Translational Neuroscience: Toward New Therapies Edited By: Karoly Nikolich, Steven E. Hyman

This is a section of
doi:10.7551/mitpress/9780262029865.001.0001
Translational Neuroscience
Toward New Therapies
Edited by: Karoly Nikolich, Steven E. Hyman
Citation:
Translational Neuroscience: Toward New Therapies
Edited by: Karoly Nikolich, Steven E. Hyman
DOI: 10.7551/mitpress/9780262029865.001.0001
ISBN (electronic): 9780262329859
Publisher: The MIT Press
Published: 2015
Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

7
Neurodegenerative Diseases
What Is to Be Done?
David M. Holtzman, Karoly Nikolich,

Menelas N. Pangalos, Karl Broich, Nancy Y. Ip,
Edward H. Koo, Walter J. Koroshetz,
Eliezer Masliah, Lennart Mucke, Pierluigi Nicotera,
Mareike Schnaars, and Maria Grazia Spillantini
Abstract
Neurodegenerative diseases, including Alzheimer disease, Parkinson disease, and many

others, lead to significant morbidity and mortality. As medical care for other disorders
(e.g., cardiovascular disease and cancer) has improved and people are living longer,
neurodegenerative diseases have become more prevalent because age is a major risk
factor for most of them. There have been tremendous advances in our understanding
of the scientific underpinnings of neurodegenerative diseases over the last thirty years.
Nonetheless, with a few exceptions, very few effective treatments are available to delay
the onset or affect the course of these diseases. This Forum brought together leaders in
the field of neurodegeneration from different disciplines and tasked them with defining
areas in need of more attention—areas where focused work is needed to reveal a better
understanding of these disorders. A time frame of 5–20 years was the goal within which
to develop more effective diagnostics and treatments. This chapter identifies eight areas
to address:
1. Major specific pathologies and circuit dysfunction in neurodegenerative dis-

eases need to be pinpointed over the life span, and dysfunctional circuits re-
quire identification.
2. Utilization of genetically well-defined patient populations will likely offer a
better chance for therapeutic success.
3. Therapies affecting neurotransmitter systems and signaling pathways should be
further explored utilizing defined patient populations and disease-affected nodes.

94 D. M. Holtzman et al.
4. Better ways need to be developed to understand protein aggregation processes,

from the formation of misfolded proteins to the critical clearance pathways
that regulate their levels and toxicity, including understanding the mechanisms
which underlie protein aggregate spreading in the brain as this could lead to
novel therapeutics.
5. Better understanding is needed on the role of human apolipoprotein E (apoE),
lipoproteins, and lipid biology under normal conditions and in neurodegenera-
tive diseases.
6. To increase understanding of disease and facilitate drug/biological delivery,
more information on the blood-brain barrier, the neurovascular unit, and other
barriers separating CNS from non-CNS compartments is required.
7. The role of the innate immune system and other immune mechanisms that con-
tribute to progression of neurodegeneration must be better understood.
8. Regardless of the upstream processes, it may be possible to activate neuropro-
tective mechanisms using defined factors, signaling pathways, or via cell-based
methods.
With significant progress in each of these eight areas, substantial changes in the diagno-
sis and treatment of neurodegenerative diseases should be possible over the next twenty
years. Given the current cost of these diseases to society and the increase in their preva-
lence with no additional progress, a major worldwide effort should be made to address
these issues immediately, with the highest of priorities.
Introduction
Neurodegenerative diseases, including Alzheimer disease (AD), Parkinson

disease (PD), dementia with Lewy bodies, frontotemporal dementia (FTD),
amyotrophic lateral sclerosis (Xia et al. 2013), and Huntington disease (HD),
have devastating effects on the nervous system that lead to progressive cogni-
tive, behavioral, and motor dysfunction. With improvement in overall medical
care, the importance of these age-related diseases to society is rising. In the
United States and Europe, it is estimated that the cost of dementia care alone
is already equal to or greater than that of cardiovascular disease and cancer
(Hurd et al. 2013). The number of affected individuals and the cost of caring
for them are expected to triple over the next forty years in the absence of ef-
fective disease-modifying treatments. Thus, developing a better understanding
and treatment for these disorders has become paramount. While several useful
treatments exist to treat PD and multiple sclerosis, no treatments have been
shown to delay the onset of these diseases, and few have significantly affected
disease progression. In our discussions, several major areas emerged which we
believe are important to make progress over the next 5–20 years, if we are to
begin to decrease the incidence of these diseases and develop better treatments
for those affected. Eight specific topics are discussed in this chapter, which we
believe are critical in making progress.

Neurodegenerative Diseases: What Is to Be Done? 95
Need to Pinpoint Major Specific Pathologies and Circuit

Dysfunction in Neurodegenerative Diseases over the
Life Span and to Identify Dysfunctional Circuits
There is clear data that aggregation of specific proteins is a signature feature

of most neurodegenerative diseases (Caughey and Lansbury 2003): amyloid-β
(Aβ) and tau in AD; α-synuclein and Aβ in dementia with Lewy bodies;
α-synuclein in PD; tau in various forms of FTD, progressive supranuclear pal-
sy, corticobasal degeneration, and chronic traumatic encephalopathy; TDP-43
in amyotrophic lateral sclerosis (Xia et al. 2013); and huntingtin in HD. For
most of these disorders, however, there is strong data that the proteins found in
aggregates play major roles in disease pathogenesis. Importantly, in AD, PD,
and HD (and probably in the other diseases), these protein aggregates begin
to accumulate many years prior to the onset of symptoms—at least 15 years,
for example, in the case of Aβ and AD (Jack and Holtzman 2013). By the
time symptoms appear, there is already significant neuronal cell loss in certain
regions of the CNS as well as synaptic, axonal, and dendritic loss and dysfunc-
tion. These findings argue that if effective treatments are going to be devel-
oped, identification of pathology and regional brain dysfunction in vivo prior to
the onset of symptoms and signs of these diseases is critical to the initiation of
primary or secondary prevention. In addition, specific neurological syndromes
can result from different underlying pathologies. Thus, in individuals with dif-
ferent symptomatic stages of a disease process, identifying the actual underly-
ing pathology is critical to an accurate diagnosis, enrollment in clinical trials,
and ultimately treatment. To establish the exact pathological process that is
taking place, as well as staging of the disease, an assessment of what is occur-
ring in the CNS (e.g., by imaging, biofluid, and electrophysiological analysis)
needs to begin during the asymptomatic period (e.g., at middle age), to identify
preclinical disease, as well as once symptoms and signs emerge for both prog-
nosis and to monitor disease progression (Holtzman et al. 2011; Sperling et al.
2011a). In addition to genetic factors, aging plays a big role in disease risk and
thus needs to be accounted for in the assessment and interpretation of the bio-
markers for neurodegenerative diseases. As testing of asymptomatic individu-
als for the presence of disease becomes more widespread, in both research and
for clinical use, appropriate education and informed consent will be required.
In regard to the detection of protein aggregates, use of molecular imaging
has been very useful. There are now good ways to detect the presence of fibril-
lar Aβ by both imaging and by measuring cerebrospinal fluid Aβ42 (Klunk et
al. 2004; Fagan et al. 2006). In recent studies, tau aggregates are now detect-
able with molecular imaging (Maruyama et al. 2013; Xia et al. 2013). For
the other major aggregates that occur in these diseases, including α-synuclein,
huntingtin, and TDP-43, molecular imaging is not yet possible and should be
developed. Cerebrospinal fluid or blood tests for the presence of any of these

