minutes, and thus that hour period was used for drug effect comparisons (Figure

4.1).

167

168

4.4.2.2 Test

Behaviours following administration of L-DOPA in combination with saline, UWA-

101 (1, 3, or 10 mg/kg), riluzole, yohimbine, or rimonabant in reserpinised rats were analysed with one-way ANOVAs with antidyskinetic drug/dose as the factor, followed by pairwise comparisons.

ANOVAs revealed significant treatment effects for horizontal movements (F (7, 82) =

9.64, p < 0.001), elevation time (F (7, 82) = 6.83, p < 0.001), elevation bout (F (7, 82) = 4.32, p < 0.01), repetitive movements while elevated (F (7, 82) = 9.61, p < 0.001), non-repetitive movements while elevated (F (7, 82) = 11.35, p < 0.001), EMC (F (7,82) = 3.79, p < 0.05), while treatment effect for total elevated movements did not reach significance (F (7,82) =

2.01, p = 0.063).

Planned one-tailed pairwise comparisons showed L-DOPA significantly increased horizontal movement, elevation time, elevation bout, and repetitive and non-repetitive movements compared to saline (p < 0.001, Figure 4.2).

UWA-101 (1 and 3 mg/kg) significantly reversed L-DOPA-induced horizontal movement, elevation time, elevation bout, and repetitive and non-repetitive movements while elevated (p < 0.001), but had no effect on EMC as shown by post-ANOVA pairwise comparisons. In contrast, UWA-101 (10 mg/kg) had no significant effects on any activity measurements after L-DOPA administration (p > 0.1).

Riluzole significantly reduced L-DOPA-induced horizontal movement, elevation time, elevation bout, EMC, and repetitive and non-repetitive movements while elevated (p <

0.001).

One-tailed Bonferroni pairwise comparisons after ANOVAs revealed yohimbine significantly reduced L-DOPA-induced horizontal movement, elevation time, elevation bout,

169 and repetitive and non-repetitive movements while elevated (p < 0.001), but had no significant effect on EMC (p > 0.05).

Rimonabant significantly reduced L-DOPA-induced horizontal movement, elevation bout and repetitive movements while elevated (p < 0.001), but not elevation time, non- repetitive movements while elevated, or EMC (p > 0.1).

170

4.4.2.3 Effect of antidyskinetic drugs on L-DOPA-treated reserpinised rats

The percentage of change for horizontal movement, elevation time, elevation bout, and repetitive and non-repetitive movements while elevated for UWA-101 (1 and 3 mg/kg), riluzole, rimonabant, and yohimbine were compared to L-DOPA (Table 4.2 and Table 4.3).

The ratios of the changes in horizontal movements to elevation time, elevation bout, and repetitive or non-repetitive movements while elevated are also included. Yohimbine caused the greatest reduction of elevation-based measurements compared to horizontal movements reductions. In contrast, rimonabant appeared to have the mildest effect across all measurements and ratios. All drugs were most effective in reducing repetitive movements while elevated compared to reducing horizontal movement, with yohimbine and UWA-101 (3 mg/kg) inducing the greatest changes.

Drug HM (%) ET (%) EB (%) RME (%) NMRE (%) Riluzole 48.69 84.87 53.76 49.11 32.08 Rimonabant 85.16 150.94 92.78 83.51 57.54 UWA-101 (1 mg/kg) 79.94 123.96 76.29 76.07 53.49 UWA-101 (3 mg/kg) 72.19 132.33 72.09 73.19 37.89 Yohimbine 68.19 148.95 34.53 57.28 17.16

171

Drug HM/ET HM/EB HM/RME HM/NMRE Riluzole 1.76 1.17 0.52 0.41 Rimonabant 1.73 0.98 0.94 0.67 UWA-101 (1 mg/kg) 1.52 1.01 0.49 0.45 UWA-101 (3 mg/kg) 1.59 0.90 0.52 0.41 Yohimbine 0.81 0.27 0.43 0.36

4.4.3 Effect of antidyskinetic drugs on reserpinised rats

Omnibus one-way treatment ANOVAs were conducted between reserpinised rats administered saline, UWA-101 (1 or 3 mg/kg), yohimbine, riluzole or rimonabant, but not L-

DOPA. These ANOVAs revealed significant treatment effects for horizontal movements (F

(6, 73) = 15.79, p < 0.001), elevation time (F (6, 73) = 2.64, p < 0.05), repetitive movements while elevated (F (6,73) = 8.50, p < 0.01), and non-repetitive movements while elevated (F

(6,73) = 7.22, p<0.001). Planned two-tail comparisons were then conducted between the reserpinised rats treated with either antidyskinetic drugs or saline to determine specific antiparkinsonian-like drug effects. Rimonabant and yohimbine significantly increased horizontal movements and EMC in reserpinised rats (p < 0.05) compared to saline, while only rimonabant increased non-repetitive movements while elevated. UWA-101 (3 and 10 mg/kg) significantly increased horizontal movement, non-repetitive movements while elevated, total elevated movements and EMC (p < 0.05, Figure 4.3). Additionally, UWA-101

(10 mg/kg) increased elevation time and repetitive movements while elevated (p < 0.05), but not elevation bout (p > 0.05). In contrast, riluzole decreased repetitive movements while elevated (p < 0.05), but had no significant effect on any other behavioural measurement (p >

0.1).

172

173

4.5 Discussion

The validity of a reserpinised rat model of LID was tested by administering compounds previously shown to reduce L-DOPA-induced abnormal hyperkinetic behaviours in reserpinised rats (Johnston et al., 2012; Segovia et al., 2003), reduce dyskinetic-like behaviour in 6-OHDA-lesioned rats (Dekundy et al., 2007) or MPTP-treated primates

(Johnston et al., 2012: Segovia et al., 2003), and/or reduce L-DOPA-induced dyskinesias in humans (Bara-Jimenez et al., 2006; Merims, 1999). In the current study, L-DOPA significantly reversed reserpine-induced reduction of movement in agreement with previous studies (Goldstein et al., 1975; Johnston et al., 2005; Johnston et al., 2012; Segovia et al.,

2003). The current study also demonstrated that L-DOPA induces selective increases in horizontal movement and elevation time, but not elevated movement, regardless of location in the testing box compared to saline in reserpinised rats. Although these results are similar to previous studies, they are not directly comparable as other studies in the reserpinised rat model of LID reported L-DOPA-induced changes in vertical activity counts (Johnston et al.,

2005; Johnston et al., 2012; Segovia et al., 2003), whereas we observed changes in time spent elevated, but not the number of beam breaks during elevation. Nonetheless, these results support previous findings that high dose L-DOPA increases both horizontal and vertical activity in reserpinised rats, which may correlate with antiparkinsonian and dyskinetic behaviours, respectively.

4.5.1 Effect of antidyskinetic drugs on L-DOPA-induced movement

One aim of the current study was to replicate the anti-hyperkinetic effects of UWA-

101 in combination with L-DOPA as reported previously in reserpinised rats (Johnston et al.,

2012; Segovia et al., 2003). UWA-101 is known to have affinity for both the SERT and DAT

(Huot et al., 2014; Johnston et al., 2012) and was shown in functional assays to inhibit

174 reuptake of the monoamine neurotransmitters. The reduction of dyskinetic-like activity such as elevation time, elevation bout, and repetitive and non-repetitive movements while elevated by UWA-101 (1 and 3 mg/kg) in the current study is consistent with previous studies of compounds with affinity for the SERT (Bishop et al., 2006; Durif et al., 1995). In contrast, the negligible effect of UWA-101 (10 mg/kg) may be due to excessive monoaminergic release overwhelming the inhibitory effects of pre-synaptic autoreceptors and leading to excessive stimulation of the post-synaptic dopamine or serotonin receptors as seen with amphetamine (Sulzer, 1995) and MDMA (Rudnick, 1992).

Furthermore, this study aimed to replicate the anti-hyperkinetic effects of rimonabant

(3 mg/kg) combined with L-DOPA in reserpinised rats (Johnston et al., 2012; Segovia et al.,

2003). Indeed, rimonabant caused significant reductions of elevation bout and repetitive movements while elevated, without compromising the antiparkinsonian-like effect of L-

DOPA on horizontal movement. These results are in agreement with previous studies of rodent models in which rimonabant induced anti-hyperkinetic-like effects without compromising the antiparkinsonian effects of L-DOPA (Kelsey et al., 2009; Segovia et al.,

2003; Walsh et al., 2010).

In the present study, riluzole significantly reduced L-DOPA-induced elevation time and bout, and repetitive and non-repetitive movements while elevated, although these anti- hyperkinetic effects were accompanied by a significant reduction in L-DOPA-induced horizontal movements, indicating a reduction in efficacy of L-DOPA’s antiparkinsonian effects. However, species-dependent inconsistencies have been seen in the antidyskinetic efficacy of riluzole in rodent models of LID (Dekundy et al., 2007; Lundblad et al., 2005) and double-blind, placebo-controlled clinical trials in which no antidyskinetic effects were found

(Bara-Jimenez et al., 2006).

175

The results from the current study show yohimbine (10 mg/kg) reversed L-DOPA- induced elevation time, elevation bout, and repetitive and non-repetitive movements while elevated, consistent with an anti-hyperkinetic-like effect. In agreement with studies of 6-

OHDA-lesioned rats (Dekundy et al., 2007), yohimbine also reduced L-DOPA antiparkinsonian efficacy as measured by a reduction in horizontal movement. However, this may be a dose-dependent effect as yohimbine (0.1 mg/kg) significantly reduces LID in

MPTP-lesioned primates without affecting L-DOPA’s antiparkinsonian effects (Henry et al.,

1999). Indeed, other α2 adrenergic receptor antagonists such as fipamezole (90 mg dose) reduced LID without exacerbating parkinsonian symptoms in 115 patients during a phase 2B double-blind, placebo-controlled study (Lewitt et al., 2012).

4.5.2 Effect of antidyskinetic drugs on reserpine- induced movement

Another aim of the current study was to determine the effects of each antidyskinetic drug alone on reserpine-induced behaviour in rats, that is in the absence of L-DOPA.

Rimonabant increased horizontal movement, non-repetitive movements while elevated and

EMC, suggesting antiparkinsonian properties of rimonabant. Indeed, rimonabant (up to 0.1 mg/kg) as a monotherapy improves contralateral forepaw stepping testing in 6-OHDA- lesioned rats (Kelsey et al., 2009). Furthermore, in MPTP-lesioned primates, rimonabant (3 mg/kg) increased activity counts and range of movement, but had no effect on bradykinesia or posture (van der Stelt et al., 2005).

Double-blind, placebo-controlled clinical trials have reported improvements in

Unified Parkinson’s Disease Rating Scale (UPDRS) motor scores in PD patients after the noradrenergic ligand idazoxan was administered (Peyro-Saint-Paul et al., 1997; Ruzicka,

1997). Antiparkinsonian properties of yohimbine have not yet been studied in humans; however, in vivo animal studies have shown yohimbine increases intracellular DA levels in

176 the corpus striatum of male Sprague Dawley rats (Brannan et al., 1991) which could underlie the yohimbine-induced increase in horizontal movement seen in the current study. These results suggest yohimbine possesses antiparkinsonian effects in reserpinised rats.

Serotonin is known to have an influential role in the major motor symptoms of PD.

Previous evidence shows that UWA-101 has affinity for both the SERT and DAT, but not the noradrenaline transporter (NAT) (Huot et al., 2012b; Huot et al., 2014; Johnston et al., 2012).

In the current study, UWA-101 increased locomotive activity in reserpinised rats in a dose- dependent manner. Specifically, while UWA-101 (1 mg/kg) had no effect on reserpine- induced behaviour, UWA-101 (3 mg/kg) increased horizontal movement, non-repetitive movements while elevated, elevated movement, and EMC. These results are consistent with the evidence discussed above that UWA-101 (3 and 10 mg/kg) may induce antiparkinsonian effects in monoamine-depleted rats by reversing the transporters of serotonin and/or dopamine. Indeed, the effect of UWA-101 (10 mg/kg) on horizontal movement, non- repetitive and repetitive movements while elevated, elevated movement, EMC, and elevation time suggests this dose induces both antiparkinsonian and stereotypic/dyskinetic behaviours in reserpinised rats. Although this is the first study to assess the behavioural effects of UWA-

101 on rats in the absence of L-DOPA, studies in healthy Sprague Dawley rats show both horizontal movements and stereotypy are significantly increased after MDMA administration

(Kalivas et al., 1998). Thus, the increase in repetitive and non-repetitive movements while elevated may be due to increased stereotypic-like effects of UWA-101.

Although patients in the early stages of PD showed no improvement on monotherapy with the glutamate release inhibitor riluzole (Jankovic & Hunter, 2002), studies show riluzole administered before MPTP-lesioning in primates prevents the development of parkinsonian symptoms, suggesting riluzole has neuroprotective properties rather than antiparkinsonian properties (Benazzouz et al., 1995). This may explain the mild effect of riluzole on reserpine-

177 induced behaviour in the current study, excepting a significant decrease in repetitive- movements while elevated. This decrease in repetitive-movements while elevated may be explained by either greater dopamine denervation in reserpinised rats compared to dopamine levels in early stage PD patients or an interspecies difference.

4.5.3 Conclusion

In conclusion, the results of the present study support the validity of L-DOPA- induced hyperkinesia in the reserpinised rat as an acute model of LID. Drugs previously shown to exert antidyskinetic properties in human and animal studies significantly reduced L-

DOPA-induced elevated activity. However, L-DOPA-induced horizontal activity, a measure of antiparkinsonian effects, was also reduced by these compounds, although they remained above control levels. Further research is needed to assess the validity of this model as a predictive tool in screening potentially antidyskinetic compounds and determine additional parameters and limitations of the model, which may play a role in the interpretation of experimental findings. Specifically, other measures, such as biochemical analysis of neural tissue or neurotransmitters or the introduction of a dopamine agonist that does not induce dyskinesias, may be introduced to further explore the correlation between L-DOPA-induced hyperkinesia in reserpinised rats and LID in humans. Furthermore, additional analysis regarding time may be of use in this model as previous research has focused on therapies reducing peak-dose dyskinesia. However, the L-DOPA-treated reserpinised rat model of LID shows promise as a time-efficient and inexpensive screening test for potentially therapeutic novel compounds.

The current study also reinforced the findings of earlier studies (Huot et al., 2012b;

Johnston et al., 2012) in showing that the novel MDMA-analogue UWA-101 is particularly effective in reducing L-DOPA-induced hyperkinetic activity. As LID is a major complication

178 of current PD therapy and treatment options are limited, and considering the likely novel mode of action of UWA-101, further investigation of this and similar compounds is warranted.

179

180

Chapter 5 - Reduction of L-DOPA-induced hyperkinesia by MDMA analogue, UWA-101, in the reserpine-treated rat is attenuated by antagonism of the 5-HT1A receptor, blockade of the serotonin transporter, or inhibition of glutamate release 5 Preface Chapter 4 aimed to replicate the anti-hyperkinetic effect of UWA-101 previously found in reserpinised rats acutely challenged with high-dose L-DOPA. However, the pharmacological mechanism behind this effect is unknown. This chapter aims to identify mechanisms involved in the anti-hyperkinetic effect on UWA-101 in L-DOPA-treated reserpinised rats. By combining UWA-101 with drugs of a known pharmacological mechanism, we assessed the impact of these drugs on the anti-hyperkinetic effect of UWA-

101 in L-DOPA-treated reserpinised rats to indicate the underlying mechanism of this effect.

Publication: McLelland, AE, Winkler, CE, Piggott, MJ, Martin-Iverson, MT. Reduction of L- DOPA-induced hyperkinesia by MDMA analogue, UWA-101, in the reserpine-treated rat is attenuated by antagonism of the 5-HT1A receptor, blockade of the serotonin transporter, or inhibition of glutamate release. Publication in process.

5.1 Abstract

Chronic treatment with the dopamine-replacing agent, L-3,4-dihydroxyphenylalanine

(L-DOPA), in patients with Parkinson’s disease can result in debilitating motor side effects known as L-DOPA-induced dyskinesia (LID). The novel 3,4-methylenedioxy-N- methylamphetamine (MDMA) analogue UWA-101 reduces L-DOPA-induced hyperkinesia and dyskinesia in reserpine-treated rats and 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine

(MPTP)-treated primates, respectively. In addition, UWA-101 increases on-time of L-DOPA without compromising its antiparkinsonian effects. UWA-101 has affinity for dopamine and serotonin transporters, but not the noradrenaline transporter. However, no investigations have

181 yet been conducted to determine the mechanism underlying the antidyskinetic or anti- hyperkinetic actions of UWA-101 in vivo. The aim of this study, therefore, was to administer compounds with known pharmacological actions to attenuate the antihyperkinetic effects of

UWA-101. To test this, male Sprague-Dawley rats were administered reserpine (5 mg/kg, i.p.) 18 hours before treatment with riluzole (3 mg/kg, i.p., a glutamate release inhibitor), yohimbine (3 mg/kg, i.p., an α2 adrenoceptor antagonist), fluoxetine (10 mg/kg, i.p., a SERT inhibitor), WAY 100635 (3 mg/kg, i.p., a 5-HT1A receptor antagonist), or saline (1 ml/kg, i.p.) prior to pre-test habitation for 30 min. Rats were then administered L-DOPA (125 mg/kg, plus benserazide: 31.5 mg/kg, i.p.) in combination with UWA-101 (3 mg/kg, i.p.) immediately prior to the testing session. Horizontal movement and elevated behaviour were measured using enclosures lined with infrared emitters and sensors at two different heights.

Fluoxetine, WAY 100635, and riluzole all blocked UWA-101 reversal of L-DOPA-induced horizontal movement and elevated behaviour, while yohimbine + UWA-101 increased L-

DOPA-induced horizontal movement and further decreased selected elevated behaviours, compared to UWA-101 alone. These data suggest that UWA-101 reversal of L-DOPA- induced hyperkinesia, as indicated by the level of elevated behaviour, may occur through reversal of the direction of transport of the SERT, and the subsequent increased release of serotonin acts on 5-HT1A receptors. This process is dependent on glutamate release, consistent with the view that 5-HT1A receptors regulate glutamate release in the striatum.

5.2 Introduction

Although the dopamine replacing agent, L-3,4-dihydroxyphenylalanine (L-DOPA), is considered the most efficacious treatment available for the major motor symptoms of

Parkinson’s Disease (PD), chronic treatment with L-DOPA can result in potentially debilitating motor side effects known as dyskinesia (Cenci & Lindgren, 2007). Amantadine is

182 the only clinically effective medication currently used to treat L-DOPA-induced dyskinesia

(LID) in PD patients (Del Dotto et al., 2001; Metman et al., 1999; Metman et al., 1998; Snow et al., 2000) and has recently been approved for dyskinesia treatment as a slow-release preparation (Adamas Pharmaceuticals, 2017; Pahwa et al., 2015; Oertel et al., 2017; Elkurd et al., 2018). Nevertheless, studies have also reported inconsistent results in terms of duration and efficacy of amantadine’s antidyskinetic action (Thomas et al., 2004; Wolf et al., 2010).

Therefore, alternative treatments for LID are needed.

It has been established that 3,4-methylenedioxy-N-methamphetamine (MDMA, ecstasy) reduces LID in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated primates (Huot et al., 2011), 6-hydroxydopamine (6-OHDA)-lesioned rats (Bishop et al.,

2006), and possibly humans (BBC, 2001). Unfortunately, MDMA is not suitable as an ongoing treatment for LID because of its profound psychoactivity. Nonetheless, a range of recently discovered MDMA-analogues show potential as antidyskinetic agents, apparently without inducing psychoactivity associated with MDMA (Huot et al., 2014; Johnston et al.,

2012).

In particular, one analogue, UWA-101 (3,4-methylenedioxy-α-cyclopropyl-N- methylphenethylamine), reduces L-DOPA-induced elevated activity in reserpine-treated rats and L-DOPA-induced dyskinesia in MPTP-treated primates, without compromising the antiparkinsonian effects of L-DOPA (Huot et al., 2014; Johnston et al., 2012). These studies suggest that UWA-101 may be a candidate for LID treatment in a clinical context. UWA-101 has high affinities for the SERT and DAT, but not the noradrenaline transporter (NAT)

(Johnston et al., 2012), but the underlying mechanism for its antidyskinetic-like effects in vivo has not yet been determined. The aim of the current study, therefore, was to investigate potential mechanisms underlying UWA-101’s ability to reduce L-DOPA-induced hyperkinesia, as indicated by elevated behaviour, in reserpine-treated rats.

183

In addition to the dopaminergic neural system (Aubert et al., 2005; Darmopil et al.,

2009), the serotonergic (Cheshire & Williams, 2012), adrenergic (Henry et al., 1999; Lewitt et al., 2012), and glutamatergic (Bibbiani et al., 2005) systems have all been implicated in

LID development . We hypothesised that modulating these systems through administration of compounds with known pharmacological actions on these neurotransmitter systems would significantly inhibit UWA-101-induced effects in reserpine-treated rats administered L-

DOPA, if that system is important to the mechanism of action of UWA-101. We also hypothesised that if the administered drugs employed a mechanism of action independent to that of UWA-101, then the reduction of L-DOPA-induced hyperkinesia by UWA-101 would be augmented, or not influenced at all.

As UWA-101 has a 500 nM affinity (Ki) for the SERT, it is likely that the serotonin pathway is involved in the mechanism of UWA-101’s antidyskinetic effects (Johnston et al.,

2012). Additionally, 5-HT1A receptor antagonism in serotonergic neurons reverses antidyskinetic effects of serotonin transporter inhibitors (Bishop et al., 2006) and therefore may also influence UWA-101’s antidyskinetic effects. 5-HT1A receptors also modulate glutamate release (Czyrak et al., 2003; Mignon & Wolf, 2005). Drugs such as glutamate release inhibitors (Dekundy et al., 2007) and α2 adrenergic receptor antagonists (Henry et al.,

1999; Lewitt et al., 2012) decrease LID-like symptoms in animal models or humans and their administration in combination with UWA-101 is likely to alter the behavioural effects of

UWA-101 and provide insight into its underlying mechanism of action. We therefore tested the effects of riluzole (3 mg/kg, i.p., a glutamate release inhibitor), yohimbine (3 mg/kg, i.p., an α2 adrenoceptor antagonist), fluoxetine (10 mg/kg, i.p., a SERT inhibitor), WAY 100635

(3 mg/kg, i.p., a 5-HT1A receptor antagonist), or saline (1 ml/kg, i.p.) on UWA-101 induced behaviour in reserpinised rats treated with L-DOPA and the peripheral DOPA-decarboxylase inhibitor benserazide.

184

5.3 Methods

5.3.1 Animals

Experimental procedures were performed with approval from the University of

Western Australia Animal Ethics Committee (RA/3/100/1182) and in accordance with the

Australian Code of Practice of the Care and Use of Animals for Scientific Purposes and the

Animal Welfare Act (2002) of Western Australia. A total of 100 male Sprague Dawley rats, weighing 180-275 g upon arrival from the Animal Resource Centre of Western Australia, were housed in pairs, in clear polypropylene boxes (47 x 37 x 20 cm high). The animal housing room was temperature controlled and maintained at approximately 22 + 2°C under a

12 h light-dark cycle (lights on at 6:00 a.m.). Rats had free access to food and water except during testing. Each rat was handled for 2 min daily for 5 days prior to behavioural testing.

After reserpine administration, rats were given free access to soft feed and Necta gel packs to minimize health effects of akinesia.

5.3.2 Drugs

All drugs were administered as solutions, prepared immediately before use, and administered intraperitoneally. Reserpine (Sigma-Aldrich) was freshly dissolved in 30-40% w/v solution of Captisol® (Ligand Technology, La Jolla, CA, USA) in distilled water.

Solutions of L-3,4-dihydroxyphenylalanine methyl ester hydrochloride (Sigma Aldrich) with benserazide (a peripheral aromatic amino decarboxylase inhibitor to prevent the peripheral metabolism of L-DOPA to dopamine [Sapphire]), WAY 100635 (Sapphire), fluoxetine

(Sapphire), and UWA-101 (synthesised as described previously, Gandy et al., 2010) were prepared using saline. Yohimbine (Sapphire) solutions were prepared using distilled water and riluzole (Sapphire) was dissolved in 5% DMSO-saline solution. The pH of all solutions was approximately 6.5-7.0. Saline was used as a vehicle control treatment. All drug and

185 saline solutions were administered intraperitoneally at 1 ml/kg. Drug doses were selected based on previous research (Zhang et al., 2000; Dekundy et al., 2007; Harrison & Markou,

2001; Segovia et al., 2003; Ishibashi & Ohno et al., 2005).

5.3.3 Apparatus

The five testing boxes used in this experiment were custom-made and have been described elsewhere (McLelland et al., 2015). Briefly, the boxes were constructed from poly(methyl methacrylate), also known as Plexiglas®. The boxes (L: 32 cm x W: 18 cm x H:

25 cm) were equipped with upper and lower levels of infrared (IR) sensors and emitters attached at even intervals to the outside of the boxes. All five boxes were connected to a 72

MHz Fez Panda II microcontroller (obtained from Australian Robotics), which fed into a standard desktop computer via USB port. Custom-made software written with Visual C#

Express was used to collect and extract data.

5.3.4 Procedure

Rats were injected with either reserpine (5 mg/kg) or saline (1 ml/kg) between 4:00-

20:30. Rats were habituated to the locomotion enclosures for 5 min, 18 h post-injection. Rats were then injected with either saline, or a solution of riluzole (3 mg/kg), fluoxetine (10 mg/kg), WAY 100635 (3 mg/kg), or yohimbine (3 mg/kg) and observed for 30 min (the pre- test session), after which time the rats were injected with L-DOPA/benserazide (125/31.5 mg/kg), hereafter referred to as L-DOPA, and UWA-101 (3 mg/kg). The 2 h test session commenced immediately after those injections. All rats were tested between 9:00-17:00.

Beam breaks on the higher and lower levels of the testing boxes measured the animals’ movement. Beam breaks detected by IR sensors on the bottom level measured horizontal movement (number of beam breaks to indicate locomotion). Elevated movements were

186 measured through upper level beam breaks and elevated movements in the centre of the test box (EMC) were counted when an upper level beam break occurred while peripheral beams on the lower level remained unbroken. Repetitive movements while elevated were counted when one lower level beam was consecutively broken at least twice within a 3 s time frame

(1 sample/s sampling rate) while an upper level beam was broken. Non-repetitive movements while elevated were calculated after data collection using the following equation: (Total horizontal movements while elevated) – (Repetitive movements while elevated). Elevation time was measured as the length of time an upper beam was broken. Mean elevation bout duration, a subset measure of elevation time, was calculated after data collection using the equation: Elevation Time/(Elevated Movements +1).

5.3.5 Statistical analysis

Data was analysed using ‘R studio’ statistical program version 0.98.1028. Results from rats treated only with saline or reserpine were analysed using a between-subjects

Student’s t-test. Results from reserpinised rats treated with L-DOPA or saline were analysed using a between-subjects Student’s t-test, as were results from reserpinised rats treated with

L-DOPA + UWA-101 or saline. One-way, independent measures ANOVAs were used to analyse data between rats treated with UWA-101 and saline, riluzole, or yohimbine and two- way, independent ANOVAs were used to assess differences between rats treated with UWA-

101 and saline, fluoxetine, or WAY 100635. Pairwise planned comparisons for each measurement were then made with exact Bonferroni’s tests between all UWA-101-treated groups. Differences were accepted as being significant when p < 0.05, after the Bonferroni correction.

187

5.4 Results

Initial analysis of data revealed violation of assumptions of homogeneity of variance using Levene’s test. To avoid these violations, each data set was transformed using a square root formula, which resulted in homogeneity of variance within an acceptable range.

5.4.1 Effect of reserpine on healthy rats

Locomotion activity during the pre-treatment session for rats treated with reserpine or saline is shown in Table 5.1. Student’s t-tests showed that horizontal movement (t118 = 2.11, p

< 0.05), elevated time (t118 = 3.41, p < 0.001), elevated bout (t103.61 = 2.15, p < 0.05), centre elevated movement (t118 = 2.76, p < 0.01), and non-repeated movement while elevated (t118 =

2.49, p < 0.01) were all significantly lower in the reserpine-treated group (p < 0.05).

Differences in elevated movement, or repeated movement while elevated, did not reach significance (p > 0.1).

Table 5.1. Mean (+ SD) of all measurements during habituation session between saline- and reserpine-treated Sprague Dawley rats. Locomotion Measurement Mean (+ SD) Significance Saline Reserpine Horizontal Movement 9.60 (+ 1.66) 7.96 (+ 0.82) p < 0.05 Elevated Time 11.29 (+ 3.42) 8.02 (+ 1.90) p < 0.05 Elevated Bout 3.41 (+ 3.07) 2.18 (+ 1.37) p < 0.05 Elevated Movement 3.86 (+ 1.44) 3.39 (+ 0.84) NS Centre Elevated Movement 2.81 (+ 1.15) 1.92 (+ 0.45) p < 0.05 Repeated Movement While 1.86 (+ 0.71) 1.66 (+ 0.58) NS Elevated Non-Repeated Movement While 6.30 (+ 1.37) 4.62 (+ 1.21) p < 0.05 Elevated

188

5.4.2 Effect of L-DOPA on reserpine-treated rats

L-DOPA significantly increased horizontal movement (t438 = 8.54, p < 0.001), elevated time (t438 = 12.76, p < 0.001), elevated bout (t358.53 = 9.67, p < 0.001), elevated movement (t438 = 3.20, p < 0.001), repeated movement while elevated (t264.23 = 10.29, p <

0.001), and non-repeated movement while elevated (t413.79 = 7.02, p < 0.001), but did not induce significant changes in centre elevated movement (p > 0.1).

5.4.3 Effect of UWA-101 on reserpine-treated rats administered L-DOPA

Student’s t-tests revealed that administration of UWA-101 significantly decreased L-

DOPA-induced elevated time (t438 = 3.77, p < 0.001), elevated bout (t438 = 3.23, p < 0.01), and repeated movement while elevated (t417.36 = 2.52, p < 0.05), as shown in Figure 5.1.

However, UWA-101 did not significantly alter L-DOPA-induced horizontal movement, elevated movement, non-repeated movement while elevated (p > 0.1), or centre elevated movement (t416.07 = 1.89, p = 0.059).

5.4.4 Effect of experimental drugs on UWA-101-induced effects in rats

5.4.4.1 Yohimbine and riluzole

Yohimbine (3 mg/kg) or riluzole (3 mg/kg) were administered to rats treated with UWA-

101 + L-DOPA combination to determine their effects on horizontal and elevated behaviour.

Groups administered yohimbine, riluzole, or saline were compared using independent measures one-way ANOVAs. Significant between-groups treatment effects were found in horizontal movement (F (2, 27) = 3.63, p < 0.05), elevated movement (F (2, 27) = 3.38, p <

0.05), and centre elevated movement (F (2, 27) = 4.18, p < 0.05). In contrast, no significant

189 treatment effects were found for elevated time, elevated bout, repeated movement while elevated, or non-repeated movement while elevated (all p > 0.05).

Pairwise Bonferroni-corrected comparisons revealed that yohimbine significantly increased horizontal movement, elevated movement, centre elevated movement, and non- repeated movement while elevated, while decreasing elevated bout compared to saline in reserpine-treated rats administered UWA-101 + L-DOPA (all p < 0.05) but had no effect on elevated time or repeated movement while elevated (p > 0.05). Planned comparisons also revealed that riluzole significantly increased elevated movement, elevated time, centre elevated movement, and non-repeated movement while elevated (all p < 0.05).

5.4.4.2 WAY 100635

The specific 5-HT1A receptor antagonist WAY 100635 was also tested in reserpine- treated rats administered UWA-101 + L-DOPA to investigate the contribution of 5-HT1A receptors to the anti-hyperkinetic effects of UWA-101 (Figure 5.1). No significant treatment effects were found for horizontal movement, elevated movement, or center elevated movement in reserpinised rats administered L-DOPA in combination with UWA-101, and

WAY 100635, or saline (p > 0.05).

Two-way ANOVAs revealed that, while no treatment effects were found for WAY

100635 or UWA-101, significant treatment interactions were observed for elevated time (F1,36

= 10.78, p < 0.01), elevated bout (F1,36 = 6.38, p < 0.05), repeated movement while elevated

(F1,36 = 5.22, p < 0.05), and non-repeated movement while elevated (F1,36 = 6.06, p < 0.05).

Pairwise comparisons also revealed that UWA-101 significantly decreased elevated time, elevated bout, and repeated and non-repeated movement while elevated, compared to the saline + L-DOPA group (p < 0.01). Furthermore, these UWA-101-induced decreases in elevated time, elevated bout, repeated and non-repeated movement while elevated were

190 reversed by co-administration of WAY 100635 (p < 0.01, Figure 5.1). Pairwise comparisons also revealed that L-DOPA-induced elevated time, elevated bout, repeated and non-repeated movement while elevated were significantly reduced by WAY 100635 + saline (p < 0.05).

Figure 5.1. Effect of various experimental drugs on the reduction of L-DOPA-induced behaviour by UWA-101 (3 mg/kg) in reserpine-treated Sprague Dawley rats. *p < 0.05 vs. Veh/Veh group, ^p < 0.05 vs. L-DOPA/Veh group, +p < 0.05 vs. L-DOPA/UWA/Veh group

191

5.4.4.3 Fluoxetine

As UWA-101 is known to have affinity for the SERT, the SERT inhibitor fluoxetine was administered to reserpine-treated rats in combination with L-DOPA and UWA-101 to investigate the role of the SERT in the antidyskinetic effects of UWA-101. Independent fluoxetine x UWA-101 ANOVAs revealed that, while there were no significant treatment effects across measurements, significant treatment interactions occurred for elevated time

(F1,36 = 5.90, p < 0.05), elevated bout (F1,36 = 6.19, p < 0.05), and repeated movement while elevated (F1,36 = 6.54, p < 0.05) were observed. However, no significant effects or interactions were observed for horizontal movement, elevated movement, center elevated movement, or non-repeated movement while elevated. Planned pairwise comparisons showed that L-DOPA-induced elevated time, elevated bout, and repeated movement while elevated was significantly reduced by UWA-101 (Figure 5.1). Furthermore, administration of fluoxetine reversed this effect by UWA-101 for elevated time (p < 0.05), elevated bout (p <

0.01), and repeated movement while elevated (p < 0.05). Fluoxetine + saline also significantly attenuated L-DOPA-induced elevated time (p < 0.05), elevated bout (p < 0.05), and repeated movement while elevated (p < 0.01).

5.5 Discussion

Previous reports have described the reduction of L-DOPA-induced elevated activity and dyskinesia by the MDMA-analogue UWA-101 in reserpine-treated rats and MPTP-treated primates, respectively (Huot et al., 2012b; Huot et al., 2014; Johnston et al., 2012). UWA-101 has affinities for the DAT and SERT, but not NAT (Johnston et al., 2012), and its enantiomers differ in their affinities for these two transporters, with UWA-121 (the R- enantiomer) having a high affinity for the DAT and moderate affinity for the SERT, while

UWA-122 (the S-enantiomer) exhibits no appreciable affinity for the DAT and high affinity

192 for the SERT (Huot et al., 2014). Nonetheless, no research has yet investigated the mechanism by which UWA-101 exerts antidyskinetic effects in vivo. Therefore, this study investigated the mechanism underlying the effects of UWA-101 in rats that have been proposed to reflect antidyskinetic-like effect in vivo.

The main findings of this study are two-fold. Firstly, UWA-101 attenuated L-DOPA- induced elevated behaviour in reserpine-treated rats. As UWA-101 has affinity for the SERT, as determined by transporter binding studies, these anti-hyperkinetic-like effects are likely to be mediated through serotonergic actions of this compound and are consistent with rodent and primate studies (Huot et al., 2012b; Huot et al., 2014; Johnston et al., 2012). Secondly, we demonstrate that the reduction of L-DOPA-induced elevated activity by UWA-101 was reversed by the specific 5-HT1A receptor antagonist WAY 100635, the SERT inhibitor fluoxetine, or the glutamate inhibitor riluzole. The involvement of 5-HT1A receptors, the

SERT, and glutamate release in the effects of UWA-101 supports prior clinical and experimental work demonstrating that direct or indirect agonist action at the 5-HT1A receptor

(Bara-Jimenez et al., 2005; Bibbiani et al., 2001; Durif et al., 1995), and glutamate receptor antagonism (Bibbiani et al., 2005; Blanchet et al., 1998; Del Dotto et al., 2001; Metman et al.,

1999), results in antidyskinetic effects.

