University of Southern California Dissertations and Theses

Allopregnanolone promotes cholesterol and amyloid-beta clearance mechanisms

(USC Thesis Other)

Allopregnanolone promotes cholesterol and amyloid-beta clearance mechanisms

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Allopregnanolone Promotes Cholesterol and Amyloid-Beta Clearance Mechanisms

By

Haley M. Swanson

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTERS OF SCIENCE
(MOLECULAR PHARMACOLOGY AND TOXICOLOGY)

May 2016

ii

ACKNOWLEDGEMENTS

First and foremost I would like to thank my family for helping finance this rather
expensive journey, and always supporting my dreams. To my brother who still was
willing to edit my papers even in grad school. Hopefully this is the last one brother!

I would like to thank my thesis advisor, Dr. Brinton. Her ideas and knowledge were the
basis of my research. I would also like to thank Dr. Davies and Dr. Okamoto for serving
on my thesis committee and providing me with valuable insights.

I owe my deepest gratitude to Dr. Irwin for being my lab mentor. Thank you for all of
your help, and pushing me to get the most out of this experience as possible.

Lastly, I am thankful for my fellow lab members whose help and support have made this
a really great experience. Thank you everyone!

iii

TABLE OF CONTENTS

Acknowledgements

ii
List of Tables

iv
List of Figures

v
Abbreviations

vi
Abstract

viii
Chapter 1 Introduction 1
1.1) Alzheimer’s Disease: Etiology and the need for new therapies 1
1.2) Allopregnanolone: Effects in the aging brain 3
1.3) LXR and PXR: Structure and function 4
1.4) Cholesterol Homeostasis: The importance and implications for disease 7
1.5) Specific Aims

11
Chapter 2 In Silico Analyses to Determine if Allopregnanolone Can Bind to
LXR and PXR
12
1. Rationale 12
2. Methods 14
3. Results 16
4. Conclusion

23
Chapter 3 LXR and PXR Gene Products Are Altered With
Allopregnanolone Administration
24
1. Rationale 24
2. Methods 25
3. Results 29
4. Conclusion

36
Chapter 4 Allopregnanolone Regulation of Cholesterol Homeostasis and
Aβ Clearance Pathways: Gene Expression Analyses
39
1. Rationale 39
2. Methods 40
3. Results 42
4. Conclusion

51
Chapter 5 Overall Discussion

54
Bibliography 62

iv

LIST OF TABLES

Table 1 Structure of ligands for docking study

13
Table 2 Inhibition Constant (μM) and established free energy of
binding (kcal/mol) for each ligand to PXR and LXR
22

v

LIST OF FIGURES

Figure 1 Cholesterol homeostasis in the brain and how it relates to
Alzheimer’s disease.

10
Figure 2 Overview of specific aims.

11
Figure 3 Crystal structures of LXRβ and PXR/RXRα.

18
Figure 4 Binding affinity (Ki) for each ligand to PXR and LXR.

22
Figure 5 Allopregnanolone increased LXRβ, LDL-R, SORL1,
ABCA1, Cyp46a1, and decreased NF-κB expression in
3xTgAD male mice.

34
Figure 6 Actions of the enzyme Cyp46a1 in the brain.

35
Figure 7 Allopregnanolone decreases the gene expression of
OLR1 and PCSK9 that promote cholesterol efflux and
reduce inflammation.

48
Figure 8 Allopregnanolone increases the gene expression of
CXCL16, LRP6, HMGSC1, and Cyp51 involved in
stimulation of the PI3K/Akt/Erk and Wnt pathways as
well as the cholesterol synthesis pathway.
Allopregnanolone also decreases the gene expression of
PCSK9 and AKR1D1 to promote cholesterol efflux.

49
Figure 9 Allopregnanolone increases the gene expression of
STAB2 and Idi2 potentially reducing inflammation and
increasing cholesterol synthesis. Allopregnanolone also
decreases the gene expression of ANGPTL3 a potential
mechanism for reducing plasma cholesterol levels.

50
Figure 10 Allopregnanolone alters genes that are involved in
cholesterol synthesis, transport, and efflux.
Allopregnanolone also changes expression of genes
involved in inflammation and Aβ

53
Figure 11 Allopregnanolone promotes cholesterol and amyloid-
beta clearance mechanisms.
61

vi

ABBREVIATIONS

Allo Allopregnanolone
AD Alzheimer's disease
Aβ Amyloid-beta peptides
ABCA1 ATP-binding cassette type A1
APOE Apolipoprotein E
APP Amyloid precursor protein
BBB Blood-brain barrier
CNS Central nervous system
CXCL16 Chemokine (C-X-C motif) ligand 16
Cyp46a1 Cholesterol 24-hydroxylase
GABA
A
γ-aminobutyric acid type-A receptor
HMGCS HMG-CoA synthase
HMGCS1 HMG-CoA synthase 1
HMGCR HMG-CoA reductase
24OHC 24(S)-hydroxycholesterol
Ki Inhibition constant
IM Intramuscularly
Idi2 Isopentenyl-diphosphate delta isomerase 2
kcal/mol Kilocalorie/mole
kg Kilogram
Cyp51 Lanosterol 14 α-demethylase
LBD Ligand binding domain
LXR Liver-X-receptor
LPL Lipoprotein lipase
LDL-R Low-density lipoprotein receptor
LRP Low-density lipoprotein receptor-related protein
LRP6 LDLR-related protein 6
μM Micromolar
mg Milligram
NPC Niemann-Pick type C
NF-κB Nuclear factor κB
OLR1 Oxidized low density lipoprotein (lectin-like) receptor 1
PXR Pregnane-X-receptor

vii

PCSK9 Proprotein convertase subtilisin/kexin type 9
RXR Retinoid-X-receptor
SORL1 Sortilin-related receptor
SC Subcutaneously
STAB2 Stabilin-2
3xTgAD Triple transgenic Alzheimer’s disease mouse model
Wnt Wingless

viii

ABSTRACT

As the population of elderly Americans increases in the coming years so will the number
of Alzheimer’s disease (AD) patients, which is already listed as the sixth-leading cause of
death in the United States. There is currently “no cure” for AD and the current treatments
only temporarily treat the disease without taking into account the neuronal damage and
cholesterol dysregulation caused by the disease. Cholesterol is a vital component in the
brain for structure and function. Inadequate or excessive amounts can have a catastrophic
consequence, which is why the brain has a strict regulation of cholesterol production,
transport, and removal. It has previously been confirmed that in a mouse model of AD,
Allopregnanolone (Allo) reduces amyloid-beta (Aβ) and neuroinflammatory burden
while promoting cholesterol homeostasis via mechanisms that influence the liver-X-
receptor (LXR) and pregnane-X-receptor (PXR). We sought out to further investigate
these pathways and how Allo promotes cholesterol homeostasis in the cortex and
hippocampus of 3xTgAD mice treated with Allo for 24 hours using methods such as
western blot, gene array, and molecular docking. We established that Allo has either the
capability to bind to LXR and PXR thus modifying protein expression of target genes
directly or by increasing oxysterols production via Cyp46a1, which in turn can activate
the receptors. This ultimately leads to a decrease in inflammation, increases cholesterol
transport and clearance, and enhances Aβ clearance. Additional genes promoting
cholesterol homeostasis were altered with Allo treatment to protect the brain from
inflammation, apoptosis, decrease Aβ, and to regulate cholesterol clearance, production,
and their equivalent receptors.

1

CHAPTER 1
Introduction
1.1) Alzheimer’s Disease: Etiology and the need for new therapies
It is estimated that 5.3 million Americans currently have Alzheimer’s disease (AD) in
2015. One in nine people age 65 and older have the disease (Alzheimer’s Association,
2015). This number will only increase in the coming years as the baby boomer generation
ages, and since Americans live longer due to advances in medicine. Despite the vast
amount of people living with the disease, there is currently no “cure” for AD. The disease
is still poorly understood, which is why Alzheimer’s research is essential to better help
the growing older generation. Research has revealed a great deal about the disease in the
past decade, yet there is still so much that is unknown and needs to be discovered to
improve treatment of the disease.

AD is a degenerative brain disease that is characterized by a decline in memory and
cognitive skills. The hallmark pathologies of the disease are accumulation of the protein
fragment amyloid-beta (Aβ) forming plaques outside neurons and the protein tau forming
tangles inside neurons (Alzheimer’s Association, 2015). The amyloid precursor protein
(APP) is cleaved by different secretases of different biochemical pathways in the brain. In
the amyloidogenic harmful pathway, APP is cleaved by beta-secretase prior to cleavage
with gamma-secretase, generating the Aβ fragments that are associated with amyloid
plaque formation (Reddy, Ganley, & Pfeffer, 2006). The function of APP is still poorly
understood, however since it is a membrane protein concentrated at the synapses of
neurons it is likely to function in axonal growth, synaptic transmission, and neuronal

2

migration (De Strooper & Annaert, 2002). Once Aβ is generated and accumulated it is
believed to interfere with neuron-to-neuron communication at synapses, ultimately
leading to cell death. The formation of Aβ is irreversible and the product is insoluble and
deleterious (Kang & Rivest, 2012). Tau is a microtubule associated protein that when
hyperphosphorylated forms neurofibrillary tangles (Vance, 2012). Inside neurons, tau
blocks the transport of molecules and nutrients, which causes damage to the cells
(Alzheimer’s Association, 2015). Both the Aβ fragment plaques and tau neurofibrillary
tangles contribute to the cell loss and widespread debris from dead and dying neurons
found in the brains of AD patients. The healthy adult brain contains 100 billion neurons
and 100 trillion synapses that function to communicate with each other, release
chemicals, and send signals to create movements, emotions, memories, and all other
bodily feelings (Alzheimer’s Association, 2015). In addition to the two hallmark
pathologies described above, AD is also associated with inflammation and dysregulated
lipid homeostasis (Kang & Rivest, 2012).

The mean duration between onset of clinical symptoms and death is about 8.5 years. The
brain regions that are mostly altered are the cortex and hippocampus, which participate in
higher brain functions such as thought, action, and memory (Francisa, Palmer, Snape, &
Wilcock, 1999). Currently there are only six medications approved by the U.S. Food and
Drug Administration (FDA) to treat AD. The two different strategies are to increase
select neurotransmitters using cholinesterase inhibitors and to target the N-Methyl-D-
aspartate (NMDA) receptor to increase communication in the brain (Alzheimer’s
Association, 2015). Unfortunately, these drugs only temporarily improve symptoms of

3

the disease (Irwin et al., 2015). None of the treatments available today slow or stop the
neuronal damage (Alzheimer’s Association, 2015). There is, however, a novel therapeutic
approach, which targets the regenerative capacity of the human brain to prevent, treat,
and delay neurodegenerative diseases like AD (Irwin et al., 2015).

1.2) Allopregnanolone: Effects in the aging brain
It has been shown that the neurosteroid Allopregnanolone (Allo) declines with age and
neurodegenerative diseases (Irwin & Brinton, 2014; Mellon, Gong, & Schonemann,
2008). Studies have shown that when comparing AD patients to age-matched controls
Allo was reduced in the blood and brain (Marx et al., 2006; Naylor et al., 2010; Weill-
Engerer et al., 2002). Findings have also reported decreased neurogenesis in AD mouse
models, although this has not yet been proven in humans despite all of the strong
evidence (Lazarov & Marr, 2010).

Allo is a neurosteroid, which is synthesized de novo in the brain from cholesterol (Marx
et al., 2006). It is very similar in structure to progesterone and pregnenolone, which are
earlier steroids in the pathway to produce Allo. Neurosteroids have many proposed roles
in the brain: affecting neuronal function and differentiation, myelination, decreasing
inflammation, reducing apoptosis, neuroprotection, proliferation of neuronal stem cells,
and neurogenesis (Mellon et al., 2008; Naylor et al., 2010). The Brinton lab (University
of Southern California) has demonstrated that Allo stimulates promotion of neurogenesis
and survival of newly generated cells, which reduces AD pathology in an AD mouse
model (Chen et al., 2011). Allo is a potent positive allosteric activator of γ-aminobutyric

4

acid type-A (GABA
A
) receptor channels. Binding causing an efflux of chloride ions and
an influx of calcium consequently activating a signaling cascade, which ultimately leads
to the stimulation of mitotic genes and downregulation of anti-mitotic genes in neural
stem cells (Irwin & Brinton, 2014). It is believed that the rise in calcium and activation of
the cell cycle activates neurogenesis (Irwin & Brinton, 2014). Chronic Allo
administration promotes a reduction in neuroinflammation and Aβ in addition to
increased white matter generation. Findings have shown that Aβ plays a role in
GABAergic dysfunction that may contribute to neural disruption of cognitive functions
(Palop & Mucke, 2010).

1.3) LXR and PXR: Structure and function
Interestingly, Chen et al. discovered that Allo increased expression of liver-X-receptor
(LXR), pregnane-X-receptor (PXR), and HMG-CoA reductase (HMGCR), which are
proteins that regulate cholesterol homeostasis and clearance from the brain (Chen et al.,
2011). From this discovery it has been proposed that Allo also helps regulates cholesterol
homeostasis via the LXR and PXR pathways. Further studies have demonstrated that
Allo increases expression of LXR and PXR target genes (Mellon et al., 2008).
Dysregulation of cholesterol homeostasis is associated with the generation of Aβ and is
therefore a very important process in the aging brain.

