MECHANISM OF U18666A-MEDIATED CELL DEATH IN CULTURED NEURONS
KOH CHOR HUI VIVIEN (B.Sc.)
A THESIS SUBMITTED
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
DEPARTMENT OF BIOCHEMISTRY
NATIONAL UNIVERSITY OF SINGAPORE
2006 Acknowledgements
I would like to thank:
My supervisor, Assistant Professor Steve Cheung Nam Sang, for the opportunity to work in his lab and his guidance in this research project.
The Biomedical Research Council (BMRC) of the Agency for Science, Technology and Research (A*STAR) of Singapore for the Research Assistant position and financial support (R-183-000-082-305). This work was also financially supported by the National Medical Research Council (NMRC) of Singapore (R-183-000-075-213) and the Academic Research Fund of the National University of Singapore (R-183- 000-060-214).
Associate Professor Matthew Whiteman from the Department of Biochemistry for his guidance in oxidative stress studies.
Dr. Ou Keli and members of his lab from Agenica Research Pte. Ltd., and Dr. Zhang Dao Hai from the Department of Laboratory Medicine of the National University Hospital for their expertise in proteomics and the use of their facilities.
Dr. Li Qiao-Xin from the Department of Pathology of the University of Melbourne for her assistance in Aβ studies; Dr. Qu Dianbo from the Institute of Molecular and Cell Biology for his assistance in radioactive work; Ms Chan Yee Gek from the Department of Anatomy for her assistance in electron microscopy; Jayapal Manikandan from the Department of Physiology for his assistance in GeneSpring™.
Ms Seet Sze Jee, Ms Huang Shan Hong and Ms Siau Jialing from the Department of Biochemistry for their technical assistance.
I am especially grateful to:
Dr. Peng Zhao Feng for his immense patience in teaching and demonstrating all proteomics techniques, as well as the innumerable encouragements and tremendous help given in more ways than one.
Choy Meng Shyan for his wise and sensible advices, innovative suggestions, quick and perceptive remarks, and selfless guidance in many things throughout these years.
My parents and friends for their understanding and support.
This thesis is dedicated to my pet chicken, Jeffrey (deceased), with love.
i Table of Contents
Acknowledgements i
Table of Contents ii
Summary ix
List of Tables xii
List of Figures xiii
List of Abbreviations xvi
List of Publications xix
Chapter 1 Introduction 1
1.1 General Introduction 2
1.2 Literature Review 7
1.2.1 Overview of the central nervous system 7
1.2.2 Neuronal Cell Death 7 1.2.2.1 Cell death in neurodegeneration 8
1.2.3 Niemann-Pick disease type C (NPC) 9 1.2.3.1 Overview of Niemann-Pick diseases 9 1.2.3.2 Characteristics of NPC 9 1.2.3.3 NPC and inhibition of intracellular cholesterol 10 transport 1.2.3.4 Niemann-Pick C proteins 11
1.2.4 Cholesterol 12 1.2.4.1 Cholesterol biosynthesis 13 1.2.4.2 Cholesterol homeostasis in the brain 13
1.2.5 Alzheimer’s disease (AD) 15 1.2.5.1 Molecular pathology of AD 15 1.2.5.2 Cleavage of APP into Aβ 16 1.2.5.3 Cholesterol and AD 16
1.2.6 Bridging NPC and AD 18
ii 1.2.7 U18666A: The widely-studied inhibitor of cholesterol 18 transport 1.2.7.1 Overview of hydrophobic amines and steroids 18 1.2.7.2 Class-2 amphiphiles 19 1.2.7.3 Studied effects of U18666A 19
Chapter 2 Materials and Methods 23
2.1 Materials 24
2.1.1 Buffers and solutions 24
2.1.2 Consumables 24
2.1.3 Drugs 25
2.1.4 Primary antibodies 25
2.1.5 Secondary antibodies 26
2.1.6 Protein assay kits 27
2.2 Methods 28
2.2.1 Preparation of mouse primary cortical neurons 28
2.2.2 Culture and maintenance of adherent cell lines 29
2.2.3 MTT assay 30
2.2.4 Release of lactate dehydrogenase (LDH) 31
2.2.5 LIVE/DEAD® viability and cytotoxicity assay 31
2.2.6 Annexin V staining 32
2.2.7 Hoechst staining 32
2.2.8 Transmission electron microscopy 33
2.2.9 Preparation of protein lysates for Western blotting 33
2.2.10 Sodium dodecyl sulfate polyacrylamide gel electrophoresis 34 (SDS-PAGE)
2.2.11 Western blotting 35
iii 2.2.12 Filipin staining 35
2.2.13 Cholesterol measurement 36
2.2.14 Fluorometric caspase activity measurement 37
2.2.15 Measurement of intracellular ATP and glutathione (GSH) 38
2.2.16 Measurement of proteasome activity 39
2.2.17 Measurement of mitochondrial membrane potential 39
2.2.18 Determination of intracellular oxidative stress by 2’,7’- 40 dichlorofluorescin diacetate (DCFH-DA)
2.2.19 Measurement of lipid peroxidation 41
2.2.20 Assessment of oxidized proteins 42
2.2.21 Analysis of DNA base modifications by gas chromatography- 43 mass spectrometry (GC-MS)
2.2.22 Measurement of β-amyloid (Aβ) 45
2.2.23 Determination of amyloid precursor protein (APP) derivatives 46
2.2.24 Immunocytochemistry 46
2.2.25 Measurement of cyclin-dependent kinase 5 (cdk5) activity 47
2.2.26 Isolation of total RNA 48
2.2.27 Quantification of total RNA and determination of RNA 48 integrity
2.2.28 Agarose gel electrophoresis 49
2.2.29 One-cycle complementary DNA (cDNA) synthesis 49
2.2.30 Cleanup of double-stranded cDNA 50
2.2.31 Synthesis of biotin-labeled complementary RNA (cRNA) 50
2.2.32 Cleanup of biotin-labeled cRNA 50
2.2.33 Fragmentation of cRNA for target preparation 51
iv 2.2.34 Microarray analysis 52
2.2.35 Eukaryotic target hybridization 52
2.2.36 Washing, staining and scanning of probe arrays 53
2.2.37 Microarray data analysis 54
2.2.38 Sample preparation for proteomics analysis 55
2.2.39 Cleanup and quantification of protein for proteomics analysis 55
2.2.40 Immobilized pH gradient (IPG) strip rehydration 56
2.2.41 First-dimension isoelectric focusing (IEF) 57
2.2.42 Equilibration of IPG strips 58
2.2.43 Second-dimension SDS-PAGE 58
2.2.44 Silver staining 59
2.2.45 Analysis of 2D gels 60
2.2.46 Enzymatic digestion of protein spots 61
2.2.47 Mass spectrometry 61
2.2.48 Database searching and identification of proteins 62
2.2.49 Statistical analyses 63
Chapter 3 Cellular Signaling of U18666A-Mediated Cell Death 64
Part I Apoptotic cell death and accumulation of intracellular free 65 cholesterol
3.1 Introduction 66
3.2 Results 71
3.2.1 U18666A induces significant cell death only in primary 71 cortical neurons
3.2.2 U18666A-mediated cell death in primary cortical neurons is 73 apoptotic
v 3.2.3 Caspase-3 activation is correlated with U18666A-mediated 74 neuronal apoptosis
3.2.4 Cholesterol accumulates in primary cortical neurons treated 75 with U18666A
3.2.5 Depletion of intracellular cholesterol attenuates cell death in 76 U18666A-treated cortical neurons
3.3 Discussion 91
Part II Intracellular free radical production and β-amyloid 96 accumulation
3.4 Introduction 97
3.5 Results 101
3.5.1 U18666A treatment leads to loss of intracellular ATP and 101 glutathione (GSH), decrease in proteasome activity, and mitochondrial depolarization
3.5.2 U18666A leads to increased intracellular reactive oxygen 102 species (ROS) and lipid peroxidation in primary cortical neurons
3.5.3 U18666A induces protein oxidation and DNA damage in 103 primary cortical neurons
3.5.4 Co-treatment of U18666A with vitamin E attenuates cell death 104 during U18666A-induced oxidative stress in primary cortical neurons
3.5.5 U18666A leads to cleavage of APP and accumulation of Aβ in 105 primary cortical neurons
3.6 Discussion 117
Part III Role of caspases, calpains, kinases and cell cycle 125 regulatory proteins
3.7 Introduction 126
3.8 Results 130
vi 3.8.1 U18666A induces caspase and calpain activation in primary 130 cortical neurons
3.8.2 U18666A induces hyperphosphorylation of tau in primary 132 cortical neurons
3.8.3 U18666A decreases the kinase activity of Cdk5 and the 133 protein level of p35 in primary cortical neurons
3.8.4 U18666A decreases the protein levels of phosphorylated 133 GSK3, p44/42 MAPK and SAPK/JNK in primary cortical neurons
3.8.5 U18666A increases the protein levels of phospho-p53 and 134 activates cell cycle machinery in primary cortical neurons
3.9 Discussion 147
Chapter 4 Global Gene Expression Profile of U18666A-Mediated 156 Neuronal Apoptosis
4.1 Introduction 157
4.2 Results 162
4.2.1 Determination of RNA integrity and cRNA fragmentation 162 before target hybridization onto GeneChip® probe arrays
4.2.2 Differential gene expression after U18666A treatment in 163 primary cortical neurons
4.2.3 Cluster analysis of genes differentially expressed after 165 U18666A treatment in primary cortical neurons
4.2.4 Validation of differential gene expression through Western 170 blot analysis
4.3 Discussion 192
Chapter 5 A Proteomics Approach to the Study of U18666A- 207 Mediated Neuronal Apoptosis
5.1 Introduction 208
5.2 Results 214
vii 5.2.1 Optimization of experimental conditions for 2D gel 214 electrophoresis
5.2.2 Global protein expression profiles after U18666A treatment in 215 primary cortical neurons
5.2.3 Differentially-expressed proteins after U18666A treatment in 216 primary cortical neurons
5.2.4 Correlation of proteomics changes to corresponding gene 217 alterations in U18666A-treated cortical neurons
5.3 Discussion 233
Chapter 6 General Discussion and Future Work 242
6.1 General Discussion 243
6.2 Future Work 249
References 251
Appendices 289
Appendix I Media for preparation of mouse primary cortical neurons 290
Appendix II Buffers and reagents for SDS-PAGE and Western blotting 291
Appendix III Master mixes and buffers for microarray analysis 293
Appendix IV Gel formulations and solutions for proteomics analysis 295
viii Summary
U18666A (3-β-[2-(diethylamino)ethoxy]androst-5-en-17-one) is a well-known
cholesterol transport-inhibiting drug widely used to mimic the cellular effects of
Niemann-Pick disease type C (NPC). NPC is a juvenile neurodegenerative disorder
characterized by premature neuronal cell loss and altered cholesterol metabolism.
Studies suggest that neuronal cell death in NPC mouse brains occur via necrosis.
However, no cell injury was previously reported in in vitro studies using U18666A.
Here, the chronic effect of U18666A in primary cortical neurons and various cell
lines was first investigated. Results showed significant cell injury only in U18666A-
treated cortical neurons and not the cell lines. Further work using only the primary cortical neurons revealed caspase-3 activation and other indices of apoptosis. Filipin staining and measurement of intracellular cholesterol levels also demonstrated accumulation of free cholesterol in U18666A-treated cortical neurons. These data showed for the first time that U18666A induces apoptosis only in primary cortical neurons and suggested a novel in vitro model system to study NPC. Oxidative stress has recently been implicated in neurological diseases. To characterize the effects of
U18666A on cellular metabolism in primary cortical neurons, levels of intracellular
ATP and glutathione were measured after 72 h of treatment and found to significantly decrease. Similarly, proteasome activity decreased after 72 h of treatment. Markers associated with oxidative damage indicated mitochondrial depolarization, increased intracellular reactive oxygen species production, lipid peroxidation, DNA oxidation and protein oxidation in U18666A-treated cortical neurons. These results suggested
ix that oxidative stress may contribute to U18666A-mediated neuronal apoptosis.
Cortical neurons treated with U18666A showed increased intracellular accumulation of β-amyloid (Aβ). U18666A also induced tau hyperphosphorylation, which occurred only in primary cortical neurons undergoing apoptosis. These mechanistic features reminiscent of Alzheimer’s disease (AD), together with the increased oxidative stress markers observed, may provide further clues to the etiology and pathogenesis of AD.
Chronic exposure of primary cortical neurons to U18666A also led to a potential crosstalk between the caspase and calpain pathways during the neuronal apoptosis mediated by U18666A. Stress in the endoplasmic reticulum might also contribute to
U18666A-mediated neuronal apoptosis, as activation of caspase-12 occurred at an earlier time-point before activation of calpain and other caspases. U18666A treatment might also activate cell cycle machinery, leading to a conflict of signals which resulted in neuronal apoptosis. The microarray approach, when used in conjunction with proteomics analysis, identified eleven differentially-expressed proteins which correlated at the gene expression level in a time-dependent manner during U18666A treatment in primary cortical neurons. The identification of these differentially- expressed proteins, shown to play a role in lipid metabolism and transport, responses to cell death, protein folding and trafficking, and regulation of transcription, might provide clues to decipher the intracellular biochemical changes contributing to
U18666A-mediated neuronal apoptosis. These results provided, for the first time, a combined approach of microarray and proteomics analyses to study the mechanism of neuronal apoptosis due to dysfunction of intracellular cholesterol transport.
Collectively, findings from this research project may facilitate the study of
x neurodegenerative diseases where apoptosis might occur through a similar mechanism as observed in U18666A-mediated neuronal apoptosis.
xi List of Tables
Table 2.1 IEF conditions for cup-loading application method using 58 Immobiline™ DryStrip IPG strips pH 3-10 Non-Linear
Table 3.1 Levels of oxidized DNA base products in vehicle- (Control) and 113 U18666A-treated cortical neuronal DNA, as determined by GC- MS analysis
Table 3.2 Tau phosphorylation in vehicle- (Control) or U18666A-treated 139 cortical neurons
Table 4.1 Differentially-expressed genes in cortical neurons treated with 1 175 μg/ml U18666A
Table 5.1 List of differentially-expressed proteins identified by mass 227 spectrometry in cortical neurons treated with 1 μg/ml U18666A
Table 5.2 List of identified proteins with matching corresponding genes 232 from microarray analysis in cortical neurons treated with 1 μg/ml U18666A
xii List of Figures
Figure 1.1 A simplified cholesterol biosynthesis pathway 14
Figure 1.2 Proteolytic cleavages of APP into Aβ peptides 17
Figure 3.1 Primary cell culture of murine cortical neurons 78
Figure 3.2 Effect of U18666A in primary cortical neurons 79
Figure 3.3 Effect of U18666A in non-neuronal cell lines 80
Figure 3.4 Effect of U18666A in neuronal cell lines 82
Figure 3.5 Measurement of LDH released from U18666A-treated cortical 83 neurons
Figure 3.6 U18666A-induced apoptosis in primary cortical neurons 84
Figure 3.7 Activation of caspase-3 during U18666A-mediated apoptosis in 86 primary cortical neurons
Figure 3.8 Accumulation of intracellular cholesterol in cortical neurons treated 87 with U18666A
Figure 3.9 Effect of statins in primary cortical neurons 88
Figure 3.10 Inhibition of cholesterol biosynthesis with pravastatin attenuates 89 cell death in U18666A-treated cortical neurons
Figure 3.11 Effect of cyclodextrin in primary cortical neurons 90
Figure 3.12 U18666A-mediated loss of intracellular ATP and GSH in primary 107 cortical neurons
Figure 3.13 Measurement of proteasome activity in U18666A-treated cortical 108 neurons
Figure 3.14 Assessment of mitochondrial membrane potential in U18666A- 109 treated cortical neurons
Figure 3.15 U18666A-mediated increase in intracellular reactive oxygen species 110 (ROS) in primary cortical neurons
Figure 3.16 U18666A-mediated lipid peroxidation in primary cortical neurons 111
xiii Figure 3.17 Detection of protein oxidation in U18666A-treated cortical neurons 112
Figure 3.18 Effect of antioxidants in U18666A-treated cortical neurons 114
Figure 3.19 Determination of Aβ levels and APP proteolytic processing 115
Figure 3.20 Measurement of caspase activity in primary cortical neurons 136
Figure 3.21 Activation of caspase-12 and cleavage of PARP during U18666A- 137 mediated apoptosis in primary cortical neurons
Figure 3.22 Cleavage of α-fodrin in U18666A-treated cortical neurons 138
Figure 3.23 Distribution of hyperphosphorylated tau in primary cortical neurons 140 by immunocytochemistry
Figure 3.24 Cdk5 in U18666A-treated cortical neurons 142
Figure 3.25 GSK3 in U18666A-treated cortical neurons 143
Figure 3.26 Phosphorylated and non-phosphorylated p44/42 MAPK and 144 SAPK/JNK in U18666A-treated cortical neurons
Figure 3.27 Effect of Y-27632 in U18666A-treated cortical neurons 145
Figure 3.28 Cell cycle regulatory proteins in U18666A-treated cortical neurons 146
Figure 4.1 Determination of RNA integrity 173
Figure 4.2 Biotin-labeled cRNA before and after fragmentation 174
Figure 4.3 Venn diagrams showing the classification of differentially- 187 expressed genes in cortical neurons after 24 h, 48 h and 72 h of U18666A (1 μg/ml) treatment
Figure 4.4 Cluster analysis of microarray data obtained from cortical neurons 188 treated with vehicle (Control) or 1 μg/ml U18666A for a maximum of 72 h
Figure 4.5 Validation of selected gene products by Western blot analysis 191
Figure 5.1 The central dogma 208
Figure 5.2 Strategy for proteomics analysis in U18666A-treated cortical 213 neurons
xiv Figure 5.3 Optimization of sample preparation conditions 220
Figure 5.4 Optimization of sample application method 221
Figure 5.5 Screening of sample quality and protein profile 222
Figure 5.6 Representative silver-stained 2D gels of total proteins from primary 223 cortical neurons
Figure 5.7 Computer-aided differential display 226
Figure 5.8 Enlarged views of 2D gel profiles for 1 μg/ml U18666A treatment 229 from Figure 5.6
Figure 6.1 Proposed mechanism of U18666A-mediated neuronal apoptosis 248
xv List of Abbreviations
2D two-dimensional 3-ATA 3-amino-9-thio(10H)-acridone 8-OH guanine 8-hydroxyguanine oC degree Celsius Aβ β-amyloid AD Alzheimer’s disease AFC 7-amido-4-trifluoromethylcoumarin AMC 7-amido-4-methylcoumarin ANOVA analysis of variance APP amyloid precursor protein APS ammonium persulfate ATM ataxia telangiectasia-mutated protein kinase ATP adenosine triphosphate ATPase adenosine triphosphatase ATR ATM and Rad3-related protein kinase BCA bicinchoninic acid BODIPY 4,4-difluoro-3a,4a-diaza-s-indacene BSA bovine serum albumin BSTFA bis(trimethylsilyl)trifluoracetamide calcein AM calcein acetoxymethylester Cdk cyclin-dependent kinase cDNA complementary DNA CHAPS 3-(3-cholamidoproplyl-dimethylammonio)-1-propanesulfonate CHO Chinese hamster ovary CKI cyclin-dependent kinase inhibitor CoA coenzyme A cRNA complementary RNA DCF 2’,7’-dichlorofluorescein DCFH 2’,7’-dichlorofluorescin DCFH-DA 2’,7’-dichlorofluorescin diacetate DEPC diethylpyrocarbonate DMEM Dulbecco’s modified Eagle’s medium DMSO dimethyl sulfoxide DNA deoxyribonucleic acid DNase deoxyribonuclease DNPH 2,4-dinitrophenylhydrazine dNTP deoxyribonucleotide triphosphate DTT dithiothreitol EDTA ethylenediaminetetraacetate EGTA ethyleneglycol-bis-(β-aminoethylether)N,N,N’,N’-tetraacetate ELISA enzyme-linked immunosorbent assay ER endoplasmic reticulum EthD-1 ethidium homodimer-1
xvi g gram(s) GC-MS gas chromatography-mass spectrometry GSH glutathione GSK glycogen synthase kinase h hour(s) HBSS Hanks’ balanced salts solution HCl hydrochloric acid HEK human embryonal kidney HEPES 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid HMG-CoA 3-hydroxy-3-methylglutaryl coenzyme A IAA iodoacetamide IC50 drug concentration at which 50% of the cultured cells are dead IEF isoelectric focusing IgG immunoglobulin G IPG immobilized pH gradient IVT in vitro transcription JNK c-Jun N-terminal kinase kb kilobase(s) kDa kilodalton(s) l liter(s) LDH lactate dehydrogenase m meter(s) M molar m/z mass-to-charge ratio mA milliampere(s) MALDI-TOF matrix-assisted laser desorption ionization-time of flight MAPK mitogen-activated protein kinase MES 2-(N-morpholino)ethanesulfonic acid mg milligram(s) min minute(s) ml milliliter(s) mm millimeter(s) mM millimolar MOPS 4-morpholinepropanesulfonic acid mRNA messenger ribonucleic acid MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide NaCl sodium chloride NaOH sodium hydroxide NFT neurofibrillary tangles nm nanometer(s) nM nanomolar nmol nanomole(s) NPC Niemann-Pick disease type C NPC1 Niemann-Pick C1 protein NPC2 Niemann-Pick C2 protein PAGE polyacrylamide gel electrophoresis
xvii PARP poly(ADP-ribose)polymerase PBS phosphate-buffered saline pI isoelectric point PVDF polyvinylidene fluoride RIPA radioimmunoprecipitation assay RNA ribonucleic acid RNAi RNA interference RNase ribonuclease ROS reactive oxygen species rpm revolutions per minute of rotor RPMI Roswell Park Memorial Institute s second(s) SAPE streptavidin phycoerythrin SAPK stress-activated protein kinase SCAP SREBP cleavage-activating protein SDS sodium dodecyl sulfate SEM standard error of the mean siRNA short interfering RNA SREBP sterol regulatory element-binding protein TAE Tris-acetate-EDTA TBS Tris-buffered saline TBST Tris-buffered saline containing Tween® 20 TCA trichloroacetic acid TEMED N,N,N’,N’-tetramethyl-ethylenediamine TFA trifluoroacetic acid TMCS trimethylchlorosilane TMRM tetramethylrhodamine methyl ester Tris tris(hydroxymethyl)aminomethane Tween® 20 polyoxyethylene (20) sorbitan monolaurate U18666A 3-β-[2-(diethylamino)ethoxy]androst-5-en-17-one μA microampere(s) μg microgram(s) μl microliter(s) μm micrometer(s) μM micromolar V volts Z-VAD-FMK Z-Val-Ala-Asp(OMe)-fluoromethylketone % (v/v) milliliter(s) per 100 ml % (w/v) gram(s) per 100 ml
xviii List of Publications
Part of this work has been published or submitted for publication:
1. Cheung NS, Koh CHV, Bay BH, Qi RZ, Choy MS, Li Q-T, Wong KP, and Whiteman M. Chronic exposure to U18666A induces apoptosis in cultured murine cortical neurons. Biochemical and Biophysical Research Communications 315:408-417, 2004. Impact factor: 3.000 (2005)
2. Koh CHV, Qi RZ, Qu D, Melendez A, Manikandan J, Bay BH, Duan W, and Cheung NS. U18666A-mediated apoptosis in cultured murine cortical neurons: Role of caspases, calpains and kinases. Cellular Signalling, 2006 (in press). Impact factor: 4.398 (2005)
3. Koh CHV, Whiteman M, Li Q-X, Halliwell B, Jenner AM, Wong BS, Laughton KM, Wenk M, Masters CL, Beart PM, Bernard O, and Cheung NS. Chronic exposure to U18666A is associated with oxidative stress in cultured murine cortical neurons. Journal of Neurochemistry, 2006 (in press). Impact factor: 4.604 (2005)
4. Koh CHV and Cheung NS. Cellular mechanism of U18666A-mediated apoptosis in cultured murine cortical neurons: Bridging Niemann-Pick disease type C and Alzheimer’s disease. Cellular Signalling, 2006 (in press). Impact factor: 4.398 (2005)
5. Koh CHV, Peng ZF, Ou K, Melendez A, Manikandan J, Qi RZ, and Cheung NS. A microarray and proteomics analysis of neuronal apoptosis induced by inhibition of intracellular cholesterol transport in cultured murine cortical neurons. Submitted to Journal of Cellular Physiology, 2006 (under review). Impact factor: 4.362 (2005)
6. Koh CHV, Whiteman M, Halliwell B, Teo TS, Choy MS, Lee AYW, Qi RZ, Wong BS, Jenner AM, Khanna S, and Cheung NS. U18666A-induced neuronal apoptosis is associated with oxidative stress. Conference poster presented at the Society for Neuroscience Annual Meeting, San Diego, 2004.
xix
CHAPTER 1
INTRODUCTION
1 1 Introduction
1.1 General Introduction
The brain is the most cholesterol-rich organ in the body but little is known about the
mechanisms that regulate cholesterol homeostasis in the brain (Pfrieger, 2003; Vance
et al., 2005). Studies suggest that cholesterol imbalance in the brain might be related to the development of neurological disorders such as Niemann-Pick disease type C
(NPC) and Alzheimer’s disease (AD) (Ohm et al., 2003; Michikawa, 2004; Burns and
Duff, 2002; Wolozin, 2004).
NPC is a fatal juvenile neurodegenerative disorder characterized by premature
neuronal cell death and somatically altered cholesterol metabolism (Pentchev et al.,
1995). In the early childhood, patients with NPC exhibit swift neurological decline,
which eventually leads to death in the teen years. Histopathologically, ballooned
neurons and massive neuronal cell loss can be observed in the cortices and cerebellum of brains from NPC patients (Pentchev et al., 1995). The main
biochemical manifestation in NPC is elevated intracellular accumulation of free
cholesterol caused by a genetic deficit in cholesterol trafficking (Ory, 2000;
Mukherjee and Maxfield, 2004). However, the molecular mechanisms of
neurodegeneration in NPC are largely unknown.
2 The pharmacological agent, U18666A (3-beta-[2-(diethylamino)ethoxy]androst-5-en-
17-one), is a well-known class-2 amphiphile which inhibits cholesterol transport
(Liscum and Faust, 1989; Liscum, 1990; Underwood et al., 1996). U18666A has been used in a number of biological systems to mimic the cellular effects of NPC through dysfunction of lipid storage and inhibition of cholesterol movement from the plasma membrane to the endoplasmic reticulum, and from the lysosome to the plasma membrane (Härmälä et al., 1994; Underwood et al., 1998; Lange et al., 1998). Cells treated with this agent accumulate intracellular cholesterol to massive levels, similar to that observed in cells from NPC patients (Lange et al., 2000; Lange et al., 2002;
Neufeld et al., 1999; Liscum and Sturley, 2004). The association between U18666A toxicity and cholesterol accumulation may provide clues to the pathogenesis of NPC.
NPC and AD have some pathological similarities which may share a common underlying cause. AD is one of the most common types of dementia affecting the elderly. A hallmark of AD pathology is the presence of intracellular neurofibrillary tangles (NFT), which are formed by paired helical filaments (Maccioni et al., 2001a).
NFT have also been found to be present in NPC brains (Suzuki et al., 1995; Love et al., 1995). The core component of the paired helical filaments is the microtubule- associated protein tau. It is believed that cellular mechanisms by which tau becomes hyperphosphorylated and assembles into paired helical filaments may actively contribute to the pathogenesis of AD (Kobayashi et al., 2003). Nevertheless, the precise mechanism that leads to the formation of NFT in AD is still not known.
Although tau is phosphorylated by numerous kinases in vivo, recent attention has
3 focused on the ubiquitously-expressed cyclin-dependent kinase-5 (Cdk5) as an important tau kinase, whose activity is enhanced in response to β-amyloid (Aβ) in cultured neurons and AD brains (Alvarez et al., 1999; Li et al., 2003). The dysregulation of Cdk5 causes hyperphosphorylation of tau, thereby contributing to the formation of NFT in AD (Maccioni et al., 2001b).
Neuronal cell death can occur by means of either necrosis or apoptosis. Both necrosis
and apoptosis are generally believed to be distinct mechanisms of cell death with
different characteristic features distinguished on the basis of their morphological and
biochemical properties (Searle et al., 1982; Dive et al., 1992; Allen et al., 1997).
Recent advances in molecular genetics and neurochemistry have also tied neuronal
cell death to oxidative stress. The brain is more vulnerable to oxidative damage than
other organs in the body (Halliwell and Gutteridge, 1999; Emerit et al., 2004).
Exposure to high levels of oxidative stress may lead to cell death via apoptosis or
necrosis. Findings that antioxidant treatment may be beneficial in AD indicate that
oxidative damage is an important factor in the pathological events occurring in this
neurodegenerative disease (Behl and Moosmann, 2002; Casetta et al., 2005; Gibson
and Huang, 2005; Smith et al., 2000; Markesbery, 1997; Pappolla et al., 2002).
Global gene expression profiling in cells or tissues may help to elucidate the
molecular basis of disease pathology, drug treatment or phenotype (Schulze and
Downward, 2001). Recently, the study of gene expression on a global scale using
microarray analysis has received significant attention in the analysis of diseases and
4 the unraveling of cellular signaling pathways in neurobiology (Lockhart and Barlow,
2001; Geschwind, 2000). Gene expression profiling through microarray analysis may provide useful starting points for more in-depth investigation to explore the molecular changes that occur after U18666A treatment in cultured neurons.
Proteomics has become increasingly popular in the study of quantitative changes in global differential protein expression (Blackstock and Weir, 1999; Phizicky et al.,
2003; Tyers and Mann, 2003). A key element of proteomics is the capability to identify protein post-translational modifications, such as phosphorylations and glycosylations, which can affect the biological functions of a protein (Pandey and
Mann, 2000; Dove, 1999). Moreover, proteomics can be used to study protein structures, functions and localizations (Tyers and Mann, 2003; Graves and Haystead,
2002). To further understand the effect of U18666A in cultured neurons, a proteomics approach will be employed by utilizing two-dimensional (2D) gel electrophoresis and mass spectrometry to profile the proteins induced by U18666A. Changes in the 2D snapshots of protein expression will be monitored from different time-points to obtain an overview of U18666A-induced protein expression changes in the cultured neurons.
Analysis of the proteome and the identification of differentially-expressed proteins can more clearly reflect the actual state of biochemical changes in U18666A-treated cultured neurons. Results from the proteomics analysis will be further used to complement those from the microarray analysis to gain an insight into the molecular effects of U18666A and its mechanism of action in cultured neurons.
5 The objectives of this research project are:
1. To examine the chronic effect of U18666A in primary cortical neurons and cell
lines;
2. To investigate the mechanism of neuronal cell injury mediated by U18666A
through inhibition of intracellular cholesterol transport;
3. To study the global gene expression profile in cultured neurons treated with
U18666A through microarray analysis; and
4. To study the global protein expression profile in cultured neurons treated with
U18666A through proteomics analysis.
6 1.2 Literature Review
1.2.1 Overview of the central nervous system
The central nervous system coordinates all neural activities (Brodal, 1998). In vertebrates, it consists of the brain and spinal cord. The brain functions as the main coordinating center for the transmission of electrical nerve impulses to convey information from one part of the body to another. It is also the seat of intelligence and memory. A neuron is one of the basic functional units of the brain. Each neuron has an enlarged cell body, also known as the perikaryon, an axon and several dendrites which extend from the cell body. Another special type of supporting cell, the glia, appears to be responsible for controlling the neuronal environment. However, much of the properties of the glial cell are still not known.
Elucidating the cellular and molecular mechanisms of neuronal cell death can be difficult in vivo due to the confounding effects of intact preparations of the central nervous system. Neuronal in vitro cell cultures thus serve as an alternative means to study specific events pertinent to the neurons in the in vivo state. Primary cultured neurons may provide a reliable model to examine the cellular changes during disease states of the central nervous system.
1.2.2 Neuronal cell death
Neuronal cell death is a process which may be either physiological or pathological.
There are two distinct forms of neuronal cell death, namely apoptosis and necrosis.
7 These types of cell death can be distinguished by morphological criteria and biochemical properties (Searle et al., 1982; Dive et al., 1992; Allen et al., 1997).
Apoptosis, also known as programmed cell death, was originally described in lymphoid cells (Wyllie et al., 1980) but appears to be the primary process by which normal cell death occurs during embryonic development or tissue turnover (Nijhawan et al., 2000; Yeo and Gautier, 2004; Yuan and Yankner, 2000). It is a complex form of cell death which requires energy, activation of cell signaling pathways, gene transcription, and protein synthesis (Hengartner, 2000). Apoptosis involves chromatin and cytoplasmic condensation, progressive contraction of cell volume, and preservation of the integrity of membrane-bound particles (Allen et al., 1997; Wyllie et al., 1980). Apoptosis affects single cells and does not incur an inflammatory response.
On the other hand, necrosis is generally characterized by early swelling of the cell and culminates in the rupture of organelles and the plasma membrane, releasing intracellular contents into the extracellular space. Necrosis occurs following severe cellular injury and is usually associated with an inflammatory response (Searle et al.,
1982; Dive et al., 1992).
1.2.2.1 Cell death in neurodegeneration
The vulnerability of the central nervous system to damage by various insults and neurodegenerative conditions is the result of multiple specific processes, many of which are related to the cellular signaling events specific to the brain. Neuronal cell death via apoptosis is believed to be one of the chief events involved in the excessive
8 neuronal cell loss and neurodegeneration observed in Alzheimer’s disease (AD) and other neurodegenerative disorders (Mattson, 2000; Friedlander, 2003; Eckert et al.,
2003; Cotman and Anderson, 1995; Nijhawan et al., 2000; Yuan and Yankner, 2000).
Given that tissues from the central nervous system have almost no regenerative capability, it is of utmost importance to limit the damage caused by neuronal cell death. Delineating the exact processes can likely facilitate the development of therapeutic interventions for neuroprotection.
1.2.3 Niemann-Pick disease type C (NPC)
1.2.3.1 Overview of Niemann-Pick diseases
Niemann-Pick diseases are a group of clinically similar lysosomal lipidoses that have been divided into three different subtypes. Niemann-Pick disease types A (NPA) and
B (NPB) are caused by a deficiency of the enzyme sphingomyelinase. NPA represents a classical acute neuropathic form of the disease, whereas NPB is a chronic form without nervous system involvement. Niemann-Pick disease type C (NPC) is clinically similar to NPA and NPB; however, sphingomyelinase is functional in NPC
(Kolodny, 2000).
1.2.3.2 Characteristics of NPC
NPC is a panethnic autosomal recessive neurodegenerative disorder distinguished by a unique error in cellular trafficking of exogenous cholesterol that is associated with lysosomal accumulation of unesterified cholesterol (Vanier and Millat, 2003). Most
9 patients with NPC have progressive neurological disease, although hepatic damage is prominent in certain cases. The age of onset can vary from early infancy to adulthood and clinical manifestations are heterogeneous. The infantile form of NPC progresses rapidly and patients usually die from cholestasis or liver failure before two years old.
In contrast, the late-onset form of NPC is slowly progressive, with the first symptoms occurring in the adolescence or adulthood (Pentchev et al., 1995).
NPC patients with the classic phenotype present symptoms such as variable hepatosplenomegaly, vertical supranuclear ophthalmoplegia, ataxia, dystonia, seizures, and progressive dementia. Other phenotypes include fatal neonatal liver disease, early infantile onset with hypotonia, and delayed motor development.
Neuronal storage with cytoplasmic ballooning and a variety of inclusions are also present throughout the nervous system. In addition, levels of unesterified cholesterol, sphingomyelin, phospholipids and glycolipids are elevated. In the brain, progressive neuronal cell loss, demyelination, and formation of neurofibrillary tangles (NFT) also occur (Pentchev et al., 1995).
1.2.3.3 NPC and inhibition of intracellular cholesterol transport
Intracellular cholesterol transport and regulatory mechanisms of cholesterol homeostasis are severely impaired in NPC. These defects appear to affect not just lysosomal cholesterol, but also the cholesterol in the endoplasmic reticulum (ER)
(Ory, 2000; Mukherjee and Maxfield, 2004). Due to the block in transport, unesterified free cholesterol accumulates in large amounts in the late endosomes or lysosomes, and as a consequence, cannot reach the regulatory sites of cellular
10 cholesterol homeostasis in the ER (Ory, 2000; Mukherjee and Maxfield, 2004).
Cholesterol is found to accumulate in the late endosomes and trans-Golgi network of the NPC cells (Mukherjee and Maxfield, 2004). As a secondary effect, which probably results from cholesterol accumulation, other lipid species may also accumulate in the NPC cells. Although the total cholesterol mass in the cell bodies of
NPC neurons was higher than in normal cells, the mass in the distal axons was reduced (Karten et al., 2002).
1.2.3.4 Niemann-Pick C proteins
The Niemann-Pick C1 protein (NPC1) is a 1278 amino-acid membrane glycoprotein located primarily in the late endosomes and contains a putative sterol-sensing domain common to other proteins involved in cholesterol homeostasis (Scott and Ioannou,
2004). The NPC1 protein is also post-translationally modified to produce the final tertiary structure (Carstea et al., 1997; Loftus et al., 1997). The NPC1 protein resides in distinct sets of vesicular compartments which interact with cholesterol-filled lysosomes (Neufeld et al., 1999). This interaction between NPC1-containing vesicles and cholesterol-filled lysosomes is required for proper trafficking of cholesterol
(Scott and Ioannou, 2004). Based on the cellular phenotype of NPC, NPC1 is evidently critical for the movement of free cholesterol and other cargo out of the late endosomes or lysosomes (Neufeld et al., 1999). Despite this, the precise function of
NPC1 remains a mystery.
The Niemann-Pick C2 protein (NPC2), initially characterized as a secretory protein in the human epididymis (HE1), codes for a 151 amino-acid glycoprotein
11 which binds cholesterol (Naureckiene et al., 2000). NPC2 resides in the trans-Golgi network and late endosomes. The exact mechanism of NPC2 is still not known, although it is postulated to act as a carrier that regulates the storage of cholesterol in the plasma membrane.
Cells with dysfunctional NPC1 or NPC2 accumulate unesterified cholesterol in the late endosomes, which reflect a failure of cholesterol to efficiently exit this compartment and be transported to the plasma membrane and ER (Liscum and
Sturley, 2004). This may result in membrane lipid redistribution, changes in membrane elasticity and fluidity, and aberrant protein sorting and processing
(Liscum, 2000; Mukherjee and Maxfield, 1999). Much needs to be learned about the functions and interactions of NPC1 and NPC2. Elucidating these may help to resolve
the mysteries of the molecular basis of NPC pathology and lead to the development of treatment strategies for this neurodegenerative disease.
1.2.4 Cholesterol
Cholesterol, which can exist as a free sterol or esterified with a long-chain fatty acid,
has received the most attention in studies of NPC. Over past decades, this sterol has been intensively studied and many important findings related to its functions have been made. Cholesterol is an essential structural component of cell membranes and may act as a co-factor for signaling molecules. It is also the precursor of bile acids and many biologically active steroids such as steroid hormones and vitamin D
(Pfrieger, 2003; Horton et al., 1996).
12 1.2.4.1 Cholesterol biosynthesis
The first milestone in the elucidation of cholesterol synthesis was the discovery that all the carbons in cholesterol arises from acetyl-coenzyme A (CoA). It was also discovered that squalene, a 30-carbon linear hydrocarbon, is an intermediate in the biosynthesis of the cholesterol molecule and that squalene itself is formed from 5- carbon units (Horton et al., 1996). Cholesterol is a planar four-ring structure composed of 27 carbon atoms with a hydroxyl functional group attached to carbon-3 in β-configuration (Horton et al., 1996). The biosynthesis of cholesterol from acetyl-
CoA takes place in the ER. The rate-limiting enzyme which catalyzes the conversion of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) to mevalonate is HMG-
CoA reductase (Rudney and Sexton, 1986). This reaction is followed by several steps catalyzed by over 30 enzymes, leading to the final product cholesterol (Figure 1.1).
1.2.4.2 Cholesterol homeostasis in the brain
Cholesterol, a molecule that is essential for life, is also associated with a leading cause of death. The brain expresses the machinery required for cholesterol uptake, intracellular trafficking, de novo synthesis, and metabolic degradation (Pfrieger,
2003). To satisfy its massive cholesterol requirements, the brain produces its own cholesterol via the standard cholesterol biosynthesis pathway (Figure 1.1).
The correct intracellular distribution of cholesterol is crucial for proper functioning of the neurons (Pfrieger, 2003). The concentration of cholesterol in the
ER, where it is synthesized, is low. The level increases from the cis- to the trans-
Golgi apparatus. Compared to other cellular membranes, the plasma membrane is
13 highly enriched in cholesterol. Excessive amounts of cholesterol are toxic to neurons and thus cholesterol content has to be tightly regulated (Pfrieger, 2003). Significant progress has been made in the understanding of cholesterol homeostasis and the cellular mechanisms that protect neuronal cells from excess cholesterol. Neurons have evolved several ways to regulate their cholesterol content through transcriptional, translational and post-translational mechanisms (Pfrieger, 2003).
Intracellular cholesterol transport is essential for the maintenance of cholesterol homeostasis (Lange and Steck, 1996). Recently, several clinical and biochemical studies suggest that cholesterol imbalance and failure of homeostatic responses in the brain can lead to the development of neurological disorders such as AD and NPC
(Wolozin, 2004; Burns and Duff, 2002; Nixon, 2004).
Figure 1.1: A simplified cholesterol biosynthesis pathway (modified from Rudney and Sexton, 1986).
14 1.2.5 Alzheimer’s disease (AD)
1.2.5.1 Molecular pathology of AD
Alzheimer’s disease (AD) was first described in 1911 by Alois Alzheimer as a neuropsychiatric disorder affecting the elderly, and is characterized by a gradual age- related decline of mental functions that ultimately results in memory loss, disorientation, neglect of routine tasks, loss of language skills, and personality changes (Maccioni et al., 2001a).
Patients with AD are quite similar in brain pathology and clinical manifestations. However, they may differ in terms of causal factors. Genetics appears to play a variable role, but may be especially important in patients with an early onset of symptoms, who have inherited AD dominantly (Maccioni et al., 2001a). In a subgroup of patients, a mutation of the gene coding for APP might suffice to cause
AD. Another candidate is the gene coding for apolipoprotein, which plays a role in cholesterol transport. Although apolipoproteins are present in genetically different forms, only the apolipoprotein E4 variety is associated with AD (Ji et al., 2002;
Dolev and Michaelson, 2004).
The two pathological hallmarks of AD are intracellular NFT and extracellular amyloid plaques. The NFT consist of thickened intraneuronal fibrils whereas the senile plaques consist of extracellular deposits of β-amyloid (Aβ). Aβ is formed from a large membrane protein known as the amyloid precursor protein (APP) (Selkoe,
2000). The intraneuronal fibrils seen in NFT are made up of paired helical filaments,
15 in which a major component is the microtubule-associated protein tau in hyperphosphorylation.
1.2.5.2 Cleavage of APP into Aβ
The major constituent of the amyloid plaque is a peptide known as Aβ, which arises from the cleavage of APP, a glycosylated, single transmembrane protein (Selkoe,
2000). APP is first cleaved by either α- or β-secretase in the extracellular space to yield a secreted N-terminal fragment, as well as a membrane-bound C-terminal fragment (Figure 1.2). The C-terminal fragment is subsequently cleaved in the Golgi- endosomal network by γ-secretase within the intramembrane domain between 37-43 amino acids C-terminal to the β-secretase site. The γ-secretase products are amyloidogenic Aβ or non-amyloidogenic p3 peptide. The most abundant β/γ- secretase-cleaved products are Aβ40 and Aβ42. The initial choice between α- or β- secretase has dramatic consequences for AD. Cleavage by β/γ-secretase results in Aβ production, whereas cleavage by α-secretase within the Aβ sequence results in the production of non-amyloidogenic p3.
1.2.5.3 Cholesterol and AD
The fact that APP processing occurs within cholesterol-rich membrane domains, and that APP and β- and γ-secretases are localized in cholesterol-rich lipid rafts, suggest that alterations in cholesterol homeostasis may affect the development of AD via an increase in Aβ production and accumulation (Wolozin, 2001; Puglielli et al., 2001;
Wolozin, 2004). The role of cholesterol in the pathogenesis of AD came to the
16 forefront in 2000, when two groups independently reported that subjects treated with
statins, also known as HMG-CoA reductase inhibitors, had a significant decreased
prevalence of AD (Jick et al., 2000; Wolozin et al., 2000). Statins inhibit the
synthesis of mevalonate, which is a rate-limiting metabolite produced early in the
cholesterol biosynthesis pathway (Figure 1.1). Studies suggest that elevated levels of
cholesterol might be associated with the development of AD (Jick et al., 2000;
Wolozin et al., 2000; Fassbender et al., 2001).
Figure 1.2: Proteolytic cleavages of APP into Aβ peptides (modified from Selkoe, 2000).
17 1.2.6 Bridging NPC and AD
Although NPC differs in major aspects from AD, intriguing parallels exist in the cellular pathology of these two diseases. In the juvenile neurodegenerative disease
NPC, NFT develop in significant numbers (Suzuki et al., 1995; Love et al., 1995) and contain the paired helical form of hyperphosphorylated tau (Sawamura et al., 2001;
Distl et al., 2003), which is structurally and immunologically similar to that in AD tangles (Ohm et al., 2003; Treiber-Held et al., 2003). Some neuronal populations in
NPC also develop endosome abnormalities resembling those seen in the earlier stages of AD, and accumulate cleaved APP and Aβ40 and Aβ42 peptides within the endosomes (Yamazaki et al., 2001; Jin et al., 2004; Nixon, 2004). Many lines of evidence suggest that alterations in cholesterol homeostasis modulate Aβ production and accumulation (Wolozin, 2001; Puglielli et al., 2001; Wolozin, 2004). It is conceivable that inhibition of cholesterol trafficking and disturbed membrane lipid distribution in NPC lead to aberrant APP processing and Aβ generation. The convergence of many factors that promote neurodegeneration in NPC and AD suggests an overlap of similar cellular signaling mechanisms operating in these two neurodegenerative diseases.
1.2.7 U18666A: The widely-studied inhibitor of cholesterol transport
1.2.7.1 Overview of hydrophobic amines and steroids
Several pharmacological agents have been reported to interfere with the trafficking of cholesterol when tested in intact cells. Three of these agents are progesterone (Butler
18 et al., 1992), the androstene derivative U18666A (Liscum and Faust, 1989), and imipramine (Rodriguez-Lafrasse et al., 1990). Other compounds capable of blocking cholesterol transport include stearylamine, RV-538 and sphinganine (Roff et al.,
1991), which cause unesterified cholesterol to accumulate in perinuclear vesicles.
These cholesterol-laden vesicles can be stained with filipin to give a staining pattern indistinguishable from that seen in NPC fibroblasts. A target for these hydrophobic amines and steroids that inhibit intracellular cholesterol transport may be the NPC1 protein (Neufeld et al., 1999).
1.2.7.2 Class-2 amphiphiles
A wide variety of pharmacological agents grouped under the class-2 amphiphiles can cause cells to elicit an increase in intracellular free cholesterol, presumably through a decrease in cholesterol esterification (Lange and Steck, 1994; Lange et al., 1999;
Liscum and Faust, 1989). However, the molecular mechanism through which the class-2 amphiphilic compounds exert their effects is unknown. On the other hand, class-1 agents include exogenous oxysterols which stimulate cholesterol esterification and down-regulate the activity of HMG-CoA reductase (Lange and Steck, 1994).
1.2.7.3 Studied effects of U18666A
The best characterized drug inducing the NPC phenotype in normal cells is the hydrophobic amine U18666A (3β-[2-(diethylamino)ethoxy]androst-5-en-17-one).
U18666A is a chemical initially developed by Upjohn Company (Kalamazoo, MI,
USA) as a potential hypocholesterolemic drug, but was withdrawn due to serious
19 side-effects. U18666A contains a hydrophobic ring structure associated with a hydrophilic side-chain and a charged cationic amine group.
U18666A is first reported to inhibit desmosterol reductase (Phillips and
Avigan, 1963; Cenedella and Bierkamper, 1979), the enzyme responsible for the reduction of desmosterol to cholesterol (Figure 1.1). Besides this, U18666A also inhibits oxidosqualene cyclase (Cenedella, 1980; Sexton et al., 1983; Duriatti et al.,
1985). U18666A has been intensively used in the studies of HMG-CoA reductase
(Boogaard et al., 1987; Cohen and Griffioen, 1988) and squalene synthetase (Cohen et al., 1989; Cohen et al., 1992). The mechanism of each enzyme inhibition remains unclear, although it has been reported that the effects of U18666A on the cholesterol biosynthesis pathway is dependent on its concentration (Cenedella, 1980).
U18666A is well-known to inhibit intracellular cholesterol transport (Liscum and Faust, 1989; Liscum, 1990; Härmälä et al., 1994; Sparrow et al., 1999;
Underwood et al., 1996). The accumulation of free cholesterol by U18666A has been exploited in the macrophage model system for foam cell death (Tangirala et al., 1993;
Kellner-Weibel et al., 1998). U18666A decreases the cholesterol content of the ER and impedes the movement of cholesterol between the lysosomal compartment and the plasma membrane, and from the plasma membrane to other intracellular compartments (Lange et al., 1999; Lange and Steck, 1994). Although U18666A has been assumed to interact with the NPC1 protein (Neufeld et al., 1999), the inhibition in cholesterol trafficking may occur through a more general effect due to the structural diversity of the amphiphiles that induce the NPC phenotype (Lange et al.,
2000).
20 Apart from cholesterol, U18666A potently alters the trafficking of a variety of intracellular membrane proteins, including ganglioside GM1 (Sofer and Futerman,
1995), NPC1 (Lange et al., 2000), insulin growth factor 2/mannose 6-phosphate receptor (Kobayashi et al., 1999) and CD63 lysosome-associated membrane protein-3
(Kobayashi et al., 2000). In melanocytes, it has been shown to alter the subcellular localization of tyrosinase and inhibit melanin synthesis (Hall et al., 2003). U18666A
has also been shown to disrupt Sonic hedgehog signaling (Incardona et al., 2000).
Treatment of the opossum (Jurgelski Jr et al., 1973) and neonatal rats
(Bierkamper and Cenedella, 1978) with U18666A has been shown to result in the
development of chronic epileptiform activity, which is associated with alterations in
the lipid composition of critical neuronal structures (Sarkar et al., 1982; Cenedella
and Sarkar, 1984), localization of U18666A in lipid-rich membranes (Cenedella et
al., 1982), and lipid peroxidation (Ehler and Bierkamper, 1986). The potent inhibition
of cholesteryl ester formation by U18666A has been proposed to contribute to the
induction of epilepsy (Jeng et al., 1985). Ubiquinone biosynthesis was previously
reported to be insensitive to U18666A and unlikely to be responsible for the
epileptogenicity of this drug (Jeng et al., 1984), even though it was reported that
U18666A induced an abnormally high accumulation of ubiquinone in glioblastoma
cells (Volpe and Obert, 1982).
In addition to altering brain chemistry, U18666A has also been reported to
induce permanent cataracts in rats (Cenedella and Bierkamper, 1979), recently
hypothesized to be due to direct perturbation of the lens membrane structure
(Cenedella et al., 2004). Lipid peroxidation (Yadav and Rawal, 1992), calcium-
21 activated proteolysis and protein modification (Chandrasekher and Cenedella, 1993), increased concentrations of insoluble proteins (Alcala et al., 1985), and loss of identifiable gap junctions (Kuszak et al., 1988; Fleschner and Cenedella, 1988) are observed in the U18666A cataract. In contrast, U18666A treatment has been found to cause no major changes in the concentration or composition of phospholipids in the lens (Cenedella, 1985). U18666A has also been reported to induce myotonia, an abnormally delayed relaxation of the skeletal muscle fibers, in rats (Winer et al.,
1966). One interesting study in plants even showed that U18666A caused rapid necrosis of newly-emerged leaves and halted new leaf formation, in addition to inhibition of sterol biosynthesis (Fenner and Raphiou, 1995).
In vivo and in vitro studies using U18666A generate at least three different models of human diseases, namely NPC, epilepsy and cataracts. Multiple mechanisms may underlie the induction of these diverse conditions. Understanding how a single inhibitor of cholesterol transport and sterol metabolism can induce such varied models of disorders is a long-term goal of these investigations.
22
CHAPTER 2
MATERIALS AND METHODS
23 2 Materials and Methods
2.1 Materials
2.1.1 Buffers and solutions
Distilled water passed through a purification system (Milli-Q water; Millipore
Corporation, Bedford, MA, USA) was used for all purposes. Most stock solutions were purchased from the Laboratory Supplies Store at the National University
Medical Institutes (NUMI): phosphate-buffered saline (PBS) was supplied as 10× stock solution, Tris-HCl stock solutions were supplied as 1 M, sodium chloride
(NaCl) stock solution was supplied as 5 M, ethylenediaminetetraacetate (EDTA) stock solution was supplied as 0.5 M, and sodium dodecyl sulfate (SDS) stock solution was supplied as 10% (w/v). Stock 5× electrophoresis buffer and 10× Tris- acetate-EDTA (TAE) buffer were also from NUMI.
2.1.2 Consumables
Multiwell plates for cell culture were from Nunc (Roskilde, Denmark). Disposable 15
ml and 50 ml centrifuge tubes (FALCON), and disposable 3 ml syringes with 22½-
gauge needles were from Becton Dickinson (Franklin Lakes, NJ, USA). Disposable
1.5 ml and 2 ml microfuge tubes were from Eppendorf (Hamburg, Germany).
Nalgene® disposable filter units for sterile filtration were from Nalge Nunc
International (Rochester, NY, USA). Cell scrapers were from Techno Plastic Products
(TPP; Zollstrasse, Schweiz).
24 2.1.3 Drugs
U18666A (#S-200) and Z-VAD-FMK (a broad spectrum caspase inhibitor; #P-416) were from Biomol Research Laboratories (Plymouth Meeting, PA, USA). Methyl-β- cyclodextrin (#C4555), lovastatin (also known as mevinolin; #M2147), lithium chloride (an inhibitor of GSK3; #L7026) and (+/-)-α-tocopherol (vitamin E; #T3251) were from Sigma (Saint Louis, MO, USA). Pravastatin was from LKT Laboratories
(Saint Paul, MN, USA, #P6801). Caspase-3 inhibitor IV (#235421), γ-secretase inhibitor IX (#565770) and Y-27632 (an inhibitor of Rho-associated protein kinases;
#688000) were from Calbiochem (La Jolla, CA, USA). Calpeptin, a selective inhibitor of calpain, was from Merck (Darmstadt, Germany, #03-34-0051). PD98059, a selective inhibitor of mitogen-activated protein kinase (MAPK) kinase, was from
Promega Corporation (Madison, WI, USA, #V1191). The Cdk4 inhibitor, 3-amino-9- thio(10H)-acridone (3-ATA), was from Alexis® Biochemicals (Carlsbad, CA, USA,
#350-273-M005). Roscovitine, a specific inhibitor of Cdk5, was generously provided
by Dr. Robert Qi (Department of Biochemistry, Hong Kong University of Science
and Technology).
2.1.4 Primary antibodies
Anti-active caspase-3 (#557035) was from PharMingen (Becton Dickinson). Anti-
cleaved poly(ADP-ribose)polymerase (PARP; #9544), anti-p44/42 MAPK (#9102),
anti-phospho-p44/42 MAPK (Thr202/Tyr204; #9101), anti-stress-activated protein
kinase/c-Jun N-terminal kinase (SAPK/JNK; #9252), anti-phospho-SAPK/JNK
(Thr183/Tyr185; #9251), anti-phospho-p53 (Ser15; #9284), anti-Cdk4 (#2906), anti-
25 p27 Kip1 (#2552) and anti-cyclin D3 (#2936) were from Cell Signaling Technology
(Beverly, MA, USA). Anti-p35 (C-19; #sc-820), anti-Cdk5 (#sc-173), anti-Atf3 (#sc-
188), anti-histone 1 (#sc-8030) and anti-GADD 153 (#sc-575) were from Santa Cruz
Biotechnology (Santa Cruz, CA, USA). Anti-AT8 (#90206) and anti-AT180 (#90337) were from Innogenetics (Gent, Belgium). Anti-Tau-1 (#MAB3420) and anti-spectrin
(#MAB1622) were from Chemicon International (Temecula, CA, USA). Anti-GSK-
3α/β [pY279/216] (#44-604) was from BioSource International (Camarillo, CA, USA).
Anti-caspase-12 (#3182-100) was from BioVision (Mountain View, CA, USA). Anti-
cathepsin B (#IM27L) was from Calbiochem. Anti-cyclophilin A (#07-313) was from
Upstate (Charlottesville, VA, USA). Anti-β-tubulin (#ATN01) was from
Cytoskeleton (Denver, CO, USA). Anti-PHF1 was generously provided by Dr. Peter
Davies (Department of Pathology and Neuroscience, Albert Einstein College of
Medicine). Anti-α-synuclein was generously provided by Professor Colin Masters
(Department of Pathology, University of Melbourne). All were used at a working
dilution of 1:1000 in Western blotting or 1:100 in immunocytochemistry, unless
otherwise stated.
2.1.5 Secondary antibodies
Blotting grade goat anti-rabbit IgG conjugated with horseradish peroxidase was from
Bio-Rad Laboratories (Hercules, CA, USA, #170-6515) and ImmunoPure® goat anti-
mouse IgG conjugated with horseradish peroxidase was from Pierce Biotechnology
(Rockford, IL, USA, #31430). Both were used at a working dilution of 1:10,000 in
Western blotting. Alexa Fluor® 488 goat anti-mouse IgG was from Molecular Probes
26 (Eugene, OR, USA, #A-11029) and used at a working dilution of 1:1500 in immunocytochemistry.
2.1.6 Protein assay kits
The RC DC protein assay kit based on the Lowry method (#500-0120) and protein assay dye reagent concentrate based on the Bradford method (#500-0006) were from
Bio-Rad Laboratories. The bicinchoninic acid (BCA) protein assay reagent kit
(#23227) was from Pierce Biotechnology. The 2D Quant Kit (#80-6483-56) was from
Amersham Biosciences (Piscataway, NJ, USA). Protein standards were freshly prepared by diluting 2 mg/ml bovine serum albumin (BSA; Sigma, #A4503) stock standard in the same diluent as the samples. Assays were performed according to their respective manufacturer’s instructions.
27 2.2 Methods
2.2.1 Preparation of mouse primary cortical neurons
Cortices were aseptically microdissected from the brains of embryonic (gestational day 15-16) Swiss-white mice and collected in Solution 1 (refer to Appendix I-B) by centrifugation (Eppendorf 5810R, Hamburg, Germany) at 1000 rpm for 1 min at
15oC. The cortices were then subjected to trypsin digestion at 37oC for 5 min in
Solution 2 (refer to Appendix I-C), followed by partial trypsin inactivation upon
addition of Solution 4 (refer to Appendix I-E) and centrifugation at 1000 rpm for 5
min at 15oC. After which, the pelleted tissues were subjected to mechanical trituration
in Solution 3 (refer to Appendix I-D) and the dissociated cells were harvested by
centrifugation at 1000 rpm for 5 min at 15oC. The cells were then resuspended in
Neurobasal™ medium (Invitrogen, Carlsbad, CA, USA, #21103-049) supplemented
with 10% (v/v) dialyzed heat-inactivated fetal bovine serum (Sigma, #F0392), 2.5%
(v/v) B27 (Invitrogen, #17504-044; refer to Appendix I-F for composition), 0.25%
(v/v) GlutaMAX™-1 (Gibco, Grand Island, NY, USA, #35050-061) and 1% (v/v) penicillin-streptomycin (Sigma, #P3539) (Brewer et al., 1993). Cells were seeded at densities of 2.25 × 106 cells/well in 6-well plates, 0.5 × 106 cells/well in 24-well
plates, 0.125 × 106 cells/well in 96-well plates or 0.5 × 106 cells/well in 4-well Lab-
Tek chambered coverglasses (Nalge Nunc International, #155383), all previously coated with 0.1 mg/ml sterile poly-D-lysine (Sigma, #P0899). After 24 h in vitro, the
culture medium was replaced with Neurobasal™ medium containing 2.5% (v/v) B27,
0.25% (v/v) GlutaMAX™-1 and 1% (v/v) penicillin-streptomycin. Cultures were
28 maintained in a humidified incubator (Binder, Tuttlingen, Germany) at 37oC in 5%
CO2 until drug treatment in neat Neurobasal™ medium on day 5 in vitro.
Immunocytochemistry performed for microtubule-associated protein-2 and glial
fibrillary acidic protein indicated that more than 95% of the cells were neurons with
minimal contamination by glia (Cheung et al., 1998). All experiments involving
animals were approved by the Institutional Animal Care and Use Committee of the
National University of Singapore.
2.2.2 Culture and maintenance of adherent cell lines
COS-7 from African green monkey (American Type Culture Collection, ATCC;
Manassas, VA, USA, #CRL-1651), human embryonal kidney (HEK) 293 (ATCC,
#CRL-1573), NIH/3T3 mouse fibroblasts (ATCC, #CRL-1658) and Neuro-2a mouse
neuroblastoma cells (ATCC, #CCL-131) were cultured in Dulbecco’s modified
Eagle’s medium (DMEM; Sigma, #D1152) containing 10% (v/v) heat-inactivated
fetal bovine serum (Invitrogen, #10270-098) and 1% (v/v) penicillin-streptomycin.
Wild-type, mutant 25-RA and mutant CT43 Chinese hamster ovary (CHO) cell lines,
generously provided by Professor Chang Ta-Yuan (Department of Biochemistry,
Dartmouth Medical School), were cultured in F-12 nutrient mixture (Ham; Gibco,
#21700-075) containing 10% (v/v) heat-inactivated fetal bovine serum (Invitrogen)
and 10 μg/ml gentamycin (Gibco, #15710-064). The 25-RA CHO contains a gain-of-
function mutation in the sterol regulatory element-binding protein (SREBP) cleavage- activating protein (SCAP), while the CT43 CHO is isolated as one of the cholesterol-
trafficking mutants from mutagenized 25-RA cells and produces a non-functional
29 NPC1 protein (Cadigan et al., 1990; Dahl et al., 1992). The rat pheochromocytoma- derived PC12 (ATCC, #CRL-1721) was cultured in DMEM containing 10% (v/v) heat-inactivated fetal bovine serum (Invitrogen), 5% (v/v) heat-inactivated horse serum (Sigma, #H1270) and 1% (v/v) penicillin-streptomycin. All cell lines were maintained in their respective culture media in a humidified incubator (Binder) at
o 37 C in 5% CO2 and used not more than ten passages. Cells were seeded overnight at
a density of 0.5 × 106 cells/well in 24-well plates before drug treatment.
Differentiation of PC12, seeded overnight at a density of 0.25 × 106 cells/well in 24-
well plates previously coated with 0.1 mg/ml sterile poly-D-lysine, was induced in
DMEM containing 1% (v/v) heat-inactivated fetal bovine serum (Invitrogen), 0.5%
(v/v) heat-inactivated horse serum, 100 ng/ml nerve growth factor (Calbiochem,
#480354) and 1% (v/v) penicillin-streptomycin before drug treatment.