The reserpine-treated rat model of PD is a useful tool for studying the movement deficits that occur following dopamine depletion and identifying antiparkinsonian compounds through the reversal of these deficits (Finberg & Youdim, 2002; Goldstein et al., 1975; Skuza et al., 1994). Furthermore, L-DOPA-induced hyperkinesia in the reserpine-treated rat is thought to correlate with LID in humans and primates (Johnston et al., 2005; Segovia et al.,

2003). L-DOPA-induced vertical activity in reserpine-treated rats is also selectively reduced by compounds shown to have antidyskinetic effects in MPTP-treated primates and PD patients (Johnston et al., 2005; Segovia et al., 2003). As it is likely that this hyperkinesia and

193 vertical activity involves repetitive, purposeless (stereotypic) behavior, we aimed to include a larger range of stereotypic behaviours in measurements of L-DOPA-induced ‘elevated activity’, which were selectively reduced by UWA-101.

The 5-HT1A antagonist, WAY 100635, selectively attenuated the reduction in L-DOPA- induced elevated activity by UWA-101. This blockade of UWA-101 action is in agreement with previous studies arguing that the antidyskinetic actions of serotonergic compounds are primarily influenced by 5-HT1A receptor stimulation (Bishop et al., 2006; Eskow et al., 2007;

Ostock et al., 2011). Indeed, there are two clinical trials, “Buspirone treatment of iatrogenic dyskinesias in advanced Parkinson’s disease (BUSPARK)” and “Buspirone, in combination with amantadine, for the treatment of levodopa-induced dyskinesia”, in the USA of the 5-

HT1A receptor agonist, buspirone, as a treatment for iatrogenic dyskinesia in Parkinson’s disease (NIH, 2017). Although we did not determine whether 5-HT1A receptor inhibition by

WAY 100635 was pre- or post-synaptic, there is evidence that stimulation of presynaptic 5-

HT1A autoreceptors inhibits neuronal firing and reduces serotonin release (Rudnick & Wall,

1992). Therefore, it is possible that UWA-101 inhibits presynaptic SERT thereby indirectly stimulating presynaptic 5-HT1A receptors through increased synaptic serotonin levels leading to reduced firing of the serotonergic neurons and consequently reduced L-DOPA-induced elevated activity. If so, antagonism of 5-HT1A receptors by WAY 100635 would likely be presynaptic, as this would likely increase serotonergic neural firing and consequently increase

L-DOPA-induced elevated activity as was found in the study. These findings are consistent with the view that UWA-101 indirectly influences dopaminergic mechanisms via effects on the SERT, as well as direct effects via the DAT transporter. Depletions of central 5-HT have been shown to potentiate the effects of amphetamine on locomotor activity (Martin-Iverson et al., 1983), so presumably increasing 5-HT release will dampen dopaminergic effects.

194

As fluoxetine is a selective serotonin transporter inhibitor (SSRI), its ability to reverse the effects of UWA-101 on L-DOPA-treated reserpinised rats in the current study, along with the previously reported binding affinity, supports the hypothesis that the antidyskinetic mechanism of UWA-101 involves the SERT. Reserpine blocks the increased synaptic levels on monoamines produced by monoamine transporter inhibitors (Fairbrother et al., 1990), but not by monoamine transporter reversers such as tyramine and the amphetamines, including

MDMA. The presence of significant effects of UWA-101 in reserpinised rats suggests that

UWA-101 is acting primarily as a transporter reverser rather than a classic inhibitor like methylephenidate, nomifensine, cocaine or fluoxetine (Fleckenstein et al., 2000). Of course, as the amphetamines are also substrates for the transporters; UWA-101 may also act as a competitive inhibitor. Nevertheless, the primary action is by reversal of the transporters, explaining why the effects of amphetamines are attenuated by classic transporter inhibitors, instead of being additive.

In the current study, fluoxetine (10 mg/kg) also inhibited L-DOPA-induced increases in selective elevated activity measurements in reserpine-treated rats. Fluoxetine (10 mg/kg) has also been shown to reduce total locomotive distance and vertical activity when administered alone to Long Evans rats (Stanford et al., 2002), while forebrain serotonin depletions increase amphetamine-induced locomotion (Martin-Iverson et al., 1982). Fluoxetine has also been found to have significant antidyskinetic effects in people (Durif et al., 1995) and rats (Bishop et al., 2012).

In addition to serotonergic activity, excessive glutamate transmission has been strongly implicated in the pathophysiology of both PD motor symptoms and LID. Indeed, enhanced glutamate release in the striatum of rats has been associated with development of dyskinetic- like behaviour (Robelet et al., 2004), and increased expression (Calon et al., 2003; Hallett et al., 2005) and phosphorylation (Chase & Oh, 2000) of ionotropic glutamate receptors. The

195 main action of riluzole consists of blocking voltage-dependent sodium channels with preferential influence on glutamatergic terminals, and blockade of glutamine transport, necessary for glutamate synthesis (Erickson, 2017), although it can also exert non- competitive antagonism at NMDA receptors (Doble, 1996). Riluzole has been shown to alleviate LID without worsening parkinsonism in 6-OHDA-lesioned rats (Dekundy et al.,

2007; Lundblad et al., 2002), and an open-label study in advanced PD patients (Merims et al.,

1999). However, this effect has not been demonstrated in double-blinded, placebo-controlled studies (Bara-Jimenez et al., 2006). One explanation given by Bara-Jiminez et al. (2006) was that the glutamatergic effects of riluzole were too widespread to produce therapeutic effects without disrupting other critical glutamatergic functions. Interestingly, combined administration of riluzole and UWA-101 in the current study attenuated the effects of UWA-

101 on L-DOPA-induced elevated time and movement while elevated, with no effect of horizontal activity. As with blockade of UWA-101 effects by WAY 100635 and fluoxetine, the reversal of UWA-101’s effects on elevated activity by riluzole suggests a mechanism or neural pathway common to the action of both drugs. Based on one model of the basal ganglia in PD and LID development (Nambu, 2004; Robertson, 1992), the disinhibition of the indirect pathway of the basal ganglia may be one of the overlapping mechanisms of action that riluzole and UWA-101 may act through to exert their antidyskinetic effects, which may then be reversed when the two drugs are combined in vivo.

An unexpected finding of the current study was the increase in L-DOPA-induced horizontal activity and elevated movement when the α2 adrenergic antagonist yohimbine was combined with UWA-101 administration in reserpine-treated rats. As reported in other studies, the drug-induced decrease in akinesia and increase in horizontal movement in the reserpine-treated rat is a representative measure of antiparkinsonian effects (Carlsson et al.,

1957; Goldstein et al., 1975; Johnston et al., 2005; Johnston et al., 2012; Segovia et al.,

196

2003). Other animal models and clinical trials have also shown that α2 adrenergic receptor antagonists, both alone and as an adjunct to L-DOPA therapy, have antidyskinetic properties

(Colpaert et al., 1991; Henry et al., 1999). The combined treatment of UWA-101 and yohimbine increased horizontal and elevated movement and reduced elevated bout compared to UWA-101 alone, suggesting that these drugs act through independent mechanisms. As seen in the combinations of UWA-101 with fluoxetine or WAY 100635 in the current study, the antagonism of either SERT and 5-HT1A receptors attenuates UWA-101-induced behavioural effects, indicating that UWA-101 acts via these mechanisms. Unfortunately, it is unclear in this study whether the increase in activity by yohimbine combined with UWA-101 is an additive effect due to both drugs acting on the adrenergic system together or due to each drug acting on different neurotransmitter systems, serotonergic and adrenergic for example, individually.

One important limitation of this study was that not all of the controls were replicated.

Indeed, only fluoxetine and WAY100635 were the expimental drugs paired with both UWA-

101 and vehicle controls while yohimbine and riluzole were tested only in rats also administered L-DOPA and UWA-101. This was done for two reasons: 1) to minimise the use of animals and statistical complexity and 2) to emphasise the focus of investigating serotonergic mechanisms underlying the anti-dyskinetic effect of UWA-101 while also assessing the involvement of other neurotransmitter systems. To improve on this limitation of the current study, future research may include an increased number of controls to increase the robustness of the data; however, this would likely result in having to narrow the focus of the study in terms of neurotransmitter systems and their relevant receptors to avoid exponentially increasing the number of animals required and the complexity of the statistical analysis.

An additional point to note is that, unlike the results in Chapter 4, the current study did not demonstate a significant difference between L-DOPA and UWA-101 in horizontal

197 movement. However, significant differences in elevated time, elevated bout, and repeated movement while elevated were replicated compared to the results in Chapter 4. It is uncertain what may be driving this difference in effect on horizontal movement, but one possibility is the difference in drug administration. In Chapter 4, UWA-101 was administered 30 minutes prior to L-DOPA, while in the current study UWA-101 and L-DOPA were administered at the same time and may represent a pharmacokinetic effect of the two compounds. While this difference in drug administration is important to acknowledge, it does not account for why only horizontal movement may have been impacted compared to other behavioural and locomotive measurements, the differences of which were replicated between the two studies.

Overall, the current study demonstrates that SERT inhibition, 5-HT1A receptor antagonism, and glutamate-release inhibition all inhibit anti-hyperkinetic-like effects induced by UWA-101 in reserpine-treated rats administered L-DOPA. As UWA-101 has affinities for both the SERT and DAT, it is likely that administration of a SSRI such as fluoxetine prevents

UWA-101 from reversing the SERT, similar to the action of MDMA. It has been suggested in previous studies that exogenous L-DOPA-derived dopamine is released from both dopaminergic and serotonergic neurons and that stimulation of pre-synaptic autoreceptors reduces this output. Indeed, the results of the current study do indicate that antagonism of 5-

HT1A receptors, which have been shown to be located pre-synaptically in the serotonergic system (Bishop et al., 2006; Cheshire & Williams, 2012; Rudnick & Wall, 1992), blocks

UWA-101-induced anti-hyperkinetic-like effects. As UWA-101 was found not to have appreciable affinity for 5-HT1A receptors (Johnston et al., 2012), it is likely that a reversal in serotonin transport by UWA-101 may increase stimulation of (most likely pre-synaptic) 5-

HT1A receptors, by synaptic serotonin, leading to inhibited neuronal firing and reduced exogenous L-DOPA-derived dopamine release. This mechanism has been proposed to explain the antidyskinetic effect of serotonin ligands in animals and humans (Bishop et al.,

198

2006; Cheshire & Williams, 2012; Eskow et al., 2007; Ostock et al., 2011). Lastly, the attenuation of UWA-101’s effects via glutamatergic activity inhibition indicates that the antidyskinetic effects of the MDMA-analogue may be dependent on glutamate transmission in neural pathways in addition to more local synaptic mechanisms, as riluzole and UWA-101 have no known mutual receptor affinities, although more research would be needed to clarify this.

The results of the current study further demonstrate the complexity of the mechanism underlying LID development. Not only are multiple receptors involved in the modulation of motor symptoms in both PD and LID, but these receptors affect behaviour across a range of overlapping and interacting neurotransmitter systems. Although the findings of the current study suggest the involvement of several receptor types in the actions of UWA-101 on L-

DOPA-induced abnormal movements in reserpine-treated rats, it also provides evidence to suggest that targeting multiple receptors as a form of LID therapy may have improved efficacy compared to current, largely single-target, treatments. As the combination of yohimbine and UWA-101 created an additive effect of reducing L-DOPA-induced abnormal movements, it is possible that other combination therapies might be more suited to LID therapy rather than targeting one receptor subtype. This may introduce a new avenue for future research in terms of combination therapies targeting known neural substrates of LID and utilising candidate compounds such as UWA-101. However, careful consideration would be needed in such studies to avoid adverse drug interactions.

199

200

Chapter 6 - Rewarding and anxiolytic effects of the MDMA analogue UWA-101 in healthy Sprague Dawley rats

6 Preface After investigating the anti-hyperkinetic effect of UWA-101 in the L-DOPA-treated reserpinised rat in Chapters 4 and 5, Chapter 6 aims to assess the rewarding and anxiolytic effects of UWA-101 in healthy rats. Many medications are misused or abused in patient populations due to their rewarding or anxiolytic properties. Therefore, it is important, as part of a complete pharmacological profile of a new drug, to investigate potential reinforcing effects of a new drug that may lead to misuse or abuse in patients. In this chapter, we describe the effect of UWA-101 on rats in Conditioned Place Preference, Elevated Plus Maze tests, and assess its effect on ultrasonic vocalisations using the capsule microphones described in

Chapter 3.

Publication: McLelland, AE, Winkler, CE, Piggott, MJ, Martin-Iverson, MT. Rewarding and anxiolytic effects of the MDMA analogue UWA-101 in healthy Sprague Dawley rats. Publication in process.

6.1 Abstract

There is growing concern over the increasing trend in non-medical use of prescription drugs. This trend highlights the importance of determining whether novel drugs with potential for clinical use exert unintended reinforcing effects to assess the risk of misuse and abuse of these drugs. UWA-101 is a novel MDMA-analogue that has shown potential as a therapeutic agent for L-DOPA-induced dyskinesias in animal models of PD, but has not yet been investigated as to whether it exerts reinforcing effects. Using healthy Sprague Dawley rats, we assessed whether UWA-101 (3 mg/kg) induces conditioned place preference (CPP), alters ultrasonic vocalisations, or exerts anxiolytic effects in the elevated plus maze test compared to saline and dexamphetamine (1.5 mg/kg). Rats treated with UWA-101 showed no

201 place preference or aversion in the CPP test, and showed no change in the number of USVs or anxiolytic effects in the EPM test. However, UWA-101 did decrease the number of rears and increase immobility time compared to saline in the USV test. In contrast, dexamphetamine induced a place preference, increased the number of 50 kHz USVs and increased percentage open arm time and entries in the EPM test. However, rearing frequency and immobility time in dexamphetamine-treated rats did not differ from that of saline controls. These findings suggest that UWA-101 does not exert rewarding or anxiolytic properties in healthy rats, but attenuates mobility in specific contexts. These findings also suggest that UWA-101 may be used clinically with a lower risk of non-medical use.

6.2 Introduction

Recent decades have seen increased misuse and abuse of prescription drugs by patient and general populations, with rates remaining near peak levels (Johnston et al., 2011). Non- medical use of prescription drugs is second only to recreational use of cannabis in persons aged 12 years and older in the United States (SAMHSA, 2013) and Australia (AIHW, 2011).

Some of the most commonly misused prescription drugs include stimulants such as Adderall

(Teter et al., 2006), benzodiazepines such as diazepam (Caplan et al., 2007) or alprazolam

(O'Brien, 2005), and opiate analgesics such as morphine (Joranson et al., 2000; Gilson et al.,

2004). Previous studies suggest that the legitimate use of these drugs encourages the illusion that the health risk of consuming prescription drugs is lower than consuming illicit drugs

(Boyed et al., 2006; Compton & Volkow, 2006; Johnston et al., 2011), and that prescription drugs such as stimulants, sedatives, and analgesics can exert powerful reinforcing effects leading to regular misuse or abuse (Pulvirenti et al., 1994; Caplan et al., 2007). The risk of excessive use and even addiction are increased in recreational users of prescription drugs because of this inaccurate representation.

202

Drugs that exert reinforcing effects in humans also tend to induce rewarding or anxiolytic effects in animals. Examples of these drugs include methylphenidate, a psychostimulant commonly prescribed for attention deficit/hyperactivity disorder (ADHD)

(Martin-Iverson et al., 1985; Kollins et al., 2001), diazepam (Nomikos et al., 1988; DeWit et al., 1989), and morphine (Spyraki et al., 1985; Lamb et al., 1991; Preston et al., 1991).

Animal models of stimulus-induced reinforcing effects include conditioned place preference

(CPP) and ultrasonic vocalisations (USVs). Models such as the elevated plus maze (EPM) are pertinent indicators of anxiety, particularly in rodents. The presence of CPP is based on the preference of a previously neutral environment by an animal due to the associations of a rewarding stimuli that environment has been paired with (Nazarian et al., 2004; Tzschentke,

2007; Liao, 2008). USVs in rodents are thought to indicate affective states. Positive affective states are indicated through the emission of 50 kHz USVs as they are generally emitted in a rewarding context compared to a neutral or aversive context (Burgdorf et al., 2001; Knutson et al., 2002; Litvin et al., 2007; Ma et al., 2010; Browning et al., 2011). The EPM test is founded on the conflict between exploration and thigmotaxis in rats, with greater exploration expressed as increased open arm activity indicating lower levels of anxiety-like behaviour

(Pellow et al., 1985; Privette et al., 1995; Schulteis et al., 1998). These tests can be used to assess the reinforcing effects of prescription drugs and indicate the potential for abuse or misuse of these drugs.

Amphetamine has been repeatedly investigated in these rodent behaviour tests as it reliably induces place preference, indicating a positive association to a previously neutral environment. Amphetamine also increases rates of 50 kHz USVs in rats, indicating increase in positive affect. Both of these behaviours suggest the presence of rewarding effects

(Gerdjikov et al., 2004; Ahrens et al., 2009). However, amphetamine generally induces

203 anxiogenic-like effects in the EPM (He et al., 2005; McLelland et al., 2014), demonstrating the distinction between rewarding and anxiolytic effects.

These types of animal models are commonly employed to assess the reinforcing effects of drugs with therapeutic effects in clinical contexts (Milekic et al., 2006; Quisenberry et al., 2013; McLelland et al., 2014). This assessment allows more detailed profiles of newly discovered drugs with therapeutic potential to be constructed regarding both their efficacy in treating disease and in inducing reinforcing or anxiolytic effects.

One newly discovered therapeutic candidate is the methylene-3,4- dioxymethamphetamine (MDMA) analogue UWA-101. Previously tested in non-human primate and rat models of Parkinson’s disease, UWA-101 is thought to have potential as a treatment for L-DOPA-induced dyskinesia (LID) (Huot et al., 2012b; Johnston et al., 2012;

Huot et al., 2014). LID is defined as abnormal involuntary movements resulting from chronic exposure to L-DOPA, one of the most effect treatments for the major motor symptoms of PD

(Ahlskog & Muenter, 2001).

The current study aimed to determine any reinforcing or anxiety-related effects induced by UWA-101, compared to saline, at a dose shown to significantly decrease LID in animal models of PD. Reinforcing and anxiety-related effects of amphetamine were used as a positive control for comparison. The current study investigated the potential for UWA-101 to induce rewarding or anxiolytic effects in healthy rats using CPP, USV and EPM tests.

6.3 Materials

6.3.1 Animals

Experimental procedures were performed with approval from the University of

Western Australia Animal Ethics Committee and in accordance with the Australian Code of

Practice of Use of Animals for Scientific Purposes and relevant Western Australia State law.

204

A total of 40 male Sprague Dawley rats, originally weighing 180-220 grams upon arrival from the Animal Resource Centre of Western Australia, were housed in pairs, in clear polypropylene boxes (47 x 37 x 20 cm high). The animal housing room was temperature controlled and maintained at approximately 22 + 2 oC under a 12 hour light-dark cycle (lights on at 6:00 a.m.). Rats had free access to food and water except during testing. Each rat was handled for 2 min daily for 5 days prior to behavioural testing.

6.3.2 Drugs

(+)-Amphetamine sulphate (dexamphetamine) was purchased from Sigma-Aldrich,

Australia. UWA-101 was kindly supplied by the University of Western Australia. Drugs were dissolved in saline solution before administration. Saline was used for saline injections. All injection were intraperitoneal.

6.3.3 Conditioned place preference (CPP)

Apparatus: CPP conditioning was conducted using four plastic boxes with removable plastic covers. Each box had two chambers (38 x 27 x 34 cm) that could be blocked by plastic guillotine-style doors. The chambers had two distinct floor textures; each chamber consisted of either plastic (1 cm squares) or galvanized steel (elongated diamond pattern).

Conditioning boxes were built with a tilting mechanism so that movement of the animals into each chamber would tilt the box accordingly. Infrared sensor and emitter pairs were arranged next to the conditioning boxes so that the tilting of the box in one direction would block the infrared beam to the sensor, while tilting of the conditioning box in the opposite direction would unblock the beam. Therefore, time spent in each chamber during pre-conditioning and test days was measured through the summation of time the infrared beam was either

205 interrupted by the tilt of the conditioning box or detected by the infrared sensor. Time spent in each chamber was recorded manually for later analysis.

Procedure: Conditioning and testing occurred between 9:00 a.m. and 5:00 p.m. Rats were tested in groups of two or four, with one rat per conditioning box. The 13-day experimental period comprised 3 pre-conditioning sessions, 8 conditioning sessions and 2 tests. The first test (day 12) was drug-free, while a second, state-dependent test occurred on day 13. Rats were randomly assigned to receive saline, dexamphetamine (1.5 mg/kg), or

UWA-101 (3 mg/kg). All injections were administered intraperitoneally at 1 ml/kg.

Pre-Conditioning Phase: In each session, rats were placed in counter-balanced starting chambers. The guillotine doors were removed prior to pre-conditioning sessions to allow the animals free movement between the two chambers for 15 minutes. Infrared device pairs recorded the length of time spent in each chamber.

Conditioning Phase: Each rat was conditioned during one 30-min session daily for 8 days. Rats were injected with the test drug on days 1, 3, 5 and 7 and saline on days 2, 4, 6 and 8. Half the rats were confined to the left chamber on drug days and the right side on saline days and vice versa for the other half. Conditioning was counter-balanced across all chambers. Drugs or saline were injected immediately before each session.

Testing Phase: Two tests were conducted. The day after the last conditioning session the 15 minute drug-free test was identical to pre-conditioning sessions. For the second (state- dependent) 15 minute test conducted on the day immediately following the drug-free test, the same drugs were administered that had been given on drug conditioning days.

6.3.4 Ultrasonic vocalisations (USVs)

Apparatus: A custom-made microphone and a commercial bat detector connected to

Raspberry Pi mini-computers and controlled by a personal computer (PC) through a TP-

206

LINK router were used to record USVs by rats. The custom-made microphone consisted of a

Panasonic WM-61A electret condenser MIC capsules soldered to standard commercial wiring.

Procedure: Before EPM testing began, rats were isolated immediately after injection in a clear polypropylene box containing soiled bedding from each rat’s housing box. Rats were habituated to these boxes for 20 min after which USVs were recorded for 10 minutes in the boxes before EPM testing. Video recordings of behaviour were also collected for later analysis. Acoustic data were recorded with a sampling rate of 140,000 Hz as a wav file by the

Raspberry Pi mini-computers (Python software). For acoustical analysis, recordings were transferred to Audacity and transformed to high resolution spectrograms (Window size: 1024,

Window type: Hanning, Gain (dB): 10, Range (dB): 60) where USVs were counted by a blinded experimenter after noise removal. USVs that appeared on the custom-made microphone and the commercial bat detector within 18-30 kHz and were consistent with the shape characteristics described by previous studies (Wright et al., 2010) were categorised as

22 kHz USVs. USVs that appeared on the commercial bat detector but not the custom-made microphone within 30-60 kHz and were consistent with the shape characteristics described by previous studies (Wright et al., 2010) were categorised as 50 kHz USVs. Video recordings of behaviour were also analysed by an experimenter blinded to the condition of each group.

6.3.5 Elevated plus maze test (EPM)

Apparatus: The wooden maze had an acrylic coating and two sets of perpendicular interlocking arms (50 cm long x 10 cm wide). The central region bisected the maze into two pairs of arms. The two closed arms had 40 cm-high walls, and the two open arms had none.

The entire maze was elevated 80 cm above the floor.

207

Procedure: On experimental day 15, rats were injected with test drug or saline 30 min before being place individually into the maze centre facing a closed arm and recorded on camera for 5 min for later analysis. Entry into an arm was defined as all four paws crossing from the central region into the arm. Open and closed arm entries or time were coded on a

PC. The percentage of open arm entries or time was expressed as open arm entries or time divided by open arm + closed arm entries or time. Recordings were coded before analysis to mask the treatment of the group and were analysed by an experienced experimenter.

6.3.6 Data analysis

To assess the possibility of a side bias in the CPP test, time spent in the to-be-drug- paired side was compared to time spent in the to-be-saline-paired side during preconditioning using pair Student’s t tests for each group. Place conditioning was assessed using paired Student’s t tests for paired samples comparing time spent in the drug-paired compartment before and after conditioning. For the EPM, saline groups were pooled and percentage open arm time and percentage open arm entries each were analysed with one-way

ANOVAs. USVs between groups were analysed using Pearson’s Chi-squared test with Yates’ continuity correction. Behaviour during USV testing for each drug group were analysed with independent Student’s t tests compared to respective saline controls.

208

6.4 Results

6.4.1 CPP

6.4.1.1 Preconditioning

Mean time spent in the to-be-drug-paired chamber was averaged over the three preconditioning days and compared to averaged time spent in the to-be-saline-paired chamber. Paired sample t tests showed no significant increase in time spent in either chamber

(Table 6.1). Thus, the CPP paradigm during conditioning was unbiased.

Table 6.1. Time spent in drug- or saline-paired chambers of conditioned place preference boxes during three daily 15-min preconditioning sessions Drug N Time (s) Significance Drug Paired Saline Paired Dexmphetamine (1.5 10 420.14 (+ 40.83) 419.93 (+ 35.28) NS mg/kg) Saline 10 454.85 (+ 30.16) 415.15 (+ 34.76) NS UWA-101 (3 mg/kg) 10 437.43 (+ 28.93) 402.57 (+ 29.61) NS Saline 10 450.05 (+ 15.69) 449.94 (+ 15.69) NS Data are Mean (+ S.E.M.) values for all session time (seconds). Dependent Student’s t tests for paired sample revealed no significant differences.

6.4.1.2 Drug-free test

The paired sample t test for rats treated with dexamphetamine (1.5 mg/kg) showed increased time spent in the drug-paired chamber (t(9) = 3.19, p < 0.05) after conditioning compared to preconditioning sessions (Figure 6.1). Saline-treated rats serving as controls for dexamphetamine- or UWA-101 groups did not increase time spent in either chamber (t(9) =

0.02, p = 0.98 and t(9) = 0.51, p = 0.62, respectively). Time spent in the drug-paired chamber did not increase in rats treated with UWA-101 (3 mg/kg) after conditioning compared to preconditioning sessions (t(9) = 0.55, p = 0.59).

209

* *

Figure 6.1. Time spent in the drug-paired chamber of conditioned place preference (CPP) boxes during pre-conditioning testing (baseline), drug-free (DF) testing, and state-dependent (SD) testing by healthy Sprague Dawley rats. *p < 0.05 compared to respective baseline

6.4.1.3 State-dependent test

Rats were re-tested after administration of their respective drugs that had been given on drug conditioning days. The dexamphetamine group spent a significantly greater amount of time in the drug-paired chamber compared to the preconditioning phase (t(9) = 2.32, p <

0.05) (Figure 6.1).

Rats treated with UWA-101 (3 mg/kg) showed no significant difference in time spent in the drug-paired chamber (t(9) = 0.49, p = 0.64). Rats in saline-treated groups as controls for dexamphetamine or UWA-101 controls did not differ in the time spent in either chamber (t(9)

= 0.57, p = 0.58 and t(9) = 1.02, p = 0.33, respectively).

6.4.2 USVs

6.4.2.1 Audio recordings

Initial analysis of the data revealed violations of assumptions of equal variance. To avoid these violations, all analysis was conducted using Pearson’s Chi Square test.

A chi square of independence test revealed that dexamphetamine-treated rats emitted significantly more 50 kHz USVs compared to saline-treated rats (χ2 (1, N = 20) = 5.21, p <

210

0.05) as seen in Figure 6.2. In contrast, the frequency of 50 kHz USVs by UWA-101-treated rats did not differ to that of saline-treated rats (χ2 (1, N = 20) = 2.14, p = 0.14).

Figure 6.2. Frequency of ultrasonic vocalisations emitted by dexamphetamine-, UWA-101- or saline-treated Sprague Dawley. (+)-Amph = dexamphetamine

No significant differences in the number of 22 kHz USVs were found between dexamphetamine-treated rats compared to saline-treated rats (χ2 (1, N = 20) = 0.00, p = 1.00) or UWA-101-treated rats compared saline controls rats (χ2 (1, N = 20) = 0.00, p = 1.00).

6.4.2.2 Behaviour

Behaviours analysed included rears and immobility time. Independent t tests found no significant differences in rears or immobility time between dexamphetamine- and saline- treated rats (p > 0.1). In contrast, UWA-101-treated rats displayed significantly fewer rears

(t18 = 3.82, p < 0.01) and increased immobility time (t13.06 = 2.87, p < 0.05) compared to saline controls (Figure 6.3).

211

Figure 6.3. Rearing and immobility time during USV testing between saline, dexamphetamine and UWA-101 groups. (+)-Amph = dexamphetamine, *p<0.05 compared to respective saline group

6.4.3 EPM

Due to the multiple violations of Levene’s test for homogeneity of variance, dexamphetamine- and UWA-101-treated rats were compared to a pooled saline-treated group and analysed using between-subject one-way ANOVAs.

A between-subject one-way ANOVA revealed a significant treatment effect for percentage of OAT (F (2, 37) = 11.46, p < 0.001). Planned pairwise comparisons revealed that dexamphetamine increased percentage of OAT compared to saline (p < 0.001), but

UWA-101 did not (p > 0.1, Figure 6.4A). Similarly, the one-way ANOVA for percentage of

OAE revealed a significant treatment effect among groups (F (2, 37) = 4.93, p < 0.05).

Pairwise comparisons showed that dexamphetamine-treated rats exhibited significantly greater percentage of OAE compared to saline-treated rats (p < 0.05), while percentage of

OAE in UWA-101-treated rats was not significantly different to controls (p > 0.1, Figure

6.4B).

212

* *

Figure 6.4. Comparisons between saline- and drug-treated groups of A) Percentage open arm time (OAT) B) Percentage open arm entries (OAE). (+)-Amph = dexamphetamine, *p<0.05 compared to saline group

A significant treatment effect was observed for total CAE (F (2, 37) = 5.95, p < 0.01).

Pairwise comparisons showed that dexamphetamine increased total CAE than saline rats (p <

0.05), while rats treated with UWA-101 showed no significant difference in total CAE compared to saline (p > 0.1, Table 6.2). A significant treatment effect was observed for total arm entries in a one-way ANOVA (F (2, 37) = 11.41, p < 0.001). Pairwise comparisons using

Bonferroni’s corrections revealed that dexamphetamine increased total arm entries compared to saline (p < 0.01), but that the total arm entries for the UWA-101-treated rats was not significantly different from saline controls (p = 0.28, Table 6.2).

The one-way ANOVA for number of rears revealed a significant treatment effect (F =

(2, 37) = 5.98, p < 0.01). Pairwise comparisons showed that dexamphetamine-treated rats displayed fewer rears compared to saline (p < 0.01), but not significant differences were found in UWA-101-treated rats compared to saline controls (p > 0.1, Table 6.2).

213

Table 6.2. Mean (+ SEM) of additional behaviours measured during the EPM test between saline, dexamphetamine and UWA-101 groups. (+)-Amph = dexamphetamine Behaviours Groups (Mean + SEM) Saline (+)-Amph UWA-101 Total Closed Arm Entries 6.5 (+ 0.69) 10.0 (+ 1.10) 4.9 (+ 1.25) Total Arm Entries 8.1 (+ 0.93) 14.3 (+ 1.35) 5.7 (+ 1.34) Rears 13.45 (+ 0.81) 8.6 (+ 1.89) 15.3 (+ 1.45)

6.5 Discussion Recent reports have described the antidyskinetic properties of UWA-101 in MPTP- treated primates and reducing L-DOPA-induced hyperkinesia in reserpine-treated rats while showing no effects on PPI in healthy rats (Huot et al., 2012a; Johnston et al., 2012).

However, these studies provide no indication of reinforcing effects that UWA-101 may induce. Therefore, the primary aim of this study was to investigate whether UWA-101 induces rewarding or anxiolytic effects in healthy rats through CPP, USV and EPM tests.

CPP is a reliable method of assessing the presence of reward, either drug-or naturally- induced (Tzschentke, 2007). UWA-101-treated rats did not show an increase (or decrease) in the amount of time spent in the drug-paired chamber in the current study. Increases in time spent in the drug-paired chamber indicate the presence of subjective rewarding effects of the drug due to an association between these effects and distinct external cues. This suggests that

UWA-101 (3 mg/kg) does not induce rewarding effects in healthy Sprague Dawley rats.

In contrast, MDMA establishes significant place preference in rats providing support for the presence of MDMA-induced reinforcing properties (Bilsky et al., 1991; Bilsky et al.,

1998; Aberg et al., 2007). Although these effects are generally robust, they tend to occur at specific doses depending on the route of administration. For example, MDMA (5 and 6.3 mg/kg, s.c.) induces CPP in rats (Bilsky et al., 1998; Herzig et al., 2005), while MDMA (2.5,

5, and 10 mg/kg, i.p) induces no place preference in rats (Cole et al., 2003). It is therefore possible that higher doses of UWA-101 may exert rewarding effects and therefore induce a

214

CPP compared to the 3 mg/kg dose tested in the current study. This 3 mg/kg dose of UWA-

101 was chosen for this study as this dose exerts anti-hyperkinetic-like and antidyskinetic- like effects in both reserpine-treated rats and MPTP-treated marmosets, respectively

(Johnston et al., 2012).

Unlike UWA-101, dexamphetamine reliably induced a CPP in the current study. This preference was seen in the increased amount of time dexamphetamine-treated rats spent in the drug-paired chamber during both drug-free and state-dependent tests. Similar to MDMA, dexamphetamine-induced CPP has been described by previous studies in both drug-free and state-dependent tests and is often used as a positive CPP control (Budygin et al., 2004;

Gerdjikov et al., 2004; Tzschentke, 2007). Positive controls are often used to ensure that any lack of effect when a new variable is introduced is not due to a procedural or mechanical factor. The dexamphetamine-induced CPP in the current study acts as a positive control which suggests that the lack of place preference in UWA-101-treated rats is due to the lack of psychoactivity of UWA-101 rather than another variable preventing the detection of a place preference.

Research into USVs of rodents is relatively recent. Studies into the context of USVs suggest that 50 kHz USVs generally indicate positive affect, while 22 kHz USVs generally indicate negative affect (Knutson et al., 2002; Mallo et al., 2007; Wright et al., 2010;

Burgdorf et al., 2011). Dexmphetamine, but not UWA-101, significantly increased the number of 50 kHz USVs compared to saline in the current study. However, dexamphetamine and UWA-101 did not influence the number of 22 kHz USVs. These results suggest that while dexamphetamine induces a positive affective state in rats, UWA-101 does not influence the affective state of rats.

One important feature to note here is that, while not statistically significant, rats administered UWA-101 did exhibit a substantial decrease in the number of 55 kHz USVs

215 emitted compared to their respective controls, without demonstrating a change in the number of 22kHz USVs emitted.

This effect of dexamphetamine is in agreement with previous studies that report an increase in the number of 50 kHz USVs after dexamphetamine administration compared to saline-treated rats (Ahrens et al., 2009; Wright et al., 2010; Wright et al., 2012b). Indeed, it is well established that psychostimulants such as cocaine (Wright et al., 2012b, Barker et al.,

2010; Barker et al., 2014; Browning et al., 2011; Ma et al., 2010; Meyer et al., 2012), methamphetamine (Mahler et al., 2013), and methylphenidate (Simola et al., 2012), and even their associated cues increase 50-kHz USV emission rates. It has been noted that pharmacological agents that primarily target dopamine and noradrenaline monoamines tend increase the emission rates of 50kHz USVs in rats, while drugs that focus on the serotonin system, such as MDMA, fail to induce this effect both on first exposure (Simola et al., 2012,

2014) and after repeated administration (Simola et al, 2014).

Indeed, MDMA (5-15 mg/kg, i.p.) has no significant effect on the number of 50 kHz

USVs emitted by rats in a drugged state compared to saline-treated rats (Simola et al., 2012;

Simola et al., 2014). This is consistent with the findings that MDMA (5-10 mg/kg, i.p.) does not induce CPP and is therefore unlikely to induce rewarding effects at these doses (Cole et al., 2003). This lack of effect of MDMA in rat 50 kHz USVs is similar to that of UWA-101 in the current study. The reason for this decrease is unclear, but may involved UWA-101 blunting positive affect without actively inducing negative effects. Further research may be useful in first replicating the finding and demonstrating statistical significance and then assessing the reason for any differences in 50 kHz USV emissions found. Paired with the lack of place preference or aversion in UWA-101-treated rats, these findings suggest that UWA-

101 does not exert subjective effects that are either reinforcing or aversive.

216

The lack of reinforcing or aversive effects by UWA-101 has important implications should this drug be tested clinically. An absence in behaviour associated with adverse subjective side effects in UWAA-101-treated rats supports the possibility that UWA-101 would not induce aversive effects in humans. This lack of effect may increase patient compliance for medication consumption of UWA-101 should this drug show antidyskinetic efficacy in PD patients.