aggregates in the CNS are also not yet possible and would be very helpful. In
addition, for both molecular imaging and biofluid assessment, it would be ex-
tremely helpful to be able to detect oligomeric forms of these proteins, which
may be the forms that mediate toxicity (Benilova et al. 2012). There is strong
evidence that many of these proteins that aggregate form different types of ag-
gregates in different diseases. For example, aggregated tau has both different
ratios of 3 or 4 repeat isoforms as well as different structure in AD versus pro-
gressive supranuclear palsy (Mandelkow and Mandelkow 2012). For each of
the proteins that aggregate, development of methods to detect the specific type
of aggregate as well as the cell type within which they are present (glia versus
neurons) would help classify the specific molecular features that are present for
each proteinopathy. Other types of biomarkers that would be potentially very
helpful to assess and develop, to add to the information about protein aggre-
gates, involve the status of microRNA in biofluids, given recent data showing
their role in disease pathogenesis (Mendell and Olson 2012). It is also worth
determining if fluid biomarkers or the earliest cellular triggers precede protein
aggregation.
Important issues to consider in biomarker development include the necessity
to develop reproducible, reliable, and cost-effective tests. Before widespread
use, biomarkers need to be well validated in large populations in longitudinal
studies. As effective treatments emerge, the risk and cost-benefit ratio of the
test versus the treatment will emerge. The more invasive the test, the more
demand there will be for having effective treatments. In non-Mendelian neuro-
degenerative disorders, it will be necessary to develop sensitive measures for
population screening which can then be paired with more specific secondary
tests to identify persons for treatment. A measure that is not as costly (e.g., a
blood test) is certainly preferable, but there is already a precedent for more
invasive measures (e.g., colonoscopy for colon cancer screening); dementia is
certainly in a similar category in terms of burden of illness. Of course, a less
invasive test will increase acceptance and utilization. Along those lines, assess-
ments of plasma, serum, peripheral blood cells, and skin cells should be further
assessed with the more advanced “–omics” and other technologies. Finally, as
the cost of complete genome sequencing has reduced dramatically, the assess-
ment of each person’s genome in combination with the other imaging and bio-
fluid tests could prove very useful: the integration of genetic and biochemical
information from an individual offers greater predictive value.
In all the neurodegenerative diseases under discussion, specific regions and
circuits in the brain are selectively vulnerable. It is thus critical to both utilize
and expand upon current methods to detect synaptic and circuit dysfunction,
as well as to pick up changes that are occurring prior to, during, and after the
development of protein aggregation in the CNS, but still prior to the onset of
clinically detectable symptoms and signs of disease. Both structural and func-
tional connectivity magnetic resonance imaging combined with assessments
of brain metabolism with FDG-PET are showing changes in some of these

diseases in asymptomatic individuals (Jack and Holtzman 2013). It would thus

be very useful to optimize both of these methods, especially their quantitative
aspects. Data coming from the Human Connectome Project may provide new
methods for quantitative tract tracing, which would prove useful (Van Essen
et al. 2013). Using these and other imaging methods, it will be important to
determine if any changes occur in brain regions that precede protein aggrega-
tion. This will permit a better understanding of pathogenesis as well as early
detection of disease-related cause or dysfunction (e.g., precuneus/posterior
cingulate cortex in AD; Greicius et al. 2004).
In addition to current neuroimaging methodologies for assessing brain
function, assessment of synaptic and circuit function via other methods might
provide important new ways to get at pathogenesis and diagnosis. For ex-
ample, neuroimaging ligands that would permit detection of specific synaptic
markers would be very useful. This has been developed for the dopaminergic
system (Varrone and Halldin 2012) but not to similar degrees for other major
neurotransmitter systems or for proteins present in excitatory versus inhibitory
synapses. Other methods that would detect activity within specific neuronal
populations would also be very useful. Knowledge from methods currently uti-
lized in animals that allow for expression profiling of specific neuronal popula-
tions (e.g., BAC-TRAP; Heiman et al. 2008) may provide information to allow
assessment of specific neuronal populations in vivo. Electrophysiological test-
ing with qEEG during alert and sleep phases as well as other methods should
be further studied to determine how early circuit and synaptic changes might
be detectable. In many cases, current application of neuropsychological testing
in humans may not be able to detect some of the earliest functional changes.
Longitudinal cognitive assessments and cognitive tests in areas such as spa-
tial memory and attention in presymptomatic populations might permit earlier
changes to be detected (Sperling et al. 2011a).
To achieve the aspirational goal of disease prevention, a number of steps
should be taken in preparation for the testing of promising therapies. To dem-
onstrate prevention benefit, it is necessary to be able to identify clear, quan-
tifiable end points, often of conversion from presymptomatic to symptomatic
stages. It is also critical to be able to identify a cohort for study in whom there is
high probability of conversion within the time period of a study, generally 1–5
years. Natural history studies are therefore required ahead of a trial to construct
and validate end points and formulae that accurately predict expected time to
reach these end points in a specific population. The PREDICT and PHAROS
studies in Huntington disease provide a model for this exercise: imaging and
neuropsychological changes have been followed over time in individuals with
known cytosine-adenine-guanine expansion who are close to expected time
of conversion to symptomatic HD (Biglan et al. 2013). Given the time frames
needed to follow individuals in a prevention trial, the current patent lifetimes
are a major disincentive and these need to be revisited by policy makers.

Utilization of Genetically Well-Defined Patient Populations

May Increase Chance for Therapeutic Success
Once we are able to diagnose and assess neurodegenerative diseases better,

