Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Long-term treadmill exercise delays the onset of motor dysfunction, cognitive impairments, and mood disturbances in the CAG 140 mouse model of Huntington's disease via restoration of dopamine neu...
(USC Thesis Other)
Long-term treadmill exercise delays the onset of motor dysfunction, cognitive impairments, and mood disturbances in the CAG 140 mouse model of Huntington's disease via restoration of dopamine neu...
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
LONG-TERM TREADMILL EXERCISE DELAYS THE ONSET OF MOTOR
DYSFUNCTION, COGNITIVE IMPAIRMENTS, AND MOOD DISTURBANCES IN THE
CAG 140 MOUSE MODEL OF HUNTINGTON’S DISEASE VIA RESTORATION OF
DOPAMINE NEUROTRANSMISSION AND ATTENTUATION OF STRIATAL
PATHOLOGY
by
Daniel Patrick Stefanko
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(NEUROSCIENCE)
May 2016
Advisory Committee:
Giselle Petzinger, MD
Michael Jakowec, PhD
John Walsh, PhD
Ruth Wood, PhD
Enrique Cadenas, MD, PhD
Copyright 2016 Daniel P Stefanko
ii
ACKNOWELDGEMENTS
I would like to thank my mentor Dr. Giselle Petzinger and my co-mentor Dr. Michael
Jakowec for taking me into their lab back in 2011 and allowing me pursue a project that differed
so drastically from what they have already done in the lab. From the very beginning of my
project they gave me free rein to come up with my own hypotheses and provided me with the
means to become the scientist that I am today. My appreciation and respect for them goes well
beyond what I can put into words and I am confident that I would have never made it to this
point in my PhD without their support and guidance. It’s truly been an honor to be a part of their
lab and a part of their research family.
I would also like to thank the other members of my dissertation committee. Dr. John
Walsh, Dr. Ruth Wood, and Dr. Enrique Cadenas have truly been amazing to work with during
the past few years and always treated me much more like a colleague than a student, which is an
honor that I may only now begin to deserve. While we haven’t met formally as a committee too
many times throughout my career, I have been fortunate to have fostered a unique relationship
with all three of these incredible scientists at some point during my PhD work through
collaborations between our labs. I know that I have learned so much from each of them and I
hope to continue to strive to be the caliber of scientist that they have expected from me as I
continue beyond my PhD years.
I would also like to acknowledge and thank the Dr. Daniel Holscheider and the members
of his lab, specifically Zhuo Wang, for their extensive contributions to Chapter 5 of my
dissertation. All of the blood flow experiments in Chapter 5 were performed by Zhuo and
members of Dr. Holschneider’s lab. The collaborations with Zhuo has not only allowed me to
learn much more about my HD mice than I could have on my own, but has also challenged me to
iii
form a more complete scientific perspective through the many brainstorming sessions and
meetings that we had to discuss and interpret our findings.
I have had the distinct pleasure to work with some of the finest and brightest graduate
students that the neurograduate program has to offer. The graduate students in my lab, Dr.
Natalie Kintz, William Toy, and Matthew Halliday have supported me and pushed me to always
be at my best in and outside of the lab. Natalie in particular has been a role model of mine
throughout my PhD. The dedication and hard work that she had while we were working together
was truly inspirational and I have tried to make my star shine as bright as hers. Will and Matt
have always been quick to lend a hand, even with menial tasks and I have learned so much from
both of them. Honestly, I couldn’t thank all three of my fellow lab graduate students enough for
getting me here. Additionally, I want to thank a few other graduate students that have helped to
shape me into the scientist (and person) that I am today. First and foremost, Amy Patterson has
been a loyal friend and confidant since we first took Neuro Core together and has been one of my
best friends these last few years. I also would like to thank Jonathan Tucci, a fellow MD/PhD
student and another of my closest friends in the last few years. It’s been a real pleasure getting to
know Jon and I am so thankful for his support and friendship throughout this long process. Katie
Wallin, from Dr. Ruth Wood’s lab, has also been an exceptional collaborator and friend.
In addition to amazing graduate students, the Jakowec and Petzinger lab have had some
of the most incredible undergraduate researchers, many of whom I had the pleasure of working
with directly and several of whom contributed extensively to the work in Chapters 2, 3 and 4 of
this dissertation. Vivek Shah and Damaris Garcia both joined the lab in undergrad and remained
with the lab as lab techs after graduation while applying for medical school. Our lab does not
currently have a lab manager and so many of the duties of both lab manager and lab tech fell to
iv
Vivek and Damaris, in addition to the work they did for my project. I especially wanted to thank
them for putting up with me at my worst and bringing out my best. I also would like to
acknowledge Wendi Yamasaki, Power Lee, Amanda Tran, Zaira Gasanova and Spencer
Patzman, all of whom played a huge role in assisting me with my project during their
undergraduate years and contributing directly to experiments found in this dissertation. I
additionally wanted to thank Brian Leyshon, Brenton Keller, Katherine Arellano, Katherine
Choi, Sarah French, Tiffany Chen, James Tavornwattana, Richard Ponce, Sara Pstasnik, Nicole
Kashani, Allison Najafi, Edward Chau and any other member of the Jakowec and Petzinger lab I
may have omitted for being incredible friends and colleagues these last few years.
Finally, and most importantly, I would like to thank my family for their support,
particularly my parents Paula and Michael Stefanko for the countless things they have done
throughout my entire life and for shaping me into the man I am today. My grandmother Adeline
Zullo has been a role model and an inspiration for me my entire life and I hope to become even
half as incredible as she is. My siblings Adrianna, Brian and Lauren have been extremely
supportive and have always challenged me to be my best at everything I do. Finally, my nieces
Sydney and Taylor and nephew Carter have been wonderful sources of joy during some of the
harder parts of the last few years. I’m a better person for having all of you in my life.
Funding for the studies presented here primarily came from a Graduate Student Research
Assistantship/USC Parkinson’s Graduate Research Fellowship funded by the USC Parkinson’s
Disease Research Group including George and Mary Lou Boone, Walter and Susan Doniger, and
the family of Don Gonzalez Barrera (2011-2012 and 2015). These studies would not be possible
without the generous support of the Roberto Gonzales Family Foundation and their interest in
PD research and the importance of exercise / healthy lifestyle for patients and families.
v
TABLE OF CONTENTS
LIST OF FIGURES vi
LIST OF TABLES viii
ABSTRACT ix
CHAPTER 1: INTRODUCTION 1
CHAPTER 2: Treadmill Running Delays the Onset of Depression-like Behavior and
Striatal Pathology in the CAG140 Knock-in Mouse Model of Huntington’s Disease 35
CHAPTER 3: Treadmill running delays the onset of cognitive dysfunction and restores
dopamine neurotransmission in the CAG140 mouse model of Huntington’s disease 66
CHAPTER 4: Treadmill Running Delays the Onset of Motor Dysfunction in the CAG140
Mouse Model of Huntington’s Disease 98
CHAPTER 5: Evidence of Functional Brain Reorganization Based on Blood Flow
Changes in the CAG140 Knock-In Mouse Model of Huntington’s Disease 118
CHAPTER 6: CONCLUSIONS 134
REFERENCES 150
vi
LIST OF FIGURES
Figure 1: Comparison of WT and HD Brains. 2
Figure 2: Cognitive Impairments in Clinical and Pre-clinical HD Patients. 7
Figure 3: Mutant Huntingtin Aggregates Affect Synaptic Plasticity. 9
Figure 4: Mutant Huntingtin Cell Conformations. 11
Figure 5: Mutated Huntingtin Inclusion Body. 12
Figure 6: Voluntary Running Wheels in the R6/2 HD Mouse Model. 16
Figure 7: Four Main Dopamine Pathways in the Brain. 19
Figure 8: Distribution of Dopamine Receptors in the Brain. 20
Figure 9: Dopamine Signaling Pathways in the Brain. 22
Figure 10: Direct and Indirect Pathways in the Basal Ganglia (BG). 23
Figure 11: Medium Spiny Neuron (MSN) Cell Types in the Striatum. 25
Figure 12: DA Pathway Involvement in T-Maze Reversal Learning Task. 27
Figure 13: Basal Ganglia DA Pathway Changes in Early and Late HD. 28
Figure 14. The CAG
140
knock-in mouse model of Huntington’s disease (HD KI) shows a
deficit in the distance run per hour on a voluntary running wheel as compared to wildtype
(WT) controls. 46
Figure 15. HD KI mice display no motor function impairment at 4 months and 6 months
of age. 48
Figure 16. High intensity exercise delays the onset of depression-like behavior observed in
HD KI mice. 52
Figure 17. Exercise decreased the number and intensity of intranuclear htt aggregates in
dorsostriatal MSNs. 57
Figure 18. Exercise decreased the number and intensity of intranuclear htt aggregates in
ventrostriatal MSNs. 58
Figure 19. HD KI mice display no cognitive impairment in the novel object recognition
(NOR) task at 4 months and 6 months of age. 77
vii
Figure 20. Long-term treadmill exercise delays cognitive impairments observed in HD
KI mice. 79
Figure 21. Exercise decreases the number of trials required by HD KI mice to learn a new
strategy in the t-maze reversal learning task at 4 months and 6 months of age. 82
Figure 22 Exercise decreases the number of perverative errors made by HD KI mice in
learning a new strategy in the t-maze reversal learning task at 4 months and 6 months of
age. 85
Figure 23. Long-term treadmill exercise restored dopamine D2 receptor concentrations in
dorsostriatal MSNs. 92
Figure 24. Long-term treadmill running restored dopamine D2 receptor concentrations in
ventrostriatal MSNs. 94
Figure 25. Long-term treadmill exercise delays the onset of motor function impairment
observed in HD KI mice 110
Figure 26. HD KI animals do not exhibit changes in striatal or lateral ventricle areas or
cortical thickness at 12 months. 111
Figure 27. Exercise had no effect on the number and intensity of intranuclear htt
aggregates in dorsostriatal and ventrostriatal MSNs. 114
Figure 28: Significant genotypic differences in regional cerebral blood flow between
CAG
140
KI mice compared to wild-type controls in the resting state. 126
Figure 29: Histological staining of CAG
140
KI and WT mice shows no gross evidence of
atrophy in the dorsolateral striatum. 129
Figure 30: Summary of the significant differences in regional cerebral blood flow in the
basal ganglia–thalamic–cortical (BTC) and cerebellar–thalamic–cortical (CbTC) circuits
between CAG
140
KI mice and wild-type controls in the resting state. 131
viii
LIST OF TABLES
Table 1: Motor Function in HD Patients. 4
Table 2: DA and DA Receptor Levels in HD Patients and Animal Models. 30
Table 3: HPLC Analysis of Dopamine, Norepinephrine, Serotonin and DA turnover. 61
Table 4: Significant changes in rCBF of CAG
140
KI mice compared to wild-type controls
in the resting state in the cortex and subcortex of the left and right hemispheres (L/R). 128
ix
ABSTRACT
Huntington’s disease (HD) is an autosomal dominant neurodegenerative disorder caused
by an excessive polyglutamine (CAG) expansion in the Huntingtin (Htt) gene resulting in a
mutated form of the huntingtin (htt) protein (The Huntington's Disease Collaborative Research
Group, 1993). HD is characterized by progressive decline in cognitive and motor functions with
neuropsychiatric disturbances leading ultimately to premature death 10 to 15 years after onset of
motor symptoms. The major pathological findings include severe degeneration of striatal
medium spiny neurons (MSNs) and the cerebral cortex, particularly the prefrontal and frontal
cortex (Gil and Rego, 2008). In the striatum, there is a preferential loss of the dopamine D2
receptor (DA-D2R)-containing MSNs that mediate the indirect pathway compared to direct
pathway dopamine D1 receptor (DA-D1R)-containing MSNs.
There is currently no cure for HD. One strategy for treatment predominately aims to
attenuate symptoms after their onset and prolong patient quality of life. Elucidating the
mechanisms that can modify disease prevention could identify potential therapeutic targets for
improved treatments and could eventually lead to a cure. While long-term studies examining
lifestyle factors such as exercise, diet, and cognitive engagement have shown neuroprotective
benefits in degenerative disorders such as Parkinson’s disease (PD) and Alzheimer’s disease
(AD), similar studies have not been carried out in HD (Chen et al., 2005; Pitkala et al., 2013).
However, studies on predictors of phenotypic progression in HD (the TRACK-HD Study)
suggest that lifestyle factors may have a role in the modification of HD progression (Tabrizi et
al., 2013).
Our research group is interested in determining the underlying molecular mechanisms by
which long-term intensive exercise results in enhanced neuroplasticity in the basal ganglia and
x
cortical circuitry. Specifically, our lab has demonstrated that motorized treadmill running
increases neuroplasticity and attenuates pathophysiological mechanisms in a mouse model of
Parkinson’s disease by enhancing dopamine neurotransmission (Petzinger et al., 2007),
glutamate synaptic neurotransmission (Kintz et al., 2013), and reversing dendritic spine loss
(Toy et al., 2014), culminating in an overall rescue of behavioral motor deficits in PD mouse
models.
This dissertation will provide evidence of the benefits of long-term treadmill running in
the CAG
140
HD mouse model, chosen for its slow progression with motor symptoms emerging at
12 months of age. We speculate that this long prodromal phase presents an opportunity to
investigate interventions such as exercise that can potentially modify disease progression. To
this end we began running mice on motorized treadmills at 1 month of age and continued for 11
months probing for effects on phenotype and pathophysiology at 2, 4, 6, 9 and 12 months of age.
No study to date has explored CAG
140
KI mice in the prodromal period for potential non-motor
(mood) disturbances and cognitive impairments. Previous studies in other HD mouse models
have shown little to no benefits of exercise in modifying disease progression (Pang et al., 2006;
Kohl et al., 2007; van Dellen et al., 2008; Potter et al., 2010; Renoir et al., 2011; Wood et al.,
2011; Harrison et al., 2013). However, several recent studies have challenged the validity of
using voluntary wheel running in long-term studies of HD mouse models, as over time HD KI
animals display a decreased preference towards the running wheels, running significantly less
than WT counterparts despite the absence of motor symptoms (Hickey et al., 2008; Cepeda et al.,
2010; Stefanko et al., 2016a). No study has directly compared motorized treadmill running,
rather than voluntary wheel running, with sedentary CAG
140
KI mice for its potential as a
xi
therapeutic intervention to ameliorate behavioral symptoms and attenuate pathological
abnormalities.
To address these gaps in knowledge, Chapters 2 through 4 of this dissertation focus on
the role that long-term treadmill exercise plays in modifying the progression the three main
symptoms of HD: psychiatric disturbances in the form of depression, cognitive impairments, and
motor dysfunction, respectively. Chapter 5 examines the differences between sedentary HD KI
and WT animals in functional brain mapping via regional cerebral blood flow analysis to
determine changes in whole-brain networks.
Chapter 2 of this dissertation investigates the effects of exercise on mood disturbances in
CAG
140
mice. Additionally, this study follows the progression of intranuclear inclusion body
formation in striatal MSNs in exercised and sedentary animals. Motorized treadmill running was
initiated at 4 weeks of age (1 hour per session, 3 times per week) and continued for 6 months.
Analysis of depression-like behavior using the tail-suspension and forced-swim tests showed a
significant increased in depression-like behavior in sedentary HD KI mice as compared with WT
controls at 4 and 6 months of age. HD KI mice that underwent long-term treadmill running
displayed a significant attenuation of this feature. In addition, we examined the pattern of
expression of huntingtin (Htt) protein in the striatum and showed that the number and
immunohistochemical staining of intranuclear aggregates was also significantly reduced.
Exercise also restored decreases in dopamine, norepinephrine, and serotonin concentrations
observed in the striatum. Together these findings begin to address the potential impact of life-
style and early interventions such as exercise on modifying disease progression.
Chapter 3 focuses on cognitive impairments and the role that long-term treadmill exercise
plays in modifying the onset and progression of these symptoms in the CAG
140
HD mouse
xii
model. In HD patients, cognitive inflexibility presents early in the disease progression, as
evidenced by impaired performance on the Wisconsin Card Sorting Test and attentional set-
shifting tasks (Owen et al., 1993; Lawrence et. al., 1996; Ho et al., 2003). Several studies have
associated impairments in cognitive flexibility with altered dopamine neurotransmission in both
HD patients and HD mouse models. Specifically, the loss of DA receptors in the striatum has
been shown to strongly correlate with decline executive function, learning, and memory in HD
(Backman 1997). The studies in Chapter 3 investigates the underlying mechanism of any
beneficial effect of exercise on dopamine neurotransmission by measuring striatal levels of
tyrosine hydroxylase and dopamine D1 and D2 receptors in exercised and sedentary animals. As
in the studies in Chapter 2, motorized treadmill running was initiated at 4 weeks of age (1 hour
per session, 3 times per week) and continued for 6 months. Sedentary HD KI animals displayed a
significant impairment in the t-maze reversal learning task, suggesting an impairment in
cognitive flexibility. This impairment could be explained in part by a downregulation of the
dopamine D2 receptor observed in the dorsal and ventral striatum, resulting in aberrant dopamine
neurotransmission in the indirect pathway. Exercise restored dopamine D2 levels and rescued
the cognitive disturbances observed in reversal learning.
Chapter 4 examines the beneficial role that exercise plays in CAG
140
mice to delay the
onset of motor symptoms. While the majority of studies in the CAG
140
mouse model have
focused on motor symptoms, to date, no study has attempted to use motorized treadmill running,
as opposed to voluntary wheel running, in CAG
140
animals to examine the potential beneficial
effects of long-term exercise on motor function and neuropathological symptoms of HD. This
study attempts to investigate the effects of long-term motorized treadmill running on motor
function in the accelerating rotarod task and through gait analysis. Additionally, this study
xiii
investigates intranuclear inclusion body formation in striatal MSNs in exercised and sedentary
animals in an attempt to determine whether the attenuation of striatal pathology observed early in
the prodromal period (discussed in Chapter 2) persists up to 12 months of age. Motorized
treadmill running was initiated at 4 weeks of age (1 hour per session, 3 times per week) and
continued for 12 months. At 12 months of age, sedentary HD KI animals displayed a significant
impairment in the accelerating rotarod task and changes in gait were observed. Long-term
treadmill exercise resulted in a delay in onset of these deficits in motor ability, balance, and
coordination. While we previously showed that exercise delays the formation of htt intranuclear
inclusions in the prodromal period, by 12 months both sedentary and exercised animals exhibited
nearly identical htt aggregation in striatal MSNs.
The studies in Chapter 5 investigate the differences between 6 month old sedentary HD
KI and WT animals in functional brain mapping via regional cerebral blood flow (rCBF)
analysis. Our results showed significant changes in rCBF between CAG
140
KI and WT mice,
such that CAG
140
KI animals demonstrated hypo-perfusion of the basal ganglia motor circuit and
hyper-perfusion of cerebellar-thalamic and somatosensory regions. Significant hypo-perfusion
was noted also in CAG
140
KI mice in the prelimbic and cingulate cortex (medial prefrontal area)
and the hippocampus – areas associated with cognitive processing and mood. Changes in rCBF
were apparent in the absence of motor deficits (rotarod test) or atrophy in the striatum (caudate-
putamen) or hemispheric volume. Our results suggest a functional reorganization of whole-
brain networks at a presymptomatic stage in the life of the CAG
140
KI mouse.
Collectively, the findings of the studies outlined in this dissertation suggest that long-
term treadmill running, beginning well before onset of motor symptoms, results in a delay of
depression-like behavior, cognitive impairments, and striatal pathology observed during the
xiv
prodromal period in CAG
140
mice. Long-term exercise was also sufficient to delay the onset of
motor symptoms. Additionally, long-term exercise restored the decreased striatal levels of
dopamine, norepinephrine, and serotonin observed in sedentary HD KI animals. The
downregulation of the D2 receptor observed at 4 and 6 months was rescued after long-term
treadmill running, suggesting a restoration of dopamine neurotransmission. Together these
results indicate a potential mechanism for symptomatic improvement observed in these animals.
Analysis of cerebral regional blood flow comparing sedentary HD KI animals and WT controls
has shown a functional reorganization of whole brain networks in the presymptomatic HD brain,
particularly in the striatum and prefrontal cortex, where significant hypoperfusion of basal
ganglia motor circuits occurs in CAG
140
mice. Determining the mechanism(s) by which long-
term treadmill running has its effect in delaying symptom onset may provide potential
pharmacological and genetic interventions targeting neuroplasticity in the basal ganglia and
prefrontal cortex resulting in delaying the onset and progression of HD as well as to improving
the quality of life of HD patients.
1
CHAPTER 1: INTRODUCTION
The goal of the studies presented in this dissertation was to elucidate any behavioral and
neuropathological benefits of long-term treadmill exercise in the CAG
140
mouse model of
Huntington’s disease (HD). Chapter 1 will provide an introduction to Huntington’s disease and
other background information underlying the research discussed in this dissertation. Chapter 2
will focus on describing potential pathological and neurochemical mechanisms by which
exercise has a beneficial effect on the psychiatric component of Huntington’s disease,
specifically depression-like behavior. Chapter 3 will discuss the molecular mechanisms by
which exercise restores deficits in dopamine (DA) neurotransmission in the striatum, and its link
to ameliorating cognitive impairments observed in the CAG
140
mice. Chapter 4 will examine the
beneficial role that exercise plays in CAG
140
mice to delay the onset of motor symptoms.
Chapter 5 will discuss changes in regional cerebral blood flow observed in the CAG
140
mouse
brain, and will explain how these changes suggest a functional reorganization of whole-brain
networks in this prodromal model of HD. Finally, Chapter 6 will summarize all of the findings
in this dissertation as well as provide insight into the significance of these studies and possible
future directions. The studies included in this dissertation provide evidence that furthers our
understanding of the role that exercise plays in delaying symptom onset, restoring DA
neurotransmission, and delaying striatal pathology. Determining the underlying mechanism(s)
by which long-term treadmill running has these effects may provide potential pharmacological
and genetic interventions targeting neuroplasticity in the basal ganglia and prefrontal cortex
resulting in delaying the onset and progression of HD as well as to improving the quality of life
of HD patients.
2
1.1 Huntington’s Disease
Huntington’s disease (HD) is an autosomal dominant neurodegenerative disorder caused
by an excessive polyglutamine expansion in exon 1 of the Huntington (htt) protein (The
Huntington’s Disease Collaborative Research Group, 1993). Pathologically, HD is characterized
by progressive neuronal loss of striatal medium spiny neurons (MSNs) in the caudate nucleus
and putamen as well as the cerebral cortex due to the aberrant intracellular accumulation of htt
protein (Gil and Rego 2008). Structural analysis has shown that the mutated htt protein
accumulates in MSNs and aggregates in the nucleus to form inclusion bodies (Rubinsztein and
Carmichael 2003; Bates 2003; Zheng and Diamond 2011). These large intranuclear inclusions
have been implicated as playing a role in HD pathophysiology (Chai et al., 1999; Bates 2003;
Zhang et al., 2005; Imarsio et al, 2008). While the exact pathogenic mechanism of the mutated
htt protein is largely unknown, several recent studies implicate the involvement of transcriptional
dysregulation and mitochondrial dysfunction, ultimately leading to neuronal fragility and death
(Kim et al., 2010; Costa and Scorrano, 2012; Zheng and Diamond, 2012).
Figure 1: Comparison of WT and HD Brains.
The atrophy of the caudate nucleus and putamen observed in the HD brain results
in enlargement of the anterior horns of the lateral ventricles (hydrocephalus ex
vacuo). Additionally, as HD progresses, there is extensive loss of cortical
neurons as well. Figure taken from Kelly et al., 2009.
3
HD typically presents as a triad of motor dysfunction, cognitive impairments, and mood
disturbances. While genetic screening can identify patients at risk, clinical diagnosis is not
confirmed until patients display overt motor features. The motor symptoms that lead to the
diagnosis of clinical HD can be split into involuntary and voluntary impairments. Prior to
diagnosis, patients remain in a preclinical, or prodomal, phase. Recent studies have identified
cognitive and psychiatric disturbances in the prodomal phase, some beginning as early fifteen
years before diagnosis (Julien et al., 2007; Tabrizi et al., 2013; Epping et al., 2013).
While genetic screening can identify the presence of the mutated Htt gene in patients that
will develop HD, clinical diagnosis has traditionally been made at the onset of overt motor
symptoms. Prior to disease onset, patients may exhibit fidgetiness and small twitches (Sturrock
and Leavitt 2010). The motor symptoms that lead to the diagnosis of clinical HD can be split
into involuntary and voluntary impairments. Early involuntary symptoms include a hyperactive
state where the twitches seen in the prodomal phase progress to irregular flailing movements of
the limbs, head, and trunk known as chorea (Kirkwood et al., 2001). These jerky movements
generally worsen throughout the early to middle phases of the disease progression. However,
towards the end of the disease, the chorea typically subsides and bradykinesia predominates as
patients become Parkinsonian, displaying hypokinesia and rigid movements (Sturrock and
Leavitt 2010). This deterioration is thought to be the result of further atrophy of the brain in the
later course of the disease (Zheng and Diamond 2011). Typically, at this point of the disease
patients can no longer care for themselves and require a full-time caregiver or
institutionalization.
The impairments in voluntary motor function are also devastatingly progressive. Early
dysfunction includes impaired handwriting, which is lost completely when the patients lose fine
4
movement control to chorea in the disease state. The most typical voluntary deficits are gait
abnormalities, as patients gradually lose mobility, coordination, and balance until they are no
longer ambulatory (Sturrock and Leavitt 2010). Rao and coworkers characterized the loss of
movement in HD patients in early, middle, and late phases of the disease and demonstrated that
patients exhibit decreased abilities in a variety of motor tasks in advanced stages of the disease
(2009). Patients displayed the inability to reach forward without losing balance; increased time
to stand up, walk three meters, and sit down back down in a chair; and impaired scores on a test
of balance and coordination originally used in stroke patients (Table 1). In addition to
difficulties in gait, patients in later stages of HD have impaired voluntary control of the mouth,
tongue and larynx. Eventually patients are no longer able to communicate through speech and
are unable to swallow, leading to the inability to receive nutrition orally (Sturrock and Leavitt
2010).
Table 1: Motor Function in HD Patients.
Motor function was assessed by a variety of tasks in patients in early and
advanced stages of HD. As HD progresses, patients perform significantly worse
on every motor test.
a
denotes p < 0.01,
b
denotes p < 0.001 and
c
denotes p <
0.0001. Table from Rao et al. 2009.
5
The cognitive impairments observed after HD diagnosis are thought to be due to loss of
frontal-subcortical connectivity and typically becomes progressively worse along with its motor
counterpart (Snowden et al. 2002). One of the earliest cognitive deficits in patients in the clinical
stages of HD is a reduction in attention and concentration, causing subjects to be easily distracted
during tasks (Pillon et al. 1991). Patients can also exhibit slowness of thought and impairments
in executive functions, dual tasking, and recall. These symptoms are thought to be the result of
dysfunctions in the frontal lobe and frontal-subcortical connectivity (Sturrock and Leavitt 2010).
As HD progresses, both implicit and explicit memory become impaired. Patients have
difficulties in learning new tasks and spatial memory is affected. In advanced stages of HD,
cortical loss results in a global dementia, agnosias, and complete cognitive decline.
Interestingly, one of the last areas affected are the language centers of the brain. However, these
regions can also be affected towards the end of the disease progression, resulting in aphasia.
While cognitive function has a similar progressive decline as motor symptoms
throughout the course of HD, recent studies have shown that the onset of cognitive impairments
begins as early as fifteen years prior to clinical diagnosis. In most cases, these findings are made
retrospectively and thus, studies following cohorts during the prodromal phase to the onset of
motor symptoms and clinical diagnosis have only recently been published. Duff and coworkers
demonstrated that at least 38% of HD patients in the prodromal phase exhibited mild cognitive
impairments (2010). One of the earliest cognitive impairments observed in the prodromal phase
is the inability to determine emotions in facial expressions and from verbal cues (Snowden et al.
2008). Other modest impairments such as the inability to estimate the time to complete tasks and
loss of punctuality are also observed as early as fifteen years prior to the onset of motor
symptoms.
6
As patients get closer to disease onset, cognitive decline worsens. Stout and colleagues
performed a large-scale study that examined 19 separate cognitive tasks testing executive
functioning, dual tasking, and response times (2011). Patients were split into three groups by
proximity to disease onset. The “Far” group was estimated to be over 15 years away from
clinical diagnosis, The “Mid” group between 9 and 15 years away and the “Near” group within 9
years. Patients in the “Near” group were impaired in comparison to controls in 16 out of the 19
tasks and the “Mid” group in 13 of the 19 tasks. These results suggest strong cognitive deficits
prior to the onset of motor symptoms (Stout et al. 2011). In addition, Maroof and colleagues
demonstrated that patients showed robust impairments within five years of diagnosis on the
Stroop word and color trials (2011). In these tests, patients are shown the name of colors in font
colors that are dissimilar to the name. For example the word “blue” would be the color green.
Patients would then have to identify either the word or the color and the time taken to go through
a list of words and the mistakes made would be analyzed. As patients drew nearer to clinical
diagnosis, especially within five years, they exhibited impaired scores in identifying both the
colors and the words in the Stroop test. In the TRACK-HD study, HD patients that were
estimated to be near the onset of symptoms performed poorer than normal subjects on several
cognitive tasks including symbol digit modality test and the Stroop test (Tabrizi et al. 2013).
Both early-HD and late-HD performed much worse on these tasks than normal subjects and
those in the prodromal stage (Figure 2). Taken together, these studies suggest a role for cognitive
tasks in the early detection of prodromal HD patients who are within a decade of disease onset.
7
Figure 2: Cognitive Impairments in Clinical and Pre-clinical HD Patients.
Patients were assessed in a variety of cognitive tasks including the symbol digit
modality task. Patients with early and advanced HD both displayed cognitive
impairments compared to controls. Interestingly cognitive impairments were also
observed in pre-clinical patients. Figure taken from Tabrizi et al. 2013. Lower
scores indicate impairment.
Psychiatric symptoms, including mood disturbances and changes in personality, have also
been shown to begin well before the traditional clinical diagnosis is made and may worsen over
time. The changes in personality seen in the prodromal phase in HD patients are often reported
by immediate family and caregivers retrospectively. This is due in part to the fact that patients
are usually unaware that any changes have taken place. Depression and irritability can begin to
become apparent around the time cognitive changes are observed (Sturrock and Leavitt 2010).
While these mood changes are not substantial enough to merit clinical diagnosis based on DSM
criteria, they can affect the social functioning of prodromal patients. These psychiatric
disturbances may worsen after clinical diagnosis, especially due to the increased stress that the
disease onset brings. As patients begin to lose the ability to take care of themselves, depression
may lead to suicide ideation. The risk of suicide in HD patients has been shown to be 4 to 6
8
times more than that observed in the general population and typically occurs in the early stages
of disease when patients can carry out their suicides before complete loss of motor function
(Nehl and Paulsen 2004). In addition, a subset of patients may develop psychoses, including
paranoia and hallucinations in late stages. While these symptoms are uncommon, they are
poorly defined and lead to an even more severe loss of normal function. Together with the motor
and cognitive aspects of HD, the psychiatric symptoms observed lead to a devastating disorder
that requires the development of multifaceted treatment strategies to increase quality of life and
combat disease progression.
1.2 HD Neuropathology
While it is has been well established that the causative agent of HD is mutated htt, the
mechanism of its pathogenesis is not well understood. This is due in part to the fact that the
functions of the normal htt protein are widely unknown as well. It has been shown that htt
interacts with a variety of intracellular organelles including the plasma membrane, endoplasmic
reticulum, endosomes, Golgi apparatus, mitochondria, and the nucleus (Zheng and Diamond
2011). This evidence, along with the fact that htt is such a large protein (≈350 kDA), has lead to
the suggestion that htt functions as a scaffolding protein. However, this hypothesis has been
difficult to test as it is not yet possible to make a full-length crystalline structure of htt due to its
massive size. Knockouts of htt in mice have proved to be embryonic lethal, suggesting a vital
role of the protein in development (Imarsio et al. 2008). Analysis of specific domains on the htt
protein has lead to additional theories of its functions. Htt is thought to play a role in regulating
brain derived neurotrophic factor (BDNF) expression by regulating several transcriptional
factors. In addition, htt has been implicated as being involved in both long and short range
9
axonal transport through its interactions with a variety of proteins involved in vesicle trafficking
(Rubinsztein and Carmichael 2003). Despite several hypotheses, the true functions of htt in
neurons are still not fully understood. Since the normal htt protein is thought to play a role in a
variety of cellular processes in neurons, it is not surprisingly that the mutated form of htt has
been implicated in interfering many of these processes (Li et al., 2004). Specifically, some of the
effects that the mutant htt protein is thought to have include: activation of proteases leading to
proteolysis, reduction of protein degradation, transcriptional deregulation, altered mitochondrial
function, interference with axonal transport, and altered synaptic transcription. Taken together,
the disruption of these cellular processes ultimately affects the neuron’s potential for synaptic
plasticity (Figure 3).
Figure 3: Mutant Huntingtin Aggregates Affect Synaptic Plasticity.
While the role of the mutated htt protein has not been fully elucidated, it has been
implicated as playing a part in a variety of cellular processes leading to disruption
of synaptic plasticity. Figure taken from Li et al., 2004.
10
Since the functions of the normal htt protein remain elusive, it is not surprising that the
molecular mechanisms of the mutated form of htt in HD have continued to be controversial. The
majority of this controversy lies in whether the nuclear inclusion bodies that are one of the
hallmark signs of HD play a causative role in the pathogenesis of the disease or have a
neuroprotective function. The mutated htt protein has several different structural conformations
in the cell, including misfolded monomers, microaggregates (oligomers), and macroaggregates
(nuclear inclusion bodies). Weiss and coworkers studied the aggregation kinetics of the htt
protein and showed that the rate limiting step of aggregation is the formation of the aggregation
nucleus, with extension of the aggregate size occurring rapidly thereafter. The nucleation step
requires a single monomer to become misfolded in a way that recruits additional misfolded
proteins to attach. It is currently thought that in vivo multiple nucleation sites occur in neurons
such that several microaggregates, called oligomers, are formed and then accumulate into the
macroaggregates (Weiss et al. 2012). While it was initially thought that the nuclear inclusions
are the toxic species in HD, the discovery of oligomers as being a possible precursor to the
inclusion bodies in neurons have led several researchers to suggest that they cause the
cytotoxicity that ultimately leads to neuronal death (Figure 4).
11
Figure 4: Mutant Huntingtin Cell Conformations.
Mutated huntingtin protein has several conformations in cells, including
misfolded monomers, soluble oligomers, and inclusion bodies (macroaggregates).
While macroaggregates were considered to be the toxic species, new evidence has
suggested a neuroprotective role and that the previously unknown oligomers may
be responsible for neuronal cell death. Figure taken from Wolfe and Cyr, 2011.
While it has recently been challenged, there is an abundance of evidence that the
aggregation of the mutant htt protein into nuclear inclusions is directly related to the
pathophysiology of HD. Early studies into the role of htt aggregation in HD suggested that the
nuclear aggregates were the causative agent, since they appear prior to the signs of disease in
several mouse models (Davies et al. 1997). More specifically, it was determined that the number
of CAG repeats was correlated with the onset of disease. Increased CAG repeats correlated with
earlier symptom onset in HD patients. These findings were also associated with the number of
inclusion bodies that formed in the nucleus (Becher et al. 1998). Several mouse lines generated
to specifically target htt to the nucleus showed an acceleration of phenotypic onset with
12
increasing amounts of inclusion bodies (Benn 2005). Perhaps the most striking evidence for htt
aggregation playing a vital role in HD pathogenesis comes from therapeutic studies that inhibit
aggregation leading to beneficial effects. Chaperone proteins function in healthy cells to assist in
the folding of naïve proteins and have been demonstrated to prevent the aggregation seen in HD.
Overexpression of chaperones has been shown to decrease htt aggregation and neuronal death in
HD models (Imarsio et al. 2008). Along a similar line, overexpression of heat shock proteins,
which also play a role in the folding and unfolding of proteins in healthy cells have been shown
to reduce the amount of nuclear inclusions and decrease cell death in vitro (Chai et al. 1999).
Taken together, these studies have provided substantial evidence that htt aggregation plays some
role in HD pathogenesis.
Figure 5: Mutated Huntingtin Inclusion Body.
The mutated form of htt forms macroaggregates which can make their way into
the nucleus and form inclusion bodies as seen here. Figure taken from Ross et al.,
2004.
