University of Southern California Dissertations and Theses

From risk mitochondrial and metabolic phenotype towards a precision medicine approach for Alzheimer's disease

(USC Thesis Other)

From risk mitochondrial and metabolic phenotype towards a precision medicine approach for Alzheimer's disease

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

FROM RISK MITOCHONDRIAL AND METABOLIC PHENOTYPE TOWARDS A
PRECISION MEDICINE APPROACH FOR ALZHEIMER’S DISEASE

by
Yiwei (Yvette) Wang

A Dissertation Presented to the
FACULTY OF THE USC SCHOOL OF PHARMACY
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CLINICAL AND EXPERIMENTAL THERAPEUTICS)

DECEMBER 2018

Copyright 2018 Yiwei Wang
i

DEDICATION

To over 1825 sun rises and sun sets that will never return,

To my parents for their love and support from across the Pacific Ocean,

To my love of this life for always being by my side,
regardless of the space and distance between us,

To my alma mater, my eternal pride, and Fight On!
ii

ACKNOWLEDGEMENTS

I thank my mentor Dr. Roberta Diaz Brinton for her support and guidance during my PhD study.
I also thank my dissertation committee chair Dr. Enrique Cadenas, committee members Dr.
Kathleen Rodgers, and Dr. Christian Pike for their insightful advice and suggestions.

For the clinical studies, I thank the clinical trial participants for their kind contribution to science.
I also thank Dr. Christine Solinsky for sharing her data on NSC respiration and proliferation.

I am also thankful to all of my lab mates, especially Aarti Mishra, Jennifer Mao, Dr. Shuhua
Chen, Dr. Fei Yin, and my friends Maira Soto and Sachin Jadhav, for their support and
assistance throughout this journey.

iii

ABBREVIATIONS INDEX
A, amyloid beta
AD, Alzheimer’s disease
APOE, apolipoprotein E
APP, amyloid precursor protein
Cybrid, cytoplasmic hybrid cell
E2, estradiol
ER, estrogen receptor
ERE, estrogen response element
ETC, electron transport chain
iPSC, induced pluripotent stem cell
MAM, mitochondrial associated membrane
MRI, magnetic resonance imaging
MPP, ER selective antagonist, 1,3-Bis(4-hydroxyphenyl)-4-methyl-5-[4-(2-
piperidinylethoxy)phenol]-1H-pyrazole dihydrochloride
mtDNA, mitochondrial DNA
NSC, neural stem cell
OXPHOS, oxidative phosphorylation
OVX, ovariectomy
PET, positron emission tomography
PHTPP, ER selective antagonist, 4-[2-Phenyl-5,7-bis(trifluoromethyl)pyrazolo[1,5-
a]pyrimidin-3-yl]phenol
TCA cycle, tricarboxylic acid cycle
iv

TABLE OF CONTENTS
Dedication ...................................................................................................................................................................... i
Acknowledgements ....................................................................................................................................................... ii
Abbreviations Index .................................................................................................................................................... iii
Table of Contents ........................................................................................................................................................ iv
Table of Figures ........................................................................................................................................................... vi
List of Tables ............................................................................................................................................................. viii
Abstract ......................................................................................................................................................................... 1
Chapter I Overview ............................................................................................................................................. 3
ALZHEIMER ’S DISEASE ...................................................................................................................................................... 3
PRODROMAL PHASE OF AD AND BRAIN GLUCOSE HYPOMETABOLISM ...................................................................................... 5
MITOCHONDRIA AS A KEY PLAYER IN THE ETIOLOGY OF ALZHEIMER ’S DISEASE ........................................................................... 6
ESTROGEN REGULATION OF GLUCOSE METABOLISM AND MITOCHONDRIAL BIOENERGETICS ......................................................... 8
ESTROGEN REPLACEMENT THERAPY AND COGNITIVE FUNCTION .............................................................................................. 9
ESTROGEN RECEPTORS ................................................................................................................................................... 12
MITOCHONDRIAL GENOME ............................................................................................................................................. 20
MITOCHONDRIAL GENETIC VARIANCE AND BIOENERGETICS .................................................................................................. 22
MITOCHONDRIAL GENETIC VARIANCE AND RISK OF LATE ONSET ALZHEIMER ’S DISEASE ............................................................. 25
APOE GENOTYPE AND RISK OF ALZHEIMER ’S DISEASE ......................................................................................................... 30
SEX DIFFERENCE AND RISK OF ALZHEIMER ’S DISEASE ........................................................................................................... 33
STUDY HYPOTHESIS: FROM MITOCHONDRIAL AND METABOLIC PHENOTYPE TOWARDS A PRECISION MEDICINE APPROACH FOR
ALZHEIMER ’S DISEASE .................................................................................................................................................... 35
Chapter II Dynamic metabolic aging of the brain during endocrinological and chronological aging....... 38
ABSTRACT .................................................................................................................................................................... 38
INTRODUCTION ............................................................................................................................................................. 38
MATERIALS AND METHODS............................................................................................................................................. 40
RESULTS ...................................................................................................................................................................... 45
DISCUSSION AND CONCLUSION ........................................................................................................................................ 55
ACKNOWLEDGEMENT .................................................................................................................................................... 59
Chapter III Cell Type and Estrogen Receptor Subtype Speicific Contribution to Estradiol Regulation of
Mitochondrial Gene Expression ............................................................................................................................... 60
ABSTRACT .................................................................................................................................................................... 60
INTRODUCTION ............................................................................................................................................................. 61
MATERIALS AND METHODS............................................................................................................................................. 63
RESULTS ...................................................................................................................................................................... 69
DISCUSSION AND CONCLUSION ........................................................................................................................................ 90
ACKNOWLEDGEMENT .................................................................................................................................................... 94
v

Chapter IV Targeting Mitochondrial Genetic Variances as a Precision Medicine Opportunity for
Alzheimer’s Disease Therapies ................................................................................................................................. 95
EFFECT OF MITOCHONDRIAL GENETIC VARIANCE AND APOE GENOTYPE ON THERAPEUTIC OUTCOMES OF PHYTOSERM ................ 95
Abstract................................................................................................................................................................ 95
Introduction ........................................................................................................................................................ 95
Materials and Methods ....................................................................................................................................... 98
Results ............................................................................................................................................................... 102
Discussion and Conclusion .............................................................................................................................. 108
Acknowledgement ............................................................................................................................................. 111
EFFECT OF MITOCHONDRIAL GENETIC VARIANCE, APOE GENOTYPE, AND SEX ON THERAPEUTIC OUTCOMES OF ALLOPREGNANOLONE
................................................................................................................................................................................ 112
Abstract.............................................................................................................................................................. 112
Introduction ...................................................................................................................................................... 113
Materials and Methods ..................................................................................................................................... 115
Results ............................................................................................................................................................... 122
Discussion and Conclusion .............................................................................................................................. 130
Acknowledgement ............................................................................................................................................. 135
Chapter V Discussion and Concluding Remarks ........................................................................................... 136
References ................................................................................................................................................................. 140

vi

TABLE OF FIGURES
Figure I-1. Human mitochondrial DNA. ........................................................................................................ 21
Figure I-2. Modified mitochondrial cascade of late-onset Alzheimer’s disease (LOAD). ................................ 25
Figure II-1. Mitochondrial gene expression during endocrinological and chronological aging. ....................... 46
Figure II-2. Gene expression of nuclear encoded ETC subunits during endocrinological and chronological
transition. ...................................................................................................................................................... 47
Figure II-3. Electron transport chain complex activity. ................................................................................... 48
Figure II-4.Change in metabolites during endocrinological and chronological aging. ..................................... 50
Figure II-5. Expression of genes involved in glycolysis, TCA cycle, ketogenesis, and fatty acid beta oxidation
throughout endocrinological and chronological aging. ................................................................................... 52
Figure II-6. Change in mitochondrial encoded OXPHOS gene expression due to ovariectomy. ...................... 53
Figure II-7. Change in ovarian hormone level due to ovariectomy.................................................................. 55
Figure III-1. effect of ovariectomy and estradiol replacement treatment on hippocampal mitochondrial gene
expression. .................................................................................................................................................... 70
Figure III-2. Change of estradiol level in serum and cortex from reproductively competent young rats to
reproductively competent middle-aged rats. ................................................................................................... 71
Figure III-3. Estrogen receptor subtype-specific effect on estradiol-mediated increased in mitochondrial
respiratory capacity in neurons. ..................................................................................................................... 72
Figure III-4. Estrogen receptor subtype-specific effect on estradiol-mediated increased in mitochondrial
respiratory capacity in astrocytes. .................................................................................................................. 74
Figure III-5. Effect of ER and ER on mitochondrial gene expression in response to estradiol in neurons. ... 76
Figure III-6. Effect of ER and ER on mitochondrial gene expression in response to estradiol in astrocytes. 77
Figure III-7. Estrogen receptor subtype-specific effect on estradiol-mediated increased in mitochondrial
respiratory capacity in neuroblastoma cells. ................................................................................................... 78
Figure III-8. In wildtype SH-SY5Y cells, estradiol treatment activated both ER and ER. ........................... 79
Figure III-9. Neuronal differentiation of SH-SY5Y resulted in upregulation of mitochondrial encoded
OXPHOS genes (highlighted in blue outline) and down-regulation of nuclear encoded OXPHOS genes. ....... 85
Figure III-10. Estrogen receptor subtype-specific effect on estradiol-mediated increased in mitochondrial
respiratory capacity in neuronal differentiated SH-SY5Y cells. ...................................................................... 86
Figure IV-1. Baseline hot flash frequency by demographic data. .................................................................. 103
Figure IV-2. Impact of different doses of PhytoSERM on hot flash frequencies. .......................................... 104
Figure IV-3. Initial hot flash frequency predicts therapeutic outcomes of PhytoSERM on hot flash frequency
reduction in the PS50 group only. ................................................................................................................ 105
Figure IV-4. Change in hot flash frequency from week 1 to week 12 in participants when stratified by
mitochondrial haplogroup. ........................................................................................................................... 105
Figure IV-5. Change in hot flash frequency stratified by APOE genotype. ................................................... 106
Figure IV-6. Change in participants’ ability to learn the RAVLT trial stratified by mitochondrial haplogroup.
................................................................................................................................................................... 107
Figure IV-7. Effect of PhytoSERM on cognitive flexibility. ......................................................................... 108
Figure IV-8. Identification of potential Alzheimer’s disease treatment responders based on three genetic risk
factors: mitochondrial genetic variances, APOE genotype, and chromosomal sex. ....................................... 115
vii

Figure IV-9. Summary of experiment scheme. ............................................................................................. 117
Figure IV-10. Change in maximal mitochondrial respiratory capacity following allopregnanolone treatment
stratified by sex and APOE genotype. .......................................................................................................... 123
Figure IV-11. Percent change in maximal mitochondrial respiratory capacity in all participants. .................. 124
Figure IV-12. Neural stem cell (NSC) proliferation stratified by sex and APOE genotype. NSCs with male
APOE3/3 genetic background had significantly higher proliferation rate. *p<0.05. ...................................... 125
Figure IV-13. NSC proliferation rate by mitochondrial haplogroup and APOE genotype.............................. 126
Figure IV-14. In vitro mitochondrial respiratory capacity is predictive of in patient allopregnanolone
therapeutic outcome. ................................................................................................................................... 134

viii

LIST OF TABLES
Table I-1. Mitochondrial haplogroups/superhaplogroups differentially associated with respiratory phenotypes.
..................................................................................................................................................................... 23
Table I-2. Observed effects of mitochondrial superhaplogroup HV and haplogroup H on risk of AD. ............ 27
Table I-3. Observed effects of mitochondrial superhaplogroup KU and haplogroups K and U on risk of AD. . 28
Table I-4. Observed effects of mitochondrial superhaplogroup JT and haplogroups J and T on risk of AD. .... 29
Table I-5. Association between some Asian mitochondrial haplogroups and the risk of Alzheimer’s disease. . 30
Table I-6. Three modes of interactions between APOE4 status and mitochondrial haplogroups in modulating
the risk of AD. .............................................................................................................................................. 33
Table I-7. Sex differentiates effects of mitochondrial haplogroups on risk of Alzheimer’s disease. ................. 35
Table II-8. List of statistically significantly altered metabolites during endocrinological and chronological
aging. ............................................................................................................................................................ 49
Table II-9. Upstream regulators predicted to be significantly activated or inhibited during endocrinological and
chronological aging. ...................................................................................................................................... 53
Table II-10. Change in nuclear encoded OXPHOS gene expression due to ovariectomy. ................................ 54
Table III-1. Upstream regulator activities differentially activated or inhibited by ER or ER selective
antagonists. ................................................................................................................................................... 83
Table III-2. Gene markers of immature neurons, mature neurons, and cholinergic neurons. ............................ 84
Table IV-1. Primers and locations to amplify mitochondrial DNA before mitochondrial haplotype sequencing.
................................................................................................................................................................... 100
Table IV-2. Primers and locations for PCR reactions required for dye-terminator sequencing. ..................... 100
Table IV-3. Participants by treatment groups and mitochondrial haplogroups. ............................................. 102
Table IV-4. Participants by treatment and APOE genotype. ......................................................................... 103
Table IV-5. Amplification and sequencing primers for mitochondrial HVR1 and HVR2. ............................. 119
Table IV-6. Participants by sex and APOE genotype. .................................................................................. 122
Table IV-7. Participants by mitochondrial haplogroups. ............................................................................... 122
Table IV-8. NSC cell lines used for RNA-Seq by sex and APOE genotype. ................................................. 126
Table IV-9. NSC cell lines used for RNA-Seq by APOE genotype and mitochondrial haplogroup. .............. 126
Table IV-10. Summary of sex difference in response to Allopregnanolone by RNA-Seq. ............................. 128
Table IV-11. Summary of effect of APOE genotype on response to Allopregnanolone by RNA-Seq. ........... 129
Table IV-12. Summary of effect of mitochondrial genetic variance on response to Allopregnanolone by RNA-
Seq. ............................................................................................................................................................. 130

1

ABSTRACT
Brain glucose hypometabolism and mitochondrial dysfunction is a key signature of late onset
Alzheimer’s disease (AD). Similar metabolic pattern was also observed in natural aging and during
the perimenopausal transition in the female brain, making age and chromosomal sex two risk
factors for late onset AD. Comprehensive understanding of the dynamic metabolic aging process
in the female brain can shed light on potential interventions and prevention windows for AD.
Further, mitochondrial bioenergetics is central to the brain metabolic aging, and estrogen was
demonstrated to be a master regulator of mitochondrial bioenergetics. While much is known about
the effect of estrogen on electron transport chain activities, the contribution of, and the
coordination between mitochondrial and nuclear genome in response to estrogen stimulation were
not as clear. The regulator effect of estrogen receptor subtypes in different cell types in the context
of brain aging and AD is also not clear. Moreover, the maternal inheritance pattern of both late
onset AD and mitochondrial genome suggest its key involvement in the etiology of AD, most
likely due to differential mitochondrial bioenergetic phenotypes associated with mitochondrial
genetic variances. Thus, a comprehensive understanding of the metabolic and respiratory profiles
at key transition points, the underlying mechanism, and key players involved can help identify
potential therapeutic targets and windows.
Using a rat model recapitulating the fundamental characteristics of human perimenopausal
transition, we generated a detailed road map of metabolic profile changes throughout different
stages of chronological and endocrinological aging. Using rat primary neurons, astrocytes, and
differentiated and wildtype human neuroblastoma cells mimicking neuronal and astrocytic
respiratory phenotypes, we demonstrated the cell type and estrogen receptor subtype specific
regulatory effect of estrogen on mitochondrial transcription and respiratory capacity. With the aid
2

of transcriptome analysis, we further identified roles of unliganded estrogen receptors in regulating
mitochondrial bioenergetics. And by conducting retrospective analysis on two clinical studies, we
demonstrated that for therapeutics targeting mitochondrial bioenergetics, three genetic risk factors
of late onset Alzheimer’s disease: mitochondrial genetic variances, APOE genotype, and
chromosomal sex can modulate their therapeutic outcomes or even be used to predict and
differentiate responders from non-responders.
Collectively, this dissertation study generated profiles of at risk bioenergetic phenotypes to help
better understand the mechanism leading to late onset AD in the female brain, provided deeper
mechanistic understanding of estrogen regulation of brain bioenergetics, and further demonstrated
a practical, genetic marker-based precision medicine approach to identify potential responders to
AD therapeutics. These outcomes can be further translated into strategies for predicting at-risk
phenotypes, and into guidance for therapeutics development as well as clinical trial design of future
neurodegenerative therapeutics.

3

CHAPTER I

OVERVIEW
Alzheimer’s Disease
Alzheimer’s disease (AD) is a devastating neurodegenerative disease, with one in three seniors
affected in the United States (Alzheimer's Association, 2018). Moreover, in the past decade, while
the death due to heart disease decreased by 11%, death due to Alzheimer’s disease increased by
over 120%, making it more deadly than breast cancer and prostate cancer combined (Alzheimer's
Association, 2018). Besides cognitive problems such as issues with memories, language, and
orientation, Alzheimer’s disease is also associated mood swings and behavior issues. Symptoms
gradually worsen overtime, accompanied by substantial and progressive brain atrophy in the
hippocampus, temporal lobe, parietal lobe, and the frontal lobe (Scahill et al., 2002). Decline in
physical, psychological, and cognitive functions continues until ultimate death.
However, Alzheimer’s disease is a systemic disease with multiple etiologies, and the underlying
mechanism is still not clearly understood (Morris et al., 2014). Clinical diagnosis of Alzheimer’s
disease is mostly based on neuropsychological tests that examine a wide array of cognitive
domains, and diagnosis can only be made for possible or probable AD (Dubois et al., 2007).
Clinical imaging such as magnetic resonance imaging (MRI) and positron emission tomography
(PET) can also be used to aide diagnosis and predict conversion from mild cognitive impairment
(MCI) to clinical AD (Catana et al., 2012; Schroeter et al., 2009). Nevertheless, at present, a
definite diagnose of Alzheimer’s disease can only be made post-mortem based on these two
pathological hallmarks—amyloid beta (A) plaques and hyperphosphorylated tau tangles.
Given the almost certain presence of amyloid beta plaques and tau tangles in the AD brain, the
amyloid and tau cascade hypothesis has long been the most commonly accepted mechanistic
4

theory for AD. This theory proposed that amyloid beta and tau tangle are fundamental causes and
the initiators of AD (Hardy and Allsop, 1991; Mudher and Lovestone, 2002). The presence of
these molecules destroys cellular cytoskeleton, blocks neuronal transport system, and disrupts cell
communications, before finally leads to degeneration of neurons and Alzheimer’s disease (Chun
and Johnson, 2007; Iqbal et al., 2005; Lacor et al., 2007).
However, multiple proposed AD therapies targeting these two hallmark molecules have failed in
late phase clinical trials. These therapies targeted wide range processes, from production of A
and tau to clearance of these molecules. Tarenflurbil and Semagacestat targeted  secretase, which
cleaves amyloid precursor protein (APP) to form A oligomers, but failed to improve cognitive
function in phase III trials (Doody et al., 2013b; Green et al., 2009). Similarly, Verubecestat and
Lanabecestat were developed as a BASE-1 inhibitor, which is involved in cleavage of APP
upstream of  secretase, but failed to demonstrate clinical benefits (Egan et al., 2018). Tramiprosate
and Methylthioninium chloride were developed as an inhibitor of A and tau aggregation
respectively, but both failed to improve cognitive function in phase III trial (Paul et al., 2007).
While immunotherapies such as Bapineuzumab, Solanezumab, Aducanumab, and Crenezumab
showed promising results in clearing A, all failed to alter disease progression (Cummings et al.,
2014; Doody et al., 2013a; The Lancet, 2017; Vandenberghe et al., 2016). Some of these studies
were also done in early stage AD or even in mild cognitive impairment patients, yet the outcomes
were still not promising. These failed clinical trials suggested two things. First, factors other than
amyloid beta plaque and tau tangles may play a key role in the development of AD. And second,
potential therapeutic efficacy may be dampened by a mixture of responders and non-responders.
5

Prodromal Phase of AD and Brain Glucose Hypometabolism
Brain is the most energy consuming organ in the body. It makes up only 2% of total body mass in
human yet consumes 20% of total oxygen and 25% or glucose (Bélanger et al., 2011; Kety, 1957).
Within the brain over two thirds of energy was spent by neurons, while the rest divided by glial
cells (Harris et al., 2012; Hyder et al., 2013). As a result, the brain is very susceptible to disruption
of energy homeostasis (Bozek et al., 2014; Fu et al., 2011; Magistretti and Allaman, 2015).
Indeed, by measuring cerebral metabolic rate for glucose (CMRglc) using 2[18F]fluoro-2-deoxy-
D-glucose (18F-FDG) and PET imaging, researchers observed that even in healthy aging
individuals, hippocampal glucose hypometabolism is associated with declined cognitive function
(Mosconi et al., 2008b). Further, severe AD patients have significantly reduced brain glucose
uptake in comparison to healthy, age-matched controls (Ishii et al., 1997). In a longitudinal study,
researchers observed significantly declined brain glucose metabolism as patients converted from
MCI to AD (Chetelat et al., 2003). And in a cross-sectional study, researchers observed that
reduced glucose uptake in different brain regions was associated with transitioning into different
stages of AD. Specifically, declined entorhinal cortex glucose uptake best distinguished MCI
patients from normal controls, whereas declined temporal neocortex glucose uptake best separated
AD patients from MCI patients (De Santi et al., 2001).
Moreover, late onset AD has a 20-year prodromal phase, during which reduced brain glucose
metabolism can be readily observed in at risk populations. For example, individuals with at risk
genotypes (such as APOE4 allele) demonstrated lower brain glucose uptake compared to their
counterparts (Mosconi et al., 2008a; Mosconi et al., 2009; Reiman et al., 2001; Reiman et al., 2004;
Small et al., 2000; Willette et al., 2015). Similarly, since late onset Alzheimer’s disease has a
pattern of maternal inheritance, where inheritance of AD from mothers are more frequent than
6

from fathers (Duara et al., 1993; Edland et al., 1996; Liu et al., 2013; Mosconi et al., 2011),
cognitively normal individuals with a maternal history of late onset Alzheimer’s disease showed
decline in platelet cytochrome c oxidase activity and significantly lower brain glucose metabolism
compared to those with a paternal or no family history of the disease (Edland et al., 1996; Mosconi
et al., 2011).
Together, these studies support brain glucose hypometabolism as a risk factor / phenotype for late
onset Alzheimer’s disease. Given the function of mitochondria in cellular bioenergetics and
glucose metabolism, it is expected to play a key role in the etiology of Alzheimer’s disease.
Mitochondria as A Key Player in the Etiology of Alzheimer’s Disease
Early evidence linking mitochondrial dysfunction to Alzheimer’s disease dated back to a 1987
study by Sims and colleagues (Sims et al., 1987), who found a reduced rate of oxygen uptake in
the presence and absence of ADP in frontal neocortex of postmortem confirmed Alzheimer’s
disease cases, indicating potential mitochondrial uncoupling in AD patients (Sims et al., 1987).
More direct evidence for the association between mitochondrial dysfunction and Alzheimer’s
disease came later from studies on the activity of electron transport chain enzymes in AD patients
and post mortem AD brain tissues. In the early 1990s, reduced cytochrome c oxidase (complex
IV) activity was observed in both platelets and postmortem AD brain (Parker et al., 1990; Parker
et al., 1994). The reduction of cytochrome c activity in AD brain was later refined to the temporal
cortex and hippocampus (Maurer et al., 2000). Supporting these findings, reduced mRNA levels
of cytochrome c oxidase subunit 1 and 3 were observed in AD mid temporal gyrus and cytochrome
c oxidase subunit 2 in AD hippocampus (Aksenov et al., 1999; Chandrasekaran et al., 1994).
Furthermore, in both AD temporal and parietal cortices, protein levels of cytochrome c oxidase
subunits, especially those encoded by mitochondria DNA, were reduced (Kish et al., 1999).
7

Although no reduction of cytochrome c oxidase content was observed in platelets of AD patients,
cytoplasmic hybrids (cybrids) containing exogenous mitochondria extracted from platelets of AD
patients showed less cytochrome c oxidase activity compared to cybrids harboring mitochondria
from age-matched controls (Cardoso et al., 2004; Davis et al., 1997; Sheehan et al., 1997).
Alternations in gene expression, protein level, and activity of other electron transport chain
complexes have also been reported in AD tissues, though the evidence was less compelling and
sometimes contradicted (Aksenov et al., 1999; Bosetti et al., 2002; Bubber et al., 2005;
Chandrasekaran et al., 1997; Kim et al., 2000; Schagger and Ohm, 1995; Valla et al., 2006).
Upstream to electron transport chain, impairment of tricarboxylic acid (TCA) cycle enzymes was
observed in AD brain. Autopsy-confirmed AD brain had significantly reduced activity in pyruvate
dehydrogenase complex, isocitrate dehydrogenase, and a-ketoglutarate dehydrogenase complex,
whereas the activity of succinate dehydrogenase and malate dehydrogenase were increased
(Bubber et al., 2005). The severity of enzyme activity impairment was also correlated with clinical
severity of Alzheimer’s pathology (Bubber et al., 2005). Decreased mitochondrial respiratory
capacity was consistent with the higher level of lactate and lower key substrates for TCA cycle in
cerebrospinal fluid and blood in AD patients (Mancuso et al., 2003; Redjems-Bennani et al., 1998).
Consistent with deficits in mitochondrial respiratory capacity, AD brain had elevated levels of
peroxidation products in the frontal cortex, as well as decreased levels of superoxide dismutase, a
radical defensive enzyme, in the frontal cortex, hippocampus, and cerebellum (Richardson, 1993).
Increased oxidative stress is consistent with free radical damage of mitochondrial components and
loss of mitochondrial membrane potential as observed in cybrids harboring mitochondria with
Alzheimer’s disease origin (Cassarino et al., 1998; Trimmer et al., 2000).
8

Besides altered bioenergetic capacity, the crosstalk between mitochondria and the endoplasmic
reticulum (ER) via mitochondrial associated membrane (MAM), which regulates many key
functions of mitochondria such as calcium uptake, phospholipid exchange, intracellular trafficking,
ER stress, and mitochondrial biogenesis, was also disrupted in AD (Burte et al., 2015; Paillusson
et al., 2013).
Beyond changes in mitochondrial function, distribution and morphology of mitochondria were
also different in AD patients. Neurons from AD brain harbored mitochondria of smaller sizes and
disrupted mitochondrial cristae morphology, and cybrids created using mitochondria from
sporadic AD patients also showed swollen mitochondria and less cristae (Baloyannis et al., 2004;
Trimmer et al., 2000). The fission and fusion cycle of mitochondria also seemed disrupted in the
hippocampus of AD brain (Zhang et al., 2016). The number of mitochondrial was also significantly
reduced, likely as a result of increased mitochondria degradation, turnover, and autophagy (Hirai
et al., 2001).
All of these observations support a central role of mitochondrial in Alzheimer’s disease and
potentially in other in age-related metabolic and neurodegenerative diseases (Beal, 1996; Brinton,
2008b, 2009; Brinton et al., 2015; Bubber et al., 2005; Coskun et al., 2012; Gibson et al., 2000;
Khusnutdinova et al., 2008; Lin and Beal, 2006; Simpkins et al., 2008a; Swerdlow and Khan,
2004; Trimmer et al., 2000; Wallace, 2005).
Estrogen Regulation of Glucose Metabolism and Mitochondrial Bioenergetics
Estrogen has been shown to regulate brain glucose metabolism and mitochondrial function in a
systematic way (Brinton, 2008a). It can promote glucose uptake by increasing protein expression
of glucose transporters in the endothelial cells of blood brain barrier as well as on neurons (Cheng
et al., 2001; Shi and Simpkins, 1997). Downstream of glucose uptake, estradiol can increase
9

protein expression and enhance activities of key enzymes involved in glycolysis, including
hexokinase, phosphofructokinase, and pyruvate dehydrogenase (Kostanyan and Nazaryan, 1992;
Nilsen et al., 2007). Besides enzymes involved in glycolysis, estrogen treatment can also increase
protein expression of electron transport chain (ETC) complex IV (cytochrome c oxidase) subunits
I-IV, and ATP synthase subunits F1α and β (Bettini and Maggi, 1992; Nilsen et al., 2007; Stirone
et al., 2005). As a result of these systematic changes, in ovariectomized rats, estrogen treatment
successfully prevented loss of mitochondrial respiratory capacity (Yao et al., 2012b).
In vitro studies using rat embryonic neurons and glial cells also revealed increased maximal
respiratory capacity in response to estrogen treatment (Yao et al., 2011a). Not only can estrogen
promote ATP production in healthy neurons in vitro, it can also preserve ATP production capacity
in neurons insult by Aβ1-42 (Diaz Brinton et al., 2000).
Beyond promoting mitochondrial bioenergetics in the brain, estrogen can further reduce reactive
oxygen species production (Cadenas, 2004), promote calcium homeostasis, protect cells from
apoptosis (Nilsen and Diaz Brinton, 2003), thus systematically promote mitochondrial function.
Estrogen Replacement Therapy and Cognitive Function
The positive effect of estrogen on glucose metabolism and mitochondrial bioenergetics can be
further translated into preserved cognitive function.
In human, a longitudinal study based on the Baltimore Longitudinal Study of Aging cohort
revealed that estrogen therapy users had increased cerebral blood flow in key brain regions such
as hippocampus, para-hippocampal gyrus, and temporal lobe compared to nonusers (Maki and
Resnick, 2000; Rasgon et al., 2005). The improvement of brain metabolism was further
10

accompanied by better cognitive function in estrogen therapy users, an observation supported by
clinical studies on hormone replacement therapy during the past three decades.
Several longitudinal observational studies found that in post-menopausal females, estrogen or
hormone replacement therapy users had reduced incidence rate of Alzheimer’s disease at all ages
compared to non-users (Kawas et al., 1997; Paganini-Hill and Henderson, 1996; Tang et al., 1996;
Waring et al., 1999; Zandi et al., 2002). One study identified delayed onset of Alzheimer’s disease
in hormone users (Tang et al., 1996), and others found a positive associated between dosage and
duration of hormone usage and reduction of AD risk (Paganini-Hill and Henderson, 1996; Tang et
al., 1996; Waring et al., 1999; Zandi et al., 2002). However, no benefits of hormone therapy on
risk of AD was observed if the treatment was initiated long after menopause or near onset of
dementia (Zandi et al., 2002). It seems that late initiation and short duration of estrogen therapy
can limit its therapeutic effect (Brenner et al., 1994; Resnick and Henderson, 2002; Seshadri et al.,
2001). In fact, in female AD patients, short-term estrogen therapy (12-week or 16-week) failed to
produce a meaningful effect on cognitive performance, dementia severity, AD symptoms, or
cerebral perfusion (Henderson et al., 2000; Wang et al., 2000). A one-year estrogen replacement
therapy in women with mild to moderate AD even showed worsening of clinical dementia rating
scale (Mulnard et al., 2000). These studies suggested the importance of starting estrogen therapy
at the early phase of menopause, or well before preclinical stage of AD.
In support of this hypothesis, females who initiated long-term hormone therapy between 50 to 55
years old had enhanced verbal memory compared to controls (Maki et al., 2001). In females
younger than age 65, short-term hormone therapy can enhance verbal memory comparing to
placebo (Maki, 2006; Shaywitz et al., 2003). And short-term hormone therapy administrated for 2
11

to 3 years in the early postmenopausal years could even provide long-term protection against the
risk of cognitive impairment 5 to 15 years later (Bagger et al., 2005).
However, this is not case if estrogen therapy was initiated in older postmenopausal women (>65
years old). In healthy females aged 65 years and older, 3 years of daily estrogen treatment did not
affect cognition (Pefanco et al., 2007). In females over 75 years without dementia or depression,
short-term (9-months) estrogen replacement therapy combined with trimonthly progestin did not
improve cognitive performance (Binder et al., 2001). Furthermore, in females aged over 65 years
old with coronary disease, 4 years of postmenopausal hormone therapy even resulted in worse
verbal fluency (Grady et al., 2002). In accordance to these studies, the Women's Health Initiative
Memory Study (WHIMS) found that in females aged 65 or older, estrogen therapy, with or without
progesterone, increased the risk for dementia and global cognitive decline (Rapp et al., 2003). The
ancillary WHI Study of Cognitive Aging (WHISCA) observed decremented verbal memory
(Coker et al., 2010), which persisted even 4 years after trial stopped (Espeland et al., 2010). These
data support a “critical window” hypothesis, where initiating estrogen therapy shortly after
menopause is prerequired for a preventative effect on cognitive function.
Further, combining progesterone with estrogen seemed to dampen the protective effect of estrogen.
In the Women’s Health Initiative Memory Study in Younger Women (WHIMSY) study, which
was conducted on females between 50 to 55 years old, no evidence for overall benefit or harm on
cognition was observed the hormone replacement therapy users compared to placebos (Espeland
et al., 2013). Similarly, in the Kronos Early Estrogen Prevention Study (KEEPS) Cognitive and
Affective ancillary study (KEEPS Cog), four-year hormone therapy initiated in recently
postmenopausal women (6 months but no more than 36 months postmenopausal) did not alter
cognition (Gleason et al., 2015).
12

For women undergone surgical menopause, a similar “critical window” of therapy exists. In
females received hysterectomy and bilateral oophorectomy around perimenopause, those who
received 2 months of postoperative estrogen treatment did better on verbal recall test than those
on placebo (Phillips and Sherwin, 1992). A 3-month estrogen treatment initiated within four
months post-operation also had positive effect on cognition compared to nonusers (Sherwin, 1988).
On the same axis, patients who had a hysterectomy but retained ovaries (undisrupted estrogen
supply) showed stable cognitive function (Sherwin, 1988). On the other hand, women who started
hormone replacement therapy on an average of 8 years after surgical menopause did not show
benefit in cognitive function (Espeland et al., 2013). Similarly, in a subset of WHIMS study
participants who have had prior hysterectomy, 3 years of estrogen therapy initiated after age of 65
did not improve cognitive functioning (Resnick et al., 2009).
Collectively, the data above suggested that estrogen therapy initiated within the “critical window”
following menopause, whether natural or surgical, has a protective effect on cognition and other
brain function. The reduced estrogen therapeutic effect with aging may be due to decreased
expression of estrogen receptors (Bean et al., 2014).
Estrogen Receptors
There are two major estrogen receptor subtypes: estrogen receptor alpha (ER) and estrogen
receptor beta (ER) Both of them have three major functional domains: the COOH-terminal
ligand domain (LBD) that binds to estrogen or other estrogenic molecules; a zinc finger containing
DNA binding domain (DBD) that binds to estrogen response elements (ERE) on target genes as
either homodimers or heterodimers; and the NH2-terminal domain (NTD) that contains activation
function-1 (AF-1), which is involved in ligand-independent activation of transcription. Another
activation function-2 (AF-2) can be found in LBD and is involved in ligand-dependent activation
13

of transcription (Hewitt and Korach, 2002; Jia et al., 2015). The DBD is highly conserved between
ER and ER, but the LBD and AF-1 have only 59% and 17% amino acid identify, indicating
preferential activation by different ligands and co-transcriptional factors (Jia et al., 2015).
Besides the two nuclear receptors, membrane estrogen receptors have also been identified (Qiu et
al., 2003; Revankar et al., 2005; Toran-Allerand et al., 2002). GPR30 / GPER is a member of the
family of G-protein coupled receptors, and is localized specifically to endoplasmic reticulum
(Revankar et al., 2005). Two other putative membrane estrogen receptors, Gq-mER and ER-x have
been described, however, little is known about their sequences and functions (Qiu et al., 2003;
Toran-Allerand et al., 2002).
Estrogen receptor alpha (ER)
ER is encoded by gene ESR1 on chromosome 6. The full-length human ERα protein has 595
amino acids and a molecular size of 66 kDa. ER has four splice variants including the wild type
(ER66). Splice variant ER46 lacks exon 1 and consequently the AF-1 domain (Flouriot et al.,
2000). Splice variant ERE3 has a truncated exon 3, thus lacks a full DBD, and is a dominant
negative receptor (Wang and Miksicek, 1991). Another spice variant, ER36 lacks both AF-1 and
a part of AF-2 domain, but has an additional C-terminal sequence, which can direct the splice
variant to the plasma membrane (Wang et al., 2005b, 2006).
Estrogen receptor beta (ER)
ERβ gene is encoded by gene ESR2 on chromosome 14. The full-length human ERβ protein has
530 amino acids and a molecular size of 54kDa. Comparing to ER, ER has a truncated AF-1
domain. ERβ has five splice variants including the wild type (ERβ1). Splice variant ERβ2 possess
an in-frame insertion between exons 5 and 6 that encodes an additional 18 amino acids (AAs) in
14

the ligand-binding domain (Chu and Fuller, 1997; Maruyama et al., 1998). ERβ2 retains the ability
to bind to palindromic estrogen response element (ERE), but the insertion results in a loss of ligand
binding ability (Maruyama et al., 1998). Furthermore, ERβ2 can suppress the ERα and ERβ1
mediated transcriptional activation in a dose dependent manner, indicating it may act as a negative
regulator of estrogen activity (Maruyama et al., 1998). ERβ1δ3 and ERβ2δ3 contain a deletion of
117bp encompassing the region encoding the second zinc finger of the DNA binding domain of
ERβ1 and ERβ2 respectively (Petersen et al., 1998). ERβ1δ4 encodes an ERβ that is missing in
exon 4 and does not appear to bind estrogen or localize to nucleus, but was the dominantly
expressed variant in rat hippocampus (Price Jr et al., 2000).
Estrogen receptors and gene expression
Ligand-dependent, estrogen response element (ERE)-dependent transcription
The ligand and ERE-dependent mechanism is the classical mode of action of nuclear ERs. In the
absence of ligands, ERs remain inactive by binding to a stabilizing complex involving heat-shock
protein 90 (Hsp90) (Lee et al., 2012). Upon ligand binding, ER and ER are dissociated from the
stabilizing complex, and undergo conformation changes to form either homodimers or
heterodimers (Lee et al., 2012). The activated ER dimer binds to the palindromic ERE (5’-
GGTCAnnnTGACC-3’) via DBD, which induces further conformation changes and promotes ERs
to recruit transcription co-factors to AF-2 site to induce or suppress target gene transcription
(Cowley and Parker, 1999; Hewitt and Korach, 2002).
ERE-independent transcription
Nuclear ERs can also affect transcription of genes that do not have functional ERE sequences.
This effect is accomplished by ER tethering to other transcription factors such as AP-1, Sp1,
15

cJun/cFos, STAT5, and NFB, with AP-1 and Sp1 being the most studied ones (Bjornstrom et al.,
2001; Blobel et al., 1995; de Medeiros et al., 1997; Faulds et al., 2001; Kushner et al., 2000; Paech
et al., 1997; Porter et al., 1997; Qin et al., 1999; Ray et al., 1994; Stein and Yang, 1995; Stoecklin
et al., 1999; Webb et al., 1999).
For genes lack ERE but have GC-rich promoter sequences, nuclear ERs may induce their
expression by physically interacting with transcription factor Sp1 to enhance Sp1-DNA binding
(de Medeiros et al., 1997; Porter et al., 1997; Qin et al., 1999).
For genes with AP-1 sites, ER-regulated gene expression is mediated by interaction between AP-
1 and c-Jun/c-Fos. ERs achieve this regulation by two distinct mechanisms: the AF-mediated/AP-
1 pathway or the AF-independent/AP-1 pathway. The AF-mediated/AP-1 pathway requires both
AF-1 and AF-2 sites, and thus is specific to ER. In the presence of estrogen, the AF-mediated/AP-
1 pathway is activated, and AF sites interact with p160 family of co-activators that are recruited
by the c-Jun/c-Fos dimer, to activate AP-1 mediated gene expression. Because ER lacks a
functional AP-1 site, it regulates AP-1 mediated gene expression via the AF-independent/AP-1
pathway. This pathway is activated when antiestrogens (such as Tamoxifen) is present. The DBD
interacts with co-suppressors recruited by c-Jun/c-Fos dimer to suppress AP-1 mediated gene
expression (Kushner et al., 2000; Paech et al., 1997; Webb et al., 1999). These data indicate ligand
and receptor subtype specific regulation of gene expression by estrogen receptors (Kushner et al.,
2000; Paech et al., 1997; Webb et al., 1999).
Ligand-independent transcription
In the absence of ligands, ER and ER can still be transcriptionally activated through other
signaling pathways. For example, ER was shown to be activated by epidermal growth factor
16

(EFG), insulin-like growth factor (IGF), and dopamine (Ignar-Trowbridge et al., 1992; Kato et al.,
1995; Ma et al., 1994; Olesen et al., 2005). The ligand-independent activation of ER by growth
hormones requires mitogen-activated protein kinase (MAPK) mediated activation of AF-1.
Activation of MAPK signaling pathway activates AF-1 by promoting phosphorylation of the serine
residue at position 118 located in AF-1, and in turn activates ER transcription activity (Bunone
et al., 1996; Kato et al., 1995). In contrast, dopamine induced ER activation involves cAMP/PKA
signaling pathway through phosphorylation of AF-2 (El-Tanani and Green, 1997). Furthermore,
phosphorylation of two serine residues in the truncated AF-1 region of ER by MAPK enhanced
recruitment of transcription co-factors SRC-1, which is required for ER transcription activity
(Tremblay et al., 1999).
Membrane-initiated, indirect transcription
Upon ligand binding, membrane ERs such as GPR30 and ER36 activate various protein kinases
and phospholipases, and lead to the activation of signaling cascades such as PI3K, cAMP, and
MAPK signaling pathways, resulting in rapid, non-genomic responses such as intracellular
mobilization and other events involved in cell proliferation and apoptosis (Aronica et al., 1994;
Chaudhri et al., 2014; Filardo et al., 2002; Revankar et al., 2005; Wang et al., 2006). However,
activation of MAPK and PI3K can also regulate transcription of genes such as FOS, which as can
interact with AP-1 to regulate expression of additional genes (Maggiolini et al., 2004). Activation
of GPR30 by estrogen also lead to trans-activation of EFGR, which further activates MAPK/ERK
signaling pathway to modulate downstream gene transcription (Edwin et al., 2006; Filardo et al.,
2000; Prenzel et al., 1999).