LXR and PXR belong to the nuclear hormone receptor superfamily. Nuclear receptors are
ligand-dependent transcription factors that regulate the expression of many critical genes.
Upon agonist binding, conformational changes occur in the ligand-binding domain

5

(LBD), which lead to recruitment of a coactivator and dissociation of a corepressor (Ai,
Krasowski, Welsh, & Ekins, 2009). Nuclear receptors have the ability to interact with
DNA transcription regulation sites which results in the up or downregulation of gene
expression. From an evolutionary prospective, LXR and PXR are closely related and
most similar in the L11-12 loop (Svensson et al., 2003). Both LXR and PXR form
heterodimers with retinoid-X-receptor (RXR), which are both dependent on dimerization
to trigger transcriptional activity (Hong & Tontonoz, 2014; Jakobsson, Treuter,
Gustafsson, & Steffensen, 2012). There are three different isoforms of RXR (α, β and γ),
and little is known about their differences. Polyunsaturated fatty acids have been reported
to activate RXR, and interestingly they have been associated with protection from AD
(Ohyama et al., 2005). Both receptors have evolved a mechanism to sense multiple lipid
metabolites rather than one (Williams et al., 2003). Loss of LXR expression increased
AD-related pathology in an AD mouse model, including an increase in Aβ deposition
(Irwin & Brinton, 2014; Hong & Tontonoz, 2014). In LXR double knockout mice it has
been shown that loss of these receptors results in neurodegenerative disease and
disturbances in lipid homeostasis (Koldamova et al., 2005).

Two isoforms of LXR have been identified. LXRβ is detected throughout the brain,
whereas LXRα is expressed in other tissues such as the liver, small intestine, and kidney
(Hoerer, Schmid, Heckel, Budzinski, & Nar, 2003). It is believed that both LXRs are
located in the central nervous system (CNS) though LXRβ is detected in higher levels,
and the two share 77% sequence identity in both their DNA and LBDs (Wang et al.,
2002). Due to the conservation in sequences both isoforms show similar binding affinities

6

for a range of substances (Svensson et al., 2003). LXRα in the brain seems to be detected
selectively in particular cell types like microglia (Hong & Tontonoz, 2014). Additionally,
there is a high degree of conservation between rodent and human LXRs (Wang et al.,
2002). The LBD is a large hydrophobic cavity, which can accommodate a variety of
different ligands. The natural ligands of the receptor are oxidized forms of cholesterol
known as oxysterols such as 24-S-hydroxycholesterol (24OHC) and 27-
hydroxycholesterol, which are the two primary sterol metabolites in the brain. The
function of LXR in the brain is to regulate pathways that maintain cholesterol balance
and activate anti-inflammatory pathways (Hong & Tontonoz, 2014). LXRβ functions to
“maintain cholesterol homeostasis through promotion of efflux, downregulation of
synthesis and absorption, and suppression of uptake and thus prevent cholesterol
accumulation” (Kang & Rivest, 2012). Furthermore, LXR activation inhibits
inflammatory responses (Zelcer et al., 2007).

PXR is mainly found in the liver where it functions as a xenobiotic sensor activating drug
detoxification genes (Lamba et al., 2004). Studies have shown that it is also expressed in
the brain where it may have a protective role to sense neurosteroid concentrations to
enhance their metabolism and transporters. They may also play a role in protecting
neurons (Lamba et al., 2004). The receptor is very promiscuous due to its large LBD and
ability to bind many chemically and structurally diverse ligands at low affinity. PXR
detects endogenous pregnanes such as estradiol and progesterone (Wallace et al., 2013).
Unlike many nuclear receptors, PXR exhibits a consistent basal transcriptional activity

7

even in the absence of ligand (Xue et al., 2007). Clotrimazole, an imidazole, serves as a
PXR agonist, and fluconazole, an azole, is an antagonist.

1.4) Cholesterol Homeostasis: The importance and implications for disease
The CNS only accounts for 2.1% of body weight although it contains 23% of total body
cholesterol, which makes it the most cholesterol-enriched organ (Dietshy & Turley, 2001;
Hong & Tontonoz, 2014). Cholesterol in the brain mainly resides in myelin sheets (70–
90%) and the rest in the plasma membranes on astrocytes and neurons (Bjorkhem &
Meaney, 2004; Vance, 2012). In the CNS, cholesterol functions in growth factor
signaling, axon guidance, and synaptic transmission (Vance, 2012). Plasma lipoproteins
cannot cross the intact blood brain barrier (BBB); therefore nearly all cholesterol in the
brain is synthesized in situ (Vance, 2012). Brain cholesterol is predominantly synthesized
by glial cells, which are also the most abundant cell type in the brain (Kang & Rivest,
2012). The brain has evolved two strategies to ensure it has enough cholesterol: de novo
synthesis and internalization of lipoprotein particles (Leduc, Jasmin-Bélanger, & Poirier,
2010). De novo synthesis of cholesterol occurs mainly in astrocytes via the HMGCR
pathway, which involves over 20 reactions and is one of the most regulated pathways in
the body (Leduc et al., 2010) (Figure 1). This pathway, also known as the mevalonate
pathway, involves the key enzyme HMGCR that was found to be upregulated by Allo by
Chen et al. The other pathway involves the utilization of lipoprotein-derived cholesterol.
The ATP-binding cassette transporter types A1 and G1 (ABCA1/ABCG1) coordinate the
mobilization of cholesterol from the cytoplasm to the cell surface membrane, where it
combines with Apolipoprotein E (APOE) to produce a functional lipoprotein-like particle

8

through the action of the lipoprotein lipase (LPL) on astrocytes. APOE is considered the
most important transport protein for cholesterol (Bjorkhem & Meaney, 2004). The newly
synthesized lipoproteins are then free to migrate to receptors such as low-density
lipoprotein receptor-related protein (LRP) and low-density lipoprotein receptor (LDL-R)
on neuronal and glial cells where they are internalized and degraded. The free cholesterol
can be used for synapse formation and terminal proliferation (Leduc et al., 2010).

Cholesterol homeostasis in the brain is cautiously executed through a sequence of
practices that include synthesis, storage, degradation, and transport. To maintain
homeostasis, an equivalent amount of the cholesterol made needs to leave the brain.
Since cholesterol cannot leave the brain it is converted to 24OHC by cholesterol 24-
hydroxylase (Cyp46a1), which can cross the BBB (Vance, 2012). This flow is
unidirectional and serves as a key mechanism to sustain cholesterol homeostasis (Hong &
Tontonoz, 2014). Cholesterol elimination occurs in neurons.

Low levels of APOE correlate with an increased risk of AD since APOE associates with
Aβ to enhance its degradation and clearance across the BBB (Vance, 2012; Castellano et
al., 2012). The strongest identified genetic risk factor for AD is the APOE4 isoform,
which clears Aβ the slowest, increases the risk of AD, and lowers the age of onset
(Castellano et al., 2012; Fan et al., 2015). Brain cholesterol levels are considerably
reduced in hippocampal and cortical areas in AD patients compared to age-matched
controls (Leduc et al., 2010). Furthermore, HMGCR activity in the brain of AD patients
is also significantly reduced. LRP is the main receptor for Aβ transport across the BBB,

9

and it is also found to be downregulated and linked to AD patients. Polymorphisms result
in accumulation of Aβ in the brain (Deane, Sagare, & Zlokovic, 2008; Deane et al.,
2004). Hence therapies focusing on upregulation of lipoprotein receptors represent
promising new methods to control Aβ levels in the brain and to treat AD.

Niemann-Pick Type C (NPC) is a neurodegenerative disease associated with the inability
to process cellular cholesterol resulting in accumulation. It is mainly caused by mutations
in NPC1, which is critical in cholesterol trafficking across the plasma membrane (Reddy
et al., 2006; Mellon et al., 2008; Altmann et al., 2004). The disorder is similar to AD in
many ways including tau formation, Aβ accumulation, and myelin breakdown (Naylor et
al., 2010). NPC cells are defective in oxysterol generation and subsequent LXR/PXR
activation (Reddy et al., 2006). These cells consequently have less neurosteroidogenesis
resulting in a decrease in Allo in the brain compared to WT mice (Langmade et al.,
2006). The increase in cholesterol interferes with transport of proteins, thus NPC cells
compensate by upregulating genes involved in membrane trafficking like ABCA1, LDL-
R, and LPL. Allo has been shown to be a beneficial treatment for this disease by
activating PXR and LXR pathways and resulting in delayed neurodegeneration that has
been shown in vivo in mouse models (Langmade et al., 2006; Mellon et al., 2008).
Mellon et al. hypothesized that the short half-life of Allo in brain (mouse t
1/2
~ 30
minutes, human t
1/2
~ 45 minutes) would not result in continued elicitation of GABA
A

receptor-mediated effects weeks later; therefore, the lasting effects are due to induction of
PXR and LXR (Mellon et al., 2008). This further validates the importance of cholesterol

10

homeostasis in the brain and demonstrates how Allo may be used to treat
neurodegenerative diseases through the stimulation of LXR/PXR pathways.

Figure 1. Cholesterol homeostasis in the brain and how it relates to Alzheimer’s
disease. The mevalonate pathway is the main way that cholesterol is produced in
astrocytes. Once cholesterol is made it is associated with apoE containing lipoproteins
with the help of LPL, which can be secreted by interacting with transporters like ABCA1.
Uptake of lipoproteins by neurons occurs through receptors such as LDL-R and SORL1.
A buildup of cholesterol is thought to increase the production of soluble amyloid-beta,
which is one of the hallmarks of Alzheimer’s disease. Inside the neuron, cholesterol is
converted to 24-OHC that can cross the BBB, and also serves as a ligand for LXR and
PXR.
apoE
cholesterol 24-OHC
LDL-R/
SORL1
Alzheimer’s
Disease
cholesterol
apoE
ABCA1
mevalonate
pathway
BBB
astrocyte
24-OHC
apoE
Aβ
APP
24-OHC
gamma-secretase
LXRβ!
PXR
neuron
Brain
LPL

11

1.5) Specific Aims
Cholesterol homeostasis in the brain is rather perplexing and not well understood. An
increase in cholesterol is believed to promote Aβ, but AD is associated with a decrease in
total cholesterol levels and increased Aβ levels (Kang & Rivest, 2012). LXR and PXR
agonists have proved therapeutic in mouse models of AD, reducing Aβ and improving
memory, implying that increased cholesterol flux is beneficial in treating the disease
(Hong & Tontonoz, 2014). Cholesterol, Aβ, and AD are interconnected, thus the aim of
my research is to better understand the role of cholesterol in the brain and how it is
related to AD. Allo is known to decrease AD pathology by helping cholesterol
homeostasis, but the mechanism by which it is doing so is unclear. We sought out to
discover these mechanisms and how they relate to LXR and PXR receptors as well as
their target genes.
Figure 2. Overview of specific aims. Three different studies were implemented to
better understand Allo’s role in cholesterol homeostasis.
Allopregnanolone*
Docking(Study:(Can(Allo(
bind(to(LXR(and(PXR?(
Western(Blot(Study:(Does(
Allo(alter(protein(levels(of(
LXR(and(PXR(gene(
products?(
Gene(Array(Study:(What(is(
the(mechanism(by(which(
Allo(promotes(cholesterol(
homeostasis(and(amyloidE
beta(clearance?(

12

CHAPTER 2
In Silico Analyses to Determine if Allopregnanolone Can Bind to LXR and PXR
1. Rationale
Since the Brinton lab revealed that Allo increased expression of LXR and PXR proteins,
we first wanted to see if Allo has the ability to bind to these proteins. If Allo can activate
each receptor then desired protein products can be induced to decrease AD morphology.
To see if Allo has the ability to bind to each receptor a molecular docking study was
performed. Former studies have concluded that Allo is a ligand for PXR and downstream
PXR target genes are increased when given to mice (Mellon et al., 2008). Thirteen
ligands were selected to dock to each receptor, which included analogues of Allo,
cholesterol, and synthetic and natural agonists and antagonists (Table 1). It has been
shown that LXR ligands frequently activate PXR since they are considered promiscuous
nuclear hormone receptors with large binding pockets (Chen et al., 2011). From this
rational the same ligands were selected to bind to each receptor.

13

Table 1. Structure of ligands for docking study. Cholesterol derivatives include 24-S-
hydroxycholesterol and 27-hydroxycholesterol. Ganaxolone is an Allopregnanolone
derivative. Estradiol, Progesterone, and Pregnenolone are all natural steroid hormones.
T0901317 and SR12813 are synthetic compounds. Fluconazole, Clotrimazole, and St.
John’s Wort are clinically used and known to bind to each receptor.

1. Allopregnanolone 8. T0901317
2. Cholesterol 9. 27-hydroxy-
cholesterol

3. 24-S-
hydroxycholesterol
10. Fluconazole

4. Ganaxolone 11. Clotrimazole

5. Estradiol ! 12. SR12813
!
!
6. Progesterone 13. St John’s Wort

!
7. Pregnenolone
!

14

2. Methods
Crystal structures were downloaded from the Protein Data Bank (PDB) for PXR/RXRα,
LXRβ, and LXRα/RXRβ. Crystal Structure of the apo-PXR/RXRα LBD Heterotetramer
Complex (PDB: 4J5W) (Wallace et al., 2013). Crystal structure of the human LXRβ
ligand binding domain in complex with a synthetic agonist T0901317 (PDB: 1UPW)
(Hoerer et al., 2003). Crystal structure of the LXRα/RXRβ LBD heterodimer (PDB:
1UHL) (Svensson et al., 2003). Ligand structures were downloaded from PubChem.

Docking calculations were carried out using DockingServer (Bikadi & Hazai, 2009). The
MMFF94 force field (Halgren, 1998) was used for energy minimization of ligand
molecules using DockingServer. Gasteiger partial charges were added to the ligand
atoms. Non-polar hydrogen atoms were merged, and rotatable bonds were defined.
Docking calculations were carried out on ligand protein model. Essential hydrogen
atoms, Kollman united atom type charges, and solvation parameters were added with the
aid of AutoDock tools (Morris et al., 1998). Affinity (grid) maps of 20x20x20 Å grid
points and 0.375 Å spacing were generated using the Autogrid program (Morris et al.,
1998). AutoDock parameter set- and distance-dependent dielectric functions were used in
the calculation of the van der Waals and the electrostatic terms, respectively. Docking
simulations were performed using the Lamarckian genetic algorithm (LGA) and the Solis
& Wets local search method (Solis & Wets, 1981). Initial position, orientation, and
torsions of the ligand molecules were set randomly. All rotatable torsions were released
during docking. Each docking experiment was derived from 100 different runs that were
set to terminate after a maximum of 2,500,000 energy evaluations. The population size

15

was set to 150. During the search, a translational step of 0.2 Å, and quaternion and
torsion steps of 5 were applied.