2.2.3 MTT assay
The MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay is
based on the conversion of the yellow tetrazolium salt to the purple formazan product
by mitochondrial enzymes present in viable cells (Sladowski et al., 1993). MTT
(Duchefa, Haarlem, the Netherlands, #M1415) was dissolved in RPMI-1640 without
phenol red (Gibco, #11835-030) at a stock concentration of 5 mg/ml and filtered to
remove any insoluble residues. MTT solution equal to 10% (v/v) of the culture medium volume in each well was directly added to treated cells (0.5 mg/ml final
MTT concentration) and incubated for 30 min at 37oC. The solution was then
removed and the formazan product was dissolved in 200 μl dimethyl sulfoxide
30 (DMSO; Sigma, #D8418). Absorbance, which is proportional to the number of viable cells, was read at 570 nm (Tecan Ultra 384, Grödig/Salzburg, Austria) and results were expressed as percent MTT reduction as compared to the vehicle-treated Control.
2.2.4 Release of lactate dehydrogenase (LDH)
Culture media from cortical neurons treated with vehicle or 0.1-2.5 μg/ml U18666A for 72 h were transferred into a 96-well flat-bottomed plate, LDH reagent (Sigma,
#228-10) was added, and absorbances were immediately measured at 340 nm
(SPECTRAmax® 190; Molecular Devices, Sunnyvale, CA, USA). The rate of
increase in absorbance at 340 nm is directly proportional to LDH activity in the
sample. Results were compared as a percentage of total LDH released from cells
lysed with 0.1% (v/v) Triton X-100 (BDH, Dorset, England, #306324N) for 20 min at
37°C. Complete lysis in cortical neurons treated with Triton X-100 was confirmed
through observation under the microscope.
2.2.5 LIVE/DEAD® viability and cytotoxicity assay
The LIVE/DEAD® viability and cytotoxicity assay kit (Molecular Probes, #L-3224)
provides a two-color fluorescence assay based on the simultaneous determination of
live and dead cells using two probes: calcein acetoxymethylester (calcein AM) and
ethidium homodimer-1 (EthD-1). The cell-permeable calcein AM is converted by
intracellular esterases to produce a green fluorescence in live cells. On the other hand,
EthD-1 only enters cells with damaged membranes and binds to nucleic acids to produce a red fluorescence in dead cells. Cortical neurons treated with vehicle or
31 2.5 μg/ml U18666A for 72 h were loaded with 0.1 μM each of calcein AM and EthD-
1, and incubated for 1 h at 37oC. Digital images were acquired using a fluorescence
microscope (Leica DM IRB, Cambridge, United Kingdom).
2.2.6 Annexin V staining
Cortical neurons treated with vehicle or 1 μg/ml U18666A for 72 h were fixed with
4% (w/v) paraformaldehyde (Sigma, #P6148) for 30 min at room temperature and
washed thrice with 1× PBS before labeling with 0.5 μg/ml annexin V-FITC conjugate
(Sigma, #A9210) for 10 min at room temperature. During apoptosis, loss of
phospholipid asymmetry results in phosphatidylserine translocation to the external
portion of the plasma membrane. The phosphatidylserine thus becomes available for the annexin V-FITC fluorescent probe to bind to. Digital images were acquired using
a fluorescence microscope (Leica DM IRB).
2.2.7 Hoechst staining
Nuclear morphology. Cortical neurons treated with vehicle or 1 μg/ml U18666A for
72 h were fixed with 4% (w/v) paraformaldehyde (Sigma) for 30 min at room
temperature and washed thrice with 1× PBS before labeling with 0.5 μg/ml Hoechst
33258 (Sigma, #B2883) for 20 min at room temperature. Stained nuclei were
observed and digital images were acquired using a fluorescence microscope (Leica
DM IRB).
Quantitative assessment. Cortical neurons treated with vehicle or 0.1-2.5 μg/ml
U18666A for 72 h were fixed and labeled as described above. Nuclear morphology
32 was observed and the number of apoptotic nuclei, which look smaller and have
condensed chromatin as compared to the normal, were counted and expressed as a
percent of the total number of cells in each field. Four optical fields containing
approximately 200 cells each were selected for the cell counting.
2.2.8 Transmission electron microscopy
Cortical neurons in chambered coverglasses treated with vehicle or 1 μg/ml U18666A
for 72 h were fixed with 3% (w/v) glutaraldehyde (Agar Scientific, Cambridge,
England, #R1010) and 2% (w/v) paraformaldehyde (Merck, #1.04005) in 0.1 M phosphate buffer for 30 min at room temperature. After osmication in 2% (w/v) osmium tetroxide (Electron Microscopy Sciences, Hatfield, PA, USA, #19100), specimens were dehydrated in an ascending series (25%, 50%, 75%, 95% and 100%) of ethanol and embedded in araldite (Electron Microscopy Sciences, #10900). Ultra- thin sections (70 nm) were cut, mounted on formvar-coated copper grids (Canemco
and Marivac, Quebec, Canada, #18086), and double-stained with uranyl acetate
(TAAB Laboratories Equipment, Berkshire, England, #U006) and lead citrate (BDH,
#007147015) before viewing using a Phillips BioTwin CM120 transmission electron microscope (Phillips Electron Optics, Eindhoven, the Netherlands).
2.2.9 Preparation of protein lysates for Western blotting
Cytosolic protein. Culture media from vehicle- and drug-treated cortical neurons
were removed after 24 h, 48 h and 72 h of treatment, and 2.25 × 106 cells were
scraped into 150 μl of cold radioimmunoprecipitation assay (RIPA) buffer (refer to
33 Appendix II-A). Lysates were then transferred to microfuge tubes, incubated at -20oC
overnight, thawed on ice, and centrifuged (Eppendorf 5415R, Hamburg, Germany) at
maximum speed for 10 min at 4oC. The supernatants were retained and protein
concentrations were determined by the Lowry method (Bio-Rad Laboratories).
Total protein. Culture media from vehicle- and drug-treated cortical neurons were
removed after 24 h, 48 h and 72 h of treatment, and cells were washed with cold 1×
PBS. After which, 2.25 × 106 cells were scraped into 200 μl of boiling lysis buffer
comprising 10 mM Tris-HCl pH 7.4, 1 mM sodium orthovanadate (Alexis®
Biochemicals, #400-032-G005) and 1% (v/v) SDS. Lysates were then transferred to microfuge tubes, heated at 100oC for 30 s, and homogenized using a 22½-gauge
needle to shear cellular DNA. Protein concentrations were determined using the BCA
kit (Pierce Biotechnology).
2.2.10 Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)
Denaturing SDS-PAGE was performed based on the Laemmli method (Laemmli,
1970). Discontinuous polyacrylamide gels (1.0 mm thick) made up of a stacking gel and a resolving gel (refer to Appendix II) were cast and assembled into a Mini-
PROTEAN® 3 cell (Bio-Rad Laboratories) with chambers filled with 1× electrophoresis buffer. Equal amounts of protein were mixed with 5× SDS loading buffer (refer to Appendix II-F), heated at 100oC for 5 min, and centrifuged
(Eppendorf 5415R) at maximum speed for 2 min at room temperature. After which,
10 μg of each sample and 3 μl of Precision Plus protein standards (Bio-Rad
34 Laboratories, #161-0374) were loaded into separate wells and electrophoresed first at
40 V for 1 h, then at 100 V till the end of the run.
2.2.11 Western blotting
Proteins separated in a polyacrylamide gel after SDS-PAGE were electroblotted onto a pre-wet polyvinylidene fluoride (PVDF) membrane (Pall Gelman Laboratory, Ann
Arbor, MI, USA, #66543) at 100 V for 1 h in cold 1× transfer buffer (refer to
Appendix II-H). Non-specific sites were blocked with Western blocking reagent
(Roche Diagnostics, Penzberg, Germany, #1921673) for 1 h at room temperature with shaking. After washing thrice with Tris-buffered saline containing Tween® 20
(TBST; refer to Appendix II-J) for 5 min each, the blot was incubated in the primary antibody at 4oC overnight. After washing thrice with TBST for 5 min each, the blot
was incubated in the appropriate secondary antibody conjugated with horseradish
peroxidase for 1 h at room temperature with shaking. After washing thrice with TBST
for 5 min each, the blot was exposed to SuperSignal® West Femto (Pierce
Biotechnology, #34095) chemiluminescent substrate. Signals were detected using the
ChemiGenius2 image acquisition system (Syngene, Cambridge, United Kingdom).
When required, the blot was stripped using Re-Blot Plus strong stripping solution
(Chemicon International, #2504) and then reprobed with another primary antibody.
2.2.12 Filipin staining
Cortical neurons treated with vehicle or 1 μg/ml U18666A for 72 h were fixed with
4% (w/v) paraformaldehyde (Sigma) for 30 min at room temperature and washed
35 thrice with 1× PBS before labeling with 125 μg/ml filipin (Sigma, #F9765) for 1 h at room temperature. After washing once with 1× PBS, digital images were acquired using a fluorescence microscope (Leica DM IRB).
2.2.13 Cholesterol measurement
Culture media from vehicle- and drug-treated cortical neurons were removed after 72 h of treatment, and cells were washed with 1× PBS and air-dried. Intracellular lipids were then extracted by incubation for 30 min in a mixture of 3:2 (v/v) hexane (Merck,
#1.04367) and isopropanol (Merck, #1.00994). The solvent was transferred to poly(tetrafluoroethylene)-capped hypovials (Pierce Biotechnology) and evaporated under nitrogen gas. After extraction of lipids, 200 μl of 1 M sodium hydroxide
(NaOH; Merck, #1.06498) was added to solubilize cellular proteins and protein concentration was determined by the Bradford method (Bio-Rad Laboratories). The amount of cholesterol in each sample was determined using the Amplex® Red cholesterol assay kit (Molecular Probes, #A-12216) following manufacturer’s instructions (Amundson and Zhou, 1999). Reaction mixtures were incubated for 30 min at 37oC before fluorescence measurement (excitation 563 nm, emission 587 nm)
using a spectrofluorometer (SPECTRAmax® Gemini XS; Molecular Devices).
Background fluorescence was subtracted from each corresponding well and the
amount of cholesterol per mg of protein in each sample was calculated according to a
freshly-prepared standard curve generated using purified cholesterol (supplied in the
kit). Results were then expressed as the percent difference in cholesterol content as
compared to the vehicle-treated Control.
36 2.2.14 Fluorometric caspase activity measurement
Caspase-3 substrate. Vehicle- and drug-treated cortical neurons after 72 h of treatment were washed with cold 1× PBS and lysed in cold buffer comprising 10 mM
HEPES (Gibco, #15630-080), 2 mM EDTA, 5 mM dithiothreitol (DTT; Sigma,
#D9779) and 0.1% (v/v) Nonidet P-40 (Fluka, Buchs, Switzerland, #56741). Lysates were centrifuged (Eppendorf 5415R) at maximum speed for 30 min at 4oC and
protein concentrations in the supernatants were determined by the Lowry method
(Bio-Rad Laboratories). Equal amounts of protein (10 μg) were incubated with 50 μM
of acetyl-DEVD-AFC (Alexis® Biochemicals, #ALX-260-032) for 15 min at 37oC
before fluorescence measurement (excitation 400 nm, emission 505 nm) using a
spectrofluorometer (SPECTRAmax® Gemini XS; Molecular Devices, Sunnyvale,
CA, USA). The fluorescence intensity is proportional to the amount of cleaved
substrate, which in turn, is dependent on caspase activity and related to the percentage
of apoptotic cells.
Caspase-family substrate kit. Cortical neurons treated with vehicle or 1 μg/ml
U18666A for 24 h, 36 h, 48 h and 72 h were lysed in cold Cell Lysis Buffer following
manufacturer’s instructions for the Caspase-Family Fluorometric Substrate Set II kit
(BioVision, #K133-9-25). Lysates were incubated on ice for 10 min, centrifuged, and
protein concentrations determined as described above. Ten μg protein and 50 μM of
each of the AFC-conjugated substrates for caspase-1, -2, -3, -4, -5, -6, -8, -9 and -10 were incubated for 1 h at 37oC before fluorescence measurement as described above.
All unspecified materials were supplied in the kit.
37 2.2.15 Measurement of intracellular ATP and glutathione (GSH)
Cortical neurons treated with vehicle or 0.1-2.5 μg/ml U18666A for 72 h were first washed twice with cold 1× PBS, followed by the addition of 250 μl of 6.5% (w/v) cold trichloroacetic acid (TCA; J.T. Baker, Phillipsburg, NJ, USA, #0414) according to procedures as described previously (Whiteman et al., 2002). After incubation on ice for 10 min, the TCA extract was removed. NaOH solution (200 μl of 1 M stock;
Merck) was added to solubilize cellular proteins. Protein concentration was then determined by the Bradford method (Bio-Rad Laboratories). The TCA extract was used immediately to assess intracellular ATP and GSH.
Intracellular ATP measurement. Loss of intracellular ATP was determined using firefly lantern extract. Three μl of TCA extract was mixed with 200 μl of sodium arsenite buffer comprising 26.67 mM MgSO4·7H2O, 3.33 mM KH2PO4 and 33.33
mM Na2HASO4·7H2O (pH 7.4). After adding 10 μl of firefly lantern extract (Sigma,
#FLE-50) per sample, luminescence was then measured for 10 s per sample using a
Lumi-One luminometer (Trans Orchid Enterprises, Tampa, FL, USA). Concentrations of ATP were then determined by comparing the values obtained with a freshly-
prepared standard curve of ATP after normalization of protein content.
Intracellular GSH measurement. Analysis of intracellular GSH was determined by first mixing 7.5 μl of TCA extract with 227.5 μl of 100 mM KH2PO4-KOH buffer
(pH 10), followed by the addition of 15 μl of freshly-prepared (10 mg/ml in
methanol) o-phthaldialdehyde (Sigma, #P1378). Samples were incubated in the dark
at room temperature for 25 min before fluorescence measurement (excitation 350 nm,
emission 420 nm) using a spectrofluorometer (SPECTRAmax® Gemini XS).
38 Concentrations of GSH were then determined by comparing the values obtained with a freshly-prepared standard curve of GSH after normalization of protein content
(Hissin and Hilf, 1976).
2.2.16 Measurement of proteasome activity
Fluorogenic proteasome substrates II (Z-LLE-AMC; Calbiochem, #539141) and III
(Suc-LLVY-AMC; Calbiochem, #539142) were used to assay for the postglutamyl and chymotrypsin-like peptidase activities of the proteasome respectively. Cortical neurons treated with vehicle or 0.1-2.5 μg/ml U18666A for 24 h, 48 h and 72 h were washed twice with cold 1× PBS and lysed in cold homogenization buffer comprising
10 mM Tris-HCl pH 7.4, 5 mM EDTA, 1 mM DTT (Sigma), 5 mM ATP (Sigma,
#A2383), 20% (v/v) glycerol (J.T. Baker, #2136-1) and 0.04% (v/v) Nonidet P-40
(Fluka). Protein concentrations were then determined by the Lowry method (Bio-Rad
Laboratories). Equal amounts of protein (5 μg) were incubated with 50 μM of
Substrate II or Substrate III in assay buffer (50 mM Tris-HCl pH 7.4 and 0.5 mM
DTT) before fluorescence measurement (excitation 350 nm, emission 465 nm) using a plate-reader (Tecan Ultra384) pre-warmed to 37oC. Background fluorescence was
subtracted from each corresponding well and results were expressed as the percent
difference in fluorescence as compared to the vehicle-treated Control.
2.2.17 Measurement of mitochondrial membrane potential
Cortical neurons treated with vehicle or 0.1-2.5 μg/ml U18666A for 24 h, 48 h and 72
h were loaded with 50 nM tetramethylrhodamine methyl ester (TMRM; Molecular
39 Probes, #T668) or 100 nM RedoxSensor™ Red CC-1 (Molecular Probes, #R14060) for 1 h or 10 min respectively in the dark at 37°C, and then washed twice with 1×
PBS. Fluorescence of TMRM (excitation 550 nm, emission 590 nm) and CC-1
(excitation 540 nm, emission 600 nm) were measured (SPECTRAmax® Gemini XS).
Background fluorescence was subtracted from each corresponding well and results
were expressed as the percent difference in fluorescence as compared to the vehicle-
treated Control.
2.2.18 Determination of intracellular oxidative stress by 2’,7’-dichlorofluorescin
diacetate (DCFH-DA)
Quantitative assay. Cortical neurons treated with vehicle or 0.1-2.5 μg/ml U18666A
for 24 h, 48 h and 72 h were loaded with 1 μM DCFH-DA (Molecular Probes,
#C6827) for 1 h at 37oC before fluorescence measurement (excitation 485 nm,
emission 535 nm) using a spectrofluorometer (SPECTRAmax® Gemini XS) pre-
warmed to 37oC. The non-fluorescent DCFH-DA is cell-permeable and hydrolyzed by intracellular esterases to yield dichlorofluorescin (DCFH), another non-fluorescent
compound. DCFH can then be oxidized by various oxidants to the fluorescent
dichlorofluorescein (DCF) (LeBel et al., 1992). Background fluorescence was
subtracted from each corresponding well and results were expressed as the percent
difference in fluorescence as compared to the vehicle-treated Control. The overall
oxidative stress level is quantified by quantifying the fluorescence emitted by DCF
(Wang and Joseph, 1999).
40 Confocal microscopy. Cortical neurons in chambered coverglasses treated with vehicle or 1 μg/ml U18666A for 72 h were loaded with DCFH-DA as described above. Digital images were acquired using laser scanning confocal microscope (Carl
Zeiss LSM 510, Jena, Germany) at the Confocal Microscopy Unit.
2.2.19 Measurement of lipid peroxidation
Lipid peroxidation was assessed using two established fluorescent probes: cis-
581/591 parinaric acid and C11-BODIPY . Cis-parinaric acid is a fluorescent fatty acid that accumulates in cell membranes and loses fluorescence upon reaction with lipid
peroxides (Drummen et al., 1999; Laranjinha et al., 1992). Oxidation of the
581/591 ratiometric C11-BODIPY probe by peroxyl radicals also results in a loss of
fluorescence (Drummen et al., 2002; Pap et al., 1999). Cortical neurons treated with
vehicle or 0.1-2.5 μg/ml U18666A for 24 h, 48 h and 72 h were loaded with 1 μM cis-
581/591 parinaric acid (Molecular Probes, #P1901) or C11-BODIPY (Molecular Probes,
#D3861) for 1 h in the dark at 37°C, washed twice with 1× PBS and fluorescence
measured at excitation 325 nm and emission 405 nm for cis-parinaric acid or
581/591 excitation 581 nm and emission 591 nm for C11-BODIPY using a
spectrofluorometer (SPECTRAmax® Gemini XS). Background fluorescence was
subtracted from each corresponding well and results were expressed as the percent
difference in fluorescence as compared to the vehicle-treated Control.
41 2.2.20 Assessment of oxidized proteins
Western blotting. Protein oxidation was assayed using the OxyBlot™ Protein
Oxidation Detection Kit (Intergen Company, Purchase, NY, USA, #S7150) following manufacturer’s instructions. Lysates from cortical neurons treated with vehicle or
1 μg/ml U18666A for 24 h, 48 h and 72 h were prepared using RIPA buffer. Protein concentrations were then determined by the Lowry method (Bio-Rad Laboratories).
To prevent the oxidation of proteins that may occur after cell lysis, 2% (v/v) of β- mercaptoethanol (Sigma, #M6250) was added into each sample. Two aliquots of each sample to be analyzed were treated. One aliquot was subjected to the derivatization reaction while the other served as a negative control without derivatization. Samples were derivatized with 2,4-dinitrophenylhydrazine (DNPH), separated in a 4-20%
Tris-HCl linear gradient gel (Bio-Rad Laboratories, #161-1105) and transferred to
PVDF. Oxidation-modified proteins were probed with a rabbit antiserum against the derivatized carbonyl groups. Signals were visualized using chemiluminescence
(Pierce Biotechnology) and images acquired using ChemiGenius2 (Syngene).
Enzyme-linked immunosorbent assay (ELISA). Quantitative measurement of
protein carbonyl content was determined using the Zentech ELISA kit (Zenith
Technology, Dunedin, New Zealand, #ZPCK01) following manufacturer’s
instructions. Lysates from cortical neurons treated with vehicle or 1 μg/ml U18666A
for 72 h were prepared as described above and their protein concentrations determined. Samples were concentrated by precipitation with 28% (w/v) TCA (J.T.
Baker) and resuspension in 5 μl EIA buffer. After which, the samples, together with oxidized protein standards and carbonyl controls, were derivatized with DNPH. One
42 μg of derivatized protein was applied to each well of the ELISA plate and incubated at 4oC overnight. The adsorbed protein was probed with biotinylated antibody,
followed by streptavidin-linked horseradish peroxidase. Chromatin reagent (200 μl)
was then added for 5 min for color development before stopping the reaction with 100
μl stopping reagent. Absorbance at 450 nm was immediately determined for each well
using a spectrophotometer (SPECTRAmax® 190). The readings were related to a
standard curve prepared from increasing proportions of oxidized protein standards
that has been calibrated colorimetrically. All unspecified materials were supplied in
the kit.
2.2.21 Analysis of DNA base modifications by gas chromatography-mass
spectrometry (GC-MS)
DNA was isolated from 1.35 × 107 cortical neurons for each replicate (n = 3 per time-
point) for vehicle-treated Control and 1 μg/ml U18666A after 24 h, 48 h and 72 h of
treatment, according to procedures as described previously (Spencer et al., 1995) with
slight modifications. Cells were first washed with cold 1× PBS and lysed in 1 ml of
cell lysis buffer comprising 20 mM Tris-HCl pH 8.0, 100 mM NaCl, 10 mM EDTA
and 0.5% (v/v) SDS. The lysate was then transferred to a disposable 15 ml centrifuge
tube, and 100 μg/ml RNase A (Roche Diagnostics, #109142) and 10 units/ml RNase
T1 (Roche Diagnostics, #109193; supplied as 100 000 units/ml stock solution) were
added, mixed and incubated at 37oC for 1 h. After which, 0.1 mg/ml proteinase K
(Roche Diagnostics, #3115836) was added, mixed and further incubated at 37oC for 2
h to ensure digestion of all cellular and nuclear proteins. One ml of 6 M NaCl (Merck,
43 #1.06404) was added, followed by centrifugation at 4000 rpm for 20 min at room temperature to pellet the protein precipitate. The supernatant containing the DNA was removed and subjected to another NaCl treatment to obtain a clear and protein-free supernatant. DNA was precipitated with 5 ml of cold absolute ethanol (Merck,
#1.00983) and transferred to a 1.5 ml microfuge tube. The precipitated DNA was washed twice with 1 ml of cold 70% (v/v) ethanol each, dried under vacuum, and redissolved in 250 μl of cold Milli-Q water. The concentration of DNA (μg/ml) in every sample was measured spectrophotometrically (A260 1.0 = 50 μg/ml; Beckman
Coulter DU® 640B, Fullerton, CA, USA) and aliquots of 100 μg DNA were
lyophilized after addition of 0.5 nmol each of 6-azathymine (Sigma, #A1507) and
2,6-diaminopurine (Sigma, #D3289) as internal standards. Hydrolysis, derivatization
and analysis of samples were performed as described previously (Spencer et al.,
2000) with slight modifications. The samples were first hydrolyzed in evacuated glass
tubes at 145°C for 45 min using 0.5 ml of 60% (v/v) formic acid (Merck, #1.00264),
and finally derivatized under nitrogen in poly(tetrafluoroethylene)-capped hypovials
(Pierce Biotechnology) by adding 60 μl of bis(trimethylsilyl)trifluoracetamide
(BSTFA) containing 1% (v/v) trimethylchlorosilane (TMCS) (Pierce Biotechnology,
#38833) and 15 μl of a mixture of 3:1 (v/v) acetonitrile (Pierce Biotechnology,
#20062) and ethanethiol (Fluka, #04290), and incubating at room temperature for 6 h.
This method can prevent any artifactual formation of modified DNA base products
(Jenner et al., 1998). The derivatized samples were then transferred to autoinjector vials and analyzed by GC-MS using a Hewlett-Packard (Palo Alto, CA, USA) 5890II gas chromatograph interfaced with a Hewlett-Packard 5917A mass selective detector.
44 The injector port and the GC-MS interface were kept at 250°C and 290°C respectively. Separations were carried out on a fused silica capillary column (12 m length × 0.2 mm internal diameter) coated with phenylmethylsiloxane (Hewlett-
Packard) of film thickness 0.33 μm. The carrier gas was helium at a flow-rate of 0.93 ml/min. Approximately 0.5 μg of each derivatized sample was injected into the column. The temperature of the column was increased from 125°C to 175°C at
8°C/min after 2 min at 125°C, then from 175°C to 220°C at 30°C/min and held at
220°C for 1 min, and finally from 220°C to 290°C at 40°C/min and held at 290°C for
20 min. Quantitation of the modified bases was achieved by relating the peak area of the compound with the peak area of the internal standards and calculating from calibration curves constructed using known concentrations of the internal standards.
2.2.22 Measurement of β-amyloid (Aβ)
Culture media from vehicle- and drug-treated cortical neurons were collected after 24 h, 48 h and 72 h of treatment. Cells were washed with 1× PBS and lysed in cold buffer comprising 1% (v/v) Triton X-100 (BDH), 0.5% (w/v) deoxycholate (Sigma,
#D6750) and protease inhibitors (Complete™ Mini; Roche Diagnostics, #1836153).
Aβ levels were determined using the DELFIA® Double Capture ELISA kit
(PerkinElmer, Boston, MA, USA) as described previously (George et al., 2004).
Plates were coated with anti-13E9 (for Aβ40) or anti-G211 (for Aβ42) antibodies and
non-specific sites were blocked with SuperBlock™ blocking buffer in PBS (Pierce
Biotechnology, #37515). After washing the plates, 1E8-biotin was added to each
well. Aβ40 and Aβ42 peptide standards, as well as 50 μl of culture media (for
45 measurement of secreted Aβ) or cell lysates (for measurement of accumulated Aβ) were assayed in triplicates and incubated at 4oC overnight. After which, the plates
were washed, europium-labeled streptavidin added, and then developed with
enhancement solution. Analysis was carried out using the Wallac 1420 VICTOR2 TM plate-reader (PerkinElmer) at excitation 340 nm and emission 613 nm. All unspecified materials were supplied in the kit.
2.2.23 Determination of amyloid precursor protein (APP) derivatives
Cell lysates (20 μg) were prepared as described in 2.2.22, denatured in Laemmli sample buffer containing 8% (v/v) β-mercaptoethanol (Sigma), and electrophoresed in a NuPAGE® Novex 10-20% tricine gel (Invitrogen, #EC6625). Proteins were
electroblotted onto pure nitrocellulose (Bio-Rad Laboratories, #162-0112), and the
blot was boiled in 1× PBS for 5 min and blocked with Western blocking reagent
(Roche Diagnostics). The blot was then incubated in anti-369 against the C-terminal
sequence of APP, followed by the secondary antibody conjugated with horseradish
peroxidase. Signals were detected using enhanced chemiluminescence (ECL™;
Amersham Biosciences, #RPN2135) and the GeneGnome chemiluminescence imager
(Syngene).
2.2.24 Immunocytochemistry
Cortical neurons treated with vehicle or 1 μg/ml U18666A for 24 h, 48 h and 72 h
were first fixed with 4% (w/v) paraformaldehyde (Sigma) for 30 min at room
temperature and washed thrice with 1× PBS. After which, fixed cells were incubated
46 in 100 mM ammonium chloride (BDH, #100173D) and subsequently 0.2% (v/v)
Triton X-100 (BDH) before blocking with 10% (v/v) goat serum (Gibco, #16210-
072) for 1 h at room temperature. Cells were then incubated in monoclonal primary antibodies to AT8, AT180, tau-1 or PHF1 at 4°C overnight, followed by labeling with
Alexa Fluor® secondary antibody for 1 h at room temperature. After washing thrice
with 1× PBS, digital images were acquired using a fluorescence microscope (Leica
DM IRB).
2.2.25 Measurement of cyclin-dependent kinase 5 (cdk5) activity
Cortical neurons treated with vehicle or 1 μg/ml U18666A for 24 h, 48 h and 72 h
were first washed with cold 1× PBS and lysed in 200 μl of cold lysis buffer
comprising 30 mM MOPS pH 7.4 (Sigma, #M1254), 20 mM magnesium chloride
(Merck, #5833), 1 mM DTT (Sigma), 1 mM EDTA, 1 mM EGTA (Sigma, #E4378),
0.1 μM calpastatin (Calbiochem, #208900) and protease inhibitors (Complete™
Mini). Protein concentration was determined by the Bradford method (Bio-Rad
Laboratories). Cdk5 activity was determined as described previously (Qu et al.,
2002). Total cell lysate (16 μg) was incubated with 50 μM synthetic histone peptide
HS(9-18) (PKTPKKAKKL) and 100 μM [γ-32P]ATP in 600 μl of buffer comprising
o 20 mM MOPS pH 7.4, 10 mM MgCl2 and 1 mM DTT for 30 min at 30 C. The kinase
activity was then determined by phosphate incorporation into HS(9-18), as quantitated using a liquid scintillation counter (Beckman Coulter LS 6500, Fullerton,
CA, USA).
47 2.2.26 Isolation of total RNA
Total RNA was isolated using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA,
#74104) from 1.35 × 107 cortical neurons for each replicate (n = 3 per time-point) for vehicle-treated Control and 1 μg/ml U18666A after 24 h, 48 h and 72 h of treatment.
Cells were lysed in buffer containing guanidine isothiocyanate and β-mercaptoethanol to denature RNases to prevent the breakdown of RNA. The lysate was then homogenized using a 22½-gauge needle and mixed with an equal volume of 70%
(v/v) ethanol to optimize binding conditions. The sample was applied to a mini column containing a silica-gel-based membrane to adsorb total RNA. The column was further washed to remove contaminants such as salts, proteins and other cellular impurities. Total RNA was finally eluted with RNase-free water. All unspecified materials were supplied in the kit.
2.2.27 Quantification of total RNA and determination of RNA integrity
The concentration of RNA (μg/ml) was measured spectrophotometrically (Beckman
® Coulter DU 640B) at 260 nm (A260 1.0 = 40 μg/ml). Purity of the RNA was also
estimated based on its A260/A280 ratio. Pure RNA is expected to have a ratio of 1.8-
2.0. The integrity of RNA was checked using agarose gel electrophoresis as described
in 2.2.28. Two distinct bands representing the 28S and 18S ribosomal RNA should be
observed, with the upper 28S band having approximately twice the intensity of the
lower 18S band.
48 2.2.28 Agarose gel electrophoresis
A 1% (w/v) agarose (Bio-Rad Laboratories, #162-0133) gel was cast, and the samples were mixed with Blue/Orange 6× loading dye (Promega Corporation, #G1881), loaded into the agarose gel, and electrophoresed at 80 V for 1 h in 1× TAE buffer.
The gel was then soaked in 0.5 μg/ml ethidium bromide stain solution (Bio-Rad
Laboratories, #161-0433) for 15 min at room temperature with shaking. After washing the gel twice with Milli-Q water for 15 min each, the RNA bands were visualized using an ultraviolet transilluminator (ChemiGenius2, Syngene).
2.2.29 One-cycle complementary DNA (cDNA) synthesis
Positive controls to monitor the entire GeneChip® eukaryotic target labeling process
were serial-diluted from the GeneChip® eukaryotic poly-A RNA control kit
(Affymetrix, #900433) and spiked directly into 10 μg of total RNA in each sample.
Double-stranded cDNA was then synthesized using the one-cycle cDNA synthesis kit
(Affymetrix, #900431) according to the GeneChip® Expression Analysis technical
manual. The sample RNA (10 μg) was mixed with T7-oligo(dT) promoter primer and
incubated at 70oC for 10 min in a thermal cycler (PTC-100® Peltier Thermal Cycler,
MJ Research, Hercules, CA, USA), followed by cooling to 4oC for at least 2 min to allow primer hybridization. The First-Strand Master Mix (refer to Appendix III-A) was added and the mixture further incubated at 42oC for 2 min. For the first-strand cDNA synthesis, 200 units/μl of SuperScript™ II reverse transcriptase was added and the mixture was incubated at 42oC for 1 h, followed by cooling to 4oC for at least 2 min before adding the Second-Strand Master Mix (refer to Appendix III-B). After
49 incubation at 16oC for 2 h, T4 DNA polymerase and 0.5 M EDTA were added. All
unspecified materials were supplied in the kit.
2.2.30 Cleanup of double-stranded cDNA
The GeneChip® Sample Cleanup Module (Affymetrix, #900371) was used according to the GeneChip® Expression Analysis technical manual. Each double-stranded
cDNA synthesis preparation was mixed with the cDNA binding buffer and applied to a cDNA cleanup spin column. After centrifugation (Eppendorf 5415R) at maximum speed for 1 min, the flow-through was discarded and the spin column was rinsed with the cDNA wash buffer and further centrifuged with the cap opened to allow complete drying of the membrane. The cDNA was then eluted with 14 μl of cDNA elution buffer. All unspecified materials were supplied in the kit.
2.2.31 Synthesis of biotin-labeled complementary RNA (cRNA)
The cDNA after cleanup was used as a template to synthesize biotin-labeled cRNA
using the GeneChip® in vitro transcription (IVT) labeling kit (Affymetrix, #900449)
according to the GeneChip® Expression Analysis technical manual. Six μl of cDNA
was mixed with the IVT Reaction Master Mix (refer to Appendix III-C) and
incubated at 37oC for 16 h in a thermal cycler (PTC-100® Peltier Thermal Cycler).
2.2.32 Cleanup of biotin-labeled cRNA
The GeneChip® Sample Cleanup Module (Affymetrix) was used according to the
GeneChip® Expression Analysis technical manual. RNase-free water (60 μl), 350 μl
50 of IVT cRNA binding buffer and 250 μl of absolute ethanol (Merck) were added in sequence to each cRNA sample. The mixture was applied to an IVT cRNA cleanup spin column and centrifuged (Eppendorf 5415R) at maximum speed for 15 s. The flow-through was discarded, and the spin column was rinsed separately with the IVT cRNA wash buffer and 80% (v/v) ethanol and further centrifuged at maximum speed for 5 min with the cap opened. The cRNA was then eluted with 21 μl of RNase-free water and quantified spectrophotometrically as described in 2.2.27. An aliquot was saved for agarose gel electrophoresis as described in 2.2.28 before proceeding to the next step. All unspecified materials were supplied in the kit.
2.2.33 Fragmentation of cRNA for target preparation
Fragmentation of the cRNA products was performed according to the GeneChip®
Expression Analysis technical manual before hybridizing to GeneChip® probe arrays.
The cRNA (20 μg) was mixed with 5× fragmentation buffer diluted with RNase-free
water and incubated at 94oC for 35 min in a thermal cycler (PTC-100® Peltier
Thermal Cycler), followed by cooling to 4oC. This fragmentation procedure should
produce a distribution of fragment sizes ranging from 35 to 200 bases. An aliquot was
saved for agarose gel electrophoresis as described in 2.2.28 before proceeding to the
next step. All unspecified materials were supplied in the GeneChip® Sample Cleanup
Module (Affymetrix).
51 2.2.34 Microarray analysis
Microarray analysis was carried out using 18 GeneChip® murine genome U74A set
(version 2) probe arrays (Affymetrix, Santa Clara, CA, USA, #900343) according to
the GeneChip® Expression Analysis technical manual. The arrays were distributed as
follows: vehicle-treated Control for 24 h, 48 h and 72 h (n = 3 for each time-point),
and 1 μg/ml U18666A treatment for 24 h, 48 h and 72 h (n = 3 for each time-point).
Lee et al. (2000b) demonstrated that at least three replicates are required to give a
more reliable microarray analysis. In addition, 6 GeneChip® Test3 arrays
(Affymetrix, #900341), distributed as above (n = 1 for each time-point), were first
used as assessment tools to determine target quality and labeling efficiency before
utilizing the genome arrays.
2.2.35 Eukaryotic target hybridization
The hybridization cocktail was made up according to the GeneChip® Expression
Analysis technical manual for the 49 Format (Standard)/64 Format Array.
Fragmented cRNA (15 μg) was mixed with control oligonucleotide B2, 20×
eukaryotic hybridization controls, herring sperm DNA (Promega Corporation,
#D1811), BSA (Invitrogen, #15561-020), 2× hybridization buffer (refer to Appendix
III-E), DMSO (Sigma) and diethylpyrocarbonate (DEPC)-treated water (Ambion,
Austin, TX, USA, #9920) to a final volume of 300 μl, and heated at 99oC for 5 min.
The probe arrays were equilibrated to room temperature and wetted by filling it with
1× hybridization buffer and incubated at 45oC for 10 min with rotation. The hybridization cocktails at 99oC were further incubated at 45oC for 5 min and
52 centrifuged at maximum speed for 5 min to remove insoluble materials. The hybridization buffer in each probe array was then replaced with the hybridization cocktail. All arrays were incubated in Hybridization Oven 640 (Affymetrix) at 45oC
for 16 h with rotation. The GeneChip® Instrument System (Affymetrix) is located at
the Biosensors Focused Interests Group (BFIG) Core Facility Lab at the Clinical
Research Center. All unspecified materials were supplied in the GeneChip® eukaryotic hybridization control kit (Affymetrix, #900454).
2.2.36 Washing, staining and scanning of probe arrays
Washing, staining and scanning were performed according to the GeneChip®
Expression Analysis technical manual. The streptavidin phycoerythrin (SAPE) solution mix (refer to Appendix III-G) was prepared and divided into two aliquots of
600 μl each to be used as stains 1 and 3. The antibody solution mix (refer to
Appendix III-H) to be used as stain 2 was prepared in a separate tube. After 16 h of hybridization, the hybridization cocktail was removed and each probe array was filled completely with wash buffer A (refer to Appendix III-I). The Fluidics Station 400
(Affymetrix) was first primed to fill the appropriate lines with wash buffer A and wash buffer B (refer to Appendix III-J). After which, the probe arrays and stains 1, 2 and 3 were loaded into the respective holders on Fluidics Station 400. The appropriate fluidics protocol (Micro_1v1 for Test3 arrays; EukGE-WS2v5 for murine genome
U74A probe arrays) was chosen to automate the washing and staining of the probe arrays. At the end of the run, the probe arrays were removed and Fluidics Station 400 was shut down. The probe arrays were then scanned using the GeneChip® Scanner
53 3000 (Affymetrix) and the scanned images were inspected for the presence of artifacts or scratches before analysis.
2.2.37 Microarray data analysis
To minimize discrepancies due to variables such as sample preparation, hybridization conditions, staining or array lot, the raw expression data from each probe array were scaled to an average intensity of 500 and the relative mRNA expression levels were expressed as signal log ratios as compared to the controls using the Affymetrix
Microarray Suite (version 5.0) software. Genes whose average expression level increased more than 2-fold or decreased more than 2-fold after one-way analysis of variance (ANOVA) for multiple comparisons at p < 0.01 were identified using
GeneSpring™ software version 6.1 (Silicon Genetics, Redwood City, CA, USA).
These genes were also subjected to a conservative Bonferroni multiple testing correction using GeneSpring™ to control for false positives (Jung et al., 2005).
Differentially-expressed genes were then classified based on their known biological functions using the NetAffx™ Analysis Center at http://www.affymetrix.com (Liu et al., 2003) and Database for Annotation, Visualization, and Integrated Discovery
(DAVID) at http://www.DAVID.niaid.nih.gov (Dennis Jr et al., 2003), and further clustered using GeneSpring™ (Silicon Genetics) according to their changes in expression level over time.
54 2.2.38 Sample preparation for proteomics analysis
Protein for proteomics analysis was isolated using TRIZOL® reagent (Invitrogen,
#15596-026) from 2.0 × 107 cortical neurons for each replicate (n = 2 per time-point)
for vehicle-treated Control and 1 μg/ml U18666A after 24 h, 48 h and 72 h of
treatment. The TRIZOL®-based method is well-suited to separate cellular proteins
from DNA and RNA contamination (Giorgianni and Beranova-Giorgianni, 2005).
Cells were lysed in 1 ml TRIZOL® reagent, and the lysate was homogenized and separated into aqueous and organic phases upon the addition of 200 μl chloroform
(Merck, #1.02445) and centrifugation (Eppendorf 5415R) at maximum speed for 15 min at 4oC. The aqueous phase containing the RNA was discarded. Absolute ethanol
(300 μl; Merck) was added to the interphase and organic phase containing the DNA
and proteins, mixed by inversion, and centrifuged for 5 min at 4oC. The supernatant
containing the proteins was removed and isopropanol (Sigma, #I9516) 5× the volume
of the supernatant was added, mixed by inversion, and centrifuged at maximum speed
for 10 min at 4oC. The protein pellet was washed thrice with 0.3 M guanidine
hydrochloride (Sigma, #G3272) in 95% (v/v) ethanol and once with absolute ethanol
(5 min each), vacuum-dried, and resolubilized in buffer comprising 7 M urea (USB
Corporation, Piscataway, NJ, USA, #75826), 2 M thiourea (Fluka, #88810) and 4%
(w/v) CHAPS (USB Corporation, #13361).
2.2.39 Cleanup and quantification of protein for proteomics analysis
Cleanup of the protein samples can improve the quality of 2D electrophoresis results
by removing interfering contaminants. The ReadyPrep™ 2D Cleanup Kit (Bio-Rad
55 Laboratories, #163-2130) was used following manufacturer’s instructions.
Precipitating agent 1 (300 μl) was mixed with each protein sample and incubated for
15 min on ice. After which, 300 μl of precipitating agent 2 was added, mixed, and centrifuged (Eppendorf 5415R) at maximum speed for 5 min at 4oC. The supernatant
was completely removed and the pellet was washed in sequence with wash reagent 1,
Milli-Q water and wash reagent 2. After centrifugation at the same conditions, the
pellet was air-dried and resuspended in the same buffer as described in 2.2.38. The
2D Quant Kit (Amersham Biosciences), based on the specific binding of copper ions
to proteins, was used to determine the protein concentration of each sample. All unspecified materials were supplied in the respective kits.
2.2.40 Immobilized pH gradient (IPG) strip rehydration
The rehydration solution comprising 7 M urea (USB Corporation), 2 M thiourea
(Fluka), 4% (w/v) CHAPS (USB Corporation), 40 mM DTT (Amersham Biosciences,
#17-1318-02), 0.5% (v/v) IPG buffer pH 3-10 Non-Linear (Amersham Biosciences,
#17-6000-88) and 0.002% (w/v) bromophenol blue (Amersham Biosciences, #17-
1329-01) was prepared and applied into each slot of the Immobiline™ DryStrip
Reswelling Tray for 7-24 cm IPG strips (Amersham Biosciences, #80-6465-32).
Next, the Immobiline™ DryStrip IPG strips pH 3-10 Non-Linear of 7 cm length
(Amersham Biosciences, #17-6001-12) or 18 cm length (Amersham Biosciences,
#17-1235-01) were positioned with their gels faced down onto the rehydration solution in separate slots and completely overlaid with Immobiline™ DryStrip cover
fluid mineral oil (Amersham Biosciences, #17-1335-01) to minimize evaporation and
56 urea crystallization. The apparatus was then left at room temperature for 8 h to allow the gels to hydrate. Each protein sample was performed in duplicates (n = 2).
2.2.41 First-dimension isoelectric focusing (IEF)
The rehydrated IPG strips were transferred to the Ettan™ IPGphor™ cup loading strip holders (Amersham Biosciences, #80-6459-43) with their gels faced up. Sample cups were then placed on the sample cup bars near to the anode, as sample application at the anode proved to be superior to cathodic applications (Görg et al., 2000). After which, the strip holders except the sample cups were flooded with Immobiline™
DryStrip cover fluid mineral oil (Amersham Biosciences) to check for leakage into the sample cups. When no leakage was detected, protein samples (10 μg for each 7 cm IPG strip; 100 μg for each 18 cm IPG strip) were applied into the sample cups and overlaid with the cover fluid. The apparatus was then placed onto the Ettan™
IPGphor™ IEF system (Amersham Biosciences) and IEF was conducted overnight according to the protocol parameters in Table 2.1.
57 Table 2.1: IEF conditions for cup-loading application method using Immobiline™ DryStrip IPG strips pH 3-10 Non-Linear
Length Mode Voltage (V) Duration (h) 7 cm S1 Step-and-hold 200 1.5 S2 Gradient 1000 0.5 S3 Gradient 5000 0.5 S4 Step-and-hold 5000 2 18 cm S1 Step-and-hold 200 3 S2 Step-and-hold 500 1 S3 Gradient 8000 7 S4 Step-and-hold 8000 5
Voltage gradient and step-and-hold mode, 50 μA per IPG strip, 0.5% (v/v) IPG buffer, 20oC for IEF
2.2.42 Equilibration of IPG strips
The equilibration step saturates the IPG strips with the SDS buffer system required
for the second-dimension separation. The equilibration buffer comprising 50 mM
Tris-HCl pH 8.8, 6 M urea (USB Corporation), 30% (v/v) glycerol (J.T. Baker), 2%
(v/v) SDS and 0.002% (w/v) bromophenol blue (Amersham Biosciences) was
prepared. The IPG strips in separate tubes were first incubated in equilibration buffer
containing 10 mg/ml DTT (Amersham Biosciences) for 15 min at room temperature
with shaking. After which, the strips were incubated in equilibration buffer containing
25 mg/ml iodoacetamide (IAA; Fluka, #57670) at the same conditions.
2.2.43 Second-dimension SDS-PAGE
Homogeneous 8% polyacrylamide gels (1.0 mm thick) were cast (refer to Appendix
IV-A) for the 7 cm IPG strips, whereas large precast 8-16% Tris-HCl gels with IPG
combs (Bio-Rad Laboratories, #161-1453) were used for the 18 cm IPG strips.
58 Gradient gels were used for the longer strips to achieve a wider and larger separation and to obtain sharper spots because the decreasing pore size can minimize diffusion.
The IPG strips after equilibration were loaded onto each vertical second-dimension gel and sealed with overlay agarose (Bio-Rad Laboratories, #163-2111). The gels with the 7 cm IPG strips were assembled into a Mini-PROTEAN® 3 cell (Bio-Rad
Laboratories) and electrophoresed first at 5 mA per gel for 20 min, then at 15 mA per
gel till the end of the run. The gradient gels with the 18 cm IPG strips were assembled
into a PROTEAN® II xi 2D cell (Bio-Rad Laboratories) and electrophoresed first at 8
mA per gel for 1 h, then at 24 mA per gel till the end of the run. In the case of the
PROTEAN® II xi 2D cell, a thermostatic regulator was used to cool the gels during
electrophoresis. After electrophoresis, the gels were removed from the glass plates
and incubated in fixative solution (refer to Appendix IV-B) at room temperature
overnight with shaking.
2.2.44 Silver staining
Gels fixed overnight were washed twice with Milli-Q water for 30 min each and
sensitized in 0.02% (w/v) sodium thiosulfate (Merck, #1.06516) for 2 min. After
rinsing twice with Milli-Q water for 1 min each, the gels were incubated in cold 0.1%
(w/v) silver nitrate solution (Merck, #1.01512) for 40 min with shaking. After rinsing
twice again with Milli-Q water for 1 min each, the gels were incubated in developing solution (refer to Appendix IV-C) with shaking and visually monitored. The gels were immediately transferred to the stop solution (refer to Appendix IV-D) when the spots have reached the desired intensity and before the staining background becomes
59 too dark. After 10 min incubation in the stop solution, the gels were washed thrice with Milli-Q water for 5 min each and scanned using a GS-800 calibrated densitometer (Bio-Rad Laboratories).
2.2.45 Analysis of 2D gels
Only the large gels were subjected to analysis, since the smaller gels only serve as a rapid screening tool to confirm the profile and quality of the samples. A total of 12 gradient gels were obtained for vehicle-treated Control and 1 μg/ml U18666A after
24 h, 48 h and 72 h of treatment (n = 2 each). Gel images were analyzed using the
Discovery Series PDQuest™ 2D analysis software version 7.3 (Bio-Rad
Laboratories). Using the Spot Detection Wizard function, the original gel images were processed to remove streaks and speckles and smoothened to clarify spots.
Three-dimensional Gaussian spots were then created in artificial Gaussian images for further spot matching and analysis in PDQuest™. Subsequently, the intensities of the spots in the different gels were normalized and placed in a MatchSet for comparison and statistical analysis (p < 0.05 by Student’s t-test). The best resolved gel from each time-point was chosen to create a master gel and representative image of each time- point. Spots that were consistent in all of the samples were chosen and labeled on the master gel. These normalized synthetic master gels were used for differential comparison to identify differentially-expressed protein spots with PDQuest™. To avoid ambiguity, only spots that displayed prominent differences were analyzed for differential expression.
60 2.2.46 Enzymatic digestion of protein spots
Differentially-expressed protein spots were manually excised from the silver-stained gels and transferred to a 96-well polypropylene microtiter plate (Nunc) using a self- made plunger. The gel plugs were destained with a mixture of 1:1 (v/v) 30 mM potassium ferricyanide (Aldrich, Saint Louis, MO, USA, #24,402-3) and 100 mM sodium thiosulfate (Merck), washed with acetonitrile (Applied Biosystems, Foster
City, CA, USA, #40-4050-50) for 5 min, and dried in a Savant Automatic
Environmental SpeedVac AES2010 centrifugal concentrator (Holbrook, NY, USA) for 5 min. Protein in-gel digestion was performed at 30oC overnight with the addition
of 10 μg/ml trypsin (Promega Corporation, #V5280) in 50 mM ammonium bicarbonate pH 8.0 (Sigma, #A6141). To maximize the recovery of the peptides,
0.5% (v/v) trifluoroacetic acid (TFA; Fluka, #09746) in 50% (v/v) acetonitrile was
added and the digested samples were sonicated in an ultrasonic water-bath (Crest
Ultrasonics, Trenton, NJ, USA) for 10 min. The extracted peptides were dried using
SpeedVac, resuspended in 0.5% (v/v) TFA, and spotted onto a sample plate. Matrix
solution comprising 10 mg/ml α-cyano-4-hydroxycinnamic acid (Sigma, #C8982) and
0.5% (v/v) TFA in 50% (v/v) acetonitrile was immediately added and the plate air-
dried before loading onto the mass spectrometer for analysis.
2.2.47 Mass spectrometry
Peptide mass fingerprinting was performed using an Axima CFR Plus matrix-assisted
laser desorption ionization-time of flight (MALDI-TOF) mass spectrometer
(Shimadzu/Kratos, Manchester, UK) in the reflectron and delayed extraction mode.
61 Mass spectra and the results were acquired using the auto-experiment method. Each spectrum was an average of 20 profiles. The resulting mass spectra were automatically calibrated upon acquisition using a two-point internal calibration with the trypsin autolytic fragment peaks appearing at mass-to-charge ratio (m/z) 842.509 and 2211.104. Assignment of the peaks was done manually. The measured peptide masses were excluded if their masses corresponded to trypsin autodigestion products or were from identified proteins adjacent to the spot being analyzed.
2.2.48 Database searching and identification of proteins
To identify proteins, database search was performed using the MS-Fit software
(Protein Prospector; San Francisco, CA, USA). MS-Fit is a typical peptide mass fingerprinting program which compares the experimentally determined masses of tryptic peptides with the theoretical masses of all tryptic peptides that can be calculated from sequences of all proteins in the genomic databases. The following search parameters were used: carboxyamidomethylation of cysteine, mouse and human species, mass window of 1-200 kDa, at least four peptides required to match, oxidation of methionine, N-terminal acetylation, acrylamide-modified cysteine, and phosphorylation of serine, threonine and tyrosine. All of the proteins and database entries matching the input data and parameters were listed in a simple ranking system in which the entries with the least number of unmatched masses were ranked higher.
The SWISS-PROT non-redundant database was first queried, followed by the NCBI
(National Center for Biotechnology Information) database if no significant protein matches were found.
62 2.2.49 Statistical analyses
Results for all methods, except those for microarray and proteomics analyses, were analyzed using SPSS® 13.0 statistics software (Chicago, IL, USA) and are presented
as mean ± SEM. The mean values were calculated from data taken from at least three
separate experiments conducted in triplicates, unless otherwise stated. Where
significance testing was performed, data were analyzed using one-way ANOVA with
Tukey post-hoc test to determine significant differences in multiple comparisons.
Values of p < 0.01 were considered as statistically significant.
63 CHAPTER 3
CELLULAR SIGNALING OF U18666A-MEDIATED CELL DEATH
64 Part I
Apoptotic Cell Death and Accumulation of Intracellular Free Cholesterol
65 3 Cellular Signaling of U18666A-Mediated Cell Death
I Apoptotic cell death and accumulation of intracellular free
cholesterol
3.1 Introduction
Niemann-Pick disease type C (NPC), a fatal hereditary juvenile dementia, is an autosomal recessive neurodegenerative disorder marked by premature neuronal cell death and somatically altered cholesterol metabolism (Vanier and Millat, 2003). In the early childhood, patients with NPC exhibit swift neurological decline, which eventually leads to death in the teen years. Histopathologically, ballooned neurons and massive neuronal cell loss can be observed in the brains of NPC patients
(Pentchev et al., 1995). The main biochemical manifestation in NPC is elevated intracellular accumulation of unesterified cholesterol caused by a genetic deficit in cholesterol trafficking (Ory, 2000; Mukherjee and Maxfield, 2004). The NPC1 gene, mutated in the majority of NPC cases, encodes an integral membrane glycoprotein known as Niemann-Pick C1 protein (NPC1) which is important in controlling the intracellular levels of free cholesterol during cholesterol homeostasis (Scott and
Ioannou, 2004; Ioannou, 2000). The cloned gene, with its predicted protein, contains a sterol-sensing domain and other structural motifs suggestive of a direct causative role for the mutant product in altered cholesterol movement in NPC
(Carstea et al., 1997; Loftus et al., 1997). NPC cells may then underestimate their intracellular cholesterol levels and build up an excess due to the sensor defect. The
66 remaining minority of NPC cases is due to dysfunctional Niemann-Pick C2 protein
(NPC2), which is a ubiquitously expressed lysosomal protein (Naureckiene et al.,
2000).
The pharmacological agent, U18666A (3-β-[2-(diethylamino)ethoxy]androst-5-en-17- one), is a well-known class-2 amphiphile which inhibits cholesterol transport (Liscum and Faust, 1989; Liscum, 1990; Underwood et al., 1996). This amphiphilic drug and other hydrophobic amines and steroids, for example imipramine (Rodriguez-Lafrasse et al., 1990; Underwood et al., 1996) and progesterone (Butler et al., 1992), have been shown to alter the trafficking and cellular distribution of several intracellular membrane proteins. U18666A may act on the activity or synthesis of another protein or lipid which facilitates cholesterol movement and possibly alters the cellular distribution of the NPC1 protein, thus accumulating intracellular cholesterol to massive levels in U18666A-treated cells, similar to that observed in cells from NPC patients (Lange et al., 2000; Lange et al., 2002; Neufeld et al., 1999; Liscum and
Sturley, 2004). U18666A is therefore widely used to mimic the cellular effects of
NPC through dysfunction of lipid storage and inhibition of cholesterol movement from the plasma membrane to the endoplasmic reticulum (ER) and from the lysosome to the plasma membrane (Härmälä et al., 1994; Underwood et al., 1998; Lange et al.,
1998). Besides cholesterol trafficking, U18666A and other steroidal compounds can disrupt the Sonic hedgehog/Patched/Smoothened signaling pathway (Incardona et al.,
2000). Reports have shown that U18666A is used in a number of biological systems to manipulate the localization of certain intracellular components. U18666A has been
67 shown to block cholesterol transport and cause a steady rise in cellular cholesterol in treated fibroblasts (Lange et al., 2000; Lange et al., 2002). U18666A also impairs intracellular transport of cholesterol and inhibits cholesterol biosynthesis in cultured
Chinese hamster ovary (CHO) cells and rat intestinal epithelial cells (Liscum and
Faust, 1989; Liscum, 1990; Sexton et al., 1983). U18666A is reported to prevent cycling of CD63 lysosome-associated membrane protein-3 from late endosomes to specialized secretory granules known as Weibel-Palade bodies in human endothelial cells incubated with U18666A (Kobayashi et al., 2000). Accumulation of tyrosinase in the endolysosomal compartment of murine-derived melanocytes is also induced after U18666A treatment (Hall et al., 2003). In addition, U18666A has been shown to cause insulin growth factor 2/mannose 6-phosphate receptor primarily localized in the trans-Golgi network to relocate to late endosomes in baby hamster kidney cells
(Kobayashi et al., 1999) and inhibit the internalization of ganglioside GM1-associated
cholera toxin to the Golgi apparatus (Sofer and Futerman, 1995). Recent studies have
also reported that U18666A treatment perturbs the cation-independent mannose 6-
phosphate receptor trafficking pathway (Tomiyama et al., 2004) and increases
intracellular calcium concentration (Ikeda et al., 2005).
Neuronal cell death can occur by means of either necrosis or apoptosis. Both necrosis
and apoptosis are generally believed to be distinct mechanisms of cell death with
different characteristic features distinguished on the basis of their morphological and
biochemical properties (Searle et al., 1982; Dive et al., 1992; Allen et al., 1997).
Necrosis is characterized by swelling of the cytoplasm and disintegration of cellular
68 structures, which ultimately lead to cell lysis and subsequent release of cellular debris. In contrast, apoptosis, also known as programmed cell death, is the regulated destruction of a cell and plays an important role in many physiological and diseased conditions (Wyllie et al., 1980). Neuronal cell death by apoptosis is believed to be one of the chief events involved in Alzheimer’s disease (AD), which is a major dementia in the elderly, and other neurodegenerative diseases (Mattson, 2000;
Friedlander, 2003; Eckert et al., 2003; Cotman and Anderson, 1995). Deregulation of apoptosis may be the cause of excessive neuronal cell loss observed in these neurodegenerative diseases.
The brain is the most cholesterol-rich organ in the body but not much is known about the mechanisms that regulate cholesterol homeostasis in the brain (Pfrieger, 2003;
Vance et al., 2005). Recently, several clinical and biochemical studies suggest that cholesterol imbalance in the brain may be a risk factor related to the development of neurological disorders such as AD and NPC (Ohm et al., 2003; Michikawa, 2004;
Burns and Duff, 2002; Wolozin, 2004). Accumulation of cholesterol is observed in fibroblasts from NPC patients and cells treated with cholesterol transport-inhibiting class-2 amphiphiles. Previous in vivo studies of neuronal injury and degenerating cerebral cortices and cerebellums in NPC mouse brains suggest that cell death may be via necrosis (Erickson and Bernard, 2002). A dose-dependent reduction in β-amyloid
(Aβ) deposition and secretion was noted after U18666A treatment in Semliki Forest virus-infected neurons and neuroblastoma cells where the amyloid precursor protein
(APP) was overexpressed, but no cell injury was reported (Runz et al., 2002). The
69 effect of U18666A in cultured murine cortical neurons and cell lines is thus investigated in the present study.
70 3.2 Results
3.2.1 U18666A induces significant cell death only in primary cortical neurons
The primary cell culture method for murine cortical neurons is well-established and provides more than 95% of neurons with minimal contamination by glia (Cheung et al., 1998). As shown in Figure 3.1A, the cells were growing in a monolayer on day 1 in vitro. On day 2 in vitro, the cells began to form neurites (Figure 3.1B). On day 3 in vitro, the neurites have extended between cell bodies (Figure 3.1C) and formed a complex network by day 5 in vitro (Figure 3.1D) when drug treatment in neat
Neurobasal™ medium will be performed. U18666A has been widely used to mimic the effects of NPC via inhibition of cholesterol transport. Previous in vitro studies have been limited to cultures of cell lines. There have been no reports on the use of
U18666A in primary cortical neurons. As such, the effect of U18666A in primary cortical neurons was determined. Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 0.1-2.5 μg/ml of U18666A for a maximum of 72 h. This range encompasses the concentrations used by most groups in cell lines. The MTT assay indicated significant MTT reduction in U18666A-treated cortical neurons in a time- and concentration-dependent manner (Figure 3.2A). Cell viability of the cortical neurons was reduced by approximately 50% after treatment with 1 μg/ml (equivalent to 2.36 μM) of U18666A for 72 h. Microscopic evaluation also revealed major morphological changes characterized by cell shrinkage and membrane blebbing in cortical neurons treated with 2.5 μg/ml of U18666A for 72 h (Figure 3.2B). To further study the effect of U18666A in various cell lines, cultures of each respective
71 cell line, namely COS-7, HEK 293, NIH/3T3, wild-type CHO, mutant 25-RA CHO and mutant CT43 CHO, were treated with vehicle (Control) or 0.1-2.5 μg/ml of
U18666A for a maximum of 72 h. The MTT assay was performed after every 24 h and morphology of cells treated with 2.5 μg/ml of U18666A for 72 h was also microscopically evaluated. No significant decrease in cell viability and changes in morphology were observed in all the cell lines studied (Figure 3.3). Although there was an approximately 10% decrease in cell viability in mutant 25-RA CHO after 72 h of treatment with 2.5 μg/ml of U18666A, no morphological changes were observed
(Figure 3.3E). The 25-RA CHO contains a gain-of-function mutation in the sterol
regulatory element-binding protein (SREBP) cleavage-activating protein (SCAP)
involved in transcriptional control of sterol-sensitive genes (Cadigan et al., 1990;
Dahl et al., 1992), which may contribute to the toxic response to U18666A after 72 h.
As the non-toxicity observed in all the cell lines studied might be due to the non-
neuronal nature of these cell lines, the effect of U18666A in neuronal cell lines was
next investigated. No significant MTT reduction and morphological changes were
observed in naïve mouse neuroblastoma Neuro-2a (Figure 3.4A) and naïve rat
pheochromocytoma-derived PC12 (Figure 3.4B). In contrast, when differentiated
PC12 cells were subjected to the same treatment conditions, a significant decrease in
cell viability was observed after 48 h and 72 h of U18666A treatment (Figure 3.4Ci),
although the toxicity observed was less severe than that in primary cortical neurons.
However, no observable morphological changes were detected in U18666A-treated
differentiated PC12 (Figure 3.4Cii). Since primary cortical neurons exhibited the
72 most response to U18666A treatment, subsequent experiments were performed using only the primary cortical neurons.
3.2.2 U18666A-mediated cell death in primary cortical neurons is apoptotic
As shown in Figure 3.5, there was no significant release of lactate dehydrogenase
(LDH) from cortical neurons treated with 0.1-2.5 μg/ml of U18666A for 72 h. Acute rapid swelling of the cortical neurons indicative of necrotic cell death was also not observed within the first 5 h of exposure to U18666A. To examine whether
U18666A-mediated cell death in cortical neurons was apoptotic, characteristic features of apoptosis were first evaluated through double-staining of cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A for 72 h with calcein acetoxymethylester (calcein AM) and ethidium homodimer-1 (EthD-1). Live cells are represented by a green fluorescence from calcein AM while dead cells are represented by a red fluorescence from EthD-1. As shown in Figure 3.6A, non-viable or severely- damaged cells represented by EthD-1-positive red fluorescence were more prevalent in cortical neurons treated with 1 μg/ml of U18666A for 72 h, as compared to the vehicle-treated Control which possessed a higher number of viable cells represented by calcein AM-positive green fluorescence. Phosphatidylserine translocation attributable to the loss of phospholipid asymmetry during apoptosis was next assessed through staining with an annexin V-FITC fluorescent probe. Annexin V-FITC fluorescence was predominant only in cortical neurons treated with 1 μg/ml of
U18666A for 72 h, as compared to the vehicle-treated Control (Figure 3.6B). In addition, nuclear morphological evaluation and quantitative assessment of apoptotic
73 nuclei using Hoechst 33258 and fluorescence microscopy revealed an increase in the percentage of cortical neurons with DNA condensation and fragmentation after 72 h of treatment with 0.1-2.5 μg/ml of U18666A, as compared to the vehicle-treated
Control (Figure 3.6C). Further examination of cortical neurons treated with vehicle
(Control) or 1 μg/ml U18666A using transmission electron microscopy after 72 h of treatment verified the typical morphological features of apoptosis, such as nuclear condensation and intact cytoplasmic membrane, only in the U18666A-treated cortical neuron, whereas the vehicle-treated cortical neuron exhibited normal morphology
(Figure 3.6D).