UWA-101 significantly decreased the numbers of rears and increased the time spent immobile during the USV test compared to saline-treated controls in the current study. In contrast, rearing and immobile time seen in dexamphetamine-treated rats did not differ from saline-treated controls. MDMA (7.5-15 mg/kg, i.p.) does not influence the number of rears or locomotor activity compared to saline in acutely treated rats in open field studies under red light (Mechan et al., 2002; Ho et al., 2004). MDMA (12.5 mg/kg, i.p.) also does not influence freezing or immobility behaviour in the open field under red light conditions (Mechan et al.,

2002) and does not impact on locomotion compared to saline-treated rats (Ho et al., 2004).

However, MDMA (7.5 mg/kg, i.p.)-treated rats show higher locomotor counts compared to saline when tested in cages (Simola et al., 2014).

Although not directly comparable, the effects of UWA-101 in rats seems to differ from that of MDMA. While the current study analysed behaviour of UWA-101-treated rats in cages rather than open fields, the significant decreases in rearing and mobility are intriguing as these effects do not occur in MDMA-treated rats. These findings suggest that UWA-101 may have some inhibitory or sedatory effects in rodents. Movement inhibition by UWA-101 may be useful in treating conditions other than L-DOPA-induced dyskinesias in humans.

Interestingly, dexamphetamine did not induce significant effects on rearing or immobility compared to saline-treated rats during the USV test. In USV testing, other studies have found amphetamine (2.5 mg/kg, i.p.) increases percentage of time spent rearing and

217 locomotion activity compared to saline in red light conditions (Natusch et al., 2010).

Although these effects occurred at a higher dose of amphetamine than used in the current study (1.5 mg/kg), other researchers have also noted increased locomotion and rearing, particularly after repeated exposure (Ahrens et al., 2009) . Amphetamine can also influence rodent behaviour at specific time points after repeated administration. For example, rats administered amphetamine (1.75 mg/kg) on four separate occasions in seven days showed decreased locomotion during the first hour of testing on the fourth test day compared to the first hour of testing on the first test day before increasing significantly compared to saline during the remaining time of testing on the fourth day (Segal et al., 1987). Therefore, the

USV test in the current study may not have been long enough to detect a significant difference in amphetamine-induced behaviour.

Sedation and movement inhibition are common effects of benzodiazepines, antidepressants and anti-anxiety drugs. Although UWA-101 significantly reduced motor activity during the USV test, these sedatory effects are unlikely to be anxiolytic. No significant differences in percentage of open arm time or entries were found between UWA-

101- and saline-treated rats in the EPM test of the current study. Increases in open arm entries or time are associated with decreases in anxiety, whereas increases in anxiety are associated with greater thigmotaxic behaviour as seen by an increase in closed arm entries or time

(Pellow et al., 1985; Treit et al., 1993). These findings suggest that UWA-101 does not exert anxiolytic or anxiogenic properties.

Rats acutely exposed to MDMA do not show an increase in percentage of open arm time or entries in the EPM test (Mechan et al., 2002; Sumnall et al., 2004). However, there are strain differences in the EPM test up to 12 weeks after exposure to MDMA. Wistar rats tested 12 weeks after MDMA exposure show decreased open arm time (Mechan et al., 2002) and entries while Dark Agouti rats showed an increase in percentage of open arm time and

218 entries (Morley et al., 2001). Importantly, these measurements differ slightly as the behaviour of the Wistar rats was analysed using raw numbers and the Dark Agouti rat data was analysed as proportions. Nonetheless, both of these studies show that MDMA impacts on anxiety levels in rodent long after exposure to the drug. Therefore, investigating the impact of UWA-

101 on chronic anxiety levels in the EPM test is a potential avenue for future studies of this drug.

Furthermore, UWA-101 had no significant influence on closed arm entries, total arm entries or number of rears. The lack of effect of UWA-101 on these behaviours is generally consistent with the behavioural measurements during the USV test. Studies have reported no acute effect of MDMA on behaviour in the EPM including rearing and closed arm entries

(Mechan et al., 2002). However, rats tested 80 days after MDMA exposure did rear more often than saline controls (Mechan et al., 2002). The findings of the current study suggest that

UWA-101 does not influence anxiety or reward and only mildly influences locomotor activity in healthy Sprague Dawley rats, although long-term exposure effects are unknown.

Interestingly, dexamphetamine increased percentage open arm time and entries, total arm entries and total closed arm entries while reducing the number of rears during the EPM test in the current study. Such an increase in percentage open arm time and entries indicates a decrease in anxiety, while the increased total closed arm entries are a measurement of increased locomotor activity. These results are in contrast to previous EPM validation studies which report significant reductions in both percentage of open arm time and entries by amphetamine (1 and 2 mg/kg) (Pellow et al., 1985; McLelland et al., 2014). However, other researchers have described increased open arm activity induced by anxiogenic drugs such as amphetamine as a ‘false positive’ due to the increase in locomotor activity (Dawson et al.,

1995). Indeed, total arm entries and total closed arm entries were increased in amphetamine- treated rats in the current study which may account for this increase in open arm activity.

219

This explanation of a ‘false positive’ may also account for dexamphetamine-induced decrease in rearing frequency. As amphetamine is generally anxiogenic, rearing in the open arms is unlikely as this behaviour is associated with a decrease in anxiety. Because dexamphetamine- treated rats spent a larger portion of time in the open arms compared to saline controls, rearing frequency would have been reduced as many of the rats in the current study reared most often in the closed arms.

While the lack of effects of UWA-101 in the current study are encouraging, only one dose of UWA-101 (3 mg/kg) was used to investigate these behaviours. This dose was chosen specifically as it reverses selective L-DOPA-induced effects in animal models of Parkinson’s disease (Huot et al., 2012a; Johnston et al., 2012). As mentioned previously, MDMA induces

CPP at specific doses depending on administration route (Bilsky et al., 1991; Bilsky et al.,

1998; Cole et al., 2003; Herzig et al., 2005). UWA-101 may induce rewarding or anxiety- related effects at higher or lower doses and depending on the route of administration.

Therefore, further investigation into the effects of UWA-101 is needed to ensure that this novel MDMA analogue does not have misuse or abuse potential at higher doses. It is also important to note that UWA-101 was tested in a non-diseased model to assess rewarding and/or anxiolytic effects. The obvious next step in the research would be to further assess

UWA-101 in a PD animal model to better replicate the clinical context in which the drug could hopefully be used. Indeed, there could be a greater risk of abuse liability in PD patients, as both drug abuse and Parkinson’s disease both heavily involve the dopaminergic system.

The findings of the current study demonstrate that UWA-101 (3 mg/kg) does not induce place preference, alter USVs emission frequency or exert effects on open arm activity in rats. However, UWA-101 does reduce rearing and increase immobility time in specific contexts. These results suggest that UWA-101 does not exert rewarding properties, induce a positive affective state or influence anxiety levels, but does have an inhibitory effect on

220 movement. This effect on movement is likely to be related to its antidyskinetic properties as seen in reserpine-treated rats and MPTP-treated primates. The absence of reinforcing effects by UWA-101 at an antidyskinetic dose are encouraging as this indicates that the risk of misuse or abuse of UWA-101 is low. It is important to assess the reinforcing properties of any drug with potential for clinical use as the trend in non-medical use of prescription drugs is growing. Continued research into UWA-101-induced effects is needed to further assess this at different doses and to further investigate the inconsistent results on locomotion and mobility as demonstrated in the current study. However, much of the research into UWA-101 supports the potential for UWA-101 to be used clinically as an antidyskinetic agent in PD patients.

221

222

Chapter 7 – General Discussion

7 Summary of experimental aims

Dyskinesias continue to be a concerning side effect of L-DOPA treatment in PD patients.

Due to the chronic nature of PD, most patients will develop LID unless new treatments are introduced that are devoid of dyskinesia induction. Failing this, more efficacious antidyskinetic treatments are required as the current options, such as amantadine, are inconsistent at best. The antidyskinetic effect of MDMA-analogues is a pharmaceutical strategy being investigated in both rodent and primate models of LID as an adjunct to L-

DOPA therapy. One of the MDMA-analogues, UWA-101, has been shown to reduce LID- like behaviours in these models. The novelty of this drug and the limited research regarding its potential to dose-dependently reduce LID in animal models highlighted the need for further investigation into these effects. Therefore, it was the general aim of this thesis to add to the available literature regarding the reproducibility, dose response curve, antidyskinetic mechanism and reinforcing effects of UWA-101.

7.1 Significance of findings

The aims of this thesis were achieved with three successful studies conducted, using custom-made equipment as described in Chapter 2 and Chapter 3, interpreted and presented.

7.1.1 Reserpinised rat model of LID This thesis contains two studies of the reserpinised rat model of LID employing inexpensive, custom-made IR locomotion detection systems

7.1.1.1 Custom-made locomotion detection systems Chapter 2 outlined the design and construction of a system of testing enclosures able to detect the locomotion of rats by the interruption of infrared (IR) beams. This system was built using simple IR beams purchased online, hardware materials and accessible computer

223 components. The cost of constructing this locomotion detection system was substantially less than that of commercial suppliers and the components are easily accessible. The validation of this system demonstrated that reserpinised rats administered L-DOPA interrupted significantly more IR beams on the lower and upper levels compared to reserpinised rats administered saline, indicating an effect of L-DOPA-induced hyperkinesia.

Previous studies have shown that L-DOPA significantly increases locomotor activity in rats (Johnston et al., 2005; Segovia et al., 2003). Therefore, these findings demonstrate that the custom-made locomotion detection system effectively recorded the significantly different locomotor activity between groups administered L-DOPA or saline. This locomotion detection system is an affordable alternative for expensive commercial systems and construction can be replicated by smaller laboratories and graduate students with appropriate electrical and programming knowledge.

7.1.1.2 Validation of the reserpinised rat as a model of LID The study presented in Chapter 4 adequately characterised the reduction of select L-

DOPA-induced hyperkinetic behaviours by various drugs in reserpinised rats. L-DOPA- induced hyperkinesia in the form of excessive elevated locomotion and behaviour in reserpinised rats is thought to correlate with dyskinetic-like behaviour (Johnston et al., 2005;

Johnston et al., 2012). Not only was this study conducted to validate the reserpine-rat model of LID by administering antidyskinetic drugs and explore behavioural parameters, but no known literature had yet reproduced the anti-hyperkinetic effects of UWA-101 (3 mg/kg) in the reserpinised rat, nor had any known study described the behavioural effects induced by higher (10 mg/kg) or lower (1 mg/kg) doses of UWA-101 in this model.

L-DOPA-induced hyperkinesia was significantly reduced by UWA-101 (1 and 3 mg/kg), yohimbine, riluzole and rimonabant. All drugs have previously shown antidyskinetic properties in other animal models of PD (Dekundy et al., 2007; Johnston et al., 2012; Segovia

224 et al., 2003). However, all drugs excepting UWA-101 (10 mg/kg) also reduced L-DOPA- induced horizontal movement, but not below reserpine-induced levels. This horizontal movement is thought to be a behavioural correlate of parkinsonian symptoms, suggesting that an increase in horizontal movement by L-DOPA in reserpinised rats is representative of antiparkinsonian effects. These results, therefore, suggest that antiparkinsonian effects of L-

DOPA were reduced by the addition of UWA-101 (1 and 3 mg/kg), yohimbine, riluzole, and rimonabant.

The study also demonstrated the reproducibility of the anti-hyperkinetic-like properties of UWA-101 (3 mg/kg) in the reserpinised rat after L-DOPA administration by significantly reducing L-DOPA-induced elevated behaviour. These anti-hyperkinetic-like properties were also seen in rats treated with UWA-101 (1 mg/kg), although to a lesser extent. UWA-101 (10 mg/kg) exerted no such anti-hyperkinetic-like effects and had no significant impact on L-DOPA-induced behaviour overall.

The study also investigated new behavioural measurements in addition to horizontal and elevated locomotion. These measurements were introduced as subsets of elevated activity due to the expansive range of behaviours that were incorporated into elevated activity.

Elevated activity was divided into six separate measurements including elevated movement, centre elevated movement, elevated time, elevated bout, repeated movement while elevated and non-repeated movement while elevated. While elevated movement was not significantly affected by any of the drugs administered, increases in the other behaviours such as centre elevated movement, elevation time, bout and repetitive and non-repetitive movements while elevated by L-DOPA were selectively decreased by the antidyskinetic drugs.

It is important to acknowledge that further replication and assessment of these behavioural measurements is requirements given their novelty. In particular, given the range of behaviours described in this thesis, further work would be useful in assessing which

225 behaviours are most predictive of real dyskinetic or anti-parkinsonian responses. This could be done through a range of avenues including but not limited to 1) analysis of post-mortem neural tissue in rats shown exhibiting significant changes in horizontal and elevated activity after administration of L-DOPA and dyskinetic compounds, 2) an in vivo biochemical assessment of neurotransmitters during locomotive and behavioural testing, or 3) validation of the behavioural measurements using non-dyskinesia inducing drugs, such as dopamine agonists used in Parkinson’s disease, to further assess the correlation between rodent horizontal and elevated behaviour and parkinsonian and dyskinetic behaviours in humans.

Furthermore, this study investigated the effect of these drugs on reserpine-induced behaviour without administration of L-DOPA as an indication of antiparkinsonian effects.

The significant increases in horizontal movement and select elevated behaviours by UWA-

101 (3 and 10 mg/kg) and increases in horizontal movement and centre elevated movement by rimonabant and yohimbine indicate that these drugs may exhibit antiparkinsonian-like effects in reserpinised rats. However, the increases in elevated activity also suggest that doses would have to be titrated to prevent excessive movement fluctuations in the model.

Overall, the results of this study and their reproducibility by selectively decreasing L-

DOPA-induced elevated locomotion and behaviours and only moderately decreasing horizontal movement support the validity of L-DOPA-induced hyperkinesia in the reserpinised rat as an acute model of LID. The anti-hyperkinetic-like effect of UWA-101 (3 mg/kg) was also reproduced in this study and further anti-hyperkinetic-like effects at lower (1 mg/kg), but not higher (10 mg/kg), doses were also demonstrated. This study also identified new parameters of behaviour and locomotion that may be investigated to increase the validity of this model as a screening test for antidyskinetic drugs.

226

7.1.1.3 The mechanism underlying the anti-hyperkinetic effect of UWA-101 The study in Chapter 5 described how the anti-hyperkinetic-like effect of UWA-101 was blocked by administering drugs with known mechanisms of action in L-DOPA-treated reserpinised rats. There were two main findings in this study, the first of which was that we reproduced the attenuation of L-DOPA-induced hyperkinesia with UWA-101 administration as demonstrated in Chapter 4. Secondly, the results showed a reduction of L-DOPA-induced elevated activity by UWA-101 which was reversed with the administration of the specific 5-

HT1A receptor antagonist WAY 100635, the serotonin transporter (SERT) inhibitor fluoxetine, and the glutamate inhibitor riluzole.

The 5-HT1A receptor antagonist WAY100635 successfully blocked UWA-101’s attenuation of elevated time, elevated bout and repeated and non-repeated movement while elevated. The serotonin transporter (SERT) inhibitor fluoxetine blocked UWA-101’s the attenuation of elevated time, elevated bout and repeated movement while elevated.

Interestingly, yohimbine and riluzole increased elevated movement, centre elevated movement and non-repeated movement in reserpinised rats administered UWA-101 in combination with L-DOPA. Yohimbine also increased UWA-101-induced horizontal movement while riluzole increased UWA-101-induced elevated time. Both WAY100635 and fluoxetine had no effect on horizontal movement, elevated movement or centre elevated movement.

The attenuation of the anti-hyperkinetic-like effect of UWA-101 by drugs with a known mechanism helps to identify the underlying mechanism of UWA-101 itself. The attenuation of this effect by a 5-HT1A receptor antagonist and a SERT inhibitor indicates that this anti-hyperkinetic-like effect of UWA-101 is dependent upon the function of these two proteins. This is also consistent with previous studies that have shown UWA-101 to have an affinity for the SERT (Johnston et al., 2012), but the results of this study are the first to suggest an involvement of the 5-HT1A receptor in the mechanism of UWA-101. However, the

227 results after administration of yohimbine and riluzole also suggest that this serotonergic mechanism of UWA-101 may also interact with adrenergic and glutamatergic neurotransmitter systems in LID models.

7.1.2 Reinforcing effects of UWA-101 in reward and anxiety models This thesis contained two chapters investigating the reinforcing effects of drugs in rats using a custom-made ultrasonic microphone system.

7.1.2.1 Custom-made ultrasonic microphones In Chapter 3, results from a study of affordable alternatives to commercial ultrasonic microphones were presented. A recording system capable of recording the ultrasonic vocalisations (USVs) of four rats simultaneously was constructed from Panasonic electret microphone capsule soldered to cables and stereo jacks and plugged into sound cards connected to Raspberry Pi mini-computers. The Panasonic microphones successfully recorded artificial tones up to 40 kHz as an initial validation test. The Panasonic microphones detected and recorded 22 kHz USVs of saline-treated rats, but did not adequately record 50 kHz USVs emitted by dexamphetamine-treated rats. The overall cost of the recording system was substantially less than quotes obtained from six commercial suppliers.

The minimal expense of the Panasonic microphones provides an affordable alternative to commercial products for researchers interested in 22 kHz USVs of rodents as a model of anxiety and fear conditioning. The design of this system is flexible and can accommodate a range of experimental requirements even for studies of 50 kHz rat USVs in which the

Panasonic microphones could be replaced with commercial ultrasonic microphones.

228

7.1.2.2 The effects of UWA-101 on reward and anxiety Chapter 6 of this thesis presented the results of UWA-101-induced reinforcing or anxiolytic effects in three separate tests. Heathy rats administered UWA-101 (3 mg/kg) were exposed to conditioned place preference (CPP), USV, and elevated plus maze (EPM) tests.

No CPP was found in rats treated with UWA-101 in either the drug-free or state-dependent tests, while dexamphetamine-treated rats exhibited a robust CPP in both tests. The number of

50 or 22 kHz USVs emitted by UWA-101-treated rats did not differ from saline controls. In contrast, dexamphetamine-treated rats emitted significantly more 50 kHz USVs than saline controls. During the USV test, UWA-101 increased immobility time and reduced rearing compared to saline, although dexamphetamine did not alter behaviours significantly compared to saline. UWA-101 did not induce any changes in EPM test measurements compared to saline, although dexamphetamine increased percentage open arm time and entries compared to saline.

The lack of CPP in UWA-101-treated rats suggests that UWA-101 does not induce rewarding effects in rats as rats were not conditioned to favour the drug-paired chamber. The emission of 50 kHz USVs in rats is associated with rewarding experiences and a positive affective state (Ahrens et al., 2009; Browning et al., 2011; Burgdorf et al., 2011; Mallo et al.,

2007), while 22 kHz USVs are thought to indicate a negative affective state in aversive environments (Berger et al., 2013; Graham et al., 2009; Kim et al., 2010; Litvin et al., 2007).

Therefore, the lack of USVs emitted by UWA-101-treated rats indicates that the rats did not experience a positive or negative affective state during the USV test. Interestingly, the general reduction in movement is consistent with the reduced L-DOPA-induced hyperkinesia by UWA-101 (3 mg/kg) seen in the Chapter 4 and Chapter 5 of this thesis and animal models of LID in previous studies (Huot et al., 2012b; Huot et al., 2014; Johnston et al., 2012).

However, UWA-101 (3 mg/kg) also increased movement in rats treated only with reserpine, highlighting some behavioural inconsistencies between these studies which may be due to

229 methodical differences. Furthermore, the lack of influence of UWA-101 on anxiety-related behaviours in the EPM test suggests that this drug does not induce anxiolytic or anxiogenic effects in rats.

Overall, these findings support those of previous studies which suggest that UWA-

101 does not have psychoactive properties, unlike its parent drug MDMA (Johnston et al.,

2012). The lack of rewarding or anxiolytic effects of UWA-101 make it an attractive candidate for clinical trials in PD patients with LID as the current findings indicate it has a low risk of misuse or abuse potential compared to other clinically used drugs (Brown et al.,

2003; Pellowet al., 1985; Teteret al., 2006).

7.2 Discussion of findings

7.2.1 Custom-made locomotion detection systems The custom-made locomotion detection system successfully detected the movements of reserpinised rats administered either saline or L-DOPA. Moreover, the rats administered L-

DOPA broke the IR beams significantly more often than saline controls, as expected. The ease with which these simple infrared devices can be constructed and implemented, and their substantially lower cost compared to commercial IR devices are appealing reasons why investigators may consider constructing their own equipment rather than purchasing functionally equivalent equipment.

IR technology has several advantages over other methods of quantifying animal behaviours. Apart from affordable costs, the implementation of IR components is often unknown to the animal and requires little time to quantify beam breaks with the appropriate software compared to video recordings (Lacroix et al., 2000) or mechanical switches (Aujard et al., 2001). This means that the animals’ behaviour, either ethological or stimulus-induced, is rarely influenced by the method of data collection and that analysis can be completed efficiently.

230

Because the beam break data output is dependent on the accompanying software, the versatility of IR devices is limited only by the analysis to be performed. As with video tracking programs, the co-ordinates of the beam breaks in this IR locomotion detection system can be used to measure mobility time, travel patterns, distance travelled and movement in central or peripheral areas such as those of the open field test or chambers in

CPP testing enclosures. However, the application of IR devices for detecting rodent qualitative behaviours is limited in tests such as the social interaction and play (Lukas &

Wohr, 2015), the cat odour test (Arnold et al., 2010) and the defensive burying test

(Anderson et al., 2018). Therefore, IR devices are most versatile in studies of quantitative behaviour in a range of research fields.

While useful in the current thesis, the specific design of IR devices on the locomotion testing enclosures used here may be a limitation in other studies. The permanent attachment of the IR devices to the testing enclosures may reduce the design’s versatility in other experiments with different requirements. However, due to the flexibility of the components of the IR system, the authors encourage future researchers adapt the design used in Chapter 2 to accommodate the specific requirements of individual studies.

7.2.2 The reserpinised rat model of LID

All drugs that exert antidyskinetic properties in other animal models of LID successfully reduced L-DOPA-induced elevated activity in reserpinised rats in Chapter 4.

However, these drugs also reduced L-DOPA-induced horizontal activity. As horizontal activity is thought to correlate with parkinsonian symptoms, this reduction may represent a compromise of L-DOPA’s antiparkinsonian effects in this model. One explanation for this effect may be the regimen of drug administration as all drugs were administered 30 minutes before L-DOPA. This regimen was altered slightly in Chapter 5 where UWA-101 was

231 administered immediately after L-DOPA rather than 30 minutes before. In Chapter 5, UWA-

101 selectively reduced L-DOPA-induced elevated activity without influencing L-DOPA- induced horizontal activity. Indeed, this is consistent with previous studies in which the antidyskinetic drug was administered in combination with or immediately after L-DOPA, although precise times were not specified (Johnston et al., 2005; Johnston et al., 2012;

Segovia et al., 2003b). The differences in the effect of antidyskinetic drugs on L-DOPA- induced hyperkinesia based on different times of drug administration suggest that this model requires almost simultaneous drug administration to accurately represent the correlation in anti-hyperkinetic- and antidyskinetic-like drug effects.

Although the antidyskinetic drugs did successfully reduced elevated activity in general, not all subset measurements of elevated activity were affected in Chapters 2, 4 and 5.

Elevated movement was not affected by drug administration in Chapter 2. In contrast, L-

DOPA significantly increase elevated movement in the study described in Chapter 5, although no additional effect was seen after UWA-101 administration. However, yohimbine and riluzole did significantly increase elevated movement after L-DOPA administration. This inconsistency suggests that elevated movement is unlikely to be a reliable indicator of dyskinetic-like behaviours.

Other subsets of elevated activity were almost consistently influenced by drugs administered in Chapters 2, 4 and 5. These subset measurements of elevated activity include elevated time, elevated bout and repeated movement while elevated and non-repeated movement while elevated. It was noticed that drugs that exert behavioural effects through mechanisms implicated in LID development in humans generally reduced non-repeated movement while elevated. It is possible that non-repeated movement while elevated is a measurement that further discriminates efficacious and non-efficacious antidyskinetic drugs, but this would have to be verified through further investigation. Nonetheless, the findings

232 from Chapters 2, 4 and 5 suggest that elevated time, elevated bout and repeated movement while elevated are most likely to correlate with dyskinetic-like behaviour in other animal models of LID.

Previous studies have used only one measurement, vertical activity, to assess the antidyskinetic-like effects of drugs (Johnston et al., 2005; Johnston et al., 2012; Segovia et al., 2003b). However, there are concerns with using this one measurement as a behavioural correlated of LID. Previous reports have described several sub-categories of dyskinesias and abnormal involuntary movements in animal models and humans including dystonia, chorea, ballism and stereotypy (Iderberg et al., 2012; Luquin et al., 1992). It is unlikely that vertical activity as a single measurement in the reserpinised rat would be the best quantifier to represent all these dyskinetic sub-categories. Furthermore, vertical activity does not readily provide an estimation of time spent in a hyperkinetic state. The three novel behavioural measurements described in this thesis attempt to remedy these concerns. Elevated time is an indicator of total time spent in a hyperkinetic state (breaking the upper level IR beams), while elevated bout is an indicator of average length of each bout of hyperkinesia. Repeated movement while elevated provides an indication of the intensity of or hyperkinetic-like behaviour during elevated time. As a result, these three measurements help identify both the time spent in a hyperkinetic state and the intensity of hyperkinesia in that state.

While hyperkinesia is addressed in the measurements employed in Chapters 2, 4 and

5, other sub-categories such as chorea, ballism and dystonia are not. This is primarily due to the acute nature of the reserpine-rat model as there is insufficient time to develop AIMs after reserpine treatment that closely resemble dyskinetic behaviours. This contrasts with the

AIMS found in the 6-OHDA-lesioned rodent model of LID which are based on weeks of L-

DOPA exposure (Cenci & Lundblad, 2007; Dekundy et al., 2007; Monville et al., 2005).

Therefore, hyperkinesia is likely to be the only useful behavioural correlate of dyskinesia in

233 this acute reserpinised rat model. This is not necessarily a drawback as there is substantial overlap both behaviourally and pharmacologically between hyperkinesia and dyskinetic movements. Both behaviours are modulated by dopamine, serotonin and glutamate neurotransmitter systems (Bentall & Herberg, 1980; Blanchet et al., 1998; Fritts et al., 1997), and depend on D1 receptor activation for expression and dopamine receptor antagonism for attenuation (Aubert et al., 2005; Bhide et al., 2013; Guigoni et al., 2005; Segovia et al., 2003;

Taylor et al., 2005), and both can be induced be L-DOPA administration (Gerlach, 1977).

Therefore, these measurements allow for a multi-faceted analysis of L-DOPA-induced hyperkinesia in rats in an acute setting.

One important limitation of the reserpinised rat model of LID is the acute nature of the reserpine and L-DOPA exposure. Parkinson’s disease is characterised pathologically as a chronic degradation of dopaminergic neurons in the substantia nigra (Obeso et al., 2000) and the development of LID occurs after long-term treatment with L-DOPA (Ahlskog &

Muenter, 2001). In contrast, the reserpinised rat model involves a single, high dose of L-

DOPA 18 hours after a single reserpine treatment. This reserpinised rat model of LID attempts to imitate the chronic state of PD patients with LID in less than 24 hours.

Consequently, there are differences in the pathology between the two conditions.

Although dopamine depletion and akinesia are commonalities between the reserpinised rat and PD in humans, the actual dopamine neurons do not degenerate in the reserpine model.

Instead, reserpine acts deplete monoamines in the central nervous system through irreversible inhibition of the VMAT2 (Scherman, 1986). Furthermore, the acute high-dose L-DOPA induces hyperkinesia more reminiscent of stereotypy rather than specific dyskinetic behaviours seen in 6-OHDA-lesioned rats (Dekundy et al., 2007), MPTP-treated primates

(Riahi et al., 2011) or PD patients induced by long-term moderate-dose L-DOPA

(Constantinescu et al., 2007). Therefore, it is important to note that acute L-DOPA-induced

234 behaviour is thought to be a correlate of antiparkinsonian and dyskinetic effects rather than a behavioural representation as found in the 6-OHDA-lesioned rat model and MPTP-treated primate.

An important similarity between the reserpinised rat model of LID and other animal models of LID is that the mechanism of L-DOPA conversion to dopamine appears to be the same. In reserpine-treated rats, significant behavioural differences occur in rats administered

L-DOPA plus a peripherally acting AADC inhibitor with or without a centrally acting AADC inhibitor (Alachkar et al., 2010). Furthmore, diasylate concentrations of dopamine increase in the substantia nigra and corpus striatum after L-DOPA administration in rats pretreated with reserpine (Fisher et al., 1999). These studies suggest that in reserpine-treated rats, central

AADC is still involved in L-DOPA metabolism into dopamine.

As this thesis has investigated the effect of UWA-101 on L-DOPA-induced hyperkinesia in the reserpine-treated rat, an effect which appears to involve the serotonin system, it is important to assess the effect of reserpine on the serotinergic system. Indeed, acute reserpine treatment (90 min) increases serotonin levels within the limbic, striatal and cortical areas of rats, an effect which is blocked by administration of 5-HT1A autoreceptor agonist 8-OH-DPAT (Hjorth, 1992). Chronic administration of low-dose (0.2 mg/kg, i.p.) reserpine also reduces serotonin concentration significantly in the ventral tegmental area

(VTA) after an initial increase with acute reserpine administration (Antkiewicz-Michaluk et al., 2015). This mechanism of monoamine depletion, rather than lesioning of neurons, may have an impact on serotinergic neuroplasticity, but is unlikely to occur through the same mechanism as in neurotoxic models of Parkinson’s disease such as the 6-OHDA-lesioned rat model or the MPTP-treated non-human primate model.

Although the acute nature of the reserpinised rat model of LID limits its face validity in both PD and LID, the short time period required to deplete dopamine levels and assess

235 drug-induced locomotion makes it an inexpensive and time-efficient model of PD and LID.

Required resources such as animal housing, drug quantities, and equipment are substantially reduced compared to those for 6-OHDA-lesioned rat or MPTP-treated primate models of PD

(Dekundy et al., 2007; Johnston et al., 2012). Therefore, it is possible that the reserpinised rat model of LID could be used as a screening test for novel drugs with antidyskinetic-like properties before investigating these drugs with more accurate but more expensive and resource-heavy animal models. Screening antidyskinetic drugs through this model could reduce costs of research projects and the number of animals, particularly primates, used in

LID models. Thus, drugs that demonstrate promising antidyskinetic-like efficacy in the reserpinised rat could be further tested in models more closely resembling PD and LID, while screening out those that demonstrate reduced antidyskinetic potential.

7.2.3 Anti-hyperkinetic mechanism of UWA-101 The blockade of UWA-101-induced anti-hyperkinetic-like effects by WAY-100635 and fluoxetine in L-DOPA-treated reserpinised rats suggests mechanisms involved in exerting anti-hyperkinetic-like effects of UWA-101. As WAY-100635 is a 5-HT1A receptor antagonist, these findings suggest that 5-HT1A receptor activation is involved in anti- hyperkinetic-like properties of UWA-101, while the inhibition of the SERT by fluoxetine indicates SERT activation is also involved in exerting the anti-hyperkinetic-like properties of

UWA-101. These findings are consistent with antidyskinetic effects of MDMA which are also reduced after WAY-100635 administration, indicating that MDMA exerts its antidyskinetic properties via 5-HT1A receptor activation (Bishop et al., 2006). Although there is little research regarding SERT involvement in MDMA-induced antidyskinetic effects, it is possible that the SERT would be involved due to the large degree of overlap in pharmacological properties between UWA-101 and MDMA (Johnston et al., 2012).

236

These findings support the involvement of the serotonin system in the development of dyskinesias after L-DOPA exposure in Parkinson’s disease. Previous findings suggest that excess dopamine derived from exogenous L-DOPA may be released by serotonergic neurons as well as dopaminergic neurons (Bibbiani, Oh, & Chase, 2001; Carta et al., 2007; Lindgren et al., 2010). This excess in synaptic dopamine levels is thought to result in overstimulation of the dopamine receptors leading to excessive movements peripherally (Carta et al., 2007).

Dopamine is not known to have affinity for pre-synaptic 5-HT1A receptors which, when stimulated, inhibit further neurotransmitter release (Davidson & Stamford, 1995).

Therefore, should serotonergic neurons release dopamine, they would likely do so in an unregulated manner (Lindgren et al., 2010). However, stimulation of the presynaptic 5-HT1A receptors by serotonin can prevent further neurotransmitter release from vesicles and may inhibit excessive dopamine release from these neurons (Bara-Jimenez et al., 2005; Bibbiani et al., 2001; Durif, et al., 1995). Stimulation of these presynaptic receptors can also be increased by maintaining synaptic serotonin levels through SERT inhibition (Bishop et al., 2006; Durif et al., 1995). Based on previous studies, it is possible that UWA-101 reverses the SERT, leading to greater indirect stimulation of the 5-HT1A receptors due to increased synaptic serotonin levels (Bishop et al., 2006; Johnston et al., 2012), a mechanism which may then have been reduced by fluoxetine, a SERT inhibitor, in Chapter 5. Although more research is needed to support this possible mechanism of UWA-101 in exerting antidyskinetic effects, the finding that the anti-hyperkinetic-like properties of UWA-101 involve the SERT and 5-

HT1A receptors in the reserpinised rat is encouraging.

As UWA-101 has considerable affinity for the dopamine transporter (DAT) as well as the SERT (Huot et al., 2014; Johnston et al., 2012), the study described in Chapter 5 may have been improved with the inclusion of a DAT inhibitor drug. This would have investigated the possibility that UWA-101 exerts antidyskinetic properties through DAT

237 reversal or inhibition. In parallel to the serotoninergic system, it is possible that dyskinesia development involves excessive release of dopamine derived from exogenous L-DOPA from dopaminergic neurons. DAT inhibition would likely result in increased stimulation of the presynaptic dopamine autoreceptors, inhibiting further dopamine release. As UWA-101 has affinity for the DAT, it is possible that this transporter would also be involved in the antidyskinetic-like effects of UWA-101.

7.2.4 Custom-made ultrasonic microphones The Panasonic microphones described in Chapter 3 detected artificial frequencies up to 40 kHz and recorded 22, but not 50 kHz USVs in healthy rats. As 22 kHz USVs are indicative of negative affect (Brudzynski & Ociepa, 1992; Pohorecky, 2008), custom-made

Panasonic microphones would be an inexpensive alternative to commercial equipment in research concerned with anxiety or negative affect in rats. As USVs are thought to indicate the affect of rodents, USV tests add a new facet to reward, anxiety and even psychoactivity research that may explain behavioural tendencies. For example, anxiety models such as social interaction and aggression (Litvin et al., 2007) as well as exposure to cat odour (Arnold et al.,

2010) can be better understood as 22 kHz USVs and some categories of 50 kHz USVs are emitted in these contexts. USVs have also been used to explain unexpected phenomena reversal of PPI in phencyclidine-treated rats (Tunstall et al., 2009).

However, it must be noted that USV research is almost exclusively an indication of affective states in adult rats and not in adult mice (Portfors, 2007). However, mouse pups are known to emit USVs thought to indicate affect; the lack of fur and subcutaneous fat in a mouse pup makes it susceptible to colder temperatures, such as when the pup is outside the nest, which can endanger the pup (Portfors, 2007). As mother mice quickly retrieve the pups after they emit USVs, it is thought these vocalisations serve as a distress signal to elicit a

238 response in the mother (Ehret & Haack, 1984). This explanation for pup USVs is also applied to rat pups which vocalise when separated from the mother. Indeed, research into rat development such as schizophrenia models have utilised rat USVs to analyse social interactions in addition to behavioural measurements.

There is no evidence to suggest that the USVs of adult mice consistently indicates affect (Portfors, 2007). Rather, USVs in mice are often emitted in a range of experimental and social contexts and are increasingly analysed in behavioural phenotyping throughout the life-span of the mouse and in mouse models of neurodevelopmental disorders (Scattoni et al.,

2009). Thus, the Panasonic microphones are best utilised in the research of rat USVs. As

Chapter 6 demonstrates, these analysis of 22kHz USVs using Panasonic microphones would be best suited to anxiety models, but the entire recording system consisting of the Raspberry

Pis is a versatile system that could be used in 50 kHz USV research if a different model of ultrasonic microphone was employed.

7.2.5 Reinforcing effects of UWA-101 When tested in healthy rats, UWA-101 did not induce place preference, increase 50 kHz USVs or increase time spent in the open areas of the EPM. These results are encouraging as they suggest that UWA-101 does not induce rewarding or anxiety-related properties. These results support previous research that suggests UWA-101 does not induce psychogenic effects in rats (Johnston et al., 2012). This lack of psychogenic effect by UWA-101 increases the possibility of testing this drug in clinical trials, particularly in PD patients with LID. The verification of whether UWA-101 exerts reinforcing, anxiety-related psychoactive effects is important as its parent drug MDMA is known to induce a range of subjective effects in animals and humans and would compromise the potential of UWA-101 to be used as a clinical therapy (Aberg, et al., 2007; Banks et al., 2008; Curran & Travill, 1997).