13
Despite the abundance of evidence supporting a causal link between htt aggregation HD
pathogenesis, this hypothesis has been contested by several recent studies that show that the
presence of these inclusion bodies does not correlate with excitotoxicity and cell death. As
previously discussed, one contention with this theory lies in the fact that the most vulnerable
neurons in HD, the medium spiny neurons in the striatum, do not have as much nuclear
inclusions as cortical neurons which display less neuronal death (Bates 2003). A variety of
explanations, including an increase of excitotoxicity or a decrease of necessary trophic factors
from those cortical neurons to the striatum, has been proposed as a means to explain this
phenomenon without shifting the paradigm away from inclusion bodies playing a causal role in
HD. However, these justifications have yet to be proven. In fact, recent evidence has suggested
that these inclusion bodies may have a neuroprotective role by sequestering more toxic
cytoplasmic oligomers of htt (Kuemmerle et al 1999; Lee et al. 2003; Wolfe and Cyr 2011). In
one such study, Saudou and coworkers created mutant htt proteins with 68 repeats and wild-type
proteins with 17 repeats, infused these proteins into both striatal and hippocampal neurons in
vitro, and demonstrated evidence that mutant huntingtin aggregates function in the nucleus to
cause neuronal cell death. However, their observations also suggest an inverse relationship
between the length of mutant huntingtin protein (171 and 480 amino acids) and the number of
inclusions bodies present. The amount of inclusion bodies did not correlate with the degree of
apoptosis in either the hippocampus or the striatum. In the striatum, cell survival was
comparable for two mutant proteins of different lengths, despite having different intranuclear
aggregate loads (Saudou et al. 1998). More importantly, in hippocampal neurons, the mutant htt
had no effect on neuronal death compared to infused wild-type proteins, despite being able to
form inclusion bodies. In another study, Arrasate and coworkers created mutant htt protein of
14
varying sizes and transfected them into striatal neurons. The authors determined which cells had
inclusion body formation and followed them two and four days post-transfection to measure cell
survival. After two days, the overall risk of cell death was the same in cells with and without
inclusion bodies. In addition, measurements taken four days post-transfection showed a
statistically significant increase in cell survival for those cells with inclusion bodies as compared
to those without (Arrasate et al. 2004). Together, this evidence directly contradicts the htt
aggregation hypothesis and even suggests a neuroprotective role of inclusion bodies.
1.3 Animal Models of Huntington’s Disease
Animal models, especially those in rodents, play an important role in elucidating
phenotypic progression, investigating potential underlying mechanisms, and to serve as a
template to screen new therapeutic modalities. In HD research a wide spectrum of rodent models
are available including knock-outs, knock-ins (KI), and transgenics (Hickey and Chesselet 2003;
Ramaswamy et al., 2007; Gil and Rego 2008). Many of these different models share features that
reflect HD pathophysiology and neuropathology. The major differences between these models
are symptom onset and disease progression.
The first animal models of HD used neurotoxins to lesion striatal neurons. When
administered directly to the striatum, the NMDA receptor agonist quinolinic acid (QA) lesioned
specific neurons while sparing others, a pattern that closely resembles the lesions observed in
human HD patients (Beal et al., 1991). The naturally occurring neurotoxin 3-nitroproprionic acid
(3-NP) inhibits succinate dehydronase in complex II of the electron transport chain, resulting in
impaired mitochondrial function and bilateral neuronal death in the striatum when administered
systemically. Patients exposed to this toxin developed HD-like symptoms of choreiform
15
movements of the limbs, head, and trunk which progressed to a Parkinsonian state similar to late
stage HD. These same symptoms were exhibited by healthy rats after chronic administration of
3-NP
(Borlongan et al., 1995). These models, however, lack the slow progression of disease, as
neuronal death appears rapidly post-infusion. Although displaying widespread striatal cell death
that mimics the pattern observed in human patients, rats do not exhibit the motor component that
is the hallmark of HD
(Furtado and Mazurek 1996).
Transgenic strains including R6/1 and R6/2 carrying an insertion of the human exon 1
expansion of CAG display onset of symptoms (motor and non-motor) within 1 to 2 months after
birth (Gil and Rego 2009; Cepeda-Prado et al. 2012). The extremely early onset of disease
phenotype in the R/2 mice makes it difficult to determine whether cognitive or motor
impairments appear first, as both appear concurrently. The rapidness of disease progression in
these mice does not accurately mimic the time course observed in patients with HD. In addition,
the lack of full-length mutant htt makes interpretation of molecular mechanisms difficult, as the
lack of the C-terminus means that htt cleavage is not modeled in these mice. However, the R6
model remains a popular model in studies of possible therapeutics due to the fact that they
present multiple outcome measures, including early death. In addition, R6 mice effectively
model many of the events thought to be relevant to pathogenesis, such as mitochondrial
dysfunction, excitotoxic cell death, and protein aggregate formation (Gil and Rego 2009;
Cepeda-Prado et al. 2012). A small number of studies in the R6/1 and R6/2 models have shown
little to no benefit of exercise on motor symptoms (Pang et al., 2006; van Dellen et al., 2008;
Wood et al., 2011; Harrison et al., 2013). In all of these studies, mice were exposed to voluntary
wheel running. Recent studies have challenged the utilization of voluntary running wheels as an
intervention in R/1 mice, where wild type mice ran much further than HD mice after 18 weeks of
16
age (Hickey et al. 2008; Cepeda et al. 2010). Thus, it is possible that the lack of observed
benefits of exercise in these models could be due to reduced motivation for wheel running.
Figure 6: Voluntary Running Wheels in the R6/2 HD Mouse Model.
HD animals run significantly less than WT animals on voluntary running wheels.
The distance run in voluntary running wheels was assayed in R6/2 mice and WT
controls prior to motor symptom onset. It was observed that R6/2 mice ran
significantly less than WT controls after 26 days of age. Figure taken from
Cepeda et al., 2010.
KI strains such as CAG
140
have a longer time frame until onset of symptomology
(Menalled et al. 2003). CAG
140
animals show early onset of HD pathology with nuclear
inclusions and changes in electrophysiology beginning as early as 2 months of life. Motor
impairments, however, may not present until the mice are a year old
(Brooks et al., 2012). The
delayed onset of motor dysfunction most accurately resembles the progression of HD in patients
as compared to other mouse models. Thus, KI mice provide a longer preclinical period before
motor symptom onset which can serve as a window of opportunity to investigate long-term
interventions (exercise, diet, etc) and their impact on disease onset, progression, and severity
17
(Hickey et al., 2002; Menalled et al., 2003; Rising et al., 2011). To date, studies in the CAG
140
KI mouse have focused on the motor impairments observed in the around 1 year of age
(Menalled et al., 2003; Dorner et al., 2007; Hickey et al., 2008; Rising et al., 2011). No study to
date has explored CAG
140
KI mice in the prodromal period for potential non-motor (mood)
disturbances. Additionally, as with the R6/1 and R6/2 mouse models, no study has directly
compared motorized treadmill running, rather than voluntary wheel running, with sedentary
CAG
140
KI mice for its potential as a therapeutic intervention to ameliorate behavioral symptoms
and attenuate pathological abnormalities.
1.4 Dopamine (DA)
Dopamine (DA), also referred to as 3-hydroxytyramine, is a catecholaminergic
neurotransmitter that generally exerts its effects on neuronal circuitry through the modulation of
fast-acting neurotransmitter systems including glutamate and γ- aminobutyric acid (GABA). All
catecholamines, including epinephrine (adrenaline), norepinephrine (noradrenaline), and DA are
derived from the amino acid tyrosine. L-tysosine is converted to L-DOPA by the enzyme
tysosine hydroxylase (TH). L-DOPA is subsequently converted into DA by the enzyme aromatic
L-amine acid decarboxylase. In a subset of neurons the process ends there and DA is the
primary neurotransmitter, but in other neurons DA is converted to norepinephrine and
epinephrine. TH is the rate-limiting enzyme in DA synthesis, and, as such, it is often used as a
molecular marker for DA levels in brain tissue (Daubner et al., 2011).
Once synthesized, DA is stored in vesicles by vesicular monoamine transporter (VMAT-
2). Upon stimulation, vesicles within the nerve terminals fuse with the presynaptic terminal and
release DA into the synaptic cleft, where DA activates receptors on pre- or post-synaptic
18
neurons, until it is removed from the synaptic cleft by DAT. DA is metabolized by monoamine
oxidase (MAO), aldehyde dehydrogenase (ALDH) and catechol-O-methyl transferase (COMT)
into 2 main metabolites: DOPAC (3,4-Dihydroxyphenyl-acetic acid) and HVA (homovanillic
acid). DA levels and DA metabolites can be measured via High-Performance Liquid
Chromatography (HPLC). The sum of the concentration of DA metabolites divided by the
concentration of DA gives the turnover ratio which it a helpful tool to analyze that amount of DA
available in the synaptic cleft to initiate an effect.
1.4.1 Dopamine Pathways in Brain
There four major DA pathways in the brain including (1) the nigrostriatal pathway, (2)
mesocortical pathway, (3) mesolimbic pathway, (4) tuberoinfundibular pathway (Figure 7). The
nigrostriatal pathway is primarily involved in the coordination of motor movement and is
thought to be altered in neurodegenerative movement disorders such as HD. It originates in DA
neurons in the substantia nigra pars compacta (SNpc), and projects to the striatum (Smith and
Villalba, 2008). The mesolimbic pathway is associated with pleasure, reward and goal directed
behavior, and therefore plays a role in many reward-based cognitive behavioral tasks. It
originates from DA neurons in the ventral tegmental area (VTA), and projects to the limbic area
via the Nucleus accumbens (NAcc) in the ventral striatum. The mesocortical pathway is
associated with motivational and emotional responses. It also originates from DA cells in the
VTA, and projects to the frontal cortex, specifically to the anterior cingulate cortex, entorhinal
cortex, prefrontal cortex (PFC), and olfactory bulbs. The tuberoinfundibular pathway regulates
the secretion of prolactin by the pituitary gland and is involved in maternal behavior. This
pathway originates from DA cells in the hypothalamus (arcuate nucleus) and projects to the
pituitary gland (Beaulieu and Gainetdinov, 2011).
19
Figure 7: Four Main Dopamine Pathways in the Brain.
The four major DA pathways in the brain are (1) nigrostriatal pathway (in blue,
SNpc compacta to the dorsal striatum (2) mesolimbic pathway (in red, VTA to
limbic area via the vStr/NAcc, (3) mesocortical pathway (in grey, ventral
tegmentum area to frontal cortex), and (4) tuberoinfundibular pathway (in green,
hypothalamus to pituitary). Figure taken from (Leucht et al., 2011).
Dopamine neurotransmission through these four pathways has been shown to play a
prominent role in a variety of important central nervous system functions including motor
control, arousal, cognition, reward, sleep, attention, and learning (Beaulieu and Gainetdinov,
2011). Perturbations to the DA-system are linked to a number of disorders including, HD,
Parkinson’s disease (PD), schizophrenia, ADHD, and Tourette’s syndrome. Specifically of
interest to this dissertation is the role that DA has on dopamine receptors in the striatum and
frontal cortex, such that impairments in dopamine neurotransmission in HD lead to problems
with motor and cognitive functions as well as mood disturbances (Chen et al., 2013).
1.4.2 Dopamine Receptors
DA exerts its physiological effects through the activation of one of five G-protein
coupled DA receptors (D
1
R, D
2
R, D
3
R, D
4
R, D
5
R) (Beaulieu and Gainetdinov, 2011). DA
20
receptors are divided into two major classes based on structural, pharmacological, and signaling
properties (Thompson et al., 2010). D
1
-like receptors, which include the D
1
R and D
5
R subtypes,
are functionally distinct from D
2
-like receptors, which include the D
2
R, D
3
R and D
4
R subtypes
(Figure 8). The work discussed in this dissertation will focus on the D
1
R and the D
2
R because
they are the most abundantly expressed DA receptors in brain, particularly in the regions
associated with the nigrostriatal, mesolimbic, and mesocortical pathways, such as the straitum,
NAcc, olfactory bulb and frontal cortex (Jaber et al., 1996, Lidow et al., 2003, Araki et al.,
2007). These receptors are also found in much lower levels in the hippocampus, hypothalamus,
SN and VTA.
Figure 8: Distribution of Dopamine Receptors in the Brain.
The work in this dissertation focuses on the D
1
and D
2
DA receptors in the brain.
These recetpros are found throughout the brain, but their expression levels are
particularly high in the striatum where DA plays a prominent role in motor
functioning and the frontal cortex where dopamine plays a role in cognition.
Figure taken from Brichta et al. 2013.
21
1.4.3 Dopamine Signaling Pathways
The D
1
- and D
2
-like receptors modulate functionally distinct intracellular signaling
pathways. While DA signaling exerts numerous downstream effects, DA receptor function is
often associated with the regulation of cyclic AMP (cAMP) and protein kinase A (PKA)
signaling. DA binding to D
1
-like receptors causes the receptor to couple to Gα
s/olf
proteins, which
activates adenylate cyclase (AC). AC then stimulates cAMP production, which subsequently
enhances PKA activation. On the other hand when DA binds to D
2
-like receptors, they couple to
Gα
i/o
proteins that inhibit AC, and thus cause a decrease cAMP and PKA activity (Beaulieu and
Gainetdinov, 2011). PKA activation has a variety of downstream effects on several substrates
including cAMP response element-binding protein (CREB), ionotropic glutamate receptors (i.e.
AMPA receptors), GABA receptors, and voltage-gated potassium (K
+
), sodium (Na
+
), and
calcium (Ca
2+
) ion channels (Gerfen et al., 1990; Kebabian & Calne, 1979). Of these substrates,
perhaps the most extensively studied is the 32-kDa dopamine and cAMP-regulated
phosphoprotein (DARPP-32), a multifunctional phosphoprotein involved in the modulation of
cell signaling in response to DA stimulation. DARPP-32 is of particular interest because it has
been shown to modulate glutamate receptors, which is the main excitatory neurotransmitter in
striatal signaling pathways (Figure 9). D
1
R stimulation results in synaptic GluA1 insertion, and
enhances Glu2-lacking AMPAR expression. Conversely, D
2
R stimulation reduces GluA1
synaptic insertion, and restores GluA2-containing AMPAR expression. This modulation of
glutamate receptors is necessary for complex behaviors such as motor function, learning and
memory. Therefore, aberrant DA signaling via DARPP-32 can play a profound role in altering
glutamate receptors and downstream signaling, ultimately resulting in phenotypic impairments.
22
Figure 9: Dopamine Signaling Pathways in the Brain.
Dopamine D
1
- and D
2
-like signaling exerts opposing effects on intracellular
signaling pathways that regulate a number of downstream targets including
GluA1 surface expression. Green lines indicate stimulatory signaling, red lines
indicate inhibitory signaling, Green circles are PKA-dependent phosphorylation
sites (DARPP32thr34; GluA1ser845). AC = adenylate cyclase; PKA = protein
kinase A, DARRP32 = 32-kDa dopamine and cAMP-regulated phosphoprotein.
Figure courtesy of Dr. Natalie Kintz.
1.5 Direct and Indirect DA Pathways in the Basal Ganglia
As mentioned briefly above, alterations in dopamine neurotransmission in the basal
ganglia (BG) and frontal cortex are thought to play a prominent role in the emergence of motor
and cognitive symptoms in HD. The BG is composed of a group of subcortical nuclei involved
in the coordination of information from sensorimotor, motivational, and cognitive brain areas
(Figure 10). This coordination allows the BG to modulate complex behaviors, including
movement and motor learning. The BG is comprised of the striatum (which in humans consists
of the caudate nucleus and putamen) as well as the external and internal segments of the Globus
Pallidus (GPe and GPi respectively), the subthalamic nucleus (STN), and the substania nigra pars
reticulata (SNr) (Blandini et al., 2000). In the classic model of the basal ganglia, the dorsal and
D
1
-like
(D
1
, D
5
)
D
2
-like
(D
2
, D
3,
D
4
)
AC
PKA
DARPP32
Thr
34
P
PP-1
GluA1
Ser
845
P
GluA1
GluA1
GluA2-lacking
AMPARs
= stimulatory signaling
= inhibitory signaling
P
= PKA-dependent
phosphorylation site
23
ventral striatum are the main input nuclei, while the GPi and SNr are the principal output nuclei.
Input to the striatum is processed in two main pathways: (1) a monosynaptic “direct” pathway
from the striatum to the GPi/SNr; and (2) a polysynaptic “indirect” pathway that projects from
the striatum to the GPi/SNr via the GPe and the STN. Stimulation of the direct pathway
facilitates motor movement, while stimulation of the indirect pathway inhibits motor movement.
Neurons in the GPi and SNr have projections to several thalamic nuclei, which in turn innervate
neurons in prefrontal, premotor, and motor cortical areas (Redgrave et al., 2010).
Figure 10: Direct and Indirect Pathways in the Basal Ganglia (BG).
The striatum is the main input nuclei, while the Globus Pallidus (GPi) and
Substantia Nigra pars reticulata (SNr) are the main output nuclei of the BG.
Striatal signaling is divided into the direct and indirect pathways, which exert
opposing effects on the BG circuit. The direct pathway originates from D
1
R-
MSNs that project directly to the GPi and SNr output nuclei. The indirect pathway
originates from D
2
R-MSNs that project to the external segment of the Globus
Pallidus (GPe) and the subthalamic nucleus (STN), which together project to the
GPi and SNr output nuclei. The GPi and SNr project to the thalamus, which is
turn projects to prefrontal, premotor, and motor cortical areas. Figure taken from
Gerfen and Surmeier, 2011.
24
Medium spiny neurons (MSNs) are the main neuronal population in the striatum,
representing 90-95% of all striatal cells. MSNs utilize γ-aminobutyric acid (GABA) as a
neurotransmitter and they express high levels of DA and glutamate receptors (Beaulieu and
Gainetdinov, 2011).
There are two structurally and functionally different MSN populations in the striatum,
characterized by their axonal projections and by differential expression of neurotransmitter
receptors in each cell type (Figure 11). Direct pathway MSNs (D
1
R-MSNs, also referred to as
striatonigral MSNs) preferentially express the D
1
DA receptor while indirect pathway MSNs
(D
2
R-MSNs, also referred to as striatopallidal MSNs) preferentially express the D
2
DA receptor
(Gerfen et al., 1990, Gerfen and Surmeier, 2011, Gittis et al., 2011). Importantly, D
1
R-MSNs and
D
2
R-MSNs respond differently to DA. D
1
R-MSNs are excited by dopamine and result in
increased GABAergic signaling, while D
2
R-MSNs are inhibited by DA and thus suppress
GABAergic signaling. This differential effect of DA on striatal MSNs leads to opposing effects
on motor behavior. Briefly, activation of the direct pathway through stimulation of D
1
R-MSNs
facilitates motor behavior, whereas indirect pathway activation suppresses motor behavior
(Kreitzer and Malenka, 2008, Cerovic et al., 2013). Changes in DA signaling through the direct
and indirect pathways have been implicated to a number of disorders including, HD, Parkinson’s
disease, schizophrenia, and ADHD (Beaulieu and Gainetdinov, 2011). Those alterations in
striatal signaling observed in HD that are specifically relevant to this dissertation will be
discussed in more detail below.
25
Figure 11: Medium Spiny Neuron (MSN) Cell Types in the Striatum.
D
1
R MSNs, also called direct pathway or striatonigral pathway MSNs, primarily
express D
1
Rs and are activated by DA to facilitate motor behavior. D
2
R-MSNs,
also called indirect pathway or striatopallidal MSNs are de-activated by DA to
facilitate motor behavior. Adenosine A
2A
Rs are selectively expressed on D
2
R-
MSNs in the striatum. D
2
R-MSNs are activated by adenosine A
2A
Rs to suppress
motor behavior. Figure courtesy of Dr. Natalie Kintz
1.6 DA Alterations and Behavioral Inflexibility in HD
Behavioral inflexibility is defined as the inability to change behaviors in order to adapt to
new environmental stimuli, such that one disengages from a previous non-rewarded behavior in
lieu of a new behavioral strategy. In intact cortico-basal ganglia pathways, repetitive behaviors
lead to habit formation through learning (Aron and Poldrack 2006; Aron et al., 2007).
Neuropsychiatric disorders, addiction, and neurodegenerative diseases that alter basal ganglia
pathways may result in extremely repetitive behaviors, or stereotypies. These stereotypies range
from behavioral inhibition and attentional disturbances observed in attentional deficit
hyperactivity disorder and obsessive compulsive disorder to the inability to suppress movement
as seen in tics and vocalizations in Tourette’s and chorea in HD. While these behaviors are
Receptor Pathway Result of Activation
D
2
R Indirect
Inhibits D
2
R-MSN firing
Facilitates behavior
A
2A
R Indirect
Facilitates D
2
R-MSN firing
Suppresses behavior
D
1
R Direct
Facilitates D
1
R-MSN firing
Facilitates behavior
D
2
-MSN
Enkephalin
A
2A
R
D
2
R
D
1
-MSN
Substance P
dynorphin
D
1
R
26
strikingly different, the central feature of all of them lies in compromised dopamine
neurotransmission in the striatum (Frank et al., 2004; Beste et al, 2010).
Many studies have associated impairments in cognitive flexibility with aberrant
dopamine (DA) neurotransmission in the striatum. Specifically, patients with DA impairments
in the basal ganglia display difficulties with set-shifting tasks (Cools et al., 2006; Dang et al.,
2012) and rats with lesions of the medial striatum have impairments in spatial and reversal
learning tasks (Whishaw et al, 1987; Pisa and Cyr, 1990). Disruptions of DA signaling in the
direct and indirect pathways impair set-shifting and reversal learning in different ways (Yawata
et al., 2012). Specifically, a reversible blockade of D1 containing MSNs in the direct pathway
resulted in impaired learning in the initial phase of the task as well as in the late half of the
relearning phase, suggesting deficits in the acquisition of reward based learning. On the other
hand, a reversible blockade in D2 containing MSNs in the indirect pathway resulted in impaired
learning in the early part of the relearning phase, suggesting deficits in cognitive flexibility
(Yawata et al., 2012). Taken together, these studies demonstrate that DA signaling plays a
significant role in behavioral flexibility, specifically in cognition.
27
Figure 12: DA Pathway Involvement in T-Maze Reversal Learning Task.
The initial learning phase of the T-maze is driven by D1R containing MSNs in the
direct pathway to go to the goal (reward). When the reward is switched in the
reversal learning phase of the task, the early portion, when the animal must learn
to disregard the previous strategy, is modulated by the indirect pathway and D2R
containing MSNs. The later portion when the mouse learns to find the reward is
once again driven by the direct pathway and D1R containing neurons. Figure
taken from Yawata et al. 2012.
1.6.1. Dopamine Alterations in HD Patients
Aberrant DA neurotransmission in the direct and indirect pathways in the basal ganglia
has a significant role in the motor dysfunction, cognitive impairments, and mood disturbances
observed in HD. One of the earliest studies to show that altered DA signaling is an underlying
mechanism modulating motor, cognitive, and mood symptoms linked the stimulation of DA
receptors by levadopa (L-DOPA) to the development of chorea in the offspring of HD patients
(Klawans et al., 1970). Early analysis of the brains of post-mortem HD patients demonstrated an
increase in DA levels and showed that DA-depleting agents as well as DA receptor agonists had
beneficial therapeutic effects (Bird 1980; Spokes 1980). Later on, however, it was discovered
28
that this increase in DA levels occurs early in HD disease progression (Garrett and Soares-Da-
Silva, 1992), and the analysis of post-mortem brains of late-stage HD patients indicated that the
levels of TH, DA and HVA were all reduced in the caudate (Bernheimer et al., 1973; Kish et al.,
1987; Bedard et al., 2011). Together, these studies suggested that DA levels may show biphasic
changes that are dependent on the stage of disease progression, with increases in DA levels early
in the disease followed by decreases in DA levels later on (Figure 13). These biphasic changes
can also explain the progression of motor symptoms in HD patients, as increased DA release has
been shown to induce chorea and decreased DA levels have been linked to akinesia in HD
patients (Bird 1980; Spokes 1980).
Figure 13: Basal Ganglia DA Pathway Changes in Early and Late HD.
Early HD disease progression is marked by an increase in involuntary motor
movements (chorea). These symptoms can be explained by decreased firing in
the indirect pathway, leading to an overall decreased inhibition of the thalamus
and overexcitement of the motor cortex. Later in HD, patients exhibit
Parkinsonian akinesia, due to decreased firing in both pathways resulting in
increased inhibition of the thalamus and decreased stimulation of the motor
cortex. Figure taken from Schwab et al., 2015
29
In addition to changes in the levels of DA itself, several studies focusing on changes in
DA receptors have shown decreases in the expression of DA receptors in the striatum that
worsens with disease progression (Table 2). In HD patients, presymptomatic HD carriers have
been shown to have reduced D2 receptor binding (van Oostrom et al., 2005; van Oostrom et al.,
2009). Symptomatic HD patients exhibit a decrease in both D1 and D2 receptor binding that
progressively worsens over time (Ginovart et al., 1997; Pavese et al., 2010). Positron emission
topography (PET) imaging has demonstrated a reduction in dopamine receptors in the striatum in
both presymptomatic and symptomatic HD patients (Richfield et. al, 1991). This loss of striatal
DA receptors has been further confirmed in imaging studies (Antonini et al., 1996; Weeks et al.,
1996). Loss of DA receptors in the striatum has been shown to strongly correlate with declines
in executive function, learning, and memory in HD, which may be due to altered synaptic
plasticity (Backman et al., 1997; Backman and Forde, 2001).
1.6.2. Dopamine Alterations in HD Mouse Models
The changes in DA levels and DA receptor expression observed in HD patients have been
investigated further in animal models (Table 2). The vast majority of studies have investigated
changes in DA neurotransmission in the R6/2, R6/1, and YAC 128 HD mouse models, which
show overt motor symptoms at 5-7 weeks, 16-20 weeks, and 6 months of age respectively
(Schwab et al., 2015). DA levels have been shown to be reduced in all three of these models. In
R6/2 mice, DA levels are significantly reduced (74% of control) at 6 weeks of age and by 10-11
weeks drops to 30% of that seen in WT controls (Johnson et al., 2006). Dopamine release is also
affected early in these mouse models as evidenced when induced by amphetamines (Johnson et
al., 2006; Callahan et al., 2011). Similar results have been shown in R6/1 mice (Peterson et al.,
30
2002). The YAC128 mice also exhibit diminished DA levels which reach 50% of WT controls
by 10 months of age (Callahan et al., 2011).
Table 2: DA and DA Receptor Levels in HD Patients and Animal Models.
Changes in DA levels in HD patients and HD animal models follow a biphasic
time-dependent pattern with early increases in DA followed by decreased DA in
later stages. DA receptors levels gradually decrease throughout HD disease
progression, with changes in D2 receptor seen even in presymptomatic HD
patients. Decreased DA receptor levels have also been observed in animals
models. Table taken from Chen et al., 2013.
31
Changes in DA receptor levels in HD mouse models have also been observed to mimic
those seen in HD patients. Several studies have demonstrated that D1 and D2 receptor binding is
reduced in the striatum early in disease progression in R6/1 and R6/2 mouse models, leading to
altered dopamine pathway signaling (Cha et al., 1998; Bibb et al., 2000; Ariano et al, 2002;
Peterson et al., 2002). Additionally, YAC128 mice have been shown to have reduced mRNA
expression of D1 and D2 receptors (Pouladi et al., 2012). Dramatic decreases in DA metabolites
(DOPAC and HVA) have also been observed in R6/1, R6/2, and YAC128 mice, which correlate
strongly to the emergence of motor impairments in these models (Hickey et al., 2008; Callahan
et al., 2011; Ortiz et al., 2011). While changes in DA neurotransmission have been extensively
studied in the aforementioned transgenic mouse models, few if any studies have attempted to
observe similar changes in knock-in models such as the CAG
140
model utilized in all of the
studies covered in this dissertation.
1.7 Brief Summary
Huntington’s disease (HD) is a devastating and progressive neurodegenerative disorder
characterized by a triad of motor dysfunction, cognitive impairments, and mood disturbances.
The hallmark pathological sign is the progressive neuronal loss of striatal medium spiny neurons
(MSNs) in the caudate nucleus and putamen as well as the cerebral cortex due to the aberrant
intracellular accumulation of Htt protein (Gil and Rego 2008). While genetic screening can
identify patients at risk, clinical diagnosis is not confirmed until patients display overt motor
features, however, several recent studies have identified cognitive and psychiatric disturbances in
as early fifteen years before diagnosis (Julien et al., 2007; Tabrizi et al., 2013; Epping et al.,
2013). HD has no cure and treatment is limited to an attempt to attenuate symptoms.
32
In our research group, we are interested in determining the underlying molecular mechanisms
by which experiences in the form of intensive exercise can enhance neuroplasticity within the
basal ganglia and cortical circuitry, important for both HD and PD. Even though HD is inherited
in an autosomal dominant fashion, the time frame of clinical manifestation is influenced by non-
genetic factors suggesting that factors of lifestyle such as diet and physical activity influence
may modify disease progression and severity. Exercise has been shown to have a beneficial
effect on disease progression and symptoms in several neurodegenerative disorders, including
Alzheimer’s disease (Verghese et al., 2003; Abbott et al., 2004; Yu et al., 2011; Venturelli et al.,
2011) and Parkinson’s disease (Frazzitta et al., 2013; Corcos et al., 2013; Petzinger et al., 2010;
Petzinger et al., 2013; Speelman et al. 2011). Specifically, our lab has demonstrated that
motorized treadmill running increases neuroplasticity and attenuates pathophysiological
mechanisms in a mouse model of Parkinson’s disease by enhancing dopamine neurotransmission
(Petzinger et al., 2007) and glutamate synaptic neurotransmission (Kintz et al., 2013), and by
reversing dendritic spine loss (Toy et al., 2014), culminating in an overall rescue of behavioral
motor deficits.
The studies described in this dissertation examine the effects of intensive treadmill running
(forced exercise) on experience-dependent neuroplasticity in a rodent model of HD. We have
selected the CAG
140
HD mouse model because this model shows slow progression with motor
symptoms emerging at 12 months of age (Menalled et al., 2003; Hickey et al., 2008; Rising et
al., 2011). Mice will begin running at 1 month of age and continue for 11 months. We speculate
that this long prodromal phase presents an opportunity for intervention and alterations in lifestyle
that may impact disease progression, a characteristic not available in fast-progressing models
such as the R6/1, R6/2, and other transgenic strains (Hickey et al., 2003).
33
Chapter 1 will describe our investigation into the effects of long-term treadmill exercise on
depression-like behaviors in the CAG
140
mouse prior to the onset of any motor symptoms. We
first explored for prodromal mood disturbances and upon finding an increase in depression-like
behaviors at 4 and 6 month, we attempted to attenuate those effects through treadmill exercise.
We also attempted to determine the underlying mechanisms of any beneficial effect that exercise
has on mood disturbances by analyzing striatal pathology and neurochemical changes in
dopamine, norepinephrine, and serotonin.
Chapter 2 focuses on dopamine neurotransmission and cognitive impairments in HD, again
in the prodromal period before motor symptom onset. A probe for cognitive deficits showed
impairments in cognitive flexibility, specifically those related to aberrant dopamine signaling in
the indirect pathway. As in Chapter 1, we explored the role of long-term treadmill exercise in an
attempt to rescue the cognitive deficits observed. In an attempt to link the cognitive deficits with
dopamine neurotransmission, we also analyzed the dorsal and ventral striatum for TH, D1
receptor, and D2 receptor protein levels in sedentary and exercised animals.
Chapter 3 moves past the prodromal period studied in Chapters 1 and 2 and instead
investigates motor symptom onset in the CAG
140
animals. As before, mice were exercised from
an early age and were assessed for any beneficial role in delaying the onset of motor symptoms
observed in sedentary CAG
140
mice. As in Chapter 1, we analyzed striatal pathology to
determine if the benefits seen in Chapter 1 carried over to the one year time point. Additionally,
we measured the area of the lateral ventricles and striatum as an indirect way to determine if cell
loss had begun to noticeably occur.
Finally in Chapter 4, we employ imaging techniques to search for areas in the CAG
140
mouse
brain with hyper- or hypoperfusion. Specifically, the regional cerebral blood flow was mapped
34
and comparisons were drawn between sedentary CAG
140
animals and WT controls. Several
striking changes in blood flow were observed, suggesting that HD causes a functional
reorganization of brain networks.
As a whole, this dissertation proposes that long-term exercise delays the onset of motor,
cognitive, and neuropsychiatric symptoms in the CAG
140
HD mouse model. We provide
evidence of dopamine pathway dysfunction in the striatum through imaging and immunoblot
techniques. Furthermore, we propose that the beneficial effects of exercise in delaying symptom
onset come as a result of delaying htt aggregation events and restoring aberrant dopamine
neurotransmission in the striatum. The goal of this dissertation is to provide possible targets for
pharmacological and genetic interventions targeting neuroplasticity in the basal ganglia in an
attempt to translate these studies in the mouse model to HD patients with the hope of delaying
symptom onset and improving the overall quality of life.
35
CHAPTER 2: Treadmill Running Delays the Onset of Depression-like Behavior and
Striatal Pathology in the CAG140 Knock-in Mouse Model of Huntington’s Disease
Author List: DP Stefanko, WK Yamasaki, P Lee, DN Garcia, GM Petzinger, and MW Jakowec
ABSTRACT
Depression, and other neuropsychiatric disturbances are common during the prodromal
phase of Huntington’s disease well before the onset of classical motor symptoms of this
degenerative disorder. The purpose of this study was to examine the potential impact of exercise
in the form of motorized treadmill running on depression-like behavior in the CAG
140
knock-in
mouse model of Huntington’s disease. This model has a long lifespan compared to other rodent
models of this disease including the late stage onset of motor features after 12 months of age.
This provides the opportunity to investigate the effects of interventions such as exercise initiated
early in life. Motorized treadmill running was initiated at 4 weeks of age (1 hour per session, 3
times per week) and continued for 6 months. Analysis of depression-like behavior using the tail-
suspension and forced-swim tests showed a significant attenuation of this feature. In addition, we
examined the pattern of expression of huntingtin (htt) protein in the striatum and showed that the
number and immunohistochemical staining of intranuclear aggregates was also significantly
reduced. Exercise also restored decreases in dopamine, norepinephrine, and serotonin
concentrations observed in the striatum. Together these findings begin to address the potential
impact of life-style and early interventions such as exercise on modifying disease progression.
KEYWORDS: huntingtin, exercise, Htt protein, tail suspension, forced swim, Q140, behavior
36
INTRODUCTION
Huntington’s disease (HD) is an autosomal dominant neurodegenerative disorder caused
by an excessive polyglutamine expansion in exon 1 of the Huntington (HTT) gene (The
Huntington's Disease Collaborative Research Group, 1993). Pathologically, HD is characterized
by progressive neuronal loss of striatal medium spiny neurons (MSNs) in the caudate nucleus
and putamen as well as the cerebral cortex due to the aberrant intracellular accumulation of Htt
protein (Gil and Rego 2008). Clinically, HD is characterized by a progressive decline in motor
function and cognitive ability as well as neuropsychiatric disturbances culminating in premature
death 15 to 20 years after diagnosis. Affected individuals typically display motor symptoms
(chorea, dyskinesia, and tremors) in the fourth decade of life. Recent studies have shown that
impairments in cognition and mood appear up to a decade prior to motor symptom onset, a stage
deemed the prodromal period (Julien et al., 2007; Tabrizi et al., 2013; Epping et al., 2013). The
mood disturbances observed in HD, particularly depression, make a significant contribution to
the overall morbidity of the disorder (Hamilton et al, 2003; Beglinger et al., 2010), as well as
early mortality due to suicide (Cummings 1995; Fiedorowicz et al., 2011).
The discovery of the mHtt gene for HD has allowed for the development of a wide
spectrum of rodent models including transgenic, gene knockout, and gene knock-in models. The
heterogeneity of the constructs in terms of gene expression has resulted in a wide spectrum of
disease progression and severity in these models. For example, the first two HD transgenic lines,
R6/1 and R6/2, developed in 1996 by Mangiarini and colleagues are to date the most studied
models in the HD field and have a relatively early onset of motor symptoms (tremors,
dyskinesia, gait abnormalities, and paw clasping) with onset at about 15 weeks in R6/1 model
and 5 to 6 weeks for the R6/2 model (Carter et al., 1999; Mangiarini et al., 1996). These features
37
are in contrast to other models, such as the knock-in (KI) mouse model with 140 CAG repeats
(CAG
140
) in exon 1 of the mouse huntingtin gene ortholog (hdh) developed by Menalled and
colleagues (Menallled et al., 2003). This mouse termed the CAG
140
KI model develops overt
motor symptom onset (tremors, paw clasping, and dyskinesia) at around 1 year of age (Dorner et
al., 2007; Hickey et al., 2008). Mutant Htt protein aggregates in the CAG
140
KI mice model are
present well before motor symptom onset, with progressive accumulation of Htt protein into
intranuclear inclusions first observed at 2 months in striatal MSNs (Menalled et al., 2003).
