Striatal neurons in the development of levodopainduced dyskinesias in Parkinson s disease

Transcription

1 University of Iowa Iowa Research Online Theses and Dissertations Fall 2017 Striatal neurons in the development of levodopainduced dyskinesias in Parkinson s disease Stephanie Lorraine Alberico University of Iowa Copyright 2017 Stephanie Lorraine Alberico This dissertation is available at Iowa Research Online: Recommended Citation Alberico, Stephanie Lorraine. "Striatal neurons in the development of levodopa-induced dyskinesias in Parkinson s disease." PhD (Doctor of Philosophy) thesis, University of Iowa, Follow this and additional works at: Part of the Neuroscience and Neurobiology Commons

2 STRIATAL NEURONS IN THE DEVELOPMENT OF LEVODOPA-INDUCED DYSKINESIAS IN PARKINSON S DISEASE by Stephanie Lorraine Alberico A thesis submitted in partial fulfillment of the requirements for the Doctor of Philosophy degree in Neuroscience in the Graduate College of The University of Iowa December 2017 Thesis Supervisor: Assistant Professor Nandakumar Narayanan

3 Copyright by STEPHANIE LORRAINE ALBERICO 2017 All Rights Reserved

4 Graduate College The University of Iowa Iowa City, Iowa CERTIFICATE OF APPROVAL This is to certify that the Ph.D. thesis of PH.D. THESIS Stephanie Lorraine Alberico has been approved by the Examining Committee for the thesis requirement for the Doctor of Philosophy degree in Neuroscience at the December 2017 graduation. Thesis Committee: Nandakumar Narayanan, Thesis Supervisor Steven Anderson Raghuraman Mudumbai Amy Lee Stefan Strack

5 To the Smith family, without whom my passion for this disease would not have amounted to what it has, influencing not only me personally but also my scientific career. To my mother, without whom I would not be half the person I am today. And to my father, with whom I could have been so much more ii

6 ACKNOWLEDGEMENTS I would like to first acknowledge my thesis advisor, Dr. Nandakumar Narayanan. I am sure that taking in your first graduate student was no small feat, but you did it with such poise and without hesitation. I have learned so much from you in the past few years. This project would not have amounted to what it did without you continuously pushing me further than I thought possible and without your ideas. I would also like to thank the other members of the Narayanan lab, especially Dr. Young-Cho Kim, Eric Emmons, and Ryan Kelley. Young-Cho, I cannot thank you enough for all the time you spent explaining things to me, talking through some of my ideas out loud, and providing advice not only about science but also on life. Eric and Ryan, I could not have asked for better lab mates turned friends. Our conversations and discussions helped me see things differently and clearly. I would also like to thank the members of my thesis committee Drs. Steven Anderson, Amy Lee, Stefan Strack, and Raghu Mudumbai. Thank you for the fruitful conversations and the guidance and support. Your continued dedication to my success and scientific development provided me with the perfect mentoring and learning environment. I would like to extend my gratitude to my previous mentors Drs. Rodney Clark and Juanita Anders for their encouragement to follow my passion and continue my education by pursuing a PhD in Neuroscience. During my undergraduate career Dr. Clark provided key mentoring that has truly stood the test of time, as he continues to advise and support me and my endeavors. Dr. Anders beautifully complemented my undergraduate mentor and taught me much of what I needed to be a good scientist and a good candidate iii

7 during the application process. She also provided the perfect opportunities for me to begin my scientific career. Her mentoring also turned into a lifelong process, from which I have learned so much. I would finally like to thank my family and friends, not only those that were with me from the start but also those that I have made along the way. I would especially like to thank my mother, my father, and Alexander Borton for their unwavering support throughout the years. iv

8 ABSTRACT Levodopa-induced dyskinesias (LIDs) are abnormal involuntary movements that limit the effectiveness of treatments for Parkinson s disease. Although dyskinesias involve the striatum, it is unclear how striatal neurons are involved in dyskinetic movements. Here we record from striatal neurons in mice during levodopa-induced axial dyskinesias. We developed an automated 3-dimensional motion tracking system to capture the development of axial dyskinesias at ~10 ms resolution, and correlated these movements with neuronal activity of striatal medium spiny neurons and fast spiking interneurons. The average firing rate of medium spiny neurons increased as axial dyskinesias developed, and both medium spiny neurons and fast spiking interneurons were modulated around axial dyskinesias. We also found that delta field potential power increased in the striatum with dyskinesia, and that this increased delta power coupled with striatal neurons. Secondly, we studied the role of the two main types of dopamine receptors. We pharmacologically inhibited either the D1 or D2 receptors while recording from neuronal ensembles in the striatum and measuring LIDs in high temporal resolution. We found that inhibiting the D1, but not the D2, receptor led to a decrease in axial dyskinesias. Interestingly, both types of antagonist attenuated the strong modulation of MSNs around axial dyskinesias when compared to levodopa alone. These results suggest that LIDs are modulated through activity in D1-MSNs. Lastly, we selectively targeted the D1 receptor expressing neurons (D1-MSNs) with optogenetics. With this technique, we can specifically activate or inhibit certain neuronal populations. We found that stimulating the D1-MSNs led to dyskinetic events v

9 only after levodopa priming. However, inhibiting these neurons was not sufficient to attenuate dyskinesias following levodopa administration. We also found that putative D1- MSNs are more strongly modulated around axial dyskinesias than other MSNs. Together, our findings provide novel insight into how striatal networks change as LIDs develop, and suggest that increased medium spiny neuron firing, that D1-MSNs are strongly modulated around LIDs, and that D1-MSN activity is sufficient to drive dyskinesias. These data could help clarify the role of the striatum in the pathogenesis of dyskinesias in Parkinson s disease. vi

10 PUBLIC ABSTRACT In Parkinson s disease (PD) the neurons that make the neurotransmitter dopamine, which is important for movement, degenerate. To compensate for this depletion of dopamine, PD patients are most commonly treated with levodopa, a dopamine precursor. However, up to 80% of patients develop levodopa-induced dyskinesias (LIDs). LIDs are abnormal involuntary movements and they limit the effectiveness of treatments for PD. This project aimed at determining how neurons in the striatum dysfunction around dyskinetic movements. We recorded from striatal neurons in mice during levodopa-induced axial dyskinesias. To better analyze striatal activity around these events, we developed an automated 3-dimensional motion tracking system to capture precisely when axial dyskinesias occurred. Secondly, we studied the role of the two main types of dopamine receptors. We pharmacologically inhibited either the D1 or D2 dopamine receptors while recording from neuronal ensembles in the striatum and measuring LIDs. Lastly, we selectively targeted the D1 receptor expressing neurons (D1-MSNs) with optogenetics. With this technique, we can specifically activate or inhibit certain neuronal populations. Together, our findings provide novel insight into how striatal networks change as LIDs develop, and suggest that increased medium spiny neuron firing, that D1-MSNs are strongly modulated around LIDs, and that D1-MSN activity is sufficient to drive dyskinesias. These data could help clarify the role of the striatum in the pathogenesis of dyskinesias in Parkinson s disease. vii

11 TABLE OF CONTENTS TABLE OF CONTENTS... viii LIST OF FIGURES... x LIST OF ABREVIATIONS... xii CHAPTER 1: INTRODUCTION... 1 The Basal Ganglia... 1 Parkinson s Disease... 3 Levodopa and Levodopa-induced Dyskinesias... 7 Animal Models Rodent Model Non-human Primate Model Possible Mechanisms Serotonin Medium Spiny Neurons Fast-spiking interneurons Cholinergic Interneurons Experimental Design CHAPTER 2: AUTOMATED COMPUTER TRACKING Introduction Methods Results Summary CHAPTER 3: STRIATAL NEURONS IN THE DEVELOPMENT OF LIDs Introduction Methods Results Summary CHAPTER 4: THE ROLE OF D1- AND D2-DOPAMINE RECEPTORS IN LIDs Introduction Methods Results Summary viii

12 CHAPTER 5: SELECTIVE ACTIVATION OF D1-MSNs DRIVES LIDs Introduction Methods Results Summary CHAPTER 6: DISCUSSION AND FUTURE DIRECTIONS Discussion Conclusion REFERENCES APPENDIX ix

13 LIST OF FIGURES Figure 1: The basal ganglia and movement circuitry... 2 Figure 2: Example of midbrain sections from control and PD subjects depicting dopamine cell loss... 4 Figure 3. Cell count in dopaminergic midbrain nuclei in PD cases and controls Figure 4. Molecular structures of levodopa (L-DOPA), carbidopa, and dopamine... 8 Figure 5: Timeline of experimental design for LID development Figure 6. Example of abnormal involuntary movement scores as LIDs develop Figure 7. Automated tracking in mouse model of levodopa-induced dyskinesias Figure 8. Tracking of axial position with infrared cameras Figure 9. Automated tracking of dyskinetic events Figure 10. Computer identified axial dyskinesias are correlated with AIM scores Figure 11. Mouse model of levodopa-induced dyskinesias Figure 12. Classification of striatal neurons Figure 13. Striatal MSNs increase firing rate as LIDs develop Figure 14. Striatal neurons are modulated around levodopa-induced dyskinesias Figure 15. Delta power in striatal local field potentials increases as LIDs develop Figure 16. Striatal FSI delta spike-field coherence increases as LIDs develop Figure 17: Blocking D1R decreases LIDs Figure 18: MSN firing rate Figure 19: MSN firing rate in 15 minute bins Figure 20: MSNs are less modulated around LIDs when inhibiting the direct and indirect pathways Figure 21: Example of optogenetically tagged putative D1-MSN x

14 Figure 22: Automatically detected LIDs Figure 23: Stimulation of D1-MSNs increases dyskinesia Figure 24: Mean firing rate across days of LID development Figure 25: D1-MSNs are strongly modulated around axial LIDs xi

15 LIST OF ABREVIATIONS 6-OHDA: 6-hydroxy dopamine AIM: abnormal involuntary movement ChI: cholinergic interneurons ChR2: channelrhodopsin D1/2DR: D1/D2 dopamine receptor FSI: fast-spiking interneurons GPe: globus pallidus external segment GPi: globus pallidus internal segment LFP: local field potential LIDs: levodopa-induced dyskinesias MFB: medial forebrain bundle MPTP: 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine MSN: medium spiny neuron NpHR: halorhodopsin PCA: principal component analysis PD: Parkinson s disease SN: substantia nigra SNc: substantia nigra pars compacta STN: subthalamic nucleus TH: tyrosine hydroxylase VTA: ventral tegmental area xii

16 CHAPTER 1: INTRODUCTION The ease with which humans and other animals move is for the most part taken for granted. Much is known about the motor pathway and how the firing of neurons in the brain leads to action. However, what happens when these pathways are disrupted is still much debated. One such example is Parkinson s disease. Even more controversial remains the mechanism of the development of levodopa-induced dyskinesias, a disabling side-effect of the main treatment for Parkinson s disease. Below, I will outline the movement circuit, depict how it goes awry in Parkinson s disease, and provide supporting evidence for the involvement of a specific neuronal population (D1-medium spiny neurons) in driving levodopa-induced dyskinesias. The Basal Ganglia The biggest component of the motor circuit is the basal ganglia comprised of the substantia nigra (SN), putamen, caudate nucleus, subthalamic nucleus, and globus pallidus that all work to control motor activity (Figure 1) (Calabresi et al., 2014). The caudate nucleus and putamen are collectively called the dorsal striatum, in contrast to the nucleus accumbens referred to as the ventral striatum, which will not be in the scope of this project. The dorsal striatum is the main input of the basal ganglia receiving heavy input from the cortex, thalamus, and the substantia nigra pars compacta (SNc). Cortical and thalamic projections are glutamatergic, excitatory, while the SNc provides dopaminergic input (Lanciego et al., 2012). Dopamine is one of the main neuromodulatory neurotransmitters, meaning that it does not simply activate or inactivate its target. The striatum is composed of approximately 95% medium spiny neurons 1