239

Animal models have repeatedly demonstrated the rewarding effects of MDMA. In contrast to UWA-101, CPP effects have been described in MDMA-treated rats at 1.0-6.3 mg/kg doses depending on the route of administration (Bilsky et al., 1991; Cole et al., 2003;

Herzig et al., 2005). While rats do not increase emission of 50 kHz USVs after injections of

MDMA, they have been shown to emit more 50 kHz USVs when re-exposed to MDMA- associated cages in a drug-free state after repeated administration of MDMA on alternate days compared to vehicle (Simola et al., 2014). These results suggest that the subjective effects of MDMA can induce positive associations that are still active when the animals are in a drug-free state. Furthermore, self-administration of MDMA has been described in rats

(Schenk et al., 2003; Schenk et al., 2007), mice (Trigo et al., 2006) and monkeys (Banks et al., 2008). Studies have also noted that, compared to cocaine self-administration, MDMA self-administration occurs less frequently in rats (100% and 60% of rats tested, respectively), but latency to acquisition of self-administration is not different between 0.25 and 1.0 mg/kg doses of MDMA (Schenk et al., 2007).

MDMA has also been shown to exert a mixture of anxiogenic and anxiolytic properties in animals. This mixture of anxiety-related effects is primarily dependent on the context in which the animals are tested such as cat odour, footshock, and EPM tests. For example, rats treated with MDMA (5 mg/kg i.p.) spend significantly less time in close proximity to cat odour and exhibit increased anxiogenic behaviours in the emergence and

EPM tests (Morley & McGregor, 2000). However, rats administered MDMA (5 mg/kg i.p.) emit fewer footshock-induced 22 kHz USVs and spent more time in social interaction with other rats while exhibiting fewer aggressive behaviours (Morley & McGregor, 2000). These behaviours in the footshock and social interaction test suggest anxiolytic effects of MDMA.

MDMA induces a range of subjective and psychoactive effects in humans, both reinforcing and aversive. Reports have described anxiety, panic attacks, insomnia, flashbacks,

240 chronic psychoses (McGuire & Fahy, 1991), recurrent acute paranoid psychosis and cognitive disturbances (Curran & Travill, 1997). However, users often report positive moods after ingestion of MDMA, but feel significantly more depressed up to 48 hours after the

MDMA-induced euphoria (Parrott & Lasky, 1998).

The similar effects of MDMA and other reward- or anxiety-modulating drugs in animal models and humans demonstrates that models such as CPP, self-administration,

USVs, EPM and social interaction have substantial predictive validity. Indeed, potential anti- anxiety drugs are tested in reward and anxiety animal models before being used clinically

(Lynch et al., 2010). It is important for such drugs to be tested in a range of models because as the findings previously discussed demonstrate, the subjective effects of drugs can be influenced by the context of the animal or human.

Although informative, there are limitations to this study of rewarding and anxiolytic effects in UWA-101. As mentioned previously, the context of the test can influence the subjective effects of the drug in animals and humans (Curran & Travill, 1997; McGuire &

Fahy, 1991; Morley & McGregor, 2000; Parrott & Lasky, 1998; Tzschentke, 2007). While the findings show that UWA-101 does not induce a place preference, induce USVs or alter anxiety-related behaviours in the EPM, there are a range of other reward, anxiety or other psychoactive models in which UWA-101 may also be tested. Additionally, UWA-101 was tested at only one dose. This dose exerts antidyskinetic properties in MPTP-treated primates with LID and reserpinised rats (Johnston et al., 2012), but many drugs recreationally misused are taken at doses higher than those clinically prescribed. Therefore, it is possible that UWA-

101 might induce rewarding effects at higher doses and that these effects may also depend on the route of administration as seen in MDMA. Therefore, further research is would still be beneficial to investigate the effects of UWA-101 at different doses in a range of models and contexts, both in animals and in humans.

241

7.3 Improvements and future directions

7.3.1 Custom-made locomotion detection system

Although the current design of the locomotion detection system was effective in measuring locomotion in rats and accommodated specific requirements of the studies outlined in Chapters 4 and 5, these requirements may not be suitable for all investigators and experimental designs. Firstly, the placement of the IR devices resulted in the beam of IR light emitters overlapping across multiple sensors. This overlap prevented the IR sensors from detecting light beam breaks from their corresponding IR emitters due to IR beams projected from neighbouring emitters. Although shielding the IR emitters to narrow the beam width as done in Chapter 2 addressed this issue, alternating the placement of the IR devices along the walls of the testing enclosures to avoid IR light overlap (Figure 7.1) may also be another solution.

Figure 7.1. Alternating IR emitters (light grey) and sensors (dark grey) along two walls of a locomotion testing box to avoid IR light from emitters overlapping with multiple sensors.

Future studies could investigate this alternative placement pattern of IR devices on locomotion testing enclosures to determine the efficiency of such placements and the necessity of device shielding with tubing as used in the current design.

242

Secondly, the permanent placement of the IR devices onto the locomotion testing enclosures was not a limitation for the studies in Chapters 4 and 5, but may be inefficient in other experiments. Commercial locomotion detection equipment similar to our custom-made testing enclosures addressed this potential issue by placing IR devices on adjustable frames to allow for modification of height of the beams in the testing enclosures. While this may be useful for standard size testing enclosures, these frames would not be able to accommodate a wider range of larger testing enclosures. Therefore, an alternative method of height adjustment may be to attach the IR devices to a fastener able to slide along a frame vertically attached to the external wall of the testing box. In this way, the height of the IR devices could be adjusted while ensuring they are securely in place on the wall of the testing box without the accommodation of the technology being dependent on the width and length of the enclosures. Future studies may employ these suggestions to improve the design of the locomotor testing enclosures in such a way that would accommodate a range of box sizes. By increasing the range of enclosure sizes that this system design can accommodate, the versatility of this custom-made locomotion detection system may be useful in other models of animal research.

7.3.2 Validation of the reserpinised rat model of LID

The reserpinised rat model study described in Chapter 4 demonstrated that antidyskinetic drugs including UWA-101 reduced both horizontal movement and various measurements of elevated activity when administered 30 minutes before L-DOPA. However, the results from Chapter 5 showed that UWA-101 only reduced elevated activity and not horizontal activity when administered immediately after L-DOPA. This discrepancy highlights that importance of the pharmacokinetics involved in this model as the metabolism of drugs at different time points may influence effects on L-DOPA-induced hyperkinesia.

243

Studies in other animal models of LID have described simultaneous antidyskinetic drug administration with L-DOPA (Tahar et al., 2004) and administration of antidyskinetic drugs at specified time periods before L-DOPA based on pharmacokinetics of these compounds

(Dekundy et al., 2007). Therefore, the validation of the reserpinised rat as a model of LID could be improved on by investigating the time-dependent effects of antidyskinetic drugs and assessing the impact of different administration regimens on the L-DOPA-induced hyperkinesia. Studies of antidyskinetic drugs in this model could also include dose response curves for each drug tested as this aspect was not addressed in Chapters 4 and 5.

Regarding the actual reserpinised rat model of LID, future research may also focus on expanding this model into a chronic model of progressive dyskinesia development from long- term L-DOPA exposure. Indeed, previous studies have investigated using the reserpinised rat model as a chronic model of Parkinson’s disease, but these have focused on replicating the development of a parkinsonian tremor as represented by oral dyskinesia (Fernandes et al.,

2012), rather than investigating the additional element of dyskinesia develop through chronic and repeated L-DOPA exposure as found in the 6-OHDA-lesioned rat model (Dekundy et al.,

2007). There are no known studies that have investigated the reserpinised rat as a chronical model for LID, highlighting a new avenue for future investigation.

7.3.3 Anti-hyperkinetic mechanism of UWA-101

While the blockage of the anti-hyperkinetic actions of UWA-101 as described in

Chapter 5 are encouraging, further research is still needed to both reproduce the findings and assess the involvement of other neurotransmitter systems in these anti-hyperkinetic actions.

Previous studies have shown that UWA-101 has affinity for both the SERT and DAT

(Johnston et al., 2012) and, while Chapter 5 did describe the involvement of the SERT in the anti-hyperkinetic effect of UWA-101, involvement of the DAT was not assessed.

244

Administration of a DAT inhibitor may have been informative in investigating the role of the

DAT in UWA-101-induced anti-hyperkinetic effects in reserpinised rats, also possibly alluding to the mechanisms underlying mechanism of UWA-101 in MPTP-treated primates administered L-DOPA as seen in previous studies (Huot et al., 2014; Johnston et al., 2012).

Therefore, future studies could investigate whether the DAT modulates the anti- hyperkinetic effect of UWA-101 in the reserpinised rat through administration of a compound known to inhibit or reverse the DAT. These mechanisms could then be further reproduced or investigated in other models of LID such as the 6-OHDA-lesioned rat and the MPTP-treated primate. As demonstrated in Chapter 5, UWA-101 may also exert antiparkinsonian effects in reserpinised rats. The mechanism of this reduction in reserpine-induced akinesia was not investigated and therefore may be a possible avenue of research for future studies.

7.3.4 Custom-made ultrasonic microphones

While the custom-made microphones successfully recorded 22 kHz USVs in adult rats, the study could have been improved by assessing the detection of USVs in a range of microphone models in addition to the Panasonic WM-61A electret condenser microphone capsules. Alternatively, the study could also have investigated the construction of USV microphones by investigating whether the model of microphone capsule used in the commercial bat detector could be purchased separately. Other avenues for improving the construction and effectiveness of USV detectors include assessing the detection of 40 kHz rodent pup USVs using the Panasonic microphones as they recorded frequencies up to 40 kHz in Chapter 2.

245

7.3.5 Reinforcing effects of UWA-101

The dose of UWA-101 (3 mg/kg) shown to induce anti-hyperkinetic effects did not exert any reinforcing properties in the CPP, USV or EPM tests. However, the dose of drugs in a recreational context tends to be higher than doses used clinically, or drugs may be mixed to get a more intense ‘high’ (Compton & Volkow, 2006; Lessenger & Feinberg, 2008). While the anti-hyperkinetic dose of UWA-101 in rats does not influence behaviours in animal reward and anxiety models, higher doses of this drug may modulate such behaviours. This is the primary limitation of the study in Chapter 6, as only one dose of UWA-101 was tested in all three models. A dose response curve of UWA-101 in CPP, USV or EPM tests would be a useful pursuit in future studies of this drug. This pursuit could then be expanded to include other animal models of reward or anxiety that investigate other aspects of behaviour.

Furthermore, drugs can induce substantially different effects that are dependent on context and administration route. MDMA is known to induce CPP at 5 mg/kg given intraperitoneally but not at 5 mg/kg given subcutaneously (Cole et al., 2003; Herzig et al.,

2005). The effect of MDMA on anxiety tests is also inconsistent. MDMA (5 mg/kg, i.p.) increases social interaction and decreases 22 kHz USVs in the footshock test, indicating reduced anxiety (Morley & McGregor, 2000). Other studies of MDMA show that 7.5 mg/kg

(i.p.) induces anxiogenic behaviour in the EPM while 15 mg/kg (i.p.) induces anxiolytic behaviour (Ho et al., 2004). These results demonstrate that multiple tests and models are often required to build a detailed profile of the drugs investigated as their effects of behaviour are often context-dependent. One step further in the research of UWA-101 would be to investigate withdrawal effects in animals chronically exposed to UWA-101.

246

7.3.6 Overall future directions

While these improvements and avenues for future research are important, the transition of UWA-101 research from animal models of PD and LID to assess its antidyskinetic efficacy in PD patients suffering from LID is the most pertinent. To enable this transition, studies into toxicology, dose titration, and chronic exposure to UWA-101 in animal models are necessary and should be reproducible. Future studies should also assess the safety of UWA-101 in healthy human volunteers in terms of adverse effects and abuse liability. Such studies can better assess the likelihood for patient compliance UWA-101 is well tolerated and can assess the likelihood of non-medical use of UWA-101 occurring.

Pharmacokinetic and pharmacodynamics studies in PD patients are also required in addition to behavioural assessment determine the antidyskinetic efficacy of UWA-101 to treat LID.

Studies may also investigate the efficacy of UWA-101 in treating other conditions such as parkinsonian motor symptoms as the results in Chapter 5 suggest that UWA-101 may exert antiparkinsonian effects as well as having antidyskinetic properties.

7.4 Conclusion The administration of 3 mg/kg UWA-101 resulted in reduced L-DOPA-induced hyperkinesia in reserpinised rats, an effect modulated by serotonin receptors. Furthermore, administration of 3 mg/kg UWA-101 alone reversed akinesia in reserpinised rats and induced no reinforcing effects in healthy rats. Although the reduction in L-DOPA-induced hyperkinesia by UWA-101 has been demonstrated previously (Johnston et al., 2012), this thesis demonstrates the reproducibility of this effect and demonstrates the effects of other doses of UWA-101 in the reserpinised rat for the first time.

The administration of 1 mg/kg UWA-101 had similar effects on L-DOPA-induced hyperkinesia in the reserpinised rat compared to 3 mg/kg, but exerted no effects on reserpine- induced akinesia. In contrast, 10 mg/kg UWA-101 reversed reserpine-induced akinesia but

247 had no effect on L-DOPA-induced activity. These results suggest that 3 mg/kg UWA-101 may be an efficacious dose for research into the reserpinised rat either as a model of PD or as a model of LID and suggest that UWA-101 may have antiparkinsonian benefits. These findings also support the validity of the reserpinised rat model of LID as an inexpensive and efficient screening test for potentially antidyskinetic drugs to reduce the number of higher order animals used in dyskinesia research.

The mechanism underlying the reversal of L-DOPA-induced hyperkinesia, but not reserpine-induced akinesia, by UWA-101 (3 mg/kg) was blocked by SERT inhibition and 5-

HT1A receptor antagonism. This suggests that UWA-101 acts to attenuate L-DOPA-induced hyperkinesia through the serotonin system in a similar fashion to its parent drug MDMA

(Bishop et al., 2012) and provides further evidence for the involvement of the serotonin system in the development of LID (Fox et al., 2009). However, research is needed to verify this mechanism underlying the antidyskinetic effect of UWA-101 in more detail as the reserpinised rat model of LID is not a pathologically representative model of dyskinesia.

This thesis also demonstrated that an antidyskinetic dose of UWA-101 (3 mg/kg) in healthy rats failed to induce effects of modulated reward or anxiety in a range of tests. This is in contrast to MDMA (Ho et al., 2004; Morley & McGregor, 2000), providing supportive evidence that clinical use of UWA-101 is unlikely to result in non-medical use or abuse.

The findings of the thesis overall support the use of UWA-101 as a potential antidyskinetic treatment in PD patients treated with L-DOPA. However, further research of this drug is still required in a range of areas before this drug is deemed suitable to be clinically trialled as an efficacious treatment of LID in PD patients.

248

References Abercrombie ED, Bonatz AE, Zigmond & MJ (1990). Effects on L-DOPA on extracellular dopamine in striatum of normal and 6-hydroxydopamine-treated rats. Brain Res, 525: 36-44. Aberg M, Wade D, Wall E & Izenwasser S (2007). Effect of MDMA (ecstasy) on activity and cocaine conditioned place preference in adult and adolescent rats. Neurotoxicol Teratol, 29: 37-46. doi:10.1016/j.ntt.2006.09.002 Adamas Pharmaceuticals (2017). Adamas announces U.S. FDA acceptance of ADS-5102 new drug application for the treatment of levodopa-induced dyskinesia in patients with Parkinson's disease: Emeryville, California p2. Aghajanian GK & Marke GJ (1999). Serotonin, via 5-HT2A receptors, increases EPSCs in layer V pyramidal cells of prefrontal cortex by an asynchronous mode of glutamate release. Brain Res, 825: 161-171. Ahlenius S & Engel J (1971). Behavioural and biochemical effects of L-DOPA after inhibition of dopamine-B-hydroxylase in reserpine pretreated rats. Nunyn-Schmiedelbergs Ach Pharmak, 270: 349-360. Ahlskog JE & Muenter MD (2001). Frequency of levodopa-related dyskinesias and motor fluctuations as estimated from the cumulative literature. Mov Disord, 16: 448-458. Ahrens AM, Nobile CW, Page LE, Maier EY, Duvauchelle CL & Schallert T (2013). Individual differences in the conditioned and unconditioned rat 50-kHz ultrasonic vocalizations elcitied by repeated amphetamine exposure. Psychopharamcol, 229: 687-700. Ahrens AM, Ma ST, Maier EY, Duvauchelle CL & Schallert T (2009). Repeated intravenous amphetamine exposure: rapid and persistent sensitization of 50-kHz ultrasonic trill calls in rats. Behav Brain Res, 197: 205-209. doi:10.1016/j.bbr.2008.08.037 AIHW (2011). 2010 National Drug Strategy Household Survey Report. Canberra: Australian Institute of Health and Welfare Alachkar A, Brotchie JM & Jones OT (2010). Locomotor response to L-DOPA in reserpine-treated rats following central inhibiton of aromatic L-amino acid decarboxylase: Further evidence for non-dopaminergic actions of L-DOPA and its metabolites. Neurosci Res, 68: 44-50. Allen TA, Furtak SC & Brown TH (2007). Single-unit responses to 22 kHz ultrasonic vocalizations in rat perirhinal cortex. Behav Brain Res, 182: 327-336. doi:10.1016/j.bbr.2007.03.009 Allred RP, Adkins DL, Woodlee MT, Husbands LC, Maldonado MA et al. (2008). The vermicelli handling test: a simple quantitative measure of dexterous forepaw function in rats. J Neurosci Methods, 170: 229-244. Almirall H, Pigarev I, de la Calzada MD, Pigareva M, Herrero MT & Sagales T (1999). Nocturnal sleep structure and temperature slope in MPTP treated monkeys. J Neural Transm, 106: 1125-1134. American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders. Fourth Edition Text Revision (DSM-IV-TR). American Psychiatric Association, Washington, D.C. Andersen, JK, Frim DM, Isacson O, Beal MG & Breakefield XO (1994). Elevation of neuronal MAO-B activity in a transgenic mouse model does not increase sensitivity to the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). Brain Res, 656: 108-114. Anderson EM, McWaters ML, McFadden LM & Matuszewich L (2018). Defensive burying as an ethological approach to studying anxiety: Influence of juvenile methamphetamine on adult defensive burying behavior in rats. Learn Motiv, 61: 97-106. doi:10.1016/j.lmot.2017.02.001 Anderson JW (1954). The production of ultrasonic sounds by laboratory rats and other mammals. Science, 119: 808-809. Andersson M, Hilberston A & Cenci MA (1999). Striatal fosB expression is causally linked with L- DOPA-induced abnormal involuntary movements and the associated upregulation of striatal prodynorphin mRNA in a rat model of Parkinson's disease. Neurobiol Dis, 6: 461-474.

249

Anson LC, Chen PE, Wyllie DJA, Colquhoun D & Schoepfer R (1998). Identification of amino acid residues of the NR2A subunity that control glutamate potency in recombinant NR1/Nr2A NMDA receptors. J Neurosci, 18: 581-589. Antkiewicz-Michaluk L, Wasik A, Mozdzen E, Romanska I, Michaluk J (2015). Withdrawal from repeated administration of a low dose of reserpine-induced opposing adaptive changes in the noradrenaline and serotonin system function: A behavioural and neurochemical ex vivo and in vivo studies in the rat. Prog Neuropsychpharm Biol Psychiat, 57: 146-154. Antonini A (2007). Continuous dopaminergic stimulation--from theory to clinical practice. Parkinsonism Relat Disord, 13: Suppl S24-28. Antonini A, Mancini F, Canesi M, Zangaglia R, Isaias IU, Manfredi L, et al. (2008). Duodenal levodopa infusion improves quality of life in advanced Parkinson's disease. Neurodegener Dis, 5: 244-246. Antonini A, Isaias IU, Canesi M, Zibetti M, Mancini F, Manfredi L, et al. (2007). Duodenal levodopa infusion for advanced Parkinson's disease: 12-month treatment outcome. Mov Disord, 22: 1145-1149. Apostolakis EM, Garai J, Fox C, Smith CL, Watson SJ, Clark SH, et al. (1996). Dopaminergic regulation of progesterone receptors: Brain D5 dopamine receptors mediate induction of lordosis by D1-like agonists in rats. J Neurosci, 16: 4823-4834. Arai R, Karasawa N, Geffard M & Nagatsu I (1995). L-DOPA is converted to dopamine in serotonergic fibers of the striatum of the rat: a double-labeling immunofluorescence study. Neurosci Lett, 195:195-198. Arai R, Karasawa N, Geffard M, Nagatsu T & Nagatsu I (1994). Immunohistochemical evidence that central serotonin neurons produce dopamine from exogenous L-DOPA in the rat, with reference to the involvement of aromatic L-amino acid decarboxylase. Brain Res, 667: 295- 299. Arnold JC, Dielenberg RA & McGregor IS (2010). Cannabinoids increase conditioned ultrasonic vocalisations and cat odour avoidance in rats: strain differences in drug-induced anxiety. Life Sci, 87: 572-578. doi:10.1016/j.lfs.2010.09.018 Arnt J (1985) Hyperactivity induced by stimulation of separate dopamine D-1 and D-2 receptors in rats with bilateral 6-OHDA lesions. Life Sci, 37: 717-723. Ator NA & Griffiths RR (2003). Principles of drug abuse liability assessment in laboratory animals. Drug Alcohol Dep, 70: S55-S72. Aubert I, Guigoni C, Hakansson K, Li Q, Dovero S, Barthe N, et al. (2005). Increased D1 dopamine receptor signaling in levodopa-induced dyskinesia. Ann Neurol, 57: 17-26. doi:10.1002/ana.20296 Australian Institute of Health and Welfare [AIHW] (2005). 2004 national drug strategy household survey: first results. AIHW cat. no PHE 57. Canberra: AIHW (Drug statistics series no. 13). Aujard F, Herzog ED & Block GD (2001). Circadian rhythms in firing rate of individual suprachiasmatic nucleus neurons from adult and middle-aged mice. Neuroscience, 106: 255- 261. Bals-Kubik R, Ableitner A, Herz A & Shippenberg TS (1993). Neuroanatomical sites mediating the motivational effects of opioids as mapped by the conditioned place preference paradigm in rats. J Pharmacol Exp Therap, 264: 489-495. Banahan C, Rogerson Z, Rousseau C, Ramnarine KV, Evans DH & Chung EM (2014). An in vitro comparison of embolus differentiation techniques for clinically significant macroemboli: dual-frequency technique versus frequency modulation method. Ultrasound Med Biol, 40: 2642-2654. doi:10.1016/j.ultrasmedbio.2014.06.003 Banks ML, Czoty PW, Gage HD, Bounds MC, Garg PK, Garg S & Nader MA (2008). Effects of cocaine and MDMA self-administration on serotonin transporter availability in monkeys. Neuropsychopharm, 33: 219-225. doi:10.1038/sj.npp.1301420

250

Bara-Jimenez W, Dimitrova TD, Sherzai A, Aksu M & Chase TN (2006). Glutamate release inhibition ineffective in levodopa-induced motor complications. Mov Disord, 21: 1380-1383. Bara-Jimenez W, Bibbiani F, Morris M.J, Dimitrova T, Sherzai A, Mouradian MM & Chase, T. N. (2005). Effects of serotonin 5-HT1A agonist in advanced Parkinson's disease. Mov Disord, 20: 932-936. doi:10.1002/mds.20370 Bara-Jimenez A, Dimitrova T, Sherzai A, Favit A, Mouradian MM & Chase TN (2004). Effect of monoamine reuptake inhibitor NS 2330 in advanced Parkinson’s disease” Mov DIsord, 19: 1183-1186. Barfield RJ & Geyer LA (1972). Sexual behavior: ultrasonic postejeculatory song of the male rate. Science, 176: 1349-1350. Barker DJ, Simmons SJ, Servilio LA, Bercovicz D, Ma S, Root DH, Pawlak AP & West MO (2014). Ultrasonic vocalizations: evidence for an affective opponent process during cocaine self- administration. Psychopharmacol, 231: 909-918. Barker DJ, Root DH, Ma S, Jha S, Megehee L, Paelak AP, West MO (2010). Dose-dependent differences in short ultrasonic vocalizations emitted by rats during cocain self-administration. Psychopharmacol, 211: 435-442. Barnum CJ, Bhide N, Lindenbach D, Surrena MA, Goldenberg AA, Tignor S, Klioueva A, Walters H & Bishop C (2012). Effects of noradrenergic denervation on L-DOPA-induced dyskinesia and its treatment by a- and B-adrenergic receptor antagonists in hemiparkinsonian rats. Pharmacol Biochem Behav, 100: 607-615. Batson JD & Turner JD. (1986). A simple infrared photocell device and its interface to an Apple computer. Behavior Research Methods, Instruments & Computers, 18: 447-51. Battaglia G, Brooks BP, Kulsakdinun C & De Souza EB (1988). Pharmacologic profile of MDMA (3,4-methylenedioxy-methamphetamine) at various brain recognition sites”. Eur J Pharmacol, 149: 159-163. BBC (2001). Ecstasy 'relieves Parkinson's disease'. BBC News. Beal MF (2002). Oxidatively modified proteins in aging and disease. Free Radic Biol Med, 32: 797- 803. Beaulieu JM & Gainetdinov RR (2011). The physiology, signaling, and pharmacology of dopamine receptors. Pharmacol Rev, 63: 182-217. Bedard PJ, Di Paolo T, Falardeau P & Boucher R (1986). Chronic treatment with L-DOPA, but not bromocriptine, induces dyskinesia in MPTP-parkinsonian monkeys. Correlation with [3H]Spiperone binding. Brain Res, 379: 294-299. Benazzouz A, Boraud T, Dubedat P, Boireau A, Stutzmann JM & Gross C (1995). Riluzole prevents MPTP-induced parkinsonism in the rhesus monkey: a pilot study. Eur J Pharmacol, 284: 299- 307. Benazzouz A, Gross C, Dupont J & Bioulac B (1992). MPTP induced hemiparkinsonism in monekys: behavioral, mechanographic, electromyographic and immunohistochemical studies. Exp Brain Res, 90: 116-120. Bentall ACC & Herberg LJ (1980). Blockade of amphetamine-induced locomotor activity and stereotypy in rats by spiroperidol but not by atypical neuroleptic, thioridazine. Neuropharmacology, 19: 699-703. Berg D, Godau J, Trenkwalder C, Eggert K, Csoti I, Storch A, Huber H, Morelli-Canelo M, Stamelow M, Ries V, Wolz M, Schneider C, di Paolo T, Gasparini F, Hariry S, Vandemeulebroecke M, Abi-Saab W, Cooke K, Johns D & Gomez-Mancilla B (2011). AFQ056 treatment of levodopa-induced dyskinesias: results of 2 randomized controlled trials. Mov Disord, 26: 1243-1250. Berger AL, Williams AM, McGinnis MM & Walker BM (2013). Affective cue-induced escalation of alcohol self-administration and increased 22-kHz ultrasonic vocalizations during alcohol

251

withdrawal: role of kappa-opioid receptors. Neuropsychopharmacology, 38: 647-654. doi:10.1038/npp.2012.229 Betancourt J, Rios JL, Pagan I, Fabian C, Gonzalez AM, Cruz SY, Gonzalez MJ, Rivera WT & Palacious C (2013). Non-medical use of prescription drugs and its associated with socio- demographic characteristics, dietary pattern, and perceived academic load and stress in college students in Puerto Rico. PRHSJ, 32: 89-94. Bez F, Francardo V, Cenci MA (2016). Dramatic differences in susceptibility to L-DOPA-induced dyskinesia between mice that are aged before or after a nigrostriatal dopamine lesion. Neurobiol Dis, 94: 213-225. Bezard E, Ferry S, Mach U, Stark H, Leriche L, Boraud T, Gross C & Sokoloff P (2003). Attenuation of levodopa-induced dyskinesia by normalizing dopamine D3 receptor function. Nat Med, 9: 762-767. Bezard E, Brotchie JM, & Gross CE (2001). Pathophysiology of levodopa-induced dyskinesia: potential for new therapies. Nat Rev Neurosci 2: 577-588. Bhide N, Lindenbach D, Surrena MA, Goldenberg AA, Bishop C, Berger SP & Paquette MA (2013). The effects of BMY-14802 against L-DOPA- and dopamine agonist-induced dyskinesia in the hemiparkinsonian rat. Psychopharmacology (Berl), 227: 533-544. doi:10.1007/s00213- 013-3001-4 Biala G, Kruk M & Budzynska B (2009). Effects of the cannabinoid receptor ligands on anxiety- related effects of d-amphetamine and nicotine in the mouse elevated plus maze test. J Physiol Pharmacol, 60: 113-122. Bibbiani F, Oh JD, Kielaite A, Collins MA, Smith C & Chase TN (2005). Combined blockade of AMPA and NMDA glutamate receptors reduces levodopa-induced motor complications in animal models of PD. Exp Neurol 196: 422-429. Bibbiani F, Oh JD & Chase TN (2001). Serotonin 5-HT1A agonist improves motor complications in rodent and primate parkinsonian models. Neurology, 57: 1829-1834. Bilska A, Dubiel M, Sokolowska-Jezewicz M, Lorenc-Koci E & Wlodek L (2007) Alpha-lipoic acid differently affects the reserpine-induced oxidative stress in the striatum and prefrontal cortex of rat brain. Neuroscience, 146: 1758-1771. Bilsky EJ, Montegut MJ, Nichols ML & Reid LD (1998). CGS 10746B, a novel dopamine release inhibitor, blocks the establishment of cocaine and MDMA conditioned place preference. Pharmacology, biochemistry, and behavior 59: 215-220. Bilsky EJ, Hubbell CL, Delconte JD & Reid LD (1991). MDMA produces a conditioned place preference and elicits ejaculation in male rats: a modulatory role for the endogenous opioids. Pharmacol Biochem Behav, 40: 443-447. Bisaga A, Comer SD, Ward AS, Popik P, Kleber HD & Fischman MW (2001). The NMDA antagonist memantine attenuates the expression of opioid physical dependence in humans. Psychopharmacol, 157: 1-10. Bishop C, George JA, Buchta W, Goldenberg AA, Mohamed M, Dickinson SO., Eissa S, Eskow Jaunarajs KL (2012). Serotonin transporter inhibition attenuates l-DOPA-induced dyskinesia without compromising l-DOPA efficacy in hemi-parkinsonian rats. Eur J Neurosci, 36: 2839- 2848. doi:10.1111/j.1460-9568.2012.08202.x Bishop C, Krolewski DM, Eskow KL, Barnum CJ, Dupre KB, Deak T & Walker PD (2009). Contribution of the striatum to the effects of 5-HT1A receptor stimulation in L-DOPA-treated hemiparkinsonian rats. J Neurosci Res, 87: 1645-1658. Bishop C, Taylor JL, Kuhn DM, Eskow KL, Park JY & Walker PD (2006). MDMA and fenfluramine reduce L-DOPA-induced dyskinesia via indirect 5-HT1A receptor stimulation. Eur J Neurosci, 23: 2669-2676. doi:10.1111/j.1460-9568.2006.04790.x

252

Bishop C, Tessmer JL, Ullrich T, Rice KC & Walker PD (2004). Serotonin 5-HT2A receptors underlie increased motor behaviors induced in dopamine-depleted rats by intrastriatal 5- HT2A/2C agonism. J Pharmacol Exp Ther, 310: 687-694. Blanchard RJ, Blanchard DC, Agullana R & Weiss SM (1991). Twenty-teo kHz alarm cries to presentation of a predator, by laboratory rats living in visible burrow systems. Physiol Behav, 50: 967-972. Blanchet PJ, Konitsiotis S, Whittemore ER, Zhou ZL, Woodward RM & Chase TN (1999) Differing effects of N-methyl-D-aspartate receptor subtype selective antagonists on dyskinesias in levodopa-treated 1-Methyl-4-phenyl-tetrahydropyridine monkeys. J Pharmacol Exp Therap, 290: 1034-1040. Blanchet PJ, Konitsiotis S & Chase TN (1998). Amantadine reduces levodopa-induced dyskinesias in parkinsonian monkeys. Mov Disord, 13: 798-802. Block G, Liss C, Reines S, Irr J, & Nibbelink D (1997). Comparison of immediate-release and controlled release carbidopa/levodopa in Parkinson's disease. Eur Neurol 37: 23-27. Borta A, Wohr M & Schwarting RK (2006). Rat ultrasonic vocalization in aversively motivated situations and the role of individual differences in anxiety-related behavior. Behav Brain Res, 166: 271-280. doi:10.1016/j.bbr.2005.08.009 Boyce S, Rupniak NMJ, Steventon MJ & Iversen SD (1990). Characterisation of dyskinesias induced by L-DOPA in MPTP-treated squirrel monkeys. Psychopharmacology, 102: 21-27. Boyd CJ, McCabe SE, Cranford JA & Young A (2006). Adolescents' motivations to abuse prescription medications. Pediatrics 118: 2472-2480. Bozarth MA (1990). Evidence for the rewarding effects of ethanol using the conditioned place preference method. Pharmacol Biochem Behav, 35: 485-487. Brannan T, Martinez-Tica J & Yahr MD (1991). Effect of yohimbine on brain monoamines: An in vivo study. Journal of Neural Transmission - Parkinson's Disease and Dementia Section, 3: 81-87. doi:10.1007/bf02260883 Bredberg E, Nilsson D, Johansson K, Aquilonius SM, Johnels B, Nystrom C, Paalzow L (1993). Intraduodenal infusion of a water-based levodopa dispersion for optimisation of the therapeutic effect in severe Parkinson's disease. Eur J Clin Pharmacol 45: 117-122. Breger LS, Dunnett SB & Lane EL (2013). Comparison of rating scales used to evaluate L-DOPA- induced dyskinesia in the 6-OHDA lesioned rat. Neurobiol Dis, 50: 142-150. Brodsky MA, Park BS & Nutt JG (2010). Effects of a dopamine agonist on the pharmacodynamics of levodopa in Parkinson disease. Arch Neurol 67: 27-32. Brown ES, Nejtek VA, Perantie DC, Rajan Thomas N & Rush AJ (2003). Cocaine and amphetamine use in patients with psychiatric illness: a randomized trial of typical antipsychotic continuation or discontinuation. J Clin Psychopharmacol, 23: 384-388. doi:10.1097/01.jcp.0000085412.08426.08 Browning JR, Browning DA, Maxwell AO, Dong Y, Jansen HT, Panksepp J & Sorg BA (2011). Positive affective vocalizations during cocaine and sucrose self-administration: a model for spontaneous drug desire in rats. Neuropharmacology, 61: 268-275. doi:10.1016/j.neuropharm.2011.04.012 Brudzynski SM, Kehor P & Callahan M (1999). Sonographic structure of isolation-induced ultrasonic calls of rat pups. Dev Psychbiol, 34: 195-204. Brudzynski SM, Bihari F, Ociepa D & Fu X-W (1993). Analysis of 22 kHz ultrasonic vocalization in laboratory rats: long and short calls. Physiol Behav, 54: 215-221. Brudzynski SM & Ociepa D (1992). Ultrasonic vocalization of laboratory rats in response to handling and touch. Physiol Behav, 52: 655-660. Brus R, Kostrzewa RM, Perry KW & Fuller RW (1994). Supersensitization of the oral response to SKF 38393 in neonatal 6-hydroxydopamine-lesioned rats is eliminated by neonatal 5,7- dihydroxytryptamine treatment. J Pharmacol Exp Ther 268: 231-237.