17

Tissue specific distribution of ERs
Estrogen receptors have tissue specific distribution pattern. (Brandenberger et al., 1997; Couse et
al., 1997; Shughrue et al., 1998).
In the peripheral, ER is dominantly expressed in pituitary gland and mammary gland, highly
expressed in stromal cells in ovary, uterus, kidney, spleen, skeletal muscle, liver and bone marrow,
and moderately expressed in testis, lung and skin, and gut (Brandenberger et al., 1997; Couse et
al., 1997; Shughrue et al., 1998). ER is dominantly expressed in lung, highly expressed in prostate,
testes and epididymis, spleen, and adrenals, and moderately expressed in ovary (mainly in
granulosa cells of developing follicles), uterus, oviduct, testes, skin, lung, and kidney
(Brandenberger et al., 1997; Couse et al., 1997; Shughrue et al., 1998). GPR30 is highly expressed
in the pituitary gland, adrenal medulla, renal pelvis and ovary in mice and rats (Hazell et al., 2009).
In the central nervous system, ER is predominantly expressed in the hippocampus, preoptic area,
and most of the hypothalamus, but barely in the cerebral cortex and cerebellum (Mitra et al., 2003;
Shughrue et al., 1997). ER is primarily expressed in olfactory bulb, cerebral cortex, septum,
preoptic area, bed nucleus of the stria terminalis, amygdala, paraventricular hypothalamic nucleus,
thalamus, ventral tegmental area, substantia nigra, dorsal raphe, locus coeruleus, and cerebellum
(Gonzalez et al., 2007; Mitra et al., 2003; Shughrue et al., 1997). GPR30 is highly expressed in
cortex, hypothalamus, hippocampus, the pontine nuclei and locus coeruleus in the midbrain, and
the trigeminal nuclei and cerebellum Purkinje layer of the hindbrain (Hazell et al., 2009).
Males and females have different distribution of ER in the brain. Most notably is the higher
expression of ER in pyramidal cells of CA3 and CA4, and the dentate gyrus in the females (Zhang
et al., 2002).
18

Interaction between ERs and its effect on gene expression
In tissues where both ER and ER are present, they are likely expressed in different cell types
within the same tissue (for example both ER and ER were detected in the uterus, but ER was
mainly detected in the stromal cells while ER in the follicle). In vitro study found that when
ER and ER are co-expressed in the same cell, ERα had significantly higher transcriptional
activity than ERβ (Yi et al., 2002). This is likely due to higher affinity of ER to ERE (Cowley et
al., 1997; Hyder et al., 1999). However, in case of co-expression, ER / ER heterodimer
formation is preferred over homodimer formation (Pettersson et al., 1997). Because ER / ER
homodimer had the highest ERE binding affinity, followed by ER / ER heterodimer, and ER /
ER homodimer (Cowley et al., 1997), these data indicated that ER may act to attenuate ER
transcription activity. In cells where ER and ER are co-expressed, the ratio of the two subtypes
may modulate the overall transcription activity.
As summarized above, in the case of ERE-independent transcription via AP-1 site, ER can oppose
ER mediated transcription activity induced by estrogen. This is evident in estrogen induced
cyclin D1 gene expression, where ER inhibits E2 induced cyclin D1 transcription by opposing
ER activation (Liu et al., 2002). Furthermore, in mice, ER knockout resulted in an increase in
gene expression of estrogen-responsive genes in both bone and liver by an average of 85%
(Lindberg et al., 2003).
Similarly, for genes with GC-rich promoters, ER-Sp1 complex effectively activated transcription
of target gene while ER-Sp1 had much weaker transcription activity (Saville et al., 2000).
Together these data support that when ER and ER are both present, ER can inhibit or attenuate
ER transcription activity.
19

ER splice variants can also modulate ER-mediated transcription activity. ER46 and ER36 are
ER splice variants that lack functional AF-1 site. They were found to heterodimerize with ER
wildtype and inhibit AF-1 dependent transcription activity (Flouriot et al., 2000). ER is an ER
splice variant found in human cancer cells. It has deletions of exons 5 and 6, and lacks a functional
LBD ER can reduce estrogen-dependent-ERE-dependent transcription activity when co-
expressed with wild type ER or ER, indicating its function as an ER inhibitor (Vladusic et al.,
1998). Similarly, ERβ2, which has an in frame insertion between exon 5 and exon 6, retains the
ability to bind to ERE but not ligands (Maruyama et al., 1998). ERβ2 can suppress the wild type
ERα and ERβ mediated transcriptional activation in a dose dependent manner (Maruyama et al.,
1998). Furthermore, ER2 preferentially dimerize with ER, and have a dominant-negative effect
on ligand-dependent ERα reporter gene activity (Peng et al., 2003).
Together, these data suggest that interaction between ER subtypes and splice variants can modulate
ER mediated transcription activity. In tissues where ER and ER are co-expressed, ER has a
dominance over ER, possibly due to its higher affinity for ERE. However, ER can attenuate or
inhibit ER activity. Given that ER subtypes and their splice variants are differentially distributed
in different tissues and cell types, cells with higher ER expression are likely to have more
suppression of ER transcription activity. Furthermore, because both ER and ER transcriptional
activities require a diverse set of co-regulators, whose availability differ among different cell types,
the overall ER-mediated transcription activity is thus tissue dependent, and dependent upon the
ratio of the ER subtypes and their variants, as well as the availability of transcription co-regulators
in a particular cell. It should be kept in mind that the distribution of ER and ER differs
significantly among different cell types, thus their mode of action may also be cell type specific.
20

Mitochondrial Genome
Unlike many other organelles, mitochondria have their own genome. The human mitochondrial
genome is a circular set of 16569 base pairs encoding 37 genes. Thirteen of these genes encode
protein subunits required for four of the five electron transport chain complexes: complex I
(NADH ubiquinone oxidoreductase), complex III (cytochrome bc1 complex), complex IV
(cytochrome c oxidase), and complex V (ATP synthase); 2 encode rRNAs for mitochondrial
ribosomes (12S and 16S), and 22 encode tRNAs (Figure I-1). Mitochondrial retention of their own
genome throughout evolution solves two cell biology problems. First, the 13 electron transport
subunits coded by mitochondrial DNA (mtDNA) solves the problem that if they were generated
by nuclear DNA, they would not cross the inner mitochondria membrane due to their high
hydrophobicity (Popot and Vitry, 1990). Second, the eukaryotic mitochondrial genome is
transcribed and translated quite differently than the nuclear genome (Mercer et al., 2011). The
genetic system of the mitochondria is transcribed as precursor polycistronic transcripts that are
subsequently cleaved to generate mRNAs, tRNAs and rRNAs (Mercer et al., 2011). The mRNAs
devoted to generating the 13 catalytic subunits required for oxidative phosphorylation are further
translated using several non-universal codons unique to the mitochondrial translation machinery
(Watanabe, 2010; Watanabe and Yokobori, 2011).

21

Figure I-1. Human mitochondrial DNA. Orange indicates protein-encoding genes (13), purple indicates
mitochondrial rRNAs (2), blue indicates tRNAs (22), and gray indicates the D-loop. O H: heavy strand origin,
O L: light strand origin, HSP1: major heavy strand promoter, HSP2: minor heavy strand promoter, LSP, light
strand promoter.
Within the mitochondrial DNA, although no consensus estrogen response element (ERE) were
detected, the presence of putative EREs (half EREs and non-consensus EREs) was confirmed
(Demonacos et al., 1996). It was also demonstrated that both human recombinant ER and ER
can bind to human mitochondrial putative ERE sequences extracellularly, although only ER was
found to bind to mtDNA in the mitochondria (of MCF-7 cells) (Chen et al., 2004b). An ER
binding region within the mitochondrial control region was also identified in breast cancer MCF-
7 cells (Gertz et al., 2013). One study demonstrated that only ER and not ER has mitochondrial
targeting sequences (Chen et al., 2004a). Another study further suggested that ER can interact
with TOM70 and HSP70 to facilitate its internalization into mitochondria (Simpkins et al., 2008b),
suggesting potential direct regulatory effect of ER on mitochondrial gene transcription.
22

Like nuclear DNA, mitochondrial DNA undergoes mutation, though at a much higher rate (Miyata
et al., 1982; Wallace et al., 1987), likely due to higher replication rate, a more mutagenic
environment, and less efficient DNA repair (Xu et al., 2012). Unlike the nucleus, the mitochondrial
DNA repair mechanism is largely limited to base excision repair. Mismatch as a result of either
recombination or repair can lead to single nucleotide polymorphisms (SNPs). Accumulation of
SNPs throughout human evolution may be a result of adaptation, while accumulation of SNPs
during aging may lead to pathological function.
Mitochondrial Genetic Variance and Bioenergetics
Clusters of specific SNPs in the mitochondrial genome define mitochondrial haplogroups that
reflects maternal lineage (Giles et al., 1980; Torroni et al., 1992). For example, the four lineages
specific for sub-Saharan Africa are L0, L1, L2 and L3, and haplogroups A, B, C, D, G, F are
common in Asia (Stewart and Chinnery, 2015). The major haplogroups within descendants of
European ancestry are haplogroups H, I, J, K, M, T, U, V, W, and X (Torroni et al., 1996). These
haplogroups can be further classified into subhaplogroups or clustered together into
superhaplogroups, such as superhaplogroup HV, JT, UK.
Mitochondrial genetic variance has been shown to affect metabolism and mitochondrial
bioenergetics. Resting metabolic rate (RMR) and total energy expenditure (TEE) were measured
in the health, aging and body composition study (Health ABC) (Tranah et al., 2011). Compared to
cluster N, cluster L had significantly lower RMR and TEE. Specifically, haplogroups L0, L2, and
L3 had significantly lower RMR than haplogroup H and superhaplogroups UK and JT; haplogroup
L3 had significantly lower TEE than haplogroup H and superhaplogroups UK and JT; haplogroup
L2 had significantly lower TEE than haplogroup H and superhaplogroup JT (Tranah et al., 2011)
(Table ). In a cohort of healthy Spanish males, haplogroup J participants had significantly lower
23

maximum oxygen consumption (VO2max) than non-J participants (Marcuello et al., 2009). This
difference was later confirmed in an independent cohort of healthy Spanish males, where
haplogroup H was determined to be the driving force for the difference (Martinez-Redondo et al.,
2010) (Table I-1).
Table I-1. Mitochondrial haplogroups/superhaplogroups differentially associated with respiratory phenotypes.
Individuals of haplogroup/superhaplogroups H, UK, and JT had higher rest metabolism rate (RMR), total
energy differences (TEE), and/or VO 2max compared to individuals of cluster L and haplogroup J.
Relatively high Relatively low References
RMR H, UK, JT L2, L3, L3 Tranah et al., 2011
TEE H, UK, JT L0, L2 Tranah et al., 2011
VO 2max H J Marcuello et al., 2009; Martinez-Redondo et al., 2010

The underlying cellular mechanism of the observed differences across different mitochondrial
haplogroups was primarily elucidated using trans-mitochondrial cytoplasmic hybrids, or cybrids,
which controlled for the nuclear genetic background to reveal mitochondrial variances. An early
cybrid study using cultured A539 human lung carcinoma cells harboring either mitochondrial of
haplogroup H or T failed to identify any differences in bioenergetics function (Amo et al., 2008).
However, differences in bioenergetics and mitochondrial function were identified in multiple later
studies using different cell lines and mitochondrial haplogroups. Cybrids constructed from
osteosarcoma 143B rho0 cells and platelets from healthy Spanish donors of either haplogroup H
or superhaplogroup UK were investigated for mtDNA content (Gomez-Duran et al., 2010).
Cybrids harboring UK superhaplogroup were found to have lower mtDNA content, lower mt-
rRNA level, reduced protein synthesis, and decreased cytochrome c oxidase amount (Gomez-
Duran et al., 2010). UK cybrids also had lower mitochondrial inner membrane potential and higher
mitochondrial uncoupling, indicating potentially lower respiratory capacity and reduced ATP
24

production (Gomez-Duran et al., 2010). Similar results were obtained in a later study in middle-
age Caucasian males, where OXPHOS capacity normalized to citrate synthase content was found
to be reduced by 24% in subjects with haplogroup U background comparing to those with
haplogroup H background (Larsen et al., 2014). In 2013, cybrids constructed form human retinal
epithelial cell line ARPE-19 and either haplogroup H or J mitochondria showed reduced ATP
production and glycolysis in J cybrids (Kenney et al., 2013). In accordance with the observed
reduction in mitochondrial respiration, J cybrids also showed lower ROS production (Kenney et
al., 2013). Haplogroup J cybrids with chondrocyte nuclear genetic background also demonstrated
lower NO levels than non-J cybrids (Fernandez-Moreno et al., 2011). Similarly, major Asian
mitochondrial haplogroups are also differentially associated with bioenergetic function (Lin et al.,
2012). These associations identified in the human is also evident in animal models ranging from
drosophila to mice such that different mitochondrial genetic background is associated with
differences in respiratory and metabolic phenotypes, electron transport chain enzyme activities and
mitochondrial functions (Latorre-Pellicer et al., 2016; Pichaud et al., 2012; Scheffler et al., 2012).
These observations support a modified “mitochondrial cascade hypothesis” of late onset AD,
where mitochondrial genetic variations and mutations initiated deficient electron transport chain
function, resulting in less ATP production, disrupted calcium homeostasis, increased free radical
production, ER stress, MAM disfunction, beta amyloid plaque deposition, and tau tangle formation.
These results in turn lead to further damage of mitochondrial DNA, proteins, and lipids, and the
opening of mitochondrial permeability transition pore, which ultimately leads to cell death and
neurodegeneration (Swerdlow and Khan, 2004) (Figure I-2).
25

Figure I-2. Modified mitochondrial cascade of late-onset Alzheimer’s disease (LOAD). Inherited mitochondrial
genetic variations and accumulated mutations during aging lead to deficiency in mitochondrial functions,
initiating a cascade of events including reduced ATP production, increased free radical formation and ER
stress, disrupted mitochondrial associated membrane (MAM) function and mitochondrial dynamics, as well as
amyloid beta plaque and tau tangle formation, which result in further damages of mitochondria and ultimately
leads to apoptosis and neurodegeneration, such as LOAD.

Mitochondrial Genetic Variance and Risk of Late Onset Alzheimer’s Disease
In an early effort to assess the contribution of mitochondrial DNA variances to pathologies of
neurodegenerative diseases, researchers identified a non-synonymous SNP in tRNA
Gln
, mt4336C,
that had an increased frequency in a Caucasian cohort of late onset Alzheimer’s and Parkinson
disease patients (Shoffner et al., 1993). The contribution of mt4336C was later confirmed in a
different North American Caucasian cohort of Alzheimer’s disease patients (Hutchin and
Cortopassi, 1995). Individuals harboring the mt4336C SNP also tended to harbor the mt16304C
SNP, and had a more closely related D-loop sequence, which could be traced back to a single
phylogenetic node (Hutchin and Cortopassi, 1995; Shoffner et al., 1993). However, two other
26

studies did not confirm either mt4336C or mt16304C as a risk factor for developing Alzheimer’s
disease in similar populations, when blood samples and leukocytes from clinically diagnosed
patients were used instead of histopathological confirmed postmortem AD brain tissues (Wragg et
al., 1995; Zsurka et al., 1998). Today, mt16304C is known as a defining SNP for subhaplogroup
H5, and mt4336C a defining SNP for subhaplogroup H5a. The above studies constituted the
earliest debate over whether mitochondrial genetic variances can modify the risk of developing
AD.
In fact, haplogroup H and superhaplogroup HV, which contains haplogroup H and its
subhaplogroups, have been the most reported haplogroup in association with increased risk of
developing AD (Chinnery et al., 2000; Coto et al., 2011; Edland et al., 2002; Elson et al., 2006;
Fachal et al., 2015; Fesahat et al., 2007; Mancuso et al., 2007; Maruszak et al., 2009; Maruszak et
al., 2011; Ridge et al., 2012; Santoro et al., 2010; van der Walt et al., 2004; van der Walt et al.,
2005) (Table I-2). A study based on 30 Iranian late onset Alzheimer’s patients and 100 controls
found that haplogroup H was significantly more abundant in the disease group (Fesahat et al.,
2007) (Table I-2Table I-2). In a Spanish-Caucasian group, haplogroup H and its defining SNP
mt7028C were enriched in late onset AD patients compared to controls (Coto et al., 2011) (Table
I-2). In a large Caucasian cohort containing 422 late-onset Alzheimer’s disease patients and 318
neurologically healthy controls, researchers found that superhaplogroup HV, which contains
haplogroup H, had significantly higher presence in LOAD than in control (Maruszak et al., 2011)
(Table I-2). Finally, a meta-analysis pooling data from five previous studies (some studies
including early-onset Alzheimer’s disease patients) also confirmed the association between
haplogroup H and superhaplogroup HV and the risk of developing Alzheimer’s disease (Maruszak
et al., 2011) (Table I-2).
27

When mitochondrial DNA was sequenced in greater depth, sub-haplogroup H5 was significantly
associated with AD compared to haplogroup H of central-northern Italians, (Santoro et al., 2010)
(Table I-2). However, from the Cache county study on aging and memory in Utah residents, sub-
haplogroups H6a1a and H6a1b were found to be protective against AD (Ridge et al., 2012) (Table
I-2). While the protective role of sub-haplogroups H6a1a and H6a1b seems contradictory to the
overall risk of haplogroup H, the data predict that the observed risks within haplogroup H may be
driven by its sub-haplogroups with H5 increases risk of LOAD whereas H6 reduces risk.
Table I-2. Observed effects of mitochondrial superhaplogroup HV and haplogroup H on risk of AD. Effects of
haplogroup H defining SNPs and its subhaplogroups are listed under haplogroup H.
Haplogroups Observations References
HV Increased risk, especially in
females
Maruszak, 2011
No effect Elson, 2006; mazuszak, 2009; Fachal,
2015
H Increased risk Fasahat, 2007; mazuszak 2009; Coto,
2011; Maruszak, 2011
Defining SNP mt7028C increased
risk
Coto, 2011
Defining SNP mt7028C increased
risk in females only
Van der Walt 2004
H5 increased risk, especially in
females
Santoro, 2010
H5 and APOE4 synergistically
increased risk
Maruszak, 2011
H5a defining SNP mt4336C
increased risk in APOE4 carriers
Edland, 2002;
H6a1a and H6a1b decreased risk Ridge,2012
No effect Chinnery, 2000; Van der Walt, 2005;
Mancuso, 2007; Fachal, 2015

The second most studied superhaplogroup is UK, and its member haplogroups U and K, including
their subhaplogroups (While haplogroup K is currently recognized as a branch of haplogroup U,
early studies classified haplogroups U and K as two parallel haplogroups under superhaplogroup
28

UK. For consistency of referring to previous studies, we will use the earlier classification system
throughout this thesis) (Table I-3). In a Utah based ADNI cohort, superhaplogroup UK was
identified as a risk factor for AD (Lakatos et al., 2010) (Table I-3). These findings are in contrast
to an earlier study conducted in a Poland-based Caucasian population, where no effect of
superhaplogroup UK was observed (Maruszak et al., 2009) (Table I-3). The disparity may be
explained by differences in the distribution of specific subhaplogroups or SNPs in the studied
populations. Specifically, while each of the three defining SNPs for haplogroup U (mt11467G,
mt12308G, and mt12372A) has been identified as a risk factor for AD (Lakatos et al., 2010),
subhaplogroup U5a1 and SNP mt16224C, a haplogroup K defining SNP, were shown to be
protective (Maruszak et al., 2011) (Table I-3).
Table I-3. Observed effects of mitochondrial superhaplogroup KU and haplogroups K and U on risk of AD.
Haplogroups Observations References
UK Increased risk Lakatos, 2010
Decreased risk in males Maruszak, 2011
No effect mazuszak, 2009; Fachal, 2015
K Defining SNP mt16224C
decreased risk
Maruszak, 2011
Decreased risk in APOE4 carriers Carrieri, 2001; Maruszak, 2011
No effect Chinnery, 2000; Van der Walt, 2005;
Elson, 2006; Fasahat, 2007; Mancuso,
2007; mazuszak 2009
U Increased risk Fasahat, 2007
Increased risk in males Van der Walt 2004
Defining SNPs mt11467G,
mt12308G, and mt12372A
individually increased risk
Lakatos, 2010
Decreased risk in females Van der Walt 2004
Decreased risk in APOE4 carriers Carrieri, 2001
U5a1 decreases risk Maruszak, 2011
No effect Chinnery, 2000; Van der Walt, 2005;
Elson, 2006; Mancuso, 2007;
mazuszak 2009; Fachal, 2015

29

Another common European haplogroup studied for its association with AD is haplogroup T, where
one study in French-Canadians found that the frequency of SNPs mt709A and mt15928A, both
defining SNPs for haplogroup T, were three times higher in controls than in AD patients,
suggesting a protective role of haplogroup T (Chagnon et al., 1999) (Table I-4). However, the
Health, Aging, and Body Composition (Health ABC) study found that haplogroup T had increased
risk for dementia when compared to haplogroup H (Tranah et al., 2012). The Health ABC study
also found that haplogroup J, also under superhaplogroup JT, had significant decline in cognitive
function compared to haplogroup H (Tranah et al., 2012).
Table I-4. Observed effects of mitochondrial superhaplogroup JT and haplogroups J and T on risk of AD.
Haplogroups Observations References
JT Decreased risk in females Maruszak, 2011
No effect mazuszak, 2009; Fachal, 2015
J Decline in cognitive function
comparing to H
Tranah, 2012
J2b defining SNP mt7476T,
mt5633T, and mt15812A
increased risk
Chagnon, 1999
No effect Chinnery, 2000; Van der Walt, 2005;
Elson, 2006; Fasahat, 2007; Mancuso,
2007; mazuszak 2009; Fachal, 2015
T Increased risk for dementia
comparing to H
Tranah, 2012
T defining SNP mt709A and
mt15928A decreased risk
Chagnon, 1999
Decreased risk in females Maruszak, 2011
No effect Chinnery, 2000; Van der Walt, 2005;
Elson, 2006; Fasahat, 2007; Mancuso,
2007; mazuszak 2009; Fachal, 2015

In addition to major European haplogroups, several African and Asian haplogroups have also been
reported to be associated with the AD or risk of dementia (Table I-5). For example, in an African
American population, haplogroup L1 was found to be have increased risk for developing dementia
30

(Tranah et al., 2014). In Asians, subhaplogroups G2a, B4c1, and N9b1 were reported to be
associated with AD in Japanese populations, and haplogroup B5 was reported to be associated
with AD in Han Chinese (Bi et al., 2015; Takasaki, 2008, 2009) (Table I-5).
Table I-5. Association between some Asian mitochondrial haplogroups and the risk of Alzheimer’s disease.
Studies listed in chronological order.
Haplogroups Observations References
L1 Increased risk Tranah, 2014
G2a Increasesd risk Takasaki, 2008; Takasaki, 2009
B4c1 Increased risk Takasaki, 2009
N9b1 Increased risk Takasaki, 2009
B5 Increases risk Bi, 2015

APOE Genotype and Risk of Alzheimer’s Disease
APOE4 genotype is a widely recognized risk factor for Alzheimer’s disease, and has been
repeatedly confirmed in the studies reviewed herein (Carrieri et al., 2001; Corder et al., 1993; Coto
et al., 2011; Edland et al., 2002; Maruszak et al., 2011; Poirier et al., 1993; Rebeck et al., 1993;
Saunders et al., 1993). Further, APOE4 has been associated with mitochondrial dysfunction and
glucose hypometabolism in brain (Mosconi et al., 2008a; Mosconi et al., 2005; Mosconi et al.,
2004a; Mosconi et al., 2004b; Mosconi et al., 2004c; Reiman et al., 2001; Reiman et al., 2004,
2005; Valla et al., 2010; Wolf et al., 2013). Compared to non-carriers, APOE4 carriers showed
reduced cerebral parietal glucose metabolism among cognitive normal elderlies with family history
of AD (Small et al., 2000; Small et al., 1995). In APOE4 positive MCI patients, reduced regional
cerebral metabolic rate of glucose consumption (rCMRglc) was detected in temporoparietal and
posterior cingulate cortex (Mosconi et al., 2004b). In AD patients, more severe hypometabolism
was detected in the parietal, temporal, and cingulate areas in APOE4 carriers than non-carriers
(Drzezga et al., 2005; Mosconi et al., 2004c). Brain glucose hypometabolism was also more
widespread in APOE4 positive AD patients (Mosconi et al., 2004a). On the therapeutic side, mild-
31

to-moderate AD patients who are APOEε4 carriers were shown to be less responsive to
rosiglitazone, which can improve mitochondrial efficiency and glucose metabolism (Risner et al.,
2006; Roses et al., 2007).
In longitudinal studies, APOE4 carriers had significantly greater rCMRglc decline in the vicinity
of temporal, posterior cingulate, and prefrontal cortex, basal forebrain, parahippocampal gyrus,
and thalamus (Mosconi et al., 2008a; Reiman et al., 2001). Decrease of glucose metabolism was
also evident in young and middle-aged APOE4 carriers in posterior cingulate, parietal, temporal,
and prefrontal cortex, as well as thalamus (Mosconi et al., 2008a; Reiman et al., 2004). The effect
of APOE4 allele has also been shown to be gene dose dependent with APOE4 homozygote
carriers showing greater hypometabolic deficit relative to APOEe3/4 heterozygote carriers
(Reiman et al., 2005).
At the cellular level, APOE4 gene expression in human was associated with down-regulation of
genes involved in mitochondrial oxidative phosphorylation and energy metabolism (Xu et al.,
2007; Xu et al., 2006). APOE4 gene expression was also found to be associated with lower
mitochondrial cytochrome oxidase activity in posterior cingulate cortex among young adults with
family history of AD (Valla et al., 2010). Neurons from humanized APOE4 knock-in mice had
significantly lower amount of all five electron transport chain complexes comparing to those from
APOE3 knock-in mice (Chen et al., 2011a). Proteomic analysis revealed decreased expression of
proteins involved in the TCA cycle, glucose, lipid and amino acid metabolism in APOE4 knock-
in mice (Shi et al., 2014). Further, cytochrome c levels were significantly lower in ApoE4 mice
compared with ApoEe3 mice (Shi et al., 2014). In vitro studies also suggested that truncated
APOE4 fragment can interact directly with mitochondrial and cause mitochondrial dysfunction
32

and neurotoxicity (Chang et al., 2005; Mahley et al.). Given the association between decreased
bioenergetic capacity in brain and the risk of AD, an interaction between APOE genotype and
mitochondrial haplotypes is possible.
Three interesting modes of interactions between APOE4 status and mitochondrial haplogroups
can be identified in modulating the risk of AD were apparent (Table I-6). The first mode is a
neutralizing effect of mitochondrial haplogroup on the effect of APOE4 on risk of AD (Table I-
6). Early studies identified a non-random association between mitochondrial haplogroup and
APOEe4 status in AD patients (Carrieri et al., 2001). Specifically, while APOE4 carriers had
significantly higher odds ratio for AD, those belonging to haplogroups K and U did not, indicating
a neutralizing effect of haplogroups K and U on the risk of APOE4 gene status (Carrieri et al.,
2001). The non-random distribution of mitochondrial haplogroups associated with APOE4 status
and the neutralizing effect of mitochondrial haplogroup K on APOE4 were later confirmed by
Maruszak et al (Maruszak et al., 2011). The second mode is an enabling effect of APOEe4 on
mitochondrial genetic variances as risk factors for AD (Table I-6). In non-APOE4 carriers, SNP
mt4336C (a defining SNP for sub-haplogroup H5a) was not an AD risk factor, however, in
APOE4 carriers, the same SNP was a risk factor for AD (Edland et al., 2002). This study indicated
that APOE genotype could explain the earlier disparity regarding the association between SNP
mt4336C and AD (Hutchin and Cortopassi, 1995; Shoffner et al., 1993; Wragg et al., 1995; Zsurka
et al., 1998). A synergistic effect was also observed between APOE4 and mitochondrial
haplotypes (Table I-6). For example, SNP mt7028C, a defining SNP for haplogroup H, and sub-
haplogroup H5 were suggested to act synergistically with APOE4 to increase risk for AD (Coto
et al., 2011; Maruszak et al., 2011).
33

Table I-6. Three modes of interactions between APOE4 status and mitochondrial haplogroups in modulating
the risk of AD. Haplogroups K and U could neutralize the risk of APOE4 on AD. Haplogroup H defining SNP
mt4336C was associated with late onset AD in APOE4 carriers only. Subhaplogroup H5 and haplogroup H
defining SNP mt7028C could act synergistically with APOE4 to increase the risk of late onset AD.
Interactions Haplogroups/SNPs References
Neutralizing K Carrieri et al., 2001; Maruszak et al., 2011
U Carrieri et al., 2001
Enabling mt4336C (H) Edland et al., 2002
Synergistic H5 Maruszak et al., 2011
mt7028C (H) Coto et al., 2011

As with the association between mitochondrial haplotype and the risk of late onset AD, the
interaction between APOE4 and mitochondrial genetic variances in modulating the risk of late
onset AD remains debatable. Multiple studies failed to identify any correlation between
mitochondrial haplogroup and APOE4 status or failed to an interaction between the two on the
risk of developing AD (Chinnery et al., 2000; Lakatos et al., 2010; Mancuso et al., 2007; Ridge et
al., 2013; Santoro et al., 2010; van der Walt et al., 2005; Zsurka et al., 1998). Collectively, these
disparate findings on the association between mitochondrial haplotype, APOE genotype, and risk
of Alzheimer’s disease emphasize the importance of a precision medicine approach that considers
mitochondrial genetic variance in combination with nuclear genetics.
Sex Difference and Risk of Alzheimer’s Disease
Females are at greater lifetime risk for late onset Alzheimer’s disease, and also have higher
prevalence and incidence rate than all age-matched males (Brookmeyer et al., 1998; Grimm et al.,
2016; Mielke et al., 2014). The higher risk for female is also evident in faster disease progression
and greater burden of AD pathology (Aguero-Torres et al., 1998; Barnes et al., 2005; Corder et al.,
2004; Grimm et al., 2016; Kelly et al., 2013; Mielke et al., 2014; Skup et al., 2011). While the
underlying mechanism remains to be elucidated, increased mitochondrial oxidative stress may play
34

a role (Mandal et al., 2012; Schuessel et al., 2004). Given the effect of mitochondrial genetic
variances on mitochondrial function and risk of AD, it is of interest to identify any interaction
between mitochondrial haplotypes and sex difference on risk of late onset AD.
Indeed, some mitochondrial genetic variances were found to be associated with AD in females
only (Maruszak et al., 2009; Maruszak et al., 2011; Santoro et al., 2010; van der Walt et al., 2004).
Superhaplogroup JT, haplogroup T, a haplogroup T defining SNP mt13368A, a haplogroup U
defining SNP 12308G, and a non-H defining SNP mt7028T were found to exert protective effects
only in females (Maruszak et al., 2011; van der Walt et al., 2004) (Table I-7). Superhaplogroup
HV, haplogroup H, sub-haplogroup H5, and a haplogroup H defining SNP mt7028C were
identified as risk factors for females only (Maruszak et al., 2009; Maruszak et al., 2011; Santoro
et al., 2010) (Table I-7). In contrast, some variances affected only males (Table I-7).
Superhaplogroup UK, and SNP mt9055A, a defining SNP for haplogroup K, were found to be
associated with reduced risk of AD in males, while SNP mt13708G (for many non-J haplogroups),
and SNP mt10398A, a defining SNP for some subhaplogroups of U, were associated with
increased risk in males only (Maruszak et al., 2009; Maruszak et al., 2011; van der Walt et al.,
2004) (Table I-7). Certain mitochondrial genetic variances also showed opposite effects in each
sex. For example, haplogroup U was associated with increased risk in males but decreased risk in
females (van der Walt et al., 2004) (Table I-7).

35

Table I-7. Sex differentiates effects of mitochondrial haplogroups on risk of Alzheimer’s disease.
Haplogroups and SNPs identified to have sex differences from literatures are listed below. A possible
haplogroup defined by the SNPs listed is indicated in the parenthesis following the SNP.
Haplogroups/SNPs Female Male Authors
H Increased risk No effect Maruszak, 2009; Mazuszak, 2011
H5 Increased risk No effect Santoro, 2010
HV Increased risk No effect Maruszak, 2009; Mazuszak, 2011
T Decreased risk No effect Maruszak, 2011
JT Decreased risk No effect Maruszak, 2011
U Decreased risk Increased risk Van der Walt, 2004
KU No effect Decreased risk Maruszak, 2011
mt7028C (H) Increased risk No effect Maruszak, 2009
mt13368A (T) Decreased risk No effect Maruszak, 2011
mt13708G (non-J) No effect Increased risk Maruszak, 2009
mt9055A (K) No effect Decreased risk Maruszak, 2011

Study Hypothesis: From Mitochondrial and Metabolic Phenotype Towards a Precision
Medicine Approach for Alzheimer’s Disease
Overall, literature and our previous studies demonstrated that alternation in brain metabolic and
mitochondrial respiratory phenotypes precedes clinical onset of Alzheimer’s disease. Having a
comprehensive understanding of the metabolic and respiratory profiles at key transition points, the
underlying mechanism, and key players can help identify potential therapeutic targets and
windows. As summarized here, estrogen is a master regulator of brain glucose metabolism and
mitochondrial function. These cellular function as well as risk of late onset AD can be further
modulated by genetic factors such as mitochondrial genetic variance, APOE genotype, and
chromosomal sex difference.
Herein is a program of research that focuses on a heretofore unexamined aspect, which is highly
relevant to understanding mechanisms leading to LOAD in female brain, estrogen regulation of
mitochondrial function, and implications for personalized medicine based on genetic markers. We
36

hypothesize the dismantling of the female brain bioenergetics system during perimenopause is
associated with changes in metabolic profile and mitochondrial OXPHOS gene expression, which
could be driven by loss of E2. Because ER and ER have different structure, function, and
intracellular distribution pattern, we propose to elucidate how each of these estrogen receptors
modulates mitochondrial gene expression and bioenergetics. Given that mitochondrial genetic
variances, APOE genotype, and chromosomal sex were associated with differential risk for late
onset AD and bioenergetics capacities, we further hypothesize that these three genetic risk factors
can modulate potential Alzheimer’s disease therapies targeting mitochondrial bioenergetics.
In chapter II, we analyze and determine the metabolic and mitochondrial respiratory phenotype in
aging female brain. The goal is to provide a comprehensive and detailed roadmap of changes in
metabolic profile during chronological aging and at each stage of the perimenopausal transition.
This can help identify potential therapeutic windows and targets to preserve brain metabolic and
bioenergetic capacity.
In chapter III, we elucidated the role of estrogen receptor alpha and estrogen receptor beta in
regulating OXPHOS gene expression and mitochondrial function. The goal is to establish estrogen
receptor subtype specific contribution to estrogen regulation of mitochondrial respiration and gene
expression.
In chapter IV, we aimed to establish whether the three genetic modulators for mitochondrial
function and late onset AD: mitochondrial genetic variance, APOE genotype, and chromosomal
sex can affect therapeutic outcomes of molecules targeting mitochondrial bioenergetics. To
accomplish this, we conducted retrospective responsive analysis on two clinical studies previously
completed in our lab – the PhytoSERM study and the Allopregnanolone study.
37

Together, this program of studies deepened our understanding of bioenergetics transformation in
the aging female brain and the underlying regulatory effect by specific estrogen receptors. These
results can be translated into strategies for predicting at-risk phenotypes and preventing late onset
AD. Our responsive analysis of two clinical trials further shed light on a genetic marker-based
precision medicine approach for Alzheimer’s disease therapeutics.

38

CHAPTER II

DYNAMIC METABOLIC AGING OF THE BRAIN DURING ENDOCRINOLOGICAL
AND CHRONOLOGICAL AGING

Abstract
Brain glucose hypometabolism and mitochondrial dysfunction is a key signature of late onset
Alzheimer’s disease (AD). Similar metabolic patterns are also observed in natural aging and during
the perimenopausal transition in the female brain, making age and chromosomal sex two risk
factors for late onset AD. Comprehensive understanding of the dynamic metabolic aging process
in the female brain can shed light on potential interventions and prevention windows for AD. Using
a rat model recapitulating fundamental characteristics of human menopausal transition, we
observed systematic alternations in gene expression of both mitochondrial and nuclear encoded
electron transport chain (ETC) subunits, as well as changes in complex I and complex IV enzyme
activities. Using an unbiased, discover-based metabolomics analysis, we outlined the dynamic
metabolic signature from glucose centric to utilization of fatty acids, lipids, and ketone bodies, and
finally to anaerobic glycolysis during pre-, peri-, and post-menopausal aging respectively.
Observations in metabolomics were further backed up by analysis of transcriptomics. Collectively,
we present the first detailed metabolic profile in the aging female brain that covers both the
endocrinological and chronological transitions.
Introduction
Late onset AD is a complex disease with approximately a 20-year prodromal period (Alzheimer's
Association, 2015). It is associated with brain glucose hypometabolism, which can be detected in
at risk groups before diagnosis of the disease, and is predictive of disease progression (Chetelat et
al., 2003; De Santi et al., 2001; Ishii et al., 1997; Mosconi et al., 2008a; Mosconi et al., 2008b;
39

Mosconi et al., 2009; Reiman et al., 2001; Reiman et al., 2004; Small et al., 2000; Willette et al.,
2015).
Because aging is also the associated reduced glucose metabolism, mitochondrial dysfunction, and
oxygen flow in the brain, it is considered a primary risk factor for AD (de Leon et al., 1983; Hoyer,
1982; Hyder and Rothman, 2012; Lin and Rothman, 2014; Rothman et al., 2011). On the cellular
level, aging is associated with reduced glucose transporter expression, compromised hexokinase
activity, phosphorylated (inactivated) PDH, and altered levels and activities of key enzymes
involved in oxidative phosphorylation (Boveris and Navarro, 2008; Bowling et al., 1993; Ding et
al., 2013a; Ding et al., 2013b; Irwin et al., 2008; Jones and Brewer, 2010; Klosinski et al., 2015;
Maklashina and Ackrell, 2003; Meier-Ruge et al., 1980; Navarro and Boveris, 2007; Rettberg et
al., 2014; Ulfert et al., 1982; Yao and Brinton, 2012; Yao et al., 2010; Yao et al., 2012a; Yao et
al., 2009; Yao et al., 2011c; Yin et al., 2015). On the molecular level, aging is associated with
significant down regulation of nuclear encoded OXPHOS genes (Mastroeni, 2017; Yin et al.,
2015) and disrupted balance of NAD/NADH, AMP/ ATP, purine and pyrimidine pool (Ivanisevic
et al., 2016; Zhu et al., 2015).
In aging females, the perimenopausal transition serves as a double risk for late onset AD. This
unique endocrine transition is also linked to deficits in brain glucose metabolism and mitochondrial
dysfunction, (Mosconi et al., 2017; Yin et al., 2015), which helps explain the two-fold greater
lifetime risk of AD in females (Brinton, 2008b; Paganini-Hill and Henderson, 1994; Seshadri et
al., 2006).
While the brain primarily utilizes glucose as its fuel source, glucose is not the sole energy substrate
for the brain — it has certain flexibility to adapt to alternative fuel sources in response to energy
crisis. Under conditions of restricted nutrient access, alternative energy substrates such as lactate
40

and ketone body can also be used to generate ATP (Guzman and Blazquez, 2004; Lin et al., 2015;
Morris, 2005; Pellerin, 2010). Previous study in the lab also demonstrated that ketone bodies are
utilized as an alternative fuel in aging female brain in response to deficits in glucose metabolism
(Klosinski et al., 2015; Yao et al., 2011c).
Yet, a detailed and comprehensive metabolic profile is not available for the female brain during
endocrinological and chronological aging, partly because most aging models included only males,
and also that most studies utilized a targeted approach and focused on particular metabolites. In
this study, we outline the dynamic metabolic signature at each stage of the perimenopausal and
chronological transition in the female brain. Given the central role of mitochondria and electron
transport chain (ETC) in brain bioenergetics, we first investigated the gene expression of all
mitochondrial and nuclear encoded OXPHOS genes, followed by changes in activities of key
electron transport chain complexes. We also report here observations from global metabolomics
and transcriptomes, detailing the dynamic metabolic profile in the female brain at each stage of
aging.
Materials and Methods
Animals
All animal studies were performed following National Institutes of Health guidelines on use of
laboratory animals and all protocols were approved by the University of Southern California
Institutional Animal Care and Use Committee.
Female Sprague Dawley rats were obtained from Harlan Laboratories (now part of Envigo) at
either 5-month or 8-month of age. Their estrous cycle status was monitored and evaluated by
vaginal cytology obtained through daily lavage between 9am to 11am. The stage of estrous cycle
41

was determined by the proportion of different cell types (epithelial cells, cornified cells, and
leukocytes) presented in the vaginal secretions. Details of the procedure and classification of
estrous stages were previously described in detail by Yin et al. (Yin et al., 2015). Female Sprague
Dawley rats normally cycle through the four stages of estrous cycle in 4 to 5 days (regular). As
they age, their reproductive system becomes incompetent, and the estrous cycle becomes
unpredictable and prolonged, usually between 6 to 9 days (irregular), before they finally become
reproductive senescent and stay constantly in the estrous stage (acyclic). This transition occurs in
rats at around 9-month to 10-month of age (Brinton et al., 2015; Yin et al., 2015). And to capture
this endocrinological transition, we included 9-10 months old regular cycling rats (Reg 9 mo), 9-
10 months old irregular cycling rats (Irreg 9 mo), and 9-10 months old acyclic rats (Acyc 9 mo).
And to test for age effect, we also included 6 months old regular cycling rats (Reg 6 mo) and 15
months old acyclic rats (Acyc 15 mo). This model was used as an animal model that recapitulates
the menopausal transition in human.
To eliminate confounding effect of estrous cycle, all animals were euthanized on the estrous day
of their estrous cycle. Rats that did not meet the endocrine status criteria were excluded from the
study. For each assay, and an N of 5-6 was used for each group.
Dissection of the Brain
Rats were euthanized per animal protocol at University of Southern California, and brains were
dissected quickly on ice to prevent degradation. Briefly, meninges were completely removed,
followed by removal of hypothalamus, cerebellum, and brain stem. The two hemispheres were
then separated, and hippocampus was peeled off from each hemisphere carefully. Brain tissues
were snap frozen in liquid nitrogen before being stored in -80°C for subsequent assays.
42

RNA isolation
Frozen hippocampus tissue was directly homogenized in TRIzol® Reagent (Invitrogen, 15596026)
using The Bullet Blender® and silicon beads. Chloroform was used to extract RNA from the
homogenate at a volume ratio of 1:5 to that of the TRIzol® Reagent. Ethanol was then used to
precipitate nucleic acids from the aqueous phase. RNA was further purified using PureLink™
RNA Mini Kit (Invitrogen™, 12183018A) following manufacturer’s instructions. Purelink™
DNase (Invitrogen™, 12185010) was used to eliminate DNA contamination. Purified RNA was
eluded in RNase-free, diH2O. RNA concentration and quality were checked by NanoDrop™ One.
Real-time quantitative PCR assays
Gene expression of the 12 mitochondrially encoded genes and mitochondrial transcriptional
factors (TFAM, TFB1M, and TFB2M) was determined using single tube Taqman gene expression
assays (ThermoScientific, Rn03296764_s1, Rn03296765_s1, Rn03296781_s1, Rn03296792_s1,
Rn03296799_s1, Rn03296815_s1, Rn03296746_s1, Rn03296721_s1, Rn03296734_s1,
Rn03296820_s1, Rn03296710_s1, Rn03296716_s1, Rn00667869_m1, Rn03296716_s1,
Rn00580051_m1, Rn00710690_m1, Rn01412502_m1). Briefly, RNA was converted to cDNA
using SuperScript VILO cDNA Synthesis Kit (ThermoFisher, 11754050). A total of 6.25ng of
cDNA was used per rt-PCR reaction along with TaqMan™ Universal PCR Master Mix (Applied
Biosystems, 4304437). MT-ND3 was excluded due to unsatisfying primer performance. Target
mRNA was amplified using the Applied Biosystems™ QuantStudio™ 12K Flex system. Relative
gene expression level (fold change) to reference group was calculated by the comparative Ct
(ΔΔCt) method. Statistical significance was calculated by ANOVA followed by non-paired t-test.

43

RNA Sequencing (RNA-Seq)
RNA-Seq was conducted on hippocampal RNA at Vanderbilt Technologies for Advanced
Genomics (VANTAGE). Only RNA samples with an acceptable RNA quality indicator score (RQI
>7) were used for sequencing. Enrichment of mRNA and library preparation of cDNA were done
using a stranded mRNA (poly(A) - selected) sample preparation kit. Sequencing was performed at
100bp paired-end on NovaSeq600, targeting 30 million reads per sample. Transcripts were mapped
to rat genome (ensemble release 90) using Kallisto 0.4.3 (Bray et al., 2016). Tximport V1.6.0
(Soneson et al., 2015) was used to generate a counts table from Kallisto output, and DESeq2
V1.18.1 (Love et al., 2014) was used to calculate normalized read counts for each gene and/or
transcript and to perform expression analysis.
Ingenuity Pathway Analysis (IPA) of RNA-Seq
Differentially expressed gene lists were processed using the core analysis function of IPA. Only
genes with p value smaller than 0.05 was considered. The outputs are lists of altered canonical
pathways and upstream regulators. The canonical pathways are identified based on enrichment of
qualified genes. The upstream regulator analysis predicted activation or inhibition of regulatory
molecules based on expression of respective downstream genes and networks compiled from
literature and IPA’s Ingenuity knowledge base.
Enzyme activity assays
Complex I Enzyme Activity Assay
Complex I enzyme activity was measured using Complex I Enzyme Activity Microplate Assay
Kit (ABCAM, ab109721) following manufacturer’s instructions. Briefly, 100g of hippocampal
protein homogenate was used per sample in duplicates. Complex I was captured by capture
44

antibody in pre-coated microplate wells. After immobilization of the target, Complex I activity
was determined by following the oxidation of NADH to NAD
+
and the simultaneous reduction of
a dye which leads to increased absorbance at optical density (OD) of 450 nm. Complex I activity
was calculated as change of absorbance per minute using the “slope” function in Microsoft Excel,
and results expressed as mili-OD per minute (mOD/min) or percentage of a reference group.
Statistical significance was calculated by ANOVA followed by un-paired t-test.
Complex IV Enzyme Activity Assay
Complex IV enzyme activity was measured by Complex IV Rodent Enzyme Activity Microplate
Assay Kit (ABCAM, ab109911) following manufacturer’s instructions. Briefly, 100g of
hippocampal protein homogenate was used per sample in duplicates. Cytochrome c oxidase was
immunocaptured in pre-coated microplate wells. After immobilization, cytochrome c oxidase
activity was determined colorimetrically by following the oxidation of reduced cytochrome c by
the absorbance change at optical density of 550nm. Complex IV activity was calculated as change
of absorbance per minute using the “slope” function in Microsoft Excel, and results expressed as
percentages of a reference group. Statistical significance was calculated by ANOVA followed by
un-paired t-test.
Metabolome Analysis
Changes in metabolomic profile during chronological and endocrinological aging processes was
determined at each aging and endocrinological transition point. Both cortex (200g) and plasma
(100L) samples were included to identify differences and correlations between the central
nervous system and the peripheral system. Metabolomics analysis was performed by
METABOLON® utilizing their Global Metabolomics platform and the Complex Lipids Panel™
45

to identify changes in metabolic pathways. Briefly, the Global Metabolomics Platform uses mass
spectrometry to identify and quantify over 1000 metabolites, covering classes of metabolites
including amino acids, carbohydrates, lipids, nucleotides, microbiota metabolism, cofactors and
vitamins, and xenobiotics. The Complex Lipids Panel focuses on the lipidomic, and determines
absolute quantitation, molecular species concentration, and complete fatty acid composition of 14
lipid classes, including principle phospholipid, sphingolipid and neutral lipid classes.
Hormone panels
Serum estradiol level was measured by BIOCRATES Life Sciences using their AbsoluteIDQ®
Stero17 Kit. Samples were prepared and extracted using solid phase extraction and steroids
quantified on an HPLC-MS/MS platform (MRM, precursor scans and neutral loss scans).
Statistical analysis
Statistical significance between animal groups was calculated by one-way ANOVA followed by
unpaired t-test. Significance was defined as p<0.05.
Results
Mitochondrial Gene Expression
Before onset of perimenopausal transition, as female rats aged from 6-month-old to 9-month-old
and reproductive competent, we observed no significant effect of aging on mitochondrial gene
expression in the hippocampus. As female rats transitioned from reproductive competent to
reproductive incompetent around 9-month-old, mitochondrial encoded genes first slightly
increased as the transition started (from Reg 9 mo to Irreg 9 mo), and then dropped as the animals
become reproductively senescent (from Irreg 9 mo to Acyc 9 mo), with MT-ND1 and MT-CO2
significantly down-regulated (Figure II-1). Post-menopausal aging had non-uniform effect on
46

mitochondrial gene expression in ovary intact rats, with MT-CYB expression significantly
decreased (Figure II-1).