17 (MSNs), also referred to as spiny projection neurons (Kreitzer, 2009). These neurons use gamma- Aminobutyric acid (GABA) as their neurotransmitter, providing inhibition to target nuclei, and are divided into two categories based on the type of dopamine receptor they express, either D1- or D2-dopamine receptor. These neurons are the main output of the striatum and follow two main pathways: D1 in the striatonigral, or direct pathway, and D2 in the Figure 1: The basal ganglia and movement circuitry. This schematic represents all of the components of the basal ganglia including the dorsal striatum (caudate and putamen; grey), globus pallidus (both internal and external, GPi and GPe; green), the subthalamic nucleus (STN; yellow), and the thalamus (blue). This schematic also provides the projections of the direct (red) and indirect (blue) pathways. From Calabresi et al., striatopallidal, or indirect pathway (Smith et al., 1998; Kreitzer, 2009; Lanciego et al., 2012). Activity in the direct pathway is important for promoting movement while activity in the indirect pathway prevents unnecessary movements. Typically, dopamine release increases activity of D1R expressing MSNs (D1-MSNs) while it lowers activity of D2R expressing MSNs (D2- MSNs) (Kreitzer, 2009). D1-MSNs in the direct pathway project to the globus pallidus internal segment (GPi) which in turn sends GABAergic projections to the thalamus and the thalamus provides glutamatergic input to the cortex (Lanciego et al., 2012). When dopamine is released in the striatum, D1-MSNs increase their activity which inhibits the GPi allowing for disinhibition of the thalamus. On the other hand, D2-MSNs project to 2

18 the globus pallidus external segment (GPe), this nucleus sends inhibitory projections to the subthalamic nucleus (STN; a tonically active and excitatory nucleus) which in turn projects to the tonically active GPi and then the thalamus (Lanciego et al., 2012). Activity of D2-MSNs leads to inhibition of the GPe, which disinhibits the STN leading to the activation of the GPi and its inhibition of the thalamus, thus reducing movement. Parkinson s Disease In the United States alone one million individuals suffer from Parkinson s disease (PD), the second most common neurodegenerative disease (Thomas and Beal, 2007). This disorder mainly affects individuals over the age of 50, 2% of the population over 65, thus, as our average lifespan is projected to increase, so is the prevalence of PD. The progressive neurodegenerative disorder is characterized by the loss of dopaminergic neurons in the substantia nigra pars compacta (SNc) and the presence of Lewy bodies (Hornykiewicz, 1998; Eeden et al., 2003; Factor and Weiner, 2008). Lewy bodies are abnormal clusters of aggregated protein that are found in a number of structures of the brain, such as the substantia nigra and hypothalamus (Gibb and Lees, 1988). Lewy bodies are comprised of alpha-synuclein fibrils that cause disturbances to affected areas (Fahn et al., 1990). Unfortunately, patients don t typically present with motor symptoms until about 70% of dopamine is lost. In advanced disease stages, the loss of dopamine observed in the striatum is at least 80% (Hornykiewicz, 1998). The degeneration of dopaminergic neurons disrupts striatal activity, contributing to the devastating motor disabilities that characterize PD. The onset of motor problems related to PD may be overwhelming and leads to a decrease in the quality of life of patients (Karlsen et al., 1999). 3

19 Dopaminergic cell bodies are in the SN (both the pars compacta and reticulata) and in the ventral tegmental area (VTA). These nuclei project diffusely throughout the brain in separate pathways. The SN gives rise to the nigrostriatal pathway sending projections to the striatum and the VTA sends projections to the cortex and limbic system giving rise to the mesocortical and mesolimbic pathways, respectively. During the progression of Figure 2: Example of midbrain sections from control and PD subjects depicting cell loss. When stained for tyrosine hydroxylase the SNc and VTA PD cases (bottom) have lower cell counts and loss of dark pigmentation typical of control SNc (top). From Alberico et al., PD, degeneration of dopaminergic neurons occurs in the SN and VTA causing motor and non-motor problems (Figure 2) (Marsden and Parkes, 1977). The cardinal symptoms of PD include a resting tremor, rigidity of muscles and joints, postural instability, and bradykinesia, which express in patients as problems with movement initiation and slowing of movement (Marsden and Parkes, 1977; Jankovic, 2008). PD can be difficult to diagnose and typically relies on exclusion criteria. In the clinic several motor tests and rating scales are used. The most widely used and established measure of PD is the United Parkinson s Disease Rating scale which is a measure of disability and impairment. Bradykinesia and tremor at rest (4-16 Hz range which disappears during action and sleep) are the most identifying symptoms of PD; rigidity is one of the sources of pain and postural rigidity is widely seen in the later stages 4

20 of the disease; postural instability is the most common cause of falls in PD patients; while freezing seems to be the most disabling symptom (Jankovic, 2008). In addition to these motor tests, clinicians will typically order a positron emission tomography scan or a single photon emission computed tomography scan to image the dopamine transporter as these can provide insight into how much dopamine is left in the midbrain (Thobois et al., 2001; Marshall and Grosset, 2003). However, without a positive identification of Lewi bodies (typically done post mortem) it is nearly impossible to diagnose true PD. Though the physical symptoms alone are disabling, other major changes occur. Crucially, many PD patients have non-motor symptoms that include disorders associated with the VTA, which also undergoes substantial cell loss in PD (Figure 3; Alberico et al., 2015). Figure 3. Cell count in dopaminergic midbrain nuclei in PD cases and controls. Average tyrosine hydroxylase (TH) positive cell counts in sections of the VTA and SNpc of controls (black) and PD cases (grey). * indicated p < From Alberico et al., For instance, 25% of PD patients have anxiety and/or depression and nearly 30% of PD patients have executive dysfunction (Elgh et al., 2009; Aarsland et al., 2010). Over 80% develop non-motor symptoms affecting mood, cognition, and sleep (Hely et al., 5

21 2008; Aarsland et al., 2010). Goetz et al. and Forsaa et al. identified the development of psychotic symptoms as a risk factor for mortality, as those patients with psychotic symptoms died at earlier ages following PD diagnosis (Goetz et al., 2006; Forsaa et al., 2010). PD patients demonstrating executive impairments exhibited lower prefrontal cortex activity as measured by fmri during a working memory task (Lewis et al., 2003). An idea that grows out of this insight is that VTA dopamine loss contributes to nonmotor symptoms such as depression (Frosini et al., 2015). Many of the PD patients in pathological studies in the literature, die before non-motor symptoms are routinely investigated, thus rendering it impossible to assess the significance of VTA cell loss in PD. Neuroimaging studies, however have shed some light on this question by documenting abnormalities in mesocortical and mesolimbic circuits. For instance, Frosini et al. demonstrated decreased dopaminergic innervation of the anterior cingulate cortex, part of the limbic system, in PD patients with depression as compared to PD patients without depression, measured by dopamine transporter density (Frosini et al., 2015). Moreover, extensive animal work has indicated that dopamine dysfunction in mesocortical and mesolimbic pathways can consistently impair behaviors that are impaired in PD patients (Parker et al., 2013; Kim et al., 2017). PD is most commonly a sporadic or idiopathic disease having no singular cause, thus the etiology of this disease remains unclear as the disorder has proven to be complex and different causes have been identified. Varying underlying causes such as genetic mutations and environmental factors have been explored. Mutations in six genes that code for different proteins have been found among some individuals with PD. In particular, mutations in the Parkin, DJ-1, and PINK1 genes result in early-onset familial 6

22 PD (Gasser, 2005, 2009). Additionally, a few studies have reported that exposure to pesticides and wood preservatives is correlated to increased risk of developing PD, suggesting possible environmental causes (Jiménez-Jiménez et al., 1992; Le Couteur et al., 1999; Giasson and Lee, 2000). Pesticides such as rotenone and paraquat have been used to model the disease in rodents (Bové et al., 2005; Tanner et al., 2011). Regardless of the amount of information known to date and in question on the actual causes of this disorder, the pathways affected by PD are generally well known and studied. Levodopa and Levodopa-induced Dyskinesias From the description of PD by James Parkinson in 1817 it took over 100 years for the development and use of levodopa, a dopamine precursor, as the main dopamine replacement therapy in 1957 by Carlson and colleagues, who were later awarded the Nobel Prize in Medicine in Levodopa has turned out to be the most effective treatment for the motor symptoms of PD. Levodopa is readily transported across the blood-brain barrier, unlike dopamine itself, and taken up by the remaining dopaminergic terminals in the striatum where it is converted to dopamine by the aromatic L-amino acid decarboxylase enzyme and stored in vesicles. However, levodopa is also readily metabolized peripherally, thus each dose must be supplemented with a peripheral decarboxylase inhibitor, such as carbidopa, allowing more levodopa to reach the site of action in the brain (Figure 4). 7

23 Figure 4. Molecular structures of levodopa (L-DOPA), carbidopa, and dopamine. The left-side of the schematic depicts the inability of dopamine to cross the blood-brain barrier (BBB). However, levodopa can cross the BBB and is metabolized into dopamine in the brain. (From Anon, n.d.) Once in the brain, levodopa is synthesized into dopamine and is either kept or synthesized into norepinephrine via dopamine β-hydroxylase. In turn, phenylethanolamine N-methyltransferase converts norepinephrine into epinephrine (McKim, 2007). Dopamine has been studied extensively due to its neuromodulator role. The basal ganglia and dopaminergic neurons have been studied for their significant presence in motor function, memory, learning, and reinforcement effects (Wachter et al., 2009). As mentioned above, certain pathways in the brain use dopamine as their main neurotransmitter, including the nigrostriatal, the mesolimbic, and the mesocortical systems. In view of PD as a movement disorder, the nigrostriatal system is the main dopaminergic pathway studied. The somas of the dopaminergic cells in this system are located in the SN and they project their axons to the striatum, which is composed of the caudate nucleus, putamen, and nucleus accumbens. For the scope of this project, I will focus on the caudate and putamen as these are associated with movement, and will often 8

24 be referred to as the dorsal striatum versus the ventral striatum which contains the nucleus accumbens. The connection between the SN and the striatum is of importance because the striatum is a key component of the basal ganglia and it is implicated in motor function, which is disturbed in PD (Carlson, 2010). In addition, the mesolimbic system begins in the VTA and terminates in the nucleus accumbens, amygdala, and hippocampus. This system is concerned with the effects of reinforcement and memory forming (Wachter et al., 2009; Chakravarthy, Joseph, & Bapi, 2010). Similarly, the mesocortical system has somas in the VTA but the axons project to the prefrontal cortex playing a role in executive function which is disrupted in PD. Disrupting this system either by depleting the VTA of dopamine or by blocking dopamine receptors in the prefrontal cortex led to disruptions in timing tasks in rodents, a measure of executive function (Parker et al., 2013; Kim et al., 2017) Though levodopa remains the most effective drug treatment, a major limitation is that long-term use leads to levodopa-induced dyskinesias (LIDs). LIDs are disruptive and debilitating involuntary movements that drastically impair patients quality of life. As with PD, there are several scales used to measure LIDs: 1) modified abnormal involuntary movement (m-aim) scale, 2) Rush Dyskinesia Rating Scale, 3) Lang-Fahn Scale, 4) United Parkinson s Disease Rating Scale Part IV, 5) Obeso Scale, and most recently developed 5) the United Dyskinesia Rating Scale. Some of these scales rely largely on patient or caregiver recounting of events in a journal (Parkinson Study Group, 2001; Siderowf et al., 2002). Other rating scales require the patient to go to a neurology clinic and be seen and tested by a physician (Guy and National Institute of Mental Health (U.S.). Psychopharmacology Research Branch. Division of Extramural Research 9