16

3. Results
LXRβ was used in the majority of this study due to the fact that it is located throughout
the brain. Hoerer et al. determined the three-dimensional structure of the human LXRβ
LBD in complex with T0901317, a synthetic LXR-selective agonist (Figure 3a) (Hoerer
et al., 2003). The LBD (830 Å
3
) is rather large compared to the classic steroid hormone
receptors (420–550 Å
3
) (Williams et al., 2003). The most important interactions in the
binding pocket are formed between the ligand and His435 and Trp457 (Williams et al.,
2003). This is referred to as the “His-Trp switch” which is required to stimulate the
nuclear receptor, and mutations in either residue result in no response from agonists
(Williams et al., 2003; Svensson et al., 2003). To ensure the validity of this experiment
only docking results that had the ligand binding to both key residues were used.
Additionally, eleven of the ligands were docked to LXRα/RXRβ crystalized by Svensson
et al., to compare the results with the more prominent isoform in the brain (Svensson et
al., 2003). The LXRα/RXRβ combination was the only one from which crystals were
obtained even though a number of LXR–RXR heterodimers were screened (Svensson et
al., 2003). The LXRα isoform also has a critical “His-Trp switch” between the ligand and
His421 and Trp443 (Svensson et al., 2003).

First, we docked T0901317 and compared the results to the crystal structure of LXRβ
with T0901317 in the binding pocket (Hoerer et al., 2003). The docked ligand from our
run (blue) docked in a similar manner to the paper (red), and the interactions with His435
and Trp457 were maintained (Figure 3c). The LXRβ LBD is able to accommodate many
different ligand conformations due to its large size and flexibility of substrates. The

17

discrepancy in the alignment of the docking results is due to the large pocket. The key
interactions were maintained; therefore it is proposed that the ligand is able to activate the
receptor. Since T0901317 docked in a correct conformation we continued with the rest of
the ligands.

18

Binding

pocket
a.
LXRβ

Binding

pocket
b.
PXR/RXRα

Figure 3. Crystal structures of LXRβ and PXR/RXRα. A. Crystal
structure of LXRβ (PDB: 1UPW). B. Crystal structure of PXR/RXRα
(PDB: 4J5W). C. Comparison of experimentally observed binding
mode of T0901317 in the ligand binding pocket of LXRβ from Hoerer
et al. (red), and Docking Server solution (blue). All of the key
interactions are maintained. Images created with PyMOL.
c.

19

Wallace et al. crystalized the human PXR/RXRα LBD heterotetramer complex, which
has a 1294 Å
3
LBD (Figure 3b) (Wallace et al., 2013). PXR can bind a diverse set of
chemicals due to the unique composition of its ligand pocket, which also allows it to
permit a single ligand to dock in multiple orientations (Watkins et al., 2001). PXR has
also been shown to bind endogenous ligands of the estrogen receptor (ER) in a similar
manner, and the two LBDs share 16% sequence identity (Xue et al., 2007). There are
many critical polar residues in the binding pocket that form interactions with ligands.
These residues include: Met243, Ser247, Gln285, Trp299, His407, and Arg410 (Watkins
et al., 2001; Timsit & Negishi, 2007). Xue et al. concluded “all compounds capable of
maintaining contacts to Ser247 and the Arg410 residues were established PXR agonists”
(Xue et al., 2007). Taken this into consideration all of the ligands docked made key
hydrogen bonds to Ser247 and His407, which stabilized the ligands in the large flexible
pocket. Furthermore, all of the ligands made at least two or more other important
interactions and half of the ligands made all six key interfaces.

The established inhibition constant (Ki) was measured for each ligand in μM, which is
the concentration of ligand required to decrease the maximal rate of reaction to half
(Figure 4 & Table 2). The smaller the Ki correlates with a greater binding affinity and the
smaller amount of ligand to inhibit or activate the receptor. The Ki for Allo (LXRβ 0.12
μM, PXR 1.17 μM) was comparable to all of the natural ligands for each receptor
indicating that the configuration and composition is compatible for each of the receptors.
Interestingly, the synthetic agonists T0901317 and SR12813 bound best to LXR.
T0901317 is known as a potent PXR ligand, but it is used in vivo to activate both nuclear

20

receptors. 24OHC, a natural LXR ligand, bound the best to LXRβ (0.0054 μM), which is
consistent with other studies attesting it as a potent activator (Wang et al., 2002; Janowski
et al., 1999). The Ki for 24OHC is reported to be 0.1 μM for LXRβ and 0.11 μM for
LXRα (Janowski et al., 1999). We found them to be 0.0054 μM (LXRβ) and 0.56 μM
(LXRα), which are comparable and makes the study effective. Progesterone (0.22 μM)
and cholesterol (0.23 μM) bound the strongest to PXR/RXRα. This is interesting since
cholesterol derivatives are natural LXR agonists but nevertheless estrogen derivatives are
derived from cholesterol. In that progesterone and cholesterol have the same steroid
structure backbone with three 6-sided carbon rings and one 5-sided carbon ring, it was
not surprising we found that they ended up binding in a similar configuration with
analogous binding strengths. Unexpectedly, the natural ligand that activates both
receptors, St. John’s Wort, ended up having the poorest binding affinity to either
receptor. This could be due to its large molecular structure that because of size was
unable to achieve enough robust interactions as compared to the other smaller ligands.
Altogether the ligands preferred LXR, which could be due to the fact that the LBD of
PXR is larger and the ligands are capable of making stronger interactions in the smaller
binding pocket.

The established free energy of binding (kcal/mol) was also calculated for each docking
experiment (Table 2). This measure describes the protein-ligand association. A large and
negative value accounts for a favorable interaction. All of the docking trials ranged from
-5 kcal/mol to -11 kcal/mol signifying a favorable binding interaction for each protein-
ligand association. The most favorable interaction occurred between 24OHC and LXRβ

21

(-11.28 kcal/mol), which also correlates with the highest binding affinity. Fluconazole (-6
kcal/mol) and St. John’s Wort (-5 kcal/mol) were the least favorable interactions, most
likely due to their complicated structures.

22

Allopregnanolone
Cholesterol
24-S-hydroxycholesterol
Ganaxolone
Estradiol
Progesterone
Pregnenolone
T0901317
27-hydroxycholesterol
Fluconazole
Clotrimazole
SR12813
St. John's Wort
0
1
2
3
100
200
Inhibition Constant, Ki (μM)
Binding Affinity
PXR/RXRα
LXRβ
LXRα/RXRβ
Figure 4. Binding affinity (Ki) for each ligand to PXR and LXR. Binding
affinities for 13 ligands to PXR/RXRα and LXRβ are shown. Only 11 ligands were
docked to LXRα/RXRβ. Best docking run was based on proper configuration with
key interactions being maintained.
Table 2. Inhibition Constant (μM) and established free energy of binding (kcal/mol) for each
ligand to PXR and LXR. A smaller Ki represents a stronger binding affinity, and a large negative
free energy of biding signifies a favorable interaction.
Ki (μM)
Est. Free Energy
of Binding
(kcal/mol) Ki (μM)
Est. Free Energy
of Binding
(kcal/mol) Ki (μM)
Est. Free Energy
of Binding
(kcal/mol)
Allopregnanolone 1.17 -8.49 0.11985 -9.44 7.86 -6.96
Cholesterol 0.23565 -10.97 0.00659 -11.16 1.16 -8.1
24-S-hydroxycholesterol 1.01 -10.56 0.0054 -11.28 0.56393 -8.52
Ganaxolone 0.49451 -8.52 0.07054 -9.76 1.27 -8.04
Estradiol 8.71 -8.45 1.34 -8.01 1.16 -8.1
Progesterone 0.22451 -9.33 0.09436 -9.58 0.04837 -9.98
Pregnenolone 0.93678 -8.87 0.08435 -9.65 0.50343 -8.59
T0901317 284.96 -8.33 0.09922 -9.55 0.10414 -9.53
27-hydroxycholesterol 0.74259 -10.87 0.0109 -10.86 0.34321 -8.82
Fluconazole 11.59 -6.85 10.48 -6.79 11.5 -6.74
Clotrimazole 0.48167 -8.52 0.06963 -9.76 1.41 -7.98
SR12813 19.13 -9.15 0.2335 -9.05 - -
St. John's Wort 83.3 -6.93 95.19 -5.49 - -
PXR/RXRα LXRβ LXRα/RXRβ
Ligands

23

4. Conclusion
From this experiment we can successfully conclude that Allo has the ability to bind to
LXR and PXR. The key interfaces were maintained in the LBD for each receptor. Allo
bound comparable if not better than the estrogen derivatives that naturally bind to PXR.
A similar correlation was observed when comparing the binding affinity of Allo and the
natural cholesterol derivatives to LXR (Table 2). Since the binding cavity for both
receptors are so massive, each ligand could take on multiple configurations (Figure 3).
Allo was shown to bind with a higher affinity to LXRβ (0.12 μM) as compared to LXRα
(7.86 μM) (Table 2). This is significant since the β isoform is primarily expressed in the
brain, and the α isoform is located throughout the body. In that AD is characterized by
deficits in the brain, these findings suggest that targeting LXRβ is more beneficial in the
cholesterol homeostasis process. We found that Allo bound best to LXRβ as indicated by
the significantly lower Ki, suggesting that the neurosteroid has preference for the brain
specific isoform. Additionally, Allo showed a slight preference for LXRβ over PXR.
Although both receptors play important roles in the brain, LXR gene products seem to be
more related to AD processes and thus should be considered further for targets of
pharmacology development for AD.

24

CHAPTER 3
LXR and PXR Gene Products Are Altered With Allopregnanolone Administration
1. Rationale
As presented above, the data indicates that Allo has the ability to bind to LXR and PXR.
We next sought out to see if the protein levels of the gene products were altered in vivo.
We hypothesized that since Allo has the ability to bind to LXR/PXR that it should be able
to activate each receptor. This is important because we know that activation of each
receptor leads to an increase or decrease in many target genes, which can be transcribed
and ultimately translated into new proteins. We also know that Allo has a long beneficial
effect and a short half-life, suggesting that it is doing considerably more than just binding
to GABA
A
receptors. Altering protein expression has a more profound effect by
modifying a variety of signaling cascades in the brain.

25

2. Methods
a. Transgenic Mice
Colonies of 3xTgAD (homozygous mutant of human APPswe, tauP301L, and
PS1M146V) were bred and maintained at the University of Southern
California (Los Angeles, CA) following National Institutes of Health
guidelines on use of laboratory animals, and an approved protocol by the
University of Southern California Institutional Animal Care and Use
Committee. Mice were group-housed 2–5 per cage on 12-h light/dark cycles
and provided ad libitum access to food and water. The 3xTgAD mice
progressively develop AB plaques and tau hyperphosphorylation mainly in the
hippocampus, amygdala, and cortex, which is consistent with human AD
pathology. The mice were genotyped regularly to confirm the purity of the
colony. To ensure the stability of AD-like phenotype in the 3xTgAD mouse
colony, we performed routine immunohistological assays every three to four
generations.
b. Chemicals
Allopregnanolone (APα 3α-hydroxy-5α-pregnan-20-one) (aka AP, Allo or
THP) used in this study was provided by Dr. M.A. Rogawski (University of
California, Davis). (2-hydroxypropyl)-beta-cyclodextrin (HBCD) was
obtained from Cyclodextrin Technologies Development, Inc. (High Springs,
FL). Sulfobutylether-beta-cyclodextrin (SBECD) obtained from CycloLab
Cyclodextrin Research and Development Laboratory, Ltd (Budapest,
Hungary) was used for the intramuscular mouse efficacy study.

26

c. Drug Preparation
Male mice were randomly assigned to treatment groups by an age and body
weight stratification procedure. Allo was dissolved in 20%w/v HBCD solution
at 2.5 mg/ml by brief sonication and was subcutaneously (SC) injected to
mice at 1 mg/kg (0.5 mg/ml) and 10 mg/kg (5 mg/ml). As a positive control,
SC 10 mg/kg 2.5 mg/ml; PBS/5%EtOH was administered as a suspension
formulation. Additionally, Allo was dissolved in 6%w/v SBECD solution at
0.5 mg/ml and injected intramuscularly (IM) to mice at 2 mg/kg (1.5 mg/ml),
and 10 mg/kg (2.5 mg/ml) 5% EtOH. Vehicle groups received 20% HBCD in
0.9% NaCl or 6% SBECD. Each treatment group received equal volume of 50
microliter/25g body weight. The thymidine analogue, 5-Bromo-2’-
deoxyuridine (BrdU), incorporated into newly synthesized DNA of replicating
cells during the S-Phase of the cell cycle, was dissolved in PBS and
intraperitoneally injected (IP) at 100 mg/kg 1 h following Allo treatment.
Male 3xTgAD mice 5-7 months old were used with 6-8 mice per treatment
group for the SC injection. 2.5-7 month old 3xTgAD mice were used for the
IM injection.
d. Treatment Paradigms
Paradigm 1
24 hr I.M.
50 microliter/25g body weight
2.5-7 month old male
3xTgAD mice
BrdU Used
Treatment Abbreviation
AP 0 mg/kg I.M. 6% SBECD Veh
AP 2 mg/kg I.M. 6% SBECD AP2
AP 10 mg/kg I.M. PBS/5% EtOH AP10