These mice also show atrophy and neuronal loss in the striatum in adult mice (ages 20-26
months), a hallmark of advanced HD in human patients. Taken together, the CAG
140
KI
resembles the progressive behavioral and pathological disease progression HD in patients as
compared to other mouse models.
Even though HD is inherited in an autosomal dominant fashion the time frame of clinical
manifestation is influenced by non-genetic factors suggesting that factors of lifestyle such as diet
and physical activity may modify disease progression and severity. Exercise has been shown to
have a beneficial effect on disease progression and symptoms in several neurodegenerative
disorders, including Alzheimer’s disease (Verghese et al., 2003; Abbott et al., 2004; Yu et al.,
2011; Venturelli et al., 2011) and Parkinson’s disease (Frazzitta et al., 2013; Corcos et al., 2013;
Petzinger et al., 2010; Petzinger et al., 2013; Speelman et al. 2011). Specifically, our lab has
demonstrated that motorized treadmill running increases neuroplasticity and attenuates
pathophysiological mechanisms in a mouse model of Parkinson’s disease by enhancing
dopamine neurotransmission (Petzinger et al., 2007), glutamate synaptic neurotransmission
(Kintz et al., 2013), and reversing dendritic spine loss (Toy et al., 2014), culminating in an
overall rescue of behavioral motor deficits.
38
A small number of studies in the R6/1 and R6/2 models have shown little to no benefit of
exercise on motor symptoms (Pang et al., 2006; van Dellen et al., 2008; Wood et al., 2011;
Harrison et al., 2013). In all of these studies, mice were exposed to voluntary wheel running.
Recent studies have challenged the utilization of voluntary running wheels as an intervention in
R/1 mice, where wild type mice ran much further than HD mice after 18 weeks of age (Cepeda et
al. 2010). Thus, it is possible that the lack of observed benefits of exercise in these models could
be due to reduced motivation for wheel running.
To date, studies in the CAG
140
KI mouse have focused on the motor impairments
observed in the around 1 year of age (Menalled et al., 2003; Dorner et al., 2007; Hickey et al.,
2008; Rising et al., 2011). No study to date has explored CAG
140
KI mice in the prodromal
period for potential non-motor (mood) disturbances. Additionally, as with the R6/1 and R6/2
mouse models, no study has directly compared motorized treadmill running, rather than
voluntary wheel running, with sedentary CAG
140
KI mice for its potential as a therapeutic
intervention to ameliorate behavioral symptoms and attenuate pathological abnormalities. To
this end, this study uses long-term motorized treadmill running in the CAG
140
KI mouse model to
observe its role in depression-like behavior in the forced swim and tail suspension tests.
Additionally, this study follows the progression of intranuclear inclusion body formation in
striatal MSNs in exercised and sedentary animals. The results reported here provide evidence of
the benefits of motorized exercise to delay HD symptom onset and point to potential
pharmacological targets for future studies in the CAG
140
model.
39
MATERIALS AND METHODS
Animals and Groups
For these studies we used knock-in mice that contained a chimeric mouse/human exon 1
with 140 CAG repeats inserted into the mouse gene by homologous targeting (Menalled et al.,
2003). CAG
140
KI mice were produced in-house using lines descended from heterozygous
pairing. Wild-type littermates were used for controls. The founder mice for our colony were a
generous gift of Drs. Michael Levine and Carlos Cepeda (UCLA) with permission from Dr. Scott
Zeitlin (University of Virginia) through a Material Transfer Agreement. These mice were
backcrossed onto the C57BL/6J background annually to maintain vigor. Mouse genotypes from
tail biopsies were determined using real time PCR (Transnetyx, Inc., Cordova, TN). Mice were
randomly assigned to one of four groups based on WT or homozygous for CAG KI genotype
including: (i) wildtype, (ii) wildtype + exercise, (iii) CAG KI, and (iv) CAG KI + exercise. Only
male mice were utilized in these studies. Mice were group housed with a reverse light cycle
(lights off from 7 a.m. to 7 p.m.) and were allowed access to food and water ad libitum.
Experimental procedures were approved by the University of Southern California’s Institutional
Animal Care and Use Committee (IACUC) and were conducted in accordance with the National
Research Council Guide for the Care and Use of Laboratory Animals (DHEW Publication 80-23,
2011, Office of Laboratory Animal Welfare, DRR/NIH, Bethesda, MD). All efforts were made
to minimize animal suffering and to reduce the number of animals used to achieve statistical
significance.
40
Exercise Regimen
Mice were weaned at postnatal day 28 and subjected to treadmill running on a Model
EXER-6M Treadmill (Columbus Instruments, Columbus, Ohio). The treadmill exercise protocol
was conducted based on our previously publications with modifications (Fisher et al., 2004).
Briefly, exercise was initiated on postnatal day 28. In week one, mice in the exercise groups ran
at a speed of 8.0 ± 0.5 m/min for 40 minutes and they were closely monitored for any adverse
reaction to the treadmill, inability to run, or failure to learn the task. No mice needed to be
excluded. Exercise mice started at a velocity of 10.0 ± 1.5 m/min, and ran 3 times per week for
the 6-month duration of this study. Treadmill speed was gradually increased to 20 ± 1.5 m/min
by the final month. A non-noxious stimulus (metal beaded curtain) was used as a tactile incentive
to prevent animals from drifting back on the treadmill. All mice were weighed at the end of each
week and closely assessed for adverse reactions including stress. In the past we have found
treadmill running to not be stressful based on the evaluation of anxiety, depression, and
corticosterone levels (Gorton et al., 2010).
Behavioral Testing
Behavioral testing was carried out to evaluate (i) depression-like behavior and the effects
of motorized treadmill running and (ii) motor behaviors in this HD model. For depression-like
behavior, the forced swim test and tail suspension test were employed at 2, 4 and 6 months of
age in all four groups of mice. To evaluate motor behavior and to confirm that motor deficits
were not evident in HD mice at these early time points the accelerating rotarod was utilized as
well as an analysis of gait examining painted paw prints on a paper runway. To determine the
ability and motivation of mice to run on a voluntary running wheel a subset of mice independent
41
of those used for motorized treadmill running were subjected to running wheels in their home
cage.
Forced Swim Test
The forced swim test was based on the method of Porsolt et al. (1977). Briefly, mice
were placed individually into a glass cylinder (20 cm x 14 cm) filled with 10 cm deep fresh water
and were allowed to swim freely for 360 s. An experienced experimenter blind to treatment and
genotype of each mouse manually scored total immobility time of each mouse. A mouse was
considered to be immobile when it ceased struggling and moved only to maintain balance or to
remain afloat.
Tail Suspension Test
The tail suspension test was based on the method of Steru et al. (1985). Briefly, mice
were suspended by their tail with their head 5 cm over a table and videotaped for 360 s. An
experienced experimenter blind to treatment and genotype of each mouse manually scored total
immobility time of each mouse. A mouse was considered to be immobile when it ceased
struggling and its limbs remained motionless.
Accelerating Rotarod
Mice were tested for motor function on a rotating spindle 6.0 cm in diameter (Orchid
Scientifics, Nashik, India) according to the methods of Rozas et al. (1998). Prior to the test, mice
were acclimatized to the rotarod for a period of 10 minutes at 5 rpm. During acclimatization,
mice that fell were replaced onto the spindle. Each mouse was subjected to 3 trials at speeds
42
increasing from 5 rpm to 35 rpm over 300 seconds at 1 rpm intervals, with 300 second rest
between trials. Motor function and coordination was assessed as the average latency to fall of
the second and third trials.
Gait Analysis
Stride length and paw overlap were measured according to the methods of Fernagut et al.
(2002). Briefly, mice were placed at the end of an illuminated pathway (4.5 cm wide, 42 cm
long, with walls 12 cm high) and were allowed to run on a white strip of paper towards a dark
goal box (20 x 17 x 10 cm). Mice were placed into the goal box for 120 s and were acclimatized
into the apparatus for two trials, after which gait was measured in a single trial. Prior to being
placed in the apparatus, the forelimbs and hind limbs of the mice were painted two different
colors to differentiate between front and back paw prints. Stride length was measured as the
distance between successive hind limb paw prints. Paw overlap was measured as the distance
between the center of overlapping forelimb and hind limb paw prints. The average of three
stride lengths and paw overlaps were taken for each animal.
Voluntary Running Wheel
Voluntary wheel running was conducted in separate cages each containing a running
wheel equipped with a bicycle odometer (CatEye, Boulder, CO) to measure distance and speed.
Mice were habituated to the running wheel over a 6-hour period for 2 days. On the third day, the
total distance run in a 6-hour period was measured. During habituation and testing trials each
mouse had access to the running wheel for an equal amount of time. Running wheel distance
43
was obtained every two weeks for a cohort of HD KI mice and WT controls beginning at 2
months of age.
Immunohistochemical Staining for Htt Aggregates
A total of 32 animals (8 per experimental group) were used for immunohistochemical
analysis. Immunohistochemistry was performed on 20 µm-thick slices containing the striatum
between 0.14 mm to 0.62 mm anterior to bregma. 4 tissue sections per animal were washed 3
times for 15 min in TBS 7.2 and blocked for 90 min at 4° C in 5% normal goat serum (NGS) and
TBS with 0.2% Triton X-100 (TX). The anti-htt EM48 antibody (MAB5374, 1:250, EMD
Millipore, Bilerica, MA) was used to detect intranuclear Htt aggregate inclusions. Antibody
specification was validated by subjecting tissue slices to the same IHC protocol but without the
addition of primary or secondary antibodies. The EM48 primary antibody was diluted in TBS
with 2% NGS and 0.2% TX. Following incubation with the EM48 antibody overnight at 4° C,
the slices were washed 3 times in TBS for 15 min. From this point on the tissue was protected
from light. The tissue was subsequently incubated with Alexa Flour 594-conjugated goat anti-
mouse IgG (1:500, Invitrogen, Grand Island, NY) diluted in TBS with 2% NGS and 0.05% TX
for 90 minutes at 4°C. Sections were mounted on gelatin-subbed slides, and cover-slipped using
Vectashield Mounting Medium with DAPI (Vector Laboratories, Inc., Burlingame, CA).
Immunofluorescence intensity was captured at 60x magnification with an Olympus BX61
microscope (Shinjuku, Tokyo, Japan) equipped with a Disk Scanning Unit (spinning disk
confocal) and a 100 W mercury light source (U-LH100HG) using a Hamamatsu Photogenics
ORCA-R2 camera (Hamamatsu, Japan). Images were analyzed using Metamorph Advanced
7.7.20 (Molecular Device, LLC, Sunnyvale, CA). Images were captured from the dorsolateral or
44
ventrolateral striatum of each brain slice. To account for striatal bundles, cell bodies, and
unfocused areas, data was collected from three regions of interest placed randomly on areas of
fluorescing tissue.
High Performance Liquid Chromatography
Neurotransmitter concentrations were determined according to an adaptation of Irwin et
al. (1992) from the method of Kilpatrick et al. (1986). Tissues for analysis were homogenized in
0.4 N perchloric acid and centrifuged at 12,000 × g to separate precipitated protein. The protein
pellet was resuspended in 0.5 N NaOH and the total protein concentration determined using the
Coomassie Plus protein assay system (Pierce, Rockford, IL) using a Biotek Model Elx800
microplate reader (Biotek Instruments Wincoski, VT) and KCjunior software. The
concentrations of dopamine norepinephrine, serotonin, 3,4-dihydroxyphenylacetic (DOPAC),
and homovanillic acid (HVA) were assayed by HPLC with electrochemical detection. Samples
were injected with an ESA (Chelmsford, MA) autosampler. Dopamine, norepinephrine,
serotonin, and the dopamine metabolites were separated by a 150 × 3.2 mm reverse phase 3-μm-
diameter C-18 column (ESA) regulated at 28°C. The mobile phase MD-TM (ESA) consisted of
acetylnitrile in phosphate buffer and an ion-pairing agent delivered at a rate of 0.6 ml/min. The
electrochemical detector was an ESA model Coularray 5600A with a four-channel analytical cell
with three set potentials at −100, 50, and 220 mV. The HPLC was integrated with a Dell GX-280
computer with analytical programs including ESA Coularray for Windows software and the
statistics package InStat (GraphPad Software, San Diego, CA).
45
Statistical Methods
All data are reported as mean SEM. SSPS Statistics 21 (IBM, Armonk, NY) was used
to compare means of acquired data. For comparisons between two groups Student t-tests with
alpha levels of 0.05 were used. All other behavioral, immunochemistry and immunoblot data
were analyzed by 2-factor ANOVA. Where effects of genotype or exercise were identified,
Tukey’s honest significance test was employed to perform post-hoc analysis to determine
significance. Figures were made in GraphPad Prism 5 (Graphpad Software, Inc., San Diego,
CA).
RESULTS
All mice in the exercise groups were able to complete the treadmill running regimen.
Mice form all groups ran on the motorized treadmill starting at 28 days postnatal, 3 times per
week, for 6 months. Both the HD KI and WT exercise groups ran at equal speeds and duration
throughout the entire 6-month study. At the end of each week their body weight was determined.
There were no significant differences in body weight between the 4 groups at any time point
throughout the 6-month study. A two-way ANOVA of 5.5 month old sedentary and exercised KI
mice and their WT controls showed no statistically significant difference between the groups
(F(3,23) = 0.71, p = 0.56). There was no significant effect of genotype (F(1,23) = 0.59, p = 0.45)
or exercise (F(1, 23) = 1.93, p = 0.18) and no significant interaction between the variables
(F(1,23) = 0.06, p = 0.80). Average weights were wildtype sedentary 28.36 ± 0.45 g, HD KI
sedentary 26.95 ± 0.31 g, wildtype + Exercise 28.80 ± 0.28 g, and HD KI + Exercise 27.83 ±
0.20 g. Typically reports in the literature do not indicate weight loss in this strain of HD KI mice
until 12 months of age (Menalled et al., 2003)).
46
Voluntary Running Wheel
A subset of HD KI and WT mice distinct from the treadmill running groups were
subjected to voluntary running wheel to assess their capacity to run since others have shown a
decline in running in a number of HD genetic models (Hickey et al., 2008; Cepeda et al., 2010).
Figure 14 shows the average distance per hour of HD KI mice and WT controls in the voluntary
running wheel during a 6-hour period beginning at 2 months of age. HD KI mice exhibit a
significant decrease (p < 0.001) in distance run beginning at 4 months of age and continuing until
the termination of the study at 6 months (repeated measures ANOVA, genotype effect, F
(1,143)
=
14.87). There was no significant effect of age (F
(8,143)
= 0.85, p = 0.56) and no significant
interaction between the variables (F
(8,143)
= 0.98, p = 0.46). While HD KI mice showed a decline
in voluntary running wheel usage compared to WT littermates we did not observe any
differences in treadmill running between these 2 groups during the entire 6-month period.
Figure 14. The CAG
140
knock-in mouse model of Huntington’s disease (HD
KI) shows a deficit in the distance run per hour on a voluntary running
wheel as compared to wildtype (WT) controls.
Mice were allowed free access to a running wheel for 2 consecutive days for a 6-
hour period and on day 3, the distance ran in 6 hours was recorded.
Measurements were taken every 2 weeks beginning at 2 months of age. HD KI
mice (n = 8) ran significantly less (ANOVA; p < 0.001) than WT controls (n = 8)
beginning at an age of 3.5 months and continuing until the end of the study. Error
bars indicate SEM. Asterisks represent statistical significance: *** p < 0.001.
47
Rotarod Performance
Latency to fall from the accelerating rotarod was assessed at 4 and 6 months of age in
both the HD KI and WT no exercise mice to determine any deficiency in motor function,
balance, and coordination in the prodromal period (Figure 15A and 15B). At 4 months of age,
there was no statistically significant difference (t = 0.10; p = 0.92) in the performance (latency to
fall) of HD KI mice (154.4 ± 8.5 s; n = 10; mean ± SEM) compared to WT mice (152.7 ± 14.1 s;
n = 10). At 6 months of age, there was no statistically significant difference (t = 1.24; p = 0.24)
in latency to fall for HD KI (152.7 ± 9.3 s; n= 8) and WT mice (175.1 ± 11.9 s; n = 8).
Gait Analysis
A gait analysis task was performed at 4 and 6 months of age in both the HD KI and WT
no exercise mice to assess motor coordination based on stride length (Figure 15C and 15D) and
forepaw/hindpaw overlap (Figure 2E and 2F). At 4 months of age, there was no statistically
significant difference (t = 0.59; p = 0.57) in stride length between HD KI mice (6.46 ± 0.37 cm;
n = 9) and WT mice (6.69 ± 0.17 cm; n = 11). Also, at 6 months of age there was no statistically
significant difference (t = 0.96; p = 0.36) in stride length between HD KI mice (6.54 ± 0.32 cm;
n = 8) and WT mice (6.91 ± 0.22cm; n = 8). At 4 months of age, there was no statistically
significant difference in forepaw/hindpaw overlap (t = 0.93; p = 0.38) between HD KI mice
(0.83 ± 0.13 cm; n = 9) and WT mice (0.75 ± 0.04 cm; n = 11). Similarly at 6 months of age,
there was no statistically significant difference (t = 0.46; p = 0.65) in paw overlap between HD
KI (0.72 ± 0.03 cm; n = 8) and WT mice (0.75 ± 0.06 cm; n = 8).
48
Figure 15. HD KI mice display no motor function impairment at 4 months
and 6 months of age.
Motor function was assessed in the accelerating rotarod task (A-B) and gait
analysis (C-F). There was no significant difference between the performance of
HD KI (n = 8-10) and WT controls (n = 8-10) on an accelerating rotarod at 4
months (A) or 6 months (B) of age. Analysis of gait was performed by painting
the forepaws and hindpaws of the mice two different colors and allowing the mice
to walk on a white strip of paper towards a dark goal box. No significant
differences in stride length (C-D) or forepaw/hindpaw overlap (E-F) were
observed in HD KI (n = 8) and WT controls (n = 8). Error bars indicate SEM.
49
Analysis of Depression-like Behavior with the Forced Swim Test
Depression-like behavior was evaluated by determining the percent time a mouse
remained immobile during the forced swim test in all 4 groups of mice at 2, 4, and 6 months of
age (N = 8 to 10 mice per group) (Figure 16, upper panels). At 2 months of age, a two-way
ANOVA analysis of sedentary and exercised HD KI and WT mice showed no statistically
significant difference between at 4 groups (F
(3,32)
= 0.32, p = 0.81). There was no statistically
significant effect of genotype (F
(1,32)
= 0.93, p = 0.34) or exercise (F
(1, 32)
= 0.02, p = 0.89) and no
statistically significant interaction between the variables (F
(1,32)
= 0.01, p = 0.91). At 4 months of
age HD KI and WT mice showed a statistically significant difference between the groups (F
(3,33)
= 5.06, p < 0.01) as well as significant effects of genotype (F
(1,33)
= 4.81, p < 0.05) and exercise
(F
(1, 33)
= 6.64, p < 0.05) and a trend towards significant interaction between the variables (F
(1,33)
= 3.04, p = 0.09). Post hoc analysis revealed the following differences between specific
treatment groups. Sedentary HD KI mice had a statistically significant increase (p < 0.05) in the
percent immobility time compared to sedentary WT mice (HD KI, 52.96 ± 2.2%; n = 9;
compared to WT, 39.46 ± 3.69%; n = 8). Exercise led to a statistically significant decrease in
percent immobility (p < 0.05) between the HD KI mice (HD KI + Exercise, 40.66 ± 3.68%; n =
8) compared to HD KI sedentary mice. Exercise did not have a statistically significant effect (p
= 0.99) between WT mice (WT + Exercise, 38.06 ± 2.90%; n= 9). The was no statistically
significant change in percent immobility between KI + Exercise and either WT sedentary or
WT+ exercise mice (WT p = 0.99; WT + Exercise p = 0.94). At 6 months of age HD KI and WT
mice showed a statistically significant difference between the groups (F
(3,35)
= 9.70, p < 0.001) as
well as a statistically significant effect of genotype (F
(1,35)
= 19.48, p < 0.001) and exercise (F
(1,
35)
= 5.11, p < 0.05). There was a statistically significant interaction between the variables (F
(1,35)
50
= 6.77, p < 0.05). Post hoc analysis revealed the following differences between specific
treatment groups. Sedentary HD KI mice had a statistically significant increase (p < 0.01) in the
percent immobility time compared to sedentary WT mice (HD KI, 58.58 ± 3.00%; n = 8; WT,
41.64 ± 3.46%; n = 10). Exercise led to a statistically significant decrease in percent immobility
(p < 0.001) between the HD KI groups (KI + Ex, 34.14 ± 3.11%; n = 8). Exercise did not have a
statistically significant effect (p = 0.53) between WT sedentary and WT + exercise mice (WT +
Ex, 35.33 ± 3.85%; n= 10). There was no statistically significant change in percent immobility
between HD KI + Ex and either of the WT sedentary or WT + exercise mice (WT p = 0.44; WT
+ Exercise p = 1.00).
Analysis of Depression-like Behavior with the Tail Suspension Test
Depression-like behavior was also evaluated by determining the percent time a mouse
remained immobile during the tail suspension test at 2, 4, and 6 months of age (Figure 16, lower
panel). At 2 months of age a two-way ANOVA showed no statistically significant difference
between sedentary and exercised HD KI mice and WT mice (F
(3,32)
= 0.55, p = 0.65). There was
no statistically significant effect of genotype (F
(1,32)
= 1.38, p = 0.25) or exercise (F
(1, 32)
= 0.24, p
= 0.63) and no significant interaction between the variables (F
(1,32)
= 0.07, p = 0.80). At 4 months
of age HD KI and WT mice showed a statistically significant difference between the groups
(F
(3,31)
= 0.7.07, p < 0.005) as well as a significant effect of genotype (F
(1,31)
= 14.71, p < 0.001)
and significant interaction between the variables (F
(1,31)
= 6.24, p < 0.05), but not an exercise
effect (F
(1, 31)
=0.26 , p = 0.62). Post hoc analysis revealed the following differences between
specific treatment groups. Sedentary HD KI mice had a statistically significant increase (p <
0.05) in the percent immobility time compared to sedentary WT mice (HD KI, 58.60 ± 1.72%; n
51
= 9; WT, 49.45 ± 3.13%; n = 8). Exercise led to a statistically significant decrease in percent
immobility (p < 0.001) between the HD KI sedentary and HD KI + exercise mice (HD KI +
Exercise, 39.30 ± 4.61%; n = 8). Exercise did not have a statistically significant effect (p = 0.78)
between WT sedentary and WT + exercise mice (WT + Exercise, 45.37 ± 1.78%; n= 8). There
was no statistically significant difference in percent immobility between HD KI + Exercise and
either of the WT sedentary or WT + exercise mice (WT p = 0. 11; WT + Exercise p = 0.50). At 6
months of age HD KI and WT mice showed a statistically significant difference between the
groups (F
(3,32)
= 5.88, p < 0.005) as well as a statistically significant effect of exercise (F
(1,32)
=
9.28, p < 0.005) and a statistically significant interaction between the variables (F
(1,32)
= 6.01, p <
0.05), but not a genotype effect (F
(1, 32)
=1.48 , p = 0.23). Post hoc analysis revealed the
following differences between groups. Sedentary HD KI mice had a statistically significant
increase (p < 0.005) in the percent immobility time compared to sedentary WT mice (HD KI,
61.48 ± 3.24%; n = 9; WT, 45.59 ± 2.45%; n = 8). Exercise led to a statistically significant
decrease in percent immobility (p < 0.05) between the KI groups (KI + Exercise, 50.87 ± 2.57%;
n = 8). Exercise did not have a statistically significant effect (p = 0.83) between WT sedentary
and WT + exercise mice (WT + Exercise, 49.16 ± 3.08%; n= 8). There was no statistically
significant difference in percent immobility between HD KI + Exercise and either the WT
sedentary or WT + Exercise mice (WT p = 0.59; WT + Exercise, p = 0.98).
Taken together, these data demonstrate the onset of depression-like behavior between 2
and 4 months of age as assessed in both the forced swim and tail suspension tests. In both tests,
exercise ameliorated this depression-like behavioral impairment, effectively delaying its onset to
at least the 6 months of age time point.
52
Figure 16. High intensity exercise delays the onset of depression-like
behavior observed in HD KI mice.
(A-C) Mice were placed in a beaker of water for 6 minutes and the total time they
were immobile was recorded as a percent of the total time. At 2 months (A) no
significant effect of genotype or exercise was observed. At both 4 months (B)
and 6 months (C), sedentary HD KI animals (n = 8-9) exhibited significantly
increased immobility time compared to sedentary WT controls (n = 8). This
depression-like behavior was rescued by long-term treadmill running (HD KI +
Exercise; n = 8) at both time points. (D-F) Mice were suspended by their tails for
6 minutes and the total time they were immobile was recorded as a percent of the
total time. At 2 months (D) no significant effect of genotype or exercise was
observed. At both 4 months (E) and 6 months (F), sedentary HD KI animals (n =
8-9) exhibited significantly increased immobility time compared to sedentary WT
controls (n = 8). This depression-like behavior was rescued by long term
treadmill running (HD KI + Exercise; n = 8) at both 4 months (E) and 6 months
(F). In both tasks, long-term treadmill running delayed the onset of depression-
like behavior up to 6 months of age. Error bars indicate SEM. Asterisks represent
statistical significance: * p < 0.05, ** p < 0.01, *** p < 0.001
53
Immunohistochemical Staining for mHTT aggregates in the Dorsal Striatum
Immunohistochemical staining of the mutant huntingtin (mHTT) protein was used to
analyze exercise effects on intranuclear aggregation in both the dorsal and ventral striatum in
mice from all 4 groups at 2, 4, and 6 months of age. We observed both quantitative and
qualitative changes between HD KI and WT mice with aging and between HD KI sedentary and
HD KI + exercise (Figures 17 and 18).
At 2 months of age two-way ANOVA comparing the percentage of dorsal striatal
neurons containing intranuclear inclusions in HD KI and WT mice revealed a statistically
significant difference between the groups (F
(3,59)
= 51.96, p < 0.001) as well as significant effects
of genotype (F
(1,59)
= 116.27, p < 0.001) and exercise (F
(1, 59)
= 19.22, p < 0.001) and a
statistically significant interaction between the variables (F
(1,59)
= 20.40, p < 0.001). Post hoc
analysis revealed the following differences between specific treatment groups. Sedentary KI
animals had a statistically significant increase (p < 0.001) in the percentage of neurons
containing intranuclear inclusions compared to sedentary WT mice (KI, 46.91 ± 3.64%; n = 15;
WT, 2.13 ± 0.73%; n = 15; mean% ± SEM). Exercise led to a statistically significant decrease in
intranuclear inclusions (p < 0.001) between the HD KI groups (HD KI + Exercise, 20.86 ±
4.36%; n = 15). Exercise did not have a statistically significant effect (p = 1.00) between WT
and WT + Exercise mice (WT + Exercise, 2.52 ± 1.22%; n= 15). HD KI + Exercise mice had a
statistically significant increase in the number of cells showing intranuclear inclusions compared
to both WT and WT + Exercise mice (WT p < 0.001; WT + Exercise p < 0.001). At 4 months of
age comparing the percentage of dorsal striatal neurons containing intranuclear inclusions in HD
KI and WT mice demonstrated a statistically significant difference between groups (F
(3,59)
=
209.79, p < 0.001), a statistically significant effect of genotype (F
(1,59)
= 580.77, p < 0.001) and
54
exercise (F
(1,59)
= 26.46, p < 0.001), and a statistically significant interaction between the
variables (F
(1,59)
= 22.12, p < 0.001). Post hoc analysis revealed the following differences
between specific treatment groups. Sedentary HD KI mice had a statistically significant increase
(p < 0.001) in the percentage of neurons containing intranuclear inclusions compared to
sedentary WT mice (HD KI, 62.99 ± 2.15%; n = 15; WT, 4.94 ± 1.64%; n = 15). Exercise led to
a statistically significant decrease in intranuclear inclusions (p < 0.001) in HD KI mice (HD KI +
Exercise, 43.14 ± 2.66%; n = 15). Exercise did not have a statistically significant effect (p =
0.99) between WT and WT + Exercise mice (WT + Ex, 4.05 ± 1.36%; n= 15). HD KI +
Exercise mice had a statistically significant increase in intranuclear inclusions compared to both
of the WT groups (WT p < 0.001; WT + Ex p < 0.001). At 6 months of age the percentage of
dorsal striatal neurons containing mHtt intranuclear inclusions in HD KI and WT mice showed a
statistically significant difference between groups (F
(3,59)
= 191.50, p < 0.001), a statistically
significant effect of genotype (F
(1,59)
= 469.26, p < 0.001) and exercise (F
(1, 59)
= 58.72, p <
0.001), and a statistically significant interaction between the variables (F
(1,59)
= 46.51, p < 0.001).
Post hoc analysis revealed the following differences between specific treatment groups.
Sedentary HD KI mice had a statistically significant increase (p < 0.001) in the percentage of
neurons containing intranuclear mHtt inclusions compared to sedentary WT mice (HD KI, 72.28
± 2.50%; n = 15; WT, 10.58 ± 1.19%; n = 15). Exercise led to a statistically significant decrease
in intranuclear inclusions (p < 0.001) between the HD KI groups (HD KI + Exercise, 40.91 ±
3.09%; n = 15). Exercise did not have a statistically significant effect (p = 0.93) between WT
groups (WT + Exercise, 8.75 ± 1.26%; n= 15).
Similar exercise-induced changes in mHtt intranuclear aggregates as observed in the
dorsal striatum were also seen in the ventral striatum (Figure 18). At 2 months of age HD KI and
55
WT mice revealed a statistically significant difference between the groups (F
(3,59)
= 60.31, p <
0.001) as well as significant effects of genotype (F
(1,59)
= 151.04, p < 0.001) and exercise (F
(1, 59)
= 12.00, p < 0.001) as well as a statistically significant interaction between the variables (F
(1,59)
=
17.89, p < 0.001). Post hoc analysis revealed the following differences between specific
treatment groups. Sedentary HD KI animals had a statistically significant increase (p < 0.001) in
the percentage of neurons containing intranuclear inclusions compared to sedentary WT mice
(HD KI, 52.47 ± 3.39%; n = 15; WT, 2.81 ± 1.11%). Exercise led to a statistically significant
decrease in intranuclear inclusions (p < 0.001) between the HD KI groups (KI + Exercise, 29.34
± 4.69%; n = 15). Exercise did not have a statistically significant effect (p = 0.95) between WT
sedentary and exercise mice (WT + Ex, 5.11 ± 1.19%; n= 15). At 4 months of age HD KI and
WT mice revealed a statistically significant difference between groups (F
(3,59)
= 109.08, p <
0.001) as well as significant effects of genotype (F
(1,59)
= 262.33, p < 0.001) and exercise (F
(1, 59)
= 27.52, p < 0.001) and a statistically significant interaction between the variables (F
(1,59)
=
37.40, p < 0.001). Post hoc analysis revealed that sedentary HD KI mice had a statistically
significant increase (p < 0.001) in the percentage of neurons containing intranuclear inclusions
compared to sedentary WT mice (HD KI, 62.49 ± 2.84%; n = 15; WT, 3.25 ± 0.99%; n = 15).
Exercise led to a statistically significant decrease in intranuclear inclusions (p < 0.001) between
the HD KI groups (HD KI + Exercise, 32.32 ± 4.09%; n = 15). Exercise did not have a
statistically significant effect (p = 0.93) between WT groups (WT + Exercise, 5.55 ± 1.57%; n=
15). At 6 months, showed a statistically significant difference between the groups (F
(3,59)
=
77.31, p < 0.001) as well as significant effects of genotype (F
(1,59)
= 183.90, p < 0.001) and
exercise (F
(1, 59)
= 21.49, p < 0.001) as well as a statistically significant interaction between the
variables (F
(1,59)
= 26.53, p < 0.001). Post hoc analysis revealed that sedentary HD KI mice had
56
a statistically significant increase (p < 0.001) in the percentage of neurons containing
intranuclear inclusions compared to sedentary WT mice (HD KI, 63.58 ± 3.69%; n = 15; WT,
5.01 ± 1.16%; n = 15). Exercise led to a statistically significant decrease in intranuclear
inclusions (p < 0.001) between the HD KI sedentary and HD KI exercise mice (HD KI +
Exercise, 32.95 ± 4.73%; n = 15). Exercise did not have a statistically significant effect (p =
0.98) between WT groups (WT + Exercise, 6.63 ± 1.35%; n= 15).
In addition to the quantitative changes showing that exercise reduced the number of
striatal neurons with mHtt aggregates in HD KI + Exercise mice, there were also qualitative
changes in the appearance and distribution of mHTT protein immunoreactivity within nuclei. In
both the dorsal and ventral striatum, at 2 months of age mHtt immunoreactivity is evident as a
diffusely distributed material throughout the nucleus. At 4 months of age the mHtt
immunoreactivity increases throughout the nucleus with an increase in puncta appearance which
appear as small aggregates. At 6 months of age the mHtt immunoreactivity predominates in large
intracellular aggregates, typically 1 per cell nuclei with a diameter of approximately 2 to 3
microns. In the HD KI + Exercise mice there is a statistically significant reduction of the
accumulation of mHtt immunoreactivity compared to HD KI sedentary mice at 2, 4, and 6
months by 60, 69, and 47% respectively (p < 0.01 for all time points compared). Interestingly, at
6 months the HD KI + Exercise mice a reduction by 31% (P < 0.05) in cells with large
intracellular aggregates are seen compared to sedentary HD KI mice at this same time point.
57
Figure 17. Exercise decreased the number and intensity of intranuclear htt
aggregates in dorsostriatal MSNs.
(A-C) Immunohistochemical analysis of the mutated htt protein was evaluated in
dorsostriatal MSNs of sedentary and exercised HD KI and WT mice at 2, 4, and 6
months of age. Images of coronal sections were taken at 60x magnification. Each
panel displays an overlapped image of MSNs (DAPI) and htt aggregate staining
(RFP) across the different HD KI and WT groups. (D-F) The total number of
intranuclear aggregates in a captured image were counted and recorded as a
percentage of the total cells. A significant increase in the number of intranuclear
aggregates was observed in sedentary HD KI animals (n = 15) at 2, 4, and 6
months compared to their sedentary WT counterparts (n = 15). There was a
significant reduction in the amount of intranuclear aggregates in HD KI animals
that underwent treadmill running (HD KI + Ex; n = 15). Error bars indicate SEM.
Scale bar: 10 µm. Asterisks represent statistical significance: *** p < 0.001
58
Figure 18. Exercise decreased the number and intensity of intranuclear htt
aggregates in ventrostriatal MSNs.
(A-C) Immunohistochemical analysis of the mutated htt protein was evaluated in
ventrostriatal MSNs of sedentary and exercised HD KI and WT mice at 2, 4, and
6 months of age. Images of coronal sections were taken at 60x magnification.
Each panel displays an overlapped image of MSNs (DAPI) and htt aggregate
staining (RFP) across the different HD KI and WT groups. (D-F) The total
number of intranuclear aggregates in a captured image were counted and recorded
as a percentage of the total cells. A significant increase in the number of
intranuclear aggregates was observed in sedentary HD KI animals (n = 15) at 2, 4,
and 6 months compared to their sedentary WT counterparts (n = 15). There was a
significant reduction in the amount of intranuclear aggregates in HD KI animals
that underwent treadmill running (HD KI + Ex; n = 15). Error bars indicate SEM.
Scale bar: 10 µm. Asterisks represent statistical significance: *** p < 0.001
59
HPLC Analysis of Striatal Dopamine
High performance liquid chromatography was used to examine exercise effects on the
concentrations of dopamine, norepinephrine, and serotonin in the dorsal striatum in all 4
experimental groups at 6 months of age. We observed changes in all three of these
neurotransmitters in sedentary HD KI animals as compared to sedentary WT controls as well as
between sedentary and exercised HD KI animals (Table 3).
Specifically, a 2 way ANOVA comparing the concentration of dopamine revealed a
statistically significant difference between the experimental groups (F
(3,24)
= 5.64, p < 0.01).
There was no significant effect of genotype (F
(1,24)
= 1.77, p = 0.20) or exercise (F
(1,24)
= 1.17, p
= 0.30), but there was a significant interaction between the two variables (F
(1,24)
= 14.05, p <
0.01). Post hoc tests revealed that sedentary HD KI animals had a significant decrease (p < 0.01)
in striatal dopamine concentration as compared to sedentary WT controls (HD KI, 263.2 ± 17.6
ng/mg; n = 5; WT, 433.9 ± 26.7 ng/mg; n = 5). Exercise led to a significant increase (p < 0.05)
in dopamine concentration between HD KI groups (HD + Exercise 425.5 ± 38.7 ng/mg; n = 5).
Exercise did not have a statistically significant effect (p = 0.21) between WT groups (WT +
Exercise 344.3 ± 41.8 ng/mg; n = 5).