25 Programs, 1976; Goetz et al., 1994). Three subtypes of LIDs exist: axial, limb, and orofacial. Axial dyskinesia is dystonic in nature whereas limb and orofacial are hyperactive and present as chorea. These uncontrollable movements may also interfere with daily life. LIDs are correlated with levodopa dose and treatment duration. Additionally, quality of life in PD patients is impaired by disease progression and how long they have suffered with the disease (Karlsen et al., 2000; Zach et al., 2004). Therefore, it is not surprising that PD patients with LIDs report lower quality of life and LID severity negatively impacts quality of life even further (Péchevis et al., 2005). As disease progresses, levodopa treatment continues, and LIDs worsen patients begin to experience interference with day to day activities, loss of autonomy due to mobility impairments that can lead to falls, and issues with communication all attributing to decreased quality of life (Damiano et al., 2000; Miyasaki et al., 2002; Chapuis et al., 2005). As discussed above, PD is a progressive disease. Higher doses of levodopa have been correlated with more severe LIDs; thus, physicians typically begin treatment at the lowest effective dose and titrate up as patients endogenous dopamine levels continue to decrease (Fahn et al., 2004). Despite efforts to mitigate this debilitating side effect, up to 80% of patients develop LIDs (Ahlskog and Muenter, 2001). Once LIDs develop, further treatment options are extremely limited. To date, the only drug available to treat LIDs is amantadine, a glutamatergic antagonist, but the effects are variable (Snow et al., 2000; Thomas et al., 2004; Rodnitzky et al., 2014). Other drugs such as serotonin blockers have also been introduced in clinical trials but, again, the results are inconclusive (Durif et al., 1995; Chung et al., 2005). Another alternative for patients who do not respond to 10

26 pharmacological treatment or develop severe side-effects is deep brain stimulation (DBS). However, not all patients qualify for DBS and it requires an invasive procedure where patients undergo neurosurgery to have electrodes placed either in the subthalamic nucleus or globus pallidus (internal segment). Additionally, after several years of DBS patients begin to have postural instability and adverse effects (Baba et al., 2012). DBS is thought to work by inserting high enough frequencies into the subthalamic nucleus, thereby inhibiting (or jamming ) the output of this nucleus and alleviating parkinsonian symptoms. It remains to be studied what really occurs in the brain when high frequencies are added. Animal Models The long-term effects of levodopa in PD patients remain inconclusive. Whether levodopa is detrimental or beneficial to PD was not clear in a human trial of PD (Fahn et al., 2004). A few risk factors for the development of LIDs have been reported, mainly age of onset, disease severity, levodopa dose, and length of levodopa treatment (Fahn et al., 2004). Though these factors are difficult to control for in humans, animal models of LIDs have proved to be essential in furthering the knowledge of these side effects. Much like PD patients treated with levodopa, animal models of LIDs develop abnormal involuntary movements (AIMS) divided into three subtypes and are dependent on the extent of dopamine depletion and levodopa dose (Lundblad et al., 2002, 2005). An overview of the most common animal models of LIDs is outlined below. 11

27 Rodent Model The rat and mouse 6-hydroxydopamine (6-OHDA) models are very similar so they will be discussed together. Though the rat 6-OHDA model is the most commonly used by research groups around the world, the mouse model has become increasingly popular with the development of transgenic lines. Two main strategies are used to acutely deplete animals of dopamine where a neurotoxin, 6-OHDA, is intracranially injected in the SNc, striatum, or medial forebrain bundle (MFB). The latter is used to model later stages of the disease with a near complete depletion of dopamine, yet lesions of the striatum are sufficient to lead to LIDs. 6-OHDA is a selective catecholaminergic neurotoxin that is readily taken up by the dopamine transporter, as it closely resembles dopamine. To protect other catacholaminergic (i.e. norepinephrine) neurons from the toxin, rodents are typically injected with desipramine, a norepinephrine and serotonin uptake inhibitor, before delivery of 6-OHDA preventing up-take in these neurons. Following recovery of 1-3 weeks, rodents are tested for lesion with various drug-induced rotation tests and other behavioral test that probe at preferences of limb use, that is, whether they preferentially use those on the ipsilateral or contralateral side of the lesion. Once lesion is confirmed animals are treated daily with levodopa for 2-3 weeks as AIMs develop. AIMs are divided into three categories: axial, limb, and orofacial (the 6-OHDA mouse model is discussed in more detail in Chapter 2). These AIMs are typically scored by human raters. Importantly, these are the same subtypes of LIDs described in PD patients that develop these side-effects (see Appendix for a full description of scoring). The 6-OHDA rodent model is useful in the study of LIDs as it unilaterally depletes 12

28 dopamine and provides a stark comparison between the lesioned and non-lesioned side leading to one-sided movements. Non-human Primate Model Though the rodent models have been instrumental in furthering the knowledge of PD and LIDs, they do have their limitations, mainly that they don t mimic symptom manifestation or the progressive degeneration of dopaminergic neurons. Another PD model that is commonly used and tends to resemble the disease more closely is the 1- methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) primate model. MPTP is also a neurotoxin selective for dopaminergic neurons, but unlike 6-OHDA, this compound can cross the blood brain barrier. The ability for MPTP to cross the blood brain barrier is beneficial in that it can be administered systemically but it also presents danger for the experimenters, thus more protective precautions must be taken. To induce dopaminergic cell death, animals are repeatedly administered with MPTP. This PD model has led to significant advances in the study of neuroprotection. Importantly, the non-human primate LID research along with the rodent models also heavily rely on human raters scoring the severity and relative presence of AIMs. Recently, a group studied the use of a quantitative video-based scoring method in the MTPT primate model to more objectively study LIDs (Potts et al., 2015). Studies such as this highlight how technological advances may aid in the clinic as not all patients will present with the same LID severity and expression of LID subtypes. 13

29 Possible Mechanisms Though progress has been made in exploring the cause of LIDs in these animal models, it remains unknown how they develop. Below is an overview of several mechanisms that have been proposed to be responsible for the development of LIDs. Serotonin During the progression of PD, there is a substantial loss of dopaminergic SNc neurons innervating the striatum (Hornykiewicz, 1998; Thomas and Beal, 2007; Factor and Weiner, 2008). However, serotonergic neurons are relatively spared (Halliday et al., 1990; Kish et al., 2008). There is a 70-90% decrease in dopamine levels in the striatum of post mortem PD brains (Hornykiewicz, 1998; Kish et al., 2008). By contrast, a 60% decrease in serotonin levels is observed (Kish et al., 2008). These data imply that serotonergic neurons are moderately spared in comparison (Kish et al., 2008). Interestingly, a number of striking similarities exist between dopaminergic and serotonergic neuronal populations. Both serotonergic and dopaminergic neurons contain an aromatic L-amino acid decarboxylase (Arai et al., 1994; Yamada et al., 2007; Kitahama et al., 2009). In serotonergic neurons, this decarboxylase is necessary for metabolizing tryptophan into serotonin, while in dopamine neurons this same enzyme converts DOPA into dopamine. In addition, dopaminergic neurons contain a vesicular monoamine transporter which allows for the storage of monoamines, specifically dopamine. Similarly, serotonergic neurons express this transporter, providing them with the machinery to store monoamines (Narboux-Nême et al., 2011). Importantly, as dopaminergic neurons degenerate during PD, the remaining serotonergic neurons retain 14

30 the ability to convert levodopa into dopamine and, subsequently, release the neurotransmitter (Arai et al., 1994; Maeda et al., 2005; Yamada et al., 2007). Microdialysis studies have demonstrated that under non-physiological conditions, such as dopamine depletion, serotonergic neurons release dopamine following L-dopa administration (Maeda et al., 2005; Carta et al., 2007). This capacity raises the intriguing possibility that serotonergic neurons may play a role in compensating for the loss of dopaminergic neurons in the midbrain of PD patients. Anatomical evidence lends credence to this hypothesis. For example, others have found that the rostral portion of the dorsal raphe nucleus projects to the striatum (Waselus et al., 2006). Further, ablation of serotonergic neurons projecting to the striatum significantly reduces LIDs in a rat model of PD (Carta et al., 2007). While the broad effects of serotonergic neurons on LIDs have been studied, the mechanistic underpinnings have not yet been established. Medium Spiny Neurons PD involves degeneration of midbrain dopaminergic neurons which heavily project to the striatum (Hornykiewicz, 1998; Alberico et al., 2015). Rodent and primate PD models also develop LIDs with chronic levodopa exposure; even in animals with dopamine depletion restricted to the dorsal striatum (Lundblad et al., 2004, 2005; Pavón et al., 2006; Santini et al., 2007). As mentioned above, the dorsal striatum is largely comprised of MSNs, which are the primary output neuron of the striatum and these express one type of dopamine receptors (Gerfen, 2000; Kreitzer, 2009). These two neuronal populations not only differ in the dopamine receptors they express, but also in the peptides they contain. For instance, D1-MSNs contain the peptides dynorphin and substance P while D2-MSNs contain enkephalin (Gerfen et al., 1990). Additionally, D1-15

31 MSNs communicate via Gs/olf while D2-MSNs communicate via Go/i proteins. In D1- MNSs, Gs/olf activity will lead to adenylyl cyclase activation which in turn increases cyclic adenosine monophosphate (camp) levels activating protein kinase A (PKA). On the other hand, in D2-MSNs activity of Gi proteins activate phospholipase C (PLC) and inhibits adenylyl cyclase. During the progression of PD, a loss of dopaminergic input to the striatum is evidenced, affecting both pathways and leading to the motor symptoms characteristic of PD. Decreased activity at the D1R and increased activity of D2-MSNs results in decreased movement (Jenner, 2008). In a disease state, that is with dopamine depleted from the SNc, studies have shown alterations in mrna expression encoding the two receptor types and the different peptides as well as morphology of the two MSN populations. In an in situ hybridization study, Gerfen and colleagues found that following dopamine depletion substance P mrna levels were decreased while mrna levels for enkephalin were decreased (Gerfen et al., 1990). Interestingly, researchers reversed these changes by administering either a D1 or a D2 agonists. Additionally, D2-MSNs lose dendritic spines from excitatory input and have a decreased intrinsic excitability but no similar change in dendritic spines was observed in D1-MSNs yet their intrinsic excitability was increased, possibly a homeostatic plasticity to compensate for the loss of inhibition from dopamine D2R and a loss of excitation to D1R (Day et al., 2006; Fieblinger et al., 2014). Furthermore, in dyskinetic mice these shift in intrinsic excitability were closer to normal levels observed in non-dopamine depleted animals (Fieblinger et al., 2014). 16

32 Although both pathways are modulated by dopamine and are affected throughout the course of PD, previous work strongly implicates the direct pathway in the development of LIDs. In MPTP-treated monkeys that develop LIDs heightened D1R mrna expression is reported when compared to monkeys that did not develop LIDs (Aubert et al., 2005). In this study, they also found that in the brain of these dyskinetic animals cyclin-dependent protein kinase 5 (Cdk5) and DARRP-32 levels were increased. These authors concluded that levodopa administration increases D1R expression and sensitivity as well as D1 binding in the striatum (Aubert et al., 2005). Hanrieder et al. demonstrated increases in proteins specifically expressed by D1 neurons in the striatum of rats with severe dyskinesias (Hanrieder et al., 2011). Similarly, the increase in D1R expression in dyskinetic animals is accompanied by increases in postsynaptic density 95 protein (PSD95) which have been shown to modulate glutamatergic receptor trafficking (Nash et al., 2005). A study in non-human primates demonstrated that downregulation of PSD95 significantly decreased LID expression and that PSD95 and the D1R coimmunoprecipitate (Porras et al., 2012). Finally, Darmopil et al. suggested that mice lacking the D1R do not develop LID (Darmopil et al., 2009). Fast-spiking interneurons Although comparatively rare, fast spiking interneurons (FSIs) are positioned to powerfully influence MSNs and in turn affect movement (Berke, 2011). Though, FSIs comprise fewer than 5% of the neurons in the striatum, one FSI can make inhibitory synapses with hundreds of MSNs, on the other hand, one MSN can receive inhibitory input from about 30 FSIs. FSIs are also modulated by dopamine as they are thought to express the D5-dopamine receptor and by a D2-mediated inhibition. Both MSNs and 17