which populations should clinical trials focus on over the next 5–20 years?
The negative results obtained through the testing of drugs for the treatment
of neurodegenerative disorders from the general non-enriched population, in
which genetics or biomarkers have not been utilized for enrollment criteria,
has been both disappointing and discouraging. This continuing approach re-
flects the standard conventional practice of testing of neurological drugs and
market-driven decisions: to seek as broad an indication as possible, if suc-
cessful, and to position the drug in ways that differentiate it from competing
drugs. The recent testing of the anti-Aβ monoclonal bapineuzumab in apoE4
carriers and apoE4 noncarriers provided very useful information (Salloway et
al. 2014a). It was one of the first times that a potential Alzheimer treatment
was tested in genetically defined groups (by apoE status). If the treatment had
been successful in one but not both subgroups, this would have set the prec-
edent that a treatment is not intended for the affected population at large. While
still novel in the Alzheimer field, this practice follows the lead from oncology,
where the genetic makeup of a particular tumor can show predictable outcomes
to targeted therapies. The recent start of several secondary prevention trials in
individuals with various rare hereditary mutations in the APP and Presenilin
genes (Moulder et al. 2013; Ayutyanont et al. 2014) as well as in apoE4 homo-
zygotes continues this new trend to examine genetically defined populations to
test mechanism-based therapies in subjects who may either benefit most from
the treatment or offer insights not available from unselected populations.
Genetically well-defined populations can offer a clearer route to demon-
strate pathophysiological mechanisms or provide proof-of-concept for the
mechanism of drug action. For example, the recent positive response to either
tafimidis (Coelho et al. 2012) or diflunisal (Berk et al. 2013) in the rare famil-
ial amyloid polyneuropathy, due to transthyretin accumulation in peripheral
tissues, was a seminal success in a previously untreatable neurodegenerative
disease. This trial offers a number of instructive lessons:
1. Use of objective measurements (in this case, tests of nerve and muscle
function) can provide meaningful assessment of disease progression.
2. Treatment can be initiated after onset of disease and still demonstrate
a positive effect.
3. Early treatment demonstrated greater benefit than when treatment was
instituted later in disease course.
4. Results represent a dramatic proof-of-concept of the proposed underly-
ing disease mechanism (inhibition of aggregation of transthyretin).
This benefit may, in the future, extend to similar success in the more common
systemic amyloidosis, which does not have a genetic basis. Therefore, this

example suggests that targeting a “purer” population can provide more clear-
cut answers, albeit in a very select group of individuals. This idea could be
expanded to a much larger nongenetic disease-affected group of individuals.
Following the genetic enrichment lead, one can envision similar approaches
applied to genetic forms of other neurodegenerative diseases, such as with
the LRRK2 and GBA mutations associated with PD and mutations in various
genes in FTDs. In the latter, progranulin mutations are particularly interesting
as disease pathophysiology appears to be related to haploinsufficiency (Cenik
et al. 2012). Thus, one can readily envision boosting progranulin levels as a
therapeutic approach.
Genetically well-defined groups offer quick validation of surrogate mark-
ers that track disease stages or progression. They also provide confirmation of
validity of various end points used to assess treatment effects. Further, new
biomarkers could be discovered that represent changes in brain circuits. The
latter are particularly informative as they may represent new clinical measures
of disease symptomatology useful for quantification of treatment outcomes.
One shortcoming of this approach is the rarity of individuals carrying vari-
ous mutations. Mounting a successful biomarker or treatment study will re-
quire a consortium of national or international sites, a precedent established
by the Dominantly Inherited Alzheimer Network (DIAN) study (Bateman et
al. 2012; Moulder et al. 2013). In the DIAN study, a number of PSEN and
APP mutations were included, since any one mutation is exceptionally rare.
Fortunately, the different mutations all appear to induce perturbations in Aβ
metabolism, aggregation, or production as the underlying disease pathophysi-
ology. This may not be the case with other neurodegenerative diseases, such as
in the ALS-FTD spectrum, where mutations have been identified in different
genes that seemingly appear to affect very different cellular pathways (Perry
and Miller 2013).
In AD, in addition to apoE4 homozygotes, trisomy 21 (Down syndrome)
offers another genetically well-defined population to study Alzheimer patho-
physiology that arguably is a resource that has not been taken advantage of
fully. As these individuals invariably develop Alzheimer brain changes by the
fourth decade of life and dementia in later years, they offer the possibility
to correlate markers of disease initiation and progression as well as to pro-
vide another genetically defined group for treatment trials. Further, drugs af-
fecting other neurotransmitter pathways, such as pentylenetetrazol acting as a
GABA-A antagonist, may also provide cognitive benefit in these individuals,
perhaps distinct from an effect on dementia (Fernandez et al. 2007). Of course,
use of different instruments to assess cognition will likely be required, given
that all individuals suffer mental retardation to varying degrees at baseline.
Finally, many neurodegenerative disorders have both genetic and nongenet-
ic causes and are likely to share multiple disease mechanisms. Complicating
this is the tendency for the diseases to be age related. As such, comorbidities
are frequently present in the elderly, and these concurrent systemic disorders

complicate drug testing. These comorbid conditions can contribute to disease

risk, confound treatment outcomes (as these conditions can be nonresponsive
to the treatments) and, in so doing, contaminate the treatment pool or decrease
diagnostic accuracy. The use of genetically well-defined populations can over-
come some of these complicating factors. On the other hand, some of these
comorbid diseases (e.g., hypertension, cardiovascular disease, or diabetes) can
be effectively treated, and treatment may well reduce risk of, for example,
dementia or AD (Schrijvers et al. 2012). Thus, for multiple reasons, comorbidi-
ties should not be ignored.
Explore Therapies That Affect Neurotransmitter

Systems and Signaling Pathways through Defined
Patient Populations and Disease-Affected Nodes
Over the past few years, efforts have been directed toward the development
of strategies to address the pathogenic mechanisms underlying diseases such
as AD and PD. Strategies to prevent or remove accumulation of Aβ, tau, and
α-synuclein are in the early stages of testing. Utilizing these approaches, a
clear-cut success in humans has not yet been obtained, but newer trials of
agents directed against these targets have identified potentially more appropri-
ate populations of patients earlier in the course of disease, optimizing chances
for success. The reason that previous trials of potential disease-modifying
therapies targeting Aβ in individuals with mild to moderate dementia have not
yet been successful may be due to a fundamental problem with targeting Aβ
accumulation in patients with overt manifestations of disease—the interven-
tion may be already too late. However, there is reason to believe that delaying
disease is possible in individuals at early clinical stages of diseases such as AD
(mild cognitive impairment, very mild dementia) or in asymptomatic subjects
with preclinical AD. Progress has been made in targeting tau pathology in ani-
mal models (Boutajangout et al. 2011; Yanamandra et al. 2013), which may
be applicable to a variety of neurodegenerative disorders. Similar strategies
are thought possible for PD, and common approaches include antibodies to
prevent the spreading of protein aggregates in brain.
In addition to potential therapies that target molecules involved directly in
disease pathogenesis, several other possibilities address symptoms of disease
in AD, PD, and other disorders, and have the potential to delay disease pro-
gression and improve quality of life, even in patients where disease is already
advanced. Arguably, these can be considered “disease-modifying” therapies.
Delaying or ameliorating disease progression can produce substantial benefits.
Historically, symptomatic treatments that have been developed and widely uti-
lized include cholinergic/dopaminergic agents. While useful in AD, cholinergic
agents have only been moderately effective (Birks 2006). New formulations
and combinations with other therapeutics might increase effectiveness. Testing