A 2 way ANOVA comparing the concentration of norepinephrine reveals a statistically
significant difference between the experimental groups (F
(3,24)
= 4.16, p < 0.05). There was no
significant effect of genotype (F
(1,24)
= 0.70, p = 0.41) or exercise (F
(1,24)
= 1.38, p = 0.26), but
there was a significant interaction between the two variables (F
(1,24)
= 10.39, p < 0.01). Post hoc
tests revealed that sedentary HD KI animals had a significant decrease (p < 0.05) in striatal
norepinephrine concentration as compared to sedentary WT controls (HD KI, 8.7 ± 1.3 ng/mg; n
= 5; WT, 13.7 ± 1.1 ng/mg; n = 5). Exercise resulted in a significant increase (p < 0.05) in
norepinephrine concentration between HD KI groups (HD + Exercise 14.1 ± 1.7 ng/mg; n = 5).
60
Exercise did not have a statistically significant effect (p = 0.49) between WT groups (WT +
Exercise 11.2 ± 0.9 ng/mg; n = 5).
A 2 way ANOVA comparing the concentration of serotonin in the dorsal striatum
revealed a statistically significant difference between the experimental groups (F
(3,24)
= 6.09, p <
0.01). There was no significant effect of genotype (F
(1,24)
= 2.54, p = 0.13) or exercise (F
(1,24)
=
0.30, p = 0.59), but there was a significant interaction between the two variables (F
(1,24)
= 15.37,
p < 0.001). Post hoc tests revealed that sedentary HD KI animals had a significant decrease (p <
0.01) in striatal serotonin concentration as compared to sedentary WT controls (HD KI, 12.4 ±
1.1 ng/mg; n = 5; WT, 23.6 ± 3.1 ng/mg; n = 5). Exercise led to a significant increase (p < 0.05)
in serotonin concentration between HD KI groups (HD + Exercise 21.5 ± 0.9 ng/mg; n = 5).
Exercise did not have a statistically significant effect (p = 0.14) between WT groups (WT +
Exercise 16.7 ± 1.8 ng/mg; n = 5).
Additionally, while there was a trends towards significant difference in dopamine
turnover between the groups (F
(3,24)
= 3.05, p = 0.053) and no significant effect of genotypes
(F
(1,24)
= 0.07, p = 0.79), there was a statistically significant effect of exercise on DA turnover
(F
(1,24)
= 8.97, p < 0.01).
61
HPLC Analysis of Dopamine, Norepinephrine, Serotonin and DA turnover
Dopamine Norepinephrine Serotonin DA Turnover
No Exercise WT 433.9 ± 26.7 13.7 ± 1.1 23.6 ± 3.1 0.32 ± 0.03
HD KI 263.2 ± 17.6** 8.7 ± 1.3* 12.4 ± 1.1** 0.32 ± 0.03
Exercise WT 344.3 ± 41.8 11.2 ± 0.9 16.7 ± 1.8 0.25 ± 0.01
HD KI 425.5 ± 38.7
#
14.1 ± 1.7
#
21.5 ± 0.9
#
0.24 ± 0.01
Table 3: HPLC Analysis of Dopamine, Norepinephrine, Serotonin and DA
turnover.
The concentration of dopamine, norepinephrine, serotonin and dopamine turnover
ratio in the striatum were analyzed in sedentary and exercise HD KI and WT mice
(n = 5 per group) at 6 months of age. The turnover ratio is defined as [(DOPAC +
HVA)/dopamine]. Sedentary HD KI mice had a significant decrease in the
concentrations of dopamine (p < 0.01), norepinephrine (p < 0.05), and serotonin
(p < 0.05). Long-term treadmill exercise led to a significant increase in the levels
of all three (p < 0.05 for each). Additionally, while there was no effect of
genotype on dopamine turnover (p = 0.79), there was an effect of exercise
observed (p < 0.01). Data is reported as Mean ± SEM ng/mg. Asterisks represent
statistical significance between sedentary WT and HD KI groups: * p < 0.05, ** p
< 0.01. # indicates statistical significance between sedentary and exercised HD KI
groups: # p < 0.05.
62
DISCUSSION
In this study, we found that long-term treadmill running rescues depression-like behavior
and attenuates pathological abnormalities observed in the CAG
140
mouse in the prodromal
period, prior to motor symptom onset. Specifically, depression-like behavior in two separate
tasks (forced swim test and tail suspension test) was observed in HD KI sedentary mice
beginning at an age of 4 months and worsening at 6 months as compared to their WT
counterparts (Figure 16). While mood disturbances have been reported in other HD animal
models (Pang et al., 2009; Renoir et al., 2012), to the best of our knowledge, this is the first
report of mood disturbances observed in the CAG
140
mouse. Long term treadmill exercise
rescued the mood disturbances observed in HD KI sedentary mice, effectively delaying the onset
of depression-like behavior up to 6 months of age. These findings are consistent with previous
findings that demonstrate beneficial effects of voluntary wheel running on depression-like
behavior in the R6/1 HD mouse model (Renoir et al., 2012).
While some studies (Renoir et al., 2012) have shown beneficial effects of exercise, a
small number of studies in the R6/1 and R6/2 models have shown little to no benefit of exercise
on motor and cognitive symptoms (Pang et al., 2006; van Dellen et al., 2008; Wood et al., 2011;
Harrison et al., 2013). In all of these studies, mice were exposed to voluntary wheel running.
Recent studies have challenged the utilization of voluntary running wheels as an intervention in
R/1 mice, where wild type mice ran much further than HD mice after 18 weeks of age (Cepeda et
al. 2010). This study replicates those results in the CAG
140
mouse model (Figure 14). A cohort
of sedentary HD KI and WT animals were probed once every two weeks from age 2 months to 6
months for performance on a voluntary running wheel. HD KI animals exhibited a significant
decrease in the distance run on the voluntary running wheel beginning at 4 months and
63
continuing until the termination of the study at 6 months of age. At 4 months, this reduction in
voluntary wheel running observed in the CAG
140
mouse occurs well before motor symptom
onset. Thus, it is possible that the lack of observed benefits of exercise observed in the R6
mouse models could be due to reduced motivation for wheel running. Our study therefore
attempts to avoid this confounder through the use of motorized treadmill running.
Elucidating the onset of disease pathology could point to a crucial time period by which
HD patients must begin long-term exercise in order for it to have a beneficial effect. For
example, our study demonstrates that early therapeutic interventions must be initiated during the
prodromal period to successfully attenuate impairments brought on by HD disease progression.
Previous studies from other groups that reported neutral or detrimental effects of exercise began
their intervention when their mice were several months old (Pang et al., 2006; Potter et al.,
2010). Thus it is possible that there is a critical time period for the initiation of exercise in order
for it to have a beneficial effect on HD disease onset and progression.
To this end, we investigated the effect of long-term treadmill running on the formation of
intranuclear inclusions in striatal MSNs. Intranuclear inclusions of htt aggregates were first
observed in the dorsal and ventral striatum at 2 months of age and increased in number of
neurons affected and overall intensity over time (Figure 17 and 18). Intensive treadmill exercise
significantly decreased the total number of cells with intranuclear inclusions, as well as the
overall intensity of htt aggregate staining within affected neuronal nuclei, at 2, 4, and 6 months
in HD KI mice compared to their sedentary counterparts. Additionally, in both the dorsal and
ventral striatum, the percent of neurons with inclusions observed in 6-month old HD KI
exercised animals was less than the percent inclusions measured in 2 month old HD KI sedentary
64
animals. This suggests a delay in htt aggregate formation in HD KI exercised mice up to 6
months of age.
These beneficial effects observed in the CAG
140
mice conflict with the effects of
voluntary wheel running observed in R6/1 mice (Harrison et al., 2013) where exercised animals
were shown to have a greater number and larger inclusions than R6/1 control mice. Importantly,
this study was performed on mice that already had substantial neuronal loss and motor symptom
onset when exercise was initiated. CAG
140
animals in the study presented here began motorize
treadmill exercise well before htt aggregation caused intranuclear inclusions. We therefore are
observing the effects of long-term exercise on delaying the onset of these aggregation events.
Importantly, while this study demonstrates the amelioration of behavioral impairment and
morphological abnormalities, the underlying mechanisms regulating the improvements brought
on by long-term treadmill exercise remains to be determined. One possible mechanism could lie
in changes in the amount of available neurotransmitters in the synaptic cleft. Dopamine,
norepinephrine, and serotonin levels in the striatum have long been associated with playing a
role in depression (Pare, 1972). We present evidence here that dopamine, norepinephrine, and
serotonin are all reduced in the striatum in sedentary HD KI animals (Table 3). Long-term
treadmill running significantly increased the availability of all three of these neurotransmitters.
These results are consistent with previous studies (Renoir et al., 2012) which have shown
beneficial effects of selective serotonin reuptake inhibitors (SSRIs) on improving depression-like
behavior in the R6/1 HD mouse model.
65
CONCLUSIONS
Collectively, our findings suggest that long-term treadmill running, beginning well before
onset of motor symptoms, results in a delay of depression-like behavior and striatal pathology
observed during the prodromal period in CAG
140
mice. Additionally, long-term exercise restored
the decreased striatal levels of dopamine, norepinephrine, and serotonin observed in sedentary
HD KI animals, indicating a potential mechanism for symptomatic improvement observed in
these animals. Determining the mechanisms by which long-term treadmill running has its effect
in delaying symptom onset and htt aggregate formation may provide potential pharmacological
and genetic interventions targeting neuroplasticity in the basal ganglia resulting in delaying the
onset and progression of HD as well as to improving the quality of life of HD patients.
66
CHAPTER 3: Treadmill running delays the onset of cognitive dysfunction and restores
dopamine neurotransmission in the CAG140 mouse model of Huntington’s disease
Author List: DP Stefanko, VD Shah, AL Tran, Z Gasanova, GM Petzinger, and MW Jakowec
ABSTRACT
Cognitive impairments are common during the prodromal phase of Huntington’s disease,
and can occur up to a decade before the onset of classical motor symptoms of this degenerative
disorder. The purpose of this study was to examine the potential impact of exercise in the form of
motorized treadmill running on cognition in the CAG
140
knock-in mouse model of Huntington’s
disease. This model has a long lifespan compared to other rodent models of this disease
including the late stage onset of motor features after 12 months of age, allowing for investigation
of prodromal behavior disturbances. Additionally, this also provides the opportunity to
investigate the effects of interventions such as exercise initiated early in life. Motorized treadmill
running was initiated at 4 weeks of age (1 hour per session, 3 times per week) and continued for
6 months. Sedentary HD KI animals displayed a significant impairment in the t-maze reversal
learning task, suggesting an impairment in cognitive flexibility. This impairment could be
explained in part by a downregulation of the dopamine D2 receptor observed in the dorsal and
ventral striatum, resulting in aberrant dopamine neurotransmission in the indirect pathway.
Exercise restored dopamine D2 levels and rescued the cognitive disturbances observed in
reversal learning. Together these findings begin to address the potential impact of life-style and
early interventions such as exercise on modifying disease progression.
KEYWORDS: huntingtin, exercise, dopamine, tmaze, Q140, behavior
67
INTRODUCTION
Huntington’s disease (HD) is a devastating neurodegenerative disorder caused by a
polyglutamine expansion in the huntingtin (htt) protein (The Huntington's Disease Collaborative
Research Group, 1993). In patients, HD is characterized by a progressive loss of motor function,
impairments in cognitive ability, and psychiatric disturbances, resulting in premature death 15 to
20 years after diagnosis. Affected individuals typically begin to display motor function
impairments (chorea, tremors, dyskinesia, etc) in the 4
th
or 5
th
decade of life. Recent studies
have demonstrated that cognitive impairments and neuropsychiatric symptoms can begin to
emerge up to a decade prior to the onset of motor symptoms (Julien et al., 2007; Tabrizi et al.,
2013; Epping et al., 2013). Cognitive inflexibility presents early in the disease progression, as
evidenced by impaired performance on the Wisconsin Card Sorting Test and attentional set-
shifting tasks (Owen et al., 1993; Lawrence et. al., 1996; Ho et al., 2003).
Many studies have associated impairments in cognitive flexibility with aberrant
dopamine (DA) neurotransmission. Specifically, patients with DA impairments in the basal
ganglia display difficulties with set-shifting tasks (Cools et al., 2006; Dang et al., 2012) and rats
with lesions of the medial striatum have impairments in spatial and reversal learning tasks
(Whishaw et al., 1987; Pisa and Cyr, 1990). Dopaminergic input into the striatum terminates on
two distinct types medium spiny neurons (MSN), resulting in two different pathways: the direct
pathway with D1 receptor containing MSNs and the indirect pathway with D2 receptor
containing MSNs (Albin et al., 1989; Mink, 1996; Surmeier et al., 2007). Disruptions of DA
signaling in the direct and indirect pathways impair set-shifting and reversal learning in different
ways (Yawata et al., 2012). Specifically, in the t-maze reversal learning task, animals with a
reversible blockade of the D1 MSNs in the striatum displayed impairments in acquiring reward
68
based behaviors. A reversible blockade of D2 MSNs, however, resulted in impairments in
cognitive flexibility (Yawata et al., 2012).
Aberrant DA neurotransmission in the striatum has been implicated in playing a
significant role in the motor and cognitive impairments observed in HD. In HD patients,
presymptomatic HD carriers have been shown to have reduced D2 receptor binding (van
Oostrom et al., 2005; van Oostrom et al., 2009). Symptomatic HD patients exhibit a decrease in
both D1 and D2 receptor binding that progressively worsens over time (Ginovart et al., 1997;
Pavese et al., 2010). Positron emission topography (PET) imaging has demonstrated a reduction
in dopamine receptors in the striatum in both presymptomatic and symptomatic HD patients
(Richfield et. al, 1991). Loss of DA receptors in the striatum has been shown to strongly
correlate with decline executive function, learning, and memory in HD (Backman 1997).
Exercise has been shown to have a beneficial effect on disease progression and symptoms
in several neurodegenerative disorders, including Alzheimer’s disease (Verghese et al., 2003;
Abbott et al., 2004; Yu et al., 2011; Venturelli et al., 2011) and Parkinson’s disease (Frazzitta et
al., 2013; Corcos et al., 2013; Petzinger et al., 2010, 2013; Speelman et al., 2011). Specifically,
our lab has demonstrated that motorized treadmill running increases neuroplasticity and
attenuates pathophysiological mechanisms in a mouse model of Parkinson’s disease by
enhancing dopamine neurotransmission (Petzinger et al., 2007) and glutamate synaptic
neurotransmission (Kintz et al., 2013), and reversing dendritic spine loss (Toy et al., 2014),
culminating in an overall rescue of behavioral phenotype. A small number of studies in mouse
models of HD have shown little to modest improvements in motor and cognitive function after
exercise (Pang et al., 2006; van Dellen et al., 2008; Wood et al., 2011; Harrison et al., 2013).
These studies have all used voluntary wheel running, which may be ineffectual in long-term
69
studies as studies in different HD mouse models have shown that HD animals experience
reduced running wheel performance over time (Hickey et al., 2008; Cepeda et al., 2010). It is
therefore possible that the lack of observed benefits of exercise in these studies could be
confounded by reduced motivation of voluntary wheel running.
This study was designed to observe the effect of long-term motorized treadmill running
in the CAG
140
HD mouse model. To date, all studies of the CAG
140
mice have focused on the
motor impairments observed in the mouse model around the age of 1 year (Menalled et al., 2003;
Dorner et al., 2007; Hickey et al., 2008; Rising et al., 2011). No study to date has probed the
CAG
140
mice in the prodromal period for potential cognitive impairments. Additionally, no
study to date has attempted to use motorized treadmill running, as opposed to voluntary wheel
running, in CAG
140
animals to examine the potential beneficial effects of long-term exercise on
the behavioral and neuropathological symptoms of HD. To this end, this study uses long-term
motorized treadmill running in the CAG
140
mouse model to observe its role in cognitive behavior
in the novel object recognition and t-maze reversal learning tasks. Additionally, this study
investigates the underlying mechanism of any beneficial effect of exercise by measuring striatal
levels of tyrosine hydroxylase and dopamine D1 and D2 receptors in exercised and sedentary
animals. The results reported here provide evidence of the benefits of motorized treadmill
exercise to delay HD symptom onset and point to potential pharmacological targets for future
studies in the CAG
140
model.
70
MATERIALS AND METHODS
Animals and Groups
For these studies we used knock-in mice that contained a chimeric mouse/human exon 1
with 140 CAG repeats inserted into the mouse gene by homologous targeting (Menalled et al.,
2003). CAG
140
KI mice were produced in-house using lines descended from heterozygous
pairing. Wild-type littermates were used for controls. The founder mice for our colony were a
generous gift of Drs. Michael Levine and Carlos Cepeda (UCLA) with permission from Dr. Scott
Zeitlin (University of Virginia) through a Material Transfer Agreement. These mice were
backcrossed onto the C57BL/6J background annually to maintain vigor. Mouse genotypes from
tail biopsies were determined using real time PCR (Transnetyx, Inc., Cordova, TN). Mice were
randomly assigned to one of four groups based on WT or homozygous for CAG KI genotype
including: (i) wildtype, (ii) wildtype + exercise, (iii) CAG KI, and (iv) CAG KI + exercise. Only
male mice were utilized in these studies. Mice were group housed with a reverse light cycle
(lights off from 7 a.m. to 7 p.m.) and were allowed access to food and water ad libitum.
Experimental procedures were approved by the University of Southern California’s Institutional
Animal Care and Use Committee (IACUC) and were conducted in accordance with the National
Research Council Guide for the Care and Use of Laboratory Animals (DHEW Publication 80-23,
2011, Office of Laboratory Animal Welfare, DRR/NIH, Bethesda, MD). All efforts were made
to minimize animal suffering and to reduce the number of animals used to achieve statistical
significance.
71
Exercise Regimen
Mice were weaned at postnatal day 28 and subjected to treadmill running on a Model
EXER-6M Treadmill (Columbus Instruments, Columbus, Ohio). The treadmill exercise protocol
was conducted based on our previously publications with modifications (Fisher et al., 2004).
Briefly, exercise was initiated on postnatal day 28. In week one, mice in the exercise groups ran
at a speed of 8.0 ± 0.5 m/min for 40 minutes and they were closely monitored for any adverse
reaction to the treadmill, inability to run, or failure to learn the task. No mice needed to be
excluded. Exercise mice started at a velocity of 10.0 ± 1.5 m/min, and ran 3 times per week for
the 6-month duration of this study. Treadmill speed was gradually increased to 20 ± 1.5 m/min
by the final month. A non-noxious stimulus (metal beaded curtain) was used as a tactile incentive
to prevent animals from drifting back on the treadmill. All mice were weighed at the end of each
week and closely assessed for adverse reactions including stress. In the past we have found
treadmill running to not be stressful based on the evaluation of anxiety, depression, and
corticosterone levels (Gorton et al., 2010).
Behavioral Testing
Behavioral testing was carried out to evaluate cognitive ability and the effects of forced
treadmill running. For cognitive behaviors, novel object recognition and the t-maze reversal
learning task were employed at 4 and 6 months of age in all four groups of mice.
Novel Objection Recognition
The object recognition task consisted of a training phase and a testing phase. Before
training, all mice were handled 1–2 min a day for 5 d and were habituated to the experimental
72
apparatus 15 min a day for 3 consecutive days in the absence of objects. The experimental
apparatus was a white rectangular open field (30 x 23 x 21.5 cm). During the training phase,
mice were placed in the experimental apparatus with two identical objects (150 ml beakers, 1
inch circumference x 1.5 inch height; large blue Lego blocks, 1 x 1 x 2 inches) and were allowed
to explore for either 3 or 10 min. The objects were thoroughly cleaned between trials to make
sure no olfactory cues were present. Retention was tested at 90 min for short-term memory and
24 h for long-term memory. During these retention tests, mice explored the experimental
apparatus for 5 min in the presence of 1 familiar and 1 novel object. The location of the object
was counterbalanced so that one-half of the animals in each group saw the novel object on the
left side of the apparatus, and the other half saw the novel object on the right side of the
apparatus.
All training and testing trials were videotaped and analyzed by individuals blind to the
treatment condition and the genotype of subjects to determine the amount of time the mouse
spent exploring the novel and familiar objects. A mouse was scored as exploring an object when
its head was oriented toward the object within a distance of 1 cm or when the nose was touching
the object. The relative exploration time was recorded and expressed by a discrimination index
[D.I. = (t
novel
- t
familiar
)/( t
novel
+ t
familiar
) x 100%]. Mean exploration times were calculated and the
discrimination indexes between treatment groups were compared.
T-Maze Reversal Learning
The t-maze reversal learning task was carried out based on the methods described in
Yawata et al. (2012). Briefly, a four-arm cross maze was made of a clear plastic wall and a white
floor and placed 90 cm above the floor. Each arm was 25 cm long and 5 cm wide, and the center
73
platform was 5 × 5 cm. Visual cues cut out of construction paper such as stars, triangles, and
puzzle pieces were hung outside the maze. Two of the four sides of the testing room had white
curtains while the other two had black curtains for further visual cues. The position of a mouse
was detected by video camera suspended over the maze. One week prior to testing, animals were
food-restricted to reach approximately 85% of their original baseline weight. Each animal went
through 3 days of habituation. On each day of habituation, three non-flavored sugar pellets were
placed in the food well of each arm. A mouse was allowed to freely navigate and consume the
pellets within 15 minutes. If all pellets were consumed before the end of the 15 minutes, the
mouse we removed, new pellets were placed in the well, and the mouse was placed back into the
apparatus to continue habituation. During the training phase, a mouse was started in either the
north or south arm and had to make a 90° turn to the left or to the right on the basis of visual
cues. Each start arm was used with an equal number of trials in a pseudorandom fashion. Two
sessions, each consisting 10 trials, were carried out per day. Between trials, the mouse was
placed back in the holding cage. The maze arms were wiped down with a sponge isopentanol
solution. The intertrial interval was ~10 seconds.
Performance was evaluated as: (i) percent of correct responses per session during initial
learning and reversal learning; (ii) the number of trials an individual animal required to reach the
learning criterion during initial learning and reversal learning, defined as 9 out of 10 correct
choices in consecutive trials; and (iii) errors made during the reversal-learning phase, categorized
as perseverative errors or regressive errors. The acquisition criterion was defined as more than 9
correct choices in consecutive trials. Regardless of reaching this criterion, at least five sessions
were performed in each test. Perseverative errors, defined as incorrect choices made until the
animal chose the correct arm three out of four consecutive times, occur early in the reversal
74
learning phase and reflect deficits in suppressing previous strategies for finding the food reward.
Regressive errors are defined as an incorrect choice made after the animal chose the correct arm
three out of four consecutive times. Regressive errors occur after the animal begins to learn the
new rule yet occasionally regresses to the previous response strategy, demonstrating a difficulty
in maintaining a new response strategy. All tests of animal behaviors were conducted in a
blinded fashion.
Western Immunoblotting
Western immunoblotting was used to determine the relative expression of proteins within
the dorsal and ventral striatum (N = 9 animals per group per time point). Tissue was dissected
from the dorsal and ventral stiratum and was immediately frozen and stored at -80
o
C until use.
Tissue samples were disrupted by sonication in 300 μl of homogenization buffer (50mM Tris pH
7.4, 150 mM NaCl, 1 mM EDTA, 0.1% sodium dodecyl sulfate, and 1/100 dilution protease
inhibitor cocktail (Cat#: 539134; Calbiochem/EMD Millipore, Billerica, MA). Protein
concentration was determined by the BCA method (Pierce; Rockford, IL). The immunoblotting
technique used was previously described (Jakowec et al., 2004, VanLeeuwen et al., 2010,
Vuckovic et al., 2010, Kintz et al., 2013) with slight modifications. Briefly, samples were
separated on a precast PAGEr Tris-glycine gel (Lonza Rockland Inc; Rockland, ME) and were
transferred to nitrocellulose membranes in Towbin buffer. Nitrocellulose membranes were
washed with PBS, and then blocked in LI-COR blocking buffer (LI-COR, Inc.; Lincoln,
Nebraska) for one hour at room temperature. Membranes were then exposed to the primary
antibody solution overnight at 4
o
C. Primary antibodies used in this study include: D
1
R (1:500,
Cat#: SC-14001, Santa Cruz Biotechnology; Dallas, TX ), D
2
R (1:500, Cat#: SC-5303, Santa
75
Cruz Biotechnology; Dallas, TX), and TH (1:1000, Cat#: MAB318, EMD Millipore; Billerica,
MA). Membranes were washed three times in PBS+0.05% Tween, and exposed to the secondary
antibody solution for two hours at room temperature. Secondary antibodies used in this study
include: Goat anti-Mouse IgG IRDye 800 CW (1:10,000, Cat#: 926-32210, LI-COR Inc.;
Lincoln, Nebraska) and Goat anti-Rabbit IgG IRDye 680 LT (1:15,000; Cat#: 926-68021, LI-
COR Inc.; Lincoln, Nebraska). Nitrocellulose membranes were scanned with an Odyssey
Infrared Imaging System 3.0 (LI-COR, Inc.; Lincoln, Nebraska). Densiometric quantification of
immunopositive bands was determined with Image Studio Software Version 3.1.4 (LI-COR,
Inc.; Lincoln, Nebraska), and expressed as relative optical density (O.D.). The O.D. of each
band was quantified relative to the O.D. of Beta-actin, serving as the protein loading control. For
comparison across groups, the relative O.D. levels for each sample was compared to the
averaged value of WT sedentary mice analyzed on the same blot and normalized. Data are
presented as mean ± SEM.
Statistical Methods
All data are reported as mean SEM. SSPS Statistics 21 (IBM, Armonk, NY)
was used to compare means of acquired data. For the t-maze reversal learning test, performance
differences (percent correct) across testing sessions were analyzed with a repeated-measures
ANOVA, with the within-subjects factor between the test session (Initial Learning Sessions: 1, 2,
3, 4, 5; Reversal-learning Sessions: 7, 8 ,9, 10, 11, 12) and the between-subjects factor being the
experimental group (WT, HD KI, WT + Exercise, HD KI + Exercise). Post-hoc contrasts using
the Bonferroni correction were used to examine the locus of any significant differences across all
groups. The novel objection recognition task and all other t-maze performance measures (trials to
76
criterion, errors) were analyzed via a 2 factor ANOVA. Additionally, all immunoblot data were
analyzed by a 2 factor ANOVA. Where effects of genotype or exercise were identified, the
Bonferroni correction was employed to perform post-hoc analysis to determine significance.
Figures were made in GraphPad Prism 5 (Graphpad Software, Inc., San Diego, CA).
RESULTS
Novel Object Recognition
The discrimination index between novel and familiar objects in the novel object
recognition task was assessed at 4 and 6 months of age in all 4 groups to determine cognitive
capabilities at these time points (Figure 19). At both 4 and 6 months of age, all 4 groups
successfully discriminated between the novel and familiar objects. At 4 months of age, a two-
way ANOVA analysis of sedentary and exercised HD KI and WT mice showed no statistically
significant difference between all 4 groups (F
(3,36)
= 0.006, p = 0.99). There was no statistically
significant effect of genotype (F
(1,36)
= 0.008, p = 0.93) or exercise (F
(1, 36)
= 0.008, p = 0.93) and
no statistically significant interaction between the variables (F
(1,36)
= 0.002, p = 0.97). Similarly,
at 6 months of age, a two-way ANOVA analysis of sedentary and exercised HD KI and WT
mice showed no statistically significant difference between all 4 groups (F
(3,32)
= 0.17 p = 0.92).
There was no statistically significant effect of genotype (F
(1,32)
= 0.17, p = 0.68) or exercise (F
(1,
32)
= 0.0001, p = 0.99) and no statistically significant interaction between the variables (F
(1,32)
=
0.33, p = 0.57).
77
Figure 19. HD KI mice display no cognitive impairment in the novel object
recognition (NOR) task at 4 months and 6 months of age.
Mice were tested for their ability to recognize and ignore a familiar object in lieu
or exploring a novel object in their environment at 4 months (A) and 6 months
(B). There was no significant difference between the performance of HD KI (n =
8-10) and WT controls (n = 8-10) in the NOR task and exercise had no effect on
performance as well (WT + Exercise; n =; HD KI + Exercise, n =). Error bars
indicate SEM.
78
T-Maze Reversal Learning
The ability to adopt a new learning strategy in the t-maze reversal learning task was
assessed at 4 and 6 months of age in all 4 groups to determine cognitive flexibility at these time
points (Figures 20-22). Performance was evaluated as: (i) percent of correct responses per
session during initial learning and reversal learning; (ii) the number of trials an individual animal
required to reach the learning criterion during initial learning and reversal learning, defined as 9
out of 10 correct choices in consecutive trials; and (iii) errors made during the reversal-learning
phase, categorized as perseverative errors or regressive errors.
At 4 months, all 4 groups successfully learned the initial placement of the food reward
with no significant difference in the number of sessions to reach 9 out of 10 correct choices
(F
(3,32)
= 0.67, p = 0.58). During the relearning phase, the percent correct choices within a session
were significantly reduced in sedentary HD KI animals as compared to sedentary WT controls
(F
(3,32)
= 7.04, p < 0.01). Pairwise comparisons revealed a statistically significant decrease in
correct responses by sedentary HD KI animals in sessions 7 – 10 (p < 0.05). On average, 4
month sedentary HD KI animals required 2 more sessions than their WT counterparts to reach
the required 9 out of 10 correct responses. Long-term exercise reduced the number of sessions
necessary to correctly find the food reward 9 out of 10 times in a session. Similarly at 6 months,
all 4 groups also successfully learned the initial placement of the food reward with no significant
difference in the number of sessions to reach 9 out of 10 correct choices (F
(3,32)
= 0.69, p = 0.57).
During the relearning phase, the percent correct choices within a session were significantly
reduced in sedentary HD KI animals as compared to sedentary WT controls (F
(3,32)
= 9.35, p <
0.001). Pairwise comparisons revealed a statistically significant decrease in correct responses by
sedentary HD KI animals in sessions 7 – 10 (p < 0.01). On average, 6 month sedentary HD KI
79
animals required 2 more sessions than their WT counterparts to reach the required 9 out of 10
correct responses. As observed at the 4 month time point, long-term exercise reduced the
number of sessions required by 6 month HD KI animals to correctly find the food reward 9 out
of 10 times within a session.
Figure 20. Long-term treadmill exercise delays cognitive impairments
observed in HD KI mice.
Cognitive ability was assessed in mice via the t-maze reversal learning task at 4
months (A) and 6 months (B) of age. At both time points, all 4 groups
successfully learned the initial placement of the food reward with no significant
difference between the 4 groups in the number of sessions to reach 9 out of 10
correct choices. On average, 4 and 6 month old sedentary HD KI animals
required 2 additional sessions to correctly choose the new location of the food
reward 9 out of 10 times as compared to sedentary WT controls (n = 8). Long-
term treadmill exercise rescued this impairment in HD KI animals (HD KI + Ex;
n= 8) at both 4 months (A) and 6 months (B) of age. Error bars indicate SEM.
Asterisks represent statistical significance: * p < 0.05, ** p < 0.01.
80
At 4 months of age, a two-way ANOVA analysis of sedentary and exercised HD KI and
WT mice showed no statistically significant difference between all 4 groups in the number of
trials required to reach 9 out of 10 consecutive correct responses (F
(3,32)
= 1.38, p = 0.27). There
was no statistically significant effect of genotype (F
(1,32)
= 0.45, p = 0.51) or exercise (F
(1, 32)
=
3.17, p = 0.09) and no statistically significant interaction between the variables (F
(1,32)
= 0.51, p =
0.48). During the relearning phase of the task, a two-way ANOVA analysis of sedentary and
exercised HD KI and WT mice comparing the number of trials required to reach learning criteria
showed a statistically significant difference between all 4 groups (F
(3,32)
= 7.88, p < 0.001) as
well as significant effects of genotype (F
(1,32)
= 6.64, p < 0.05) and exercise (F
(1, 32)
= 9.42, p <
0.01) and a significant interaction between the variables (F
(1,32)
= 7.59, p < 0.01). Post hoc
analysis revealed the following differences between specific treatment groups. Sedentary HD KI
mice had a statistically significant increase (p < 0.01) in the number of trials required to reach
learning criteria compared to sedentary WT mice (HD KI, 46.88 ± 3.81; n = 8; compared to WT,
32.00 ± 3.05; n = 8). Exercise led to a statistically significant decrease (p < 0.01) in the number
of trials required to reach learning criteria between the HD KI mice (HD KI + Exercise, 30.63 ±
2.15; n = 8) compared to HD KI sedentary mice. Exercise did not have a statistically significant
effect (p = 0.99) between WT mice (WT + Exercise, 31.13 ± 1.65; n= 8). There was no
statistically significant change in the number of trials required to reach learning criteria between
KI + Exercise and either WT sedentary or WT+ exercise mice (WT p = 0.99; WT + Exercise p =
0.99).
Similarly at 6 months of age, a two-way ANOVA analysis of sedentary and exercised HD
KI and WT mice showed no statistically significant difference between all 4 groups in the
number of trials required to reach 9 out of 10 consecutive correct responses (F
(3,32)
= .55, p =
81
0.65). There was no statistically significant effect of genotype (F
(1,32)
= 0.09, p = 0.77) or
exercise (F
(1, 32)
= 0.06, p = 0.81) and no statistically significant interaction between the variables
(F
(1,32)
= 1.50, p = 0.23). During the relearning phase of the task, a two-way ANOVA analysis of
sedentary and exercised HD KI and WT mice comparing the number of trials required to reach
learning criteria showed a statistically significant difference between all 4 groups (F
(3,32)
= 8.60,
p < 0.001) as well as significant effects of genotype (F
(1,32)
= 9.13, p < 0.01) and exercise (F
(1, 32)
= 9.13, p < 0.01) and a significant interaction between the variables (F
(1,32)
= 7.54, p < 0.01). Post
hoc analysis revealed the following differences between specific treatment groups. Sedentary
HD KI mice had a statistically significant increase (p < 0.01) in the number of trials required to
reach learning criteria compared to sedentary WT mice (HD KI, 52.13 ± 4.32; n = 8; compared
to WT, 33.88 ± 2.66; n = 8). Exercise led to a statistically significant decrease (p < 0.01) in the
number of trials required to reach learning criteria between the HD KI mice (HD KI + Exercise,
33.88 ± 2.52; n = 8) compared to HD KI sedentary mice. Exercise did not have a statistically
significant effect (p = 0.99) between WT mice (WT + Exercise, 33.00 ± 2.83; n= 8). There was
no statistically significant change in the number of trials required to reach learning criteria
between KI + Exercise and either WT sedentary or WT+ exercise mice (WT p = 0.99; WT +
Exercise p = 0.99).
82
Figure 21. Exercise decreases the number of trials required by HD KI mice
to learn a new strategy in the t-maze reversal learning task at 4 months and 6
months of age.
At both 4 months (A) and 6 months (B), there was no significant difference in the
amount of trials needed by all 4 groups to learn the initial phase of the task.
When the food reward was switched after initial learning, there was a statistically
significant increase in the amount of trials required by sedentary HD KI animals
(n = 8) to learn the new strategy as compared to sedentary WT controls (n =8).
This impairment in cognitive flexibility was rescued by long-term treadmill
exercise in HD KI animals (HD KI + Ex; n= 8) at both 4 months (A) and 6
months (B) of age. Error bars indicate SEM. Asterisks represent statistical
significance: ** p < 0.01.
83
In addition to decreasing the number of trials required by HD KI mice to reach learning
criteria, long-term treadmill running also decreased the amount of perseverative errors made by
HD KI mice at 4 and 6 months of age (Figure 22). Specifically, at 4 months of age, a two-way
ANOVA analysis of sedentary and exercised HD KI and WT mice comparing the number of
perseverative errors made in the relearning phase showed a statistically significant difference
between all 4 groups (F
(3,32)
= 5.44, p < 0.01) as well as a trends towards significant effect of
genotype (F
(1,32)
= 3.29, p = 0.08), a significant effect of exercise (F
(1, 32)
= 8.21, p < 0.01), and a
significant interaction between the variables (F
(1,32)
= 4.83, p < 0.05). Post hoc analysis revealed
the following differences between specific treatment groups. Sedentary HD KI mice had a
statistically significant increase (p < 0.05) in perseverative errors compared to sedentary WT
mice (HD KI, 14.38 ± 1.08; n = 8; compared to WT, 9.13 ± 1.84; n = 8). Exercise led to a
statistically significant decrease (p < 0.01) in perseverative errors between the HD KI mice (HD
KI + Exercise, 7.75 ± 1.25; n = 8) compared to HD KI sedentary mice. Exercise did not have a
statistically significant effect (p = 0.99) between WT mice (WT + Exercise, 8.25 ± 0.86; n= 8).