33 FSIs receive glutamatergic input from the cortex and thalamus. Pharmacological studies show that inhibition of glutamatergic input to FSIs is sufficient to cause abnormal involuntary movements, similar to the axial dyskinesias seen in animal models of LIDs and in humans (Gittis et al., 2011b). FSIs can powerfully modulate the firing of MSNs (Koós and Tepper, 1999; Planert et al., 2010). As FSIs have dopamine receptors and provide strong feedforward inhibition of MSN activity, they may play a role in dyskinesias (Bracci et al., 2002; Centonze et al., 2003; Koos et al., 2004; Gittis et al., 2011b). Dopamine depletion can have complex effects on MSNs and FSIs as a result of altered input to striatum and/or remodeling of intrastriatal networks (Prosperetti et al., 2013; Corbit et al., 2016; Kondabolu et al., 2016). Prior work on FSIs describe that under physiological conditions FSIs tend to preferentially target D1 MSNs (Gittis et al., 2011b), but following dopamine depletion FSIs showed higher connectivity to D2 MSNs (Gittis et al., 2011a). Cholinergic Interneurons Though cholinergic interneurons (ChI) in the striatum are relatively few (1-3%), they are comprised of large and dense axons that perfectly position them to modulate MSNs (Zhou et al., 2002; Bonsi et al., 2011). Cholinergic interneurons have large cell bodies and their axons mainly synapse on MSNs, though some synapses are formed with FSIs (Bolam et al., 1984; Koós and Tepper, 2002). Two types of cholinergic receptors exist, one is simply an ion channel, the nicotinic receptor, while the muscarinic receptor is a G-coupled protein receptor. Both of these receptors are expressed in the striatum. Five muscarinic receptor subtypes have been established, M1-5, and these are widely 18

34 expressed in MSNs, ChIs, and GABAergic interneurons. The M1 receptor, a Gq-coupled receptor, is expressed in all MSNs while the M4 receptor, a Gi-coupled receptor, is expressed exclusively in the D1- MSNs. Interestingly, ChIs express both the D2- and D5- dopamine receptors, where the D2 has its typical inhibitory effect while the D5 receptor leads to excitation. Because of their fortunate location and presence in the striatum, researchers have explored the role of these neurons in PD and LIDs. Interestingly, selective ablation of cholinergic interneurons in the striatum of a ChAT-cre mouse led to a decrease of LIDs in the 6-OHDA model (Won et al., 2014). Additionally, one of the cardinal symptoms of PD is postural instability, which typically worsens as disease progresses leading to falls. Approximately 66% of PD patients experience falls each year, which leads to what has been coined fear of falling (Ashburn et al., 2001; Wood et al., 2002). Naturally, these experiences further decrease the quality of life of patients. Acetylcholine is also significantly decreased as the nucleus basalis of meynert degenerates, further proving that PD is a multifaceted disease (Nakano and Hirano, 1984; Fujita et al., 2006). Researchers have shown that anticholinergic drugs worsened balance and increased falls. Researchers have looked at the potential of cholinergic drugs in ameliorating the postural instability seen in PD patients and found that acetylcholinesterase inhibitors significantly decreased the number of falls experienced by patients when compared to a placebo (Chung et al., 2010). At the molecular level, phosphorylation of ERK1/2, which occurs following administration of a dopaminergic agonist such as levodopa, has been shown to occur in MSNs during the development of LIDs, but once established, this ERK phosphorylation 19

35 occurs in ChIs instead. Consequently, inhibiting ERK phosphorylation attenuates LIDs (Ding et al., 2011). Suggesting that ChI may be modulating LIDs. Experimental Design To further the knowledge on the mechanism of LIDs, we used the 6-OHDA MFB lesion mouse model (Figure 5). This model mimics the substantial loss of DA which is common in an advanced disease and allows researchers to see the effects of different treatments as it is unilateral and severe. Following lesion, animals are primed with levodopa for two weeks (Cenci and Lundblad, 2007). Figure 5: Timeline of experimental design for LID development. Mice undergo dopamine depletion via an injection of 6-OHDA into the medial forebrain bundle (MFB). One to three weeks later mice undergo a battery of motor tests to assess lesion. Once lesion is confirmed mice are injected daily with levodopa for 13 days after which LIDs should be stable. Following LID development,t a series of pharmacological or optogenetics sessions were run to establish the role of the D1-MSNs in LIDs. In the first part of this project we investigated the development of LIDs and the role of the MSNs. We tested the hypothesis that MSNs were modulated around times of LIDs. We explored this possibility by recording from neuronal ensembles in the rodent dorsal striatum as animals developed LIDs; and found that MSNs significantly change their firing rates from the first day of levodopa to day 13 of levodopa whereas FSIs did not. Additionally, both neuronal populations are strongly modulated around the time of LIDs. 20

36 In the second set of experiments, we investigated the role of D1R and D2R expressing neurons in LIDs. Based on the literature, we hypothesized that inhibiting D1- MSNs, but not D2-MSNs, would decrease LIDs. Again, we interrogated this hypothesis with striatal neuronal ensembles in mice with LIDs. Indeed, we found that blocking D1R decreased LIDs strongly suggesting that the neurons expressing D1R are responsible for LIDs. In the last part of this project, we tested the hypothesis that D1-MSNs are sufficient to cause dyskinesia. On the other hand, other neurons expressing the D1- dopamine receptor could be contributing to LIDs. We distinguished between these possibilities by recording from neuronal ensembles in the striatum while stimulating or inhibiting these neurons, and found that activating D1R expressing neurons in the striatum led to abnormal involuntary movements. To date, there are no clinically approved drugs targeting the D1 dopamine receptor. The purpose of this project was to define the role of D1-MSNs in LIDs to provide a novel therapeutic target in the treatment of LIDs. We hypothesized that stimulation of D1-MSNs would be sufficient to cause dyskinesias and that inhibiting these neurons would decrease LIDs. These data may not only be important to helping PD patients with LIDs, but may also further our understanding of other movement disorders and other disorders with disrupted dopamine signaling. Furthermore, the results from this thesis shed light on the circuitry of the development of LIDs and, thus, on the striatal changes in PD broadening the knowledge of the disease allowing for more optimal treatment. Crucially, the knowledge gained about the LID circuitry will aid in understanding disease progression in hopes of one day finding the causes of PD. 21

37 CHAPTER 2: AUTOMATED COMPUTER TRACKING Introduction A significant challenge to studying the mechanism of LIDs is that prior studies have human raters score abnormal involuntary movements as a measure of LIDs in rodents (Lundblad et al., 2002; Winkler et al., 2002; Smith et al., 2011; Breger et al., 2013). In this method, humans grade axial, limb, and orofacial dyskinesias over a 1-2 minute window every several minutes (Figure 6; Lundblad et al., 2005; Cenci and Lundblad, 2007; Santini et al., 2007). Figure 6. Example of abnormal involuntary movement scores as LIDs develop. Time profile of individual abnormal involuntary movement (AIM) scores measures by axial, limb, and orofacial (ALO) ratings following levodopa (L-DOPA) administration. Each line represents one testing session. From Cenci & Lundblad, As raters only get a 1-2 minute window, this technique is not suited to capture modulation of striatal neurons, which can be correlated with movements on the scale of milliseconds (Lundblad et al., 2004; Cenci and Lundblad, 2007; Breger et al., 2013; Sgroi 22

38 et al., 2014). To address these issues, we developed a 3-dimensional position tracking technique to capture animals position at ~10 ms and ~ 1 mm resolution. Methods We used 9 C57/BL6 male mice (Harlan, Madison, WI), weighing >25g at the time of dopamine depletion. All procedures were approved by the Animal Care and Use Committee at the University of Iowa. We induced dyskinesias as previously described (Figure 7; Lundblad et al., 2004; Cenci and Lundblad, 2007). Briefly, mice were anesthetized using ketamine (100 mg/kg) and xylazine (10 mg/kg) and injected with desipramine (25 mg/kg; i.p.) to protect catecholaminergic neurons. First, we depleted dopamine in the medial forebrain bundle (MFB ) using the neurotoxin 6-OHDA bromide (Sigma, St. Louis, MO) made each surgery day at a concentration of 1 µg/µl dissolved in 0.02% ascorbic acid (AVANTOR, Central Valley, PA). Animals were unilaterally depleted of dopamine with 1µg of 6- OHDA stereotaxically injected into the MFB (AP: -1.2, ML: -1.2, DV: -4.7 from dura). Figure 7. Automated tracking in mouse model of levodopa-induced dyskinesias. A) To model dyskinesias, we depleted dopamine with a medial forebrain bundle (MFB) 6-OHDA lesion and injected levodopa (20 mg / kg) for two weeks. We recorded striatal neuronal ensemble activity on Day 1, 7, and 13 of levodopa administration. From Alberico et al., Following recovery, lesions were confirmed with the amphetamine-induced rotation test. Animals were then injected with 20 mg/kg of levodopa for 13 days (Figure 23

39 7). Starting on Day 1 with baseline recording before levodopa administration, behavior was recorded via OptiTrack Prime 13 cameras (NaturalPoint Inc, Corvallis, OR) using Motive OptiTrack software and a video camera at the beginning and end of treatment (Day 1 and 13) to measure the development of LIDs. The OptiTrack cameras were mounted on tripods with a height of 34 cm from recording surface (Figure 8A). These cameras were calibrated on each recording day using a wand (250 mm; NaturalPoint Inc) consisting of three infrared reflective spheres at a fixed distance from one another in a straight line. The filter switch on Motive OptiTrack was set to infrared spectrum, the gain was set to low for short range, and the cameras were set to object mode. At least 5000 samples were gathered to calibrate the cameras. A sample is registered when two or more cameras can view all three spheres. Following wand calibration, the ground plane (X, Y, Z directions) was set by placing a calibration square (NaturalPoint Inc) with three infrared reflective spheres. Wand calibration was applied and converted to meters and the output had a Figure 8. Tracking of axial position with infrared cameras. A) Inset: Picture of two infrared reflective spheres and recording electrode. Right: Diagram of 4 infrared cameras calibrated to track movements at 120 frames/s. B) Tracking examples (2 s) in a dopaminedepleted animal and C) the same animal with initial levodopa administration in the X (blue), Y (green), and Z (black) axes with corresponding still images. From Alberico et al.,

40 mean 3-dimensional error of <0.03 mm (calculated by Motive Optitrack from the wanding results of each camera). Two 4-mm infrared reflective spheres were attached to the headstage to determine animals head position. The animals were placed in a transparent cylinder (15 cm diameter) 42 cm from each camera (Figure 8A). Motive OptiTrack was synced with the video camera. The Motive Optitrack software tracks movements along the X, Y, and Z axes. Position data were exported as CSV files using Motive OptiTrack. Basic tracking of two infra-red reflective spheres on the animal s head was done in Motive. When Motive lost tracking of the spheres due to reflection from the recording chamber, coordinates of the spheres were co-registered in MATLAB using a nearest neighbor algorithm. The distance between the spheres in 3D space was calculated between n and n+1 frame, and minimal distance change per each detected sphere was selected as n+1 tracked sphere. Frames with no detected spheres were linearly interpolated when continuous missing frames were less than 1 second; otherwise these frames were excluded from behavioral analysis. A total of 96.9±0.8% frames were tracked per each session and linear interpolation filled in 2.8±0.7% of missing frames. Rapid and continuous head rotation movements were identified as characteristic of axial dyskinesia-associated movements based on hand-scored dyskinesias. Head rotation was identified by simultaneous velocity changes in X and Y axes. Tracked sphere coordinates were filtered with a low-pass FIR filter at 4Hz to remove high frequency noise. Velocity was calculated between frames on the X and Y axes, then dyskinesia-associated head rotations were identified using findpeaks function in MATLAB. The threshold parameters for rotational movement were minimum peak 25