early symptomatic treatments in patients without overt disease manifestations

might offer better chances for success. In particular, delaying the onset of dis-
ease by targeting early synaptic dysfunction might significantly delay other
major disease symptoms. Some of the major neurotransmitter systems in the
CNS that have not been fully explored therapeutically in neurodegenerative
diseases include agents to modify the glutaminergic, GABAergic, serotoner-
gic, and noradrenergic systems. In addition, development of better cholinergic
and dopaminergic agents is likely still possible. One potential example of a
novel type of neurotransmitter/excitability modulator comes from the use of
levetiracetam, a drug used for treatment of epilepsy. At lower doses, it has
shown cognitive benefits in mouse models of Aβ-amyloidosis that have epilep-
tiform features and brain network abnormalities (Sanchez et al. 2012). Here,
the difference between symptomatic and disease-modifying therapies, if simi-
lar effects are seen in humans, would only become apparent when neurode-
generative processes progress beyond the manifestations of the initial synaptic
dysfunction observed. Symptomatic treatments might also be directed to early
symptoms which are dependent or independent of disease, such as sleep, which
could optimize memory performance (Ju et al. 2014). Critical to a research
roadmap is careful attention to the problems that are most important to quality
of life in affected individuals and their caregivers. More effective means of
ameliorating symptoms is a critical area for research. Symptomatic trials could
be performed more quickly and potentially at lower cost. For instance in PD,
nonmotor symptoms have been identified as most troublesome in the modern
age of effective treatments for bradykinesia (Chaudhuri et al. 2006). Cognitive
impairment, gait disturbance, and dyskinesias rise to prominence. In AD and
HD, disorders of circadian rhythm and behavioral disorders frequently exhaust
a family’s ability to care for an affected individual at home. Pharmacologic or
nonpharmacologic means of addressing these concrete problems should be a
focus of research and could lead to success in the short term. It will also be im-
portant to determine whether responses to treatments are seen only in patients
with specific characteristics, so as not to miss effects.
Another major area for investigation that can offer a strong therapeutic pro-
spective is the modification of APOE. This area is largely underfunded and
understudied, despite it being recognized as the major genetic risk factor for
AD and other diseases (see below). Inflammation and the innate immune sys-
tem offer another symptomatic/disease-modifying treatment target in both AD
and PD. Inflammation appears to be involved in disease progression in AD,
PD, ALS, FTD, and HD at different disease stages. Identifying the window of
opportunity to intervene with drugs that reduce or prevent activation of infec-
tion-related inflammation or innate immunity could also prove fruitful. Finally,
treatments that would enhance or restore plasticity including transcranial mag-
netic stimulation, deep brain stimulation, or nonpharmacological interventions
(exercise, diet, sleep) have the potential to produce significant benefits and
should be further explored.

Some limitations to the development of new symptomatic or disease-mod-

ifying therapies include:
• Current design of clinical trials (which have so far been considering
individuals with overt disease over presymptomatic individuals with
high risk).
• Poor outcome measures (e.g., better tools to assess cognition).
• Lack of multiple biomarkers: Better and more molecular, pathogenesis-
linked biomarkers need to be identified and followed over time during,
before, and after disease onset to detect dynamic changes. To this end,
new useful biomarkers need to correlate with existing biomarkers re-
lated to disease progression.
• Funding agencies need to support areas of research which, although
not widely represented within the scientific community, offer strong
potential for therapies (e.g., APOE). Regulatory agencies need to be
more open to the acceptance of preventive trials and new functional
biomarkers.
Understand Protein Aggregation Processes and

Spreading Mechanisms to Spur Novel Therapeutics
In terms of ultimately developing disease-modifying therapies for neurodegen-

erative diseases, it is critical to develop a better understanding of the normal
metabolism as well as the pathophysiology of key proteins that form aggre-
gates and accumulate in the brain. Most proteins implicated in the pathogen-
esis of neurodegenerative disorders can exist in a dazzling array of different
assemblies and conformational states, and for most, if not all, it remains to be
determined which conformation has the greatest impact on neuronal function
and survival. To fill this important knowledge gap, there is an urgent need to
develop methods that can reliably quantitate distinct assemblies of pathogenic
proteins in brain tissues of humans and experimental animal models, ideally
through noninvasive means. Relevant protein species include soluble oligo-
mers, which have emerged as potentially critical but rather elusive culprits in
a number of diseases (Benilova et al. 2012). With such methods, it should be
possible to define not only how many of these forms of proteins are present but
also how much the most pathogenic subspecies have to be reduced to improve
significantly brain function at the synaptic, network, and clinical levels. This
information could have an important impact on the design and interpretation of
clinical trials. For example, none of the anti-Aβ trials completed to date have
shown whether treatment significantly lowered Aβ oligomer levels in relevant
brain regions, nor is it clear whether, and by how much, the levels of these as-
semblies would have to be lowered to improve synaptic function or slow down
neuronal loss.

Based on a large body of published data, it is likely that pathogenic protein

assemblies or conformations can trigger diverse molecular cascades (Hayden
and Teplow 2013). The most proximal downstream mechanisms may involve
both receptor-dependent and receptor-independent processes. This diversity
of protein species and downstream pathways makes it unlikely that targeting
any one of them by itself will have a major impact on the disease overall.
Blocking the production of the abnormal proteins, inhibiting their aggregation,
or enhancing their clearance may be more effective strategies. Far greater ef-
forts appear to have been invested to date in inhibiting production of specific
proteins, such as Aβ, as opposed to enhancing their clearance. Indeed, modula-
tion of clearance mechanisms clearly deserves greater attention as this may be
one of the main overall mechanisms by which aging leads to elevated risk of
protein aggregation. Some of the abnormalities in clearance pathways which
may give rise to disease and be targetable include degrading enzymes, chaper-
one-mediated clearance, autophagy, uptake into glial cells, better utilization of
interstitial fluid or cerebrospinal fluid flow pathways, and transfer across the
blood-brain barrier into the circulation. Determining which of these mecha-
nisms are most effective, most druggable, and most affected by disease could
result in the identification of novel entry points for therapeutic interventions.
Closely related to this quest is the issue of whether the spread of pathogenic
proteins, such as tau and α-synuclein, from one cell to another, might involve
unique processes that are amenable to specific therapeutic interventions. In cel-
lular and animal models, cell-to-cell spread appears to involve converting the
normal, non-aggregated form of a protein into an aggregated form with high
β-sheet content by interacting with an already seeded/aggregated form of the
same protein (Frost and Diamond 2010; Lee et al. 2014). In the case of tau and
α-synuclein, this process appears to move from cell to cell in what some scien-
tists call a prion-like manner. However, to what extent this type of spread via
template seeding contributes to the progression of neurodegenerative disorders
in both animals and humans remains uncertain and is currently under intense
investigation. One condition that may be a particularly good example to test
more extensively is chronic traumatic encephalopathy, where there appears to
be spreading of tau aggregates over many years between different brain re-
gions (Stein et al. 2014). There also is a need to understand the mechanisms by
which pathogenic proteins impair the function of brain cells and the networks
they form. In particular, in AD and dementia with Lewy bodies, there is strong
evidence that Aβ accumulation leads in some way to further accumulation and
spread of tauopathy in AD (Musiek and Holtzman 2012) and synuclein in de-
mentia with Lewy bodies (Masliah et al. 2001). The mechanisms (e.g., cel-
lular signaling pathways) that underlie these effects are unknown and if sorted
out, could lead to new ideas for treatment. From a treatment standpoint, it
would also be valuable to complement ongoing efforts to detect the accumula-
tion of abnormal proteins in the brain with intensified efforts to develop and

incorporate novel methods into clinical trials to detect the emergence, progres-
sion, and potential therapeutic reversal of synaptic and network dysfunction.
Understand the Role of ApoE, Lipoproteins, and Lipid Biology