Figure II-1. Mitochondrial gene expression during endocrinological and chronological aging. A,
mitochondrial gene expression during natural aging.

Nuclear encoded ETC Gene Expression
Contrary to the mitochondrial encoded ETC genes, nuclear encoded ETC subunits had decreased
expression at the onset of perimenopausal transition (from Reg 9 mo to Irreg 9 mo), but rebounded
as the endocrine transition completes (Irreg 9 mo to Acyc 9 mo). Chronological aging post-
menopausal led to decreased nuclear encoded ETC subunit gene expression across all complexes
(Figure II-2).
Reg 6 mo
Reg 9 mo
Irreg 9 mo
Acyc 9 mo
Acyc 15 mo
MT-ND1
MT-ND2
MT-ND3
MT-ND4
MT-ND4L
MT-ND5
MT-ND6
MT-CYB
MT-CO1
MT-CO2
MT-CO3
MT-ATP6
MT-ATP8
47

Figure II-2. Gene expression of nuclear encoded ETC subunits during endocrinological and chronological
transition.

Electron Transport Chain Complex Activities
Complex I NADH dehydrogenase enzyme activity followed gene expression of its nuclear
encoded subunits. Specifically, complex I activity significantly reduced from Reg 9 mo to Irreg 9
mo (p=0.01), but bounced back as animals completed their perimenopausal transition (p=0.04).
Post-menopausal aging is associated with a trend of reduced complex I activity (Figure II-3A).
On the other hand, while complex IV activity also had significant decrease from Reg 9 mo to Irreg
9 mo (p=0.02), it did not bounce back until Acyc 15 mo, instead of at Acyc 9 mo (p=0.04) (Figure
II-3B).
48

Figure II-3. Electron transport chain complex activity. A, complex I activity change throughout
endocrinological and chronological aging. B. Complex IV activity change throughout endocrinological aging.

Metabolomic Changes
We looked for metabolomic changes during the perimenopausal transition and aging in rat brain.
The most significant change occurred at the onset of perimenopausal transition, where we observed
systematic, significant down-regulation of glucose metabolism, especially glycolysis in the Irreg
9 mo group compared to the Reg 9 mo group. Fructose-6-phosphate, 3-phosphoglycerate,
phosphoenolpyruvate, and pyruvate levels were all significantly lower with p value smaller than
0.05, and glucose 6-phosphate and dihydroxyacetone phosphate were marginally significant with
p value smaller than 0.1 (Figure II-4 and Table II-8). We also surveyed changes in other metabolite
groups, including those involved in TCA cycle, peptides, carnitines, fatty acids, ketone body, lipids,
and sterols. Although changes in these groups were not significant, we did observe some
interesting trends in these metabolites. First, TCA cycle seemed the most active in regular 9 months
old animals. Similarly, amino acids levels were also the highest in the Reg 9 mo group. Fatty acids
and lipids are different from glycolysis and TCA cycle metabolites, but shared similar trends. For
them, the peri-menopausal transition was associated with reduced metabolite levels, and post-
49

menopausal aging was associated with slight accumulation of fatty acids and lipids. These changes
suggested increased utilization of these substrates as fuels during endocrinological aging. The
rebound during post-menopausal aging may be due to either reduced overall mitochondrial
respiratory capacity or adapted energetic circuit to provide more such fuel during post-menopausal
aging. Unlike fatty acids and lipids, ketone body was gradually depleted during these transitions.
On the contrary, acylcarnitine levels gradually built, either due to increased production or reduced
utilization capacity (Table II-8 and Figure II-4 and).
Table II-8. List of statistically significantly altered metabolites during endocrinological and chronological
aging.
Up-regulated Down-regulated
Reg 9 mo vs
Reg 6 mo
3-methylhistidine, homocarnosine, 1-methyl-
5-imidazoleacetate, oxalate (ethanedioate), N-
acetyltaurine, 4-acetamidobutanoate, alpha-
ketoglutarate, argininate, 4-
guanidinobutanoate, sphingadienine
2'-deoxyinosine, thymidine, 1-
methylnicotinamide, N2,N2-
dimethylguanosine, ribitol, gluconate, N,N,N-
trimethyl-5-aminovalerate, 1-
methylhistamine, 1-oleoyl-GPS (18:1), allo-
threonine
Irreg 9 mo vs
Reg 9 mo
Anserine, adenosine 2'-monophosphate (2'-
AMP), leucylglycine, ophthalmate, inositol 1-
phosphate (I1P), butyrylcarnitine (C4)
3-hydroxy-3-methylglutarate, 1-
carboxyethylphenylalanine, Pyruvate, N-
acetylaspartate (NAA), phosphoenolpyruvate
(PEP), glucosamine-6-phosphate, erythronate,
lysine, fructose-6-phosphate, 1-stearoyl-2-
arachidonoyl-GPS (18:0/20:4), uracil,
allantoin, N6-methyladenosine, phenyllactate
(PLA), 3-phosphoglycerate
Acyc 9 mo vs
Irreg 9 mo
phenylalanine 1-carboxyethyltyrosine
Acyc 15 mo vs
Acyc 9 mo
trigonelline (N'-methylnicotinate), (N(1) +
N(8))-acetylspermidine, urea,
ribulose/xylulose, acetylcarnitine (C2), 1,2-
dipalmitoyl-GPG (16:0/16:0), glycylleucine,
trimethylamine N-oxide, mead acid (20:3n9),
docosatrienoate (22:3n6), 1-
methylnicotinamide, N6-
carbamoylthreonyladenosine, adenosine 2'-
monophosphate (2'-AMP)
1,5-anhydroglucitol (1,5-AG), Campesterol,
beta-sitosterol, 1-carboxyethyltyrosine, chiro-
inositol, N-acetylleucine, N-acetylthreonine

50

Figure II-4.Change in metabolites during endocrinological and chronological aging. Metabolites were classified into those involved in glycolysis, TCA
cycle, peptides, carnitines, fatty acids, ketone body, lipids, and sterols.
Reg 6 mo
Reg 9 mo
Irreg 9 mo
Acyc 9 mo
Acyc 15 mo
1,5-anhydroglucitol (1,5-AG)
glucose
glucose 6-phosphate
fructose-6-phosphate
fructose 1,6-diphosphate/glucose 1,6-diphosphate
dihydroxyacetone phosphate (DHAP)
3-phosphoglycerate
phosphoenolpyruvate (PEP)
pyruvate
lactate
glycerate Glycolysis and Gluconeogenesis
Reg 6 mo
Reg 9 mo
Irreg 9 mo
Acyc 9 mo
Acyc 15 mo
citrate
aconitate [cis or trans]
alpha-ketoglutarate
succinylcarnitine (C4-DC)
fumarate
malate
2-methylcitrate/homocitrate
TCA Cycle
Reg 6 mo
Reg 9 mo
Irreg 9 mo
Acyc 9 mo
Acyc 15 mo
acetylcarnitine (C2)
hexanoylcarnitine (C6)
laurylcarnitine (C12)
myristoylcarnitine (C14)
palmitoylcarnitine (C16)
stearoylcarnitine (C18)
arachidoylcarnitine (C20)*
myristoleoylcarnitine (C14:1)*
palmitoleoylcarnitine (C16:1)*
oleoylcarnitine (C18:1)
eicosenoylcarnitine (C20:1)*
Acyl Carnitine
Reg 6 mo
Reg 9 mo
Irreg 9 mo
Acyc 9 mo
Acyc 15 mo
valerate (5:0)
caproate (6:0)
pelargonate (9:0)
caprate (10:0)
pentadecanoate (15:0)
palmitate (16:0)
margarate (17:0)
stearate (18:0)
nonadecanoate (19:0)
myristoleate (14:1n5)
palmitoleate (16:1n7)
10-heptadecenoate (17:1n7)
oleate/vaccenate (18:1)
10-nonadecenoate (19:1n9)
eicosenoate (20:1)
erucate (22:1n9)
tetradecadienoate (14:2)*
eicosapentaenoate (EPA; 20:5n3)
docosapentaenoate (n3 DPA; 22:5n3)
docosahexaenoate (DHA; 22:6n3)
docosatrienoate (22:3n3)
nisinate (24:6n3)
hexadecadienoate (16:2n6)
linoleate (18:2n6)
linolenate [alpha or gamma; (18:3n3 or 6)]
dihomo-linoleate (20:2n6)
dihomo-linolenate (20:3n3 or n6)
arachidonate (20:4n6)
docosatrienoate (22:3n6)*
adrenate (22:4n6)
docosapentaenoate (n6 DPA; 22:5n6)
docosadienoate (22:2n6)
mead acid (20:3n9)
(16 or 17)-methylstearate (a19:0 or i19:0)
Fatty acid (short chain, medium chain, long chain, and branched)
Reg 6 mo
Reg 9 mo
Irreg 9 mo
Acyc 9 mo
Acyc 15 mo
3-hydroxybutyrate (BHBA)
Ketone Bodies
Reg 6 mo
Reg 9 mo
Irreg 9 mo
Acyc 9 mo
Acyc 15 mo
choline
choline phosphate
cytidine 5'-diphosphocholine
glycerophosphorylcholine (GPC)
phosphoethanolamine
cytidine-5'-diphosphoethanolamine
glycerophosphoethanolamine
trimethylamine N-oxide
1-palmitoyl-2-oleoyl-GPS (16:0/18:1)
1-stearoyl-2-oleoyl-GPS (18:0/18:1)
1-stearoyl-2-arachidonoyl-GPS (18:0/20:4)
1,2-dipalmitoyl-GPG (16:0/16:0)
1-palmitoyl-2-oleoyl-GPG (16:0/18:1)
1-stearoyl-2-oleoyl-GPG (18:0/18:1)
glycerol
glycerol 3-phosphate
glycerophosphoglycerol
sphinganine
sphingadienine
phytosphingosine
sphingosine
Lipids
Reg 6 mo
Reg 9 mo
Irreg 9 mo
Acyc 9 mo
Acyc 15 mo
desmosterol
cholesterol
4-cholesten-3-one
beta-sitosterol
campesterol
7-hydroxycholesterol (alpha or beta)
corticosterone
Sterol
Reg 6 mo
Reg 9 mo
Irreg 9 mo
Acyc 9 mo
Acyc 15 mo
histidine
leucine
isoleucine
lysine
methionine
phenylalanine
tryptophan
valine
threonine
arginine
cysteine
glutamine
glycine
proline
tyrosine
alanine
aspartate
asparagine
glutamate
serine
Essential
Conditionally
Essential Non-essential
Amino Acids
51

RNA Sequencing
We first looked at genes that are involved in or known to affect different aspects of metabolism:
glycolysis, TCA cycle, ketogenesis, and fatty acid -oxidation (Yin et al., 2015). Expression of
genes involved in TCA cycle and ketogenesis uniformly and gradually decreased as animals went
through menopausal transition and aging, consistent with the gradually decreased levels of
involved metabolite (Figure II-5). This corresponded well with our metabolomic observations,
where metabolites of TCA cycle and ketogenesis decreased from Reg 6 mo to Acyc 15 mo overall
(Figure II-4).
On the other hand, while about two thirds of genes involved in glycolysis and fatty acid -oxidation
were gradually downregulated following menopausal transition and aging, a small portion of genes
were gradually upregulated during the process, almost like a compensatory effect (Figure II-5).
With a closer look, we observed that genes involved in fatty acid transportation and long-chain
fatty acid degradation were upregulated while those involved in short-chain and medium-chain
fatty acid metabolism were downregulated or unaltered during post-menopausal aging (Figure II-
5). The decreased expression of short-chain fatty acid metabolism genes and lower levels of short-
chain fatty acids, coupled with increased expression of long-chain fatty acid metabolism genes and
accumulated long-chain fatty acid level, again suggested either a shift to long-chain fatty acid
centric metabolic profile, or overall reduced ability of mitochondrial energy production during
post-menopausal aging (Figure II-4).
52

Figure II-5. Expression of genes involved in glycolysis, TCA cycle, ketogenesis, and fatty acid beta oxidation
throughout endocrinological and chronological aging.

RNA-Seq results suggested that the most prominent change during pre-menopausal aging from
Ret 6 mo to Reg 9 mo is increased inflammation. Based on predicted upstream regulators, this
process is also associated with reduced glucose metabolism, lipid metabolism, and mitochondrial
bioenergetics (Table II-9). During the perimenopausal transition, IPA predicted limited changes
in the transcriptome, with activation of MAPK signaling and alterations in DNA methylation
(Table II-9), potentially as a prelude for later transcriptome alternations. Indeed, post-menopausal
aging is associated with a lot of transcriptome alternations. IPA predicted increased STAT3
signaling, RXR/PPAR activation, EGF signaling, IGF1R expression, mTORC1 signaling,
ERK/MAPK signaling, cAMP/CREB signaling, cytoskeleton modulation, and androgen signaling;
reduced oxidative phosphorylation, TCA cycle activity, PTEN signaling, and cell proliferation;
and altered circadian clock, and calcium homeostasis. Upstream regulators predicted to be
significantly activated or inhibited are listed in Table II-9. Overall, the common theme of
endocrinological and chronological aging is increased inflammation, altered metabolism, and
ultimately reduced bioenergetics in the brain and potential risk of insulin resistance.
53

Table II-9. Upstream regulators predicted to be significantly activated or inhibited during endocrinological
and chronological aging.
Predicted Activation Predicted Inhibition

Reg 9 mo vs Reg 6 mo
MAVS, STAT1, IFNAR1, IRF7,
TICAM1, MAP4K4, IRF3, IFNB1,
IRF5,
TRAF3, TP73, SIRT1, IRGM, VEGF,
PPARD, SIRT1, TCF7L2, TP73
Irreg 9 mo vs Reg 9 mo MAPK8, DNMT3A, DNMT3B NONE
Acyc 9 mo vs Irreg 9 mo NONE DNMT3A

Acyc 15 mo vs Acyc 9 mo
RICTOR, CREBBP, PRKAR1A,
TSC1, MGEA5, SREBF1, TCF7L2,
KDM5A, DGAT2, DTNBP1, CNR1,
NR3C2, SYVN1, EP300, ARNTL,
A2M, EBF1, FAAH, TRAP1, NFASC,
JAK1/2, MKNK1
RB1, INSIG1, NFE2L2, INSIG2,
ELOVL5, AGPAT2, PTP4A3, NRF1,
IL24, NRG4, BTRC, LH, HTT, VBP1,
EPHA2

Effect of Ovarian Hormone on OXPHOS Gene Expression
In human, menopausal transition is associated with diminished ovarian hormone level. In rats,
although menopausal transition resulted in disrupted estrous cycle, there was still substantial level
of estradiol and progesterone in post-menopausal aged animals. To overcome this difference and
to more accurately mimic the post-menopausal aging in females, we ovariectomized 9-month-old
reproductively senescent rats, and aged them to 15 months old. We observed that compared to
Acyc 15 mo rats, OVX 15 mo rats had significantly reduced expression of OXPHOS genes, both
encoded by the mitochondria (Figure II-6) and by the nucleus (Table II-10).

Figure II-6. Change in mitochondrial encoded OXPHOS gene expression due to ovariectomy. Loss of ovarian
hormone significantly reduced gene expression of MT-ND4, MT-ND5, MT-ND6, MT-CO3, MT-ATP6, and
MT-ATP8, and had a trend to decrease expression of other subunits. * p<0.05 compared to Acyc 15 mo.
54

Table II-10. Change in nuclear encoded OXPHOS gene expression due to ovariectomy. Loss of ovarian
hormone uniformly reduced nuclear encoded OXPHOS gene expression, especially significantly reduced
expression of multiple genes for complex I, III, IV, and ATP synthase. Genes significantly down-regulated were
highlighted in green.
Complex I Fold change p Complex II Fold change p
NDUFA10 1.02 0.59 SDHA 0.96 0.34
NDUFA11 0.86 0.03 SDHB 0.93 0.08
NDUFA12 0.92 0.22 SDHC 0.93 0.20
NDUFA2 0.84 0.14 SDHD 0.96 0.41
NDUFA3 0.88 0.10 Complex III Fold change p
NDUFA4 0.89 0.12 UQCR10 0.87 0.05
NDUFA5 0.92 0.20 UQCR11 0.82 0.00
NDUFA6 0.88 0.07 UQCRC1 0.93 0.14
NDUFA7 0.90 0.09 UQCRC2 1.04 0.47
NDUFA8 0.84 0.00 UQCRFS1 0.92 0.06
NDUFA9 0.93 0.12 UQCRQ 0.81 0.01
NDUFAB1 0.94 0.37 Complex IV Fold change p
NDUFB10 0.85 0.05 COX4I1 0.88 0.03
NDUFB11 0.91 0.11 COX4I2 0.94 0.69
NDUFB2 0.79 0.00 COX5A 0.87 0.04
NDUFB3 0.86 0.05 COX5B 0.85 0.02
NDUFB4 0.90 0.15 COX6A1 0.90 0.06
NDUFB5 0.94 0.31 COX6A2 0.79 0.42
NDUFB6 0.90 0.12 COX6B1 0.92 0.23
NDUFB7 0.81 0.01 COX6B2 0.92 0.61
NDUFB8 0.92 0.16 COX7B 0.91 0.17
NDUFB9 0.84 0.01 COX8A 0.85 0.01
NDUFS1 1.03 0.53 Complex V Fold change p
NDUFS2 0.96 0.21 Complex V fold change p
NDUFS3 0.88 0.03 ATP5C1 1.02 0.75
NDUFS4 0.99 0.83 ATP5F1 1.01 0.79
NDUFS6 0.83 0.00 ATP5H 0.86 0.03
NDUFS7 0.85 0.01
NDUFS8 0.92 0.13
NDUFV1 0.89 0.01
NDUFV2 0.92 0.13
NDUFV3 0.84 0.00

We further observed that ovariectomy diminished peripheral estradiol but left progesterone level
unchanged (Figure II-7), suggesting that the reduction in OXPHOS gene expression may be linked
to estradiol but not progesterone. This is consistent with clinical hormone therapy studies where
combined estrogen and progesterone therapy did not seem to have better results than estrogen
replacement therapy alone (Espeland et al., 2013; Gleason et al., 2015).
55

Figure II-7. Change in ovarian hormone level due to ovariectomy. Left panel corresponds to estradiol level,
and right panel corresponds to progesterone level.

Discussion and Conclusion
In this study, we examined the dynamic metabolic changes in the female brain. One significant
turning point of female aging is the perimenopausal transition, which was proposed to be partly
responsible for the two-fold life-time risk of late onset Alzheimer’s disease (Brinton, 2008b;
Paganini-Hill and Henderson, 1994).
We observed that the mitochondrial encoded OXPHOS genes and nuclear encoded OXPHOS
genes exhibited different patterns during this transition. Mitochondrial encoded genes were
expressed at their highest level when animals transited from reproductive competent to
incompetent, before being down-regulated as the animals became reproductive senescent. On the
contrary, nuclear encoded OXPHOS genes were slightly down-regulated during the transition and
bounced back as the animals became reproductive senescent. This seemed to be a coordination
between the two genomes and a compensation in their gene expression, and that appropriate
expression of OXPHOS genes from both genomes were required for proper mitochondrial electron
transport chain function. While most of the gene expression changes did not reach statistical
significant, we believe this was due to the nature of this natural aging model. The value of this
56

study was not to demonstrate drastic differences between each aging stage, but the to reveal
systematic shift of multiple components serving the same function.
Our observation in post-menopausal ovariectomized rats further suggested that loss of ovarian
hormones, specifically estradiol but not progesterone, may be responsible for transcriptome
alternations. This was in line with previous studies from our lab demonstrating that the decline of
ovarian hormones during menopausal transition and aging led to a systemic deficit in glucose
metabolism and a shift to alternative fuel pathways, as evident by decreased neuronal glucose
transporter, compromised hexokinase activity, phosphorylated (inactivated) PDH, and increased
ketone body metabolism (Ding et al., 2013a; Ding et al., 2013b; Irwin et al., 2008; Klosinski et al.,
2015; Rettberg et al., 2014; Yao and Brinton, 2012; Yao et al., 2010; Yao et al., 2012a; Yao et al.,
2009; Yao et al., 2011c). The more prominent role of estradiol than progesterone in this study may
be due to the already reduced level of progesterone in post-menopausal animals compared to young,
reproductively competent animals.
We also observed reduced electron transport chain complex I and complex IV activity in the brain
of 9-month-old reproductive incompetent rats, consistent with our previous observation in a
separate cohort (Yin et al., 2015). We further observed that complex I activity temporarily
rebounded in 9 months old reproductive senescent rats, following the gene expression of nuclear
encoded OXPHOS subunits.
We then explored the effect of endocrinological and chronological aging on the metabolomics in
female brain. We observed that different fuel sources were differentially preferred at each stage of
aging, and the perimenopausal transition is a turning point of brain bioenergetics and metabolic
profile. Before the onset of perimenopausal transition, the female brain majorly utilized glucose
as its fuel. Compared to young, reproductively competent animals, middle aged, reproductively
57

active animals increased utilization of amino acids as fuel sources. Given the glycogenic nature of
multiple amino acids, pre-menopausal aging is also associated with increased gluconeogenesis, as
evident by increased cortical lactate and glycerate levels, as well as increased glucose-6-phosphate
and glucose-6-biphosphoate gene expression. As the endocrine transition starts, reduced electron
transport chain activity stimulated the brain to seek additional fuel source, thus fatty acids and
lipids were gradually consumed as energy source. Long-chain fatty acids were the first to be
consumed, followed medium-chain, then short-chain fatty acids. During post-menopause aging,
the mitochondria became even less efficient in energy production and fatty acid beta oxidation,
which led to increased production of long-chain fatty acids potentially in a feedback mechanism,
and as a result more accumulation of fatty acids (especially long-chain fatty acids) and lipids in
15-month-old animals. The effect is two-fold. First, it stimulated production of acylcarnitine,
which is a key step in transporting long-chain acyl-CoA into mitochondria for -oxidation and also
serves as a marker of incomplete -oxidation (Adams et al., 2009; Zhang et al., 2017). Second, as
a feedback mechanism it stimulated glucose metabolism as a compensation, which explained why
genes involved in glycolysis were up-regulated in aged reproductive senescent rats compared to
newly menopaused rats. However, deficiency in TCA cycle and electron transport chain activity
restricted energy production through oxidative phosphorylation, thus promoted anaerobic
respiration in the brain. The accumulation of glucose at this stage also indicates potential insulin
resistance. Throughout the whole transition, ketone body level was gradually diminished, due to
reduced ketogenesis and potentially increased consumption. The above proposed aging
metabolism profile fits well with both the metabolomic and the transcriptomic data. It is also
consistent with our previous studies and literature showing that brain aging is associated with
increased glucose hypometabolism and deficient mitochondrial bioenergetics (Boveris and
58

Navarro, 2008; Ding et al., 2013a; Ding et al., 2013b; Jones and Brewer, 2010; Maklashina and
Ackrell, 2003; Navarro and Boveris, 2007; Rettberg et al., 2014; Yao and Brinton, 2012; Yao et
al., 2010; Yao et al., 2009; Yao et al., 2011c; Yin et al., 2015). The shifting of fuel source during
endocrinological and chronological aging also echoes with our previous study demonstrating that
increased myelin degradation as an adaptive approach of the aging female brain to generate
alternative energy source (Klosinski et al., 2015; Yao et al., 2011c).
We are aware that we used cross sectional observations to infer longitudinal changes. However,
given the type of samples required for this kind of study, there are limited alternatives. Although
non-invasive imaging techniques such as PET may lend insight into brain metabolism status, they
are usually limited by the availability of radiotracers and lack the throughput necessary for a
comprehensive metabolism profile.
This study also raised more questions to be answered. The brain is composed of multiple cell types.
We know that neurons primarily rely on glucose as its fuel source (Bélanger et al., 2011; Schönfeld
and Reiser, 2013), whereas astrocytes are the primary source of fatty acid -oxidation and the only
producer of ketone body in the brain when glucose level is limited (Blazquez et al., 1999; Edmond
et al., 1998; Edmond et al., 1987). Similarly, astrocyte is also responsible for glycogen storage and
for lactate generation in the brain to support neurons and oligodendrocytes (reviewed by (Riske et
al.)). So what is the cell type-specific metabolic profile? Single cell RNA-Seq can help provide an
answer. Given the tight control in supply and demand in brain bioenergetics, how do different cell
types adapt, communicate with, and support each other? The communication between brain and
peripheral metabolomics should also be elucidated. Finally, since estrogen is a master regulator of
mitochondrial bioenergetics, does its regulatory effect differ among different cell types?
59

In summary, we demonstrated here that endocrinological transition and chronological aging in the
female brain is associated with differential regulation of mitochondrial and nuclear encoded
OXPHOS genes. We also observed ovarian estrogen as a potential regulator of OXPHOS gene
expression. While many studies have demonstrated aging is associated with declined brain
bioenergetics and shift in energy substrate, to our knowledge, this is the first study that mapped
out the bioenergetic changes during perimenopausal transition and aging in the female brain in
details. Given the parallel metabolic phenotype between aging female brain and prodromal AD,
our observations can further provide insight to preventative interventions and therapeutic windows.
Acknowledgement
I thank “Raymond” Yuan Shang for his assistance in generating DEG files from raw RNA-Seq
FASTQ files. I also thank Gregory Branigan for his assistance in complex I activity assay. This
study was supported by NIA 5P01AG026572 to RDB; Project 1 to RDB & EC.

60

CHAPTER III

CELL TYPE AND ESTROGEN RECEPTOR SUBTYPE SPEICIFIC CONTRIBUTION
TO ESTRADIOL REGULATION OF MITOCHONDRIAL GENE EXPRESSION

Abstract
Brain glucose hypometabolism and mitochondrial dysfunction are hallmarks of late onset
Alzheimer’s disease (AD). Estrogen was demonstrated to be a master regulator of mitochondrial
bioenergetics, and loss of estrogen due to menopause was implicated as a risk factor for females
in developing late onset AD. Estrogen replacement therapy in post-menopausal women is
associated with improved brain glucose metabolism and preserved cognitive function. Previous
studies demonstrated that estrogen was able to upregulate electron transport chain complex
activities and subunit protein expression, likely through regulation of their gene expression.
Mammalia mitochondria has its own genome and encodes 13 subunits of the electron transport
chain complexes, all of which are considered core subunits. The contribution and coordination of
mitochondrial and nuclear genome in response to estrogen treatment is not fully understood.
Further, given that the two major cell types in the brain – neurons and astrocytes have different
respiratory phenotype in response to estrogen treatment, and that estrogen receptor alpha (ER)
and estrogen receptor beta (ER) have different intracellular localizations in neurons, we
hypothesized that the regulatory effect of estradiol on mitochondrial gene expression is cell type
and estrogen receptor subtype-specific. Using rat embryonic neurons, astrocytes, human wildtype
neuroblastoma cells and neuronal differentiated cells, we demonstrate that estrogen regulates
mitochondrial gene expression in a cell type-dependent and ER subtype-specific manner, and that
transcriptomic changes can be translated to mitochondrial respiratory capacity differences. We
further surveyed the transcriptome and determined the role of ER and ER under both unliganded
61

and liganded conditions. Outcomes of this study provided further insight into development of safer
and more effective hormone therapies for Alzheimer’s disease.
Introduction
Alzheimer’s disease (AD) is the most common form of dementia, and affects more than 5 million
Americans, with a projected prevalence to triple by 2050 (Alzheimer's Association, 2018). While
the mechanism underlying late onset AD etiology remains undetermined, it is clear that during the
20-year prodromal period, mitochondrial dysfunction and brain glucose hypometabolism are
evident. In humans, the two-fold greater lifetime risk of developing late onset AD in females was
partially attributed to declined brain glucose metabolism and mitochondrial bioenergetics due to
menopause (Seshadri et al., 2006). Indeed, estrogen replacement therapy in postmenopausal
women has been shown to improve brain glucose metabolism and cognitive function (Kawas et
al., 1997; Maki, 2006; Maki et al., 2001; Paganini-Hill and Henderson, 1996; Resnick et al., 1998;
Shaywitz et al., 2003; Tang et al., 1996; Waring et al., 1999; Zandi et al., 2002), indicating a
protective effect of estrogen on brain bioenergetics. Similarly, in ovariectomized mice and rats,
estrogen therapy preserved mitochondrial respiration capacity and efficiency, supporting the role
of E2 as a master regulator of brain bioenergetics via mitochondrial function (Irwin et al., 2008;
Yao et al., 2012a).
On the mechanistic side, estrogen has been shown to regulate glucose metabolism, mitochondrial
respiration, and ATP production (Irwin et al., 2008; Yao et al., 2012a; Yao et al., 2009). Estrogen
can also regulate mitochondrial proteome in vivo, including both the nuclear and mitochondrial
encoded proteins (Nilsen et al., 2007). Furthermore, previous studies in our lab demonstrated that
in the female brain, during aging and menopausal transition, the decline of ovarian hormones leads
to a systemic deficit in glucose metabolism and a shift to alternative fuel pathways, as evident by
62

decreased neuronal glucose transporter, compromised hexokinase activity, phosphorylated
(inactivated) PDH, and increased ketone body metabolism (Ding et al., 2013a; Ding et al., 2013b;
Rettberg et al., 2014; Yao and Brinton, 2012; Yao et al., 2010; Yao et al., 2009; Yao et al., 2011c).
Our analysis from the previous chapter also suggested that loss of ovarian estrogen in post-
menopausal females led to significant downregulation of OXPHOS gene expression.
Interestingly, mammalian mitochondria have their own genome, which encodes 13 (~10%)
subunits for the electron transport chain complexes, all of which are considered core subunits. The
coordination of the mitochondrial and nuclear genome and their individual contribution to estradiol
regulation of brain bioenergetics were not fully elucidated, and several key aspects remain to be
investigated. First, previous data from our lab suggested that major brain cell types, such as
neurons and glial cells, have different bioenergetic demands and respiratory phenotypes following
E2 treatment (Yao et al., 2012a), indicating potential cell type specific regulation by E2. Second,
in rodent primary hippocampal neuron and neuronal cell lines, while both ERα and ERβ are present
and can promote mitochondrial respiration (Wu et al., 2011; Zhao and Brinton, 2007a), only ERβ
is specifically localized to mitochondria (Alvarez-Delgado et al., 2010; Chen et al., 2007; Chen et
al., 2009; Guo et al., 2012; Irwin et al., 2012; Yager and Chen, 2007; Yang et al., 2004; Yang et
al., 2009; Yao et al., 2013; Zhao and Brinton, 2005b, 2007a; Zhao et al., 2009; Zhao et al., 2013;
Zhao et al., 2011; Zhao et al., 2006), indicating potential differential contribution of ER and ER
in regulation of mitochondrial gene expression.
In this study, we confirmed that estrogen regulates mitochondrial gene expression in a cell type
specific and estrogen receptor subtype dependent manner. Using ER and ER specific
antagonists and transcriptome analysis, we further identified the role of these two nuclear ERs
under unliganded condition and in response to estradiol stimulation.
63

Materials and Methods
Animals
To determine the effect of ovariectomy (OVX) and estradiol replacement therapy on brain
mitochondrial gene expression, female Sprague Dawley rats were either sham ovariectomized
(Sham) or ovariectomized (OVX) at 6-months of age. OVXed rats were subcutaneously treated at
time of OVX with either vehicle or E2 via silastic tubes (30μg/kg/day) for 5. E2 treatment dosage
and duration were based on our extensive previous analyses documenting prevention of OVX-
induced decline in mitochondrial respiration in mice and rats (Ding et al., 2013a; Ding et al.,
2013b; Nilsen and Diaz Brinton, 2003; Nilsen et al., 2007; Yao et al., 2011b; Yao et al., 2012a;
Zhao et al., 2012).
All animal studies were performed following National Institutes of Health guidelines on use of
laboratory animals and all protocols were approved by the University of Southern California
Institutional Animal Care and Use Committee.
Cell culture
Primary neurons
Sprague Dawley rat embryonic day 18 (E18) hippocampal neurons were cultured in poly-D-lysine
coated culture vessels, with phenol red – free Neurobasal medium (Gibco, 12348017),
supplemented with 2% B-27 (Gibco, 17504044), 0.5mM L-glutamine (Gibco, 25030081), 25M
glutamate (MP Biomedicals, 02101800), and 10U/mL penicillin/streptomycin (Gibco, 15140122).
On in vitro day 4, glutamate was removed from culture medium. On in vitro day 10, neurons were
treated with desired reagents, and downstream assays were performed on in vitro day 11.
64

Primary astrocytes
Sprague Dawley rat embryonic day 18 (E18) hippocampal mixed glial cells were cultured in
phenol red – free DMEM/F-12 (Gibco, 11039021) supplemented with 10% FBS (ATCC, 30-2020).
Upon 80% confluency, astrocytes were enriched by shaking microglia and oligodendrocytes off
on an orbital shaker. Enriched astrocytes were then trypsinized, seeded onto poly-D-lysine coated
vessels, and allowed to recover. Astrocytes were starved in phenol red – free DMEM/F-12
supplemented with 10% charcoal stripped FBS (Gibco, 12676029) for 24 hours before treatment
with various reagents.
Human neuroblastoma cell line
Human neuroblastoma cell line SH-SY5Y was purchased from ATCC® (CRL-2266™). Cells
were maintained in DMEM/F-12 (Gibco, 11039021) supplemented with 10% FBS (ATCC, 30-
2020), and passaged at 80% confluency. Cells were starved in DMEM/F-12 supplemented with
10% charcoal stripped FBS (Gibco, 12676029) for 48 hours before treatment with various reagents.
Differentiation of human neuroblastoma cell line
To differentiate SH-SY5Y cells towards a more neuronal phenotype, cells were seeded onto poly-
D-lysine and laminin double coated vessels. After recovery overnight, on differentiation day 0,
cells were treated with 10M all-trans retinoic acids (Sigma-Aldrich, 2625) in DMEM/F-12
supplemented with 0.5% heat inactivated FBS for 5 days (Korecka et al., 2013; Kovalevich and
Langford, 2013b). Differentiation medium was replenished every other day. On differentiation day
5, culture medium was changed to starvation medium containing 10M all-trans retinoic acids in
DMEM/F-12 supplemented with 0.5% charcoal stripped FBS. On differentiation day 6, cells were
treated with reagents of interest and were ready for downstream assays and analysis.
65

In vitro treatment
To test for the effect of estradiol, cells will be treated with either vehicle (equally diluted EtOH, to
0.001%) or E2 (10nM for primary neurons and astrocytes, and 10ng/mL for SH-SY5Y cells) for
24 hours before downstream experiments. These concentrations were determined based on
previous experiments and functional assay outcomes (Irwin et al., 2011b; Yao et al., 2013).
To test for the contrition of estrogen receptor alpha (ERα) and estrogen receptor beta (ERβ), ERα
selective antagonist 1,3-Bis(4-hydroxyphenyl)-4-methyl-5-[4-(2-piperidinylethoxy)phenol]-1H-
pyrazole dihydrochloride (MPP) (TOCRIS, 1991), and ERβ selective antagonist IV-[2-Phenyl-
5,7-bis(trifluoromethyl)pyrazolo[1,5-a]pyrimidin-3-yl]phenol (PHTPP) (TOCRIS, 2002) were
used to inhibit the two ER subtypes respectively. Cells were pre-incubated with MPP (100nM) or
PHTPP (100nM) for 1 hour before being treated with vehicle or E2 for 24 hours. These
concentrations were selected based on literature (Adams et al., 2010; López-González et al., 2011;
Ma et al., 2016; Xing et al., 2016).
RNA isolation
To extract total RNA from rat hippocampal tissues, frozen hippocampus was directly homogenized
in TRIzol® Reagent (Invitrogen, 15596026) using The Bullet Blender® and silicon beads. To
extract total RNA from cultured cells, cells were lysed in TRIzol® Reagent. Chloroform was used
to extract RNA from the homogenate at a volume ratio of 1:5 to that of the TRIzol® Reagent.
Ethanol was then used to precipitate nucleic acids from the aqueous phase. RNA was further
purified using PureLink™ RNA Mini Kit (Invitrogen™, 12183018A) following manufacturer’s
instructions. Purelink™ DNase (Invitrogen™, 12185010) was used to eliminate DNA
66

contamination. Purified RNA was eluded in RNase-free, diH2O. RNA concentration and quality
were checked by NanoDrop™ One.
Real-time quantitative PCR
For rat primary neurons and astrocytes, gene expression of the 13 mitochondrially encoded genes
and mitochondrial transcriptional factors (TFAM, TFB1M, and TFB2M) was determined using
single tube Taqman gene expression assays (ThermoScientific, Rn03296764_s1, Rn03296765_s1,
Rn03296825_s1, Rn03296781_s1, Rn03296792_s1, Rn03296799_s1, Rn03296815_s1,
Rn03296746_s1, Rn03296721_s1, Rn03296734_s1, Rn03296820_s1, Rn03296710_s1,
Rn03296716_s1, Rn00667869_m1, Rn03296716_s1, Rn00580051_m1, Rn00710690_m1,
Rn01412502_m1). RNA was converted to cDNA using SuperScript VILO cDNA Synthesis Kit
(ThermoFisher, 11754050). A total of 6.25ng of cDNA was used per rt-PCR reaction in TaqMan™
Universal PCR Master Mix (Applied Biosystems, 4304437). Target mRNA was amplified using
the Applied Biosystems™ QuantStudio™ 12K Flex system.
For SH-SY5Y cells, gene expression of 8 representative mitochondrially encoded genes was
determined using single tube Taqman gene expression assays (ThermoScientific, Hs02596873_s1,
Hs02596874_g1, Hs02596878_g1, Hs02596879_g1, Hs02596867_s1, Hs02596864_g1,
Hs02596866_g1, Hs02596863_g1, Hs99999903_m1, and Hs02758991_g1). RNA was converted
to cDNA using SuperScript VILO cDNA Synthesis Kit. A total of 12.5ng of cDNA was used per
rt-PCR reaction in TaqMan™ Universal PCR Master Mix (Applied Biosystems, 4304437). Target
mRNA was amplified using the Applied Biosystems™ QuantStudio™ 6 Flex system. Relative
gene expression level (fold change) to reference group was calculated by the comparative Ct
(ΔΔCt) method. Statistical significance was calculated by ANOVA followed by non-paired t-test.
67

RNA-Sequencing (RNA-Seq)
RNA-Seq was conducted at Vanderbilt Technologies for Advanced Genomics (VANTAGE). Only
RNA samples with an acceptable RNA quality indicator score (RQI >7) was used for sequencing.
mRNA enrichment and cDNA library preparation were done using a stranded mRNA (poly(A) -
selected) sample preparation kit. Sequencing was performed at 150bp paired-end on NovaSeq600,
targeting 30 million reads per sample. Transcripts were mapped to rat genome (ensemble release
90) using kallisto 0.4.3 (Bray et al., 2016), or to human genome using (ensemble release 92) using
salmon 0.9.1 for human (Patro et al., 2017). Tximport V1.6.0 (Soneson et al., 2015) was used to
generate a counts table from kallisto/salmon output, and DESeq2 V1.18.1 (Love et al., 2014) was
used to calculate normalized read counts for each gene and/or transcript and to perform expression
analysis.
Ingenuity Pathway Analysis (IPA) of RNA-Seq
Differentially expressed gene lists were processed using the core analysis function of IPA. Only
genes with p value smaller than 0.05 was considered (or the top 8000 most significant genes, as
per limit of IPA analysis). The outputs are lists of significantly altered canonical pathways and
upstream regulators. The canonical pathways are identified based on enrichment of qualified genes.
The upstream regulator analysis predicted activation or inhibition of regulatory molecules based
on expression of respective downstream genes and networks compiled from literature and IPA’s
Ingenuity knowledge base.
Comparison analysis was used to determine similarities and differences among multiple age and
endocrine groups. Top canonical pathways and upstream regulators were also identified.

68

Metabolic flux assay
Cellular respiratory capacity will be determined by XF24 metabolic flux analyzer (for primary
neurons and astrocytes) or XF96e metabolic flus analyzer (for SH-SY5Y cells) as previously
described (Irwin et al., 2011; Irwin et al., 2012; Yao et al., 2013). Briefly, cells were seeded into
appropriately coated cell plates at desired density (50,000/well for primary neurons, 75,000/well
for primary astrocytes, and 20,000/well for wildtype or differentiated SH-SY5Y cells), cultured
and treated as described above. On the day of assay, culture medium was changed to unbuffered
DMEM (Sigma-Aldrich, D5030) medium supplemented with 25mM glucose, 1mM sodium
pyruvate, and 2mM GlutaMAX for primary cells (Gibco, 35050061) or 2mM glutamine for SH-
SY5Y cells (Gibco, 25030081). Medium pH was adjusted to 7.4. Cells were incubated at 37°C in
a non-CO2 incubator for 1 hour. Oxygen consumption rate (OCR) was used as an indicator of
mitochondrial oxidative phosphorylation, and mitochondrial basal respiration, ATP production,
maximal respiratory capacity, and proton leak were determined by sequential acute injection of
mitochondrial electron transport chain inhibitors and un-couplers: oligomycin (MP Biomedicals,
02151786), FCCP (carbonyl cyanide4(trifluoromethoxy)- phenylhydrazone) (TOCRIS Bioscience,
0453), rotenone (MP Biomedicals, 02150154), and antimycin (Sigma-Aldrich, A-8674). For
primary neurons and astrocytes, 4M oligomycin, 1M FCCP, and 1M rotenone/antimycin were
used. For wildtype and differentiated SH-SY5Y cells, 1M oligomycin, 0.5M FCCP, and 1M
rotenone/antimycin were used.
Results were normalized to protein reading of each plate to eliminate variation due to cell seeding
and loss during assay procedures.

69

Statistical analysis
Statistical significance for each assay was determined by one-way ANOVA followed by unpaired
t-test. Comparisons with p value smaller than 0.05 were considered statistically significant.
Results
Ovarian Hormone and Mitochondrial Gene Expression
We showed in previous study that loss of ovarian hormone in post-menopausal rats led to reduced
mitochondrial gene expression. In this study, we made similar observation in pre-menopausal,
reproductive competent rats, where ovariectomy in regularly cycling 9 months old rats
systematically reduced mitochondrial gene expression and reached statistical significance in
multiple genes (Figure III-A). In 6 months old reproductive competent rats, however, loss of
ovarian hormones did not result in such significant changes, although restoration of peripheral
estradiol level tended to increase mitochondrial gene expression in OVXed rats (Figure III-B).

70

Figure III-1. effect of ovariectomy and estradiol replacement treatment on hippocampal mitochondrial gene
expression. A, ovariectomy in reproductively competent, middle-aged rats reduced mitochondrial gene
expression. B, ovariectomy in reproductively competent young rats did not have significant effect on
mitochondrial gene expression, although restoring estradiol level still tend to increase mitochondrial gene
expression. *, p<0.05 to sham, #, p<0.05 to OVX.
A closer look at both serum and cortical estradiol revealed a sharp reduction in cortical estradiol
level from 6-month-old female rats to 9-month-old rats (Figure III-2B). We also looked into the
gene expression of aromatase (CYP19A1) in the hippocampus, where rtPCR revealed very low
expression level of this gene, and RNA-Seq revealed a 5% non-significant upregulation of its
expression during pre-menopausal aging, consistent with a pattern of compensatory effect for the
drop of cortical estrogen level. No the other hand, serum estradiol level remained largely
unchanged (Figure III-2A). These observations suggested that in young females, the effect of
peripheral estradiol may be secondary to the endogenous estradiol level in the brain. As a result,
loss of peripheral estradiol may not have a significantly negative impact on brain mitochondrial
71

gene expression, but addition of estradiol may still have a positive impact on mitochondrial gene
expression.

Figure III-2. Change of estradiol level in serum and cortex from reproductively competent young rats to
reproductively competent middle-aged rats. A, serum estradiol level steadily increased during the transition.
B, cortical estradiol level significantly dropped during the transition. *, p<0.05.

Cell Type-Specific Contribution to Mitochondrial Respiratory Capacity
Neurons
In primary embryonic rat hippocampal neurons, we observed that regardless of treatment, neurons
had high reserved respiratory capacity. The reserved (spare) respiratory capacity was further
increased following 24-hour estradiol treatment, along with increased basal respiration, ATP
production, and mitochondrial maximal respiratory capacity. ER antagonist MPP did not block
the effect of estradiol on mitochondrial respiration (Figure III-3A and Figure III-3B). On the other
hand, ER antagonist PHTPP treatment successfully prevented E2 mediated enhancement of
mitochondrial respiration, and further led to significantly reduced mitochondrial respiratory
capacity compared to vehicle treated control (Figure III-3A and Figure III-3B). Neither MPP nor
PHTPP treatment alone could alter mitochondrial respiratory capacity (Figure III-3C and Figure
III-3D).
72

Figure III-3. Estrogen receptor subtype-specific effect on estradiol-mediated increased in mitochondrial
respiratory capacity in neurons. A, raw oxygen consumption rate showing MPP and PHTPP antagonizing E2
effect. B, calculated categorized mitochondrial respiratory capacities demonstrating MPP and PHTPP
antagonizing E2 effect. C and D, raw data and calculated categorized mitochondrial respiratory capacities
respectively showing no effect of MPP or PHTPP alone on mitochondrial respiratory capacity. * p<0.05
compared to control, error bar shows standard error of the mean.