41 prominence (continuous velocity changes) at and minimum peak height (velocity) at mm/sec. We used amplitude based scoring to characterize LIDs in our mouse model (Lundblad et al., 2002; Winkler et al., 2002; Smith et al., 2011; Breger et al., 2013). Briefly, animals were assessed for two minutes every 10 minutes for an hour and each LID subtype was given two scores, one for duration of dyskinesias (0-4) and one for severity (0-4). These scores were then multiplied and axial, orofacial, and limb LID subtypes were added to obtain the integrated AIM scores at each time point (Figure 9A, line graph). To obtain a global score, integrated AIM scores were summed for each recording session (Figure 9A). Results All mice had two 4 mm infrared-reflective spheres attached to the recording headstage in the anterior-posterior dimension (Figure 8A). Four infrared cameras recorded the X (right and left), Y (forward-back), and Z (up-down) coordinates of the mouse s head at 120 frames/per second (frames/s) to track head position (Figure 8A). Automated computer tracking data was synchronized with a video camera at 30 frames/s and Plexon neurophysiological recording hardware (Figure 8B-C). To examine how movement tracking captured dyskinetic movements, we analyzed movement videos to identify dyskinetic movements by hand, and AIM scoring was performed by human raters in LID sessions (Figure 9A) (Winkler et al., 2002; Lundblad et al., 2004, 2005; Cenci and Lundblad, 2007; Bido et al., 2011; Smith et al., 2011). AIM scoring indicated that LIDs developed and increased severity over two weeks of levodopa administration in 6-OHDA-lesioned mice (t(4)= 4.71, p< 0.01; Day 1 vs. Day 13) but not in sham-lesioned 26

42 mice (Figure 9A). We recorded the timestamps of hand-coded LIDs along with tracking data (Figure 9B). By aligning automated tracking to video and hand-scoring, we found that axial LIDs involved large changes in the X and Y axes, with smaller changes in the Z axis (Figure 9B-C). Figure 9. Automated tracking of dyskinetic events. A) With levodopa administration for two weeks, levodopa-induced dyskinesias increased over time by abnormal-involuntary movement (AIM) scoring (N= 5 mice; **: p<0.01). B) Example traces from one mouse 2 s before and after a single hand-coded dyskinetic movement on all three axes (X: blue, Y: green, Z: black). C) Displacement observed in position around the computer-identified axial dyskinetic movements based on X- and Z- axis characteristics in all three axes (X: blue, Y: green, Z: black) compared to shuffled events (noise, in grey; based on randomly selected timestamps). Error bars are mean ± SEM. From Alberico et al., By examining patterns of movement in 3D around hand coded dyskinetic events, we observed that axial dyskinesias were associated with rapid changes in velocity and angular position, identified by peaks in the velocity record in X and Y axes. These were denoted as computer-identified axial dyskinesias. Around these computer-identified axial dyskinesias, there were large changes in X/Y/Z position (Figure 9C). 27

43 As dyskinesias developed with two weeks of levodopa administration, average velocity increased, but not acceleration or angular velocity (F(2,14)= 4.8, p<0.05; Figure 10A-B). Consistent with AIM scoring, there were more computer-identified axial dyskinesias on Day 13 vs. Day 1 of levodopa (F(2,14)= 4.6, p<0.05; Figure 10C) and these increases were significantly correlated with AIM scores (R=0.73, p<0.01; Figure 10D). These data indicate that computer-identified axial dyskinesias correlate with AIM scores, which have low temporal resolution (Breger et al., 2013; Sgroi et al., 2014). Figure 10. Computer identified axial dyskinesias are correlated with AIM scores. A) Around computer-identified events, there were large changes in velocity, acceleration, and angular velocity. B) Average velocity increased as LIDs developed, while acceleration and angular velocity did not. C) Computer-identified dyskinetic events increased as LIDs developed. D) Computer-detected LIDs and AIM scores were significantly correlated. These data indicate that automated motion tracking can capture dyskinesias. Error bars are mean ± SEM. (*: p<0.05) From Alberico et al., Indeed, around axial dyskinesias, automated tracking captured patterns of large displacements and faster movements. Automated tracking revealed that dyskinetic sessions are characterized by more overall movement and higher average velocity. Thus, such automated tracking techniques could be useful in capturing dyskinetic movements in 28

44 mouse models of LIDs with high temporal resolution and linking these movements to striatal neuronal activity. Summary We developed a novel automated tracking system to capture dyskinetic movements with high temporal resolution, allowing us to correlate dyskinesias with neuronal activity. One important advance in our work is automated tracking with high temporal (<10 ms) and high spatial (<1 mm) resolution. Previous studies have largely relied on AIM scores, which can be laborious and temporally limited (Lundblad et al., 2004). Tracking techniques have been used to correlate movements with striatal neurophysiology (Barter et al., 2015a,b), and here we extend this work to LIDs. Our results are biased toward axial dyskinesias as the tracking device was placed on the animals midline. However, because axial dyskinesias are a large component of LIDs, the present results are representative of striatal changes with the development of LIDs (Winkler et al., 2002). 29

45 CHAPTER 3: STRIATAL NEURONS IN THE DEVELOPMENT OF LIDs Introduction Dopamine increases excitability in MSNs expressing D1-type dopamine receptors, promoting movement, and decreases excitability in MSNs expressing D2-type dopamine receptors (Surmeier et al., 2007; Cui et al., 2013). PD involves a progressive loss of dopaminergic input to the striatum, disrupting both D1 and D2 MSNs leading to the motor symptoms of PD (Bunney et al., 1973; Hernandez et al., 2013). Here, we examine the neuronal activity of dorsal striatal neurons in LIDs. We record from MSNs, FSIs, and striatal local field potentials (LFPs) as dopamine-depleted mice develop LIDs. Striatal LFPs are of particular translational significance because they can be directly measured in humans via intracranial recordings. While these LFPs have been implicated in PD it is unknown how they change as LIDs develop (Brown et al., 2001; Williams et al., 2002; Brown and Williams, 2005). A significant challenge to resolving these issues is that prior studies have human raters score abnormal involuntary movements as a measure of LIDs in rodents (Lundblad et al., 2002, 2002; Smith et al., 2011; Breger et al., 2013). In this method, humans grade axial, limb, and orofacial dyskinesias over a 1-2 minute window every several minutes (Lundblad et al., 2005; Cenci and Lundblad, 2007; Santini et al., 2007). This technique is not suited to capture modulation of striatal MSNs, which can be correlated with movements on the scale of milliseconds (Breger et al., 2013; Sgroi et al., 2014; Barter et al., 2015b). To address these issues, we developed a 3-dimensional position tracking technique to capture animals position at ~10 ms and ~ 1 mm resolution while recording neuronal activity from the dorsal striatum of mice during the development of LIDs (Chapter 2). 30

46 Methods We used 9 C57/BL6 male mice (Harlan, Madison, WI), weighing >25g at the time of dopamine depletion. Since males develop more quickly and gain weight more quickly than females and had lower mortality rates, I used males for all of these experiments. All procedures were approved by the Animal Care and Use Committee at the University of Iowa. We induced dyskinesias as previously described in Chapter 2 (Lundblad et al., 2004; Cenci and Lundblad, 2007). Briefly, mice were anesthetized using ketamine (100 mg/kg) and xylazine (10 mg/kg) and injected with desipramine (25 mg/kg; i.p.) to protect Figure 11. Mouse model of levodopa-induced dyskinesias. Location of MFB 6-OHDA lesion (left) via immunohistochemistry resulting in large loss of striatal dopamine in the dorsal striatum, where electrodes were implanted (right; TH in red; DAPI in blue). Each white circle represents electrode placement of one animal. Bar graph shows quantification of TH in the substantia nigra pars compacta. On average, animals had 71±5% less TH positive cells on the lesioned side (red) when compared to the contralateral side (black). Data from 5 lesioned mice. (*) p<0.05. From Alberico et al., other catecholaminergic neurons. Animals were unilaterally depleted of dopamine with 1µg of 6-OHDA stereotaxically injected into the MFB (AP: -1.2, ML: -1.2, DV: -4.7 from dura). We found that 1 µg of 6-OHDA reliably caused 71±5% (mean±sem) dopamine depletion in the striatum (Figure 11). With levodopa administration this dopamine-depletion protocol maximized LIDs and minimized mortality of mice with 31

47 electrode implants. In these animals, a 16 channel stainless steel microwire array (50 µm, 4x4; MicroProbes, Gaithersburg, MD) was lowered into the dorsal striatum (AP: +0.1, ML: -2.0, DV: -3.0; Figure 11) while measuring neuronal activity. To ground the electrode, the stainless steel ground wire was wrapped around two skull screws, which were placed over the contralateral lateral surface of the skull between the bregmatic and lambdoid sutures. The craniotomy was sealed and implants were fixed in place with cyanoacrylate ('SloZap', Pacer Technologies, Rancho Cucamonga, CA) and methyl methacrylate (dental cement; AM Systems, Port Angeles, WA). In addition, an implant was designed to attach two 4 mm infrared reflective spheres to the recording headstage (Chapter 2). Following a week of recovery, animals were tested for unilateral depletion with the amphetamine-induced rotation test. Animals were injected with amphetamine (5 mg/kg; i.p.) and ipsilateral rotations were recorded 30 min post-injection (Healy-Stoffel et al., 2012; Chotibut et al., 2013). Animals that did not show >5 ipsilateral rotations per minute were not included in this study (Chang et al., 1999; Paquette et al., 2009); animals used in the study performed 11±2 ipsilateral rotations per minute. Only complete rotations (360 ) were counted. On the day following amphetamine-induced rotation testing, we began administering 20 mg/kg of L-DOPA-methylester (levodopa; Sigma) dissolved in 0.09% NaCl sterile saline at a concentration of 4 mg/ml with Benserazide (12 mg/kg; 2 mg/ml; Sigma) daily. Striatal neuronal activity was recorded on Day 1 and Day 13 of levodopa administration while tracking movements. 32

48 Neuronal ensemble recordings in the striatum were made using a multi-electrode recording system (Plexon, Dallas, TX). Putative single neuronal units were identified online using an oscilloscope and audio monitor. Plexon Off-line Sorter was used to analyze the signals after the experiments and to remove artifacts. Spike activity was analyzed for all cells that fired at rates above 0.1 Hz. Statistical summaries were based on all recorded neurons. Principal component analyses (PCA) and waveform shapes were used for spike sorting. Single units were identified as having 1) consistent waveform shape, 2) separable Figure 12. Classification of striatal neurons. A) Example waveforms of one MSN (red) and one FSI (blue) recorded from a single electrode. B) Clustering across peak-to-trough duration and half-peak-width identified 240 MSNs (red) and 146 FSIs (blue). From Alberico et al., clusters in PCA space, and 3) a consistent refractory period of at least 1 ms in interspike interval histograms. All neurons were analyzed offline using Off Line Sorter (Plexon) and MSNs and FSIs were identified by waveform shape, half-peak-width, peak-to-trough duration, and firing rate (Figure 12). Analysis of neuronal activity and quantitative analysis of basic firing properties were carried out using NeuroExplorer (Nex Technologies, Littleton, MA), and MATLAB. LFP channels were filtered between 0.7 and 1000 Hz online, sampled at 1000 Hz and recorded in parallel with single unit 33