under Normal Conditions and in Neurodegenerative Diseases
Of the common diseases in humans, one of the strongest genetic risk factors
is the link between APOE and AD. ApoE4 dose dependently increases risk
for AD and cerebral amyloid angiopathy whereas apoE2 protects against AD
(Strittmatter and Roses 1996; Kim et al. 2009). APOE genotype may also mod-
ulate risk for other CNS disorders, such as recovery after head injury, stroke,
and other forms of neurodegeneration (Mahley and Huang 2012; Wolf et al.
2013). Despite the strong effects of APOE genotype on AD, our understand-
ing of exactly how apoE has these effects is incomplete, and this has limited
development of ways to target apoE-related pathways. Despite the importance
of apoE in AD risk, there is a paucity of academic and industry-related efforts
and investment in this important topic. Significantly more effort and resources
need to be prioritized here, given the strong impact of this gene which appears
to be mediated by the protein product.
There are two general categories by which apoE proteins appear to modu-
late AD and neurodegeneration: Aβ-dependent and Aβ-independent. For Aβ-
dependent mechanisms, there is overwhelming evidence that apoE isoforms
influence whether and when Aβ aggregates and accumulates in the brain and
contributes to its toxicity (E4 promotes Aβ accumulation, E2 retards it relative
to E3) (Holtzman et al. 2012). The general mechanisms appear to be due to the
ability of apoE to interact directly with Aβ to influence its aggregation as well
the ability of apoE to influence soluble Aβ clearance. In terms of which recep-
tors modulate apoE-related Aβ clearance, this has not been entirely resolved.
However, there is strong genetic and biochemical evidence that increasing
apoE lipidation (Koldamova et al. 2010) as well as decreasing levels of apoE3
and apoE4 can decrease Aβ accumulation (Kim et al. 2011; Bien-Ly et al.
2012). ApoE lipidation may be targetable by liver X receptor and retinoid X
receptor agonists (Cramer et al. 2012), although side effects from this approach
would need to be overcome. Decreasing apoE levels can be accomplished by
increasing activity of apoE receptors, such as low-density lipoprotein recep-
tor (LDLR) and LDLR-related protein (LRP1) (Holtzman et al. 2012). There
may be ways to increase function of these receptors using a biological or small
molecule approach. Knocking down apoE3 and apoE4 via an antisense oligo-
nucleotide approach should also be considered. Experiments with anti-apoE
antibodies show promise in decreasing Aβ deposition (Kim et al. 2012) and
should be further explored and developed. Inhibiting the interaction between
apoE and Aβ is another target that has been attempted successfully in animal
models, using peptides that might be approachable with small molecules. In

addition to apoE’s effect on Aβ, it may also influence tauopathy via direct
interactions with tau or via an indirect mechanism. This needs further explora-
tion as a biological mechanism as it offers new ways to target this interaction.
In regard to other non-Aβ-related mechanisms, apoE has been shown in
various in vitro and animal models to have the following effects, which may
be relevant to neurodegeneration (Mahley and Huang 2012; Wolf et al. 2013):
• effects on neurite outgrowth and synaptic plasticity,
• formation of apoE fragments in neurons that can be toxic,
• effects on mitochondrial function,
• effects on neuroinflammation, and
• effects on reverse cholesterol transport and lipid scavenging.
Further exploration of these effects in vivo in different models of neurodegen-
eration and in different laboratories will be critical to determine how important
each mechanism is. Investigators at the Gladstone Institute (Chen et al. 2012)
suggest that small molecules act as “structure correctors” to convert an apoE4
structure close to that of apoE3 or apoE2. This should be further explored in
regard to the influence on both Aβ- and non-Aβ-dependent effects of apoE.
Two areas of apoE biology should be fruitful: further characterization of
apoE structure and neuroprotective effects of apoE2. Although the structure of
apoE has been determined, it was done in the nonlipidated state. Since almost
all apoE in vivo is lipidated, understanding apoE structure on high-density
lipoprotein-like lipoproteins, as it occurs in the brain, should assist in model-
ing its various interactions. ApoE2 is strongly protective against AD, yet our
understanding of the mechanism underlying this effect is poor. ApoE2 results
in lower Aβ deposition; increased expression of apoE2 in the brain (via a viral
vector approach) decreases Aβ deposition (Hudry et al. 2013; Dodart et al.
2005). Further experiments utilizing such an approach should be pursued in
animal studies and potentially taken into humans if successful.
To Facilitate Disease and Drug/Biological Delivery,

More Information Is Needed on the Blood-Brain
Barrier, Neurovascular Unit, and Other Barriers
Separating CNS from Non-CNS Compartments
In AD, PD, NeuroAIDS, and in other neurodegenerative disorders, there is

evidence of dysfunction of the neurovascular unit (NVU), which is composed
of neurons, astrocytes, pericytes, and endothelial cells that surround blood ves-
sels in the brain (Iadecola 2004). This dysfunction may contribute to CNS
dysfunction as well as accumulation of neurotoxic molecules. It has been sug-
gested that NVU abnormalities might contribute or even trigger the neuropa-
thology of AD and other diseases (Zlokovic 2008). The NVU plays a key role
in regulating the permeability of the blood-brain barrier, cerebral blood flow,

and the activity of neuronal circuits. Better understanding of how the NVU and
blood-brain barrier normally function as well as how alterations in the NVU
are involved in clearance of misfolded proteins, neuroinflammation, and other
aspects of degeneration will provide important new insights. In addition, the
brain’s intrinsic drainage pathway, the recently named gliolymphatic system
(Iliff et al. 2012), needs to be better understood both under normal and disease
conditions, in relation to all the issues just mentioned. How the neurons are in-
teracting with endothelial cells and astroglia at the NVU is very important and
a key area for further research. Clarifying the nature of the alterations in the
NVU in neurodegeneration is likely to be important in the elucidation of treat-
ments designed to target the NVU as well as in understanding how this alters
the traffic of drugs and biologics into and out of the CNS. The concept that the
blood-brain barrier is damaged in CNS disorders has been around for a while,
but whether this contributes to disease pathophysiology remains controver-
sial. Still, recent evidence has shown that AD changes affect pericytes and the
blood-brain barrier (Zlokovic 2008), because vascular amyloid is a known as-
sociation that leads to hemorrhage and potentially to impaired autoregulation.
In this context, there is a need for better tools to quantify blood-brain barrier
leakage in both animals and humans.
The traditional drug release systems that deliver drugs systemically fail to
transport biologics and drugs effectively into the brain. As most therapeutic
agents for neurodegenerative disorders need to likely reach the CNS to be ef-
fective, approaches that will selectively target the CNS more efficiently need
urgent development. Growth factors, neuroprotective and anti-inflammatory
peptides, antibodies directed against neurotoxic protein aggregates, and en-
zymes are some examples of biologics with potential application in the treat-
ment of neurodegenerative disorders. Major challenges when considering the
delivery of such biologics into the CNS are how to target them specifically to
the brain. This has the potential to reduce off-target effects by lowering the
drug dose as well as by other mechanisms. Trafficking of proteins and peptides
into the CNS involves the interactions of the macromolecules of interest with
receptors located on the luminal and/or abluminal surfaces of the brain endo-
thelial cells. Both endocytosis and transcytosis play key roles in the trafficking
of such macromolecules across the blood-brain barrier.
Multiple strategies deliver proteins of therapeutic interest into the CNS,
including the use of peptides or antibodies that bind to endothelial cell re-
ceptors as well as nanoparticles, liposomes, and other container-type of car-
riers and viral vectors (e.g., adeno-associated virus, rhabdovirus). All have
advantages and disadvantages. In the case of charged nanoparticles, ensuring
specificity has been a challenge. With viral vectors, questions have been raised
as to side effects and long-term toxicity. Overall, each strategy needs to be
categorized according to specificity, capacity, mechanism of action, and ap-
plication. The so-called “one-fit-for-all” approach is unlikely to work; thus
we recommend that each strategy be considered independently and specific