27

Paradigm 2
24 hr S.C.
50 microliter/25g body weight
5-7 month old male
3xTgAD mice
BrdU Used
Treatment Abbreviation
AP 0 mg/kg S.C. 20% HBCD Veh
AP 10 mg/kg S.C. 20% HBCD AP1
AP 10 mg/kg S.C. 20% HBCD AP10
AP 10 mg/kg S.C. suspension PBS/5% EtOH AP10susp

e. Animal Dissection and Tissue Collection
Mice were euthanized 24 hours after Allo administration. Mice were
anesthetized with 100 mg/kg ketamine and 10 mg/kg xylazine and euthanized
by cervical dislocation. The mice were perfused with pre-chilled PBS and
brains were immediately collected, dissected, and stored at -80°C for
biochemical analysis
f. Western Blotting
Protein was extracted from the cortex and hippocampus using T-per (Pierce)
under reducing and denaturing conditions. Protein concentrations were
determined by DC protein assay kit (Bio-Rad). Equal amounts of protein (20-
30 μg depending on the protein of interest) were separated by SDS-Page on a
12% Criterion TGX (Bio-Rad), and transferred to 0.2 μm PVDF membrane
(Bio-Rad). Nonspecific binding sites were blocked with blocking buffer (5%
nonfat milk in Tris-buffered saline, TBS, containing 0.1% Tween-20). After
blocking, primary antibodies were incubated with membrane in Superblock
(Thermo) overnight at 4°C. The following antibodies were used in this study:
Rabbit polyclonal anti-Cyp46a1 (1:1000, Pierce), Rabbit polyclonal anti-

28

LXRβ (1:500, Lifetechnologies), Goat polyclonal anti-LDLR (1:500, R&D
Systems), Mouse monoclonal anti-LR11/SORLA (1:500, BD Biosciences),
Mouse monoclonal anti-NFκB (1:1000, Novus), Mouse monoclonal anti-
ABCA1 (1:1000, Abcam). The membranes were then incubated with a
horseradish peroxidase-conjugated goat anti-rabbit, horse anti-goat, or horse
anti-mouse secondary antibody (1:5000-10,000) complementary to the
primary antibody for 1 hr. Blots were developed with the SuperSignal West
Pico and Femto Substrate (Thermo). Results were visualized with the
Chemidoc System (Bio-Rad). Quantitative analyses of optical band intensity
were performed by Un-Scan-It software. Protein expression was normalized
by loading house keeping protein β-actin (1:3000, Millipore). All comparisons
were made within blots.
g. Statistical Analysis
Statistical significance for group comparison was performed by a Student's t-
Test. The difference between groups was considered significant when the P-
value was <0.05.

29

3. Results
The triple transgenic AD mouse model (3xTgAD) progressively develops plaques and
tangles consistent with the amyloid cascade hypothesis. These changes have been shown
to appear mainly in the hippocampus and cortex, which is consistent with AD pathology
(Oddo et al., 2003). There are numerous genes affected by LXR and PXR activation. The
target genes of LXR are those involved in cholesterol transport, efflux, metabolism,
synthesis, and the inhibition of inflammation (Hoerer et al., 2003; Zelcer et al., 2007;
Kang & Rivest, 2012). In LXR double knockout mice it has been shown that several
LXR target genes (such as ABCA1, LDL-R, and HMGCR) are significantly decreased in
their brains (Wang et al., 2002). Additionally, deletion of either LXR isoform in a
transgenic mouse model of AD led to a marked increase in Aβ and AD-like pathology
(Hong & Tontonoz, 2014). PXR gene products on the other hand activate drug and
steroid metabolizing enzymes and receptors as well as those involved in protective
cholesterol metabolism and processing (Lamba et al., 2004; Xue et al., 2007). It was
previously confirmed that there was an increase in LXR and PXR expression in 3xTgAD
mice injected with Allo once a week for 6 months (Chen et al., 2011). Allo has an
inverted U-shaped dose response profile meaning too high or low of a dose leads to non-
ideal reactions (Irwin & Brinton, 2014). The dose of Allo 1mg/kg and 10mg/kg were
selected as optimal based on previous findings of neurogenesis (Irwin & Brinton, 2014).
We wanted to look at the acute effects of Allo administration in brain samples from mice
injected subcutaneously (SC) and intramuscularly (IM) once and euthanized 24 hours
later. Different routes of administration and doses were used for this study. We chose to
use samples from the cortex and hippocampus since these are the specific brain regions

30

that are most vulnerable to AD pathology (Chen et al., 2011). First we confirmed that
there was a significant increase in LXRβ protein expression in the hippocampus of the
SC injection samples (Figure 5a). This further validated the findings of Chen et al., and
showed that Allo has beneficial effects after a short 24 hour exposure time. Additionally,
the protein was not detected in the liver further validating that the isoform is brain
specific.

Next we looked at the protein expression of several known target genes that are affected
by LXR/PXR activation. In the IM samples we looked at ABCA1 and the nuclear factor
kappa-light-chain-enhancer of activated B cells (NF-κB). In the SC injection paradigm
we looked at the protein expression of LDL-R, sortilin-related receptor (SORL1), and
Cyp46a1.

The protein LDL-R functions to regulate the removal of lipoprotein particles from the
blood by endocytosis (Kang & Rivest, 2012). APOE binds to this receptor and plays a
role in cholesterol uptake (Figure 1). Overexpression of LDL-R in the brain has been
shown to decrease Aβ accumulation and increase its clearance in mice (Kang & Rivest,
2012). It is believed that LDL-R assisting in removal of extracellular APOE leads to an
increase in Aβ clearance across the BBB (Castellano et al., 2012). LXR knockout mice
have a significant decrease in LDL-R further validating that LDL-R is a LXR target gene
(Wang et al., 2002). Additionally, in AD brains the defective transport of cholesterol
could be due to low levels of LDL-R (Wang et al., 2002). Allo administration
considerably increased the expression of LDL-R in the 10mg/kg group in the cortex

31

(Figure 5b). Thus Allo has another beneficial effect on AD pathology via enhancement of
Aβ and cholesterol clearance by increasing LDL-R protein expression (Wollmer, 2010).

SORL1, also known as SORLA, is a neuronal sorting receptor for APP that is measured
in neurons of the cortex and hippocampus. SORL1 is a member of the LDL-R family of
receptors (Castellano et al., 2012). The receptor functions in interacting with APP
molecules and preventing the movement of the molecules into cellular compartments
where secretases reside. As stated prior, Aβ is generated when APP is cleaved by
gamma-secretases. SORL1 can also bind Aβ and promote its catabolism. Additionally, it
functions in serving as an APOE receptor and mediating endocytotic uptake of
lipoproteins (Wollmer, 2010) (Figure 1). It has been shown that some individuals with
sporadic AD have reduced expression of this receptor in the brain, showing that it is
causally implicated in the disease. Additionally SORL1 mutations are considered a major
risk factor for AD (Caglayan et al., 2014). In a mouse model with an overexpression of
SORL1, there was a significant decrease in Aβ concentrations in the mouse brain
(Caglayan et al., 2014). In the cortex, SORL1 was significantly increased in all Allo
treatment groups compared to the vehicle from the SC samples (Figure 5c). As a negative
control, SORL1 was not expressed in the liver validating that it is brain specific and that
the antibody is indeed SORL1 specific. Treatment with Allo increased the expression of
SORL1, which could be one of the mechanisms by which the neurosteroid reduces Aβ
formed in AD.

32

ABCA1 is a major regulator of cholesterol efflux and Aβ transport and clearance, and its
transcriptional activation is controlled by LXR (Kang & Rivest, 2012; Koldamova et al.,
2005). It has been shown that a lack of ABCA1 promotes amyloid deposition and thus it
is beneficial to increase this protein when treating AD. Studies have shown that with the
administration of T0901317, ABCA1 protein levels increase and Aβ levels decrease in
mice on a high fat diet or transgenic AD mice (Kang & Rivest, 2012; Koldamova et al.,
2005). Additionally, polymorphisms in this gene are associated with an increased risk for
AD, further showing how essential it is (Zelcer et al., 2007). Our data showed a
statistically significant increase of ABCA1 in the 2mg/kg IM Allo treated group in the
hippocampus compared to vehicle (Figure 5d). ABCA1 is a very imperative protein in
cholesterol efflux since it transports APOE and Aβ. This could explain why Allo has the
capability to decrease AD pathology.

As presented above, Cyp46a1 is a key enzyme that converts excess cholesterol into
24OHC, which can cross the BBB (Kang & Rivest, 2012). The extra hydroxyl group on
24OHC promotes its movement across the BBB to facilitate cholesterol efflux since
cholesterol can not leave (Hong & Tontonoz, 2014). 24OHC is also a ligand to LXR and
serves as a precursor to other oxysterols, which are also ligands to LXR (Figure 6).
Neuronal overexpression of Cyp46a1 has shown to reduce Aβ pathology in mouse
models of AD (Kang & Rivest, 2012). Furthermore, polymorphisms in Cyp46a1 are
associated with increased risk for AD (Zelcer et al., 2007). Our data has shown that Allo
increases expression of Cyp46a1 in the cortex of all treatment groups compared to the
vehicle in the SC paradigm (Figure 5e). Increasing the expression of Cyp46a1 is

33

beneficial to treat AD by removing Aβ and activating LXR pathways to further decrease
deficits caused by the disease.

AD is associated with microglia-mediated inflammation and therefore it is beneficial to
target neuroinflammation when treating the disease (Kang & Rivest, 2012). Elevated
levels of inflammatory mediators in AD brains are proposed to contribute to neuronal
loss (Zelcer et al., 2007). A study demonstrated that when the LXR agonist GW3965 was
administered to rats following global cerebral ischemia, NF-κB and its target gene COX-
2 were decreased in the hippocampus to reduce neural damage (Cheng, Ostrowski, Liu, &
Zhang, 2010). NF-κB is known to be repressed by LXR activation, and is consequently a
target gene of the receptor (Calkin & Tontonoz, 2010). Our data showed that in the
hippocampus Allo notably decreased NF-κB expression in the 10mg/kg group from the
IM paradigm (Figure 5f). NF-κB is considered one of the main regulators of
inflammatory processes hence this pathway is a beneficial target in lowering the
inflammation caused by the disease.

34

Figure 5. Allopregnanolone increased LXRβ, LDL-R, SORL1, ABCA1, Cyp46a1,
and decreased NF-κB expression in 3xTgAD male mice. Western blot images from
hippocampus and cortex samples. Bars represent mean relative expression ± SEM. *
p<0.05 and ** p<0.01 compared to vehicle control group.
Vehicle 20% HBCD
Allo 1mg/kg SC 20% HBCD
Allo 10mg/kg SC 20% HBCD
Allo 10mg/kg SC PBS/5% EtOH
0
50
100
150
200
24 hr
LXRb/Actin protein expression
(% vs Vehicle)
LXRb/Actin Increase
Hippocampus
*
*
**
LXRb
Actin 43kDa
52kDa
Liver
a.
Vehicle 20% HBCD
Allo 1mg/kg SC 20% HBCD
Allo 10mg/kg SC 20% HBCD
Allo 10mg/kg SC PBS/5% EtOH
0
50
100
150
200
250
SORL1/Actin Increase
Cortex
24 hr
SORL1/Actin protein expression
(% vs Vehicle)
*
**
*
SORL
Actin 43kDa
250kDa
Liver
c.
Vehicle 20% HBCD
Allo 1mg/kg SC 20% HBCD
Allo 10mg/kg SC 20% HBCD
Allo 10mg/kg SC PBS/5% EtOH
0
50
100
150
200
24 hr
Cyp46a1/Actin protein expression
(% vs Vehicle)
Cyp46a1/Actin Increase
Cortex
**
**
*
Cyp46a1
Actin 43kDa
52kDa
e.
Vehicle 20% HBCD
Allo 1mg/kg SC 20% HBCD
Allo 10mg/kg SC 20% HBCD
Allo 10mg/kg SC PBS/5% EtOH
0
50
100
150
200
LDL-R/Actin Increase
Cortex
24 hr
LDL-R/Actin protein expression
(% vs Vehicle)
**
*
LDL-R
Actin 43kDa
95kDa
b.
Vehicle: 6% Dexolve
Allo 2mg/kg: 1.5mg/ml, 6% Dexolve
Allo 10mg/kg: Allo 2.5mg/ml, 5% EtOH
0
50
100
150
200
ABCA1/Actin Increase
Hippocampus
24 hr
ABCA1/Actin protein expression
(% vs Vehicle)
ABCA1
Actin 43kDa
254kDa
*
d.
Vehicle: 6% Dexolve
Allo 2mg/kg: 1.5mg/ml, 6% Dexolve
Allo 10mg/kg: Allo 2.5mg/ml, 5% EtOH
0
50
100
150
NFKB/Actin Decrease
Hippocampus
24 hr
NFKB/Actin protein expression
(% vs Vehicle)
*
NFKB
Actin 43kDa
54kDa
f.

35

Figure 6. Actions of the enzyme Cyp46a1 in the brain.
Cyp46a1 converts cholesterol into the oxysterol 24OHC that
can cross the BBB and activate LXR. Activation of LXR
results in the gene transcription of various proteins involved in
cholesterol homeostasis. Adapted from
“Hypercholesterolaemia-induced oxidative stress at the blood-
brain barrier, ” by I. Dias, M. Polidori, and H. Griffiths, 2014,
Biochemical Society Transactions, 42(4), p. 1003. Copyright
2014 by the Biochemical Society.