3.2.3 Caspase-3 activation is correlated with U18666A-mediated neuronal
apoptosis
Given the abovementioned indications of the apoptotic nature of U18666A-mediated cell death, the level of caspase-3 was detected to ascertain if procaspase-3 is activated during U18666A treatment. Cortical neurons treated with vehicle (Control) or 0.5-2.5
μg/ml of U18666A were harvested after every 24 h for Western blot analysis. Results showed that caspase-3 was activated in U18666A-treated cortical neurons in a time- dependent manner (Figure 3.7A). Co-treatment of cortical neurons with 1 μg/ml
U18666A and 25 μM Z-VAD-FMK, a broad spectrum caspase inhibitor, for 72 h resulted in significant attenuation of cell death mediated by U18666A, as well as the inhibition of active caspase-3 (Figure 3.7B). Interestingly, although there was an
inhibition of active caspase-3 upon co-treatment of cortical neurons with 1 μg/ml
U18666A and 25 μM Z-VAD-FMK for 72 h, the level of procaspase-3 was greatly
74 reduced (Figure 3.7Bii). This non-reconcilable data might be due to a direct cleavage of procaspase-3 mediated by calpain (Neumar et al., 2003).
3.2.4 Cholesterol accumulates in primary cortical neurons treated with
U18666A
To determine if U18666A caused an accumulation of intracellular cholesterol, cortical neurons treated with vehicle (Control) or 1 μg/ml of U18666A for 72 h were stained with filipin, a dye which forms a fluorescent complex with unesterified cholesterol (Miller, 1984). In the absence of U18666A, only the plasma membrane and neurites were stained intensely with filipin (Figure 3.8A). On the other hand, cortical neurons treated with U18666A induced an intracellular accumulation of cholesterol-laden vesicles, as represented by a punctate filipin staining pattern (Figure
3.8A). To investigate if this increased filipin fluorescence in U18666A-treated cortical neurons reflected an increase in intracellular cholesterol content, the levels of free cholesterol in cortical neurons treated with vehicle (Control) or 0.1-1 μg/ml of
U18666A for 72 h were measured. As shown in Figure 3.8B, the levels of intracellular free cholesterol gradually increase with increasing concentrations of
U18666A. These data implied that, besides the accumulation of cholesterol in the cell bodies of U18666A-treated cortical neurons, the level of intracellular free cholesterol also increased.
75 3.2.5 Depletion of intracellular cholesterol attenuates cell death in U18666A-
treated cortical neurons
Statins were used to inhibit the biosynthesis of cholesterol to investigate the effect of
U18666A in cholesterol-depleted cortical neurons. Cortical neurons were first treated with vehicle (Control) or 1-10 μM of lovastatin, also known as mevinolin, for 72 h.
As shown in Figure 3.9A, treatment of cortical neurons with lovastatin alone induced severe cell death, as assessed by the MTT assay. Lovastatin toxicity has been reported in murine neuroblastoma cells (Kumar et al., 2002), rat brain neuroblasts (García-
Román et al., 2001) and human glioblastoma cells (Jiang et al., 2004). Thus, lovastatin could not be used for further experiments with U18666A. The possibility of another statin, pravastatin, was next considered. As shown in Figure 3.9B, cortical neurons treated with 10-100 μM of pravastatin for 72 h did not lead to any cell injury, as compared to the vehicle-treated Control. The different effects of these two statins may be related to their physicochemical and pharmacological properties. Pravastatin
(IC50 = 1.9 μM) is a hydrophilic compound with minimal adverse effects upon treatment, whereas lovastatin (IC50 = 24 nM), of a lipophilic nature, may generate
toxicity (Gaw et al., 2004). Cortical neurons further incubated with pravastatin (10
μM and 25 μM) in the presence of 0.5 μg/ml of U18666A for 72 h exhibited significant attenuation of cell death, as determined by the MTT assay (Figure 3.10A).
To ascertain if caspase-3 activation is attenuated by pravastatin, the level of caspase-3
activity in cortical neurons exposed to 1 μg/ml of U18666A in the presence or
absence of 10 μM of pravastatin for a maximum of 72 h was determined.
Fluorometric caspase-3 activity measurement indicated an increase in caspase-3
76 activity in cortical neurons treated with U18666A alone and a significant decrease in the activity in cortical neurons co-treated with U18666A and pravastatin for 24-72 h
(Figure 3.10B). To determine if pravastatin has an effect on intracellular cholesterol content, the level of free cholesterol in cortical neurons co-treated with U18666A (1
μg/ml) and pravastatin (10 μM) for 72 h was measured. As shown in Figure 3.10C, treatment with pravastatin in the presence or absence of U18666A decreased intracellular free cholesterol levels in cortical neurons. Cyclodextrin, a water-soluble compound which can remove cholesterol from the plasma membrane, was next employed to obtain a cholesterol-lowering effect in primary cortical neurons.
Treatment of cortical neurons with 0.25-1 mM cyclodextrin alone did not lead to toxicity (Figure 3.11A). As such, 250 μM of cyclodextrin was used in co-treatment of cortical neurons with 1 μg/ml of U18666A for a maximum of 72 h. Surprisingly, the
MTT assay showed significant attenuation of cell death in U18666A-treated cortical neurons only after 48 h of co-treatment (Figure 3.11B). No neuroprotection was observed in cortical neurons co-treated with U18666A and cyclodextrin for 24 h and
72 h.
77
Figure 3.1: Primary cell culture of murine cortical neurons. Primary cortical neurons were obtained from embryonic Swiss-white mice and maintained in Neurobasal™ medium containing B27, GlutaMAX™-1 and penicillin-streptomycin, as described in the Methods. Microscopic observations of the cell culture on (A) day 1, (B) day 2, (C) day 3 and (D) day 5 in vitro showed that neurite outgrowth and complexity of the neuronal network increase with the number of days in vitro. Scale bar: 20 μm.
78
A
B
Figure 3.2: Effect of U18666A in primary cortical neurons. (A) Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 0.1-2.5 μg/ml of U18666A for a maximum of 72 h. The MTT assay was performed after every 24 h. Depicted are MTT reductions of cortical neurons treated with U18666A as a percentage of the total reductions in vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 3); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h. (B) Images of cortical neurons treated with vehicle (Control) or 2.5 μg/ml of U18666A for 72 h were acquired. Arrows: cell shrinkage and membrane blebbing. Scale bar: 20 μm.
79 A
B
C
Figure 3.3: Effect of U18666A in non-neuronal cell lines. The cell lines (A) COS-7, (B) HEK 293, (C) NIH/3T3, (D) wild-type CHO, (E) 25-RA CHO and (F) CT43 CHO were subjected to the MTT assay and microscopic evaluation. (i) The respective cell line was treated with vehicle (Control) or 0.1-2.5 μg/ml of
80 U18666A for a maximum of 72 h. The MTT assay was performed after every 24 h. Depicted are MTT reductions of each respective cell line treated with U18666A as a percentage of the total reductions in the corresponding vehicle-treated (Control) cell line. Values are the mean ± SEM (n = 3); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h. (ii) Images of the respective cell line treated with vehicle (Control) or 2.5 μg/ml of U18666A for 72 h were acquired. Scale bar: 20 μm. D
E
F
Figure 3.3: Continued. 81 A
B
C
Figure 3.4: Effect of U18666A in neuronal cell lines. The cell lines (A) naïve Neuro-2a, (B) naïve PC12 and (C) differentiated PC12 were studied. (i) The respective cell line was treated with vehicle (Control) or 0.1- 2.5 μg/ml of U18666A for a maximum of 72 h. The MTT assay was performed after every 24 h. Depicted 82 are MTT reductions of each respective cell line treated with U18666A as a percentage of the total reductions in the corresponding vehicle-treated (Control) cell line. Values are the mean ± SEM (n = 3); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h. (ii) Images of the respective cell line treated with vehicle (Control) or 2.5 μg/ml of U18666A for 72 h were acquired. Scale bar: 20 μm. Figure 3.5: Measurement of LDH released from U18666A-treated cortical neurons. Release of LDH into the medium was determined after 72 h of U18666A (0.1-2.5 μg/ml) treatment. Data are presented as an increase in LDH released from cortical neurons treated with U18666A as a percentage of total LDH released from cortical neurons lysed with 0.1% (v/v) Triton X- 100. Values are the mean ± SEM (n = 4); p < 0.01 compared with the value of total release.
83 A
B
C i
Figure 3.6: Pages 84 and 85.
84 C ii
D
Figure 3.6: U18666A-induced apoptosis in primary cortical neurons. (A) Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A for 72 h were double-stained with calcein AM and EthD-1, as described in the Methods. Calcein AM-positive cells (green fluorescence) indicate viable or healthy cells with an intact plasma membrane, whereas EthD-1-positive cells (red fluorescence) represent non-viable or severely-damaged cells. Scale bar: 10 μm. (B) Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A for 72 h were stained with annexin V-FITC probe and viewed using a fluorescence microscope. Relative to the Control, the number of cortical neurons stained with annexin V-FITC increased significantly after U18666A treatment. Scale bar: 10 μm. (C) Alterations in the nuclear morphology of cortical neurons after 72 h of vehicle (Control) or U18666A (1 μg/ml) treatment were observed through Hoechst 33258 staining and fluorescence microscopy. (i) Fragmented nuclei with condensed chromatin were displayed only in the U18666A-treated cortical neurons. Scale bar: 10 μm. (ii) The number of apoptotic nuclei was counted and expressed as a percent of the total number of cortical neurons in each randomly-chosen field containing approximately 200 cells. Values are the mean ± SEM (n = 4); *p < 0.01 compared with the Control value. (D) Transmission electron micrographs showing U18666A-mediated apoptosis in primary cortical neurons. Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A for 72 h were fixed before routine processing for transmission electron microscopy, as described in the Methods. Scale bar: 2 μm. 85 A
B i
B ii
Figure 3.7: Activation of caspase-3 during U18666A-mediated apoptosis in primary cortical neurons. (A) Cortical neurons treated with vehicle (Control) or 0.5-2.5 μg/ml of U18666A were harvested using RIPA buffer at each time-point indicated. Proteins (20 μg per lane) were separated by 15% SDS-PAGE and transferred to PVDF membrane. Caspase-3 was detected with an active caspase-3 antibody. Arrows: pro-caspase 3 (32 kDa) and active caspase-3 (17 kDa). Internal control: β-tubulin (qualitative observation showed activation of caspase-3). (B) Effect of the broad spectrum caspase inhibitor Z-VAD-FMK was investigated by co-treating cortical neurons with U18666A (1 μg/ml) and Z-VAD-FMK (25 μM). (i) The MTT assay was performed after 72 h. Depicted are MTT reductions of cortical neurons treated with U18666A in the presence or absence of Z-VAD-FMK as a percentage of the total reduction in vehicle- treated (Control) cortical neurons. Values are the mean ± SEM (n = 6); *p < 0.01 compared with the U18666A value. (ii) Cortical neurons exposed to U18666A with or without Z-VAD- FMK were harvested using RIPA buffer after 72 h. Proteins (20 μg per lane) were separated by 15% SDS-PAGE and subjected to Western blotting with an active caspase-3 antibody. Arrows: pro-caspase 3 (32 kDa) and active caspase-3 (17 kDa). Internal control for equal loading: β- tubulin. 86
A
B
Figure 3.8: Accumulation of intracellular cholesterol in cortical neurons treated with U18666A. (A) Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A for 72 h were fixed with 4% (w/v) paraformaldehyde and stained with filipin, as described in the Methods. Enhanced accumulation of cholesterol was detected only in the U18666A- treated cortical neurons. Scale bar: 10 μm. Arrow: intracellular cholesterol-laden vesicles. (B) The level of free cholesterol in each sample was determined using the Amplex® Red cholesterol assay kit, as described in the Methods. Fluorescence (excitation 563 nm, emission 587 nm) was measured and background fluorescence was subtracted from each value. Cholesterol was calculated according to a freshly-prepared standard curve generated using purified cholesterol. Results were then expressed as the percent difference in cholesterol content as compared to the vehicle-treated Control. Values are the mean ± SEM (n = 3); *p < 0.01 compared with the Control value.
87
A
B
Figure 3.9: Effect of statins in primary cortical neurons. Cortical neurons at day 5 in vitro were treated with (A) 1-10 μM of lovastatin or (B) 10-100 μM of pravastatin and their respective vehicle (Control) for 72 h, and the MTT assay was performed. Depicted are MTT reductions of cortical neurons treated with each statin as a percentage of the total reduction in their respective vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 3); *p < 0.01 compared with the respective Control value.
88 A
B
C
Figure 3.10: Inhibition of cholesterol biosynthesis with pravastatin attenuates cell death in U18666A- treated cortical neurons. (A) Effect of pravastatin was investigated by concurrently treating cortical neurons with pravastatin (10 μM and 25 μM) and U18666A (0.5 μg/ml). The MTT assay was performed after 72 h. Depicted are MTT reductions of cortical neurons treated with each condition as a percentage of the total reductions in vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 3); *p < 0.01 compared with the U18666A value. (B) Lysates from cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A in the presence or absence of 10 μM pravastatin for a maximum of 72 h were subjected to caspase-3 activity measurement. Fluorescence (excitation 400 nm, emission 505 nm) was measured and background fluorescence was subtracted from each value. Results were then expressed as the percent difference in caspase-3 activity as compared to the vehicle-treated Control. Values are the mean ± SEM (n = 3); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h. (C) Measurement of free cholesterol in each sample was determined using the Amplex® Red cholesterol assay kit. Fluorescence (excitation 563 nm, emission 587 nm) was measured and background fluorescence was subtracted from each value. Cholesterol was calculated according to a freshly-prepared standard curve generated using purified cholesterol. Results were then expressed as the percent difference in cholesterol content as compared to the vehicle-treated Control. Values are the mean ± SEM (n = 6); *p < 0.01 compared with the Control value. 89
A
B
Figure 3.11: Effect of cyclodextrin in primary cortical neurons. (A) Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 0.25-1 mM of cyclodextrin for 72 h and the MTT assay was performed. Depicted are MTT reductions of cortical neurons treated with cyclodextrin as a percentage of the total reduction in vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 3); p < 0.01 compared with the Control value. (B) Cortical neurons treated with 1 μg/ml of U18666A in the presence or absence of 250 μM cyclodextrin were subjected to the MTT assay after every 24 h. Values are the mean ± SEM (n = 3); *p < 0.01 compared with the respective Control value.
90 3.3 Discussion
U18666A, a well-known amphiphilic drug which inhibits cholesterol transport
(Liscum and Faust, 1989; Liscum, 1990), is commonly used to mimic the cellular effects of NPC due to the inhibition of cholesterol trafficking and accumulation of cholesterol (Lange et al., 2000; Lange et al., 2002). U18666A appears to interact with the NPC1 protein or its pathway involved, leading to the inhibition of cholesterol transport (Ory, 2000; Liscum and Faust, 1989).
Apoptosis and necrosis are distinct mechanisms of cell death with different characteristic features that can be distinguished on the basis of their morphological and biochemical properties (Searle et al., 1982; Dive et al., 1992; Allen et al., 1997).
In the present study, cell viability was first evaluated using the MTT assay after
U18666A insult in cell lines and cultured mouse cortical neurons. Non-neuronal and naïve neuronal cell lines were impartial to U18666A treatment. Cell death was observed only in differentiated neuronal PC12 cell line and primary cortical neurons, suggesting the vulnerability of neuron-like cells to U18666A exposure. It is also likely that U18666A can trigger apoptotic signals only in non-dividing cells. Since primary cortical neurons may more accurately provide responses which can mirror cellular changes in the central nervous system, only the primary culture system was used in subsequent experiments to study the effect of chronic exposure to U18666A.
The minimal release of LDH after exposure to U18666A for 72 h is negligible and most probably not due to necrosis. Apoptosis is characterized by a variety of
91 morphological features (Wyllie et al., 1980; Allen et al., 1997). Changes in the plasma membrane are one of the earliest of these features. Annexin V is a 35-36 kDa calcium-dependent phospholipid-binding protein that has a high affinity for the membrane phospholipid phosphatidylserine. In early apoptosis, phosphatidylserine is redistributed from the inner to the outer surface of the plasma membrane, thereby exposing phosphatidylserine to the external cellular environment. Once on the cell surface, phosphatidylserine becomes exposed for binding to annexin V (van Engeland et al., 1998). The LIVE/DEAD® viability and cytotoxicity assay also indicates prevalent non-viable and severely-damaged cells in U18666A-treated cortical
neurons. One of the later steps in apoptosis is DNA fragmentation, a process which
begins with the activation of endonucleases during the apoptotic program (Robertson
et al., 2000). These nucleases degrade chromatin into small fragments. Hoechst
staining and transmission electron microscopy confirmed DNA condensation and
fragmentation, which are typical nuclear morphological features of apoptosis, in
U18666A-treated cortical neurons. Transmission electron microscopy also verified
ultrastructural features that typified apoptosis, such as neurite blebbing and cellular
shrinkage with an intact plasma membrane.
Caspases have been shown to play a crucial role in apoptosis induced by various
deleterious and physiologic stimuli (Earnshaw et al., 1999; Grütter, 2000; Budihardjo
et al., 1999). In particular, caspase-3 has been implicated as a key cell-death protease involved in the execution phase of apoptosis, where cells undergo morphological changes such as DNA fragmentation, chromatin condensation and apoptotic body
92 formation (Porter and Jänicke, 1999; Earnshaw et al., 1999). Active caspase-3 is found in cells undergoing apoptosis and can proteolytically cleave and activate other downstream caspases, as well as relevant targets in the cytoplasm and nucleus
(Earnshaw et al., 1999; Orth et al., 1996; Slee et al., 1999). Inhibition of caspases has been shown to delay apoptosis (Ekert et al., 1999). U18666A-mediated cell death in primary cortical neurons is associated with activation of caspase-3 in a time- dependent manner, and co-treatment with the pan-caspase inhibitor Z-VAD-FMK has been shown to provide significant neuroprotection in cultures given U18666A.
Therefore, cell death caused by administration of U18666A in primary cortical neurons is exclusively by apoptosis.
NPC is associated with an intracellular accumulation of free cholesterol and presently has no therapeutic cure. A low cholesterol diet combined with cholesterol-lowering agents has been proposed as a potential treatment (Camargo et al., 2001; Sylvain et al., 1994). To address if cholesterol might be involved in U18666A-mediated neuronal apoptosis, assessments of cholesterol levels in cultured cortical neurons were conducted. Free cholesterol can be specifically and sensitively labeled with filipin, a fluorescent polyene antibiotic that binds to cholesterol (Miller, 1984). Filipin staining of U18666A-treated mouse cortical neurons showed a pattern of circumferential staining around cholesterol-containing vesicular structures, which is similar to the observations in U18666A-treated CHO/NPC1 cells (Mohammadi et al.,
2001). Other studies of filipin staining in NPC1-deficient neurons revealed numerous cholesterol-filled endosomes and lysosomes with a greater mass of cholesterol in cell
93 bodies as compared to wild-type neurons, but no injury was observed (Karten et al.,
2002). Intracellular cholesterol measurement showed marked elevation of free cholesterol in primary cortical neurons exposed to U18666A. These indicate that a major part of the storage material in U18666A-treated cortical neurons was indeed unesterified cholesterol.
The possibility of cholesterol contributing to cell death in cortical neurons was addressed by the inhibition of cholesterol biosynthesis using pravastatin and depletion of intracellular cholesterol using cyclodextrin to investigate the effect of U18666A in cholesterol-depleted cortical neurons. Administration of statins, also known as 3- hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors, has been shown to dramatically reduce brain cholesterol and may be effective in treating neurological diseases (Cucchiara and Kasner, 2001; Lütjohann et al., 2004; Menge et al., 2005). Cortical neurons co-treated with U18666A and pravastatin for 72 h showed significant attenuation of cell death, which is in association with a decrease in caspase-3 activity and intracellular free cholesterol level. Similarly, cortical neurons co-treated with U18666A and cyclodextrin showed significant attenuation of cell death only at 48 h of treatment. This might suggest that efflux of cholesterol from the cortical neurons after 24 h of treatment was not sufficient to provide neuroprotection from U18666A toxicity. On the other hand, prolonged (72 h) incubation with cyclodextrin might have led to the extensive removal of neuronal sterol which resulted in membrane instability (Kilsdonk et al., 1995; Yancey et al., 1996).
94 Cyclodextrins have been used in ameliorating neurological symptoms in a mouse model of NPC (Camargo et al., 2001).
Application of U18666A to cultured neurons has been reported to result in a dose- dependent reduction in Aβ secretion and intraneuronal Aβ deposition, indicating that
Aβ generation and secretion are reduced in neurons during inhibition of intracellular cholesterol transport (Runz et al., 2002). High levels of cholesterol seem to correlate with an increased risk of AD with modulation in the generation of Aβ (Puglielli et al.,
2001; Haley and Dietschy, 2000). As such, studies involving Aβ secretion in chronic exposure of cortical neurons to U18666A are underway to elucidate the role of cholesterol in Aβ production. No reports have demonstrated U18666A-mediated cell death via apoptosis. In the present study, data demonstrate for the first time that
U18666A induces cell death by apoptosis rather than necrosis, and that this cell injury is only observed in primary cortical neurons but not in cell lines or neuroblastoma cells. The association between U18666A-mediated neuronal apoptosis and cholesterol accumulation may provide clues to the pathogenesis of NPC. These results raise the possibility of an important in vitro model system to study how inhibition of cholesterol transport can lead to apoptosis in NPC cells.
95 Part II
Intracellular Free Radical Production and β-Amyloid Accumulation
96 II Intracellular Free Radical Production and β-Amyloid
Accumulation
3.4 Introduction
The brain is more susceptible to oxidative damage than other organs in the body. The
vulnerability of neurons to oxidative damage may be associated with a high rate of
oxygen consumption, high levels of polyunsaturated fatty acids, and the presence of
iron, a metal involved in the generation of free radicals (Halliwell and Gutteridge,
1999; Emerit et al., 2004). Neurons also contain low levels of antioxidants and
detoxifying enzyme activities, thus increasing their susceptibility to damage induced
by free radicals. Moreover, neurons are post-mitotic non-dividing cells and any
alterations in cellular processes may bring on irreversible damage to the central
nervous system (Halliwell and Gutteridge, 1999; Andersen, 2004; Emerit et al.,
2004).
A free radical in any molecule is defined as an unpaired electron that occupies an
atomic or molecular orbital on its own and confers considerable reactivity on that molecule (Halliwell and Gutteridge, 1999). Since the reactive molecule seeks another electron to pair, this can initiate an uncontrolled chain reaction that can damage the natural function of the living cell, resulting in various consequences. Oxidative stress, a form of cellular stress, results from an overproduction of reactive oxygen species
(ROS) (Halliwell and Gutteridge, 1999). Growing evidence supports the hypothesis
97 that oxidative stress produced from different conditions can trigger neuronal apoptosis in physiological settings, as well as in neurodegenerative disorders
(Andersen, 2004; Barnham et al., 2004; Emerit et al., 2004). Recent studies have reported an increased production of ROS during induction of neuronal apoptotic cell death (Andersen, 2004; Emerit et al., 2004). Cells may sustain increased levels of intracellular ROS following the initiation of an apoptotic signal. ROS may act as mediators of neuronal apoptosis during development of the central nervous system or in neurodegenerative disorders (Andersen, 2004; Halliwell and Gutteridge, 1999).
Although free radicals are produced in neurons as normal intermediates in metabolic processes, undesirable consequences such as cell death ensue during unchecked overproduction of intracellular levels of ROS (Andersen, 2004; Emerit et al., 2004).
An increase in ROS is postulated to be an early event during neuronal apoptosis.
Elevation of ROS may activate signaling pathways and expression of genes that induce apoptosis in neurons. As the incidence of many neurodegenerative diseases increases with age, it has also been proposed that oxidative stress may be involved in the aging process (Andersen, 2004; Emerit et al., 2004; Halliwell and Gutteridge,
1999). In addition, free radicals and other reactive species have been reported to play important roles in the mechanisms of action of several toxins (Halliwell and
Gutteridge, 1999).
ROS, for example hydroxyl radicals, singlet oxygen, superoxide anions and peroxynitrite, have been shown to damage chromosomal DNA and other cellular components, resulting in DNA degradation, protein denaturation and lipid
98 peroxidation (Halliwell and Gutteridge, 1999). Agents that prevent ROS formation are expected to exert protective actions in neurons subjected to oxidative stress
(Casetta et al., 2005; Moosmann and Behl, 2002; Behl and Moosmann, 2002). The search for agents with antioxidant neuroprotective action is, therefore, an important issue in neurological research. The use of antioxidants has recently emerged as a possibly useful therapy for neurodegenerative diseases (Moosmann and Behl, 2002;
Casetta et al., 2005; Behl and Moosmann, 2002). An antioxidant is any substance which significantly delays or inhibits oxidation under physiological conditions when present in a lower concentration as compared to that of an oxidizable substrate
(Halliwell and Gutteridge, 1999). In recent years, several antioxidant drugs have been examined for their ability to prevent ROS-induced neuronal cell death, with studies placing emphasis on vitamin E (Tucker and Townsend, 2005; Singh et al., 2005).
Vitamin E was first discovered in 1922 as a fat-soluble dietary substance essential for normal reproduction in rats before its molecular structure as α-tocopherol was elucidated (Bell, 1987).
An increasing body of evidence indicates that oxidative damage plays a key role in the pathological events occurring in Alzheimer’s disease (AD), one of the most common types of dementia affecting the elderly (Gibson and Huang, 2005; Smith et al., 2000; Markesbery, 1997; Pappolla et al., 2002; Eckert et al., 2003). Several oxidative stress markers are found to increase in brains from patients with AD, as compared to age-matched controls (Smith et al., 2000; Markesbery, 1997).
Furthermore, levels of oxidation end-products are elevated in AD brain tissues (Lyras
99 et al., 1997). As antioxidants are used in AD prevention and therapy (Behl and
Moosmann, 2002; Halliwell, 2001), oxidative stress is proposed to be one of the major factors contributing to the pathogenesis and progression of this disease. An initial source of oxidative stress, such as a high level of cholesterol, may initiate amyloid formation with increased oxidative stress markers (Pappolla et al., 2002).
The influences of β-amyloid (Aβ) and other genetic factors on the pathogenesis of
AD may be mediated through their effects on oxidative homeostasis (Miranda et al.,
2000; Butterfield et al., 2002; Butterfield and Lauderback, 2002). The production of
Aβ, which is derived from the proteolytic processing of the amyloid precursor protein
(APP), leads to the aggregation of Aβ peptides and deposition of amyloid plaques, which are major hallmarks of AD (Maccioni et al., 2001a). The pathway of Aβ production is thought to be responsible for the pathophysiology of AD. High cholesterol levels are also apparently correlated with an increased risk of AD through modulation of Aβ generation (Puglielli et al., 2001; Haley and Dietschy, 2000).
However, the exact mechanism by which cholesterol affects Aβ production is still unknown.
In the previous study, U18666A has been found to cause apoptotic cell death and intracellular accumulation of free cholesterol in primary cortical neurons (Chapter 3
Part I). However, the mechanism of U18666A-mediated neuronal apoptosis remains unknown. Here, the effect of U18666A in primary cortical neurons is further explored to find out if ROS are generated during U18666A treatment.
100 3.5 Results
3.5.1 U18666A treatment leads to loss of intracellular ATP and glutathione
(GSH), decrease in proteasome activity, and mitochondrial depolarization
To characterize the effects of U18666A on cellular metabolism, the levels of intracellular ATP and GSH in primary cortical neurons were examined. Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 0.1-2.5 μg/ml of
U18666A for 72 h. Intracellular ATP and GSH were measured using firefly lantern extract and o-phthaldialdehyde respectively. As shown in Figure 3.12, the levels of intracellular ATP and GSH significantly decreased in a concentration-dependent manner in cortical neurons treated with U18666A, although the U18666A-dependent
GSH loss was observed only in the presence of high (0.75-2.5 μg/ml) U18666A concentrations (Figure 3.12B). Loss of intracellular ATP and GSH was concomitant with a decrease in both the postglutamyl and chymotrypsin-like activities of the proteasome after 72 h of treatment with 0.1-2.5 μg/ml of U18666A in primary cortical neurons (Figure 3.13). To determine if the reduction in intracellular ATP level is due to depolarization of the mitochondrial membrane, changes in the mitochondrial membrane potential were monitored using TMRM and CC-1 probes. A time- and concentration-dependent decline in both the TMRM and CC-1 fluorescence indicates a loss of mitochondrial membrane potential from 24-72 h of U18666A (0.1-2.5
μg/ml) treatment in primary cortical neurons (Figure 3.14).
101 3.5.2 U18666A leads to increased intracellular reactive oxygen species (ROS)
and lipid peroxidation in primary cortical neurons
Studies have demonstrated that disturbances in the mitochondrial membrane potential could be induced by oxidative stress. U18666A-induced generation of intracellular
ROS was monitored by the fluorescence emission of DCFH-DA, which is hydrolyzed by intracellular esterases to a non-fluorescent product that is readily oxidized to the highly fluorescent DCF by ROS (LeBel et al., 1992). Measurement of ROS through
DCF fluorescence showed that U18666A (0.1-2.5 μg/ml) treatment in primary cortical neurons led to a significant time- and concentration-dependent increase in intracellular DCF fluorescence from 24-72 h, indicative of ROS production (Figure
3.15A). Further evaluation of intracellular ROS production by confocal microscopy revealed an increased intensity of DCF localized in cortical neurons treated with 1
μg/ml of U18666A for 72 h, as compared to the vehicle-treated (Control) cortical neurons (Figure 3.15B). As intense oxidative stress can lead to oxidative damage in cellular targets, the possibility of neuronal lipid peroxidation during U18666A- induced oxidative stress in primary cortical neurons was determined using two
581/591 established fluorescent probes, cis-parinaric acid and C11-BODIPY . Cis-
parinaric acid is a natural polyunsaturated fatty acid that loses its fluorescence when
oxidized by ROS (Drummen et al., 1999; Laranjinha et al., 1992). As shown in
Figure 3.16A, a significant decrease in cis-parinaric acid fluorescence was observed
in a concentration-dependent manner after 48 h of U18666A (0.1-2.5 μg/ml)
581/591 treatment. Similarly, C11-BODIPY can be used to measure lipid peroxidation by
exploiting its loss of fluorescence upon interaction with peroxyl radicals (Drummen
102 et al., 2002; Pap et al., 1999). The significant decrease in fluorescence assessed using
581/591 C11-BODIPY (Figure 3.16B) was concomitant with the observation from cis-
parinaric acid. These findings indicate an increase in lipid peroxidation in U18666A-
treated cortical neurons.
3.5.3 U18666A induces protein oxidation and DNA damage in primary cortical
neurons
Generation of ROS usually leads to the post-translational oxidative modification of
cellular proteins, such as the introduction of site-specific reactive carbonyl groups, in
which the level of formation reflects the intensity of oxidative stress (Halliwell and
Gutteridge, 1999). Protein lysates from cortical neurons treated with vehicle (Control)
or 1 μg/ml of U18666A for a maximum of 72 h were analyzed for carbonyl contents indicative of protein oxidation. Western blotting analysis using the OxyBlot™
Protein Oxidation Detection Kit indicated an increase in the level of carbonyl groups from 48 h of U18666A treatment (Figure 3.17A). Likewise, the concentration of
protein carbonyls significantly increased in cortical neurons treated with 1 μg/ml of
U18666A for 72 h, as measured quantitatively using a protein carbonyl ELISA kit
(Figure 3.17B). Free radicals have multifarious effects that include oxidative damage
to chromosomal DNA, resulting in DNA degradation and nucleotide base oxidative
modifications reported in a wide variety of neurodegenerative diseases (Halliwell and
Gutteridge, 1999). DNA was isolated from cortical neurons treated with vehicle
(Control) or 1 μg/ml of U18666A for a maximum of 72 h for analysis of DNA base
modifications by GC-MS. Exposure to U18666A resulted in significant increases in
103 the levels of 5-hydroxymethyl-hydantoin and 5-formyl-uracil from 24-72 h of treatment, 5-hydroxyuracil at 24 h and 72 h of treatment, and 5-hydroxycytosine from
48-72 h of treatment (Table 3.1). There was a marked time-dependent increase in the level of the base oxidation product 8-hydroxyguanine, which is a common biomarker of oxidative DNA damage. There were however no significant changes in the levels of 5-hydroxymethyl-uracil and FAPy-guanine from 24-72 h of U18666A treatment.
The oxidized purines FAPy-adenine and 8-hydroxyadenine substantially increased in levels at only 24 h and 48 h of U18666A treatment respectively, and might indicate that these are not the major bases to be affected by U18666A treatment.
3.5.4 Co-treatment of U18666A with vitamin E attenuates cell death during
U18666A-induced oxidative stress in primary cortical neurons
There has been increasing concern in recent years relating to neuroprotection against oxidative damage by ROS. Specific antioxidants which prevent the detrimental consequences of ROS are considered to be a promising approach for neuroprotection in neurodegenerative diseases associated with oxidative stress. To determine if antioxidants can provide neuroprotection in U18666A-mediated neuronal apoptosis, cortical neurons treated with 1 μg/ml of U18666A were further co-incubated with
2.5% (v/v) B27, a Neurobasal™ medium supplement containing a cocktail of antioxidants (Appendix I-F), or 20 μM α-tocopherol, also known as vitamin E, for a maximum of 72 h. As shown in Figure 3.18, B27 was unable to rescue cell death from 24-72 h of U18666A treatment, as determined by the MTT assay. In contrast, vitamin E exhibited significant attenuation of cell death from 48-72 h of U18666A
104 treatment (Figure 3.18), suggesting neuroprotection of cortical neurons from oxidative damage induced by U18666A.
3.5.5 U18666A leads to cleavage of APP and accumulation of Aβ in primary
cortical neurons
Several in vitro studies suggest that aggregates of Aβ, which are found in senile plaques as one of the hallmarks of AD, can directly or indirectly induce oxidative stress in cultured cells. The general neurotoxicity of Aβ appears to be mediated, at least in part, by oxidative stress. To find out if the oxidative stress observed in
U18666A-treated cortical neurons is related to Aβ, the DELFIA® Double Capture
ELISA kit specific for both Aβ40 and Aβ42 was used to determine Aβ levels.
Cortical neurons were treated with vehicle (Control) or 1 μg/ml of U18666A for a
maximum of 72 h. After every 24 h, culture medium from each treatment condition was collected for measurement of secreted Aβ40 and Aβ42, before preparation of cell lysates from the respective treatment condition for measurement of intracellular Aβ40 and Aβ42. Figure 3.19A shows that exposure of cortical neurons to 1 μg/ml of
U18666A significantly increased intracellular accumulation of both Aβ40 (Figure
3.19Ai) and Aβ42 (Figure 3.19Aiii) after 72 h of treatment, whilst significantly decreased secretion of Aβ40 from 48-72 h of treatment (Figure 3.19Aii) and Aβ42 after 72 h of treatment (Figure 3.19Aiv). As the two forms of Aβ, Aβ40 and Aβ42,
are believed to be generated from the cellular proteolytic processing of the APP, cleavage of the APP into its derivatives was investigated. Immunoblot analysis using anti-369 against the C-terminal sequence of APP revealed that cleavage of APP, with
105 the appearance of APP fragments of approximately 60-80 kDa, started at 24 h of
U18666A treatment and was most prominent after 48 h of treatment (Figure 3.19B).
To determine if inhibition of the cleavage of APP into Aβ40 and Aβ42 can rescue cell death, cortical neurons were co-treated with 1 μg/ml U18666A and 1 μM γ-secretase inhibitor IX for 48 h. Cleavage of APP by γ-secretase yields amyloidogenic Aβ peptides. However, as shown in Figure 3.19C, γ-secretase inhibitor IX was unable to attenuate U18666A toxicity.
106
A
B
Figure 3.12: U18666A-mediated loss of intracellular ATP and GSH in primary cortical neurons. Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 0.1-2.5 μg/ml of U18666A for 72 h. Intracellular ATP and GSH were measured using firefly lantern extract and o- phthaldialdehyde respectively, as described in the Methods. Depicted are intracellular levels of (A) ATP and (B) GSH as a percentage of the levels in vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 4); *p < 0.01 compared with the respective Control value.
107 A
B
Figure 3.13: Measurement of proteasome activity in U18666A-treated cortical neurons. Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 0.1-2.5 μg/ml of U18666A for a maximum of 72 h. Five μg of total proteins were incubated with the fluorogenic peptides (A) substrate II for postglutamyl activity or (B) substrate III for chymotrypsin-like activity. Fluorescence (excitation 350 nm, emission 465 nm) was measured and background fluorescence was subtracted from each value. Values are the mean ± SEM (n = 6); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h.
108 A
B
Figure 3.14: Assessment of mitochondrial membrane potential in U18666A- treated cortical neurons. Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 0.1-2.5 μg/ml of U18666A for a maximum of 72 h. The mitochondrial membrane potential was monitored using (A) the potential-sensitive probe TMRM and (B) the oxidant probe Redox Sensor Red CC-1. Fluorescence was measured after incubation and background fluorescence was subtracted from each value. Values are the mean ± SEM (n = 4); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h.
109 A
B
Figure 3.15: U18666A-mediated increase in intracellular reactive oxygen species (ROS) in primary cortical neurons. Measurement of intracellular ROS was performed with the fluorescence of DCF. (A) Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 0.1-2.5 μg/ml of U18666A for a maximum of 72 h. Fluorescence was measured after incubation with DCFH-DA and background fluorescence was subtracted from each value. Values are the mean ± SEM (n = 8); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h. (B) Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A for 72 h were loaded with DCFH-DA. Images using laser scanning confocal microscopy were then acquired, showing localization of the fluorescent DCF in U18666A-treated cortical neurons. Scale bar: 10 μm.
110 A
B
Figure 3.16: U18666A-mediated lipid peroxidation in primary cortical neurons. Cortical neurons treated with vehicle (Control) or 0.1-2.5 μg/ml of U18666A for a maximum of 72 h were loaded with (A) cis-parinaric acid 581/591 or (B) C11-BODIPY . Fluorescence was measured and background fluorescence was subtracted from each value. Values are the mean ± SEM (n = 8); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h.
111 A
B
Figure 3.17: Detection of protein oxidation in U18666A-treated cortical neurons. (A) Cortical neurons treated with vehicle (Control) or 1 μg/ml of U18666A were harvested using RIPA buffer at each time-point indicated. The DNP-derivatized protein samples were separated by 4-20% Tris-HCl linear gradient gel and transferred to PVDF membrane. Oxidation-modified proteins were detected with a DNP antibody. Internal control for equal loading: β-tubulin. (B) Concentration of protein carbonyls in cortical neurons treated with vehicle (Control) or 1 μg/ml of U18666A for 72 h was measured using an ELISA method which enables carbonyls to be determined quantitatively. Values are the mean ± SEM (n = 3); *p < 0.01 compared with the Control value.
112
Table 3.1: Levels of oxidized DNA base products in vehicle- (Control) and U18666A-treated cortical neuronal DNA, as determined by GC-MS analysis.
Oxidized base Yield (nmol/mg DNA), mean ± SEM 24 h 48 h 72 h
Control U18666A Control U18666A Control U18666A 5-OH, Me Hydantoin 0.25 ± 0.09 * 0.94 ± 0.37 0.23 ± 0.09 * 0.58 ± 0.28 0.10 ± 0.05 * 0.79 ± 0.38 5-Formyl Uracil 0.14 ± 0.05 * 0.70 ± 0.35 0.17 ± 0.07 * 0.54 ± 0.30 0.12 ± 0.06 * 0.54 ± 0.18 5-OH Uracil 0.05 ± 0.02 * 0.27 ± 0.08 0.05 ± 0.02 0.10 ± 0.03 0.02 ± 0.00 * 0.12 ± 0.05 5-OH, Me Uracil 0.02 ± 0.01 0.09 ± 0.03 0.04 ± 0.02 0.08 ± 0.03 0.03 ± 0.01 0.04 ± 0.01 5-OH Cytosine 0.14 ± 0.08 0.19 ± 0.06 0.21 ± 0.07 * 0.62 ± 0.16 0.15 ± 0.05 * 0.41 ± 0.11 FAPy Adenine 0.08 ± 0.02 * 0.24 ± 0.05 0.10 ± 0.05 0.18 ± 0.05 0.13 ± 0.06 0.20 ± 0.05 8-OH Adenine 0.05 ± 0.02 0.06 ± 0.02 0.07 ± 0.03 * 0.20 ± 0.03 0.05 ± 0.02 0.13 ± 0.03 FAPy Guanine 0.09 ± 0.02 0.04 ± 0.01 0.05 ± 0.01 0.09 ± 0.03 0.08 ± 0.07 0.05 ± 0.01 8-OH Guanine 0.17 ± 0.06 * 0.27 ± 0.03 0.15 ± 0.06 * 0.36 ± 0.09 0.07 ± 0.02 * 0.64 ± 0.21
Effect of U18666A on oxidative DNA base damage was determined by treating cortical neurons at day 5 in vitro with vehicle (Control) or 1 μg/ml of U18666A for a maximum of 72 h. Depicted are the significance of the increase in changes in DNA base oxidation products; *p < 0.01 compared with the Control value of each respective time-point. Data are the mean ± SEM of six separate experiments, each performed in triplicate, and are expressed as units of nmol/mg DNA.
113 Figure 3.18: Effect of antioxidants in U18666A-treated cortical neurons. Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 1 μg/ml of U18666A in the presence or absence of 2.5% (v/v) B27 or 20 μM vitamin E for a maximum of 72 h. The MTT assay was performed after every 24 h. Depicted are MTT reductions of cortical neurons treated with each condition as a percentage of the total reductions in vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 4); p < 0.01 compared with the respective U18666A value of *24 h, #48 h and §72 h.
114 A
B
Figure 3.19: Pages 115 and 116.
115 C
Figure 3.19: Determination of Aβ levels and APP proteolytic processing. Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 1 μg/ml of U18666A for a maximum of 72 h. (A) Levels of Aβ40 and Aβ42, both accumulated within or secreted from the cortical neurons, were measured using the DELFIA® Double Capture ELISA kit specific for both Aβ40 and Aβ42. Data are represented as the mean ± SEM (n = 8); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h. (B) APP derivatives from proteolytic processing in cortical neurons were detected using Western blotting, as described in the Methods. Reactive bands were probed with anti-369 against the C-terminal sequence of APP. Internal control for equal loading: β-tubulin. (C) Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 1 μg/ml of U18666A in the presence or absence of 1 μM γ-secretase inhibitor IX for 72 h. The MTT assay was performed. Depicted are MTT reductions of cortical neurons treated with each condition as a percentage of the total reductions in vehicle- treated (Control) cortical neurons. Values are the mean ± SEM (n = 3); *p < 0.01 compared with the Control value.
116 3.6 Discussion
Oxidative stress can produce cell injury by multiple pathways. Often, these pathways overlap and interact in complex ways. There has been heightened interest in the role of oxidative stress in neurological disorders, particularly in AD, where there is evidence that free radicals play a role in the pathogenesis and progression of this neurodegenerative disease (Gibson and Huang, 2005; Smith et al., 2000; Markesbery,
1997; Pappolla et al., 2002; Eckert et al., 2003).
U18666A is a well-known pharmacological agent which inhibits cholesterol transport
(Liscum and Faust, 1989; Liscum, 1990; Underwood et al., 1996). Besides causing apoptotic cell death and intracellular accumulation of free cholesterol (Chapter 3 Part
I), U18666A has been found to lead to a depletion of intracellular ATP and GSH in primary cortical neurons after chronic exposure. GSH is one of the most abundant intracellular thiols in the central nervous system and functions as a major cellular antioxidant with a crucial role as a scavenger of toxic free radicals (Cooper and
Kristal, 1997). GSH provides protection against several toxic oxygen-derived chemical species, acts as an essential cofactor of many enzymes, and is also involved in many other cellular functions, including regulation of protein and DNA synthesis.
Loss of GSH in oxidative damage has been suggested to increase the vulnerability of neurons towards a wide array of insults (Monks et al., 1999). A decrease in the level of intracellular GSH might indicate that the capacity of U18666A-treated cortical
117 neurons to compensate oxidative stress was diminished and oxidative stress could therefore contribute to U18666A-mediated neuronal apoptosis.
Decreased GSH availability in the brain is believed to promote mitochondrial damage via increases in the levels of ROS to this organelle (Cooper and Kristal, 1997).
Cumulative oxidative damage is often proposed as the major cause of mitochondrial dysfunction, as the mitochondrial respiratory chain is a major source of damaging intracellular free radicals (Sullivan and Brown, 2005; Sastre et al., 2000). The main function of the mitochondria is energy production through oxidative phosphorylation and lipid oxidation (Alberts et al., 2002). Mitochondria are responsible for generating most of the ATP required for all functions of the cell. Other functions performed by the mitochondria include urea production, control of the cellular redox state, and intracellular calcium homeostasis (Alberts et al., 2002). Oxidative damage to the mitochondria might interfere with all these functions. Defects in oxidative phosphorylation of the mitochondria inhibit ATP production and also increase production of mitochondrial ROS (Green and Reed, 1998). A change in the mitochondrial membrane potential is a key feature of the early stages of mitochondrial dysfunction and may contribute to the execution of programmed cell death (Kroemer and Reed, 2000; Bernardi et al., 1999). In the present study,
U18666A led to a loss of mitochondrial membrane potential, which might cause mitochondrial dysfunction, decrease in ATP production and ultimately neuronal apoptosis, in the treated cortical neurons. Preservation of the mitochondrial membrane potential is therefore essential during normal conditions, as well as
118 conditions of stress and disease (Kroemer and Reed, 2000). Progressive loss of the mitochondrial membrane potential, in addition to mitochondrial dysfunction via oxidative stress, have been attributed to cell death and may be causative of the aging process in several neurodegenerative disorders (Sullivan and Brown, 2005; Beal,
2005).
The proteasome has been implicated in aging and the pathogenesis of a variety of neurodegenerative diseases (Keller et al., 2002; Gaczynska et al., 2001; Keller et al.,
2000b; Halliwell, 2002). Impaired proteasome function has been reported in AD
(Keller et al., 2000a; Lam et al., 2000; Checler et al., 2000). The present study suggests that U18666A treatment led to loss of postglutamyl peptidase and chymotrypsin-like peptidase activities of the proteasome in primary cortical neurons.
Inhibition of proteasome activity has been postulated to be partially due to enhanced oxidative stress, which may cause disruption of the proteasome complex (Halliwell,
2002; Keller et al., 2000b; Lee et al., 2001). Inhibition of proteasome function is also widely reported to activate the cell death program (Drexler, 1997). Ubiquitination of protein substrates before degradation via the ubiquitin-proteasome pathway is an energy-dependent process and consumes relatively large amounts of ATP
(Hochstrasser, 1996; Tanaka, 1998). Decreased proteasome activity during U18666A treatment might affect ubiquitin-dependent protein degradation by the proteasome, leading to an accumulation of oxidized proteins within U18666A-treated cortical neurons.
119 Oxidative modification of cellular protein may involve generation of increased levels of ROS, thereby promoting further modifications to the protein population and altering the normal functions of these proteins (Stadtman and Levine, 2000; Levine,
2002). Damage to proteins can occur by direct attack of ROS upon specific amino acids or by secondary damage involving attack by end-products of lipid peroxidation
(Stadtman and Levine, 2000; Butterfield and Lauderback, 2002). The biological importance of oxidative damage to proteins has only recently been considered in detail. The delay in placing attention on studies involving protein oxidation may indicate that many proteins can actually sustain considerable damage before impairment of their observed functions. Protein oxidation has been increasingly implicated in the pathology of neurodegenerative disorders associated with protein aggregation, such as in AD (Butterfield and Kanski, 2001; Halliwell and Jenner,
1998; Butterfield and Lauderback, 2002). Protein carbonyls are formed by a variety of oxidative mechanisms and are sensitive indices reflecting damage to proteins from oxidative injury (Levine, 2002; Berlett and Stadtman, 1997). The carbonyl assay is based on the fact that ROS attack amino acid residues, predominantly histidine, arginine, lysine and proline, to produce proteins with carbonyl groups. The increase in protein carbonyls detected in cortical neurons after U18666A treatment suggests that increased oxidative stress during U18666A treatment could result in protein oxidation. The decrease in proteasome activity after U18666A treatment might also impair clearance of oxidative-damaged proteins.
120 Oxidative damage to proteins can also contribute to secondary damage in other biomolecules, such as inactivation of DNA repair enzymes and loss of fidelity of
DNA polymerases, which could increase oxidative DNA damage levels and raise mutation frequency (Cadet et al., 2000). A variety of ROS are also capable of causing oxidative damage to DNA, culminating in chemical changes to both pyrimidine and purine bases, as well as a variety of other lesions (Halliwell and Gutteridge, 1999).
Here, as shown by GC-MS, U18666A caused a significant increase in the purine- derived DNA base damage product, 8-OH guanine, in isolated DNA from primary cortical neurons. The level of 8-OH guanine, which is the most frequently measured indicator of oxidative DNA damage, has been reported to be elevated in AD (Lyras et al., 1997). Increases in other oxidized purines and pyrimidines were also detected in
DNA from U18666A-treated cortical neurons. Analysis of DNA damage by GC-MS enables identification of the whole profile of DNA base oxidation products (Jenner et al., 1998), which can help to indicate the possible damaging reactive species involved during U18666A treatment. The pattern of DNA oxidation to purines and pyrimidines suggests that destructive hydroxyl radicals might be involved during U18666A- mediated neuronal apoptosis.
Lipids are also susceptible to damage by ROS. U18666A has been found to lead to an increase in lipid peroxidation in treated cortical neurons. Lipid peroxidation is initiated when activated oxygen reacts with the double bonds on the lipid hydrocarbon chains (Niki et al., 2005). Vitamin E or more correctly, α-tocopherol, is a scavenger of peroxyl radicals and probably the most important, but not the only,
121 inhibitor of the chain reactions of free radicals leading to lipid peroxidation (Halliwell and Gutteridge, 1999). Tocopherols are able to inhibit lipid peroxidation largely because they scavenge lipid peroxyl radicals much faster than these radicals can react with adjacent fatty acid side-chains or with membrane proteins (Halliwell and
Gutteridge, 1999; Witting et al., 1998). The effectiveness of vitamin E in attenuating cell death suggests the involvement of peroxyl radicals in U18666A-mediated neuronal apoptosis. Vitamin E has previously been shown to exert neuroprotection against oxidative stress and neurotoxicity in cultured neurons and cell lines (Osakada et al., 2003; Butterfield et al., 1999; Behl, 2000; Kölsch et al., 2001).
Aβ peptides can be neurotoxic and may increase neuronal sensitivity to oxidative stress. Studies have shown that Aβ-induced oxidative damage of neuronal membrane proteins and lipids are involved in cell degeneration (Mark et al., 1997a; Mark et al.,
1997b). In the present study, U18666A treatment showed that intracellular accumulation of Aβ40 and Aβ42 increased, whereas the secretion of Aβ40 and Aβ42 decreased. Inhibition of intracellular cholesterol transport in U18666A-treated cortical neurons seemed to block Aβ secretion, as less Aβ was measured from the culture media, as compared to the accumulation of Aβ inside the cortical neurons after 72 h of treatment. Intracellular Aβ may be the initial site of Aβ aggregation and neurotoxicity. In particular, accumulation of intraneuronal Aβ42, which is less soluble and thus more resistant to degradation (Skovronsky et al., 1998; Knauer et al.,
1992), has been proposed to be an early and toxic event in the pathogenesis of AD
(Selkoe, 2000; Gouras et al., 2000). Accumulation of intraneuronal Aβ42 may
122 partially contribute to the neurotoxicity induced by U18666A. The implication that alterations in cholesterol metabolism may be involved in the modulation of Aβ generation has led to a further study of a potential relationship between inhibition of intracellular cholesterol transport and APP metabolism in U18666A-treated cortical neurons. U18666A treatment seemed to promote proteolytic processing of full-length
APP, generating fragments which could be further cleaved into Aβ peptides. In contrast to these results, observations by Runz et al. (2002) suggested a dose- dependent reduction in Aβ deposition and secretion with no cell injury after U18666A treatment in Semliki Forest virus-infected neurons and neuroblastoma cells, in which
APP has been overexpressed. The γ-secretase inhibitor IX was unable to attenuate
U18666A toxicity. Further work should be performed to determine if co-treatment with γ-secretase inhibitor IX could suppress the generation of Aβ40 and Aβ42 during
U18666A treatment, or that the Aβ40 and Aβ42 were cleaved from APP by other proteases instead. Studies have suggested that calpains are activated in the brains of
AD patients (Saito et al., 1993) and may also participate in the proteolytic processing of APP into Aβ peptides (Kuwako et al., 2002; Figueiredo-Pereira et al., 1999; Siman et al., 1990).
In the present study, U18666A induced decreased intracellular ATP and GSH levels, decreased proteasome activity, mitochondrial depolarization, increased intracellular
ROS production, and the occurrence of other oxidative events, such as lipid peroxidation, DNA oxidation and protein oxidation, in primary cortical neurons.
Although the mechanism of U18666A-mediated neuronal apoptosis remains elusive,
123 the observation of a sizeable increase of markers associated with oxidative damage demonstrates that the consequences of oxidative stress may contribute to the apoptotic process in primary cortical neurons during U18666A treatment. The association of oxidative stress and inhibition of intracellular cholesterol transport with
Aβ accumulation in U18666A-treated cortical neurons may provide clues to AD pathogenesis, and may also be useful for the study of other neurodegenerative disorders associated with oxidative stress. Indeed, Yu et al. (2005a) recently reported with similar findings as the present study that altered cholesterol metabolism in
Niemann-Pick disease type C (NPC) mouse brains induced mitochondrial dysfunction and ATP deficiency. Taken together, AD and NPC may share similarities in the neuropathological features through a common pathway involving cholesterol metabolism, leading to neurodegeneration.
124 Part III
Role of Caspases, Calpains, Kinases and Cell Cycle Regulatory Proteins
125 III Role of Caspases, Calpains, Kinases and Cell Cycle Regulatory
Proteins
3.7 Introduction
Apoptosis plays a central role in tissue remodeling during the normal development of
the mammalian nervous system (Yuan and Yankner, 2000; Nijhawan et al., 2000).
There is increasing evidence that neurons often die by apoptosis in cell culture and animal models of chronic human neurodegenerative disorders. Deregulation of apoptosis has been well-documented in several human pathologies. Under pathophysiological conditions, the inappropriate activation of apoptosis may be responsible for the neuronal cell loss seen in neurodegenerative diseases such as
Alzheimer’s, Parkinson’s and Huntington’s diseases (Mattson, 2000; Friedlander,
2003; Eckert et al., 2003; Cotman and Anderson, 1995).
Alzheimer’s disease (AD), one of the most common types of dementia affecting the
elderly, is a progressive neurodegenerative disease characterized by the accumulation of neurofibrillary tangles and amyloid plaques. Recent studies suggest that caspase activation occurs in AD and leads to the proteolytic cleavage of several neuronal proteins (Cotman et al., 2005; Rissman et al., 2004; Kang et al., 2005; Chung et al.,
2001). Caspases are cysteine proteases that mediate neuronal injury, including apoptosis, in a wide range of cell systems. Because they bring about most of the visible changes that characterize apoptotic cell death, it is largely accepted that
126 caspases play a key role in both the initiation and execution pathways of apoptosis
(Earnshaw et al., 1999; Grütter, 2000; Budihardjo et al., 1999). Apart from caspases, activation of other proteases such as calpains also contributes to apoptosis. Calpains are a family of calcium-dependent thiol-proteases which have been proposed to participate in many physiological processes, as well as pathological conditions
(Sorimachi et al., 1997; Wang, 2000). Several studies have shown that calpains can contribute to neuronal cell death (Neumar et al., 2003; Nakagawa and Yuan, 2000;
Harwood et al., 2005), and calpain inhibitors have been used to successfully block apoptosis (Villa et al., 1998; McGinnis et al., 1998; Jordán et al., 1997). Although widely studied, the precise mechanisms involved in calpain activation and the regulation of its activity remain poorly understood.
The main component of neurofibrillary tangles (NFT) is the microtubule-associated protein tau in hyperphosphorylation (Iqbal and Grundke-Iqbal, 2004). In the normal brain, tau plays a major role in the assembly and disassembly of microtubules, and in bridging microtubules with other cytoskeletal filaments (Kurt et al., 1997).
Hyperphosphorylated tau is the major component of paired helical filaments, the building block of NFT observed in brains from patients with AD (Kobayashi et al.,
2003). The precise mechanism that leads to the formation of NFT in AD is still not known. However, it is believed that cellular mechanisms by which tau becomes hyperphosphorylated and assembles into paired helical filaments may actively contribute to the pathogenesis of AD. Although tau is phosphorylated by numerous kinases such as the glycogen synthase kinase 3 (GSK3) and mitogen-activated protein
127 kinase (MAPK) in vivo, recent attention has focused on the ubiquitously-expressed cyclin-dependent kinase 5 (Cdk5) as an important tau kinase whose activity is enhanced in response to β-amyloid (Aβ) in cultured neurons and AD brains (Alvarez et al., 1999; Li et al., 2003). The dysregulation of Cdk5 causes hyperphosphorylation of tau, thereby contributing to the formation of NFT in AD (Maccioni et al., 2001b).
Increased tau phosphorylation by Cdk5 leads to decreased microtubule stabilization
and cytoskeleton disruption (Saito et al., 2003). Cdk5 and its specific cyclin-related
activator protein p35, which is exclusively expressed in neuronal cells, are required
for the migratory behavior of post-mitotic neurons during brain development and
function of the mature nervous system (Saito et al., 2003). The GSK3 is another
protein kinase originally identified due to its role in glycogen metabolism regulation
(Woodgett, 1990). It is highly expressed in neural tissues and its activity is subjected to intense regulation which involves phosphorylation of its two isoforms, GSK3α and
GSK3β. The GSK3β isoform accumulates in the cytoplasm of AD pretangle neurons and tau deposits in most tauopathies, and has thus been implicated as another key enzyme in regulating tau hyperphosphorylation (Ferrer et al., 2002). Recent studies have also suggested the involvement of MAPK and stress-activated protein kinase/c-
Jun N-terminal kinase (SAPK/JNK) in the aberrant phosphorylation of tau. These kinases function in protein kinase cascades that play a critical role in the regulation of cell growth and differentiation, as well as in the control of cellular responses to cytokines and stress (Ferrer et al., 2005). Given that more than half of the phosphorylation sites in tau are serine and threonine residues followed by proline
128 (Hanger et al., 1998), it is conceivable that members of the MAPK family play an important role in tau hyperphosphorylation.
Inappropriate cell cycle entry in response to defects in the cell cycle control pathway may be a general mechanism through which cell death occurs. Recent findings indicate the expression of cell cycle-related proteins in the neurons of AD patients and in cultured primary neurons undergoing programmed cell death, suggesting the potential involvement of cell cycle markers and breach of cell cycle checkpoints in
AD and neuronal apoptosis (McShea et al., 1999; Nagy et al., 1998). Progression through the cell cycle is mediated by the activation of a series of cyclin-dependent kinases (Cdks) associated with regulatory subunits termed cyclins (Grana and Reddy,
1995). These cyclin/Cdk complexes are further regulated by changes in the cyclin- dependent kinase inhibitor (CKI) levels (Grana and Reddy, 1995; Sherr and Roberts,
1999). Cell cycle disturbances can be deleterious and may promote many of the pathological events in neurodegenerative diseases. However, the purpose of cell cycle activation remains elusive.
Previously, it has been demonstrated that treatment of primary cortical neurons with
U18666A results in the accumulation of cholesterol, intracellular free radical and Aβ production, and induction of apoptotic cell death (Chapter 3 Parts I and II). Here, the potential roles which caspases and calpains play in U18666A-mediated neuronal apoptosis in primary cortical neurons were examined. Kinases reported to regulate tau phosphorylation in vivo were also studied.
129 3.8 Results
3.8.1 U18666A induces caspase and calpain activation in primary cortical
neurons
Previous data established that caspase-3 was activated in U18666A-induced apoptosis and cell viability of primary cortical neurons exposed to U18666A (1 μg/ml) for 72 h is reduced by about 50% (Chapter 3 Part I). In the present study, cortical neurons at day 5 in vitro were treated with 1 μg/ml U18666A for a maximum of 72 h. Caspase activities for caspase-1, -2, -3, -4, -5, -6, -8, -9 and -10 were measured using cell lysates prepared after 24 h, 36 h, 48 h and 72 h of treatment. No significant change in activity was present for each caspase after 24 h of U18666A treatment, as compared to their respective vehicle-treated Control (Figure 3.20A). However, activities of caspase-2, -3, -4, -5, -6, -8, -9 and -10 were significantly increased in U18666A- treated cortical neurons from 36 h of treatment onwards (Figure 3.20B-D). Using
Western blotting, increased activation of caspase-12 and increased cleavage of poly(ADP-ribose)polymerase (PARP) were also observed in cortical neurons treated with 1 μg/ml U18666A from 24-72 h of treatment (Figure 3.21). No full length PARP
was detected as the antibody recognizes only the apoptosis-related cleavage fragment
of PARP. In addition, α-fodrin, also known as spectrin and is an endogenous substrate
of calpain, was cleaved to 145 kDa and 150 kDa fragments from 48 h of U18666A (1
μg/ml) treatment, suggesting an activation of calpain, as observed through Western
blot analysis (Figure 3.22A, lane denoted U). An additional cleavage product of 120
kDa, together with the 150 kDa fragment, also indicated caspase-3 cleavage of α-
130 fodrin after 1 μg/ml U18666A treatment from 48 h (Figure 3.22A, lane denoted U).
Co-treatment of cortical neurons with 1 μg/ml U18666A and 50 μM caspase-3 inhibitor IV after 48 h resulted in a partial inhibition of the 120 kDa fragment from caspase-3 cleavage of α-fodrin induced by U18666A (Figure 3.22A, lane denoted U +
C3). A separate co-treatment of 1 μg/ml U18666A with 25 μM Z-VAD-FMK, a broad spectrum caspase inhibitor, was able to attenuate the formation of the 120 kDa cleavage product from 48 h onwards (Figure 3.22A, lane denoted U + Z). Another co- treatment of 1 μg/ml U18666A with 10 μM calpeptin, a selective inhibitor of calpain, led to a decrease in the 145 kDa and 150 kDa cleavage products from calpain cleavage of α-fodrin after 48 h of treatment, but was unable to inhibit formation of the
120 kDa fragment from the caspase-3 cleavage (Figure 3.22A, lane denoted U + CP).