There was no statistically significant change in the number of perseverative errors between KI +
Exercise and either WT sedentary or WT+ exercise mice (WT p = 0.99; WT + Exercise p =
0.99). A two-way ANOVA analysis of sedentary and exercised HD KI and WT mice showed no
statistically significant difference between all 4 groups in the number of regressive errors made
in the relearning phase (F
(3,32)
= 1.32, p = 0.29). There was no statistically significant effect of
genotype (F
(1,32)
= 0.66, p = 0.42) or exercise (F
(1, 32)
= 1.36, p = 0.25) and no statistically
significant interaction between the variables (F
(1,32)
= 1.95, p = 0.17).
Similarly, at 6 months of age, a two-way ANOVA analysis of sedentary and exercised
HD KI and WT mice comparing the number of perseverative errors made in the relearning phase
84
showed a statistically significant difference between all 4 groups (F
(3,32)
= 9.60, p < 0.001) as
well as significant effects of genotype (F
(1,32)
= 9.97, p < 0.01) and exercise (F
(1, 32)
= 4.04, p <
0.05) and a significant interaction between the variables (F
(1,32)
= 14.79, p < 0.001). Post hoc
analysis revealed the following differences between specific treatment groups. Sedentary HD KI
mice had a statistically significant increase (p < 0.001) in perseverative errors compared to
sedentary WT mice (HD KI, 15.00 ± 1.57; n = 8; compared to WT, 7.38 ± 0.46; n = 8). Exercise
led to a statistically significant decrease (p < 0.01) in perseverative errors between the HD KI
mice (HD KI + Exercise, 8.63 ± 1.09; n = 8) compared to HD KI sedentary mice. Exercise did
not have a statistically significant effect (p = 0.99) between WT mice (WT + Exercise, 9.38 ±
0.94; n= 8). There was no statistically significant change in the number of perseverative errors
between KI + Exercise and either WT sedentary or WT+ exercise mice (WT p = 0.99; WT +
Exercise p = 0.99). A two-way ANOVA analysis of sedentary and exercised HD KI and WT
mice showed no statistically significant difference between all 4 groups in the number of
regressive errors made in the relearning phase (F
(3,32)
= 1.38, p = 0.27). There was no statistically
significant effect of genotype (F
(1,32)
= 0.45, p = 0.51) or exercise (F
(1, 32)
= 3.17, p = 0.09) and no
statistically significant interaction between the variables (F
(1,32)
= 0.51, p = 0.48).
Taken together, these data demonstrate an impairment in cognitive flexibility in sedentary
HD KI as compared to sedentary WT controls. Sedentary HD KI animals not only require more
trials to adopt a new learning strategy, but they also make a significantly larger amount of
perseverative errors, indicating an inability to suppress previous strategies that are no longer
rewarded in lieu of adopting a new positively rewarded strategy. This impairment of cognitive
flexibility is rescued by long-term treadmill exercise in HD KI animals, which display a reduced
85
number of trials required to reach learning criteria in the relearning phase and a reduced number
of perseverative errors.
Figure 22 Exercise decreases the number of perverative errors made by HD
KI mice in learning a new strategy in the t-maze reversal learning task at 4
months and 6 months of age.
Perseverative errors made early in the relearning phase of the t-maze task reflect
cognitive flexibility in adopting a new strategy. There was a statistically
significant increase in the number of perseverative errors made by sedentary HD
KI animals (n = 8) as compared to sedentary WT controls (n = 8). Long-term
treadmill exercise significantly reduced the number of perseverative errors made
by HD KI animals (HD KI + Ex; n= 8) at both 4 months (A) and 6 months (B) of
age. At both time points, there was no significant difference between all 4 groups
in the number of regressive errors made while learning the new strategy. Error
bars indicate SEM. Asterisks represent statistical significance: * p < 0.05, ** p <
0.01, *** p < 0.001
86
Western Immunoblotting
Western Immunoblotting was conducted to examine the effect of exercise on D1
receptor, D2 receptor, and TH concentrations within the dorsal and ventral striatum in mice in all
4 groups at 4 and 6 months of age. Measurements were normalized to sedentary WT controls.
While we observed no change in the concentrations of the D1 receptor or TH, we observed a
significant decrease in the concentration of the D2 receptor in both the dorsal and ventral
striatum in sedentary HD KI animals as compared to sedentary WT controls, which was rescued
by long-term treadmill exercise (Figures 23 and 24).
Specifically, at 4 months of age, a two-way ANOVA analysis of sedentary and exercised
HD KI and WT mice comparing D1 receptor concentration in the dorsal striatum showed no
statistically significant difference between all 4 groups (F
(3,36)
= 0.20, p =0.99). There was no
statistically significant effect of genotype (F
(1,36)
= 0.05, p = 0.82) or exercise (F
(1, 36)
= 0.004, p =
0.95) and no statistically significant interaction between the variables (F
(1,36)
= 0.002, p = 0.97).
Similarly, at 6 months of age, a two-way ANOVA analysis of sedentary and exercised HD KI
and WT mice showed no statistically significant difference in D1 receptor concentration between
all 4 groups (F
(3,36)
= 0.81, p =0.50). There was no statistically significant effect of genotype
(F
(1,36)
= 0.47, p = 0.50) or exercise (F
(1, 36)
= 0.45, p = 0.51) and no statistically significant
interaction between the variables (F
(1,36)
= 1.50, p = 0.23).
Interestingly, at both 4 and 6 months of age there was a significant decrease in the D2
receptor concentration in the dorsal striatum. At 4 months of age, a two-way ANOVA analysis
of sedentary and exercised HD KI and WT mice comparing D2 receptor concentration in the
dorsal striatum showed a statistically significant difference between all 4 groups (F
(3,36)
= 8.98, p
< 0.001) as well as significant effects of genotype (F
(1,36)
= 6.70, p < 0.05) and exercise (F
(1, 36)
=
87
9.40, p < 0.01) and a significant interaction between the variables (F
(1,36)
= 15.51, p < 0.001).
Post hoc analysis revealed the following differences between specific treatment groups.
Sedentary HD KI mice had a statistically significant decrease (p < 0.001) in D2 receptor
concentration compared to sedentary WT mice (HD KI, 0.85 ± 0.02; n = 9; compared to WT,
1.00 ± 0.01; n = 9). Exercise led to a statistically significant increase (p < 0.001) in D2 receptor
concentration between the HD KI mice (HD KI + Exercise, 1.01 ± 0.02; n = 9) compared to HD
KI sedentary mice. Exercise did not have a statistically significant effect (p = 0.99) between WT
mice (WT + Exercise, 0.98 ± 0.02; n= 9). There was no statistically significant change in D2
receptor concentration between KI + Exercise and either WT sedentary or WT+ exercise mice
(WT p = 0.99; WT + Exercise p = 0.99).
Similarly, at 6 months of age, a two-way ANOVA analysis of sedentary and exercised
HD KI and WT mice comparing D2 receptor concentration in the dorsal striatum showed a
statistically significant difference between all 4 groups (F
(3,36)
= 6.48, p < 0.001) as well as a
trends towards significant effect of genotype (F
(1,36)
= 3.50, p = 0.06) and a significant effect of
exercise (F
(1, 36)
= 14.54, p < 0.001). There was no significant interaction between the variables
(F
(1,36)
= 2.12, p = 0.15). Post hoc analysis revealed the following differences between specific
treatment groups. Sedentary HD KI mice had a statistically significant decrease (p < 0.001) in
D2 receptor concentration compared to sedentary WT mice (HD KI, 0.76 ± 0.05; n = 9;
compared to WT, 1.00 ± 0.01; n = 9). Exercise led to a statistically significant increase (p <
0.05) in D2 receptor concentration between the HD KI mice (HD KI + Exercise, 1.15 ± 0.11; n =
9) compared to HD KI sedentary mice. Exercise did not have a statistically significant effect (p
= 0.99) between WT mice (WT + Exercise, 1.17 ± 0.08; n= 9). There was no statistically
88
significant change in D2 receptor concentration between KI + Exercise and either WT sedentary
or WT+ exercise mice (WT p = 0.99; WT + Exercise p = 0.72).
Like with the D1 receptor, there was no significant difference in TH levels in the dorsal
striatum between WT and HD KI animals at both 4 and 6 months. At 4 months of age, a two-
way ANOVA analysis of sedentary and exercised HD KI and WT mice showed no statistically
significant difference in TH concentration between all 4 groups (F
(3,36)
= 0.91, p =0.44). There
was no statistically significant effect of genotype (F
(1,36)
= 1.10, p = 0.30) or exercise (F
(1, 36)
=
1.53, p = 0.22) and no statistically significant interaction between the variables (F
(1,36)
= 0.04, p =
0.84). Similarly, at 6 months of age, a two-way ANOVA analysis of sedentary and exercised
HD KI and WT mice showed no statistically significant difference in TH concentration between
all 4 groups (F
(3,36)
= 1.31, p =0.28). There was no statistically significant effect of genotype
(F
(1,36)
= 0.26, p = 0.61) or exercise (F
(1, 36)
= 1.49, p = 0.23) and no statistically significant
interaction between the variables (F
(1,36)
= 2.18, p = 0.15).
As in the dorsal striatum, a two-way ANOVA analysis of sedentary and exercised HD KI
and WT mice comparing D1 receptor concentration in the ventral striatum at 4 months of age
showed no statistically significant difference between all 4 groups (F
(3,36)
= 0.14, p =0.94). There
was no statistically significant effect of genotype (F
(1,36)
= 0.12, p = 0.73) or exercise (F
(1, 36)
=
0.001, p = 0.98) and no statistically significant interaction between the variables (F
(1,36)
= 0.30, p
= 0.59). Similarly, at 6 months of age, a two-way ANOVA analysis of sedentary and exercised
HD KI and WT mice showed no statistically significant difference in D1 receptor concentration
between all 4 groups (F
(3,36)
= 1.43, p =0.25). There was no statistically significant effect of
genotype (F
(1,36)
= 0.07, p = 0.79) or exercise (F
(1, 36)
= 0.06, p = 0.81). There was a statistically
significant interaction between the variables (F
(1,36)
= 4.22, p < 0.05).
89
Interestingly, at both 4 and 6 months of age there was a significant decrease in the D2
receptor concentration in the ventral striatum. At 4 months of age HD KI and WT mice showed
a statistically significant difference in ventrostriatal D2 receptor concentration between all 4
groups (F
(3,36)
= 13.33, p < 0.001) as well as significant effects of genotype (F
(1,36)
= 7.40, p <
0.01) and exercise (F
(1, 36)
= 23.00, p < 0.01) and a significant interaction between the variables
(F
(1,36)
= 11.17, p < 0.01). Post hoc analysis revealed the following differences between specific
treatment groups. Sedentary HD KI mice had a statistically significant decrease (p < 0.001) in
D2 receptor concentration compared to sedentary WT mice (HD KI, 0.75 ± 0.07; n = 9;
compared to WT, 1.00 ± 0.02; n = 9). Exercise led to a statistically significant increase (p <
0.001) in D2 receptor concentration between the HD KI mice (HD KI + Exercise, 1.08 ± 0.05; n
= 9) compared to HD KI sedentary mice. Exercise did not have a statistically significant effect
(p = 0.99) between WT mice (WT + Exercise, 1.06 ± 0.02; n= 9). There was no statistically
significant change in D2 receptor concentration between KI + Exercise and either WT sedentary
or WT+ exercise mice (WT p = 0.91; WT + Exercise p = 0.99).
Similarly, at 6 months of age HD KI and WT mice showed a statistically significant
difference in ventrostriatal D2 receptor concentration between all 4 groups (F
(3,36)
= 17.08, p <
0.001) as well as significant effects of genotype (F
(1,36)
= 7.48, p < 0.01) and exercise (F
(1, 36)
=
21.90, p < 0.001) and a significant interaction between the variables (F
(1,36)
= 21.08, p < 0.001).
Post hoc analysis revealed the following differences between specific treatment groups.
Sedentary HD KI mice had a statistically significant decrease (p < 0.001) in D2 receptor
concentration compared to sedentary WT mice (HD KI, 0.67 ± 0.04; n = 9; compared to WT,
1.00 ± 0.01; n = 9). Exercise led to a statistically significant increase (p < 0.001) in D2 receptor
concentration between the HD KI mice (HD KI + Exercise, 1.09 ± 0.05; n = 9) compared to HD
90
KI sedentary mice. Exercise did not have a statistically significant effect (p = 0.99) between WT
mice (WT + Exercise, 1.00 ± 0.06; n= 9). There was no statistically significant change in D2
receptor concentration between KI + Exercise and either WT sedentary or WT+ exercise mice
(WT p = 0.99; WT + Exercise p = 0.99).
Like with the D1 receptor, there was no significant difference in TH levels in the ventral
striatum between WT and HD KI animals at both 4 and 6 months. At 4 months of age, a two-
way ANOVA analysis of sedentary and exercised HD KI and WT mice showed no statistically
significant difference in TH concentration in the ventral striatum between all 4 groups (F
(3,36)
=
0.25, p = 0.86). There was no statistically significant effect of genotype (F
(1,36)
= 0.37, p = 0.55)
or exercise (F
(1, 36)
= 0.06, p = 0.80) and no statistically significant interaction between the
variables (F
(1,36)
= 0.22, p = 0.64). Similarly, at 6 months of age, a two-way ANOVA analysis
of sedentary and exercised HD KI and WT mice showed no statistically significant difference in
TH concentration between all 4 groups (F
(3,36)
= 0.48, p = 0.70). There was no statistically
significant effect of genotype (F
(1,36)
= 0.07, p = 0.79) or exercise (F
(1, 36)
= 1.25, p = 0.27) and no
statistically significant interaction between the variables (F
(1,36)
= 0.11, p = 0.74).
Taken together, this data shows a disruption of dopamine neurotransmission via a
downregulation of the D2 receptor in the dorsal and ventral striatum at 4 and 6 months of age.
This downregulation is rescued by long-term treadmill exercise in HD KI animals.
91
92
Figure 23. Long-term treadmill exercise restored dopamine D2 receptor
concentrations in dorsostriatal MSNs.
Concentrations of dopamine D1 and D2 receptors and tyrosine hydroxylase (TH)
were measured by western immunoblot at 4 and 6 months of age. (A-B, previous
page) Sample immunoblots at 4 and 6 months of age. The optical density of D1,
D2, and TH bands were quantified relative to beta-actin and normalized to the
WT sedentary group. (C-D) There was no significant difference in D1
concentrations of HD KI (n = 9) and WT controls (n = 9) at either 4 or 6 months.
Exercise had no effect on the relative concentration of the D1 receptor at these
time points (HD + Exercise, n = 9; WT + Exercise, n = 9). (E-F) At both 4 and 6
months, there was a statistically significant decrease in concentration of the D2
receptor in sedentary HD KI (n = 9) as compared the sedentary WT controls (n =
9). This downregulation was rescued by long-term treadmill running (HD KI +
Exercise, n = 9). (G-H) There was no significant difference in TH concentrations
of HD KI (n = 9) and WT controls (n = 9) at either 4 or 6 months. . Exercise had
no effect on TH concentration at these time points (HD + Exercise, n = 9; WT +
Exercise, n = 9). Error bars indicate SEM. Asterisks represent statistical
significance: * p < 0.05, *** p < 0.001
93
94
Figure 24. Long-term treadmill running restored dopamine D2 receptor
concentrations in ventrostriatal MSNs.
Concentrations of dopamine D1 and D2 receptors and tyrosine hydroxylase (TH)
were measured by western immunoblot at 4 and 6 months of age. (A-B, previous
page) Sample immunoblots at 4 and 6 months of age. The optical density of D1,
D2, and TH bands were quantified relative to beta-actin and normalized to the
WT sedentary group. (C-D) There was no significant difference in D1
concentrations of HD KI (n = 9) and WT controls (n = 9) at either 4 or 6 months.
Exercise had no effect on the relative concentration of the D1 receptor at these
time points (HD + Exercise, n = 9; WT + Exercise, n = 9). (E-F) At both 4 and 6
months, there was a statistically significant decrease in concentration of the D2
receptor in sedentary HD KI (n = 9) as compared the sedentary WT controls (n =
9). This downregulation was rescued by long-term treadmill running (HD KI +
Exercise, n = 9). (G-H) There was no significant difference in TH concentrations
of HD KI (n = 9) and WT controls (n = 9) at either 4 or 6 months. Exercise had no
effect on TH concentration at these time points (HD + Exercise, n = 9; WT +
Exercise, n = 9). Error bars indicate SEM. Asterisks represent statistical
significance: *** p < 0.001.
95
DISCUSSION
In this study, we found that long-term treadmill exercise rescues impairments in cognitive
flexibility observed in sedentary HD KI mice. We first probed HD KI animals for cognitive
ability at 4 and 6 months of age in the novel object recognition task. Both WT and HD KI
animals successfully learned this task with no differences in learning between sedentary HD KI
animals and their sedentary WT controls (Figure 19). The novel object recognition task is a
good probe for general cognitive ability, suggesting that HD KI animals at this age are able to
form long-term memories for environmental stimuli. Given that these animals are not impaired
in long-term memory, we then attempted to elicit more subtle cognitive behaviors in the t-maze
reversal learning task, specifically probing for cognitive flexibility. In the initial phase of this
task, animals must learn to correctly choose between two arms of the maze based on external
visual cues to find a food reward. Not surprisingly, both WT and HD KI animals successfully
learned the initial phase of the task at the same rate regardless of whether or not they exercised.
Upon learning the initial location of the food reward, the location was switched to the opposite
arm. Mice then had to disregard their previous strategy which was no longer rewarded and adopt
a new strategy for finding the reward based on the external cues. At both 4 and 6 months, HD
animals struggled early in the relearning phase to disengage from the previously rewarded
strategy. This is evident in the significantly larger amount of perseverative errors that the HD KI
animals made as compared to their WT counterparts (Figure 22). This impairment early in the
relearning phase resulted in a rightward shift in the learning curve for HD KI animals (Figure 20)
and a significantly larger amount of trials necessary to reach the 9 out of 10 consecutive correct
responses required in this task (Figure 3). Long-term treadmill exercise resulted in a correction
of the learning curve and a decrease in the amount of trials required to learn the task back to the
96
amount observed in sedentary WT controls. Thus, exercise rescued the cognitive inflexibility
observed in sedentary HD KI animals.
Changes in striatal DA signaling have been shown to play a significant role in
determining the performance in the t-maze reversal learning task (Yawata et al. 2012).
Specifically, a reversible blockade of D1 containing MSNs in the direct pathway resulted in
impaired learning in the initial phase of the task as well as in the late half of the relearning phase,
suggesting deficits in the acquisition of reward based learning. On the other hand, a reversible
blockade in D2 containing MSNs in the indirect pathway resulted in impaired learning in the
early part of the relearning phase, suggesting deficits in cognitive flexibility. The results of our
study in HD KI animals closely mirror those observed by Yawata et al. in animals with impaired
striatal DA transmission in the indirect pathway (2012).
To determine if aberrant striatal DA transmission was the underlying mechanism for the
impairments observed in HD KI animals at 4 and 6 months, we measured the levels of dopamine
receptors and tyrosine hydroxylase. We found a significant decrease in the concentration of D2
receptors in sedentary HD KI animals in both the dorsal and ventral striatum at both time points
as compared to their sedentary WT counterparts (Figures 23 and 24). This downregulation of D2
receptors was ameliorated after long term exercise. Taken together, these results suggest that
the indirect pathway was compromised in sedentary HD KI animals, which could have played a
significant role in the impairments in cognitive flexibility observed in the t-maze reversal
learning task. Additionally, exercise was observed to rescue the decrease in D2 receptors
observed in HD KI animals, and exercised animals were able to perform at the same level as
sedentary WT controls in the t-maze reversal learning task, suggesting that a restoration of DA
97
signaling in the striatum played a role in rescuing cognitive inflexibility observed in HD KI
animals.
CONCLUSION
In conclusion, this study suggests that long-term treadmill running, beginning well before
onset of motor symptoms, results in a delay of the cognitive dysfunction observed during the
prodromal period in CAG
140
mice. Additionally, the downregulation of the D2 receptor observed
at 4 and 6 months was rescued after long-term treadmill running, suggesting a restoration of
dopamine neurotransmission and indicating a potential mechanism for symptomatic
improvement observed in these animals. Determining the mechanism(s) by which long-term
treadmill running has its effect in delaying symptom onset may provide potential
pharmacological and genetic interventions targeting neuroplasticity in the basal ganglia resulting
in delaying the onset and progression of HD as well as to improving the quality of life of HD
patients.
98
CHAPTER 4: Treadmill Running Delays the Onset of Motor Dysfunction in the CAG140
Mouse Model of Huntington’s Disease
Author List: DP Stefanko, Z Gasanova, S Patzman, AL Tran, GM Petzinger, and MW Jakowec.
ABSTRACT
Huntington’s disease is a devastating neurodegenerative disorder characterized by
progressive decline in cognitive and motor function and pervasive psychiatric disturbances. The
motor symptoms that lead to the diagnosis of clinical HD begin as subtle abnormalities but
progress to irregular flailing of the head, limbs and trunk known as chorea. The purpose of this
study was to examine the potential impact of exercise in the form of motorized treadmill running
on motor function in the CAG
140
knock-in mouse model of Huntington’s disease. This model
has a long lifespan compared to other rodent models of this disease including the late onset of
motor features after 12 months of age, providing an opportunity to investigate the effects of
interventions such as exercise initiated early in life. Motorized treadmill running was initiated at
4 weeks of age (1 hour per session, 3 times per week) and continued for 12 months. At 12
months of age, sedentary HD KI animals displayed a significant impairment in the accelerating
rotarod task and changes in gait were observed. Long-term treadmill exercise resulted in a delay
in onset of these deficits in motor ability, balance, and coordination. While we previously
showed that exercise delays the formation of htt intranuclear inclusions in the prodromal period,
by 12 months both sedentary and exercised animals exhibited nearly identical htt aggregation in
striatal MSNs. Together these findings begin to address the potential impact of life-style and
early interventions such as exercise on modifying disease progression.
99
KEYWORDS: huntingtin, exercise, Htt protein, accelerating rotarod, gait analysis, Q140,
behavior
INTRODUCTION
Huntington’s disease is a devastating neurodegenerative disorder caused by an excessive
polyglutamine expansion in exon 1 of the huntingtin (htt) protein (The Huntington's Disease
Collaborative Research Group, 1993). In patients, HD is characterized by a progressive decline
triad of motor dysfunction, cognitive impairments, and mood disturbances until premature death.
While cognitive and mood symptoms may begin up to a decade before the onset of motor
symptoms, clinical diagnosis of HD is not confirmed until patients display overt motor features
(Julien et al., 2007; Tabrizi et al., 2013; Epping et al., 2013). Motor symptoms typically emerge
in the 4
th
or 5
th
decade of life as a variety of subtle abnormalities including involuntary tics and
twitches, as well as difficulties with handwriting and grip strength (Kirkwood et al., 2001;
Tabrizi et al., 2013). Motor symptoms worsen with disease progression, as subtle twitches
progress to involuntary rapid non-repetitive movements of the head, limbs, and trunk known as
chorea (Sturrock and Leavitt, 2010; Zheng and Diamond, 2012). The most typical voluntary
deficits are gait abnormalities, as patients gradually lose mobility, coordination, and balance until
they are no longer ambulatory. Towards the end of disease progression, chorea typically
subsides and patients display Parkinsonian features such as bradykinesia and motor rigidity
(Sturrock and Leavitt, 2010; Zheng and Diamond, 2012).
Exercise has been shown to have a beneficial effect on disease progression and symptoms
in several neurodegenerative disorders, including Alzheimer’s disease (Verghese et al., 2003;
Abbott et al., 2004; Yu et al., 2011; Venturelli et al., 2011), Parkinson’s disease (Frazzitta et al.,
2013; Corcos et al., 2013; Petzinger et al., 2010; Petzinger et al., 2013; Speelman et al., 2011),
100
and Huntington’s disease (Pang et al., 2006; van Dellen et al., 2008; Wood et al., 2011; Harrision
et al., 2013; Stefanko et al., 2016a; Stefanko et al., 2016b). Specifically, our lab has
demonstrated that motorized treadmill running attenuates striatal pathology and enhances
dopamine neurotransmission in the CAG
140
mouse model of Huntington’s disease up to 6 months
of age, prior to motor symptoms, resulting in delayed onset of prodromal cognitive impairments
and depression-like behavior (Stefanko et al., 2016a; Stefanko et al., 2016b).
A small number of studies in transgenic mouse models of HD have examined the effect
of exercise but have not shown benefits (Kohl et al., 2007; Potter et al., 2010; Renoir et al.,
2011), while other studies have shown modest improvements in motor function after exercise
(Pang et al., 2006; van Dellen et al., 2008; Wood et al., 2011; Harrison et al., 2013). The vast
majority of studies showing little to no benefit of exercise in HD mouse models have employed
voluntary wheel running. Several recent studies have challenged the validity of using voluntary
wheel running as a long-term intervention, as studies in several different HD models have
demonstrated that HD animals experience reduced running wheel performance over time
(Hickey et al., 2008; Cepeda et al., 2010; Stefanko et al., 2016a). It is therefore possible that the
lack of observed benefits of exercise in these studies could be confounded by reduced motivation
of voluntary wheel running.
This study was designed to observe the effect of long-term exercise on motor function in
the CAG
140
mouse model. This model has a long lifespan compared to other rodent models of
this disease including the late onset of motor features after 12 months of age, providing an
opportunity to investigate the effects of interventions such as exercise initiated early in life. To
avoid the potential confounder of voluntary wheel running, we utilized motorize treadmill
running as an intervention, beginning at 4 weeks of age. To date, no study has attempted to use
101
motorized treadmill running, as opposed to voluntary wheel running, in CAG
140
animals to
examine the potential beneficial effects of long-term exercise on motor function and
neuropathological symptoms of HD at 12 months. This study attempts to investigate the effects
of long-term motorized treadmill running on motor function in the accelerating rotarod task and
through gait analysis. Additionally, this study investigates intranuclear inclusion body formation
in striatal MSNs in exercised and sedentary animals in an attempt to determine whether the
attenuation of striatal pathology observed early in the prodromal period (Stefanko et al., 2016a)
persists up to 12 months of age. The results presented here provide evidence of the benefits of
motorized exercise to delay HD motor symptom onset.
MATERIALS AND METHODS
Animals and Groups
For these studies we used knock-in mice that contained a chimeric mouse/human exon 1
with 140 CAG repeats inserted into the mouse gene by homologous targeting (Menalled et al.,
2003). CAG
140
KI mice were produced in-house using lines descended from heterozygous
pairing. Wild-type littermates were used for controls. The founder mice for our colony were a
generous gift of Drs. Michael Levine and Carlos Cepeda (UCLA) with permission from Dr. Scott
Zeitlin (University of Virginia) through a Material Transfer Agreement. These mice were
backcrossed onto the C57BL/6J background annually to maintain vigor. Mouse genotypes from
tail biopsies were determined using real time PCR (Transnetyx, Inc., Cordova, TN). Mice were
randomly assigned to one of four groups based on WT or homozygous for CAG KI genotype
including: (i) wildtype, (ii) wildtype + exercise, (iii) CAG KI, and (iv) CAG KI + exercise. Only
male mice were utilized in these studies. Mice were group housed with a reverse light cycle
102
(lights off from 7 a.m. to 7 p.m.) and were allowed access to food and water ad libitum.
Experimental procedures were approved by the University of Southern California’s Institutional
Animal Care and Use Committee (IACUC) and were conducted in accordance with the National
Research Council Guide for the Care and Use of Laboratory Animals (DHEW Publication 80-23,
2011, Office of Laboratory Animal Welfare, DRR/NIH, Bethesda, MD). All efforts were made
to minimize animal suffering and to reduce the number of animals used to achieve statistical
significance.
Exercise Regimen
Mice were weaned at postnatal day 28 and subjected to treadmill running on a Model
EXER-6M Treadmill (Columbus Instruments, Columbus, Ohio). The treadmill exercise protocol
was conducted based on our previously publications with modifications (Fisher et al., 2004).
Briefly, exercise was initiated on postnatal day 28. In week one, mice in the exercise groups ran
at a speed of 8.0 ± 0.5 m/min for 40 minutes and they were closely monitored for any adverse
reaction to the treadmill, inability to run, or failure to learn the task. No mice needed to be
excluded. Exercise mice started at a velocity of 10.0 ± 1.5 m/min, and ran 3 times per week for
the 12-month duration of this study. Treadmill speed was gradually increased to 20 ± 1.5 m/min
by the final month. A non-noxious stimulus (metal beaded curtain) was used as a tactile incentive
to prevent animals from drifting back on the treadmill. All mice were weighed at the end of each
week and closely assessed for adverse reactions including stress. In the past we have found
treadmill running to not be stressful based on the evaluation of anxiety, depression, and
corticosterone levels (Gorton et al., 2010).
103
Behavioral Testing
Behavioral testing was carried out to evaluate motor behaviors in this HD model. To
evaluate motor behavior the accelerating rotarod was utilized as well as an analysis of gait
examining painted paw prints on a paper runway at 9 and 12 months of age in all four groups of
mice.
Accelerating Rotarod
Mice were tested for motor function on a rotating spindle 6.0 cm in diameter (Orchid
Scientifics, Nashik, India) according to the methods of Rozas et al. (1998). Prior to the test, mice
were acclimatized to the rotarod for a period of 10 minutes at 5 rpm. During acclimatization,
mice that fell were replaced onto the spindle. Each mouse was subjected to 3 trials at speeds
increasing from 5 rpm to 35 rpm over 300 seconds at 1 rpm intervals, with 300 second rest
between trials. Motor function and coordination was assessed as the average latency to fall of
the second and third trials.
Gait Analysis
Stride length and paw overlap were measured according to the methods of Fernagut et al.
(2002). Briefly, mice were placed at the end of an illuminated pathway (4.5 cm wide, 42 cm
long, with walls 12 cm high) and were allowed to run on a white strip of paper towards a dark
goal box (20 x 17 x 10 cm). Mice were placed into the goal box for 120 s and were acclimatized
into the apparatus for two trials, after which gait was measured in a single trial. Prior to being
placed in the apparatus, the forelimbs and hind limbs of the mice were painted two different
colors to differentiate between front and back paw prints. Stride length was measured as the
104
distance between successive hind limb paw prints. Paw overlap was measured as the distance
between the center of overlapping forelimb and hind limb paw prints. The average of three
stride lengths and paw overlaps were taken for each animal.
Nissl Staining
At the conclusion of behavioral testing at 12 months of age, a total of 32 animals (8 per
experimental group) were anaesthetized and perfused intracardially with 25 ml of normal saline
then with 75 ml of 4% paraformaldehyde in 0.1 M NaH2PO4 (pH 7.1). Brains were removed,
postfixed in the same fixative for 2 weeks at 4°C, immersed in 20% sucrose in 0.1 M NaH2PO4
(pH 7.1) for 24--48 h at 4°C, frozen by slow immersion in isopentane cooled on dry-ice, then
stored at -80°C until sectioning. For each mouse, cryostat-cut sections (20 µm thick)
encompassing the entire mesencephalon were collected free floating and used for Nissl staining
as well as immunohistochemical staining for htt aggregates.
For Nissl staining, sections were mounted on acid-washed gelatin-coated slides, dried
overnight at room temperature, stained with thionin, dehydrated in alcohols, cleared in xylenes,
and coverslipped. Cortical thickness and the area of the lateral ventricles and striatum were
defined bilaterally in stained slices at bregma by manual tracing using Fiji (Schindelin et al.,
2012). Cortical thickness was assessed at bregma at the primary/secondary motor cortex
transition. A line perpendicular from the cortical surface was extended to the external capsule
(cingulum) and its length assessed. The striatum was defined using the boundary of the external
capsule, lateral ventricles and anterior commissures as landmarks. Bilateral measurements were
taken and averaged for each animal. Means and standard errors of the means (SEM) were
calculated.
105
Immunohistochemical Staining for Htt Aggregates
A total of 32 animals (8 per experimental group) were used for immunohistochemical
analysis. Immunohistochemistry was performed on 20 µm-thick slices containing the striatum
between 0.14 mm to 0.62 mm anterior to bregma. 4 tissue sections per animal were washed 3
times for 15 min in TBS 7.2 and blocked for 90 min at 4° C in 5% normal goat serum (NGS) and
TBS with 0.2% Triton X-100 (TX). The anti-htt EM48 antibody (MAB5374, 1:250, EMD
Millipore, Bilerica, MA) was used to detect intranuclear Htt aggregate inclusions. Antibody
specification was validated by subjecting tissue slices to the same IHC protocol but without the
addition of primary or secondary antibodies. The EM48 primary antibody was diluted in TBS
with 2% NGS and 0.2% TX. Following incubation with the EM48 antibody overnight at 4° C,
the slices were washed 3 times in TBS for 15 min. From this point on the tissue was protected
from light. The tissue was subsequently incubated with Alexa Flour 594-conjugated goat anti-
mouse IgG (1:500, Invitrogen, Grand Island, NY) diluted in TBS with 2% NGS and 0.05% TX
for 90 minutes at 4°C. Sections were mounted on gelatin-subbed slides, and cover-slipped using
Vectashield Mounting Medium with DAPI (Vector Laboratories, Inc., Burlingame, CA).
Immunofluorescence intensity was captured at 60x magnification with an Olympus BX61
microscope (Shinjuku, Tokyo, Japan) equipped with a Disk Scanning Unit (spinning disk
confocal) and a 100 W mercury light source (U-LH100HG) using a Hamamatsu Photogenics
ORCA-R2 camera (Hamamatsu, Japan). Images were analyzed using Metamorph Advanced
7.7.20 (Molecular Device, LLC, Sunnyvale, CA). Images were captured from the dorsolateral or
ventrolateral striatum of each brain slice. To account for striatal bundles, cell bodies, and
unfocused areas, data was collected from three regions of interest placed randomly on areas of
fluorescing tissue.
106
Statistical Methods
All data are reported as mean SEM. SSPS Statistics 21 (IBM, Armonk, NY) was used
to compare means of acquired data. All other behavioral and immunochemistry data were
analyzed by 2-factor ANOVA. Where effects of genotype or exercise were identified, the
Bonferroni correction was employed to perform post-hoc analysis to determine significance.
Figures were made in GraphPad Prism 5 (Graphpad Software, Inc., San Diego, CA).
RESULTS
Rotarod Performance
Latency to fall from the accelerating rotarod was assessed at 9 and 12 months of age in
all four groups to determine any deficiency in motor function, balance, and coordination (Figures
25A and 25B). At 9 months of age, a two-way ANOVA analysis of sedentary and exercised HD
KI and WT mice showed no statistically significant difference between all 4 groups (F
(3,33)
=
0.21, p = 0.89). There was no statistically significant effect of genotype (F
(1,33)
= 0.55, p = 0.46)
or exercise (F
(1, 33)
= 0.2, p = 0.90) and no statistically significant interaction between the
variables (F
(1,33)
= 0.07, p = 0.79). At 12 months of age HD KI and WT mice showed a
statistically significant difference in latency to fall between the groups (F
(3,34)
= 5.60, p < 0.01) as
well as significant effects of genotype (F
(1,34)
= 9.85, p < 0.01) and a trend towards significant
effect of exercise (F
(1, 34)
= 3.76, p = 0.06) and a trends towards significant interaction between
the variables (F
(1,34)
= 4.01, p = 0.06). Post hoc analysis revealed the following differences
between specific treatment groups. Sedentary HD KI mice had a statistically significant decrease
(p < 0.01) in latency to fall compared to sedentary WT mice (HD KI, 101.28 ± 8.60 s ; n = 8;
compared to WT, 157.11 ± 13.06 s; n = 8). Exercise led to a statistically significant increase (p <
107
0.05) in latency to fall between the HD KI mice (HD KI + Exercise, 144.09 ± 7.00 s; n = 10)
compared to HD KI sedentary mice. Exercise did not have a statistically significant effect (p =
0.99) between WT mice (WT + Exercise, 156.41 ± 13.64 s; n= 8). There was no statistically
significant change in latency to fall between KI + Exercise and either WT sedentary or WT+
exercise mice (WT p = 0.99; WT + Exercise p = 0.99).
Gait Analysis
A gait analysis task was performed at 9 and 12 months of age in all four groups of to
assess motor coordination based on stride length (Figures 25C and 25D) and forepaw/hindpaw
overlap (Figures 25E and 25F).