253

Buck K, Voehringer P & Ferger B (2010). The a2 adrenoceptor antagonist idazoxan alleviates L- DOPA-induced dyskinesia be reduction of striatal dopamine levels: an in vivo microdialysis study in 6-hydroxydopamine-lesioned rats. J Neurochem, 112: 444-452. Buck K & Ferger B (2008). Intrastriatal inhibition of aromatic amino acid decarboxylase prevents l- DOPA-induced dyskinesia: a bilateral reverse in vivo microdialysis study in 6- hydroxydopamine lesioned rats. Neurobiol Dis, 29: 210-220. doi:10.1016/j.nbd.2007.08.010 Budygin EA, Brodie MS, Sotnikova TD, Mateo Y, John CE, Cyr M, Gainetdinov RR, Jones SR, (2004). Dissociation of rewarding and dopamine transporter-mediated properties of amphetamine. Proceedings of the National Academy of Sciences of the United States of America 101: 7781-7786. Burgdorf J, Panksepp J & Moskal JR (2011). Frequency-modulated 50 kHz ultrasonic vocalizations: a tool for uncovering the molecular substrates of positive affect. Neurosci Biobehav Rev, 35: 1831-1836. doi:10.1016/j.neubiorev.2010.11.011 Burgdorf J, Kroes RA, Moskal JR, Pfaus JG, Brudzynski SM & Panksepp J (2008). Ultrasonic vocalizations of rats (Rattus norvegicus) during mating, play, and aggression: Behavioral concomitants, relationship to reward, and self-administration of playback. J Comp Psychol, 122: 357-367. doi:10.1037/a0012889 Burgdorf J, Wood PL, Kroes RA, Moskal JR & Panksepp J (2007). Neurobiology of 50-kHz ultrasonic vocalizations in rats: electrode mapping, lesion, and pharmacology studies. Behav Brain Res, 182: 274-283. doi:10.1016/j.bbr.2007.03.010 Burgdorf J, Knutson B, Panksepp J, Shippenberg TS (2001). Evaluation of rat ultrasonic vocalizations as predictors of the conditioned aversive effects of drugs. Psychopharmacology 155: 35-42. Burgess C, O'Donohoe AO & Gill M (2000). Agony and ecstasy: a review of MDMA effects and toxicity. Eur Psychiatry, 15: 287-294. Calon F, Rajput AH, Hornykiewicz O, Bedard PJ & Di Paolo T (2003). Levodopa-induced motor complications are associated with alterations of glutamate receptors in Parkinson's disease. Neurobiol Dis 14: 404-416. Caplan JP, Epstein LA, Quinn DK, Stevens JR & Stern TA (2007). Neuropsychiatric effects of prescription drug abuse. Neuropsychology review 17: 363-380. Carlsson A, Lindqvist M & Magnusson T (1957). 3,4-dihydroxyphenylalanine and 5-hydroxytryptophan as reserpine antagonists. Nature 180: 1200. Camargos ST, Dornas LO, Momeni P, Lees A, Hardy J, Singleton A, Cardoso F (2009). Familial Parkinsonism and early onset Parkinson's disease in a Brazilian movement disorders clinic: phenotypic characterization and frequency of SNCA, PRKN, PINK1, and LRRK2 mutations. Mov Disord 24: 662-666. Cai R, Crane E & Poneleit K (2010). Emergency department visits involving nonmedical use of selected prescription drugs - Unites States, 2004-2008. Morb Mort Week Rep, 59: 705-709. Cain SW, Ko CH, Chalmers JA & Ralph MR (2004). Time of day modulation of conditioned place preference in rats, depends on the strain of rat used. Neurbiol Learn Mem, 81: 217-220. Cao X, Liang L, Hadcock JR, Iredale PA, Griffith DA, Menniti FS, Factor S, Greenamyre JT & Papa SM (2007). Blockade of cannabinoid type 1 receptors augments the antiparkinsonian action of levodopa without affecting dyskinesias in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine- treated rhesus monkeys. J Pharmacol Exp Ther, 323: 318-326. Carr GD & White NM (1983). Conditioned place preference from intra-accumbins but not intra- caudate amphetamine injections. Life Sci, 33: 2551-2557. Carta M & Bjorklund A (2018). The serotonergic system in L-DOPA-induced dyskinesia: pre- clinical evidence and clinical perspective. J Neur Trans, 125: 1195-1202.

254

Carta M, Carlsson T, Kirik D & Bjorklund A (2007). Dopamine released from 5-HT terminals is the cause of L-DOPA-induced dyskinesia in parkinsonian rats. Brain, 130: 1819-1833. doi:10.1093/brain/awm082 Carta M, Lindgren HS, Lundblad M, Stancampiano R, Fadda F & Cenci MA (2006). Role of striatal L-DOPA in the production of dyskinesia in 6-hydroxydopamine lesioned rats. J Neurochem, 96: 1718-1727. doi:10.1111/j.1471-4159.2006.03696.x Carter CJ & Pycock CJ (1979) The effects of 5,7-dihydroxytryptamine lesions of extrapyramidal and mesolimbic sites on spontaneous motor behaviour, and amphetamine-induced stereotypy. Naunyn Schmiedebergs Arch Pharmacol, 308: 51-54 Carter CJ & Pycock CJ (1978) Differential effects of central serotonin manipulation on hyperactive and stereotyped behaviour. Life Sci, 23: 953-960 Cash R, Ruberg M, Raisman R & Agid Y (1984). Adrenergic receptors in Parkinson's disease. Brain Res, 322: 269-275. Cassel JC, Cassel S, Galani R, Kelche C, Will B, Jarrard L (1998) Fimbria-fornix vs selective hippocampal lesions in rats: effects on locomotor activity and spatial learning and memory. Neurobiol Learn Mem, 69: 22-45. Castaneda E, Whishaw IQ & Robinson TE (1990). Changes in striatal dopamine neurotransmission assessed with microdialysis following recovery from a bilateral 6-OHDA lesion: variation as a function of lesion size. J Neurosci, 10: 1847-1854. Caudle WM, Colebrooke RE, Emson PC & Miller GW (2008) Altered vesicular dopamine storage in Parkinson's disease: a premature demise. Trends Neurosci, 31: 303-308. Cenci MA & Crossman AR (2018). Animal models of L-DOPA induced dyskinesia in Parkinson’s disease. Mov Disord, 33: 889-899. Cenci MA, Lee CS & Bjorklund A (1998). L-DOPA-induced dyskinesia in the rat is associated with striatal overexpression of prodynorphin- and glutamic acid decarboxylase mRNA. Eur J Neurosci, 10: 2694-2706. Cenci MA & Lindgren HS (2007). Advances in understanding L-DOPA-induced dyskinesia. Curr Opin Neurobiol 17: 665-671. Cenci MA & Lundblad M (2007). Ratings of L-DOPA-induced dyskinesia in the unilateral 6-OHDA lesion model of Parkinson's disease in rats and mice. Curr Protoc Neurosci, Chapter 9, Unit 9 25. doi:10.1002/0471142301.ns0925s41 Cenci MA, Whishaw IQ & Schallert T (2002). Animal models of neurological deficits: how relevant is the rat? Nat Rev, 3: 574-579. Cervo L & Samanin R (1995). Effects of dopaminergic and glutamatergic receptor antagonists on the acquisition and expression of cocaine conditioning place preference. Brain Res, 673: 242- 250. Chang J-W, Wachtel SR, Young D & Kang U-J (1999). Biochemical and anatomical characterization of forepaw adjusting steps in rat models of Parkinson's disease: studies on medial forebrain bundle and striatal lesions. Neurosci, 88: 617-628. Chapuis S, Ouchchane L, Metz O, Gerbaud L, Durif F (2005). Impact of the motor complications of Parkinson’s disease on quality of life. Mov Disord, 20: 224-230. Charney DS, Heninher GR & Redmon DE Jnr (1983). Yohimbine induced anxiety and increased noradrenergic function in humans: effects of diazepam and clonidine. Life Sci, 33: 19-29. Chase TN (1998). The significance of continuous dopaminergic stimulation in the treatment of Parkinson’s disease. Drugs 55: 1-9. Chase TN & Oh JD (2000). Striatal dopamine- and glutamate-mediated dysregulation in experimental parkinsonism. Trends Neurosci 23: S86-S91. Chen Q, Niu Y, Zhang R, Guo H, Gao Y, Li Y, Liu R (2010). The toxic influence of paraquat on hippocampus of mice: involvement of oxidative stress. Neurotoxicology, 31: 310-316.

255

Cheshire PA & Williams DR (2012). Serotonergic involvement in levodopa-induced dyskinesias in Parkinson's disease. J Clin Neurosci 19: 343-348. Christine CW, Bankiewicz KS, Van Larr AD, Richardson RM, Ravina B, Kells AP, Boot B, Martin AJ, Nutt J, Thompson ME, Larson PS et al. (2019). Magnetic resonance imaging-guided phase 1 trial of putaminal AADC gene therapy for Parkinson’s disease. Ann Neurol, 85: 704- 714. Ciancio D & Helinski M (2010). The use of shear wave velocity for assessing strength development in Fibre Reinforced Shotcrete. Shotcrete Elements of a System, 65-70. Cilia R, Akpalu A, Sarfo FS, Cham M, Amboni M, Cereda E, Fabbri M, Adjei P, Akassi J, Bonetti A & Pezzoli G (2014). The modern pre-levodopa era of Parkinson’s disease: insights into motor complications from sub-Saharn Africa. Brain, 137: 2731-2742. Ciliax BJ, Nash N, Heilman C, Sunahara R, Hartney A, Tiberi M, Rye DB, Caron MG, Niznik HB & Levey AI (2000). Dopamine D5 receptor immunolocalization in rat and monkey brain. Synapse, 37: 125-145. Clarke CE & Deane K (2001). Cabergolic versus bromocriptine for levodopa-induced complications in Parkinson's disease. Cochrane Database Syst Rev (1), 1-25. Cohen JY, Haesler S, Vong L, Lowell BB & Uchida N (2012). Neuron-type-specific signals for reward and punishment in the ventral tegmental area. Nature, 482: 85-90. Cole JC, Sumnall HR, O'Shea E & Marsden CA (2003). Effects of MDMA exposure on the conditioned place preference produced by other drugs of abuse. Psychopharmacology (Berl), 166: 383-390. doi:10.1007/s00213-002-1374-x Colpaert FC, Degryse AD & Van Craenendonck H (1991). Effects of an alpha2 antagonist in a 20- year-old java monkey with MPTP-induced parkinsonian signs. Brain Res, 26: 627-631. Colpaert FC (1987). Pharmacological characteristics of tremor, rigidity and hypokinesia induced by reserpine in rat. Neuropharmacology, 26: 1431-1440 Colzi A, Turner K & Lees AJ (1998). Continuous subcutaneous waking day apomorphine in the long term treatment of levodopa induced interdose dyskinesias in Parkinson's disease. J Neurol Neurosurg Psychiatry 64: 573-576. Commissaris RL, Palmer A, Neophytou S, Graham M, Beckett S & Marsden CA (2000). Acoustically elicited behaviours in Lister hooded and Wistar rats. Physiol Behav, 68: 521- 531. Compton WM & Volkow ND (2006). Abuse of prescription drugs and the risk of addiction. Drug Alcohol Depend, 83: Suppl 1 S4-7. doi:10.1016/j.drugalcdep.2005.10.020 Constantinescu R, Romer M, McDermott MP, Kamp C, Kieburtz K & Group C-P, I. o. t. P. S. (2007). Impact of pramipexole on the onset of levodopa-related dyskinesias. Mov Disord, 22: 1317-1319. doi:10.1002/mds.21292 Cooper JA, Sagar HJ, Doherty SM, Jordan N, Tidswell P & Sullivan EV (1992). Different effects of dopaminergic and anticholinergic therapies on cognitive and motor function in Parkinson's disease: a follow-up study of untreated patients. Brain, 115: 1701-1725. Cosford NDP, Tehrani L, Roppe J, Schweigher E, Smith ND, Anderson J, Bristow L, Brodkin J, Jiang X, MacDonald I, Rao S, Washburn M & Varney MA (2003). 3-[(2-Methyl-1,3-thiazol- 4-yl)ethynyl]-pyridine: a potent and highly selective metabotropic glutamate subtype 5 receptor antagonist with anxiolytic activity. J Med Chem, 46: 204-206. Cotzias GC, Papavasiliou PS & Gellene R (1969). Modification of parkinsonism - chronic treatment with L-DOPA. New Eng J Med 280: 337-345. Crabbe M, Van der Perran A, Weerasekera A, Himmelreich U, Baekelandt V, Van Laere K, Casteels C (2018). Altered mGluR5 binding potential and glutamine concentration in the 6-OHDA rat model of acute Parkinson’s disease and levodopa-induced dyskinesia. Neurobiol Aging, 61: 82-92.

256

Crabtree DM & Zhang J (2012). Genetically engineered mouse models of Parkinson's disease. Brain Res Bull, 88: 13-32. Crespi D, Mennini T & Gobbi M (1997). Carrier-dependent and Ca(2+)-dependent 5-HT and dopamine release induced by (+)-amphetamine, 3,4-methylendioxymethamphetamine, p- chloroamphetamine and (+)-fenfluramine. Br J Pharmacol, 121: 1735-1743. Crosby NJ, Deane K & Clarke CE (2003). Amantadine for dyskinesia in Parkinson's disease (Review). Cochrane Database of Systematic Reviews: 1-14. Curran HV & Travill RA (1997). Mood and cognitive effects of +3,4- methylenedioxymethamphetamine (MDMA, 'ecstasy'): a week-end 'high' followed by mid- week low. Addiction, 92: 821-831. Czyrak A, Mackowiak M, Chocyk A, Fijal K, Gadek-Michalska A & Wedzony K (2003). 8- OHDPAT-induced disruption of prepulse inhibition in rats is attenuated by prolonged corticosterone treatment. Neuropsychopharmacology 28: 1300-1310. Darmopil S, Martin AB, De Diego IR, Ares S, & Moratalla R (2009). Genetic inactivation of dopamine D1 but not D2 receptors inhibits L-DOPA-induced dyskinesia and histone activation. Biol Psychiatry 66: 603-613. Dauer W & Prezedborski S (2003). Parkinson's disease: mechansims and models. Neuron 39: 889- 909. Davidson C & Stamford JA (1995). Evidence that 5-hydroxytryptamine release in rat dorsal raphe nucleus is controlled by 5-HT1A , 5-HT1B and 5-HT1D autoreceptors British Journal of Pharmacology, 114: 1107-1109. Davis GC, Williams AC, Markey SP, Ebert MH, Caine ED, Reichert CM & Kopin IJ (1979). Chronic parkinsonism secondary to intravenous injection of meperidine analogues. Psychiatry Res, 1: 249-254. Dawson GR & Tricklebank MD (1995). Use of the elevated plus maze in the search for novel anxiolytic agents. TiPS, 16: 33-36. Dekundy A, Lundblad M, Danysz W & Cecilia MA (2007). Modulation of L-DOPA-induced abnormal involuntary movements by clinically tested compounds: further validation of the rat dyskinesia model. Behav Brain Res, 179: 76-89. doi:10.1016/j.bbr.2007.01.013 Del Dotto P, Pavese N, Gambaccini G, Barnardini S, Metman LV, Chase TN & Bonucelli U (2001). Intravenous amantadine improves levodopa-induced dyskinesias: An acute double-blind placebo-controlled study. Mov Disord 16: 515-520. Delfino MA, Stefano AV, Ferrario JE, Taravini IRE, Murray MG & Gershanik OS (2004). Behavioral sensitization to different dopamine agonists in a parkinsonian rodent model of drug-induced dyskinesias. Behav Brain Res, 152: 297-306. DeLong MR & Wichmann T (2007). Circuits and circuit disorders of the basal ganglia. Arch Neurol 64: 20-24. DePaulis A, Keay KA & Bandler R (1992). Longitudinal neuronal organization of defensive reactions in the midbrain periaqueductal gray region of the rat. Ex Brain Res, 90: 307-318. DeWit H, Pierri J & Johanson CE (1989). Reinforcing and subjective effects of diazepam in nondrug-abusing volunteers. Pharmacol Biochem Behav, 33: 205-213 Dietz D, Wang H & Kabbaj M (2007). Corticosterone fails to produce conditioned place preference or conditioned place aversion in rats. Behavioural brain research, 181: 287-91. Di Marzo V, Hill MP, Bisogna T, Crossman A.R & Brotchie JM (2000). Enhanced levels of endogenous cannabinoids in the globus pallidus are associated with a reduction in movement in an animal model of Parkinson's disease. FASEB J, 14: 1432-1438. Ding Y, Won L, Britt JP, Lim SAI, McGehee DS, Kang UJ & Palmiter RD (2011). Enhanced striatal cholinergic neuronal activity mediates L-DOPA-induced dyskinesia in parkinsonian mice. Proc Nat Acad Sci USA, 108: 840-845.

257

Dingledine R, Borges K, Bowei D & Traynelis SF (1999) The glutamate receptor ion channels. Pharmacol Rev, 51: 7-61. Doble A (1996). The pharmacology and mechansim of action of riluzole. Neurology, 47: S233-S241 Drake JD, Kibuuka LN, Dimitrov KD & Pollack AE (2013) Abnormal involuntary movement (AIM) expression following D2 dopamine agonist challenge is determined by the nature of prior dopamine receptor stimulation (priming) in 6-hydroxydopamine lesioned rats. Pharmacol Biochem Behav, 105: 26-33. Drevets WC, Gautier C, Price JC, Kupfer DJ, Kinahan PE, Grace AA, Price JL & Mathis CA (2001). Amphetamine-induced dopamine release in human ventral striatum correlates with euphoria. Psychiatry, 49: 81-96. Droge W (2002). Free radicals in the physiological control of cell function. Physiol Rev, 82: 47-95. Du J, Zhang S, Jin X & Yan G (2011). The Design of Ultrasonic Distance Measurement System Based of SOPC. 2011 Third International Conference on Intelligent Human-Machine Systems and Cybernetics, 166-168. doi:10.1109/ihmsc.2011.111 Dunn F & Fry WJ (1959). Ultrasonicn absorption microscope. Journal of the Acoustical Society of America, 31: 632. Durif F, Vidailhet M, Bonnet AM, Blin J & Agid Y (1995). Levodopa-induced dyskinesias are improved by fluoxetine. Neurology, 45: 1855-1858. Dziewczapolski G, Menalled LB, Garcia MC, Mora MA, Gershanik OS, & Rubinstein M (1998). Opposite roles of D1 and D5 dopamine receptors in locomotion revealed by selective antisense oligonucleotides. NeuroReport 9: 1-5. Ehret G & Haack B (1984). Motivation and arousal influence soundinduced maternal pup-retrieving behavior in lactating house mice. Z Tierpsychol, 65: 25–39. Elkurd MT, Bahroo LB & Pahwa R (2018). The role of extended-release amantadine for the treatment of dyskinesia in Parkinson’s disease patients. Neurodegener Dis Manag, 8: Elliot PJ (1988). Place aversion induced by the substance P analogue, dimethyl-C7, is not state dependent: implication of substance P in aversion. Exp Brain Res, 73: 354-356. Elmer LW, Juncos JL, Singer C, Truong DD, Criswell SR, Parashos S, Felt L, Johnson R & Patni R (2018). Pooled analyses of Phase III studies of ADS-5102 (amantadine) extended-release capsules for dyskinesia in Parkinson’s disease. CNS Drugs, 32: 387-398. Erickson JD (2017). Functional identification of activity-regulated, high-affinity glutamine transport in hippocampal neurons inhibited by riluzole. J Neurochem, 142: 29-40. Erickson JD, Schafer MKH, Bonner TI, Eiden LE & Weihe E (1996). Distinct pharmacological properties and distribution in neurons and endocrine cells of two isoforms of the human vesicular monoamine transporter. Proc Natl Acad Sci U S A, 93: 5166-5171. Eskow KL, Dupre K, Barnum CJ, Dickinson SO, Park JY & Bishop C (2009). The role of the dorsal raphe nucleus in the development, expression, and treatment of L-DOPA-induced dyskinesia in hemiparkinsonian rats. Synapse, 63: 610-620. Eskow KL, Gupta V, Alam S, Park JY & Bishop C (2007). The partial 5-HT(1A) agonist buspirone reduces the expression and development of l-DOPA-induced dyskinesia in rats and improves l-DOPA efficacy. Pharmacol Biochem Behav, 87: 306-314. Fabbrini G, Brotchie JM, Grandas F, Nomoto M & Goetz CG (2007). Levodopa-induced dyskinesias. Mov Disord, 22: 1379-1389. Fahn S, Oakes D, Shoulson I, Kieburtz K, Rudolph A, Lang A, Olanow CW, Tanner C, Marek K & Fahn S (2004). Levodopa and the progression of Parkinson’s disease. N Engl J Med, 351: 2498-2508. Fahn S (2003). Description of Parkinson's disease as a clinical syndrome. Ann NY Acad Sci, 991: 1- 14. Fahn S (2000). The spectrum of levodopa-induced dyskinesias. Ann Neurol, 47: S2-9.

258

Fairbrother IS, Arbuthnott GW, Kelly JS & Butcher SP (1990). In vivo mechanisms underlying dopamine release from rat nigrostriatal terminals: II. Studies using potassium and tyramine. J Neurochem, 54: 1844-1851. Faria RR, Abilio VC, Grassl C, Chinen CC, Negrao LTR, de Castro JPMV, Fukushiro DF, Rodrigues MSD, Gomes PHZ, Registro S, de Carvalho RC, D'Almeida V, Silve RH, de Alencar Riberio R & Frussa-Filho R (2005). Beneficial effects of vitamin C and vitamin E on reserpine-induced oral dyskinesia in rats: Critical role of striatal catalase activity. Neuropharmacology, 48: 993-1001. Fernandes VS, Santos JR, Leao AH, Medeiros AM, Melo TG, Izidio GS, Cabral A, Riveiro RA, Abilio VC, Ribeiro AM & Silva, R. H. (2012). Repeated treatment with a low dose of reserpine as a progressive model of Parkinson's disease. Behav Brain Res, 231: 154-163. doi:10.1016/j.bbr.2012.03.008 Fernandes C & File SE (1996). The influence of open arm ledges and maze experience in the elevated plus-maze. Pharmacol Biochem Behav, 54: 31-40. Ferro MM, Bellissimo MI, Anselmo-Franci JA, Angellucci MEM, Canteras NS & Da Cunha C (2005). Comparison of bilaterally 6-OHDA- and MPTP-lesioned rats as models of the early phase of Parkinson's disease: Histological, neurochemical, motor and memory alterations. J Neurosci Meth, 148: 78-87. Fibiger HC, Lynch GS & Cooper HP (1971). A biphasic action of central cholinergic stimulation on behavioral arousal in the rat. Psychopharmacologia, 20: 366-82. Fields HL, Hjelmstad GO, Margolis EB & Nicola SM (2007). Ventral tegmental area neurons in learned appetitive behavior and positive reinforcement. Ann Rev Neurosci, 30: 289-316. File SE (1986). Aversive and appetitive properties of anxiogenic and anxiolytic agents. Behav Brain Res, 21: 189-194. Finberg JPM & Youdim MBH (2002). Pharmacological properties of the anti-Parkinson drug rasagiline; modification of endogenous brain amines, reserpine reversal, serotonergic and dopaminergic behaviours. Neuropharmacology, 43: 1110-1118. doi:10.1016/s0028- 3908(02)00216-2 Fischer BA & Boggs DL (2010). The role of antihistaminic effects in the misuse of quetiapine: a case report and review of the literature. Neurosci Biobehav Rev, 34: 555-558. Fisher A, Biggs, CS, Eradiri O, Starr MS (1999). Dual effects of L-3,4-dihydrophenylalanine on aromatic L-amino acid decarboxylase, dopamine release and motor stimulation in the reserpine-treated rat: evidence that behaviour is dopamine independent. Neurosci, 95: 97-111. Flaquer C, Torre I & Arrizabalaga A (2007). Comparison of sampling methods for inventory of bat communities. Journal of Mammology, 88: 526-533. Fleckenstein AE, Gibb JW & Hanson GR (2000). Differential effects of stimulants on monoaminergic transporters: Pharmacological consequences and implications for neurotoxicity. Eur J Pharmacol, 406: 1-13. Foltynie T, Cheeran B, Williams-Gray CH, Edwards MJ, Schneider SA, Weinberger D, Rothwell JC, Barker RA & Bhatia KP (2008). BDNF val66met influences time to onset of levodopa induced dyskinesia in Parkinson's disease. J Neurol Neursurg Psychiatry, 80: 141-144. Foord SM, Bonner TJ, Neubig RR, Rosser EM, Pin JP, Davenport AP, Spedding M & Harmar AJ (2005). International union of pharmacology.XLVI. G protein-coupled receptor list. Pharmacol Rev, 57: 279-288. Fox SH, Katzenschlager R, Lim S-Y, Barton B, De Bie, R, Seppi K, Coelho M & Sampaio C (2018). International Parkinson and movement disorder society evidence-based medicine review: Update on treatments for the motor symptoms of Parkinson’s disease. Mov Disord, 33: 1248- 1266. Fox SH, Chuang R & Brotchie JM (2009). Serotonin and Parkinson's disease: On movement, mood, and madness. Mov Disord, 24: 1255-1266. doi:10.1002/mds.22473

259

Fox SH, Henry B, Hill M, Crossman A & Brotchie JM (2001). Stimulation of cannabinoid receptors reduces levodopa-induced dyskinesia in the MPTP-lesioned nonhuman primate model of Parkinson's disease. Mov Disord, 17: 1180-1187. Fornaguera J & Schwarting RK (1999). Early behavioural changes after nigro-striatal system damage can serve as predictors of strial dopamine deplation. Prog Neuropsychopharmacol Biol Psychitry, 23: 1353-1368. Francardo V, Recchia A, Popovic N, Andersson D, Nissbrandt H & Cenci MA (2011). Impact of the lesion procedure on the profiles of motor impairment and molecular responsiveness to L- DOPA in the 6-hydroxydopamine mouse model of Parkinson's disease. Neurbiol Dis, 42: 327-340. Fritts ME, Mueller K & Morris L (1997). Amphetamine-induced locomotor stereotypy in rats is reduced by a D1 but not a D2 antagonist. Pharmacol Biochem Behav, 58: 1015-1019. Fudala PJ, Teoh KW & Iwamoto ET (1985). Pharmacologic characterization of nicotine-induced conditioned place preference. Pharmacol Biochem Behav, 22: 237-241. Gandy MN, McIldowie M, Lewis K, Wasik AM, Salomonczyk D, Wagg K, Millar SA, Tindiglia D, Hunt P, Johnston T, Thiele S, Nguyen B, Barnes NM, Brotchie JM, Martin-Iverson MT, Nash J, Gordon J & Piggott MJ (2010) Redesigning the designer drug ecstasy: non- psychoactive MDMA analogues exhibiting Burkitt's lymphoma cytotoxicity. Med Chem Commun, 1: 287-293. Gerdeman G & Lovinger DM (2000). CB1 cannabinoid receptor inhibits synaptic release of glutamate in rat dorsolateral striatum. J Neurophysiol, 85: 468-471. Gerdeman GL, Ronesi J & Lovinger DM (2002). Postsynaptic endocannabinoid release is critical to long-term depression in the striatum. Nature Neurosci, 5: 446-451. Gerdjikov TV, Ross GM & Beninger RJ (2004). Place preference induced by nucleus accumbens amphetamine is impaired by antagonists of ERK or p38 MAP kinases in rats. Behavioral neuroscience 118(4): 740-750. Gerfen CR, McGinty JF & Young WS III (1991). Dopamine differentially regulates dynorphin, substance P, and enkephalin expression in striatal neurons: in situ hybridization histochemical analysis. J Neurosci, 11: 1016-1031. Gerfen CR, Engber TM, Mahan LC, Susel Z, Chase TN, Mansma FJ Jnr & Sibley DR (1990). D1 and D2 dopamine receptor-regulated gene expression of striatonigral and striatopallidal neurons. Science, 250: 1429-1432. Gerlach M, Double K, Arzberger T, Leblhuber F, Tatschner T & Riederer P (2003). Dopamine receptor agonists in current clinical use: comparative dopamine receptor binding profiles defined in the human striatum. J Neural Trans, 110: 1119-1127. Gerlach J (1977). Relationship between tardive dyskinesia, L-DOPA-induced hyperkinesia and parkinsonism. Psychopharmacol, 51: 259-263. Geyer MA, Russo PV, Segal DS & Kuczenski R (1987). Effects of apomorphine and amphetamine on patterns of locomotor and investigatory behavior in rats. Pharmacology Biochemistry and Behavior, 28: 393-9. Gilson AM, Ryan KM, Joranson DE & Dahl JL (2004). A reassessment of trends in the medical use and abuse of opioid analgesics and implications for diversion control: 1997-2002. J Pain Symp Manag, 28: 176-188. Goetz CG, Laska E, Hicking C, Stat D, Damier P, Muller T, Nutt J, Olanow CW, Rascol O & Russ H (2008). Placebo influences on dyskinesia in Parkinson's disease. Mov Disord, 23: 700-707. Goetz CG, Damier P, Hicking C, Laska E, Muller T, Olanow CW, Rascol O & Russ H (2007). Sarizotan as a treatment for dyskinesias in Parkinson's disease: a double-blind placebo- controlled trial. Mov Disord, 22: 179-186. Golbe LI (1991). Young-onset Parkinson's disease: A clinical review. Neurology, 41: 168-173.

260

Goldstein JM, Barnett A & Malick JB (1975). The evaluation of anti-parkinson drugs on reserpine- induced rigidity in rats. Eur J Pharmacol, 33: 183-188. Gomez-Mancilla B & Bedard PJ (1991). Effect of D1 and D2 agonists and antagonists on dyskinesia produced by L-DOPA in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated monkeys. J Pharmacol Exp Ther, 259: 409-413. Gomez-Ramirez J, Johnston TH, Visanji NP, Fox SH & Brotchie JM (2006). Histamine H3 receptor agonists reduce L-dopa-induced chorea, but not dystonia, in the MPTP-lesioned nonhuman primate model of Parkinson's disease. Mov Disord, 21: 839-846. doi:10.1002/mds.20828 Goole J & Amighi K (2009). Levodopa delivery systems for the treatment of Parkinson's disease: an overview. Int J Pharm, 380: 1-15. Gottwald MD & Aminoff MJ (2011). Therapies for dopaminergic-induced dyskinesias in Parkinson disease. Ann Neurol, 69: 919-927. Graff-Guerrero A, Redden L, Abi-Saab W, Katz DA, Houle S, Barsoum P, Bhathena A, Palaparthy R, Saltarelli MD, Kapur, S. (2010). Blockade of [11C](+)-PHNO binding in human subjects by the dopamine D3 receptor antagonist ABT-925. Int J Neuropsychopharmacol, 13(3), 273- 287. doi:10.1017/S1461145709990642 Graham L, Yoon T, Lee H & Kim J (2009). Strain and sex differences in fear conditioning: 22 kHz ultrasonic vocalizations and freezing in rats. Psychology and Neuroscience, 2: 219-225. doi:10.3922/j.psns.2009.2.015 Graham WC, Clarke CE, Boyce S, Sambrook MA, Crossman AR & Woodruff GN (1990). Autoradiographic studies in animal models of hemi-parkinsonism reveal dopamine D2 but not D1 receptor supersensitivity. II. Unilateral intra-carotid infusion of MPTP in the monkey (Macaca fascicularis). Brain Res, 514: 103-110. Grandy DK, Zhang Y, Bouvier C, Zhou Q, Johnson RA, Allen L, Buck K, Bunzow JR, Salon J Civelli O (1991). Multiple human D5 dopamine receptor genes: A functional receptor and two pseudogenes. Proc Natl Acad Sci USA, 88: 9175-9179. Gregoire L, Morin N, Ouatarra B, Gasparini F, Bilbe G, Johns D, Vranesic I, Sahasranaman S, Gomez-Mancilla B & di Paolo T (2011). The acute antiparkinsonian and anktidyskinetic effect of AFQ056, a novel metabotropic glutamate receptor type 5 antagonist, in L-DOPA- treated parkinsonian monkeys. Parkinsonism Relat D, 17: 270-276. Gresack JE, Risbrough VB, Scott CN, Coste S, Stenzel-Poore M, Geyer MA & Powell SB (2010). Isolation rearing-induced deficits in contextual fear learning do not require CRF(2) receptors. Behavioural brain research, 209: 80-4. Grondin R, Doan VD, Gregoire L & Bedard PJ (1999). D1 receptor blockade improves L-dopa- induced dyskinesia but worsens parkinsonism in MPTP monkeys. Neurology, 52: 771-776. Gubellini P, Picconi B, Bari M, Battista N, Calabresi P, Centonze D, Bernadi G, Finazzi-Agro A & Maccarrone M (2002). Experimental parkinsonism alters endocannabinoid degradation: implications for striatal glutamatergic transmission. J Neurosci, 22: 6900-6907. Guigoni C, Aubert I, Li Q, Gurevich VV, Benovic JL, Ferry S, Mach U, Stark H, Leriche L., Hakansson K, Bioulac BH, Gross CE, Sokoloff P, Fisone G, Gurevich EV, Bloch B & Bezard, E. (2005). Pathogenesis of levodopa-induced dyskinesia: focus on D1 and D3 dopamine receptors. Parkinsonism Relat Disord, 11: Suppl 1, S25-29. doi:10.1016/j.parkreldis.2004.11.005 Hall H, Sedvall G, Magnusson O, Kopp J, Halldin C & Farde L (1994). Distribution of D1- and D2- like dopamine receptors, and dopamine and its metabolites in the human brain. Neuropsychopharmacology, 11: 245-256. Hallett PJ, Dunah AW, Ravenscroft P, Zhou S, Bezard E, Crossman AR, Brotchie JM & Standaert DG (2005). Alterations of striatal NMDA receptor subunits associated with the development of dyskinesia in the MPTP-lesioned primate model of Parkinson's disease. Neuropharmacology 48: 503-516.

261

Halliday GM, Blumberg PC, Cotton RGH, Blessing WW & Geffen LB (1990). Loss of brainstem serotonin- and substance P-containing neurons in Parkinson's disease. Brain Res, 510: 104- 107. Han DD & Gu HH (2006). Comparison of the monoamine transporters from human and mouse in their sensitivities to psychostimulant drugs. BMC Pharmacol, 6:6. Hamed A, Taracha E, Szyndler J, Krzascik P, Lehner M, Maciejak P, Skorzewska A & Plaznik A (2012). The effects of morphine and morphine conditioned context on 50 kHz ultrasonic vocalisation in rats. Behav Brain Res, 229: 447-450. doi:10.1016/j.bbr.2012.01.053 Handley SL & McBlane JW (1993). An assessment of the elevated X-maze for studying anxiety and anxiety-modulating drugs. JPM, 29: 129-138. Hansard MJ, Smith LA, Jackson MJ, Cheetham SC & Jenner P (2004). The monoamine reuptake inhibitor BTS 74398 fails to evoke established dyskinesia but does not synergise with levodopa in the MPTP-treated primates. Mov Disord, 19: 15-21. Hansard MJ, Smith LA, Jackson MJ, Cheetham SC & Jenner P (2002). Dopamine, but not norepinephrine or serotonin, reuptake inhibition reverses motor deficits in 1-methyl-4- phenyl-1,2,3,6-tetrahydropyridine-treated primates. J Pharmacol Exp THer, 303: 952-958. Hara M, Fukui R, Hieda E, Kuroiwa M, Bateup HS, Kano T, Greengard P and Nishi A (2010) Role of adrenoceptors in the regulation of dopamine/DARPP-32 signaling in neostriatal neurons. J Neurochem 113: 1046-1059. Harms AS, Barnum CJ, Ruhn KA, Varghese S, Trevino I, Blesch A & Tansey MG (2011). Delayed dominant-negative TNF gene therapy hals progressive loss of nigral dopaminergic neurons in a rat model of Parkinson’s disease. Mol Ther, 19: 46-52. Harrison AA & Markou A (2001). Serotonergic manipulations both potentiate and reduced brain stimulation reward in rats: Involvement of serotonin-1A receptors. J Pharmacol Exp Ther, 297: 316-325. Hashimoto H, Moritani N, Aoki-Komori S, Tanaka M & Saito TR (2004). Comparison of ultrasonic vocalizations emitted by rodent pups. Exp Anim, 53: 409-416. Hauser RA (2009). New considerations in the medical management of early Parkinson’s disease: Impact of recent clinical trials on treatment strategy. Parkinsonism Relat Disord, 1553: S17- S21. He J, Xu H, Yang Y, Zhang X & Li XM (2005). Chronic administration of quetiapine alleviates the anxiety-like behavioural changes induced by a neurotoxic regimen of dl-amphetamine in rats. Behav Brain Res, 160: 178-187. Heeringa MJ & Abercrombie ED (1995). Biochemistry of somatodendritic dopamine release in substantia nigra: an in vivo comparison with striatal dopamine release. J Neurochem, 65: 192-200. Hefti F, Melamed E & Wurtman RJ (1980). Partial lesions of the dopaminergic nigrostriatal system in rat brain: biochemical characterization. Brain Res, 195: 123-137. Heikkila RE, Shapiro BS & Duboisin RC (1981). The relationship between loss of dopamine nerve terminals, striatal [3H]spiroperidol binding and rotational behavior in unilaterally 6- hydroxydopamine-lesioned rats. Brain Res, 211: 285-292. Hely MA, Morris JG, Reid WG & Trafficante R (2005). Sydney Multicenter Study of Parkinson's disease: non-L-dopa-responsive problems dominate at 15 years. Mov Disord, 20: 190-199. doi:10.1002/mds.20324 Henry B, Fox SH, Peggs D, Crossman AR & Brotchie JM (1999). The a2-adrenergic receptor antagonist idazoxan reduces dyskinesias and enhances anti-parkinsonian actions of L-Dopa in the MPTP-lesioned primate model of Parkinson's disease. Mov Disord, 14: 744-753. Henry JA, Jeffreys KJ & Dawling S (1992). Toxicity and deaths from 3,4- methylenedioxymethamphetamine ("ecstasy"). Lancet, 340: 384-387.