U18666A (1 μg/ml) treatment with both 25 μM Z-VAD-FMK and 10 μM calpeptin abolished the formation of the 120 kDa cleavage product (Figure 3.22A, lane denoted
U + Z + CP), as similarly observed in the co-treatment of U18666A with Z-VAD-
FMK alone. However, the 145 kDa and 150 kDa fragments could not be blocked. No cleavage of α-fodrin was observed in the vehicle-treated Controls (Figure 3.22A, lane denoted C), as well as 24 h of drug treatment. A further MTT reduction assay after 72 h (Figure 3.22B) confirmed that 25 μM Z-VAD-FMK was able to significantly rescue cell death from U18666A (1 μg/ml) toxicity. However, Z-VAD-FMK (25 μM) and calpeptin (10 μM) with 1 μg/ml U18666A could only partially reverse the toxicity, whereas calpeptin (10 μM) with U18666A (1 μg/ml) showed no significant difference from the U18666A alone treatment.
131 3.8.2 U18666A induces hyperphosphorylation of tau in primary cortical
neurons
Immunofluorescence microscopy was employed to monitor the effect of 1 μg/ml
U18666A on tau phosphorylation in primary cortical neurons. The distribution and phosphorylation of tau were monitored using four specific monoclonal anti-tau antibodies, AT8, AT180, PHF1 and Tau-1, which are commonly used in the diagnosis of tau in AD. AT8, AT180 and PHF1 react with epitopes that are phosphorylated respectively at serine 202/threonine 205, threonine 231 and serine 396/404. On the other hand, Tau-1 reacts with epitopes that are dephosphorylated at serine 195/199 or serine 199/202. Positive tau signals were observed only in cortical neurons treated with 1 μg/ml U18666A (Table 3.2). An intraneuronal accumulation of hyperphosphorylated tau was observed with AT8 (Figure 3.23Ai) and AT180 (Figure
3.23Aii) in U18666A-treated cortical neurons from 24-72 h, whereas tau phosphorylation detected by PHF1 (Figure 3.23Aiii) was moderately induced from
48-72 h of U18666A treatment. Likewise, dephosphorylation of tau was detected with
Tau-1 only from 48-72 h of U18666A treatment (Figure 3.23Aiv). The possibility that tau phosphorylation occurs in cells undergoing apoptosis was further examined using
Hoechst 33258 staining after immunocytochemistry. As shown in Figure 3.23B, staining with AT8, AT180 and Tau-1 in U18666A-treated cortical neurons was restricted to cells with apoptotic nuclei after 72 h of treatment, suggesting that
U18666A-mediated neuronal apoptosis is correlated with the presence of hyperphosphorylated tau.
132 3.8.3 U18666A decreases the kinase activity of Cdk5 and the protein level of
p35 in primary cortical neurons
Cortical neurons were treated at day 5 in vitro with 1 μg/ml U18666A in the presence or absence of 5 μM roscovitine, a specific Cdk5 inhibitor, for a maximum of 72 h.
Roscovitine was unable to rescue cell death mediated by U18666A, as determined by the MTT assay (Figure 3.24A). In addition, the Cdk5 kinase activity in cortical neurons treated with 1 μg/ml U18666A for a maximum of 72 h decreased significantly in a time-dependent manner (Figure 3.24B). To examine whether the loss of Cdk5 activity is due to a decrease in its protein level, cell lysates of cortical neurons exposed to vehicle (Control) or 1 μg/ml of U18666A from 24-72 h were subjected to Western blot analysis. Results showed that, while the level of Cdk5 protein remained unchanged after treatment with U18666A, the level of the activator protein p35 decreased upon U18666A treatment (Figure 3.24C).
3.8.4 U18666A decreases the protein levels of phosphorylated GSK3, p44/42
MAPK and SAPK/JNK in primary cortical neurons
To determine if the protein levels of other kinases are affected, cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A were harvested after every 24 h for Western blot analysis. The protein level of phosphorylated GSK3 decreased upon
U18666A treatment (Figure 3.25A). Cortical neurons treated at day 5 in vitro with 1
μg/ml U18666A and 1 mM lithium chloride, an inhibitor of GSK3, was unable to reverse U18666A-mediated toxicity and resulted in decreased cell viability, as determined by the MTT assay after 48 h (Figure 3.25B). A further Western blot
133 analysis showed that there was also a decrease in the protein levels of phosphorylated p44/42 MAPK and phosphorylated SAPK/JNK after U18666A treatment, even though there was no significant change in the non-phosphorylated levels of these kinases (Figure 3.26A). To determine if an upstream kinase could be involved in the activation of other downstream kinases not investigated in the present study, cortical neurons treated with 1 μg/ml of U18666A were co-incubated with 1 μM of PD98059, a selective inhibitor of MAPK kinase, for 72 h. As shown in Figure 3.26B, PD98059 moderately attenuated U18666A-mediated neuronal apoptosis, as determined by the
MTT assay. In contrast, 10 μM of Y-27632, an inhibitor of Rho-associated protein kinases, was unable to reverse U18666A-mediated neurotoxicity in primary cortical neurons when co-treated with 1 μg/ml of U18666A (Figure 3.27).
3.8.5 U18666A increases the protein levels of phospho-p53 and activates cell
cycle machinery in primary cortical neurons
To determine if proteins involved in cell cycle regulation play a role in U18666A- mediated neuronal apoptosis, cortical neurons treated with vehicle (Control) or 1
μg/ml U18666A were harvested after every 24 h for Western blot analysis. The tumor-suppressor protein p53 phosphorylated at serine 15 increased in protein level in a time-dependent manner upon U18666A treatment (Figure 3.28A). Western blot analysis using antibodies representing cell cycle regulatory proteins showed that there was also an increase in the protein level of cyclin D3 in U18666A-treated cortical neurons from 24-72 h of treatment (Figure 3.28A). On the other hand, despite an increase in the protein level of p27 Kip1 from 24-48 h of U18666A treatment, its
134 level decreased after 72 h (Figure 3.28A). No significant change in the protein level of Cdk4 was detected (Figure 3.28A). To determine if inhibition of cell cycle proteins can provide neuroprotection to U18666A-treated cortical neurons, cortical neurons treated with 1 μg/ml of U18666A were co-incubated with 1 μM of 3-ATA, a Cdk4 inhibitor, for 72 h. As shown in Figure 3.28B, 3-ATA was not able to attenuate
U18666A-mediated neuronal apoptosis, as determined by the MTT assay.
135
A
B
C
D
Figure 3.20: Measurement of caspase activity in primary cortical neurons. Cell lysates from cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A for (A) 24 h, (B) 36 h, (C) 48 h and (D) 72 h were subjected to caspase activity measurement for caspase-1, -2, -3, -4, -5, -6, -8, -9 and -10. Fluorescence (excitation 400 nm, emission 505 nm) was measured and background fluorescence was subtracted from each value. Results were then expressed as the percent difference in each caspase activity as compared to the respective vehicle-treated Control. Data are represented as the mean ± SEM (n = 3); *p < 0.01 compared with the respective Control value.
136
Figure 3.21: Activation of caspase-12 and cleavage of PARP during U18666A- mediated apoptosis in primary cortical neurons. Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A were harvested using the total protein isolation method at each time-point indicated. Proteins (10 μg per lane) were separated by 10% SDS-PAGE and transferred to PVDF membrane. Caspase-12 and cleaved PARP were detected with the respective antibodies. Arrows: pro- caspase-12 (55 kDa) and cleaved caspase-12 (42 kDa). Internal control for equal loading: β-tubulin.
137 A
B
Figure 3.22: Cleavage of α-fodrin in U18666A-treated cortical neurons. (A) Cortical neurons treated with vehicle (Control, lane denoted C), 1 μg/ml U18666A (lane denoted U), 1 μg/ml U18666A and 50 μM caspase-3 inhibitor IV (lane denoted U + C3), 1 μg/ml U18666A and 25 μM Z-VAD-FMK (lane denoted U + Z), 1 μg/ml U18666A and 10 μM calpeptin (lane denoted U + CP) or 1 μg/ml U18666A, 25 μM Z- VAD-FMK and 10 μM calpeptin (lane denoted U + Z + CP) were harvested using RIPA buffer at each time-point indicated. Proteins (10 μg per lane) were separated by 6% SDS-PAGE and transferred to PVDF membrane. Arrows: α-fodrin (250 kDa) and its cleaved fragments (150 kDa, 145 kDa and 120 kDa). Internal control for equal loading: β-tubulin. (B) Effect of 1 μg/ml U18666A with 25 μM Z-VAD-FMK, 10 μM calpeptin or both was monitored by treating cortical neurons at day 5 in vitro with these drugs. The MTT assay was performed after 72 h. Depicted are the MTT reductions of cortical neurons treated with the respective drugs as a percentage of the total reduction in the vehicle-treated (Control) cortical neurons. Data are represented as the mean ± SEM (n = 3); *p < 0.01 compared with the U18666A value.
138
Table 3.2: Tau phosphorylation in vehicle- (Control) or U18666A-treated cortical neurons.
Tau phosphorylation 24 h 48 h 72 h
Antibody Site(s) Control U18666A Control U18666A Control U18666A Phosphorylated AT8 Ser202/Thr205 – + – + – + AT180 Thr231 – + – + – + PHF1 Ser396/Ser404 – – – + – +
Dephosphorylated Tau-1 Ser199/Ser202 – – – + – +
Phosphorylation of tau in cortical neurons treated with vehicle (Control) or 1 μg/ml of U18666A for a maximum of 72 h was observed through fluorescence microscopy after immunocytochemistry with the various phosphorylation- or dephosphorylation-dependent anti-tau antibodies. (+) denotes positive labeling of the respective tau within the cell bodies of cortical neurons, whereas (–) denotes absence of labeling within the cell bodies.
139 A
Figure 3.23: Pages 140 and 141.
140 B
Figure 3.23: Distribution of hyperphosphorylated tau in primary cortical neurons by immunocytochemistry. (A) Cortical neurons at day 5 in vitro were treated with vehicle (Control) or 1 μg/ml of U18666A for a maximum of 72 h before fixing. Images using fluorescence microscopy were then acquired after labeling with antibodies to (i) AT8, (ii) AT180, (iii) PHF1 or (iv) Tau-1. Scale bar: 20 μm. Arrows: intraneuronal accumulation of hyperphosphorylated tau. (B) Cortical neurons were fixed after 72 h of U18666A (1 μg/ml) treatment, followed by double immunofluorescence with (i) AT8, (ii) AT180 or (iii) Tau-1 and Hoechst 33258 to visualize nuclear morphology. Digital images were acquired using fluorescence microscopy. Scale bar: 10 μm. Arrows: co-localization of hyperphosphorylated tau with apoptotic nucleus.
141 A
B
C
Figure 3.24: Cdk5 in U18666A-treated cortical neurons. (A) Effect of the Cdk5 inhibitor, roscovitine (5 μM), was monitored concurrently with 1 μg/ml U18666A for a maximum of 72 h. The MTT assay was performed after every 24 h. Depicted are the MTT reductions of cortical neurons treated with U18666A in the presence or absence of roscovitine as a percentage of the total reduction in the vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 3); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h. (B) The activity of Cdk5 in cortical neurons treated with vehicle (Control) or 1 μg/ml of U18666A for a maximum of 72 h was determined. Background activity was then deducted from the assay data. Values are the mean ± SEM (n = 6); p < 0.01 compared with the respective Control value of *24 h, #48 h and §72 h. (C) Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A were harvested using RIPA buffer at each time-point indicated. Proteins (10 μg per lane) were separated by 10% SDS-PAGE and analyzed by Western blotting. Cdk5 and its specific activator protein p35 were detected with anti-Cdk5 (C-8) and anti-p35 (C-19) respectively. Internal control for equal loading: β-tubulin.
142 A
B
Figure 3.25: GSK3 in U18666A-treated cortical neurons. (A) Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A were harvested using RIPA buffer at each time-point indicated. Proteins (10 μg per lane) were separated by 10% SDS-PAGE and analyzed by Western blotting. Phosphorylated GSK3 was detected with anti-GSK-3α/β [pY279/216]. Arrows: GSK3α (51 kDa) and GSK3β (47 kDa). Internal control for equal loading: β-tubulin. (B) Effect of the GSK3 inhibitor, lithium chloride (1 mM), was monitored concurrently with 1 μg/ml U18666A for 48 h. Depicted are the MTT reductions of cortical neurons treated with U18666A in the presence or absence of lithium chloride as a percentage of the total reduction in the vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 3); *p < 0.01 compared with the Control value.
143 A
B
Figure 3.26: Phosphorylated and non-phosphorylated p44/42 MAPK and SAPK/JNK in U18666A-treated cortical neurons. (A) Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A were harvested using RIPA buffer at each time-point indicated. Proteins (10 μg per lane) were separated by 10% SDS-PAGE and analyzed by Western blotting. Phosphorylated and non-phosphorylated kinases were detected with the respective antibodies for p44/42 MAPK and SAPK/JNK. Arrows: double bands of p44/42 MAPK (44 kDa and 42 kDa) and SAPK/JNK (54 kDa and 46 kDa). Internal control for equal loading: β-tubulin. (B) Effect of the MAPK kinase inhibitor, PD98059 (1 μM), was monitored concurrently with 1 μg/ml U18666A for 72 h. Depicted are the MTT reductions of cortical neurons treated with U18666A in the presence or absence of PD98059 as a percentage of the total reduction in the vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 3); *p < 0.01 compared with the Control value.
144 Figure 3.27: Effect of Y-27632 in U18666A-treated cortical neurons. Effect of Y-27632 (10 μM), an inhibitor of Rho-associated protein kinases, was monitored concurrently with 1 μg/ml U18666A for 72 h. Depicted are the MTT reductions of cortical neurons treated with U18666A in the presence or absence of Y-27632 as a percentage of the total reduction in the vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 3); *p < 0.01 compared with the Control value.
145 A
B
Figure 3.28: Cell cycle regulatory proteins in U18666A-treated cortical neurons. (A) Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A were harvested using the total protein isolation method at each time-point indicated. Proteins (10 μg per lane) were separated by 10% SDS-PAGE and analyzed by Western blotting. Cell cycle regulatory proteins were detected with anti-phospho-p53 (Ser15), anti-Cdk4, anti-p27 Kip1 and anti-cyclin D3. Internal control for equal loading: β-tubulin. (B) Effect of the Cdk4 inhibitor, 3-ATA (1 μM), was monitored concurrently with 1 μg/ml U18666A for 72 h. Depicted are the MTT reductions of cortical neurons treated with U18666A in the presence or absence of 3-ATA as a percentage of the total reduction in the vehicle-treated (Control) cortical neurons. Values are the mean ± SEM (n = 3); *p < 0.01 compared with the Control value.
146 3.9 Discussion
It is well documented that caspases, a family of cysteine acid proteases, are the central executioners of apoptosis induced by various deleterious and physiologic stimuli (Earnshaw et al., 1999; Grütter, 2000; Budihardjo et al., 1999). Caspase proteases play an essential role in apoptosis by degrading specific structural regulatory and DNA repair proteins within a cell. U18666A-induced cell death in cortical neuronal cultures is associated with the activation of caspase-12 after 24 h, and activation of caspase-2, -3, -4, -5, -6, -8, -9 and -10 after 36 h of treatment.
Caspase-12 is predominantly found as a pro-enzyme in the endoplasmic reticulum
(ER), and it has been shown that activation of calpain is required for pro-caspase-12 processing (Nakagawa and Yuan, 2000). Stress in the ER can also promote cleavage and activation of caspase-12 to initiate activation of other caspases, resulting in apoptosis (Nakagawa et al., 2000; Siman et al., 2001). Recent data have also indicated that excess free cholesterol may exert its cytotoxic effects via the ER (Feng et al., 2003). As cleavage of α-fodrin leading to the formation of the 145 kDa and 150 kDa fragments, a well-recognized marker for the calpain-generated protein breakdown, only occurred after 48 h of U18666A treatment, cleavage of pro-caspase-
12 at an earlier time-point before calpain activation in U18666A treatment may suggest that ER stress could play a role in U18666A-mediated neuronal apoptosis.
Activation of other initiator caspases such as caspase-8, -9 and -10 could cleave and lead to the activation of downstream effector caspases including caspase-2, -3 and -6, which in turn can then cleave PARP. PARP cleavage is widely known as a marker for
147 apoptosis and has been shown to correlate with chromatin condensation in apoptotic cells (Earnshaw et al., 1999). Caspase-2 has been reported to mediate neuronal cell death induced by Aβ (Troy et al., 2000). In addition, caspase-8 has been earlier found to be activated in brains from AD patients (Rohn et al., 2001). Recent studies have also suggested the involvement of caspase-8 in the progression of neuronal cell death in Niemann-Pick disease type C (NPC) mice (Wu et al., 2005). Likewise, caspase-9 has been found to be activated in AD brains (Rohn et al., 2002).
Calpains are calcium-dependent proteases that have also been proposed to participate in neuronal apoptosis (Neumar et al., 2003; Nakagawa and Yuan, 2000; Harwood et al., 2005). Caspase and calpain inhibitors have been employed to block apoptosis in the U18666A-treated neuronal cell system. Treatment with the broad spectrum caspase inhibitor Z-VAD-FMK inhibited the activation of caspases, but was unable to completely prevent U18666A-induced apoptosis in primary cortical neurons. This indicates the likely contribution of an additional caspase-independent pathway. The fact that calpains might also participate in this neuronal apoptosis and were still active in the presence of Z-VAD-FMK could explain why the inactivation of caspases by Z-
VAD-FMK alone did not lead to inhibition of apoptosis. The addition of a cell- permeable calpain inhibitor, calpeptin, was also not able to completely block the cell death. This may reflect that there could be multiple mechanisms of cell death which can be triggered after U18666A treatment. There is growing evidence of significant crosstalk between caspases and calpains in the literature (Neumar et al., 2003;
Nakagawa and Yuan, 2000; Siman et al., 2001; Blomgren et al., 2001; Harwood et
148 al., 2005). Uncontrolled calpain activity may be playing a downstream role from the caspases in the U18666A apoptotic phase. These results suggest a synergistic role for both the caspase and calpain proteolytic pathways in the U18666A-mediated apoptotic process.
Accumulation of the highly phosphorylated microtubule-associated protein tau in the
form of NFT has been well-documented in AD. Tau phosphorylation occurs at serine
and threonine residues, and hyperphosphorylation of the protein promotes it to self-
assemble into tangles of paired helical filaments (Kurt et al., 1997; Kobayashi et al.,
2003). It has been reported that the abnormal phosphorylation of tau associated with
AD is preferentially located in the neurites and occurs at serine 202 (Su et al., 1994), which can be recognized by the AT8 monoclonal antibody (Goedert et al., 1995).
Other phosphorylation sites of the tau protein are threonine 231 detectable by AT180
(Goedert et al., 1994) and serine 396/404 detectable by PHF1 (Otvos Jr et al., 1994).
On the contrary, Tau-1 detects serine 199/202 when these sites are unphosphorylated
(Rosner et al., 1994). The presence of positive labeling with AT8, AT180 and PHF1 antibodies in U18666A-treated cortical neurons suggests that the accumulation of intracellular cholesterol may lead to site-specific phosphorylations of tau similar to those observed in AD and other neurodegenerative diseases such as NPC. The slightly less intense labeling with Tau-1 might be due to enhanced phosphorylation of tau at site serine 202, which is recognized by AT8, rather than dephosphorylation at this site recognizable by Tau-1. Studies have also previously reported that neurons with altered dephosphorylated tau distribution are linked to apoptosis due to tau
149 cleavage and not tau dephosphorylation (Rametti et al., 2004). It has been indicated that intraneuronal elevated free cholesterol plays an important role in the formation of
NFT, which are present both in the senile dementia of AD and juvenile dementia of
NPC (Distl et al., 2001; Treiber-Held et al., 2003). The understanding of tau phosphorylation under physiological conditions might help to elucidate possible mechanisms involved in the hyperphosphorylation in AD and NPC. There has been evidence suggesting similarities in the neuropathological features of AD and NPC.
Recently, Yu et al. (2005b) reported that neurodegeneration in NPC mice is associated with enhanced phosphorylation of tau and impaired cholesterol trafficking.
Therefore, it may be possible that AD and NPC share a common pathway involving changes in cholesterol metabolism. The finding that nuclear fragmentation correlates with tau-positive U18666A-treated cortical neurons suggests that aberrant tau phosphorylation occurs in conjunction with cell death, and implies a signaling pathway involving tau and the apoptotic machinery.
Although many signaling pathways are likely to modulate tau phosphorylation in vivo, little is known about the specific enzymes involved. Identification of the protein kinases that phosphorylate tau is important for understanding both the normal and pathological regulation of tau function. One possible protein kinase is the Cdk5, which plays important roles in several neuronal processes occurring within the central nervous system (Maccioni et al., 2001b). The involvement of Cdk5 in cytoskeleton impairment resulting from abnormal tau phosphorylation is strongly supported by the presence of this protein kinase in the NFT from AD brains (Alvarez et al., 1999; Li et
150 al., 2003; Maccioni et al., 2001b). Results presented in the present study demonstrate that the level of Cdk5 remained unchanged while its activity decreased significantly in apoptotic cortical neurons treated with U18666A. It is interesting to note that in contrast to these data, findings by others revealed that an activation of Cdk5 induces apoptosis in neuronal cells (Ahuja et al., 1997; Henchcliffe and Burke, 1997). As the protein level of the Cdk5 activator p35 decreased upon U18666A treatment, this could explain the consequent decrease in Cdk5 activity. The decrease in p35 levels might indicate degradation of this neuronal specific Cdk5 activator by the ubiquitin- proteasome pathway, as reported in other studies (Patrick et al., 1998; Saito et al.,
1998; Saito et al., 2003).
Many other protein kinases, such as the GSK3, MAPK and SAPK/JNK, can also phosphorylate tau in vitro. The level of phosphorylated GSK3 was found to decrease after U18666A treatment. However, GSK3 is not involved in the U18666A-induced tau phosphorylation response, as treatment with lithium chloride as a potent inhibitor of GSK3 activity did not eliminate U18666A-mediated neuronal apoptosis, even though the concentration used (1 mM) was that reported with neuroprotection in other studies (Phiel et al., 2003; Nonaka et al., 1998; Hashimoto et al., 2002b). Lithium chloride might have interfered with another upstream intracellular transduction pathway leading to neuronal apoptosis in the cultured neurons. MAPK and
SAPK/JNK could also be key enzymes in the pathological phosphorylation of the tau protein (Ferrer et al., 2005). The p44/42 MAPK, also known as Erk1/2, is activated by growth and neurotrophic factors. On the other hand, SAPK/JNK is activated by
151 inflammatory cytokines and a wide variety of cellular stresses. Collectively, these kinases function in protein kinase cascades that play a critical role in the regulation of cell growth, and the differentiation and control of cellular responses to cytokines and stress (Ferrer et al., 2005). Abnormal activation of these kinases occurs through phosphorylation of threonine and tyrosine residues, and could contribute to tau hyperphosphorylation characteristic of AD brains. Previous studies indicate that
MAPK is activated in NPC mouse brains (Sawamura et al., 2001; Yu et al., 2005b).
However, neither the p44/42 MAPK nor SAPK/JNK was activated in U18666A- treated cortical neurons. The moderate neuroprotection observed with PD98059, a potent inhibitor of an upstream MAPK kinase in the MAPK signaling cascade
(Dudley et al., 1995), suggests that other kinases, other than those studied here, might be involved in the U18666A-mediated tau hyperphosphorylation. As MAPK has been postulated to phosphorylate and activate calpain, the neuroprotection exerted by
PD98059 might also be due to the suppression of MAPK-mediated calpain activation
(Huang et al., 2004). However, the Rho-associated family of protein serine/threonine kinases are unlikely to be involved in the U18666A-induced tau phosphorylation response as their specific inhibitor, Y-27632 (Narumiya et al., 2000), was unable to attenuate U18666A-mediated cell death. On the other hand, the abnormal phosphorylation of tau after U18666A treatment might arise from a decrease in the activity of a tau phosphatase, instead of from the activation of a kinase (Hunter,
1995). Further work will be necessary to elucidate which kinase or phosphatase is responsible for U18666A-induced tau phosphorylation.
152 The vast majority of neurons is not proliferating but rather, exists in a quiescence state of the cell cycle at any given time. Studies indicate that neurodegeneration in
AD is associated with cell cycle disturbances in terminally-differentiated post-mitotic neurons (McShea et al., 1999; Nagy et al., 1998). Recently, DNA damage has been found to activate a cell death program in post-mitotic neurons (Kruman et al., 2004).
The p53 tumor-suppressor protein plays a major role in cellular response to DNA damage and other genomic aberrations (Lakin and Jackson, 1999; Miller et al., 2000).
Activation of p53 can lead to either apoptosis or cell cycle arrest and DNA repair.
The level of p53 phosphorylated at serine 15 increased in U18666A-treated cortical neurons, probably in response to DNA damage. Oxidative DNA damage in
U18666A-treated cortical neurons has been elucidated previously through GC-MS analysis in Chapter 3 Part II. Phosphorylation of p53 at serine 15, possibly induced by ataxia telangiectasia-mutated protein kinase (ATM), ATM and Rad3-related protein kinase (ATR) or the DNA-activated protein kinase (Lewin, 1994), is thought to reduce interaction of p53 with its negative regulator Mdm2, thereby stabilizing p53 by preventing its ubiquitination and proteasomal degradation (Lakin and Jackson,
1999).
Binding of a Cdk to its positive regulatory cyclin subunit is essential for the activation of the kinase. The different Cdks are generally abundant throughout the cell cycle, therefore it is mainly the regulated synthesis and destruction of each cyclin that controls the activity of the cyclin/Cdk complexes (Sherr, 1996). The expression of cyclin D begins in response to mitogenic stimulation and controls exit from the cell
153 cycle together with Cdk4 or Cdk6 (Sherr, 1995). The Cdk inhibitor (CKI), p27 Kip1, plays an important role in cell cycle control. In quiescent cells such as post-mitotic neurons, CKIs are present in excess of the cyclin/Cdk complexes. The amount of p27
Kip1 tends to be high in quiescent cells, decreases upon entry of cells into the cell cycle, and is controlled predominantly by the rate of its degradation in a proteasome- dependent manner (Hengst and Reed, 1996; Pagano et al., 1995). Upon receiving mitotic signals, p27 Kip1 becomes phosphorylated and is targeted for degradation, hence reducing the overall level of CKIs (Pagano et al., 1995). The increase in the level of cyclin D3 concomitant with the decrease in the level of p27 Kip1 observed after 72 h of U18666A treatment suggest that U18666A might activate cell cycle machinery. Studies have previously suggested that a possible mechanism mediating neuronal apoptosis could be the activation of cell cycle regulatory proteins, a non- permissive condition for the neurons due to their post-mitotic state (Copani et al.,
1999; Padmanabhan et al., 1999). This situation can create a conflict of signals that result in cell death by apoptosis (Nagy et al., 1998; Giovanni et al., 1999; Raina et al.,
2000). It has been demonstrated that Aβ can also induce an increase in the expression of cell cycle proteins in cortical neurons (Copani et al., 1999; Copani et al., 2001).
The levels of intracellular Aβ40 and Aβ42 have previously been found to increase upon U18666A treatment (Chapter 3 Part II) and may also indirectly contribute, at least in part, to the activation of cell cycle regulatory proteins observed in U18666A- treated cortical neurons. Although treatment with 3-ATA as a potent inhibitor of
Cdk4 activity did not attenuate U18666A-mediated neuronal apoptosis, this might indicate that another Cdk, such as Cdk6, is implicated during U18666A exposure.
154 Nonetheless, more studies are required to determine the role of these cell cycle regulatory proteins in U18666A-mediated neuronal apoptosis.
In summary, the present study shows that neuronal apoptosis mediated by U18666A leads to a potential crosstalk between the caspase and calpain pathways, suggesting a loop that may act to amplify the apoptotic cascade. Both pathways are implicated, but the exact contribution of each individual protease in the neuronal apoptosis remains unclear. It is also demonstrated that U18666A treatment leads to hyperphosphorylation of tau in cultured neurons. Nevertheless, it is still to be clarified which kinase or phosphatase is responsible for the increased levels of tau phosphorylation in U18666A-treated cortical neurons. Results in the present study have also suggested that chronic exposure to U18666A might activate cell cycle machinery, leading to a conflict of signals which resulted in neuronal apoptosis. The association of intracellular cholesterol accumulation and tauopathies may provide a novel in vitro model system to elucidate the etiology and pathogenesis of AD, and the cellular mechanism of the inhibition of cholesterol transport-mediated cell death in other neurodegenerative diseases with similar pathological features, including NPC.
Although in vitro studies undoubtedly cannot reproduce the physiological conditions of an organism, they constitute a useful tool to characterize molecular events taking place under tightly controlled conditions.
155 CHAPTER 4
GLOBAL GENE EXPRESSION PROFILE OF U18666A-MEDIATED NEURONAL APOPTOSIS
156 4 Global Gene Expression Profile of U18666A-Mediated
Neuronal Apoptosis
4.1 Introduction
Traditional molecular research tools for gene expression studies have been limited to
the analysis of one or a small group of genes at a time. The availability of the probes and the interest and expertise of the researcher used to dictate which genes would be chosen for the study. However, with the development of microarray technology, the expression levels of thousands of genes can now be simultaneously examined at the same time in a single experiment (Lockhart and Winzeler, 2000). Due to the thousands of distinct reporters on one array, each microarray experiment is therefore equivalent to the same number of genetic tests carried out in parallel. Since microarrays have the possibility to incorporate thousands of genes, it would not be impossible to scan the entire set of genes of a particular organism, thus enabling a complete comparison of the expression levels of almost all transcribed genes on a genomic scale (Brown and Botstein, 1999). The influence of microarray analysis has been powerful in both basic and applied biology. This technology has dramatically accelerated many scientific investigations by altering the way biologists approach complex problems. Unmanageable amounts of microarray data have even stimulated the development of bioinformatics as a new science area (Schulze and Downward,
2001; Gershon, 2002).
157 The use of microarrays for expression profiling was first published in 1995 by the laboratory of Patrick Brown from Stanford University (Schena et al., 1995).
Microarrays consist of ordered sets of DNA probes carefully chosen to record the expression of specific genes and are immobilized onto a solid glass support about 1.5 cm2 in size at precise physical locations. Microarrays can be fabricated using a variety
of technologies, including printing with fine-pointed pins onto glass slides,
photolithography using pre-made masks or dynamic micromirror devices, ink-jet
printing, and electrochemistry on microelectrode arrays (Fodor et al., 1993; Lipshutz
et al., 1999). The in situ synthesis of oligonucleotide arrays was originally developed
by Fodor et al. (1991) and is now commonly referred to as the Affymetrix GeneChip® system. In these oligonucleotide microarrays, DNA oligonucleotides are synthesized in situ onto each GeneChip® array using photolabile protecting groups and
photolithographic masks to add selective sequences of nucleotides (Lipshutz et al.,
1999; Schulze and Downward, 2001). Up to 1.3 million different oligonucleotide
probes are synthesized onto each array in a series of cycles and affixed in a known
location referred to as a probe cell. Basically, the Affymetrix GeneChip® system first
uses in vitro transcription of the complimentary RNA (cRNA) to incorporate biotin-
labeled nucleotides before hybridization onto the GeneChip®. The cRNA, having a
sequence complementary to one of the single-stranded probe sequences on the array,
will hybridize via base pairing to the spot at which the complementary reporters are
affixed. Only one sample can be hybridized onto each array and comparisons of the
one-color hybridization are made among multiple arrays.
158 Global gene expression profiles in cells or tissues may help to elucidate the molecular basis of disease pathology, drug treatment or phenotype. Microarray analysis enables these through the detection of global changes of gene expression in samples derived from normal and diseased tissues, treated and non-treated time courses, or different stages of differentiation and development (Schulze and Downward, 2001). Significant data on gene expression profiles under different conditions can be obtained with the addition of appropriate controls and repeated experiments. Advances in the microarray field have also led to many other potential applications including the identification of gene copies in a genome (Pollack et al., 1999; Pinkel et al., 1998), detection and analysis of mutations and polymorphisms (Wang et al., 1998), development of diagnostic tools for diseases (Gershon, 2005), and drug discovery
(Marton et al., 1998; Geschwind, 2003). In addition, microarray technology has helped to bridge the gap between sequence information and functional genomics
(Lockhart and Winzeler, 2000; Brown and Botstein, 1999). Microarray analysis can also be combined with chromatin immunoprecipitation to perform genome-wide identification of transcription factors and their respective binding sites (Buck and
Lieb, 2004; Hanlon and Lieb, 2004). Computational analysis of all microarray data further permits reliable interpretation after the classification of known and unknown genes based on their expression patterns (Armstrong and van de Wiel, 2004).
Recently, the study of gene expression on a global scale using microarrays has received a great deal of attention in the analysis of diseases and the unraveling of cellular signaling pathways. One area in which microarray analysis has received
159 significant attention is in neurobiology (Lockhart and Barlow, 2001; Geschwind,
2000). Gene expression mapping in multiple brain regions has been used to determine the genetic causes and molecular mechanisms responsible for neurobehavioral differences in mice (Sandberg et al., 2000). Studies using microarrays to determine gene expression changes occurring in the neocortex and cerebellum of aging mice have shown that brain aging in the mice might be comparable to changes in human neurodegenerative disorders at the transcriptional level (Lee et al., 2000a). Similarly, microarrays have been extensively used to measure transcript expression profiles or search for molecular markers and pathways involved in the pathogenesis of
Alzheimer’s disease (AD) (Ginsberg et al., 2000; Colangelo et al., 2002; Emilsson et al., 2006). Studies for the development of new therapies for diseases without suitable animal models, such as schizophrenia, also involved microarray analysis of gross brain samples to reveal alterations in specific metabolic pathways (Mirnics et al.,
2000; Hakak et al., 2001; Middleton et al., 2002).
Gene expression profiling through microarray analysis may provide useful starting points for more in-depth investigation to explore the molecular changes that occur after U18666A treatment and search for signature pathways leading to U18666A- mediated neuronal apoptosis. Studying which genes are expressed and which are not after U18666A treatment at different time-points can help to unravel these mysteries.
It is very likely that time-dependent microarray data will be reflective of the phases of
U18666A activity in primary cortical neurons. Initial responses of cortical neurons to
U18666A within 24 h of exposure, which has yet to present acute toxicity, may
160 provide data on specific genes involved in the pharmacological action of the drug. As the exposure to U18666A is increased in time to a maximum of 72 h, cellular injury or toxicity becomes progressively obvious and various adaptive functions may be expressed. Such data will allow insight into the mode or mechanism of U18666A toxicity and also provide a means of distinguishing array patterns indicative of the adverse effects of the agent.
161 4.2 Results
4.2.1 Determination of RNA integrity and cRNA fragmentation before target
hybridization onto GeneChip® probe arrays
Quality assessments are vital in obtaining highly reproducible GeneChip® probe array
results. Cortical neurons at day 5 in vitro were treated with vehicle or 1 μg/ml
U18666A for a maximum of 72 h and total RNA was isolated after every 24 h of
treatment. After spectrophotometric quantification, integrity of the RNA was checked
using agarose gel electrophoresis to detect for any degradation of the samples. Two
distinct bands representing the larger 28S and smaller 18S ribosomal RNA were
identified (Figure 4.1), indicating that the total RNA samples were of good quality
and could be used as starting material for microarray analysis. To ensure that the
biotin-labeled cRNA products were unfragmented during the in vitro transcription
reaction and fragmented prior to target preparation, the respective samples were
assessed by agarose gel electrophoresis. As shown in Figure 4.2, the unfragmented
cRNA was confirmed by the presence of a smear, whereas the fragmented cRNA was
represented by a single thick band. Fragmentation of the cRNA target before
hybridization onto GeneChip® probe arrays is required to achieve optimal assay
sensitivity.
162 4.2.2 Differential gene expression after U18666A treatment in primary cortical
neurons
To understand the mechanism of U18666A-mediated apoptosis in primary cortical neurons, gene expression profiling was performed using GeneChip® murine genome
U74A set (version 2) probe arrays from Affymetrix to study changes in gene
transcription during 24-72 h of U18666A (1 μg/ml) treatment. High-quality arrays
were obtained in all of the 6 GeneChip® Test3 arrays and 18 GeneChip® murine
genome arrays, as determined by visual inspection of each array image and
monitoring fundamental parameters such as the ratios of housekeeping genes. Each of
the GeneChip® murine genome array contains 12,488 probes for genes and expressed sequence tags. The raw expression data from each probe array were scaled to an
average intensity of 500 and the expression levels for all probes were expressed as
signal log ratios obtained from the Affymetrix Microarray Suite (version 5.0)
software. The spiked Affymetrix controls and probes absent across all of the 18 probe
arrays were deleted before subsequent data analysis.
One-way ANOVA for multiple comparisons. Out of the 12,488 probes on each of
the GeneChip® murine genome array, 626 genes were found with at least 2-fold
change after one-way ANOVA at p < 0.01 upon U18666A treatment, as identified
using GeneSpring™ software. These genes were then classified according to their
biological functions using NetAffx™ and DAVID (Table 4.1). Out of the 626 genes,
517 genes exhibited known biological functions whereas 109 genes were with unknown functions. The Venn diagram generated by GeneSpring™ after one-way
ANOVA for multiple comparisons (Figure 4.3A) indicated that 105 (red region), 71
163 (green region) and 138 (blue region) genes with at least 2-fold change were exclusively differentially expressed in cortical neurons after 24 h, 48 h and 72 h of
U18666A treatment respectively. Besides these, 134 genes were differentially expressed after 24 h and 48 h of U18666A treatment (yellow region), 9 genes were differentially expressed after 24 h and 72 h of U18666A treatment (pink region), and
76 genes were differentially expressed after 48 h and 72 h of U18666A treatment
(cyan region). A total of 93 genes were differentially expressed upon U18666A treatment at all three time-points (white region).
Bonferroni multiple testing correction. After applying a conservative Bonferroni multiple testing correction at *p < 0.05 to control the number of false-positive genes identified by chance (Jung et al., 2005), only 130 genes were selected by
GeneSpring™ (Table 4.1). Among which, 111 genes were with known biological functions while 19 were with unknown functions. The Venn diagram generated by
GeneSpring™ after the Bonferroni correction (Figure 4.3B) indicated that 7 (red region), 11 (green region) and 25 (blue region) genes with at least 2-fold change were exclusively differentially expressed in cortical neurons after 24 h, 48 h and 72 h of
U18666A treatment respectively. In addition, 36 genes were differentially expressed after 24 h and 48 h of U18666A treatment (yellow region), 4 genes were differentially expressed after 24 h and 72 h of U18666A treatment (pink region), and 24 genes were differentially expressed after 48 h and 72 h of U18666A treatment (cyan region). A total of 23 genes were differentially expressed upon U18666A treatment at all three time-points (white region).
164 4.2.3 Cluster analysis of genes differentially expressed after U18666A
treatment in primary cortical neurons
Differentially-expressed genes with at least 2-fold change classified under categories of interest were further clustered using GeneSpring™ according to their changes in expression level over time. Clustering was performed on the average of normalized expression in each condition representing cortical neurons treated with vehicle
(Control) or 1 μg/ml U18666A for a maximum of 72 h. Genes were organized on the basis of overall similarity in their expression. Other categories of differentially- expressed genes not selected for clustering include genes encoding for proteins involved in genesis and development, cell growth and maintenance, protein
biosynthesis, protein folding, carbohydrate metabolism, and enzymatic activity (Table
4.1). Genes with biological functions not relevant to all the categories in Table 4.1
were grouped under “Others” while genes with currently no known biological functions were grouped under “Unknown” (Table 4.1). Both of these groups were also not subjected to further cluster analysis.
Cell adhesion. From Figure 4.4B representing 18 differentially-expressed genes
encoding for proteins involved in cell adhesion, more than half of the genes were
down-regulated by 72 h of U18666A treatment. In particular, the gene encoding for
follistatin (Fst) was significantly down-regulated in a time-dependent manner upon
U18666A treatment. The remaining genes, except those encoding for cadherin 13
(Cdh13) and cell adhesion molecule with homology to L1CAM (Chl1), were only
prominently up-regulated after 72 h of U18666A treatment.
165 Cytoskeleton organization. From Figure 4.4C representing 24 genes grouped under cytoskeleton organization, majority of the genes, such as those encoding for tubulin isotypes, were up-regulated from 24 h of U18666A treatment. On the other hand, the gene encoding for cytoplasmic β-actin (Actb) was significantly down-regulated throughout U18666A treatment.
Cell cycle. From the 16 genes encoding for proteins involved in cell cycle (Figure
4.4D), most were generally up-regulated by 72 h of U18666A treatment. The gene encoding for FBJ osteosarcoma oncogene (Fos) was significantly up-regulated after
48 h and 72 h of U18666A treatment. Among the few genes that were down-regulated
by U18666A in this category, the ras homolog gene family member U (Rhou) was the
most significant after 72 h of treatment.
Cell death. As shown in Figure 4.4E representing 12 genes grouped under cell death,
only two genes were down-regulated from 24-72 h of U18666A treatment. The other
genes were notably up-regulated, particularly the genes encoding for BCL2- associated transcription factor 1 (Bclaf1), clusterin (Clu), DAZ associated protein 2
(Dazap2) and growth arrest specific 2 (Gas2) after 72 h of U18666A treatment.
Defense response. More than two-thirds of the 16 genes grouped under defense
response were up-regulated after 48 h of U18666A treatment (Figure 4.4F). However,
5 genes were down-regulated after U18666A treatment, particularly for the gene
encoding for the M region locus 9 of histocompatibility 2 (H2-M9).
Response to stress. All of the 13 genes grouped under response to stress were up- regulated upon U18666A treatment (Figure 4.4G), especially the genes encoding for
166 DNA-damage inducible transcript 3 (Ddit3) and immediate early response 2 (Ier2), which peaked at 48 h of treatment.
Cholesterol biosynthesis. Ten out of 11 genes grouped under cholesterol biosynthesis were down-regulated after 72 h of U18666A treatment (Figure 4.4H), supporting the role of U18666A in the inhibition of intracellular cholesterol transport.
The only gene which remained up-regulated was the one encoding for 3-hydroxy-3- methylglutaryl-coenzyme A lyase (Hmgcl).
Lipid metabolism and transport. Almost all of the 15 genes grouped under lipid metabolism and transport were up-regulated after 72 h of U18666A treatment (Figure
4.4I), except those encoding for elongation of very long chain fatty acids-like 2
(Elovl2), phospholipase C β1 (Plcb1) and stearoyl-coenzyme A desaturase 1 (Scd1)
which were instead down-regulated upon treatment.
Protein amino acid modification. About half of the 29 genes grouped under protein
amino acid modification, made up mostly of genes encoding for kinases and
phosphatases, were up-regulated while the other half were down-regulated after 48 h
and 72 h of U18666A treatment (Figure 4.4J).
Proteolysis and peptidolysis. In Figure 4.4K representing 14 genes grouped under
proteolysis and peptidolysis, more than half of the genes were up-regulated upon
U18666A treatment. Among the genes encoding for ring finger proteins, that for ring finger protein 20 (Rnf20) was significantly down-regulated after 24 h and 48 h of
U18666A treatment while the rest were up-regulated. Contrastingly, the genes encoding for a disintegrin and metalloproteinase domain 15 (Adam15) and calpain 2
(Capn2) were up-regulated while their family members, a disintegrin and
167 metalloproteinase domain 19 (Adam19) and calpain 3 (Capn3), were down-regulated from 24-72 h of U18666A treatment.
Ubiquitin-proteasome system. Two of the 15 genes, namely ubiquitin specific
protease 34 (Usp34) and ubiquitin-conjugating enzyme E2E3 (Ube2e3), grouped
under ubiquitin-proteasome system were down-regulated from 24-72 h of U18666A
treatment (Figure 4.4L). The remaining genes from this category were up-regulated,
particularly the gene encoding for F-box only protein 8 (Fbxo8) which showed
significant up-regulation in a time-dependent manner upon U18666A treatment.
Regulation of transcription. From Figure 4.4M representing 50 differentially-
expressed genes encoding for proteins involved in the regulation of transcription,
most were up-regulated from 24-72 h of U18666A treatment. Among which, genes
encoding for nuclear factor I/A (Nfia), polymerase (RNA) II (DNA directed)
polypeptide K (Polr2k) and SKI interacting protein (Skiip) exhibited most
significance in a time-dependent up-regulation. In contrast, the genes encoding for
AE binding protein 2 (Aebp2), LIM homeobox protein 8 (Lhx8) and myocyte
enhancer factor 2C (Mef2c) significantly decreased after 72 h of U18666A treatment.
DNA and RNA processing. From Figure 4.4N representing 22 genes grouped under
DNA processing and Figure 4.4O representing 14 genes grouped under RNA
processing, most genes were up-regulated from 24-72 h of U18666A treatment. The
up-regulated genes from DNA processing were significantly increased at a later time-
point while those from RNA processing were significantly increased at an earlier
time-point. Out of the down-regulated genes, those encoding for nucleoplasmin 3
(Npm3) under DNA processing and heterogeneous nuclear ribonucleoprotein U
168 (Hnrpu) under RNA processing were the most significantly down-regulated after 48 h of U18666A treatment.
Transport. More than half of the 29 genes grouped under transport were up- regulated from 24-72 h of U18666A treatment (Figure 4.4P). In particular, the genes encoding for synaptotagmin 4 (Syt4) and synaptotagmin 11 (Syt11) were significantly up-regulated after 24 h and 48 h of U18666A treatment. On the other hand, genes encoding for solute carrier family members made up the majority of the genes down- regulated by the U18666A treatment.
Ion transport. About half of the 30 genes grouped under ion transport were up- regulated while the other half were down-regulated from 24-72 h of U18666A treatment (Figure 4.4Q). Significant up-regulation or down-regulation was observed in most genes after 72 h of U18666A treatment.
Electron transport. As shown in Figure 4.4R representing 19 genes grouped under electron transport, those encoding for cytochrome P450 family members and
cytochrome c oxidase subunits, except subunit VIIIc (Cox8c), were up-regulated from
24-72 h of U18666A treatment. Alternatively, genes encoding for cytochrome c
oxidase subunit VIIIc (Cox8c) and cytochrome c-1 (Cyc1) showed significant down-
regulation from 24-72 h of U18666A treatment.
Signal transduction. From Figure 4.4S representing 13 genes grouped under signal transduction, more than half of the genes were up-regulated upon U18666A
treatment. The gene encoding for gap junction membrane channel protein α1 (Gja1)
was most significantly up-regulated after 72 h of U18666A treatment. The genes
encoding for ankyrin 1 (Ank1), polymeric immunoglobulin receptor (Pigr) and malate
169 dehydrogenase 1 (Mdh1) were significantly down-regulated from 24-72 h of
U18666A treatment.
G-protein coupled receptor protein signaling pathway. In Figure 4.4T representing 23 genes grouped under G-protein coupled receptor protein signaling pathway, almost equal numbers of genes were up-regulated or down-regulated upon
U18666A treatment. The gene encoding for centaurin γ2 (Centg2) was most significantly down-regulated after 72 h of U18666A treatment. Another gene, encoding for G protein-coupled receptor 12 (Gpr12), was similarly down-regulated.
However, the gene encoding for another receptor subtype, G protein-coupled receptor
19 (Gpr19), was up-regulated instead. The gene encoding for GTP binding protein
(Gem), which is usually over-expressed only in skeletal muscles, was significantly
up-regulated in a time-dependent manner after U18666A treatment.
4.2.4 Validation of differential gene expression through Western blot analysis
Although microarray analysis has been shown to be a powerful tool for generating large amounts of data for gene expression studies, confirmation of the data by other means, including real-time quantitative reverse transcription-polymerase chain reaction and Western blotting, is recommended to overcome technical limitations such as cross-hybridization between closely-related genes, as well as biological
variances. To verify results of the microarray analysis in the present study, a few
genes differentially expressed after U18666A treatment were selected from different
functional categories and their gene products analyzed by Western blotting. The
selected genes were: activating transcription factor 3 (Atf3) from regulation of
170 transcription, α-synuclein (Snca) from the category representing genes with other biological functions, cathepsin B (Ctsb) from proteolysis and peptidolysis, peptidylprolyl isomerase A (Ppia), also known as cyclophilin A, from protein folding,
DNA-damage inducible transcript 3 (Ddit3), also known as GADD 153, from response to stress, and histone 1 H2bc (Hist1h2bc) from DNA processing. Protein levels of the gene products from the Western blot analysis (Figure 4.5A) generally confirmed the mean fold differences of their corresponding genes from the microarray analysis (Figure 4.5B). The protein levels of Atf3 coincided with the mean fold differences of its gene at 3.05 and 2.76 after 48 h and 72 h of U18666A treatment respectively. The protein levels of α-synuclein coincided with the mean fold differences of its gene at 2.44, 2.04 and 1.87 after 24 h, 48 h and 72 h of U18666A treatment respectively. The protein levels of cathepsin B comparatively corresponded with the mean fold differences of its gene at 2.51, 1.90 and 2.24 after 24 h, 48 h and
72 h of U18666A treatment respectively. The protein levels of cyclophilin A coincided with the mean fold differences of the gene for Ppia at 2.43, 2.07 and 1.63 after 24 h, 48 h and 72 h of U18666A treatment respectively. The protein levels of
GADD 153 only matched with the mean fold differences of the gene for Ddit3 at 4.39 and 2.24 after 48 h and 72 h of U18666A treatment respectively. Although a mean fold difference of 1.35 was measured for Ddit3 gene after 24 h of U18666A treatment, no observable GADD 153 protein could be detected. This might be due to the fold change difference at the gene level was too low to be detected at the protein level. It might also indicate that there was still no significant change in the protein level even though there was an increase in the gene expression at the same time-point.
171 Similarly, the protein levels of histone 1 matched with the mean fold differences of its gene at 2.38 and 1.93 after 48 h and 72 h of U18666A treatment respectively. A mean fold difference of 1.26 was measured for the histone 1 gene after 24 h of U18666A treatment, but no significant histone 1 protein was detected at that time-point, probably for the same reasons as proposed above.
172
Figure 4.1: Determination of RNA integrity. Total RNA was isolated from cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A after every 24 h. An aliquot of RNA was subjected to electrophoresis in a 1% (w/v) agarose gel. After staining with ethidium bromide, the RNA bands were visualized using an ultraviolet transilluminator. Arrows: bands representing the 28S and 18S ribosomal RNA. Only one representative replicate out of a total of n = 3 for each sample was depicted.
173
Figure 4.2: Biotin-labeled cRNA before and after fragmentation. Aliquots of cleaned biotin-labeled cRNA before and after fragmentation were collected from each sample representing vehicle (Control) or U18666A (1 μg/ml) treatment for each time-point shown. Unfragmented and fragmented cRNA products were checked by electrophoresis in a 1% (w/v) agarose gel. The RNA bands were visualized using an ultraviolet transilluminator after staining with ethidium bromide. Only one representative replicate out of a total of n = 3 for each sample was depicted.