Astrocytes
Unlike neurons, enriched rat embryonic hippocampal astrocytes had barely any reserved
mitochondrial respiratory capacity. They tended to function at their maximal respiratory capacity
73

even under normal growth condition (Figure III-4). Although 24 hours of estradiol treatment still
increased mitochondrial basal respiratory capacity and ATP production capacity, it had limited
effect on maximal respiratory capacity and spare capacity (Figure III-4). Unlike in neurons, where
estradiol mediated mitochondrial respiration primarily through ER, in astrocytes, both ER and
ER shared the responsibility. Inhibiting either ER or ER blocked the ability of estradiol to
significantly increased mitochondrial basal respiration and ATP production (Figure III-4A and
Figure III-4B). We also show that neither of the drugs alone could significantly alter mitochondrial
respiration in astrocytes. Although MPP seemed to increased ATP production in astrocytes, this
can be attributed to reduced proton leak in MPP treated cells (Figure III-4C and Figure III-4D).

74

Figure III-4. Estrogen receptor subtype-specific effect on estradiol-mediated increased in mitochondrial
respiratory capacity in astrocytes. A, raw oxygen consumption rate showing MPP and PHTPP antagonizing
E2 effect. B, calculated categorized mitochondrial respiratory capacities demonstrating MPP and PHTPP
antagonizing E2 effect. C and D, raw data and calculated categorized mitochondrial respiratory capacities
respectively showing no effect of MPP or PHTPP alone on mitochondrial respiratory capacity. * p<0.05
compared to control, error bar indicates standard error of the mean.

75

Estrogen Receptor Subtype-specific Contribution to Estradiol Induced Mitochondrial Gene
Expression
Neurons
In neurons, estradiol treatment systematically and significantly increased mitochondrial encoded
OXPHOS gene expression. Neither ER antagonist MPP nor ER antagonist PHTPP was able to
block estradiol induced mitochondrial gene expression (Figure III-5A). We observed that this may
be due to the net effect of MPP and PHTPP alone on mitochondrial gene expression (Figure III-
5B). However, both MPP and PHTPP are well documented to not activate genes downstream of
ER and ER respectively. Given the unexpected net effect of MPP and PHTPP on mitochondrial
gene expression, we investigated at the expression of three major mitochondrial transcriptional
factors: TFAM, TFBIM, and TFB2M, as well as NRF1, which is known to regulate mitochondrial
biogenesis. We observed that neurons pretreated with MPP or PHTPP displayed comparable
express pattern of mitochondrial transcriptional factors compared to that of mitochondrial
OXPHOS genes (Figure III-5C). MPP treatment without estradiol was also able to significantly
upregulate all three mitochondrial transcriptional factors, explaining the up-regulation of
mitochondrial OXPHOS genes observed (Figure III-5D). However, PHTPP by itself had no effect
on the gene expression of mitochondrial transcriptional factors (Figure III-5D), suggesting that the
upregulation of mitochondrial OXPHOS genes when treated with PHTPP alone may be due to a
different signaling pathway.
76

Figure III-5. Effect of ER and ER on mitochondrial gene expression in response to estradiol in neurons. A,
antagonizing effect of MPP and PHTPP on gene expression of mitochondrial OXPHOS genes. B, net effect of
MPP and PHTPP on mitochondrial gene expression. C, antagonizing effect of MPP and PHTPP on gene
expression of mitochondrial transcriptional factors. D, net effect of MPP or PHTPP on gene expression of
mitochondrial transcriptional factors. * p<0.05.
77

Astrocytes
In astrocytes, estradiol treatment had no significant effect on mitochondrial OXPHOS gene
expression. As a result, no antagonizing effect of MPP or PHTPP was observed (Figure III-6A).
These observations supported the ineffectiveness of estradiol on astrocytic mitochondrial reserved
respiratory capacity. Interestingly, while we observed significant net effect of MPP on
mitochondrial gene expression in neurons, it did not seem to have any effect on mitochondrial
gene expression in astrocytes (Figure III-6B), suggesting the upregulation of mitochondrial gene
expression by MPP alone in neurons was not due to an artifact of MPP acting as an ER agonist.
Similarly, nor did PHTPP had any significant effect on mitochondrial gene expression in astrocytes,
similar to what we observed in neurons (Figure III-6B).

Figure III-6. Effect of ER and ER on mitochondrial gene expression in response to estradiol in astrocytes.
A, antagonizing effect of MPP and PHTPP on gene expression of mitochondrial OXPHOS genes. B, net effect
of MPP and PHTPP on mitochondrial gene expression.
78

Estrogen and Wildtype SH-SY5Y
Mitochondrial Respiratory Capacity
Undifferentiated SH-SY5Y cells have a respiratory phenotype similar to that of astrocytes, with
very little spare respiratory capacity. Estradiol treatment did not seem to alter basal mitochondrial
respiratory capacity but was able to increase its reserved capacity (Figure III-7A). Also mirroring
our observations in astrocytes, inhibiting either ER or ER diminished the positive effect of
estradiol (Figure III-7A). Neither MPP or PHTPP alone had any positive effect on mitochondrial
respiratory capacity (Figure III-7B).

Figure III-7. Estrogen receptor subtype-specific effect on estradiol-mediated increased in mitochondrial
respiratory capacity in neuroblastoma cells. A, estradiol did not affect basal respiration but increased spare
capacity. B. MPP and PHTPP had no net effect on mitochondrial respiration. * p<0.05 compared to control,
error bar indicates standard error of the mean.

Effect of Estradiol
Estradiol treatment did not significantly alter OXPHOS gene expression in wildtype SH-SY5Y
cells, either mitochondrial encoded or nuclear encoded, although the subunits tended to be
upregulated. In accordance, PGC1a was significantly upregulated. The single most prominent
effect of estradiol on WT SH-SY5Y cells was promotion of proliferation. Estradiol induced
79

expression of MYC, which in turn activated cyclin A, cyclin D, cyclin E, CDK2, and CDC4, to
promote mitosis. This observation was supported by activation of upstream regulators involved in
transcription and translation (EIF4E, E2F3, FOXM1, EP400), cell cycle progression (S100A6),
cytoskeleton and chromatin remodeling (ACTL6A), as well as oncogenes and genes associated
with cancer (SPDEF, MYC, ERBB2, ELAVL1, RABL6). In the meanwhile, estradiol suppressed
expression of upstream regulators with opposite effects (Brd4, CDKN1A, PAX6, TGFB1,
KDM5B and TP53). On the other hand, while IPA predicted significant activation of both ER
and ER following estradiol treatment (Figure III-8), signaling pathways associated with estrogen
receptor activation induced bioenergetics such as ERK, MAPK, PI3K, and AKT signaling
pathways were not significantly affected. Together, these observations suggested that the lack of
effect of estradiol to increase mitochondrial basal respiratory capacity was due to its preferential
effect on mitosis.

Figure III-8. In wildtype SH-SY5Y cells, estradiol treatment activated both ER and ER.

80

Ligand-independent Effect of ER
MPP was designed to block ER transcriptional activity through EREs, non-conventional EREs,
half EREs, or tethering to DNA via other proteins (TGFβ3 and AP-1), potentially by blocking its
binding site for co-transcriptional factors (Harrington et al., 2003).
In wildtype SH-SY5Y cells, transcriptome analysis revealed that gene expression of ESR1 was
non-significantly reduced, and that ER was not predicted to be activated. In line with these
observations, multiple co-activators (SP1, p300, AP1) associated with ER at half EREs were
significantly upregulated, supposedly due to a feedback mechanism. Non-genomic action of
estradiol via signaling pathways including AMPK signaling, ERK1/2 signaling, JNK/MAPK
signaling, and CREB signaling were also predicted to be inhibited. IPA predicted significant
inhibition of multiple growth hormone signaling pathways, including insulin, EGF, and PDGF
signaling (as evident by significant repression of IGF2, IGFR, INSR, EGFR, and PDGFR
expression).
However, MPP treatment significantly upregulated expression of majority of nuclear encoded
OXPHOS genes but not mitochondrial OXPHOS gene expression. Consistently, TFAM
expression was not altered. One possibility is that ER had some genomic activity even at normal
culture condition, and the upregulation of nuclear OXPHOS gene expression was a compensatory
response to disruption of such activities. Transcriptome analysis suggested that MPP treatment led
to significant increase of cellular and oxidative stress (PCGEM1, NFE2L, XBP1, TFEB, EIF2AK3,
SYVN1), mitochondrial fragmentation (FIS1), cytochrome c level and activation of apoptosis
(increased CASP3, AIF, BID, BAX, DIABLO, ENDOG, and decrease of BCL-2). In response, the
NRF2 antioxidation signaling pathway was activated, and ESRRA, NRF1, mTOR, PPARGC1A
81

were significantly upregulated as alternative mechanisms, and increased OXPHOS gene
expression to maintain cellular bioenergetic homeostasis.
Ligand-independent Effect of ER
Unlike MPP, PHTPP alone had no significant effect on the expression of either mitochondrial or
nuclear encoded OXPHOS genes, while tended to decrease their expression. Consistent with these
observations, PHTPP significantly downregulated PGC1 expression while had no significant
effect on signaling pathways associated with bioenergetics, such as AMPK, ERK1/2, CREB, and
PI3K signaling pathways. PHTPP also reduced translation machinery (EIF1A, EIF2B, EIF4A,
ribosome 40s, ribosome 60s), and altered cell cycle progression (increased cyclin D1, CCND1,
ATF1, MYC, P73, CDK2, PLN1, BRCA1, and RXRA, downregulation of CAV1, CCND2, and
CDKN2B) accompanied by altered cytoskeleton and extracellular matrix modulation
(upregulation of CDH1, downregulation of collagen, integrin, fibronectin, ACTC1, PDGFA,
PDGFB, PDGFC), suggesting a potential ligand-independent role in cell proliferation and cell
growth (downregulation of BDNF and GDNF). Blocking ER genomic action also seemed to
reduce inflammation (downregulation of TRL4, TNFRSF1B, BTK, HLA-DRA). Further, PHTPP
significantly reduced expression of EGF, GFG, PDGF, and TGF, suggesting an interaction of ER
at unliganded state with growth factor signaling.
Role of ER in Estradiol Response
In the presence of estradiol, blocking the effect of estradiol by ER antagonist MPP significantly
reduced expression of multiple mitochondrial encoded OXPHOS genes (MT-ND1, MT-ND3, MT-
ND4, MT-CYB, MT-CO1, MT-CO2, and MT-CO3), while non-significantly but systematically
reduced expression nuclear encoded OXPHOS genes. This is consistent with significant
82

downregulation of PGC1 upregulation of RPTOR, a negative regulator of mTOR, and
significant activation of RICTOR. This is also consistent with functional assay outcomes.
Signaling pathways and upstream regulators associated with bioenergetics such as p38 MAPK,
PI3K, MAP2K1/2 were non-significantly inhibited, while IGF1, AMPK, and CREB were not
significantly altered. Beyond mitochondrial bioenergetics, inhibiting ER mediated estradiol
response reduced extracellular matrix and cytoskeleton modulation (downregulation of collagen,
laminin, integrin, FN1, and F-actin and G-actin), reduced transcription and translation
(downregulation of EIF1, EIF2B, EIF4A, EIF4E, ribosome 40s, ribosome 60s), and potentially
reduced proliferation (upregulation of TP73, downregulation of MYC). Counteracting estradiol
mediated increase of EGF level, MPP significantly reduced EGF expression, as well as PDGF and
FGF expression. IPA further predicted inhibition of EGF, TGF, and insulin receptor signaling.
Role of ER in Estradiol Response
Blocking the effect of estradiol by ER antagonist PHTPP led to non-significantly but
systematically decreased expression of both mitochondrial and nuclear OXPHOS genes. This
observation was consistent with significant downregulation of PGC1, which was significantly
upregulated following estradiol treatment. Similar to inhibition of ER, inhibition of ER also
significantly activated RICTOR. Significant activation of TSC1 and TSC2 indicated potential
inhibition of mTOR signaling. Inhibiting ER also significantly decreased EGF, PDGF, and FGF
levels. IPA further predicted inhibition of signaling pathways associated with EGF, PDGF, growth
hormone, insulin, IGF1, IGF2, and TGF1 signaling. PI3K signaling and AKT signaling were also
predicted to be inhibited, although not statistically significant.
83

Similar to ER inhibition, ER inhibition also significantly reduced cytoskeleton and extracellular
matrix modulation (downregulation of collagen, laminin, integrin, FN1, and F-actin and G-actin),
and reduced translation (downregulation of EIF1, EIF4E, ribosome 40s). However, inhibiting ER
did not seem to affect cell cycle or proliferation as observed in ER inhibition.
Blocking either ER or ER in the presence of estradiol was associated with activation of estrogen
receptor (as a whole) and MAPK1, although gene expression levels were not significantly altered.
The prediction also incorporated many contradictory findings, a profile consistent with inhibition
of one ER subtype and compensation from another. Further, while inhibiting either ER or ER
counteracted on estradiol mediated increase in mitochondrial respiratory capacity, ER seemed to
exert its effect more on the non-genomic side, interacting with various growth factors. Selective
upstream regulators related to bioenergetics that were differentially affected by ER or ER were
listed in Table III-1.
Table III-1. Upstream regulator activities differentially activated or inhibited by ER or ER selective
antagonists. Red indicates upregulation and green indicates downregulation. Bolded regulators were predicted
to reach statistical significance, whereas regular font indicates non-significant changes. * indicates significant
activation by PHTPP but non-significant activation by MPP.
Upregulated Downregulated
Common
regulators
RICTOR, estrogen receptor, TSC2,
MAPK1*, TSC1*
PGC1, EGFR, TGFB1, INSR, ERBB2,
PI3K
ER only SIRT3, CREBBP, PSEN1, TP73 TGF, TGFBR1, RB1, ESRRB, p38 MAPK,
MAP2K1/2, PDGFBB
ER only

VEGF, PDGF, growth hormone, IGF2BP1,
JUN, ESRRG, PI3KR, IGF1, GPER, AKT1,
ESRRA, ESR1

84

Neuronal Differentiation of SH-SY5Y Cells
SH-SY5Y is a widely used human neuroblastoma cell line for its neuronal like characteristics, and
literature also suggested that wildtype (WT) SH-SY5Y cells can be differentiated towards a more
mature cholinergic neuronal phenotype (Diff) by all-trans retinoic acid treatment. Our RNA-Seq
data confirmed significantly decreased expression of immature neuronal markers such as
NEUROD1 and STMN1, and increased expression of mature neuronal markers including NEFM,
SYN, and DLG4. Retinoic acid treated SH-SY5Y cells also had significantly increased
cholinesterase expression level, accompanied by a trend of increased level of choline o-
acetyltransferase and vesicular acetylcholine transporter, supporting a differentiation lineage
towards cholinergic neurons. However, the differentiation is not complete, as evident by non-
changed Choline O-Acetyltransferase (CHAT) expression and strong presence of doublecortin, a
marker of immature neurons (Table III-2).
Table III-2. Gene markers of immature neurons, mature neurons, and cholinergic neurons. Retinoic acid
differentiated SH-SY5Y cells towards mature cholinergic neurons. Significantly upregulated genes were
highlighted in red, and significantly down-regulated genes were highlighted in green.
Gene Fold change p
Immature neuronal markers
DCX 3.07 6.47E-86
NEUROD1 0.26 4.85E-41
STMN1 0.62 1.15E-33
TUBB3 0.96 0.34
TBR1 1.55 0.43
Mature neuronal markers
DLG4 1.84 3.07E-43
NEFM 1.35 9.33E-07
SYP 2.16 9.75E-102
MAP2 1.16 1.25E-03
NEFH 0.60 4.52E-39
RBFOX3 0.42 1.57E-03
Cholinergic Neuron Marker
ACHE 7.24 2.44E-292
SLC18A3 1.24 1.46E-03
CHAT 1.42 0.18

85

Compared to wild type SH-SY5Y cells, retinoic acid differentiated cells had systematically
increased expression of mitochondrial encoded OXPHOS genes and decreased expression of
nuclear encoded OXPHOS genes. Out of 10 RNA-Seq detected mitochondrial subunits, 8 were
significantly upregulated, and out of 77 nuclear encoded OXPHOS genes detected, 70 were
significantly down-regulated (Figure III-9).

Figure III-9. Neuronal differentiation of SH-SY5Y resulted in upregulation of mitochondrial encoded
OXPHOS genes (highlighted in blue outline) and down-regulation of nuclear encoded OXPHOS genes. Red
indicates upregulated genes and green indicates down-regulated genes.

Estrogen and Differentiated SH-SY5Y
Mitochondrial Respiratory Capacity
Retinoic acid differentiation of SH-SY5Y cells led to increased spare capacity in comparison to
wildtype cells (from ~10% to ~60%), a respiratory phenotype closer to that of neurons. Estradiol
treatment systematically increased mitochondrial basal respiration, ATP production, maximum
respiratory capacity, and spare respiratory capacity, although these changes did not reach statistical
significance (Figure III-10A). Mirroring our observations in neurons, only ER inhibition was able
to counteract the effect of estradiol (Figure III-10A). Neither MPP or PHTPP alone had any
positive effect on mitochondrial respiratory capacity (Figure III-10B).
86

Figure III-10. Estrogen receptor subtype-specific effect on estradiol-mediated increased in mitochondrial
respiratory capacity in neuronal differentiated SH-SY5Y cells. A, estradiol tended to increase all aspects of
mitochondrial respiratory capacity, although effect was not statistically significant. Inhibition of ER
significantly counteracted effect of estradiol. B. MPP and PHTPP had no net effect on mitochondrial
respiration. # p<0.05 compared to estradiol, error bar indicates standard error of the mean.

Effect of Estradiol
In differentiated SH-SY5Y cells (Diff cells), estradiol treatment did not significantly alter
mitochondria or nuclear OXPHOS gene expression. While estradiol tended to upregulate
mitochondrial encoded subunits, it had mixed effect on nuclear encoded subunits. This was in
agreement with the non-significant effect of estradiol on functional assays.
In comparison to the effect of estradiol on wildtype SH-SY5Y cells, the scope is much limited in
differentiated cells. Interestingly, the gene expression of ER was significantly decreased. The
effect of estradiol can be classified into two functions. First, it mildly increased glucose
metabolism through predicted activation of mTOR signaling and IGF1 signaling, supported by
activation of PIK3R1, FOXO1, and predicted inhibition of RICTOR. Second, it promoted cellular
homeostasis through activation of unfolded protein response to promote cell survival instead of
apoptosis (upregulation of BIP, XBP1, ATF6, ERN1, HSP90B1, CREB3, EIF4B, E2F2S3, EIF1,
RPS6, and downregulation of TP53), and telomere maintenance (TERF, POT1).
87

Ligand-independent Effect of ER
As a proof of concept, IPA predicted inhibition of ER signaling.
Unlike in neurons and wildtype SH-SY5Y cells, in differentiated SH-SY5Y cells, MPP did not
systematically significantly upregulate OXPHOS gene expression. While mitochondrial encoded
subunits for complex I and IV had a trend to be upregulated, nuclear encoded subunits had mixed
response. While PGC1a was significantly upregulated, its regulatory effect was not predicted to
be significantly activated, which may explain the inconsistent trend of OXPHOS gene expression.
On the other hand, HDAC6 was predicted to be significantly activated. This molecule is known to
deacetylate mitochondrial proteins and serve as a positive regulator of mitochondrial gene
expression (Jagannathan et al., 2012). Together, this helped explain the more uniform upregulation
of mitochondrial encoded OXPHOS genes compared to the nuclear encoded counterparts.
Unlike what we observed in wildtype SH-SY5Y cells, blocking unliganded ER did not lead to
alternations in co-transcriptional factor or growth factors expression. The effect of MPP on
differentiated SH-SY5Y cells can be summarized as decreased ketogenesis (downregulation of
HMGCS2), phospholipid secretion (downregulation of ABCC2), reduced cholesterol
transportation (downregulation of RON1, APOA1, HDL, LDL, SLCS1A), accompanied by
increased gluconeogenesis (upregulation of HNF4A, PGC1a, G6PC). While downstream gene
expression-based analysis did not predict PKC inhibition, its expression was significantly
downregulated,
Ligand-independent Effect of ER
PHTPP alone did not have uniform or significant effect on either mitochondrial or nuclear
OXPHOS gene expression. The few significantly upregulated genes include upregulated nuclear
88

subunits NDUFA4, NDUFA6, NDUFS4, NDUFV3, SDHA, UQCRB, COX5B, COX7B, COX7C,
ATP5ME, ATP5MF, ATP5PB, ATP5F1E, and significantly downregulated mitochondrial MT-
CYB. PGC1 expression was not significantly altered.
Similar to inhibition of unliganded ER, inhibition of unliganded ER also led to upregulation of
genes involved in glycolysis (G6PC, PCK1) and downregulation of ketogenesis (HMGCS2) as
well as LDL level. RICTOR was also predicted to be significantly inhibited. Inhibition of
unliganded ER also led to activation of EGF signaling, similar to what we observed in wildtype
SH-SY5Y cells. L-type calcium channel was also significantly downregulated.
Of interest, PHTPP alone inhibited activity of LONP1, which is a mitochondrial targeted protein
involved in degradation of misfolded, unassembled or oxidatively damaged polypeptides in the
mitochondrial matrix. It was also known to regulate expression of mitochondrial genome. Together
with non-significantly upregulated PGC1, this might explain the pattern of mitochondrial gene
expression observed here.
Role of ER in Estradiol Response
In the presence of estradiol, inhibiting ER did not seem to affect OXPHOS gene expression,
except upregulation of a few nuclear encoded subunits. In accordance, PGC1 and PPAR were
not significantly altered. This matches the changes in mitochondrial respiratory capacity. Further,
antagonizing ER in differentiated SH-SY5Y cells seemed to antagonize the differentiation
process and promote mitosis (activation of TGF, RABL6, PDGF, ELAVL1, AREG, CDKN1A,
KDM5B). Similar to what we observed in wildtype SH-SY5Y cells, MPP inhibition significantly
upregulated EGF, which might be in response to activation of PELP1, an estradiol-dependent
coactivator of nuclear estrogen receptors but co-repressor of other hormone receptors (including
89

EGFR) that can facilitate PI3K signaling. This treatment also led to significant activation of TH1R,
PTGER2 (cAMP production), downregulation of AKT, CREBBP (a subunit of CREB complex),
although signaling pathways such as PI3K/AKT, ERK, and CREB signaling were not altered in
the grand view. MPP also blocked TNF expression induced by estradiol, though mechanism was
not clear.
Role of ER in Estradiol Response
In the presence of estradiol, inhibiting ER did not significant effect on OXPHOS gene expression.
PGC1 expression was not significantly altered.
PHTPP treatment significantly inhibited estradiol biosynthesis pathway, inhibited SAFB2 (a
repressor of ER), and NCOA1 (hormone-dependent coactivator for ER, GR, and RXR).
Inhibition of liganded ER also further promoted action of retinoic acid on SH-SY5Y (activation
of OTX2, BRAC1, upregulation of CRABP2, downregulation of RARA and RARB). As the cells
further transformed away from cancerous cell phenotype, they presented a transcriptome profile
with enhanced cellular homeostasis and reduced inflammation (activation of TERF2, SYVN1,
IL10RA, TLR4, upregulation of ANXA1, downregulation of TNFR2, IL13). PI3K/AKT, JUNK,
NFB signaling may be compromised as evident by their decreased expression level and predicted
inhibition of upstream regulators such as NUPR1 and FAS. PKC and MAPK expression were also
significantly downregulated, although downstream gene expression analysis could not confirm
inhibition of their activities. Unlike inhibition of liganded ER, inhibition of liganded ER did not
led to upregulation of glycolysis genes. These observations helped explain the significantly
reduced mitochondrial respiratory capacity in PHTPP pretreated cells compared to estradiol treated
90

cells. PHTPP treatment resulted in activation of EGF, TGF, and growth hormone signaling, as well
as upregulation of ERRA as a compensatory response to the loss of ER genomic action.
Discussion and Conclusion
Our previous studies in female rats demonstrated that loss of ovarian hormone due to menopause
is associated with reduced brain glucose metabolism, electron transport chain enzyme activities,
and OXPHOS subunit protein expression (Ding et al., 2013a; Ding et al., 2013b; Rettberg et al.,
2014; Yao and Brinton, 2012; Yao et al., 2010; Yao et al., 2009; Yao et al., 2011c). In the previous
chapter, we demonstrated that post-menopausal females, loss of estradiol but not progesterone is
associated with declined OXPHOS gene expression. In this study, we further confirmed that in
pre-menopausal, reproductively competent rats, loss of estradiol systematically downregulated
expression of mitochondrial encoded OXPHS genes. In younger animals, however, loss of ovarian
hormone did not significantly reduce mitochondrial gene expression, potentially due to
significantly higher level of brain estradiol compared to that in older rats. Nevertheless, restoration
of peripheral estradiol still tended to upregulate mitochondrial gene expression, suggesting a
positive role of estradiol in regulating mitochondrial bioenergetics via regulation of OXPHOS
gene expression.
However, the brain is composed of multiple cell types, each with a distinct metabolic profile, we
proposed to determine the effect of estradiol on two major brain cell types: neurons and astrocytes.
Using rat embryonic cells, we observed that neurons and astrocytes had different respiratory
phenotypes, where neurons have a large reserved mitochondrial respiratory capacity, but astrocytes
did not. While estradiol increased multiple aspects of respiratory capacity in neurons, its effect
was more limited to basal respiration in astrocytes. Given the differential distribution of two
nuclear estrogen receptors (Irwin et al., 2012; Milner et al., 2005; Simpkins et al., 2008b; Spencer-
91

Segal et al., 2012; Yang et al., 2004; Yang et al., 2009), we hypothesized that ER and ER may
have differential responsibilities in mediating estrogen response in each of the two cell types. We
observed that in neurons, estradiol mediated increase in mitochondrial respiratory capacity was
primarily mediated by ER, whereas in astrocytes, both ER and ER shared the responsibility.
This is consistent with our previous observation that in rat hippocampus, mitochondrial encoded
complex IV subunit I is selectively activated by ERβ agonist (diarylpropionitrile; DPN) but not by
ER agonist (propylpyrazoletriol; PPT) (Irwin et al., 2012). Observations from functional assays
were further confirmed by gene expression profiles of mitochondrial genes and their
transcriptional factors. From the gene expression perspective, neurons seemed more responsive to
estradiol than astrocytes. Intriguingly, we observed that both MPP and PHTPP alone significantly
increased mitochondrial gene expression in neurons. It is possible that unliganded ERs have some
basal function under normal culture condition, and inhibiting these functions led to upregulation
of gene expression as a compensatory mechanism. Another possibility is that MPP or PHTPP
bound ERs elicited some non-genomic activities to increase mitochondrial gene expression. The
best approach would be a transcriptome analysis (analysis underway).
We were also concerned that our observations might be specific for rat embryonic cells and not
translatable to human studies. However, human neurons and astrocytes are scarce sources. While
they can be induced and differentiated from induced pluripotent stem cells (iPSCs), the huge
individual variance will likely obscure the treatment effect, not to mention the potential replication
issue if using different cell lines. To overcome this problem, we used the human neuroblastoma
cell line SH-SY5Y, which is of female origin, expresses both ER and ER, and can be
differentiated towards a more mature neuronal phenotype with retinoic acid treatment (Kovalevich
and Langford, 2013a). Using a cell line instead of primary cells also made it possible to generate
92

stable ER or ER knockout cells as a genetic validation of our pharmacological approach
(experiment in progress).
In wildtype SH-SY5Y cells, extracellular flux assay revealed a similar respiratory profile to that
of astrocytes, where cells had limited reserved respiratory capacity, and inhibition of either ER
or ER can abolish the effect of estradiol on mitochondrial respiration. Further, neuronal
differentiated SH-SY5Y cells had increased reserved respiratory capacity, resembling a respiratory
phenotype of neurons. And mirroring the observations in neurons, inhibition of ER but not ER
antagonized E2 effect on maximum respiratory capacity. Transcriptome analysis confirmed
differential action of ER and ER in response to estrogen stimulation in different cell types. In
wildtype cells, ER was able to exert its action via co-transcriptional factors, non-genomic
pathways (such as PI3K, ERK signaling), and interaction with other growth factors, whereas ER
only seemed to interact with other growth factors. On the contrary, in differentiated cells, ER
only acted through non-genomic signaling pathways, while ER acted through both non-genomic
signaling pathways, interaction with various growth factors, as well as co-regulators. The increased
responsibility of ER is consistent with significant upregulation of ESR2 gene expression and 25%
down regulation of ESR1 gene expression following neuronal differentiation. RNA-Seq analysis
also confirmed ligand-independent activity of ER and ER, especially in wildtype cells. This
study further confirmed PGC1 as a primary regulator of OXPHOS gene expression, and HDAC6
and LONP1 as potential modulator of mitochondrial gene expression.
The reason that estradiol treatment in differentiated cells did not led to statistically different change
may be due to the presence of retinoic acid as the differentiation reagent. As shown here, retinoic
acid significantly upregulated mitochondrial gene expression and increased reserved respiratory
93

capacity of the cells. This may have pushed cells towards their maximal capacity and left limited
capacity for improvement. The lack of significant effect may also be due to incomplete
differentiation of SH-SY5Y cells, as confirmed by transcriptome data.
The biggest limitation to this study is the pharmacological approach we used. While both MPP
and PHTPP have been tested to prevent ER and ER from interacting with EREs and potentially
half EREs and non-consensus EREs (Compton et al., 2004; Harrington et al., 2003), these
molecules could not completely abolish all known modes of reaction of estrogen receptor (non-
genomic, interaction with growth factors, etc.). A genetic approach with ER or ER knockout
cells can address this question. Further, it has been speculated that estrogen receptors can directly
regulate mitochondrial gene expression by binding to the mitochondrial DNA. This hypothesis is
supported by the presence of putative EREs (half EREs and non-consensus EREs) in mitochondrial
genome (Demonacos et al., 1996). Further, it was demonstrated that both human recombinant ER
and ER can bind to mitochondrial putative ERE sequences, although only ER was detected to
be bound to them in the mitochondria (of MCF-7 cells) (Chen et al., 2004b). An ER binding
region within the mitochondrial control region using was also identified in breast cancer MCF-7
cells (Gertz et al., 2013). Unfortunately, this current study was not equipped to determine the
validity of this hypothesis, and future studies are warranted to further understand the involvement
of cell membrane estrogen receptor in estrogen regulation of mitochondrial bioenergetics, as well
as the mechanism of direct regulatory effect of estrogen receptor(s) on mitochondrial gene
expression through ER-mtDNA binding.
Together, in this study, we demonstrated the role of estrogen in regulating mitochondrial
bioenergetics and OXPHOS gene expression. We further demonstrated that this regulatory effect
is cell type specific, and detailed the mechanism of cell type-dependent, estrogen receptor specific
94

response to estradiol stimulation. Given the unique and primary role of ER in mediating neuronal
estrogen induced bioenergetics enhancement, future hormone replacement therapy may consider
estrogen receptor beta specific modulators to preserve neuronal bioenergetics and hence cognitive
function.
Acknowledgement
I thank “Raymond” Yuan Shang for his assistance in generating DEG files from raw RNA-Seq
FASTQ files. This study was supported by NIA R01 AG032236 to RDB; NIA P01AG026572 to
RDB, Project 1 to RDB & EC.

95

CHAPTER IV

TARGETING MITOCHONDRIAL GENETIC VARIANCES AS A PRECISION
MEDICINE OPPORTUNITY FOR ALZHEIMER’S DISEASE THERAPIES

Effect of Mitochondrial Genetic Variance and APOE Genotype on
Therapeutic Outcomes of PhytoSERM

Abstract
PhytoSERM is a selective estrogen receptor beta (ERβ) modulator comprised of three clinically
relevant phytoestrogens: genistein, daidzein, and S-equol. This formulation captures the
advantageous effect of estradiol in the brain while eliminates its harmful effect on the reproductive
system. Our recent phase Ib / IIa clinical trial demonstrated its safety and feasibility
(ClinicalTrial.gov ID: NCT01723917). While this study was not powered for efficacy analysis,
our retrospective responsive analysis demonstrated that participants on 50mg of daily PhytoSERM
had significantly reduced hot flash frequency compared their own baseline and to the placebo
group. Participants on 50mg of daily PhytoSERM also had advantageous performance on verbal
learning and cognitive flexibility. We further identified that participants of a certain mitochondrial
haplogroup and APOE genotype were particularly responsive to PhytoSERM treatment, whereas
the therapeutic benefit can be generalized to other mitochondrial haplogroups and APOE
genotypes.
Introduction
Loss of ovarian hormones during the menopausal transition is associated with vasomotor
symptoms (hot flashes and night sweats), brain glucose hypometabolism, and cognitive decline
(Erlik et al., 1982; Farrer et al., 1997; Gold et al., 2000; Greendale et al., 2011; Kim and Jung,
2015; Mosconi, 2017). This natural endocrinological transition is also considered a contributor to
96

the two-fold higher life-time risk of late onset Alzheimer’s disease (AD) (Brinton, 2008b; Erlik et
al., 1982; Paganini-Hill and Henderson, 1994; Seshadri et al., 2006), given their shared metabolic
and cognitive phenotypes (Chetelat et al., 2003; De Santi et al., 2001; Ishii et al., 1997; Mosconi
et al., 2008a; Mosconi et al., 2008b; Mosconi et al., 2009; Reiman et al., 2001; Reiman et al., 2004;
Willette et al., 2015).
A conventional solution to this is hormone replacement therapy, which was shown to relieve hot
flash symptoms, improve brain glucose metabolism and cognitive function in post-menopausal
females (Kawas et al., 1997; Maki, 2006; Maki et al., 2001; Paganini-Hill and Henderson, 1996;
Resnick et al., 1998; Santen et al., 2010; Shaywitz et al., 2003; Tang et al., 1996; Waring et al.,
1999; Zandi et al., 2002). However, traditional estrogen or combined hormone replacement has
overshadowing side effects such as elevated risks for stroke, heart attack, and breast cancer
(Chlebowski et al., 2010a; Chlebowski et al., 2010b; Chlebowski et al., 2009; LaCroix et al., 2011;
Rossouw et al., 2002; Rossouw et al., 2007), and deterred post-menopausal females from taking
advantage of the therapy. For this reason, it is of great importance to develop safe and effective
alternatives that activates estradiol action in the brain but not in the reproductive system.
As a result, naturally occurring, plant derived phytoestrogens, such as soy-derived isoflavones,
which were not associated with breast cancer (Wu et al., 1996), attracted much attention as
alternatives to estrogen replacement therapy. Earlier studies found that physiologically relevant
levels of soy isoflavones can promote neurogenesis in vitro (Perez-Martin et al., 2005; Zhao et al.,
2002), and provide benefits in memory and cognitive functions in some clinical studies (Casini et
al., 2006; Duffy et al., 2003; File et al., 2005; File et al., 2001; Kritz-Silverstein et al., 2003; Zhao
and Brinton, 2007b). Although some studies using phytoestrogens in post-menopausal women
found positive effects on hot flashes, bone mineral density, risks of cardiovascular diseases, and
97

cognitive function, results were generally mixed and inconclusive (reviewed by (Zhao and Brinton,
2007b)). The conflicted results can be explained by the complex effects of estrogen and different
compositions of phytoestrogens, which could activate both agonist and antagonistic signaling
pathways. For example, while activation of either ER or ERβ can promote neuroprotection
against various neurodegenerative insults, co-administration of ERα-selective agonist and ERβ-
selective agonists was less efficacious (Zhao and Brinton, 2007a; Zhao et al., 2004). Because ERβ
seems to play a more central role in estrogen mediated neuronal plasticity and memory function,
a selective estrogen receptor beta (ERβ) targeting formula can be a novel and plausible solution
for menopause related vasomotor symptoms and cognitive impairment (Brinton, 2009; Rissman et
al., 2002; Zhao and Brinton, 2005a).
For this reason, we developed the PhytoSERM formulation, which combines three selective
estrogen receptor beta (ERβ) modulators —genistein, daidzein, and S-equol, in equal parts (Zhao
et al., 2009). Earlier in vitro and in vivo studies demonstrated neuroprotective effect of
PhytoSERM, without side effect on the reproductive system (Zhao et al., 2002; Zhao et al., 2009;
Zhao et al., 2013; Zhao et al., 2011). Recent phase Ib / IIa clinical trial on PhytoSERM for
management of menopause associated vasomotor symptoms and cognitive decline
(NCT01723917) also validated its safety and feasibility profiles. While this study was not powered
for efficacy analysis, it was possible to conduct a retrospective responsive analysis to identify
potential responders to PhytoSERM treatment, and to determine the optimal populations to pursue
for a phase II clinical trial on the PhytoSERM formulation.
Because mechanistic studies revealed that PhytoSERM can potentiating mitochondrial function
and bioenergetics (Yao et al., 2013), and two genetic risk factors for late onset Alzheimer’s disease
– mitochondrial haplogroup (Chinnery et al., 2000; Coto et al., 2011; Edland et al., 2002; Elson et
98

al., 2006; Fachal et al., 2015; Fernandez-Moreno et al., 2011; Fesahat et al., 2007; Gomez-Duran
et al., 2010; Kenney et al., 2013; Larsen et al., 2014; Lin et al., 2012; Mancuso et al., 2007;
Maruszak et al., 2009; Maruszak et al., 2011; Ridge et al., 2012; Santoro et al., 2010; Tranah et al.,
2011; van der Walt et al., 2004; van der Walt et al., 2005; Wang and Brinton, 2016) and APOE
genotype (Carrieri et al., 2001; Corder et al., 1993; Coto et al., 2011; Edland et al., 2002; Maruszak
et al., 2011; Mosconi et al., 2008a; Mosconi et al., 2005; Mosconi et al., 2004a; Mosconi et al.,
2004b; Mosconi et al., 2004c; Poirier et al., 1993; Rebeck et al., 1993; Reiman et al., 2001; Reiman
et al., 2004, 2005; Saunders et al., 1993; Valla et al., 2010; Wolf et al., 2013) are known to be
differentially associated with mitochondrial bioenergetics and respiratory efficiency, it is of
interest to see if these two factors can modulate therapeutic effects of PhytoSERM. Herein are
outcomes of the retrospective responsive analysis, based on mitochondrial haplogroups and APOE
genotypes.
Materials and Methods
Study design
The study design and participant characteristics have been previously described in detail. In
summary, it was a double-blind, parallel-group, placebo-controlled phase Ib / IIa trial of
PhytoSERM in peri- to postmenopausal women. Eligible participants were generally healthy
women between 45 and 60 years of age, with intact uteri and ovaries, who had at least one cognitive
complaint and one vasomotor-related symptom. Study participants were randomized to receive
either one 50mg tablet of PhytoSERM (PS50, N=18), one 100mg tablet of PhytoSERM (PS100,
N=12), or matching placebo tablet (N=16) per day for 12 weeks.
99

All participants were instructed to keep daily diaries of their hot flash (frequency and severity)
throughout the 12-week trial period. Diaries were collected at each visit at IV-week intervals. For
this retrospective analysis, only participants with complete hot flash diaries were included. Those
with more than 7 consecutive missing entries and those with overall more than 25% missing entries
were excluded to ensure participants compliance and data consistency.
Neuropsychological tests were administered at baseline, weeks 4, 8, and 12. The following tests
were included in this analysis: the Verbal Fluency (FAS) test for verbal fluency (Harold Goodglass,
1972); the Rey Auditory Verbal Learning Test (RAVLT) as an assessment of multiple cognitive
parameters associated with verbal learning and memory (Rey, 1964); the Trail Making Test Parts
A and B as a measurement of vasomotor and perceptual-scanning skills and cognitive flexibility
respectively (Reitan, 1958); the Logical Memory Test I and II (immediate and delayed paragraph
recall) modified from the Wechsler Memory Scale-Revised (WMS ‐R) as a measure of the episodic
memory (Wechsler, 1987); and Mini–Mental State Examination (MMSE) score as a measurement
of global cognitive function. The institutional review board at University of Southern California
approved the study, which started in July 2012 and ended in January 2015 (ClinicalTrials.gov
Identifier NCT01723917). All participants provided written informed consent.
Mitochondrial DNA Haplotyping
Total DNA was extracted from whole blood samples using QIAGEN QIAamp DNA Mini Kit
(QIAGEN, 51304) following manufacturer’s instructions. Isolated DNA was quantified by
PicoGreen® dsDNA quantitation assay (Invitrogen, P7589). Mitochondrial DNA sequencing
was done by University of Arizona Genomics Core. Briefly, DNA samples were first enriched
for mitochondrial DNA by PCR reaction (see Table IV- for primers and locations). Amplified
segments were sequenced by dye-terminator sequencing on a 96-capillary 3730xl DNA Analyzer
100

(see Table IV-2 for primers and locations). Sequencing results were assembled and aligned to
revised Cambridge Reference Sequence (rCRS, GenBank number NC_012920) using the CLC
Main WorkBench software. Mitochondrial haplogroup for each sample was classified using
mthap (Aissani et al., 2014; Wang et al., 2016), which provided a detailed assignment for each
sample showing the SNP path and a list of alternatives.
Table IV-1. Primers and locations to amplify mitochondrial DNA before mitochondrial haplotype sequencing.

Table IV-2. Primers and locations for PCR reactions required for dye-terminator sequencing.

Primer-forward
Nucleotide
positions
Primer sequence
Primer-
reverse
Nucleotide
positions
Primer sequence
Product
size (bp)
A-forward 503-520 ATCCTACCCAGCACACAC A-reverse 2484-2463 GATTTGCCGAGTTCCTTTTACT 1982
B-forward 2364-2386 CTGACAATTAACAGCCCAATATC B-reverse 4249-4228 GAATGCTGGAGATTGTAATGGG 1886
C-forward 4155-4175 CCAACTCATACACCTCCTATG C-reverse 6220-6199 GGTAAGAGTCAGAAGCTTATGT 2066
D-forward 6113-6133 AATACCCATCATAATCGGAGG D-reverse 8017-7996 GAGTACTACTCGATTGTCAACG 1905
E-forward 7925-7944 GGCGGACTAATCTTCAACTC E-reverse 9884-9863 GTGAAATATTAGTTGGCGGATG 1960
F-forward 9767-9784 CATTTCCGACGGCATCTA F-reverse 11748-11727 GCTAGGCAGAATAGTAATGAGG 1982
G-forward 11614-11635 CATTGCATACTCTTCAATCAGC G-reverse 13638-13617 TTGACCTGTTAGGGTGAGAAGA 2025
H-forward 13539-13559 ATCATACACAAACGCCTGAGC H-reverse 15431-15409 CGTCTTTGATTGTGTAGTAAGGG 1893
I-forward 15331-15350 CCACCTCCTATTCTTGCACG I-reverse 836-815 TGCTAAAGGTTAATCACTGCTG 2075
Primer-forward
Nucleotide
positions
Primer sequence Primer-reverse
Nucleotide
positions
Primer sequence
Product
size (bp)
1-forward 516–534 CACACACACCGCTGCTAAC 1-reverse 1190–1172 GATATGAAGCACCGCCAGG 675
2-forward 1138–1156 GAACACTACGAGCCACAGC 2-reverse 1801–1782 TCATCTTTCCCTTGCGGTAC 664
3-forward 1756–1776 AATTGAAACCTGGCGCAATAG 3-reverse 2444–2426 TGAGCATGCCTGTGTTGGG 689
4-forward 2395–2415 ACCAACAAGTCATTATTACCC 4-reverse 3074–3054 TGAACTCAGATCACGTAGGAC 680
5-forward 2995–3013 GGATCAGGACATCCCGATG 5-reverse 3645–3627 AACGGCTAGGCTAGAGGTG 651
6-forward 3536–3553 TAGCTCTCACCATCGCTC 6-reverse 4239–4219 GATTGTAATGGGTATGGAGAC 704
7-forward 4184–4202 TCCTACCACTCACCCTAGC 7-reverse 4869–4852 GTCATGTGAGAAGAAGCA 686
8-forward 4832–4849 CACCCCTCTGACATCCGG 8-reverse 5570–5551 AGTATTGCAACTTACTGAGG 739
9-forward 5526–5545 AATACAGACCAAGAGCCTTC 9-reverse 6188–6171 GGGAAACGCCATATCGGG 663
10-forward 6115–6134 TACCCATCATAATCGGAGGC 10-reverse 6781–6761 AATATATGGTGTGCTCACACG 667
11-forward 6730–6750 CTATGATATCAATTGGCTTCC 11-reverse 7398–7379 GGCATCCATATAGTCACTCC 669
12-forward 7349–7369 CCTAATAGTAGAAGAACCCTC 12-reverse 8009–7990 CTCGATTGTCAACGTCAAGG 661
13-forward 7960–7979 ATTATTCCTAGAACCAGGCG 13-reverse 8641–8621 TGATGAGATATTTGGAGGTGG 682
14-forward 8563–8581 ACAATCCTAGGCCTACCCG 14-reverse 9231–9212 GATAGGCATGTGATTGGTGG 669
15-forward 9181–9198 AGCCTCTACCTGCACGAC 15-reverse 9867–9848 GGATGAAGCAGATAGTGAGG 687
16-forward 9821–9841 ACTTCACGTCATTATTGGCTC 16-reverse 10516–10497 AGTGAGATGGTAAATGCTAG 696
17-forward 10394–10414 CTGAACCGAATTGGTATATAG 17-reverse 11032–11013 TCGTGATAGTGGTTCACTGG 639
18-forward 10985–11004 ACAATCATGGCAAGCCAACG 18-reverse 11708–11689 TTATGAGAATGACTGCGCCG 724
19-forward 11633–11651 AGCCACATAGCCCTCGTAG 19-reverse 12361–12341 TGGTTATAGTAGTGTGCATGG 729
20-forward 12284–12302 CTATCCATTGGTCTTAGGC 20-reverse 13005–12987 TTTGCCTGCTGCTGCTAGG 722
21-forward 12951–12969 CGCTAATCCAAGCCTCACC 21-reverse 13614–13595 TATTCGAGTGCTATAGGCGC 664
22-forward 13568–13587 TTACTCTCATCGCTACCTCC 22-reverse 14276–14258 GGTTGATTCGGGAGGATCC 709
23-forward 14227–14246 CCCATAATCATACAAAGCCC 23-reverse 14928–14911 GTTGAGGCGTCTGGTGAG 702
24-forward 14732–14752 ACTACAAGAACACCAATGACC 24-reverse 15419–15400 TGTAGTAAGGGTGGAAGGTG 688
25-forward 15372–15391 TAGGAATCACCTCCCATTCC 25-reverse 16067–16048 GTCAATACTTGGGTGGTACC 696
d1-forward 15879–15897 AATGGGCCTGTCCTTGTAG d1-reverse 16545–16526 AACGTGTGGGCTATTTAGGC 667
d2-forward 16495–16514 CGACATCTGGTTCCTACTTC d2-reverse 389–370 CTGGTTAGGCTGGTGTTAGG 446
d3-forward 315–332 CGCTTCTGGCCACAGCAC d3-reverse 803–786 GGTGTGGCTAGGCTAAGC 489
101

APOE genotyping
Total DNA was extracted from whole blood samples using QIAGEN QIAamp DNA Mini Kit
following manufacturer’s instructions. APOE genotyping was done as previously described with
some modifications (William Rebeck et al., 1993). Briefly, the following primers sequences were
used to amplify the DNA: FWD_TAAGCTTGGCACGGCTGTCCAAGGA and
REV_ACAGAATTCGCCCCGGCCTGGRACACTGCC. Amplification was performed in a final
volume of 25L containing 25ng/L of DNA solution, 400nM of each primer, and 1x RT2
SYBR® Green qPCR Mastermix (QIAGEN, 330500). Reactions were done using Bio-Rad
MyCycler Thermal cycler using the following conditions: initial denaturation at 95°C for 10
minutes, followed by 40 cycles of amplification (94°C for 30 seconds, 58°C for 30 seconds, and
72°C for 1 minute), and a final extension step at 72°C for 7 minutes. Amplification products were
digested with HhaI restriction endonuclease. APOE genotype for each sample was identified based
on agarose gel electrophoresis results.
Statistical Analysis
Post-hoc analysis of changes in hot flash frequency and cognitive function within each treatment
group were analyzed using non-parametric paired t-test between week 1 and week 12, and changes
among treatment groups were analyzed using non-parametric ANOVA followed by unpaired t-test.
Results were then stratified based on the APOE genotype and mitochondrial haplogroup of the
participants to identify responder groups.