262

Herkenham M, Lynn AB, Little MD, Johnson MR, Melvin LS, de Costa BR & Rice K (1990). Cannabinoid receptor localization in brain. Proc Natl Acad Sci USA, 87: 1932-1936. Herman BH, Vocci F & Bridge P (1995). The effects of NMDA receptor antagonists and nitric oxide synthase inhibitors on opioid tolerance and withdrawal. Neuropsychopharm, 13: 269-293. Herzig V, Capuani EM, Kovar KA & Schmidt WJ (2005). Effects of MPEP on expression of food-, MDMA- or amphetamine-conditioned place preference in rats. Addict Biol, 10: 243-249. doi:10.1080/13556210500223272 Hill MP, Girard F, Keywood C, Poli SM, Crossman AR, Ravenscroft P, Li Q, Bezard E & Mutel, V. (2010). ADX48621, a novel mGlu5 negative allosteric modulator alleviates L-DOPA- induced chorea and dystonia in the MPTP-lesioned macaque model of Parkinson's disease. Mov Dis, 25: S281-S282. Hiroi N & White NM (1991). The amphetamine conditioned place preference: differential involvement of dopamine receptor substypes and two dopaminergic terminal areas. Brain Res, 552: 141-152. Hjorth S (1992). Acute reserpine treatment increases rat brain serotonin synthesis via a nerve impulse-dependent mechanism. J Neurochem, 58: 772-775 Ho Y-J, Pawlak CR, Guo L & Schwarting RKW (2004). Acute and long-term consequences of single MDMA administration in relation to individual anxiety levels in the rat. Behavioural Brain Research, 149: 135-144. doi:10.1016/s0166-4328(03)00220-1 Hohmann AG & Herkenham M (2000). Localization of cannabinoid CB1 receptor mRNA in neuronal subpopulations of rat striatum: a double-label in situ hybridization study. Synapse, 37: 71-80. Holloway RG, Shoulson I, Fahn S, Kieburtz K, Lang A, Marek K, McDermitt M, Seibyl K, Weiner W, Musch B, Kamp C, Welsg M, Shinaman A, Pahwa R, Barclay L, Hubble J, LeWitt P, Miyasaki J, Suchowersky O, Stacy M, Russell DS, Ford B, Hammerstad J, Riley D, Standaert D, Wooten F, Factor S, Jankovic J, Atassi F, Kurlan R, Panisset M, Rajput A, Rodnitzky R, Shults C, Petsinger G, Waters C, Pfeiffer R, Biglan K, Borchert L, Montgomery A, Sutherland L, Weeks C, DeAngelis M, Sime E, Wood S, Pantella C, Harrigan M, Fussell B, Dillon S, Alexander-Brown B, Rainey P, Tennis M, Rost-Ruffner E, Brown D, Evans S, Berry D, Hall J, Shirley T, Dobson J, Fontaine D, Pfeiffer B, Brocht A, Bennett S, Daigneault S, Hodgeman K, O’Connell C, Ross T, Richard K, & Watts A (2004). Pramipexole vs levodopa as initial treatment for Parkinson disease: a 4-year randomized controlled trial. Arch Neurol, 61: 1044-1053. Holloway RG (2000). Pramipexole vs levodopa as initial treatment for Parkinson disease. JAMA, 284: 1931-1938. Horan B, Smith M, Gardner EL, Lepore M & Ashby CR Jnr (1997). (-)-Nicotine produces conditions place preference in Lewis, but not Fischer 344, rats. Synapse, 26: 93-94. Hubert JP, Delumeau JC, Glowinski J, Premont J & Doble A (1994). Antagonism by riluzole of entry of calcium evoked by NMDA and veratridine in rat cultured granule cells: evidence for a dual mechanism of action. Br J Pharmacol, 113: 261-267. Hueter TF & Bolt RH (1951). An ultrasonic method for outlining the cerebral ventricles. Journal of the Acoustical Society of America, 23: 160-167. Hung SW, Adeli GM, Arenovich T, Fox SH & Lang AE (2010). Patient perception of dyskinesia in Parkinson’s disease. J Neurol Neurosurg Psychiatry, 81: 1112-1115. Huot P, Johnston TH, Lewis KD, Koprich JB, Reyes MG, Fox SH, Piggott MJ & Brotchie, J. M. (2014). UWA-121, a mixed dopamine and serotonin re-uptake inhibitor, enhances L-DOPA anti-parkinsonian action without worsening dyskinesia or psychosis-like behaviours in the MPTP-lesioned common marmoset. Neuropharmacology, 82: 76-87. doi:10.1016/j.neuropharm.2014.01.012

263

Huot P & Fox SH (2013). The serotonergic system in motor and non-motor manifestations of Parkinson's disease. Exp Brain Res, 230: 463-476. Huot P, Johnston TH, Koprich JB, Aman A, Fox SH & Brotchie JM (2012a). L-745,870 reduces L- DOPA-induced dyskinesia in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-lesioned macaque model of Parkinson's disease. J Pharmacol Exp Ther, 342: 576-585. doi:10.1124/jpet.112.195693 Huot P, Johnston TH, Gandy MN, Reyes MG, Fox SH, Piggott MJ & Brotchie JM (2012b). The monoamine re-uptake inhibitor UWA-101 improves motor fluctuations in the MPTP-lesioned common marmoset. PLoS One, 7: e45587 1-7. doi:10.1371/journal.pone.0045587.g001 Huot P, Johnston TH, Lewis KD, Koprich JB, Reyes MG, Fox SH, Piggott MJ & Brotchie JM (2011). Characterization of 3,4-methylenedioxymethamphetamine (MDMA) enantiomers in vitro and in the MPTP-lesioned primate: R-MDMA reduces severity of dyskinesia, whereas S-MDMA extends duration of ON-time. J Neurosci, 31: 7190-7198. doi:10.1523/JNEUROSCI.1171-11.2011 Huot P, Johnston TH, Lewis KD et al. (2010). Effects of R- and S-MDMA on L-DOPA_induced dyskinesia and duration of on-time in the MPTP-lesioned common marmoset. Soc Neurosci. Hwang D-Y, Fleming SM, Ardayfio P, Moran-Gates T, Kim H, Tarazi FI, Chesselet MF & Kim K-S (2005). 3,4-dihydroxyphenylalanine reverses the motor deficits in Pitx3-deficient Aphakia mice: behavioral characterization of a novel genetic model of Parkinson's disease. J Neurosci, 25: 2132-2137. Hwang DY, Ardayfio P, Kang UJ, Semina EV & Kim KS (2003). Selective loss of dopaminergic neurons in the substantia nigra of Pitx3-deficient aphakia mice. Brain Res Mol Brain Res, 114: 123-131. Iderberg H, Francardo V & Pioli EY (2012). Animal models of L-DOPA-induced dyskinesia: an update on the current options. Neuroscience, 211: 13-27. doi:10.1016/j.neuroscience.2012.03.023 Insel TR, Hill JL & Mayor RB (1986). Rat pup ultrasonic isolation calls: possible mediation by the benzodiazepine receptor complex. Pharmacol Biochem Behav, 24: 1263-1267. Iravani MM, Jackson MJ, Kuoppamaki M, Smith LA & Jenner P (2003). 3,4- methylenedioxymethamphetamine (Ecstasy) inhibits dyskinesia expression and normalizes motor activity in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated primates. J Neurosci, 23: 9107-9115. Ishibashi T & Ohno Y (2004). Antiparkinsonian actions of a selective 5-HT1A agonist, tandospirone, in rats. Biogenic Amines, 18: 3-6. Jackson-Lewis, V. & Przedborski, S. (2005). Protocol for the MPTP mouse model of Parkinson's disease. Nat Prot, 2, 141-151. Jahng JW, Houpt TA. MK801 increases feeding and decreases drinking in nondeprived, freely feeding rats. Pharmacology Biochemistry and Behavior, 2001; 68: 181-6. Jankovic J & Hunter C (2002). A double-blind, placebo-controlled and longitudinal study of riluzole in early Parkinson's disease. Parkinsonism Relat Disord, 8: 271-276. Jasinski DR, Martin WR & Haertzen CA (1967). The human pharmacology and abuse potential of n- aalylnoroxymorphone (naloxone). J Pharmacol Exp Therap, 157: 420-426. Jenner P (2008). Molecular mechanisms of L-DOPA-induced dyskinesia. Nat Rev Neurosci 9: 665- 677. Jennings JH, Sparta DR, Stamatakis AM, Ung RL, Pleil KE, Kash TL & Stuber GD (2013). Distinct extended amygdala circuits for divergent motivational states. Nature, 496: 224-230. Johansen CE & Uhlenhuth EH (1980). Drug preference and mood in humans: Diazepam. Pyschopharmacology, 71: 269-273. Johnston TH, Millar Z, Huot P, Wagg K, Thiele S, Salomonczyk D, Yong-Kee CJ, Gandy MN, McIldowie M, Lewis KD, Gomes-Ramirez J, Lee J, Fox SH, Martin-Iverson M, Nash JE,

264

Piggott MJ, Brotchie JM (2012). A novel MDMA analogue, UWA-101, that lacks psychoactivity and cytotoxicity, enhances L-DOPA benefit in parkinsonian primates. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 26: 2154-2163. Johnston LD, O’Malley PM, Bachman JG & Schulenburg JE (2011). Monitoring the Future national results on adolescent drug use: Overview of key findings, 2010. Bethesda: National Institute on Drug Abuse. Johnston TH, Fox SH, Mclldowie MJ, Piggott MJ & Brotchie JM (2010). Reduction of L-DOPA- induced dyskinesia by the selective metabotropic glutamate receptor 5 antagonist 3-[(2- methyl-1,3-thiazol-4-yl)ethynyl]pyridine in the 1-methyl-4-phenyl-1,2,3,6- tetrahydropyridine-lesioned macaque model of Parkinson's disease. J Pharmacol Exp Therap, 333: 865-873. Johnston TH, Lee J, Gomez-Ramirez J, Fox SH & Brotchie JM (2005). A simple rodent assay for the in vivo identification of agents with potential to reduce levodopa-induced dyskinesia in Parkinson's disease. Exp Neurol, 191: 243-250. doi:10.1016/j.expneurol.2004.10.002 Joranson DE, Ryan KM, Gilson AM & Dahl JL (2000). Trends in medical use and abuse of opioid analgesics. JAMA, 283: 1710-1714. Joyce EM & Iversen SD (1984). Dissociable effects of 6-OHDA-induced lesions of neostriatum on anorexia, locomotor activity and stereotypy: the role of behavioral competition. Psychopharm, 83: 363-366. Kaiser R, Hofer A, Grapengiesser A, Gasser T, Kupsch A, Roots I & Brockmoller J (2003). L- DOPA-induced adverse effects in PD and dopamine transporter gene polymorphism. Neurology, 60: 1750-1755. Kahn ZU, Gutierrez A, Martin R, Penafiel A, Rivera A & De La Calle A (2000). Dopamine D5 receptors of rat and human brain. Neuroscience, 100: 689-699. Kalivas PW, Duffer P & White SR (1998). MDMA elicits behavioral and neurochemical sensitizaton in rats. Neuropsychopharmacology, 18: 469-479. Kannari K, Kurahashi K, Tomiyama M, Maeda T, Arai A, Baba M, Suda T & Matsunaga M (2002). Tandospirone citrate, a selective 5-HT1A agonist, alleviates L-DOPA-induced dyskinesia in patient with Parkinson's disease. No To Shinkei [Brain and Nerve], 54: 133-137. Kannari K, Yamato H, Shen H, Tomiyama M, Suda T & Matsunaga M (2001). Activation of 5- HT1A but not 5-HT1B receptors attentuates an increase in extracellular dopamine derived from exogenously administered L-DOPA in the striatum with nigrostriatal denervation. J Neurochem, 76; 1346-1353. Katona I, Urban GM, Wallace M, Ledent C, Jung K-M, Piomella D, Mackie K & Freud TF (2006). Molecular composition of the endocannabinoid system at glutaminergic synapses. J Neurosci, 26: 5628-5637. Katz J, Nielsen KM & Soghomonian JJ (2005). Comparative effects of acute or chronic administration of levodopa to 6-hydroxydopamine-lesioned rats on the expression of glutamic acid decarboxylase in the neostriatum and GABAA receptors subunits in the substantia nigra, pars reticulata. Neuroscience, 132: 833-842. Katzenschlager R, Hughes A, Evans A, Manson AJ, Hoffman M, Swinn L, Watt H, Bhatia K, Qinn N & Lees AJ (2005). Continuous subcutaneous apomorphine therapy improves dyskinesias in Parkinson's disease: a prospective study using single-dose challenges. Mov Disord, 20: 151- 157. Kelsey JE, Harris O & Cassin J (2009). The CB(1) antagonist rimonabant is adjunctively therapeutic as well as monotherapeutic in an animal model of Parkinson's disease. Behav Brain Res, 203: 304-307. doi:10.1016/j.bbr.2009.04.035

265

Khlebtovsky A, Rigbi A, Melamed E, Ziv I, Steiner I, Gad A & Djaldetti J (2012). Patient and caregiver perceptions of the social impact of advanced Parkinson’s disease and dyskinesias. J Neural Transm, 119: 1367-1371. Kikusui T, Takeuchi Y & Mori Y (2001). Pharmacological manipulations of the extinction process of fear-induced ultrasonic vocalization in rats. J Vet Med Sci, 63: 591-595. Kim EJ, Kim ES, Covery E & Kim JJ (2010). Social transmission of fear in rats: the role of 22-kHz ultrasonic distress vocalization. PLoS One, 5: 1-8. doi:10.1371/journal.pone.0015077.g001 Kimura K, Sela S, Bauvier C, Grandy DK, & Sidhu A (1995). Differential coupling of D1 and D5 dopamine receptors to guanine nucleotide binding proteins in transfected GH4C1 rat somatomammotrophic cells. J Neurochem 64: 2118-2124. Kirkpatrick MG, Johanson C-E & de Wit H (2013). Personality and the acute subjective effects of d- amphetamine in humans. J Psychopharmacol, 27: 256-264. Kirshner N (1962). Uptake of Catecholamines by a Particulate Fraction of the Adrenal Medulla J Biol Chem, 237: 2311-2317. Klawans HL Jr & Weiner WJ (1974). Attempted use of haloperidol in the treatment of L-DOPA induced dyskinesias. J Neurol Neurosurg Psychiatry, 37: 427-430. Kleckner NW & Dingledine R (1998). Requirement for glycine activation of NMDA-receptors expressed in xenopus oocytes. Science, 241: 835-837. Klein C & Westenberger A (2012). Genetics of Parkinson's disease. Cold Spring Harb Perspect Med 2: 1-15. Kling-Petersen T, Ljung E & Svensson K (1995). Effects on locomotor activity after local application of D3 preferring compounds in discrete areas of the rat brain. J Neural Transm (GenSect), 102: 209-220. Knutson B, Burgdorf J & Panksepp J (2002). Ultrasonic vocalizations as indices of affective states in rats. Psychological Bulletin 128: 961-977. Knutson B, Burgdorf J & Panksepp J (1999). High-frequency ultrasonic vocalizations index conditioned pharmacological reward in rats. Physiol Behav, 66: 639-643. Kobylecki C, Cenci MA, Crossman AR & Ravenscroft P (2010). Calcium-permeable AMPA receptors are involved in the induction and expression of l-DOPA-induced dyskinesia in Parkinson's disease. J Neurochem, 114: 499-511. doi:10.1111/j.1471-4159.2010.06776.x Kofalvi A, Rodrigues RJ, Ledent C, Mackie K, Vizi ES, Cunha RA & Sperlagh B (2005). Involvement of cannabinoid receptors in the regulation of neurotransmitter release in the rodent striatum: a combined immunochemical and pharmacological analysis. J Neurosci, 25: 2874-2884 Kollins SH, MacDonald EK & Rush CR (2001). Assessing the abuse potential of methylphenidate in nonhuman and human subjects. Pharmacol Biochem Behav, 68: 611-627. Konitsiotis S, Blanchet PJ, Verhagen L, Lamers E & Chase TN (2000) AMPA receptor blockade improves levodopa-induced dyskinesia in MPTP monkeys. Neurology, 58: 1589-1595. Kordower JH, Emborg ME, Bloch J, Ma SY, Chu Y, Leventhal L, McBride J, Chen E-Y, Palfi S, Roitberg BZ, Brown WD, Holden JE, Pyzalski R, Talery MD, Carvey P, Ling Z, Trono D, Hantraye P, Deglon N & Aebischer P (2000) Neurodegeneration prevented by lentiviral vector delivery of GDNF in primate models of Parkinson's disease. Science, 290: 767-773. Kraebel KS, Brasser SM, Campbell JO, Spear LP & Spear NE (2002). Developmental differences in temporal patterns and potentiation of isolation-induced ultrasonic vocalizations: influence of temperature variables. Dev Psychobiol, 40: 147-159. Kretzschmar M, Brust P, Zessin J, Cumming P, Bergmann R & Johannsen B (2003). Autoradiographic imaging of the serotonin transporter in the brain of rats and pigs using S- ([18F]fluoromethyl)-(+)-McN5652. Eur Neuropsychopharm, 13: 387-397. Kumar N, Van Gerpen JA, Bower JH & Ahlskog JE (2005). Levodopa-dyskinesia incidence by age of Parkinson’s disease onset. Mov Disord, 20: 342-366.

266

Lacroix L, Broersen LM, Feldon J & Weiner I (2000). Effects of local infusions of dopaminergic drugs into the medial prefrontal cortex of rats on latent inhibition, prepulse inhibition and amphetamine induced activity. Behav Brain Res, 107: 111-121. Lamb RJ, Preston KL, Schindler CW, Meisch RA, Davis F, Katz JL, Henningfield JE & Goldberg SR (1991). The reinforcing and subjective effects of morphine in post-addicts: a dose- response study. J Pharmacol Exp Ther, 259: 1165-1173. Lammel S, Lim BK, Ran C, Huang KW, Betley MJ, Tye K, Deisseroth K & Malenka RC (2012). Input-specific control of reward and aversion in the ventral tegmental area. Nature, 491: 212- 217. Lane E & Dunnett S (2008). Animal models of Parkinson’s disease and L-dopa induced dyskinesia: how close are we to the clinic? Psychopharmacology, 199: 303-312. Langston JW (2017). The MPTP story. J Parkinsons Dis, 7: S11-S19. Langston JW, Ballard P, Tetrud JW & Irwin I (1983). Chronic parkinsonism in humans due to a product of meperidine-analog synthesis. Science, 219: 979-980 Laube B, Kuhse J & Betz H (1999) Evidence for a tetrameric structure of recombinant NMDA receptors. J Neurosci, 18: 2954-2961. Laube B, Hirai H, Sturgess M, Betz H & Kuhse J (1997) Molecular determinents of agonist discrimination by NMDA receptor subunits: analysis of the glutamate binding site on the NR2B subunit. Neuron, 18: 493-503. Laviola G & Adriana W (1998). Evaluation of unconditioned novelty-seeking and d-amphetamine- conditioned motivation in mice. Pharmacol Biochem Behav, 59: 1011-1020. Lee J, di Marzo V & Brotchie JM (2006). A role for vanilloid receptor 1 (TRPV1) and endocannabinoid signalling in the regulation of spontaneous and L-DOPA induced locomotion in normal and reserpine-treated rats. Neuropharmacology, 51: 557-565. Lessenger JE & Feinberg SD (2008). Abuse of prescription and over-the-counter medications. J Am Board Fam Med, 21: 45-54. doi:10.3122/jabfm.2008.01.070071 Lewitt PA, Hauser RA, Lu M, Nicholas AP, Weiner W, Coppard N, Leinonen M & Savola JM (2012). Randomized clinical trial of fipamezole for dyskinesia in Parkinson disease (FJORD study). Neurology, 79: 163-169. doi:10.1212/WNL.0b013e31825f0451 Li SJ, Jiang HK, Stachowiak MS, Hudson PM, Owyang V, Nanry K, Tilson HA & Hong JS (1990). Influence of nigrostriatal dopaminergic tone on the biosynthesis of dynorphine and enkephalin in rat striatum. Mol Brain Res, 8: 219-225. Liao RM (2008). Development of conditioned place preference induced by intra-accumbens infusion of amphetamine is attenuated by co-infusion of dopamine D1 and D2 receptor antagonists. Pharmacology Biochem Behav, 89: 367-373. Lindenbach D, Ostock CY, Eskow Jaunarajs KL, Dupre KB, Barnum CJ, Bhide N & Bishop C (2011). Behavioral and cellular modulation of L-DOPA-induced dyskinesia by B- adrenoceptor blockade in the 6-hydroxydopamine-lesioned rat. J Pharmacol Exp Ther, 337: 755-765. Lindgren HS, Andersson DR, Lagerkvist S, Nissbrandt H & Cenci MA (2010). L-DOPA-induced dopamine efflux in the striatum and the substantia nigra in a rat model of Parkinson's disease: temporal and quantitative relationship to the expression of dyskinesia. J Neurochem, 112: 1465-1476. doi:10.1111/j.1471-4159.2009.06556.x Litvin Y, Blanchard DC & Blanchard RJ (2007). Rat 22kHz ultrasonic vocalizations as alarm cries. Behav Brain Res, 182: 166-172. doi:10.1016/j.bbr.2006.11.038 Lloyd KG, Davidson L & Hornykiewicz O (1975). The neurochemistry of Parkinson's disease: effect of L-DOPA therapy. J Pharmacol Exp Ther, 195: 453-464. Lohr KM, Bernstein AI, Stout KA, Dunn AR, Lazo CR, Alter SP, Wang M, Li Y, Fan X, Hess EJ, Yi H, Vecchio LM, Goldstein DS, Guillot TS, Salahpour A & Miller GW (2014). Increased vesicular monoamine transporter enhances dopamine release and opposes Parkinson disease-

267

related neurodegeneration in vivo. Proc Natl Acad Sci U S A, 111: 9977-9982. doi:10.1073/pnas.1402134111 Lopez A, Munoz A, Guerra MJ & Labanderia-Garcia JL (2001). Mechanisms of the effects of exogenous levodopa on the dopamine-denervated striatum. Neurosci, 103: 639-651. Lucas G & Spampinato U (2000). Role of striatal serotonin 2A and serotonin 2C receptor substypes in the control of in vivo dopamine outflow in the rat striatum. J Neurochem, 74: 693-701. Luginger E, Wenning GK, Bosch S, & Poew W (2000). Beneficial effects of amantadine on L- DOPA-induced dyskinesias in Parkinson's disease. Movement disorders, 15: 873-878. Luk KC, Rymar VV, van den Munckhof P, Nicolau S, Steriade C, Bifsha P, Drouin J & Sadikot AF (2013). The transcription factor Pitx3 is expressed selectively in midbrain dopaminergic neurons susceptible to neurodegenerative stress. J Neurochem, 125: 932-943. Lukas M & Wohr M (2015). Endogenous vasopressin, innate anxiety, and the emission of pro-social 50-kHz ultrasonic vacolizations during social play behavior in juvenile rats. Psychoneuroendocrinology, 56: 35-33. doi:0.1016/j.psyneuen.2015.03.005 Lundblad M, Usiello A, Carta M, Hakansson K, Fisone G & Cenci MA (2005). Pharmacological validation of a mouse model of l-DOPA-induced dyskinesia. Exp Neurol, 194: 66-75. doi:10.1016/j.expneurol.2005.02.002 Lundblad M, Picconi B, Lindgren H & Cenci MA (2004). A model of L-DOPA-induced dyskinesia in 6-hydroxydopamine lesioned mice: relation to motor and cellular parameters of nigrostriatal function. Neurolbiol Dis, 16: 110-123. Lundblad M, Andersson M, Winkler C, Kirik D, Wierup N, & Cenci MA (2002). Pharmacological validation of behavioural measures of akinesia and dyskinesia in a rat model of Parkinson's disease. Eur J Neurosci 15: 120-132. Luquin MR, Scipioni O, Vaamonde J, Gershanik O & Obeso JA (1992). Levodopa-induced dyskinesias in Parkinson's disease: clinical and pharmacological classification. Mov Disord, 7: 117-124. Lynch WJ, Nicholson KL, Dance ME, Morgan RW & Foley PL (2010). Animal models of substance abuse and addiction: Implications for science, animal welfare, and society. Comp Med, 60: 177-188. Ma ST, Maier EY, Ahrens AM, Schallert T & Duvauchelle CL (2010). Repeated intravenous cocaine experience: development and escalation of pre-drug anticipatory 50-kHz ultrasonic vocalizations in rats. Behavioural brain research 212: 109-114. Mackie K (2005). Distribution of cannabinoid receptors in the central and peripheral nervous system. In: Pertwee R.G. (eds) Cannabinoids. Handbook Exp Pharmacol, 168: 299-325. Maeda T, Nagata K, Yoshida Y & Kannari K (2005). Serotonergic hyperinnervation into the dopaminergic denervated striatum compensates for dopamine conversion from exogenously administered L-DOPA. Brain Res, 1045: 230-233. Maggio R, Scarselli M, Novi F, Millan MJ & Corsini GU (2003). Potent activation of dopamine D3/D2 heterodimers by the antiparkinsonian agents, S32504, pramipexole and ropinirole. Journal of Neurochemistry, 87: 631-641. doi:10.1046/j.1471-4159.2003.02038.x Mahler, SV, Moorman DE, Feltenstein MA, Cox BM, Ogburn KB, Bachar M, McGonigal JT, Ghee SM & See RE (2013). A rodent “self-report” measure of methamphetamine craving? Rat ultrasonic vocalizations during methamphetamine self-administration, extinction, and reinstatement. Behav Brain Res, 236: 78-89. Mallo T, Matrov D, Herm L, Koiv K, Eller M, Rinken A & Harro J (2007). Tickling-induced 50-kHz ultrasonic vocalization is individually stable and predicts behaviour in tests of anxiety and depression in rats. Behav Brain Res, 184: 57-71. doi:10.1016/j.bbr.2007.06.015 Manson AJ, Turner K & Lees AJ (2002). Apomorphine monotherapy in the treatment of refractory motor complications of Parkinson's disease: long-term follow-up study of 64 patients. Mov Disord, 17: 1235-1241.

268

Manson AJ, Brown P, O’Sullivan JD, Asselman P, Buckwell D & Lees AJ (2000). An ambulatory dyskinesia monitor. J Neurol Neurosurg Psychiatry, 68: 196-201. Marongiu R, Arango-Lievano M, Francardo V, Morgenstern P, Zhang X, Cenci MA, Svenningsson P, Greengard P & Kaplitt MG (2016). Gene therapy blockade of dorsal striatal p11 improves motor function and dyskinesia in parkinsonian mice. PNAS, 113: 1423-1428. Martin-Iverson MT, Iversen SD & Stahl SM (1988a). Long-term motor stimulant effects of (+)-4- propyl-9-hydroxynapthoxazine (PHNO), a dopamine D-2 receptor antagonist: interactions with a dopamine D-1 receptor antagonist and agonist. Eur J Pharmacol, 149: 1-2. Martin-Iverson MT & Dourish CT (1988b). Role of dopamine D-1 and D-2 receptor substypes in mediating dopamine agonist effects on food consumption in rats. Psychopharm, 96: 370-374. Martin-Iverson MT, Wilkie D & Fibiger HC (1987). Effects of haloperidol and d-amphetamine on perceived quantity of foods and tones. Psychopharm, 93: 374-381. Martin-Iverson MT, Ortmann R & Fibiger HC (1985). Place preference conditioning with methylphenidite and nomifensine. Brain Res, 332: 59-67. Martin-Iverson MT, Leclere JF & Fibiger HC (1983). Cholinergic-dopaminergic interactions and the mechanisms of action of antidepressants. Eur J Pharmacol, 94: 193-201. Martin-Iverson MT, Pisa M, Chan E & Fibiger HC (1982). Enhanced neophobia but normal plasma corticosterone levels in rats with dorsal noradrenergic bundle lesions. Pharmacol Biochem Behav, 17: 639-643. Mayeux R (2003). Epidemiology of neurodegeneration. Annu Rev Neurosci, 26: 81-104. doi:10.1146/annurev.neuro.26.043002.094919 McGuire P & Fahy T (1991). Chronic paranoid psychosis after misuse of MDMA ("ecstasy"). BMJ, 302: 697. McLelland AE, Winkler CE, Piggott MJ & Martin-Iverson MT (2019). Influence of antidyskinetic compounds on L-DOPA-induced locomotion and behavior in reserpinized rats: Manuscript submitted for publication. McLelland AE, Winkler CE & Martin-Iverson MT (2015). A simple and effective method for building inexpensive infrared equipment used to monitor animal locomotion. J Neurosci Methods, 243: 1-7. doi:10.1016/j.jneumeth.2015.01.006 McLelland AE, Martin-Iverson MT & Beninger RJ (2014). The effect of quetiapine (Seroquel) on conditioned place preference and elevated plus maze tests in rats when administered alone and in combination with (+)-amphetamine. Psychopharmacology, 231, 4349-4359. Meador-Woodruff JH, Damask SP, Wang J, Haroutunian V, Davis KL & Watson SJ (1996). Dopamine receptor mRNA expression in human striatum and neocortex. Neuropsychopharmacology, 15: 17-29. Mechan AO, Moran PM, Elliott M, Young AJ, Joseph MH & Green R (2002). A study of the effect of a single neurotoxic dose of 3,4-methylenedioxymethamphetamine (MDMA; "ecstasy") on the subsequent long-term behaviour of rats in the plus maze and open field. Psychopharmacology, 159: 167-175. Meco G, Fabrizio E, Alessandri A, Vanacore N & Bonifati V (1998). Risperidone in levodopa induced dyskinesias. J Neurol Neurosurg Psychiatry, 64: 135-142. Meissner W, Ravenscroft P, Reese R, Harnack D, Morgenstern R, Kupsch A, Klitgaard H, Bioulac B, Gross CE, Bezard E & Boraud, T. (2006). Increased slow oscillatory activity in substantia nigra pars reticulata triggers abnormal involuntary movements in the 6-OHDA-lesioned rat in the presence of excessive extracellular striatal dopamine. Neurobiol Dis, 22: 586-598. doi:10.1016/j.nbd.2006.01.009 Mela F, Marti M, Dekundy A, Danysz W, Morari M & Cenci MA (2007). Antagonism of metabotropic glutamate receptor type 5 attenuates L-DOPA-induced dyskinesia and its molecular and neurochemical correlates in a rat model of Parkinson's disease. J Neurochem, 101: 483-497.

269

Merims D, Ziv I, Djaldetti R & Melamed E (1999). Riluzole for levodopa-induced dyskinesias in advanced Parkinson's disease. The Lancet, 353: 1764. Metman LV, Del Dotto P, LePoole K, Konitsiotis S, Fang J & Chase TN (1999). Amantadine for levodopa-induced dyskinesias: A 1-year follow-up year. Arch Neurol, 56: 1383-1386. Metman LV, Del Dotto P, van der Munckhof P, Fang J, Mouradian MM & Chase TN (1998). Amantadine as treatment for dyskinesias and motor fluctuations in Parkinson's disease. Neurology, 50: 1323-1326. Metz GA, Tse A, Ballermann M, Smith LK & Fouad K (2005). The unilateral 6-OHDA rat model of Parkinson’s disease revisited: an electromyographic and behavioural analysis. Eur J Neurosci 22: 735-744. Metz GA & Whishaw IQ (2002) Cortical and subcortical lesions impair skilled walking in the ladder rung walking test: a new task to evaluate fore- and hindlimb stepping, placing, and co- ordination. J Neurosci Meth, 115: 169-179. Meyer PJ, Ma ST & Robinson TE (2012). A cocaine cue is more preferred and evokes more frequency-modulated 50-kHz ultrasonic vocalizations in rats prone to attribute incentive salience to a food cue. Psychopharmacol, 219: 999-1009. Mignat C, Wille U & Zeigler A (1995). Affinity profiles of morphine, codeine, dihydrocodeine and their glucuronides at opioid receptor subtypes. Life Sci, 56: 793-799. Mignon LJ & Wolf WA (2005). 8-hydroxy-2-(di-N-propylamino)tetralin reduces striatal glutamate in an animal model of Parkinson's disease NeuroReport, 16: 699-703. Milekic MH, Brown SD, Castellini C & Alberini CM (2006). Persistent disruption of an established morphine conditioned place preference. The Journal of neuroscience : the official journal of the Society for Neuroscience 26: 3010-3020. Mills R, Revell S, Bahr D, Williams H, Johnson A & Friedman JH (2008). A double-blind, placebo- controlled, dose-escalation trial of pimavanserin in Parkinson's disease and psychosis. Mov Disord, 23: S221-S222. Missale C, Nash SR, Robinson SW, Jaber M & Caron MG (1998). Dopamine receptors: From structure to function. Physiol Rev, 78: 189-225. Mitamura H, Arai N, Mitsunaga Y, Yokota T, Takeuchi H, Tsuzkai T & Itani M (2005). Directed movements and diel burrow fidelity patterns of red tilefish Branchiostegus japonicus determined using ultrasonic telemetry. Fisheries Sci, 71: 491-498. Mizuno T, Schmauss C & Rayport S (2007) Distinct roles of presynaptic dopamine receptors in the differential modulation of the intrinsic synapses of medium-spiny neurons in the nucleus accumbens. BMC Neurosci 8; 8. Montel S, Bonnet AM & Bungener C (2009). Quality of life in relation to mood, coping strategies, and dyskinesia in Parkinson’s disease. J Geriat Psychiat Neurol, 22: 95-102. Montgomery KC (1958). The relation between fear induced by novel stimulation and exploratory behavior. J Comp Physiol Psychol, 48: 254-260. Monville C, Torres EM & Dunnett SB (2005). Validation of the L-dopa-induced dyskinesia in the 6- OHDA model and evaluation of the effects of selective dopamine receptor agonists and antagonists. Brain Res Bull, 68: 16-23. doi:10.1016/j.brainresbull.2004.10.011 Morgese MG, Cassano T, Cuomo V & Giuffrida A (2007). Anti-dyskinetic effects of cannabinoids in a rat model of Parkinson's disease: role of CB1 and TRPV1 receptors. Exp Neurol, 208: 110-119. Morin N, Gregoire L, Gomez-Mancilla B, Gasparini F & di Paolo T (2010). Effect of the metabotropic glutamate receptor type 5 antagonists MPEP and MTEP in parkinsonian monkeys. Neuropharm, 58: 981-986. Morley KC, Gallate JE, Hunt GE, Mallet PE & McGregor IS (2001). Increased anxiety and impaired memory in rats 3 months after administration of 3,4-methylenedioxymethamphetamine ("Ecstasy"). Eur J Pharmacol, 433: 91-99.

270

Morley KC & McGregor IS (2000). (+)-3,4-Methylenedioxymethamphetamine (MDMA, 'Ecstasy') increases social interaction in rats. Eur J Pharmacol, 408: 41-49. Mucha RF & Iversen SD (1984). Reinforcing properties of morphine and naloxone revealed by conditioned place preferences: a procedural examination. Psychopharmacol, 82: 241-247. Mucha RF, van der Kooy D, O’Shaughnessy M & Bucenieks P (1982) Drug reinforcement studied by the use of place conditioning in rat. Brain Res, 243: 91-105. Mueller D & Stewart J (2000). Cocaine-induced conditioned place preference: reinstatement by priming injections of cocaine after extinction. Behav Brain Res, 115: 39-47. Muller T, Olanow CW, Nutt J, Hicking C, Laska E & Russ S (2006). The PADDY-2 study: the evaluation of sarizotan for treatment-associated dyskinesia in Parkinson's disease patients. Mov Disord, 21: S15. Munoz A, Carlsson T, Tronci E, Kirik D, Bjorkland A & Carta M (2009). Serotonin neuron- dependent and -independent reduction of dyskinesia by 5-HT1A and 5-HT1B receptor agonists in the rat Parkinson model. Exp Neurol, 219: 298-307. Munoz A, Li Q, Gardoni F, Marcello E, Qin C, Carlsson T, Kirik D, di Luca M, Bjorkland A, Bezard E & Carta M (2008). Combined 5-HT1A and 5-HT1B receptor agonists for the treatment of L-DOPA-induced dyskinesia. Brain, 131: 3380-3394. Nadjar A, Gerfen CR & Bezard E (2009). Priming for l-dopa-induced dyskinesia in Parkinson's disease: a feature inherent to the treatment or the disease? Prog Neurobiol, 87: 1-9. doi:10.1016/j.pneurobio.2008.09.013 Naito H & Tonoue T (1987). Sex difference in ultrasound distress call by rat pups. Behav Brain Res, 25: 13-21. Nambu A (2004). A new dynamic model of the cortico-basal ganglia loop. Prog Brain Res, 143: 461- 466. Nambu A, Tachibana Y & Chiken S (2015). Cause of parkinsonian symptoms: Firing rate, firing pattern or dynamic activity changes? Basal Ganglia, 5: 1-6. doi:10.1016/j.baga.2014.11.001 Nambu A, Kaneda K, Tokuno H & Takada M (2000). Abnormal pallidal activity evoked by cortical stimulation in the parkinsonian monkey. Soc Neurosci Abstr, 26: 960. Nash JF & Brodkin J (1991). Microdialysis studies on 3,4-methylenedioxymethamphetamine- induced dopamine release: effect of dopamine uptake inhibitors. J Pharmacol Exp Ther, 259: 820-825. Natusch C & Schwarting RK (2010). Using bedding in a test environment critically affects 50-kHz ultrasonic vocalizations in laboratory rats. Pharmacol Biochem Behav, 96: 251-259. doi:10.1016/j.pbb.2010.05.013 Nazarian A, Russo SJ, Festa ED, Kraish M & Quinones-Jenab V (2004). The role of D1 and D2 receptors in the cocaine conditioned place preference of male and female rats. Brain research bulletin 63: 295-299. Neuman RS & Rahman S (1996). Ca2+ mobilizing agents mimic serotonin 5-HT2A facilitation of N- methyl-D-aspartate depolarization. Behav Brain Res, 73: 273-275. Newman-Tancredi A, Varney MA & McCreary AC (2018). Effects of the serotonin 5-HT1A receptor biased agonists, F13714 and F15599, on striatal neurotransmitter levels following L- DOPA administration in hemi-pkarinosnian rats. Neurochem Res, 43: 1035-1046. NIH (2017) www.clinicaltrails.gov. [Online] [Accessed: December 14 2017]. Nicholas AP (2007). Levodopa-induced hyperactivity in mice treated with 1-methyl-4-phenyl- 1,2,3,6-tetrahydropyridine. Mov Disord, 22: 99-104. Nilsson D, Hansson L-E, Johansson K, Nystrom C, Paalzow L & Aquilonius SM (1998). Long-term intraduodenal infusion of a water based levodopa-carbidopa dispersion in very advanced Parkinson's disease. Acta Neurol Scand, 97: 175-183.