174 Table 4.1: Differentially-expressed genes in cortical neurons treated with 1 μg/ml U18666A.
GenBank Gene Title Symbol 24 h U18666A 48 h U18666A 72 h U18666A
Genesis and development M95603 achaete-scute complex homolog-like 1 (Drosophila) Ascl1 2.16 ± 0.32 2.26 ± 0.31 3.55 ± 0.36 M64068 B lymphoma Mo-MLV insertion region 1 Bmi1 2.59 ± 0.32 2.17 ± 0.31 1.57 ± 0.30 AL009226 bromodomain containing 2 Brd2 2.15 ± 0.32 1.43 ± 0.31 1.26 ± 0.30 AF071315 COP9 (constitutive photomorphogenic) homolog, subunit Cops6 2.20 ± 0.32 1.76 ± 0.30 1.27 ± 0.30 6 (Arabidopsis thaliana) AF240179 COP9 (constitutive photomorphogenic) homolog, subunit Cops7a 1.85 ± 0.32 2.20 ± 0.31 2.10 ± 0.30 7a (Arabidopsis thaliana) X80903 delta-like 1 (Drosophila) Dll1 1.06 ± 0.32 1.41 ± 0.31 2.43 ± 0.30 M16724 immunoglobulin heavy chain (S107 family) Igh-VS107 -3.06 ± 0.33 -4.25 ± 0.31 -2.36 ± 0.32 V00756 * interferon-related developmental regulator 1 Ifrd1 -1.08 ± 0.33 -1.76 ± 0.31 -2.10 ± 0.30 L31959 intraflagellar transport 88 homolog (Chlamydomonas) Ift88 -1.82 ± 0.33 -2.11 ± 0.31 -1.76 ± 0.30 U69270 * LIM domain binding 1 Ldb1 1.26 ± 0.32 1.76 ± 0.31 2.09 ± 0.30 U89489 LIM domain binding 2 Ldb2 2.63 ± 0.32 1.79 ± 0.31 -1.29 ± 0.30 U28068 neurogenic differentiation 1 Neurod1 -1.02 ± 0.33 -2.05 ± 0.31 -2.00 ± 0.30 X83569 neuronatin Nnat 2.27 ± 0.32 1.76 ± 0.31 1.95 ± 0.30 X74760 Notch gene homolog 3 (Drosophila) Notch3 -2.06 ± 0.32 -2.59 ± 0.31 -2.86 ± 0.31 X63963 paired box gene 6 Pax6 1.68 ± 0.32 1.67 ± 0.31 1.93 ± 0.30 M57683 platelet derived growth factor receptor, alpha polypeptide Pdgfra 1.46 ± 0.33 1.59 ± 0.30 2.33 ± 0.30
D86949 plexin A2 Plxna2 1.86 ± 0.32 2.01 ± 0.30 -1.16 ± 0.30 U44940 * quaking Qk 1.27 ± 0.32 1.43 ± 0.31 2.25 ± 0.30 X85992 sema domain, immunoglobulin domain (Ig), Sema4b 4.58 ± 0.32 4.28 ± 0.33 2.53 ± 0.30 transmembrane domain (TM) and short cytoplasmic domain, (semaphorin) 4B AF032967 spermatogenesis associated 6 Spata6 -2.46 ± 0.33 -1.73 ± 0.32 -1.11 ± 0.30 X94127 * SRY-box containing gene 2 Sox2 1.78 ± 0.32 2.35 ± 0.31 2.80 ± 0.30 X94125 SRY-box containing gene 3 Sox3 1.54 ± 0.34 2.31 ± 0.32 4.06 ± 0.40 AF009414 SRY-box containing gene 11 Sox11 2.28 ± 0.32 1.87 ± 0.31 -1.12 ± 0.30 X73360 transducin-like enhancer of split 3, homolog of Drosophila Tle3 -1.63 ± 0.32 -2.12 ± 0.30 -1.34 ± 0.30 E(spl) X12521 * transition protein 1 Tnp1 2.25 ± 0.32 2.17 ± 0.31 1.04 ± 0.30 M55512 Wilms tumor homolog Wt1 -2.61 ± 0.33 -3.15 ± 0.31 -2.89 ± 0.30
Cell growth and maintenance D89613 cytokine inducible SH2-containing protein Cish 2.13 ± 0.32 1.81 ± 0.31 1.29 ± 0.30 X69619 inhibin beta-A Inhba -2.63 ± 0.33 -3.79 ± 0.31 -4.50 ± 0.30 L12447 insulin-like growth factor binding protein 5 Igfbp5 1.74 ± 0.32 2.04 ± 0.31 3.13 ± 0.30 Y09257 nephroblastoma overexpressed gene Nov 2.61 ± 0.32 2.58 ± 0.31 2.80 ± 0.30 AB033921 * N-myc downstream regulated gene 2 Ndrg2 1.95 ± 0.32 2.49 ± 0.31 3.43 ± 0.30 AF303894 * opioid growth factor receptor Ogfr 2.00 ± 0.32 2.05 ± 0.30 1.89 ± 0.30 AK012457 sarcoma amplified sequence Sas 1.76 ± 0.32 2.10 ± 0.31 1.86 ± 0.30 Z21858 trefoil factor 1 Tff1 -1.70 ± 0.32 -2.07 ± 0.30 -1.77 ± 0.30
Cell adhesion X87096 brevican Bcan 1.62 ± 0.32 1.91 ± 0.31 2.64 ± 0.30 AK168301 bystin-like Bysl 1.45 ± 0.33 -1.38 ± 0.31 -2.40 ± 0.30 X60961 cadherin 1 Cdh1 -2.31 ± 0.33 -2.18 ± 0.31 -2.01 ± 0.30 AB022100 cadherin 13 Cdh13 2.59 ± 0.53 1.80 ± 0.31 -1.45 ± 0.30 L18888 calnexin Canx -2.51 ± 0.32 -3.30 ± 0.31 -1.96 ± 0.30 X59990 catenin alpha 1 Catna1 -1.12 ± 0.33 1.25 ± 0.31 2.16 ± 0.30 Z17804 catenin src Catns -2.24 ± 0.32 -2.25 ± 0.31 -1.72 ± 0.30 L08115 * CD9 antigen Cd9 2.72 ± 0.32 2.07 ± 0.30 2.55 ± 0.30 X76085 CEA-related cell adhesion molecule 2 Ceacam2 -2.46 ± 0.33 -2.84 ± 0.31 -2.40 ± 0.30 X94310 * cell adhesion molecule with homology to L1CAM Chl1 3.58 ± 0.32 3.01 ± 0.31 1.22 ± 0.30 Z29532 follistatin Fst -1.06 ± 0.32 -4.57 ± 0.31 -11.91 ± 0.30 C77966 * laminin B1 subunit 1 Lamb1-1 1.33 ± 0.33 -1.11 ± 0.31 -2.12 ± 0.30 X16834 lectin, galactose binding, soluble 3 Lgals3 1.72 ± 0.35 1.80 ± 0.47 2.79 ± 0.31 X58251 procollagen, type I, alpha 2 Col1a2 -3.17 ± 0.32 -2.83 ± 0.31 -2.19 ± 0.30 L02918 * procollagen, type V, alpha 2 Col5a2 -2.01 ± 0.32 -3.15 ± 0.31 -3.40 ± 0.30 X13986 * secreted phosphoprotein 1 Spp1 1.20 ± 0.32 1.56 ± 0.31 3.99 ± 0.33 M62470 thrombospondin 1 Thbs1 -2.17 ± 0.33 -1.97 ± 0.31 -2.27 ± 0.30 M84487 vascular cell adhesion molecule 1 Vcam1 1.11 ± 0.32 1.55 ± 0.31 2.44 ± 0.30
Cytoskeleton organization AI835883 actin related protein 2/3 complex, subunit 2 Arpc2 2.10 ± 0.32 1.74 ± 0.31 1.39 ± 0.30
175 X13297 actin, alpha 2, smooth muscle, aorta Acta2 -5.18 ± 0.32 -3.89 ± 0.31 -1.65 ± 0.30 J04181 actin, beta, cytoplasmic Actb -4.59 ± 0.32 -5.52 ± 0.30 -4.64 ± 0.30 M21495 * actin, gamma, cytoplasmic 1 Actg1 2.15 ± 0.32 2.30 ± 0.31 1.89 ± 0.30 L40632 ankyrin 3, epithelial Ank3 2.30 ± 0.33 1.62 ± 0.31 1.13 ± 0.30 Z93101 capping protein (actin filament) muscle Z-line, beta Capzb -2.32 ± 0.32 -3.27 ± 0.30 -2.73 ± 0.30 D88793 cysteine and glycine-rich protein 1 Csrp1 1.13 ± 0.32 1.61 ± 0.31 2.87 ± 0.30 D86419 keratin associated protein 6-1 Krtap6-1 -1.29 ± 0.35 -2.52 ± 0.31 -1.58 ± 0.31 X74784 keratin complex 2, basic, gene 17 Krt2-17 -2.58 ± 0.32 -1.87 ± 0.32 -1.31 ± 0.30 D12645 * kinesin family member 3A Kif3a 1.85 ± 0.32 2.19 ± 0.31 1.94 ± 0.30 AW122699 kinesin family member 16B Kif16b 1.88 ± 0.37 2.03 ± 0.38 1.58 ± 0.33 U92949 kinesin family member C2 Kifc2 2.73 ± 0.40 2.62 ± 0.39 2.77 ± 0.33 M18775 * microtubule-associated protein tau Mapt 2.66 ± 0.32 2.17 ± 0.31 1.03 ± 0.30 M86390 moesin Msn -1.82 ± 0.32 -2.10 ± 0.31 -1.19 ± 0.31 L00923 myosin IB Myo1b -1.04 ± 0.32 -1.87 ± 0.31 -2.42 ± 0.30 M13016 neurofilament, light polypeptide Nefl -1.02 ± 0.33 -1.41 ± 0.31 -2.02 ± 0.31 L20255 stathmin 1 Stmn1 2.28 ± 0.32 1.88 ± 0.30 1.61 ± 0.30 U38967 thymosin, beta 4, X chromosome Tmsb4x 2.27 ± 0.32 2.02 ± 0.30 1.73 ± 0.30 M28729 * tubulin, alpha 1 Tuba1 2.75 ± 0.32 2.26 ± 0.30 1.75 ± 0.30 M28727 tubulin, alpha 2 Tuba2 2.69 ± 0.32 2.00 ± 0.31 1.59 ± 0.30 M13441 * tubulin, alpha 6 Tuba6 3.06 ± 0.32 2.55 ± 0.31 1.99 ± 0.30 M28739 tubulin, beta 2 Tubb2 2.22 ± 0.32 1.74 ± 0.31 1.29 ± 0.30 X04663 tubulin, beta 5 Tubb5 2.06 ± 0.32 1.54 ± 0.30 1.25 ± 0.30 Y07738 vimentin Vim -4.74 ± 0.34 -3.47 ± 0.31 -2.90 ± 0.30
Cell cycle AK013213 CDC16 cell division cycle 16 homolog (S. cerevisiae) Cdc16 2.01 ± 0.32 -1.11 ± 0.31 -1.45 ± 0.30 AK052473 checkpoint with forkhead and ring finger domains Chfr -1.12 ± 0.32 -1.79 ± 0.31 -2.11 ± 0.30 M83749 cyclin D2 Ccnd2 1.73 ± 0.32 2.93 ± 0.48 -1.04 ± 0.30 L06864 epidermal growth factor receptor Egfr -1.31 ± 0.33 1.44 ± 0.31 2.72 ± 0.30 AF076681 eukaryotic translation initiation factor 2 alpha kinase 3 Eif2ak3 1.43 ± 0.33 2.81 ± 0.31 1.98 ± 0.30 V00727 * FBJ osteosarcoma oncogene Fos -1.03 ± 0.33 4.59 ± 0.31 5.86 ± 0.35 U00937 growth arrest and DNA-damage-inducible 45 alpha Gadd45a 1.51 ± 0.33 2.26 ± 0.32 2.55 ± 0.30 AF055638 * growth arrest and DNA-damage-inducible 45 gamma Gadd45g 1.00 ± 0.32 2.51 ± 0.31 3.51 ± 0.30 AK083453 growth arrest specific 5 Gas5 1.60 ± 0.32 1.71 ± 0.31 2.16 ± 0.30 J04509 Jun proto-oncogene related gene d1 Jund1 1.52 ± 0.32 2.19 ± 0.31 1.80 ± 0.30 X67735 MAS1 oncogene Mas1 -2.35 ± 0.33 -2.30 ± 0.31 -1.55 ± 0.30 AF070970 par-6 (partitioning defective 6,) homolog alpha (C. Pard6a 1.74 ± 0.32 2.01 ± 0.30 1.53 ± 0.30 elegans) AI844711 * polymerase (RNA) III (DNA directed) polypeptide D Polr3d 2.17 ± 0.33 2.16 ± 0.32 1.69 ± 0.30 AB051827 * ras homolog gene family member U Rhou -1.44 ± 0.32 -2.83 ± 0.31 -4.09 ± 0.30 Z75332 stromal antigen 1 Stag1 1.11 ± 0.32 1.53 ± 0.31 2.21 ± 0.30 M92420 transforming growth factor alpha Tgfa -1.05 ± 0.32 -1.05 ± 0.31 2.00 ± 0.30
Cell death AA797604 angiopoietin-like 4 Angptl4 -1.05 ± 0.34 1.00 ± 0.31 2.01 ± 0.30 AK011802 BCL2-associated transcription factor 1 Bclaf1 -1.06 ± 0.34 2.01 ± 0.33 3.17 ± 0.31 AA275665 BCL2-like 11 (apoptosis facilitator) Bcl2l11 1.18 ± 0.32 1.97 ± 0.31 1.78 ± 0.30 X55573 brain derived neurotrophic factor Bdnf 1.27 ± 0.33 2.40 ± 0.31 1.73 ± 0.31 D14077 * clusterin Clu 1.45 ± 0.32 1.84 ± 0.31 3.50 ± 0.30 AF085348 DAZ associated protein 2 Dazap2 1.67 ± 0.32 1.83 ± 0.31 2.74 ± 0.30 D50494 DEAD (Asp-Glu-Ala-Asp) box polypeptide 6 Ddx6 2.08 ± 0.32 1.94 ± 0.31 1.36 ± 0.30 M31314 Fc receptor, IgG, high affinity I Fcgr1 2.10 ± 0.34 1.64 ± 0.32 1.29 ± 0.30 M21828 growth arrest specific 2 Gas2 1.68 ± 0.46 2.12 ± 0.31 2.90 ± 0.31 X04480 * insulin-like growth factor 1 Igf1 -2.45 ± 0.32 -2.81 ± 0.31 -2.04 ± 0.30 AF263364 SH3-domain GRB2-like B1 (endophilin) Sh3glb1 2.26 ± 0.32 2.28 ± 0.31 1.54 ± 0.30 AK017719 transformed mouse 3T3 cell double minute 4 Mdm4 -2.05 ± 0.33 -2.01 ± 0.31 -2.02 ± 0.30
Defense response Y16258 basigin Bsg 2.22 ± 0.32 1.72 ± 0.30 1.54 ± 0.30 X01838 beta-2 microglobulin B2m -1.23 ± 0.32 1.33 ± 0.31 2.55 ± 0.30 Z80112 chemokine (C-X-C motif) receptor 4 Cxcr4 -1.27 ± 0.33 1.32 ± 0.31 2.16 ± 0.30 X97720 Friend virus susceptibility 1 Fv1 1.55 ± 0.32 2.20 ± 0.31 1.18 ± 0.30 U21906 histocompatibility 2, blastocyst H2-Bl 1.39 ± 0.34 2.18 ± 0.32 2.31 ± 0.30 X52490 histocompatibility 2, D region locus 1 H2-D1 2.13 ± 0.32 2.10 ± 0.31 1.68 ± 0.30 U47329 histocompatibility 2, K1, K region H2-K1 2.08 ± 0.33 2.11 ± 0.31 1.70 ± 0.30 AF016308 histocompatibility 2, M region locus 9 H2-M9 -3.58 ± 0.32 -4.00 ± 0.33 -4.83 ± 0.30 X57330 histocompatibility 2, Q region locus 7 H2-Q7 1.85 ± 0.32 2.33 ± 0.31 2.14 ± 0.30 AB024538 immunoglobulin superfamily containing leucine-rich Islr -1.18 ± 0.33 1.21 ± 0.31 2.06 ± 0.30 repeat AI644072 immunoglobulin superfamily, member 10 Igsf10 -1.56 ± 0.32 -2.32 ± 0.31 -2.07 ± 0.30
176 U43085 interferon-induced protein with tetratricopeptide repeats 2 Ifit2 1.57 ± 0.35 1.65 ± 0.34 2.24 ± 0.31
AF053005 interleukin 27 receptor, alpha Il27ra -1.63 ± 0.33 -2.17 ± 0.31 -1.49 ± 0.30 D86232 * lymphocyte antigen 6 complex, locus C Ly6c -2.35 ± 0.32 -2.27 ± 0.31 -2.12 ± 0.30 M38337 milk fat globule-EGF factor 8 protein Mfge8 1.70 ± 0.32 1.82 ± 0.31 2.41 ± 0.30 AF043943 TAP binding protein Tapbp -2.59 ± 0.32 -2.53 ± 0.31 -1.59 ± 0.30
Response to stress X67083 * DNA-damage inducible transcript 3 Ddit3 1.35 ± 0.33 4.39 ± 0.31 2.24 ± 0.30 AA874130 DnaJ (Hsp40) homolog, subfamily B, member 6 Dnajb6 1.67 ± 0.33 2.08 ± 0.31 1.90 ± 0.30 D87896 glutathione peroxidase 4 Gpx4 1.91 ± 0.32 2.12 ± 0.31 1.66 ± 0.30 L06047 glutathione S-transferase, alpha 4 Gsta4 1.11 ± 0.33 1.45 ± 0.31 2.29 ± 0.30 U24428 glutathione S-transferase, mu 5 Gstm5 1.94 ± 0.33 2.29 ± 0.35 1.89 ± 0.31 U50631 heat-responsive protein 12 Hrsp12 1.33 ± 0.33 1.22 ± 0.31 2.38 ± 0.30 M20567 * heat shock protein 2 Hspa2 2.26 ± 0.34 2.47 ± 0.32 3.39 ± 0.31 M59821 immediate early response 2 Ier2 1.68 ± 0.32 4.21 ± 0.31 2.71 ± 0.30 V00835 metallothionein 1 Mt1 1.49 ± 0.32 1.57 ± 0.30 2.65 ± 0.30 AF159050 microsomal glutathione S-transferase 1 Mgst1 1.49 ± 0.33 1.78 ± 0.31 2.88 ± 0.32 AF093853 * peroxiredoxin 6 Prdx6 1.27 ± 0.32 1.76 ± 0.31 2.67 ± 0.30 AK013663 protein-kinase, interferon-inducible double stranded RNA Prkrir 2.54 ± 0.32 1.95 ± 0.31 1.37 ± 0.30 dependent inhibitor, repressor of (P58 repressor) AI839138 * thioredoxin interacting protein Txnip -1.02 ± 0.33 2.16 ± 0.30 2.49 ± 0.30
Cholesterol biosynthesis AI256750 farnesyl diphosphate synthetase Fdps 2.65 ± 0.32 1.68 ± 0.31 -1.36 ± 0.30 U49878 3-hydroxy-3-methylglutaryl-Coenzyme A lyase Hmgcl 1.86 ± 0.32 2.61 ± 0.31 1.66 ± 0.30 M62766 * 3-hydroxy-3-methylglutaryl-Coenzyme A reductase Hmgcr -1.30 ± 0.32 -1.80 ± 0.31 -2.39 ± 0.30 AA716963 isopentenyl-diphosphate delta isomerase Idi1 2.18 ± 0.32 1.57 ± 0.30 -1.28 ± 0.30 AI663199 lanosterol synthase Lss 1.40 ± 0.33 -1.78 ± 0.31 -2.10 ± 0.30 Z19521 low density lipoprotein receptor Ldlr -1.03 ± 0.32 -1.08 ± 0.31 -2.42 ± 0.30 AW259788 * low density lipoprotein receptor-related protein 2 Lrp2 -2.92 ± 0.33 -2.61 ± 0.31 -2.70 ± 0.31 AF137598 mevalonate kinase Mvk 3.21 ± 0.32 1.68 ± 0.31 -1.48 ± 0.30 AI747449 * NAD(P) dependent steroid dehydrogenase-like Nsdhl 2.50 ± 0.33 1.82 ± 0.31 -1.29 ± 0.30 AB016248 * sterol-C5-desaturase (fungal ERG3, delta-5-desaturase) Sc5d 2.12 ± 0.32 2.40 ± 0.31 -1.21 ± 0.31 homolog (S. cerevisae) AA408956 very low density lipoprotein receptor Vldlr 1.69 ± 0.34 2.36 ± 0.32 -1.36 ± 0.33
Lipid metabolism and transport U21489 acetyl-Coenzyme A dehydrogenase, long-chain Acadl 1.16 ± 0.34 1.80 ± 0.35 4.14 ± 0.30 AW124201 * 1-acylglycerol-3-phosphate O-acyltransferase 3 Agpat3 2.25 ± 0.32 1.59 ± 0.31 2.20 ± 0.30 D00466 * apolipoprotein E Apoe 1.64 ± 0.32 2.02 ± 0.30 3.32 ± 0.30 AF017175 * carnitine palmitoyltransferase 1a, liver Cpt1a 1.03 ± 0.32 1.28 ± 0.31 2.75 ± 0.30 Z14050 dodecenoyl-Coenzyme A delta isomerase (3,2 trans-enoyl- Dci 1.75 ± 0.32 1.66 ± 0.31 2.38 ± 0.30 Coenyme A isomerase) AI317360 elongation of very long chain fatty acids (FEN1/Elo2, Elovl2 -1.86 ± 0.33 -2.24 ± 0.31 -1.81 ± 0.30 SUR4/Elo3, yeast)-like 2 X13135 fatty acid synthase Fasn 2.17 ± 0.32 1.92 ± 0.31 1.43 ± 0.30 U09816 GM2 ganglioside activator protein Gm2a 1.11 ± 0.32 1.52 ± 0.31 2.13 ± 0.35 AF153613 peroxisomal delta3, delta2-enoyl-Coenzyme A isomerase Peci 1.58 ± 0.32 1.65 ± 0.30 2.55 ± 0.30
AV025606 phosphatidic acid phosphatase type 2B Ppap2b 1.04 ± 0.32 1.38 ± 0.30 2.41 ± 0.30 U34277 * phospholipase A2, group VII (platelet-activating factor Pla2g7 1.26 ± 0.32 2.25 ± 0.31 4.58 ± 0.30 acetylhydrolase, plasma) AK003183 * phospholipase A2, group XIIA Pla2g12a 1.33 ± 0.32 2.31 ± 0.31 1.52 ± 0.30 U85714 phospholipase C, beta 1 Plcb1 -2.29 ± 0.32 -2.11 ± 0.31 -1.59 ± 0.30 M21285 * stearoyl-Coenzyme A desaturase 1 Scd1 1.21 ± 0.32 -1.29 ± 0.31 -2.38 ± 0.30 M91458 * sterol carrier protein 2, liver Scp2 1.23 ± 0.33 1.73 ± 0.36 2.94 ± 0.39
Protein amino acid modification Y17343 acid phosphatase 1, soluble Acp1 1.92 ± 0.32 1.56 ± 0.31 2.18 ± 0.30 AB020202 adenylate kinase 2 Ak2 2.19 ± 0.33 2.21 ± 0.38 1.62 ± 0.30 D83002 anaplastic lymphoma kinase Alk -1.39 ± 0.33 -2.02 ± 0.32 -1.99 ± 0.30 AF033566 CDC like kinase 4 Clk4 -1.45 ± 0.32 -2.13 ± 0.31 -1.24 ± 0.30 AK005062 dual specificity phosphatase 6 Dusp6 -1.47 ± 0.32 -2.36 ± 0.30 -2.27 ± 0.30 M68513 Eph receptor A3 Epha3 2.02 ± 0.33 1.93 ± 0.31 -1.23 ± 0.30 L25890 Eph receptor B2 Ephb2 3.08 ± 0.45 2.56 ± 0.31 1.45 ± 0.31 M90470 leukocyte tyrosine kinase Ltk -2.50 ± 0.32 -2.05 ± 0.31 -1.67 ± 0.31 M32017 lysosomal membrane glycoprotein 2 Lamp2 1.54 ± 0.32 1.97 ± 0.31 2.93 ± 0.40 M73491 mannoside acetylglucosaminyltransferase 1 Mgat1 -3.15 ± 0.33 -3.37 ± 0.31 -2.52 ± 0.31 X76850 MAP kinase-activated protein kinase 2 Mapkapk2 1.31 ± 0.34 2.61 ± 0.33 3.63 ± 0.31 Z14249 mitogen activated protein kinase 3 Mapk3 2.39 ± 0.49 3.21 ± 0.31 4.08 ± 0.73
177 L35236 mitogen activated protein kinase 10 Mapk10 1.04 ± 0.32 -1.22 ± 0.31 -2.02 ± 0.30 AF155142 mitogen activated protein kinase kinase kinase 11 Map3k11 -2.25 ± 0.33 -2.24 ± 0.32 -2.27 ± 0.31 U16162 procollagen-proline, 2-oxoglutarate 4-dioxygenase P4ha1 2.23 ± 0.34 1.27 ± 0.31 1.22 ± 0.31 (proline 4-hydroxylase), alpha 1 polypeptide AB011812 protein kinase C, delta Prkcd -2.03 ± 0.32 -2.50 ± 0.31 -1.70 ± 0.30 AB062122 protein kinase C, theta Prkcq 1.17 ± 0.37 1.90 ± 0.33 2.62 ± 0.31 M12303 protein kinase, cAMP dependent, catalytic, alpha Prkaca -2.37 ± 0.32 -3.20 ± 0.31 -2.54 ± 0.30 AF067129 protein phosphatase 1, regulatory (inhibitor) subunit 7 Ppp1r7 2.21 ± 0.32 1.49 ± 0.30 -1.01 ± 0.30 AF088911 protein phosphatase 4, catalytic subunit Ppp4c 2.05 ± 0.32 2.18 ± 0.31 2.43 ± 0.30 AI844911 protein tyrosine phosphatase, receptor type D Ptprd 1.06 ± 0.33 -1.26 ± 0.30 -2.93 ± 0.30 D83484 protein tyrosine phosphatase, receptor type E Ptpre -1.36 ± 0.32 -2.15 ± 0.31 -1.51 ± 0.30 D83203 * protein tyrosine phosphatase, receptor type J Ptprj -1.97 ± 0.32 -2.05 ± 0.31 -2.30 ± 0.30 L10106 * protein tyrosine phosphatase, receptor type K Ptprk 1.28 ± 0.32 -1.41 ± 0.30 -2.05 ± 0.30 AJ133130 protein tyrosine phosphatase, receptor type Z, polypeptide Ptprz1 1.22 ± 0.32 2.38 ± 0.31 2.79 ± 0.61 1 AK029923 ribophorin I Rpn1 -1.16 ± 0.32 -1.62 ± 0.30 -2.27 ± 0.30 AJ007938 ribosomal protein S6 kinase, polypeptide 2 Rps6kb2 2.11 ± 0.33 1.71 ± 0.31 1.66 ± 0.30 AA691445 serine/threonine kinase 38 Stk38 -2.06 ± 0.32 -2.73 ± 0.30 -1.87 ± 0.30 X86000 sialyltransferase 8 (alpha-2, 8-sialyltransferase) D Siat8d -1.36 ± 0.32 -1.35 ± 0.31 -2.04 ± 0.31
Protein biosynthesis X15267 acidic ribosomal phosphoprotein P0 Arbp 2.41 ± 0.32 2.14 ± 0.31 1.75 ± 0.30 M31690 argininosuccinate synthetase 1 Ass1 -1.48 ± 0.33 -3.33 ± 0.31 -1.40 ± 0.30 M22432 eukaryotic translation elongation factor 1 alpha 1 Eef1a1 2.26 ± 0.32 2.01 ± 0.31 1.85 ± 0.30 AK045250 * eukaryotic translation initiation factor 4B Eif4b 1.48 ± 0.32 2.18 ± 0.31 2.82 ± 0.30 U87965 * GTP binding protein 1 Gtpbp1 -3.11 ± 0.32 -2.74 ± 0.30 -2.54 ± 0.30 AK003886 mitochondrial ribosomal protein L3 Mrpl3 -1.33 ± 0.32 -2.12 ± 0.31 -2.51 ± 0.30 AF185591 mitochondrial ribosomal protein L20 Mrpl20 2.02 ± 0.32 1.41 ± 0.30 -1.14 ± 0.30 AK083366 mitochondrial ribosomal protein S5 Mrps5 1.68 ± 0.33 2.10 ± 0.32 2.07 ± 0.30 Y11682 mitochondrial ribosomal protein S12 Mrps12 1.86 ± 0.32 2.08 ± 0.31 1.69 ± 0.30 AI875598 mitochondrial translational initiation factor 2 Mtif2 2.03 ± 0.32 1.98 ± 0.31 1.76 ± 0.30 Y00225 ribosomal protein L3 Rpl3 2.24 ± 0.32 1.86 ± 0.31 1.64 ± 0.30 X80699 ribosomal protein L26 Rpl26 2.14 ± 0.32 1.96 ± 0.30 1.71 ± 0.30 U58105 * ribosomal protein L36a Rpl36a 1.68 ± 0.32 2.08 ± 0.30 1.74 ± 0.30 U93862 ribosomal protein L41 Rpl41 2.47 ± 0.32 2.10 ± 0.30 1.90 ± 0.30 U29402 ribosomal protein, large, P1 Rplp1 2.06 ± 0.32 1.77 ± 0.30 1.63 ± 0.30 Z83368 ribosomal protein S3a Rps3a 2.16 ± 0.32 1.91 ± 0.30 1.73 ± 0.30 AF043285 ribosomal protein S7 Rps7 2.16 ± 0.32 1.87 ± 0.30 1.55 ± 0.30 M76762 ribosomal protein S18 Rps18 2.04 ± 0.32 2.06 ± 0.30 1.84 ± 0.30 U67770 ribosomal protein S26 Rps26 2.07 ± 0.32 1.98 ± 0.30 1.68 ± 0.30 AK014294 ribosomal protein S27 Rps27 1.98 ± 0.32 2.22 ± 0.31 1.95 ± 0.30
Protein folding AA408703 aryl-hydrocarbon receptor-interacting protein Aip 1.76 ± 0.33 1.99 ± 0.31 1.52 ± 0.30 AL024092 chaperonin subunit 3 (gamma) Cct3 -2.47 ± 0.32 -1.92 ± 0.31 -2.26 ± 0.30 AA798624 * ERO1-like (S. cerevisiae) Ero1l -2.53 ± 0.32 -2.31 ± 0.31 -1.82 ± 0.30 AA589416 peptidylprolyl isomerase (cyclophilin)-like 2 Ppil2 1.92 ± 0.32 2.74 ± 0.31 2.15 ± 0.30 X52803 * peptidylprolyl isomerase A Ppia 2.43 ± 0.32 2.07 ± 0.31 1.63 ± 0.30 X58990 peptidylprolyl isomerase B Ppib 1.99 ± 0.32 1.85 ± 0.30 1.50 ± 0.30 AB009692 protein (peptidyl-prolyl cis/trans isomerase) NIMA- Pin1 2.00 ± 0.32 2.22 ± 0.31 1.47 ± 0.30 interacting 1
Proteolysis and peptidolysis AF006196 a disintegrin and metalloproteinase domain 15 Adam15 2.57 ± 0.33 1.62 ± 0.31 1.51 ± 0.30 (metargidin) AA726223 a disintegrin and metalloproteinase domain 19 (meltrin Adam19 -1.15 ± 0.33 -1.85 ± 0.31 -2.21 ± 0.30 beta) D38117 calpain 2 Capn2 2.30 ± 0.32 2.02 ± 0.31 2.07 ± 0.30 AF091998 calpain 3 Capn3 -1.49 ± 0.32 -1.63 ± 0.31 -2.10 ± 0.30 AF077738 * carboxypeptidase X 1 (M14 family) Cpxm1 2.15 ± 0.34 1.56 ± 0.31 2.15 ± 0.31 X54966 cathepsin B Ctsb 2.51 ± 0.34 1.90 ± 0.31 2.24 ± 0.31 AF179369 protease, serine, 11 (Igf binding) Prss11 1.27 ± 0.32 1.13 ± 0.33 2.17 ± 0.30 D89871 protease, serine, 12 neurotrypsin (motopsin) Prss12 -1.13 ± 0.32 -1.85 ± 0.31 -2.54 ± 0.30 AF164513 protease, serine, 25 Prss25 -1.85 ± 0.32 -2.24 ± 0.30 -1.96 ± 0.30 AF037206 * ring finger protein 13 Rnf13 1.88 ± 0.33 2.13 ± 0.31 2.47 ± 0.30 X71642 ring finger protein (C3HC4 type) 19 Rnf19 2.00 ± 0.33 1.80 ± 0.32 1.30 ± 0.30 AK048862 ring finger protein 20 Rnf20 -4.64 ± 0.33 -2.56 ± 0.31 -1.82 ± 0.30 AF434815 * ring finger protein 34 Rnf34 2.03 ± 0.32 1.41 ± 0.31 1.18 ± 0.30 X81323 tripeptidyl peptidase II Tpp2 2.31 ± 0.33 1.84 ± 0.32 1.73 ± 0.31
178 Ubiquitin-proteasome system AJ130975 ariadne homolog 2 (Drosophila) Arih2 2.08 ± 0.33 2.32 ± 0.31 2.07 ± 0.30 AF176527 F-box only protein 8 Fbxo8 1.31 ± 0.33 2.55 ± 0.35 3.12 ± 0.53 AW124512 F-box only protein 37 Fbxo37 2.38 ± 0.33 1.30 ± 0.31 -1.17 ± 0.30 U39302 protease (prosome, macropain) 26S subunit, ATPase 1 Psmc1 2.10 ± 0.32 1.67 ± 0.30 1.16 ± 0.30 D49686 proteasome (prosome, macropain) 26S subunit, ATPase 3 Psmc3 2.17 ± 0.32 1.82 ± 0.30 1.35 ± 0.30
AA407360 proteasome (prosome, macropain) 26S subunit, non- Psmd8 2.18 ± 0.32 1.81 ± 0.30 1.36 ± 0.30 ATPase, 8 AB022022 proteasome (prosome, macropain) 26S subunit, non- Psmd10 1.90 ± 0.33 2.36 ± 0.31 1.56 ± 0.30 ATPase, 10 AF055983 proteasome (prosome, macropain) subunit alpha type 3 Psma3 2.16 ± 0.32 1.88 ± 0.30 1.44 ± 0.30
AK165563 proteasome (prosome, macropain) subunit alpha type 7 Psma7 2.01 ± 0.32 1.55 ± 0.30 1.13 ± 0.30
AK002331 proteasome (prosome, macropain) subunit beta type 3 Psmb3 2.41 ± 0.32 1.67 ± 0.31 1.04 ± 0.30 X51703 * ubiquitin B Ubb 2.80 ± 0.32 2.34 ± 0.31 1.90 ± 0.30 D50527 ubiquitin C Ubc 2.80 ± 0.32 2.50 ± 0.30 2.13 ± 0.30 X92664 ubiquitin-conjugating enzyme E2E 3, UBC4/5 homolog Ube2e3 -1.15 ± 0.32 -1.85 ± 0.30 -2.04 ± 0.30 (yeast) U82122 ubiquitin protein ligase E3A Ube3a 2.58 ± 0.32 2.38 ± 0.31 1.31 ± 0.30 AK033182 ubiquitin specific protease 34 Usp34 -1.26 ± 0.32 -1.73 ± 0.30 -2.37 ± 0.30
Regulation of transcription U19118 * activating transcription factor 3 Atf3 -1.12 ± 0.33 3.05 ± 0.31 2.76 ± 0.33 AW120832 AE binding protein 2 Aebp2 -1.54 ± 0.32 -2.72 ± 0.31 -3.26 ± 0.32 AB046714 * brain abundant, membrane attached signal protein 1 Basp1 2.63 ± 0.32 2.15 ± 0.31 1.77 ± 0.30 U19892 CCAAT/enhancer binding protein zeta Cebpz 2.21 ± 0.34 2.37 ± 0.36 2.11 ± 0.32 X63866 cellular nucleic acid binding protein 1 Cnbp1 2.63 ± 0.32 2.09 ± 0.31 1.90 ± 0.30 AF084524 cellular repressor of E1A-stimulated genes Creg 2.07 ± 0.33 1.61 ± 0.41 2.08 ± 0.30 U70736 * COP9 (constitutive photomorphogenic) homolog, subunit Cops5 2.06 ± 0.32 2.03 ± 0.30 1.67 ± 0.30 5 (Arabidopsis thaliana) AI461938 forkhead box P1 Foxp1 2.83 ± 0.32 2.23 ± 0.31 -1.11 ± 0.30 AI047735 homeo box D8 Hoxd8 -1.50 ± 0.33 -2.33 ± 0.31 -2.64 ± 0.31 D49658 LIM homeobox protein 8 Lhx8 -1.33 ± 0.33 -1.49 ± 0.32 -4.76 ± 0.32 L38822 Max interacting protein 1 Mxi1 1.50 ± 0.32 1.61 ± 0.31 2.21 ± 0.30 AJ243608 melanoma antigen, family L, 2 Magel2 2.29 ± 0.34 1.22 ± 0.31 -1.29 ± 0.30 C76717 metal response element binding transcription factor 2 Mtf2 -1.69 ± 0.32 -2.38 ± 0.30 -1.49 ± 0.30 U86338 myelin transcription factor 1-like Myt1l 1.23 ± 0.32 1.01 ± 0.31 -2.09 ± 0.30 L13171 * myocyte enhancer factor 2C Mef2c 2.14 ± 0.32 -1.05 ± 0.30 -3.19 ± 0.30 D90173 * nuclear factor I/A Nfia 2.24 ± 0.32 4.60 ± 0.38 3.94 ± 0.30 Y07688 nuclear factor I/X Nfix 2.33 ± 0.32 2.38 ± 0.31 1.41 ± 0.30 U36576 nuclear factor of activated T-cells, cytoplasmic, Nfatc2 -1.37 ± 0.33 -2.33 ± 0.32 -2.29 ± 0.31 calcineurin-dependent 2 U70475 nuclear factor, erythroid derived 2, like 2 Nfe2l2 -1.56 ± 0.33 -1.12 ± 0.31 2.02 ± 0.30 AF053062 nuclear receptor interacting protein 1 Nrip1 1.43 ± 0.32 1.09 ± 0.31 -2.20 ± 0.30 U09504 nuclear receptor subfamily 1, group D, member 2 Nr1d2 1.41 ± 0.34 1.53 ± 0.31 2.25 ± 0.30 U09419 nuclear receptor subfamily 1, group H, member 2 Nr1h2 1.62 ± 0.33 1.75 ± 0.31 2.11 ± 0.30 U33196 * nuclease sensitive element binding protein 1 Nsep1 2.99 ± 0.32 2.54 ± 0.31 1.98 ± 0.30 AI845735 polymerase (RNA) II (DNA directed) polypeptide E Polr2e 1.82 ± 0.32 1.84 ± 0.30 2.01 ± 0.30 AA880275 polymerase (RNA) II (DNA directed) polypeptide K Polr2k 5.80 ± 0.32 5.38 ± 0.31 4.62 ± 0.30 M88299 POU domain, class 3, transcription factor 3 Pou3f3 2.41 ± 0.33 3.26 ± 0.41 3.72 ± 0.30 U11248 ribosomal protein S28 Rps28 2.33 ± 0.32 2.22 ± 0.31 1.71 ± 0.30 D31966 RNA polymerase 1-1 Rpo1-1 -4.25 ± 0.33 -3.85 ± 0.32 -2.69 ± 0.30 U58280 RNA polymerase 1-2 Rpo1-2 -1.42 ± 0.32 -1.83 ± 0.31 -2.20 ± 0.30 D86609 * RNA polymerase 1-3 Rpo1-3 1.62 ± 0.32 1.80 ± 0.31 2.05 ± 0.30 U40930 sequestosome 1 Sqstm1 2.34 ± 0.32 2.17 ± 0.31 1.86 ± 0.30 X67863 * simple repeat sequence-containing transcript Srst 3.77 ± 0.33 2.85 ± 0.31 2.20 ± 0.31 C79000 SKI interacting protein Skiip 3.77 ± 0.32 5.48 ± 0.50 3.85 ± 0.30 AA673500 TAF9 RNA polymerase II, TATA box binding protein Taf9 2.01 ± 0.32 1.69 ± 0.31 1.34 ± 0.30 (TBP)-associated factor AK034044 thyroid hormone receptor associated protein 3 Thrap3 2.10 ± 0.32 2.19 ± 0.31 2.50 ± 0.31 X61385 transcription factor 7, T-cell specific Tcf7 -2.83 ± 0.35 1.27 ± 0.34 -1.40 ± 0.32 D43643 * transcription factor-like 1 Tcfl1 1.61 ± 0.32 2.23 ± 0.31 1.82 ± 0.30 AI845438 zinc finger DHHC domain containing 6 Zdhhc6 2.34 ± 0.32 1.80 ± 0.31 1.54 ± 0.30 D76432 zinc finger homeobox 1a Zfhx1a 1.32 ± 0.32 1.34 ± 0.30 2.09 ± 0.30 M36146 zinc finger protein 35 Zfp35 2.75 ± 0.33 1.40 ± 0.31 -1.08 ± 0.30 M58566 * zinc finger protein 36, C3H type-like 1 Zfp36l1 1.10 ± 0.32 1.78 ± 0.31 2.87 ± 0.30 L28167 zinc finger protein 61 Zfp61 1.65 ± 0.32 2.14 ± 0.31 1.59 ± 0.30 AJ005350 zinc finger protein 125 Zfp125 -1.63 ± 0.32 -1.58 ± 0.31 -2.46 ± 0.30 AW048709 zinc finger protein (C2H2 type) 276 Zfp276 1.08 ± 0.32 2.01 ± 0.31 1.53 ± 0.30
179 AK077401 zinc finger protein 277 Zfp277 2.15 ± 0.35 1.90 ± 0.31 1.48 ± 0.30 AF194030 zinc finger protein 288 Zfp288 1.82 ± 0.32 2.13 ± 0.31 1.33 ± 0.30 AB012725 zinc finger protein 326 Zfp326 -1.77 ± 0.32 -2.09 ± 0.31 -1.47 ± 0.30 AF397208 * zinc finger protein 358 Zfp358 1.93 ± 0.32 2.55 ± 0.31 2.53 ± 0.30 AF118566 zinc finger protein 385 Zfp385 -2.12 ± 0.33 -2.12 ± 0.31 -1.47 ± 0.31 AK082970 * zinc finger protein 422 Zfp422 1.41 ± 0.32 2.00 ± 0.31 2.47 ± 0.30
DNA processing X55038 centromere autoantigen B Cenpb 2.26 ± 0.32 1.95 ± 0.31 1.84 ± 0.30 X56690 chromobox homolog 1 (Drosophila HP1 beta) Cbx1 3.73 ± 0.32 4.17 ± 0.61 2.03 ± 0.30 AK005364 * CXXC finger 5 Cxxc5 1.27 ± 0.32 2.12 ± 0.31 2.69 ± 0.30 AF128879 * ectonucleotide pyrophosphatase/phosphodiesterase 2 Enpp2 1.48 ± 0.33 2.42 ± 0.31 4.06 ± 0.30
U97572 excision repair cross-complementing rodent repair Ercc2 -2.10 ± 0.32 -2.14 ± 0.30 -1.70 ± 0.30 deficiency, complementation group 2 AJ002366 general transcription factor II H, polypeptide 1 Gtf2h1 1.97 ± 0.32 1.98 ± 0.30 1.56 ± 0.30 X67668 * high mobility group box 2 Hmgb2 1.61 ± 0.32 3.23 ± 0.31 3.70 ± 0.54 AK011516 histone 1, H2bc Hist1h2bc 1.26 ± 0.32 2.38 ± 0.31 1.93 ± 0.30 X16148 histone 2, H2aa1 Hist2h2aa1 2.11 ± 0.32 2.23 ± 0.31 1.33 ± 0.30 M32459 * histone 2, H3c2 Hist2h3c2 1.50 ± 0.33 3.06 ± 0.31 2.69 ± 0.30 AW210014 HIV-1 Rev binding protein 2 Hrb2 2.27 ± 0.34 1.49 ± 0.31 1.89 ± 0.31 AY183136 N-acetyltransferase ARD1 homolog (S. cerevisiae) Ard1 2.39 ± 0.32 1.51 ± 0.31 1.28 ± 0.30 U64450 nucleoplasmin 3 Npm3 -1.71 ± 0.32 -3.28 ± 0.31 -1.15 ± 0.30 AF038939 paternally expressed 3 Peg3 1.81 ± 0.32 3.07 ± 0.41 1.04 ± 0.30 M18070 prion protein Prnp 2.01 ± 0.32 1.58 ± 0.31 1.25 ± 0.31 X92410 RAD23a homolog (S. cerevisiae) Rad23a 2.10 ± 0.33 2.56 ± 0.31 1.58 ± 0.31 AK154336 * single stranded DNA binding protein 4 Ssbp4 2.08 ± 0.32 1.86 ± 0.31 2.25 ± 0.30 AA982124 SWI/SNF related, matrix associated, actin dependent Smarca5 1.68 ± 0.35 1.75 ± 0.31 2.16 ± 0.30 regulator of chromatin, subfamily a, member 5 X69942 SWI/SNF-related, matrix-associated, actin-dependent Smarcad1 -1.80 ± 0.33 -2.63 ± 0.31 -1.76 ± 0.30 regulator of chromatin, subfamily a, containing DEAD/H box 1 U85614 SWI/SNF related, matrix associated, actin dependent Smarcc1 1.99 ± 0.32 1.94 ± 0.31 -1.34 ± 0.30 regulator of chromatin, subfamily c, member 1 AF051911 telomerase reverse transcriptase Tert -2.17 ± 0.33 -1.18 ± 0.31 -1.62 ± 0.30 M38700 thyroid autoantigen G22p1 1.82 ± 0.32 1.98 ± 0.31 2.17 ± 0.30
RNA processing AF017153 DEAH (Asp-Glu-Ala-His) box polypeptide 15 Dhx15 -1.53 ± 0.32 -2.19 ± 0.32 -1.82 ± 0.30 AK168269 EBNA1 binding protein 2 Ebna1bp2 1.99 ± 0.32 1.75 ± 0.31 1.69 ± 0.30 AF073991 heterogeneous nuclear ribonucleoprotein U Hnrpu -1.60 ± 0.32 -3.00 ± 0.31 -2.33 ± 0.31 L17076 * hnRNP-associated with lethal yellow Raly 2.81 ± 0.32 4.22 ± 0.31 2.29 ± 0.30 AK009617 LSM5 homolog, U6 small nuclear RNA associated (S. Lsm5 2.59 ± 0.38 2.38 ± 0.36 2.00 ± 0.33 cerevisiae) S64860 * non-POU-domain-containing, octamer binding protein Nono 2.01 ± 0.32 2.37 ± 0.31 2.24 ± 0.30 AA690061 RNA binding motif protein 8a Rbm8a 2.12 ± 0.32 1.61 ± 0.30 1.30 ± 0.30 AF031568 RNA binding motif protein, X chromosome retrogene Rbmxrt 2.11 ± 0.32 1.44 ± 0.31 1.07 ± 0.30 AI837853 small nuclear ribonucleoprotein D2 Snrpd2 3.23 ± 0.32 2.42 ± 0.30 1.67 ± 0.30 AK005341 small nuclear ribonucleoprotein D3 Snrpd3 3.03 ± 0.32 2.23 ± 0.30 2.08 ± 0.30 AK168185 speckle-type POZ protein Spop -1.82 ± 0.32 -2.16 ± 0.30 -2.02 ± 0.30 X83733 * splicing factor 3a, subunit 2 Sf3a2 -2.41 ± 0.32 -2.08 ± 0.31 -2.12 ± 0.30 AB037890 splicing factor 3b, subunit 1 Sf3b1 2.07 ± 0.32 1.88 ± 0.30 -1.03 ± 0.30 X96767 U1 small nuclear ribonucleoprotein 1C Snrp1c 2.00 ± 0.32 1.91 ± 0.30 1.63 ± 0.30
Transport AK034591 achalasia, adrenocortical insufficiency, alacrimia Aaas 2.21 ± 0.33 2.36 ± 0.31 1.13 ± 0.30 D87898 ADP-ribosylation factor 1 Arf1 2.02 ± 0.32 2.19 ± 0.31 1.83 ± 0.30 D87902 ADP-ribosylation factor 5 Arf5 -1.61 ± 0.32 -2.38 ± 0.31 -1.89 ± 0.30 U88623 * aquaporin 4 Aqp4 1.17 ± 0.32 2.12 ± 0.31 2.30 ± 0.30 X15789 cellular retinoic acid binding protein I Crabp1 1.65 ± 0.33 2.37 ± 0.32 2.65 ± 0.30 X14961 fatty acid binding protein 3, muscle and heart Fabp3 2.07 ± 0.32 1.30 ± 0.30 -1.21 ± 0.30 M73748 glycoprotein 38 Gp38 -1.07 ± 0.32 1.14 ± 0.31 2.39 ± 0.30 X61455 N-ethylmaleimide sensitive fusion protein attachment Napb 2.44 ± 0.32 2.91 ± 0.31 1.72 ± 0.31 protein beta U69171 peroxisome biogenesis factor 7 Pex7 1.35 ± 0.33 1.37 ± 0.31 1.98 ± 0.30 AF408432 RAB4B, member RAS oncogene family Rab4b 2.02 ± 0.32 1.66 ± 0.31 1.12 ± 0.31 M79313 RAB6, member RAS oncogene family Rab6 -2.09 ± 0.32 -2.18 ± 0.32 -1.91 ± 0.30 AF084642 retinaldehyde binding protein 1 Rlbp1 1.61 ± 0.32 1.70 ± 0.31 2.31 ± 0.30 AF022962 SEC8-like 1 (S. cerevisiae) Sec8l1 2.31 ± 0.33 2.70 ± 0.31 1.67 ± 0.30 AK015706 secretory carrier membrane protein 1 Scamp1 -2.08 ± 0.32 -2.06 ± 0.30 -1.72 ± 0.30
180 AY572835 solute carrier family 5 (sodium-dependent vitamin Slc5a6 -2.10 ± 0.32 -1.77 ± 0.31 -1.94 ± 0.30 transporter), member 6 AF257711 solute carrier family 15 (H+/peptide transporter), member Slc15a2 -2.08 ± 0.33 1.04 ± 0.31 1.40 ± 0.30 2 M73696 solute carrier family 20, member 1 Slc20a1 -1.25 ± 0.32 -2.02 ± 0.31 -2.26 ± 0.30 U25739 solute carrier family 23 (nucleobase transporters), member Slc23a3 -1.79 ± 0.32 -2.08 ± 0.31 -2.03 ± 0.30 3 U27315 * solute carrier family 25 (mitochondrial carrier, adenine Slc25a4 2.05 ± 0.32 1.69 ± 0.30 1.43 ± 0.30 nucleotide translocator), member 4 AJ006341 solute carrier family 25 (mitochondrial carrier, Slc25a17 -2.10 ± 0.32 -2.05 ± 0.30 -1.90 ± 0.30 peroxisomal membrane protein), member 17 X86682 solute carrier family 29 (nucleoside transporters), member Slc29a2 2.47 ± 0.34 -1.46 ± 0.33 -1.43 ± 0.31 2 AL591430 solute carrier family 35, member C2 Slc35c2 -2.52 ± 0.32 -2.35 ± 0.31 -1.97 ± 0.30 AA407151 solute carrier organic anion transporter family, member Slco5a1 -2.36 ± 0.33 -2.56 ± 0.31 -2.53 ± 0.30 5A1 AF062484 sorting nexin 12 Snx12 1.98 ± 0.33 1.86 ± 0.31 2.42 ± 0.30 AK077650 sorting nexin 17 Snx17 1.44 ± 0.33 2.15 ± 0.31 1.73 ± 0.30 U10355 * synaptotagmin 4 Syt4 2.75 ± 0.32 3.01 ± 0.30 1.34 ± 0.30 AB026808 synaptotagmin 11 Syt11 3.10 ± 0.32 3.48 ± 0.31 2.55 ± 0.36 AI846109 * transmembrane emp24 protein transport domain Tmed7 2.20 ± 0.32 1.60 ± 0.31 1.23 ± 0.30 containing 7 AJ001598 * vesicular inhibitory amino acid transporter Viaat -2.74 ± 0.32 -2.62 ± 0.30 -2.07 ± 0.30
Ion transport AF039405 arsA (bacterial) arsenite transporter, ATP-binding, Asna1 2.35 ± 0.32 2.07 ± 0.30 1.57 ± 0.30 homolog 1 U03434 * ATPase, Cu++ transporting, alpha polypeptide Atp7a 2.75 ± 0.36 2.24 ± 0.31 1.48 ± 0.30 AK007610 ATPase, H+ transporting, V0 subunit Atp6v0e 2.05 ± 0.32 2.09 ± 0.31 2.41 ± 0.30 U13837 ATPase, H+ transporting, V1 subunit A, isoform 1 Atp6v1a1 -1.82 ± 0.32 -2.28 ± 0.31 -1.62 ± 0.30 AK137602 ATPase, Na+/K+ transporting, alpha 2 polypeptide Atp1a2 1.74 ± 0.32 2.02 ± 0.31 2.99 ± 0.30 AJ012569 calcium channel, voltage-dependent, T type, alpha 1G Cacna1g 2.22 ± 0.33 1.15 ± 0.31 -1.40 ± 0.30 subunit AF029347 * chloride channel 3 Clcn3 2.21 ± 0.32 3.01 ± 0.31 1.43 ± 0.30 AK167765 chloride intracellular channel 1 Clic1 1.36 ± 0.32 1.54 ± 0.31 1.95 ± 0.30 J03941 * ferritin heavy chain Fth1 -2.01 ± 0.32 -2.06 ± 0.31 -1.85 ± 0.30 L32372 glutamate receptor, ionotropic, AMPA2 (alpha 2) Gria2 2.01 ± 0.32 1.49 ± 0.31 -1.25 ± 0.30 AF182040 glutamate receptor, ionotropic, N-methyl D-asparate- Grina 1.62 ± 0.32 1.66 ± 0.31 2.08 ± 0.30 associated protein 1 (glutamate binding) M21530 inositol 1,4,5-triphosphate receptor 1 Itpr1 -1.40 ± 0.33 -1.84 ± 0.31 -3.69 ± 0.30 L16912 potassium large conductance calcium-activated channel, Kcnma1 -1.07 ± 0.32 -1.13 ± 0.31 -2.25 ± 0.30 subfamily M, alpha member 1 M30441 potassium voltage-gated channel, shaker-related Kcna3 1.92 ± 0.35 2.38 ± 0.32 1.62 ± 0.31 subfamily, member 3 U70068 * potassium voltage-gated channel, subfamily Q, member 1 Kcnq1 -2.80 ± 0.32 -2.87 ± 0.31 -2.24 ± 0.30
AB000503 potassium voltage-gated channel, subfamily Q, member 2 Kcnq2 -1.32 ± 0.32 -2.31 ± 0.31 -3.54 ± 0.30
AF089751 purinergic receptor P2X, ligand-gated ion channel 4 P2rx4 2.72 ± 0.58 3.03 ± 0.37 2.77 ± 0.45 X17320 Purkinje cell protein 4 Pcp4 1.89 ± 0.32 2.03 ± 0.31 1.61 ± 0.30 AF004941 S100 calcium binding protein A3 S100a3 -2.81 ± 0.34 -2.22 ± 0.31 -1.59 ± 0.31 U41341 S100 calcium binding protein A11 (calizzarin) S100a11 -2.33 ± 0.32 -2.36 ± 0.31 -2.13 ± 0.30 X99921 S100 calcium binding protein A13 S100a13 -2.11 ± 0.32 -1.98 ± 0.31 1.05 ± 0.30 AK003669 S100 calcium binding protein A14 S100a14 -3.48 ± 0.33 -2.78 ± 0.31 -2.84 ± 0.31 X68837 * secretogranin II Scg2 1.76 ± 0.32 2.69 ± 0.31 3.08 ± 0.30 AF004666 solute carrier family 8 (sodium/calcium exchanger), Slc8a1 -1.09 ± 0.32 -1.33 ± 0.31 -2.37 ± 0.30 member 1 L33415 solute carrier family 11 (proton-coupled divalent metal ion Slc11a2 -1.64 ± 0.33 -1.70 ± 0.32 -2.21 ± 0.30 transporters), member 2 AF058055 solute carrier family 16 (monocarboxylic acid Slc16a1 1.49 ± 0.35 2.23 ± 0.36 2.86 ± 0.32 transporters), member 1 U76009 * solute carrier family 30 (zinc transporter), member 3 Slc30a3 -1.99 ± 0.32 -2.12 ± 0.30 -1.79 ± 0.30 AI848508 SPARC related modular calcium binding 1 Smoc1 -1.02 ± 0.32 1.66 ± 0.31 2.90 ± 0.30 AB021665 transient receptor potential cation channel, subfamily V, Trpv2 -1.19 ± 0.32 -1.27 ± 0.31 -2.11 ± 0.30 member 2 AF061346 tumor necrosis factor, alpha-induced protein 1 Tnfaip1 1.81 ± 0.32 2.55 ± 0.31 2.14 ± 0.30 (endothelial)
Electron transport M31775 cytochrome b-245, alpha polypeptide Cyba -2.08 ± 0.33 -1.22 ± 0.31 1.55 ± 0.31 X54691 cytochrome c oxidase subunit IV isoform 1 Cox4i1 2.00 ± 0.32 1.57 ± 0.30 1.38 ± 0.30
181 AK012602 * cytochrome c oxidase, subunit VIc Cox6c 2.35 ± 0.32 2.11 ± 0.30 1.70 ± 0.30 AK002593 * cytochrome c oxidase subunit VIIb Cox7b 2.87 ± 0.32 2.58 ± 0.30 2.22 ± 0.30 AK005708 * cytochrome c oxidase, subunit VIIIc Cox8c -2.39 ± 0.33 -2.70 ± 0.30 -2.07 ± 0.30 AK002815 cytochrome c-1 Cyc1 -2.36 ± 0.32 -2.90 ± 0.31 -1.89 ± 0.31 AA212964 cytochrome P450, family 4, subfamily v, polypeptide 3 Cyp4v3 1.73 ± 0.37 1.53 ± 0.34 2.59 ± 0.31 U36993 cytochrome P450, family 7, subfamily b, polypeptide 1 Cyp7b1 1.41 ± 0.33 1.75 ± 0.32 2.05 ± 0.30 AF080580 demethyl-Q 7 Coq7 2.03 ± 0.32 1.68 ± 0.31 1.44 ± 0.30 AK028118 * electron transferring flavoprotein, alpha polypeptide Etfa 1.64 ± 0.32 1.60 ± 0.31 2.00 ± 0.30 D16215 flavin containing monooxygenase 1 Fmo1 -2.15 ± 0.32 -1.66 ± 0.31 -1.50 ± 0.30 AK003133 NADH dehydrogenase (ubiquinone) 1 alpha subcomplex 3 Ndufa3 2.35 ± 0.32 2.04 ± 0.30 1.60 ± 0.30
AK005339 * NADH dehydrogenase (ubiquinone) 1 alpha subcomplex Ndufa10 -1.60 ± 0.32 -2.06 ± 0.31 -1.91 ± 0.30 10 U51908 * neurotensin receptor 2 Ntsr2 1.35 ± 0.32 1.82 ± 0.31 2.68 ± 0.30 D17571 P450 (cytochrome) oxidoreductase Por 1.97 ± 0.32 1.97 ± 0.31 1.56 ± 0.30 AK003217 * thioredoxin domain containing 4 (endoplasmic reticulum) Txndc4 2.31 ± 0.32 2.42 ± 0.31 2.63 ± 0.30
AW045202 thioredoxin domain containing 7 Txndc7 2.30 ± 0.32 2.42 ± 0.31 1.62 ± 0.30 AF052660 * thioredoxin-like 1 Txnl1 2.14 ± 0.32 1.76 ± 0.31 -1.05 ± 0.30 AK005151 * ubiquinol cytochrome c reductase core protein 2 Uqcrc2 2.61 ± 0.32 2.32 ± 0.31 1.41 ± 0.30
Signal transduction U70210 amyloid beta (A4) precursor protein-binding, family B, Apbb2 -1.08 ± 0.33 -1.23 ± 0.31 -2.25 ± 0.30 member 2 U76758 ankyrin 1, erythroid Ank1 -2.40 ± 0.32 -2.67 ± 0.31 -2.13 ± 0.30 AB007141 5-azacytidine induced gene 2 Azi2 2.64 ± 0.38 1.64 ± 0.32 1.51 ± 0.30 AB012693 CD47 antigen (Rh-related antigen, integrin-associated Cd47 1.90 ± 0.32 2.26 ± 0.30 1.91 ± 0.30 signal transducer) M63801 gap junction membrane channel protein alpha 1 Gja1 -1.03 ± 0.32 1.18 ± 0.31 2.80 ± 0.30 AB019479 homer homolog 1 (Drosophila) Homer1 2.51 ± 0.34 2.66 ± 0.49 1.06 ± 0.33 AJ223156 immunoglobulin (CD79A) binding protein 1 Igbp1 1.91 ± 0.32 2.08 ± 0.31 2.32 ± 0.30 M29462 malate dehydrogenase 1, NAD (soluble) Mdh1 -2.12 ± 0.33 -1.80 ± 0.32 -1.80 ± 0.30 AB001489 * polymeric immunoglobulin receptor Pigr -2.28 ± 0.33 -2.22 ± 0.31 -1.84 ± 0.30 AK005340 * signal sequence receptor, beta Ssr2 1.97 ± 0.32 2.19 ± 0.31 1.71 ± 0.30 AY178580 synaptojanin 2 binding protein Synj2bp 2.41 ± 0.34 2.69 ± 0.34 1.67 ± 0.32 AI852645 thyroid hormone receptor interactor 13 Trip13 1.96 ± 0.35 -1.44 ± 0.31 -2.05 ± 0.31 U59864 TRAF family member-associated Nf-kappa B activator Tank 1.53 ± 0.32 1.67 ± 0.31 2.13 ± 0.30
G-protein coupled receptor protein signaling pathway AF143241 ADP-ribosylation factor-like 3 Arl3 2.12 ± 0.32 1.84 ± 0.30 1.65 ± 0.30 AA822412 ADP-ribosylation factor-like 10C Arl10c -1.13 ± 0.33 1.21 ± 0.31 2.28 ± 0.30 AK052588 arrestin, beta 1 Arrb1 -1.94 ± 0.32 -2.07 ± 0.30 -1.61 ± 0.30 M19380 calmodulin 3 Calm3 2.08 ± 0.32 1.41 ± 0.31 1.46 ± 0.30 AK035494 centaurin, gamma 2 Centg2 -2.07 ± 0.33 -2.23 ± 0.31 -2.56 ± 0.30 U96963 diaphanous homolog 1 (Drosophila) Diap1 -1.71 ± 0.32 -1.85 ± 0.31 -2.10 ± 0.30 D21061 G protein-coupled receptor 12 Gpr12 -1.53 ± 0.32 -2.31 ± 0.31 -2.45 ± 0.30 U46923 G protein-coupled receptor 19 Gpr19 1.63 ± 0.32 2.20 ± 0.31 1.96 ± 0.30 AK088490 G protein pathway suppressor 1 Gps1 2.03 ± 0.32 1.51 ± 0.31 1.02 ± 0.30 AF042783 galanin receptor 3 Galr3 -1.20 ± 0.33 -2.15 ± 0.31 -1.69 ± 0.30 AF107848 GNAS (guanine nucleotide binding protein, alpha Gnas 2.23 ± 0.32 1.91 ± 0.31 1.70 ± 0.30 stimulating) complex locus U10551 GTP binding protein (gene overexpressed in skeletal Gem 1.15 ± 0.34 2.09 ± 0.55 4.30 ± 0.31 muscle) U37413 guanine nucleotide binding protein, alpha 11 Gna11 2.02 ± 0.32 1.22 ± 0.31 -1.03 ± 0.30 U38501 guanine nucleotide binding protein, alpha inhibiting 1 Gnai1 1.69 ± 0.33 2.10 ± 0.31 1.74 ± 0.30 X94908 5-hydroxytryptamine (serotonin) receptor 1D Htr1d -1.40 ± 0.32 -2.49 ± 0.31 -1.72 ± 0.30 AJ133429 olfactory receptor 71 Olfr71 -2.20 ± 0.32 -2.47 ± 0.31 -1.65 ± 0.30 AI180687 phosphodiesterase 4B, cAMP specific Pde4b -1.95 ± 0.33 -2.50 ± 0.31 -1.57 ± 0.30 D86066 * rabaptin, RAB GTPase binding effector protein 1 Rabep1 4.57 ± 0.32 3.64 ± 0.31 1.94 ± 0.38 AF011360 regulator of G protein signaling 7Rgs7 1.22 ± 0.32 -1.28 ± 0.31 -2.25 ± 0.30 U85055 regulator of G-protein signaling 14 Rgs14 -2.11 ± 0.32 -1.80 ± 0.31 -1.68 ± 0.30 AF104358 * regulator of G-protein signaling 19 interacting protein 1 Rgs19ip1 2.08 ± 0.32 2.00 ± 0.31 1.61 ± 0.30
L07918 Rho, GDP dissociation inhibitor (GDI) beta Arhgdib -1.34 ± 0.33 -1.80 ± 0.31 -2.24 ± 0.30 U67160 Rho GTPase activating protein 5 Arhgap5 1.43 ± 0.34 1.79 ± 0.31 2.53 ± 0.31
Carbohydrate metabolism AK002386 glyoxalase 1 Glo1 1.06 ± 0.32 1.35 ± 0.30 2.02 ± 0.30 AA939571 hexose-6-phosphate dehydrogenase (glucose 1- H6pd -2.52 ± 0.32 -2.47 ± 0.31 -2.28 ± 0.30 dehydrogenase) AJ223966 N-acetyl galactosaminidase, alpha Naga -1.14 ± 0.33 1.21 ± 0.31 2.22 ± 0.30
182 X97047 pyruvate kinase, muscle Pkm2 -2.37 ± 0.33 -2.86 ± 0.30 -1.88 ± 0.31 M32599 similar to glyceraldehyde-3-phosphate dehydrogenase LOC14433 2.68 ± 0.32 2.28 ± 0.30 1.66 ± 0.30
Enzymatic activity U07235 * aldehyde dehydrogenase 2, mitochondrial Aldh2 1.47 ± 0.32 1.56 ± 0.31 2.27 ± 0.30 AK016213 CTD (carboxy-terminal domain, RNA polymerase II, Ctdp1 -2.26 ± 0.33 -1.56 ± 0.31 -1.66 ± 0.30 polypeptide A) phosphatase, subunit 1 X95281 dehydrogenase/reductase (SDR family) member 3 Dhrs3 1.64 ± 0.34 2.60 ± 0.33 2.97 ± 0.37 AK002230 fucosidase, alpha-L- 1, tissue Fuca1 2.11 ± 0.32 1.89 ± 0.31 1.70 ± 0.30 U09114 glutamate-ammonia ligase (glutamine synthase) Glul -2.27 ± 0.33 -1.91 ± 0.31 -1.82 ± 0.30 D50430 * glycerol phosphate dehydrogenase 2, mitochondrial Gpd2 3.51 ± 0.32 2.96 ± 0.31 1.03 ± 0.30 U27295 inositol polyphosphate-1-phosphatase Inpp1 -2.01 ± 0.33 -2.56 ± 0.31 -2.14 ± 0.30 AK076290 * N-acetylneuraminic acid synthase (sialic acid synthase) Nans 2.64 ± 0.32 2.26 ± 0.31 1.46 ± 0.30
L48514 paraoxonase 2 Pon2 1.07 ± 0.33 1.17 ± 0.31 2.18 ± 0.30 AY039032 retinol dehydrogenase 11 Rdh11 2.04 ± 0.32 1.54 ± 0.30 -1.10 ± 0.30 AK002261 thioesterase superfamily member 2 Them2 2.77 ± 0.33 1.96 ± 0.31 1.62 ± 0.30
Others AB023957 adenomatosis polyposis coli down-regulated 1 Apcdd1 1.50 ± 0.32 2.48 ± 0.31 2.18 ± 0.30 AK133934 ADP-ribosylation factor-like 6 interacting protein 4 Arl6ip4 2.17 ± 0.32 1.55 ± 0.31 1.29 ± 0.30 AK146202 amyloid beta precursor protein binding protein 1 Appbp1 2.49 ± 0.32 1.77 ± 0.30 1.07 ± 0.30 AK142858 AP1 gamma subunit binding protein 1 Ap1gbp1 2.21 ± 0.34 2.30 ± 0.31 1.25 ± 0.31 AB041609 brain protein 17 Brp17 2.08 ± 0.33 2.10 ± 0.31 1.61 ± 0.30 AK005825 breast carcinoma amplified sequence 2 Bcas2 2.03 ± 0.32 2.04 ± 0.31 1.94 ± 0.30 X59047 * CD 81 antigen Cd81 1.37 ± 0.32 1.55 ± 0.31 2.33 ± 0.30 AK047941 COMM domain containing 2 Commd2 2.09 ± 0.32 1.63 ± 0.30 -1.02 ± 0.30 AW049394 COMM domain containing 6 Commd6 2.15 ± 0.32 1.78 ± 0.31 1.56 ± 0.30 AK007640 * COMM domain containing 9 Commd9 1.97 ± 0.32 2.15 ± 0.31 1.69 ± 0.30 AK004009 cysteine and histidine rich 1 Cyhr1 2.27 ± 0.32 1.75 ± 0.31 1.16 ± 0.30 X53929 decorin Dcn 2.16 ± 0.33 2.67 ± 0.31 1.45 ± 0.30 Y08460 degenerative spermatocyte homolog (Drosophila) Degs 1.78 ± 0.32 1.75 ± 0.30 2.13 ± 0.30 Z79787 dystrobrevin alpha Dtna -1.53 ± 0.32 -1.46 ± 0.31 -2.23 ± 0.31 AJ005985 endosulfine alpha Ensa 2.49 ± 0.32 3.14 ± 0.31 2.86 ± 0.30 X91043 erythrocyte protein band 7.2 Epb7.2 -2.35 ± 0.32 -2.69 ± 0.32 -1.62 ± 0.30 AF153451 expressed in non-metastatic cells 4, protein Nme4 1.68 ± 0.33 2.06 ± 0.32 2.45 ± 0.31 AI047076 fidgetin-like 1 Fignl1 -2.27 ± 0.33 -1.79 ± 0.31 -1.61 ± 0.30 AK005061 FIP1 like 1 (S. cerevisiae) Fip1l1 1.88 ± 0.33 2.37 ± 0.31 2.63 ± 0.30 AK146018 * FK506 binding protein-like Fkbpl 1.73 ± 0.33 2.34 ± 0.33 2.11 ± 0.31 AI931876 golgi phosphoprotein 4 Golph4 1.34 ± 0.33 1.81 ± 0.32 2.36 ± 0.31 L04961 inactive X specific transcripts Xist 2.06 ± 0.33 2.35 ± 0.37 1.18 ± 0.31 AK006977 inner membrane protein, mitochondrial Immt -2.12 ± 0.33 -2.33 ± 0.31 -1.89 ± 0.30 AJ011565 leptin receptor overlapping transcriptLeprot 1.98 ± 0.32 2.01 ± 0.30 1.45 ± 0.30 D45913 leucine rich repeat protein 1, neuronal Lrrn1 1.23 ± 0.33 1.74 ± 0.31 2.06 ± 0.30 M82903 Myb protein P42POP P42pop -2.54 ± 0.32 -2.76 ± 0.31 -2.39 ± 0.30 AK046219 * NudC domain containing 3 Nudcd3 -1.86 ± 0.32 -2.20 ± 0.30 -1.81 ± 0.30 AB025413 odd Oz/ten-m homolog 4 (Drosophila) Odz4 1.43 ± 0.32 1.07 ± 0.31 -2.17 ± 0.30 X02966 per-hexamer repeat gene 5 Phxr5 2.46 ± 0.33 2.46 ± 0.32 1.73 ± 0.31 AI849556 * pleckstrin homology domain containing, family A Plekha1 -1.84 ± 0.32 -2.04 ± 0.30 -2.02 ± 0.30 (phosphoinositide binding specific) member 1 X78682 * prohibitin Phb 2.22 ± 0.32 2.13 ± 0.31 1.63 ± 0.30 AF006465 * RAN binding protein 9 Ranbp9 1.22 ± 0.32 1.72 ± 0.30 2.44 ± 0.30 AF051347 * reduced expression 3 Rex3 2.09 ± 0.32 2.00 ± 0.30 1.37 ± 0.30 Z27088 relaxin 1 Rln1 -1.19 ± 0.33 1.19 ± 0.33 2.03 ± 0.30 AA600645 renin binding protein Renbp -1.76 ± 0.33 -2.20 ± 0.31 -1.23 ± 0.30 X70296 * serine (or cysteine) proteinase inhibitor, clade E, member Serpine2 1.26 ± 0.32 1.50 ± 0.31 2.04 ± 0.30 2 AK007881 SGT1, suppressor of G2 allele of SKP1 (S. cerevisiae) Sugt1 1.99 ± 0.32 2.03 ± 0.31 1.59 ± 0.30 AW125253 small EDRK-rich factor 2 Serf2 2.85 ± 0.33 2.77 ± 0.31 2.62 ± 0.30 AF068921 soc-2 (suppressor of clear) homolog (C. elegans) Shoc2 1.59 ± 0.33 1.91 ± 0.31 2.09 ± 0.31 X92864 * sparc/osteonectin, cwcv and kazal-like domains Spock1 -1.73 ± 0.32 -3.85 ± 0.31 -2.45 ± 0.30 proteoglycan 1 C79529 * subcommissural organ spondin Sspo -2.13 ± 0.32 -2.12 ± 0.31 -1.49 ± 0.30 U40375 suppressor of Ty 6 homolog (S. cerevisiae) Supt6h -2.00 ± 0.32 -2.37 ± 0.31 -2.04 ± 0.30 L32025 synapsin I Syn1 2.27 ± 0.32 1.94 ± 0.31 1.80 ± 0.30 AF532969 * synovial sarcoma, X breakpoint 2 interacting protein Ssx2ip 2.37 ± 0.32 2.38 ± 0.31 2.23 ± 0.30 AF044672 * synuclein, alpha Snca 2.44 ± 0.32 2.04 ± 0.30 1.87 ± 0.30 AI037493 TBC1 domain family, member 15 Tbc1d15 1.58 ± 0.33 1.84 ± 0.31 2.37 ± 0.30 X78989 testis derived transcript Tes -1.10 ± 0.32 -2.12 ± 0.31 -5.59 ± 0.30 AK076586 trans-golgi network protein Tgoln1 1.57 ± 0.33 2.22 ± 0.31 2.08 ± 0.30 AK008175 transmembrane 4 superfamily member 13 Tm4sf13 2.08 ± 0.33 2.06 ± 0.31 -1.13 ± 0.30
183 AW047445 transmembrane 7 superfamily member 2 Tm7sf2 2.53 ± 0.32 1.68 ± 0.31 -1.47 ± 0.30 AW125880 transmembrane protein 4 Tmem4 3.03 ± 0.32 2.64 ± 0.31 2.99 ± 0.30 AW125346 transmembrane protein 14C Tmem14c 1.97 ± 0.32 1.80 ± 0.31 1.67 ± 0.30 AF220124 * tripartite motif protein 11 Trim11 -2.35 ± 0.32 -1.90 ± 0.30 -1.47 ± 0.30 AF272895 UBX domain containing 1 Ubxd1 2.28 ± 0.32 2.18 ± 0.31 1.78 ± 0.30 AI838094 UPF3 regulator of nonsense transcripts homolog B (yeast) Upf3b -1.41 ± 0.32 -2.08 ± 0.31 -1.59 ± 0.30
U92454 WW domain binding protein 5 Wbp5 1.92 ± 0.32 2.18 ± 0.30 1.67 ± 0.30 AF071186 WW domain binding protein 11 Wbp11 -2.23 ± 0.33 -1.89 ± 0.30 -1.45 ± 0.30
Unknown AI849678 Adult male testis cDNA, RIKEN full-length enriched --- 2.16 ± 0.32 2.06 ± 0.31 1.33 ± 0.30 library, clone:1700092M07 product:unknown EST, full insert sequence AV100072 AV100072 Mus musculus C57BL/6J ES cell Mus --- -2.22 ± 0.32 -1.42 ± 0.32 1.08 ± 0.30 musculus cDNA clone 2410054H03, mRNA sequence. AV237354 AV237354 RIKEN full-length enriched, 10 day neonate --- -1.89 ± 0.32 -2.00 ± 0.31 -2.18 ± 0.31 skin Mus musculus cDNA clone 4732420J08 3', mRNA sequence. AV349686 AV349686 RIKEN full-length enriched, adult male --- 1.63 ± 0.32 1.64 ± 0.31 2.43 ± 0.30 cerebellum Mus musculus cDNA clone 6530414M01 3', mRNA sequence. AV377060 AV377060 RIKEN full-length enriched, adult male cecum --- -2.07 ± 0.33 -2.28 ± 0.31 -2.16 ± 0.30 Mus musculus cDNA clone 9130221I18 3', mRNA sequence. AI854144 cDNA sequence BC004636 BC004636 -1.70 ± 0.32 -2.19 ± 0.31 -1.69 ± 0.30 AI838889 cDNA sequence BC005624 BC005624 1.77 ± 0.32 2.02 ± 0.31 1.77 ± 0.30 AW060889 cDNA sequence BC005662 BC005662 2.48 ± 0.32 2.14 ± 0.31 1.64 ± 0.31 AV265258 cDNA sequence BC037034 BC037034 -1.38 ± 0.32 -1.67 ± 0.31 -2.47 ± 0.30 AW047237 cDNA sequence BC052328 BC052328 1.49 ± 0.35 2.32 ± 0.33 1.84 ± 0.31 AV341577 DNA segment, Chr 1, ERATO Doi 161, expressed D1Ertd161e 2.23 ± 0.34 2.00 ± 0.32 1.21 ± 0.31 AI853668 DNA segment, Chr 10, ERATO Doi 610, expressed D10Ertd610e -2.36 ± 0.32 -1.56 ± 0.31 -1.13 ± 0.30 AI847163 DNA segment, Chr 11, Wayne State University 68, D11Wsu68e 1.83 ± 0.33 2.03 ± 0.31 1.59 ± 0.30 expressed AI844034 DNA segment, Chr 16, Wayne State University 109, D16Wsu109e 2.28 ± 0.33 1.83 ± 0.31 1.08 ± 0.30 expressed C79676 DNA segment, Chr 18, ERATO Doi 232, expressed D18Ertd232e 2.19 ± 0.33 2.07 ± 0.32 1.72 ± 0.31 AI851365 * DNA segment, Chr 19, Wayne State University 12, D19Wsu12e 1.78 ± 0.32 1.56 ± 0.31 2.25 ± 0.30 expressed N28141 * DNA segment, Chr 6, Brigham & Women's Genetics 1452 D6Bwg1452e 2.11 ± 0.32 2.22 ± 0.31 1.26 ± 0.30 expressed AA796868 DNA segment, Chr 6, ERATO Doi 365, expressed D6Ertd365e 2.35 ± 0.33 1.46 ± 0.31 1.05 ± 0.30 AW048884 DNA segment, Chr 8, ERATO Doi 319, expressed D8Ertd319e -1.83 ± 0.32 -2.27 ± 0.31 -2.03 ± 0.30 AW214136 EST X83313 X83313 -12.23 ± 0.32 -15.39 ± 0.31 -7.65 ± 0.30 AI604013 expressed sequence AA408985 AA408985 -1.19 ± 0.32 -2.08 ± 0.31 -1.56 ± 0.30 AA623587 expressed sequence AA536743 AA536743 1.41 ± 0.32 2.27 ± 0.31 2.48 ± 0.30 AW125330 expressed sequence AL024210 AL024210 -1.02 ± 0.32 1.36 ± 0.33 2.19 ± 0.31 C80068 expressed sequence C80068 C80068 1.38 ± 0.32 -1.31 ± 0.31 -2.22 ± 0.31 M17551 * hypothetical protein LOC280487 LOC280487 1.29 ± 0.32 1.92 ± 0.31 2.49 ± 0.30 Z22552 * M.musculus membrane glycoprotein gene. --- -2.09 ± 0.32 -2.00 ± 0.31 -1.63 ± 0.30 AI852553 M.musculus mRNA for testis-specific thymosin beta-10 --- 2.70 ± 0.32 2.35 ± 0.30 2.03 ± 0.30
D18865 * MUSGS01047 Mouse 3'-directed Mus musculus --- -1.91 ± 0.33 -5.74 ± 0.32 -1.69 ± 0.31 domesticus cDNA clone mc0129 3', mRNA sequence. AW125185 open reading frame 18 ORF18 1.50 ± 0.32 1.78 ± 0.31 2.23 ± 0.30 AI226264 RIKEN cDNA 0610009D07 gene 0610009D07Rik 2.01 ± 0.32 1.92 ± 0.30 1.60 ± 0.30 AI845987 RIKEN cDNA 0610009E20 gene 0610009E20Rik -2.23 ± 0.32 -2.50 ± 0.31 -2.30 ± 0.30 AW047746 RIKEN cDNA 1110005L13 gene 1110005L13Rik 2.98 ± 0.32 1.99 ± 0.30 1.33 ± 0.30 AA693236 RIKEN cDNA 1110007M04 gene 1110007M04Rik 1.07 ± 0.32 -1.69 ± 0.31 -2.15 ± 0.30 AI851206 RIKEN cDNA 1110032A03 gene 1110032A03Rik 3.19 ± 0.33 3.00 ± 0.34 1.85 ± 0.40 AI839522 RIKEN cDNA 1110033C18 gene 1110033C18Rik 1.73 ± 0.32 2.01 ± 0.31 1.49 ± 0.30 AW046496 RIKEN cDNA 1110033J19 gene 1110033J19Rik 1.95 ± 0.32 2.04 ± 0.31 1.36 ± 0.30 AW122061 RIKEN cDNA 1190017O12 gene 1190017O12Rik 2.17 ± 0.32 2.25 ± 0.31 1.68 ± 0.30 AW060656 RIKEN cDNA 1200004M23 gene 1200004M23Rik -2.32 ± 0.33 -2.36 ± 0.31 -2.03 ± 0.30 AA815795 RIKEN cDNA 1200007D18 gene 1200007D18Rik -1.63 ± 0.32 -2.53 ± 0.30 -1.13 ± 0.30 AV248951 RIKEN cDNA 1300011C24 gene 1300011C24Rik -2.48 ± 0.33 -3.08 ± 0.31 -2.16 ± 0.31 AW124941 RIKEN cDNA 1500001L15 gene 1500001L15Rik 2.17 ± 0.33 1.71 ± 0.32 1.53 ± 0.31 AI851081 RIKEN cDNA 1500001L20 gene 1500001L20Rik -2.10 ± 0.32 -2.30 ± 0.31 -1.38 ± 0.31 AW047875 RIKEN cDNA 1500005K14 gene 1500005K14Rik 1.07 ± 0.32 -1.74 ± 0.31 -2.10 ± 0.30 AI853412 * RIKEN cDNA 1500011J06 gene 1500011J06Rik 2.17 ± 0.33 2.06 ± 0.31 2.28 ± 0.30 AI848821 RIKEN cDNA 1500034J20 gene 1500034J20Rik 2.90 ± 0.32 3.24 ± 0.31 2.90 ± 0.30
184 AW120867 RIKEN cDNA 1700013H19 gene 1700013H19Rik 2.36 ± 0.34 2.95 ± 0.32 1.98 ± 0.31 AW122255 RIKEN cDNA 1810009A16 gene 1810009A16Rik -1.63 ± 0.32 -1.70 ± 0.31 -2.25 ± 0.30 AW048976 RIKEN cDNA 1810020G14 gene 1810020G14Rik -2.00 ± 0.32 -1.80 ± 0.30 -1.84 ± 0.30 AI853323 RIKEN cDNA 1810030M08 gene 1810030M08Rik 2.21 ± 0.32 1.44 ± 0.31 1.17 ± 0.30 AW047207 * RIKEN cDNA 1810037I17 gene 1810037I17Rik 2.58 ± 0.50 5.21 ± 0.32 7.19 ± 0.55 AI882440 RIKEN cDNA 1810045K06 gene 1810045K06Rik -2.06 ± 0.32 -2.23 ± 0.31 -1.65 ± 0.30 AI852949 RIKEN cDNA 2010008E23 gene 2010008E23Rik 1.59 ± 0.32 1.93 ± 0.31 1.94 ± 0.30 AI853864 RIKEN cDNA 2010100O12 gene 2010100O12Rik 2.15 ± 0.32 1.77 ± 0.30 1.27 ± 0.30 AA615161 RIKEN cDNA 2010315L10 gene 2010315L10Rik 2.00 ± 0.32 2.01 ± 0.31 1.78 ± 0.30 AI835662 RIKEN cDNA 2010315L10 gene 2010315L10Rik 2.40 ± 0.32 1.97 ± 0.30 1.74 ± 0.30 AF109906 RIKEN cDNA 2210409B01 gene 2210409B01Rik -2.00 ± 0.32 -2.21 ± 0.31 -2.44 ± 0.30 AI840191 RIKEN cDNA 2210409B22 gene 2210409B22Rik 1.34 ± 0.32 1.82 ± 0.31 2.54 ± 0.30 AW046093 RIKEN cDNA 2210412K09 gene 2210412K09Rik 1.55 ± 0.32 1.95 ± 0.31 2.20 ± 0.30 AW045317 RIKEN cDNA 2300006M17 gene 2300006M17Rik -2.06 ± 0.33 -2.09 ± 0.31 -1.46 ± 0.31 AW121406 RIKEN cDNA 2310014H01 gene 2310014H01Rik -1.49 ± 0.33 -1.64 ± 0.31 -2.04 ± 0.30 AI325791 RIKEN cDNA 2310032M22 gene 2310032M22Rik 2.05 ± 0.32 2.37 ± 0.31 1.02 ± 0.30 AI507524 RIKEN cDNA 2310032M22 gene 2310032M22Rik 1.56 ± 0.32 3.44 ± 0.32 1.80 ± 0.35 AW228316 * RIKEN cDNA 2310046G15 gene 2310046G15Rik -1.07 ± 0.33 1.97 ± 0.31 4.37 ± 0.30 AI788543 * RIKEN cDNA 2310058J06 gene 2310058J06Rik -2.07 ± 0.32 -1.27 ± 0.31 1.64 ± 0.30 AI853654 RIKEN cDNA 2310061I09 gene 2310061I09Rik 2.70 ± 0.32 2.09 ± 0.31 2.09 ± 0.30 AI845915 RIKEN cDNA 2310075E07 gene 2310075E07Rik -2.19 ± 0.33 -2.49 ± 0.31 -1.95 ± 0.30 AW046723 * RIKEN cDNA 2400003B06 gene 2400003B06Rik -3.76 ± 0.32 -6.25 ± 0.30 -1.28 ± 0.30 AV207612 RIKEN cDNA 2410004C24 gene 2410004C24Rik 3.11 ± 0.33 2.25 ± 0.31 2.08 ± 0.30 AI853819 RIKEN cDNA 2410015M20 gene 2410015M20Rik 2.89 ± 0.33 2.56 ± 0.31 1.69 ± 0.30 AI504100 * RIKEN cDNA 2410019G02 gene 2410019G02Rik 2.46 ± 0.33 2.01 ± 0.31 1.44 ± 0.31 AV113045 * RIKEN cDNA 2410022L05 gene 2410022L05Rik -2.24 ± 0.32 -2.45 ± 0.31 -2.45 ± 0.30 AW208818 RIKEN cDNA 2410022L05 gene 2410022L05Rik 2.23 ± 0.33 1.61 ± 0.32 1.41 ± 0.30 AI837786 RIKEN cDNA 2500002E12 gene 2500002E12Rik 1.40 ± 0.32 1.50 ± 0.31 2.42 ± 0.31 AW258842 RIKEN cDNA 2510049I19 gene 2510049I19Rik 2.64 ± 0.32 3.55 ± 0.31 3.42 ± 0.30 AW045323 RIKEN cDNA 2600002E23 gene 2600002E23Rik 1.70 ± 0.32 2.78 ± 0.31 3.01 ± 0.30 AW122012 RIKEN cDNA 2610001E17 gene 2610001E17Rik 1.18 ± 0.33 2.05 ± 0.32 2.70 ± 0.30 AA815890 RIKEN cDNA 2610103J23 gene 2610103J23Rik 3.93 ± 0.42 3.41 ± 0.84 2.92 ± 0.35 AW121984 RIKEN cDNA 2610205H19 gene 2610205H19Rik 2.36 ± 0.32 2.47 ± 0.30 1.85 ± 0.30 AI648018 RIKEN cDNA 2610207I16 gene 2610207I16Rik 1.77 ± 0.34 1.20 ± 0.32 2.50 ± 0.30 AA982595 RIKEN cDNA 2610318G08 gene 2610318G08Rik 2.00 ± 0.32 2.14 ± 0.31 1.93 ± 0.30 AW124052 RIKEN cDNA 2700059D21 gene 2700059D21Rik 2.02 ± 0.32 1.79 ± 0.30 1.37 ± 0.30 AA989957 RIKEN cDNA 2700084L22 gene 2700084L22Rik 2.06 ± 0.33 2.17 ± 0.32 1.74 ± 0.32 AI843655 RIKEN cDNA 2810429C13 gene 2810429C13Rik 2.51 ± 0.33 1.55 ± 0.31 1.27 ± 0.31 AW047026 RIKEN cDNA 2810443J12 gene 2810443J12Rik 2.70 ± 0.32 1.80 ± 0.31 1.32 ± 0.31 AW121847 RIKEN cDNA 2900091E11 gene 2900091E11Rik 1.76 ± 0.32 2.67 ± 0.31 1.57 ± 0.31 AI047107 RIKEN cDNA 3732413I11 gene 3732413I11Rik 3.22 ± 0.32 2.26 ± 0.30 2.48 ± 0.30 AW122483 RIKEN cDNA 4631416I11 gene 4631416I11Rik 2.48 ± 0.33 1.66 ± 0.32 1.17 ± 0.31 AI843949 RIKEN cDNA 4632413K17 gene 4632413K17Rik 4.09 ± 0.33 4.92 ± 0.31 4.00 ± 0.42 AI846236 RIKEN cDNA 4732477C12 gene 4732477C12Rik 1.89 ± 0.33 1.34 ± 0.31 -1.74 ± 0.31 AW125109 RIKEN cDNA 4732477C12 gene 4732477C12Rik -1.69 ± 0.32 -1.75 ± 0.31 -2.04 ± 0.30 AV298145 * RIKEN cDNA 4833420N02 gene 4833420N02Rik 2.98 ± 0.33 2.31 ± 0.32 2.02 ± 0.31 AA409766 RIKEN cDNA 4833441D16 gene 4833441D16Rik 2.38 ± 0.33 1.29 ± 0.34 1.08 ± 0.30 AI835622 RIKEN cDNA 4930578F06 gene 4930578F06Rik 1.67 ± 0.36 2.45 ± 0.32 1.76 ± 0.32 AI595812 RIKEN cDNA 4933428G09 gene 4933428G09Rik 1.37 ± 0.33 1.66 ± 0.31 1.99 ± 0.31 AI426400 * RIKEN cDNA 5430401D19 gene 5430401D19Rik 1.54 ± 0.32 -1.16 ± 0.31 -2.05 ± 0.30 AI843959 RIKEN cDNA 5730403B10 gene 5730403B10Rik 2.37 ± 0.32 2.21 ± 0.31 1.85 ± 0.31 AI116222 RIKEN cDNA 5730494N06 gene 5730494N06Rik 2.01 ± 0.32 1.70 ± 0.31 1.65 ± 0.30 AI324972 * RIKEN cDNA 5730497N03 gene 5730497N03Rik 3.34 ± 0.35 3.33 ± 0.34 3.44 ± 0.31 AW125272 * RIKEN cDNA 5930418K15 gene 5930418K15Rik 3.67 ± 0.40 2.90 ± 0.31 1.70 ± 0.30 AW122948 RIKEN cDNA 9430020E02 gene 9430020E02Rik 2.24 ± 0.32 2.17 ± 0.30 1.37 ± 0.30 AI846811 * RIKEN cDNA 9430099J10 gene 9430099J10Rik 2.31 ± 0.32 3.37 ± 0.44 2.57 ± 0.30 AA014745 RIKEN cDNA A930001N09 gene A930001N09Rik 1.17 ± 0.34 2.18 ± 0.33 2.98 ± 0.31 AW047329 RIKEN cDNA C330007P06 gene C330007P06Rik 2.10 ± 0.32 1.52 ± 0.31 1.55 ± 0.30 AI846304 * RIKEN cDNA C630002B14 gene C630002B14Rik -1.15 ± 0.32 -1.62 ± 0.31 -2.13 ± 0.30 AW121683 RIKEN cDNA G431001I09 gene G431001I09Rik 2.27 ± 0.32 2.05 ± 0.31 1.75 ± 0.30 C79108 Transcribed sequences --- -1.34 ± 0.32 -2.28 ± 0.32 -2.41 ± 0.31 AI463421 * uc45a07.x1 Soares_mammary_gland_NMLMG Mus --- -2.04 ± 0.32 -2.58 ± 0.31 -2.12 ± 0.30 musculus cDNA clone IMAGE:1400916 3', mRNA sequence. AI837905 UI-M-AL0-abq-c-01-0-UI.s1 NIH_BMAP_MCO Mus --- 2.90 ± 0.32 1.96 ± 0.31 1.09 ± 0.30 musculus cDNA clone UI-M-AL0-abq-c-01-0-UI 3', mRNA sequence. X16670 unnamed protein product; open reading frame (461 AA) --- 2.29 ± 0.32 3.65 ± 0.31 4.16 ± 0.30 containing a duplicated 66bp region of integrase; Mouse RNA for type IIB intracisternal A-particle (IAP) element encoding integrase, clone 106.