36

4. Conclusion
We have demonstrated that Allo alters multiple proteins involved in cholesterol
homeostasis, inflammation, and products of LXR activation. Allo treatment reduced
amounts of Aβ in 3xTgAD mouse brains (Chen et al., 2011). The mechanism by which
Allo reduces Aβ could be through increased expression of LDL-R, SORL1, ABCA1, and
Cyp46a1. These proteins promote cholesterol efflux and transport both of which are
compromised in AD. SNPs in all of these proteins are associated with increased risk for
AD (Zelcer et al., 2007; Caglayan et al., 2014; Vance, 2012). ABCA1 increases
cholesterol efflux by coordinating the mobilization of cholesterol to combine with APOE
(Leduc et al., 2010). APOE then delivers its lipid cargo to target cells by binding to LDL-
R and SORL1 (Figure 1, Figure 11). Once taken up by cells, cholesterol can be broken
down by Cyp46a1 to cross the BBB (Figure 6). Additionally, LDL-R, SORL1, and
ABCA1 promote mechanisms of Aβ degradation and clearance. The lipidation of APOE
by ABCA1 stimulates degradation of Aβ (Fan et al., 2015). Aβ phagocytosis and
clearance is dependent on APOE (Jakobsson et al., 2012). APOE is a known direct target
of LXR, and a specific variant in the protein is the largest known genetic risk factor for
AD (Kang & Rivest, 2012; Jakobsson et al., 2012). LDL-R and SORL1 affect the
accumulation of Aβ by regulating the trafficking and processing of APP (Castellano et
al., 2012). Collectively, increased expression of ABCA1, APOE, LDL-R and SORL1
support the hypothesis that Allo treatment promotes the clearance of cholesterol and Aβ.
Individuals with the APOE4 allele have impaired Aβ clearance and subsequently a higher
risk of getting AD (Fan et al., 2015). Allo activation of LXR provides a plausible

37

therapeutic strategy to increase Aβ degradation and clearance in these individuals by
utilizing the functions of ABCA1 and LDL-R to remove Aβ (Figure 11).

Inflammation plays a major role in AD, and it has been proposed to give patients anti-
inflammatory drugs to alleviate some of the problems. Inflammation during the disease
causes neurodegeneration, breakdown of the BBB, reduction in phagocytic activity, and
an increase in Aβ. A growth in inflammatory cytokines in the AD brain causes Aβ to
accumulate at a faster rate than the cells can remove the toxic Aβ product. Previous
studies have shown that Allo treatment reduces inflammatory cytokines after traumatic
brain injury, however the mechanism of action is unclear (He, Evans, Hoffman, Oyesiku,
& Stein, 2004). NF-κB controls many genes involved in inflammation, and is found in
the nervous system. LXR activation inhibits NF-κB, and our data revealed that Allo
reduces this protein. The mechanism by which Allo lowers inflammation could be
through preventing NF-κB from producing cytokines.

There are many other targets genes of LXR activation, which we did not show in this
study, that could be beneficial towards treating AD. The enzyme LPL is necessary in the
formation of functional APOE particles, and it is also a LXR target gene (Figure 11)
(Leduc et al., 2010; Dias, Polidori, & Griffiths, 2014). LXR activation results in
inhibition of many other inflammatory genes in addition to NF-κB. Others include
CXCL10, Ccl5, and Il1b (Zelcer et al., 2007). Inhibition results in a reduction of
neuroinflammation, enhances phagocytic activity of microglia, and increases Aβ
clearance (Irwin & Brinton, 2014). LXR signaling affects the development of AD

38

pathology through multiple pathways that promote cholesterol and Aβ efflux. Allo
activation of LXR and its subsequent target genes is a valuable way to reduce some of the
pathology caused by the disease.

39

CHAPTER 4
Allopregnanolone Regulation of Cholesterol Homeostasis and Aβ Clearance
Pathways: Gene Expression Analyses
1. Rationale
Since we know that Allo has many roles in the brain we wanted to investigate other
pathways involved. We were interested in determining more about the mechanism of
action of Allo as it pertains to treating AD and more specifically cholesterol homeostasis.
A gene array was performed on the SC paradigm cortex samples to investigate what other
genes are modified when mice are treated with Allo. Since our data has demonstrated that
increased cholesterol flux is beneficial in AD the Lipoprotein Signaling and Cholesterol
Metabolism array was performed to further identify key players. Genes involved in
cholesterol metabolism, transport, and lipoprotein receptors were analyzed.

40

2. Methods
a. Gene Array
Total RNA was isolated from mouse cortex tissues using the PureLink RNA
Mini Kit (Ambion). RNA was qualified using a nanodrop machine. RNA was
converted to cDNA using the RT² First Strand Kit (Qiagen) on a MyCycler
Thermal Cycler (Bio-Rad). Real-Time PCR for RT² Profiler Array format
384-well (4 × 96) Plate for mouse lipoprotein signaling and cholesterol
metabolism (Qiagen) was ran for 4 samples per 4 treatment groups (Qiagen).
Fluorescence was detected on an ABI 7900HT Fast Real-Time PCR System
equipped with the Sequence Detection System Software Version 2.3 (Applied
Biosystems). Relative gene expression levels or fold changes relative to the
vehicle group were calculated by the comparative Ct (ΔΔCt) method, with Ct
denoting threshold cycle (Schmittgen and Livak, 2008). Data was analyzed
using RT² Profiler PCR Array Data Analysis version 3.5. Housekeeping gene
normalization was selected automatically. Relative gene expression levels or
fold changes relative to vehicle group were calculated. Data from the third
sample in each group was discarded due to low expression, resulting in an n=3
for each treatment group. Fold change values greater than one indicate a
positive expression or up-regulation relative to the comparison group. Fold
change values less than one indicate a negative expression or down-regulation
relative to the comparison group.

41

b. Statistical Analysis
The difference between each group compared to vehicle was considered
significant when the P-value was <0.05. Boundary was set at 1.2 (20%
change).

42

3. Results
Genes that were altered in each treatment group that were statistically significant are
shown in Figures 7, 8, and 9. Genes that are involved in LDL receptors that were altered
when compared to vehicle include CXCL16, OLR1, LRP6, STAB2, and PCSK9.
Chemokine (C-X-C motif) ligand 16 (CXCL16) is produced in the brain endothelium in a
variety of cells, and is released upon shedding. CXCL16 upregulates its own receptor
(CXCR6) and induces the PI3K/Akt and Erk pathways. The stimulation of these
pathways results in activation of the transcription factor AP-1, increases cell
proliferation, stimulates cell migration, and helps in wound healing. Insulin has been
proven to be beneficial to treat AD, and it works by activating PI3K/Akt/Erk to assist in
regulation of Aβ production (Kang & Rivest, 2012). A former study determined that the
CXCL16-CXCR6 system recruits glial precursor cells to sites of brain damage to play a
crucial role in brain development and homeostasis “as well as in pathological conditions
such as inflammation, neuronal degeneration, ischemia or malignant transformation”
(Hattermann, Ludwig, Gieselmann, Held-Feindt, & Mentlein, 2008). Therefore, this
chemokine could be beneficial when released in the brain to help with damage caused by
AD. Comparing the Allo 10mg/kg group to vehicle, CXCL16 was significantly
upregulated (2.16 fold). Additionally, CXCL16 functions to uptake modified lipoproteins
in chronically inflamed tissue to assist in healing (Aslanian & Charo, 2006).

Oxidized LDL Receptor 1 (OLR1) is a cell surface receptor copiously identified in the
brain (Serpente et al., 2011). Activation by a variety of ligands has been shown to initiate
several transcription factors including NF-κB resulting in inflammation (Khaidakov et

43

al., 2011). Deletion of OLR1 in an atherosclerotic mouse model has shown a drastic
reduction of inflammation, and OLR1 has also been shown to be overexpressed in cancer
cells (Khaidakov et al., 2011). Furthermore, a single nucleotide polymorphism in the
OLR1 gene has been shown to be associated with AD (Serpente et al., 2011). The single
nucleotide polymorphism results in a deregulated micoRNA (miRNA) which contributes
to neurodegeneration (Serpente et al., 2011). The OLR1 gene was downregulated (-11.6
fold) and statistically significant when comparing Allo 1mg/kg to vehicle. It is beneficial
to downregulate this receptor to decrease inflammation via NF-κB, which has already
shown to be harmful, as well as protect the brain from oxidative stress forming oxidized
LDL.

Low-density lipoprotein receptor-related protein 6 (LRP6) is a coreceptor for the
canonical wingless (Wnt) pathway, which regulates diverse developmental processes in
the CNS such as neurogenesis and synaptic function (Liu et al., 2014). Activated LRP6
leads to protection of β-catenin from proteasomes, allowing it to travel to the nucleus and
regulate transcription of essential target genes involved in the previously stated processes
(Joiner, Ke, Zhong, Xu, & Williams, 2013). Loss of LRP6 has been shown in AD, which
results in a decrease in Wnt signaling and causes synaptic dysfunction, elevated Aβ, and
an accelerated AD progression (Liu et al., 2014). Restoring LRP6-mediated Wnt
signaling has been suggested as a viable strategy for AD therapy. LRP6 is related to
LRP1, which serves as a receptor for APOE and is involved in clearance of Aβ. The gene
for LRP6 was significantly upregulated (1.3 fold or 30%) in Allo treated 10mg/kg cortex

44

samples when compared to vehicle. It is beneficial to increase expression of this gene in
order to decrease Aβ and activate the advantageous Wnt signaling pathway.

Stabilin-2 (Stab2) is a phosphatidylserine (PS) receptor that functions in phagocytosis of
apoptotic cells and helps in the production of anti-inflammatory cytokines (Kim et al.,
2012). Removal of unwanted cells by phagocytosis is critical for cellular homeostasis and
to fight disease (Kim et al., 2012). Inefficient engulfment of apoptotic cells is detrimental
to the brain and is related to AD (Kim et al., 2012). Apoptosis occurs during the
progression of AD leading to cell death, and it is necessary to have an effective way to
remove these dead cells. Stab2 was upregulated (1.47 fold or 47%) in the Allo 10mg/kg
EtOH group compared to vehicle.

Proprotein convertase subtilisin/kexin type 9 (PCSK9) is involved in the degradation of
LDL-R and associated proteins plus cholesterol homeostasis (Kysenius, Muggalla,
Mätlik, Arumäe, & Huttunen, 2012). It was found to be significantly reduced when
comparing the Allo 1mg/kg group to vehicle (-1.8 fold or -80%) and Allo 10mg/kg to
vehicle (-1.6 fold or -60%). PCSK9 also functions in cell apoptosis, inflammation, and
tumor metastasis (Wu et al., 2014). It may promote neuronal apoptosis through NF-κB,
which we have previously shown to be decreased by Allo in a beneficial way. It is
favorable to downregulate this gene in order to increase LDL-R to lower circulating
lipoprotein and cholesterol levels as well as decrease inflammation and apoptosis. We
have shown earlier that Allo increases LDL-R protein expression in the cortex and it
could be doing so through this pathway in addition to LXR activation.

45

Angiopoietin-like 3 (ANGPTL3) regulates the clearance of circulating lipids and is also
involved in cholesterol homeostasis. ANGPTL3 inhibits LPL, which functions to uptake
lipoproteins (Figure 10), and therefore increases plasma lipids (Zou et al., 2009). This
gene was downregulated when treated with Allo 10mg/kg EtOH compared to vehicle (-
1.39 fold or -39%) to help reduce cholesterol levels in circulation and better manage
clearance. High cholesterol levels in circulation may further the progression of AD due to
the fact that aggregation leads to toxic effects. This result indicates that it is advantageous
for Allo to target this pathway to reduce cholesterol levels in circulation and help reverse
the negative effects of excessive cholesterol.

Genes that had modified expression that are involved in cholesterol biosynthesis are
HMGCS1, Cyp51, and Idi2. HMG-CoA synthase 1 (HMGCS1) is a key enzyme involved
in cholesterol synthesis in the mevalonate pathway (Goldsetin & Brown, 1990). This
pathway generates oxysterol LXR agonists and cholesterol. HMGSC1 was shown to be
significantly increased by Allo (1.2 fold or 20%) in the Allo 10mg/kg group compared to
vehicle. There are two different isoforms of the HMGSC enzyme. They are very similar
however, the cytosolic form (HMGCS1) is the starting point in the mevalonate pathway,
and the mitochondrial form (HMGCS2) is responsible for the biosynthesis of ketone
bodies (Goldsetin & Brown, 1990). Too much cholesterol in the CNS can be detrimental
and cause neurotoxicity in addition to tau hyperphosphorylation as seen in AD patients.
This key enzyme in cholesterol synthesis was only slightly upregulated by Allo showing
that you need some synthesis for brain function and healing but not so much as to cause
problems.

46

Cyp51 is the most common cytochrome P450 in the brain. It is the only cytochrome
involved in cholesterol synthesis, and plays an essential role in the synthesis of sterols
(Lepesheva & Waterman, 2004). The reaction that it catalyzes leads to the initial
substrate in steroid hormone biosynthesis from lanosterol (Lepesheva & Waterman,
2004). Cyp51 was slightly upregulated (1.2 fold or 20%) when comparing Allo 10mg/kg
to vehicle. Cholesterol synthesis is necessary in the brain to survive, and also to generate
oxysterols to activate LXR target genes to regulate cholesterol homeostasis.

Isopentenyl-diphosphate delta isomerase 2 (Idi2) catalyzes the isomerization of IPP to
DMAPP, an early step in steroid hormone and cholesterol synthesis (Clizbe, Owens,
Masuda, Shackelford, & Krisans, 2007). Isoprenoids and their derived compounds are
necessary for functioning in membrane structure and signal transduction (Clizbe et al.,
2007). Idi2 was significantly upregulated (1.9 fold or 90%) in the Allo 10mg/kg EtOH
group when compared to vehicle, highlighting how important cholesterol homeostasis
and biosynthesis is in the brain.

The last gene affected by Allo treatment is involved in cholesterol catabolism. Aldo-keto
reductase family 1, member D1 (AKR1D1) encodes the enzyme responsible for the
catalysis of steroid hormones carrying a delta(4)-3-one structure (Chen, Jin, & Penning,
2015). It generates active 5β-steroids such as progesterone, which are involved in
neurotransmission as well as initiating steroid clearance by conjugation (Chen et al.,
2015). Neurosteroids derived from this pathway also act as ligands for PXR. This enzyme
is mainly found in the liver and deficiency contributes to hepatic dysfunction (Chen &

47

Penning, 2014). AKR1D1 was significantly reduced (-8.66 fold) when comparing Allo
10mg/kg to vehicle. I think that increasing this enzyme would be beneficial to AD by
aiding in neurotransmission. However, since Allo itself is a neurosteroid there might be
potent feedback inhibition of the same pathway that may contribute to Allo generation.