271

Nisenbaum LK, Crowley WR & Kltai ST (1996). Partial striatal dopamine depletion differentially affects striatal substance P and enkephaline messenger RNA expression. Brain Res Mot Brain Res, 37: 209-216. Nisson D, Nyholm D & Aquilonius SM (2001). Duodenal levodopa infusion in Parkinson's disease: Long-term experience. Acta Neurol Scand, 104: 343-348. Nomikos GG & Spyraki C (1988). Effects of ritanserin on the rewarding properties of d- amphetamine, morphine and diazepam revealed by conditioned place preferece in rats. Pharmacol Biochem Behav, 30: 853-858. de Noordhout AM & Delwaide PJ (1986). Open pilot trial of ritanserin in parkinsonism. Clin Neuropharm, 9: 480-484. Noyes K, Dick AW, Holloway RG & Parkinson Study G (2006). Pramipexole versus levodopa in patients with early Parkinson's disease: effect on generic and disease-specific quality of life. Value Health, 9: 28-38. Nunes I, Tovmasain LT, Silva RM, Burke RE & Goff SP (2003). Pitx3 is required for development of substantia nigra dopaminergic neurons. PNAS, 100: 4245-4250. Nutt JG, Carter JH & Sexton GJ (2004). The dopamine transporter: importantance in Parkinson’s disease. Ann Neurol, 55: 766-773. Nutt JG (2003). Long-term l-DOPA therapy: challenges to our understanding and for the care of people with Parkinson's disease. Exp Neurol, 184: 9-13. doi:10.1016/s0014-4886(03)00304-2 Nutt JG (1990). Levodopa-induced dyskinesia: review, observations, and speculations. Neurology 40: 340-345. Nyholm D, Nilsson Remahl AIM, Dizdar N, Constantinescu R, Holmberg B, Jansson R, Aquilonius S-M & Askmark H (2005). Duodenal levodopa infusion monotherapy vs oral polypharmacy in advanced Parkinson disease. Neurology, 64: 216-223. O'Brien CP (2005). Benzodiazepine use, abuse and dependence. J Clin Psychiatry, 66: 28-33. O’Carrol AM, Fowler CJ, Phillip JP, Tobia I & Tipton KF (1983). The deamination of dopamine by human brain monoamine oxidase. Specificity for the two forms in seven brain regions. Naunyn Schmiedebergs Arch Pharmacol, 322: 198-207. O’Hearn E, Battaglia, De SouzaEB, Kuhar MJ & Molliver ME (1988). Methylenedioxyamphetamine (MDA) and methylenedioxymethamphetamine (MDMA) cause selective ablation fo serotonergic axon terminals in forebrain: immunocytochemical evidence for neurotoxicity. J Neurosci, 8: 2788-2803. Oberling P, Rocha B, Di Scala G & Sandner G (1993). Evidence for state-dependent retrieval in conditioned place aversion. Behav Neural Biol, 60: 27-32. Obeso JA, Marin C, Rodriguez-Oroz C, Blesa J, Benitez-Temino B, Mena-Segovia J, Rodrigues M & Olanow CW (2008). The basal ganglia in Parkinson's disease: current concepts and unexplained observations. Ann Neurol, 64 Suppl 2: S30-46. Obeso JA, Rodriguez-Oroz MC, Rodriguez M, Lanciego JL, Artieda J, Gonzalo N & Olanow CW (2000). Pathophysiology of the basal ganglia in Parkinson's disease. Trends Neurosci, 23: S8- S19. Oertel W, Eggert K, Pahwa R, Tanner CM, Hauser RA, Trankwalder C, Ehret R, Azulay JP, Isaacson S, Felt L & Stempien MJ (2017). Randomized, placebo-controlled trial of ADS- 5102 (amantadine) extended-release capsules for levodopa-induced dyskinesia in Parkinson’s disease (EASE LID 3). Mov Disord, 32: 1701-1709. Oertel WH, Wolters E, Sampaio C, Gimenez-Roldan S, Bergamasco B, Dujardin M, Grosset DG, Arnold G, Leenders KL, Hundermer H-P, Lledo A, Wood, A, Frewer P & Schwarz J (2006). Pergolide versus levodopa monotherapy in early Parkinson's disease patients: The PELMOPET study. Mov Disord, 21: 343-353. Oertel WH & Quinn NP (1997). Parkinson's disease: Drug therapy. Baillieres Clin Neurol, 6: 89- 108.

272

Oh JD, Russell D, Vaughan CL & Chase TN (1998). Enhanced tyrosine phosphorylation of striatal NMDA receptor subunits: effect of dopaminergic denervation and L-DOPA administration. Brain Res, 813: 150-159. Ohi K, Mikuni M & Takahashi K (1989). Stress adaptation and hypersensitivity in 5-HT neuronal systems after repeated foot shock. Pharmacol Biochem Behav, 34: 603-608. Olanow CW, Kieburtz K, Odin P, Espay AJ, Standaert DG, Fernandez HH, Vanagunas A, Othman AA, Widnell KL, Robieson WZ, Pritchett Y, Chatamra K, Benesh J, Lenz RA, Antonini A, LCIG Horizon Study Group (2014). Continuous intrajejunal infusion of levodopa-carbidopa intestinal gel for patients with advanced Parkinson's disease: a randomised, controlled, double-blind, double-dummy study. The Lancet Neurology, 13: 141-149. Olanow CW, Hauser RA, Gauger L, Malapira T, Koller W, Hubble J, Bushenbark K, Lilienfeld D & Esterlitz J (1995). The effect of deprenyl and levodopa on the progression of Parkinson's disease. Ann Neurol, 38: 771-777. Orman JS & Keating GM (2009). Spotlight on buprenorphine/naloxone in the treatment of opioid dependence. CNS Drugs, 23: 899-902. Ostock CY, Dupre KB, Jaunarajs KL, Walters H, George J, Krolewski D, Walker PD & Bishop C (2011). Role of the primary motor cortex in L-Dopa-induced dyskinesia and its modulation by 5-HT1A receptor stimulation. Neuropharmacology, 61: 753-760. Ouatarra B, Gregoire L, Morissette M, Gasparini F, Vranesic I, Bilbe G, Johns DR, Rajput A, Hornykiewicz O, Rajput AH, Gomez-Mancilla B & di Paolo T (2011). Metabotropic glutamate receptor type 5 in levodopa-induced motor complications. Neurobiol Aging, 32: 1286-1295. Overton DA (1978). Basic mechanisms of state-dependent learning. Psychopharmacol Bull, 14: 67- 68. Padovan-Neto FE, Patterson S, Voelkner NMF, Altwal F, Beverley, JA, West AR & Steiner H (2019). Selective regulation of 5-HT1B serotonin receptor expression in the striatum by dopamine depletion and repeated L-DOPA treatment: relationship to L-DOPA-induced dyskinesias. Mol Neurobiol, 1-16. Pahwa R, Tanner CM, Hauser RA, Isaacson SH, Nausieda PA, Truong DD, Agarwal P, Hull KL, Lyons KE, Johnson R & Stempien MJ (2017). ADS-5102 (Amantadine) extended-release capsules for levodopa-induced dyskinesia in Parkinson disease (EASE LID study): A randomized clinical trial. JAMA Neurol, 74: 941-949. Pahwa R, Tanner CM, Hauser RA, Sethi K, Isaacson S, Truong D, Struck L, Ruby AE, McClure NL, Went GT & Stempien MJ (2015). Amantadine extended release for levodopa-induced dyskinesia in Parkinson’s disease (EASED Study). Mov Disord, 30: 788-795. Panksepp J & Burgdorf J (2000). 50-kHz chirping (laughter?) in response to conditioned and unconditioned tickle-induced reward in rats: effects of social housing and genetic variables. Behav Brain Res, 115: 25-38. Papa SM & Chase TN (1996). Levodopa-induced dyskinesias improved by a glutamate antagonist in parkinsonian monkeys. Ann Neurol, 39: 574-578. Parkes JD, Zilkha KJ, Marsden P, Baxter RCH & Knill-Jones RP (1970). Amantadine dosage in treatment of Parkinson's disease. Lancet, 295: 1130-1133. Parkin SG, Gregory RP, Scott R, Bain P, Silburn P, Hall B, Boyle R, Joint C & Aziz TZ (2002). Unilateral and bilateral pallidotomy for idiopathic Parkinson's disease: a case series of 115 patients. Mov Disord, 17: 682-692. Parkinson Study Group (2009). Long-term effect of initiating pramipexole vs levodopa in early Parkinson disease. Arch Neurol, 66: 563-570. Parkinson Study Group (2000). Pramipexole vs levodopa as initial treatment for Parkinson disease: A randomized controlled trial. JAMA, 284: 1931-1938.

273

Parkinson Study Group (1993). Effects of tocopherol and deprenyl on the progression of disability in early Parkinson disease. NEJM, 328: 176-183. Parkinson Study Group (1989). Effect of deprenyl on the progression of disability in early Parkinson disease. NEJM, 321: 1364-1371. Parrott AC & Lasky J (1998). Ecstasy (MDMA) effects upon mood and cognition: before, during and after a Saturday night dance. Psychopharmacology (Berl), 139: 261-268. Paoletti P & Neyton J (2007). NMDA receptor subunits: function and pharmacology. Curr Opin Pharmacol, 7: 39-47. Pavon N, Martin AB, Mendialdua A & Moratalla R (2005). ERK phosphorylation and FosB expression are associated with L-DOPA-induced dyskinesia in hemiparkinsonian mice. Biol Psychiatry, 59: 64-74. Peaston MJT & Bianchine JR (1970). Metabolic studies and clinical observations during L-DOPA treatment of Parkinson's disease. Brit Med J, 1: 400-403. Pechevis M, Clarke CE, Vieregge P, Khoshnood B, Deschaseaux-Voinet C, Berdeaux G & Ziegler M (2005). Effects of dyskinesias in Parkinson’s disease on quality of life and health-related costs: a prospective European study. Eur J Neurol, 12: 956-963. Pehek EA, Nocjar C, Roth BL, Byrd TA & Mabrouk OS (2006). Evidence for the preferential involvement of 5-HT2A serotonin receptors in stress- and drug-induced dopamine release in the rat medial prefrontal cortex. Neuropsychopharm, 31: 265-277. Pellow S, Chopin P, File SE & Briley M (1985). Validation of open : closed arm entries in an elevated plus-maze as a measure of anxiety in the rat. J Neurosci Methods, 14: 149-167. Perry EK, McKeith I, Thompson P, Marshall E, Kerwin J, Jabeen S, Edwardson JA, Ince P, Blessed G & Irving D (1991). Topography, extent, and clinical relevance of neurochemical deficits in dementia of Lewy body type, Parkinson's disease, and Alzheimer's disease. Ann NY Acad Sci 640: 197-202. Peyro-Saint-Paul HDF, Pollak P, Bonnet AM, Payen I, Vidailhet M, Ple'tan Y & Agid Y (1997). Short term oral administration of idazoxan in mild stable parkinsonian patients treated with L-DOPA. J Pharmacol Clin Ther, 1: 172. Pierce GW & Griffin DR (1938). Experimental determination of supersonic notes emitted by bats. J Mammal, 19:154. Piercey MF, Hoffmann WE, Smith MW & Hyslop DK (1996). Inhibition of dopamine neuron firing by pramipexole, a dopamine D3 receptor-preferring agonist: comparison to other dopamine receptor agonists. Eur J Pharmacol, 312: 35-44. Pohorecky LA (2008). Psychosocial stress and chronic ethanol ingestion in male rats: effects on elevated plus maze behavior and ultrasonic vocalizations. Physiol Behav, 94: 432-447. doi:10.1016/j.physbeh.2008.02.010 Pompeiano M, Palacios JM & Mengod G (1994). Distribution of the serotonin 5-HT2 receptor family mRNAs: comparison between 5-HT2A and 5-HT2C receptors. Mol Brain Res, 23: 163-178. Portfors CV (2007). Types and functions of ultrasonic vocalizations in laboratory rats and mice. JAALAS, 46: 28-34. Porter CC, Totaro JA & Stone CA (1963). Effect of 6-hydroxydopamine and some other compounds on the concentration of norepinephrine in the hearts of mice. J Pharm Exp Ther, 140: 308-316 Prescott IA, Dostrovsky JO, Moro E, Hodaie M, Lozano AM & Hutchison WD (2009). Levodopa enhances synaptic plasticity in the substantia nigra pars reticulata of Parkinson's disease patients. Brain, 132: 309-318. Preston KL, Jasinski DR & Testa M (1991). Abuse potential and pharmacological comparison of tramadol and morphine. Drug Alc Dep, 27: 7-17. Privette TH & Terrian DM (1995). Kappa opioid agonists produce anxiolytic-like behaviour on the elevated plus maze. Psychopharmacol, 118: 444-450.

274

Proft J, Faraji J, Robbins JC, Zucchi FCR, Zhao X, Metz GA & Braun JEA (2011). Identification of bilateral changes in TID1 expression in the 6-OHDA rat model of Parkinson’s disease. Plos One, 6: 1-10. Pulvirenti L & Koob GF (1994). Dopamine receptor agonists, partial agonists and psychostimulant addiction. Trends Pharmacol Sci, 15: 374-379. Quik M, Cox H, Parameswaran N, O'Leary K, Langston JW & Di Monte D (2007). Nicotine reduces levodopa-induced dyskinesias in lesioned monkeys. Ann Neurol, 62: 588-596. Quik M, Police S, Langston JW & Di Monte DA (2002). Increases in striatal preproenkephalin gene expression are associated with nigrostriatal damage but not L-DOPA-induced dyskinesias in the squirrel monkey. Neurosci, 113: 213-220. Quisenberry AJ, Prisinzano TE & Baker LE (2013). Modafinil alone and in combination with low dose amphetamine does not establish conditioned place preference in male Sprague-Dawley rats. Exp Clin Psychpharm, 21: 252-258. Rascol O, Fitzer-Attas CJ, Hauser R, Jankovic J, Lang A, Langston JW, Melamed E, Poewe W, Stocchi F, Tolosa E, Eyal E, Weiss YM & Olanow CW (2011). A double-blind, delayed-start trial of rasagiline in Parkinson's disease (the ADAGIO study): prespecified and post-hoc analyses of the need for additional therapies, changes in UPDRS scores, and non-motor outcomes. Lancet Neurol, 10: 415-423. Rascol O, Poewe W, Lees A, Aristin M, Salin L, Juhel N, Waldhauser, L & Schindler T (2008). Tesofensine (NS 2330), a monoamine reuptake inhibitor , in patients with advanced Parkinson disease and motor fluctuactions. Arch Neurol, 65: 577-583. Rascol O, Brooks DJ, Korczyn AD, De Deyn PP, Clarke CE, Lang AE, et al. (2006). Development of dyskinesias in a 5-year trial of ropinirole and L-dopa. Mov Disord 21: 1844-1850. Rascol O, Brooks DJ, Melamed E, Oertel W, Poewe W, Stocchi F, et al. (2005). Rasagiline as an adjunct to levodopa in patients with Parkinson's disease and motor fluctuations (LARGO, Lasting effect in Adjunct therapy with Rasagiline Given Once daily, study): a randomised, double-blind, parallel-group trial. The Lancet 365: 947-954. Rascol, O., Payoux, P., Ory, F., Ferreira, J. J., Brefel-Courbon, C., & Montastruc, J. L. (2003). Limitations of current Parkinson's disease therapy. Ann Neurol, 53 Suppl 3, S3-12; discussion S12-15. doi:10.1002/ana.10513 Rascol O, Nutt JG, Blin O, Goetz CG, Trugman JM, Soubrouillard C, et al. (2001a). Induction by dopamine D1 receptor agonist ABT-431 of dyskinesia similar to levodopa in patients with Parkinson's disease. Arch Neurol 58: 249-254. Rascol, O., Amulf, I., Peyro-Saint-Paul, H., Brefel-Courbon, C., Vidailhet, M., Thalams, C., Bonnet, A.M., Descombes, S., Bejjani, B., Fabre, N., Mantastuc, J.J. & Agrid, Y. (2001b). Idazoxan, an alpha-2 antagonist, and L-DOPA-induced dyskinesias in patients with Parkinson's disease. Mov Disord, 16, 708-713. Rascol O, Brooks DJ, Korczyn AD, De Deyn PP, Clarke CE, & Lang AE (2000). A five-year study of the incidence of dyskinesia in patients with early Parkinson's disease who were treated with ropinirole or levodopa. 056 Study Group. New Eng J Med, 342: 1484-1491. Rezak M (2007). Current pharmacotherapeutic treatment options in Parkinson's disease. Dis Mon, 53: 214-222. doi:10.1016/j.disamonth.2007.05.002 Riachi, NJ, Behmand RA & Harik SI (1991). Correlation of MPTP neurotoxicity in vivo with oxidation of MPTP by the brain and blood-brain barrier in vitro in five rat strains. Brain Res, 555: 19-24. Riahi G, Morissette M, Parent M & Di Paolo T (2011). Brain 5-HT(2A) receptors in MPTP monkeys and levodopa-induced dyskinesias. Eur J Neurosci, 33: 1823-1831. doi:10.1111/j.1460- 9568.2011.07675.x

275

Ricaurte GA, Forno LS, MA Wilson et al. (1988) (+)3,4-methylenedioxymethamphetamine selectively damages central serotonergic neurons in nonhuman primates. J Am Med Ass, 260: 51-55. Riederer P & Youdim MBH (1986). Monoamine oxidase activity and monoamine metabolism in brain of Parkinson patients treated with l-deprenyl. J Neurochem, 46: 1359-1365. Riley EP, Shapiro NR & Lochry EA (1979). Nose-poking and head-dipping behaviors in rats prenatally exposed to alcohol. Pharmacology Biochemistry and Behavior, 11: 513-9. Rinne UK, Bracco F, Chouza C, Dupont E, Gershanik O, Marti Masso JF, Montastruc JL & Marsden CD (1998). Early treatment of Parkinson's disease with cabergoline delays the onset of motor complications. Drugs, 55: 23-30. Riobo NA, Schopper FJ, Boveris AD, Cadenas E & Poderoso JJ (2002). The reaction of nitric oxide with 6-hydroxydopamine: implications for Parkinson's disease. Free Radic Biol Med, 32: 115-121. Rivera A, Cuellar B, Giron FJ, Grandy DK, de La Calle A & Moratalla R (2002). Dopamine D4 receptors are heterogeneously distributed in the striosomes/matrix compartments of the striatum. J Neurochem, 80: 219-229. Robelet S, Melon C, Guillet B, Salin P & Kerkerian-Le Goff L (2004). Chronic L-DOPA treatment increases extracellular glutamate levels and GLT1 expression in the basal ganglia in a rat model of Parkinson's disease. Eur J Neurosci, 20: 1255-1266. Roberst C (2006). ACP-103, a 5-HT2A receptor inverse agonist. Curr Opin Investig Drugs, 7: 653- 660. Robertson HA (1992). Synergistic interactions of D1- and D2-selective dopamine agonists in animal models for Parkinson's disease: Sites of action and implications for the pathogenesis of dyskinesias. Can J Neurol Sci, 19: 147-152. Rodgers RJ & Dalvi A (1997). Anxiety, defence and the elevated plus maze. Neurosci Biobehav Rev, 21: 801-810. Rogalski A (2012). History of infrared detectors. Opto-Electronics Review, 20: 279-308. Rothman RB, Baumann MH, Dersch CM et al. (2001). Amphetamine-type central nervous system stimulants release norepinephrine more potently than they release dopamine and serotonin, Synapse, 39: 32-41. Rozenbroek K & Rothstein WC (2011). Medical and nonmedical users of prescription drugs among college students. J Am Coll Health, 59: 358-363. Rudnick G & Wall SC (1992). The molecular mechanisms of "ecstasy" [3,4-methylenedioxymethamphetamine (MDMA)]: Serotonin transporters are targets for MDMA-induced serotonin release. Proc Natl Acad Sci USA, 89: 1817-1821. Ruzicka E, Ladure P, Roth J, Jech R, Mecir P & Peyro-Saint-Paul H (1997). Efficacy and safety of efaroxan, an alpha-2 adrenoceptor antagonist, in Parkinson's disease: An oral short-term study. XIIth International Symposium on Parkinson's disease. Rylander D, Recchia A, Mela F, Dekundy A, Danysz W & Cenci MA (2009). Pharmacological modulation of glutamate transmission in a rat model of L-DOPA-induced dyskinesia: effects on motor behavior and striatal nuclear signaling. J Pharmacol Exp Therap, 330: 227-235. Salamone J & Baskin P (1996). Vacuous jaw movements induced by acute reserpine and low-dose apomorphine: possible model of parkinsonian tremor. Pharmacol Biochem Behav, 53: 179- 183. Samadi P, Gregoire L, Morisette M, Calon F, Tahar AH, Dridi M, Belanger N, Meltzer LT, Bedard PJ & di Paolo T (2008). mGluR5 metabotropic glutamate receptors and dyskinesias in MPTP monkeys. Neurbiol Aging, 29: 1040-1051. Samarra FI, Deecke VB, Vinding K, Rasmussen MH, Swift RJ & Miller PJ (2010). Killer whales (Orcinus orca) produce ultrasonic whistles. J Acoust Soc Am, 128: EL205-210. doi:10.1121/1.3462235

276

Savola J-M, Hill M, Engstrom M, Merivuori H, Wurster S, McGuire SG, Fox SH, Crossman AR & Brotchie JM (2003). Fipamezole (JP-1730) is a potent a2 adrenergic receptor antagonist that reduces levodopa-induced dyskinesia in the MPTP-lesioned primate model of Parkinson's disease. Mov Disord, 18: 872-883. Scatton B, Javoy-Agid F, Rouquier L, Dubois B & Agid Y (1983). Reduction of cortical dopamine, noradrenaline, serotonin and their metabolites in Parkinson's disease. Brain Res, 275: 321- 328. Scattoni ML, Crawley J & Ricceri L (2009). Ultrasonic vocalizations: A tool for behavioural phenotyping of mouse models of neurodevelopmental disorders. Neurosci Biobehav Rev, 33: 508-515. Schallert T & Hall S (1988). ‘Disengage’ sensorimotor deficit following apparent recovery from unilateral dopamine depletion. Behav Brain Res, 30: 15-24. Schenk S, Gittings D, Johnstone M & Daniela E (2003). Development, maintenance and temporal pattern of self-administration maintained by ecstasy (MDMA) in rats. Psychopharmacology (Berl), 169: 21-27. doi:10.1007/s00213-003-1407-0 Schenk S, Hely L, Lake B, Daniela E, Gittings D & Mash DC (2007). MDMA self-administration in rats: acquisition, progressive ratio responding and serotonin transporter binding. Eur J Neurosci, 26: 3229-3236. doi:10.1111/j.1460-9568.2007.05932.x Scherman D (1986). Dihydrotetrabenazine binding and monoamine uptake in mouse brain regions. J Neurochem, 47: 331-339. Schneider JS & Kovelowski CJ (1990). Chronic exposure to low doses of MPTP. I. Cognitive deficits in motor asymptomatic monkeys. Brain Res, 519: 122-128. Schneider JS (1990). Chronic exposure to low doses of MPTP. II. Neurochemical and pathological consequences in cognitively-impaired, motor asymptomatic monkeys. Brain Res, 534: 25-36. Scholz B, Svensson M, Alm H, Skold K, Faith M, Kultima K, Guigoni C, Doudnikoff E, Li Q, Crossman AR, Bezard E & Andren PE (2008). Strial proteomic analysis suggests that first L- DOPA dose equates to chronic exposure. Plos One, 3: e1589- Schrag A & Quinn N (2000). Dyskinesias and motor fluctuations in Parkinson’s disease. A community-based study. Brain, 123: 2297-2305. Schulteis G, Yackey M, Risbrough V & Koob GF (1998). Anxiogenic-like effects of spontaneous and naloxone-precipitated opiated withdrawal in the elevated plus maze. Pharmacol Biochem Behav 60: 727-731. Schmidt N & Ferger B (2001). Neurochemical findings in the MPTP model of Parkinson's disease. J Neural Transm, 108: 1263-1282. Segal DS & Kuczenski R (1987). Individual differences in responsiveness to single and repeated amphetamine administration: behavioural characteristis and neurochemical correlates. J Pharmacol Exp Ther, 242: 917-926. Segovia G, Mora F, Crossman AR & Brotchie JM (2003). Effects of CB1 cannabinoid receptor modulating compounds on the hyperkinesia induced by high-dose levodopa in the reserpine- treated rat model of Parkinson's disease. Mov Disord, 18: 138-149. Sharma JC, Bachmann CG & Linazasoro G (2010). Classifying risk factors for dyskinesia in Parkinson's disease. Parkinsonism Rel Disord, 16: 490-497. Shoblock JR, Welty N, Aluisio L, Fraser I, Motley ST, Korton K, Palmer J, Bonaventure P, Carruthers NI, Lovenberg TW, Boggs J & Galici R (2011). Selective blockade of the orexin- 2 receptor attentuates ethanol self-administration, place preference, and reinstatement, Psychopharmacol, 215: 191-203. Sieradazan KA, Fox SH, Hill M, Dick JR, Crossman AR & Brotchie JM (2001). Cannabinoids reduced levodopa-induced dyskinesia in Parkinson's disease: a pilot study. Neurology, 57: 2108-2111.

277 da Silva-Junior FP, Braga-Neto P, Sueli Monte F & de Bruin VM (2005). Amantadine reduces the duration of levodopa-induced dyskinesia: a randomized, double-blind, placebo-controlled study. Parkinsonism Relat Disord, 11: 449-452. Simola N, Frau L, Plumitallo A & Morelli M (2014). Direct and long-lasting effects elicited by repeated drug administration on 50-kHz ultrasonic vocalizations are regulated differently: implications for the study of the affective properties of drugs of abuse. Int J Neuropsychopharmacol, 17: 429-441. doi:10.1017/S1461145713001235 Simola N, Fenu S, Costa G, Pinna A, Plumitallo A & Morelli M (2012). Pharmacological characterization of 50-kHz ultrasonic vocalizations in rats: comparison of the effects of different psychoactive drugs and relevance in drug-induced reward. Neuropharmacology, 63: 224-234. Skuza G, Rogoz Z, Quack G & Danysz W (1994). Memantine, amantadine, and L-deprenyl potentiate the action of L-DOPA in monoamine-depleted rats. J Neural Transm (GenSect), 98: 57-67. Smidt MP, Smits SM, Bouwmeester H, Hamers FP, van der Linden AJ, Hellemons AJ, Graw J & Burbach JP (2003). Early developmental failure of substantia nigra dopamine neurons in mice lacking the homeodomain gene Pitx3. Development, 131: 1145-1155. Smith GA, Heuer A, Dunnet SB & Lane EL (2012). Unilateral nigrostriatal 6-hydroxydopamine lesions in mice II: predicting L-DOPA-induced dyskinesia. Behav Brain Res, 226: 281-292. Smits SM, Mathon DS, Burbach JP, Ramakers GM & Smidt MP (2005). Molecular and cellular alterations in the Pitx3-deficient midbrain dopaminergic system. Mil Cell Neurosci, 30: 352- 363. Snow BJ, Macdonald L, Mcauley D & Wallis W (2000). The effect of amantadine of levodopa- induced dyskinesia in Parkinson's disease: a double-blind, placebo-controlled study. Clin Neuropharmacology, 23: 82-85. Spina MB & Cohen G (1989). Dopamine turnover and glutathione oxidation: implications for Parkinson disease. Proc Natl Acad Sci, 86: 1398-1400. Stamatakis AM & Stuber GD (2012). Activation of lateral habenula inputs to the ventral midbrain promotes behavioral avoidance. Nat Neurosci, 15: 1105-1107. Stanford JA, Currier TD & Gerhardt GA (2002). Acute locomotor effects of fluoxatine, sertraline and nomifensine in young versus aged Fischer 344 rats. Pharmacol Biochem Behav, 71: 325- 332. van der Stelt M, Fox SH, Hill M, Crossman AR, Petrosino S, Di Marzo V & Brotchie JM (2005). A role for endocannabinoids in the generation of parkinsonism and levodopa-induced dyskinesia in MPTP-lesioned non-human primate models of Parkinson's disease. FASEB J, 19: 1140-1142. doi:10.1096/fj.04-3010fje Stocchi F, Roscol O, Destee A, Hattori N, Hauser RA, Lang AE, Poewe W, Stacy M, Tolosa E, Gao H, Nagal K, Merschhemke M, Graf A, Kenney C & Trankwalder C (2013). AFQ056 in Parkinson patients with levodopa-induced dyskinesia: 13-week randomized, dose-finding stufy. Mov Disord, 28: 1838-1846. Stocchi F (2006). The levodopa wearing-off phenomenon in Parkinson's disease: pharmacokinetic considerations. Expert Opin Pharmacother, 7: 1399-1407. Stocchi F, Ruggieri S, Vacca L & Olanow CW (2002). Prospective randomized trial of lisuride infusion versus oral levodopa in patients with Parkinson's disease. Brain, 125: 2058-2066. Spyraki C, Kazandjian A & Varonos D (1985). Diazepam-induced place preference conditioning: appetitive and antiaversive properties. Psychopharmacology, 87: 225-232.

Substance Abuse and Mental Health Services Administration [SAMHSA] (2013), Results from the 2012 National Survey on Drug Use and Health: Summary of National Findings, NSDUH

278

Series H-46, HHS Publication No. (SMA) 13-4795. Rockville, MD: Substance Abuse and Mental Health Services Administration.

Substance Abuse and Mental Health Services Administration [SAMHSA] (2009). Results from the 2008 National Survey on Drug Use and Health: National Findings (Office of Applied Studies, NSDUH Series H-36, HHS Publication No. SMA 09-4434). Rockville, MD.

Sulzer D, Chen T, Lau YY, Kristensen H, Rayport S & Ewing A (1995). Amphetamine redistributes dopamine from synaptic vesicles to the cytosol and promotes reverse transport. The Journal of Neuroscience, 15: 4102-4108. Sumnall HR, O'Shea E, Marsden CA, Cole JC (2004). The effects of MDMA pretreatment on the behavioural effects of other drugs of abuse in the rat elevated plus-maze test. Pharm Biochem Behav 77: 805-814. Sunahara RK, Guan H, O’Dowd BF, Seeman P, Laurier LG, Ng G, George SR, Torchia J, Van Tol HH & Niznik HB (1991). Cloning of the gene for a human dopamine D5 receptor with higher affinity for dopamine than D1. Nature, 350: 614-619. Svensson K, Carlsson A & Waters N (1994). Locomotor inhibition by the D3 ligang R-(+)-7-OH- DPAT is independent of changes in dopamine release. J Neural Transm (GenSect), 95: 71-74. Szabo B, Dorner L, Pfreundtner C, Norenberg W & Starke K (1998). Inhibition of GABAergic inhibitory postsynaptic currents by cannabinoids in rat corpus striatum. Neuroscience, 85: 395-403. Tahar AH, Gregoire L, Darre A, Belanger N, Meltzer L & Bedard PJ (2004). Effect of a selective glutamate antagonist on L-dopa-induced dyskinesias in drug-naive parkinsonian monkeys. Neurobiol Dis, 15: 171-176. doi:10.1016/j.nbd.2003.10.007 Tan KR, Yvon C, Turiault M, Mirzabekov JJ, Doehner J, Labouebe G, Deisseroth K, Tye KM & Luscher C (2012). GABA neurons of the VTA drive conditioned place aversion. Neuron, 73: 1173-1183. Tanaka H, Kannari K, Maeda T, Tomiyama M, Suda T & Matsunaga M (1999). Role of serotonergic neurons in L-DOPA-derived extracellular dopamine in the striatum of 6-OHDA-lesioned rats. NeuroReport, 10: 631-634. Tanner CM & Aston DA (2000). Epidemiology of Parkinson's disease and akinetic syndromes. Curr Opin Neuro, 13: 427-430. Taylor JL, Bishop C & Walker PD (2005). Dopamine D1 and D2 receptor contributions to L-DOPA- induced dyskinesia in the dopamine-depleted rat. Pharmacol Biochem Behav, 81: 887-893. doi:10.1016/j.pbb.2005.06.013 Tcheremissine OV (2008). Is quetiapine a drug of abuse? Reexamining the issue of addiction. Exp Op Drug Safety, 7: 739-748. Testa CM, Standaert DG, Landwehrmeyer GB, Penney Jr JB & Young AB (1995). Differential expression of mGluR5 metabotropic glutamate receptor mRNA by rat striatal neurons. J Comp Neurol, 354: 241-252. Teter CJ, McCabe SE, LaGrange K, Cranford JA & Boyd CJ (2006). Illicit use of specific prescription stimulants among college students: prevalence, motives, and routes of administration. Pharmacotherapy, 26: 1501-1510. Teixeria AM, Reckziegel P, Muller L, Pereira RP, Roos DH, Rocha JBT & Burger ME (2009). Intensive exercise potentiates oxidative stress in striatum of reserpine-treated animals. Pharmacol Biochem Behav, 92: 231-235. Thanvi B, Lo N & Robinson T (2007). Levodopa-induced dyskinesia in Parkinson's disease: clinical features, pathogenesis, prevention and treatment. Postgrad Med J, 83: 384-388.