185 Data represent the mean fold differences over the Control ± SEM of genes, whose average expression level differed by more than 2-fold at each time-point after one-way ANOVA for multiple comparisons (p < 0.01), as selected by GeneSpring™ from three microarray hybridizations for each of the treatment. Genes identified by GeneSpring™ after a conservative Bonferroni multiple testing correction are denoted by an asterisk (*). The GenBank accession number, gene title and symbol are given. Genes are classified into categories using NetAffx™ and DAVID based on their known biological functions. EST: expressed sequence tag. All GenBank numbers are correct as on 22 December 2005.
186 A
B
Figure 4.3: Venn diagrams showing the classification of differentially-expressed genes in cortical neurons after 24 h, 48 h and 72 h of U18666A (1 μg/ml) treatment. Diagrams were generated by GeneSpring™ after (A) one-way ANOVA for multiple comparisons and (B) Bonferroni multiple testing correction. Each circle represents U18666A treatment at a different time-point. The numbers in the regions between overlapping circles represent the number of genes that were differentially expressed after U18666A treatment at those particular time-points represented by the respective overlapping circles. The numbers in the non- overlapping regions of each circle represent the number of genes that were exclusively differentially expressed after U18666A treatment at that particular time-point represented by the respective circle.
187 Figure 4.4: Pages 188, 189 and 190. Cluster analysis of microarray data obtained from cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A for a maximum of 72 h. Clustering of differentially-expressed genes with at least 2-fold change after U18666A treatment (Table 4.1) was performed on the average of normalized expression in each condition. Each cell represents the average of normalized expression of three GeneChip® arrays per condition for a single gene. Rows represent individual genes; columns represent experimental conditions. Genes are organized on the basis of overall similarity in their expression. (A) Color bar representing the normalized expression scale. Blue indicates low level of expression; red indicates high level of expression. Differentially-expressed genes encoding for proteins involved in (B) cell adhesion, (C) cytoskeleton organization, (D) cell cycle, (E) cell death, (F) defense response, (G) response to stress, (H) cholesterol biosynthesis, (I) lipid metabolism and transport, (J) protein amino acid modification, (K) proteolysis and peptidolysis, (L) ubiquitin-proteasome system, (M) regulation of transcription, (N) DNA processing, (O) RNA processing, (P) transport, (Q) ion transport, (R) electron transport, (S) signal transduction and (T) G-protein coupled receptor protein signaling pathway.
188 A B F J
G C
H K
D I L
E
Figure 4.4: Continued.
189 M O R
P
S
T
Q
N
Figure 4.4: Continued.
190 A
B
Figure 4.5: Validation of selected gene products by Western blot analysis. (A) Cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A were harvested using the total protein isolation method at each time-point indicated. Proteins (10 μg per lane) were separated by 15% SDS-PAGE, transferred to PVDF membrane, and analyzed by Western blotting for selected gene products differentially expressed after U18666A treatment. Atf3, α-synuclein, cathepsin B, cyclophilin A, GADD 153 and histone 1 were detected with the respective antibodies. Arrows: pro-cathepsin B (43 kDa) and its subunits (31 kDa and 25 kDa). Internal control for equal loading: β- tubulin. (B) Mean fold differences of validated genes, as excerpted from Table 4.1.
191 4.3 Discussion
Microarray analysis is now a widely-used technology for gene expression research at high-throughput. The ability to array large numbers of individual gene fragments on small matrices has made it possible to synchronously allow quantitative measurements of gene expression for thousands of genes at a time. Microarray analysis was performed in order to profile the global alterations in gene expression involved in chronic U18666A exposure that ultimately lead to neuronal apoptosis.
The main scientific objective underpinning the use of microarray analysis was to determine biomarkers of U18666A exposure, action of the drug, and potential adverse effects during treatment.
The functional class of genes most responsive to U18666A treatment in primary cortical neurons was that of genes encoding for proteins involved in the regulation of transcription. This might most likely reflect the differential activity of transcription factors following exposure to U18666A. The activating transcription factor 3 (Atf3) is strongly induced in response to many environmental changes as an immediate early response gene, thus explaining its significant up-regulation after 48 h of U18666A treatment. Atf3 also appears to function in the regulation of cellular stress response and cell proliferation. The constitutive photomorphogenic homolog subunit 5
(Cops5), also known as Jab1, controls cell cycle progression and cell survival by regulating multiple cell cycle signaling pathways. A recent study has suggested that
Cops5 is involved in the onset of neuronal diseases through its interaction with the
192 endoplasmic reticulum (ER) stress transducer IRE1, leading to the accumulation of unfolded proteins due to ER stress (Oono et al., 2004). Cops5 was up-regulated during U18666A treatment, suggesting the possible contribution of ER stress to
U18666A-mediated neuronal apoptosis, as previously postulated in Chapter 3 Part III.
The protein encoded by the Max interacting protein 1 (Mxi1) gene, which was up- regulated in a time-dependent manner during U18666A treatment, is a transcriptional repressor thought to negatively regulate Myc function and belongs to a family of
Myc/Max/Mad proteins involved in the regulation of several other cellular processes
(Foley and Eisenman, 1999). The cytoplasmic nuclear factor of activated T-cells
(Nfatc2) regulates cell cycle progression during lymphocyte activation and may act as an inhibitor of cell proliferation in normal cells (Caetano et al., 2002). However, its gene expression was found to be down-regulated during U18666A treatment. The nuclear factor erythroid derived 2 like 2 (Nfe212), also known as Nrf2, plays a critical role in protecting neurons from oxidative stress and increased concentrations of intracellular calcium (Lee et al., 2003). Up-regulation of the gene after 72 h of
U18666A treatment suggested neuroprotection mediated by Nfe212 against neuronal apoptosis induced by reactive oxygen species (ROS). Sequestosome 1 (Sqstm1), also known as Osi or A170, possesses transcription cofactor activity and plays a role in oxidative stress-induced apoptosis (Ishii et al., 1996). The high transcript levels of
Sqstm1 from 24-72 h of U18666A treatment might indicate its contribution to
U18666A-mediated neuronal apoptosis.
193 The gene categories of main interest in the study of U18666A in the inhibition of intracellular cholesterol transport are those encoding for proteins involved in cholesterol biosynthesis and lipid metabolism and transport. Almost all of the genes grouped under cholesterol biosynthesis were down-regulated after 72 h of U18666A treatment, supporting the notion of inhibition of cholesterol biosynthesis by
U18666A. Farnesyl diphosphate synthetase (Fdps), isopentenyl-diphosphate δ isomerase (Idi1), lanosterol synthase (Lss), mevalonate kinase (Mvk), NAD(P) dependent steroid dehydrogenase-like (Nsdhl) and sterol-C5-desaturase homolog
(Sc5d) are involved in the biosynthesis of steroids. The 3-hydroxy-3-methylglutaryl- coenzyme A lyase (Hmgcl), which is involved in the synthesis and degradation of ketone bodies, remained up-regulated from 24-72 h of U18666A treatment, whereas
3-hydroxy-3-methylglutaryl-Coenzyme A reductase (Hmgcr), which is localized both in the brain peroxisomes and ER and plays a functional role in the peroxisomal isoprenoid pathway in the central nervous system (Kovacs et al., 2001), was down- regulated throughout. On the other hand, most of the genes grouped under lipid metabolism and transport were up-regulated after 72 h of U18666A treatment. The gene encoding for apolipoprotein E (Apoe) was significantly up-regulated from 24-72 h of U18666A treatment. The apoE4 isoform of apolipoprotein E is the major genetic risk factor of AD (Dolev and Michaelson, 2004) and has been found to potentiate Aβ peptide-induced apoptosis in neuronal cells (Ji et al., 2002). Studies have also shown that both apolipoprotein E and clusterin are present at similar concentrations in the central nervous system and may together influence Aβ deposition (DeMattos et al.,
2002). Interestingly, the gene encoding for clusterin (Clu) involved in cell death was
194 also significantly up-regulated from 24-72 h of U18666A treatment. Findings show increased clusterin expression in aged mice as an evidence for pathological brain aging (Shapiro et al., 2004). Clusterin is critical for neuritic toxicity in a mouse model of AD (DeMattos et al., 2002), but presents an anti-apoptosis effect when overexpressed in a mouse neuroblastoma cell line (You et al., 2003). Although studies have reported that Aβ protein up-regulates stearoyl-coenzyme A desaturase 1 in macrophages (Uryu et al., 2003), its gene (Scd1) was down-regulated from 48 h of
U18666A treatment in cortical neurons. Previous data show that expression of sterol carrier protein 2 (Scp2) increases fatty acid uptake and targets fatty acids to unique lipid pools in the regulation of hepatic lipid metabolism (Murphy, 2002). The significant up-regulation of Scp2 from 24-72 h of U18666A treatment suggests that the protein may serve as a universal fatty acid-binding and trafficking protein.
Other genes encoding for proteins involved in cell death include the growth arrest specific 2 (Gas2) and the transformed mouse 3T3 cell double minute 4 (Mdm4).
Gas2, up-regulated in a time-dependent manner from 24-72 h of U18666A treatment, can increase the activity of calpain and induce degradation of β-catenin (Benetti et al.,
2005). In contrast, Mdm4, which regulates p53 activity and cell growth (Marine and
Jochemsen, 2005), was down-regulated during exposure to U18666A. The SH3- domain GRB2-like B1 (Sh3glb1), also known as endophilin, can lead to an induction of apoptosis and was up-regulated upon U18666A treatment. In addition, the up- regulation of BCL2-associated transcription factor 1 (Bclaf1) and BCL2-like 11
195 (Bcl2l11), which is a facilitator of apoptosis, further confirms that U18666A- mediated cell death is via apoptosis.
Induction of apoptotic cell death in U18666A-treated cortical neurons led to the differential expression of genes encoding proteins involved in defense and stress responses. The chemokine receptor 4 (Cxcr4), which is substantially up-regulated from 48 h of U18666A treatment, has been found to be critical to the progression of diverse brain malignances (Rubin et al., 2003). Glutathione peroxidase 4 (Gpx4), glutathione S-transferase α4 (Gsta4), glutathione S-transferase μ5 (Gstm5) and microsomal glutathione S-transferase 1 (Mgst1) were up-regulated from 24-72 h of
U18666A treatment. The enzyme Gpx4 is known to play a protective role in the mechanism of oxidative stress-induced apoptosis that occurs through oxidative damage of mitochondrial phospholipids (Ran et al., 2004). Both Gsta4 and Gstm5 possess glutathione transferase activities in the glutathione metabolism pathway.
However, the mitochondrial Gsta4 has been found to increase in expression under oxidative stress (Raza et al., 2002). Mgst1 has been reported to display an important protective function against oxidative insult in the mouse retinal pigment epithelium and may be involved in the development of age-related diseases (Maeda et al., 2005).
The expression of DNA-damage inducible transcript 3 (Ddit3), also known as CHOP or GADD 153, is controlled by mitochondrial ROS. Studies have shown that Ddit3 induces ER stress-mediated apoptosis, which may contribute to brain ischemia and neurodegenerative diseases (Marciniak et al., 2004). The expression of Ddit3 is significantly up-regulated from 48 h of U18666A treatment, suggesting that
196 U18666A might lead to oxidative stress which further contributes to ER stress- mediated apoptosis in the treated cortical neurons. Immediate early response 2 (Ier2), also known as pip92, is known to induce cell death in primary neurons and display several hallmarks of pro-apoptotic activity upon overexpression (Schneider et al.,
2004). The up-regulation of Ier2, especially after 48 h of U18666A treatment, might promote apoptotic cell death. High metallothionein 1 (Mt1), which was up-regulated from 24-72 h of U18666A treatment, has a harmful role in aging and may cause degenerative diseases (Halliwell and Gutteridge, 1999). Peroxiredoxin 6 (Prdx6) was considerably up-regulated from 48 h of U18666A treatment. Prdx6 is a key player in removing ROS to maintain cellular homeostasis (Halliwell and Gutteridge, 1999).
Some of the differentially-expressed genes encoding for proteins with enzymatic activity may play a neuroprotective role in U18666A-mediated neuronal apoptosis.
The mitochondrial aldehyde dehydrogenase 2 (Aldh2) functions as a protector against oxidative stress (Ohsawa et al., 2003) and up-regulation of the gene during U18666A treatment could represent a selective cellular response put up by the cortical neurons to reduce oxidative burden. Similarly, stimulation of paraoxonase 2 (Pon2) during
U18666A treatment might represent a compensatory mechanism against the increase in cellular ROS production, as Pon2 has been shown to protect cells against oxidative stress (Rosenblat et al., 2003).
Microarray data have indicated a time-dependent increase in the DNA damage response genes, growth arrest and DNA-damage-inducible 45α (Gadd45a) and 45γ
(Gadd45g), after U18666A treatment. Gadd45 proteins are important regulators for
197 cell cycle arrest, apoptosis and DNA stability (Mak and Kültz, 2004). DNA damage has previously been postulated to be a consequence of U18666A treatment in primary cortical neurons (Chapter 3 Parts II and III). The FBJ osteosarcoma oncogene (Fos), which regulates neuronal excitability and survival (Zhang et al., 2002), was significantly up-regulated from 48 h of U18666A treatment, suggesting the contribution of Fos to U18666A-mediated neuronal apoptosis. Jun proto-oncogene related gene d1 (Jund1) was up-regulated from 24-72 h of U18666A treatment. Jund1 has been proposed to protect cells from p53-dependent senescence and apoptosis
(Weitzman et al., 2000). The level of phosphorylated p53 was previously found to increase upon exposure to U18666A (Chapter 3 Part III), probably in response to
DNA damage.
The proteasome is a multicatalytic proteinase complex involved in selective degradation of cellular proteins (Ciechanover, 1998; Hochstrasser, 1996). The significant up-regulation of genes encoding for proteins involved in the ubiquitin- proteasome system upon exposure to U18666A suggests that many cellular proteins might be targeted for proteasomal degradation in U18666A-treated cortical neurons.
The covalent modification of a protein substrate by ubiquitin is a critical factor in determining its degradation by the proteasome. Differential genes encoding for ubiquitinating enzymes were expressed in opposite levels after U18666A treatment, which might be due to redundant functions served by the different enzymes in the ubiquitination of a particular substrate (Hochstrasser, 1996). However, proteasome activity was found to decrease after 72 h of U18666A treatment, which has been
198 discussed earlier in Chapter 3 Part II. Calpains are calcium-dependent cysteine proteases which can contribute to neuronal apoptosis (Goll et al., 2003). Calpain 2
(Capn2) was up-regulated from 24-72 h of U18666A treatment. In contrast, calpain 3
(Capn3) was down-regulated, which might be due to its predominant expression in skeletal muscle instead. Cathepsin B (Ctsb), which was up-regulated in gene expression from 24-72 h of U18666A treatment, is an abundant and ubiquitously expressed cysteine peptidase. Intracellular cathepsin B is localized in the lysosome and is partly responsible for terminal degradation of intracellular proteins. It has also been implicated in the lysosomal pathway of apoptosis in tissue injury, inflammation and fibrogenesis, and has recently been shown to have an essential role in neuronal cell death mediated by Aβ-activated inflammatory response (Gan et al., 2004).
Although many signaling pathways are likely to modulate tau phosphorylation in vivo, little is known about the specific enzymes involved. Among the genes encoding for proteins involved in protein amino acid modification, most code for protein kinases and phosphatases. Identification of the kinases and phosphatases which might play a role in modulating the tau hyperphosphorylation observed in U18666A-treated cortical neurons (Chapter 3 Part III) may reveal the normal and pathological regulation of tau function in U18666A-mediated neuronal apoptosis. None of the kinases studied was activated in U18666A-treated cortical neurons (Chapter 3 Part
III). The abnormal phosphorylation of tau after U18666A treatment may thus arise from a decrease in the activity of a tau phosphatase, instead of from the activation of a kinase. Other kinases, besides those that have been studied, might also be involved.
199 Indeed, microarray data indicated the down-regulation of genes encoding for some phosphatases, such as dual specificity phosphatase 6 (Dusp6), protein phosphatase 1 regulatory subunit 7 (Ppp1r7) and protein tyrosine phosphatase receptor subtypes
(Ptpre, Ptprj and Ptprk), and the up-regulation of genes encoding for other kinases, such as MAP kinase-activated protein kinase 2 (Mapkapk2), mitogen activated protein kinase 3 (Mapk3) and protein kinase Cθ (Prkcq). The down-regulation of an upstream mitogen activated protein kinase kinase kinase 11 (Map3k11) might also have an effect on the activation via phosphorylation of its downstream MAPK kinase and MAPK in that signaling cascade. Further work will be necessary to find out which kinase or phosphatase is responsible for U18666A-induced tau phosphorylation. On another note, the down-regulation of protein kinase Cδ (Prkcd) and the up-regulation of protein phosphatase 4 catalytic subunit (Ppp4c) in U18666A- treated cortical neurons were not surprising, as activation of Prkcd has been reported to attenuate cell death of cortical neurons (Brodie and Blumberg, 2003) whereas
Ppp4c has a role in apoptosis (Mourtada-Maarabouni et al., 2003).
The inherent network architecture of the cytoskeleton enables protein-protein and kinase-substrate interactions to regulate and respond to diverse cellular events, including cell migration and outgrowth, mitosis and apoptosis, as well as contraction, vesicular transport and endocytosis. The cytoskeleton serves an integral function in cell-cell communication, in addition to cell adhesion and extracellular matrix signaling, that is critical for the maintenance of cell and tissue integrity (Alberts et al.,
2002). Majority of the genes encoding for proteins involved in cytoskeleton
200 organization, such as those encoding for tubulin isotypes, were up-regulated from 24 h of U18666A treatment. Tubulin constitutes a structural component of the cytoskeleton and aids in protein polymerization and microtubule-based movement and process. The α- and β-tubulins are the major components of microtubules and undergo a variety of post-translational modifications, while the γ-tubulin plays a major role in the nucleation of microtubule assembly (Lewis et al., 1985). In contrast, genes encoding for actin isotypes were generally down-regulated during U18666A treatment, such as actin-α2 (Acta2) and actin-β cytoplasmic (Actb), except actin-γ cytoplasmic 1 (Actg1), which was up-regulated instead. Actin is also a constituent of the cytoskeleton and participates in cell communication. The kinesin group of family members, up-regulated during U18666A treatment, possesses microtubule motor activity and participates in the intracellular transport of organelles and protein complexes (Miki et al., 2001). Microtubule-associated protein tau (Mapt) was significantly up-regulated after 24 h and 48 h of U18666A treatment. Tau has been reported to be essential to Aβ-induced neurotoxicity (Rapoport et al., 2002). Excess tau can compete with motor-proteins for binding to the microtubules, resulting in a rigid microtubular system which inhibits axonal transport. The gene encoding for the microtubule-destabilizing protein, stathmin 1 (Stmn1), was up-regulated in U18666A- treated cortical neurons. The possibility of stathmin as one of the contributing factors in U18666A-mediated neuronal apoptosis suggests that U18666A may lead to microtubule depolymerization and cytoskeleton disruption.
201 Among the genes encoding for proteins involved in cell adhesion, follistatin (Fst) was significantly down-regulated in U18666A-treated cortical neurons. However, its significance in U18666A-mediated neuronal apoptosis remains unresolved as the physiological importance of Fst has been reported to be a critical regulator of activin, which is a TGFβ superfamily ligand (Schneyer et al., 2004). Calnexin is a molecular chaperone which plays an essential role in the correct folding of membrane proteins in the ER (Bergeron et al., 1994). The down-regulation of calnexin (Canx) from 24-
72 h of U18666A treatment suggested that unfolded proteins might accumulate in the
ER in the absence of the chaperone to support proper protein folding. Prolonged aggregation of unfolded proteins may lead to ER stress, which may eventually contribute to cell death induced by U18666A. In addition, the gene encoding for the cytosolic chaperonin subunit 3γ (Cct3), which mediates protein folding (Kubota et al.,
1999), was also down-regulated from 24-72 h of U18666A treatment. ERO1-like
(Ero1l) gene, significantly down-regulated during U18666A treatment, encodes a protein similar to a protein required for oxidative protein folding in Saccharomyces cerevisiae (Frand and Kaiser, 1998). A similar protein found in humans is an integral membrane protein in the ER and postulated to be involved in oxidative ER protein folding. Peptidylprolyl isomerases A (Ppia) and B (Ppib) and protein NIMA- interacting 1 (Pin1) were up-regulated from 24-72 h of U18666A treatment.
Peptidylprolyl isomerases, also known as cyclophilins, are ubiquitous enzymes responsible for proline isomerization during protein synthesis and for the chaperoning of several membrane proteins in protein folding (Galat, 1993). Studies have shown that Ppia is secreted from smooth muscle cells and macrophages in response to
202 oxidative stress, and may play an important role in the pathogenesis of inflammatory diseases (Jin et al., 2000). The expression of Pin1 has been found to inversely correlate with neurofibrillary degeneration in AD (Liou et al., 2003). The up- regulation of Pin1 following U18666A treatment might act as protection against neuronal demise.
Genes encoding for proteins involved in transport, including ion transport and electron transport, were also differentially expressed during U18666A treatment.
Synaptotagmins 4 (Syt4) and 11 (Syt11) were up-regulated from 24-72 h of treatment. Proteins encoded by these genes, primarily expressed in the nervous tissues, are integral membrane proteins of synaptic vesicles with multi-domains and are thought to serve as calcium sensors in the process of vesicular trafficking and exocytosis (Yoshihara and Montana, 2004). These indicate that U18666A might increase trafficking of vesicles and synaptic transmission. Secretogranin II (Scg2) was also up-regulated from 24-72 h of treatment. Secretogranins contribute to the formation of secretory granules and function as chaperones in the sorting of other regulated secretory proteins (Taupenot et al., 2003). The sorting and intracellular trafficking of several proteins might be impaired in U18666A-treated cortical neurons, as U18666A is known to alter the localization of intracellular components
(Chapter 1). Electron transport in the respiratory chain of the mammalian mitochondria is regulated by the cytochrome c oxidase. U18666A treatment resulted in an increase in gene expression of most subunits of cytochrome c oxidase. This enzyme has been postulated to play a role in oxidative stress-induced apoptosis and
203 neurodegenerative diseases (Kadenbach et al., 2004). Thioredoxins are abundant proteins found in several isoforms that carry out essential biosynthetic reactions and regulate many biological functions. Apart from the redox regulation of protein thiols involved in signal transduction and gene regulation, thioredoxins may have a neuroprotective role in the central nervous system, as they are able to reduce disulfide bonds in proteins (Patenaude et al., 2005). The thioredoxin-like 1 (Txnl1) was up- regulated after 24 h and 48 h of U18666A exposure but down-regulated after 72 h of treatment, suggesting that Txnl1 might exert neuroprotection during the initial response of cortical neurons to U18666A but was unable to salvage cell death after 72 h of treatment.
Other differentially-expressed genes which might be implicated in U18666A- mediated neuronal apoptosis include those encoding for ectonucleotide pyrophosphatase/phosphodiesterase 2 (Enpp2), prion protein (Prnp), synapsin I
(Syn1), α-synuclein (Snca) and amyloid β precursor protein binding protein 1
(Appbp1), which were generally up-regulated during U18666A treatment. Findings suggest that Enpp2, also known as autotoxin, helps to prevent apoptosis through the
G-protein coupled receptor protein signaling pathway, leading to enhanced cell survival (Song et al., 2005). Prnp is widely expressed in most tissues and is usually rapidly degraded in the cytosol. However, when proteasome activity is compromised due to stress and aging, the increase in cytosolic Prnp protein can be toxic to the neuron. Accumulation of small amounts of cytosolic Prnp has been found to be strongly neurotoxic and is associated with neurodegeneration in cultured cells and
204 transgenic mice (Ma et al., 2002). Syn1, which may interact with the cellular prion protein, has been reported to regulate membrane fusion kinetics and recruit synaptic vesicles to a reserve pool (Gitler et al., 2004). Studies show that Snca may protect against oxidative stress in neuronal cells (Hashimoto et al., 2002a) and has a critical role in the fibrillar nature of ubiquitinated inclusions induced by proteasomal inhibition in primary neurons (Rideout et al., 2004). The amyloid precursor protein
(APP) is a membrane protein which contains a large extracellular region, a transmembrane helix and a short cytoplasmic tail (Selkoe, 2000). Toxic Aβ implicated in AD can originate from intramembrane proteolysis of the APP by a complex of secretases. Recently, Serretti et al. (2005) reviewed candidate genes which can be considered as susceptibility factors for AD. Some of the reported genes were found to be differentially expressed in the present study, suggesting that
U18666A-treated cortical neurons may act as a novel in vitro model system to elucidate the molecular mechanism of AD.
While gene profiles provide only the levels of mRNA messages and cannot be presumed to reflect functional significance at the protein level, the differential regulation of such a large number of genes upon U18666A treatment indicates the great response elicited by U18666A in primary cortical neurons and is insightful in understanding the mechanism of the apoptotic cell death. Additional work on protein expression specific to the genes identified may help to elucidate the exact effects of the drug. Over one hundred possible coding sequences which were differentially expressed after U18666A treatment have no matches in the current public databases
205 and hence remain classified as novel sequences with unknown functions.
Computational analysis of these sequences can allow annotation of putative gene functions through similarity searches in nucleic acid and protein databases.
Determination of the biological functions of these genes will be among the greatest challenge for further research into the mechanism of U18666A-mediated neuronal apoptosis.
206 CHAPTER 5
A PROTEOMICS APPROACH TO THE STUDY OF U18666A-MEDIATED NEURONAL APOPTOSIS
207 5 A Proteomics Approach to the Study of U18666A-Mediated
Neuronal Apoptosis
5.1 Introduction
The central dogma, “DNA is transcribed to RNA which is translated to protein”, is the cornerstone of molecular and cell biology (Lewin, 1994) and has now developed into genomics and proteomics (Figure 5.1). While genes are the support for the transmission of genetic information, most of the chemical reactions inside cells are carried out by proteins, which are the functional products of gene expression (Horton et al., 1996). The one gene-one protein hypothesis is oversimplified, as differential splicing of the pre-messenger RNA (mRNA) from one gene can actually produce many proteins (Lewin, 1994).
Figure 5.1: The central dogma.
208 The conversion of mRNA to its protein products generates the corresponding proteome. This term, which has since been universally adopted, was first coined in
1994 by Marc Wilkins in his work for a doctoral degree to describe the entire protein complement expressed by a genome, cell or tissue at a single time-point (Wasinger et al., 1995; Huber, 2003). Unlike the genome, which is essentially the same in all the somatic cells of an organism, the proteome is a dynamic entity that not only differs in different cell types but also changes with the physiological condition of the living cell
(Huber, 2003; Pandey and Mann, 2000; Dove, 1999; Williams, 1999; Humphery-
Smith and Blackstock, 1997). To add further complexity, proteins interact with one another or with other biomolecules (Horton et al., 1996). While these may complicate the complete analysis of any proteome, proteomics has become increasingly popular in the study of quantitative changes in global differential protein expression
(Blackstock and Weir, 1999; Phizicky et al., 2003; Tyers and Mann, 2003). A key element of proteomics is the capability to identify protein post-translational modifications, such as phosphorylations and glycosylations, which can affect biological functions of the protein (Pandey and Mann, 2000; Dove, 1999). Moreover, proteomics can be used to study protein structures, functions and localizations (Tyers and Mann, 2003; Graves and Haystead, 2002). Proteomics is used in disease expression profiling to better understand disease processes, identify disease biomarkers and develop effective therapy (Hanash, 2003; Jain, 2004). Proteomics is suitable for the study of brain tissues (Collins et al., 2004) and can be used for the discovery of drug targets to treat neurological disorders (Jain, 2004). Particularly, the application of proteomics analysis in the study of proteins thought to play a role in
209 AD has revealed new understandings of the pathology and physiology of this neurodegenerative disease (Butterfield et al., 2003; Ho et al., 2005). The proteomics approach is also widely used to elucidate global protein expression profiles and cellular signaling pathways in several neurobiology studies. A map of human brain proteins has been constructed as a reference database to investigate changes in the protein level caused by various neurological disorders (Langen et al., 1999). In another study, a database of proteins present in the porcine cerebellum has been generated (Friso and Wikström, 1999). Proteomics was applied to the tau transgenic mouse to find disturbances in the levels of expression of brain proteins (Tilleman et al., 2002). Additional studies have been performed to provide reliable databases of proteins in mice, among which one reported protein profiles of different anatomic areas of the mouse brain during aging (Tsugita et al., 2000). The platelet-derived growth factor signal transduction pathway in cultured neurons and astrocytes has also been studied using a similar approach (Zhang and Hutchins, 1997).
Proteomics is at present dominated by two technologies. Two-dimensional (2D) gel electrophoresis, introduced in 1975 for the mapping of proteins from Escherichia coli
(O’Farrell, 1975), mouse (Klose, 1975) and guinea pig (Scheele, 1975), is now often used to separate complex protein samples from cells or tissues. A protein is separated on the basis of its isoelectric point (pI) in the first dimension and molecular weight in the second dimension. From an analytical point of view, 2D gel electrophoresis is unmatched in its ability to simultaneously resolve hundreds of cellular proteins that would not normally be separated by a one-dimensional method (Graves and
210 Haystead, 2002). The resolution, loading capacity and reproducibility of this technique have significantly improved with the development of immobilized pH gradient (IPG) strips, which have contributed to the standardization of 2D gel electrophoresis and made possible the reliable comparison of 2D maps (Bjellqvist et al., 1993; Görg et al., 2000). In addition, the introduction of integrated systems for the first-dimension isoelectric focusing (IEF) has greatly simplified the process of sample loading and reduced the handling of the strips prior to the IEF.
Another major tool in proteomics is mass spectrometry, followed by database searching, to identify differentially-expressed proteins of interest from the 2D gels
(Shevchenko et al., 1996; Jonsson, 2001; Graves and Haystead, 2002). Coupled with improvements in mass spectrometry and the ever-growing protein databases, powerful tools have recently become available that make analyses of whole proteomes feasible (Aebersold and Mann, 2003; Fenyö, 2000; Henzel et al., 1993).
The most commonly used ionization method is the matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (Jonsson, 2001; Griffin et al., 2001; Shevchenko et al., 2000). The masses of a peptide mixture from a protein spot on the 2D gel after proteolytic digestion are measured by MALDI-TOF mass spectrometry to obtain a peptide mass fingerprint. Peptide mass fingerprinting is a sensitive and high-throughput method that has been extensively used for rapid screening and identification of proteins (Thiede et al., 2005; Cottrell, 1994). The basis of peptide mass fingerprinting is that the set of peptide masses obtained by the mass spectrometric analysis of a protein digested with a specific protease can provide a
211 characteristic profile or fingerprint of that protein (Cottrell, 1994). The discriminating signature is then searched against the peptide masses from a simulated digest of a known protein in a database for identification. Protein identification by peptide mass fingerprinting is a widely recognized technique for proteins from 2D gels (Thiede et al., 2005).
To further understand the mechanism of U18666A-mediated apoptosis, a proteomics approach was employed by utilizing 2D gel electrophoresis to profile the proteins induced by U18666A in primary cortical neurons. Changes in the 2D snapshots of protein expression were monitored from different time-points to obtain an overview, not only of a global but also of a temporal nature, of U18666A-induced protein expression changes in the cortical neurons. Proteome changes accompanying the progression of apoptosis from 24-72 h of U18666A treatment may reflect the mechanism of action of U18666A. The strategy for proteomics analysis is shown in
Figure 5.2. Analysis of the proteome can more clearly reflect the actual state of activity in the U18666A-treated cortical neurons. The identification of differentially- expressed proteins might provide an idea of the biochemical changes resulting from the induction of neuronal apoptosis upon U18666A treatment. Results from the proteomics analysis can be used to complement those from the microarray analysis
(Chapter 4). Both the gene and protein expression data may be used as an insight into the molecular effects of U18666A and its mechanism of action.
212 Figure 5.2: Strategy for proteomics analysis in U18666A-treated cortical neurons.
213 5.2 Results
5.2.1 Optimization of experimental conditions for 2D gel electrophoresis
As an appropriate sample preparation is one of the criteria for high-quality 2D gel results, the optimum lysis buffer constituents were first determined for primary cortical neurons. Cortical neurons treated with 1 μg/ml U18666A for 72 h were used for the optimization as earlier data indicated that cell viability of the cortical neurons is reduced by about 50% after this treatment condition (Chapter 3 Part I, Figure
3.2A). For initial experiments, lysis buffer for separation under denaturing conditions in the first dimension utilized urea as the main chaotropic agent to unfold proteins.
However, the abundance of the silver-stained protein spots observed were fewer than expected after solubilization in the buffer containing 8 M urea (Figure 5.3A), even though the maximum amount of proteins (150 μg) was loaded. Horizontal streaks and background smears were also observed in the higher molecular weight range (Figure
5.3A). To increase the number of protein spots visualized, thiourea, another chaotrope, was incorporated as another constituent of the lysis buffer to improve solubility of proteins (Rabilloud et al., 1997). As shown in Figure 5.3B, more protein spots were observed from an equal amount of proteins (150 μg) when 2 M thiourea was used in combination with 7 M urea, as compared to denaturation with 8 M urea alone. Unfortunately, the horizontal streaks remained and were somewhat intensified
(Figure 5.3B). A small amount of DNase I (50 μg/ml) then added into the samples decreased the viscosity of the protein lysate. More importantly, the problem of streaking was greatly reduced without affecting the overall recovery of protein spots
214 (Figure 5.3C). Although the addition of DNase I could improve visualization of the
2D gels, this enzyme may appear as a protein on the 2D map. The presence of RNA could also contribute to background smears after silver staining. Thus, the TRIZOL®- based method of sample preparation was next considered, as it can effectively separate cellular proteins from DNA and RNA contamination (Giorgianni and
Beranova-Giorgianni, 2005). Proteins from cortical neurons treated with vehicle
(Control) for 24 h were isolated using TRIZOL® reagent. One hundred μg of proteins
was applied via the rehydration loading method by including the sample in the
rehydration solution (Figure 5.4A). Albeit the protein profile appeared reproducible
and the combination of sample application and rehydration into one step can ease
operation, overnight rehydration with the sample may cause proteolysis or other
protein modifications which might inevitably affect downstream analysis. As a result,
sample application (100 μg of proteins) through the cup-loading method was
employed immediately prior to the IEF instead (Figure 5.4B). The proteins appeared
well-resolved and results were also reproducible.
5.2.2 Global protein expression profiles after U18666A treatment in primary
cortical neurons
Proteome changes accompanying the progression of apoptosis from 24-72 h of
U18666A treatment are of interest as these could reflect the mechanism of action of
the drug in primary cortical neurons. The TRIZOL®-based method was used to isolate proteins from cortical neurons treated with vehicle (Control) or 1 μg/ml U18666A
after every 24 h. Ten μg of proteins from each sample was first focused via the cup-
215 loading method on 7 cm long pH 3-10 Non-Linear IPG strips before separation in 8% polyacrylamide gels to serve as a rapid screening tool to confirm the protein profile and quality of the samples. Proteins were consistently well-resolved throughout 24-72 h of U18666A treatment (Figure 5.5). The same samples (100 μg of proteins) were therefore further focused via the cup-loading method on 18 cm long pH 3-10 Non-
Linear IPG strips before separation in 8-16% gradient gels to achieve a wider and larger separation with increased resolution. The global protein profiles of the vehicle- treated (Control) and U18666A-treated cortical neurons after every 24 h are as shown in Figure 5.6. Based on the results of the image analysis with PDQuest™ 2D analysis software, almost 2000 spots were visualized in each gradient gel by silver staining.
5.2.3 Differentially-expressed proteins after U18666A treatment in primary
cortical neurons
Only the protein spots whose quantities in the U18666A treatment are at least twice more than or less than half of the corresponding spots in the vehicle-treated Control were included in the subsequent PDQuest™ analysis (Figure 5.7A). Approximately
100 spots at each time-point were found to be differentially-expressed with at least 2- fold change in the U18666A-treated cortical neurons, as compared to the Control.
However, only well-separated and abundantly-expressed protein spots, as visually determined, were manually excised from the silver-stained gels and subjected to mass spectrometry analysis. Out of a total of 135 spots excised from the gels as represented by the synthetic Gaussian images shown in Figure 5.7B, good spectra were obtained for 34 spots. Proteins were identified with matched returns for 31 of these spots.
216 However, after taking into account the spots representing protein isoforms of one another (such as heat shock protein and heterogeneous nuclear ribonucleoprotein) or the same protein differentially-expressed at different time-points (such as peptidylprolyl cis-trans isomerase A and stathmin), the number of different protein species identified was only 25 in total (Table 5.1). Protein coverage was calculated based on the percentage of amino acid residues covered in a particular protein by the matched peptides. Three spots with good spectra were returned with unknown identities during the search and could potentially be novel proteins (Table 5.1). The identified protein spots, as well as those with unknown identities, are labeled on enlarged views of representative silver-stained gels for U18666A treatment to indicate their respective locations on each individual gel (Figure 5.8).
5.2.4 Correlation of proteomics changes to corresponding gene alterations in
U18666A-treated cortical neurons
The 25 proteins identified through mass spectrometry were compared with the genes differentially expressed after U18666A treatment from the microarray analysis
(Chapter 4, Table 4.1). Eleven of the proteins matched with their corresponding genes
(Table 5.2). Three were from the gene functional category representing protein folding (namely aryl-hydrocarbon receptor-interacting protein, chaperonin subunit 3 and peptidylprolyl isomerase A), and one from each of the categories representing cytoskeleton organization (stathmin 1), response to stress (peroxiredoxin 6), lipid metabolism and transport (apolipoprotein E), ubiquitin-proteasome system
(proteasome subunit α type 7), regulation of transcription (zinc finger protein 422),
217 RNA processing (heterogeneous nuclear ribonucleoprotein U), ion transport (tumor necrosis factor α-induced protein 1) and enzymatic activity (thioesterase superfamily member 2). From Table 5.2, apolipoprotein E precursor was found to be up-regulated in protein expression at 72 h whereas the corresponding gene for apolipoprotein E showed a time-dependent increase in expression from 24-72 h. Aryl-hydrocarbon interacting protein was up-regulated in protein expression at only 24 h while the corresponding gene for aryl-hydrocarbon receptor-interacting protein was up- regulated from 24-72 h. Brown fat inducible thioesterase was also up-regulated in protein expression at only 24 h. The gene representing a member of the thioesterase superfamily (thioesterase superfamily member 2) was significantly up-regulated at 24 h, and remained up-regulated at 48 h and 72 h. Heat shock protein, also known as 10 kDa chaperonin, showed a down-regulation in protein expression at 72 h while the gene for the chaperonin subunit 3 remained down-regulated from 24-72 h.
Heterogeneous nuclear ribonucleoprotein was also up-regulated in protein expression at only 24 h. On the contrary, the gene for heterogeneous nuclear ribonucleoprotein U was down-regulated from 24-72 h. Peptidylprolyl cis-trans isomerase A showed up- regulated protein expression at 24 h and 48 h, while its gene expression remained up- regulated from 24-72 h. Peroxiredoxin 6 protein expression was up-regulated at only
48 h whereas its gene increased in expression from 24-72 h. Proteasome subunit α type 6 showed an up-regulation in protein expression at 48 h. In contrast, the gene for proteasome subunit α type 7 decreased in expression in a time-dependent manner from 24-72 h. Stathmin was up-regulated in protein expression at 48 h and 72 h, while the gene for stathmin 1 showed an up-regulation in gene expression from 24-72
218 h. Tumor necrosis factor-inducible protein was only up-regulated in protein expression at 72 h whereas the gene representing tumor necrosis factor α-induced protein 1 was up-regulated in expression from 24-72 h. Zinc finger protein 624 was also up-regulated in protein expression at 72 h whereas the genes representing zinc finger proteins, in particular the zinc finger protein 422, showed a time-dependent increase in expression from 24-72 h.
219
g) from from g) μ g/ml U18666A for 72 μ homogenized and the protein C ved in C were due to scanner malfunction. ) DTT and (A) 8 M urea, (B) 7 urea 2 tions. Cortical neurons treated with 1 tions. Cortical neurons r rehydration loading, equal amounts of protein (150 equal amounts of protein r rehydration loading, g/ml DNase I. The lysates were The lysates g/ml DNase I. μ were then silver- S-PAGE on separate 12% polyacrylamide gels. Gels were then conditions of first-dimension IEF using IPG strips pH 3-10 Non-Linear of pH 3-10 Non-Linear of using IPG strips IEF of first-dimension conditions (w/v) CHAPS, 1% (w/v B e Methods, and scanned. Vertical bars obser ea, 2 M thiourea and 50 : Optimization of sample preparation condi : Optimization Figure 5.3 h were lysed in buffer comprising 4% thiourea, or (C) 7 M ur Afte in each sample was quantitated. concentration were subjected to the same each sample 18 cm length and second-dimension SD stained, as described in th A
220
treated with vehicle ing method before same conditions Methods. Equal amounts of protein h and second-dimension SDS-PAGE on in the Methods, and then scanned. ns from cortical neurons ns B reagent, as described in the ® ZOL were silver-stained, as described ps pH 3-10 Non-Linear of 18 cm lengt : Optimization of sample application method. Protei application method. : Optimization of sample g) were subjected to the (A) rehydr ation loading method or (B) cup-load μ Figure 5.4 (Control) for 24 h were isolated using TRI (100 of first-dimension IEF using IPG stri separate 12% polyacrylamide gels. Gels A
221
Figure 5.5: Screening of sample quality and protein profile. Proteins from cortical neurons treated with vehicle (Control) and 1 μg/ml U18666A were isolated after every 24 h using TRIZOL® reagent, as described in the Methods. Equal amounts of protein (10 μg) were subjected to first-dimension IEF using IPG strips pH 3-10 Non-Linear of 7 cm length via the cup-loading method and second-dimension SDS- PAGE on separate 8% polyacrylamide gels. Gels were silver-stained, as described in the Methods, and then scanned.
222
g/ml U18666A g/ml U18666A μ gel. Gels were then gel. Gels were then sample was focused sample was focused g of proteins in each μ s compared with that of 1 of with that s compared an 8-16% Tris-HCl gradient tal proteins from primary cortical neurons. The (C) 72 h. One hundred (Control) cortical neurons wa in the Methods, and scanned. h, (B) 48 h and : Representative silver-stained 2D gels of to silver-stained : Representative profile of the vehicle-treated on a pH 3-10 Non-Linear IPG strip before separation in silver-stained, as described Figure 5.6 treatment after (A) 24 A
223
Continued. : Figure 5.6 B
224
Continued. : Figure 5.6 C
225
r limit and (ii) below r limit and (ii) below the PDQuest™ analysis. (B) lly-expressed spots after (i) 24 h, (ii) ing to the PDQuest™ software. Red the PDQuest™ software. ing to ons. Protein spots eventually excised and g/ml) treatment are at least twice more than at least twice are g/ml) treatment μ howing the (i) above uppe showing differentia lay. (A) Histograms s is are marked with crosses accord is een represents down-regulated proteins. g/ml) treatment in primary cortical neur in primary g/ml) treatment μ sponding spots in the vehicle-treated Control are included in sponding spots in the vehicle-treated : Computer-aided differential disp i ii A B iii i ii Gaussian images from the PDQuest™ 2D analysis software the Gaussian images from 48 h and (iii) 72 of U18666A (1 analys mass spectrometry subjected to represents up-regulated proteins while gr Figure 5.7 (1 lower limit in which protein spots whose quantities the U18666A or less than half of the corre
226 Table 5.1: List of differentially-expressed proteins identified by mass spectrometry in cortical neurons treated with 1 μg/ml U18666A.