102

Results
Participants
A total of 46 participants with complete hot flash diaries were included in the response analysis.
Participants’ ages ranged from 47 to 60 years old, with an average of 54.2 +/- 3.3 SD years old.
Participants had on average 17.3 +/- 3.2 SD years of education. Of all participants, 4 were Asians,
2 were black or African Americans, 35 were Caucasians, and 5 were unknown.
Mitochondrial Haplotyping
Of the 40 participants for which mitochondrial haplotyping was possible, Haplogroup H had the
greatest representation in this cohort (see treatment by mitochondrial haplogroup in
Table IV-3). Because haplogroup H was the most common mitochondrial variant among European
descendants and the most represented in the study, the limited number of participants from other
haplogroups, data by haplogroup H (N=11) and non-H (N=29).
Table IV-3. Participants by treatment groups and mitochondrial haplogroups.
Placebo PS50 PS100 Total
A 1 3 0 4
B 0 1 1 2
C 0 0 1 1
D 0 1 0 1
H 5 5 1 11
K 2 1 1 4
L 0 1 0 1
M 0 1 1 2
T 3 0 3 6
U 1 2 2 5
V 1 1 0 2
W 1 0 0 1
Total 14 16 10 40

103

APOE Genotype
Thirty-two participants were APOE 3/3 carriers (67%) and 14 were APOE 3/4 carriers (33%),
which is consistent with prevalence in the general population. There were no APOE 4/4 carriers
in this analysis. See Table IV-4 for treatment by APOE genotype.
Table IV-4. Participants by treatment and APOE genotype.
Placebo PS50 PS100 Total
APOE 3/3 12 14 6 32
APOE 3/4 4 4 6 14

Effect of PhytoSERM on hot flash frequency
Daily average hot flash frequency at week 1 was used as baseline for each participant. Change in
hot flash frequency was calculated as the difference between week 12 and week 1 hot flash
frequency. No difference in baseline hot flash frequency was observed among the three treatment
groups, or between different mitochondrial haplogroups or APOE genotypes (Figure Figure IV-
A). Nor was there significant difference among different age groups (Figure IV-B).
In itia l d a ily h o t fla s h fre q u e n c y
H
n o n -H
A P O E 3 /3
A P O E 3 /4
0
5
1 0
1 5
4 6 -5 0
5 1 -5 5
5 6 -6 0
0
5
1 0
1 5
A g e
In itia l d a ily h o t fla s h fre q u e n c y
IV -1 A IV -1 B

Figure IV-1. Baseline hot flash frequency by demographic data. 1A, no difference in baseline hot flash
frequency among different mitochondrial haplogroups or APOE genotypes. 1B, age does not affect baseline
hot flash frequency.

104

Compared to placebo, 50mg of daily PhytoSERM significantly reduced hot flash frequency. While
100mg of daily PhytoSERM also had a trend to reduce hot flash frequency, the difference was not
significant compared to the placebo (p=0.0389) (Figure IV-).
C h a n g e in h o t fla s h fre q u e n c y
P la c e b o P S 5 0 P S 1 0 0
-6
-4
-2
0
2
4
*

Figure IV-2. Impact of different doses of PhytoSERM on hot flash frequencies. Participants in the PS50
group experienced significant reduction in hot flash frequency in comparison to those in the placebo group, *
p<0.05. Those in the PS100 group also tend to have less hot flash, but reduction was not significant compared
to the placebo group.

We further observed that the baseline initial frequency of hot flash was significantly positively
correlated with the effect of PhytoSERM on hot flash frequency in the PS50 group (r=-0.6740,
p=0.0038) (Figure IV-3A). No such correlation was observed in the placebo group (r=-0.2309,
p=0.4069) (Figure IV-3B).

105

P S 5 0
In itia l h o t fla s h fre q u e n c y
C h a n g e in h o t fla s h fre q u e n c y
2 4 6 8 1 0
-6
-4
-2
0
2
4
r= -0 .6 7 4 0
P la c e b o
In itia l h o t fla s h fre q u e n c y
C h a n g e in h o t fla s h fre q u e n c y
2 4 6 8
-4
-2
0
2
4 r= 0 .2 3 0 9
IV -3 A IV -3 B

Figure IV-3. Initial hot flash frequency predicts therapeutic outcomes of PhytoSERM on hot flash frequency
reduction in the PS50 group only.3A, participants with higher initial hot flash frequency showed more
reduction in hot flash frequency; 3B, no such correlation observed in the placebo group.

When stratified by mitochondrial haplogroup, those belonging to mitochondrial haplogroup H had
significantly decreased hot flash frequency when treated with 50mg of PhytoSERM per day
compared to the placebo group (p=0.0397) (Figure IV-4A). Only one haplogroup H participant
was assigned to the PS100 group and no statistical analysis was conducted. Non-H participants on
PS50 demonstrated comparable magnitude of reduction in hot flash frequency, but failed to reach
statistical significance, due to huge variation within the group (Figure IV-4B).
H a p lo g ro u p H
C h a n g e in h o t fla s h fre q u e n c y
P la c e b o P S 5 0
-6
-4
-2
0
2
4
*
H a p lo g ro u p s n o n -H
C h a n g e in h o t fla s h fre q u e n c y
P la c e b o P S 5 0 P S 1 0 0
-6
-4
-2
0
2
4
IV -4 A IV -4 B

Figure IV-4. Change in hot flash frequency from week 1 to week 12 in participants when stratified by
mitochondrial haplogroup.4A, haplogroup H in the PS50 group had significantly greater reduction in hot
flash frequency compared to those on placebo. 4B, no such difference was observed in non-H participants. *
p<0.05.
106

When stratified by APOE genotype, APOE4 non-carriers in the PS50 group had significantly
greater reduction in hot flash frequency compared to those in the placebo group (p=0.0404) (Figure
IV-5A). APOE4 carriers showed a trend of improvement but did not reach statistical significance
in comparison to the placebo group, likely due to limited number of participants (Figure IV-5B).
regardless of APOE genotype, participants in PS100 did not experience significant improvement.
A P O E 3 /3
C h a n g e in h o t fla s h fre q u e n c y
P la c e b o P S 5 0 P S 1 0 0
-6
-4
-2
0
2
4
*
A P O E 3 /4
C h a n g e in h o t fla s h fre q u e n c y
P la c e b o P S 5 0 P S 1 0 0
-6
-4
-2
0
2
4
IV -5 A IV -5 B

Figure IV-5. Change in hot flash frequency stratified by APOE genotype. 5A, APOE3/3 participants on PS50
showed non-significant reduction in hot flash frequency * p<0.05. 5B, APOE3/4 participants on PS50 showed
non-significant decline in hot flash frequency.

Effect of PhytoSERM on cognitive function
The following aspects of the audio verbal learning performance were examined in this analysis:
immediate recall, delayed recall, recognition, and learning over trials (LOT), as previously
described (Teruya et al., 2009). We observed that treatment with neither 50mg nor 100mg of
PhytoSERM per day improved immediate recall, delayed recall, or recognition. Nor were these
results affected by mitochondrial haplogroup or APOE genotype status. Intriguingly, however,
haplogroup H on placebo showed declined ability to learn the trial throughout the clinical study,
whereas treatment of 50mg of PhytoSERM per day successfully prevented such decline
(p=0.0476) (Figure IV-6A). No such preventative effect was observed in non-H haplogroups
107

(Figure IV-6B). This can be explained by the significant difference between the two placebo
groups, where haplogroup H displayed significantly worse decline (p=0.007) (Figure IV-6C).
H a p lo g ro u p H
C h a n g e in R A V L T L O T
P la c e b o P S 5 0
-2 0
-1 0
0
1 0
2 0
*
C h a n g e in R A V L T L O T
P la c e b o P S 5 0 P S 1 0 0
-2 0
-1 0
0
1 0
2 0
H a p lo g ro u p s n o n -H
IV -6 A IV -6 B
IV -6 C
C h a n g e s in R A V L T L O T
P la c e b o - H P la c e b o - n o n -H
-2 0
-1 0
0
1 0
2 0
*

Figure IV-6. Change in participants’ ability to learn the RAVLT trial stratified by mitochondrial haplogroup.
6A, in haplogroup H participants, treatment with 50mg of daily PhytoSERM preserved verbal learning
ability in comparison to those on placebo. 6B, no difference was observed among three treatment groups in
non-haplogroup H participants. 6C, haplogroup H participants on placebo displayed significantly worse
decline in verbal learning ability compared to their non-haplogroup H counterparts. * P<0.05.

In addition, while PhytoSERM treatment did not result in significant improvement in cognitive
flexibility as measured by trails making part A and B (Figure IV-7A), within the PS50 groups,
participants showed significantly enhanced cognitive flexibility compared to their own baseline
(p=0.0058) (Figure IV-7B).

108

C h a n g e in tra il B tim e (s )
P la c e b o P S 5 0 P S 1 0 0
-4 0
-2 0
0
2 0
4 0
W e e k
T ra il B tim e (s )
W e e k 1 W e e k 1 2
0
5 0
1 0 0
1 5 0
*
IV -7 A IV -7 B

Figure IV-7. Effect of PhytoSERM on cognitive flexibility. 7A, PhytoSERM treatment resulted in a trend of
improvement on trail B making test. 7B, 50mg of PhytoSERM treatment for 12 weeks significantly improved
cognitive flexibility. P<0.05.

No effect of PhytoSERM was observed on verbal fluency, episodic memories and global cognition
throughout the trial.
Discussion and Conclusion
This is a retrospective analysis based of a clinical trial for the safety and efficacy of PhytoSERM
for management of menopause-associated vasomotor symptoms and cognitive decline. The parent
study did not identify significant benefit of PhytoSERM on vasomotor symptoms using a
vasomotor composite score, or cognitive function using neuropsychological composite score
among treatment groups. The goal of this current study was to determine if PhytoSERM can
specifically reduce menopause-associated hot flash frequency, and to identify potential responders
to PhytoSERM treatment based on two genetic factors: mitochondrial haplogroup and APOE
genotype. These two factors were selected because both were demonstrated risk factors for late
onset AD, with known effect on brain glucose metabolism and mitochondrial bioenergetics.
Our analysis demonstrated that PhytoSERM reduced hot flash frequency in post-menopausal
females, which was in agreement with our preclinical study using the rat surgical menopausal
model (Zhao et al., 2011), and in accordance with literature showing benefit of estrogen or
109

phytoestrogens in reducing menopause associated vasomotor symptoms (Burke et al., 2003;
Clarkson et al., 2011; Jacobs et al., 2009; Teekachunhatean et al., 2015). Furthermore, the observed
protective effect of PhytoSERM on cognitive function was also in line with previous studies
showing that estrogen replacement treatment and/or soy isoflavones extracts significantly
enhanced verbal memory and cognitive function in post-menopausal females (Casini et al., 2006;
Duffy et al., 2003; File et al., 2001; Kritz-Silverstein et al., 2003; Maki, 2006; Maki et al., 2001;
Phillips and Sherwin, 1992; Shaywitz et al., 2003; Sherwin, 1988).
Specifically, after 12 weeks of treatment, participants on 50mg of daily PhytoSERM had reduced
hot flash frequency compared to their own baseline. Participants on 100mg of daily PhytoSERM
also had a trend of reduced hot flash frequency, though the change was not significant compared
to their own baseline. Furthermore, compared to the placebo group, the PS50 group but not the
PS100 had significantly greater reduction in hot flash frequency, suggesting that like general
estrogen, PhytoSERM has a reversed V shape dose response curve, and that 50mg per day is the
superior dosage for relieving menopause associated hot flashes.
When stratified by mitochondrial haplogroups, the therapeutic effect of 50mg of PhytoSERM was
preserved in haplogroup H, the most common haplogroup for European descendants. While no
significant therapeutic effect on hot flash was observed in non-H participants, given the big
variance we observed in non-H participants on PS50, we cannot eliminate the possibility that there
are other responding haplogroups, and that the effect of PhytoSERM may be generalized into other
haplogroups that were under-represented in this study.
When stratified by APOE genotype, APOE 3/3 participants on 50mg of daily PhytoSERM had
significantly more reduced hot flash frequency in comparison to those on placebo, and APOE 3/4
110

participants on 50mg of daily PhytoSERM displayed a trend of decline in hot flash frequency. This
was likely due to big individual differences and small number of APOE 3/4 participants.
In terms of the therapeutic effect of PhytoSERM on cognitive function, we observed no protective
effect on verbal fluency, episodic memories, and overall cognitive function for both PS50 and
PS100 group, in accordance with our primary analysis. This was possibly due to the short duration
of the clinical trial, and a lack of specificity or sensitivity of such tests in non-clinically demented
participants. Some of the tests were also vulnerable to learning effect, for example, we observed
that participants “learnt” the episodic memory test at the screening visit, and all had much better
performance at their first official trial visit (data not shown).
However, we did observe a beneficial effect of PhytoSERM on two cognitive tasks. We observed
that participants on the PS50 group had significantly improved cognitive flexibility compared to
their own baseline, although the improvement was not significant compared to the placebo group.
This effect seemed to be independent of mitochondrial haplogroups and APOE genotype. More
intriguing is the effect of PhytoSERM on verbal learning ability, as measured by the “learn of trial”
parameter of the RAVLT, where haplogroup H participants from the placebo group showed a trend
of declined compared to their own baseline, whereas treatment with 50mg of PhytoSERM for 12
weeks prevented such decline (Figure IV-6). This observation is consistent with population studies
showing haplogroup H had higher risk for late onset AD (Chinnery et al., 2000; Coto et al., 2011;
Edland et al., 2002; Elson et al., 2006; Fachal et al., 2015; Fesahat et al., 2007; Mancuso et al.,
2007; Maruszak et al., 2009; Maruszak et al., 2011; Ridge et al., 2012; Santoro et al., 2010; van
der Walt et al., 2004; van der Walt et al., 2005; Wang and Brinton, 2016). The fact that
PhytoSERM exerted protective effect only on selected aspects of cognitive function also supports
111

the notion that estrogen preferentially affect cognitive tasks of greater complexity, temporal
demand and associative challenge (Brinton, 2009).
We are aware that the statistical power of this analysis was limited by the small sample size, given
the nature of this clinical study as a phase I study. Nevertheless, results from this analysis
demonstrated beneficial effect of PhytoSERM at a daily dosage of 50mg on both hot flash
frequency and cognitive function. While the observations made in this retrospective analysis
should be confirmed by a prospective, larger scaled clinical study, the data support further
development of PhytoSERM as a therapy for ameliorating menopause-associated vasomotor
symptoms and cognitive decline. We also propose to employ a precision medicine approach to
identify and target responders to PhytoSERM for its future development.
Acknowledgement
This study was supported by NIA grants UF1-AG046148, U01-AG031115, U01-AG047222 and
P01-AG026572 to RDB.

112

Effect of Mitochondrial Genetic Variance, APOE Genotype, and Sex on
Therapeutic Outcomes of Allopregnanolone

Abstract
Late onset Alzheimer’s disease (LOAD) is a systemic disease with multiple etiologies, and is
associated with compromised brain metabolism and regenerative capacity. Allopregnanolone has
been shown to promote brain mitochondrial function, neurogenesis, and memory in mouse models,
and is currently being investigated as a regenerative therapeutic for AD (NCT02221622). While
genetic markers such as APOE genotype may predict risk of AD, there is currently no genetic
markers to predict therapeutic outcomes for AD. Because mitochondrial genetic variances and
APOE genotype are known to be differentially associated with respiratory capacity and cell
proliferation, in this study, we evaluate whether they can be used as potential genetic markers to
predict responders for Alzheimer’s disease therapeutics. Using neural stem cells (NSCs)
differentiated from patient-derived induced pluripotent stem cells (iPSCs), we observed that
allopregnanolone treatment preferentially increased mitochondrial respiratory capacity in NSCs
derived from participants of certain mitochondrial haplogroups. We also observed that APOE4
carrier NSCs and non-carrier NSCs had significantly different proliferation patterns. Further,
transcriptome analysis suggested that NSCs of female, APOE4 genotype, and certain
mitochondrial haplogroups are more responsive to allopregnanolone than their counterparts. We
concluded that sex, APOE genotype, and mitochondrial genetic variance make promising
predictive biomarkers to identify potential allopregnanolone responders. Predictive biomarkers
like these will significantly contribute to a precision medicine strategy to identify responders to
therapeutic agents for Alzheimer’s disease.
113

Introduction
Late onset Alzheimer’s disease (LOAD) is a devastating disease with multiple etiologies, yet there
is no disease progression modifying therapy. Current development of LOAD therapy focuses on
two disease hallmarks of AD: amyloid beta plaque and phosphorylated tau tangle, but all efforts
were in vain (Cummings et al., 2014; Doody et al., 2013a; The Lancet, 2017; Vandenberghe et al.,
2016). It is likely that targeting pathological symptoms is not enough to stall or potentially reverse
disease progression. Considering that AD is a typical neurodegenerative disease, where loss of
endogenous regenerative potential is exacerbated, one plausible therapy is to repair and restore the
regenerative capacity of the brain by targeting its own neurogenesis machinery.
Many endogenous neurosteroids are known to promote neural stem cell regeneration and neuronal
differentiation, including allopregnanolone (3α-hydroxy-5α-pregnan-20-one). Allopregnanolone
level is decreased in AD brain (Naylor et al., 2010). Previous study demonstrated that
allopregnanolone promotes neural stem cell proliferation and neurogenesis both in vitro and in
vivo (Wang et al., 2005a; Wang et al., 2010a). In our preclinical study, allopregnanolone was
shown to reduce amyloid beta load and restore cognitive function in triple transgenic mouse model
of AD (Chen et al., 2011b; Singh et al., 2012; Wang et al., 2010a). Chronological administration
of allopregnanolone also promoted neurogenesis, increased white matter generation, while reduced
AD pathology (Brinton, 2013; Chen et al., 2011b; Irwin and Brinton, 2014). Recently completed
phase Ib / IIa clinical trial further established the safety and feasibility profiles of allopregnanolone
in a cyclodextrin-based formulation (ClinicalTrials.gov Identifier NCT02221622) (Irwin et al.,
2015).
Yet one question remains to be addressed – individual differences. As with any therapeutics, not
all patients will respond to the treatment equally. Different individuals differ not only in
114

pharmacokinetics and pharmacodynamics, but also in their response to allopregnanolone
therapeutic effect. Thus, it is important to identify and target therapeutic responders in order to
optimize clinical trial design, maximize therapeutic effect, and minimize undesired adverse effect.
Population studies identified three genetic risk factors for late onset AD: chromosomal sex, APOE
genotype, and mitochondrial genetic variances (reviewed in (Wang and Brinton, 2016)). These
factors are also differentially associated with mitochondrial respiratory capacity (Gomez-Duran et
al., 2010; Kenney et al., 2013; Larsen et al., 2014; Lin et al., 2012; Mandal et al., 2012; Mosconi
et al., 2008a; Mosconi et al., 2005; Mosconi et al., 2004a; Mosconi et al., 2004b; Mosconi et al.,
2004c; Reiman et al., 2001; Reiman et al., 2004, 2005; Schuessel et al., 2004; Valla et al., 2010;
Wolf et al., 2013). Because allopregnanolone was also shown promote mitochondrial bioenergetics
(Grimm et al., 2014), we hypothesize that these three factors, either alone or in combination, can
affect therapeutic outcome of allopregnanolone, and be used to predict responders to
allopregnanolone treatment (Figure IV-8).

115

Figure IV-8. Identification of potential Alzheimer’s disease treatment responders based on three genetic risk
factors: mitochondrial genetic variances, APOE genotype, and chromosomal sex.
Here we present our analysis on the modulating effects of sex difference, APOE genotype, and
mitochondrial genetic variance on treatment outcomes of allopregnanolone in vitro, as well as the
validity of such genetic markers in predicting clinical outcomes. We further support our
observations from patient-derived neural stem cells-based functional assays with transcriptomic
analysis.

Materials and Methods
Study design
This study utilized clinical data and neural stem cells derived from participants from a phase Ib /
IIa clinical trial, where allopregnanolone was investigated as a regenerative therapeutic for
116

Alzheimer’s disease (NCT02221622). All participants provided written informed consent, and the
study was approved by the institutional review board at University of Southern California.
Detailed clinical study design and methods for generation of neural stem cells were described in
our previous publication (Solinsky, 2017). Briefly, Whole blood was collected at the baseline, so
that all cells used for downstream studies were naive to allopregnanolone. PBMCs were isolated
and purified by density gradient centrifugation, then reprogrammed to induced pluripotent cells
(iPSCs) via a non-integrating, non-viral method (Okita et al., 2013; Solinsky, 2017). Isolated iPSC
colonies passed strict immunocytochemistry and karyotype characterization, before being
differentiated into neural stem cells (NSCs) using dual inhibition of SMAD signaling (Chambers
et al., 2009; Solinsky, 2017; Tomishima, 2008). Successfully differentiated NSCs were treated
with either vehicle (equally diluted EtOH) or 100nM allopregnanolone for 24 hours before
downstream analysis. See Figure IV-9 for a summary of the experiment design. Established NSC
lines were cultured and maintained as described previously (Solinsky, 2017).
117

Figure IV-9. Summary of experiment scheme.

Extracellular metabolism flux assay
Cellular respiratory capacity will be determined by Seahorse XF96 metabolic flux analyzer
(Agilent Technology) as previously described (Irwin et al., 2011b; Irwin et al., 2012; Yao et al.,
2013). Briefly, neural stem cells were seeded into Matrigel (Corning, 354277) coated cell plates
at 15,000 cells per well, cultured and treated as described above. On the day of assay, culture
medium was changed to unbuffered DMEM (Sigma-Aldrich, D5030) medium supplemented with
25mM glucose, and 2mM GlutaMAX for primary cells (Gibco, 35050061). Medium pH was
adjusted to 7.4. cells were incubated at 37°C in a non-CO2 incubator for 1 hour. Oxygen
consumption rate (OCR) was used as an indicator of mitochondrial oxidative phosphorylation, and
118

mitochondrial basal respiration, ATP production, maximal respiratory capacity, and proton leak
were determined by sequential acute injection of mitochondrial electron transport chain inhibitors
and un-couplers: oligomycin (MP Biomedicals, 02151786), FCCP (carbonyl
cyanide4(trifluoromethoxy)- phenylhydrazone) (TOCRIS Bioscience, 0453), rotenone (MP
Biomedicals, 02150154), and antimycin (Sigma-Aldrich, A-8674). To test for effect of
allopregnanolone on glycolysis, 1mM sodium pyruvate was added injected into each well at the
beginning of the assay. Electron transport chain inhibitors were used at the following
concentration: 4M of oligomycin, 1M of FCCP, and 1M of rotenone / 5M of antimycin.
Results were normalized to protein reading of each plate to eliminate unevenness due to cell
seeding and loss during assay procedures.
Neural stem cell regeneration assay
Neural stem cell proliferation was assessed by flow cytometry quantification of EdU-positive cells.
Cells were labeled using Click-iT Plus EdU Alexa Fluor Flow Cytometry Assay Kit (Invitrogen,
C10632) following manufacturer’s instructions, and the procedure has been described in detail
previously (Solinsky, 2017). Six replicates of each NSC cell line were processed on a LSR II flow
cytometer (BD Biosciences). NSCs were gated by forward and side scatter, and the percentaion of
EdU-positive cells quantified. Data were analyzed using FlowJo V X 10.0.7r2.
Mitochondrial DNA Haplotyping
Total DNA was extracted from all 24 patients’ whole blood samples using QIAGEN QIAamp
DNA Mini Kit (QIAGEN, 51304) following manufacturer’s instructions. Isolated DNA was
quantified by PicoGreen
®
dsDNA quantitation assay (Invitrogen, P7589).
119

Mitochondrial DNA haplotyping was based on sequencing of hypervariable region 1 and 2 (HVR1
and HVR2) of mitochondrial DNA. HVR1 and HVR2 regions were amplified and sequenced by
University of Arizona Genomics Core using primers listed in Table IV-5. Sequencing reads were
aligned to Revised Cambridge Reference Sequence (rCRS, NC_012920.1), and SNPs and
mutations were manually identified. Mitochondrial haplogroups were then assigned using
HaploGrep2 based on identified variants (Kloss-Brandstatter et al., 2011; Weissensteiner et al.,
2016).
Table IV-5. Amplification and sequencing primers for mitochondrial HVR1 and HVR2.
Amplification
Primers:

Sequence Length
Fwd. 5' to 3' tca aag ctt aca cca gtc ttg taa acc 27
Rev. 5' to 3' ggg tga tgt gag ccc gtc ta 20
Sequencing
Primers:

HVR1
Fwd. 5' to 3' caa gga caa atc aga gaa aa 20
Rev. 5' to 3' gtg gtt aat agg gtg ata g 19
HVR2
Fwd. 5' to 3' cac agg tct atc acc cta 18
Rev. 5' to 3' gtg atg tga gcc cgt cta 18

APOE genotyping
Total DNA was extracted from whole blood samples using QIAGEN QIAamp DNA Mini Kit
following manufacturer’s instructions. APOE genotyping was done as previously described with
some modifications (William Rebeck et al., 1993). Briefly, the following primers sequences were
used to amplify the DNA: FWD_TAAGCTTGGCACGGCTGTCCAAGGA and
REV_ACAGAATTCGCCCCGGCCTGGRACACTGCC. Amplification was performed in a final
volume of 25L containing 25ngL of DNA solution, 400nM of each primer, and 1x RT
2
SYBR®
120

Green qPCR Mastermix (QIAGEN, 330500). Reactions were done using Bio-Rad MyCycler
Thermal cycler using the following conditions: initial denaturation at 95°C for 10 minutes,
followed by 40 cycles of amplification (94°C for 30 seconds, 58°C for 30 seconds, and 72°C for
1 minute), and a final extension step at 72°C for 7 minutes. Amplification products were digested
with HhaI restriction endonuclease. APOE genotype for each sample was identified based on
agarose gel electrophoresis results.
RNA-isolation
Neural stem cells were first lysed in TRIzol® Reagent. Chloroform was used to extract RNA
from the homogenate at a volume ratio of 1:5 to that of the TRIzol® Reagent. Ethanol was then
used to precipitate nucleic acids from the aqueous phase. RNA was further purified using
PureLink™ RNA Mini Kit (Invitrogen™, 12183018A) following manufacturer’s instructions.
Purelink™ DNase (Invitrogen™, 12185010) was used to eliminate DNA contamination. Purified
RNA was eluded in RNase-free, diH2O. RNA concentration and quality were checked by
NanoDrop™ One.
RNA sequencing
RNA-Seq was conducted at Vanderbilt Technologies for Advanced Genomics (VANTAGE).
Only RNA samples with an acceptable RNA quality indicator score (RQI >7) was used for
sequencing. mRNA enrichment and cDNA library preparation were done using a stranded
mRNA (poly(A) - selected) sample preparation kit. Sequencing was performed at 150bp paired-
end on NovaSeq600, targeting 30 million reads per sample. Transcripts were mapped to human
genome using (ensemble release 92) using salmon 0.9.1 for human (Patro et al., 2017). Tximport
V1.6.0 (Soneson et al., 2015) was used to generate a counts table from salmon output, and
121

DESeq2 V1.18.1 (Love et al., 2014) was used to calculate normalized read counts for each gene
and/or transcript and to perform expression analysis.
Ingenuity Pathway Analysis (IPA) of RNA-Seq
Differentially expressed gene lists were processed using the core analysis function of IPA. Only
genes with p value smaller than 0.05 was considered (or the top 8000 most significant genes, as
per limit of IPA analysis). The outputs are lists of significantly altered canonical pathways and
upstream regulators. The canonical pathways are identified based on enrichment of qualified
genes. The upstream regulator analysis predicted activation or inhibition of regulatory molecules
based on expression of respective downstream genes and networks compiled from literature and
IPA’s Ingenuity knowledge base.
Comparison analysis was used to determine similarities and differences among multiple age and
endocrine groups. Top canonical pathways and upstream regulators were also identified.
Statistical analysis
The effects of allopregnanolone on NSC mitochondrial respiratory capacity and proliferation
were measured as percentage difference between allopregnanolone treated and vehicle treated
cells of each cell line. Due to the non-parametric nature of such data, statistical significance was
calculated by non-parametric t-test, where a p value smaller 0.05 indicated statistical
significance.

122

Results
Sample Demographic data
Of all the 21 neural stem cell lines used in this study (from 21 clinical trial participants
respectively), 11 are from females and 10 are from males. Seven out of 21 cell lines come from
non-APOE4 carriers. There was only one APOE 4/4 carrier (male) and was grouped together with
APOE 3/4 as APOE4 carriers for analysis in this study. See Table IV-6 for participants by sex and
APOE genotype.
Table IV-6. Participants by sex and APOE genotype.
Female Male Total
APOE 3/3 2 5 7
APOE 3/4 and 4/4 9 5 14
Total 11 10 21

Based on mitochondrial haplotyping, 13 out of 21 participants belong to haplogroups or super
haplogroups of European origin (H, HV, J, K, S, U, and X), and 8 belong to African or the Asian
origin (A, L, M, and N). See Table IV-7 for participants by mitochondrial haplogroups.
Table IV-7. Participants by mitochondrial haplogroups.
Mitochondrial
Haplogroup
Number of
Participants
A 1
L 4
M 2
N 1
H 3
HV 3
J 2
K 1
S 1
U 2
X 1
Total 21

123

Mitochondrial Respiratory Capacity
Following 24 hours of allopregnanolone treatment, maximum mitochondrial respiratory capacity
was increased by 9.2% on average in NSCs compared to vehicle treated controls (Solinsky, 2017).
We did not observe any statistically significant difference due to sex or APOE genotype alone,
although females and APOE3/4 carriers appeared to respond better to allopregnanolone treatment
(Figure IV-10A and Figure IV-10B). We further stratified the data by sex and APOE genotype
together, and observed that it was only in APOE3/3 carriers that females responded better than
males, whereas in APOE 3/4 carriers, both females and males had equally well response to
allopregnanolone (Figure IV-10C). The only APOE4 homozygote did not respond well to
allopregnanolone treatment (Figure IV-10C).
F e m a le
M a le
-5 0
0
5 0
1 0 0
% c h a n g e to v e h ic le
A P O E 3 /3
A P O E 3 /4
A P O E 4 /4
-5 0
0
5 0
1 0 0
% c h a n g e to v e h ic le
% c h a n g e to v e h ic le
A P O E 3 /3 F
A P O E 3 /3 M
A P O E 3 /4 F
A P O E 3 /4 M
A P O E 4 /4 M
-5 0
0
5 0
1 0 0
IV -1 0 A IV -1 0 B
IV -1 0 C

Figure IV-10. Change in maximal mitochondrial respiratory capacity following allopregnanolone treatment
stratified by sex and APOE genotype.3A, females tend to have better response to allopregnanolone. 3B, APOE
3/4 carriers tend to have better response to allopregnanolone treatment than non APOE4 carrier or APOE4
homozygote.
124

When we stratified maximum mitochondrial respiration by mitochondrial haplogroups, we
observed that NSCs of mitochondrial haplogroups representing African and Asian origins (A, L,
M, and N) were enriched in the upper 50% percentile, whereas NSCs of mitochondrial haplogroups
representing major European origins (Others) were enriched in the lower 50% percentile (Figure
IV-11A). While there was no statistical significant difference between these two groups, we did
observe an interplay between mitochondrial haplogroups and APOE genotype. Specifically, in
non-APOE4 carriers, haplogroups A, L, M, N had a trend of lower mitochondrial respiration
(Figure IV-11B). In APOE4 positive NSCs, however, haplogroups A, L, M, N had significantly
higher maximal mitochondrial respiratory capacity compared to the other haplogroups. This
statistical significance was observed with or without the APOE4 homozygote (p=0.0082 and
p=0.0047 respectively) (Figure IV-11B).

A ll N S C s
% c h a n g e to v e h ic le
-6 0
-4 0
-2 0
0
2 0
4 0
6 0
O th e rs
A , L , M , N
% c h a n g e to v e h ic le
A P O E 3 A , L , M , N
A P O E 3 O th e rs
A P O E 4 A , L , M , N
A P O E 4 O th e rs
-6 0
-4 0
-2 0
0
2 0
4 0
6 0 *
IV -1 1 A IV -1 1 B

Figure IV-11. Percent change in maximal mitochondrial respiratory capacity in all participants.4A,
haplogroups A, L, M, and N (red dots) were enriched in the top 50% percentile compared to other haplogroups
(blue dots). 4B, in APOE4 carriers, haplogroups A, L, M, and N had significantly higher maximal
mitochondrial respiratory capacity, whereas in APOE3/3 cells, haplogroups A, L, M, and N tend to have lower
maximal mitochondrial respiratory capacity. APOE4/4 homozygote was denoted by empty square. *p<0.05.

125

Neural Stem Cell Proliferation
We observed no difference in neural stem cell proliferation rate, either stratified by sex or APOE
genotype alone (data not shown). However, we did observe an interplay between sex and APOE
genotype, where in males only, APOE 3/3 carriers had significantly higher proliferation rate than
APOE4 carriers Figure IV-12. No difference was observed between APOE4 heterozygote (solid
squares) and homozygote (open square) (Figure IV-12).
% c h a n g e to v e h ic le
A P O E 3 F
A P O E 3 M
A P O E 4 F
A P O E 4 M
-3 0
-2 0
-1 0
0
1 0
2 0
3 0
*

Figure IV-12. Neural stem cell (NSC) proliferation stratified by sex and APOE genotype. NSCs with male
APOE3/3 genetic background had significantly higher proliferation rate. *p<0.05.

When stratified by mitochondrial haplogroups, haplogroups A, L, M, and N had no significant
different proliferation rate than the others. When further stratified by mitochondrial haplogroups
and APOE genotypes, no statistical significant difference was observed, although APOE3/3 NSCs
with major European mitochondrial haplogroup genetic background seemed to have the highest
proliferation rate (Figure IV-13).
126

% c h a n g e to v e h ic le
A P O E 3 A , L , M , N
A P O E 3 O th e rs
A P O E 4 A , L , M , N
A P O E 4 O th e rs
-3 0
-2 0
-1 0
0
1 0
2 0
3 0

Figure IV-13. NSC proliferation rate by mitochondrial haplogroup and APOE genotype.

Transcriptome analysis
RNA samples from 16 NSC cell lines were used for RNA-Seq due to sample availability and
quality limitation. See Table IV-8 for cell lines by sex and APOE genotype, and
Table IV-9 for cell lines by APOE genotype and mitochondrial haplogroups.
Table IV-8. NSC cell lines used for RNA-Seq by sex and APOE genotype.
Female Male Total
APOE 3/3 2 4 6
APOE 4
carrier 6 4 10
Total 8 8 16

Table IV-9. NSC cell lines used for RNA-Seq by APOE genotype and mitochondrial haplogroup.
APOE 3/3 APOE 4 carrier Total
ALMN 2 5 7
Others 4 5 9
Total 6 10 16

Considering our limited sample size per group, we first compared transcriptome alternations due
to allopregnanolone treatment, then stratified the difference by sex difference, APOE genotype
effect, and mitochondrial genetic variance effect.
127

Because we observed differences in mitochondrial maximal respiratory capacity, we looked at
gene expression of electron transport chain (ETC) subunits both nuclear encoded and
mitochondrial encoded. Interestingly, following allopregnanolone treatment, majority of nuclear
encoded ETC subunits were down-regulated, while most mitochondrial encoded ETC subunits
were upregulated. This parallels with what we observed in chapter III, where neuronal
differentiation of SH-SY5Y cells was associated with significant downregulation of nuclear
OXPHOS genes but upregulation of mitochondrial OXPHOS genes. When stratified by sex, both
nuclear and mitochondrial encoded subunits across all ETC complexes were upregulated in
females compared to males following allopregnanolone treatment. Similarly, when stratified by
APOE genotype and mitochondrial haplogroups, almost all ETC subunits were more upregulated
in APOE4 carriers and in haplogroups A, L, M, and N in comparison to non-APOE4 carriers and
other haplogroups respectively.
To further elucidate the underlying mechanisms of allopregnanolone action and differential
responses among different genetic groups, we took an unbiased, exploratory approach and
investigated significantly altered canonical signaling pathways and upstream regulators predicted
by IPA.
Following allopregnanolone treatment, Ingenuity Pathway Analysis could not predict any
significantly altered signaling pathway or upstream regulator based only on significantly changed
genes. When relaxed selection criteria to p<0.25, IPA predicted that allopregnanolone treatment
resulted in upregulation or Sirtuin signaling pathway, and down-regulation of eIF2 signaling
pathway and oxidative phosphorylation. IPA also predicted activation or Rictor as an upstream
regulator.
128

When stratified by sex, allopregnanolone treatment led to more alternations in male NSCs than in
female NSCs. In female, allopregnanolone activated of multiple growth factor signaling pathways
(NGF, FGF, VEGF, EGF) and integrin signaling pathway. In males, allopregnanolone activated
PTEN signaling, while suppressed eIF2 signaling, oxidative phosphorylation, cAMP/CREB
signaling, androgen signaling, and neuregulin signaling. IPA also predicted activation of Rictor
and BACH1, and inhibition of SREBF1, TP73, UCP1, ING1 (inhibitor of growth factor family
member 1), and Tcf7L2 in male NSCs. Results were summarized in Table IV-10.
Table IV-10. Summary of sex difference in response to Allopregnanolone by RNA-Seq.
Female Male
Function Status Genes and signaling pathways Function Status Genes and signaling pathways
Oxidative
phosphorylation
N/A N/A
Oxidative
phosphorylation
Inhibition
mitochondrial and nuclear
OXPHOS genes
Neuronal growth and
maintenance
Activation NGF, FGF, VEGF, EGF Neurogenesis Inhibition
eIF2 signaling, cAMP/CREB
signaling, SREBF1, Tcf7L2
Extracellular matrix
modulation
Activation integrin signaling pathway
Extracellular matrix
modulation
Inhibition Neuregulin signaling
Proliferation N/A N/A Proliferation Activation TP73, ING1, PTEN signaling

In APOE4 NSCs, IPA predicted activation of integrin signaling, actin cytoskeleton modulation,
Rho GTPase (RhoA and Cdc42) signaling, and TGF- signaling, and at the same time inhibition
of p53 signaling. In support of observations in canonical pathways, upstream regulators include
TGF-, SMAD3, BMP6, PDGF-BB, EGFR, JUNB, FN1, IL4, GATA4, FOXO1, FOXO3,
FOXO4, FOXM1, ETS1, Tcf7L2, and BRD4 were predicted activated, whereas SPDEF, estrogen
receptor, and let-7 were predicted inhibited. In APOE3 NSCs, IPA predicted significant
suppression of eIF2 signaling, mTOR signaling, actin skeleton modulation, and oxidative
phosphorylation. No canonical pathways were upregulated. In terms of upstream regulators,
129

SIRT1, Rictor, TSC2, Max, ACOX1, TRIB3, and were predicted to be activated, whereas SYVN1,
HSPA9, PI3KR1, UCP1, ATF4, FGF2, EGR1, eIF2AK4, MMP1, NEDD9, TP53, and MYC were
predicted inhibited. Results were summarized in Table IV-11.
Table IV-11. Summary of effect of APOE genotype on response to Allopregnanolone by RNA-Seq.

In mitochondrial haplogroups A, L, M, and N (ALMN), allopregnanolone treatment resulted in
significant activation of BMP signaling pathway. Upstream regulators AURK, CTGF, and ANLN
were predicted to be activated, whereas MGEA5 were predicted to be inhibited. In mitochondrial
haplogroups other than A, L, M, and N (Others), only RhoGDI signaling was activated, while eIF2
signaling, oxidative phosphorylation, mTOR signaling, phospholipase C signaling, and NRF2
signaling were inhibited. Correspondingly, Rictor, KDM5A, and TRIB3 were predicted activated,
whereas NRF1, HSPA9, CARM1, RARB, UCP1, ATF4, RB1, ING1, BRCA1, NEDD9, and MYC
were predicted suppressed. Results were summarized in Table IV-12.

APOE 4 carriers APOE 3 participants
Function Status Genes and signaling pathways Function Status Genes and signaling pathways
Oxidative
phosphorylation
N/A N/A
Oxidative
phosphorylation
Inhibition
mitochondrial and nuclear
OXPHOS genes
NSC pool
maintenance
Activation
TGF-b, SMAD3, MBP6, EGFR,
FOXOs, FOXM, Let-7
Anabolic metabolism Inhibition
eIF2 signaling, mTOR signaling,
UCP1, PIK3R1, Sirt1, TSC2,
Rictor, ACOX1
Extracellular matrix
modulation
Activation Rho GTPase, FN1 Cell migration Inhibition NEDD9, MMP1
Proliferation Activation
FGF-BB, TGF-b, EGFR, IL-4,
Tcf7L2, and BRD4
Proliferation and
differentiation
Inhibition FGF2, EGR1, TP53, MYC
130

Table IV-12. Summary of effect of mitochondrial genetic variance on response to Allopregnanolone by RNA-
Seq.