279

Thiele SL, Warre R, Khademullah CS, Fahana N, Lo C, Lam D, Talwar S, Johnston TH, Brotchie JM & Nash JE (2011). Genertion of a model of L-DOPA-induced dyskinesia in two different mouse strains. J Neurosci Meth, 197: 193-208. Thomas A, Lacono D, Luciano AL, Armellino K, Di lorio A & Onofrj M (2004). Duration of amantadine benefit on dyskinesia of severe Parkinson's disease. J Neurol Neurosurg Psychiatry, 75: 141-143. Tiberi M, Jarvi KR, Silvia C, Falardeau P, Gingrich JA, Godinot N, et al. (1991). Cloning, molecular characterization, and chromosomal assignment of a gene encoding a second D1 dopamine receptor subtype: differential expression pattern in rat brain compared with the D1A receptor. Proc Natl Acad Sci U S A 88: 7491-7495. Treit D, Menard J, Royan C (1993). Anxiogenic stimuli in the elevated plus-maze. Pharmacology, biochemistry, and behavior 44: 463-469. Trigo JM, Panayi F, Soria G, Maldonado R & Robledo P (2006). A reliable model of intravenous MDMA self-administration in naive mice. Psychopharmacology (Berl), 184: 212-220. doi:10.1007/s00213-005-0229-7 Tronci E, Shin E, Bjorkland A & Carta M (2012). Amphetamine-induced rotation and L-DOPA- induced dyskinesia in the rat 6-OHDA model: a correlation study. Neurosci Res, 73: 168-172. Trugman JM & James CL (1992). Rapid development of dopaminergic supersensitivity in reserpine- treated rats demonstrated with 14C-2-deoxyglucose autoradiopgraphy. J Neurosci, 12: 2875- 2879. Tsang AHK & Chung KKK (2009). Oxidative and nitrosative stress in Parkinson's disease. Biochemica Biphysica Acta, 1792: 643-650. Tunstall B, Beckett S & Mason R (2009). Ultrasonic vocalisations explain unexpected effects on prepulse inhibition responses in rats chronically pre-treated with phencyclidine. Behav Brain Res, 202: 184-191. doi:10.1016/j.bbr.2009.03.035 Tye KM, Mirzabekov JJ, Warden MR, Ferenczi EA, Tsai H-C, Finkelstein J, Kim S-Y, Adhikari A, Thompson KR, Andalman AS, Gunaydin LA, Witten IB & Deisseroth K (2012). Dopamine neurons modulate neural encoding and expression of depression-related behaviour. Nature, 493: 537-543. Tzschentke TM (2007). Measuring reward with the conditioned place preference (CPP) paradigm: update of the last decade. Addict Biol, 12: 227-462. doi:10.1111/j.1369-1600.2007.00070.x Tzschentke TM & Schmidt WJ (2003). Glutamatergic mechanisms in addiction. Mol Psychiat, 8: 373-382. Ungerstedt U (1968). 6-hydroxydopamineinduced degeneration of central monoamine neurons. Eur J Pharmacol, 5: 107-110. Ungerstedt U & Arbuthnott GW (1970). Quantitative recording of rotational behavior in rats after 6- hydroxydopamine lesions of the nigrostriatal dopamine system. Brain Res, 24: 485-493. Ungerstedt U (1971). Adipsia and aphagia after 6-hydroxydopamine induced degeneration of the nigro-striatal dopamine system. Acta. Physiol. Scand. Suppl., 367: 95-122. Uhlenhuth EH, Johanson CE, Kilgore K & Kobasa SC (1981). Drug preference and mood in humans: Preferences for d-amphetamine and subject characteristics. Psychopharmacology, 74: 191-194. Utsumi H, Okuma Y, Kano O, Suzuki Y, Iijima M, Tomimitsu H, Hashida H, Kubo S, Suzuki M, Nanri K, Matsumura M, Murakami H & Hattori N (2013). Evaluation of the efficacy of pramipexole for treating levodopa-induced dyskinesia in patients with Parkinson’s disease. Int Med, 53: 325-332. Van Den Eeden SK (2003). Incidence of Parkinson's Disease: Variation by Age, Gender, and Race/Ethnicity. Am J Epidem, 157: 1015-1022.

280

Van Waes V, Beverley JA, Siman H, Tseng KY & Steiner H (2012). CB1 Cannabinoid Receptor Expression in the Striatum: Association with Corticostriatal Circuits and Developmental Regulation., Front Pharmacol, 3: 21-28. Vanover KE, Betz AJ, Weber SM, Bibbiani F, Kielaite A, Weiner DM, Davis RE, Chase TN & Salamone JD (2008). A 5-HT2A receptor inverse agonist, ACP-103, reduces tremor in a rad model and levodopa-induced dyskinesias in a monkey model. Pharmacol Biochem Behav, 90: 540-544. Verrico CD, Miller GM & Madras BK (2007). MDMA (Ecastasy) and human dopamine, norepinephrine, and serotonin transporters: Implications for MDMA-induced neurotoxicity and treatment. Psychopharmacol, 189: 489-503. Vila M & Przedborski S (2003). Targeting programmed cell death in neurodegenerative diseases. Nat Rev, 4: 1-11. Visanji NP, Fox SH, Johnston T, Reyes G, Millan MJ & Brotchie JM (2009). Dopamine D3 receptor stimulation underlies the development of L-DOPA-induced dyskinesia in animal models of Parkinson's disease. Neurobiol Dis, 35: 184-192. doi:10.1016/j.nbd.2008.11.010 Vitale C, Pellecchia MT, Grossi D, Fragassi N, Cuomo T, Di Maio L & Barone P (2001). Unawareness of dyskinesias in Parkinson’s and Huntington’s diseases. Neurol Sci, 22: 105- 106. Vivian JA & Miczek KA (1993). Morphine attentuates ultrasonic vocalization during agonistic encounters in adult male rats. Psychopharmacol, 111: 367-375. Volkmann J, Sturm V, Weiss P, Kappler J, Voges J, Koulousakis A, Lehrke R, Hefter H & Freund H-J (1998). Bilateral high-frequency stimulation of the internal globus pallidus in advanced Parkinson's disease. Ann Neurol, 44: 953-961. Voorn P, Roest G & Groenewegen HJ (1987). Increase of enkephalin and decrease of substance P immunoreactivity in the dorsal and ventral striatum of the rat after midbrain 6- hydroxydopamine lesions. Brain Res, 412: 391-396. Waeber C, Rigo M, Chinaglia G, Probst A & Palacios JM (1991). Beta-adrenergic receptor subtypes in the basal ganglia of patients with Huntington's chorea and Parkinson's disease. Synapse, 8: 270-280. Waldmeier PC (1985). Displacement of striatal 5-hydroxytryptamine by dopamine released from endogenous stores. J Pharm Pharmacol, 37: 58-60. Walsh S, Gorman AM, Finn DP & Dowd E (2010). The effects of cannabinoid drugs on abnormal involuntary movements in dyskinetic and non-dyskinetic 6-hydroxydopamine lesioned rats. Brain Res, 1363: 40-48. doi:10.1016/j.brainres.2010.09.086 Wang WW, Zhang XR, Zahng ZR, Wang XS, Chen J, Chen SY & Xie CL (2018). Effects of mGluR5 antagonists on Parkinson’s patients when L-dopa-induced dyskinesia: a systematic review and meta-analysis of randomized controlled trials. Front Aging Neurosci, 10: 262. Wang K, Wang D, Akamatsu T, Li S & Xiao J (2005). A passive acoustic monitoring method applied to observation and group size estimation of finless porpoises. J Acoust Soc Am, 118: 1180. doi:10.1121/1.1945487 Wang SJ, Wang K-Y & Wang W-C (2004). Mechanisms underlying the riluzole inhibition of glutamate release from rat cerebral cortex nerve terminals (synaptosomes). Neurosci, 125: 191-201. Watanabe M, Harada S, Nakamura T, Ohkoshi N, Yoshizawa K, Hayashi A & Shoji S (2003). Association between catechol-O-methyltransferase gene polymorphisms and wearing-off and dyskinesia in Parkinson's disease. Neuropsychobiology, 48: 190-193. Waters CH, Sethi KD, Hauser RA, Molho E, Bertoni JM & Zydis Selegiline Study G (2004). Zydis selegiline reduces off time in Parkinson's disease patients with motor fluctuations: a 3-month, randomized, placebo-controlled study. Mov Disord, 19: 426-432.

281

Waters N, Löfberg L, Haadsma-Svensson S, Svensson K, Sonesson C & Carlsson A (1994). Differential effects of dopamine D2 and D3 receptor antagonists in regard to dopamine release, in vivo receptor displacement and behaviour. J Neural Transm (GenSect), 98: 39-55. Waters N, Svensson K, Haadsma-Svensson SR, Smith MW & Carlsson A (1993). The dopamine D3- receptor: a postsynaptic receptor inhibitory on rat locomotor activity. J Neural Transm (GenSect), 94: 11-19. Weihe E, Schafer MK, Erickson JD & Eiden LE (1994). Localization of vesicular monoamine transporter isoforms (VMAT1 and VMAT2) to endocrine cells and neurons in rat. J Mol Neurosci, 5: 149-164. Westin JE, Vercammen L, Strome EM, Konradi C & Cenci MA (2007). Spatiotemporal pattern of striatal ERK1/2 phosphorylation in a rat model of L-DOPA-induced dyskinesia and the role of dopamine D1 receptors. Biol Psychiatry, 62: 800-810. Whishaw IQ, Coles BL, Pellis SM & Miklyaeva EI (1997). Impairments and compensation in mouth and limb use in free feeding after unilateral dopamine deplations in a rat analog of human Parkinson’s disease. Behav Brain Res, 84: 167-177. Williams SN & Undieh AS (2010). Brain-derived neurotrophic factor signaling modulates cocaine induction of reward-associated ultrasonic vocalization in rats. J Pharmacol Exp Therap, 332: 463-468. Williams TP & Miller BD (2003). Pharmacologic management of anxiety disorders in children and adolescents. Curr Opin Ped, 15: 483-490. Wilson MA, Ricaurte GA & Molliver ME (1989). Distinct morphologic classes of serotonergic axons in primates exhibit differential vulnerability to the psychotropic drug 3,4- methylenedioxymethamphetamine. Neurosci, 28: 121-127. Wilson WJ (2004). An inexpensive alternative to commercial infrared sensors. The Journal of Undergraduate Neurosci Ed, 3: A1-A3. Wilson WJ, Ogg JA & Marsack KZ (2000). Acute Ginkgo biloba facilitates decision-making in a working memory task in rats. Acta Neurobiol Exp, 60: 511. Wilson WJ (1996). The phi-maze: a versatile automated T-maze for learning and memory experiments in the rat. Behav Res Meth Inst Comp, 28: 360-4. Wilson WJ, Steinbronn NC, Lopshire JC, Galloway PH, Kellenberger SA, Hartman AM & Bennett RL (1992). Differences in spatial and visual working memory performance of rats in a phi- maze. Soc Neurosci Abstr, 18: 1423. Winkler C, Kirik D, Bjorklund A & Cenci MA (2002). L-DOPA-induced dyskinesia in the intrastriatal 6-hydroxydopamine model of Parkinson’s disease: relation to motor and cellular parameters of nigrostriatal function. Neurobiol Dis, 10:165-186. Wohr M, Houx B, Schwarting RKW & Spruijt B (2009). Effects of experience and context on 5-kHz vocalizations in rats. Physiol Behav, 93: 766-776. Wohr M & Schwarting RK (2008). Maternal care, isolation-induced infant ultrasonic calling, and their relations to adult anxiety-related behavior in the rat. Behav Neurosci, 122: 310-330. doi:10.1037/0735-7044.122.2.310 Wolf E, Seppi K, Katzenschlager R, Hochschorner G, Ransmayr G, Schwingenschuh P, Ott E, Kloiber I, Haubenberger D, Auff E & Poewe W (2010). Long-term antidyskinetic efficacy of amantadine in Parkinson's disease. Mov Disord, 25: 1357-1363. Wood SJ & Toth M (2001). Molecular pathways of anxiety revealed by knockout mice. Mol Neurobiol, 23: 101-119. Wright JM, Deng L & Clarke PBS (2012a). Failure of rewarding and locomotor stimulant doses of morphine to promote adult rat 50-kHrz ultrasonic vocalizations. Psychopharmacol, 224: 477- 487

282

Wright JM, Dobosiewicz MR & Clarke PB (2012b). alpha- and beta-Adrenergic receptors differentially modulate the emission of spontaneous and amphetamine-induced 50-kHz ultrasonic vocalizations in adult rats. Neuropsychopharmacology, 37: 808-821. Wright JM, Gourdon JC & Clarke PB (2010). Identification of multiple call categories within the rich repertoire of adult rat 50-kHz ultrasonic vocalizations: effects of amphetamine and social context. Psychopharmacology, 211: 1-13. Xie T, Tong L, McLane MW et al. (2006). Loss of serotonin transporter protein after MDMA and other ring-substituted amphetamines. Neuropharmacol, 31: 2639-2651. Yajima Y, Hayashi Y & Yoshii N (1980). The midbrain central gray substance as a highly sensitive neural structure for the production of ultrasonic vocalization in the rat. Brain Res, 198: 446- 452. Yamada H, Aimi Y, Nagatsu I, Taki K, Kudo M & Arai R (2007). Immunohistochemical detection of L-DOPA-derived dopamine within sertotonergic fibers in the striatum and the substantia nigra pars reticulata in Parkinsonian model rats. Neurosci Res, 59: 1-7. Yamato H, Kannari K, Shen H, Suda T & Matsunaga M (2003). Fluoxetine reduces l-DOPA-derived extracellular DA in the 6-hydroxydopamine-lesioned rat striatum. International Congress Series, 1251: 191-197. doi:10.1016/s0531-5131(03)00113-4. Young WC III, Bonner TI & Brann MR (1986). Mesencephalic dopamine neurons regulate the expression of neuropeptide mRNAs in the rat forebrain. Proc Natl Acad Sci USA, 83: 9827- 9831. Zakharova E, Leoni G, Kichko I & Izenwasser S (2009). Differential effects of methamphetamine and cocaine on conditioned place preference and locomotor activity in adult and adolescent male rats. Behav Brain Res, 198: 45-50. Zangrossi H Jnr & Graeff FG (1994). Behavioral effects of intra-amygdala injections of GABA and 5-HT acting drugs in the elevated plus maze. Braz J Med Biol Res, 27: 2453-2456. Zapata A & Shippenberg TS (2005). Lack of functional D2 receptors prevents the effects of the D3- preferring agonist (+)-PD 128907 on dialysate dopamine levels. Neuropharmacology, 48: 43- 50. van Zessen R, Phillips JL, Budygin EA & Stuber GD (2012). Activation of VTA GABA neurons disrupts reward consumption. Neuron, 73: 1184-1194. Zhang X, Andren PE, Greengard P & Svenningsson P (2008). Evidence for a role of the 5-HT1B receptor and its adaptor protein p11, in L-DOPA treatment of an animal model of Parkinsonism. PNAS, 105: 2163-2168. Zhang W, Perry KW, Wong DT, Potts BD, Bao J, Tollefson GD & Bymaster FP (2000). Synergistic effects of olanzapine and other antipsychotic agents in combination with fluoxetine on norepinephrine and dopamine release in rat prefrontal cortex. Neuropsychopharmacolo, 23: 250-262. Zhou Q-Y & Palmiter RD (1995). Dopamine-deficient mice are severely hypoactive, adipsic, and aphagic. Cell, 83: 1197-1209. Zibetti M, Merola A, Ricchi V, Marchisio A, Artusi CA, Rizzi L, Montanaro E, Reggio D, De Angelis C, Rizzone M & Lopiano L (2013). Long-term duodenal levodopa infusion in Parkinson's disease: a 3-year motor and cognitive follow-up study. J Neurol, 260: 105-114.

283

284

Appendix I – A simple and effective method for building inexpensive infrared equipment used to monitor animal locomotion

This journal article reviewing infrared equipment constructed to monitor animal locomotion was written during the period of candidature:

Publication: Mclelland AE, Winkler CE, Martin-Iverson MT (2015). A simple and effective method for building inexpensive infrared equipment used to monitor animal locomotion. J Neurosci Meth, 243: 1-7.

Highlights • We designed and constructed a flexible, inexpensive infrared rat monitoring system. • We recorded and analysed drug-induced rat behaviour using our monitoring system. • Our system reveals selective significant drug-induced differences in rat behaviour. • L-DOPA significantly increases selective rat behaviours recorded by our system. • Total cost of our system is substantially less than that of commercial suppliers.

Abstract Background: Infrared (IR) technology is a flexible and effective way of measuring animal locomotion. However, the cost of most commercial IR equipment can limit their availability. We have designed an inexpensive and effective replacement for commercial IR sensors that can be attached to enclosures to monitor animal locomotion.

New method: IR components were soldered to circuits connected to a single microcontroller. These IR components were housed inexpensively using plastic tubing and cork discs to further focus and extend detection of the IR beam. A standard personal computer recorded data from circuit boards connected to an inexpensive interface. This system may be used in a range of lighting conditions without requiring readjustment or recalibration.

Results: Validation of our equipment design was done with male Sprague Dawley rats treated with reserpine 22 h prior to administration of saline or L-DOPA (125 mg/kg). Data was collected in eight different measures: horizontal activity, immobile time, elevated activity, centre elevated activity, elevation time, elevation bout, and repeated and non- repeated movement while elevated. L-DOPA increased horizontal movement and all

285 elevated activity excepting elevated movement and centre elevated movement, demonstrating selective drug effects.

Comparison with existing methods: The total cost of our complete IR system

(US$517.45) was substantially less than the least expensive quote (US$19,666.90) obtained for a commercial IR system.

Conclusions: We have successfully designed and constructed a flexible and inexpensive IR system to monitor at least eight measures of rodent locomotion at a significantly lesser cost than quoted by commercial suppliers.

Introduction

The first infrared (IR) detector was made in 1800 by the discoverer of Pluto and IR light; William Herschel (Rogalski, 2012); and has been used to measure motor activity in rodents and other animals at least since 1970 (Fibiger et al., 1971). Despite the rapid rate of technological improvements; IR photocells remain a major method of measuring motor activity in rodents. The only real competitor to assess locomotion and rearing are camera based systems that are expensive and require equally expensive software. Camera systems do provide additional information in terms of path analysis compared to infrared technology; but these measurements are less often used in pharmacological research than quantified locomotion as this additional information may not be worth the enormous increase in both expense and time needed for data analysis.

A substantial component of animal research is composed of locomotion and behaviour analysis. These measures are the basis of many animal models and tests such as the elevated plus maze and open field tests of anxiety, the reserpine-treated rat or 6- hydroxydopamine-lesion rat or MPTP-treated primate models of Parkinson’s disease, the conditioned place preference test of reward, and exploratory/investigatory behaviour. The

286 behaviours and locomotion recorded in these tests and models include total distance travelled

(Zakharova et al., 2009), spatial patterns of locomotion (Geyer et al., 1987), immobility/freezing (Gresack et al., 2010), hole pokes (Riley et al., 1979), time spent in specific areas/compartments of the apparatus (Dietz et al., 2007), drinking and eating behaviour (Jahng and Houpt, 2001) and rearing (Gresack et al., 2010), all of which can be recorded using IR technology. Analysis of these behaviours and patterns of locomotion have been used to understand basis of many human behaviours such as drug use (McLelland et al.,

One of the greatest advantages to using IR technology is that, unlike mechanical switches, the activation of IR sensors is undetectable to the animal and therefore has minimal influence over its behaviour. Additionally, IR devices are compact, lightweight, and can be used in almost all laboratory light conditions. No newer technology has been able to compete with IR in both usefulness and price, especially as IR emitters and detectors have become extremely inexpensive, as compared to 30–40 years ago. Furthermore, the equipment needed to interface between the photobeam detectors and a computer or other data-logging device has improved considerably and become inexpensive and easily available online. However, costs of pre-made equipment from research equipment companies remains relatively high, often prohibitively so, especially considering how inexpensive the component parts have become, and how simple software is to write at present.

Some of the least expensive IR devices can range between US$50 and US$100 from research equipment supply companies (for example, one pair of IR emitters and detectors

287 from a research equipment company that is widely used have a median price of around

US$75 at the time of writing, so that IR emitters and detectors 10 per box and 5 boxes would cost US$3750), just for the IR devices, and excluding delivery. However, one IR emitter and detector pair can be bought online for US$0.33, at the time of writing, so that 40 can be bought from electronic supply stores for US$13.21, substantially less than from even the least expensive research equipment supply company. This expense does not include an interface to transmit IR beam break data to a computer for further processing (four channel IR controller from one company at the time of writing is approximately US$230, or another US$2875), and does not include the cost of software, cable, or circuit boards that are generally required to accurately measure animal locomotion. Therefore, our goal was to construct a set of five enclosures as an inexpensive alternative to commercial IR devices and a computer interface that was compatible with most modern computers to measure the activity of five rats simultaneously.

Methods

288

Construction

IR emitter/sensor pairs

The IR emitters purchased were required to reliably transmit 300 mm in a low voltage situation. The IR emitters operate at 5 V and allow up to 100 mA forward current. These specifications enabled the emitters to safely transmit over the required distance (approximately 30 cm) without operating at maximum capacity. In order to create a detectable IR beam, IR emitters were paired with IR sensors, creating a unidirectional IR beam that was sensitive to interruption.

Although only a 30 cm distance was required, the IR emitters were able to transmit up to 80cm by supplying 100mA forward current and increasing the resistance in the IR receiver circuit up to 600,000 ohms. However, transmission over this distance created a substantially larger beam width which was likely to interfere with nearby IR sensors.

We built a network of IR beams to measure horizontal and elevated activity of rats in

20 cm × 30 cm × 25 cm (w × l × h) Poly(methyl methacrylate) (PMMA, commonly known by the trade name Plexiglas®, Perspex®, or Lucite®) enclosures (US$17.81 each). We purchased IR emitters (Model TSAL 7200) from element14 (au.element14.com, US$0.08 each) and sensors (Model Lite-On LTR-301) from Mouser Electronics (au.mouser.com,

US$0.25 each), classified as IR LEDs and NPN phototransistors, respectively. A 5V DC power supply rated at 4A ($22.16) was used; however, this circuit could be constructed to accommodate a range of power supplies. Accessories for this circuit to monitor five enclosures included resistors (US$4.68), 8 core cable (US$56.13), circuit boards (US$18.71), enclosing boxes (US$5.61), and connector wires (US$18.71). Extras such as glue, power connector plugs, ferrules, solder, and croc clips totalled US$80.52.

In order to calculate the requirements of the power supply, it was necessary to verify how much power the IR emitters needed to form a reliable IR beam. The circuit voltage

289 output of the IR sensor determined the strength of the IR beam: the stronger the beam, the lower the voltage is. As the microcontroller discerns a voltage below 1.2 V as a digital zero

Housing

Shielding is another factor to consider when constructing IR LEDs and sensors.

Although the IR light emitted from an IR LED is invisible to humans (and many other animals), it emits light over a wide field. Therefore, distances larger than 2–5 cm between the

LED and sensor require shielding and housing of the LED (and sometimes the sensor to avoid LED emission overlap, see Fig. 1A and B) to focus the light towards the sensor more effectively. One solution is to add a lens to the housing unit of the LED to further focus the

IR light towards the sensor (Batson and Turner, 1986). However, this is optional as the equipment can still be effective without the lens (Wilson, 1996, 2004; Wilson et al., 1992,

2000).

Each IR emitter and sensor was placed inside a 60 mm length PVC tube with a 16 mm and 20 mm diameter opening obtained from a local hardware store (US$0.18 each, 40 PVC tubes totalling US$7.20 per enclosure). The 16 mm end of each tube was attached to the outside wall of the enclosure to ensure uniformity and focus of the IR light towards to the sensor. Before the emitters were soldered to wire, they were mounted onto the middle of individual cork discs (US$3.05 for a pack of four) by pushing the pins gently through the cork and securing the main component with glue. The sensors were mounted by glueing the head of each sensor to the middle of the cork discs. The cork was mounted onto the 20 mm end of the plumbing tube with glue, although when mounting the sensors, grooves were made

290 in the edge of the tubing to accommodate the vertical pins. The IR LEDs and sensors were then soldered to wire so that the voltage of the sensor could be measured while the LED and sensor were aligned on the wall of the enclosure. This was to ensure LED and sensor pairs were properly aligned with no IR overlap from the neighbouring emitters.

Fig.1: Schematic diagram demonstrating the effect of housing IR LEDS. IR light emitted from LEDs (light grey) in the animal enclosures (A) overlapping across three different sensors (light grey) due to wide emission field and (B) targeting aligned sensor

(dark grey) after being housed to narrow the IR beam and reduce emission field covered.

Attachment

Although versions of IR equipment include frames with IR sensors and emitters attached in which an animal enclosure can be placed (for example, AM1053 activity monitors from Linton Instru- mentation, UK), we attached the housed emitters and sensors to the enclosure for increased mobility and decreased storage. This does have the drawback of decreased flexibility regarding enclosure replacement and altered the placement of the sensors and emitters. However, this was not a disadvantage for our particular study, but should be considered for individual experiments requiring custom-made equipment. This design also has the advantage of avoiding the issue with using home cages where scratches

291 and other discolourations and dirt on the walls can impede the signal by allowing holes to be drilled where the emitters and sensors are located to avoid the impediment from the walls.

8 cm above the floor of the enclosures. These measurements were chosen to ensure that movement by the animal would break at least one IR beam in both the X and Y axes of the enclosure. The arrangement of the lower level pairs was designed to record coordinates of the rat’s movement to determine whether the rat was in the peripheral or central areas in the enclosure and record exploratory movement measured as beam breaks. The arrangement of the upper level pairs was designed to record rearing behaviour rather than coordinates of the rat’s location. The area covered by the upper level enclosures was tested and found to cover the vast majority of the enclosure area horizontally. The areas that were not reached by the upper level IR pairs were found to be substantially smaller than the size of a 225–325 g rat and thus a beam break would be registered even if the animal was rearing in the uncovered areas. Of course, more IR emitters and detectors can be added if desired.

292

Computer and software

The output voltage from the IR sensors were fed to the inputs of the Fez Panda II microcontroller (one microcontroller used, purchased for US$53.57 at the time of writing).

The microcontroller firmware (freely available from https://github.com/cewinkler/

Loco.Firmware) was programmed to translate the inputs to sensors on enclosures. While turned on, the microcontroller was programmed to read the sensors at a rate of 2 Hz and translate the readings into 8-bit binary strings which indicate what state each beam in the enclosure is in. The first four bits indicate the X- axis beams; the fifth to seventh bits indicate the Y-axis beams; and the last bit indicates if any of the upper level beams are broken. The readings are transmitted through the serial port of the microcontroller, which can be received by a PC using a standard universal serial bus (USB) connection.

Software was written using the C# programming language to receive and record the microcontroller data (freely available from https://github.com/cewinkler/Loco.Collection).

293 locomotor measures. This process generated comma separated files (CSVs) for each locomotor measure for importing in to R for statistical analysis.

Experimental procedures

Animals

Experimental procedures were performed with approval from the University of

Western Australia Animal Ethics Committee and in accordance with the Australian Code of

Practice of Use of Animals for Scientific Purposes and relevant Western Australia State law.

A total of 30 drug and test naïve male Sprague Dawley rats, originally weighing 225–275 g upon arrival from the Animal Resource Centre of Western Australia, were housed in pairs, in clear polypropylene 47 cm × 37 cm × 20 cm (l × w × h) boxes with woodchip bedding. The animal housing room was temperature controlled and maintained at approximately 22 ± 2 ◦ C under a 12 h light–dark cycle (lights on at 6:00 a.m.). Rats had free access to food and water except during testing. Each rat was handled for 2 min daily for 5 days prior to behavioural testing. After reserpine administration, rats were given free access to soft feed and Necta gel packs to minimise health effects of akinesia.

Drugs

Reserpine (Sigma–Aldrich) was freshly dissolved in 30–40% captisol solution

(captisol dissolved in distilled water) before each administration. l-3,4- dihydroxyphenylalanine methyl ester hydrochloride (Sigma Aldrich) was prepared using

294 saline. The pH of all solutions was approximately 6.5–7.0. Saline was used as a vehicle treatment. All drugs were administered at 1 ml/kg.

Testing

All rats were administered reserpine (4 mg/kg, freshly dissolved in 30% captisol solution, i.p.) 22 h before testing. All rats were randomly allocated to received L-DOPA

(125mg/kg, n=10) or saline (1 mg/kg, n = 20). Rats were habituated to the testing boxes for an hour before being injected (i.p.) with L-DOPA (125 mg/kg) or saline (1 mg/kg) and tested immediately afterwards for 80 min. All testing was conducted between 10:00 am and 5:00 pm. This testing regime was designed for a separate study currently being prepared for publication.

Alternatively, horizontal movement may be translated into distance travelled by multiplying the number of beams broken on the x or y axes by the distance between the IR devices on each separate axis. For example, the distance between IR devices along an enclosure wall was 4 and 3 cm for the X and Y axes, respectively. The relevant equation used to translate horizontal movement to distance travelled would be the following:

Distance travelled = [(distance between IR devices) ∗ (beam breaks on x axis)]

+[(distance between IR devices) ∗ (beam breaks ony axis)]; (2)

However, this equation could only be effectively implemented using equipment with more IR devices to reduce the error of assuming the entire length of a square has been travelled.

295

Elevated movement (EM) was defined by beam breaks in the upper level IR pairs, elevation time (ET) was determined by measuring how long an upper level beam was broken for, centre elevated movement (CEM) was measured similarly to EM with the addition of analysing the coordinates of the lower level beam breaks that occurred simultaneously with the upper level beam breaks. Thereby, we could determine the location of the rat on the floor of the enclosure during rearing and whether the rat was in the centre or periphery of the enclosure floor in addition to its overall movement. Repeated movement while elevated

Although not recorded by the equipment itself, a new measurement termed ‘elevation bout’ was also included in the analysis. Elevation bout (EB) is determined by the equation:

(Elevation time)/(Elevated movement + 1); (4)

An animal was considered to be in an elevated position when grooming or rearing.

Therefore, measurements such as elevation bout, elevation time, and elevated movement represent the summation of grooming and rearing behaviour.

Statistical data analysis

A mixed two-way Treatment × Time comparison of locomotor measures between of groups. Mean locomotion totals between groups were analysed with a two-tailed Student’s t- test. Statistical significance was defined as p < 0.05.

296

Results

Data collection

For the purpose of our study, the enclosures, interface, and software programme were designed to detect and record HM, IMT, ET, EB, EM, CEM, RME, and NRME either collated in 5min bins (Fig. 2A–H) or as a combined total for the 80 min testing session (Fig.

3A–H). Although the data was collated in 5 min bins, our software allows data collation in any bin length.

Fig. 2: Mean Locomotor activity + SEM of Sprague-Dawley rats treated with saline (n = 10) or L-DOPA (n = 10) A) Horizontal activity measured as number of beam breaks per 5 minute bin B) Immobile time measured as time (seconds) when the animal remained motionless per

297

5 minute bin C) Elevation time measured as length (seconds) of beam break in the top level of emitters and sensors only per 5 min bin D) Elevation bout measured as length (seconds) of upper level beam break per elevation movement per 5 minute bin E) Elevated movement measured as number of beam breaks located in the top level of IR emitters and sensors only per 5 minute bin F) Centre elevated movement measured as the number of beam breaks located in the top level of IR emitters and sensors while located in the centre of the enclosure floor per 5 minute bin G) Repeated movement while elevated as measurement by repetitive horizontal movement during upper beam break and H) Non-repeated movement while elevated as measured as remaining horizontal movement during upper beam beak during testing session in enclosures with transecting unidirectional IR beams using IR emitters and sensors.

Figure. 3: Mean locomotor activity + SEM of Sprague Dawley rats treated with saline (n = 10) or L-DOPA (n = 10) over total testing session. A) Horizontal activity B) Immobile time C) Elevation time measured D) Elevation bout E) Elevated movement F) Centre elevated

298

movement G) Repeated movement while elevated H) Non-repeated movement while elevated. * Significantly difference compared to saline group (p < 0.05).

Monetary cost of equipment

The cost of our custom-made enclosures includes the Plexiglas® enclosures, IR

devices, the Fez Panda microcontroller, power supply, housing units, resistors, circuit boards,

terminal blocks, 8 core cable, solder, glue, ferrule, power connector plugs, and wires. Table 1

shows the price comparison between the IR system built for the current report and

commercial IR systems at time of writing.

The total price for the IR system in the current study amounted to US$477.89 to

measure the activity of five rats simultaneously. This amount is substantially less than quoted

prices from commercial suppliers at the time of writing, with the least expensive commercial

system being priced at US$19,666.90 for five locomotor enclosures in addition to set up

materials purchased from commercial suppliers.

Table 1: A comparison of the prices between IR-based rodent locomotor activity monitoring systems from the current study and commercial suppliers. Three separate quotes were obtained for comparison with a price per unit (activity monitoring for one rat) with and without rodent boxes and for five units (activity monitoring for five rats) with and without rodent enclosures. All prices in USD.

Equipment for current Quote 1 Quote 2 Quote 3 study

With Without With Without With Without With Without enclosures enclosures enclosures enclosures enclosures enclosures enclosures enclosures

Per unit $156.16 $138.35 $5,595.70 - $5,095.35 $5,077.54 $5,764.98 $5,507.68

For 5 units $477.89 $388.84 $19,666.90 - $22,148.75 $22,059.70 $28,824.90 $27,538.41

Behavioural and locomotor data

Initial analysis of the data revealed violations of homogeneity of variance and thus the

data was transformed to square roots, which produced homogeneity of variance. The 80 min

testing session was divided into two 40 min periods for analysis of drug effects over time.

299

In rats treated with saline or L-DOPA (125mg/kg), the two- way mixed ANOVAs for

HM and IMT revealed a significant effect of treatment (F(1,28) = 28.77, p < 0.001 and

F(1,28) = 24.60, p < 0.001, respectively), but not of period (F(1,28) = 1.31, p = 0.26 and

F(1,28) = 2.73, p = 0.11, respectively) and a significant treatment × period interaction

(F(1,28) = 28.86, p < 0.001 and F(1,28) = 30.88, p < 0.001, respectively). Pairwise comparisons revealed that L-DOPA increased HM and decreased IMT in the second, but not the first period, compared to saline (p < 0.05).

Similarly, a two-way mixed ANOVA revealed significant treatment (F(1,28) = 11.89, p < 0.01), but not period (F(1,28) = 0.68, p = 0.42) effects and a significant treatment × period interaction (F(1,28) = 21.73, p < 0.001) for ET. Pairwise comparisons showed that L-

DOPA increased ET in both periods compared to saline (p < 0.05).

Significant effects of treatment (F(1,28) = 9.88, p < 0.01) and period (F(1,28) = 7.26, p < 0.05) were seen in the ANOVA for EB as well as a significant interaction (F(1,28) =

17.99, p < 0.001). Pairwise comparisons showed that L-DOPA increased EB in both periods compared to saline (p < 0.05).

While there was no significant treatment effect on EM (F(1,28)=1.15, p=0.29), a period effect was seen (F(1,28)=4.27, p<0.05), but without a significant interaction

(F(1,28)=0.002, p=0.96) in the two-way mixed ANOVA. However, the mixed ANOVA for

CEM showed no significant effects for treatment (F(1,28)=1.53, p=0.22) or period

(F(1,28)=2.09, p=0.15) and no significant interaction (F(1,28)=0.49, p=0.49). Pairwise comparisons showed no significant differences between L-DOPA and saline over the two periods for EM (p > 0.05).

The ANOVAs for RME and NRME both revealed similar results as significant treatment effects were seen for both behaviours (F(1,28)=38.03, p<0.001 and F(1,28)=11.02, p<0.01, respectively), a non-significant effect of period (F(1,28)=4.00, p = 0.055

300 and (F(1, 28) = 1.08, p = 0.31, respectively) and a significant interaction for RME (F(1,28) =

26.42, p < 0.001), but not for NRME (F(1,28) = 3.28, p = 0.08). Pairwise comparisons showed that L-DOPA increased RME in both periods compared to saline. Pairwise comparisons showed that L-DOPA increased NRME in the second period compared to saline, while the first period showed a trend in increased L-DOPA-induced NRME compared to saline. Overall, L-DOPA-induced locomotion and behavioural activity was significantly increased compared to saline-treated rats, and also increased significantly over time.

Independent Student’s t-tests for summated data over the entire testing session revealed that L-DOPA significantly increase HM, ET, EB, RME, and NRME (p < 0.05) and significantly decreased IMT (p < 0.05) compared to saline, while no differences were found between groups for EM and CEM (p > 0.05).

Discussion

The results of the current study demonstrated significant differences in select locomotion and behavioural measures of reserpine-treated rats administered L-DOPA or saline and tested in enclosures lined with custom-made IR devices and recorded using a custom-made data analysis programme compatible with a standard PC. Furthermore, analysis of the data collected by our equipment also revealed significant differences in all measures over time. These differences suggest that our custom-made equipment and software detect and record selective drug-induced summated and time-based changes in rodent locomotion and behaviour. Furthermore, the increases in activity induced by L-DOPA compared to saline seen in the current study are in agreement with those reported previously in reserpine-treated rats (Johnston et al., 2005; Segovia et al., 2003). Therefore, our inexpensive design of an IR system is an effective alternative to commercial IR systems.

301

Although our specific enclosure design may not be suited for all studies, the IR devices, computer interface, and data analysis programme are able to accommodate specific requirements of experiment designs in a number of different research fields including, but not limited to, pharmacology, psychology, and pathology. To highlight the inexpensive construction of our equipment, we obtained three separate quotes for complete IR-based animal locomotor monitoring systems, including five enclosures to monitor five individual rats simultaneously, from commercial research equipment suppliers. Table 1 shows the price comparison between the IR system built for the current report and commercial IR systems at time of writing.

302 outcomes when observers are not ‘blind’ to experimental treatment of the animal. However, because of the often prohibitive expense of commercial IR systems and equipment, many laboratories may find it difficult to incorporate this versatile technology into their research.

We have designed a set of enclosures to measure rodent behaviour using inexpensive and durable IR emitters and sensors constructed form basic materials that can be found in a range of retail and hardware stores. These IR devices are effective and reliable in measuring a range of rodent behaviours with no faults or breakages in the equipment or in the software programme and with significant differences found in behaviours after drug administration.

References for this book chapter are embedder within the main thesis reference section as there is large overlap.

303