At 9 months of age, a two-way ANOVA analysis of sedentary and exercised HD KI and
WT mice showed a statistically significant difference in stride length between all 4 groups (F
(3,36)
= 3.94, p < 0.05). There was a statistically significant effect of exercise (F
(1, 36)
= 10.97, p <
0.01), but no statistically significant effect of genotype (F
(1,36)
= 0.82, p = 0.37) or between the
variables (F
(1,36)
= 0.29, p = 0.60). Post hoc analysis revealed a statistically significant increase
(p < 0.05) in stride length in the HD KI + Exercise mice compared to sedentary WT mice (HD
KI + Exercise, 7.96 ± 0.24 cm; n = 8; compared to WT, 6.84 ± 0.23 cm; n = 10). There was no
other statistically significant difference between the 4 groups (HD KI, 7.12 ± 0.28; n = 8; WT +
Exercise, 7.64 ± 0.14; n = 10). At 12 months of age HD KI and WT mice showed a statistically
significant difference in stride length between the groups (F
(3,37)
= 7.78, p < 0.001) as well as
significant effects of exercise (F
(1, 37)
= 10.75, p < 0.01) and a significant interaction between the
variables (F
(1,37)
= 9.37, p < 0.01). There was no significant effect of genotype (F
(1,37)
= 1.12, p =
0.30). Post hoc analysis revealed the following differences between specific treatment groups.
108
Sedentary HD KI mice had a statistically significant decrease (p < 0.05) in stride length
compared to sedentary WT mice (HD KI, 6.45 ± 0.13 cm; n = 9; compared to WT, 7.28 ± 0.12
cm; n = 8). Exercise led to a statistically significant increase in stride length (p < 0.001) between
the HD KI mice (HD KI + Exercise, 7.72 ± 0.18 cm; n = 12) compared to HD KI sedentary mice.
Exercise did not have a statistically significant effect (p = 0.99) between WT mice (WT +
Exercise, 7.32 ± 0.32 cm; n= 8). There was no statistically significant change in stride length
between HD KI + Exercise and either WT sedentary or WT+ Exercise mice (WT p = 0.69; WT +
Exercise p = 0.92).
At 9 months of age, a two-way ANOVA analysis of sedentary and exercised HD KI and
WT mice showed no statistically significant difference in forepaw/hindpaw overlap between all 4
groups (F
(3,36)
= 2.59, p = 0.07). There was a trend towards significant effect of genotype (F
(1,36)
= 3.62, p = 0.07) and exercise (F
(1, 36)
= 4.09, p = 0.06) and no statistically significant interaction
between the variables (F
(1,36)
= 0.24, p = 0.63). At 12 months of age HD KI and WT mice
showed a statistically significant difference in forepaw/hindpaw overlap between the groups
(F
(3,32)
= 5.97, p < 0.01) as well as significant effects of genotype (F
(1,32)
= 8.45, p < 0.01) and
exercise (F
(1, 32)
= 4.64, p < 0.05) and a significant interaction between the variables (F
(1,32)
=
4.83, p < 0.05). Post hoc analysis revealed the following differences between specific treatment
groups. Sedentary HD KI mice had a statistically significant increase (p < 0.01) in
forepaw/hindpaw overlap compared to sedentary WT mice (HD KI, 1.16 ± 0.09 cm; n = 8;
compared to WT, 0.81 ± 0.05 cm; n = 8). Exercise led to a statistically significant decrease in
forepaw/hindpaw overlap (p < 0.05) between the HD KI mice (HD KI + Exercise, 0.86 ± 0.06
cm; n = 8) compared to HD KI sedentary mice. Exercise did not have a statistically significant
effect (p = 0.99) between WT mice (WT + Exercise, 0.81 ± 0.07 cm; n= 8). The was no
109
statistically significant change in forepaw/hindpaw overlap between KI + Exercise and either
WT sedentary or WT+ exercise mice (WT p = 0.99; WT + Exercise p = 0.99).
110
Figure 25. Long-term treadmill exercise delays the onset of motor function
impairment observed in HD KI mice
(A-B, previous page) Motor ability was assessed in mice on the accelerating
rotarod. At 9 months (A) no significant effect of genotype or exercise was
observed. At 12 months (B) sedentary HD KI animals (n = 8) exhibited a
statistically significant decrease in latency to fall compared to sedentary WT
controls (n = 8). This motor impairment was rescued by long-term treadmill
running (HD KI + Exercise; n = 8). (C-F) Analysis of gait was performed by
painting the forepaws and hindpaws of the mice two different colors and allowing
the mice to walk on a white strip of paper towards a dark goal box. At 9 months
(C, E) no significant effect of genotype or exercise was observed in either stride
length or forepaw/hindpaw overlap. At 12 months (D, F) sedentary HD KI
animals (n = 8-9) exhibited a statistically significant decrease in stride length (D)
and statistically significant increase in forepaw/hindpaw overlap (F) as compared
to sedentary WT controls (n=8). Both of these impairments were rescued by
long-term treadmill running (HD KI + Exercise; n = 8-12). Error bars indicate
SEM. Asterisks represent statistical significance: * p < 0.05, ** p < 0.01, *** p <
0.001
Volumetric Measurements
Cortical thickness and the areas of the lateral ventricles and striatum were measured
bilaterally at bregma in Nissl stained slices from animals in all 4 experimental groups (n = 8 per
group). We observed no qualitative differences in Nissl staining and no quantitative differences
in the above volumetric measurements between WT and HD KI animals (Figure 26).
Specifically, a 2 way ANOVA comparing cortical thickness revealed no statistically
significant difference between sedentary and exercised HD KI mice and WT mice (F
(3,24)
= 0.55,
p = 0.66). There was no statistically significant effect of genotype (F
(1,24)
= 0.32, p = 0.58) or
exercise (F
(1, 24)
= 1.07, p = 0.31) and no significant interaction between the variables (F
(1,24)
=
0.20, p = 0.66). A 2 way ANOVA comparing the area of the lateral ventricles revealed no
statistically significant difference between the 4 experimental groups (F
(3,24)
= 0.04, p = 0.99).
There was no statistically significant effect of genotype (F
(1,24)
= 0.01, p = 0.91) or exercise (F
(1,
24)
= 0.10, p = 0.75) and no significant interaction between the variables (F
(1,24)
= 0.004, p =
111
0.95). Additionally, a 2 way ANOVA comparing striatal area revealed no statistically significant
difference between sedentary and exercised HD KI mice and WT mice (F
(3,24)
= 0.41, p = 0.75).
There was no statistically significant effect of genotype (F
(1,24)
= 0.76, p = 0.39) or exercise (F
(1,
24)
= 0.24, p = 0.63) and no significant interaction between the variables (F
(1,24)
= 0.18, p = 0.67).
Figure 26. HD KI animals do not exhibit changes in striatal or lateral
ventricle areas or cortical thickness at 12 months.
(A) Brain slices of animals in all 4 experimental groups were Nissl stained (n = 8
for all groups). No qualitative differences in Nissl staining were observed at 12
months. Scale bar indicates 1 mm. (B) Slices were analyzed at bregma for striatal
and lateral ventricular areas and cortical thickness. There were no differences
observed between sedentary HD KI animals and WT controls in cortical thickness
or in the size of the striatum or lateral ventricles at 12 months of age.
Additionally, exercise did not have an effect on these measurements in HD KI or
WT animals. Error bars indicate SEM.
112
Immunohistochemical Staining for mHTT aggregates in the Dorsal Striatum
Immunohistochemical staining of the mutant huntingtin (mHTT) protein was used to
analyze exercise effects on intranuclear aggregation in both the dorsal and ventral striatum in
mice from all 4 groups at 12 months of age. We observed quantitative and qualitative changes
between HD KI and WT mice in both the dorsal and ventral striatum (Figure 27).
Specifically, a 2 way ANOVA comparing the percentage of dorsal striatal neurons
containing intranuclear inclusions in exercised and sedentary HD KI and WT mice revealed a
statistically significant difference between the groups (F
(3,59)
= 403.84, p < 0.001). There was a
significant effect of genotype (F
(1,59)
= 1210.31, p < 0.001), but no significant effect of exercise
(F
(1,59)
= 0.88, p = 0.35) or statistically significant interaction between the variables (F
(1,59)
=
0.34, p = 0.57). Post hoc analysis revealed that sedentary KI animals had a statistically
significant increase (p < 0.001) in the percentage of neurons containing intranuclear inclusions
compared to sedentary WT mice (KI, 62.82 ± 2.28%; n = 15; WT, 1.44 ± 0.50%; n = 15; mean%
± SEM). Exercise did not have a statistically significant decrease in intranuclear inclusions (p =
1.00) between the HD KI groups (HD KI + Exercise, 60.18 ± 2.55%; n = 15). Additionally,
exercise did not have a statistically significant effect (p = 1.00) between WT and WT + Exercise
mice (WT + Exercise, 0.82 ± 0.33%; n= 15). HD KI + Exercise mice had a statistically
significant increase in the number of cells showing intranuclear inclusions compared to both WT
and WT + Exercise mice (WT p < 0.001; WT + Exercise p < 0.001).
Similarly, a 2 way ANOVA comparing the percentage of ventral striatal neurons
containing intranuclear inclusions in exercised and sedentary HD KI and WT mice revealed a
statistically significant difference between the groups (F
(3,59)
= 239.10, p < 0.001). There was a
significant effect of genotype (F
(1,59)
= 716.22, p < 0.001), but no significant effect of exercise
113
(F
(1,59)
= 0.41, p = 0.52) or statistically significant interaction between the variables (F
(1,59)
=
0.68, p = 0.42). Post hoc analysis revealed that sedentary KI animals had a statistically
significant increase (p < 0.001) in the percentage of neurons containing intranuclear inclusions
compared to sedentary WT mice (KI, 62.31 ± 3.07%; n = 15; WT, 0.72 ± 0.46%; n = 15; mean%
± SEM). Exercise did not have a statistically significant decrease in intranuclear inclusions (p =
1.00) between the HD KI groups (HD KI + Exercise, 59.04 ± 3.19%; n = 15). Additionally,
exercise did not have a statistically significant effect (p = 1.00) between WT and WT + Exercise
mice (WT + Exercise, 1.12 ± 0.41%; n= 15). HD KI + Exercise mice had a statistically
significant increase in the number of cells showing intranuclear inclusions compared to both WT
and WT + Exercise mice (WT p < 0.001; WT + Exercise p < 0.001).
114
Figure 27. Exercise had no effect on the number and intensity of intranuclear
htt aggregates in dorsostriatal and ventrostriatal MSNs.
(A-B) Immunohistochemical analysis of the mutated htt protein was evaluated in
dorsostriatal (A) and ventrostriatal (B) MSNs of sedentary and exercised HD KI
and WT mice at 12 months of age. Images of coronal sections were taken at 60x
magnification. Each panel displays an overlapped image of MSNs (DAPI) and htt
aggregate staining (RFP) across the different HD KI and WT groups. Scale bar
indicates 10 µm. (C-D) The total number of intranuclear aggregates in a captured
image were counted and recorded as a percentage of the total cells. A significant
increase in the number of intranuclear aggregates was observed in sedentary HD
KI animals (n = 15) at 12 months compared to their sedentary WT counterparts (n
= 15). Exercise had no effect on in the amount of intranuclear aggregates in HD
KI animals. (HD KI + Ex; n = 15). Error bars indicate SEM. Asterisks represent
statistical significance: *** p < 0.001
115
DISCUSSION
In HD, motor symptom onset dictates clinical diagnosis and is a measure of disease
progression. This study demonstrates that long-term exercise can delay the onset of motor
symptoms in the CAG
140
HD mouse model. Sedentary HD KI animals exhibited a statistically
significant decrease in motor performance on the accelerating rotarod compared to WT controls
at 12 months of age. This is consistent with previous findings that demonstrate motor
dysfunction in CAG
140
mice at 12 months (Menalled et al., 2003; Dorner et al., 2007; Hickey et
al., 2008; Rising et al., 2011). Additionally, gait analysis revealed that sedentary HD KI animals
had decreased stride length and increased paw overlap in consecutive strides, both of which
suggest impairments in balance and coordination.
Long-term treadmill exercise rescued the motor dysfunction observed in sedentary HD
KI mice at 12 months of age. After exercise, HD KI animals performed comparably to sedentary
WT animals on the accelerating rotarod (Figure 25). Gait analysis revealed that exercised HD KI
animals had increased stride length and decreased paw overlap between strides as compared to
sedentary HD KI animals. Similarly to the accelerating rotarod task, exercised HD KI animals
exhibited similar stride lengths and paw overlaps as sedentary WT animals, suggesting a delay in
motor symptom onset beyond 12 months. These results contradict the majority of findings in
HD mouse models which used voluntary wheel running as an intervention and showed little to
no benefits of exercise on motor function (Pang et al., 2006; Kohl et al., 2007; van Dellen et al.,
2008; Potter et al., 2010; Renoir et al., 2011; Wood et al., 2011; Harrison et al., 2013). However,
these studies are potentially confounded by the fact that over time HD KI animals display a
decreased preference towards the running wheels, running significantly less than WT
counterparts despite the absence of motor symptoms (Hickey et al., 2008; Cepeda et al., 2010;
116
Stefanko et al., 2016a). It was for this reason that we previously chose to use a motorized
treadmill system for our long-term exercise intervention in the CAG
140
animals, and our previous
studies have demonstrated that motorized treadmill exercise can delay the onset of cognitive
dysfunction and mood disturbances in CAG
140
mice (Stefanko et al., 2016a; Stefanko et al.,
2016b). The results of this study suggest that motorized treadmill exercise can successfully
delay the onset of the final part of the triad of HD symptoms, as exercise rescued motor symptom
deficits in the accelerating rotarod task and gait analysis in HD KI animals.
In this study we measured cortical thickness and the areas of the lateral ventricle and
striatum at bregma in Nissl-stained brain slices of sedentary and exercised 12 month old HD KI
and WT animals (Figure 26). We observed no difference in cortical thickness, striatal area, or
lateral ventricle areas between HD KI and WT animals regardless of whether they underwent
exercise. These findings contradict some previous studies that showed a decrease in striatal
volume by 12 months of age (Lerner et al., 2012). It is possible that neuronal loss could be
present in the anterior portion of the striatum, not assessed by this study. Additionally, exercise
had no effect on striatal area or cortical thickness in HD KI or WT animals.
We have previously reported that long-term exercise initiated at 1 month of age was
sufficient to delay striatal pathology, as intranuclear inclusions were decreased in both number
and intensity at 4 and 6 months of age (Stefanko et al., 2016a). This suggests that at least in the
prodromal period, exercise can modify disease progression beyond simply ameliorating
emerging symptoms. To investigate this further at later time points, we stained for intranuclear
inclusions of aggregated htt proteins in brain slices of sedentary and exercised HD KI and WT
animals (Figure 27). By 12 months of age, the htt immunoreactivity observed was predominated
in the form of large macroaggreates, typically 1 per cell nuclei with a diameter of approximately
117
2 to 3 microns. Roughly 60 percent of striatal neurons of sedentary HD KI were observed to
have these inclusions. Unlike our previous findings in prodromal HD KI animals, exercise had
no effect on striatal pathology, as both the number and intensity of the macroaggregates were
nearly identical between sedentary and exercised groups. This suggests that a different
mechanism beyond altering striatal pathology is responsible for the improvements in motor
function observed in HD KI animals.
CONCLUSION
In conclusion, these studies suggest that long-term treadmill running, beginning well
before onset of motor symptoms, results in a delay of motor symptom onset in CAG
140
mice.
This delay in motor dysfunction occurs despite the lack or change in striatal pathology, as there
was no difference in intranuclear inclusions between exercised and sedentary HD KI animals.
This suggests that the underlying mechanism of the benefits observed after exercise in HD KI
animals is independent from htt aggregation into intranuclear inclusions. Further study is
therefore necessary to elucidate the mechanisms by which long-term treadmill exercise has its
effects in delaying symptom onset. Determining these mechanisms may provide potential
pharmacological and genetic interventions targeting neuroplasticity in the basal ganglia resulting
in delaying the onset and progression of HD as well as to improving the quality of life of HD
patients.
118
CHAPTER 5: Evidence of Functional Brain Reorganization Based on Blood Flow Changes
in the CAG140 Knock-In Mouse Model of Huntington’s Disease
Author List: Z Wang, DP Stefanko, Y Guo, WA Toy, GM Petzinger, MW Jakowec, DP
Holschneider
ABSTRACT
Neuroimaging is increasingly being recognized as an important part of the translational
research agenda. However, neuroimaging, especially functional brain mapping, has only
recently begun to be applied to existent Huntington’s disease (HD) animal models. A HD mouse
model characterized by a gene knock-in (KI) of a human exon 1 CAG
140
expansion repeat
(CAG
140
KI mice) was examined at 6 months and compared to wild-type littermates. Regional
cerebral blood flow (rCBF) was mapped in the awake, non-restrained, male mouse at rest using
[
14
C]-iodoantipyrine autoradiography, and analyzed in three-dimensionally reconstructed brains
by statistical parametric mapping. Our results showed significant changes in rCBF between
CAG
140
KI and WT mice, such that CAG
140
KI animals demonstrated hypo-perfusion of the
basal ganglia motor circuit and hyper-perfusion of cerebellar-thalamic and somatosensory
regions. Significant hypo-perfusion was noted also in CAG
140
KI mice in prelimbic and
cingulate cortex (medial prefrontal area) and the hippocampus – areas associated with cognitive
processing and mood. Changes in rCBF were apparent in the absence of motor deficits (rotarod
test) or atrophy in the striatum (caudate-putamen) or hemispheric volume. Our results suggest a
functional reorganization of whole-brain networks at a presymptomatic stage in the life of the
CAG
140
KI mouse. Functional brain mapping in animals may in the future serve as a
119
translational biomarker for identifying sites of early synaptic change in the HD brain, and for
directing targeted pharmacological and non-pharmacological therapies.
KEYWORDS: brain mapping, caudate-putamen, cortex, cerebellar, neuroplasticity,
neurodegeneration.
INTRODUCTION
Huntington’s disease (HD) is a progressive and devastating neurodegenerative disorder
caused by the abnormal expansion of CAG repeats in the huntingtin (HTT) gene located on
chromosome 4p16.3. HD follows an autosomal dominant form of inheritance, and has a
prevalence of approximately 10/100,000, with variable age of onset. Patients typically survive
15-20 years from initial clinical symptomology (Walker, 2007). HD is characterized by
intracellular aggregates of mutant Htt protein (mHtt) seen throughout the brain, but most
prominently in the basal ganglia and frontal cerebral cortex (Gusella and MacDonald, 2006).
Current treatment is symptomatic and limited in efficacy. At present there is no cure.
Recent reviews of the HD field have suggested that the severity of clinical manifestations
in HD does not depend only on neuronal loss but also on neuronal dysfunction and circuitry
reorganization, and these processes may occur at an early stage of the disease, possibly prior to
neurodegeneration (Niccolini and Politis, 2014). Neuroimaging biomarkers have been proposed
as valuable outcome measures in future clinical trials for pre-symptomatic HD patients because
of their clear relevance to the neuropathology of disease and their increased precision and
sensitivity compared with some standard functional measures (Niccolini and Politis, 2014).
120
Rodent models of HD play an important role in elucidating phenotypic progression and in
investigating potential underlying mechanisms of the clinical disorder. A wide spectrum of such
models are available, including knockout, transgenic, and knock-in (KI) mice, as well as HD rats
(Pouladi et al., 2013). Many of these different models share features that reflect HD
pathophysiology and behavioral deficits. The identification of potential biomarkers in these
animal models will aid in the preclinical screening of new therapeutic modalities. However,
neuroimaging, especially functional brain mapping, has only recently begun to be applied to
existent HD animal models. In the current study, we applied perfusion autoradiography to
examine regional cerebral blood flow (rCBF) in the CAG
140
knock-in (CAG
140
KI) mouse model
of HD in comparison to wild-type (WT) littermates. Our results suggest a functional
reorganization of whole-brain networks in motor, somatosensory and cognitive regions that
occur at a presymptomatic stage in the life of the CAG
140
KI mouse.
MATERIALS AND METHODS
Animals
Mice were derived from a colony of CAG
140
KI mice (C57BL/6 background) established
in our vivarium with 10 to 12 breeding pairs heterozygous for mHTT. The founder mice for our
colony were a generous gift of Drs. Michael Levine and Carlos Cepeda (UCLA), with
permission from Dr. Scott Zeitlin (University of Virginia) through a Material Transfer
Agreement (Menalled et al., 2002). The CAG
140
KI mouse carries a human derived expanded
CAG repeat inserted into exon 1 of the endogenous mouse HTT gene (orthologous to the human
HTT gene). CAG
140
KI mice display aggregates of mHtt protein throughout the basal ganglia
and cortex. In comparison to the R6/2 transgenic mouse model, the CAG
140
KI mouse model
121
exhibits a slower disease course, with a lifespan of about 2 years (Menalled et al., 2003; Hickey
et al., 2008). Mouse experimental groups consisted of 6 month old, male CAG
140
KI animals,
homozygous for mHTT (CAG
140
KI, n=8), as well as wild-type (WT, n=9) littermates derived
from F1 offspring of the crossing of CAG
140
KI heterozygotic mice. Abnormalities in locomotor
activity, rearing, or gait parameters in homozygotes compared to WT mice are largely absent or
minimal at this age (Menalled et al., 2003; Fowler et al., 2015). The copy number of CAG
triplicates in exon 1 of mHTT was verified by periodic genomic sequencing (Laragen Inc.,
Culver City, CA) by PCR analysis of genomic DNA from tail clips.
Mice were group housed with a reverse light cycle (lights off from 7 a.m. to 7 p.m.) and
were allowed access to food and water ad libitum. Experimental procedures were approved by
the University of Southern California’s Institutional Animal Care and Use Committee and were
conducted in accordance with the National Research Council Guide for the Care and Use of
Laboratory Animals (Animals, 2011).
Functional Brain Mapping
Mice were anesthetized with isoflurane (2.0%) and the right external jugular vein was
catheterized with a 1-French silastic catheter (SAI Infusion Technologies, Lake Villa, IL) which
was externalized through subcutaneous space to a dorsal percutaneous port as per our prior
methods (Pang et al., 2011). One week thereafter, the mice were habituated to the experimental
arena (a cylindrical Plexiglas cage with a flat Plexiglas floor, dimly lit at 300 lx) for 30 min/day
for 3 days prior to the resting-state CBF experiments. On the day of the CBF mapping, the
mouse was placed in the experimental arena for 10 min. Thereafter, the animal's percutaneous
cannula was connected to a tethered catheter containing the perfusion radiotracer [
14
C]-
122
iodoantipyrine (American Radiolabelled Chemicals, St. Louis, MO), 325 μCi/kg bodyweight in
0.18 mL of 0.9% saline and a syringe containing an euthanasia solution (50 mg/kg pentobarbital,
3M KCl). The mouse was returned to the arena, and 10 min later the tracer was injected i.v. (1
mL/min) followed immediately by the euthanasia agent, resulting in cardiac arrest within 8 to 10
s, a precipitous drop in blood pressure, termination of the radiotracer perfusion, and death
(Holschneider et al., 2002). The brain was quickly removed and flash frozen in methylbutane
over dry ice.
Autoradiography
Brains were serially sectioned at -20
o
C into 20-μm coronal sections (100-μm interslice
distance). Sections were exposed to Kodak Biomax MR film (Amersham Biosci., Piscataway,
NJ), along with 16 radioactive
14
C standards. Autoradiograms of brain slices were digitized on an
8-bit gray scale using a voltage stabilized light box and a Retiga 4000R charge-coupled device
monochrome camera (Qimaging, British Columbia, Canada). CBF related tissue radioactivity
was measured by the classic [
14
C]-iodoantipyrine method (Sakurada et al., 1978). In this method,
there is a strict linear proportionality between tissue radioactivity and CBF when data is captured
within a brief interval (~10 s) after the tracer injection (van Uitert and Levy, 1978).
Image Processing
Each 3-D brain was reconstructed from 95 digitized autoradiograms (voxel size: 40 μm x
100 μm x 40 μm) using TurboReg (Thevenaz et al., 1998) and a nonwarping geometric model
that includes rotations and translations (rigid-body transformation) and nearest-neighbor
interpolation. A brain template was created using our prior methods (Nguyen et al., 2004), in
123
which the background and ventricular spaces were thresholded based on their optical density.
Spatial normalization consists of a 12-parameter affine transformation, followed by a nonlinear
spatial normalization using 3-D discrete cosine transforms. Normalized brains were smoothed
with a Gaussian kernel (FWHM = 3 x voxel dimension). To account for any global differences in
the absolute amount of radiotracer, voxel intensities of each brain were proportionally scaled to a
single mean.
Analysis of Regional Brain Activation
Statistical parametric mapping (Friston et al., 1995) was used to identify regions showing
significant rCBF changes across the whole brain using our prior methods (Nguyen et al., 2004).
Threshold for significance was set at P < 0.05 at the voxel level and an extent threshold of 100
contiguous voxels, which reflects a balanced approach to control Type I and Type II errors.
Regions were identified by a mouse brain atlas (Franklin and Paxinos, 2007). The directionality
of rCBF changes was analyzed with nonbiased, voxel-by-voxel Student’s t-tests.
Morphometric Measurements and Immunostaining
Cerebral hemispheres, lateral ventricles and cortical thickness were defined bilaterally in
the digitized, thresholded autoradiographic images of each brain by manual tracing using Fiji
(Schindelin et al., 2012). Bi-hemispheric volumes were assessed from 58 serial coronal sections
(AP -4.1 mm AP +2.7 mm, 20 μm thickness, 100 μm inter-slice distance). Lateral ventricular
volume was assessed across 13 coronal sections (AP 0.0 mm AP +1.44 mm). Striatal volume
and cortical thickness was determined in Golgi impregnation sections (Toy et al., 2014) in a
separate dataset of 6 month old male mice (CAG
140
KI mice, WT mice, n=5/group), which
124
allowed a clearer delineation of white matter of the external capsule. The striatum was defined
using the boundary of the external capsule, lateral ventricles and anterior commissures as
landmarks (15 contiguous, 100 μm-thick coronal sections, AP 0.0 mm AP +1.50 mm).
Cortical thickness was assessed at bregma at the primary/secondary motor cortex transition using
both the autoradiographs and Golgi stained sections. A line perpendicular from the cortical
surface was extended to the external capsule (cingulum) and its length assessed. Bilateral
measurements were taken and averaged for each animal. Means and standard errors of the
means (SE) were calculated. Tukey’s HSC tests were used to examine significant group
differences in (1) bi-hemispheric volume (2) lateral ventricular volume, (3) striatal volume, (4)
cortical thickness (P < 0.05). Staining for dopamine- and cAMP-responsive protein phosphatase
of 32 kDa (DARPP-32) and tyrosine hydroxylase were performed using standard methods
(Petzinger et al., 2006).
Accelerating Rotarod Test for Motor Behavior
Mice were tested for motor function on a rotating spindle 6.0 cm in diameter (Orchid
Scientifics, Nashik, India) according to the methods of Rozas et al. (1998). Prior to the test, mice
were acclimatized to the rotarod for a period of 10 min at 5 rpm. During acclimatization, mice
that fell were replaced onto the spindle. Each mouse was subjected to 3 trials at speeds
increasing from 5 rpm to 35 rpm over 300 s at 1 rpm intervals, with 300 s rest between trials.
Motor function and coordination was assessed as the average latency to fall of the second and
third trials.
125
RESULTS
Motor Function on the Rotarod
The latency to fall from the accelerating rotarod was assessed in WT and CAG
140
KI
mice at 6 months of age to determine any deficiency in motor function, balance, and
coordination. There was no statistically significant difference (t = 1.24; P = 0.24) in latency to
fall for CAG
140
KI (152.7 ± 9.3 s; n = 8) and WT mice (175.1 ± 11.9 s; n = 8), indicating no
evidence of gross motor dysfunction at this age.
Cerebral Perfusion Mapping
CAG
140
KI mice compared to WT controls showed significant decreases in rCBF during
the resting state in the basal ganglia motor circuit (P < 0.05, >100 contiguous voxels, Figure 28).
Decreases were bilateral and noted in the striatum, substantia nigra, ventral pallidum, and to a
lesser extent in the external globus pallidus and posterior motor cortex (primary, secondary).
Significant decreases in rCBF were also noted in the nucleus accumbens and ventral pallidum.
The largest decreases were seen in the striatum, with involvement of all subsectors, across all
anterior to posterior bregma levels. The site of greatest decrease in rCBF was in the ventral
striatal/ventral pallidum area.
Significant increases in rCBF were noted in the motor thalamus (ventrolateral,
ventromedial, ventral anterior, mediodorsal, centromedial), the tegmental nucleus (dorsal,
laterodorsal), as well as the subthalamic nucleus and zona incerta area. Significant increases in
rCBF were also clearly seen in the cerebellar vermis, intermediate and lateral lobes, as well as in
the deep cerebellar nuclei.
126
Figure 28: Significant genotypic differences in regional cerebral blood flow
between CAG
140
KI mice compared to wild-type controls in the resting state.
Depicted are select coronal slices (anterior-posterior coordinates in mm relative to
bregma) of the template brain. Colored overlays show statistically significant
increased (red) and decreased (blue) differences for each comparison.
Abbreviations are from the Franklin and Paxinos mouse atlas (Franklin and
Paxinos, 2007): Acb (accumbens nucleus), Crus1 (crus 1 of the ansiform lobule),
CPu (caudate putamen), Int (interposed cerebellar n.), Lat (lateral cerebellar n.),
M2 (secondary motor cortex), Med (medial cerebellar n.), PCRtA (parvicellular
reticular n, alpha part), Pr5 (principal sensory trigeminal n.), PrL (prelimbic
cortex), S1 (primary somatosensory cortex), Sim (simple lobule), SN (substantia
nigra), STN (subthalamic n.), Tg (tegmental area), VP (ventral pallidum).
127
Table 4 outlines a comprehensive list of anatomical regions throughout the brain showing
significant changes in CBF. Included here are significant decreases in rCBF in the dorsal
cingulate (Cg1), prelimbic area (PrL), ectorhinal (Ect) and entorhinal (Ent) cortices, as well as
visual cortex (primary V1, secondary V2). Dramatic bilateral increases in rCBF were noted in
primary somatosensory cortex of the forelimb, jaw and upper lip, anterior portions of barrel field
cortex (S1FL, S1J, S1ULp, S1BF), as well as in secondary somatosensory cortex. Subcortically,
significant decreases in rCBF were seen also in the posterior hippocampal area, including the
CA1 region and dentate gyrus, as well the ventral subiculum.
Though the percentage of optical density measurements noted above were modest when
rCBF of CAG
140
KI mice were compared to WT controls (e.g. dorsal striatum -3.0%, ventral
striatum -3.1%, primary/secondary motor cortex -2.4%; primary somatosensory cortex +2.7%,
motor thalamus +2.4%, cerebellar cortex +3.6%), the results were significant, with clusters of
100 or greater contiguous voxels significant at P < 0.05 – 0.0005. Additional measures
contributing to the confidence of effects detected included the presence of left–right symmetry
and the correspondence of clusters of significant ROIs within the boundaries of known
anatomical structures.
128
Cortex rCBF
Auditory - / -
Cingulate, dorsal - / -
Ectorhinal - / -
Entorhinal, posterior / -
Insula, mid + / + *
Motor cortex, posterior (primary, secondary) - / (-)
Prelimbic - / -
Piriform + / + *
Retrosplenial, posterior (granular, dysgranular) - / -
Somatosensory, primary, forelimb, jaw, upper lip + / + *
Somatosensory, primary, barrel field + / + ant.,
- / - post. *
Somatosensory, secondary + / +
Visual, primary, secondary - / -
Subcortex
Accumbens n. - / -
Caudate putamen
dorsolat., ventrolat., - / - *
dorsomed. , lat., med., ventromed., - / -
posterior CPu (AP -1.8mm -3.0mm) - / - *
Cerebellum
lateral (Crus 1) + / +
intermediate lobule (simple lobule) + / + *
vermis, lobules 2, 3 +
deep n., interposed, medial, lateral + /
Claustrum + / +
Gigantocellular interposed n. +
Globus pallidus, external / -
Hippocampus
CA1, posterior - / -
dentate gyrus, posterior / -
Interpeduncular n. +
Interstitial n. of Cajal / +
Oculomotor n. / +
Parvicellular reticular n., alpha + / + *
Pretectal n., anterior + /
Raphe n., dorsal +
Subiculum, ventral - / -
Substantia nigra - / -
Subthalamic n. + /
Tegmental n., dorsal, laterodorsal + / + *
Tenia tecta, dorsal + /
Thalamus
centromedial n. + / + *
mediodorsal n. + / +
posterior n. + / +
ventral anterior n. + / +
ventrolateral n. + / +
ventromedial n. + / +
Ventroposterior medial n. / +
Trigeminal n., motor, sensory + / + *
Ventral Pallidum - / -
Ventral tegmental area - / -
Vestibular n. + / +
Zona Incerta + / + *
Table 4: Significant changes in rCBF of CAG
140
KI mice compared to wild-
type controls in the resting state in the cortex and subcortex of the left and
right hemispheres (L/R).
‘+’ and ‘-‘ indicate the direction of significant rCBF change in the particular area,
with significance defined for cluster of greater than 100 contiguous voxels
significant at P<0.05 - 0.005 or *P<0.05 - 0.0005. Abbreviations are from the
Franklin and Paxinos mouse atlas (Franklin and Paxinos, 2007).
129
Morphometric Measurements and Immunostaining
No evidence of genotypic differences was observed for lateral ventricular volume
(CAG
140
KI 1.0 ± 0.1 mm
3
, WT 1.1 ± 0.1 mm
3
, P = 0.59), striatal volume (CAG
140
KI 13.9 ± 0.2
mm
3
, WT 14.1 ± 0.4 mm
3
, P = 0.59), cortical thickness (CAG
140
KI 1529 ± 29 µm, WT 1555 ±
27 µm, P = 0.52, autoradiographs; CAG
140
KI 1566 ± 64 µm, WT 1475 ± 89 µm, P = 0.43 Golgi
stain). There was a non-significant trend (P < 0.11) of a decrease (2.5%) in bi-hemispheric
volume of the CAG
140
KI mice (310 ± 3.6 mm
3
) compared to the WT mice (318 ± 2.9 mm
3
). In
addition, there was no significant evidence of striatal atrophy or gross morphological changes in
the dorsolateral striatum following staining for Nissl substance or immunohistochemical staining
for tyrosine hydroxylase or DARPP-32 protein (Figure 29).
Figure 29: Histological staining of CAG
140
KI and WT mice shows no gross
evidence of atrophy in the dorsolateral striatum.
The upper panels compare staining for Nissl substance (40x, scale bar 50 µm).
The middle panels show representative sections from immunohistochemical
staining for tyrosine hydroxylase protein (20x, scale bar 100 µm). The lower
panels show representative sections from immunohistochemical staining for
DARPP-32 protein (40x, scale bar 50 µm).
130
DISCUSSION
In individuals with the HD expansion, the most striking structural and functional
pathological changes are found in the basal ganglia and longitudinal atrophy of this structure is
seen in clinical and pre-clinical patients (Niccolini and Politis, 2014). Cerebral functional
changes may precede onset of motor symptoms. In our study, the CAG
140
KI mice compared to
WT mice revealed a significant decrease in rCBF in the regions of the basal ganglia motor
circuit, with broad significance found in the striatum. Decreases in rCBF appeared in the
absence of striatal atrophy, bi-hemispheric volume loss, motor cortical thickness or gross motor
impairment. Significant, though lesser decreases in rCBF were noted also in regions, such as the
substantia nigra, external globus pallidus and motor cortex, which are directly upstream or
downstream from the caudate-putamen (Fig. 30). In contrast, the motor thalamus (ventrolateral,
ventromedial, ventral anterior, mediodorsal, centromedial) and subthalamic nucleus and zona
incerta showed significantly greater rCBF in CAG
140
KI mice than in WT littermates. Our
findings in the striatum are consistent with functional imaging in presymptomatic mHTT carriers
(Wolf et al., 2011), as well as findings in the YAC128 HD transgenic mouse (Lewandowski et
al., 2013). By contrast, prior work in the R6/2 mouse HD model demonstrated progressive
increases over time in mean regional cerebral blood volume (rCBV) values in striatum,
prefrontal cortex, temporal cortex, primary motor cortex, thalamus and hippocampus (Parievsky
et al, 2012). Differences in the genotype and phenotypic expression in these mouse models may
contribute to difference in functional neuroimaging results. In addition to striatal dysfunction,
decreases in rCBF of CAG
140
KI mice compared to WT mice were noted also in prelimbic and
cingulate cortex (medial prefrontal area) and the hippocampus – areas associated with cognitive
processing and mood (Giralt et al., 2012).
131
Figure 30: Summary of the significant differences in regional cerebral blood
flow in the basal ganglia–thalamic–cortical (BTC) and cerebellar–thalamic–
cortical (CbTC) circuits between CAG
140
KI mice and wild-type controls in
the resting state.
Red shading indicates a relative increase in regional cerebral perfusion in the
CAG
140
KI mice, while blue denotes a relative decrease. GPE (external globus
pallidus), GPI (internal globus pallidus), RN (red n.), STN (subthalamic n.), SN
(substantia nigra), VL (ventrolateral thalamic n.), VM (ventromedial thalamic n.).