49 channels using a wide-band board. MSNs and FSIs were clustered using Gaussian mixture models. These parameters Field potentials (4 per animal) were low-pass filtered at 0.5 Hz and high-pass filtered at 50 Hz using EEGlab s eegfilt. Spectrograms were calculated using EEGlab s spectopo using the entire recording session (Delorme and Makeig, 2004; Emmons et al., 2016). Power was compared via a paired-t test between sessions in four frequency bands: delta (1-4 Hz), theta (4-8 Hz), beta (12-25 Hz), and gamma (25-95 Hz). Spike-field coherence was calculated using neurospec s sp2a_m1 using type 0 analysis, and a segment power of 11 with the average LFP per each animal according to methods described in detail previously (Rosenberg et al., 1989; Halliday et al., 1995, 1998; Parker et al., 2014). Coherence values were normalized to the 95% confidence interval for each neuron to facilitate comparisons across animals. All statistics assumed each recording day was an independent sample; no statistical dependence between days was considered and the same neurons were not explicitly tracked across days. After the completion of the experiments, mice were euthanized by injections of 100 mg/kg sodium pentobarbital, and transcardially perfused with 4% paraformaldehyde. Brains were post fixed in 4% paraformaldehyde and cryoprotected in 30% sucrose before sectioning in a cryostat. Brain slices were mounted and stained for tyrosine hydroxylase (TH; polyclonal rabbit anti-th, 1:1000; Millipore, Temecula, CA) and cell bodies using DAPI. Histological reconstruction was completed using post mortem analysis of lesion, electrode placement in the striatum, and cell concentration in the substantia nigra. Images were captured on Zeiss Apotome.2 Axio Imager and cells in the substantia nigra were 34

50 counted using the optical fractionator in Stereo Investigator for dopamine depletion quantification and analysis. Results Prior work demonstrates that striatal MSNs and FSIs encode aspects of movement (Berke et al., 2004; Berke, 2008; Hernandez et al., 2013; Jin et al., 2014; Barter et al., 2015a, 2015b). To examine striatal function during dyskinesias, we recorded striatal neuronal ensembles and tracked movement. In accordance with prior work, we used peak-to-trough duration and half-peak-width spike-waveforms to identify striatal MSNs and FSIs (Berke et al., 2004; Jin et al., 2014; Figure 12). Gaussian mixture models were able to cluster striatal MSNs and striatal FSIs (Figure 12). Across five 6-OHDA-lesion animals and four sham-lesion animals in three recording sessions (Saline, Day 1, and Day 13 of levodopa) we classified 240 neurons as MSNs and 146 as FSIs. Without levodopa administration dopamine-depleted animals had higher firing rates of MSNs (4.2 Hz ± 1.8) compared to sham-lesioned animals (1.2 ± 0.6; t(7) = 3.6, p<0.01; Figure 13). Figure 13. Striatal MSNs increase firing rate as LIDs develop. Left: Mean firing rate of MSNs increased following 6-OHDA lesion (Saline: sham-lesion (grey) vs 6-OHDA-lesioned in (red)) and as LIDs developed (Day 1 vs Day 13). Right: FSIs did not change as LIDs developed; data from 240 MSNs and 146 FSIs in 9 mice (5 dopamine-depleted, 4 sham-lesion) over 3 days. Error bars are mean ± SEM; (*) p<0.05. From Alberico et al.,

51 However, the firing rate of FSIs did not differ between dopamine depleted animals and sham-lesion animals (Figure 13). For MSNs, a repeated measures ANOVA revealed main effects of lesion (F(1,171) = 5.60, p < 0.02) and levodopa administration (F(1,171) = 23.0, p < 10-5 ) without interactions. The average firing rate of striatal MSNs was significantly higher on Day 13 (4.8 Hz ± 0.7) when compared to Day 1 (3.4 Hz ± 1.3) while FSIs on average were unchanged (Figure 13). We focused on Day 13 of levodopa sessions, as this time point had the most LIDs. In LID sessions after two weeks of levodopa injection (Day 13 of levodopa), we recorded 39 MSNs and 31 FSIs. Around axial dyskinesias, both striatal MSNs and FSIs had prominent modulations (Figure 14). We found no clear evidence that MSNs and FSIs had differential patterns of activity around LIDs. Similar number of MSNs and FSIs were significantly modulated around LIDs (pre: χ 2 =0.43, p=0.51; post: χ 2 =0.16, p=0.69). These data indicate that both MSNs and FSIs could be strongly modulated around dyskinetic events. 36

52 Figure 14. Striatal neurons are modulated around levodopa-induced dyskinesias. A) Example of an MSN and B) an FSI showing prominent firing rate modulation around computer-identified axial dyskinesias (0 s, Time from LID). Top portions are raster plots where each line represents a timecoded LID and each dot represents an action potential of the neurons. Bottom portions are the average firing rate for that neuron for all time-coded LIDs. C) Normalized firing rate changes for all MSNs and D) all FSIs around axial dyskinesias on Day 13. E) Quantification of the number of MSNs and F) FSIs that changed frequency around dyskinesias. Adapted from Alberico et al.,

53 Striatal LFPs reflect the integrated activity of striatal neurons. LFPs are prominently modulated by levodopa and could be a neurophysiological signature of LIDs that can be measured in humans (Brown et al., 2001; Brown, 2003; Brown and Williams, 2005). To investigate striatal LFPs in LID sessions, we recorded striatal LFPs and measured delta (1-4 Hz), theta (5-8 Hz), beta (12-24 Hz), and gamma (25-95 Hz) frequency bands. Around computer-identified axial dyskinesias on Day 13, striatal LFPs had significant delta, theta, and beta modulations around axial dyskinesias (compared to random events; Figure 15A-B). Delta, beta, and gamma power significantly increased from Day 1 vs Day 13 of levodopa (delta: paired-t(19) = 3.4, p < 0.003; beta: paired-t(19) = 2.8, p < 0.01; gamma: paired-t(19) = 3.5, p < 0.003; data from 20 channels in 5 mice; Figure 15 C-D). Additionally, we found that the Hz power in dyskinetic animals was significantly higher (Day 13 vs Day 1; paired t(19) = 5.6, p < 10-5 ), similar to that previously reported in the motor cortex in dyskinetic rats (Halje et al., 2012). The LFPs in the sham lesion animals did not change from Day 1 to Day 13 of levodopa injections (Figure 15). These data implicate delta, beta, and gamma bands in axial LIDs. 38

54 Figure 15. Delta power in striatal local field potentials increases as LIDs develop. A) Voltage modulation of LFPs around axial dyskinesias (0 s, Time from LID). B) Time-frequency plot of LFPs around axial dyskinesias. Spectral power of LFP activity revealed significant modulations in delta, theta, and beta bands around axial dyskinesias vs. shuffled events (outlined in black). C) Spectral power of striatal LFPs on Saline (black), Day 1 (pink), and Day 13 (red) of levodopa administration. D) Error bar of average normalized power (db) of 6-OHDA-lesioned mice (red) and sham-lesioned mice (grey) in each frequency band across days (Day 1 and Day 13). Delta, beta, and gamma power significantly increased as LIDs developed in 6-OHDA-lesioned mice (red). No changes were seen in mice with sham lesions (grey). Data from five 6-OHDA-lesioned and four sham-lesioned mice. Error bars are mean ± SEM. (*) p<0.05. From Alberico et al., Next, we examined how spectral power in striatal LFPs was related to individual neuronal activity using spike-field coherence (Halliday et al., 1998). This technique allows us to investigate the evolution of coupling between individual neurons and striatal LFPs (Halliday et al., 1998; Narayanan et al., 2013; Parker et al., 2014). We normalized coherence values to the 95% confidence interval to compare raw spike-field coherence; thus a scaled coherence measure above 1 is significant at p<0.05 (Parker et al., 2015). This measure revealed significant average delta spike-field coherence for FSIs but not for MSNs around axial LIDs (Figure 16). Across all ensembles, we found that delta coherence in FSIs increased on Day 13 of levodopa administration compared to Day 1 39

55 (Figure 16A-D; χ 2 =5.8, p<0.02), whereas theta, beta, and gamma coherence did not. No changes were observed in MSNs on Day 1 vs Day 13 of levodopa, although levodopa administration did decrease delta spike-field coherence for MSNs (Saline vs Day 1, χ 2 =4.7, p<0.03; Figure 16). Taken together, these data implicate delta-band striatal LFPs and delta spike-field coherence in dyskinesias and provide some insight into striatal network function in LIDs. Figure 16. Striatal FSI delta spike-field coherence increases as LIDs develop. Normalized spike-field coherence around axial dyskinesias for A) MSNs and B) FSIs. All spike-field coherence was normalized to 95% significance; thus, yellow and red indicate significant spike-field coherence. Data from 39 MSNs and 31 FSIs in five 6-OHDA-lesioned mice on Day 13 of levodopa. C) Initial administration of levodopa (Day 1) significantly reduced delta spike field-coherence in MSNs when compared to Saline but no change was detected as dyskinesias developed. D) FSIs significantly increased delta coherence as dyskinesias developed. Data from 5 mice. (*) p<0.05. From Alberico et al.,

56 Summary We recorded from striatal neurons in unilaterally dopamine-depleted mice as they developed LIDs. We developed a novel automated tracking system to capture dyskinetic movements with high temporal resolution, allowing us to correlate dyskinesias with neuronal activity. Striatal MSNs increased activity as dyskinesias developed, but both striatal MSNs and FSIs were modulated around dyskinetic events. Furthermore, striatal LFPs increased delta and theta activity as dyskinesias developed, and striatal FSIs but not MSNs developed delta coupling during dyskinesias. To our knowledge, these are the first data to describe striatal network activity as dyskinesias develop, and suggest that dyskinesias change striatal network activity. This observation could help understand the disruption of striatal circuits during dyskinetic movements. 41

57 CHAPTER 4: THE ROLE OF D1- AND D2-DOPAMINE RECEPTORS IN LIDs Introduction As mentioned in Chapter 1, the striatum is mainly characterized by two families of dopamine receptors are the D1- and D2-type receptors, which are all G-protein coupled receptors. The D1-type consist of the D1 and D5-dopamine receptors which are excitatory in nature coupled to the Gs and Golf proteins, while the D2-type consist of the D2, D3, and D4 receptors, which decrease neuronal activity and are coupled with Go and Gi proteins (Neve et al., 2004). Nearly all neurons in the striatum express one type of dopamine receptor. As MSNs are the primary output of the striatum and consist of approximately 95% of the neurons in the striatum, here we pharmacologically interrogated the role of D1- and D2-MSNs in LIDs. The response to dopamine in these two types of neurons is different. In D1-MSNs dopamine has an excitatory effect leading to release of GABA in the SNr, inhibiting this nucleus, which in turn disinhibits the thalamus allowing thalamic input to the motor cortex. However, in D2-MSNs dopamine has an inhibitory effect leading to a disinhibition of the GPe allowing for the inhibition of the STN, which then allows the GPi to run its course inhibiting the thalamus. Thus, a D1 agonist should increase the activity of D1-MSNs while a D2R agonist would decrease the activity of D2-MSNs. Through the same idea, a D1 antagonist should decrease D1-MSN activity while a D2 antagonist will increase the firing rate of D2-MSNs. These two receptor types have given rise to the canonical direct and indirect pathways, with MSNs either expressing the D1R or D2R, respectively. Activity in the direct pathway promotes movement while activity of the indirect pathway inhibits movement. 42