applications developed. We also recommend that combinations of strategies

to carry biologics into the CNS be considered. For example, nanoparticles can
be combined with selected peptides to improve specificity. Given the potential
complications with rabdoviruses and some nanoparticles, other strategies have
been considered, in particular those involving receptor-mediated delivery.
The main receptors that mediate endocytosis and transcytosis of antibodies,
peptides, and proteins across the blood-brain barrier include LDLR, LRP1,
transferrin receptor (TfR), Fc receptor (FcRn), as well as the insulin receptor
(IR) and insulin growth factor-1 (IGF-1) receptor. Compared to other strate-
gies, the use of targeted small peptides or antibodies to transfer macromol-
ecules into the CNS has the advantage of being minimally invasive. One of the
most advanced strategies involves the use of monoclonal antibodies against
TfR or against IR. An interesting application of this strategy for the potential
treatment of AD has been the development of a divalent antibody in which one
arm of the antibody targets TfR (to facilitate blood-brain barrier permeability)
while the other arm targets β-secretase-1 (BACE1), a protease involved in the
processing of amyloid precursor protein (APP) into Aβ (Atwal et al. 2011).
These types of antibodies have been shown to facilitate the blood-brain barrier
with great efficacy and to reduce Aβ levels in both mice and nonhuman pri-
mates. Antibodies against TfR have also been used to enhance the trafficking
of antibodies in which one arm targets Aβ. This approach has been tested in
APP/PS1 transgenic models that develop Aβ deposition and was more effec-
tive than using a standard anti-Aβ antibody in reducing Aβ levels (Niewoehner
et al. 2014). More recently, utilizing phage display, two single domain anti-
bodies, FC5 and FC44, were identified that display high permeability across
the blood-brain barrier (Haqqani et al. 2013). Both have been proposed for
development as vectors for brain delivery. Overall, utilizing this or analogous
approaches offers great hope that one can increase the entry of proteins such as
antibodies in the brain that will be able to bind to and target specific proteins
and cells in the CNS.
Another approach is to use selective targeted small peptides that bind lipo-
protein receptors present on the surface of endothelial cells in the blood-brain
barrier. For example, in recent years small peptides derived from apolipopro-
teins (B and E) that bind LDLR have been shown to have the capacity of
increasing peptides and antibody entry into the CNS. Through the fusion of
the LDLR-binding domain of proteins, such as apoB (38 amino acids) and
apoE (19 amino acids), to cargo proteins, therapeutic promise has been shown
in various neurodegeneration models for AD, PD, Gaucher, and Sly disease.
For example, by coupling the LDLR-binding domain of apoB to neprilysin
(a metalloprotease that degrades Aβ), it has been shown that the systemically
administered fusion protein crosses the blood-brain barrier and accumulates in
the neocortex and hippocampus, thus reducing Aβ levels in the brain as well as
behavioral deficits and neurodegeneration (Spencer et al. 2011).

Other molecular Trojan horses, which use a lipoprotein receptor, are pep-
tides derived from the apoE sequence, Angiopep-2, and receptor-associated
proteins that target LRP1 (Demeule et al. 2008). This lipoprotein receptor has
been shown to mediate the endocytosis of Aβ peptides across the blood-brain
barrier (Zlokovic 2008). Aprotinin, a pancreatic trypsin inhibitor that contains
the Kunitz protease inhibitor (KPI) sequence, is a ligand for LRP1 and can
cross the blood-brain barrier. Aligning the sequence of aprotinin with the KPI
domain, a family of peptides (named Angiopeps) was developed. Angiopep-2,
in combination with micelles, has been tested to increase the CNS penetration
of drugs such as the antifungal amphotericin B (Shao et al. 2010). The am-
photerocin B-incorporated, angiopep-2 modified micelles showed highest effi-
ciency in penetrating into the CNS. Another peptide has been tested as a recep-
tor-specific carrier, namely cross-reacting material 197 or CRM197, which is
a nontoxin mutant of diphtheria toxin. CRM197 increased pinocytotic vesicles
and vacuoles in brain microvascular endothelial cells and enhanced caveolin-1
protein expression in brain microvessels (Wang et al. 2010b). Regarding the
apoE strategy, peptides of this molecule have been fused to α-L-iduronidase
(IDUA), a lysosomal enzyme being evaluated for efficacy in a mouse model
of mucopolysaccharidosis type I (Miller et al. 2013b). After 5 months of treat-
ment, apoE-IDUA was found to have normalized brain glycosaminoglycan
and β-hexosaminidase in an mucopolysaccharidosis type I model (Wang et
al. 2013a). These strategies offer great promise as they are further developed.
Targeting receptor-mediated transport via receptors such as TfR has potential
side effects. For example, long-term use can theoretically trigger deficits in
iron transport. Thus, due to the potential side effects for chronic use, other re-
ceptors should be considered. For example, LRP1 may have a higher capacity,
better trafficking abilities, and be more efficient than TfR. Some controversy
has emerged over whether LRP1 is present in endothelial cells in the blood-
brain barrier; further investigation is thus needed.
Nonreceptor-mediated approaches are also being considered for targeting
the CNS. The use of the HIV protein Tat as a carrier has been tested alone
as well as in combination with liposomes to deliver antibodies, peptides, and
growth factors. In addition, approaches that involve nanoparticles and lipo-
somes are being tested to deliver small molecules into the CNS. In terms of
delivering biologics to the brain, it is important to consider how the biologic
will be delivered to the correct compartment, how specific brain regions can be
targeted, and how the trafficking of the biological can be measured.
It will be crucial to develop imaging tools that allow measurement of the
trafficking of the therapeutically administered bioactive proteins more effec-
tively. Some work has shown that selectively tagged (coded) methods using
RNA sequences can be used to target neuronal populations selectively. Once
a therapeutic bioactive protein is targeted to the correct brain region and cel-
lular population, the next problem is for the protein to find the correct cellular
compartments. Using molecules that target a receptor, such as LRP1, has a