48

Gene Fold Regulations P-Value
Olr1 -11.6062 0.008721
Pcsk9 -1.8064 0.001797
!
Olr1
Pcsk9
Allo
!Olr1 !NFKB
!Inflammation
and oxidative
stress
!PCSK9 "LDLR
"Cholesterol
Efflux
Allo 1mg/kg SC 20% HBCD vs. Vehicle 20% HBCD

Figure 7. Allopregnanolone decreases the gene expression of OLR1 and PCSK9 that
promote cholesterol efflux and reduce inflammation. Comparisons were made
between the Allo 1mg/kg SC 20% HBCD group and Vehicle 20% HBCD. Significant
genes shown with fold regulation and P-value < 0.05. Genes altered by Allo are presented
with their main function in red boxes.

49

Gene Fold Regulations P-Value
Cxcl16 2.164 0.043616
Cyp51 1.2049 0.038345
Hmgcs1 1.2153 0.002836
Lrp6 1.3052 0.030385
Akr1d1 -8.6677 0.047963
Pcsk9 -1.6132 0.007486
!
Cxcl16
Cyp51
Hmgcs1
Lrp6
Akr1d1
Pcsk9
Allo
!Cxcl16
PI3K/Akt/Erk
Pathways
Cell
proliferation,
migration,
wound
healing
!Lrp6 Wnt Pathway
"AD
progression
"PCSK9 !LDLR
!Cholesterol
Efflux
!HMGCS1
!Cholesterol
synthesis
!Cyp51
!Cholesterol
synthesis
"Akr1d1 "5β-steroids
Allo 10mg/kg SC 20% HBCD vs. Vehicle 20% HBCD

Figure 8. Allopregnanolone increases the gene expression of CXCL16, LRP6,
HMGSC1, and Cyp51 involved in stimulation of the PI3K/Akt/Erk and Wnt
pathways as well as the cholesterol synthesis pathway. Allopregnanolone also
decreases the gene expression of PCSK9 and AKR1D1 to promote cholesterol efflux.
Comparisons were made between the Allo 10mg/kg SC 20% HBCD group and Vehicle
20% HBCD. Significant genes shown with fold regulation and P-value < 0.05. Genes
altered by Allo are presented with their main function in red boxes.

50

Figure 9. Allopregnanolone increases the gene expression of STAB2 and Idi2
potentially reducing inflammation and increasing cholesterol synthesis.
Allopregnanolone also decreases the gene expression of ANGPTL3 a potential
mechanism for reducing plasma cholesterol levels. Comparisons were made between
the Allo 10mg/kg SC PBS/5% EtOH group and Vehicle 20% HBCD. Significant genes
shown with fold regulation and P-value < 0.05. Genes altered by Allo are presented with
their main function in red boxes.

Gene Fold Regulations P-Value
Idi2 1.923 0.012321
Stab2 1.4751 0.046407
Angptl3 -1.3942 0.035674
!
Idi2
Stab2
Angptl3
Allo
!Stab2
!Anti-
inflammatory
cytokines
Phagocytosis of
apoptotic cells
"Angpl3
"Plasma
triglyceride and
HDL cholesterol
!Idi2
!Cholesterol
synthesis
Allo 10mg/kg SC PBS/5% EtOH vs. Vehicle 20% HBCD

51

4. Conclusion
Acute treatment with Allo in the AD mouse model, 3xTgAD, altered the expression of
multiple genes that promote cholesterol homeostasis and thereby decrease Aβ generation
in brain. Many genes involved in cholesterol synthesis and efflux were upregulated by
Allo, highlighting the importance of cholesterol regulation in the brain. HMGCS1,
Cyp51, and Idi2 genes, involved in the mevalonate pathway, increased in expression 24
hours after Allo administration (Figure 10). The mevalonate pathway is the only
intracellular component to endogenous cholesterol production and is vital for brain
function and survival (Leduc et al., 2010). Allo increased the expression of the LXR
target protein LDL-R (Figure 5b). PCSK9 is involved in the degradation of LDL-R
protein therefore altering cholesterol homeostasis. However, Allo reduced the expression
of the PCSK9 gene, preventing LDL-R removal and thus indirectly promoting cholesterol
homeostasis. Allo further promotes cholesterol efflux by downregulating the gene
ANGPTL3. ANGPTL3 inhibits the enzyme LPL that is necessary to form APOE
lipoproteins (Figure 10). APOE is vital for cholesterol transport as well as Aβ clearance.

Gene expression data suggest that Allo may influence the PI3K/Akt/Erk and Wnt
signaling pathways in the brain to decrease AD progression and increase cell proliferation
(Figure 10). Further, insulin resistance and oxidative stress inhibit Wnt signaling to
promote the formation of Aβ and thereby contribute to the pathogenesis of AD (Kang &
Rivest, 2012). Allo increased the expression of the gene LRP6, a coreceptor for the Wnt
pathway, therefore promoting the Wnt pathway to clear Aβ from the brain. Insulin
resistance also attenuates the PI3K/Akt/Erk pathway promoting the formation of Aβ. Akt

52

has been shown to inhibit proteins that are involved in Aβ generation (Kang & Rivest,
2012). CXCL16 stimulates the PI3K/Akt/Erk pathway, and it was upregulated with Allo
administration. Both pathways are advantageous to stimulate in order to help remove Aβ.

Furthermore, genes involved in inflammation were altered in the 3xTgAD mouse model
suggesting the role of Allo in reducing inflammation caused by AD. Our previous data
demonstrated that Allo administration stimulates the LXR pathway to reduce the
inflammatory mediator NF-κB (Figure 5f). OLR1 stimulates transcription of NF-κB, and
this gene was downregulated with Allo treatment. The PCSK9 protein also promotes
inflammation through NF-κB. Allo significantly reduced PCSK9 gene expression and
provides a plausible role for Allo in reducing NF-κB-mediated neuroinflammation.
Moreover, CXCL16, involved in the PI3K/Akt/Erk pathway, is an anti-inflammatory
marker and was upregulated by Allo. Lastly, Allo increased gene expression of Stab2,
involved in producing anti-inflammatory mediators as well as functions in phagocytosis.
Overall, our targeted array gene expression studies provide evidence for an anti-
inflammatory role of Allo in the brain.

Two different doses of Allo (1mg/kg and 10mg/kg) along with vehicle were
subcutaneously injected into 3xTgAD transgenic male mice. Additionally, two different
formulations of the 10mg/kg dose were used, which differ in their release of Allo. The
PBS/5% EtOH suspension had a slower release of Allo compared to the 20% HBCD
suspension (Irwin et al., 2015). Most of the genes between treatment groups varied,
highlighting how different doses and suspensions of Allo elicit varying responses.

53

Figure 10. Allopregnanolone alters genes that are involved in cholesterol synthesis,
transport, and efflux. Allopregnanolone also changes expression of genes involved in
inflammation and Aβ production. Three genes were increased that are involved in the
mevalonate pathway to produce cholesterol (HMGCS1, Cyp51, and Idi2). The gene
Angpt3 that inhibits LPL was reduced. The gene PCSK9 that inhibits LDL-R was
reduced. Genes that stimulate the Wnt pathway (LRP6) and the PI3K/Akt/Erk pathway
(Cxcl16) were increased, which inhibit Aβ formation. Lastly, two genes involved in
inflammation were altered (OLR1 and Stab2).
apoE
cholesterol 24-OHC
24-OHC
LDL-R
cholesterol
apoE
ABCA1
mevalonate
pathway
!Hmgcs1
!Cyp51
!Idi2
BBB
neuron
astrocyte
!Stab2
Gene Array Genes Altered by Allo
in Cortex
24-OHC
Brain
apoE
"PCSK9! X!
Wnt!Pathway! !LRP6
LPL
X!
"Angpt3!
Alzheimer’s
Disease
Aβ
APP
gamma-secretase
inflamma.on! NF-κB !
"OLR1
PI3K/Akt/Erk! !Cxcl16

54

CHAPTER 5
Overall Discussion
Allo has many beneficial actions in the brain. It is known that Allo administration
promotes neuron regeneration, helps cognitive function, and reduces the
neuropathological symptoms of AD. However, the mechanisms of action through which
it exerts these benefits are not clearly defined. We have shown that one of the favorable
outcomes of Allo administration is to promote lipid homeostasis. Since cholesterol cannot
cross the BBB it has to be made in the brain. Appropriate methods for production,
utilization, transport, and efflux need to be tightly regulated, as the BBB creates a strict
segregation from the peripheral tissues. Cholesterol synthesis primarily occurs in
astrocytes and is transported within the brain by local lipoproteins (Figure 11).
Variability of cholesterol distribution and metabolism may lead to the pathology of AD
(Wollmer, 2010). High blood pressure and high total plasma cholesterol is thought to
increase the risk for developing AD (Koldamova et al., 2015). Conversely, it has been
shown that blood cholesterol levels are not consistently elevated in AD patients (Dias,
Polidori, & Griffiths, 2014). This is also justified by the fact that “plasma lipoproteins are
segregated from brain cholesterol” so elevated plasma cholesterol does not correlate with
brain cholesterol levels (Dias, Polidori, & Griffiths, 2014).

Studies have shown that cholesterol overload can lead to diseases such as AD and an
increased level of Aβ in the brain (Kang & Rivest, 2012; Hong & Tontonoz, 2014). Chen
et al. discovered that HMGCR expression was induced by Allo. HMGCR is the rate-
limiting enzyme in the cascade of cellular cholesterol biosynthesis. Our results showed an

55

increase in HMGCS1 with Allo treatment from the gene array, which is the step prior in
the reaction. Allo increases cholesterol production but it also increases the mechanisms
for its transport and removal. Therefore too much cholesterol in the brain may not be
detrimental as long as the appropriate efflux and transport mechanisms are functioning
properly. It has been suggested to use statin drugs that inhibit HMGCR to reduce
cholesterol levels to prevent the onset of AD or delay its progression. Statins reduce
formation and entry of cholesterol into circulation and also upregulate LDL-R activity.
However, clinical trials with statins proved to not be useful for preventing or treating AD
and dementia (McGuinness & Passmore, 2010). Furthermore, clinical trials had mixed
results further verifying that cholesterol-lowering drugs do not have protective effects
against the manifestation or progression of AD (Wollmer, 2010; McGuinness &
Passmore, 2010).

Cholesterol is fundamental in the body and is the main component of myelin, a substance
needed for proper neuronal communication. Thus cholesterol is essential for structure and
function of the CNS (Wang et al., 2002). Depletion is associated with the development of
degenerative disorders; therefore, it is not beneficial to inhibit cholesterol synthesis
(Hong & Tontonoz, 2014). Additionally, some studies have shown age-related decrease
in major membrane lipids is accelerated in individuals with AD (Valdez, Phelix, Smith,
Perry, & Santamaria, 2011).

Since the brain cannot degrade cholesterol nor can cholesterol cross the BBB,
mechanisms are in place to rid the brain of extra cholesterol and prevent its toxic buildup.

56

The most important mechanism involves the conversion of cholesterol into oxysterols,
which can diffuse across the BBB. Cholesterol synthesized in the brain can be oxidized
through enzymatic and non-enzymatic mechanisms (Dias, Polidori, & Griffiths, 2014).
The most imperative conversion occurs through Cyp46a1 that produces 24OHC (Figure
11). Oxidation generally has a negative connotation with it being involved in free radical
generation, Aβ production, neuroinflammation, and neuronal loss. In this process
oxidation is doing something advantageous allowing for a mechanism for cholesterol to
exit the brain. The brain can eliminate excess oxysterols but it can also conversely allow
the toxic buildup of them. Cyp27a1 converts the oxysterol 27OHC from cholesterol in a
similar manner to Cyp46a1. 27OHC is thought to stimulate the amyloidogenic pathway,
which produces Aβ and hyperphosphorylated tau. On the other hand, 24OHC is thought
to reduce Aβ production by down regulating APP trafficking. From this rationale the
balance between 24OHC and 27OHC concentrations and their subsequent enzymes is
essential. Cyp46a1 plays a protective role in the brain by regulating cholesterol
homeostasis, increasing the flux of excess cholesterol in the form of 24OHC out of the
brain, and preventing Aβ production. We have demonstrated that Allo significantly
increases the amount of Cyp46a1 enzyme and therefore we hypothesize that Allo controls
an important checkpoint mechanism to convert cholesterol to 24OHC which can either
activate LXR and PXR regulated transcription pathways or efflux cholesterol, in the form
of 24OHC, from the brain (Figure 5e, Figure 6 and Figure 11).

In addition to Allo directly binding, 24OHC is a potent endogenous ligand for both LXR
and PXR. LXR and PXR are nuclear receptors that function in cholesterol clearance and

57

lead to decreased Aβ accumulation. We have shown that Allo potentially binds to both
LXR and PXR receptors, which in turn activate target genes. Previous research has also
demonstrated that chronic Allo administration increases LXR, PXR, and HMGCR. Our
data showed that the protein expression of several LXR and PXR genes were altered with
24 hour Allo administration in the 3xTgAD mouse model. The inflammatory mediator
NF-κB was found to be reduced and can result in less inflammatory burden. Oxidative
stress increases with age resulting in a variety of negative consequences on the brain,
especially neuroinflammation causing damage to proteins, lipids, and DNA (Dias,
Polidori, & Griffiths, 2014). Since age is a significant risk factor for AD it is crucial to
target inflammatory pathways. The transporter ABCA1 and receptor LDL-R were
increased, which play crucial roles in cholesterol and Aβ clearance. Furthermore, SORL1
was also increased which prevents Aβ production. Since cholesterol is a crucial
component to the brain for development and function it is important to target all aspects
of its utilization. Allo helps maintain and reverse the effects of dysregulated cholesterol
homeostasis caused from AD. It does so by insuring proper transport, removal, and
production of cholesterol as well as enhancing Aβ clearance and preventing its
catabolism.