Discussion and Conclusion
In this study, we proposed to determine whether three risk factors for late onset Alzheimer’s
disease: sex difference, APOE genotype, and mitochondrial genetic variances, can predict
therapeutic responders of allopregnanolone using a personalized, in vitro model.
Because allopregnanolone is known to promote mitochondrial bioenergetics in vitro and in rodent
models (Brinton, 2013; Brinton and Wang, 2006a, b; Chen et al., 2011b; Irwin and Brinton, 2014;
Irwin et al., 2014; Irwin et al., 2011a; Singh et al., 2012; Sun et al., 2012; Wang and Brinton, 2008;
Wang et al., 2007; Wang et al., 2005a; Wang et al., 2010a; Wang et al., 2010b), we first
investigated whether these three factors can differentially affect therapeutic outcomes of
allopregnanolone in patient derived neural stem cells. We observed that if disregard sex, APOE
genotype, and mitochondrial genetic variances, there was huge individual difference in effect of
allopregnanolone on maximal mitochondrial respiratory capacity (Solinsky, 2017). When
stratified by sex and APOE genotype, we observed that females had better response than males,
and the APOE4 carriers had better response than non-APOE4 carriers (differences were not
statistically significant). We further observed that the above differences were driven by lower
response from APOE3/3 males, although differences were not statistically significant. When
Haplogroups ALMN Haplogroups Others
Function Status Genes and signaling pathways Function Status Genes and signaling pathways
Oxidative
phosphorylation
N/A N/A
Oxidative
phosphorylation
Inhibition
mitochondrial and nuclear
OXPHOS genes
Neuronal growth and
maintenance
N/A N/A
Glucose and fatty acid
metabolism
Inhibition
mTOR signaling, Phospholipase C
signaling, NRF2 signaling, Rictor,
NRF1, UCP1
Proliferation Activation
BMP, AURK, CTGF, ANLN,
MGEA5
Proliferation and
differentiation
Prevention
KDM5A, RB1, HSPA9, CARM1,
RARB, BRCA1, NEDD9, MYC,
eIF2 signaling
131

stratified by mitochondrial haplogroups, we observed that top responders were enriched in
mitochondrial haplogroups A, L, M, and N (ALMN) in comparison to other haplogroups (Others).
This was especially significant in APOE4 carriers. Coincidently, ALMN is of African and Asian
origin, whereas Others is majorly of European origin, suggesting a potential pharmacogenetic
difference among different ethnic groups. These observations were supported on transcriptome
level, where female, APOE4, and ALMN NSCs had more increased gene expression of ETC
subunits compared to their counterparts respectively.
Because allopregnanolone is also known to improve neural stem cell proliferation in vitro and in
vivo, we also decided to investigate whether the three risk factors can differentially affect
proliferation of patient-derived neural stem cells. Similar to mitochondrial respiratory capacity,
without stratification by sex, APOE genotype or mitochondrial haplogroups, we did not observe
any significant difference following allopregnanolone treatment (Solinsky, 2017). Heterogenous
genetic background in patient-derived NSCs may lead to high noise to signal ratio and obscure the
treatment effect. When stratified by sex and APOE genotype, we observed that APOE3/3 males
had the highest NSC proliferation rate, which is opposite to what we observed for maximal
mitochondrial respiratory capacity. This is most likely because increase in mitochondrial maximal
respiratory capacity is usually accompanied by differentiation of stem cells, thus increased
mitochondrial respiration coupled with reduced proliferation in allopregnanolone treated NSCs
could indicate enhanced neurogenesis (Agostini et al., 2016).
To understand the underlying mechanism of allopregnanolone on NSCs, we conducted an unbiased,
explorative transcriptome analysis using IPA. Canonical pathways and upstream regulators
responsible for the outcomes of each of the three risk factors were identified. While the most
significant outcomes in our functional assays were predicted not by any single factor, but by
132

interplays of two of the above three factors, limitation in sample size restricted our investigation
to a single factor. Despite this limitation, we still observed a reasonable number of differentially
altered canonical pathways and upstream regulators for each dimorphism.
While distracted by large individual differences, IPA predicted that allopregnanolone exerts its
effect through metabolic and energetic circuits. Upregulation of SIRT1, which is involved in
nutrient sensing and is activated in conditions such as calorie restriction, coupled with down-
regulation of Rictor (rapamycin-insensitive companion of mTOR), a specific component of
mTORC2, and eIF2 (eukaryotic initiation factor) may be a response to increased energy demand
(in an environment with limited nutrient level) following allopregnanolone treatment.
Female and males had different response to allopregnanolone treatment. In females,
allopregnanolone may promote neural stem cell proliferation and differentiation through activation
of multiple growth factors that are involved in neuronal growth and maintenance and molecules
involved in extracellular matrix modulation. In males, allopregnanolone seemed to inhibit
differentiation while promote cell proliferation, as evident in suppression of genes involved in
dendritic formation and neurogenesis (SREBF1, Tcf7L2) (Gurok et al., 2004; Ziegler et al., 2017)
and tumor suppressor genes (TP73 and ING1). This together helped explain why female NSCs
tended to have higher mitochondrial respiratory capacity, but lower proliferation rate.
The effect of allopregnanolone on APOE4 carrier NSCs was three-fold. First, it maintained neural
stem cell pool homeostasis by preserving progenitor cell quiescence and reducing self-renewal
(TGF-, SMAD3, MBP6, EGFR, FOXOs, FOXM, and Let-7) (Aguirre et al., 2010; Hou et al.,
2015; Kandasamy et al., 2011; Kim et al., 2015; Lee et al., 2016; Paik et al., 2009; Renault et al.,
2009; Ro et al., 2013; Vilchez et al., 2013). Second, allopregnanolone actively modulates
cytoskeleton reorganization and extracellular matrix modification, preparing for growth cone
133

development (Rho GTPase, FN1) (Stankiewicz and Linseman, 2014). And last, it induced early
stage of neurogenesis and promoted survival of newly generated neurons (PDGF-BB, TGF-,
EGFR, IL-4, Tcf7L2, and BRD4) (Aguirre et al., 2010; Bhattarai et al., 2016; Butovsky et al.,
2006; Chodelkova et al., 2018; Erlandsson et al., 2006; Kandasamy et al., 2014; Li et al., 2016).
On the other hand, in APOE3 NSCs, transcriptome signature indicated cellular response to nutrient
depletion, as evident in reduced glucose and fatty acid metabolism (Sirt1, TSC2, Rictor, UCP1,
ACOX1, PIK3R1), reduced cell migration (NEDD9, MMP1), and altered stress response (TRIB3,
SYVN1, HSPA9, ATF4) (Saleem and Biswas, 2017; Yang et al., 2016). Transcriptomic profile on
NSC proliferation and differentiation is hard to predict (FGF2, EGR1, TP53, MYC) (Arsenijevic
et al., 2001; Cera et al., 2018; Taupin et al., 2000). Considering that APOE3 NSCs appeared more
proliferative in functional assay, one possible explanation is that allopregnanolone induced
APOE3 NSCs proliferation, but increased neural sphere size led to limited nutrient access for inner
cells, thus resulted in the nutrient depletion phenotype on transcriptome level.
In ALMN NSCs, allopregnanolone effect pointed to activation of NSCs through enhanced
proliferation (BMP, AURK, CTGF, ANLN, MGEA5). And in Others NSCs, allopregnanolone
promoted preserving of neural stem cell pool, as characterized by reduction in energy metabolism
and mitochondrial biogenesis (Rictor, NRF1, UCP1), prevention of differentiation (KDM5A,
RB1), and restriction of proliferation (HSPA9, CARM1, RARB, BRCA1, NEDD9, MYC). These
observations are in accordance with our observation that ALMN NSCs had higher mitochondrial
respiratory capacity. The differential effect of allopregnanolone in these two clusters of
mitochondrial haplogroups may be explained by differences in their endogenous level of
allopregnanolone (Girdler et al., 2006).
134

The lack of effect or even negative effect of allopregnanolone on NSCs with low-risk genotypes
may be due to their healthier status at baseline, as allopregnanolone is not expected to induce
neurogenesis in healthy, active NSC populations (Brinton, 2013).
Together, our data demonstrated that effect of allopregnanolone can be largely affected by
individual variances, where females, APOE4 carriers, and haplogroups A, L, M, and N benefited
more from allopregnanolone in neurogenesis than their counterparts. Interplay of these three
factors may further modulate allopregnanolone effect. These results also suggested that certain
genetic markers can be used to differentiated and predict allopregnanolone treatment effect in vitro.
We previously observed a positive correlation between maximal mitochondrial respiratory
capacity in patient-derived NSCs and left hippocampal volume change in patients (Solinsky, 2017).
If certain genetic markers can differentiate responders from non-responders in in vitro assays, there
is great potential that they can be used to predict therapeutic outcomes. Indeed, after overlaying
patient sex, APOE genotype, and mitochondrial haplogroup information onto this correlation data,
we observed that APOE genotype together with mitochondrial haplogroup in vitro is predictive of
therapeutic outcome of allopregnanolone in patient (Figure IV-14).

Figure IV-14. In vitro mitochondrial respiratory capacity is predictive of in patient allopregnanolone
therapeutic outcome.

135

While we propose to use genetic markers to identify responders and use in vitro assays to predict
therapeutic outcomes in patients, question arose as whether there are better biomarkers or in vitro
assays to use? Given that therapeutics for Alzheimer’s disease require long-term treatment to
achieve clinically relevant outcomes, how effective can short-term in vitro treatment predict actual
therapeutic effect? And further, how well can in vitro assays predict the ultimate goal of AD
therapy: improvement in cognitive function? Only a phase II efficacy trial can answer these
questions.
We are also aware of limitations of this study. First, our sample size is limited, as they are derived
from a Phase Ib / IIa clinical study, where efficacy is not a designed study endpoint. Second, NSCs
differentiated from patient-derived iPSCs may have lost their aging phenotype during
reprogramming, thus the observations made in vitro may have limitations in prediction of clinical
outcomes.
Keeping these unknowns and limitations in mind, we are still excited to report that simple genetic
markers such as sex, APOE genotype, and mitochondrial genetic variances can differentiate
allopregnanolone treatment effect in vitro and predict its therapeutic outcome in patients. And
these preliminary results can provide insights into future clinical trial design, data analysis, as well
as therapeutic regimes. It also strongly supports the potential for Alzheimer’s disease therapies
towards a personalized medicine approach.
Acknowledgement
This is a retrospective analysis. I thank Dr. Christine Solinsky for generating the NSCs, and for
sharing her data on extracellular metabolism flux assays and the NSC regeneration assays. I also
thank Dr. “Raymond” Yuan Shang for his assistance with RNA-Seq raw file processing.
136

CHAPTER V

DISCUSSION AND CONCLUDING REMARKS

In this thesis study, we proposed to address a program of questions revolved around the
mitochondrial respiratory risk phenotype and late onset Alzheimer’s disease. We believed that a
comprehensive understanding of the metabolic and respiratory profile at key transition points, their
underlying mechanism and key players can help identify potential intervention targets and
windows.
Because females have higher lifetime risk for late onset AD, and the aging female brain shares the
same glucose hypometabolism phenotype as prodromal AD brain, understanding the dismantling
process of female brain bioenergetics system during the perimenopausal transition holds key to
better elucidate the etiology of late onset AD. We hypothesized that this process is associated with
changes in the metabolic profile and mitochondrial OXPHOS gene expression, which could be
driven by the loss of estradiol during the endocrine transition.
We observed that in the perimenopausal female brain, OXPHOS genes encoded by the
mitochondrial genome and the nuclear genome have different expression patterns. At the early
stage of perimenopause, mitochondrial encoded genes tended to be upregulated, whereas nuclear
encoded genes tended to be downregulated. As the transition approaches completion, the trends
flipped. This demonstrated a coordinated and mutual compensatory pattern between the two
genomes. We further observed that ovariectomy in the post-menopausal female led to significant
reduction of both mitochondrial and nuclear OXPHOS gene expression, and that this change was
associated with diminished estradiol level but not progesterone level, supporting our hypothesis
that estradiol may regulate brain metabolic profile through regulation of OXPHOS gene expression.
137

This observation also confers with clinical studies on hormone replacement therapies, where only
estrogen replacement therapy, but not estrogen and progesterone combined therapy tended to be
neuroprotective in post-menopausal females (Espeland et al., 2013; Gleason et al., 2015).
We further examined the global metabolic profile of female brain at each stage of chronological
and endocrinological aging. We detected that glucose metabolism was hit first at onset of
perimenopausal transition, and the transitioning brain increased fatty acid and ketone body
metabolism potentially as compensatory mechanism. As the mitochondria became even less
efficient with aging, even fatty acid metabolism got stalled, and the brain turned to anaerobic
respiration for energy production. These dynamic metabolic changes in the female brain supported
the “critical window” hypothesis of estrogen therapy, where treatment need to be initiated shortly
after menopause when mitochondrial respiratory capacity is still salvageable to ensure protection
and efficiency.
Since we have confirmed that estradiol was able to regulate OXPHOS expression, the question
follows was how. Literature revealed that ER and ER has different intracellular localization
within the brain. In hippocampal neurons, for example, ER is mainly in the nuclei while ER has
been observed in the cytosol and mitochondrial (Irwin et al., 2012; Milner et al., 2005; Simpkins
et al., 2008b; Spencer-Segal et al., 2012; Yang et al., 2004; Yang et al., 2009). Further, estrogen
receptors can mediate estradiol signaling in have multiple ways, including direct DNA binding,
indirect binding with co-factors, and non-genomic interactions with growth factors. we proposed
to elucidate how the two major estrogen receptors modulates mitochondrial gene expression and
bioenergetics.
To achieve this goal, we used primary rat embryonic neurons and astrocytes, and treated them with
either ER or ER selective antagonists. We confirmed a dramatic difference in respiratory
138

phenotype between neurons and astrocytes, and observed that unlike astrocytes, neurons relied
heavily on ER to mediate estradiol induced OXPHOS gene transcriptional activity. We further
confirmed this result in human mitotic neuroblastoma cells, which mimicked astrocytic respiratory
capacity, and differentiated cells, which had a more mature neuronal phenotype. This can be
explained by the significantly increased ESR2 gene expression and the 25% decreased ESR1 gene
expression in the differentiated cells. In this study, using transcriptome analysis, we further
identified involvement of ER and ER in mitochondrial bioenergetics under unliganded
condition.
Given the presence of multiple putative hormone response elements in the mitochondrial genome
and presence of nuclear receptors and transcriptional factors such as glucocorticoid receptor (RG),
ER, NFkB, and AP1, it is highly likely that there these receptors and transcriptional factors can
directly regulate mitochondrial gene expression through mtDNA binding (Lee et al., 2008). Yet
the binding affinity is highly dependent upon the binding sequence. Given that mitochondria
harbor diverse SNPs, it was not hard to image that mitochondrial genetic variance can affect the
regulatory effect of ER direct transcriptional activity. On the same axis, studies using cytoplasmic
hybrid cells demonstrated significant difference in mitochondrial gene expression and
bioenergetics between different mitochondrial haplogroups. We hypothesized that similar results
can be observed in therapeutic outcomes of drugs targeting mitochondrial function.
To test this hypothesis, we conducted a retrospective analysis on two clinical trials previously done
in the lab. The first trial tested the effect of PhytoSERM, an ER selective modulator for its effect
on hot flash and cognitive function in early post-menopausal females. The second trial tested
allopregnanolone, an endogenous molecule as a regenerative therapy for AD. Both molecules were
known to promote mitochondrial respiration and function. Unlike studies using cell lines or animal
139

models, in human studies are associated with huge individual differences. To account for this
difference, we included two other genetic risk factors for late onset AD in these analysis: APOE
genotype and sex difference. By stratifying clinical data into different combinations of these three
genetic factors, we confirmed our hypothesis that therapeutic outcomes of potential Alzheimer’s
disease therapies targeting mitochondrial bioenergetics can be modulated by mitochondrial genetic
variances, along with nuclear genetic risk factors.
Together, the program of studies included in this doctoral thesis deepened our understanding of
bioenergetics transformation in the aging female brain and the underlying regulatory effect by
specific estrogen receptors. It also linked observations in respiratory and metabolic risk phenotype
to differential AD therapeutics outcomes. These studies further shed light on a practical genetic
marker-based precision medicine approach for development and clinical trial design of
Alzheimer’s disease therapeutics.

140

REFERENCES
Adams, S.H., Hoppel, C.L., Lok, K.H., Zhao, L., Wong, S.W., Minkler, P.E., Hwang, D.H.,
Newman, J.W., and Garvey, W.T. (2009). Plasma Acylcarnitine Profiles Suggest Incomplete
Long-Chain Fatty Acid β-Oxidation and Altered Tricarboxylic Acid Cycle Activity in Type 2
Diabetic African-American Women. J Nutr 139, 1073-1081.
Adams, S.M., Aksenova, M.V., Aksenov, M.Y., Mactutus, C.F., and Booze, R.M. (2010). ER-β
mediates 17 β-estradiol attenuation of HIV-1 Tat-induced apoptotic signaling. Synapse (New York,
NY) 64, 829-838.
Agostini, M., Romeo, F., Inoue, S., Niklison-Chirou, M.V., Elia, A.J., Dinsdale, D., Morone, N.,
Knight, R.A., Mak, T.W., and Melino, G. (2016). Metabolic reprogramming during neuronal
differentiation. Cell death and differentiation 23, 1502-1514.
Aguero-Torres, H., Fratiglioni, L., Guo, Z., Viitanen, M., and Winblad, B. (1998). Prognostic
factors in very old demented adults: a seven-year follow-up from a population-based survey in
Stockholm. Journal of the American Geriatrics Society 46, 444-452.
Aguirre, A., Rubio, M.E., and Gallo, V. (2010). Notch and EGFR pathway interaction regulates
neural stem cell number and self-renewal. Nature 467, 323-327.
Aksenov, M.Y., Tucker, H.M., Nair, P., Aksenova, M.V., Butterfield, D.A., Estus, S., and
Markesbery, W.R. (1999). The expression of several mitochondrial and nuclear genes encoding
the subunits of electron transport chain enzyme complexes, cytochrome c oxidase, and NADH
dehydrogenase, in different brain regions in Alzheimer's disease. Neurochemical research 24, 767-
774.
Alvarez-Delgado, C., Mendoza-Rodriguez, C.A., Picazo, O., and Cerbon, M. (2010). Different
expression of alpha and beta mitochondrial estrogen receptors in the aging rat brain: interaction
with respiratory complex V. Experimental gerontology 45, 580-585.
Alzheimer's Association (2015). 2015 Alzheimer's disease facts and figures. Alzheimer's &
dementia : the journal of the Alzheimer's Association 11, 332-384.
Alzheimer's Association (2018). 2018 Alzheimer's disease facts and figures. Alzheimer's &
Dementia: The Journal of the Alzheimer's Association 14, 367-429.
Amo, T., Yadava, N., Oh, R., Nicholls, D.G., and Brand, M.D. (2008). Experimental assessment
of bioenergetic differences caused by the common European mitochondrial DNA haplogroups H
and T. Gene 411, 69-76.
Aronica, S.M., Kraus, W.L., and Katzenellenbogen, B.S. (1994). Estrogen action via the cAMP
signaling pathway: stimulation of adenylate cyclase and cAMP-regulated gene transcription.
Proceedings of the National Academy of Sciences of the United States of America 91, 8517-8521.
Arsenijevic, Y., Weiss, S., Schneider, B., and Aebischer, P. (2001). Insulin-Like Growth Factor-I
Is Necessary for Neural Stem Cell Proliferation and Demonstrates Distinct Actions of Epidermal
Growth Factor and Fibroblast Growth Factor-2. The Journal of Neuroscience 21, 7194.
Bagger, Y.Z., Tanko, L.B., Alexandersen, P., Qin, G., and Christiansen, C. (2005). Early
postmenopausal hormone therapy may prevent cognitive impairment later in life. Menopause
(New York, NY) 12, 12-17.
Baloyannis, S.J., Costa, V., and Michmizos, D. (2004). Mitochondrial alterations in Alzheimer's
disease. American journal of Alzheimer's disease and other dementias 19, 89-93.
Barnes, L.L., Wilson, R.S., Bienias, J.L., Schneider, J.A., Evans, D.A., and Bennett, D.A. (2005).
SEx differences in the clinical manifestations of alzheimer disease pathology. Archives of General
Psychiatry 62, 685-691.
141

Beal, M.F. (1996). Mitochondria, free radicals, and neurodegeneration. Current opinion in
neurobiology 6, 661-666.
Bean, L.A., Ianov, L., and Foster, T.C. (2014). Estrogen Receptors, the Hippocampus, and
Memory. The Neuroscientist : a review journal bringing neurobiology, neurology and psychiatry
20, 534-545.
Bélanger, M., Allaman, I., and Magistretti, Pierre J. (2011). Brain Energy Metabolism: Focus on
Astrocyte-Neuron Metabolic Cooperation. Cell metabolism 14, 724-738.
Bettini, E., and Maggi, A. (1992). Estrogen induction of cytochrome c oxidase subunit III in rat
hippocampus. Journal of neurochemistry 58, 1923-1929.
Bhattarai, P., Thomas, A.K., Cosacak, M.I., Papadimitriou, C., Mashkaryan, V., Froc, C.,
Reinhardt, S., Kurth, T., Dahl, A., Zhang, Y., et al. (2016). IL4/STAT6 Signaling Activates Neural
Stem Cell Proliferation and Neurogenesis upon Amyloid-beta42 Aggregation in Adult Zebrafish
Brain. Cell Rep 17, 941-948.
Bi, R., Zhang, W., Yu, D., Li, X., Wang, H.Z., Hu, Q.X., Zhang, C., Lu, W., Ni, J., Fang, Y., et al.
(2015). Mitochondrial DNA haplogroup B5 confers genetic susceptibility to Alzheimer's disease
in Han Chinese. Neurobiology of aging 36, 1604 e1607-1604 e1616.
Binder, E.F., Schechtman, K.B., Birge, S.J., Williams, D.B., and Kohrt, W.M. (2001). Effects of
hormone replacement therapy on cognitive performance in elderly women. Maturitas 38, 137-146.
Bjornstrom, L., Kilic, E., Norman, M., Parker, M.G., and Sjoberg, M. (2001). Cross-talk between
Stat5b and estrogen receptor-alpha and -beta in mammary epithelial cells. Journal of molecular
endocrinology 27, 93-106.
Blazquez, C., Sanchez, C., Daza, A., Galve-Roperh, I., and Guzman, M. (1999). The stimulation
of ketogenesis by cannabinoids in cultured astrocytes defines carnitine palmitoyltransferase I as a
new ceramide-activated enzyme. Journal of neurochemistry 72, 1759-1768.
Blobel, G.A., Sieff, C.A., and Orkin, S.H. (1995). Ligand-dependent repression of the erythroid
transcription factor GATA-1 by the estrogen receptor. Molecular and cellular biology 15, 3147-
3153.
Bosetti, F., Brizzi, F., Barogi, S., Mancuso, M., Siciliano, G., Tendi, E.A., Murri, L., Rapoport,
S.I., and Solaini, G. (2002). Cytochrome c oxidase and mitochondrial F1F0-ATPase (ATP
synthase) activities in platelets and brain from patients with Alzheimer's disease. Neurobiology of
aging 23, 371-376.
Boveris, A., and Navarro, A. (2008). Brain mitochondrial dysfunction in aging. IUBMB Life 60,
308-314.
Bowling, A.C., Mutisya, E.M., Walker, L.C., Price, D.L., Cork, L.C., and Beal, M.F. (1993). Age-
dependent impairment of mitochondrial function in primate brain. Journal of neurochemistry 60,
1964-1967.
Bozek, K., Wei, Y., Yan, Z., Liu, X., Xiong, J., Sugimoto, M., Tomita, M., Paabo, S., Pieszek, R.,
Sherwood, C.C., et al. (2014). Exceptional evolutionary divergence of human muscle and brain
metabolomes parallels human cognitive and physical uniqueness. PLoS biology 12, e1001871.
Brandenberger, A.W., Tee, M.K., Lee, J.Y., Chao, V., and Jaffe, R.B. (1997). Tissue Distribution
of Estrogen Receptors Alpha (ER-α) and Beta (ER-β) mRNA in the Midgestational Human Fetus.
The Journal of Clinical Endocrinology & Metabolism 82, 3509-3512.
Bray, N.L., Pimentel, H., Melsted, P., and Pachter, L. (2016). Near-optimal probabilistic RNA-seq
quantification. Nature biotechnology 34, 525-527.
Brenner, D.E., Kukull, W.A., Stergachis, A., van Belle, G., Bowen, J.D., McCormick, W.C., Teri,
L., and Larson, E.B. (1994). Postmenopausal estrogen replacement therapy and the risk of
142

Alzheimer's disease: a population-based case-control study. American journal of epidemiology
140, 262-267.
Brinton, R.D. (2008a). Estrogen Regulation of Glucose Metabolism and Mitochondrial Function:
Therapeutic Implications for Prevention of Alzheimer’s Disease. Advanced drug delivery reviews
60, 1504-1511.
Brinton, R.D. (2008b). The healthy cell bias of estrogen action: mitochondrial bioenergetics and
neurological implications. Trends in neurosciences 31, 529-537.
Brinton, R.D. (2009). Estrogen-induced plasticity from cells to circuits: predictions for cognitive
function. Trends in pharmacological sciences 30, 212-222.
Brinton, R.D. (2013). Neurosteroids as regenerative agents in the brain: therapeutic implications.
Nature reviews Endocrinology 9, 241-250.
Brinton, R.D., and Wang, J.M. (2006a). Preclinical analyses of the therapeutic potential of
allopregnanolone to promote neurogenesis in vitro and in vivo in transgenic mouse model of
Alzheimer's disease. Current Alzheimer research 3, 11-17.
Brinton, R.D., and Wang, J.M. (2006b). Therapeutic potential of neurogenesis for prevention and
recovery from Alzheimer's disease: allopregnanolone as a proof of concept neurogenic agent.
Current Alzheimer research 3, 185-190.
Brinton, R.D., Yao, J., Yin, F., Mack, W.J., and Cadenas, E. (2015). Perimenopause as a
neurological transition state. Nature reviews Endocrinology.
Brookmeyer, R., Gray, S., and Kawas, C. (1998). Projections of Alzheimer's disease in the United
States and the public health impact of delaying disease onset. American journal of public health
88, 1337-1342.
Bubber, P., Haroutunian, V., Fisch, G., Blass, J.P., and Gibson, G.E. (2005). Mitochondrial
abnormalities in Alzheimer brain: mechanistic implications. Ann Neurol 57, 695-703.
Bunone, G., Briand, P.A., Miksicek, R.J., and Picard, D. (1996). Activation of the unliganded
estrogen receptor by EGF involves the MAP kinase pathway and direct phosphorylation. The
EMBO journal 15, 2174-2183.
Burke, G.L., Legault, C., Anthony, M., Bland, D.R., Morgan, T.M., Naughton, M.J., Leggett, K.,
Washburn, S.A., and Vitolins, M.Z. (2003). Soy protein and isoflavone effects on vasomotor
symptoms in peri- and postmenopausal women: the Soy Estrogen Alternative Study. Menopause
(New York, NY) 10, 147-153.
Burte, F., Carelli, V., Chinnery, P.F., and Yu-Wai-Man, P. (2015). Disturbed mitochondrial
dynamics and neurodegenerative disorders. Nature reviews Neurology 11, 11-24.
Butovsky, O., Ziv, Y., Schwartz, A., Landa, G., Talpalar, A.E., Pluchino, S., Martino, G., and
Schwartz, M. (2006). Microglia activated by IL-4 or IFN-gamma differentially induce
neurogenesis and oligodendrogenesis from adult stem/progenitor cells. Molecular and cellular
neurosciences 31, 149-160.
Cadenas, E. (2004). Mitochondrial free radical production and cell signaling. Molecular aspects of
medicine 25, 17-26.
Cardoso, S.M., Proenca, M.T., Santos, S., Santana, I., and Oliveira, C.R. (2004). Cytochrome c
oxidase is decreased in Alzheimer's disease platelets. Neurobiology of aging 25, 105-110.
Carrieri, G., Bonafè, M., De Luca, M., Rose, G., Varcasia, O., Bruni, A., Maletta, R., Nacmias, B.,
Sorbi, S., Corsonello, F., et al. (2001). Mitochondrial DNA haplogroups and APOE4 allele are
non-independent variables in sporadic Alzheimer's disease. Human genetics 108, 194-198.
Casini, M.L., Marelli, G., Papaleo, E., Ferrari, A., D'Ambrosio, F., and Unfer, V. (2006).
Psychological assessment of the effects of treatment with phytoestrogens on postmenopausal
143

women: a randomized, double-blind, crossover, placebo-controlled study. Fertility and sterility 85,
972-978.
Cassarino, D.S., Swerdlow, R.H., Parks, J.K., Parker, W.D., Jr., and Bennett, J.P., Jr. (1998).
Cyclosporin A increases resting mitochondrial membrane potential in SY5Y cells and reverses the
depressed mitochondrial membrane potential of Alzheimer's disease cybrids. Biochemical and
biophysical research communications 248, 168-173.
Catana, C., Drzezga, A., Heiss, W.D., and Rosen, B.R. (2012). PET/MRI for neurologic
applications. Journal of nuclear medicine : official publication, Society of Nuclear Medicine 53,
1916-1925.
Cera, A.A., Cacci, E., Toselli, C., Cardarelli, S., Bernardi, A., Gioia, R., Giorgi, M., Poiana, G.,
and Biagioni, S. (2018). Egr-1 Maintains NSC Proliferation and Its Overexpression Counteracts
Cell Cycle Exit Triggered by the Withdrawal of Epidermal Growth Factor. Developmental
neuroscience 40, 223-233.
Chagnon, P., Gee, M., Filion, M., Robitaille, Y., Belouchi, M., and Gauvreau, D. (1999).
Phylogenetic analysis of the mitochondrial genome indicates significant differences between
patients with Alzheimer disease and controls in a French-Canadian founder population. American
Journal of Medical Genetics 85, 20-30.
Chambers, S.M., Fasano, C.A., Papapetrou, E.P., Tomishima, M., Sadelain, M., and Studer, L.
(2009). Highly efficient neural conversion of human ES and iPS cells by dual inhibition of SMAD
signaling. Nature biotechnology 27, 275-280.
Chandrasekaran, K., Giordano, T., Brady, D.R., Stoll, J., Martin, L.J., and Rapoport, S.I. (1994).
Impairment in mitochondrial cytochrome oxidase gene expression in Alzheimer disease. Brain
research Molecular brain research 24, 336-340.
Chandrasekaran, K., Hatanpaa, K., Rapoport, S.I., and Brady, D.R. (1997). Decreased expression
of nuclear and mitochondrial DNA-encoded genes of oxidative phosphorylation in association
neocortex in Alzheimer disease. Brain research Molecular brain research 44, 99-104.
Chang, S., ran Ma, T., Miranda, R.D., Balestra, M.E., Mahley, R.W., and Huang, Y. (2005). Lipid-
and receptor-binding regions of apolipoprotein E4 fragments act in concert to cause mitochondrial
dysfunction and neurotoxicity. Proceedings of the National Academy of Sciences of the United
States of America 102, 18694-18699.
Chaudhri, R.A., Schwartz, N., Elbaradie, K., Schwartz, Z., and Boyan, B.D. (2014). Role of
ERalpha36 in membrane-associated signaling by estrogen. Steroids 81, 74-80.
Chen, H.K., Ji, Z.S., Dodson, S.E., Miranda, R.D., Rosenblum, C.I., Reynolds, I.J., Freedman,
S.B., Weisgraber, K.H., Huang, Y., and Mahley, R.W. (2011a). Apolipoprotein E4 domain
interaction mediates detrimental effects on mitochondria and is a potential therapeutic target for
Alzheimer disease. The Journal of biological chemistry 286, 5215-5221.
Chen, J., Russo, P.A., Cooke, C., Russo, I.H., and Russo, J. (2007). ERbeta shifts from
mitochondria to nucleus during estrogen-induced neoplastic transformation of human breast
epithelial cells and is involved in estrogen-induced synthesis of mitochondrial respiratory chain
proteins. Biochimica et biophysica acta 1773.
Chen, J.-Q., Cammarata, P.R., Baines, C.P., and Yager, J.D. (2009). Regulation of mitochondrial
respiratory chain biogenesis by estrogens/estrogen receptors and physiological, pathological and
pharmacological implications. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research
1793, 1540-1570.
144

Chen, J.Q., Delannoy, M., Cooke, C., and Yager, J.D. (2004a). Mitochondrial localization of
ERalpha and ERbeta in human MCF7 cells. American journal of physiology Endocrinology and
metabolism 286, E1011-1022.
Chen, J.Q., Eshete, M., Alworth, W.L., and Yager, J.D. (2004b). Binding of MCF-7 cell
mitochondrial proteins and recombinant human estrogen receptors α and β to human mitochondrial
dna estrogen response elements. Journal of cellular biochemistry 93, 358-373.
Chen, S., Wang, J.M., Irwin, R.W., Yao, J., Liu, L., and Brinton, R.D. (2011b). Allopregnanolone
Promotes Regeneration and Reduces β-Amyloid Burden in a Preclinical Model of Alzheimer's
Disease. PloS one 6, e24293.
Cheng, C.M., Cohen, M., Wang, J., and Bondy, C.A. (2001). Estrogen augments glucose
transporter and IGF1 expression in primate cerebral cortex. FASEB journal : official publication
of the Federation of American Societies for Experimental Biology 15, 907-915.
Chetelat, G., Desgranges, B., de la Sayette, V., Viader, F., Eustache, F., and Baron, J.C. (2003).
Mild cognitive impairment: Can FDG-PET predict who is to rapidly convert to Alzheimer's
disease? Neurology 60, 1374-1377.
Chinnery, P.F., Taylor, G.A., Howell, N., Andrews, R.M., Morris, C.M., Taylor, R.W., McKeith,
I.G., Perry, R.H., Edwardson, J.A., and Turnbull, D.M. (2000). Mitochondrial DNA haplogroups
and susceptibility to AD and dementia with Lewy bodies. Neurology 55, 302-304.
Chlebowski, R.T., Anderson, G., Manson, J.E., Pettinger, M., Yasmeen, S., Lane, D., Langer, R.D.,
Hubbell, F.A., McTiernan, A., Hendrix, S., et al. (2010a). Estrogen alone in postmenopausal
women and breast cancer detection by means of mammography and breast biopsy. Journal of
clinical oncology : official journal of the American Society of Clinical Oncology 28, 2690-2697.
Chlebowski, R.T., Anderson, G.L., Gass, M., Lane, D.S., Aragaki, A.K., Kuller, L.H., Manson,
J.E., Stefanick, M.L., Ockene, J., Sarto, G.E., et al. (2010b). Estrogen plus progestin and breast
cancer incidence and mortality in postmenopausal women. Jama 304, 1684-1692.
Chlebowski, R.T., Kuller, L.H., Prentice, R.L., Stefanick, M.L., Manson, J.E., Gass, M., Aragaki,
A.K., Ockene, J.K., Lane, D.S., Sarto, G.E., et al. (2009). Breast cancer after use of estrogen plus
progestin in postmenopausal women. N Engl J Med 360, 573-587.
Chodelkova, O., Masek, J., Korinek, V., Kozmik, Z., and Machon, O. (2018). Tcf7L2 is essential
for neurogenesis in the developing mouse neocortex. Neural Development 13.
Chu, S., and Fuller, P.J. (1997). Identification of a splice variant of the rat estrogen receptor β gene.
Molecular and Cellular Endocrinology 132, 195-199.
Chun, W., and Johnson, G.V. (2007). The role of tau phosphorylation and cleavage in neuronal
cell death. Frontiers in bioscience : a journal and virtual library 12, 733-756.
Clarkson, T.B., Utian, W.H., Barnes, S., Gold, E.B., Basaria, S.S., Aso, T., Kronenberg, F.,
Frankenfeld, C.L., Cline, J.M., Landgren, B.-M., et al. (2011). The role of soy isoflavones in
menopausal health: report of The North American Menopause Society/Wulf H. Utian
Translational Science Symposium in Chicago, IL (October 2010). Menopause (New York, NY)
18, 732-753.
Coker, L.H., Espeland, M.A., Rapp, S.R., Legault, C., Resnick, S.M., Hogan, P., Gaussoin, S.,
Dailey, M., and Shumaker, S.A. (2010). Postmenopausal hormone therapy and cognitive
outcomes: The Women's Health Initiative Memory Study (WHIMS). The Journal of Steroid
Biochemistry and Molecular Biology 118, 304-310.
Compton, D.R., Sheng, S., Carlson, K.E., Rebacz, N.A., Lee, I.Y., Katzenellenbogen, B.S., and
Katzenellenbogen, J.A. (2004). Pyrazolo[1,5-a]pyrimidines: estrogen receptor ligands possessing
estrogen receptor beta antagonist activity. Journal of medicinal chemistry 47, 5872-5893.
145

Corder, E., Saunders, A., Strittmatter, W., Schmechel, D., Gaskell, P., Small, G., Roses, A., Haines,
J., and Pericak-Vance, M.A. (1993). Gene dose of apolipoprotein E type 4 allele and the risk of
Alzheimer’s disease in late onset families. Science 261, 921-923.
Corder, E.H., Ghebremedhin, E., Taylor, M.G., Thal, D.R., Ohm, T.G., and Braak, H. (2004). The
biphasic relationship between regional brain senile plaque and neurofibrillary tangle distributions:
modification by age, sex, and APOE polymorphism. Annals of the New York Academy of
Sciences 1019, 24-28.
Coskun, P., Wyrembak, J., Schriner, S.E., Chen, H.W., Marciniack, C., Laferla, F., and Wallace,
D.C. (2012). A mitochondrial etiology of Alzheimer and Parkinson disease. Biochimica et
biophysica acta 1820, 553-564.
Coto, E., Gomez, J., Alonso, B., Corao, A.I., Diaz, M., Menendez, M., Martinez, C., Calatayud,
M.T., Moris, G., and Alvarez, V. (2011). Late-onset Alzheimer's disease is associated with
mitochondrial DNA 7028C/haplogroup H and D310 poly-C tract heteroplasmy. Neurogenetics 12,
345-346.
Couse, J.F., Lindzey, J., Grandien, K., Gustafsson, J.A., and Korach, K.S. (1997). Tissue
distribution and quantitative analysis of estrogen receptor-alpha (ERalpha) and estrogen receptor-
beta (ERbeta) messenger ribonucleic acid in the wild-type and ERalpha-knockout mouse.
Endocrinology 138, 4613-4621.
Cowley, S.M., Hoare, S., Mosselman, S., and Parker, M.G. (1997). Estrogen Receptors α and β
Form Heterodimers on DNA. Journal of Biological Chemistry 272, 19858-19862.
Cowley, S.M., and Parker, M.G. (1999). A comparison of transcriptional activation by ER alpha
and ER beta. J Steroid Biochem Mol Biol 69, 165-175.
Cummings, J.L., Morstorf, T., and Zhong, K. (2014). Alzheimer's disease drug-development
pipeline: few candidates, frequent failures. Alzheimer's research & therapy 6, 37.
Davis, R.E., Miller, S., Herrnstadt, C., Ghosh, S.S., Fahy, E., Shinobu, L.A., Galasko, D., Thal,
L.J., Beal, M.F., Howell, N., et al. (1997). Mutations in mitochondrial cytochrome c oxidase genes
segregate with late-onset Alzheimer disease. Proceedings of the National Academy of Sciences of
the United States of America 94, 4526-4531.
de Leon, M.J., Ferris, S.H., George, A.E., Reisberg, B., Christman, D.R., Kricheff, II, and Wolf,
A.P. (1983). Computed tomography and positron emission transaxial tomography evaluations of
normal aging and Alzheimer's disease. Journal of cerebral blood flow and metabolism : official
journal of the International Society of Cerebral Blood Flow and Metabolism 3, 391-394.
de Medeiros, S.R.B., Krey, G., Hihi, A.K., and Wahli, W. (1997). Functional Interactions between
the Estrogen Receptor and the Transcription Activator Sp1 Regulate the Estrogen-dependent
Transcriptional Activity of the Vitellogenin A1 ioPromoter. Journal of Biological Chemistry 272,
18250-18260.
De Santi, S., de Leon, M.J., Rusinek, H., Convit, A., Tarshish, C.Y., Roche, A., Tsui, W.H., Kandil,
E., Boppana, M., Daisley, K., et al. (2001). Hippocampal formation glucose metabolism and
volume losses in MCI and AD. Neurobiology of aging 22, 529-539.
Demonacos, C.V., Karayanni, N., Hatzoglou, E., Tsiriyiotis, C., Spandidos, D.A., and Sekeris, C.E.
(1996). Mitochondrial genes as sites of primary action of steroid hormones. Steroids 61, 226-232.
Diaz Brinton, R., Chen, S., Montoya, M., Hsieh, D., Minaya, J., Kim, J., and Chu, H.P. (2000).
The women's health initiative estrogen replacement therapy is neurotrophic and neuroprotective.
Neurobiology of aging 21, 475-496.
146

Ding, F., Yao, J., Rettberg, J.R., Chen, S., and Brinton, R.D. (2013a). Early Decline in Glucose
Transport and Metabolism Precedes Shift to Ketogenic System in Female Aging and Alzheimer's
Mouse Brain: Implication for Bioenergetic Intervention. PloS one 8, e79977.
Ding, F., Yao, J., Zhao, L., Mao, Z., Chen, S., and Brinton, R.D. (2013b). Ovariectomy induces a
shift in fuel availability and metabolism in the hippocampus of the female transgenic model of
familial Alzheimer's. PloS one 8, e59825.
Doody, R.S., Raman, R., Farlow, M., Iwatsubo, T., Vellas, B., Joffe, S., Kieburtz, K., He, F., Sun,
X., Thomas, R.G., et al. (2013a). A phase 3 trial of semagacestat for treatment of Alzheimer's
disease. The New England journal of medicine 369, 341-350.
Doody, R.S., Raman, R., Farlow, M., Iwatsubo, T., Vellas, B., Joffe, S., Kieburtz, K., He, F., Sun,
X., Thomas, R.G., et al. (2013b). A Phase 3 Trial of Semagacestat for Treatment of Alzheimer's
Disease. New England Journal of Medicine 369, 341-350.
Drzezga, A., Riemenschneider, M., Strassner, B., Grimmer, T., Peller, M., Knoll, A., Wagenpfeil,
S., Minoshima, S., Schwaiger, M., and Kurz, A. (2005). Cerebral glucose metabolism in patients
with AD and different APOE genotypes. Neurology 64, 102-107.
Duara, R., Lopez-Alberola, R.F., Barker, W.W., Loewenstein, D.A., Zatinsky, M., Eisdorfer, C.E.,
and Weinberg, G.B. (1993). A comparison of familial and sporadic Alzheimer's disease.
Neurology 43, 1377-1384.
Dubois, B., Feldman, H.H., Jacova, C., DeKosky, S.T., Barberger-Gateau, P., Cummings, J.,
Delacourte, A., Galasko, D., Gauthier, S., Jicha, G., et al. (2007). Research criteria for the
diagnosis of Alzheimer's disease: revising the NINCDS–ADRDA criteria. The Lancet Neurology
6, 734-746.
Duffy, R., Wiseman, H., and File, S.E. (2003). Improved cognitive function in postmenopausal
women after 12 weeks of consumption of a soya extract containing isoflavones. Pharmacology,
biochemistry, and behavior 75, 721-729.
Edland, S.D., Silverman, J.M., Peskind, E.R., Tsuang, D., Wijsman, E., and Morris, J.C. (1996).
Increased risk of dementia in mothers of Alzheimer's disease cases: evidence for maternal
inheritance. Neurology 47, 254-256.
Edland, S.D., Tobe, V.O., Rieder, M.J., Bowen, J.D., McCormick, W., Teri, L., Schellenberg, G.D.,
Larson, E.B., Nickerson, D.A., and Kukull, W.A. (2002). Mitochondrial genetic variants and
Alzheimer disease: a case-control study of the T4336C and G5460A variants. Alzheimer disease
and associated disorders 16, 1-7.
Edmond, J., Higa, T.A., Korsak, R.A., Bergner, E.A., and Lee, W.N. (1998). Fatty acid transport
and utilization for the developing brain. Journal of neurochemistry 70, 1227-1234.
Edmond, J., Robbins, R.A., Bergstrom, J.D., Cole, R.A., and de Vellis, J. (1987). Capacity for
substrate utilization in oxidative metabolism by neurons, astrocytes, and oligodendrocytes from
developing brain in primary culture. Journal of neuroscience research 18, 551-561.
Edwin, F., Wiepz, G.J., Singh, R., Peet, C.R., Chaturvedi, D., Bertics, P.J., and Patel, T.B. (2006).
A historical perspective of the EGF receptor and related systems. Methods in molecular biology
327, 1-24.
Egan, M.F., Kost, J., Tariot, P.N., Aisen, P.S., Cummings, J.L., Vellas, B., Sur, C., Mukai, Y.,
Voss, T., Furtek, C., et al. (2018). Randomized Trial of Verubecestat for Mild-to-Moderate
Alzheimer’s Disease. New England Journal of Medicine 378, 1691-1703.
El-Tanani, M.K., and Green, C.D. (1997). Two separate mechanisms for ligand-independent
activation of the estrogen receptor. Molecular endocrinology 11, 928-937.
147