unique advantage: once the protein is transcytosed, it is targeted to the lyso-

some, which may be useful for specific metabolic diseases. This approach has
also been used with antisense oligonucleotides tagged for delivery into the
CNS to treat neurodegenerative disorders.
Other questions to consider concern the advantages and disadvantages of
invasive versus noninvasive approaches. Is it preferable to deliver a biologi-
cal or medicinal therapy into one specific brain region site or diffusely into
the brain? For example, for some years, intranasal administration has been
considered a potential approach for getting certain compounds into the CNS
effectively. An intranasal trial of insulin in AD is currently in progress: the
advantage here is that it appears to influence solely the CNS and does not
change blood glucose levels. Is intranasal administration still a valid approach
for drug delivery? Alternatives include considering whether compounds can be
delivered into the brain more effectively in areas lacking a blood-brain barrier
(e.g., hypothalamus). It may also be possible to utilize cell-mediated delivery
of material into the brain via cells such as monocytes to deliver drugs and
biologics in the CNS. Finally, direct CNS delivery via the lumbar or ventricu-
lar cerebrospinal fluid as well as into the brain parenchyma via convection-
enhanced delivery under appropriate circumstances may be very effective and,
while more invasive, may still provide a good risk-benefit ratio. This approach
is being explored for gene delivery via adeno-associated virus (AAV) as well
as administration of antisense oligonucleotides (Miller et al. 2013b). Overall
the NVU, blood-brain barrier, and gliolymphatic systems are being actively
investigated under normal conditions and in the setting of neurodegeneration.
Clearly, more work needs to be done in this area.
Developing a Better Understanding of the Role of the

Innate Immune System and Other Immune Mechanisms
That Contribute to Progression of Neurodegeneration
The occurrence of microglial activation in AD and all of the neurodegenera-

tive diseases has been observed for many years. Over the course of these dis-
eases, microglia change their phenotype, retract their ramified processes, and
adopt an amoeboid cell shape. These morphological changes are accompanied
by the accumulation of microglia in areas of injury as well as in and around
AD amyloid-β plaques. Microglial activation is accompanied by secretion of
pro-inflammatory cytokines (Meyer-Luehmann et al. 2008; Prinz et al. 2011b).
However, questions remain as to whether and how the main innate immune
effector cells, microglia, contribute to disease progression. In addition to acti-
vated microglia and their potential role in neurodegenerative diseases, there is
a striking increase in certain complement proteins. This is interesting given the
emerging role that complement proteins such as C3 appear to play in synaptic

pruning during development (Schafer et al. 2012), suggesting that increased

complement in disease may contribute to synaptic damage. This area needs
further exploration as complement may prove to be an attractive target for
intervention.
Further highlighting the role of microglia, specifically in AD, are recent
genetic data. Genome-wide association studies as well as whole genome se-
quencing have led to the identification of additional confirmed genetic risk
factors for AD: among them are CD33 (Bertram et al. 2008; Hollingworth et al.
2011; Naj et al. 2011) and TREM-2 (Guerreiro et al. 2013; Jonsson et al. 2013).
Both receptors are specifically expressed by microglia in brain. These may of-
fer future specific insights into both mechanism and treatment opportunities.
There are a number of additional questions that will be important to
ask regarding the role of the innate immune system and microglia in
neurodegeneration:
• We need to understand the inefficient role of microglia in amyloid
clearing in AD. While antibodies to Aβ and other stimuli can enhance
microglial clearance of Aβ in vivo, can other methods be utilized to
activate this pathway?
• How do molecules released by microglia, such as cytokine and chemo-
kines, influence CNS function? These may be good therapeutic targets.
• What is the time course of inflammation? When is the right time to
intervene? Magnitude of the anti-inflammatory intervention as well as
the specific inflammatory target will be important to resolve.
• Recent evidence suggests the importance of microRNAs in modulating
microglial function. Can this be better understood to develop therapies?
In the healthy brain, microglia continuously scan their microenvironment for
pathogens and inflammatory stimuli by sending out membrane processes.
This surveillance activity is thought to support tissue homeostasis. Moreover,
microglia play a role in monitoring synaptic activity and assist in synaptic
pruning, perhaps via complement receptors which they express (Prinz et al.
2011b; Tremblay and Majewska 2011). Recent animal data show that the
chronic accumulation of Aβ within the brain drives inflammation. This effect
appears to be mediated by the activation of the microglial NALP3 inflamma-
some and the subsequent release of interleukin-1β that leads to the shift of
microglia toward a pro-inflammatory M1-type phenotype (Heneka et al. 2013).
In turn, Nalp knockout animals in the APP/PS1 model background show re-
duced interleukin-1β activation as well as enhanced Aβ clearance and a shift
of the microglial phenotype toward the anti-inflammatory M2-like phenotype
(Heneka et al. 2013). Moreover, it was shown that inflammation not only re-
sults in a gain of microglial function, the protective function that microglia ex-
ert in tissue is impaired in mouse models of amyloidosis (Krabbe et al. 2013).
These findings raise the following questions:

• Is it possible to shift microglia back toward a phagocytic, surveillance

M2-like phenotype that would improve neurodegeneration and post-
pone disease onset?
• If we understand the transition between phenotypes, what are further
trigger events that could be targeted?
• Is NALP3 a good target for intervention?
• How do we target microglia? Can nanoparticles be used to deliver
therapeutics?
Other important questions to be addressed in the future in this area include:
• In addition to microglial activation, astrocytes are activated in neurode-
generative diseases. What is the role of this activation? Do astrocytes
contribute to the pathology? Can this activation be harnessed in a pro-
tective fashion?
• Is there any role for an adaptive immune system in neurodegenerative
disease outside of multiple sclerosis? If so, how does this occur?
Since most of the mechanistic data on the effects of the immune system in
neurodegeneration has been generated through mouse models, it will be im-
portant to understand the differences in innate and adaptive immunity of mice
and humans.
Neuroprotective Mechanisms through Defined Factors,

Signaling Pathways, or Cell-Based Methods
Extensive dendritic atrophy and synapse loss correlates with the severity of
cognitive and memory impairment in neurodegenerative diseases. Neuronal
damage (e.g., the loss of synapses) is believed to be the pathophysiological
consequence of neurotoxic agents, such as oligomeric and other forms of Aβ,
tau, and α-synuclein in AD, tauopathies, and synucleinopathies. Since these
early deficits are likely reversible, strategies targeting the stabilization and po-
tential restoration of dendritic arbors and spines are expected to modify disease
progression. Based on current understanding of the pathways and regulatory
mechanisms that are disrupted by such neurotoxic agents, a variety of thera-
peutic approaches may activate neuroprotective mechanisms in the brain.
Neurotrophic factors (e.g., BDNF, NGF, GDNF, Neurturin, IGF1, and
BMP9) have been shown to be supportive of dying and injured/stressed neu-
rons in animals. Some of them are also effective in reducing Aβ and plaque
burden in animal models of AD. However, direct delivery of neurotrophic fac-
tors into the human brain has proved to be very challenging. To date, two
main approaches have been used: (a) the factors were infused into brain as
recombinant proteins and (b) viral vectors carrying the genes encoding the
trophic factors were directly injected into specific brain areas. Mixed results