Taken all of this data into consideration, the results suggest that Allo plays a role in
cholesterol homeostasis by directly activating LXR and PXR leading to the induction of
various genes or helps generate oxysterols, which serve as ligands to the receptors via
Cyp46a1. Either process results in the generation of proteins involved in treating AD.
Some LXR agonists have entered clinical trials and are being investigated for their

58

therapeutic potential for the treatment of AD and atherosclerosis. However, due to their
undesirable increases in serum and hepatic triglyceride levels there are no LXR agonists
currently on the market (Williams et al., 2003; Hong & Tontonoz, 2014). Since Allo has
many beneficial effects in the brain and can potentially activate LXR and PXR, the
effects in the liver have been looked at. Chen et al. concluded that Allo has no significant
effect on LXR or PXR expression in the liver (Chen et al., 2011). Therefore, Allo is
primarily affecting cholesterol homeostasis in the brain without leading to undesirable
consequences elsewhere. This is further shown by the preference of Allo for the brain
specific isoform of LXR (LXRβ) over the one that is ubiquitously expressed (LXRα).

This research has demonstrated that Allo may be involved in many other pathways in
addition to LXR and PXR (Figure 10, Figure 11). The gene array revealed many genes
that were upregulated and downregulated in the cortex when 3xTgAD mice were
administered Allo. For example, the PI3K/Akt and Erk pathways may be stimulated by
Allo to increase cell proliferation, cell migration, and aid in wound healing. The Wnt
pathway may also be involved which assists in neurogenesis, and this has already been a
known action of Allo. Additionally, more anti-inflammatory genes were discovered as
demonstrated, for instance we found that Allo reduced the expression of the
inflammatory mediator NF-κB (via multiple pathways). We also showed a direct
correlation between free radical generation, Aβ plaques, neuroinflammation, and AD
progression. Other pathways involved in lowering circulating lipoprotein and cholesterol
were increased following exposure to Allo. Additional pathways that increase expression
of the cholesterol receptor LDL-R were discovered. Lastly, Allo increased expression of

59

cholesterol synthesis genes involved in the mevalonate pathway. These genes are
subsequently involved in enzymatic pathways to generate oxysterol LXR agonists (Figure
11).

Allo has proved to be a valuable treatment for AD in mouse models by decreasing
pathology as well as increasing neurogenesis. Allo administered once per week has been
determined as the optimal treatment regimen since 1) the mechanism of neurogenesis and
incorporation of newly born neurons requires the cells to differentiate prior to the next
dose 2) the mechanisms of Allo reduction in AD pathology and improved cholesterol
homeostasis require a minimum of once per week dosing. Further, the dose of Allo is
critical to avoid the ascending or descending arms of its characteristic inverted U-shaped
dose-response curve. The optimal Allo dose depends on the model system and biological
activity desired which form promotion of adult neurogenesis were estimated through
pharmacokinetic and efficacy studies of Allo levels in the blood and brain (Irwin et al.,
2015; Irwin, Solinsky, & Brinton, 2014). Chen et al. showed that when Allo is given
three times a week to the AD mouse model, AD pathology was reduced the most but it
also decreased neurogenesis (Chen et al., 2011). From this rationale, administration of an
Allo analog, in an oral suspension, more frequently along with Allo once a week has been
proposed. This would maintain the increase in neurogenesis caused by Allo as well as
better treat the pathology with the analog. From the docking study we learned that the
Allo derivative Ganaxolone binds strongly to LXR (LXRβ 0.07 μM, LXRα 1.27 μM) and
PXR (0.49 μM). Ganaxolone also shows preference for the brain specific LXR isoform β.
Thus since Ganaxolone does not promote neurogenesis it can be taken more frequently to

60

help with cholesterol homeostasis through all of the mechanisms described prior (Mellon,
Gong, & Schonemann, 2011).

In conclusion, my work coupled with previous efforts by others has shown that
cholesterol homeostasis is essential for normal brain function and health. The work
presented in my thesis, taken in context with previous investigations advances the
understanding of Allo-induced regulation of cholesterol homeostasis and Aβ generation
to ameliorate AD. These efforts have demonstrated that an increase in cholesterol
production is beneficial to the patient as long as there are adequate measures for its
transport and removal from the brain. To this end, results from my current work indicate
that Allo promotes cholesterol homeostasis through processes that involve LXR and PXR
activation. Additionally LXR initiation reduces Aβ formation, promotes its clearance,
and reduces the inflammatory response (Jakobsson et al., 2012).

61

Figure 11. Allopregnanolone promotes cholesterol and amyloid-beta clearance
mechanisms. Complete representation of the western blot proteins and gene array genes
altered by Allo administration in 3xTgAD mice.
1. Allo promotes cholesterol synthesis by altering genes involved in the mevalonate
pathway. The mevalonate pathway is the main way cholesterol is synthesized in
the brain.
2. The protein levels of ABCA1, LDL-R, and SORL1 are increased with Allo
administration resulting in increased cholesterol movement, uptake, and efflux.
These proteins also promote Aβ degradation and clearance.
3. The enzyme Cyp46a1 is increased with Allo administration providing a way for
cholesterol to leave the brain. Oxysterols are also generated which serve as
ligands for LXR and PXR.
4. Allo can bind and activate LXR and PXR resulting in gene products that reduce
AD pathology by increasing cholesterol efflux, uptake, and synthesis. Activation
of LXR reduces NF-κB subsequently reducing inflammation.
5. Allo alters other genes involved in cholesterol homeostasis, inflammation, and Aβ
clearance, which are useful to reduce AD pathology and progression.
6. Allo promotes cholesterol homeostasis by increasing cholesterol efflux, uptake,
and synthesis thus decreasing AD pathology, progression. Allo also promotes Aβ
degradation and clearance as well as decreases inflammation.
apoE
cholesterol 24-OHC
24-OHC
LDL-R/
SORL1
Aβ
degradation
clearance
Cyp46a1
!LXRβ!
PXR
!LDL-R
!SORL1
!ABCA1
"NF-κB
Allo
cholesterol
apoE
ABCA1
mevalonate
pathway
Hmgcs1
Cyp51
Idi2
BBB
"AD pathology/progression
!cholesterol efflux, uptake,
synthesis
"inflammation
Allo
neuron
astrocyte
!LRP6
!Cxcl16
"PCSK9
"OLR1
!Stab2
"Angp13
W.B. Proteins Altered by Allo
in Cortex/Hippocampus
Gene Array Genes Altered by Allo
in Cortex
24-OHC
Brain
apoE
1.#
2.#
3.#
4.#
5.#
6.#
LPL

62

BIBLIOGRAPHY

Ai, N., Krasowski, M. D., Welsh, W. J., and Ekins, S. (2009) Understanding Nuclear
Receptors Using Computational Methods. Drug Discovery Today 14(9-10), 486–494.

Altmann, S. W., et al. (2004) Niemann-Pick C1 Like 1 Protein Is Critical for Intestinal
Cholesterol Absorption. Science 303(5661), 1201-1204.

Alzheimer’s Association. (2015) Alzheimer’s Disease Facts and Figures. Alzheimer’s &
Dementia 11(3), 332-384.

Aslanian, A. M., and Charo, I. F. (2006) Targeted Disruption of the Scavenger Receptor
and Chemokine CXCL16 Accelerates Atherosclerosis. Molecular Cardiology 114(6),
583-590.

Bikadi, Z., Demko, L., and Hazai, E. (2007) Functional and structural characterization of
a protein based on analysis of its hydrogen bonding network by hydrogen bonding plot.
Arch Biochem Biophys 461(2), 225-234.

Bikadi, Z., and Hazai, E. (2009) Application of the PM6 semi-empirical method to
modeling proteins enhances docking accuracy of AutoDock. J Cheminf 1(1), 15.

Bjorkhem, I., and Meaney, S. (2004) Brain Cholesterol: Long Secret Life Behind a
Barrier. Arterioscler Thromb Vasc Biol 24(5), 806-815.

Caglayan, S., et al. (2014) Lysosomal Sorting of Amyloid-b by the SORLA Receptor
Is Impaired by a Familial Alzheimer’s Disease Mutation. Science Translational Medicine
6(223), 223ra20.

Calkin, A., and Tontonoz, P. (2010) LXR signaling pathways and atherosclerosis.
Arteriosclerosis, Thrombosis, and Vascular Biology 30(8), 1-12.

Castellano, J. M., et al. (2012) Low-Density Lipoprotein Receptor Overexpression
Enhances the Rate of Brain-to-Blood Aβ Clearance in a Mouse Model of Β-Amyloidosis.
PNAS 109(38), 15502-15507.

Chen, M., and Penning, T. M. (2014) 5β-Reduced steroids and human Δ4-3-ketosteroid
5β-reductase (AKR1D1). Steroids 83(1), 17-26.

Chen, M., Jin, Y., and Penning, T. M. (2015) The rate-determining steps of aldo–keto
reductases (AKRs), a study on human steroid 5β-reductase (AKR1D1). Chemico-
Biological Interactions 234(1), 360-365.

Chen, S., Wang, J. M., Irwin, R. W., Yao, J., Liu, L., Brinton, R. D. (2011).
Allopregnanolone Promotes Regeneration and Reduces β-Amyloid Burden in a
Preclinical Model of Alzheimer's Disease. PLoS ONE 6(8), e24293.

63

Cheng, O., Ostrowski, R. P., Liu, W., and Zhang, J. H. (2010) Activation of liver X
receptor reduces global ischemia brain injury by reduction of nuclear factor-κB.
Neuroscience 144(4), 1101-1109.

Clizbe, D. B., Owens, M. L., Masuda K. R., Shackelford, J. E., and Krisans S. K. (2007)
IDI2, a Second Isopentenyl Diphosphate Isomerase in Mammals. The Journal of
Biological Chemistry 282(9), 6668-6676.

De Strooper, B., and Annaert, W. (2000) Proteolytic processing and cell biological
functions of the amyloid precursor protein. Journal of Cell Science 113(11), 1857-1870.

Deane, R., et al. (2004) LRP/Amyloid β-Peptide Interaction Mediates Differential Brain
Efflux of Aβ Isoforms. Neuron 43(3), 333-344.

Deane, R., Sagare, A., and Zlokovic, B. (2008). The Role of the Cell Surface LRP and
Soluble LRP in Blood-Brain Barrier Aβ Clearance in Alzheimer’s Disease. Current
Pharmaceutical Design 14(16), 1601–1605.

Dias, I. H., Polidori, M. C., and Griffiths, H. R. (2014) Hypercholesterolaemia-induced
oxidative stress at the blood–brain barrier. Biochem Soc Trans 42(4), 1001-1005.

Dietshy, J. M., and Turley S. D. (2004) Cholesterol metabolism in the central nervous
system during early development and in the mature animal. Journal of Lipid Research
45(8), 1375–1397.

Fan, J., Shimizu, Y., Chan, J., Wilkinson, A., Ito, A., Tontonoz, P., Dullaghan, E., Galea,
L. A., Pfeifer, T., and Wellington, C. L. (2013) Hormonal Modulators of Glial ABCA1
and apoE Levels. Journal of Lipid Research 54(11), 3139-3150.

Francisa, P. T., Palmer, A. M., Snape, M., and Wilcock, G. K. (1999) The cholinergic
hypothesis of Alzheimer’s disease: a review of progress. J Neurol Neurosurg Psychiatry
55(2), 137-147.

Goldsetin, J. and Brown, M. (1990) Regulation of the mevalonate pathway. Nature
343(1), 425-430.

Halgren, T. A. (1998) Merck molecular force field. I. Basis, form, scope, parametrization,
and performance of MMFF94. Journal of Computational Chemistry 17(5-6), 490-519

Hattermann, K., Ludwig, A., Gieselmann, V., Held-Feindt, J., and Mentlein, R. (2008)
The chemokine CXCL16 induces migration and invasion of glial precursor cells via its
receptor CXCR6. Molecular and Cellular Neuroscience 39(1), 133-141.

He, J., Evans, C. O., Hoffman, S. W., Oyesiku, N. M., and Stein, D. G. (2004)
Progesterone and allopregnanolone reduce inflammatory cytokines after traumatic brain
injury. Experimental Neurology 189(2), 404-412.

64

Hoerer, S., Schmid, A., Heckel, A., Budzinski, R. M., and Nar, H. (2003) Crystal
Structure of the Human Liver X Receptor β Ligand-binding Domain in Complex with a
Synthetic Agonist. Journal of Molecular Biology 334(5), 853-861.

Hong, C., and Tontonoz, P. (2014) Liver X receptors in lipid metabolism: opportunities
for drug discovery. Nature Reviews Drug Discovery 13(6), 433-444.

Irwin, R., and Brinton, R. D. (2014). Allopregnanolone as regenerative therapeutic for
Alzheimer's disease: translational development and clinical promise. Prog Neurobiol
113(1), 40-55.

Irwin, R., 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(203), 1-19.

Irwin, R., 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(6), e012831

Jakobsson, T., Treuter, E., Gustafsson, J. A., and Steffensen, K. R. (2012) Liver X
receptor biology and pharmacology: new pathways, challenges and opportunities. Trends
in Pharmacological Sciences 33(7), 394-404.

Janowski, B. A., Grogan, M. J., Jones, S. A., Wisely, G. B., Kliewer, S. A., Corey, E. J.,
and Mangelsdorf, D. J. (1999) Structural requirements of ligands for the oxysterol liver X
receptors LXRα and LXRβ. PNAS 96(1), 266-271.

Joiner, D. M., Ke, J., Zhong, Z., Xu, H. E., and Williams, B. O. (2013) LRP5 and LRP6
in development and disease. Trends in Endocrinology & Metabolism 24(1), 31-39.

Kang, J., and Rivest, S. (2012) Lipid Metabolism and Neuroinflammation in Alzheimer's
Disease: A Role for Liver X Receptors. Endocrine Reviews 33(5), 715-746.