Elson, J., Herrnstadt, C., Preston, G., Thal, L., Morris, C., Edwardson, J.A., Beal, M.F., Turnbull,
D., and Howell, N. (2006). Does the mitochondrial genome play a role in the etiology of
Alzheimer’s disease? Human genetics 119, 241-254.
Erlandsson, A., Brannvall, K., Gustafsdottir, S., Westermark, B., and Forsberg-Nilsson, K. (2006).
Autocrine/paracrine platelet-derived growth factor regulates proliferation of neural progenitor
cells. Cancer Res 66, 8042-8048.
Erlik, Y., Meldrum, D.R., and Judd, H.L. (1982). Estrogen levels in postmenopausal women with
hot flashes. Obstetrics and gynecology 59, 403-407.
Espeland, M.A., Brunner, R.L., Hogan, P.E., Rapp, S.R., Coker, L.H., Legault, C., Granek, I.,
Resnick, S.M., and for the Women's Health Initiative Study of Cognitive Aging Study, G. (2010).
Long-Term Effects of Conjugated Equine Estrogen Therapies on Domain-Specific Cognitive
Function: Results from the Women's Health Initiative Study of Cognitive Aging Extension.
Journal of the American Geriatrics Society 58, 1263-1271.
Espeland, M.A., Shumaker, S.A., Leng, I., and et al. (2013). LOng-term effects on cognitive
function of postmenopausal hormone therapy prescribed to women aged 50 to 55 years. JAMA
Internal Medicine 173, 1429-1436.
Fachal, L., Mosquera-Miguel, A., Pastor, P., Ortega-Cubero, S., Lorenzo, E., Oterino-Duran, A.,
Toriello, M., Quintans, B., Camina-Tato, M., Sesar, A., et al. (2015). No evidence of association
between common European mitochondrial DNA variants in Alzheimer, Parkinson, and migraine
in the Spanish population. American journal of medical genetics Part B, Neuropsychiatric
genetics : the official publication of the International Society of Psychiatric Genetics 168B, 54-65.
Farrer, L.A., Cupples, L.A., Haines, J.L., Hyman, B., Kukull, W.A., Mayeux, R., Myers, R.H.,
Pericak-Vance, M.A., Risch, N., and van Duijn, C.M. (1997). Effects of age, sex, and ethnicity on
the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis.
APOE and Alzheimer Disease Meta Analysis Consortium. Jama 278, 1349-1356.
Faulds, M.H., Pettersson, K., Gustafsson, J.A., and Haldosen, L.A. (2001). Cross-talk between
ERs and signal transducer and activator of transcription 5 is E2 dependent and involves two
functionally separate mechanisms. Molecular endocrinology 15, 1929-1940.
Fernandez-Moreno, M., Tamayo, M., Soto-Hermida, A., Mosquera, A., Oreiro, N., Fernandez-
Lopez, C., Fernandez, J.L., Rego-Perez, I., and Blanco, F.J. (2011). mtDNA haplogroup J
modulates telomere length and nitric oxide production. BMC musculoskeletal disorders 12, 283.
Fesahat, F., Houshmand, M., Panahi, M.S., Gharagozli, K., and Mirzajani, F. (2007). Do
haplogroups H and U act to increase the penetrance of Alzheimer's disease? Cellular and molecular
neurobiology 27, 329-334.
Filardo, E.J., Quinn, J.A., Bland, K.I., and Frackelton, A.R., Jr. (2000). Estrogen-induced
activation of Erk-1 and Erk-2 requires the G protein-coupled receptor homolog, GPR30, and
occurs via trans-activation of the epidermal growth factor receptor through release of HB-EGF.
Molecular endocrinology 14, 1649-1660.
Filardo, E.J., Quinn, J.A., Frackelton, A.R., Jr., and Bland, K.I. (2002). Estrogen action via the G
protein-coupled receptor, GPR30: stimulation of adenylyl cyclase and cAMP-mediated attenuation
of the epidermal growth factor receptor-to-MAPK signaling axis. Molecular endocrinology 16,
70-84.
File, S.E., Hartley, D.E., Elsabagh, S., Duffy, R., and Wiseman, H. (2005). Cognitive improvement
after 6 weeks of soy supplements in postmenopausal women is limited to frontal lobe function.
Menopause (New York, NY) 12, 193-201.
148

File, S.E., Jarrett, N., Fluck, E., Duffy, R., Casey, K., and Wiseman, H. (2001). Eating soya
improves human memory. Psychopharmacology 157, 430-436.
Flouriot, G., Brand, H., Denger, S., Metivier, R., Kos, M., Reid, G., Sonntag-Buck, V., and Gannon,
F. (2000). Identification of a new isoform of the human estrogen receptor-alpha (hER-alpha) that
is encoded by distinct transcripts and that is able to repress hER-alpha activation function 1. The
EMBO journal 19, 4688-4700.
Fu, X., Giavalisco, P., Liu, X., Catchpole, G., Fu, N., Ning, Z.B., Guo, S., Yan, Z., Somel, M.,
Paabo, S., et al. (2011). Rapid metabolic evolution in human prefrontal cortex. Proceedings of the
National Academy of Sciences of the United States of America 108, 6181-6186.
Gertz, J., Savic, D., Varley, K.E., Partridge, E.C., Safi, A., Jain, P., Cooper, G.M., Reddy, T.E.,
Crawford, G.E., and Myers, R.M. (2013). Distinct properties of cell type-specific and shared
transcription factor binding sites. Molecular cell 52, 10.1016/j.molcel.2013.1008.1037.
Gibson, G.E., Haroutunian, V., Zhang, H., Park, L.C., Shi, Q., Lesser, M., Mohs, R.C., Sheu, R.K.,
and Blass, J.P. (2000). Mitochondrial damage in Alzheimer's disease varies with apolipoprotein E
genotype. Ann Neurol 48, 297-303.
Giles, R.E., Blanc, H., Cann, H.M., and Wallace, D.C. (1980). Maternal inheritance of human
mitochondrial DNA. Proceedings of the National Academy of Sciences of the United States of
America 77, 6715-6719.
Girdler, S.S., Beth Mechlin, M., Light, K.C., and Leslie Morrow, A. (2006). Ethnic differences in
allopregnanolone concentrations in women during rest and following mental stress.
Psychophysiology 43, 331-336.
Gleason, C.E., Dowling, N.M., Wharton, W., Manson, J.E., Miller, V.M., Atwood, C.S., Brinton,
E.A., Cedars, M.I., Lobo, R.A., Merriam, G.R., et al. (2015). Effects of Hormone Therapy on
Cognition and Mood in Recently Postmenopausal Women: Findings from the Randomized,
Controlled KEEPS–Cognitive and Affective Study. PLoS Med 12, e1001833.
Gold, E.B., Sternfeld, B., Kelsey, J.L., Brown, C., Mouton, C., Reame, N., Salamone, L., and
Stellato, R. (2000). Relation of demographic and lifestyle factors to symptoms in a multi-
racial/ethnic population of women 40-55 years of age. American journal of epidemiology 152,
463-473.
Gomez-Duran, A., Pacheu-Grau, D., Lopez-Gallardo, E., Diez-Sanchez, C., Montoya, J., Lopez-
Perez, M.J., and Ruiz-Pesini, E. (2010). Unmasking the causes of multifactorial disorders:
OXPHOS differences between mitochondrial haplogroups. Human molecular genetics 19, 3343-
3353.
Gonzalez, M., Cabrera-Socorro, A., Perez-Garcia, C.G., Fraser, J.D., Lopez, F.J., Alonso, R., and
Meyer, G. (2007). Distribution patterns of estrogen receptor alpha and beta in the human cortex
and hippocampus during development and adulthood. The Journal of comparative neurology 503,
790-802.
Grady, D., Yaffe, K., Kristof, M., Lin, F., Richards, C., and Barrett-Connor, E. (2002). Effect of
postmenopausal hormone therapy on cognitive function: The Heart and Estrogen/progestin
Replacement Study. American Journal of Medicine 113, 543-548.
Green, R.C., Schneider, L.S., Amato, D.A., and et al. (2009). Effect of tarenflurbil on cognitive
decline and activities of daily living in patients with mild alzheimer disease: A randomized
controlled trial. JAMA 302, 2557-2564.
Greendale, G.A., Derby, C.A., and Maki, P.M. (2011). Perimenopause and Cognition. Obstetrics
and gynecology clinics of North America 38, 519-535.
149

Grimm, A., Mensah-Nyagan, A.G., and Eckert, A. (2016). Alzheimer, mitochondria and gender.
Neuroscience & Biobehavioral Reviews 67, 89-101.
Grimm, A., Schmitt, K., Lang, U.E., Mensah-Nyagan, A.G., and Eckert, A. (2014). Improvement
of neuronal bioenergetics by neurosteroids: Implications for age-related neurodegenerative
disorders. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 1842, 2427-2438.
Guo, J., Duckles, S.P., Weiss, J.H., Li, X., and Krause, D.N. (2012). 17beta-Estradiol prevents cell
death and mitochondrial dysfunction by an estrogen receptor-dependent mechanism in astrocytes
after oxygen-glucose deprivation/reperfusion. Free radical biology & medicine 52, 2151-2160.
Gurok, U., Steinhoff, C., Lipkowitz, B., Ropers, H.H., Scharff, C., and Nuber, U.A. (2004). Gene
Expression Changes in the Course of Neural Progenitor Cell Differentiation. The Journal of
Neuroscience 24, 5982.
Guzman, M., and Blazquez, C. (2004). Ketone body synthesis in the brain: possible
neuroprotective effects. Prostaglandins, leukotrienes, and essential fatty acids 70, 287-292.
Hardy, J., and Allsop, D. (1991). Amyloid deposition as the central event in the aetiology of
Alzheimer's disease. Trends in pharmacological sciences 12, 383-388.
Harold Goodglass, E.K. (1972). The Assessment of Aphasia and Related Disorders (Philadelphia:
Le & Febiger).
Harrington, W.R., Sheng, S., Barnett, D.H., Petz, L.N., Katzenellenbogen, J.A., and
Katzenellenbogen, B.S. (2003). Activities of estrogen receptor alpha- and beta-selective ligands at
diverse estrogen responsive gene sites mediating transactivation or transrepression. Molecular and
Cellular Endocrinology 206, 13-22.
Harris, J.J., Jolivet, R., and Attwell, D. (2012). Synaptic energy use and supply. Neuron 75, 762-
777.
Hazell, G.G.J., Yao, S.T., Roper, J.A., Prossnitz, E.R., O'Carroll, A.-M., and Lolait, S.J. (2009).
Localisation of GPR30, a novel G protein-coupled oestrogen receptor, suggests multiple functions
in rodent brain and peripheral tissues. The Journal of Endocrinology 202, 223-236.
Henderson, V.W., Paganini-Hill, A., Miller, B.L., Elble, R.J., Reyes, P.F., Shoupe, D., McCleary,
C.A., Klein, R.A., Hake, A.M., and Farlow, M.R. (2000). Estrogen for Alzheimer's disease in
women: randomized, double-blind, placebo-controlled trial. Neurology 54, 295-301.
Hewitt, S.C., and Korach, K.S. (2002). Estrogen receptors: structure, mechanisms and function.
Reviews in endocrine & metabolic disorders 3, 193-200.
Hirai, K., Aliev, G., Nunomura, A., Fujioka, H., Russell, R.L., Atwood, C.S., Johnson, A.B., Kress,
Y., Vinters, H.V., Tabaton, M., et al. (2001). Mitochondrial abnormalities in Alzheimer's disease.
The Journal of neuroscience : the official journal of the Society for Neuroscience 21, 3017-3023.
Hou, Y., Li, W., Sheng, Y., Li, L., Huang, Y., Zhang, Z., Zhu, T., Peace, D., Quigley, J.G., Wu,
W., et al. (2015). Foxm1 is essential for quiescence and maintenance of hematopoietic stem cells.
Nature immunology 16, 810-818.
Hoyer, S. (1982). The young-adult and normally aged brain. Its blood flow and oxidative
metabolism. A review--part I. Archives of gerontology and geriatrics 1, 101-116.
Hutchin, T., and Cortopassi, G. (1995). A mitochondrial DNA clone is associated with increased
risk for Alzheimer disease. Proceedings of the National Academy of Sciences of the United States
of America 92, 6892-6895.
Hyder, F., and Rothman, D.L. (2012). Quantitative fMRI and oxidative neuroenergetics.
NeuroImage 62, 985-994.
150

Hyder, F., Rothman, D.L., and Bennett, M.R. (2013). Cortical energy demands of signaling and
nonsignaling components in brain are conserved across mammalian species and activity levels.
Proceedings of the National Academy of Sciences of the United States of America 110, 3549-3554.
Hyder, S.M., Chiappetta, C., and Stancel, G.M. (1999). Interaction of human estrogen receptors α
and β with the same naturally occurring estrogen response elements. Biochemical Pharmacology
57, 597-601.
Ignar-Trowbridge, D.M., Nelson, K.G., Bidwell, M.C., Curtis, S.W., Washburn, T.F., McLachlan,
J.A., and Korach, K.S. (1992). Coupling of dual signaling pathways: epidermal growth factor
action involves the estrogen receptor. Proceedings of the National Academy of Sciences 89, 4658-
4662.
Iqbal, K., del C. Alonso, A., Chen, S., Chohan, M.O., El-Akkad, E., Gong, C.-X., Khatoon, S., Li,
B., Liu, F., Rahman, A., et al. (2005). Tau pathology in Alzheimer disease and other tauopathies.
Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 1739, 198-210.
Irwin, R.W., and Brinton, R.D. (2014). Allopregnanolone as regenerative therapeutic for
Alzheimer's disease: translational development and clinical promise. Progress in neurobiology 113,
40-55.
Irwin, R.W., Solinsky, C.M., and Brinton, R.D. (2014). Frontiers in therapeutic development of
allopregnanolone for Alzheimer’s disease and other neurological disorders. Frontiers in Cellular
Neuroscience 8.
Irwin, R.W., Solinsky, C.M., Loya, C.M., Salituro, F.G., Rodgers, K.E., Bauer, G., Rogawski,
M.A., and Brinton, R.D. (2015). Allopregnanolone Preclinical Acute Pharmacokinetic and
Pharmacodynamic Studies to Predict Tolerability and Efficacy for Alzheimer’s Disease. PloS one
10, e0128313.
Irwin, R.W., Wang, J.M., Chen, S., and Brinton, R.D. (2011a). Neuroregenerative mechanisms of
allopregnanolone in Alzheimer's disease. Frontiers in endocrinology 2, 117.
Irwin, R.W., Yao, J., Ahmed, S.S., Hamilton, R.T., Cadenas, E., and Brinton, R.D. (2011b).
Medroxyprogesterone acetate antagonizes estrogen up-regulation of brain mitochondrial function.
Endocrinology 152, 556-567.
Irwin, R.W., Yao, J., Hamilton, R.T., Cadenas, E., Brinton, R.D., and Nilsen, J. (2008).
Progesterone and estrogen regulate oxidative metabolism in brain mitochondria. Endocrinology
149, 3167-3175.
Irwin, R.W., Yao, J., To, J., Hamilton, R.T., Cadenas, E., and Brinton, R.D. (2012). Selective
oestrogen receptor modulators differentially potentiate brain mitochondrial function. J
Neuroendocrinol 24, 236-248.
Ishii, K., Sasaki, M., Kitagaki, H., Yamaji, S., Sakamoto, S., Matsuda, K., and Mori, E. (1997).
Reduction of cerebellar glucose metabolism in advanced Alzheimer's disease. Journal of nuclear
medicine : official publication, Society of Nuclear Medicine 38, 925-928.
Ivanisevic, J., Stauch, K.L., Petrascheck, M., Benton, H.P., Epstein, A.A., Fang, M., Gorantla, S.,
Tran, M., Hoang, L., Kurczy, M.E., et al. (2016). Metabolic drift in the aging brain. Aging 8, 1000-
1013.
Jacobs, A., Wegewitz, U., Sommerfeld, C., Grossklaus, R., and Lampen, A. (2009). Efficacy of
isoflavones in relieving vasomotor menopausal symptoms - A systematic review. Molecular
nutrition & food research 53, 1084-1097.
Jagannathan, R., Baseler, W.A., Thapa, D., Croston, T.L., Shepherd, D.L., Nichols, C.E., and
Hollander, J.M. (2012). HDAC6 regulates mitochondrial oxidative phosphorylation by ATP
151

synthase beta subunit acetylation in diabetic cardiomyopathy. The FASEB Journal 26, 869.813-
869.813.
Jia, M., Dahlman-Wright, K., and Gustafsson, J.A. (2015). Estrogen receptor alpha and beta in
health and disease. Best practice & research Clinical endocrinology & metabolism 29, 557-568.
Jones, T.T., and Brewer, G.J. (2010). Age-related deficiencies in complex I endogenous substrate
availability and reserve capacity of complex IV in cortical neuron electron transport. Biochimica
et Biophysica Acta (BBA) - Bioenergetics 1797, 167-176.
Kandasamy, M., Lehner, B., Kraus, S., Sander, P.R., Marschallinger, J., Rivera, F.J., Trümbach,
D., Ueberham, U., Reitsamer, H.A., Strauss, O., et al. (2014). TGF-beta signalling in the adult
neurogenic niche promotes stem cell quiescence as well as generation of new neurons. Journal of
Cellular and Molecular Medicine 18, 1444-1459.
Kandasamy, M., Reilmann, R., Winkler, J., Bogdahn, U., and Aigner, L. (2011). Transforming
Growth Factor-Beta Signaling in the Neural Stem Cell Niche: A Therapeutic Target for
Huntington's Disease. Neurology Research International 2011.
Kato, S., Endoh, H., Masuhiro, Y., Kitamoto, T., Uchiyama, S., Sasaki, H., Masushige, S., Gotoh,
Y., Nishida, E., Kawashima, H., et al. (1995). Activation of the estrogen receptor through
phosphorylation by mitogen-activated protein kinase. Science 270, 1491-1494.
Kawas, C., Resnick, S., Morrison, A., Brookmeyer, R., Corrada, M., Zonderman, A., Bacal, C.,
Lingle, D.D., and Metter, E. (1997). A prospective study of estrogen replacement therapy and the
risk of developing Alzheimer's disease: the Baltimore Longitudinal Study of Aging. Neurology 48,
1517-1521.
Kelly, R.D., Rodda, A.E., Dickinson, A., Mahmud, A., Nefzger, C.M., Lee, W., Forsythe, J.S.,
Polo, J.M., Trounce, I.A., McKenzie, M., et al. (2013). Mitochondrial DNA haplotypes define
gene expression patterns in pluripotent and differentiating embryonic stem cells. Stem Cells 31,
703-716.
Kenney, M.C., Chwa, M., Atilano, S.R., Pavlis, J.M., Falatoonzadeh, P., Ramirez, C., Malik, D.,
Hsu, T., Woo, G., Soe, K., et al. (2013). Mitochondrial DNA variants mediate energy production
and expression levels for CFH, C3 and EFEMP1 genes: implications for age-related macular
degeneration. PloS one 8, e54339.
Kety, S.S. (1957). THE GENERAL METABOLISM OF THE BRAIN IN VIVO. In Metabolism
of the Nervous System, D. Richter, ed. (Pergamon), pp. 221-237.
Khusnutdinova, E., Gilyazova, I., Ruiz-Pesini, E., Derbeneva, O., Khusainova, R., Khidiyatova,
I., Magzhanov, R., and Wallace, D.C. (2008). A mitochondrial etiology of neurodegenerative
diseases: evidence from Parkinson's disease. Annals of the New York Academy of Sciences 1147,
1-20.
Kim, D.Y., Hwang, I., Muller, F.L., and Paik, J.H. (2015). Functional regulation of FoxO1 in
neural stem cell differentiation. Cell death and differentiation 22, 2034-2045.
Kim, S.A., and Jung, H. (2015). Prevention of Cognitive Impairment in the Midlife Women.
Journal of Menopausal Medicine 21, 19-23.
Kim, S.H., Vlkolinsky, R., Cairns, N., and Lubec, G. (2000). Decreased levels of complex III core
protein 1 and complex V beta chain in brains from patients with Alzheimer's disease and Down
syndrome. Cellular and molecular life sciences : CMLS 57, 1810-1816.
Kish, S.J., Mastrogiacomo, F., Guttman, M., Furukawa, Y., Taanman, J.W., Dozic, S., Pandolfo,
M., Lamarche, J., DiStefano, L., and Chang, L.J. (1999). Decreased brain protein levels of
cytochrome oxidase subunits in Alzheimer's disease and in hereditary spinocerebellar ataxia
disorders: a nonspecific change? Journal of neurochemistry 72, 700-707.
152

Klosinski, L.P., Yao, J., Yin, F., Fonteh, A.N., Harrington, M.G., Christensen, T.A., Trushina, E.,
and Brinton, R.D. (2015). White Matter Lipids as a Ketogenic Fuel Supply in Aging Female Brain:
Implications for Alzheimer's Disease. EBioMedicine 2, 1888-1904.
Kloss-Brandstatter, A., Pacher, D., Schonherr, S., Weissensteiner, H., Binna, R., Specht, G., and
Kronenberg, F. (2011). HaploGrep: a fast and reliable algorithm for automatic classification of
mitochondrial DNA haplogroups. Hum Mutat 32, 25-32.
Korecka, J.A., van Kesteren, R.E., Blaas, E., Spitzer, S.O., Kamstra, J.H., Smit, A.B., Swaab, D.F.,
Verhaagen, J., and Bossers, K. (2013). Phenotypic Characterization of Retinoic Acid
Differentiated SH-SY5Y Cells by Transcriptional Profiling. PloS one 8, e63862.
Kostanyan, A., and Nazaryan, K. (1992). Rat brain glycolysis regulation by estradiol-17 beta.
Biochimica et biophysica acta 1133, 301-306.
Kovalevich, J., and Langford, D. (2013a). Considerations for the Use of SH-SY5Y Neuroblastoma
Cells in Neurobiology. Methods in molecular biology (Clifton, NJ) 1078, 9-21.
Kovalevich, J., and Langford, D. (2013b). Considerations for the Use of SH-SY5Y Neuroblastoma
Cells in Neurobiology. Methods in molecular biology 1078, 9-21.
Kritz-Silverstein, D., Von Muhlen, D., Barrett-Connor, E., and Bressel, M.A. (2003). Isoflavones
and cognitive function in older women: the SOy and Postmenopausal Health In Aging (SOPHIA)
Study. Menopause (New York, NY) 10, 196-202.
Kushner, P.J., Agard, D.A., Greene, G.L., Scanlan, T.S., Shiau, A.K., Uht, R.M., and Webb, P.
(2000). Estrogen receptor pathways to AP-1. The Journal of Steroid Biochemistry and Molecular
Biology 74, 311-317.
Lacor, P.N., Buniel, M.C., Furlow, P.W., Sanz Clemente, A., Velasco, P.T., Wood, M., Viola,
K.L., and Klein, W.L. (2007). Aβ Oligomer-Induced Aberrations in Synapse Composition, Shape,
and Density Provide a Molecular Basis for Loss of Connectivity in Alzheimer's Disease.
The Journal of Neuroscience 27, 796.
LaCroix, A.Z., Chlebowski, R.T., Manson, J.E., Aragaki, A.K., Johnson, K.C., Martin, L.,
Margolis, K.L., Stefanick, M.L., Brzyski, R., Curb, J.D., et al. (2011). Health outcomes after
stopping conjugated equine estrogens among postmenopausal women with prior hysterectomy: a
randomized controlled trial. Jama 305, 1305-1314.
Lakatos, A., Derbeneva, O., Younes, D., Keator, D., Bakken, T., Lvova, M., Brandon, M., Guffanti,
G., Reglodi, D., Saykin, A., et al. (2010). Association between mitochondrial DNA variations and
Alzheimer's disease in the ADNI cohort. Neurobiology of aging 31, 1355-1363.
Larsen, S., Diez-Sanchez, C., Rabol, R., Ara, I., Dela, F., and Helge, J.W. (2014). Increased
intrinsic mitochondrial function in humans with mitochondrial haplogroup H. Biochimica et
biophysica acta 1837, 226-231.
Latorre-Pellicer, A., Moreno-Loshuertos, R., Lechuga-Vieco, A.V., Sánchez-Cabo, F., Torroja, C.,
Acín-Pérez, R., Calvo, E., Aix, E., González-Guerra, A., Logan, A., et al. (2016). Mitochondrial
and nuclear DNA matching shapes metabolism and healthy ageing. Nature advance online
publication.
Lee, H., Han, S., Kwon, C.S., and Lee, D. (2016). Biogenesis and regulation of the let-7 miRNAs
and their functional implications. Protein & Cell 7, 100-113.
Lee, H.-R., Kim, T.-H., and Choi, K.-C. (2012). Functions and physiological roles of two types of
estrogen receptors, ERα and ERβ, identified by estrogen receptor knockout mouse. Laboratory
Animal Research 28, 71-76.
Lee, J., Sharma, S., Kim, J., Ferrante, R.J., and Ryu, H. (2008). Mitochondrial Nuclear Receptors
and Transcription Factors: Who’s Minding the Cell? Journal of neuroscience research 86, 961-971.
153

Li, J., Ma, J., Meng, G., Lin, H., Wu, S., Wang, J., Luo, J., Xu, X., Tough, D., Lindon, M., et al.
(2016). BET bromodomain inhibition promotes neurogenesis while inhibiting gliogenesis in
neural progenitor cells. Stem cell research 17, 212-221.
Lin, A.L., and Rothman, D.L. (2014). What have novel imaging techniques revealed about
metabolism in the aging brain? Future neurology 9, 341-354.
Lin, A.L., Zhang, W., Gao, X., and Watts, L. (2015). Caloric restriction increases ketone bodies
metabolism and preserves blood flow in aging brain. Neurobiology of aging 36, 2296-2303.
Lin, M.T., and Beal, M.F. (2006). Mitochondrial dysfunction and oxidative stress in
neurodegenerative diseases. Nature 443, 787-795.
Lin, T.K., Lin, H.Y., Chen, S.D., Chuang, Y.C., Chuang, J.H., Wang, P.W., Huang, S.T., Tiao,
M.M., Chen, J.B., and Liou, C.W. (2012). The creation of cybrids harboring mitochondrial
haplogroups in the Taiwanese population of ethnic Chinese background: an extensive in vitro tool
for the study of mitochondrial genomic variations. Oxidative medicine and cellular longevity 2012,
824275.
Lindberg, M.K., Movérare, S., Skrtic, S., Gao, H., Dahlman-Wright, K., Gustafsson, J.-Å., and
Ohlsson, C. (2003). Estrogen Receptor (ER)-β Reduces ERα-Regulated Gene Transcription,
Supporting a “Ying Yang” Relationship between ERα and ERβ in Mice. Molecular endocrinology
17, 203-208.
Liu, M.M., Albanese, C., Anderson, C.M., Hilty, K., Webb, P., Uht, R.M., Price, R.H., Jr., Pestell,
R.G., and Kushner, P.J. (2002). Opposing action of estrogen receptors alpha and beta on cyclin D1
gene expression. The Journal of biological chemistry 277, 24353-24360.
Liu, Z., Chen, H.H., Li, T.L., Xu, L., and Du, H.Q. (2013). A cross-sectional study on cerebrospinal
fluid biomarker levels in cognitively normal elderly subjects with or without a family history of
Alzheimer's disease. CNS neuroscience & therapeutics 19, 38-42.
López-González, R., Camacho-Arroyo, I., and Velasco, I. (2011). Progesterone and 17β-estradiol
increase differentiation of mouse embryonic stem cells to motor neurons. IUBMB Life 63, 930-
939.
Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and dispersion
for RNA-seq data with DESeq2. Genome Biol 15, 550.
Ma, Y., Guo, H., Zhang, L., Tao, L., Yin, A., Liu, Z., Li, Y., Dong, H., Xiong, L., and Hou, W.
(2016). Estrogen replacement therapy-induced neuroprotection against brain ischemia-reperfusion
injury involves the activation of astrocytes via estrogen receptor β. Scientific Reports 6, 21467.
Ma, Z.Q., Santagati, S., Patrone, C., Pollio, G., Vegeto, E., and Maggi, A. (1994). Insulin-like
growth factors activate estrogen receptor to control the growth and differentiation of the human
neuroblastoma cell line SK-ER3. Molecular endocrinology 8, 910-918.
Maggiolini, M., Vivacqua, A., Fasanella, G., Recchia, A.G., Sisci, D., Pezzi, V., Montanaro, D.,
Musti, A.M., Picard, D., and Ando, S. (2004). The G protein-coupled receptor GPR30 mediates c-
fos up-regulation by 17beta-estradiol and phytoestrogens in breast cancer cells. The Journal of
biological chemistry 279, 27008-27016.
Magistretti, P.J., and Allaman, I. (2015). A cellular perspective on brain energy metabolism and
functional imaging. Neuron 86, 883-901.
Mahley, R.W., Huang Y Fau - Weisgraber, K.H., and Weisgraber, K.H. Detrimental effects of
apolipoprotein E4: potential therapeutic targets in Alzheimer's disease.
Maki, P.M. (2006). Hormone therapy and cognitive function: Is there a critical period for benefit?
Neuroscience 138, 1027-1030.
154

Maki, P.M., and Resnick, S.M. (2000). Longitudinal effects of estrogen replacement therapy on
PET cerebral blood flow and cognition. Neurobiology of aging 21, 373-383.
Maki, P.M., Zonderman, A.B., and Resnick, S.M. (2001). Enhanced Verbal Memory in
Nondemented Elderly Women Receiving Hormone-Replacement Therapy. American Journal of
Psychiatry 158, 227-233.
Maklashina, E., and Ackrell, B.A.C. (2003). Is defective electron transport at the hub of aging?
Aging cell 3, 21-27.
Mancuso, M., Filosto, M., Bosetti, F., Ceravolo, R., Rocchi, A., Tognoni, G., Manca, M.L., Solaini,
G., Siciliano, G., and Murri, L. (2003). Decreased platelet cytochrome c oxidase activity is
accompanied by increased blood lactate concentration during exercise in patients with Alzheimer
disease. Experimental neurology 182, 421-426.
Mancuso, M., Nardini, M., Micheli, D., Rocchi, A., Nesti, C., Giglioli, N.J., Petrozzi, L., Rossi,
C., Ceravolo, R., Bacci, A., et al. (2007). Lack of association between mtDNA haplogroups and
Alzheimer's disease in Tuscany. Neurological sciences : official journal of the Italian Neurological
Society and of the Italian Society of Clinical Neurophysiology 28, 142-147.
Mandal, P.K., Tripathi, M., and Sugunan, S. (2012). Brain oxidative stress: Detection and mapping
of anti-oxidant marker ‘Glutathione’ in different brain regions of healthy male/female, MCI and
Alzheimer patients using non-invasive magnetic resonance spectroscopy. Biochemical and
biophysical research communications 417, 43-48.
Marcuello, A., Martinez-Redondo, D., Dahmani, Y., Casajus, J.A., Ruiz-Pesini, E., Montoya, J.,
Lopez-Perez, M.J., and Diez-Sanchez, C. (2009). Human mitochondrial variants influence on
oxygen consumption. Mitochondrion 9, 27-30.
Martinez-Redondo, D., Marcuello, A., Casajus, J.A., Ara, I., Dahmani, Y., Montoya, J., Ruiz-
Pesini, E., Lopez-Perez, M.J., and Diez-Sanchez, C. (2010). Human mitochondrial haplogroup H:
the highest VO2max consumer--is it a paradox? Mitochondrion 10, 102-107.
Maruszak, A., Canter, J.A., Styczynska, M., Zekanowski, C., and Barcikowska, M. (2009).
Mitochondrial haplogroup H and Alzheimer's disease--is there a connection? Neurobiology of
aging 30, 1749-1755.
Maruszak, A., Safranow, K., Branicki, W., Gaweda-Walerych, K., Pospiech, E., Gabryelewicz, T.,
Canter, J.A., Barcikowska, M., and Zekanowski, C. (2011). The impact of mitochondrial and
nuclear DNA variants on late-onset Alzheimer's disease risk. Journal of Alzheimer's disease : JAD
27, 197-210.
Maruyama, K., Endoh, H., Sasaki-Iwaoka, H., Kanou, H., Shimaya, E., Hashimoto, S., Kato, S.,
and Kawashima, H. (1998). A Novel Isoform of Rat Estrogen Receptor Beta with 18 Amino Acid
Insertion in the Ligand Binding Domain as a Putative Dominant Negative Regulator of Estrogen
Action. Biochemical and biophysical research communications 246, 142-147.
Mastroeni, D. (2017). Nuclear but not mitochondrial-encoded oxidative phosphorylation genes are
altered in aging, mild cognitive impairment, and Alzheimer’s disease. 13, 510-519.
Maurer, I., Zierz, S., and Moller, H.J. (2000). A selective defect of cytochrome c oxidase is present
in brain of Alzheimer disease patients. Neurobiology of aging 21, 455-462.
Meier-Ruge, W., Iwangoff, P., Reichlmeier, K., and Sandoz, P. (1980). Neurochemical findings
in the aging brain. Advances in biochemical psychopharmacology 23, 323-338.
Mercer, T.R., Neph, S., Dinger, M.E., Crawford, J., Smith, M.A., Shearwood, A.M., Haugen, E.,
Bracken, C.P., Rackham, O., Stamatoyannopoulos, J.A., et al. (2011). The human mitochondrial
transcriptome. Cell 146, 645-658.
155

Mielke, M.M., Vemuri, P., and Rocca, W.A. (2014). Clinical epidemiology of Alzheimer's disease:
assessing sex and gender differences. Clinical epidemiology 6, 37-48.
Milner, T.A., Ayoola, K., Drake, C.T., Herrick, S.P., Tabori, N.E., McEwen, B.S., Warrier, S., and
Alves, S.E. (2005). Ultrastructural localization of estrogen receptor beta immunoreactivity in the
rat hippocampal formation. The Journal of comparative neurology 491, 81-95.
Mitra, S.W., Hoskin, E., Yudkovitz, J., Pear, L., Wilkinson, H.A., Hayashi, S., Pfaff, D.W., Ogawa,
S., Rohrer, S.P., Schaeffer, J.M., et al. (2003). Immunolocalization of Estrogen Receptor β in the
Mouse Brain: Comparison with Estrogen Receptor α. Endocrinology 144, 2055-2067.
Miyata, T., Hayashida, H., Kikuno, R., Hasegawa, M., Kobayashi, M., and Koike, K. (1982).
Molecular clock of silent substitution: at least six-fold preponderance of silent changes in
mitochondrial genes over those in nuclear genes. Journal of molecular evolution 19, 28-35.
Morris, A.A. (2005). Cerebral ketone body metabolism. Journal of inherited metabolic disease 28,
109-121.
Morris, J.K., Honea, R.A., Vidoni, E.D., Swerdlow, R.H., and Burns, J.M. (2014). Is Alzheimer's
disease a systemic disease? Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease
1842, 1340-1349.
Mosconi, L. (2017). Perimenopause and emergence of an Alzheimer’s bioenergetic phenotype in
brain and periphery. 12.
Mosconi, L., Berti, V., Guyara-Quinn, C., McHugh, P., Petrongolo, G., Osorio, R.S., Connaughty,
C., Pupi, A., Vallabhajosula, S., Isaacson, R.S., et al. (2017). Perimenopause and emergence of an
Alzheimer’s bioenergetic phenotype in brain and periphery. PloS one 12, e0185926.
Mosconi, L., de Leon, M., Murray, J., E, L., Lu, J., Javier, E., McHugh, P., and Swerdlow, R.H.
(2011). Reduced mitochondria cytochrome oxidase activity in adult children of mothers with
Alzheimer's disease. Journal of Alzheimer's disease : JAD 27, 483-490.
Mosconi, L., De Santi, S., Brys, M., Tsui, W.H., Pirraglia, E., Glodzik-Sobanska, L., Rich, K.E.,
Switalski, R., Mehta, P.D., Pratico, D., et al. (2008a). Hypometabolism and altered cerebrospinal
fluid markers in normal apolipoprotein E E4 carriers with subjective memory complaints.
Biological psychiatry 63, 609-618.
Mosconi, L., De Santi, S., Li, J., Tsui, W.H., Li, Y., Boppana, M., Laska, E., Rusinek, H., and de
Leon, M.J. (2008b). Hippocampal hypometabolism predicts cognitive decline from normal aging.
Neurobiology of aging 29, 676-692.
Mosconi, L., Herholz, K., Prohovnik, I., Nacmias, B., De Cristofaro, M.T., Fayyaz, M., Bracco,
L., Sorbi, S., and Pupi, A. (2005). Metabolic interaction between ApoE genotype and onset age in
Alzheimer's disease: implications for brain reserve. Journal of neurology, neurosurgery, and
psychiatry 76, 15-23.
Mosconi, L., Mistur, R., Switalski, R., Brys, M., Glodzik, L., Rich, K., Pirraglia, E., Tsui, W., De
Santi, S., and de Leon, M.J. (2009). Declining brain glucose metabolism in normal individuals
with a maternal history of Alzheimer disease. Neurology 72, 513-520.
Mosconi, L., Nacmias, B., Sorbi, S., De Cristofaro, M.T., Fayazz, M., Tedde, A., Bracco, L.,
Herholz, K., and Pupi, A. (2004a). Brain metabolic decreases related to the dose of the ApoE e4
allele in Alzheimer's disease. Journal of neurology, neurosurgery, and psychiatry 75, 370-376.
Mosconi, L., Perani, D., Sorbi, S., Herholz, K., Nacmias, B., Holthoff, V., Salmon, E., Baron, J.C.,
De Cristofaro, M.T., Padovani, A., et al. (2004b). MCI conversion to dementia and the APOE
genotype: a prediction study with FDG-PET. Neurology 63, 2332-2340.
156

Mosconi, L., Sorbi, S., Nacmias, B., De Cristofaro, M.T., Fayyaz, M., Bracco, L., Herholz, K., and
Pupi, A. (2004c). Age and ApoE genotype interaction in Alzheimer's disease: an FDG-PET study.
Psychiatry research 130, 141-151.
Mudher, A., and Lovestone, S. (2002). Alzheimer's disease – do tauists and baptists finally shake
hands? Trends in neurosciences 25, 22-26.
Mulnard, R.A., Cotman, C.W., Kawas, C., van Dyck, C.H., Sano, M., Doody, R., Koss, E., Pfeiffer,
E., Jin, S., Gamst, A., et al. (2000). Estrogen replacement therapy for treatment of mild to moderate
Alzheimer disease: a randomized controlled trial. Alzheimer's Disease Cooperative Study. JAMA
283, 1007-1015.
Navarro, A., and Boveris, A. (2007). The mitochondrial energy transduction system and the aging
process. American Journal of Physiology-Cell Physiology 292, C670-C686.
Naylor, J.C., Kilts, J.D., Hulette, C.M., Steffens, D.C., Blazer, D.G., Ervin, J.F., Strauss, J.L.,
Allen, T.B., Massing, M.W., Payne, V.M., et al. (2010). Allopregnanolone levels are reduced in
temporal cortex in patients with Alzheimer's disease compared to cognitively intact control
subjects. Biochimica et biophysica acta 1801, 951-959.
Nilsen, J., and Diaz Brinton, R. (2003). Mechanism of estrogen-mediated neuroprotection:
regulation of mitochondrial calcium and Bcl-2 expression. Proceedings of the National Academy
of Sciences of the United States of America 100, 2842-2847.
Nilsen, J., Irwin, R.W., Gallaher, T.K., and Brinton, R.D. (2007). Estradiol in vivo regulation of
brain mitochondrial proteome. The Journal of neuroscience : the official journal of the Society for
Neuroscience 27, 14069-14077.
Okita, K., Yamakawa, T., Matsumura, Y., Sato, Y., Amano, N., Watanabe, A., Goshima, N., and
Yamanaka, S. (2013). An efficient nonviral method to generate integration-free human-induced
pluripotent stem cells from cord blood and peripheral blood cells. Stem Cells 31, 458-466.
Olesen, K.M., Jessen, H.M., Auger, C.J., and Auger, A.P. (2005). Dopaminergic activation of
estrogen receptors in neonatal brain alters progestin receptor expression and juvenile social play
behavior. Endocrinology 146, 3705-3712.
Paech, K., Webb, P., Kuiper, G.G.J.M., Nilsson, S., Gustafsson, J.-Å., Kushner, P.J., and Scanlan,
T.S. (1997). Differential Ligand Activation of Estrogen Receptors ERα and ERβ at AP1 Sites.
Science 277, 1508.
Paganini-Hill, A., and Henderson, V.W. (1994). Estrogen deficiency and risk of Alzheimer's
disease in women. American journal of epidemiology 140, 256-261.
Paganini-Hill, A., and Henderson, V.W. (1996). Estrogen replacement therapy and risk of
Alzheimer disease. Archives of internal medicine 156, 2213-2217.
Paik, J., Ding, Z., Narurkar, R., Ramkissoon, S., Muller, F., Kamoun, W.S., Chae, S.S., Zheng, H.,
Ying, H., Mahoney, J., et al. (2009). FoxOs cooperatively regulate diverse pathways governing
neural stem cell homeostasis. Cell Stem Cell 5, 540-553.
Paillusson, S., Stoica, R., Gomez-Suaga, P., Lau, D.H.W., Mueller, S., Miller, T., and Miller, C.C.J.
(2013). There's Something Wrong with my MAM; the ER-Mitochondria Axis and
Neurodegenerative Diseases. Trends in neurosciences 39, 146-157.
Parker, W.D., Filley, C.M., and Parks, J.K. (1990). Cytochrome oxidase deficiency in Alzheimer's
disease. Neurology 40, 1302.
Parker, W.D., Jr., Parks, J., Filley, C.M., and Kleinschmidt-DeMasters, B.K. (1994). Electron
transport chain defects in Alzheimer's disease brain. Neurology 44, 1090-1096.
Patro, R., Duggal, G., Love, M.I., Irizarry, R.A., and Kingsford, C. (2017). Salmon provides fast
and bias-aware quantification of transcript expression. Nat Methods 14, 417-419.
157

Paul, S.A., Serge, G., Bruno, V., Richard, B., Daniel, S., Julie, L., and Denis, G. (2007). Alzhemed:
A Potential Treatment for Alzheimers Disease. Current Alzheimer research 4, 473-478.
Pefanco, M.A., Kenny, A.M., Kaplan, R.F., Kuchel, G., Walsh, S., Kleppinger, A., and Prestwood,
K. (2007). The Effect of 3-Year Treatment with 0.25 mg/day of Micronized 17β-Estradiol on
Cognitive Function in Older Postmenopausal Women. Journal of the American Geriatrics Society
55, 426-431.
Pellerin, L. (2010). Food for thought: the importance of glucose and other energy substrates for
sustaining brain function under varying levels of activity. Diabetes & metabolism 36 Suppl 3, S59-
63.
Peng, B., Lu, B., Leygue, E., and Murphy, L.C. (2003). Putative functional characteristics of
human estrogen receptor-beta isoforms. Journal of molecular endocrinology 30, 13-29.
Perez-Martin, M., Salazar, V., Castillo, C., Ariznavarreta, C., Azcoitia, I., Garcia-Segura, L.M.,
and Tresguerres, J.A. (2005). Estradiol and soy extract increase the production of new cells in the
dentate gyrus of old rats. Experimental gerontology 40, 450-453.
Petersen, D.N., Tkalcevic, G.T., Koza-Taylor, P.H., Turi, T.G., and Brown, T.A. (1998).
Identification of estrogen receptor beta2, a functional variant of estrogen receptor beta expressed
in normal rat tissues. Endocrinology 139, 1082-1092.
Pettersson, K., Grandien, K., Kuiper, G.G., and Gustafsson, J.A. (1997). Mouse estrogen receptor
beta forms estrogen response element-binding heterodimers with estrogen receptor alpha.
Molecular endocrinology 11, 1486-1496.
Phillips, S.M., and Sherwin, B.B. (1992). Effects of estrogen on memory function in surgically
menopausal women. Psychoneuroendocrinology 17, 485-495.
Pichaud, N., Ballard, J.W., Tanguay, R.M., and Blier, P.U. (2012). Naturally occurring
mitochondrial DNA haplotypes exhibit metabolic differences: insight into functional properties of
mitochondria. Evolution; international journal of organic evolution 66, 3189-3197.
Poirier, J., Bertrand, P., Kogan, S., Gauthier, S., Davignon, J., and Bouthillier, D. (1993).
Apolipoprotein E polymorphism and Alzheimer's disease. The Lancet 342, 697-699.
Popot, J.L., and Vitry, C. (1990). On the Microassembly of Integral Membrane Proteins. Annual
Review of Biophysics and Biophysical Chemistry 19, 369-403.
Porter, W., Saville, B., Hoivik, D., and Safe, S. (1997). Functional Synergy between the
Transcription Factor Sp1 and the Estrogen Receptor. Molecular endocrinology 11, 1569-1580.
Prenzel, N., Zwick, E., Daub, H., Leserer, M., Abraham, R., Wallasch, C., and Ullrich, A. (1999).
EGF receptor transactivation by G-protein-coupled receptors requires metalloproteinase cleavage
of proHB-EGF. Nature 402, 884-888.
Price Jr, R.H., Lorenzon, N., and Handa, R.J. (2000). Differential expression of estrogen receptor
beta splice variants in rat brain: identification and characterization of a novel variant missing exon
4. Molecular Brain Research 80, 260-268.
Qin, C., Singh, P., and Safe, S. (1999). Transcriptional Activation of Insulin-Like Growth Factor-
Binding Protein-4 by 17β-Estradiol in MCF-7 Cells: Role of Estrogen Receptor-Sp1 Complexes.
Endocrinology 140, 2501-2508.
Qiu, J., Bosch, M.A., Tobias, S.C., Grandy, D.K., Scanlan, T.S., Ronnekleiv, O.K., and Kelly, M.J.
(2003). Rapid signaling of estrogen in hypothalamic neurons involves a novel G-protein-coupled
estrogen receptor that activates protein kinase C. The Journal of neuroscience : the official journal
of the Society for Neuroscience 23, 9529-9540.
Rapp, S.R., Espeland, M.A., Shumaker, S.A., Henderson, V.W., Brunner, R.L., Manson, J.E., Gass,
M.L., Stefanick, M.L., Lane, D.S., Hays, J., et al. (2003). Effect of estrogen plus progestin on
158

global cognitive function in postmenopausal women: the Women's Health Initiative Memory
Study: a randomized controlled trial. JAMA 289, 2663-2672.
Rasgon, N.L., Silverman, D., Siddarth, P., Miller, K., Ercoli, L.M., Elman, S., Lavretsky, H.,
Huang, S.C., Phelps, M.E., and Small, G.W. (2005). Estrogen use and brain metabolic change in
postmenopausal women. Neurobiology of aging 26, 229-235.
Ray, A., Prefontaine, K.E., and Ray, P. (1994). Down-modulation of interleukin-6 gene expression
by 17 beta-estradiol in the absence of high affinity DNA binding by the estrogen receptor. The
Journal of biological chemistry 269, 12940-12946.
Rebeck, G.W., Reiter, J.S., Strickland, D.K., and Hyman, B.T. (1993). Apolipoprotein E in
sporadic Alzheimer's disease: allelic variation and receptor interactions. Neuron 11, 575-580.
Redjems-Bennani, N., Jeandel, C., Lefebvre, E., Blain, H., Vidailhet, M., and Gueant, J.L. (1998).
Abnormal substrate levels that depend upon mitochondrial function in cerebrospinal fluid from
Alzheimer patients. Gerontology 44, 300-304.
Reiman, E.M., Caselli, R.J., Chen, K., Alexander, G.E., Bandy, D., and Frost, J. (2001). Declining
brain activity in cognitively normal apolipoprotein E epsilon 4 heterozygotes: A foundation for
using positron emission tomography to efficiently test treatments to prevent Alzheimer's disease.
Proceedings of the National Academy of Sciences of the United States of America 98, 3334-3339.
Reiman, E.M., Chen, K., Alexander, G.E., Caselli, R.J., Bandy, D., Osborne, D., Saunders, A.M.,
and Hardy, J. (2004). Functional brain abnormalities in young adults at genetic risk for late-onset
Alzheimer's dementia. Proceedings of the National Academy of Sciences of the United States of
America 101, 284-289.
Reiman, E.M., Chen, K., Alexander, G.E., Caselli, R.J., Bandy, D., Osborne, D., Saunders, A.M.,
and Hardy, J. (2005). Correlations between apolipoprotein E epsilon4 gene dose and brain-imaging
measurements of regional hypometabolism. Proceedings of the National Academy of Sciences of
the United States of America 102, 8299-8302.
Reitan, R.M. (1958). Validity of the Trail Making Test as an indicator of organic brain damage.
Perceptual and motor skills 8, 271-276.
Renault, V.M., Rafalski, V.A., Morgan, A.A., Salih, D.A.M., Brett, J.O., Webb, A.E., Villeda,
S.A., Thekkat, P.U., Guillerey, C., Denko, N.C., et al. (2009). FoxO3 regulates neural stem cell
homeostasis. Cell Stem Cell 5, 527-539.
Resnick, S.M., Espeland, M.A., An, Y., Maki, P.M., Coker, L.H., Jackson, R., Stefanick, M.L.,
Wallace, R., and Rapp, S.R. (2009). Effects of Conjugated Equine Estrogens on Cognition and
Affect in Postmenopausal Women with Prior Hysterectomy. The Journal of Clinical
Endocrinology & Metabolism 94, 4152-4161.
Resnick, S.M., and Henderson, V.W. (2002). Hormone therapy and risk of alzheimer disease: A
critical time. JAMA 288, 2170-2172.
Resnick, S.M., Maki, P.M., Golski, S., Kraut, M.A., and Zonderman, A.B. (1998). Effects of
Estrogen Replacement Therapy on PET Cerebral Blood Flow and Neuropsychological
Performance. Hormones and Behavior 34, 171-182.
Rettberg, J.R., Yao, J., and Brinton, R.D. (2014). Estrogen: a master regulator of bioenergetic
systems in the brain and body. Front Neuroendocrinol 35, 8-30.
Revankar, C.M., Cimino, D.F., Sklar, L.A., Arterburn, J.B., and Prossnitz, E.R. (2005). A
Transmembrane Intracellular Estrogen Receptor Mediates Rapid Cell Signaling. Science 307,
1625.
Rey, A. (1964). L'examen clinique en psychologie (Paris: Presses universitaires de France).
159