While functional and structural deficits in the motor basal ganglia circuit are dominant in
the pathology of the prodromal HD brain, human studies have also shown that prodromal HD
gene carriers may have increased activation outside the basal ganglia (Paulsen et al., 2004;
Zimbelman et al., 2007; Paradiso et al., 2008; Wolf et al., 2007; Wolf et al., 2011; Novak et al.,
2012). CAG
140
KI mice in our study showed significant increases in rCBF in the cerebellum,
including the vermis and deep cerebellar nuclei. In human prodromal HD and HD subjects,
functional imaging has revealed the presence of increased cerebellar activation compared to
normal controls in a variety of activational paradigms (Zimbelman et al., 2007; Beste et al.,
132
2011; Novak et al, 2012; Wolf et al., 2014), though others have reported a reduction in rCBF
(Paradiso et al., 2008; Georgiou-Karistianis et al., 2014).
CAG
140
KI mice compared to WT controls also demonstrated significant increases in
rCBF in sensory regions of the brain, including the primary somatosensory cortex, areas of the
sensory thalamus (ventroposterior medial, posterior), as well as the anterior pretectal nucleus
which has been proposed to play a role in somatosensory processing (Rees et al., 1993).
Whether such activations represent early primary regional brain pathology, or rather represent
early compensatory responses of the HD brain to striatal deficits remains to be investigated.
Interestingly, the above mentioned functional changes in the basal ganglia motor circuit,
the somatosensory circuit, and cerebellar-thalamic circuit of the CAG
140
KI mice largely
reproduced changes elicited by us previously in rats after bilateral lesioning of the dorsal
striatum with 6-hydroxydopamine (6-OHDA) (Wang et al., 2013). This suggests the possibility
that functional brain changes may be determined in the HD mouse by striatal pathology. An
exception to this was that in our rats, the 6-OHDA lesions spared the ventral and anterior-most
striatal regions, with a resultant increase in rCBF in these regions.
In the field of HD research there is a wide spectrum of rodent models available whose
time course of disease progression is attributed to variations in the expression level of the mHTT
gene. We selected the CAG
140
KI model since it has a long prodromal period where motor
behavior symptoms do not appear until 12 months of age. Our findings show that significant
changes in rCBF occur 6 months before motor symptoms, suggesting that alterations in brain
structure and connectivity occur early in the lifespan of this model. Currently, we do not know
the underlying synaptic mechanisms that link rCBF and neuronal dysfunction (or compensation)
in the prodromal phase of HD. However, these changes indicate that there is a significant
133
therapeutic window to test interventions such as physical activity or diet, an approach not
amenable to faster progressing models. Such interventions may target synaptic dysfunction and
consequently modify disease progression.
CONCLUSION
CAG
140
KI mice compared to WT mice showed a striking hypo-perfusion of basal
ganglia motor circuit and hyper-perfusion of cerebellar-thalamic and somatosensory regions at 6
months of age. These results suggest a functional reorganization of whole-brain networks in this
prodromal model of HD. While no animal model replicates all features of HD, the CAG
140
KI
mouse provides an alternative approach to study the functional brain changes in pre-manifest
HD. The use of select functional neuroimaging end points in animals may provide an improved
means for the bidirectional translation of biomarkers between animal models and human disease
conditions. Reverse translation of available biomarkers into rodents may provide an important
tool for evaluating pharmacological effects of candidate compounds (Parievsky et al., 2012).
Future studies will have to address the predictive validity of this approach.
134
CHAPTER 6: CONCLUSIONS
The studies in this dissertation examine the effects of long-term treadmill exercise on
modifying Huntington’s disease progression through experience-dependent neuroplasticity.
Experience-dependent neuroplasticity is comprised of a variety of processes resulting in the
incorporation of new experiences and learning of new behaviors with changing external stimuli.
Neuroplasticity occurs at many different levels in the brain, from molecular and structural
changes at the level of the synapse, to cell signaling and circuit changes between neurons, to
behavioral changes (i.e. motor, cognition, mood) of the organism as a whole. At the synapse,
neuroplasticity can be observed in changes in receptor density, neurotransmitter release, and
dendritic spine formation. Neuroplastic changes in functional connectivity and blood flow can
alter brain circuits. Neuroplasticity in the brain is observed as a result of a variety of different
experiences, including: neurodevelopment, learning, exercise/physical activity, stress, injury, and
disease.
The studies in this dissertation investigated experience-based neuroplasticity in the
CAG
140
mouse model of HD, in an attempt to determine the effect that long-term treadmill
exercise has in modifying the disease progression, as observed in sedentary animals. HD has no
cure and the predominant strategy for treatment in HD aims to attenuate symptoms after their
onset to prolong the quality of life in HD patients. Determining the mechanisms that can modify
disease progression could identify potential therapeutic targets for improved treatments in HD
and could potentially delay the onset of symptoms, resulting in prolonged life expectancy in
patients. Exercise is a promising therapeutic strategy widely used in the treatment of PD and
AD, as long-term studies have shown neuroprotective benefits in patients with these disorders.
While similar longitudinal studies on the role of long-term exercise in HD have not yet been
135
carried out, several studies that have investigated predictors of HD have suggested that lifestyle
factors (like exercise) may play a role in modifying disease progression. The studies in this
dissertation, therefore, attempt to address this gap in knowledge by investigating exercise-
induced changes in pathology and behavior in the CAG
140
HD mouse model. Chapters 2 through
4 focus on the role that long-term treadmill exercise plays in modifying the progression of the
three main symptoms of HD: psychiatric disturbances in the form of depression, cognitive
impairments, and motor dysfunction, respectively. In addition, each of these chapters
investigates potential underlying mechanisms for the benefits observed in exercised animals.
Chapter 5 examines the differences between sedentary HD KI and WT animals in functional
brain mapping via regional cerebral blood flow analysis to determine changes in whole-brain
networks observed in the disease state. These studies are of key importance as they may provide
insight into several potential therapeutic targets and/or strategies for the improved treatment in
HD patients.
Conclusions from Chapter 2: Long-term treadmill exercise delays the onset of depression-like
behavior and striatal pathology in the CAG
140
HD mouse model
Clinically, HD is characterized by a progressive decline in motor function and cognitive
ability as well as neuropsychiatric disturbances, such as depression, ultimately resulting in
premature mortality 15 to 20 years after diagnosis (Walker 2007). Recent studies have shown
that cognitive and mood symptoms can occur up to a decade prior to motor symptom onset, a
stage deemed the prodromal phase (Julien et al., 2007; Tabrizi et al. 2013; Epping et al., 2013).
Since the mood disturbances observed in HD patients, particularly depression, make a significant
136
contribution to the overall morbidity of the disorder, these studies have indicated the need for
new therapeutic strategies initiated during the prodromal phase. The work discussed in Chapter
2 builds upon previous studies in HD rodent models that have used voluntary wheel running and
have shown little to no benefits in modifying disease progression (Pang et al., 2006; Kohl et al.,
2007; van Dellen et al., 2008; Potter et al., 2010; Renoir et al., 2011; Wood et al., 2011; Harrison
et al., 2013). However, several recent studies have challenged the validity of using voluntary
wheel running in long-term studies of HD mouse models, as over time HD animals display a
decreased preference towards the voluntary running wheels, running significantly less than WT
controls despite the absence of motor symptoms (Hickey et al., 2008; Cepeda et al., 2010). This
study replicates these results in the CAG
140
mouse model (Figure 14). A cohort of sedentary HD
KI animals and WT controls were probed biweekly from age 2 months to 6 months for distance
run on voluntary wheels in a 6 hour period. HD KI animals exhibited a significant decrease in
the distance run beginning at 4 months and continuing until the termination of the study at 6
months of age. Similarly to the decrease in wheel running observed in other models, this
decrease occurred well before the onset of motor symptoms in the CAG
140
mouse. It is possible
that the lack of exercise benefits observed in previous studies using voluntary running wheels is
potentially due to reduced motivation for the running wheels themselves. In order to avoid this
potential confounder, the remaining studies in Chapter 2 employed motorized treadmill running,
rather than voluntary wheels, as a therapeutic intervention.
The results of the studies presented in Chapter 2 suggest that long-term treadmill exercise
initiated well before symptom onset rescues depression-like behavior and attenuates pathological
abnormalities observed in the CAG
140
mouse. Motor function, assessed via the accelerating
rotarod task and gait analysis, was not significantly different between sedentary HD KI and WT
137
animals at 4 or 6 months, indicating that these animals were still in the prodromal phase at these
time points (Figure 15). Depression-like behavior in two separate tasks (forced swim test and
tail suspension test) was observed in HD KI sedentary mice beginning at 4 months of age and
worsening at 6 months (Figure 16). We found that long-term treadmill exercise rescued the
depression-like behavior observed in sedentary HD KI mice, effectively delaying the onset of
mood disturbances up to 6 months of age. These findings are consistent with previous studies in
the R6/1 mouse model which demonstrated that voluntary wheel running can rescue depression-
like behavior (Renoir et al., 2012).
This study additionally investigated potential underlying mechanisms of the beneficial
effects that exercise had on depression-like behavior by analyzing htt intranuclear inclusion
formation and measuring dopamine, serotonin, and norepinephrine concentrations in the
striatum. Consistent with previous reports (Menalled et al., 2003), intranuclear inclusions were
first observed at 2 months of age in dorsostriatal and ventrostriatal MSNs and increased in
number and intensity over time (Figure 17 and 18). At both 4 and 6 months, long-term treadmill
exercise significantly decreased the percent of dorsostriatal and ventrostrital MSNs that
contained intranuclear inclusions. Interestingly, at 6 months, the percent of striatal MSNs
containing inclusions was less than the percent inclusions measured in 2 month old sedentary HD
KI animals, suggesting a delay in aggregate formation in exercised animals. We also found that
sedentary HD KI animals had significantly reduced striatal concentrations of dopamine,
serotonin, and norepinephrine (Table 3). Long-term treadmill running significantly increased the
availability of all three of these neurotransmitters. Since these neurotransmitters have long been
implicated as playing a role in depression (Pare, 1972), this study suggests that the restoration of
dopamine, serotonin, and/or norepinephrine concentrations in the striatum could play a role in
138
regulating the improvements in depression-like behaviors observed in exercised CAG
140
animals.
This is further validated in HD mouse models by previous studies (Renoir et al., 2012), which
have shown beneficial effects of selective serotonin reuptake inhibitors (SSRIs) in improving
depression-like behavior in R6/1 mice.
Collectively, the findings in Chapter 2 suggest that long-term treadmill exercise, initiated
well before the onset of motor symptoms, results in a delay of depression-like behavior and
striatal pathology observed during the prodromal period in CAG
140
mice. Additionally, exercise
restored the decreased striatal levels of dopamine, serotonin, and norepinephrine observed in
sedentary HD KI animals, indicating several potential mechanisms by which long-term treadmill
exercise could have an effect in delaying symptom onset and htt aggregate formation. Further
investigation into the underlying mechanisms of exercise benefits may provide additional
therapeutic targets.
Limitations of Chapter 2: In this study, we utilized immunohistochemistry (IHC) to
observe changes in intranuclear inclusions in sedentary and exercised HD KI and WT animals.
This technique allowed us to visualize htt aggregation within the nucleus and quantify the
percent of striatal MSNs that contained inclusions. While our results suggest that htt aggregation
is delayed in exercised HD KI animals, changes over time in the total concentration of mutant htt
were not assessed in this study. The htt protein is rather large (~350 kDA) and is therefore
difficult to analyze via immunoblotting for protein concentration, as the large size precludes it
from descending into the wells of the western blot gels. There are several techniques to solve
this problem, however, we were not successful in our attempts to do so, and therefore have not
yet measured for htt protein concentration via immunoblot. Our IHC data suggests a delay in the
formation of intranuclear inclusions, but it is not clear if this is due to a decreased amount of htt
139
protein or if the concentrations of htt are the same and htt is in a monomeric or oligomeric form
in the cytoplasm rather than macroaggreates observed in the nuclease. It has also been suggested
that htt oligomers are the toxic species of mutant htt (Kuemmerle et al 1999; Lee et al. 2003;
Wolfe and Cyr 2011). It would be interesting to determine the effect that exercise has on mutant
htt oligomerization. Further investigation is therefore needed to determine the effects of exercise
on macroaggregation of htt protein into intranuclear inclusions.
Additionally, this study points to an involvement of dopamine, serotonin, and/or
epinephrine levels in modulating the beneficial effects seen in depression-like behavior. We
found that long-term treadmill exercise was sufficient to rescue decreased striatal levels of these
neurotransmitters, and previous studies have suggested that chronic SSRI treatment can
ameliorate depression-like behavior in the R6/1 mouse model (Renoir et al., 2011). Further
studies into the mechanisms that these neurotransmitters may regulate to have beneficial effects
are required, particularly those employing monoamine agonists and antagonists to selectively
observe the effects of one neurotransmitter at a time. The combinatory effects of exercise and
chronic administration of antidepressants (SSRIs or SNRIs) would also be interesting to
investigate in the CAG
140
HD mouse model.
Conclusions from Chapter 3: Long-term treadmill exercise delays the onset of cognitive
impairments and restores dopamine receptor levels in the CAG
140
HD mouse model
Similarly to the mood disturbances focused on in Chapter 2 of this dissertation, cognitive
impairments in HD patients have also been shown to emerge well before the onset of motor
symptoms (Julien et al., 2007; Tabrizi et al., 2013; Epping et al., 2013). Early cognitive
140
impairments in HD patients predominately present as difficulties in cognitive flexibility, as
evidenced by impaired performance on the Wisconsin Card Sorting Test and set-shifting tasks
(Owen et al., 1993; Lawrence et al., 1996; Ho et al., 2003). Impairments in cognitive flexibility
have been associated with impairments in DA neurotransmission in both patients (Cools et al.,
2006; Dang et al., 2012) and HD mouse models (Whishaw et al., 1987; Pisa and Cyr, 1990). The
studies discussed in Chapter 3 focus on the role that long-term exercise has in improving
cognitive flexibility in the CAG
140
HD mouse model.
In this study we found that long term exercise rescues impairments in cognitive flexibility
observed in sedentary HD KI animals. We first confirmed that HD KI animals had the capacity
to learn during the prodromal period using the novel object recognition task, a relatively simple
task that can detect impairments in long-term memory formation for external stimuli. Both HD
KI and WT animals were successful in discerning between the novel and familiar objects at both
4 and 6 months, suggesting that HD KI animals are capable of forming long-term memories
(Figure 19). Given that HD KI animals were not globally impaired, we then probed for more
subtle cognitive impairments, particularly cognitive flexibility. To this end, we employed the t-
maze reversal learning task in which mice have to use external environmental cues to determine
the location of a food reward. Upon completion of this initial learning, the location of the food
reward is changed and mice must disregard their former strategy that is no longer rewarded and
adopt new exploratory behavior to learn the new location. At both 4 and 6 months, sedentary
HD KI and WT animals performed equally well in the initial learning phase, but sedentary HD
KI animals required significantly more trials during the reversal phase to learn the new location
of the food reward (Figures 20 and 21). HD KI animals that underwent long-term exercise
exhibited a significant decrease in the number of trials required to learn during the reversal phase
141
as compared to their sedentary counterparts. Previous studies (Yawata et al., 2012) have
suggested that the reversal phase of the t-maze task can be broken into two parts, where
perseverative errors, made early in this phase, indicate a failure to disengage from the previous
strategy in lieu of adopting a new one, and regressive errors, made late in this phase, indicate a
failure to learn the new location (similar to errors made in the initial phase). Error analysis
demonstrated that sedentary HD KI animals made significantly more perseverative errors than
WT controls, suggesting an impairment in cognitive flexibility (Figure 22). The increase in
perseverative errors results in a rightward shift of the overall learning curve in the sedentary HD
KI animals (Figure 20). Long-term exercise resulted in a significant decrease in the number of
perseverative errors in HD KI compared with their sedentary counterparts and a restoration of the
learning curve.
Changes in striatal DA neurotransmission in the direct and indirect pathways have been
shown to play a significant role in determining the performance in the t-maze reversal learning
task (Yawata et al., 2012). Specifically, a blockade of D1 receptor containing striatal MSNs
which constitute the direct pathway results in impaired learning in the initial phase and the late
half of the reversal phase of the task, while blockade of the D2 receptor containing striatal MSNs
results in impaired learning in the early half of the reversal phase. Our findings in the CAG
140
mice are therefore consistent with those seen in mice that have impaired signaling in the indirect
pathway resulting in cognitive inflexibility.
To further investigate the role of DA neurotransmission in the CAG
140
mice, we analyzed
protein concentrations of the D1 and D2 receptors as well as TH, the rate-limiting enzyme in DA
synthesis. At both 4 and 6 months of age, we found a significant decrease in D2 receptor
concentration in the dorsal and ventral striatum in sedentary HD KI animals compared to WT
142
controls (Figures 23 and 24). Long-term exercise in HD KI animals rescued these decreases and
restored D2 receptor levels to those observed in sedentary WT controls.
Together the results of the studies in Chapter 3 suggest that the D2 receptor
mediated indirect pathway is compromised early in the disease progression, well before the onset
of motor symptoms, resulting in cognitive inflexibility during the prodromal period. Long-term
exercise was sufficient to restore decreased D2 receptor protein levels in HD KI mice. After
exercise, HD KI animals were able to perform comparably to sedentary WT controls in the t-
maze reversal learning task, suggesting that a restoration of DA signaling through the indirect
pathway in the striatum played a role in ameliorating the cognitive inflexibility observed in
sedentary HD KI animals.
Limitations of Chapter 3: In this chapter, we employed immunoblotting to probe for
changes in dopamine receptor concentration. While this approach allowed us to see overall
changes in receptor levels, it is limited in its ability to determine whether these receptor changes
occur are pre-, post-, or extra-synaptic sites. Further studies could cross the HD animals with
D2-GFP animals that our lab has in house, allowing for selective analysis of the D2 receptors via
immunohistochemistry and electrophysiological studies. These studies could further delineate
the role that long-term exercise has in the indirect pathway in HD KI animals, particularly on
AMPA receptor levels and composition, which have been shown to be compromised in PD
mouse models (Kintz et al., 2013). It would be interesting to see if the impaired indirect pathway
signaling in the CAG
140
HD mouse model follows a similar mechanism.
Additionally, future studies should also consider using behavioral tasks that are not
dependent on food rewards, as these tasks may be ineffective in mouse models with reduced
motivation. The CAG
140
animals have been shown to have a reduced motivation for the
143
voluntary wheel after 4 months (discussed in Chapter 2) and could have decreased motivation for
a food reward as more mood disturbances emerge over time. Other tasks that could avoid this
potential confounder are the contextual novel object recognition task and the Morris Water
Maze, both of which have been shown to be impaired in the DA-depleted brain (Miyoshi et al.,
2002, Moriguchi et al., 2012).
Conclusions from Chapter 4: Long-term treadmill exercise delays the onset of motor
dysfunction in the CAG
140
HD mouse model
While the mood disturbances and cognitive impairments discussed in Chapters 2 and 3
respectively may emerge up to a decade before the onset of motor symptoms (Julien et al., 2007;
Tabrizi et al., 2013; Epping et al., 2013), clinical diagnosis of HD is not made until a patient
displays overt motor dysfunction, typically in the 4
th
or 5
th
decade of life. As such, the majority
of studies in HD patients investigating potential targets for modifying disease progression have
focused on delaying the onset of motor symptoms. The same holds true for HD mouse models,
including the CAG
140
mouse utilized in all of the studies in this dissertation (Menalled et al.
2003; Dorner et al. 2007; Hickey et al., 2008; Rising et al., 2011). These studies have
demonstrated that overt motor dysfunction begins to emerge around 12 months of age in HD KI
animals. Since motor symptom onset dictates clinical diagnosis and is a measure of disease
progression, finding new therapeutic interventions that can delay motor symptom onset may
provide insight into better treatment options for patients with HD.
The studies in Chapter 4 focus on the effects that long-term exercise, begun well before
the onset of motor features in HD KI animals, has on modifying disease progression using motor
144
symptom onset as an outcome measure. Consistent with previous findings, we observed motor
dysfunction in the accelerating rotarod task, as sedentary HD KI animals exhibited a significant
decrease in the latency to fall off the rod as compared to WT controls (Figure 25). Similarly, gait
analysis also revealed motor deficits in decreased stride length and increased paw overlap in
consecutive strides, suggesting impairments in balance and coordination. Long-term treadmill
exercise rescued these motor deficits as exercised HD KI animals performed comparably to
sedentary WT controls in both the accelerating rotarod task and gait analysis.
In this study we also analyzed brain slices for cortical thickness and the areas of the
lateral ventricles and striatum to indirectly probe for cell loss. We found no difference between
HD KI and WT animals in any of the brain measurements at 12 months regardless of whether or
not they were exercised (Figure 26). Additionally, we built upon the studies in Chapter 2 of this
dissertation by analyzing intranuclear inclusions in striatal MSNs at 12 months. We had
previously suggested that long-term exercise delayed the onset of striatal pathology by several
months as 6 month exercised HD KI animals exhibited fewer inclusions than 2 month sedentary
HD KI mice. Analysis at 12 months revealed a significant increase in MSNs with inclusions
when comparing sedentary HD KI animals to WT controls (Figure 27). Unlike our previous
findings in the prodromal period, there was no difference in the percent of neurons with
inclusions between exercised and sedentary HD KI animals at 12 months, suggesting a different
mechanism by which exercise has its beneficial effects in delaying the motor symptom onset.
Limitations of Chapter 4: In this chapter we present evidence that long-term exercise
can delay the onset of motor symptoms. The exercise intervention began at 1 month of age and
was continued until 12 months. The major limitation of this study comes in its feasibility in
translating to human patients, as this study suggests that they would need to be identified
145
immediately as being at risk and placed on a high intensity exercise program for their entire life.
Nevertheless, elucidating the onset of disease pathology could point to a crucial time period by
which HD patients must begin interventions in order for beneficial effects to be observed.
Previous studies in other HD mouse models that have reported negative or neutral effects of
exercise began their intervention when their mice were several months old, allowing for HD
pathology to begin to emerge (Pang et al., 2006; Potter et al., 2010). Our study suggests that
exercise must be initiated early in the prodromal period to have an effect, but it is not known
how early and for how long this intervention must be carried out for positive effects. Further
investigation can determine how long of an intervention time period is necessary for exercise to
have the neuroprotective effects displayed in the studies in this thesis. For example, in this study
mice were run from 1 month to 12 months. It would be interesting to investigate whether fewer
months of running would be sufficient to alter disease progression such that one would still
observe benefits at later time points (i.e. motor symptom delay at 12 months). Additionally,
other further studies could delay the initiation of exercise until disease progression has begun to
determine if exercise has neurorestorative properties in HD models.
In this study, the analysis of intranuclear inclusion formation in dorsostriatal and
ventrostriatal MSNs found no effect from exercise. However, as previously mentioned, other
species of mutant htt (i.e. oligomers) have been implicated as being toxic and potentially
responsible for cellular death. It is therefore possible that exercise could still have beneficial
effects of reducing striatal htt oligomers which was not assessed in this study. On the other
hand, it is equally plausible that the benefits in motor function observed after exercise are
independent of striatal pathology, in which case, further investigation would be required in order
to elucidate the mechanism(s) by which exercise results in beneficial effects.
146
Conclusions from Chapter 5: HD KI animals exhibit a functional remapping of whole-brain
networks evident early in disease progression.
In presymptomatic and symptomatic HD patients, the most prominent molecular,
pathological, and functional changes occur in the basal ganglia, and may depend not only on
neuronal loss, but also on neuronal dysfunction and circuitry reorganization (Niccolini and
Politis, 2014). Since this reorganization may occur early in disease progression, neuroimaging
biomarkers have been proposed as valuable outcome measures in future clinical trials in
preclinical HD patients. As such, the identification of these biomarkers can aid in screening new
therapeutic interventions in both HD patients and in preclinical research in HD rodent models,
such as the CAG
140
HD KI mouse. Functional brain mapping has only recently begun to be
applied to HD mouse models. The studies in Chapter 5 focus on functional brain mapping in the
CAG
140
mouse by analysis of regional cerebral blood flow (rCBF) to identify brain areas with
functional changes as a result of HD.
In this study, we found that 6 month sedentary HD KI animals compared to WT controls
exhibit a significant decrease in blood flow in the basal ganglia motor circuit, particularly in the
striatum (Figure 28). These changes occurred in the absence of striatal atrophy and motor
symptom onset. We also observed significant, though lesser, decreases in rCBF in the substantia
nigra, the external globus pallidus, and the motor cortex, which are either directly upstream or
downstream from the striatum (Figure 30). Other areas of the brain, including the motor
thalamus, (ventrolateral, ventromedial, ventral anterior, mediodorsal, centromedial) and
subthalamic nucleus and zona incerta had significant increases in rCBF in HD KI animals
compared to WT controls. Our findings in the striatum are consistent with previous studies in
147
both presymptomatic HD patients (Lewandowski et al., 2013) and in the YAC128 HD mouse
model (Parievsky et al., 2012). In addition to striatal decreases, we also observed decreases in
rCBF in HD KI animals in the prelimbic and cingulate cortex (medial prefrontal cortex) and in
the hippocampus, which have been shown to play a role in cognitive processing and mood
(Giralt et al., 2012).
Several human studies have demonstrated increases in neuronal activation outside of the
striatum (Paulsen et al., 2004; Zimbelman et al., 2007; Paradiso et al., 2008; Wolf et al., 2007;
Wolf et al., 2011; Novak et al., 2012). HD KI animals in our study exhibited increased rCBF in
the cerebellum, and in sensory regions of the brain including the primary somatosensory cortex,
areas of the sensory thalamus (ventroposterior medial, posterior), as well as the anterior pretectal
nucleus which has been implicated in playing a role in somatosensory processing. Is it unknown
whether the activation in these areas comes as a result of early HD pathology, or if it represents
early compensatory responses to the decreases observed in the basal ganglia. However, taken
together, the findings of this study suggest a functional reorganization of whole brain networks,
particularly in the striatum, in prodromal HD KI animals.
Limitations of Chapter 5: Unlike the other studies in this dissertation, the studies in
Chapter 5 were conducted only in sedentary HD KI and WT animals. The results of this study
provide evidence of different regions of the brain that have changes in rCBF, but this study
doesn’t assess the underlying mechanisms that link changes in rCBF and neuronal dysfunction or
compensation. Additionally, it is difficult to interpret some of the changes seen in this study,
such as increases in the somatosensory regions and cerebellum, as it is unknown whether these
increases come as a result of HD pathology or are in fact part of a compensatory response.
148
Further studies at earlier time points can help to identify which functional changes occur first and
which may come in response to other changes.
Within the scope of this dissertation, the most obvious follow up study would be to
examine changes in rCBF after long-term treadmill exercise. Determining the changes seen
between sedentary and exercised animals in both HD KI and WT groups could point to potential
regions where exercise may play a role in modifying disease progression. Currently, some very
preliminary findings from this exercise study suggest that long-term treadmill exercise may
attenuate some of the decreases in rCBF observed in the striatum in sedentary HD KI animals.
Further study into this and other brain areas will provide insight as to which brain circuits may
exhibit beneficial effects due to exercise and may begin to validate the predictive validity of this
approach.
149
Concluding Remarks
The studies presented in this dissertation contribute to our increasing knowledge and
understanding of the mechanisms by which experience-dependent neuroplasticity has beneficial
effects on delaying symptom onset and striatal pathology in the CAG
140
HD mouse model.
Collectively, the findings of the studies outlined in this dissertation suggest that long-term
treadmill running, beginning well before onset of motor symptoms, results in a delay of
depression-like behavior, cognitive impairments, and striatal pathology observed during the
prodromal period in CAG
140
mice. Long-term exercise was also sufficient to delay the onset of
motor symptoms. Additionally, long-term exercise restored the decreased striatal levels of
dopamine, norepinephrine, and serotonin observed in sedentary HD KI animals. The
downregulation of the D2 receptor observed at 4 and 6 months was rescued after long-term
treadmill running, suggesting a restoration of dopamine neurotransmission. Together these
results indicate a potential mechanism for symptomatic improvement observed in these animals.
Analysis of cerebral regional blood flow comparing sedentary HD KI animals and WT controls
has shown a functional reorganization of whole brain networks in the presymptomatic HD brain,
particularly in the striatum and prefrontal cortex, where significant hypoperfusion of basal
ganglia motor circuits occurs in CAG
140
mice. These results highlight the value of preclinical
research in HD mouse models especially in longitudinal studies on lifestyle modifications such
as exercise. Determining the mechanism(s) by which long-term treadmill running has its effect in
delaying symptom onset may provide potential pharmacological and genetic interventions
targeting neuroplasticity in the basal ganglia and prefrontal cortex resulting in delaying the onset
and progression of HD as well as to improving the quality of life of HD patients.
150
REFERENCES
Abbott RD, White LR, Ross GW, Masaki KH, Curb JD, Petrovitch H. 2004. Walking and
dementia in physically capable elderly men. JAMA 292: 1447-53.
Albin RL, Young AB, Penney JB. 1989. The functional anatomy of basal ganglia disorders.
Trends Neurosci 12:366-75.
Animals NRCUCftUotGftCaUoL. 2011. Guide for the Care and Use of Laboratory Animals.
Washington, DC: National Academies Press.
Antonini A, Leenders KL, Spiegel R, Meier D, Vontobel P, Weigell-Weber M, et al. 1996.
Striatal glucose metabolism and dopamine D2 receptor binding in asymptomatic gene
carriers and patients with Huntington’s disease. Brain 119: 2085–2095.
Araki KY, Sims JR, Bhide PG. 2007. Dopamine receptor mRNA and protein expression in the
mouse corpus striatum and cerebral cortex during pre- and postnatal development. Brain
research 1156:31-45.
Ariano MA, Aronin N, Difiglia M, Tagle DA, Sibley DR, Leavitt BR, et al. 2002. Striatal
neurochemical changes in transgenic models of Huntington’s disease. J Neurosci Res. 68:
716–729.
Aron AR, Poldrack RA. 2006. Cortical and subcortical contributions to Stop signal response
inhibition: role of the subthalamic nucleus. J Neurosci 26: 2424–2433.
Aron AR, Durston S, Eagle DM, Logan GD, Stinear CM, Stuphorn V. 2007. Converging
evidence for a fronto- basal-ganglia network for inhibitory control of action and
cognition. J Neurosci 27: 11860–11864.
Arrasate M, Mitra S, Schweitzer ES, Segal MR, Finkbeiner S. 2004. Inclusion body formation
reduces levels of mutant huntingtin and risk of neuronal death. Nature 431: 805-810.
Backman L. 1997. Cognitive deficits in Huntington’s disease are predicted by dopaminergic
PET markers and brain volumes. Brain 120:2207-17.
Backman L, Farde L. 2001. Dopamine and cognitive functioning: brain imaging findings in
Huntington’s disease and normal aging. Scand J Psychol 42: 287–296.
Bates GP. 2003. Huntingtin aggregation and toxicity in Huntington’s disease. Lancet 361: 1642-
1644.
Beal MF, Ferrante RJ, Swartz KJ, and Kowall NW, 1991. Chronic quinolinic acid lesions in rats
closely resemble Huntington’s disease. J Neurosci 11: 1649–1659.
151
Beaulieu JM, Gainetdinov RR. 2011. The physiology, signaling, and pharmacology of dopamine
receptors. Pharmacological reviews 63:182-217.
Becher MW, Kotzuk JA, Sharp AH, Davies SW, Bates GP, Price DL, Ross CA. 1998.
Intranuclear neuronal inlusions in Huntington’s disease and dentatorubal and
pallidoluysian atrophy: correlation between the density of inclusions and IT15 CAG
triplet repeat length. Neurobiol Disease 4: 387-397.
Bedard C, Wallman MJ, Pourcher E, Gould PV, Parent A, Parent M. 2011. Serotonin and
dopamine striatal innervation in Parkinson's disease and Huntington's chorea.
Parkinsonism Relat Disord 17: 593-598.
Beglinger LJ, O’Rourke JJF, Wang C, Langbehn DR, Duff K, Paulsen JS. 2010. Earliest
functional declines in Huntington disease. Psychiatry Res 178:414–81.
Benn CL, Landles C, Li H, Strand AD, Woodman B, Sathasivam K, Li SH, Ghazi-Noori S,
Hockly E, Faruque SMNN, Cha JHJ, Sharpe PT, Olson JM, Li XJ, Bates GP. 2005.
Contribution of nuclear and extranuclear polyQ to neurological phenotypes in mouse
models of Huntington’s disease. Human Mol Gen 14: 3065-3078.
Bernheimer H, Birkmayer W, Hornykiewicz O, Jellinger K, Seitelberger F. 1973. Brain
dopamine and the syndromes of Parkinson and Huntington. Clinical, morphological and
neurochemical correlations. J Neurol Sci 20: 415–455.
Beste C, Schuttke A, Pfleiderer B, Saft C. 2011. Music perception and movement deterioration
in Huntington's disease. PLoS Curr 3: RRN1252.
Beste C, Willemssen R, Saft C, Falkenstein M. 2010. Response inhibition subprocesses and
dopaminergic pathways:basal ganglia disease effects. Neuropsychologia 48: 366–373.
Bibb JA, Yan Z, Svenningsson P, Snyder GL, Pieribone VA, Horiuchi A, et al. 2000. Severe
deficiencies in dopamine signaling in presymptomatic Huntington’s disease mice. Proc
Natl Acad Sci USA 97: 6809–6814.
Bird ED. 1980. Chemical pathology of Huntington’s disease. Annu Rev Pharmacol Toxicol 20:
533–551.
Blandini F, Nappi G, Tassorelli C, Martignoni E. 2000. Functional changes of the basal ganglia
circuitry in Parkinson's disease. Prog Neurobiol 62:63-88.
Borlongan CV,Koutouzis TK, Freeman TB, Cahill DW, Sanberg PR, 1995. Behavioral
pathology induced by repeated systemic injections of 3-nitropropionic acid mimics the
motoric symptoms of Huntington's disease. Brain Res 697: 254-257.
Brichta L, Greengard P, Flajolet M. 2013. Advances in the pharmacological treatment of
Parkinson's disease: targeting neurotransmitter systems. Trends in Neurosci 36:543-554.
152
Brooks SP, Jones L, Dunnett SB, 2012. Comparative analysis of pathology and behavioural
phenotypes in mouse models of Huntington’s disease. Brain Res Bull 88: 81-93.
Callahan JW, Abercrombie ED. 2011. In vivo dopamine efflux is decreased in striatum of both
fragment (R6/2) and full- length (YAC128) transgenicmouse models of Huntington’s
disease. Front Syst Neurosci 5:61.
Carter RJ, Lione LA, Humby T, Mangiarini L, Mahal A, Bates GP, Dunnett SB, Morton AJ.
1999. Characterization of progressive motor deficits in mice transgenic for the human
Huntington's disease mutation. J Neurosci 19: 3248-57.
Cepeda C, Cummings DM, Hicey MA, et al. 2010. Rescuing the corticostrital synaptic
disconnection in the R6/2 mouse model of Huntington’s disease: exercise, adenosine
receptors and ampakines. PLoS Curr 2: pii: RRN1182.
Cepeda-Prado E, Popp S, Khan U, Stefanov D, Rodriguez J, et al. 2012. R6/2 Huntington's
Disease Mice Develop Early and Progressive Abnormal Brain Metabolism and Seizures.
J Neurosci 32:6456-67.
Cerovic M, d'Isa R, Tonini R, Brambilla R. 2013. Molecular and cellular mechanisms of
dopamine-mediated behavioral plasticity in the striatum. Neurobiology of learning and
memory 105:63-80.
Cha JH, Kosinski CM, Kerner JA, Alsdorf SA, Mangiarini L, Davies SW, et al. 1998. Altered
brain neurotransmitter receptors in transgenic mice expressing a portion of an abnormal
human Huntington disease gene. Proc Natl Acad Sci USA 95: 6480–6485.
Chai Y, Koppenhafer SL, Bonini NM, Paulson HL. 1999. Analysis of the role of heat shock
protein (Hsp) molecular chaperones in polyglutamine disease. J Neurosci 19: 10338-
10347.
Chen H, Zhang SM, Schwarzschild MA, Hernan MA, Aherio A. 2005. Physical activity and the
risk of Parkinson disease. Neurol 64: 664-9.
Chen JY, Wang EA, Cepeda C, Levine MS. 2013. Dopamine imbalance in Huntington’s disease:
a mechanism for the lack of behavioral flexibility. Front Neurosci 4: 114.
Cools R, Ivry RB, D’Esposito M. 2006. The human striatum is necessary for responding to
changes in stimulus relevance. J Cogn Neurosci 18: 1973–1983.
Corcos DM, Robichaud JA, David FJ, Leurgans SE, Vaillancourt DE, Poon C, et al. 2013. A
two-year randomized controlled trial of progressive resistance exercise for Parkinson’s
disease. Mov Disord 28: 1230-40.