Khaidakov, M., Mitra, S., Kang, B. Y., Wang, X., Kadlubar, S., Novelli, G., Raj, V.,
Winters, M., Carter, W. C., and Mehta, J. L. (2011) Oxidized LDL Receptor 1 (OLR1) as
a Possible Link between Obesity, Dyslipidemia and Cancer. PLoS ONE 6(5), e20277.

Kim, S., Park, S. Y., Kim, S. Y., Bae, D. J., Pyo, J. H., Hong, M., and Kim, I. S. (2012)
Cross Talk between Engulfment Receptors Stabilin-2 and Integrin αvβ5 Orchestrates
Engulfment of Phosphatidylserine-Exposed Erythrocytes. Molecular and Cellular
Biology 32(14), 2698-2708.

65

Koldamova, R. P., Lefterov, I. M., Staufenbiel, M., Wolfe, D., Huang, S., Glorioso, J. C.,
Walter, M., Roth, M. G., and Lazo, J. S. (2005) The liver X receptor ligand T0901317
decreases amyloid beta production in vitro and in a mouse model of Alzheimer's disease.
J Biol Chem 280(6), 4079-4088.

Kysenius, K., Muggalla, P., Mätlik, K., Arumäe, U., and Huttunen, H. J. (2012) PCSK9
regulates neuronal apoptosis by adjusting ApoER2 levels and signaling. Cellular and
Molecular Life Sciences 69(11), 1903-1916.

Lamba, V., Yasuda, K., Lamba, J. K., Assem, M., Davila, J., Strom, S., and Schuetz, E.
G. (2004) PXR (NR1I2): splice variants in human tissues, including brain, and
identification of neurosteroids and nicotine as PXR activators. Toxicol Appl Pharmacol
199(3), 251-265.

Langmade, S. J., Gale, S. E., Frolov, A., Mohri, I., Suzuki, K., Mellon, S. H., Walkley, S.
U., Covey, D. F., Schaffer, J. E., and Ory, D. S. (2006) Pregnane X receptor (PXR)
activation: a mechanism for neuroprotection in a mouse model of Niemann-Pick C
disease. PNAS 103(37), 13807-13812.

Lazarov, O., and Marr, R. A. (2010) Neurogenesis and Alzheimer's disease: at the
crossroads. Exp Neurol 223(1), 267–281.

Leduc, V., Jasmin-Bélanger, S., and Poirier, J. (2010) APOE and cholesterol homeostasis
in Alzheimer's disease. Trends in Molecular Medicine 16(10), 469-477.

Lepesheva, G. I., and Waterman, M. R. (2004) CYP51—the omnipotent P450. Molecular
and Cellular Endocrinology 215(1–2), 165-170.

Liu, C., et al. (2014) Deficiency in LRP6-Mediated Wnt Signaling Contributes to
Synaptic Abnormalities and Amyloid Pathology in Alzheimer’s Disease. Neuron 84(1),
63-77.

Livak, K. J., and Schmittgen, T. D. (2008) Analyzing real-time PCR data using the
comparative Ct method. Nature Protocols 3(6), 1101-1108.

Marx, C. E., et al. (2006) The neurosteroid allopregnanolone is reduced in prefrontal
cortex in Alzheimer's disease. Biol Psychiatry 60(12), 1287-1294.

McDonald, I. K., and Thornton, J. M. (1994) Satisfying Hydrogen Bonding Potential in
Proteins. JMB 238(5), 777-793.

McGuinness, B., and Passmore, P. (2010) Can Statins Prevent or Help Treat Alzheimer’s
Disease? Journal of Alzheimer’s Disease 20(3), 925-933.

66

Mellon, S. H., Gong, W., and Schonemann, M. D. (2008) Endogenous and synthetic
neurosteroids in treatment of Niemann–Pick Type C disease. Brain Research Reviews
57(2), 410-420.

Mellon, S. H., Gong, W., and Schonemann, M. D. (2011) Endogenous and Synthetic
Neurosteroids in the Treatment of Niemann–Pick Type C Disease, in Hormones in
Neurodegeneration, Neuroprotection, and Neurogenesis (eds A. G. Gravanis and S. H.
Mellon), Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany.

Morris, G. M., Goodsell, D. S., Halliday, R. S., Huey, R., Hart, W. E., Belew, R. K. and
Olson, A. J. (1998) Automated docking using a Lamarckian genetic algorithm and an
empirical binding free energy function. Journal of Computational Chemistry 19(14),
1639-1662.

Naylor, J. C., et al. (2010) Allopregnanolone levels are reduced in temporal cortex in
patients with Alzheimer's disease compared to cognitively intact control subjects.
Biochim Biophys Acta 1801(8), 951-959.

Oddo, S., Caccamo, A., Shepherd, J. D., Murphy, M. P., Golde, T.E., Kayed, R.,
Metherate, R., Mattson, M. P., Akbari, Y., and LaFerla, F. M. (2003) Triple-Transgenic
Model of Alzheimer's Disease with Plaques and Tangles: Intracellular Aβ and Synaptic
Dysfunction. Neuron 39(3), 409-421.

Ohyama, Y., et al. (2005) Studies on the Transcriptional Regulation of Cholesterol 24-
Hydroxylase (CYP46A1). The Journal of Biological Chemistry 281(7), 3810-3820.

Palop, J. J., and Mucke, L. (2010) Amyloid-β Induced Neuronal Dysfunction in
Alzheimer’s Disease: From Synapses toward Neural Networks. Nature Neuroscience
13(7), 812-818.

Reddy, J. V., Ganley, I. G., and Pfeffer, S. R. (2006) Clues to Neuro-Degeneration in
Niemann-Pick Type C Disease from Global Gene Expression Profiling. PLoS ONE 1(1),
e19.

Serpente, M., et al. (2011) Role of OLR1 and its Regulating hsa-miR369-3p in
Alzheimer’s Disease: Genetics and Expression Analysis. Journal of Alzheimer’s Disease
26(4), 787-793.

Solis, F. T., and Wets, R. J-B. (1981) Minimization by random search techniques.
Mathematics of Operations Research 6(1), 19-30.

Svensson, S., et al. (2003) Crystal structure of the heterodimeric complex of LXRα and
RXRβ ligand-binding domains in a fully agonistic conformation. The EMBO Journal
22(18), 4625-4633.

67

Timsit, Y. E., and Negishi, M. (2007) CAR and PXR: The Xenobiotic-Sensing
Receptors. Steroids 72(3), 231-246.

Valdez, C. M., Phelix, C. F., Smith, M. A., Perry, G., and Santamaria, F. (2011)
Modeling cholesterol metabolism by gene expression profiling in the hippocampus.
Molecular Biosystems 7(6), 1891-1901.

Vance, J. E. (2012) Dysregulation of cholesterol balance in the brain: contribution to
neurodegenerative diseases. Disease Models & Mechanisms 5(6), 746-755.

Wallace, B. D., Betts, L., Talmage, G., Pollet, R. M., Holman, N. S., and Redinbo, M. R.
(2013) Structural and Functional Analysis of the Human Nuclear Xenobiotic Receptor
PXR in Complex with RXRα. Journal of molecular biology 425(14), 2561-2577.

Wang, L., Schuster, G. U., Hultenby, K., Zhang, Q., Andersson, S., and Gustafsson, J. A.
(2002) Liver X Receptors in the Central Nervous System: From Lipid Homeostasis to
Neuronal Degeneration. PMC 99(21), 13878-13883.

Watkins, R. E., Wisely, G. B., Moore, L. B., Collins, J. L., Lambert, M. H., Williams, S.
P., Willson, T. M., Kliewer, S. A., and Redinbo, M. R. (2001) The Human Nuclear
Xenobiotic Receptor PXR: Structural Determinants of Directed Promiscuity. Science
292(5525), 2329-2333.

Weill-Engerer, S., David, J. P., Sazdovitch, V., Liere, P., Eychenne, B., Pianos, A.,
Schumacher, M., Delacourte, A., Baulieu, E. E., and Akwa, Y. (2002) Neurosteroid
quantification in human brain regions: comparison between Alzheimer's and
nondemented patients. J Clin Endocrinol Metab 87(11), 5138-5143.

Westergard, T., Salari, R., Martin, J. V., and Brannigan, G. (2015) Interactions of L-
3,5,3'-Triiodothyronine, Allopregnanolone, and Ivermectin with the GABAA Receptor:
Evidence for Overlapping Intersubunit Binding Modes. PLoS ONE 10(11), e0139072.

Williamas, S., et al. (2003) X-ray crystal structure of the liver X receptor beta ligand
binding domain: regulation by a histidine-tryptophan switch. J Biol Chem 278(29),
27138-27143.

Wollmer, M. A. (2010) Cholesterol-related genes in Alzheimer's disease. Biochimica et
Biophysica Acta 1801(8), 762-773.

Wu, Q., Tang, Z. H., Peng, J., Liao, L., Pan, L. H., Wu, C. Y., Jiang, Z. S., Wang, G. X.,
and Liu, L. S. (2014). The dual behavior of PCSK9 in the regulation of apoptosis is
crucial in Alzheimer's disease progression (Review). Biomedical Reports 2(2), 167-171.

Xue, Y., Moore, L. B., Orans, J., Peng, L., Bencharit, S., Kliewer, S. A., and Redinbo, M.
R. (2007) Crystal Structure of the Pregnane X Receptor-Estradiol Complex Provides
Insights into Endobiotic Recognition. Molecular Endocrinology 21(5), 1028-1038.

68

Zelcer, N., Khanlou, N., Clare, R., Jiang, Q., Reed-Geaghan, E. G., Landreth, G. E.,
Vinters, H. V., and Tontonoz, P. (2007) Attenuation of neuroinflammation and
Alzheimer's disease pathology by liver x receptors. PNAS 104(25), 10601-10606.

Zou, Y., Du, H., Yin, M., Zhang, L., Mao, L., Xiao, N., Ren, G., Zhang, C., and Pan, J.
(2009) Effects of high dietary fat and cholesterol on expression of PPARa, LXRa, and
their responsive genes in the liver of apoE and LDLR double deficient mice. Mol Cell
Biochem 323(1-2), 195-205.

Abstract (if available)

Abstract As the population of elderly Americans increases in the coming years so will the number of Alzheimer’s disease (AD) patients, which is already listed as the sixth‐leading cause of death in the United States. There is currently “no cure” for AD and the current treatments only temporarily treat the disease without taking into account the neuronal damage and cholesterol dysregulation caused by the disease. Cholesterol is a vital component in the brain for structure and function. Inadequate or excessive amounts can have a catastrophic consequence, which is why the brain has a strict regulation of cholesterol production, transport, and removal. It has previously been confirmed that in a mouse model of AD, Allopregnanolone (Allo) reduces amyloid‐beta (Aβ) and neuroinflammatory burden while promoting cholesterol homeostasis via mechanisms that influence the liver‐X‐receptor (LXR) and pregnane‐X‐receptor (PXR). We sought out to further investigate these pathways and how Allo promotes cholesterol homeostasis in the cortex and hippocampus of 3xTgAD mice treated with Allo for 24 hours using methods such as western blot, gene array, and molecular docking. We established that Allo has either the capability to bind to LXR and PXR thus modifying protein expression of target genes directly or by increasing oxysterols production via Cyp46a1, which in turn can activate the receptors. This ultimately leads to a decrease in inflammation, increases cholesterol transport and clearance, and enhances Aβ clearance. Additional genes promoting cholesterol homeostasis were altered with Allo treatment to protect the brain from inflammation, apoptosis, decrease Aβ, and to regulate cholesterol clearance, production, and their equivalent receptors.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

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

PDF

Development of glycine and GABAA ultra-sensitive ethanol receptors (USERs) as novel tools for alcohol and brain research

PDF

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

PDF

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

PDF

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

PDF

Perimenopausal transition increases blood brain permeability: implications for neurodegenerative diseases

PDF

Role of purinergic P2X7 receptors in inflammatory responses in the brain and liver: a study using a mouse model of chronic ethanol and high-fat diet exposure

PDF

Dihydromyricetin improves social isolation anxiety induced synaptic loss and astrogliosis in mice hippocampus

PDF

Age dependent modulation of synaptic plasticity and insulin mimetic effect of lipoic acid on a 3xTg-AD mouse model of Alzheimer's disease: implications as a therapeutic/nutraceutical agent

PDF

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

PDF

Macrophage-specific Stat3 deletion in a mouse model of Alzheimer’s disease

PDF

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

PDF

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

PDF

Akt1 deletion decrease proliferation in aged pancreatic beta-cells by arresting cell cycle in S phase

PDF

Role of purinergic P2X4 receptors in regulation of dopamine homeostasis in the basal ganglia and associated behaviors

PDF

Role of CD33 structure and function in Alzheimer’s disease

PDF

Beneficial effect of antibiotic treatment on alcohol-related liver pathology in mice is not due to reduction in the butyrate-producing gut microbial phyla

PDF

Role of steroidogenic acute regulatory protein (STAR) and sterol carrier protein-x (SCPx) in the transport of cholesterol and other lipids

PDF

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

PDF

Energy metabolism and inflammation in brain aging: significance of age-dependent astrocyte metabolic-redox profile

Asset Metadata

Creator Swanson, Haley M. (author)

Core Title Allopregnanolone promotes cholesterol and amyloid-beta clearance mechanisms

School School of Pharmacy

Degree Master of Science

Degree Program Molecular Pharmacology and Toxicology

Publication Date 02/18/2016

Defense Date 01/28/2016

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

Tag allopregnanolone,Alzheimer's disease,brain,cholesterol homeostasis,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Brinton, Roberta (committee chair), Davies, Daryl (committee member), Okamoto, Curtis (committee member)

Creator Email hmswany@gmail.com,hswanson@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-209823

Unique identifier UC11277978

Identifier etd-SwansonHal-4113.pdf (filename),usctheses-c40-209823 (legacy record id)

Legacy Identifier etd-SwansonHal-4113-0.pdf

Dmrecord 209823

Document Type Thesis

Format application/pdf (imt)

Rights Swanson, Haley M.

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