58 Previous work in animals with LIDs report that the D1R is upregulated making these neurons hypersensitive and pharmacological studies suggest that D1 receptor antagonists decrease LIDs (Aubert et al., 2005; Taylor et al., 2005). However, these studies did not investigate how activity in MSNs changes. Here, we hypothesized that inhibiting the D1-dopamine receptors, and not the D2 receptors, would significantly decrease LIDs and the firing rate of MSNs. Methods We used the same methods as in Chapter 3. Briefly, mice were unilaterally depleted of dopamine with the neurotoxin 6-OHDA and implanted with a 16 channel stainless steel microwire array into the dorsal striatum and with an implant to attach two 4 mm infrared reflective spheres to the recording headstage (Chapter 2), allowed to recover, and injected daily with levodopa (i.p.) for two weeks in order to establish LIDs (please refer to Chapter 3 Methods for a detailed description). Following LID development, we ran a series of pharmacological tests to interrogate the role of the two main dopamine receptors in the striatum, the D1R and D2R. Based on the results from Chapters 2 and 3 that we can reliably measure axial dyskinesias in this model of LIDs, we used the automatic tracking as the main technique to asses LIDs. Neuronal recordings and neuron clustering was done as in Chapter 3. Comparisons for LIDs and neuronal recordings were made between levodopa, levodopa with SCH23390, and levodopa with sulpiride. Additionally, controls consisted of shamlesioned animals injected with levodopa for 14 days (as with experimental animals). 43

59 Results As discussed in the introduction, the direct and indirect pathways are thought to work in opposition with the direct pathway promoting movement while the indirect pathway inhibits movement. Previous behavioral studies have shown the effects of inhibition in the two pathways (Taylor et al., 2005). Here, we investigated the role of the two main dopamine receptor types in the expression of LIDs by co-injecting levodopa either with a D1R antagonist, SCH23390 (1 mg/kg; i.p.), or a D2R antagonist, sulpiride (1 mg/kg; i.p.). We used the automatic tracking of LIDs and we recorded from striatal neuronal ensembles to measure changes in striatal neurons and LFPs. Blocking D1R expressing neurons decreased the number of automatically detected LIDs (Figure 17). Here, we report percent from levodopa alone, as each animal had a different starting point of maximal LIDs. When compared to levodopa alone, coadministration of levodopa with SCH23390 had significantly fewer LIDs detected in mice with established LIDs (t(4)=3.87, p= 0.018). However, co-administration of levodopa and sulpiride did not significantly change the number of LIDs recognized by the tracking system (t(4)= -1.85, p= 0.13). Previous work looking at hyperactivity in rats treated with methamphetamine and apomorphine found and that injections of sulpiride into the dorsal striatum increased hyperactivity providing a possible explanation for the results we observed (Koshikawa et al., 1987). To ensure that we were not reducing overall movement we measured total distance traveled and did not find that administration of D1 or D2 antagonist changed total distance traveled. Thus, the effects observed on automatically detected LIDs is not a reflection of decreased overall movement. 44

60 Figure 17: Blocking D1R decreases LIDs. Automated tracking shows that co-administration of levodopa and a D1-dopamine receptor antagonist (levodopa + SCH23390, blue) but not a D2-dopamine receptor antagonist (levodopa + sulpiride, green) decreases the number LIDs when compared to levodopa alone (dotted line). Error bars are mean ± SEM; N= 5; * p < Striatal MSNs either express D1- or D2-dopamine receptors corresponding to either the direct or indirect pathways, respectively. To examine the role of these neurons during LIDs, we recorded striatal neuronal ensembles and tracked movement. As in Chapter 3, we used peak-to-trough duration and half-peak-width spike-waveforms to identify striatal MSNs and FSIs (Berke et al., 2004; Jin et al., 2014). Across five 6- OHDA-lesion animals and four sham-lesion animals in three recording sessions (levodopa, levodopa with SCH23390, and levodopa with sulpiride) we classified 100 neurons as MSNs. When compared to levodopa alone the average firing rate of striatal MSNs significantly decreased with co-injection levodopa and SCH23390 (levodopa: 5.3 Hz ± 0.3; levodopa + SCH23390: 3.9 Hz ± 0.5; t(5)= 2.74, p= 0.04; Figure 18) but not with co-injection of sulpiride (t(4)= 1.34, p=0.25). In these mice we did not have the capability of dividing the neurons into either D1- or D2-MSNs. It is possible that not all 45

61 MSNs were reflected in our results if the antagonists we used reduced the firing rate to undetectable frequencies. Figure 18: MSN firing rate. Inhibiting D1R expressing neurons decreased MSN firing rate (+SCH23390, blue) while inhibiting D2R expressing neurons did not (+sulpiride, green). Error bars are mean ± SEM; N=5; *p<0.05. Additionally, we looked at effects of these drugs on neuronal firing rate across time. Levodopa has a typical behavioral curve of approximately minutes with a peak action around 60 min (Lundblad et al., 2004; Cenci and Lundblad, 2007). Here, I divided the average firing rate of MSNs into 15 min bins to assess their activity with the different pharmacological treatments. At the first time point, there were no differences between any of the recording sessions. However, at 30 minutes, co-administering SCH23390 with levodopa decreases the firing rate of MSNs. At 45 minutes, both sulpiride and SCH23390 successfully decreased MSN firing rate in dopamine depleted mice with established LIDs (Figure 19). 46

62 Figure 19: MSN firing rate in 15 minute bins. Blocking D1 neurons decreased the firing rate of MSNs 30 minutes after co-injection of levodopa+sch23390 (blue) but sulpiride (green) had no effect until 45 minutes post injection. Error bars are mean ± SEM. Finally, as neurons are correlated with movement on a millisecond scale, we looked at time aligned neuronal activity at times of dyskinesias. We found that while MSNs were strongly modulated around LIDs (Chapter 3) co-administration of levodopa and SCH23390 led to fewer significantly modulated MSNs around time of dyskinesia (χ 2 = 12.41, p= ; Figure 20). Similarly, when looking at co-administration of levodopa and sulpiride fewer neurons were significantly modulated (χ 2 = 9.90, p= ). Yet, this modulation was not sufficient to decrease LIDs suggesting that the D1- MSNs are responsible for the expression of LIDs. 47

63 Figure 20: MSNs are less modulated around LIDs when inhibiting the direct and indirect pathways. Left: Normalized firing rate changes for all MSNs recorded in levodopa, levodopa+sch23390, and levodopa+sulpiride sessions. Right: Quantification of the number of neurons that changed frequency during each session. 48

64 Summary We recorded neuronal ensembles in the dorsal striatum of dopamine-depleted mice and tracked dyskinetic movements with high temporal resolution while assessing the effects of dopamine antagonists on neuronal firing rate and LIDs. When looking at behavior alone, we observed that only blocking D1 neurons, levodopa with SCH23390 session, decreased LIDs while blocking D2 dopamine receptors, levodopa with sulpiride session, had no significant effect, though results did trend to an increase in LIDs. Along the same line, neuronal activity showed that blocking D1 neurons decreased the overall firing rate of MSNs while co-administering a D2-receptor antagonist did not. Interestingly, when looking at the modulation of these neurons around LIDs we found that both the D1 and D2 antagonists significantly decreased the number of modulated neurons. These data suggest that though both receptor types can modulate MSN activity, only the changes in activity driven by the D1-receptor expressing neurons are sufficient to attenuate LIDs. 49

65 CHAPTER 5: SELECTIVE ACTIVATION OF D1-MSNs DRIVES LIDs Introduction Advancement of optogenetics in neuroscience has allowed researchers to selectively manipulate specific neuronal populations with high temporal resolution using different wavelengths of light. Briefly, researchers can selectively stimulate or inhibit neurons by injecting a virus (typically an adeno-associated virus, AAV) expressing an opsin, a light sensitive protein, targeted by a specific promoter that can be selectively expressed. The first light-sensitive protein was isolated from Halobacterium halobium (Oesterhelt and Stoeckenius, 1973; Deisseroth, 2011). Initial work by Ernst Bamberg described the use of Channelrhodopsin-2 (ChR2) to drive neuronal activity with light (Nagel et al., 2003). More commonly known, Boyden et al from Karl Deisseroth s laboratory further characterized ChR2 s potential for fast, precise, and dynamic stimulation of neurons (Boyden et al., 2005). Importantly, optogenetic stimulation has almost no measurable side effects for the host tissue, proving to be minimally invasive for use in vivo (Aravanis et al., 2007). Opsins have also been engineered for the targeted inhibition of neuronal populations. The two primary inhibitory opsins are a halorhodopsin, NpHR, and archaerhodopsin, Arch (Chow et al., 2010; Gradinaru et al., 2010). Halorhodopsin is a chloride pump, which requires constant light to move through its photocycle and has slower dynamics than Arch, which is a proton pump and has more potent inactivation effects (Chow et al., 2010; Witten et al., 2010; Okazaki and Takagi, 2013). The development of this technique and that of Cre-recombinase transgenic mouse lines have been instrumental in the progression of the field. For a comprehensive review please see reference Parker et al. (Parker et al., 2016). In this part of the project, we make 50

66 use of these techniques by using a transgenic mouse line (D1-Cre; detailed below) and an AAV expressing either ChR2 or NpHR with a Cre promotor to selectively target D1- MSNs. A pivotal study in 2010 made use of these techniques to specifically and selectively stimulate either D1- or D2-MSNs. In intact animals, bilateral stimulation of the direct pathway resulted in increased movement while stimulating the indirect pathway animals spent more time in a freezing posture. In dopamine-depleted animals, stimulation of D1-MSNs recovered parkinsonian behavior to pre-lesion levels and mice spent more time moving during stimulation than without stimulation (Kravitz et al., 2010). A recent study reported that chemogenetic activation of either the D1- or D2-MSNs in intact mice led to increased or decreased mobility, respectively. In line with Kravitz et al., in dopamine-depleted mice stimulation of the direct pathway rescued bradykinesia while indirect pathway stimulation reduced mobility from baseline (Alcacer et al., 2017). These results that stimulating the direct pathway leads to movement production, support the canonical circuit of movement. As previously discussed in Chapter 1, the direct pathway has been implicated in the expression of LIDs. Administration of D1 receptor agonists (instead of levodopa) also leads to the development of dyskinesias while D1R antagonists tend to decrease LIDs. Our results from Chapter 4 support the literature in that inhibition of D1 receptors led to a reduction of LIDs measured by our automated tracking. We also observed a decrease in striatal MSN activity when we co-injected levodopa and SCH23390; however, as this drug was injected systemically it is impossible to attribute all effects observed to one population of neurons in the striatum. 51

67 Several groups have shown that optogenetic and chemogenetic stimulation striatal MSNs leads to dyskinesias (Alcacer et al., 2017; F Hernandez et al., 2017; Perez et al., 2017). First, a group studied the effects of optogenetic stimulation of MSNs (without specificity) in rats and found that stimulation alone could cause dyskinesias in both levodopa naïve and levodopa primed rats (F Hernandez et al., 2017). In line with previous work, this research group also reported higher FosB expression in D1- but not D2-MSNs. Another study optogenetically stimulated D1 MSNs in dykinetic mice and found that D1-MSN stimulation alone caused dyskinesias. They also reported that combining D1-MSN stimulation with levodopa administration resulted in higher AIM scores than either condition alone (Perez et al., 2017). Lastly, another group chemogenetically manipulated D1 and D2 MSNs. In this study, D1-MSN stimulation led to dyskinesias that were not enhanced by D2-MSN stimulation. Interestingly, D1-MSN driven dyskinesias were enhanced by D2 agonists (Alcacer et al., 2017). These studies clearly implicate MSNs in dyskinetic movements. However, some questions arise from these studies: how are these neurons changing with LID development; how are they modulated around LIDs; is inhibition of these neurons sufficient to reduce severity of LIDs? Methods We used male mice in which Cre-recombinase was driven by the D1-dopamine receptor promoter (Drd1a-cre + ; derived from Gensat strain EY262; weighing >25g at the time of dopamine depletion). Mice were bred and verified by genotyping using primers for D1-Cre recombinase transgene (D1-Cre-F: AGG GGC TGG GTG GTG AGT GAT TG, D1-Cre-R: CGC CGC ATA ACC AGT GAA ACA GC) (MMRRC, Davis, CA). All 52