Richardson, J.S. (1993). Free radicals in the genesis of Alzheimer's disease. Annals of the New
York Academy of Sciences 695, 73-76.
Ridge, P.G., Koop, A., Maxwell, T.J., Bailey, M.H., Swerdlow, R.H., Kauwe, J.S., and Honea,
R.A. (2013). Mitochondrial haplotypes associated with biomarkers for Alzheimer's disease. PloS
one 8, e74158.
Ridge, P.G., Maxwell, T.J., Corcoran, C.D., Norton, M.C., Tschanz, J.T., O’Brien, E., Kerber,
R.A., Cawthon, R.M., Munger, R.G., and Kauwe, J.S.K. (2012). Mitochondrial Genomic Analysis
of Late Onset Alzheimer’s Disease Reveals Protective Haplogroups H6A1A/H6A1B: The Cache
County Study on Memory in Aging. PloS one 7, e45134.
Riske, L., Thomas, R.K., Baker, G.B., and Dursun, S.M. Lactate in the brain: an update on its
relevance to brain energy, neurons, glia and panic disorder.
Risner, M.E., Saunders, A.M., Altman, J.F., Ormandy, G.C., Craft, S., Foley, I.M., Zvartau-Hind,
M.E., Hosford, D.A., Roses, A.D., and Rosiglitazone in Alzheimer's Disease Study, G. (2006).
Efficacy of rosiglitazone in a genetically defined population with mild-to-moderate Alzheimer's
disease. The pharmacogenomics journal 6, 246-254.
Rissman, E.F., Heck, A.L., Leonard, J.E., Shupnik, M.A., and Gustafsson, J.A. (2002). Disruption
of estrogen receptor beta gene impairs spatial learning in female mice. Proceedings of the National
Academy of Sciences of the United States of America 99, 3996-4001.
Ro, S.-H., Liu, D., Yeo, H., and Paik, J.-h. (2013). FoxOs in neural stem cell fate decision.
Archives of Biochemistry and Biophysics 534, 55-63.
Roses, A.D., Saunders, A.M., Huang, Y., Strum, J., Weisgraber, K.H., and Mahley, R.W. (2007).
Complex disease-associated pharmacogenetics: drug efficacy, drug safety, and confirmation of a
pathogenetic hypothesis (Alzheimer's disease). The pharmacogenomics journal 7, 10-28.
Rossouw, J.E., Anderson, G.L., Prentice, R.L., LaCroix, A.Z., Kooperberg, C., Stefanick, M.L.,
Jackson, R.D., Beresford, S.A., Howard, B.V., Johnson, K.C., et al. (2002). Risks and benefits of
estrogen plus progestin in healthy postmenopausal women: principal results From the Women's
Health Initiative randomized controlled trial. Jama 288, 321-333.
Rossouw, J.E., Prentice, R.L., Manson, J.E., Wu, L., Barad, D., Barnabei, V.M., Ko, M., LaCroix,
A.Z., Margolis, K.L., and Stefanick, M.L. (2007). Postmenopausal hormone therapy and risk of
cardiovascular disease by age and years since menopause. JAMA 297, 1465-1477.
Rothman, D.L., De Feyter, H.M., de Graaf, R.A., Mason, G.F., and Behar, K.L. (2011). 13C MRS
studies of neuroenergetics and neurotransmitter cycling in humans. NMR in biomedicine 24, 943-
957.
Saleem, S., and Biswas, S.C. (2017). Tribbles Pseudokinase 3 Induces Both Apoptosis and
Autophagy in Amyloid-beta-induced Neuronal Death. The Journal of biological chemistry 292,
2571-2585.
Santen, R.J., Allred, D.C., Ardoin, S.P., Archer, D.F., Boyd, N., Braunstein, G.D., Burger, H.G.,
Colditz, G.A., Davis, S.R., Gambacciani, M., et al. (2010). Postmenopausal hormone therapy: an
Endocrine Society scientific statement. The Journal of clinical endocrinology and metabolism 95,
s1-s66.
Santoro, A., Balbi, V., Balducci, E., Pirazzini, C., Rosini, F., Tavano, F., Achilli, A., Siviero, P.,
Minicuci, N., Bellavista, E., et al. (2010). Evidence for sub-haplogroup h5 of mitochondrial DNA
as a risk factor for late onset Alzheimer's disease. PloS one 5, e12037.
Saunders, A.M., Strittmatter, W.J., Schmechel, D., George-Hyslop, P.S., Pericak-Vance, M.A.,
Joo, S., Rosi, B., Gusella, J., Crapper-MacLachlan, D., and Alberts, M. (1993). Association of
160

apolipoprotein E allele ϵ4 with late ‐onset familial and sporadic Alzheimer's disease. Neurology
43, 1467-1467.
Saville, B., Wormke, M., Wang, F., Nguyen, T., Enmark, E., Kuiper, G., Gustafsson, J.A., and
Safe, S. (2000). Ligand-, cell-, and estrogen receptor subtype (alpha/beta)-dependent activation at
GC-rich (Sp1) promoter elements. The Journal of biological chemistry 275, 5379-5387.
Scahill, R.I., Schott, J.M., Stevens, J.M., Rossor, M.N., and Fox, N.C. (2002). Mapping the
evolution of regional atrophy in Alzheimer's disease: Unbiased analysis of fluid-registered serial
MRI. Proceedings of the National Academy of Sciences of the United States of America 99, 4703-
4707.
Schagger, H., and Ohm, T.G. (1995). Human diseases with defects in oxidative phosphorylation.
2. F1F0 ATP-synthase defects in Alzheimer disease revealed by blue native polyacrylamide gel
electrophoresis. European journal of biochemistry / FEBS 227, 916-921.
Scheffler, K., Krohn, M., Dunkelmann, T., Stenzel, J., Miroux, B., Ibrahim, S., von Bohlen Und
Halbach, O., Heinze, H.J., Walker, L.C., Gsponer, J.A., et al. (2012). Mitochondrial DNA
polymorphisms specifically modify cerebral beta-amyloid proteostasis. Acta neuropathologica
124, 199-208.
Schönfeld, P., and Reiser, G. (2013). Why does brain metabolism not favor burning of fatty acids
to provide energy? - Reflections on disadvantages of the use of free fatty acids as fuel for brain.
Journal of Cerebral Blood Flow & Metabolism 33, 1493-1499.
Schroeter, M.L., Stein, T., Maslowski, N., and Neumann, J. (2009). Neural correlates of
Alzheimer's disease and mild cognitive impairment: A systematic and quantitative meta-analysis
involving 1351 patients. NeuroImage 47, 1196-1206.
Schuessel, K., Leutner, S., Cairns, N.J., Muller, W.E., and Eckert, A. (2004). Impact of gender on
upregulation of antioxidant defence mechanisms in Alzheimer's disease brain. J Neural Transm
(Vienna) 111, 1167-1182.
Seshadri, S., Beiser, A., Kelly-Hayes, M., Kase, C.S., Au, R., Kannel, W.B., and Wolf, P.A. (2006).
The lifetime risk of stroke: estimates from the Framingham Study. Stroke; a journal of cerebral
circulation 37, 345-350.
Seshadri, S., Zornberg, G.L., Derby, L.E., Myers, M.W., Jick, H., and Drachman, D.A. (2001).
Postmenopausal estrogen replacement therapy and the risk of Alzheimer disease. Arch Neurol 58,
435-440.
Shaywitz, S.E., Naftolin, F., Zelterman, D., Marchione, K.E., Holahan, J.M., Palter, S.F., and
Shaywitz, B.A. (2003). Better oral reading and short-term memory in midlife, postmenopausal
women taking estrogen. Menopause (New York, NY) 10, 420-426.
Sheehan, J.P., Swerdlow, R.H., Miller, S.W., Davis, R.E., Parks, J.K., Parker, W.D., and Tuttle,
J.B. (1997). Calcium homeostasis and reactive oxygen species production in cells transformed by
mitochondria from individuals with sporadic Alzheimer's disease. The Journal of neuroscience :
the official journal of the Society for Neuroscience 17, 4612-4622.
Sherwin, B.B. (1988). Estrogen and/or androgen replacement therapy and cognitive functioning in
surgically menopausal women. Psychoneuroendocrinology 13, 345-357.
Shi, J., and Simpkins, J.W. (1997). 17 beta-Estradiol modulation of glucose transporter 1
expression in blood-brain barrier. The American journal of physiology 272, E1016-1022.
Shi, L., Du, X., Zhou, H., Tao, C., Liu, Y., Meng, F., Wu, G., Xiong, Y., Xia, C., Wang, Y., et al.
(2014). Cumulative effects of the ApoE genotype and gender on the synaptic proteome and
oxidative stress in the mouse brain. The international journal of neuropsychopharmacology /
161

official scientific journal of the Collegium Internationale Neuropsychopharmacologicum (CINP)
17, 1863-1879.
Shoffner, J.M., Brown, M.D., Torroni, A., Lott, M.T., Cabell, M.F., Mirra, S.S., Beal, M.F., Yang,
C.-C., Gearing, M., Salvo, R., et al. (1993). Mitochondrial DNA Variants Observed in Alzheimer
Disease and Parkinson Disease Patients. Genomics 17, 171-184.
Shughrue, P.J., Lane, M.V., and Merchenthaler, I. (1997). Comparative distribution of estrogen
receptor-alpha and -beta mRNA in the rat central nervous system. The Journal of comparative
neurology 388, 507-525.
Shughrue, P.J., Lane, M.V., Scrimo, P.J., and Merchenthaler, I. (1998). Comparative distribution
of estrogen receptor-alpha (ER-alpha) and beta (ER-beta) mRNA in the rat pituitary, gonad, and
reproductive tract. Steroids 63, 498-504.
Simpkins, J.W., Yang, S.-H., Sarkar, S.N., and Pearce, V. (2008a). Estrogen actions on
mitochondria—Physiological and pathological implications. Molecular and Cellular
Endocrinology 290, 51-59.
Simpkins, J.W., Yang, S.H., Sarkar, S.N., and Pearce, V. (2008b). Estrogen actions on
mitochondria--physiological and pathological implications. Mol Cell Endocrinol 290, 51-59.
Sims, N.R., Finegan, J.M., Blass, J.P., Bowen, D.M., and Neary, D. (1987). Mitochondrial function
in brain tissue in primary degenerative dementia. Brain Research 436, 30-38.
Singh, C., Liu, L., Wang, J.M., Irwin, R.W., Yao, J., Chen, S., Henry, S., Thompson, R.F., and
Brinton, R.D. (2012). Allopregnanolone restores hippocampal-dependent learning and memory
and neural progenitor survival in aging 3xTgAD and nonTg mice. Neurobiology of aging 33, 1493-
1506.
Skup, M., Zhu, H., Wang, Y., Giovanello, K.S., Lin, J.A., Shen, D., Shi, F., Gao, W., Lin, W., Fan,
Y., et al. (2011). Sex differences in grey matter atrophy patterns among AD and aMCI patients:
results from ADNI. NeuroImage 56, 890-906.
Small, G.W., Ercoli, L.M., Silverman, D.H., Huang, S.C., Komo, S., Bookheimer, S.Y., Lavretsky,
H., Miller, K., Siddarth, P., Rasgon, N.L., et al. (2000). Cerebral metabolic and cognitive decline
in persons at genetic risk for Alzheimer's disease. Proceedings of the National Academy of
Sciences of the United States of America 97, 6037-6042.
Small, G.W., Mazziotta, J.C., Collins, M.T., and et al. (1995). APolipoprotein e type 4 allele and
cerebral glucose metabolism in relatives at risk for familial alzheimer disease. JAMA 273, 942-
947.
Solinsky, C.M. (2017). An iPSC-based biomarker strategy to identify neuroregenerative
responders to allopregnanolone.
Soneson, C., Love, M.I., and Robinson, M.D. (2015). Differential analyses for RNA-seq:
transcript-level estimates improve gene-level inferences. F1000Research 4, 1521.
Spencer-Segal, J.L., Tsuda, M.C., Mattei, L., Waters, E.M., Romeo, R.D., Milner, T.A., McEwen,
B.S., and Ogawa, S. (2012). Estradiol acts via estrogen receptors alpha and beta on pathways
important for synaptic plasticity in the mouse hippocampal formation. Neuroscience 202, 131-146.
Stankiewicz, T.R., and Linseman, D.A. (2014). Rho family GTPases: key players in neuronal
development, neuronal survival, and neurodegeneration. Frontiers in Cellular Neuroscience 8.
Stein, B., and Yang, M.X. (1995). Repression of the interleukin-6 promoter by estrogen receptor
is mediated by NF-kappa B and C/EBP beta. Molecular and cellular biology 15, 4971-4979.
Stewart, J.B., and Chinnery, P.F. (2015). The dynamics of mitochondrial DNA heteroplasmy:
implications for human health and disease. Nat Rev Genet 16, 530-542.
162

Stirone, C., Duckles, S.P., Krause, D.N., and Procaccio, V. (2005). Estrogen increases
mitochondrial efficiency and reduces oxidative stress in cerebral blood vessels. Molecular
pharmacology 68, 959-965.
Stoecklin, E., Wissler, M., Schaetzle, D., Pfitzner, E., and Groner, B. (1999). Interactions in the
transcriptional regulation exerted by Stat5 and by members of the steroid hormone receptor family.
J Steroid Biochem Mol Biol 69, 195-204.
Sun, C., Ou, X., Farley, J.M., Stockmeier, C., Bigler, S., Brinton, R.D., and Wang, J.M. (2012).
Allopregnanolone increases the number of dopaminergic neurons in substantia nigra of a triple
transgenic mouse model of Alzheimer's disease. Current Alzheimer research 9, 473-480.
Swerdlow, R.H., and Khan, S.M. (2004). A "mitochondrial cascade hypothesis" for sporadic
Alzheimer's disease. Medical hypotheses 63, 8-20.
Takasaki, S. (2008). Mitochondrial SNPs associated with Japanese centenarians, Alzheimer's
patients, and Parkinson's patients. Computational biology and chemistry 32, 332-337.
Takasaki, S. (2009). Mitochondrial haplogroups associated with Japanese centenarians,
Alzheimer's patients, Parkinson's patients, type 2 diabetic patients and healthy non-obese young
males. Journal of genetics and genomics = Yi chuan xue bao 36, 425-434.
Tang, M.-X., Jacobs, D., Stern, Y., Marder, K., Schofield, P., Gurland, B., Andrews, H., and
Mayeux, R. (1996). Effect of oestrogen during menopause on risk and age at onset of Alzheimer's
disease. The Lancet 348, 429-432.
Taupin, P., Ray, J., Fischer, W.H., Suhr, S.T., Hakansson, K., Grubb, A., and Gage, F.H. (2000).
FGF-2-responsive neural stem cell proliferation requires CCg, a novel autocrine/paracrine cofactor.
Neuron 28, 385-397.
Teekachunhatean, S., Mattawanon, N., and Khunamornpong, S. (2015). Short-Term Isoflavone
Intervention in the Treatment of Severe Vasomotor Symptoms after Surgical Menopause: A Case
Report and Literature Review. Case reports in obstetrics and gynecology 2015, 962740.
Teruya, L.C., Ortiz, K.Z., and Minett, T.S. (2009). Performance of normal adults on Rey Auditory
Learning Test: a pilot study. Arquivos de neuro-psiquiatria 67, 224-228.
The Lancet, N. (2017). Solanezumab: too late in mild Alzheimer's disease? The Lancet Neurology
16, 97.
Tomishima, M. (2008). Neural induction - Dual SMAD inhibition. In StemBook (Cambridge
(MA): Harvard Stem Cell Institute
Copyright: (c) 2012 Mark Tomishima.).
Toran-Allerand, C.D., Guan, X., MacLusky, N.J., Horvath, T.L., Diano, S., Singh, M., Connolly,
E.S., Jr., Nethrapalli, I.S., and Tinnikov, A.A. (2002). ER-X: a novel, plasma membrane-
associated, putative estrogen receptor that is regulated during development and after ischemic
brain injury. The Journal of neuroscience : the official journal of the Society for Neuroscience 22,
8391-8401.
Torroni, A., Huoponen, K., Francalacci, P., Petrozzi, M., Morelli, L., Scozzari, R., Obinu, D.,
Savontaus, M.L., and Wallace, D.C. (1996). Classification of european mtDNAs from an analysis
of three European populations. Genetics 144, 1835-1850.
Torroni, A., Schurr, T.G., Yang, C.C., Szathmary, E.J., Williams, R.C., Schanfield, M.S., Troup,
G.A., Knowler, W.C., Lawrence, D.N., Weiss, K.M., et al. (1992). Native American mitochondrial
DNA analysis indicates that the Amerind and the Nadene populations were founded by two
independent migrations. Genetics 130, 153-162.
163

Tranah, G.J., Manini, T.M., Lohman, K.K., Nalls, M.A., Kritchevsky, S., Newman, A.B., Harris,
T.B., Miljkovic, I., Biffi, A., Cummings, S.R., et al. (2011). Mitochondrial DNA variation in
human metabolic rate and energy expenditure. Mitochondrion 11, 855-861.
Tranah, G.J., Nalls, M.A., Katzman, S.M., Yokoyama, J.S., Lam, E.T., Zhao, Y., Mooney, S.,
Thomas, F., Newman, A.B., Liu, Y., et al. (2012). Mitochondrial DNA sequence variation
associated with dementia and cognitive function in the elderly. Journal of Alzheimer's disease :
JAD 32, 357-372.
Tranah, G.J., Yokoyama, J.S., Katzman, S.M., Nalls, M.A., Newman, A.B., Harris, T.B., Cesari,
M., Manini, T.M., Schork, N.J., Cummings, S.R., et al. (2014). Mitochondrial DNA sequence
associations with dementia and amyloid-beta in elderly African Americans. Neurobiology of aging
35, 442 e441-448.
Tremblay, A., Tremblay, G.B., Labrie, F., and Giguere, V. (1999). Ligand-independent
recruitment of SRC-1 to estrogen receptor beta through phosphorylation of activation function AF-
1. Molecular cell 3, 513-519.
Trimmer, P.A., Swerdlow, R.H., Parks, J.K., Keeney, P., Bennett, J.P., Jr., Miller, S.W., Davis,
R.E., and Parker, W.D., Jr. (2000). Abnormal mitochondrial morphology in sporadic Parkinson's
and Alzheimer's disease cybrid cell lines. Experimental neurology 162, 37-50.
Ulfert, G., Schmidt, U., and Hoyer, S. (1982). Glucose and energy metabolism of rat cerebral
cortex during aging. Experimental brain research Suppl 5, 102-111.
Valla, J., Schneider, L., Niedzielko, T., Coon, K.D., Caselli, R., Sabbagh, M.N., Ahern, G.L.,
Baxter, L., Alexander, G., Walker, D.G., et al. (2006). Impaired platelet mitochondrial activity in
Alzheimer's disease and mild cognitive impairment. Mitochondrion 6, 323-330.
Valla, J., Yaari, R., Wolf, A.B., Kusne, Y., Beach, T.G., Roher, A.E., Corneveaux, J.J.,
Huentelman, M.J., Caselli, R.J., and Reiman, E.M. (2010). Reduced posterior cingulate
mitochondrial activity in expired young adult carriers of the APOE epsilon4 allele, the major late-
onset Alzheimer's susceptibility gene. Journal of Alzheimer's disease : JAD 22, 307-313.
van der Walt, J.M., Dementieva, Y.A., Martin, E.R., Scott, W.K., Nicodemus, K.K., Kroner, C.C.,
Welsh-Bohmer, K.A., Saunders, A.M., Roses, A.D., Small, G.W., et al. (2004). Analysis of
European mitochondrial haplogroups with Alzheimer disease risk. Neuroscience letters 365, 28-
32.
van der Walt, J.M., Scott, W.K., Slifer, S., Gaskell, P.C., Martin, E.R., Welsh-Bohmer, K., Creason,
M., Crunk, A., Fuzzell, D., McFarland, L., et al. (2005). Maternal lineages and Alzheimer disease
risk in the Old Order Amish. Human genetics 118, 115-122.
Vandenberghe, R., Rinne, J.O., Boada, M., Katayama, S., Scheltens, P., Vellas, B., Tuchman, M.,
Gass, A., Fiebach, J.B., Hill, D., et al. (2016). Bapineuzumab for mild to moderate Alzheimer's
disease in two global, randomized, phase 3 trials. Alzheimer's research & therapy 8, 18.
Vilchez, D., Boyer, L., Lutz, M., Merkwirth, C., Morantte, I., Tse, C., Spencer, B., Page, L.,
Masliah, E., Berggren, W.T., et al. (2013). FOXO4 is necessary for neural differentiation of human
embryonic stem cells. Aging cell 12, 518-522.
Vladusic, E.A., Hornby, A.E., Guerra-Vladusic, F.K., and Lupu, R. (1998). Expression of Estrogen
Receptor β Messenger RNA Variant in Breast Cancer. Cancer Research 58, 210.
Wallace, D.C. (2005). A MITOCHONDRIAL PARADIGM OF METABOLIC AND
DEGENERATIVE DISEASES, AGING, AND CANCER: A Dawn for Evolutionary Medicine.
Annual Review of Genetics 39, 359-407.
164

Wallace, D.C., Ye, J.H., Neckelmann, S.N., Singh, G., Webster, K.A., and Greenberg, B.D. (1987).
Sequence analysis of cDNAs for the human and bovine ATP synthase beta subunit: mitochondrial
DNA genes sustain seventeen times more mutations. Current genetics 12, 81-90.
Wang, J.M., and Brinton, R.D. (2008). Allopregnanolone-induced rise in intracellular calcium in
embryonic hippocampal neurons parallels their proliferative potential. BMC neuroscience 9 Suppl
2, S11.
Wang, J.M., Irwin, R.W., Liu, L., Chen, S., and Brinton, R.D. (2007). Regeneration in a
degenerating brain: potential of allopregnanolone as a neuroregenerative agent. Current Alzheimer
research 4, 510-517.
Wang, J.M., Johnston, P.B., Ball, B.G., and Brinton, R.D. (2005a). The neurosteroid
allopregnanolone promotes proliferation of rodent and human neural progenitor cells and regulates
cell-cycle gene and protein expression. The Journal of neuroscience : the official journal of the
Society for Neuroscience 25, 4706-4718.
Wang, J.M., Singh, C., Liu, L., Irwin, R.W., Chen, S., Chung, E.J., Thompson, R.F., and Brinton,
R.D. (2010a). Allopregnanolone reverses neurogenic and cognitive deficits in mouse model of
Alzheimer's disease. Proceedings of the National Academy of Sciences of the United States of
America 107, 6498-6503.
Wang, J.M., Singh, C., Liu, L., Irwin, R.W., Chen, S., Chung, E.J., Thompson, R.F., and Brinton,
R.D. (2010b). Allopregnanolone reverses neurogenic and cognitive deficits in mouse model of
Alzheimer's disease. Proceedings of the National Academy of Sciences 107, 6498.
Wang, P.N., Liao, S.Q., Liu, R.S., Liu, C.Y., Chao, H.T., Lu, S.R., Yu, H.Y., Wang, S.J., and Liu,
H.C. (2000). Effects of estrogen on cognition, mood, and cerebral blood flow in AD: a controlled
study. Neurology 54, 2061-2066.
Wang, Y., and Brinton, R.D. (2016). Triad of Risk for Late Onset Alzheimer's: Mitochondrial
Haplotype, APOE Genotype and Chromosomal Sex. Frontiers in aging neuroscience 8, 232.
Wang, Y., and Miksicek, R.J. (1991). Identification of a Dominant Negative Form of the Human
Estrogen Receptor. Molecular endocrinology 5, 1707-1715.
Wang, Z., Zhang, X., Shen, P., Loggie, B.W., Chang, Y., and Deuel, T.F. (2005b). Identification,
cloning, and expression of human estrogen receptor-alpha36, a novel variant of human estrogen
receptor-alpha66. Biochemical and biophysical research communications 336, 1023-1027.
Wang, Z., Zhang, X., Shen, P., Loggie, B.W., Chang, Y., and Deuel, T.F. (2006). A variant of
estrogen receptor-{alpha}, hER-{alpha}36: transduction of estrogen- and antiestrogen-dependent
membrane-initiated mitogenic signaling. Proceedings of the National Academy of Sciences of the
United States of America 103, 9063-9068.
Waring, S.C., Rocca, W.A., Petersen, R.C., O'Brien, P.C., Tangalos, E.G., and Kokmen, E. (1999).
Postmenopausal estrogen replacement therapy and risk of AD: a population-based study.
Neurology 52, 965-970.
Watanabe, K. (2010). Unique features of animal mitochondrial translation systems. Proceedings
of the Japan Academy, Series B 86, 11-39.
Watanabe, K., and Yokobori, S. (2011). tRNA Modification and Genetic Code Variations in
Animal Mitochondria. Journal of nucleic acids 2011, 623095.
Webb, P., Nguyen, P., Valentine, C., Lopez, G.N., Kwok, G.R., McInerney, E., Katzenellenbogen,
B.S., Enmark, E., Gustafsson, J.A., Nilsson, S., et al. (1999). The estrogen receptor enhances AP-
1 activity by two distinct mechanisms with different requirements for receptor transactivation
functions. Molecular endocrinology 13, 1672-1685.
165

Wechsler, D. (1987). Instruction Manual for the Wechsler Memory Scale Revised. New York:
Psychological Corporation.
Weissensteiner, H., Pacher, D., Kloss-Brandstatter, A., Forer, L., Specht, G., Bandelt, H.J.,
Kronenberg, F., Salas, A., and Schonherr, S. (2016). HaploGrep 2: mitochondrial haplogroup
classification in the era of high-throughput sequencing. Nucleic acids research 44, W58-63.
Willette, A.A., Bendlin, B.B., Starks, E.J., Birdsill, A.C., Johnson, S.C., Christian, B.T., Okonkwo,
O.C., La Rue, A., Hermann, B.P., Koscik, R.L., et al. (2015). Association of Insulin Resistance
With Cerebral Glucose Uptake in Late Middle-Aged Adults at Risk for Alzheimer Disease. JAMA
neurology 72, 1013-1020.
William Rebeck, G., Reiter, J.S., Strickland, D.K., and Hyman, B.T. (1993). Apolipoprotein E in
sporadic Alzheimer's disease: Allelic variation and receptor interactions. Neuron 11, 575-580.
Wolf, A.B., Caselli, R.J., Reiman, E.M., and Valla, J. (2013). APOE and neuroenergetics: an
emerging paradigm in Alzheimer's disease. Neurobiology of aging 34, 1007-1017.
Wragg, M.A., Talbot, C.J., Morris, J.C., Lendon, C.L., and Goate, A.M. (1995). No association
found between Alzheimer's disease and a mitochondrial tRNA glutamine gene variant.
Neuroscience letters 201, 107-110.
Wu, A.H., Ziegler, R.G., Horn-Ross, P.L., Nomura, A.M., West, D.W., Kolonel, L.N., Rosenthal,
J.F., Hoover, R.N., and Pike, M.C. (1996). Tofu and risk of breast cancer in Asian-Americans.
Cancer epidemiology, biomarkers & prevention : a publication of the American Association for
Cancer Research, cosponsored by the American Society of Preventive Oncology 5, 901-906.
Wu, T.W., Chen, S., and Brinton, R.D. (2011). Membrane estrogen receptors mediate calcium
signaling and MAP kinase activation in individual hippocampal neurons. Brain Res 1379, 34-43.
Xing, L., Esau, C., and Trudeau, V. (2016). Direct regulation of aromatase B expression by 17β-
estradiol and dopamine D1 receptor agonist in adult radial glial cells. Frontiers in Neuroscience
9.
Xu, P.T., Li, Y.J., Qin, X.J., Kroner, C., Green-Odlum, A., Xu, H., Wang, T.Y., Schmechel, D.E.,
Hulette, C.M., Ervin, J., et al. (2007). A SAGE study of apolipoprotein E3/3, E3/4 and E4/4 allele-
specific gene expression in hippocampus in Alzheimer disease. Molecular and cellular
neurosciences 36, 313-331.
Xu, P.T., Li, Y.J., Qin, X.J., Scherzer, C.R., Xu, H., Schmechel, D.E., Hulette, C.M., Ervin, J.,
Gullans, S.R., Haines, J., et al. (2006). Differences in apolipoprotein E3/3 and E4/4 allele-specific
gene expression in hippocampus in Alzheimer disease. Neurobiology of disease 21, 256-275.
Xu, S., Schaack, S., Seyfert, A., Choi, E., Lynch, M., and Cristescu, M.E. (2012). High Mutation
Rates in the Mitochondrial Genomes of Daphnia pulex. Molecular biology and evolution 29, 763-
769.
Yager, J.D., and Chen, J.Q. (2007). Mitochondrial estrogen receptors--new insights into specific
functions. Trends in endocrinology and metabolism: TEM 18, 89-91.
Yang, S.H., Liu, R., Perez, E.J., Wen, Y., Stevens, S.M., Jr., Valencia, T., Brun-Zinkernagel, A.M.,
Prokai, L., Will, Y., Dykens, J., et al. (2004). Mitochondrial localization of estrogen receptor beta.
Proceedings of the National Academy of Sciences of the United States of America 101, 4130-4135.
Yang, S.H., Sarkar, S.N., Liu, R., Perez, E.J., Wang, X., Wen, Y., Yan, L.J., and Simpkins, J.W.
(2009). Estrogen receptor beta as a mitochondrial vulnerability factor. The Journal of biological
chemistry 284, 9540-9548.
Yang, Y., Cheung, H.H., Tu, J.J., Miu, K.K., and Chan, W.Y. (2016). New insights into the
unfolded protein response in stem cells. Oncotarget 7, 54010-54027.
166

Yao, J., and Brinton, R.D. (2012). Estrogen regulation of mitochondrial bioenergetics:
implications for prevention of Alzheimer's disease. Advances in pharmacology (San Diego, Calif)
64, 327-371.
Yao, J., Chen, S., Cadenas, E., and Brinton, R.D. (2011a). Estrogen protection against
mitochondrial toxin-induced cell death in hippocampal neurons: Antagonism by progesterone.
Brain Research 1379, 2-10.
Yao, J., Chen, S., Cadenas, E., and Brinton, R.D. (2011b). Estrogen protection against
mitochondrial toxin-induced cell death in hippocampal neurons: antagonism by progesterone.
Brain Res 1379, 2-10.
Yao, J., Hamilton, R.T., Cadenas, E., and Brinton, R.D. (2010). Decline in mitochondrial
bioenergetics and shift to ketogenic profile in brain during reproductive senescence. Biochimica
et biophysica acta 1800, 1121-1126.
Yao, J., Irwin, R., Chen, S., Hamilton, R., Cadenas, E., and Brinton, R.D. (2012a). Ovarian
hormone loss induces bioenergetic deficits and mitochondrial beta-amyloid. Neurobiology of
aging 33, 1507-1521.
Yao, J., Irwin, R., Chen, S., Hamilton, R., Cadenas, E., and Brinton, R.D. (2012b). Ovarian
hormone loss induces bioenergetic deficits and mitochondrial β-amyloid. Neurobiology of Aging
33, 1507-1521.
Yao, J., Irwin, R.W., Zhao, L., Nilsen, J., Hamilton, R.T., and Brinton, R.D. (2009). Mitochondrial
bioenergetic deficit precedes Alzheimer's pathology in female mouse model of Alzheimer's disease.
Proceedings of the National Academy of Sciences of the United States of America 106, 14670-
14675.
Yao, J., Rettberg, J.R., Klosinski, L.P., Cadenas, E., and Brinton, R.D. (2011c). Shift in brain
metabolism in late onset Alzheimer's disease: implications for biomarkers and therapeutic
interventions. Molecular aspects of medicine 32, 247-257.
Yao, J., Zhao, L., Mao, Z., Chen, S., Wong, K.C., To, J., and Brinton, R.D. (2013). Potentiation
of brain mitochondrial function by S-equol and R/S-equol estrogen receptor beta-selective
phytoSERM treatments. Brain Res 1514, 128-141.
Yi, P., Driscoll, M.D., Huang, J., Bhagat, S., Hilf, R., Bambara, R.A., and Muyan, M. (2002). The
effects of estrogen-responsive element- and ligand-induced structural changes on the recruitment
of cofactors and transcriptional responses by ER alpha and ER beta. Molecular endocrinology 16,
674-693.
Yin, F., Yao, J., Sancheti, H., Feng, T., Melcangi, R.C., Morgan, T.E., Finch, C.E., Pike, C.J.,
Mack, W.J., Cadenas, E., et al. (2015). The perimenopausal aging transition in the female rat brain:
decline in bioenergetic systems and synaptic plasticity. Neurobiology of aging 36, 2282-2295.
Zandi, P.P., Carlson, M.C., Plassman, B.L., and et al. (2002). Hormone replacement therapy and
incidence of alzheimer disease in older women: The cache county study. JAMA 288, 2123-2129.
Zhang, J., Light, A.R., Hoppel, C.L., Campbell, C., Chandler, C.J., Burnett, D.J., Souza, E.C.,
Casazza, G.A., Hughen, R.W., Keim, N.L., et al. (2017). Acylcarnitines as markers of exercise-
associated fuel partitioning, xenometabolism, and potential signals to muscle afferent neurons.
Experimental physiology 102, 48-69.
Zhang, J.-Q., Cai, W.-Q., Zhou, D.-S., and Su, B.-Y. (2002). Distribution and differences of
estrogen receptor beta immunoreactivity in the brain of adult male and female rats. Brain Research
935, 73-80.
167

Zhang, L., Trushin, S., Christensen, T.A., Bachmeier, B.V., Gateno, B., Schroeder, A., Yao, J.,
Itoh, K., Sesaki, H., Poon, W.W., et al. (2016). Altered brain energetics induces mitochondrial
fission arrest in Alzheimer’s Disease. Scientific Reports 6, 18725.
Zhao, L., and Brinton, R.D. (2005a). Estrogen receptor β as a therapeutic target for promoting
neurogenesis and preventing neurodegeneration. Drug Development Research 66, 103-117.
Zhao, L., and Brinton, R.D. (2005b). Structure-based virtual screening for plant-based ERbeta-
selective ligands as potential preventative therapy against age-related neurodegenerative diseases.
Journal of medicinal chemistry 48, 3463-3466.
Zhao, L., and Brinton, R.D. (2007a). Estrogen receptor alpha and beta differentially regulate
intracellular Ca(2+) dynamics leading to ERK phosphorylation and estrogen neuroprotection in
hippocampal neurons. Brain Res 1172, 48-59.
Zhao, L., and Brinton, R.D. (2007b). WHI and WHIMS follow-up and human studies of soy
isoflavones on cognition. Expert review of neurotherapeutics 7, 1549-1564.
Zhao, L., Chen, Q., and Diaz Brinton, R. (2002). Neuroprotective and neurotrophic efficacy of
phytoestrogens in cultured hippocampal neurons. Experimental biology and medicine (Maywood,
NJ) 227, 509-519.
Zhao, L., Mao, Z., and Brinton, R.D. (2009). A select combination of clinically relevant
phytoestrogens enhances estrogen receptor beta-binding selectivity and neuroprotective activities
in vitro and in vivo. Endocrinology 150, 770-783.
Zhao, L., Mao, Z., Chen, S., Schneider, L.S., and Brinton, R.D. (2013). Early intervention with an
estrogen receptor beta-selective phytoestrogenic formulation prolongs survival, improves spatial
recognition memory, and slows progression of amyloid pathology in a female mouse model of
Alzheimer's disease. Journal of Alzheimer's disease : JAD 37, 403-419.
Zhao, L., Mao, Z., Schneider, L.S., and Brinton, R.D. (2011). Estrogen receptor beta-selective
phytoestrogenic formulation prevents physical and neurological changes in a preclinical model of
human menopause. Menopause (New York, NY) 18, 1131-1142.
Zhao, L., Morgan, T.E., Mao, Z., Lin, S., Cadenas, E., Finch, C.E., Pike, C.J., Mack, W.J., and
Brinton, R.D. (2012). Continuous versus cyclic progesterone exposure differentially regulates
hippocampal gene expression and functional profiles. PloS one 7, e31267.
Zhao, L., O'Neill, K., and Brinton, R.D. (2006). Estrogenic agonist activity of ICI 182,780
(Faslodex) in hippocampal neurons: implications for basic science understanding of estrogen
signaling and development of estrogen modulators with a dual therapeutic profile. The Journal of
pharmacology and experimental therapeutics 319, 1124-1132.
Zhao, L., Wu, T.W., and Brinton, R.D. (2004). Estrogen receptor subtypes alpha and beta
contribute to neuroprotection and increased Bcl-2 expression in primary hippocampal neurons.
Brain Res 1010, 22-34.
Zhu, X.-H., Lu, M., Lee, B.-Y., Ugurbil, K., and Chen, W. (2015). In vivo NAD assay reveals the
intracellular NAD contents and redox state in healthy human brain and their age dependences.
Proceedings of the National Academy of Sciences 112, 2876.
Ziegler, A.B., Thiele, C., Tenedini, F., Richard, M., Leyendecker, P., Hoermann, A., Soba, P., and
Tavosanis, G. (2017). Cell-Autonomous Control of Neuronal Dendrite Expansion via the Fatty
Acid Synthesis Regulator SREBP. Cell Reports 21, 3346-3353.
Zsurka, G., Kalman, J., Csaszar, A., Rasko, I., Janka, Z., and Venetianer, P. (1998). No
mitochondrial haplotype was found to increase risk for Alzheimer's disease. Biological psychiatry
44, 371-373.

Abstract (if available)

Abstract Brain glucose hypometabolism and mitochondrial dysfunction is a key signature of late onset Alzheimer’s disease (AD). Similar metabolic pattern was also observed in natural aging and during the perimenopausal transition in the female brain, making age and chromosomal sex two risk factors for late onset AD. Comprehensive understanding of the dynamic metabolic aging process in the female brain can shed light on potential interventions and prevention windows for AD. Further, mitochondrial bioenergetics is central to the brain metabolic aging, and estrogen was demonstrated to be a master regulator of mitochondrial bioenergetics. While much is known about the effect of estrogen on electron transport chain activities, the contribution of, and the coordination between mitochondrial and nuclear genome in response to estrogen stimulation were not as clear. The regulator effect of estrogen receptor subtypes in different cell types in the context of brain aging and AD is also not clear. Moreover, the maternal inheritance pattern of both late onset AD and mitochondrial genome suggest its key involvement in the etiology of AD, most likely due to differential mitochondrial bioenergetic phenotypes associated with mitochondrial genetic variances. Thus, a comprehensive understanding of the metabolic and respiratory profiles at key transition points, the underlying mechanism, and key players involved can help identify potential therapeutic targets and windows. ❧ Using a rat model recapitulating the fundamental characteristics of human perimenopausal transition, we generated a detailed road map of metabolic profile changes throughout different stages of chronological and endocrinological aging. Using rat primary neurons, astrocytes, and differentiated and wildtype human neuroblastoma cells mimicking neuronal and astrocytic respiratory phenotypes, we demonstrated the cell type and estrogen receptor subtype specific regulatory effect of estrogen on mitochondrial transcription and respiratory capacity. With the aid of transcriptome analysis, we further identified roles of unliganded estrogen receptors in regulating mitochondrial bioenergetics. And by conducting retrospective analysis on two clinical studies, we demonstrated that for therapeutics targeting mitochondrial bioenergetics, three genetic risk factors of late onset Alzheimer’s disease: mitochondrial genetic variances, APOE genotype, and chromosomal sex can modulate their therapeutic outcomes or even be used to predict and differentiate responders from non-responders. ❧ Collectively, this dissertation study generated profiles of at risk bioenergetic phenotypes to help better understand the mechanism leading to late onset AD in the female brain, provided deeper mechanistic understanding of estrogen regulation of brain bioenergetics, and further demonstrated a practical, genetic marker-based precision medicine approach to identify potential responders to AD therapeutics. These outcomes can be further translated into strategies for predicting at-risk phenotypes, and into guidance for therapeutics development as well as clinical trial design of future neurodegenerative therapeutics.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Neuroinflammation and ApoE4 genotype in at-risk female aging: implications for Alzheimer's disease

PDF

Estradiol regulation of cerebral metabolism: implications for neuroprotection and mitochondrial bioenergetics

PDF

Bioenergetic fuel shift of reproductive aging: implication for late-onset Alzheimer's disease

PDF

Estrogen regulation of bioenergetics and mitochondrial function: implications for Alzheimer's disease risk and therapeutics

PDF

Insulin sensitivity in cognition, Alzheimer's disease and brain aging

PDF

An iPSC-based biomarker strategy to identify neuroregenerative responders to allopregnanolone

PDF

Perimenopausal transition increases blood brain permeability: implications for neurodegenerative diseases

PDF

C-jun N-terminal Kinase (JNK) mediated inhibition of Pyruvate Dehydrogenase (PDH) activity and its effect on mitochondrial metabolism during brain aging

PDF

The regulation of fatty acid oxidation by estrogen related receptor alpha

PDF

Dihydromyricetin as a potential therapeutic for Alzheimer's disease-like changes induced by social isolation

PDF

Phenotypic and multi-omic characterization of novel C. elegans models of Alzheimer's disease

PDF

Averting dementia: renin-angiotensin system and angiogenic cells in cognitive decline

PDF

Development of dihydromyricetin (DHM) as a novel therapy for alcoholic liver disease (ALD) and alcohol use disorder (AUD)

PDF

Transcriptional regulation of lipid metabolism and exploring the specific roles of AKT isoforms

PDF

Renin-angiotensin system modulation for the prevention and treatment of metabolic dysfunction

PDF

Epigenetic regulation of endocrine aging transitions of the perimenopausal and menopausal brain

Asset Metadata

Creator Wang, Yiwei (Yvette) (author)

Core Title From risk mitochondrial and metabolic phenotype towards a precision medicine approach for Alzheimer's disease

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Clinical and Experimental Therapeutics

Publication Date 11/09/2018

Defense Date 10/16/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Alzheimer's disease,APOE,estrogen receptor,Menopause,metabolic profile,mitochondria,OAI-PMH Harvest,precision medicine

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Cadenas, Enrique (committee chair), Brinton, Roberta (committee member), Pike, Christian (committee member), Rodgers, Kathleen (committee member)

Creator Email yiweiwan@usc.edu,yvetteyww@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-103599

Unique identifier UC11676692

Identifier etd-WangYiweiY-6953.pdf (filename),usctheses-c89-103599 (legacy record id)

Legacy Identifier etd-WangYiweiY-6953.pdf

Dmrecord 103599

Document Type Dissertation

Format application/pdf (imt)

Rights Wang, Yiwei (Yvette)

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