Theoretical Matched peptides/ Accession No. Protein Name Residues of identified peptides mass/pI Protein coverage
P01009 Alpha-1-antitrypsin precursor 46.7/5.4 7/23% 35-49, 180-187, 188-198, 284-305, 307-324, 335-352, 360-367 P08226 Apolipoprotein E precursor (Apo E) 35.8/5.6 8/24% 43-48, 87-100, 130-144, 202-214, 202-216, 226-236, 244- 250, 262-270 Q9NZN9 Aryl-hydrocarbon interacting protein 43.9/5.6 7/13% 1-13, 2-13, 2-14, 14-32, 15-32, 266-272, 290-302 Q9DCX2 ATP synthase D chain, mitochondrial 18.6/5.5 6/47% 9-24, 32-40, 41-57, 123-143, 148-160, 149-160 Q8VHQ9 Brown fat inducible thioesterase (BFIT) 67.3/6.3 8/14% 1-11, 64-73, 174-182, 185-194, 315-329, 336-348, 403- 415, 537-543 Q9JLQ0 CD2-associated protein 70.4/6.1 7/13% 65-71, 292-310, 398-411, 412-425, 426-436, 437-446, 578-585 Q04447 Creatine kinase B-type 42.7/5.4 12/40% 12-32, 14-32, 33-43, 87-96, 108-130, 157-172, 224-236, 321-341, 342-358, 342-358, 359-366, 367-381 O60869 Endothelial differentiation-related factor 1 16.4/10.0 8/47% 16-25, 44-56, 51-60, 64-72, 64-79, 73-85, 89-98, 136-146 Q91VW5 Golgi autoantigen 57.4/5.3 24/9% 104-111, 158-165, 353-361, 373-382, 455-462, 524-532, 526-537, 739-749, 809-819, 881-888, 1065-1074, 1066- 1074, 1186-1198, 1477-1485, 1528-1536, 1571-1580, 1595-1604, 1634-1643, 1688-1698, 1753-1759, 1867- 1876, 1869-1876, 2020-2033, 2191-2199 Q64433 Heat shock protein (10 kDa chaperonin) 10.8/8.2 7/65% 8-19, 36-53, 40-53, 54-65, 66-79, 70-79, 92-101 O88569 Heterogeneous nuclear ribonucleoprotein 36.0/8.7 13/40% 10-26, 11-26, 27-34, 102-117, 109-125, 118-125, 118- 135, 141-156, 142-156, 162-173, 192-201, 202-216, 314- Q9JHR7 Insulin-degrading enzyme 117.7/6.1 9/10% 37-49, 75-85, 300-308, 312-327, 352-364, 471-477, 659- 668, 827-839, 1001-1018 NP_034825 Lipocalin 4 21.4/5.5 7/29% 46-59, 47-59, 72-86, 78-86, 78-89, 128-137, 157-169 NP_619611 3-mercaptopyruvate sulfurtransferase (MST) 32.9/6.1 8/30% 30-39, 52-63, 112-117, 118-132, 164-175, 188-196, 266- 281, 282-292 Q01768 Nucleoside diphosphate kinase B 17.4/7.0 4/31% 7-18, 19-27, 57-66, 89-105 Q9NX02 Nucleotide-binding site protein 1 120.4/5.7 12/12% 79-87, 191-199, 265-281, 311-324, 343-355, 437-444, 449-467, 449-472, 520-525, 627-645, 646-652, 649-659 P17742 Peptidylprolyl cis-trans isomerase A 17.8/7.9 7/35% 1-18, 19-27, 19-30, 76-90, 82-90, 131-143, 133-143 O08709 Peroxiredoxin 6 24.7/5.7 10/48% 1-21, 24-40, 67-83, 97-105, 97-107, 132-140, 132-141, 142-154, 144-154, 162-181 227 Q6PCE3 Phosphoglucomutase 2-like 1 70.4/6.8 11/16% 98-107, 151-165, 166-178, 166-179, 205-228, 239-244, 239-245, 381-392, 381-395, 419-427, 569-577 Q9QUM9 Proteasome subunit alpha type 6 27.4/6.3 7/28% 12-21, 22-30, 31-43, 60-71, 94-102, 229-245, 229-246 P54227 Stathmin 17.1/5.8 9/48% 13-26, 14-26, 27-40, 29-40, 43-51, 52-59, 76-84, 85-94, 126-133 P40142 Transketolase (TK) 67.6/7.2 8/22% 76-101, 175-186, 187-204, 265-281, 284-302, 472-493, 531-542, 604-617 P48759 Tumor necrosis factor-inducible protein 41.8/5.3 5/11% 131-140, 222-230, 253-266, 256-266, 350-360 Q9JM98 WD-repeat protein 8 52.0/6.8 7/20% 1-15, 134-146, 144-153, 158-172, 277-295, 364-377, 449- 461 Q9P2J8 Zinc finger protein 624 85.5/9.3 8/12% 1-12, 15-28, 98-119, 147-157, 329-338, 386-394, 527- 534, 721-730 N/A Hypothetical protein [Homo sapiens] 27.3/7.8 6/20% 72-82, 73-82, 158-164, 167-181, 182-190, 208-215 N/A Unnamed protein product [Mus musculus] 75.3/9.4 19/28% 21-34, 22-35, 97-112, 132-138, 218-229, 218-232, 238- 253, 250-266, 267-283, 284-300, 305-317, 339-351, 340- 351, 374-387, 389-404, 405-418, 406-418, 467-473, 616- 626 N/A Unnamed protein product [Mus musculus] 11.0/11.0 6/51% 4-20, 12-20, 44-56, 46-57, 59-70, 81-92
Accession numbers and the theoretical molecular mass and pI values of the proteins were derived from SWISS-PROT and NCBI non- redundant databases. Protein coverage is based on the percentage of amino acid residues covered in a particular protein by the matched peptides. All accession numbers are correct as on 7 December 2005.
228
A
Figure 5.8: Enlarged views of 2D gel profiles for 1 μg/ml U18666A treatment from Figure 5.6. The differentially-expressed protein spots identified by mass spectrometry (Table 5.1) after (A) 24 h, (B) 48 h and (C) 72 h of U18666A treatment are labeled. Ç represents up-regulated proteins while È represents down-regulated proteins.
229
B
Figure 5.8: Continued.
230
C
Figure 5.8: Continued.
231 Table 5.2: List of identified proteins with matching corresponding genes from microarray analysis in cortical neurons treated with 1 μg/ml U18666A.
Protein Name 24 h 48 h 72 h Gene Title 24 h 48 h 72 h
Cell death Tumor necrosis factor-inducible protein No change No change Up tumor necrosis factor alpha-induced protein 1 1.81 2.55 2.14
Cytoskeleton organization Stathmin No change Up Up stathmin 1 2.28 1.88 1.61
Lipid metabolism and transport Apolipoprotein E precursor (Apo E) No change No change Up apolipoprotein E 1.64 2.02 3.32 Brown fat inducible thioesterase (BFIT) Up No change No change thioesterase superfamily member 2 2.77 1.96 1.62
Protein folding Aryl-hydrocarbon interacting protein Up No change No change aryl-hydrocarbon receptor-interacting protein 1.76 1.99 1.52 Heat shock protein (10 kDa chaperonin) No change No change Down chaperonin subunit 3 (gamma) -2.47 -1.92 -2.26 Peptidylprolyl cis-trans isomerase A Up Up No change peptidylprolyl isomerase A 2.43 2.07 1.63 Proteasome subunit alpha type 6 No change Up No change proteasome subunit alpha type 7 2.01 1.55 1.13
Regulation of transcription Heterogeneous nuclear ribonucleoprotein Up No change No change heterogeneous nuclear ribonucleoprotein U -1.60 -3.00 -2.33 Zinc finger protein 624 No change No change Up zinc finger protein 422 1.41 2.00 2.47
Response to oxidative stress Peroxiredoxin 6 No change Up No change peroxiredoxin 6 1.27 1.76 2.67
Protein spots were indicated “Up” for up-regulation, “Down” for down-regulation or “No change” based on their normalized optical densities derived from the PDQuest™ software analysis. Genes with their respective mean fold differences were from the microarray analysis in Chapter 4 (Table 4.1).
232 5.3 Discussion
It has only been a decade since the concept of the proteome was first introduced
(Wasinger et al., 1995; Huber, 2003). However, in these ten years, proteomics has developed into a powerful approach for the exploitation of the increasing gene expression data for a wide variety of organisms (Blackstock and Weir, 1999; Phizicky et al., 2003; Tyers and Mann, 2003). In the post-genomics era, proteomics is now providing new insights into how cells, tissues and even whole organisms function at the molecular level (Pandey and Mann, 2000; Dove, 1999). Proteomics is an interdisciplinary science that includes biology, bioinformatics and protein chemistry.
It can be defined as the large-scale study of protein properties, such as their expression levels, post-translational modifications and interactions with other molecules, to obtain a global view of cellular processes at the protein level (Huber,
2003; Graves and Haystead, 2002). With its unique ability to simultaneously separate several hundred proteins, 2D gel electrophoresis using IPG in the first dimension remains to be the core technique of proteomics (O’Farrell, 1975). Mass spectrometry is also now accepted as the standard approach for the identification and characterization of proteins in proteomics projects (Aebersold and Mann, 2003;
Jonsson, 2001).
The major problems concerning the visualization of total proteins stem from the great diversity of proteins and the high dynamic range of protein expression. Although a large number of standard protocols have been published, there is no single method of
233 sample preparation that can be universally applied to all kinds of samples analyzed by
2D gel electrophoresis. These protocols have to be adapted and further optimized for the type of sample, as well as the proteins of interest to be analyzed (Herbert, 1999;
Shaw and Riederer, 2003). The most commonly used lysis buffer is based on
O’Farrell’s protocol (O’Farrell, 1975). However, this standard buffer is not ideal for the solubilization of many proteins, particularly membrane or other hydrophobic proteins. Few studies have reported success for the recovery of such proteins (Santoni et al., 1999; Friso and Wikström, 1999; Molloy, 2000), although the addition of thiourea has been shown to increase the solubility of hydrophobic proteins and have a positive effect on the number of visualized protein spots on 2D maps (Rabilloud et al., 1997). The TRIZOL®-based method of protein extraction can effectively remove
nucleic acids contamination, which may contribute to horizontal streaks and
background smears seen in 2D gels after silver staining (Shaw and Riederer, 2003).
Proteins isolated with TRIZOL® have previously been used in the proteomics analysis
of human tissues (Giorgianni and Beranova-Giorgianni, 2005) and more extensively
in Western blot analysis (Harvell et al., 2006; Psyrri et al., 2004; Vreugdenhil et al.,
2003; Eltom et al., 1999). On the other hand, there are many steps involved with the
use of this method. The more procedures there are during sample preparation, the
higher the risk of protein loss. A one-step procedure for protein extraction would thus
be more desirable with regard to simplicity and reproducibility (Herbert, 1999; Shaw
and Riederer, 2003). Protein modifications during sample preparation might cause
artifactual spots on the 2D gels. Samples can be dialyzed or pre-fractionated using
commercially available kits to obtain less complicated 2D gels for easier downstream
234 analysis (Chapman, 2005; Shaw and Riederer, 2003). Fractionation may also provide information on protein localization, which serves as an additional advantage.
The level and activity of a protein are controlled at several post-transcriptional steps and by several post-translational modifications (Horton et al., 1996). As the expression of proteins with post-translational modifications cannot be studied at the transcriptional level using gene expression studies, proteomics was employed to detect the changes of the proteome during U18666A-mediated cell death in primary cortical neurons. One of the major advantages of 2D gel electrophoresis is the ability to provide a global view of changes occurring in the cell in terms of protein synthesis, abundance, turnover rate, and reversible (for example phosphorylation) and irreversible (for example proteolysis) post-translational modifications (O’Farrell,
1975). Previous reports have attempted to relate microarray with proteomics information from yeast (Ideker et al., 2001; Washburn et al., 2003), bacteria (Ou et al., 2005; Nie et al., 2006), cancer tissues (Chen et al., 2002) and cell lines (White et al., 2004; Kuo et al., 2005). In the present study, proteomics results were used to complement with the gene expression data from primary cortical neurons to provide a better understanding of the observed phenomena in neuronal apoptosis induced during inhibition of intracellular cholesterol transport. The resulting differentially- expressed proteins identified may more accurately represent the actual potential target or cell death-associated molecules during U18666A treatment.
235 Using proteomics tools, 25 differentially-expressed proteins during U18666A treatment have been identified. The identification of the same protein in different spots at the same time-point, as in the case of heterogeneous nuclear ribonucleoprotein and peptidylprolyl isomerase A, may indicate a shift in the pI or molecular weight due to post-translational modifications after U18666A treatment.
Not surprisingly, the proteomics and microarray analyses gave only 11 partially overlapping differentially-expressed proteins and genes. The roles of all these proteins in U18666A-mediated neuronal apoptosis have yet to be completely determined. Among the identified proteins with corresponding gene alterations, apolipoprotein E mediates the binding, internalization and catabolism of lipoprotein particles. It can also serve as a ligand for the low density lipoprotein receptor and plays a role in lipid metabolism and transport (Rajavashisth et al., 1985). Another protein, the aryl-hydrocarbon receptor-interacting protein has been reported to be involved in protein folding and trafficking (Sohocki et al., 2000). The brown fat inducible thioesterase has acyl-coenzyme A thioesterase activity towards medium- and long-chain fatty acyl-coenzyme A substrates (Adams et al., 2001). The heat shock protein (10 kDa chaperonin) is essential for mitochondrial protein biogenesis and is found to be induced by stress and aids protein folding (Dickson et al., 1994).
Peptidylprolyl isomerase A, also known as cyclophilin A, accelerates the folding of proteins and catalyzes the cis-trans isomerization of proline imidic peptide bonds in oligopeptides. It also binds cyclosporin A, which mediates an inhibitory action on the isomerase (Colgan et al., 2000). The presence of proteins which play a role in cell death responses may have a considerable influence on the biological properties of a
236 cell. Peroxiredoxin 6 is involved in redox regulation of the cell and may play a role in the regulation of phospholipid turnover, as well as in the protection against oxidative injury (Munz et al., 1997). The proteasome is a multicatalytic proteinase complex characterized by its ability to cleave peptide bonds with very broad specificity in the ubiquitin-proteasome pathway (Ciechanover, 1998; Hochstrasser, 1996). The proteasome is composed of at least 15 subunits, in which some are of the α-type, to form a highly-ordered ring-shaped structure (Tanaka, 1998). Although there was an up-regulation in the protein expression of proteasome subunit α type 6, proteasome activity was earlier found to decrease after 72 h of U18666A treatment (Chapter 3
Part II). Stathmin prevents assembly and promotes destabilization of microtubules
(Okazaki et al., 1993). The tumor necrosis factor-inducible protein plays a role in the regulation of innate resistance to pathogens and inflammatory reactions (Mantovani et al., 2003). Other proteins of interest include the heterogeneous nuclear ribonucleoproteins, which are involved in pre-mRNA processing where they form complexes with heterogeneous nuclear RNA (Roshon et al., 2003), and zinc finger proteins, which belong to the nuclear zinc finger protein family and function as transcription factors (Nagase et al., 2000).
Among the remaining proteins identified but with no matching genes from the microarray analysis, the α-1-antitrypsin precursor is an inhibitor of serine proteases
(Carrell et al., 1982). The CD2-associated protein acts as an adapter protein between
membrane proteins and the actin cytoskeleton, and may play a role in receptor
clustering and cytoskeletal polarity (Lehtonen et al., 2002). Creatine kinase B-type
237 reversibly catalyzes the transfer of phosphate between ATP and various phosphogens.
Creatine kinase isoenzymes are found to play a central role in energy transduction in tissues with large energy demands, such as in the heart, brain, and skeletal muscle
(van Deursen et al., 1992). Endothelial differentiation-related factor 1 is a
transcriptional co-activator which enhances the DNA-binding activity of transcription
factors such as Atf1 and Atf2. It is ubiquitously expressed in the brain, liver and heart
at the protein level and may function in lipid metabolism (Ballabio et al., 2004). The
Golgi autoantigen may play a role in vesicular transport from the trans-Golgi (Cowan
et al., 2002). The insulin-degrading enzyme plays a role in the cellular processing of
insulin through degradation of insulin, glucagon and other polypeptides, and may be involved in intercellular peptide signaling. Recently, it was reported that this enzyme degrades amyloid peptides associated with familial dementia (Morelli et al., 2005).
Lipocalin 4 aids in the transportation of lipophilic molecules and possibly acts as a
pheromone-carrier (Suzuki et al., 2004). The 3-mercaptopyruvate sulfurtransferase is
a newly-identified enzyme localized in the mouse mitochondria (Mootha et al., 2003).
The nucleotide-binding site protein 1, also known as PYRIN-containing APAF1-like
protein 2, may be implicated in apoptosis (Bertin and DiStefano, 2000). The
phosphoglucomutase 2-like 1 belongs to the phosphohexose mutase family and
possesses catalytic activity (Mammalian Gene Collection Program Team, 2002).
Transketolase also possesses catalytic activity and binds one calcium ion and one
thiamine pyrophosphate in each subunit of a homodimer (Salamon et al., 1998). The
WD-repeat protein 8 has a variety of cellular functions (Koshizuka et al., 2001). The
mitochondrial ATP synthase D chain is one of the chains of the non-enzymatic
238 component of the mitochondrial ATPase complex while the nucleoside diphosphate
kinase B plays a major role in the synthesis of nucleoside triphosphates other than
ATP (Genome Network Project Core Group, 2005).
Taken together, these differentially-expressed proteins suggest that U18666A
treatment may lead to cytoskeleton disruption, intracellular and intercellular signal
transduction, induction of stress responses, loss of energy production, and inhibition
of critical protein functions by arylation, which could all subsequently lead to
apoptotic cell death of the primary cortical neurons when irreversible damage occurs.
Others, which may be completely unrelated to the apoptosis, may play a role in the
global effect of U18666A in cortical neurons. The detection of apolipoprotein E,
endothelial differentiation-related factor 1 and thioesterase also supports the
involvement of cholesterol transport and lipid and fatty acid metabolism in U18666A-
mediated neuronal apoptosis.
The results of the present study illustrate some inherent limitations of the proteomics
approach. Proteomics is a relatively labor-intensive method. It relies on the
satisfactory resolution of different spots from 2D gel electrophoresis for the isolation
of sufficient sample from those spots for protein identification by mass spectrometry
(Aebersold and Mann, 2003; Graves and Haystead, 2002). A spot on the gel may
represent more than one protein due to poor resolution, or that the spot may also
correspond to another protein having similar molecular mass and pI. Here, only a
relatively small number of differentially-expressed proteins were found to correlate
239 with differentially-expressed genes from the microarray analysis. Some of the spots that were not positively identified may be due to sequence variations from post- translational modifications, while a few others might represent low molecular weight proteins which could not generate a sufficient number of tryptic peptides for protein identification. The poor spectra, which indicated that the samples were contaminated with an unknown polymer, might have also complicated identification. The unidentified protein spots can be further subjected to protein sequencing by tandem mass spectrometry to obtain amino acid sequences to search in the available databases to determine if they are potentially novel proteins or are actually known proteins which could not be initially identified with peptide mass fingerprinting (Griffiths et al., 2001; Dongré et al., 1997). Nevertheless, it is possible to have the gene and protein to be differentially expressed in opposite directions under the same treatment conditions, as in the case of the heterogeneous nuclear ribonucleoprotein and proteasome subunit α type 6. In addition, one has to bear in mind that the correlation between gene expression and protein expression is not always linear. There have been conflicting results on their correlation (Chen et al., 2002; White et al., 2004; Nie et al., 2006). It was reported that the correlation coefficient between the abundance of mRNA and protein is less than 0.5 (Anderson and Seilhamer, 1997; Gygi et al., 1999;
Chen et al., 2002). Even if many of the differences at the gene level are reflected in differences at the protein level, there is not always a good correlation as gene transcript levels cannot reflect protein modifications or interactions (Pandey and
Mann, 2000; Dove, 1999). On the contrary, it should also be noted that analysis of gene expression patterns is no less powerful a concept than proteomics when it comes
240 to the identification of the characteristics of signaling pathways or disease states
(Lockhart and Winzeler, 2000; Schulze and Downward, 2001).
By integrating information from both the proteomics and microarray approaches, correlations between changes in the gene and protein expression levels may offer insights into the function of the genes during U18666A-mediated neuronal apoptosis or serve as a guide in the search of protein biomarkers for U18666A exposure and toxicity. Hopefully, the effect of U18666A on gene and protein expressions in primary cortical neurons can be related to similar changes in the gene and protein expressions observed in disease states during neurodegeneration.
241
CHAPTER 6
GENERAL DISCUSSION AND FUTURE WORK
242 6 General Discussion and Future Work
6.1 General Discussion
The pharmacological agent, U18666A (3-β-[2-(diethylamino)ethoxy]androst-5-en-17- one), is a well-known class-2 amphiphile which inhibits cholesterol transport (Liscum and Faust, 1989; Liscum, 1990; Underwood et al., 1996). U18666A-treated cells
accumulate intracellular cholesterol to massive levels, similar to that observed in cells
from patients with Niemann-Pick disease type C (NPC) (Lange et al., 2000; Lange et al., 2002; Neufeld et al., 1999; Liscum and Sturley, 2004). U18666A is therefore widely used to mimic the cellular effects of NPC in several biological and cell culture systems. NPC is a juvenile neurodegenerative disorder characterized by premature neuronal cell loss and altered cholesterol metabolism (Pentchev et al., 1995).
Previous in vivo studies of neuronal injury and degenerating cerebral cortices and cerebellums in NPC mouse brains suggest that cell death may be via necrosis
(Erickson and Bernard, 2002). Other reports applying an 8-hour exposure of
U18666A to Semliki Forest virus-infected neurons and neuroblastoma cells, in which the amyloid precursor protein (APP) was overexpressed, demonstrated a dose- dependent reduction in β-amyloid (Aβ) deposition and secretion, but with no significant cell injury reported (Runz et al., 2002).
To study the chronic effect of U18666A, primary cortical neurons and different cell
lines were treated with U18666A for a maximum of 72 h (Chapter 3 Part I). The MTT
243 reduction assay and microscopic evaluations of cell cultures showed significant loss of cell survival and major morphological changes only in the U18666A-treated cortical neurons but not in the cell lines. Further work using only the primary cortical neurons demonstrated neurite blebbing, cellular shrinkage and caspase-3 activation, suggesting cell death was via apoptosis in the U18666A-treated cortical neurons.
Rapid swelling of the neurons indicative of necrosis was not observed. Z-VAD-FMK, a broad spectrum caspase inhibitor, was able to significantly attenuate U18666A- mediated neuronal apoptosis. Filipin staining and measurement of intracellular cholesterol levels revealed the accumulation of free cholesterol in cortical neurons treated with U18666A, indicating the phenotypic mimic of NPC by U18666A in the cortical neuronal cell system and that free cholesterol was a major storage material in
U18666A-treated cortical neurons. Co-treatment with pravastatin, a 3-hydroxy-3- methylglutaryl coenzyme A (HMG-CoA) reductase inhibitor, exhibited significant attenuation of U18666A toxicity in the primary cortical neurons. Taken together, these results showed for the first time that U18666A induces cell death by apoptosis only in primary cortical neurons, and suggested an important in vitro model system to study NPC. Recent studies have also reported the induction of apoptosis by U18666A in cultured bovine lens epithelial cells (Cenedella et al., 2005; Cenedella et al., 2004).
Observations from these studies suggest that apoptosis in cultured lens cells is independent of cellular cholesterol levels and may instead be due to direct perturbation of membrane structure (Cenedella et al., 2004). The difference between the findings in this research project and those reported by others might be due to different cell type responses.
244 The brain is more susceptible to oxidative damage than other organs in the body
(Halliwell and Gutteridge, 1999; Andersen, 2004; Emerit et al., 2004). Recently, there has been heightened interest in the role of oxidative stress in neurological disorders such as NPC and Alzheimer’s disease (AD). To characterize the effects of
U18666A on cellular metabolism, levels of intracellular ATP and glutathione were measured and were both found to decrease significantly after 72 h of U18666A treatment in primary cortical neurons (Chapter 3 Part II). The postglutamyl peptidase and chymotrypsin-like peptidase activities of the proteasome also decreased after 72 h of U18666A treatment. Markers associated with oxidative damage, such as mitochondrial depolarization, intracellular reactive oxygen species (ROS) production, lipid peroxidation, DNA oxidation and protein oxidation, also increased in U18666A- treated cortical neurons. These data demonstrated that the consequences of oxidative stress may contribute to the apoptotic process in primary cortical neurons during
U18666A treatment.
AD is one of the most common types of dementia affecting the elderly. The production of Aβ, which is derived from the proteolytic processing of APP, leads to the aggregation of Aβ peptides and deposition of amyloid plaques, which are major hallmarks of AD (Maccioni et al., 2001a). Oxidative stress may initiate amyloid formation with increased oxidative stress markers (Pappolla et al., 2002). Cortical neurons treated with U18666A showed decreased secretion and increased intraneuronal accumulation of both Aβ40 and Aβ42 (Chapter 3 Part II). Intracellular
Aβ42 might be the initial site of Aβ aggregation and neurotoxicity induced by
245 U18666A. In addition, U18666A treatment induced tau hyperphosphorylation, which occurred only in cortical neurons undergoing apoptosis (Chapter 3 Part III). The presence of positive labeling with the tau antibodies in U18666A-treated cortical neurons suggested that U18666A-mediated neuronal apoptosis is correlated with the presence of hyperphosphorylated tau. These mechanistic features reminiscent of AD, in association with the increased oxidative stress markers, may provide further clues to the etiology and pathogenesis of AD.
Chronic exposure to U18666A also led to the activation of caspases and calpains, suggesting a potential crosstalk between the caspase and calpain pathways during the neuronal apoptosis mediated by U18666A (Chapter 3 Part III). As pro-caspase-12 is predominantly found in the endoplasmic reticulum (ER), activation of caspase-12 at an earlier time-point before the activation of calpain and other caspases suggested that
ER stress might play a role in initiating U18666A-mediated neuronal apoptosis.
Caspase-12 could also activate other effector caspases in the downstream of the caspase activation cascade. On the other hand, the kinases studied, namely cyclin- dependent kinase 5 (Cdk5), glycogen synthase kinase 3 (GSK3), p44/42 mitogen- activated protein kinase (MAPK) and stress-activated protein kinase/c-Jun N-terminal kinase (SAPK/JNK), were not activated during U18666A treatment. The kinase or phosphatase which mediated U18666A-induced tau phosphorylation in primary cortical neurons remains unknown. In addition, U18666A treatment might also activate cell cycle machinery in primary cortical neurons, leading to a conflict of signals which resulted in neuronal apoptosis.
246 To elucidate the molecular processes contributing to neuronal apoptosis induced by
U18666A, the microarray approach was used in conjunction with proteomics techniques to identify specific proteins which might serve as signature biomarkers during U18666A treatment. An overview of both the genes and proteins differentially expressed during U18666A treatment was generated (Chapters 4 and 5). Although attempts to correlate the microarray and proteomics data were not particularly successful, eleven differentially-expressed proteins were correlated at the gene expression level in a time-dependent manner (Chapter 5). These proteins have been shown to play a role in lipid metabolism and transport, responses to cell death, protein folding and trafficking, and regulation of transcription. The identification of these differentially-expressed proteins might provide a clue to decipher the intracellular biochemical changes during U18666A-mediated neuronal apoptosis. These results provided, for the first time, a combined microarray and proteomics analysis of neuronal apoptosis due to inhibition of intracellular cholesterol transport. This new insight may greatly facilitate the study of neurodegenerative diseases.
A proposed mechanism of U18666A-mediated neuronal apoptosis is shown in Figure
6.1. At some stage, cellular injury in U18666A-treated cortical neurons may become irreversible and the neurons undergo apoptosis. Ultimately, it is hoped that the mechanism of U18666A-mediated neuronal apoptosis can provide clues to elucidate the pathogenesis of neurodegenerative diseases where apoptosis might occur through a similar mechanism due to an accumulation of intracellular cholesterol.
247
Figure 6.1: Proposed mechanism of U18666A-mediated neuronal apoptosis. Upon chronic exposure of primary cortical neurons to U18666A, there is an early activation of caspase-12, suggesting stress in the ER (denoted by 1). Cleavage of α-fodrin and activation of caspase-3 at a later time-point suggest a potential crosstalk between the caspase and calpain proteolytic pathways (denoted by 2). U18666A treatment leads to an increased production of ROS, which may play a role in the oxidative damage of primary cortical neurons, as illustrated by an increase in various oxidative stress markers (denoted by 3). In addition, ROS may also indirectly contribute to Aβ production (denoted by 4). U18666A treatment also leads to phosphorylation of p53 and activation of cell cycle regulatory proteins (denoted by 5). Hyperphosphorylation of tau is also observed, although which kinase is involved remains to be determined (denoted by 6). Taken together, these phenomena may synergistically lead to U18666A-mediated neuronal apoptosis in primary cortical neurons. (? denotes possible effects yet to be confirmed.)
248 6.2 Future Work
Although in vitro studies constitute a useful tool to characterize molecular events taking place under tightly controlled conditions, they undoubtedly cannot reproduce the physiological conditions of an organism. The use of the NPC murine model to complement the in vitro findings from this research project might more accurately delineate neurodegeneration observed in NPC patients, as the mouse model exhibits neuropathological similarities to the human condition.
The NPC mutant mouse was first discovered in a colony of BALB/c mice at the
National Center for Toxicological Research (Little Rock, AR, USA) in 1979
(Pentchev et al., 1980). Two murine models of NPC1 disease, namely BALB/c npcnih
(NPC1-/-) and C57BL/KsJ (sphingomyelinosis) mice, are available (Morris et al.,
1982; Miyataki et al., 1986). In the homozygous NPC1-/- mice, the murine ortholog of
NPC1 is mutated and this results in protein truncation before the sterol-sensing domains (Loftus et al., 1997). NPC1-/- mouse, which has since been widely used,
develops symptoms by the 5th week of age. Initial tremor is followed by hind limb
paralysis, poor food intake, and death within 11 weeks of age. On the other hand, the
sphingomyelinosis mice are found to be allelic with the homozygous NPC1-/- mice through crossbreeding experiments, although the exact gene mutation has yet to be identified. The newly-developed NPC2 knock-out mouse also has a phenotype globally similar to that of the NPC1 model (Sleat et al., 2004). Pathological changes
249 in both the NPC1 and NPC2 murine models mimic the severe late infantile neurological form of NPC in humans.
Recently, a revolutionary new method termed RNA interference (RNAi) has been shown to be highly effective in inhibiting the expression of specific target genes in mammalian cells in vitro (Hannon, 2002; Kurreck, 2003). RNAi in mammalian cells is achieved by the delivery of short double-stranded RNA molecules, termed short interfering RNA (siRNA), which results in a sequence-specific inhibition of target mRNA translation without triggering an interferon response (Yu et al., 2002). The effectiveness of the siRNA depends on efficient delivery into the target cells (Scherer and Rossi, 2003). The siRNA of the NPC sequence can be designed and delivered into primary cortical neurons to silence the NPC gene, using the Nucleofector™ technology (Amaxa Biosystems, Gaithersburg, MD, USA) which is based on electroporation. This nucleofection method is able to deliver the siRNA straight into the nucleus, even in non-dividing cells such as neurons. Previous trials of siRNA delivery into primary cortical neurons using a number of transfection reagents have resulted in very low transfection efficiencies. However, preliminary studies using the
Nucleofector™ technology have shown a promising transfection efficiency of at least
50%. This method of transfection can thus be utilized to effectively deliver the NPC siRNA into primary cortical neurons to further study the gene silencing effect.
250
REFERENCES
251 References
Adams SH, Chui C, Schilbach SL, Yu XX, Goddard AD, Grimaldi JC, Lee J, Dowd P, Colman S, and Lewin DA (2001) BFIT, a unique acyl-CoA thioesterase induced in thermogenic brown adipose tissue: Cloning, organization of the human gene and assessment of a potential link to obesity. Biochem J 360:135-142
Aebersold R and Mann M (2003) Mass spectrometry-based proteomics. Nature 422:198-207
Ahuja HS, Zhu Y, and Zakeri Z (1997) Association of cyclin-dependent kinase 5 and its activator p35 with apoptotic cell death. Dev Genet 21:258-267
Alberts B, Johnson A, Lewis J, Raff M, Roberts K, and Walter P (2002) Molecular Biology of the Cell, Fourth edition. Garland Science, New York
Alcala J, Cenedella RJ, and Katar M (1985) Limited proteolysis of MP26 in lens fiber plasma membranes of the U18666A-induced cataract in rats. Curr Eye Res 4:1001- 1005
Allen RT, Hunter III WJ, and Agrawal DK (1997) Morphological and biochemical characterization and analysis of apoptosis. J Pharmacol Toxicol Methods 37:215-228
Alvarez A, Toro R, Cáceres A, and Maccioni RB (1999) Inhibition of tau phosphorylating protein kinase cdk5 prevents β-amyloid-induced neuronal death. FEBS Lett 459:421-426
Amundson DM and Zhou M (1999) Fluorometric method for the enzymatic determination of cholesterol. J Biochem Biophys Methods 38:43-52
Andersen JK (2004) Oxidative stress in neurodegeneration: Cause or consequence? Nat Rev Neurosci 5:S18-S25
Anderson L and Seilhamer J (1997) A comparison of selected mRNA and protein abundances in human liver. Electrophoresis 18:533-537
Armstrong NJ and van de Wiel MA (2004) Microarray data analysis: From hypotheses to conclusions using gene expression data. Cell Oncol 26:279-290
Ballabio E, Mariotti M, De Benedictis L, and Maier JAM (2004) The dual role of endothelial differentiation-related factor-1 in the cytosol and nucleus: Modulation by protein kinase A. Cell Mol Life Sci 61:1069-1074
Barnham KJ, Masters CL, and Bush AI (2004) Neurodegenerative diseases and oxidative stress. Nat Rev Drug Discov 3:205-214
252 Beal MF (2005) Mitochondria take center stage in aging and neurodegeneration. Ann Neurol 58:495-505
Behl C (2000) Vitamin E protects neurons against oxidative cell death in vitro more effectively than 17-β estradiol and induces the activity of the transcription factor NF- κB. J Neural Transm 107:393-407
Behl C and Moosmann B (2002) Antioxidant neuroprotection in Alzheimer’s disease as preventive and therapeutic approach. Free Radic Biol Med 33:182-191
Bell EF (1987) History of vitamin E in infant nutrition. Am J Clin Nutr 46:183-186
Benetti R, Copetti T, Dell’Orso S, Melloni E, Brancolini C, Monte M, and Schneider C (2005) The calpain system is involved in the constitutive regulation of β-catenin signaling functions. J Biol Chem 280:22070-22080
Bergeron JJ, Brenner MB, Thomas DY, and Williams DB (1994) Calnexin: A membrane-bound chaperone of the endoplasmic reticulum. Trends Biochem Sci 19:124-128
Berlett BS and Stadtman ER (1997) Protein oxidation in aging, disease, and oxidative stress. J Biol Chem 272:20313-20316
Bernardi P, Scorrano L, Colonna R, Petronilli V, and Di Lisa F (1999) Mitochondria and cell death: Mechanistic aspects and methodological issues. Eur J Biochem 264:687-701
Bertin J and DiStefano PS (2000) The PYRIN domain: A novel motif found in apoptosis and inflammation proteins. Cell Death Differ 7:1273-1274
Bierkamper GG and Cenedella RJ (1978) Induction of chronic epileptiform activity in the rat by an inhibitor of cholesterol synthesis, U18666A. Brain Res 150:343-351
Bjellqvist B, Pasquali C, Ravier F, Sanchez JC, and Hochstrasser D (1993) A nonlinear wide-range immobilized pH gradient for two-dimensional electrophoresis and its definition in a relevant pH scale. Electrophoresis 14:1357-1365
Blackstock WP and Weir MP (1999) Proteomics: Quantitative and physical mapping of cellular proteins. Trends Biotechnol 17:121-127
Blomgren K, Zhu C, Wang X, Karlsson J-O, Leverin A-L, Bahr BA, Mallard C, and Hagberg H (2001) Synergistic activation of caspase-3 by m-calpain after neonatal hypoxia-ischemia: A mechanism of “pathological apoptosis”? J Biol Chem 276:10191-10198
253 Boogaard A, Griffioen M, and Cohen LH (1987) Regulation of 3-hydroxy-3- methylglutaryl-coenzyme A reductase in human hepatoma cell line Hep G2. Effects of inhibitors of cholesterol synthesis on enzyme activity. Biochem J 241:345-351
Brewer GJ, Torricelli JR, Evege EK, and Price PJ (1993) Optimized survival of hippocampal neurons in B27-supplemented Neurobasal™, a new serum-free medium combination. J Neurosci Res 35:567-576
Brodal P (1998) The Central Nervous System: Structure and Function, Second edition. Oxford University Press, New York
Brodie C and Blumberg PM (2003) Regulation of cell apoptosis by protein kinase c δ. Apoptosis 8:19-27
Brown PO and Botstein D (1999) Exploring the new world of the genome with DNA microarrays. Nat Genet Suppl 21:33-37
Buck MJ and Lieb JD (2004) ChIP-chip: Considerations for the design, analysis, and application of genome-wide chromatin immunoprecipitation experiments. Genomics 83:349-360
Budihardjo I, Oliver H, Lutter M, Luo X, and Wang X (1999) Biochemical pathways of caspase activation during apoptosis. Annu Rev Cell Dev Biol 15:269-290
Burns M and Duff K (2002) Cholesterol in Alzheimer’s disease and tauopathy. Ann N Y Acad Sci 977:367-375
Butler JD, Blanchette-Mackie J, Goldin E, O’Neill RR, Carstea G, Roff CF, Patterson MC, Patel S, Comly ME, Cooney A, Vanier MT, Brady RO, and Pentchev PG (1992) Progesterone blocks cholesterol translocation from lysosomes. J Biol Chem 267:23797-23805
Butterfield DA, Boyd-Kimball D, and Castegna A (2003) Proteomics in Alzheimer’s disease: Insights into potential mechanisms of neurodegeneration. J Neurochem 86:1313-1327
Butterfield DA, Griffin S, Munch G, and Pasinetti GM (2002) Amyloid β-peptide and amyloid pathology are central to the oxidative stress and inflammatory cascades under which Alzheimer’s disease brain exists. J Alzheimers Dis 4:193-201
Butterfield DA and Kanski J (2001) Brain protein oxidation in age-related neurodegenerative disorders that are associated with aggregated proteins. Mech Ageing Dev 122:945-962
Butterfield DA, Koppal T, Subramaniam R, and Yatin S (1999) Vitamin E as an antioxidant/free radical scavenger against amyloid β-peptide-induced oxidative stress
254 in neocortical synaptosomal membranes and hippocampal neurons in culture: Insights into Alzheimer's disease. Rev Neurosci 10:141-149
Butterfield DA and Lauderback CM (2002) Lipid peroxidation and protein oxidation in Alzheimer’s disease brain: Potential causes and consequences involving amyloid β- peptide-associated free radical oxidative stress. Free Radic Biol Med 32:1050-1060
Cadet J, Bourdat A-G, D’Ham C, Duarte V, Gasparutto D, Romieu A, and Ravanat J- L (2000) Oxidative base damage to DNA: Specificity of base excision repair enzymes. Mutat Res 462:121-128
Cadigan KM, Spillane DM, and Chang TY (1990) Isolation and characterization of Chinese hamster ovary cell mutants defective in intracellular low density lipoprotein- cholesterol trafficking. J Cell Biol 110:295-308
Caetano MS, Vieira-de-Abreu A, Teixeira LK, Werneck MBF, Barcinski MA, and Viola JPB (2002) NFATC2 transcription factor regulates cell cycle progression during lymphocyte activation: Evidence of its involvement in the control of cyclin gene expression. FASEB J 16:1940-1942
Camargo F, Erickson RP, Garver WS, Hossain GS, Carbone PN, Heidenreich RA, and Blanchard J (2001) Cyclodextrins in the treatment of a mouse model of Niemann- Pick C disease. Life Sci 70:131-142
Carrell RW, Jeppsson JO, Laurell CB, Brennan SO, Owen MC, Vaughan L, and Boswell DR (1982) Structure and variation of human alpha 1-antitrypsin. Nature 298:329-334
Carstea ED, Morris JA, Coleman KG, Loftus SK, Zhang D, Cummings C, Gu J, Rosenfeld MA, Pavan WJ, Krizman DB, Nagle J, Polymeropoulos MH, Sturley SL, Ioannou YA, Higgins ME, Comly M, Cooney A, Brown A, Kaneski CR, Blanchette- Mackie EJ, Dwyer NK, Neufeld EB, Chang T-Y, Liscum L, Strauss III JF, Ohno K, Zeigler M, Carmi R, Sokol J, Markie D, O’Neill RR, van Diggelen OP, Elleder M, Patterson MC, Brady RO, Vanier MT, Pentchev PG, and Tagle DA (1997) Niemann- Pick C1 disease gene: Homology to mediators of cholesterol homeostasis. Science 277:228-231
Casetta I, Govoni V, and Granieri E (2005) Oxidative stress, antioxidants and neurodegenerative diseases. Curr Pharm Des 11:2033-2052
Cenedella RJ (1980) Concentration-dependent effects of AY-9944 and U18666A on sterol synthesis in brain. Variable sensitivities of metabolic steps. Biochem Pharmacol 29:2751-2754
Cenedella RJ (1985) Regional distribution of lipids and phospholipase A2 activity in normal and cataractous rat lens. Curr Eye Res 4:113-120
255 Cenedella RJ and Bierkamper GG (1979) Mechanism of cataract production by 3- β(2-diethylaminoethoxy)androst-5-en-17-one hydrochloride, U18666A: An inhibitor of cholesterol biosynthesis. Exp Eye Res 28:673-688
Cenedella RJ, Jacob R, Borchman D, Tang D, Neely AR, Samadi A, Mason RP, and Sexton P (2004) Direct perturbation of lens membrane structure may contribute to cataracts caused by U18666A, an oxidosqualene cyclase inhibitor. J Lipid Res 45:1232-1241
Cenedella RJ and Sarkar CP (1984) Mechanism of depression of brain phospholipid levels by an epileptogenic drug. Biochem Pharmacol 33:591-598
Cenedella RJ, Sarkar CP, and Towns L (1982) Studies on the mechanism of the epileptiform activity induced by U18666A. II. Concentration, half-life and distribution of radiolabeled U18666A in the brain. Epilepsia 23:257-268
Cenedella RJ, Sexton PS, Krishnan K, and Covey DF (2005) Comparison of effects of U18666A and enantiomeric U18666A on sterol synthesis and induction of apoptosis. Lipids 40:635-640
Chandrasekher G and Cenedella RJ (1993) Calcium activated proteolysis and protein modification in the U18666A cataract. Exp Eye Res 57:737-745
Chapman T (2005) Pure but not simple. Nature 434:795-798
Checler F, da Costa CA, Ancolio K, Chevallier N, Lopez-Perez E, and Marambaud P (2000) Role of the proteasome in Alzheimer’s disease. Biochim Biophys Acta 1502:133-138
Chen G, Gharib TG, Huang C-C, Taylor JMG, Misek DE, Kardia SLR, Giordano TJ, Iannettoni MD, Orringer MB, Hanash SM, and Beer DG (2002) Discordant protein and mRNA expression in lung adenocarcinomas. Mol Cell Proteomics 1:304-313
Cheung NS, Pascoe CJ, Giardina SF, John CA, and Beart PM (1998) Micromolar L- glutamate induces extensive apoptosis in an apoptotic-necrotic continuum of insult- dependent, excitotoxic injury in cultured cortical neurons. Neuropharmacology 37:1419-1429
Chung C-W, Song Y-H, Kim I-K, Yoon W-J, Ryu B-R, Jo D-G, Woo H-N, Kwon Y- K, Kim H-H, Gwag B-J, Mook-Jung I-H, and Jung Y-K (2001) Proapoptotic effects of tau cleavage product generated by caspase-3. Neurobiol Dis 8:162-172
Ciechanover A (1998) The ubiquitin-proteasome pathway: On protein death and cell life. EMBO J 17:7151-7160
256 Cohen LH and Griffioen M (1988) Regulation of 3-hydroxy-3-methylglutaryl-CoA reductase mRNA contents in human hepatoma cell line Hep G2 by distinct classes of mevalonate-derived metabolites. Biochem J 255:61-67
Cohen LH, Griffioen M, van Roermund CWT, and Wanders RJA (1992) Subcellular localization of squalene synthase in human hepatoma cell line Hep G2. Biochim Biophys Acta 1126:114-118
Cohen LH, van Miert E, and Griffioen M (1989) Regulation of squalene synthetase in human hepatoma cell line Hep G2 by sterols, and not by mevalonate-derived non- sterols. Biochim Biophys Acta 1002:69-73
Colangelo V, Schurr J, Ball MJ, Pelaez RP, Bazan NG, and Lukiw WJ (2002) Gene expression profiling of 12633 genes in Alzheimer hippocampal CA1: Transcription and neurotrophic factor down-regulation and up-regulation of apoptotic and pro- inflammatory signaling. J Neurosci Res 70:462-473
Colgan J, Asmal M, and Luban J (2000) Isolation, characterization and targeted disruption of mouse Ppia: Cyclophilin A is not essential for mammalian cell viability. Genomics 68:167-178
Collins MO, Husi H, and Grant SGN (2004) Chapter 4 – Proteomics in the neurosciences. In: Proteomics: Biomedical and Pharmaceutical Applications. H Hondermarck (ed.), Kluwer Academic Publishers, the Netherlands, pp. 101-121
Cooper AJ and Kristal BS (1997) Multiple roles of glutathione in the central nervous system. Biol Chem 378:793-802
Copani A, Condorelli F, Caruso A, Vancheri C, Sala A, Giuffrida Stella AM, Canonico PL, Nicoletti F, and Sortino MA (1999) Mitotic signaling by β-amyloid causes neuronal death. FASEB J 13:2225-2234
Copani A, Uberti D, Sortino MA, Bruno V, Nicolletti F, and Memo M (2001) Activation of cell-cycle-associated proteins in neuronal death: A mandatory or dispensable path? Trends Neurosci 24:25-31
Cotman CW and Anderson AJ (1995) A potential role for apoptosis in neurodegeneration and Alzheimer’s disease. Mol Neurobiol 10:19-45
Cotman CW, Poon WW, Rissman RA, and Blurton-Jones M (2005) The role of caspase cleavage of tau in Alzheimer disease neuropathology. J Neuropathol Exp Neurol 64:104-112
Cottrell JS (1994) Protein identification by peptide mass fingerprinting. Pept Res 7:115-124
257 Cowan DA, Gay D, Bieler BM, Zhao H, Yoshino A, Davis JG, Tomayko MM, Murali R, Greene MI, and Marks MS (2002) Characterization of mouse tGolgin-1 (golgin-245/trans-golgi p230/256 kD golgin) and its upregulation during oligodendrocyte development. DNA Cell Biol 21:505-517
Cucchiara B and Kasner SE (2001) Use of statins in CNS disorders. J Neurol Sci 187:81-89
Dahl NK, Reed KL, Daunais MA, Faust JR, and Liscum L (1992) Isolation and characterization of Chinese hamster ovary cells defective in the intracellular metabolism of low density lipoprotein-derived cholesterol. J Biol Chem 267:4889- 4896
DeMattos RB, O’dell MA, Parsadanian M, Taylor JW, Harmony JAK, Bales KR, Paul SM, Aronow BJ, and Holtzman DM (2002) Clusterin promotes amyloid plaque formation and is critical for neuritic toxicity in a mouse model of Alzheimer’s disease. Proc Natl Acad Sci USA 99:10843-10848
Dennis Jr G, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, and Lempicki RA (2003) DAVID: Database for annotation, visualization, and integrated discovery. Genome Biol 4:R60.1-R60.11
Dickson R, Larsen B, Viitanen PV, Tormey MB, Geske J, Strange R, and Bemis LT (1994) Cloning, expression, and purification of a functional nonacetylated mammalian mitochondrial chaperonin 10. J Biol Chem 269:26858-26864
Distl R, Meske V, and Ohm TG (2001) Tangle-bearing neurons contain more free cholesterol than adjacent tangle-free neurons. Acta Neuropathol 101:547-554
Distl R, Treiber-Held S, Albert F, Meske V, Harzer K, and Ohm TG (2003) Cholesterol storage and tau pathology in Niemann-Pick type C disease in the brain. J Pathol 200:104-111
Dive C, Gregory CD, Phipps DJ, Evans DL, Milner AE, and Wyllie AH (1992) Analysis and discrimination of necrosis and apoptosis (programmed cell death) by multiparameter flow cytometry. Biochim Biophys Acta 1133:275-285
Dolev I and Michaelson DM (2004) A nontransgenic mouse model shows inducible amyloid-β (Aβ) peptide deposition and elucidates the role of apolipoprotein E in the amyloid cascade. Proc Natl Acad Sci USA 101:13909-13914
Dongré AR, Eng JK, and Yates III JR (1997) Emerging tandem-mass-spectrometry techniques for the rapid identification of proteins. Trends Biotechnol 15:418-425
Dove A (1999) Proteomics: Translating genomics into products? Nat Biotechnol 17:233-236
258 Drexler HCA (1997) Activation of the cell death program by inhibition of proteasome function. Proc Natl Acad Sci USA 94:855-860
Drummen GPC, Op den Kamp JAF, and Post JA (1999) Validation of the peroxidative indicators, cis-parinaric acid and parinaroyl-phospholipids, in a model system and cultured cardiac myocytes. Biochim Biophys Acta 1436:370-382
Drummen GPC, van Liebergen LCM, Op den Kamp JAF, and Post JA (2002) C11- BODIPY581/591, an oxidation-sensitive fluorescent lipid peroxidation probe: (Micro)spectroscopic characterization and validation of methodology. Free Radic Biol Med 33:473-490
Dudley DT, Pang L, Decker SJ, Bridges AJ, and Saltiel AR (1995) A synthetic inhibitor of the mitogen-activated protein kinase cascade. Proc Natl Acad Sci USA 92:7686-7689
Duriatti A, Bouvier-Nave P, Benveniste P, Schuber F, Delprino L, Balliano G, and Cattel L (1985) In vitro inhibition of animal and higher plants 2,3-oxidosqualene- sterol cyclases by 2-aza-2,3-dihydrosqualene and derivatives, and by other ammonium-containing molecules. Biochem Pharmacol 34:2765-2777
Earnshaw WC, Martins LM, and Kaufmann SH (1999) Mammalian caspases: Structure, activation, substrates, and functions during apoptosis. Annu Rev Biochem 68:383-424
Eckert A, Marques CA, Keil U, Schüssel K, and Müller WE (2003) Increased apoptotic cell death in sporadic and genetic Alzheimer’s disease. Ann N Y Acad Sci 1010:604-609
Ehler KW and Bierkamper GG (1986) Oxygen free radical mechanisms in rats treated with the epileptogenic drug, U18,666A. Proc West Pharmacol Soc 29:325-328
Ekert PG, Silke J, and Vaux DL (1999) Caspase inhibitors. Cell Death Differ 6:1081- 1086
Eltom SE, Zhang L, and Jefcoate CR (1999) Regulation of cytochrome P-450 (CYP) 1B1 in mouse Hepa-1 variant cell lines: A possible role for aryl hydrocarbon receptor nuclear translocator (ARNT) as a suppressor of CYP1B1 gene expression. Mol Pharmacol 55:594-604
Emerit J, Edeas M, and Bricaire F (2004) Neurodegenerative diseases and oxidative stress. Biomed Pharmacother 58:39-46
Emilsson L, Saetre P, and Jazin E (2006) Alzheimer’s disease: mRNA expression profiles of multiple patients show alterations of genes involved with calcium signaling. Neurobiol Dis 21:618-625
259 Erickson RP and Bernard O (2002) Studies on neuronal death in the mouse model of Niemann-Pick C disease. J Neurosci Res 68:738-744
Fassbender K, Simons M, Bergmann C, Stroick M, Lütjohann D, Keller P, Runz H, Kühl S, Bertsch T, von Bergmann K, Hennerici M, Beyreuther K, and Hartmann T (2001) Simvastatin strongly reduces levels of Alzheimer’s disease β-amyloid peptides Aβ42 and Aβ40 in vitro and in vivo. Proc Natl Acad Sci USA 98:5856-5861
Feng B, Yao PM, Li Y, Devlin CM, Zhang D, Harding HP, Sweeney M, Rong JX, Kuriakose G, Fisher EA, Marks AR, Ron D, and Tabas I (2003) The endoplasmic reticulum is the site of cholesterol-induced cytotoxicity in macrophages. Nat Cell Biol 5:781-792
Fenner GP and Raphiou I (1995) Growth of Cucurbita maxima L. plants in the presence of the cycloartenol synthase inhibitor U18666A. Lipids 30:253-256
Fenyö D (2000) Identifying the proteome: Software tools. Curr Opin Biotechnol 11:391-395
Ferrer I, Barrachina M, and Puig B (2002) Glycogen synthase kinase-3 is associated with neuronal and glial hyperphosphorylated tau deposits in Alzheimer’s disease, Pick’s disease, progressive supranuclear palsy and corticobasal degeneration. Acta Neuropathol (Berl) 104:583-591
Ferrer I, Gomez-Isla T, Puig B, Freixes M, Ribé E, Dalfó E, and Avila J (2005) Current advances on different kinases involved in tau phosphorylation, and implications in Alzheimer’s disease and tauopathies. Curr Alzheimer Res 2:3-18
Figueiredo-Pereira ME, Efthimiopoulos S, Tezapsidis N, Buku A, Ghiso J, Mehta P, and Robakis NK (1999) Distinct secretases, a cysteine protease and a serine protease, generate the C termini of amyloid β-proteins Aβ1–40 and Aβ1–42, respectively. J Neurochem 72:1417-1422
Fleschner CR and Cenedella RJ (1988) Specific restriction of cholesterol from cortical lens gap junctional membrane in the U18666A cataract. Curr Eye Res 7:1029-1034
Fodor SPA, Rava RP, Huang XC, Pease AC, Holmes CP, and Adams CL (1993) Multiplexed biochemical assays with biological chips. Nature 364:555-556
Fodor SPA, Read JL, Pirrung MC, Stryer L, Lu AT, and Solas D (1991) Light- directed, spatially addressable parallel chemical synthesis. Science 251:767-773
Foley KP and Eisenman RN (1999) Two MAD tails: What the recent knockouts of Mad1 and Mxi1 tell us about the MYC/MAX/MAD network. Biochim Biophys Acta 1423:M37-M47
260 Frand AR and Kaiser CA (1998) The ERO1 gene of yeast is required for oxidation of protein dithiols in the endoplasmic reticulum. Mol Cell 1:161-170
Friedlander RM (2003) Apoptosis and caspases in neurodegenerative diseases. N Engl J Med 348:1365-1375
Friso G and Wikström L (1999) Analysis of proteins from membrane-enriched cerebellar preparations by two-dimensional gel electrophoresis and mass spectrometry. Electrophoresis 20:917-927
Gaczynska M, Osmulski PA, and Ward WF (2001) Caretaker or undertaker? The role of the proteasome in aging. Mech Ageing Dev 122:235-254
Galat A (1993) Peptidylproline cis-trans-isomerases: Immunophilins. Eur J Biochem 216:689-707
Gan L, Ye S, Chu A, Anton K, Yi S, Vincent VA, von Schack D, Chin D, Murray J, Lohr S, Patthy L, Gonzalez-Zulueta M, Nikolich K, and Urfer R (2004) Identification of cathepsin B as a mediator of neuronal death induced by Aβ-activated microglial cells using a functional genomics approach. J Biol Chem 279:5565-5572
García-Román N, Álvarez AM, Toro MJ, Montes A, and Lorenzo MJ (2001) Lovastatin induces apoptosis of spontaneously immortalized rat brain neuroblasts: Involvement of nonsterol isoprenoid biosynthesis inhibition. Mol Cell Neurosci 17:329-341
Gaw A, Packard CJ, and Shepherd J (2004) Statins: The HMG-CoA Reductase Inhibitors in Perspective, Second edition. Martin Dunitz, United Kingdom
Genome Network Project Core Group (2005) The transcriptional landscape of the mammalian genome. Science 309:1559-1563
George AJ, Holsinger RMD, McLean CA, Laughton KM, Beyreuther K, Evin G, Masters CL, and Li Q-X (2004) APP intracellular domain is increased and soluble Aβ is reduced with diet-induced hypercholesterolemia in a transgenic mouse model of Alzheimer disease. Neurobiol Dis 16:124-132
Gershon D (2002) Microarray technology: An array of opportunities. Nature 416:885- 891
Gershon D (2005) More than gene expression. Nature 437:1195-1198
Geschwind DH (2000) Mice, microarrays, and the genetic diversity of the brain. Proc Natl Acad Sci USA 97:10676-10678
261 Geschwind DH (2003) DNA microarrays: Translation of the genome from laboratory to clinic. Lancet Neurol 2:275-282
Gibson GE and Huang H-M (2005) Oxidative stress in Alzheimer’s disease. Neurobiol Aging 26:575-578
Ginsberg SD, Hemby SE, Lee VM-Y, Eberwine JH, and Trojanowski JQ (2000) Expression profile of transcripts in Alzheimer’s disease tangle-bearing CA1 neurons. Ann Neurol 48:77-87
Giorgianni F and Beranova-Giorgianni S (2005) Analysis of the proteomes in human tissues by in-gel isoelectric focusing and mass spectrometry. In: The Proteomics Protocols Handbook. JM Walker (ed.), Humana Press, Totowa, New Jersey, pp. 367- 374
Giovanni A, Wirtz-Brugger F, Keramaris E, Slack R, and Park DS (1999) Involvement of cell cycle elements, cyclin-dependent kinases, pRb, and E2F·DP, in B-amyloid-induced neuronal death. J Biol Chem 274:19011-19016
Gitler D, Takagishi Y, Feng J, Ren Y, Rodriguiz RM, Wetsel WC, Greengard P, and Augustine GJ (2004) Different presynaptic roles of synapsins at excitatory and inhibitory synapses. J Neurosci 24:11368-11380
Goedert M, Jakes R, Crowther RA, Cohen P, Vanmechelen E, Vandermeeren M, and Cras P (1994) Epitope mapping of monoclonal antibodies to the paired helical filaments of Alzheimer’s disease: Identification of phosphorylation sites in tau protein. Biochem J 301:871-877
Goedert M, Jakes R, and Vanmechelen E (1995) Monoclonal antibody AT8 recognises tau protein phosphorylated at both serine 202 and threonine 205. Neurosci Lett 189:167-170
Goll DE, Thompson VF, Li H, Wei W, and Cong J (2003) The calpain system. Physiol Rev 83:731-801
Görg A, Obermaier C, Boguth G, Harder A, Scheibe B, Wildgruber R, and Weiss W (2000) The current state of two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 21:1037-1053
Gouras GK, Tsai J, Naslund J, Vincent B, Edgar M, Checler F, Greenfield JP, Haroutunian V, Buxbaum JD, Xu H, Greengard P, and Relkin NR (2000) Intraneuronal Aβ42 accumulation in human brain. Am J Pathol 156:15-20
Grana X and Reddy EP (1995) Cell cycle control in mammalian cells: Role of cyclins, cyclin dependent kinases (CDKs), growth suppressor genes and cyclin- dependent kinase inhibitors (CKIs). Oncogene 11:211-219
262 Graves PR and Haystead TAJ (2002) Molecular biologist’s guide to proteomics. Microbiol Mol Biol Rev 66:39-63
Green DR and Reed JC (1998) Mitochondria and apoptosis. Science 281:1309-1312