68 procedures were approved by the Animal Care and Use Committee at the University of Iowa. We induced dyskinesias as similarly to that described in Chapters 2 and 3 (Lundblad et al., 2004; Cenci and Lundblad, 2007). However, in these experiments mice were depleted of dopamine in the MFB using 2µg 6-OHDA bromide (instead of 1µg). During this surgery mice were also injected with ChR2 and NpHR in the dorsal striatum ipsilateral to the lesion (AP: +0.1, ML: -2.5, DV: -2.5). We used AAV constructs with floxed double-inverted AAV-ChR2 with mcherry (UNC Viral Core; AAV5-EF1a-DIOhChR2(H134R)-mCherry) or NpHR with yellow fluorescent protein (UNC Viral Core; AAV-EF1a-DIO-eNpHR 3.0-eYFP) (Cardin et al., 2009). When delivered to D1-Cre+ mice, Cre recombination leads to high expression driven by an EF-1a promoter selectively in neurons expressing D1DRs. The injection consisted of 0.5 µl of approximately 10 infectious particles per milliliter, per construct. Following three weeks of recovery, these animals were implanted with a 16- channel stainless steel microwire array optrodes (50 µm with optic fiber in the middle of configuration; MicroProbes, Gaithersburg, MD) in the dorsal striatum (AP: +0.1, ML: - 2.5, DV: -2.5). In addition, an implant was designed to attach two 4 mm infrared reflective spheres to the recording headstage (see Chapter 2). Following 1-2 weeks of recovery and stabilization of implant, animals were tested for unilateral depletion with paw-preference, open-field, and amphetamine-induced rotation tests. Animals were injected with amphetamine (5 mg/kg; i.p.) and ipsilateral rotations were recorded 30 min post-injection (Healy-Stoffel et al., 2012; Chotibut et al., 2013). Animals that did not show >5 ipsilateral rotations per minute were not included in this study (Chang et al., 53

69 1999; Paquette et al., 2009; Alberico et al., 2017); no animals were rejected based on this criterion. On the following day, we began administering 5 mg/kg of L-DOPAmethylester (levodopa; Sigma) dissolved in 0.09% NaCl sterile saline at a concentration of 1 mg/ml with Benserazide (2 mg/ml; Sigma) daily. Once LIDs were established, the following experiments were done: levodopa, levodopa with inhibition (589 nm) 1 min on/off, inhibition without levodopa, 4 Hz stimulation (473 nm) 30 seconds on/off, 20 Hz stimulation (473 nm) 30 seconds on/off. On testing days, mice with optrodes were connected to the optical patch cable through Zirconia ferrule (Doric Lenses) without anesthesia. Light was generated from a 473 nm or 589 nm DPSS laser source (OEM Laser Systems) and an optical rotary joint (Doric Lenses) was used to facilitate animal rotation. Specific frequencies were generated through a microcontroller (Teensy). In stimulation sessions, light was delivered every 30 seconds at 4 Hz or 20 Hz with pulse width 5 ms. The power output of the 473 nm laser (blue) was adjusted to be 8 mw and 14 mw for the 589 nm laser (yellow) at the fiber tip before every experiment. At the beginning of each session we tagged putative D1- MSNs by combining optogenetics and neuronal ensemble recording. MSNs virally expressing ChR2 should fire action potentials with a short latency (<5 ms) in response to blue light with consistent waveforms (Figure 21). 54

70 Figure 21: Example of optogenetically tagged putative D1-MSN. Top: Raster plot of a single neuron with spikes evoked blue light (red bar at 0 msec). Red spikes are during stimulation trials while blue spikes are the intrinsic activity for this neuron. Bottom: Average waveforms of light-evoked (red) and spontaneous activity (blue). Pearson correlation coefficient (CR) > 0.95 indicated that light-evoked and non-light evoked spikes were the same. Additionally, waveform characteristics such as half-peak-width (HMPW) and peak-to-trough duration (P2TD) are consistent with MSN waveforms. On recording days without levodopa we used amplitude based scoring to characterize LIDs in our mouse model as described in Chapter 2 and LIDs were counted (Lundblad et al., 2002; Winkler et al., 2002; Smith et al., 2011; Breger et al., 2013). Briefly, animals were assessed for two minutes every 10 minutes for an hour and each LID subtype was given two scores, one for duration of dyskinesias (0-4) and one for severity (0-4) (see Appendix for detailed description). These scores were then multiplied and axial, orofacial, and limb LID subtypes were added to obtain the integrated AIM scores at each time point. To obtain a global score, integrated AIM scores were summed for each recording session. Number of dyskinesias were also counted where all subtypes 55

71 were counted for 10 minutes. Additionally, on recording days with levodopa (levodopa Day 1 and Day 13, levodopa+inhibition) automatic tracking was used to analyze LIDs. Neuronal ensemble recordings in the striatum were made using an open-source recording system (OpenEphys). Putative single neuronal units were identified on-line using an oscilloscope and audio monitor. Spike activity was analyzed for all cells that fired at rates above 0.1 Hz. Single units were identified as described in Chapter 3. All neurons were analyzed offline using Off Line Sorter (Plexon) and MSNs identified by waveform shape, half-peak-width, peak-to-trough duration, and firing rate. Analysis of neuronal activity and quantitative analysis of basic firing properties were carried out using NeuroExplorer and MATLAB. All statistics assumed each recording day was an independent sample; no statistical dependence between days was considered and the same neurons were not explicitly tracked across days. After the completion of the experiments, mice were euthanized by injections of 100 mg/kg sodium pentobarbital, and transcardially perfused with 4% paraformaldehyde. Brains were post fixed in 4% paraformaldehyde and cryoprotected in 30% sucrose before sectioning in a cryostat. SNc sections were mounted and stained for tyrosine hydroxylase (TH; polyclonal rabbit anti-th, 1:500; Millipore, Temecula, CA) and cell bodies using DAPI. Striatal sections were mounted and stained for DARRP-32 to assess colocalization of ChR2 and Halo. Histological reconstruction was completed using post mortem analysis of lesion, optrode placement in the striatum, viral expression, and cell concentration in the SNc. Images were captured on Zeiss Apotome.2 Axio Imager and cells in the substantia nigra were counted using the optical fractionator in Stereo Investigator for dopamine depletion quantification and analysis. 56

72 Results Pharmacologically blocking D1R in the striatum successfully attenuates LIDs (Taylor et al., 2005). We used hemiparkinsonian D1-Cre mice expressing NpHR-eYFP (an inhibitory opsin) in the dopamine-depleted striatum to explore the effect of optogenetic inhibition of the direct pathway on LIDs. Once mice exhibited stable LIDs (Day 13 of levodopa) we tested the efficiency of inhibiting D1- MSNs in decreasing LIDs following levodopa administration. Though we did find that between Day 1 and Day 13 of levodopa LIDs worsened (more LIDs detected on Day 13), unilaterally inhibiting the direct pathway did not successfully decrease LIDs (Figure 22). Figure 22: Automatically detected LIDs. Chronic levodopa administration (Day 13, red) led to more LIDs when compared to the first day of levodopa administration (Day 1, light red). Intermitent inhibition of D1-MSNs (inhibition, yellow) following levodopa administration did not decrease the number of automatically detected LIDs. Error bars are mean ± SEM; N = 4. Recent work demonstrates that selective stimulation of D1-MSNs causes AIMs in levodopa primed mice (Alcacer et al., 2017; Perez et al., 2017). To explore this effect in our mice we used the same mice as above and used a blue laser to activate an excitatory opsin (ChR2-mCherry) expressed in the striatum unilateral to the lesion. Following LID development, we tested the effects of 10 minutes constant inhibition (yellow laser), 10 minutes 4 Hz stimulation 30 seconds on/off (blue laser), and 10 minutes 20 Hz stimulation 30 seconds on/off (blue laser) without levodopa. We found that D1-MSN 4 57

73 Hz stimulation caused AIMs (5.7 ± 0.9), and 20 Hz stimulation led to even more dyskinesias (16.7 ± 4.4) (Figure 23). Figure 23: Stimulation of D1-MSNs increases dyskinesia. Right: Tracking traces from a hemiparkinsonian mouse exhibiting bradykinesia (no laser) and increased movement with stimulation of D1-MSNs at 4 and 20 Hz. Right: Quantification of dyskinesia. All AIMs subtypes (axial, orofacial, and limb) were counted for 10 min while D1-MSNs stimulated at 4 Hz (light blue), and stimulated at 20 Hz (blue). Driving the direct pathway is sufficient to cause AIMs, and higher frequencies tended to lead to more AIMs. Error bars are mean ± SEM; N= 3. To further examine the function of D1-MSNs during dyskinesias, we recorded striatal neuronal ensembles and tracked movement with high temporal precision. As in previous chapters, we used peak-to-trough duration and half-peak-width spikewaveforms to identify striatal MSNs. Across 3 6-OHDA-lesion animals in three recording sessions (Baseline, Day 1, and Day 13 of levodopa) we classified 68 neurons as MSNs and of those 24 were putative D1-MSNs. On Day 13 of levodopa administration it seems that putative D1-MSNs have a higher average firing rate when compared to untagged neurons, though it is not yet significant (5.3 Hz ± 2.1 vs. 1.6 Hz ± 0.9 ; t(4) = 1.61, p=0.18; Figure 24). From Chapter 3 we know that the firing rate of striatal MSNs in dyskinetic mice on average was higher when compared to Day 1 of levodopa. The 58

74 results in the current chapter suggest that this heightened firing rate is driven by D1- MSNs. Figure 24: Mean firing rate across days of LID development. MSNs recorded in different sessions were divided into opto-tagged (D1-tagged, red) and untagged (black) to calculate mean firing rate of each population across recording days (Baseline, Day 1 (D1), and Day 13 (D13)). Error bars are mean ± SEM; N = 4. To examine how D1-MSNs are modulated around LIDs when compared to other MSNs, we focused on Day 13 of levodopa session as this time point had the most LIDs (Chapter 2). In LID sessions after two weeks of levodopa injection (Day 13 of levodopa), we recorded 25 MSNs and of these 13 were tagged as putative D1-MSNs. When we looked at MSN activity round axial dyskinesias, we found that striatal D1-MSNs were more prominently modulated when compared to other MSNs (t(22) = 2.02, p = 0.056; Figure 25). Due to the finding in Chapter 4 that blocking D1R decreased MSN modulation around axial dyskinesias, we explored how D1-MSNs changed around the time of LIDs. We found that when compared to untagged MSNs, D1-MSNs had stronger firing rate changes during LIDs (t(22) = 2.95, p = 0.008). 59

75 Figure 25: D1-MSNs are strongly modulated around axial LIDs. Left: Normalized firing rate of putative D1-MSNs (red) and untagged MSNs (black) around axial dyskinesias (Time from LID). Right: Mean (*) and distribution of firing rate changes (in Hz) in D1-MSNs (red) and untagged MSNs (black). Error bars are mean ± SEM; N = 3. Summary We used D1-Cre mice and optogenetics to stimulate and inhibit neurons in the direct pathway in dyskinetic mice. As in previous chapters, we recorded neuronal ensembles in the dorsal striatum and tracked dyskinetic movements with high temporal resolution while assessing changes in MSN activity around LIDs and the effects of D1- MSN stimulation and inhibition on dyskinesias. While D1-Cre mice developed LIDs similar to wild-type mice, unilateral inhibition was not sufficient to attenuate LIDs. In line with previous work, stimulation of D1-MSNs (at 4 and 20 Hz) induced AIMs. Interestingly, when looking at the modulation of MSNs around LIDs we found that D1- MSNs were more significantly modulated. These data indicate that D1-MSNs are strongly modulated around dyskinetic events and these may be the drivers of LIDs. 60