have been obtained in human studies, and there is much to learn from these
past experiences if we are to improve future clinical trial design. For example,
GDNF was found to be beneficial for PD in a small, open-label study (5 pa-
tients). However, Amgen conducted a systematic study (34 patients) where no
clear activity was found (Lang et al. 2006). It is important to improve the tech-
nology for delivery of GDNF and other growth factors. Importantly, in these
and other trials, the patients treated had advanced clinical disease. Conducting
studies at earlier ages, when there is less cell death, would seem more likely
to yield benefits. More recently, gene therapy has been used to treat PD by
introducing the neurturin gene via a viral vector approach. A Phase II trial by
Ceregene did not show a statistically significant benefit; however, further work
with this type of approach seems warranted (Bartus et al. 2014). AAV-NGF is
also being tested in AD patients and appears to show some promise (Rafii et al.
2014). In addition to direct brain delivery, intranasal trophic factor administra-
tion has been tested for decades using NGF, BDNF, FGF-2, IGF-1, and insulin.
Outcome measures using nasal delivery of trophic factors has been modest and
acceptance limited and controversial. Overall, optimization of protein delivery,
better understanding of pharmacokinetics/pharmacodynamics, and beginning
trials early in disease course seem to be critical factors for future trials. Further,
implementation of larger trials with such approaches need to be justified by
careful Phase I results that show strong target engagement.
Attempts have also been made to design or identify small molecules from
screening that activate neurotrophic factor receptors, such as TrkB receptor, or
their signaling pathways (Obianyo and Ye 2013). For some small molecules,
efficacy data were not reproducible, calling for a need to use unified protocols.
While the ability of these small molecules to mimic the neurotrophic factors
remains to be validated, some have been orally effective and show promise as
a basis to develop novel therapeutics for neurodegeneration. Rigorous pharma-
cological profiling of these small molecules is warranted. In addition to clas-
sical growth factor signaling, other intracellular signaling pathways have the
potential to be exploited to prevent axonal, dendritic, and synaptic degenera-
tion. These include pathways involving Nmnat and the sirtuin pathways (Araki
et al. 2004). Further exploration of their role in preventing degeneration in the
CNS seems warranted.
Cellular transplantation offers an alternative to provide a source of growth
or other trophic factors as well as a potential cell replacement strategy. Further
experiments to sort out how iPS cells differentiated into neurons or glia might
best serve in this role should be explored. Use of neural precursor cells that can
differentiate into specific neural cell types, such as interneurons, may prove
useful in that they can migrate to the appropriate layers within regions such as
cortex as well as modulate regional neuronal activity.
The relevance of nonpharmacological manipulations (e.g., physical exer-
cise, sleep, cognitive activity, and specific lipids) to neuroprotection needs fur-
ther exploration and may be better understood as neuroprotective strategies.

Identification of genes/factors, such as apoE4 homozygotes, which allow high-

risk individuals to escape disease, might provide insight into new neuroprotec-
tive mechanisms (Jonsson et al. 2012). In addition, the use of iPS cells coupled
with genomic analysis of such subjects could provide a way to explore this area.

MIT Press Direct
© 2015 Massachusetts Institute of Technology and

the Frankfurt Institute for Advanced Studies
This volume is the result of the 17th Ernst Strüngmann Forum,
held March 16–21, 2014, in Frankfurt am Main, Germany.
Series Editor: J. Lupp
Assistant Editor: M. Turner
Photographs: U. Dettmar
Lektorat: BerlinScienceWorks
All rights reserved. No part of this book may be reproduced in any form
by electronic or mechanical means (including photocopying, recording,
or information storage and retrieval) without permission in writing from
the publisher.
MIT Press books may be purchased at special quantity discounts

for business or sales promotional use. For information, please email
special_sales@mitpress.mit.edu.
The book was set in TimesNewRoman and Arial.

Printed and bound in the United States of America.
Library of Congress Cataloging-in-Publication Data
Translational neuroscience : toward new therapies / edited by Karoly

Nikolich and Steven E. Hyman.
pages cm.— (Strüngmann forum reports)
Includes bibliographical references and index.
ISBN 978-0-262-02986-5 (hardcover : alk. paper)
1. Neuroscience—Research. 2. Nervous system—Diseases. 3. Drug devel-
opment. I. Nikolich, Karoly. II. Hyman, Steven E.
RC337.T729 2015
616.80072—dc23
2015012872
10 9 8 7 6 5 4 3 2 1

Translational Neuroscience: Toward New Therapies Edited By: Karoly Nikolich, Steven E. Hyman

Uploaded by

Copyright:

Available Formats

Translational Neuroscience: Toward New Therapies Edited By: Karoly Nikolich, Steven E. Hyman

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Translational Neuroscience: Toward New Therapies Edited By: Karoly Nikolich, Steven E. Hyman

Uploaded by

Copyright:

Available Formats

This is a section of

Edited by: Karoly Nikolich, Steven E. Hyman

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

David M. Holtzman, Karoly Nikolich,

Neurodegenerative diseases, including Alzheimer disease, Parkinson disease, and many

1. Major specific pathologies and circuit dysfunction in neurodegenerative dis-

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

4. Better ways need to be developed to understand protein aggregation processes,

Neurodegenerative diseases, including Alzheimer disease (AD), Parkinson

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Need to Pinpoint Major Specific Pathologies and Circuit

There is clear data that aggregation of specific proteins is a signature feature

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

diseases in asymptomatic individuals (Jack and Holtzman 2013). It would thus

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Utilization of Genetically Well-Defined Patient Populations

Once we are able to diagnose and assess neurodegenerative diseases better,

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

complicate drug testing. These comorbid conditions can contribute to disease

Explore Therapies That Affect Neurotransmitter

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

early symptomatic treatments in patients without overt disease manifestations

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Some limitations to the development of new symptomatic or disease-mod-

Understand Protein Aggregation Processes and

In terms of ultimately developing disease-modifying therapies for neurodegen-

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Based on a large body of published data, it is likely that pathogenic protein

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Understand the Role of ApoE, Lipoproteins, and Lipid Biology

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

To Facilitate Disease and Drug/Biological Delivery,

In AD, PD, NeuroAIDS, and in other neurodegenerative disorders, there is

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

applications developed. We also recommend that combinations of strategies

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

unique advantage: once the protein is transcytosed, it is targeted to the lyso-

Developing a Better Understanding of the Role of the

The occurrence of microglial activation in AD and all of the neurodegenera-

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

pruning during development (Schafer et al. 2012), suggesting that increased

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

• Is it possible to shift microglia back toward a phagocytic, surveillance

Neuroprotective Mechanisms through Defined Factors,

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

Identification of genes/factors, such as apoE4 homozygotes, which allow high-

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

© 2015 Massachusetts Institute of Technology and

MIT Press books may be purchased at special quantity discounts

The book was set in TimesNewRoman and Arial.

Library of Congress Cataloging-in-Publication Data

Translational neuroscience : toward new therapies / edited by Karoly

Downloaded from http://direct.mit.edu/books/oa-edited-volume/chapter-pdf/2121537/9780262329859_c000600.pdf by guest on 20 July 2023

You might also like