Costa V and Scorrano L. 2012. Shaping the role of mitochondria in the pathogenesis of
Huntington's disease. EMBO Journal 31: 1853-1864.
153
Cummings JL. 1995. Behavioral and psychiatric symptoms associated with Huntington’s
disease. Adv Neurol 65:179-86.
Dang LC, Donde A, Madison C, O’Neil JP, Jagust WJ. 2012. Striatal dopamine influences the
default mode network to affect shifting between object features. J Cogn Neurosci 24:
1960–1970.
Daubner SC, Le T, Wang S. 2011. Tyrosine hydroxylase and regulation of dopamine synthesis.
Archives of biochemistry and biophysics 508:1-12.
Davies SW, Turmaine M, Cozens BA, DiFiglia M, Sharp AH, Ross CA, Scherzinger E, Wanker
EE, Mangiarini L, Bates GP. 1997. Formation of neuronal intranuclear inclusions
underlies the neurological dysfunction in mice transgenic for the HD mutation. Cell 90:
537-548.
Dorner JL, Miller BR, Barton SJ, Brock TJ, Rebec GV. 2007. Sex differences in behavior and
striatal ascorbate release in the 140 CAG knock-in mouse model of Huntington’s disease.
Behav Brain Res 178: 90-97.
Duff K, Beglinger LJ, Theriault D, Allison J, Paulsen JS. 2010. Cognitive deficits in
Huntington’s disease on the repeatable battery for the assessment of neuropsychological
status. J Clin Exp Neuropsychol. 32: 1–9.
Epping EA, Mills JA, Beglinger LJ, et al. 2013. Characterization of depression in prodromal
Huntington disease in the neurobiological predictors of HD (PREDICT-HD) study. J
Psychiatr Res. 47:1423–1431.
Fernagut PO, Diguet E, Labattu B, Tison F. 2002. A simple method to measure stride length as
an index of nigrostriatal dysfunction in mice. J Neurosci Methods 113: 123-30.
Fiedorowicz JG, Mills JA, Ruggle A, Langbehn D, Paulsen JS, et al. 2011. Suicidal behavior in
prodromal Huntington’s disease. Neurodegener Dis 8: 483-90.
Fisher BE, Petzinger GM, Nixon K, Hogg E, Bremmer S, Meshul CK, Jakowec MW. 2004.
Exercise-induced behavioral recovery and neuroplasticity in the 1-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine-lesioned mouse basal ganglia. J Neurosci Res 77: 378-390.
Fowler SC, Muma NA. 2015. Use of a force-sensing automated open field apparatus in a
longitudinal study of multiple behavioral deficits in CAG140 Huntington's disease model
mice. Behav Brain Res 294: 7-16.
Frank MJ, Seeberger LC, O’Reilly RC. 2004. By carrot or by stick: cognitive reinforcement
learning in parkinsonism. Science 306: 1940–1943.
Franklin KBJ, Paxinos G. 2007. The Mouse Brain in Stereotactic Coordinates. San Diego, CA:
Academic Press.
154
Frazzitta G, Balbi P, Maestri R, Bertotti G, Boveri N, Pezzoli G. 2013. The beneficial role of
intensive exercise Parkinson disease progression. Am J Phys Med Rehabil 92: 523-32.
Friston KJ, Holmes A, Worsley KJ, Poline JB, Frith CD, Frackowiak RS. 1995. Statistical
parametric maps in functional imaging: A general linear approach. Human Brain
Mapping 2: 189-210.
Furtado JCS and Mazurek MF, 1996. Behavioral characterization of quinolinate-induced lesions
of the medial striatum: Relevance for Huntington’s disease. Exptl Neurobiol 138: 158-
168.
Garrett MC, Soares-Da-Silva P. 1992. Increased cerebrospinal fluid dopamine and 3,4-
dihydroxyphenylacetic acid levels in Huntington’s disease: evidence for an overactive
dopaminergic brain transmission. J Neurochem 58: 101–106.
Georgiou-Karistianis N, Stout JC, Dominguez DJ, Carron SP, Ando A, Churchyard A, et al.
2014. Functional magnetic resonance imaging of working memory in Huntington's
disease: cross-sectional data from the IMAGE-HD study. Hum Brain Mapp 35: 1847-
1864.
Gerfen CR, Engber TM, Mahan LC, Susel Z, Chase TN, Monsma Jr. FJ. 1990. D1 and D2
Dopamine Receptor-Regulated Gene Expression of Striatonigral and Striatopallidal
Neurons Science 250:4986-4989.
Gerfen CR, Surmeier DJ. 2011. Modulation of striatal projection systems by dopamine. Annual
review of neuroscience 34:441-466.
Gil JM, Rego AC. 2008. Mechanisms of neurodegeneration in Huntington's disease. Eur J
Neurosci 27:2803-20.
Gil JM, Rego AC. 2009. The R6 lines of transgenic mice: a model for screening new therapies
for Huntington's disease. Brain Res Rev 59:410-31.
Ginovart N, Lundin A, Farde L, et al. 1997. PET study of the pre- and post-synaptic
dopaminergic markers for the neurodegenerative process in Huntington’s disease. Brain
120:503-14.
Giralt A, Saavedra A, Alberch J, Perez-Navarro E. 2012. Cognitive Dysfunction in Huntington's
Disease: Humans, Mouse Models and Molecular Mechanisms. Journal of Huntington's
disease 1: 155-173.
Gittis AH, Leventhal DK, Fensterheim BA, Pettibone JR, Berke JD, Kreitzer AC. 2011.
Selective inhibition of striatal fast-spiking interneurons causes dyskinesias. J Neurosci
31:15727-15731.
155
Gorton LM, Vuckovic MG, Vertelkina N, Petzinger GM, Jakowec MW, Wood RI. 2010.
Exercise effects on motor and affective behavior and catecholamine neurochemistry in
the MPTP-lesioned mouse. Behav Brain Res 213: 253-62.
Gusella JF, MacDonald ME. 2006. Huntington's disease: seeing the pathogenic process through a
genetic lens. Trends Biochem Sci 31: 533-540.
Hamilton JM, Salmon DP, Corey-Bloom J, et al. 2003. Behavioural abnormalities contribute to
functional decline in Huntington’s disease. J Neurol Neurosurg Psychiatry 74:120–122.
Harrison DJ, Busse M, Openshaw R, Rosser AE, Dunnet SB, Brooks SP. 2013. Exercise
attenuates neuropathology and has greater benefit on cognitive than motor deficits in the
R6/1 Huntington's disease mouse model. Exp Neurol 248: 457-69.
Hickey MA, Reynolds GP, Morton AJ. 2002. The role of dopamine in motor symptoms in the
R6/2 transgenic mouse model of Huntington's disease. J Neurochem 81:46-59.
Hickey MA, Chesselet MF. 2003. The use of transgenic and knock-in mice to study Huntington's
disease. Cytogenet Genome Res 100:276-86.
Hickey MA, Kosmalska A, Enayati J, Cohen R, Zeitlin S, et al. 2008. Extensive early motor and
non-motor behavioral deficits are followed by striatal neuronal loss in knock-in
Huntington's disease mice. Neuroscience 157:280-95.
Ho AK, Sahakian BJ, Brown RG, Barker RA, Hodges JR, Ané MN, et al. 2003. Profile of
cognitive progression in early Huntington's disease. Neurology 61:1702–1706.
Holschneider DP, Maarek JM, Harimoto J, Yang J, Scremin OU. 2002. An implantable bolus
infusion pump for use in freely moving, nontethered rats. Am J Physiol Heart Circ
Physiol 283: H1713-1719.
Huntington’s Disease Collaborative Research Group. 1993. A novel gene containing a
trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes.
Cell 72:971–983.
Imarsio S, Carmichael J, Korolchuk V, Chen CW, Saiki S, Rose C, Krishna G, Davies JE, Ttofi
E, Underwood BR, Rubinsztein DC. 2008. Huntington’s disease: from pathology and
genetics to potential therapies. Biochem J 412: 191-209.
Irwin I, Finnegan KT, DeLanney LE, Di Monte D, Langston JW. 1992. The relationship between
aging, monoamine oxidase, striatal dopamine and the effects of MPTP in C57BL/6 mice:
a critical reassessment. Brain Res 572: 224 –231.
Jaber M, Robinson SW, Missale C, Caron MG. 1996. Dopamine Receptors and Brain Function.
Neuropharmacology 35:1503-1519.
156
Jahanshahi A, Vlamings R, Kaya AH, Lim LW, Janssen ML, Tan, S, et al. 2010.
Hyperdopaminergic status in experimental Huntington disease. J Neuropathol Exp
Neurol 69: 910–917.
Jakowec MW, Nixon K, Hogg L, McNeill T, Petzinger GM. 2004. Tyrosine hydroxylase and
dopamine transporter expression following 1-methyl-4-phenyl-1,2,3,6-
tetrahydropyridine-induced neurodegeneration in the mouse nigrostriatal pathway. J
Neurosci Res 76:539-550.
Johnson MA, Rajan V, Miller CE, Wightman RM. 2006. Dopamine release is severely
compromised in the R6/2 mouse model of Huntington’s disease. J Neurochem 97: 737-
46.
Joyce JN, Lexow N, Bird E, Winokur A. 1988. Organization of dopamine D1 and D2 receptors
in human striatum: receptor autoradiographic studies in Huntington’s disease and
schizophrenia. Synapse 2: 546–557.
Julien C. L., Thompson J. C., Wild S., Yardumian P., Snowden J. S., Turner G., et al. 2007.
Psychiatric disorders in preclinical Huntington's disease. J Neurol Neurosurg Psychiatry
78: 939–943.
Kebabian JW, Calne DB. 1979. Multiple receptors for dopamine. Nature 277: 93-96.
Kelly CM, Dunnett SB, Rosser AE. 2009. Medium spiny neurons for transplantation in
Huntington’s disease. Biochem Soc Trans 37: 323-8.
Kilpatrick IC, Jones MW, Phillipson OT. 1986. A semiautomated analysis method for
catecholamines, indoleamines, and some prominent metabolites in microdissected regions
of the nervous system: an isocratic HPLC technique employing coulometric detection and
minimal sample preparation. J Neurochem 46: 1865–1876.
Kim J, Moody JP, Edgerly CK, Bordiuk OL, Cormier K, Smith K, Beal MF, Ferrante RJ. 2010.
Mitochondrial loss, dysfunction and altered dynamics in Huntington’s disease. Human
Mol Gen 19: 3919-3935.
Kintz N, Petzinger GM, Akopian G, Ptasnik S, Williams C, Jakowec MW, Walsh JP. 2013.
Exercise modifies alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor
expression in striatopallidal neurons in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-
lesioned mouse. Journal of neuroscience research 91:1492-1507.
Kirkwood SC, Su JL, Conneally M, Forund T. 2001. Progression of symptoms in the early and
middle stages of Huntington disease. Arch Neurol 58: 273-278.
Kish SJ, Shannak K, Hornykiewicz O. 1987. Elevated serotonin and reduced dopamine in
subregionally divided Huntington’s disease striatum. Ann Neurol 22: 386–389.
157
Klawans HC, Paulson GW, Barbeau A. 1970. Predictive test for Huntington’s chorea. Lancet 2:
1185–1186.
Kohl Z, Kandasamy M, Winner B, Aigner R, Gross C, et al. 2007. Physical activity fails to
rescue hippocampal neurogenesis deficits in the R6/2 mouse model of Huntington's
disease. Brain Res 1155:24-33.
Kreitzer AC, Malenka RC. 2008. Striatal plasticity and basal ganglia circuit function. Neuron
60:543-554.
Kuemmerle S, Gutekunst CA, Klein AM, Li XJ, Li SH, Beal MF, Hersch SM, Ferrante RJ. 1999.
Huntingtin aggregates may not predict neuronal death in Huntington’s disease. Ann
Neurol 46: 842-849.
Lawrence, A.D., Sahakian, B.J., Hodges, J.R., Rosser, A.E., Lange, K.W., Robbins, T.W. 1996.
Executive and mnemonic functions in early Huntington’s disease. Brain 119: 1633–1645.
Leavitt BR, van Raamsdonk JM, Shehadeh J, Fernandes H, Murphy Z, Graham RK, Wellington
CL, Raymond LA, Hayden MR. 2006. Wild-type huntingtin protects neurons from
excitotoxicity. J Neurosci 96: 1121-1129.
Lee HG, Petersen RB, Zhu X, Honda K, Aliev G, Smith MA, Perry G. 2003. Will preventing
protein aggregates live up to its promise as prophylaxis against neurodegenerative
diseases? Brain Pathol 13: 630-638.
Lerner RP, Trejo Martinez LdelC, Zhu C, Chesselet MF, Hickey MA. 2012. Striatal atrophy and
dendritic alterations in a knock-in mouse model of Huntington’s disase. Brain Res Bull
87: 571-8.
Leucht S, Correll CU, Kahn RS. 2011. Research Directions in Schizophrenia Treatment:
Mechanisms of Action for Next-Generation Agents. (Medscape Education Psychiatry
and Mental Health http://img.medscape.com/article/750/960/Slide6.png, ed).
Lewandowski NM, Bordelon Y, Brickman AM, Angulo S, Khan U, Muraskin J, et al. 2013.
Regional vulnerability in Huntington's disease: fMRI-guided molecular analysis in
patients and a mouse model of disease. Neurobiol Dis 52: 84-93.
Li SH, Li XJ. 2004. Huntingtin-protein interactions and the pathogenesis of Huntington’s
disease. Trends Genet. 20:146–154.
Lidow MS, Koh PO, Arnsten AF. 2003. D1 dopamine receptors in the mouse prefrontal cortex:
Immunocytochemical and cognitive neuropharmacological analyses. Synapse 47:101-
108.
158
Mangiarini L, Sathasivam K, Seller M, Cozens B, Harper A, Hetherington C, et al. 1996. Exon 1
of the HD gene with an expanded CAG repeat is sufficient to cause a progressive
neurological phenotype in transgenic mice. Cell 87: 493–506.
Maroof DA, Gross AL, Brandt J. 2011. Modeling longitudinal change in motor and cognitive
processing speed in presymptomatic Huntington’s disease. J Clin Exptl Neuropsych 33:
901-909.
Menalled LB, Sison JD, Wu Y, Olivieri M, Li XJ, Li H, et al. 2002. Early motor dysfunction and
striosomal distribution of huntingtin microaggregates in Huntington's disease knock-in
mice. J Neurosci 22: 8266-8276.
Menalled LB, Sison JD, Dragatsis I, Zeitlin S, Chesselet MF. 2003. Time course of early motor
and neuropathological anomalies in a knock-in mouse model of Huntington's disease with
140 CAG repeats. J Comp Neurol 465:11-26.
Mink J.W. 1996. The basal ganglia: focused selection and inhibition of competing motor
programs. Prog Neurobiol 50: 381-425.
Miyoshi E, Wietzikoski S, Camplessei M, Silveira R, Takahashi RN, Da Cunha C. 2002.
Impaired learning in a spatial working memory version and in a cued version of the water
maze in rats with MPTP-induced mesencephalic dopaminergic lesions. Brain research
bulletin 58:41-47.
Moriguchi S, Yabuki Y, Fukunaga K. 2012. Reduced calcium/calmodulin-dependent protein
kinase II activity in the hippocampus is associated with impaired cognitive function in
MPTP-treated mice. Journal of neurochemistry 120:541-551.
Nehl C and Paulsen JS. 2004. Huntington Study Group: cognitive and psychiatric aspects of
Huntington disease contribute to functional capacity. J Nerv Ment Dis. 192: 72–4.
Nguyen PT, Holschneider DP, Maarek JM, Yang J, Mandelkern MA. 2004. Statistical parametric
mapping applied to an autoradiographic study of cerebral activation during treadmill
walking in rats. Neuroimage 23: 252-259.
Niccolini F, Politis M. 2014. Neuroimaging in Huntington's disease. World journal of radiology
6: 301-312.
Novak MJ, Warren JD, Henley SM, Draganski B, Frackowiak RS, Tabrizi SJ. 2012. Altered
brain mechanisms of emotion processing in pre-manifest Huntington's disease. Brain 135:
1165-1179.
Ortiz AN, Osterhaus GL, Lauderdale K, Mahoney L, Fowler SC, Von Horsten S, et al. 2012.
Motor function and dopamine release measurements in transgenic Huntington’s disease
model rats. Brain Res 1450: 148–156.
159
Owen AM, Roberts AC, Hodges JR, Summers BA, Polkey CE, Robbins TW. 1993. Contrasting
mechanisms of impaired attentional set-shifting in patients with frontal lobe damage or
Parkinson’s disease. Brain 116:1159–1175.
Pang TY, Stam NC, Nithianantharajah J, Howard ML, Hannan AJ. 2006. Differential effects of
voluntary physical exercise on behavioral and brain-derived neurotrophic factor
expression deficits in Huntington's disease transgenic mice. Neuroscience 141:569-84.
Pang TY, Du X, Zajac MS, Howard ML, Hannan AJ. 2009. Altered serotonin receptor
expression is associated with depression-related behavior in the R6/1 transgenic mouse
model of Huntington’s disease. Hum Mol Genet 18: 753-66.
Pang RD, Wang Z, Klosinski LP, Guo Y, Herman DH, Celikel T, et al. 2011. Mapping
Functional Brain Activation using [14C]-iodoantipyrine in Male Serotonin Transporter
Knockout Mice. PLoS ONE 6: e23869.
Paradiso S, Turner BM, Paulsen JS, Jorge R, Ponto LL, Robinson RG. 2008. Neural bases of
dysphoria in early Huntington's disease. Psychiatry Res 162: 73-87.
Pare CM. 1972. Clinical implications of monoamine oxidase inhibition. Adv Biochem
Psychopharmacol 5: 441-4.
Parievsky A, Cepeda C, Levine MS. 2012. Evidence from the R6/2 Mouse Model of
Huntington's Disease for Using Abnormal Brain Metabolism as a Biomarker for
Evaluating Therapeutic Approaches for Treatment. Future Neurol 7: 527-530.
Paulsen JS, Zimbelman JL, Hinton SC, Langbehn DR, Leveroni CL, Benjamin ML, et al. 2004.
fMRI biomarker of early neuronal dysfunction in presymptomatic Huntington's Disease.
AJNR Am J Neuroradiol 25: 1715-1721.
Pavese N, Politis M, Tai Y, et al. 2010. Cortical dopamine dysfunction in symptomatic and
premanifest Huntington’s disease gene carriers. Neurobiol Dis 37:356-61.
Petersen A, Puschban Z, Lotharius J, Nicniocaill B, Wiekop P, O’Connor WT, et al. 2002.
Evidence for dysfunction of the nigrostriatal pathway in the R6/1 line of transgenic
Huntington’s disease mice. Neurobiol Dis 11: 134–146.
Petzinger GM, Fisher B, Hogg E, Abernathy A, Arevalo P, Nixon K, et al. 2006. Behavioral
motor recovery in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-lesioned squirrel
monkey (Saimiri sciureus): changes in striatal dopamine and expression of tyrosine
hydroxylase and dopamine transporter proteins. J Neurosci Res 83: 332-347.
Petzinger GM, Fisher BE, Van Leeuwen JE, Vukovic M, Akopian G, Meshul CK, Holschneider
DP, Nacca A, Walsh JP, Jakowec MW. 2010. Enhancing neuroplasticity in the basal
ganglia: the role of exercise in Parkinson’s disease. Mov Disord 25 Suppl 1: S141-5.
160
Petzinger GM, Walsh JP, Akopian G, Hogg E, Abernathy A, Arevalo P, Turnquist P, Vuckovic
M, Fisher BE, Togasaki DM, Jakowec MW. 2007. Effects of treadmill exercise on
dopaminergic transmission in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-lesioned
mouse model of basal ganglia injury. The Journal of neuroscience: the official journal of
the Society for Neuroscience 27:5291-5300.
Petzinger GM, Fisher BE, McEwen S, Beeler JA, Walsh JP, Jakowec MW. 2013. Exercise-
enhanced neuroplasticity targeting motor and cognitive circuitry in Parkinson's disease.
The Lancet Neurology 12:716-726.
Pillon B, Dubois B, Ploska A, Agid Y. 1991. Severity and specificity of cognitive impairment in
Alzheimer’s, Huntington’s, and Parkinson’s diseases and progressive supranuclear palsy.
Neurology 41:634-643.
Pisa M, Cyr J. 1990. Regionally selective roles of the rat’s striatum in modality-specific
discrimination learning and forelimb reaching. Behav Brain Res 37: 281–292.
Pitkala KH, Poysti MM, Laakkonen ML, Tilvis RS, Savikko M, et al. 2013. Effects of the
Finninsh Alzheimer Disease Exercise Trial (FINALEX): A Randomized Controlled Trial.
JAMA internal medicine 173: 894-901.
Porsolt RD, Bertin A, Jalfre M. 1977. Behavioral despair in mice: a primary screening test for
antidepressants. Arch Int Pharmacodyn Ther 229: 327-36.
Potter MC, Yuan C, Ottenritter C, Mughal M, van Praag H. 2010. Exercise is not beneficial and
may accelerate symptom onset in a mouse model of Huntington’s disease. PLoS Curr 2:
RRN1201.
Pouladi MA, Stanek LM, Xie Y, Franciosi S, Southwell AL, Deng Y, et al. 2012. Marked
differences in neurochemistry and aggregates despite similar behavioural and
neuropathological features of Huntington disease in the full-length BACHD and YAC128
mice. Hum Mol Genet 21: 2219–2232.
Pouladi MA, Morton AJ, Hayden MR. 2013. Choosing an animal model for the study of
Huntington's disease. Nat Rev Neurosci 14: 708-721.
Ramaswamy S, McBride JL, Kordower JH. 2007. Animal models of Huntington's disease. ILAR
J 48: 356-73.
Rangel-Barajas C, Coronel I, Florán B. 2015. Dopamine receptors and neurodegeneration. Aging
Dis 6: 349-68.
Rao AK, Muratori L, Louis ED, Moskowitz CB, Marder KS. 2009. Clinical measure of mobility
and balance impairments in Huntington’s disease: Validity and responsiveness. Gait &
Posture 29: 433-436.
161
Redgrave P, Rodriguez M, Smith Y, Rodriguez-Oroz MC, Lehericy S, Bergman H, Agid Y,
DeLong MR, Obeso JA (2010) Goal-directed and habitual control in the basal ganglia:
implications for Parkinson's disease. Nature reviews Neuroscience 11:760-772.
Rees H, Roberts MH. 1993. The anterior pretectal nucleus: a proposed role in sensory
processing. Pain 53: 121-135.
Renoir T, Pang TY, Zajac MS, Chan G, Du X, Leang L, et al. 2012. Treatment of depressive-like
behaviour in Huntington’s disease mice by chronic sertraline and exercise. Br J
Pharmacol 165: 1375-89.
Richfield EK, O'Brien CF, Eskin T, Shoulson I. 1991. Heterogeneous dopamine receptor
changes in early and late Huntington's disease. Neurosci Lett, 132:121-126.
Rising AC, Xu J, Carlson A, Napoli VV, Denovan-Wright EM, Mandel RJ. 2011. Longitudinal
behavioral, cross-sectional transcriptional and histopathological characterization of a
knock-in mouse model of Huntington's disease with 140 CAG repeats. Exp Neurol
228:173-82.
Ross, CA and Poirier, MA. 2004. Protein aggregation and neurodegenerative disease. Nat. Med.
10(Suppl.): S10–S17.
Rozas G, López-Martín E, Guerra MJ, Labandeira-García JL. 1998. The overall rod performance
test in the MPTP-treated-mouse model of Parkinsonism. J Neurosci Methods 83: 165-75.
Rubinsztein DC and Carmichael J. 2003. Huntington’s disease: molecular basis of
neurodegeneration. Expert Reviews in Mol Med 5: 1-21.
Sakurada O, Kennedy C, Jehle J, Brown JD, Carbin GL, Sokoloff L. 1978. Measurement of local
cerebral blood flow with iodo [14C] antipyrine. Am J Physiol 234: H59-66.
Saudou F, Finkbeiner S, Devys D, Greenberg ME. 1998. Huntingtin acts in the nucleus to induce
apoptosis but death does not correlate with the formation of intranuclear inclusions. Cell
95: 55-66.
Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T et al. 2012. Fiji: an
open-source platform for biological-image analysis. Nature methods 9: 676-682.
Schwab LC, Garas SN, Drouin-Ouellet J, Mason SL, Stott SR, Barker RA. 2015. Dopamine and
Huntington’s disease. Expert Rev Neurother 15: 445-58.
Smith Y, Villalba R. 2008. Striatal and extrastriatal dopamine in the basal ganglia: an overview
of its anatomical organization in normal and Parkinsonian brains. Movement disorders :
official journal of the Movement Disorder Society 23 Suppl 3:S534-547.
162
Snowden JS, Craufurd D, Thompson JC, Neary D. 2002. Psychomotor, execute, and memory
formation in preclinical Huntington disease. J Clin Exptl Neuropsych 24: 133-145.
Snowden JS, Austin NA, Sembi S, Thompson JC, Craufurd D, Neary D. 2008. Emotion
recognition in Huntington’s disease and frontotemporal dementia. Neuropsychologia. 46:
2638–49.
Speelman AD, van de Warrenburg BP, van Nimwegen M, Petzinger GM, Munneke M, Bloem
BR. 2011. How might physical activity benefit patients with Parkinson disease? Nat Rev
Neurol 7: 528-34.
Spokes EG. 1980. Neurochemical alterations in Huntington’s chorea: a study of post-mortem
brain tissue. Brain 103: 179–210.
Stefanko DP, Yamasaki WK, Lee P, Garcia DN, Petzinger GM, Jakowec MW. 2016a. Treadmill
running delays the onset of depression-like behavior and striatal pathology in the CAG
140
knock-in mouse model of Huntington’s disease. Neurobio Dis. In Press.
Stefanko DP, Shah VD, Tran AL, Gasanova Z, Petzinger GM, Jakowec MW. 2016b. Treadmill
running delays the onset of cognitive dysfunction and restores dopamine
neurotransmission in the CAG
140
mouse model of Huntington’s disease. Neurobio Dis. In
Press.
Steru L, Chermat R, Thierry B, Simon P. 1985. The tail suspension test: a new method for
screening antidepressants in mice. Psychopharmacology (Berl) 85: 367-70.
Stout JC, Paulsen JS, Queller S, Solomon AC, Whitlock KB, Campbell JC, Carlozzi N, Duff K,
Beglinger LJ, Langbehn DR, Johnson SA, Biglan KM, Aylward EH. 2011.
Neurocognitive signs in prodromal Huntington disease. Neuropsych 25: 1-14.
Sturrock A and Leavitt BR. 2010. The clinical and genetic features of Huntington disease. J
Geriatric Psych 23: 243-259.
Surmeier D.J., Ding J., Day M., et al. 2007. D1 and D2 dopamine-receptor modulation of striatal
glutamatergic signaling in striatal medium spiny neurons. Trends Neurosci 30:228-35.
Suzuki M, Desmond TJ, Albin RL, Frey, KA. 2001. Vesicular neurotransmitter transporters in
Huntington’s disease: initial observations and comparison with traditional synaptic
markers. Synapse 41: 329–336.
Tabrizi S. J., Scahill R. I., Owen G., Durr A., Leavitt B. R., Roos R. A., et al. 2013. Predictors of
phenotypic progression and disease onset in premanifest and early-stage Huntington's
disease in the TRACK-HD study: analysis of 36-month observational data. Lancet Neurol
4422: 1–13.
163
Thevenaz P, Ruttimann UE, Unser M. 1998. A pyramid approach to subpixel registration based
on intensity. IEEE Trans Image Process 7: 27-41.
Thompson D, Martin L, Whistler JL. 2010. Altered ratio of D1 and D2 dopamine receptors in
mouse striatum is associated with behavioral sensitization to cocaine. PLoS One 5:
e11038.
Toy WA, Petzinger GM, Leyshon BJ, Akopian GK, Walsh JP, Hoffman MV, Vučković MG,
Jakowec MW. 2014. Treadmill exercise reverses dendritic spine loss in direct and
indirect striatal medium spiny neurons in the 1-methyl-4-phenyl-1,2,3,6-
tetrahydropyridine (MPTP) mouse model of Parkinson's disease. Neurobiol Dis 63: 201-
9.
van Dellen A, Cordery PM, Spires TL, Blakemore C, Hannan AJ. 2008. Wheel running from a
juvenile age delays onset of specific motor deficits but does not alter protein aggregate
density in a mouse model of Huntington's disease. BMC Neurosci 9:34.
VanLeeuwen JE, Petzinger GM, Walsh JP, Akopian GK, Vuckovic M, Jakowec MW. 2010.
Altered AMPA receptor expression with treadmill exercise in the 1-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine-lesioned mouse model of basal ganglia injury. J Neurosci Res
88:650-668.
van Oostrom JCH, Maguire RP, Verschuuren-Bemelmans CC, Veenma-van der Duin L, Pruim J,
Roos RAC, Leenders, KL. 2005. Striatal dopamine D2 receptors, metabolism, and
volume in preclinical Huntington disease. Neurol 65: 941-3.
van Oostrom JCH, Dekker M, Willemsen ATM, de Jong BM, Roos RAC, Leenders, KL. 2009.
Changes in striatal dopamine D2 receptor binding in pre-clinical Huntington’s disease.
Eur J Neurol 16: 226–231.
van Uitert RL, Levy DE. 1978. Regional brain blood flow in the conscious gerbil. Stroke 9: 67-
72.
Verghese J, Lipton RB, Katz MJ, Hall CB, Derby CA, Kuslansky G, et al. 2003. Leisure
activities and the risk of dementia in the elderly. N Engl J Med 348: 2508-16.
Venturelli M, Scarsini R, Schena F. 2011. Six-month waliing program changes cognitive and
ADL performance in patients with Alzheimer. Am J Alzheimers Dis Other Demen 26:
381-8.
Vuckovic MG, Li Q, Fisher B, Nacca A, Leahy RM, Walsh JP, Mukherjee J, Williams C,
Jakowec MW, Petzinger GM. 2010. Exercise elevates dopamine D2 receptor in a mouse
model of Parkinson's disease: in vivo imaging with [(1)(8)F]fallypride. Movement
disorders : official journal of the Movement Disorder Society 25:2777-2784.
164
Walker FO. 2007. Huntington's disease. Lancet 369: 218-228.
Wang Z, Myers KG, Guo Y, Ocampo MA, Pang RD, Jakowec MW, et al. 2013. Functional
reorganization of motor and limbic circuits after exercise training in a rat model of
bilateral parkinsonism. PLoS ONE 8: e80058.
Weeks RA, Piccini P, Harding AE, Brooks DJ. 1996. Striatal D1 and D2 dopamine receptor loss
in asymptomatic mutation carriers of Huntington’s disease. Ann Neurol 40: 49–54.
Weiss KR, Kimura Y, Lee WCM, Littleton T. 2012. Huntingtin aggregation kinetics and their
pathological role in a Drosophila Huntington’s disease model. Genetics 190: 581-600.
Whishaw IQ, Mittleman G, Bunch ST, Dunnett SB. 1987. Impairments in the acquisition,
retention, and selection of spatial navigation strategies after medial caudate-putamen
lesions in rats. Behav Brain Res 24: 125–138.
Wolf RC, Gron G, Sambataro F, Vasic N, Wolf ND, Thomann PA, et al. 2011. Magnetic
resonance perfusion imaging of resting-state cerebral blood flow in preclinical
Huntington's disease. J Cereb Blood Flow Metab 31: 1908-1918.
Wolf RC, Sambataro F, Vasic N, Baldas EM, Ratheiser I, Bernhard Landwehrmeyer G, et al.
2014. Visual system integrity and cognition in early Huntington's disease. Eur J Neurosci
40: 2417-2426.
Wolf RC, Vasic N, Schonfeldt-Lecuona C, Landwehrmeyer GB, Ecker D. 2007. Dorsolateral
prefrontal cortex dysfunction in presymptomatic Huntington's disease: evidence from
event-related fMRI. Brain 130: 2845-2857.
Wolfe KJ and Cyr DM, 2011. Amyloid in neurodegenerative disease: Friend or foe? Sem Cell &
Dev Biol 22: 476-481.
Wood NI, Glynn D, Morton AJ. 2011. "Brain training" improves cognitive performance and
survival in a transgenic mouse model of Huntington's disease. Neurobiol Dis 42: 427-37.
Yawata S, Yamaguchi T, Danjo T, Hikida T, Nakanishi S. 2012. Pathway-specific control of
reward learning and its flexibility via selective dopamine receptors in the nucleus
accumbens. PNAS 109:12764-9.
Yu F, Nelson NW, Savik K, Wyman JF, Dysken M, Bronas UG. 2011. Affecting cognition and
quality of life via aerobic exercise in Alzhemier’s disease. West J Nurs Res 35: 24-38.
Zhang X, Smith DL, Merlin AB, Engemann S, Russel DE, Roark M, Washington SL, Maxwell
MM, Marsh JL, Thompson LM, Wanker EE, Young AB, Housman DE, Bates GP,
Sherman MY, Kazantsev AG. 2005. A potent small molecule inhibits polyglutamine
aggregation in Huntington’s disease neurons and suppresses neurodegeneration in vivo.
Proc Natl Acad Sci 102: 892-897.
165
Zheng Z and Diamond MI. 2012. Huntington disease and the huntingtin protein. Prog Mol Biol
107: 189-214.
Zimbelman JL, Paulsen JS, Mikos A, Reynolds NC, Hoffmann RG, Rao SM. 2007. fMRI
detection of early neural dysfunction in preclinical Huntington's disease. J Int
Neuropsychol Soc 13: 758-769.
Abstract (if available)
Abstract
Huntington’s disease (HD) is an autosomal dominant neurodegenerative disorder caused by an excessive polyglutamine (CAG) expansion in the Huntingtin (Htt) gene resulting in a mutated form of the huntingtin (htt) protein (The Huntington's Disease Collaborative Research Group, 1993). HD is characterized by progressive decline in cognitive and motor functions with neuropsychiatric disturbances leading ultimately to premature death 10 to 15 years after onset of motor symptoms. The major pathological findings include severe degeneration of striatal medium spiny neurons (MSNs) and the cerebral cortex, particularly the prefrontal and frontal cortex (Gil and Rego, 2008). In the striatum, there is a preferential loss of the dopamine D2 receptor (DA-D2R)-containing MSNs that mediate the indirect pathway compared to direct pathway dopamine D1 receptor (DA-D1R)-containing MSNs. ❧ There is currently no cure for HD. One strategy for treatment predominately aims to attenuate symptoms after their onset and prolong patient quality of life. Elucidating the mechanisms that can modify disease prevention could identify potential therapeutic targets for improved treatments and could eventually lead to a cure. While long-term studies examining lifestyle factors such as exercise, diet, and cognitive engagement have shown neuroprotective benefits in degenerative disorders such as Parkinson’s disease (PD) and Alzheimer’s disease (AD), similar studies have not been carried out in HD (Chen et al., 2005
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The effect of treadmill running on dendritic spine density in two neurodegenerative disorders: 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) mouse model of Parkinson’s disease and CAG₁₄₀ kn...
PDF
Experience-dependent neuroplasticity of the dorsal striatum and prefrontal cortex in the MPTP-lesioned mouse model of Parkinson’s disease
PDF
Exploration and expansion of novel therapeutic strategies targeting brain metabolism in neurodegenerative diseases
Asset Metadata
Creator
Stefanko, Daniel Patrick
(author)
Core Title
Long-term treadmill exercise delays the onset of motor dysfunction, cognitive impairments, and mood disturbances in the CAG 140 mouse model of Huntington's disease via restoration of dopamine neu...
School
Keck School of Medicine
Degree
Doctor of Philosophy
Degree Program
Neuroscience
Publication Date
04/18/2016
Defense Date
03/24/2016
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
accelerating rotarod,behavior,brain mapping,caudate-putamen,cerebellar,cortex,Dopamine,Exercise,forced swim,gait analysis,huntingtin,Huntington's disease,neurodegeneration,neuroplasticity,OAI-PMH Harvest,Q140,tail suspension,t-maze
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Walsh, John (
committee chair
), Cadenas, Enrique (
committee member
), Jakowec, Michael (
committee member
), Petzinger, Giselle (
committee member
), Wood, Ruth (
committee member
)
Creator Email
dstefank@gmail.com,stefanko@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-230132
Unique identifier
UC11278237
Identifier
etd-StefankoDa-4272.pdf (filename),usctheses-c40-230132 (legacy record id)
Legacy Identifier
etd-StefankoDa-4272.pdf
Dmrecord
230132
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Stefanko, Daniel Patrick
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
accelerating rotarod
behavior
brain mapping
caudate-putamen
cerebellar
cortex
forced swim
gait analysis
huntingtin
Huntington's disease
neurodegeneration
neuroplasticity
Q140
tail suspension
t-maze