Griffin TJ, Gygi SP, Rist B, Aebersold R, Loboda A, Jilkine A, Ens W, and Standing KG (2001) Quantitative proteomic analysis using a MALDI quadrupole time-of-flight mass spectrometer. Anal Chem 73:978-986
Griffiths WJ, Jonsson AP, Liu S, Rai DK, and Wang Y (2001) Electrospray and tandem mass spectrometry in biochemistry. Biochem J 355:545-561
Grütter MG (2000) Caspases: Key players in programmed cell death. Curr Opin Struct Biol 10:649-655
Gygi SP, Rochon Y, Franza BR, and Aebersold R (1999) Correlation between protein and mRNA abundance in yeast. Mol Cell Biol 19:1720-1730
Hakak Y, Walker JR, Li C, Wong WH, Davis KL, Buxbaum JD, Haroutunian V, and Fienberg AA (2001) Genome-wide expression analysis reveals dysregulation of myelination-related genes in chronic schizophrenia. Proc Natl Acad Sci USA 98:4746-4751
Haley RW and Dietschy JM (2000) Is there a connection between the concentration of cholesterol circulating in plasma and the rate of neuritic plaque formation in Alzheimer disease? Arch Neurol 57:1410-1412
Hall AM, Krishnamoorthy L, and Orlow SJ (2003) Accumulation of tyrosinase in the endolysosomal compartment is induced by U18666A. Pigment Cell Res 16:149-158
Halliwell B (2001) Role of free radicals in the neurodegenerative diseases: Therapeutic implications for antioxidant treatment. Drugs Aging 18:685-716
Halliwell B (2002) Hypothesis: Proteasomal dysfunction – A primary event in neurodegeneration that leads to nitrative and oxidative stress and subsequent cell death. Ann N Y Acad Sci 962:182-194
Halliwell B and Gutteridge JMC (1999) Free Radicals in Biology and Medicine, Third edition. Oxford University Press, New York
Halliwell B and Jenner P (1998) Impaired clearance of oxidised proteins in neurodegenerative diseases. Lancet 351:1510
Hanash S (2003) Disease proteomics. Nature 422:226-232
263 Hanger DP, Betts JC, Loviny TLF, Blackstock WP, and Anderton BH (1998) New phosphorylation sites identified in hyperphosphorylated tau (paired helical filament- tau) from Alzheimer’s disease brain using nanoelectrospray mass spectrometry. J Neurochem 71:2465-2476
Hanlon SE and Lieb JD (2004) Progress and challenges in profiling the dynamics of chromatin and transcription factor binding with DNA microarrays. Curr Opin Genet Dev 14:697-705
Hannon GJ (2002) RNA interference. Nature 418:244-251
Härmälä A-S, Pörn MI, Mattjus P, and Slotte JP (1994) Cholesterol transport from plasma membranes to intracellular membranes is inhibited by 3β-[2- (diethylamino)ethoxy]androst-5-en-17-one. Biochim Biophys Acta 1211:317-325
Harvell DME, Richer JK, Allred DC, Sartorius CA, and Horwitz KB (2006) Estradiol regulates different genes in human breast tumor xenografts compared to the identical cells in culture. Endocrinology 147:700-713
Harwood SM, Yaqoob MM, and Allen DA (2005) Caspase and calpain function in cell death: Bridging the gap between apoptosis and necrosis. Ann Clin Biochem 42:415-431
Hashimoto M, Hsu LJ, Rockenstein E, Takenouchi T, Mallory M, and Masliah E (2002a) α-Synuclein protects against oxidative stress via inactivation of the c-Jun N- terminal kinase stress-signaling pathway in neuronal cells. J Biol Chem 277:11465- 11472
Hashimoto R, Hough C, Nakazawa T, Yamamoto T, and Chuang D-M (2002b) Lithium protection against glutamate excitotoxicity in rat cerebral cortical neurons: Involvement of NMDA receptor inhibition possibly by decreasing NR2B tyrosine phosphorylation. J Neurochem 80:589-597
Henchcliffe C and Burke RE (1997) Increased expression of cyclin-dependent kinase 5 in induced apoptotic neuron death in rat substantia nigra. Neurosci Lett 230:41-44
Hengartner MO (2000) The biochemistry of apoptosis. Nature 407:770-776
Hengst L and Reed SI (1996) Translational control of p27Kip1 accumulation during the cell cycle. Science 271:1861-1864
Henzel WJ, Billeci TM, Stults JT, Wong SC, Grimley C, and Watanabe C (1993) Identifying proteins from two-dimensional gels by molecular mass searching of peptide fragments in protein sequence databases. Proc Natl Acad Sci USA 90:5011- 5015
264 Herbert B (1999) Advances in protein solubilisation for two-dimensional electrophoresis. Electrophoresis 20:660-663
Hissin PJ and Hilf R (1976) A fluorometric method for determination of oxidized and reduced glutathione in tissues. Anal Biochem 74:214-226
Ho L, Sharma N, Blackman L, Festa E, Reddy G, and Pasinetti GM (2005) From proteomics to biomarker discovery in Alzheimer’s disease. Brain Res Rev 48:360- 369
Hochstrasser M (1996) Ubiquitin-dependent protein degradation. Annu Rev Genet 30:405-439
Horton HR, Moran LA, Ochs RS, Rawn JD, and Scrimgeour KG (1996) Principles of Biochemistry, Second edition. Prentice-Hall, New Jersey
Huang C, Jacobson K, and Schaller MD (2004) MAP kinases and cell migration. J Cell Sci 117:4619-4628
Huber LA (2003) Is proteomics heading in the wrong direction? Nat Rev Mol Cell Biol 4:74-80
Humphery-Smith I and Blackstock W (1997) Proteome analysis: Genomics via the output rather than the input code. J Protein Chem 16:537-544
Hunter T (1995) Protein kinases and phosphatases: The yin and yang of protein phosphorylation and signaling. Cell 80:225-236
Ideker T, Thorsson V, Ranish JA, Christman R, Buhler J, Eng JK, Bumgarner R, Goodlett DR, Aebersold R, and Hood L (2001) Integrated genomic and proteomic analyses of a systematically perturbed metabolic network. Science 292:929-934
Ikeda Y, Iguchi H, Nakata M, Ioka RX, Tanaka T, Iwasaki S, Magoori K, Takayasu S, Yamamoto TT, Kodama T, Yada T, Sakurai T, Yanagisawa M, and Sakai J (2005) Identification of N-arachidonylglycine, U18666A, and 4-androstene-3,17-dione as novel insulin Secretagogues. Biochem Biophys Res Commun 333:778-786
Incardona JP, Gaffield W, Lange Y, Cooney A, Pentchev PG, Liu S, Watson JA, Kapur RP, and Roelink H (2000) Cyclopamine inhibition of sonic hedgehog signal transduction is not mediated through effects on cholesterol transport. Dev Biol 224:440-452
Ioannou YA (2000) The structure and function of the Niemann-Pick C1 protein. Mol Genet Metab 71:175-181
265 Iqbal K and Grundke-Iqbal I (2004) Inhibition of neurofibrillary degeneration: A promising approach to Alzheimer’s disease and other tauopathies. Curr Drug Targets 5:495-502
Ishii T, Yanagawa T, Kawane T, Yuki K, Seita J, Yoshida H, and Bannai S (1996) Murine peritoneal macrophages induce a novel 60-kDa protein with structural similarity to a tyrosine kinase p56Ick-associated protein in response to oxidative stress. Biochem Biophys Res Commun 226:456-460
Jain KK (2004) Chapter 9 – Applications of proteomics technologies for drug discovery. In: Proteomics: Biomedical and Pharmaceutical Applications. H Hondermarck (ed.), Kluwer Academic Publishers, the Netherlands, pp. 201-227
Jeng I, Klemm N, and Proctor B (1984) Insensitivity of ubiquinone biosynthesis in glioblastoma cells to an epileptogenic drug, U18666A. J Neurochem 43:1409-1414
Jeng I, Klemm N, and Samson L (1985) Effect of a hypocholesterolemic agent on cholesteryl ester metabolism in glioblastoma cells. Biochem Pharmacol 34:1305-1309
Jenner A, England TG, Aruoma OI, and Halliwell B (1998) Measurement of oxidative DNA damage by gas chromatography-mass spectrometry: Ethanethiol prevents artifactual generation of oxidized DNA bases. Biochem J 331:365-369
Ji Z-S, Miranda RD, Newhouse YM, Weisgraber KH, Huang Y, and Mahley RW (2002) Apolipoprotein E4 potentiates amyloid β peptide-induced lysosomal leakage and apoptosis in neuronal cells. J Biol Chem 277:21821-21828
Jiang Z, Zheng X, Lytle RA, Higashikubo R, and Rich KM (2004) Lovastatin- induced up-regulation of the BH3-only protein, Bim, and cell death in glioblastoma cells. J Neurochem 89:168-178
Jick H, Zornberg GL, Jick SS, Seshadri S, and Drachman DA (2000) Statins and the risk of dementia. Lancet 356:1627-1631
Jin L-W, Maezawa I, Vincent I, and Bird T (2004) Intracellular accumulation of amyloidogenic fragments of amyloid-β precursor protein in neurons with Niemann- Pick type C defects is associated with endosomal abnormalities. Am J Pathol 164:975-985
Jin Z-G, Melaragno MG, Liao D-F, Yan C, Haendeler J, Suh Y-A, Lambeth JD, and Berk BC (2000) Cyclophilin A is a secreted growth factor induced by oxidative stress. Circ Res 87:789-796
Jonsson AP (2001) Mass spectrometry for protein and peptide characterization. Cell Mol Life Sci 58:868-884
266 Jordán J, Galindo MF, and Miller RJ (1997) Role of calpain- and interleukin-1β converting enzyme-like proteases in the β-amyloid-induced death of rat hippocampal neurons in culture. J Neurochem 68:1612-1621
Jung S-H, Bang H, and Young S (2005) Sample size calculation for multiple testing in microarray data analysis. Biostatistics 6:157-169
Jurgelski Jr W, Hudson PM, and Vogel FS (1973) Induction of a chronic somatosensory epilepsy in the opossum (Didelphis virginiana Kerr) with an inhibitor of cholesterol biosynthesis. Brain Res 64:466-471
Kadenbach B, Arnold S, Lee I, and Hüttemann M (2004) The possible role of cytochrome c oxidase in stress-induced apoptosis and degenerative diseases. Biochim Biophys Acta 1655:400-408
Kang HJ, Yoon WJ, Moon GJ, Kim DY, Sohn S, Kwon HJ, and Gwag BJ (2005) Caspase-3-mediated cleavage of PHF-1 tau during apoptosis irrespective of excitotoxicity and oxidative stress: An implication to Alzheimer’s disease. Neurobiol Dis 18:450-458
Karten B, Vance DE, Campenot RB, and Vance JE (2002) Cholesterol accumulates in cell bodies, but is decreased in distal axons, of Niemann-Pick C1-deficient neurons. J Neurochem 83:1154-1163
Keller JN, Gee J, and Ding Q (2002) The proteasome in brain aging. Ageing Res Rev 1:279-293
Keller JN, Hanni KB, and Markesbery WR (2000a) Impaired proteasome function in Alzheimer’s disease. J Neurochem 75:436-439
Keller JN, Hanni KB, and Markesbery WR (2000b) Possible involvement of proteasome inhibition in aging: Implications for oxidative stress. Mech Ageing Dev 113:61-70
Kellner-Weibel G, Jerome WG, Small DM, Warner GJ, Stoltenborg JK, Kearney MA, Corjay MH, Phillips MC, and Rothblat GH (1998) Effects of intracellular free cholesterol accumulation on macrophage viability: A model for foam cell death. Arterioscler Thromb Vasc Biol 18:423-431
Kilsdonk EPC, Yancey PG, Stoudt GW, Bangerter FW, Johnson WJ, Phillips MC, and Rothblat GH (1995) Cellular cholesterol efflux mediated by cyclodextrins. J Biol Chem 270:17250-17256
Klose J (1975) Protein mapping by combined isoelectric focusing and electrophoresis of mouse tissues. A novel approach to testing for induced point mutations in mammals. Humangenetik 26:231-243
267 Knauer MF, Soreghan B, Burdick D, Kosmoski J, and Glabe CG (1992) Intracellular accumulation and resistance to degradation of the Alzheimer amyloid A4/β protein. Proc Natl Acad Sci USA 89:7437-7441
Kobayashi K, Nakano H, Hayashi M, Shimazaki M, Fukutani Y, Sasaki K, Sugimori K, and Koshino Y (2003) Association of phosphorylation site of tau protein with neuronal apoptosis in Alzheimer’s disease. J Neurol Sci 208:17-24
Kobayashi T, Beuchat M-H, Lindsay M, Frias S, Palmiter RD, Sakuraba H, Parton RG, and Gruenberg J (1999) Late endosomal membranes rich in lysobisphosphatidic acid regulate cholesterol transport. Nat Cell Biol 1:113-118
Kobayashi T, Vischer UM, Rosnoblet C, Lebrand C, Lindsay M, Parton RG, Kruithof EKO, and Gruenberg J (2000) The tetraspanin CD63/lamp3 cycles between endocytic and secretory compartments in human endothelial cells. Mol Biol Cell 11:1829-1843
Kolodny EH (2000) Niemann-Pick disease. Curr Opin Hematol 7:48-52
Kölsch H, Ludwig M, Lütjohann D, and Rao ML (2001) Neurotoxicity of 24- hydroxycholesterol, an important cholesterol elimination product of the brain, may be prevented by vitamin E and estradiol-17β. J Neural Transm 108:475-488
Koshizuka Y, Ikegawa S, Sano M, Nakamura K, and Nakamura Y (2001) Isolation, characterization, and mapping of the mouse and human WDR8 genes, members of a novel WD-repeat gene family. Genomics 72:252-259
Kovacs WJ, Faust PL, Keller G-A, and Krisans SK (2001) Purification of brain peroxisomes and localization of 3-hydroxy-3-methylglutaryl coenzyme A reductase. Eur J Biochem 268:4850-4859
Kroemer G and Reed JC (2000) Mitochondrial control of cell death. Nat Med 6:513- 519
Kruman II, Wersto RP, Cardozo-Pelaez F, Smilenov L, Chan SL, Chrest FJ, Emokpae Jr R, Gorospe M, and Mattson MP (2004) Cell cycle activation linked to neuronal cell death initiated by DNA damage. Neuron 41:549-561
Kubota H, Yokota S-I, Yanagi H, and Yura T (1999) Structures and co-regulated expression of the genes encoding mouse cytosolic chaperonin CCT subunits. Eur J Biochem 262:492-500
Kumar B, Andreatta C, Koustas WT, Cole WC, Edwards-Prasad J, and Prasad KN (2002) Mevastatin induces degeneration and decreases viability of cAMP-induced differentiated neuroblastoma cells in culture by inhibiting proteasome activity, and mevalonic acid lactone prevents these effects. J Neurosci Res 68:627-635
268 Kuo C-C, Kuo C-W, Liang C-M, and Liang S-M (2005) A transcriptomic and proteomic analysis of the effect of CpG-ODN on human THP-1 monocytic leukemia cells. Proteomics 5:894-906
Kurreck J (2003) Antisense technologies. Improvement through novel chemical modifications. Eur J Biochem 270:1628-1644
Kurt MA, Davies DC, and Kidd M (1997) Paired helical filament morphology varies with intracellular location in Alzheimer’s disease brain. Neurosci Lett 239:41-44
Kuszak JR, Khan AR, and Cenedella RJ (1988) An ultrastructural analysis of plasma membrane in the U18666A cataract. Invest Ophthalmol Vis Sci 29:261-267
Kuwako K-I, Nishimura I, Uetsuki T, Saido TC, and Yoshikawa K (2002) Activation of calpain in cultured neurons overexpressing Alzheimer amyloid precursor protein. Mol Brain Res 107:166-175
Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685
Lakin ND and Jackson SP (1999) Regulation of p53 in response to DNA damage. Oncogene 18:7644-7655
Lam YA, Pickart CM, Alban A, Landon M, Jamieson C, Ramage R, Mayer RJ, and Layfield R (2000) Inhibition of the ubiquitin-proteasome system in Alzheimer’s disease. Proc Natl Acad Sci USA 97:9902-9906
Lange Y and Steck TL (1994) Cholesterol homeostasis. Modulation by amphiphiles. J Biol Chem 269:29371-29374
Lange Y and Steck TL (1996) The role of intracellular cholesterol transport in cholesterol homeostasis. Trends Cell Biol 6:205-208
Lange Y, Ye J, Rigney M, and Steck T (2000) Cholesterol movement in Niemann- Pick type C cells and in cells treated with amphiphiles. J Biol Chem 275:17468- 17475
Lange Y, Ye J, Rigney M, and Steck TL (1999) Regulation of endoplasmic reticulum cholesterol by plasma membrane cholesterol. J Lipid Res 40:2264-2270
Lange Y, Ye J, Rigney M, and Steck TL (2002) Dynamics of lysosomal cholesterol in Niemann-Pick type C and normal human fibroblasts. J Lipid Res 43:198-204
Lange Y, Ye J, and Steck TL (1998) Circulation of cholesterol between lysosomes and the plasma membrane. J Biol Chem 273:18915-18922
269 Langen H, Berndt P, Röder D, Cairns N, Lubec G, and Fountoulakis M (1999) Two- dimensional map of human brain proteins. Electrophoresis 20:907-916
Laranjinha JA, Almeida LM, and Madeira VM (1992) Lipid peroxidation and its inhibition in low density lipoproteins: Quenching of cis-parinaric acid fluorescence. Arch Biochem Biophys 297:147-154
LeBel CP, Ischiropoulos H, and Bondy SC (1992) Evaluation of the probe 2’,7’- dichlorofluorescin as an indicator of reactive oxygen species formation and oxidative stress. Chem Res Toxicol 5:227-231
Lee C-K, Weindruch R, and Prolla TA (2000a) Gene expression profile of the ageing brain in mice. Nat Genet 25:294-297
Lee J-M, Shih AY, Murphy TH, and Johnson JA (2003) NF-E2-related factor-2 mediates neuroprotection against mitochondrial complex I inhibitors and increased concentrations of intracellular calcium in primary cortical neurons. J Biol Chem 278:37948-37956
Lee MH, Hyun DH, Jenner P, and Halliwell B (2001) Effect of proteasome inhibition on cellular oxidative damage, antioxidant defences and nitric oxide production. J Neurochem 78:32-41
Lee M-LT, Kuo FC, Whitmorei GA, and Sklar J (2000b) Importance of replication in microarray gene expression studies: Statistical methods and evidence from repetitive cDNA hybridizations. Proc Natl Acad Sci USA 97:9834-9839
Lehtonen S, Zhao F, and Lehtonen E (2002) CD2-associated protein directly interacts with the actin cytoskeleton. Am J Physiol Renal Physiol 283:F734-F743
Levine RL (2002) Carbonyl modified proteins in cellular regulation, aging, and disease. Free Radic Biol Med 32:790-796
Lewin B (1994) Genes V. Oxford University Press, New York
Lewis SA, Lee MG-S, and Cowan NJ (1985) Five mouse tubulin isotypes and their regulated expression during development. J Cell Biol 101:852-861
Li G, Faibushevich A, Turunen BJ, Yoon SO, Georg G, Michaelis ML, and Dobrowsky RT (2003) Stabilization of the cyclin-dependent kinase 5 activator, p35, by paclitaxel decreases β-amyloid toxicity in cortical neurons. J Neurochem 84:347- 362
Liou Y-C, Sun A, Ryo A, Zhou XZ, Yu Z-X, Huang H-K, Uchida T, Bronson R, Bing G, Li X, Hunter T, and Lu KP (2003) Role of the prolyl isomerase Pin1 in protecting against age-dependent neurodegeneration. Nature 424:556-561
270 Lipshutz RJ, Fodor SPA, Gingeras TR, and Lockhart DJ (1999) High density synthetic oligonucleotide arrays. Nat Genet Suppl 21:20-24
Liscum L (1990) Pharmacological inhibition of the intracellular transport of low- density lipoprotein-derived cholesterol in Chinese hamster ovary cells. Biochim Biophys Acta 1045:40-48
Liscum L (2000) Niemann-Pick type C mutations cause lipid traffic jam. Traffic 1:218-225
Liscum L and Faust JR (1989) The intracellular transport of low density lipoprotein- derived cholesterol is inhibited in Chinese hamster ovary cells cultured with 3-β-[2- (diethylamino)ethoxy]androst-5-en-17-one. J Biol Chem 264:11796-11806
Liscum L and Sturley SL (2004) Intracellular trafficking of Niemann-Pick C proteins 1 and 2: Obligate components of subcellular lipid transport. Biochim Biophys Acta 1685:22-27
Liu G, Loraine AE, Shigeta R, Cline M, Cheng J, Valmeekam V, Sun S, Kulp D, and Siani-Rose MA (2003) NetAffx: Affymetrix probesets and annotations. Nucleic Acids Res 31:82-86
Lockhart DJ and Barlow C (2001) Expressing what’s on your mind: DNA arrays and the brain. Nat Rev Neurosci 2:63-68
Lockhart DJ and Winzeler EA (2000) Genomics, gene expression and DNA arrays. Nature 405:827-836
Loftus SK, Morris JA, Carstea ED, Gu JZ, Cummings C, Brown A, Ellison J, Ohno K, Rosenfeld MA, Tagle DA, Pentchev PG, and Pavan WJ (1997) Murine model of Niemann-Pick C disease: Mutation in a cholesterol homeostasis gene. Science 277:232-235
Love S, Bridges LR, and Case CP (1995) Neurofibrillary tangles in Niemann-Pick disease type C. Brain 118:119-129
Lütjohann D, Stroick M, Bertsch T, Kühl S, Lindenthal B, Thelen K, Andersson U, Björkhem I, von Bergmann K, and Fassbender K (2004) High doses of simvastatin, pravastatin, and cholesterol reduce brain cholesterol synthesis in guinea pigs. Steroids 69:431-438
Lyras L, Cairns NJ, Jenner A, Jenner P, and Halliwell B (1997) An assessment of oxidative damage to proteins, lipids, and DNA in brains from patients with Alzheimer’s disease. J Neurochem 68:2061-2069
271 Ma J, Wollmann R, and Lindquist S (2002) Neurotoxicity and neurodegeneration when PrP accumulates in the cytosol. Science 298:1781-1785
Maccioni RB, Muñoz JP, and Barbeito L (2001a) The molecular bases of Alzheimer’s disease and other neurodegenerative disorders. Arch Med Res 32:367-381
Maccioni RB, Otth C, Concha II, and Muñoz JP (2001b) The protein kinase Cdk5. Structural aspects, roles in neurogenesis and involvement in Alzheimer’s pathology. Eur J Biochem 268:1518-1527
Maeda A, Crabb JW, and Palczewski K (2005) Microsomal glutathione S-transferase 1 in the retinal pigment epithelium: Protection against oxidative stress and a potential role in aging. Biochemistry 44:480-489
Mak SK and Kültz D (2004) Gadd45 proteins induce G2/M arrest and modulate apoptosis in kidney cells exposed to hyperosmotic stress. J Biol Chem 279:39075- 39084
Mammalian Gene Collection Program Team (2002) Generation and initial analysis of more than 15,000 full-length human and mouse cDNA sequences. Proc Natl Acad Sci USA 99:16899-16903
Mantovani A, Garlanda C, and Bottazzi B (2003) Pentraxin 3, a non-redundant soluble pattern recognition receptor involved in innate immunity. Vaccine 21:S2/43- S2/47
Marciniak SJ, Yun CY, Oyadomari S, Novoa I, Zhang Y, Jungreis R, Nagata K, Harding HP, and Ron D (2004) CHOP induces death by promoting protein synthesis and oxidation in the stressed endoplasmic reticulum. Genes Dev 18:3066-3077
Marine J-C and Jochemsen AG (2005) Mdmx as an essential regulator of p53 activity. Biochem Biophys Res Commun 331:750-760
Mark RJ, Lovell MA, Markesbery WR, Uchida K, and Mattson MP (1997a) A role for 4-hydroxynonenal, an aldehydic product of lipid peroxidation, in disruption of ion homeostasis and neuronal death induced by amyloid β-peptide. J Neurochem 68:255- 264
Mark RJ, Pang Z, Geddes JW, Uchida K, and Mattson MP (1997b) Amyloid β- peptide impairs glucose transport in hippocampal and cortical neurons: Involvement of membrane lipid peroxidation. J Neurosci 17:1046-1054
Markesbery WR (1997) Oxidative stress hypothesis in Alzheimer’s disease. Free Radic Biol Med 23:134-147
272 Marton MJ, DeRisi JL, Bennett HA, Iyer VR, Meyer MR, Roberts CJ, Stoughton R, Burchard J, Slade D, Dai H, Bassett Jr DE, Hartwell LH, Brown PO, and Friend SH (1998) Drug target validation and identification of secondary drug target effects using DNA microarrays. Nat Med 4:1293-1301
Mattson MP (2000) Apoptosis in neurodegenerative disorders. Nat Rev Mol Cell Biol 1:120-129
McGinnis KM, Whitton MM, Gnegy ME, and Wang KKW (1998) Calcium/calmodulin-dependent protein kinase IV is cleaved by caspase-3 and calpain in SH-SY5Y human neuroblastoma cells undergoing apoptosis. J Biol Chem 273:19993-20000
McShea A, Wahl AF, and Smith MA (1999) Re-entry into the cell cycle: A mechanism for neurodegeneration in Alzheimer disease. Med Hypotheses 52:525-527
Menge T, Hartung H-P, and Stüve O (2005) Statins – A cure-all for the brain? Nat Rev Neurosci 6:325-331
Michikawa M (2004) Neurodegenerative disorders and cholesterol. Curr Alzheimer Res 1:271-275
Middleton FA, Mirnics K, Pierri JN, Lewis DA, and Levitt P (2002) Gene expression profiling reveals alterations of specific metabolic pathways in schizophrenia. J Neurosci 22:2718-2729
Miki H, Setou M, Kaneshiro K, and Hirokawa N (2001) All kinesin superfamily protein, KIF, genes in mouse and human. Proc Natl Acad Sci USA 98:7004-7011
Miller FD, Pozniak CD, and Walsh GS (2000) Neuronal life and death: An essential role for the p53 family. Cell Death Differ 7:880-888
Miller RG (1984) The use and abuse of filipin to localize cholesterol in membranes. Cell Biol Int Rep 8:519-535
Miranda S, Opazo C, Larrondo LF, Muñoz FJ, Ruiz F, Leighton F, and Inestrosa NC (2000) The role of oxidative stress in the toxicity induced by amyloid β-peptide in Alzheimer’s disease. Prog Neurobiol 62:633-648
Mirnics K, Middleton FA, Marquez A, Lewis DA, and Levitt P (2000) Molecular characterization of schizophrenia viewed by microarray analysis of gene expression in prefrontal cortex. Neuron 28:53-67
Miyawaki S, Yoshida H, Mitsuoka S, Enomoto H, and Ikehara S (1986) A mouse model for Niemann-Pick disease. Influence of genetic background on disease expression in spm/spm mice. J Hered 77:379-384
273 Mohammadi A, Perry RJ, Storey MK, Cook HW, Byers DM, and Ridgway ND (2001) Golgi localization and phosphorylation of oxysterol binding protein in Niemann-Pick C and U18666A-treated cells. J Lipid Res 42:1062-1071
Molloy MP (2000) Two-dimensional electrophoresis of membrane proteins using immobilized pH gradients. Anal Biochem 280:1-10
Monks TJ, Ghersi-Egea J-F, Philbert M, Cooper AJL, and Lock EA (1999) Symposium overview: The role of glutathione in neuroprotection and neurotoxicity. Toxicol Sci 51:161-177
Moosmann B and Behl C (2002) Antioxidants as treatment for neurodegenerative disorders. Expert Opin Investig Drugs 11:1407-1435
Mootha VK, Bunkenborg J, Olsen JV, Hjerrild M, Wisniewski JR, Stahl E, Bolouri MS, Ray HN, Sihag S, Kamal M, Patterson N, Lander ES, and Mann M (2003) Integrated analysis of protein composition, tissue diversity, and gene regulation in mouse mitochondria. Cell 115:629-640
Morelli L, Llovera RE, Alonso LG, Frangione B, de Prat-Gay G, Ghiso J, and Castaño EM (2005) Insulin-degrading enzyme degrades amyloid peptides associated with British and Danish familial dementia. Biochem Biophys Res Commun 332:808- 816
Morris MD, Bhuvaneswaran C, Shio H, and Fowler S (1982) Lysosome lipid storage disorder in NCTR-BALB/c mice. I. Description of the disease and genetics. Am J Pathol 108:140-149
Mourtada-Maarabouni M, Kirkham L, Jenkins B, Rayner J, Gonda TJ, Starr R, Trayner I, Farzaneh F, and Williams GT (2003) Functional expression cloning reveals proapoptotic role for protein phosphatase 4. Cell Death Differ 10:1016-1024
Mukherjee S and Maxfield FR (1999) Cholesterol: Stuck in traffic. Nat Cell Biol 1:E37-E38
Mukherjee S and Maxfield FR (2004) Lipid and cholesterol trafficking in NPC. Biochim Biophys Acta 1685:28-37
Munz B, Frank S, Hübner G, Olsen E, and Werner S (1997) A novel type of glutathione peroxidase: Expression and regulation during wound repair. Biochem J 326:579-585
Murphy EJ (2002) Sterol carrier protein-2: Not just for cholesterol any more. Mol Cell Biochem 239:87-93
274 Nagase T, Kikuno R, Ishikawa K-I, Hirosawa M, and Ohara O (2000) Prediction of the coding sequences of unidentified human genes. XVI. The complete sequences of 150 new cDNA clones from brain which code for large proteins in vitro. DNA Res 7:65-73
Nagy ZS, Esiri MM, and Smith AD (1998) The cell division cycle and the pathophysiology of Alzheimer’s disease. Neuroscience 87:731-739
Nakagawa T and Yuan J (2000) Cross-talk between two cysteine protease families: Activation of caspase-12 by calpain in apoptosis. J Cell Biol 150:887-894
Nakagawa T, Zhu H, Morishima N, Li E, Xu J, Yankner BA, and Yuan J (2000) Caspase-12 mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloid-β. Nature 403:98-103
Narumiya S, Ishizaki T, and Uehata M (2000) Use and properties of ROCK-specific inhibitor Y-27632. Methods Enzymol 325:273-284
Naureckiene S, Sleat DE, Lackland H, Fensom A, Vanier MT, Wattiaux R, Jadot M, and Lobel P (2000) Identification of HE1 as the second gene of Niemann-Pick C disease. Science 290:2298-2301
Neufeld EB, Wastney M, Patel S, Suresh S, Cooney AM, Dwyer NK, Roff CF, Ohno K, Morris JA, Carstea ED, Incardona JP, Strauss III JF, Vanier MT, Patterson MC, Brady RO, Pentchev PG, and Blanchette-Mackie EJ (1999) The Niemann-Pick C1 protein resides in a vesicular compartment linked to retrograde transport of multiple lysosomal cargo. J Biol Chem 274:9627-9635
Neumar RW, Xu YA, Gada H, Guttmann RP, and Siman R (2003) Cross-talk between calpain and caspase proteolytic systems during neuronal apoptosis. J Biol Chem 278:14162-14167
Nie L, Wu G, and Zhang W (2006) Correlation between mRNA and protein abundance in Desulfovibrio vulgaris: A multiple regression to identify sources of variations. Biochem Biophys Res Commun 339:603-610
Nijhawan D, Honarpour N, and Wang X (2000) Apoptosis in neural development and disease. Annu Rev Neurosci 23:73-87
Niki E, Yoshida Y, Saito Y, and Noguchi N (2005) Lipid peroxidation: Mechanisms, inhibition, and biological effects. Biochem Biophys Res Commun 338:668-676
Nixon RA (2004) Niemann-Pick type C disease and Alzheimer’s disease: The APP- endosome connection fattens up. Am J Pathol 164:757-761
275 Nonaka S, Hough CJ, and Chuang D-M (1998) Chronic lithium treatment robustly protects neurons in the central nervous system against excitotoxicity by inhibiting N- methyl-D-aspartate receptor-mediated calcium influx. Proc Natl Acad Sci USA 95:2642-2647
O’Farrell PH (1975) High resolution two-dimensional electrophoresis of proteins. J Biol Chem 250:4007-4021
Ohm TG, Treiber-Held S, Distl R, Glöckner F, Schönheit B, Tamanai M, and Meske V (2003) Cholesterol and tau protein – Findings in Alzheimer’s and Niemann Pick C’s disease. Pharmacopsychiatry 36:S120-S126
Ohsawa I, Nishimaki K, Yasuda C, Kamino K, and Ohta S (2003) Deficiency in a mitochondrial aldehyde dehydrogenase increases vulnerability to oxidative stress in PC12 cells. J Neurochem 84:1110-1117
Okazaki T, Yoshida BN, Avraham KB, Wang H, Wuenschell CW, Jenkins NA, Copeland NG, Anderson DJ, and Mori N (1993) Molecular diversity of the SCG10/stathmin gene family in the mouse. Genomics 18:360-373
Oono K, Yoneda T, Manabe T, Yamagishi S, Matsuda S, Hitomi J, Miyata S, Mizuno T, Imaizumi K, Katayama T, and Tohyama M (2004) JAB1 participates in unfolded protein responses by association and dissociation with IRE1. Neurochem Int 45:765- 772
Orth K, O’Rourke K, Salvesen GS, and Dixit VM (1996) Molecular ordering of apoptotic mammalian CED-3/ICE-like proteases. J Biol Chem 271:20977-20980
Ory DS (2000) Niemann-Pick type C: A disorder of cellular cholesterol trafficking. Biochim Biophys Acta 1529:331-339
Osakada F, Hashino A, Kume T, Katsuki H, Kaneko S, and Akaike A (2003) Neuroprotective effects of α-tocopherol on oxidative stress in rat striatal cultures. Eur J Pharmacol 465:15-22
Otvos Jr L, Feiner L, Lang E, Szendrei GI, Goedert M, and Lee VM (1994) Monoclonal antibody PHF-1 recognizes tau protein phosphorylated at serine residues 396 and 404. J Neurosci Res 39:669-673
Ou K, Ong C, Koh SY, Rodrigues F, Sim SH, Wong D, Ooi CH, Ng KC, Jikuya H, Yau CC, Soon SY, Kesuma D, Lee MA, and Tan P (2005) Integrative genomic, transcriptional, and proteomic diversity in natural isolates of the human pathogen Burkholderia pseudomallei. J Bacteriol 187:4276-4285
Padmanabhan J, Park DS, Greene LA, and Shelanski ML (1999) Role of cell cycle regulatory proteins in cerebellar granule neuron apoptosis. J Neurosci 19:8747-8756
276 Pagano M, Tam SW, Theodoras AM, Beer-Romero P, Del Sal D, Chau V, Yew PR, Draetta GF, and Rolfe M (1995) Role of the ubiquitin-proteasome pathway in regulating abundance of the cyclin-dependent kinase inhibitor p27. Science 269:682- 685
Pandey A and Mann M (2000) Proteomics to study genes and genomes. Nature 405:837-846
Pap EHW, Drummen GPC, Winter VJ, Kooij TWA, Rijken P, Wirtz KWA, Op den Kamp JAF, Hage WJ, and Post JA (1999) Ratio-fluorescence microscopy of lipid oxidation in living cells using C11-BODIPY581/591. FEBS Lett 453:278-282
Pappolla MA, Smith MA, Bryant-Thomas T, Bazan N, Petanceska S, Perry G, Thal LJ, Sano M, and Refolo LM (2002) Cholesterol, oxidative stress, and Alzheimer’s disease: Expanding the horizons of pathogenesis. Free Radic Biol Med 33:173-181
Patenaude A, Murthy MRV, and Mirault M-E (2005) Emerging roles of thioredoxin cycle enzymes in the central nervous system. Cell Mol Life Sci 62:1063-1080
Patrick GN, Zhou P, Kwon YT, Howley PM, and Tsai L-H (1998) p35, the neuronal- specific activator of cyclin-dependent kinase 5 (Cdk5) is degraded by the ubiquitin- proteasome pathway. J Biol Chem 273:24057-24064
Pentchev PG, Gal AE, Booth AD, Omodeo-Sale F, Fouks J, Neumeyer BA, Quirk JM, Dawson G, and Brady RO (1980) A lysosomal storage disorder in mice characterized by a dual deficiency of sphingomyelinase and glucocerebrosidase. Biochim Biophys Acta 619:669-679
Pentchev PG, Vanier MT, Suzuki K, and Patterson MC (1995) Niemann-Pick disease type C: A cellular cholesterol lipidosis. In: The Metabolic and Molecular Bases of Inherited Disease. CR Scriver, AL Beaudet, WS Sly, and D Valle (eds.), McGraw Hill, New York, pp. 2625-2639
Pfrieger FW (2003) Cholesterol homeostasis and function in neurons of the central nervous system. Cell Mol Life Sci 60:1158-1171
Phiel CJ, Wilson CA, Lee VM-Y, and Klein PS (2003) GSK-3α regulates production of Alzheimer’s disease amyloid-β peptides. Nature 423:435-439
Phillips WA and Avigan J (1963) Inhibition of cholesterol biosynthesis in the rat by 3β-(2-diethylamino-ethoxy)androst-5-en-17-one, hydrochloride. Proc Soc Exp Biol Med 112:233-236
Phizicky E, Bastiaens PIH, Zhu H, Snyder M, and Fields S (2003) Protein analysis on a proteomic scale. Nature 422:208-215
277 Pinkel D, Segraves R, Sudar D, Clark S, Poole I, Kowbel D, Collins C, Kuo W-L, Chen C, Zhai Y, Dairkee SH, Ljung B-M, Gray JW, and Albertson DG (1998) High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays. Nat Genet 20:207-211
Pollack JR, Perou CM, Alizadeh AA, Eisen MB, Pergamenschikov A, Williams CF, Jeffrey SS, Botstein D, and Brown PO (1999) Genome-wide analysis of DNA copy- number changes using cDNA microarrays. Nat Genet 23:41-46
Porter AG and Jänicke RU (1999) Emerging roles of caspase-3 in apoptosis. Cell Death Differ 6:99-104
Psyrri A, DeFilippis RA, Edwards APB, Yates KE, Manuelidis L, and DiMaio D (2004) Role of the retinoblastoma pathway in senescence triggered by repression of the human papillomavirus E7 protein in cervical carcinoma cells. Cancer Res 64:3079-3086
Puglielli L, Konopka G, Pack-Chung E, Ingano LAM, Berezovska O, Hyman BT, Chang TY, Tanzi RE, and Kovacs DM (2001) Acyl-coenzyme A: Cholesterol acyltransferase modulates the generation of the amyloid β-peptide. Nat Cell Biol 3:905-912
Qu D, Li Q, Lim H-Y, Cheung NS, Li R, Wang JH, and Qi RZ (2002) The protein SET binds the neuronal cdk5 activator p35nck5a and modulates cdk5/p35nck5a activity. J Biol Chem 277:7324-7332
Rabilloud T, Adessi C, Giraudel A, and Lunardi J (1997) Improvement of the solubilization of proteins in two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 18:307-316
Raina AK, Zhu X., Rottkamp CA, Monteiro M, Takeda A, and Smith MA (2000) Cyclin’ toward dementia: Cell cycle abnormalities and abortive oncogenesis in Alzheimer disease. J Neurosci Res 61:128-133
Rajavashisth TB, Kaptein JS, Reue KL, and Lusis AJ (1985) Evolution of apolipoprotein E: Mouse sequence and evidence for an 11-nucleotide ancestral unit. Proc Natl Acad Sci USA 82:8085-8089
Rametti A, Esclaire F, Yardin C, and Terro F (2004) Linking alterations in tau phosphorylation and cleavage during neuronal apoptosis. J Biol Chem 279:54518- 54528
Ran Q, Liang H, Gu M, Qi W, Walter CA, Roberts II LJ, Herman B, Richardson A, and Remmen HV (2004) Transgenic mice overexpressing glutathione peroxidase 4 are protected against oxidative stress-induced apoptosis. J Biol Chem 279:55137- 55146
278 Rapoport M, Dawson HN, Binder LI, Vitek MP, and Ferreira A (2002) Tau is essential to β-amyloid-induced neurotoxicity. Proc Natl Acad Sci USA 99:6364-6369
Raza H, Robin M-A, Fang J-K, and Avadhani NG (2002) Multiple isoforms of mitochondrial glutathione S-transferases and their differential induction under oxidative stress. Biochem J 366:45-55
Rideout HJ, Dietrich P, Wang Q, Dauer WT, and Stefanis L (2004) α-Synuclein is required for the fibrillar nature of ubiquitinated inclusions induced by proteasomal inhibition in primary neurons. J Biol Chem 279:46915-46920
Rissman RA, Poon WW, Blurton-Jones M, Oddo S, Torp R, Vitek MP, LaFerla FM, Rohn TT, and Cotman CW (2004) Caspase-cleavage of tau is an early event in Alzheimer disease tangle pathology. J Clin Invest 114:121-130
Robertson JD, Orrenius S, and Zhivotovsky B (2000) Review: Nuclear events in apoptosis. J Struct Biol 129:346-358
Rodriguez-Lafrasse C, Rousson R, Bonnet J, Pentchev PG, Louisot P, and Vanier MT (1990) Abnormal cholesterol metabolism in imipramine-treated fibroblast cultures. Similarities with Niemann-Pick type C disease. Biochim Biophys Acta 1043:123-128
Roff CF, Goldin E, Comly ME, Cooney A, Brown A, Vanier MT, Miller SPF, Brady RO, and Pentchev PG (1991) Type C Niemann-Pick disease: Use of hydrophobic amines to study defective cholesterol transport. Dev Neurosci 13:315-319
Rohn TT, Head E, Nesse WH, Cotman CW, and Cribbs DH (2001) Activation of caspase-8 in the Alzheimer’s disease brain. Neurobiol Dis 8:1006-1016
Rohn TT, Rissman RA, Davis MC, Kim YE, Cotman CW, and Head E (2002) Caspase-9 activation and caspase cleavage of tau in the Alzheimer’s disease brain. Neurobiol Dis 11:341-354
Rosenblat M, Draganov D, Watson CE, Bisgaier CL, La Du BN, and Aviram M (2003) Mouse macrophage paraoxonase 2 activity is increased whereas cellular paraoxonase 3 activity is decreased under oxidative stress. Arterioscler Thromb Vasc Biol 23:468-474
Roshon M, DeGregori JV, and Ruley HE (2003) Gene trap mutagenesis of hnRNP A2/B1: A cryptic 3’ splice site in the neomycin resistance gene allows continued expression of the disrupted cellular gene. BMC Genomics 4:2-12
Rosner H, Rebhan M, Vacun G, and Vanmechelen E (1994) Expression of a paired helical filament tau epitope in embryonic chicken central nervous system. Neuroreport 5:1164-1166
279 Rubin JB, Kung AL, Klein RS, Chan JA, Sun Y, Schmidt K, Kieran MW, Luster AD, and Segal RA (2003) A small-molecule antagonist of CXCR4 inhibits intracranial growth of primary brain tumors. Proc Natl Acad Sci USA 100:13513-13518
Rudney H and Sexton RC (1986) Regulation of cholesterol biosynthesis. Ann Rev Nutr 6:245-272
Runz H, Rietdorf J, Tomic I, de Bernard M, Beyreuther K, Pepperkok R, and Hartmann T (2002) Inhibition of intracellular cholesterol transport alters presenilin localization and amyloid precursor protein processing in neuronal cells. J Neurosci 22:1679-1689
Saito K-I, Elce JS, Hamos JE, and Nixon RA (1993) Widespread activation of calcium-activated neutral proteinase (calpain) in the brain in Alzheimer disease: A potential molecular basis for neuronal degeneration. Proc Natl Acad Sci USA 90:2628-2632
Saito T, Ishiguro K, Onuki R, Nagai Y, Kishimoto T, and Hisanaga S-I (1998) Okadaic acid-stimulated degradation of p35, an activator of CDK5, by proteasome in cultured neurons. Biochem Biophys Res Commun 252:775-778
Saito T, Onuki R, Fujita Y, Kusakawa G-I, Ishiguro K, Bibb JA, Kishimoto T, and Hisanaga S-I (2003) Developmental regulation of the proteolysis of the p35 cyclin- dependent kinase 5 activator by phosphorylation. J Neurosci 23:1189-1197
Salamon C, Chervenak M, Piatigorsky J, and Sax CM (1998) The mouse transketolase (TK) gene: Cloning, characterization, and functional promoter analysis. Genomics 48:209-220
Sandberg R, Yasuda R, Pankratz DG, Carter TA, Del Rio JA, Wodicka L, Mayford M, Lockhart DJ, and Barlow C (2000) Regional and strain-specific gene expression mapping in the adult mouse brain. Proc Natl Acad Sci USA 97:11038-11043
Santoni V, Rabilloud T, Doumas P, Rouquié D, Mansion M, Kieffer S, Garin J, and Rossignol M (1999) Towards the recovery of hydrophobic proteins on two- dimensional electrophoresis gels. Electrophoresis 20:705-711
Sarkar CP, Bierkamper GG, and Cenedella RJ (1982) Studies on the mechanism of the epileptiform activity induced by U18666A. I. Gross alteration of the lipids of synaptosomes and myelin. Epilepsia 23:243-255
Sastre J, Pallardo FV, and Vina J (2000) Mitochondrial oxidative stress plays a key role in aging and apoptosis. IUBMB Life 49:427-435
Sawamura N, Gong J-S, Garver WS, Heidenreich RA, Ninomiya H, Ohno K, Yanagisawa K, and Michikawa M (2001) Site-specific phosphorylation of tau
280 accompanied by activation of mitogen-activated protein kinase (MAPK) in brains of Niemann-Pick type C mice. J Biol Chem 276:10314-10319
Scheele GA (1975) Two-dimensional gel analysis of soluble proteins – Characterization of guinea pig exocrine pancreatic proteins. J Biol Chem 250:5375- 5385
Schena M, Shalon D, Davis RW, and Brown PO (1995) Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 270:467- 470
Scherer LJ and Rossi JJ (2003) Approaches for the sequence-specific knockdown of mRNA. Nat Biotechnol 21:1457-1465
Schneider A, Fischer A, Weber D, von Ahsen O, Scheek S, Krüger C, Rossner M, Klaussner B, Faucheron N, Kammandel B, Goetz B, Herrmann O, Bach A, and Schwaninger M (2004) Restriction-mediated differential display (RMDD) identifies pip92 as a pro-apoptotic gene product induced during focal cerebral ischemia. J Cereb Blood Flow Metab 24:224-236
Schneyer A, Sidis Y, Xia Y, Saito S, del Re E, Lin HY, and Keutmann H (2004) Differential actions of follistatin and follistatin-like 3. Mol Cell Endocrinol 225:25-28
Schulze A and Downward J (2001) Navigating gene expression using microarrays – A technology review. Nat Cell Biol 3:E190-E195
Scott C and Ioannou YA (2004) The NPC1 protein: Structure implies function. Biochim Biophys Acta 1685:8-13
Searle J, Kerr JFR, and Bishop CJ (1982) Necrosis and apoptosis: Distinct modes of cell death with fundamentally different significance. Pathol Annu 17:229-259
Selkoe DJ (2000) The origins of Alzheimer disease: A is for amyloid. JAMA 283:1615-1617
Serretti A, Artioli P, Quartesan R, and De Ronchi D (2005) Genes involved in Alzheimer’s disease, a survey of possible candidates. J Alzheimers Dis 7:331-353
Sexton RC, Panini SR, Azran F, and Rudney H (1983) Effects of 3β-[2- (diethylamino)ethoxy]androst-5-en-17-one on the synthesis of cholesterol and ubiquinone in rat intestinal epithelial cell cultures. Biochemistry 22:5687-5692
Shapiro LA, Marks A, Whitaker-Azmitia PM (2004) Increased clusterin expression in old but not young adult S100B transgenic mice: Evidence of neuropathological aging in a model of Down Syndrome. Brain Res 1010:17-21
281 Shaw MM and Riederer BM (2003) Sample preparation for two-dimensional gel electrophoresis. Proteomics 3:1408-1417
Sherr CJ (1995) D-type cyclins. Trends Biochem Sci 20:187-190
Sherr CJ (1996) Cancer cell cycles. Science 274:1672-1677
Sherr CJ and Roberts JM (1999) CDK inhibitors: Positive and negative regulators of G1-phase progression. Genes Dev 13:1501-1512
Shevchenko A, Loboda A, Shevchenko A, Ens W, and Standing KG (2000) MALDI quadrupole time-of-flight mass spectrometry: A powerful tool for proteomic research. Anal Chem 72:2132-2141
Shevchenko A, Wilm M, Vorm O, and Mann M (1996) Mass spectrometric sequencing of proteins from silver-stained polyacrylamide gels. Anal Chem 68:850- 858
Siman R, Card JP, and Davis LG (1990) Proteolytic processing of β-amyloid precursor by calpain I. J Neurosci 10:2400-2411
Siman R, Flood DG, Thinakaran G, and Neumar RW (2001) Endoplasmic reticulum stress-induced cysteine protease activation in cortical neurons. J Biol Chem 276:44736-44743
Singh U, Devaraj S, and Jialal I (2005) Vitamin E, oxidative stress, and inflammation. Annu Rev Nutr 25:151-174
Skovronsky DM, Doms RW, and Lee VM-Y (1998) Detection of a novel intraneuronal pool of insoluble amyloid beta protein that accumulates with time in culture. J Cell Biol 141:1031-1039
Sladowski D, Steer SJ, Clothier RH, and Balls M (1993) An improved MTT assay. J Immunol Methods 157:203-207
Sleat DE, Wiseman JA, El-Banna M, Price SM, Verot L, Shen MM, Tint GS, Vanier MT, Walkley SU, and Lobel P (2004) Genetic evidence for nonredundant functional cooperativity between NPC1 and NPC2 in lipid transport. Proc Natl Acad Sci USA 101:5886-5891
Slee EA, Harte MT, Kluck RM, Wolf BB, Casiano CA, Newmeyer DD, Wang H-G, Reed JC, Nicholson DW, Alnemri ES, Green DR, and Martin SJ (1999) Ordering the cytochrome c-initiated caspase cascade: Hierarchical activation of caspases-2, -3, -6, - 7, -8, and -10 in a caspase-9 dependent manner. J Cell Biol 144:281-292
282 Smith MA, Rottkamp CA, Nunomura A, Raina AK, and Perry G (2000) Oxidative stress in Alzheimer’s disease. Biochim Biophys Acta 1502:139-144
Sofer A and Futerman AH (1995) Cationic amphiphilic drugs inhibit the internalization of cholera toxin to the Golgi apparatus and the subsequent elevation of cyclic AMP. J Biol Chem 270:12117-12122
Sohocki MM, Bowne SJ, Sullivan LS, Blackshaw S, Cepko CL, Payne AM, Bhattacharya SS, Khaliq S, Mehdi SQ, Birch DG, Harrison WR, Elder FFB, Heckenlively JR, and Daiger SP (2000) Mutations in a new photoreceptor-pineal gene on 17p cause Leber congenital amaurosis. Nat Genet 24:79-83
Song J, Clair T, Noh JH, Eun JW, Ryu SY, Lee SN, Ahn YM, Kim SY, Lee SH, Park WS, Yoo NJ, Lee JY, and Nam SW (2005) Autotaxin (lysoPLD/NPP2) protects fibroblasts from apoptosis through its enzymatic product, lysophosphatidic acid, utilizing albumin-bound susbstrate. Biochem Biophys Res Commun 337:967-975
Sorimachi H, Ishiura S, and Suzuki K (1997) Structure and physiological function of calpains. Biochem J 328:721-732
Sparrow SM, Carter JM, Ridgway ND, Cook HW, and Byers DM (1999) U18666A inhibits intracellular cholesterol transport and neurotransmitter release in human neuroblastoma cells. Neurochem Res 24:69-77
Spencer JPE, Jenner A, Chimel K, Aruoma OI, Cross CE, Wu R, and Halliwell B (1995) DNA strand breakage and base modification induced by hydrogen peroxide treatment of human respiratory tract epithelial cells. FEBS Lett 374:233-236
Spencer JPE, Whiteman M, Jenner A, and Halliwell B (2000) Nitrite-induced deamination and hypochlorite-induced oxidation of DNA in intact human respiratory tract epithelial cells. Free Radic Biol Med 28:1039-1050
Stadtman ER and Levine RL (2000) Protein oxidation. Ann N Y Acad Sci 899:191- 208
Su JH, Cummings BJ, and Cotman CW (1994) Early phosphorylation of tau in Alzheimer’s disease occurs at Ser-202 and is preferentially located within neurites. Neuroreport 5:2358-2362
Sullivan PG and Brown MR (2005) Mitochondrial aging and dysfunction in Alzheimer’s disease. Prog Neuropsychopharmacol Biol Psychiatry 29:407-410
Suzuki K, Lareyre J-J, Sánchez D, Gutierrez G, Araki Y, Matusik RJ, and Orgebin- Crist M-C (2004) Molecular evolution of epididymal lipocalin genes localized on mouse chromosome 2. Genes 339:49-59
283 Suzuki K, Parker CC, Pentchev PG, Katz D, Ghetti B, D’Agostino AN, and Carstea ED (1995) Neurofibrillary tangles in Niemann-Pick disease type C. Acta Neuropathol (Berl) 89:227-238
Sylvain M, Arnold DL, Scriver CR, Schreiber R, and Shevell MI (1994) Magnetic resonance spectroscopy in Niemann-Pick disease type C: Correlation with diagnosis and clinical response to cholestyramine and lovastatin. Pediatr Neurol 10:228-232
Tanaka K (1998) Proteasomes: Structure and biology. J Biochem 123:195-204
Tangirala RK, Mahlberg FH, Glick JM, Jerome WG, and Rothblat GH (1993) Lysosomal accumulation of unesterified cholesterol in model macrophage foam cells. J Biol Chem 268:9653-9660
Taupenot L, Harper KL, and O’Connor DT (2003) Mechanisms of disease: The chromogranin-secretogranin family. N Engl J Med 348:1134-1149
Thiede B, Höhenwarter W, Krah A, Mattow J, Schmid M, Schmidt F, and Jungblut PR (2005) Peptide mass fingerprinting. Methods 35:237-247
Tilleman K, Van den Haute C, Geerts H, van Leuven F, Esmans EL, and Moens L (2002) Proteomics analysis of the neurodegeneration in the brain of tau transgenic mice. Proteomics 2:656-665
Tomiyama Y, Waguri S, Kanamori S, Kametaka S, Wakasugi M, Shibata M, Ebisu S, and Uchiyama Y (2004) Early-phase redistribution of the cation-independent mannose 6-phosphate receptor by U18666A treatment in HeLa cells. Cell Tissue Res 317:253-264
Treiber-Held S, Distl R, Meske V, Albert F, and Ohm TG (2003) Spatial and temporal distribution of intracellular free cholesterol in brains of a Niemann-Pick type C mouse model showing hyperphosphorylated tau protein. Implications for Alzheimer’s disease. J Pathol 200:95-103
Troy CM, Rabacchi SA, Friedman WJ, Frappier TF, Brown K, and Shelanski ML (2000) Caspase-2 mediates neuronal cell death induced by β-amyloid. J Neurosci 20:1386-1392
Tsugita A, Kawakami T, Uchida T, Sakai T, Kamo M, Matsui T, Watanabe Y, Morimasa T, Hosokawa K, and Toda T (2000) Proteome analysis of mouse brain: Two-dimensional electrophoresis profiles of tissue proteins during the course of aging. Electrophoresis 21:1853-1871
Tucker JM and Townsend DM (2005) Alpha-tocopherol: Roles in prevention and therapy of human disease. Biomed Pharmacother 59:380-387
284 Tyers M and Mann M (2003) From genomics to proteomics. Nature 422:193-197
Underwood KW, Andemariam B, McWilliams GL, and Liscum L (1996) Quantitative analysis of hydrophobic amine inhibition of intracellular cholesterol transport. J Lipid Res 37:1556-1568
Underwood KW, Jacobs NL, Howley A, and Liscum L (1998) Evidence for a cholesterol transport pathway from lysosomes to endoplasmic reticulum that is independent of the plasma membrane. J Biol Chem 273:4266-4274
Uryu S, Tokuhiro S, and Oda T (2003) β-Amyloid-specific upregulation of stearoyl coenzyme A desaturase-1 in macrophages. Biochem Biophys Res Commun 303:302- 305 van Deursen J, Schepens J, Peters W, Meijer D, Grosveld G, Hendriks W, and Wieringa B (1992) Genetic variability of the murine creatine kinase B gene locus and related pseudogenes in different inbred strains of mice. Genomics 12:340-349 van Engeland M, Nieland LJW, Ramaekers FCS, Schutte B, and Reutelingsperger CPM (1998) Annexin V-affinity assay: A review on an apoptosis detection system based on phosphatidylserine exposure. Cytometry 31:1-9
Vance JE, Hayashi H, and Karten B (2005) Cholesterol homeostasis in neurons and glial cells. Semin Cell Dev Biol 16:193-212
Vanier MT and Millat G (2003) Niemann-Pick disease type C. Clin Genet 64:269- 281
Villa PG, Henzel WJ, Sensenbrenner M, Henderson CE, and Pettmann B (1998) Calpain inhibitors, but not caspase inhibitors, prevent actin proteolysis and DNA fragmentation during apoptosis. J Cell Sci 111:713-722
Volpe JJ and Obert KA (1982) Interrelationships of ubiquinone and sterol syntheses in cultured cells of neural origin. J Neurochem 38:931-938
Vreugdenhil HA, Haitsma JJ, Jansen KJ, Zijlstra J, Plötz FB, van Dijk JE, Lachmann B, van Vught H, and Heijnen CJ (2003) Ventilator-induced heat shock protein 70 and cytokine mRNA expression in a model of lipopolysaccharide-induced lung inflammation. Intensive Care Med 29:915-922
Wang DG, Fan J-B, Siao C-J, Berno A, Young P, Sapolsky R, Ghandour G, Perkins N, Winchester E, Spencer J, Kruglyak L, Stein L, Hsie L, Topaloglou T, Hubbell E, Robinson E, Mittmann M, Morris MS, Shen N, Kilburn D, Rioux J, Nusbaum C, Rozen S, Hudson TJ, Lipshutz R, Chee M, and Lander ES (1998) Large-scale identification, mapping, and genotyping of single-nucleotide polymorphisms in the human genome. Science 280:1077-1082
285 Wang H and Joseph JA (1999) Quantifying cellular oxidative stress by dichlorofluorescein assay using microplate reader. Free Radic Biol Med 27:612-616
Wang KKW (2000) Calpain and caspase: Can you tell the difference? Trends Neurosci 23:20-26
Washburn MP, Koller A, Oshiro G, Ulaszek RR, Plouffe D, Deciu C, Winzeler E, and Yates III JR (2003) Protein pathway and complex clustering of correlated mRNA and protein expression analyses in Saccharomyces cerevisiae. Proc Natl Acad Sci USA 100:3107-3112
Wasinger VC, Cordwell SJ, Cerpa-Poljak A, Yan JX, Gooley AA, Wilkins MR, Duncan MW, Harris R, Williams KL, and Humphery-Smith I (1995) Progress with gene-product mapping of the Mollicutes: Mycoplasma genitalium. Electrophoresis 16:1090-1094
Weitzman JB, Fiette L, Matsuo K, and Yaniv M (2000) JunD protects cells from p53- dependent senescence and apoptosis. Mol Cell 6:1109-1119
White SL, Gharbi S, Bertani MF, Chan H-L, Waterfield MD, and Timms JF (2004) Cellular responses to ErbB-2 overexpression in human mammary luminal epithelial cells: comparison of mRNA and protein expression. Br J Cancer 90:173-181
Whiteman M, Hooper DC, Scott GS, Koprowski H, and Halliwell B (2002) Inhibition of hypochlorous acid-induced cellular toxicity by nitrite. Proc Natl Acad Sci USA 99:12061-12066
Williams KL (1999) Genomes and proteomes: Towards a multidimensional view of biology. Electrophoresis 20:678-688
Winer N, Klachko DM, Baer RD, Langley PL, and Burns TW (1966) Myotonic response induced by inhibitors of cholesterol biosynthesis. Science 153:312-313
Witting PK, Upston JM, and Stocker R (1998) The molecular action of α-tocopherol in lipoprotein lipid peroxidation: Pro- and antioxidant activity of vitamin E in complex heterogeneous lipid emulsions. Subcell Biochem 30:345-390
Wolozin B (2001) A fluid connection: Cholesterol and Aβ. Proc Natl Acad Sci USA 98:5371-5373
Wolozin B (2004) Cholesterol and the biology of Alzheimer’s disease. Neuron 41:7- 10
Wolozin B, Kellman W, Ruosseau P, Celesia GG, and Siegel G (2000) Decreased prevalence of Alzheimer disease associated with 3-hydroxy-3-methyglutaryl coenzyme A reductase inhibitors. Arch Neurol 57:1439-1443
286 Woodgett JR (1990) Molecular cloning and expression of glycogen synthase kinase- 3/factor A. EMBO J 9:2431-2438
Wu Y-P, Mizukami H, Matsuda J, Saito Y, Proia RL, and Suzuki K (2005) Apoptosis accompanied by up-regulation of TNF-α death pathway genes in the brain of Niemann-Pick type C disease. Mol Genet Metab 84:9-17
Wyllie AH, Kerr JFR, and Currie AR (1980) Cell death: The significance of apoptosis. Int Rev Cytol 68:251-306
Yadav S and Rawal UM (1992) Cholesterol and lipid peroxidation in 3β-(2- diethylaminoethoxy)-androst-5-en-17-oneHCl (U18666A) induced cataractogenesis in rats. Indian J Exp Biol 30:147-148
Yamazaki T, Chang T-Y, Haass C, and Ihara Y (2001) Accumulation and aggregation of amyloid β-protein in late endosomes of Niemann-Pick type C cells. J Biol Chem 276:4454-4460
Yancey PG, Rodrigueza WV, Kilsdonk EPC, Stoudt GW, Johnson WJ, Phillips MC, and Rothblat GH (1996) Cellular cholesterol efflux mediated by cyclodextrins. Demonstration of kinetic pools and mechanism of efflux. J Biol Chem 271:16026- 16034
Yeo W and Gautier J (2004) Early neural cell death: Dying to become neurons. Dev Biol 274:233-244
Yoshihara M and Montana ES (2004) The synaptotagmins: Calcium sensors for vesicular trafficking. Neuroscientist 10:566-574
You KH, Ji YM, and Kwon OY (2003) Clusterin overexpression is responsible for the anti-apoptosis effect in a mouse neuroblastoma cell line, B103. Z Naturforsch 58:148-151
Yu J-Y, DeRuiter SL, and Turner DL (2002) RNA interference by expression of short-interfering RNAs and hairpin RNAs in mammalian cells. Proc Natl Acad Sci USA 99:6047-6052
Yu W, Gong J-S, Ko M, Garver WS, Yanagisawa K, and Michikawa M (2005a) Altered cholesterol metabolism in Niemann-Pick type C1 mouse brains affects mitochondrial function. J Biol Chem 280:11731-11739
Yu W, Ko M, Yanagisawa K, and Michikawa M (2005b) Neurodegeneration in heterozygous Niemann-Pick type C1 (NPC1) mouse: Implication of heterozygous NPC1 mutations being a risk for tauopathy. J Biol Chem 280:27296-27302
287 Yuan J and Yankner BA (2000) Apoptosis in the nervous system. Nature 407:802- 809
Zhang FX and Hutchins JB (1997) Protein phosphorylation in response to PDGF stimulation in cultured neurons and astrocytes. Dev Brain Res 99:216-225
Zhang J, Zhang D, McQuade JS, Behbehani M, Tsien JZ, and Xu M (2002) c-fos regulates neuronal excitability and survival. Nat Genet 30:416-420
288
APPENDICES
289 Appendices
Appendix I Media for preparation of mouse primary cortical neurons
A Hanks’ balanced salts solution (HBSS) 1 M glucose (Merck Biosciences, #8342) 7.4 ml 100 mM sodium pyruvate (Sigma, #S8636) 10 ml 1 M HEPES buffer solution (Gibco, #15630-080) 10 ml 7.5% (w/v) sodium bicarbonate (Gibco, #25080-094) 4.7 ml Hanks’ balanced salts (Sigma, #H2387) 9.5 g Reconstitute in Milli-Q water to a final volume of 1 l, filter-sterilize and store at 4oC
B Solution 1 HBSS 250 ml 150 mM magnesium sulfate (Sigma, #M2773) 1.94 ml Bovine serum albumin (BSA; Sigma, #A4503) 0.75 g Filter-sterilize and incubate on ice
C Solution 2 Solution 1 20 ml 10 mg/ml deoxyribonuclease I (DNase I; Sigma, #D5025; 150,000 units) 80 μl Trypsin type IX-S from porcine pancreas (Sigma, #T0303) 4 mg Filter-sterilize and incubate at 37oC
D Solution 3 Solution 1 20 ml 150 mM magnesium sulfate (Sigma, #M2773) 200 μl 10 mg/ml DNase I 80 μl Trypsin inhibitor type II-S from soybean (Sigma, #T9128) 10.4 mg Filter-sterilize and incubate on ice
E Solution 4 Solution 1 16.8 ml Solution 3 3.2 ml Filter-sterilize and incubate on ice
F Composition of B27 supplement (Brewer et al., 1993) Biotin Linoleic acid Insulin Selenium Linolenic acid Albumin, bovine L-carnitine T3 (triodo-1-thyronine) Progesterone Corticosterone DL-α-tocopherol (vitamin E) Transferrin Ethanolamine DL-α-tocopherol acetate Retinyl acetate D(+)-galactose Superoxide dismutase Putrescine Glutathione (reduced) Catalase
290 Appendix II Buffers and reagents for SDS-PAGE and Western blotting
A Components of RIPA buffer 10 mM Tris-HCl pH 7.4 1% (v/v) Nonidet P-40 (Fluka, #56741) 0.5% (w/v) deoxycholate (Sigma, #D6750) 150 mM NaCl 1 mM EDTA 2 mM EGTA (Sigma, #E4378) 0.1% (v/v) SDS 25 mM sodium fluoride (Sigma, #S7920) 2 mM sodium orthovanadate (Alexis® Biochemicals, #400-032-G005) 10 mM sodium pyrophosphate (Sigma, #P8010) Protease inhibitors (Complete™ Mini; Roche Diagnostics, #1836153)
B Stacking gel formulation (Per 1.0 mm thick gel) 4% Milli-Q water 1.525 ml 30% Acrylamide/Bis solution (Bio-Rad Laboratories, #161-0158) 325 μl Stacking gel buffer 625 μl 10% (w/v) ammonium persulfate (APS; Bio-Rad Laboratories, #161-0700) 12.5 μl TEMED (Fluka, Buchs, Switzerland, #87689) 3.75 μl
C Resolving gel formulations (Per 1.0 mm thick gel) 6% 10% 15% Milli-Q water 2.7 ml 2.05 ml 1.2 ml 30% Acrylamide/Bis solution (Bio-Rad Laboratories) 1 ml 1.65 ml 2.5 ml Resolving gel buffer 1.25 ml 1.25 ml 1.25 ml 10% (w/v) APS (Bio-Rad Laboratories) 25 μl 25 μl 25 μl TEMED (Fluka) 5 μl 5 μl 5 μl
D Stacking gel buffer (4× Laemmli, pH 6.8) Tris base (J.T. Baker, Phillipsburg, NJ, USA, #4109-02) 12 g 10% (w/v) SDS 8 ml Adjust to pH 6.8 with 5 M HCl (Merck, Darmstadt, Germany, #1.00317) Top to 200 ml with Milli-Q water
E Resolving gel buffer (4× Laemmli, pH 8.8) Tris base (J.T. Baker) 36.3 g 10% (w/v) SDS 8 ml Adjust to pH 8.8 with 5 M HCl Top to 200 ml with Milli-Q water
291 F Components of 5× SDS loading buffer 0.5 M Tris-HCl pH 6.8 20% (v/v) glycerol (J.T. Baker, #2136-1) 10% (w/v) SDS (Sigma, #L5750) 0.01% (w/v) bromophenol blue (Sigma, #B5525) 20% (v/v) β-mercaptoethanol (Sigma, #M6250; add fresh)
G 10× Transfer buffer Tris base (J.T. Baker) 30.285 g Glycine (BDH, Dorset, England, #101196X) 144.13 g In 1 l Milli-Q water
H 1× Transfer buffer 10× Transfer buffer 100 ml Milli-Q water 700 ml Methanol (Mallinckrodt, Phillipsburg, NJ, USA, #3016) 200 ml
I 5× Tris-buffered saline (TBS) pH 7.5 Tris base (J.T. Baker) 30.285 g NaCl (Merck, Darmstadt, Germany, #1.06404) 43.83 g Adjust to pH 7.5 with 5 M HCl Top to 1 l with Milli-Q water
J Tris-buffered saline containing Tween® 20 (TBST) 5× TBS pH 7.5 200 ml Milli-Q water 799 ml Tween® 20 (Sigma, #P1379) 1 ml
292 Appendix III Master mixes and buffers for microarray analysis
A First-Strand Master Mix (Per sample) 5× First-strand reaction mix 4 μl 0.1 M DTT 2 μl 10 mM dNTP 1 μl All materials were supplied in the one-cycle cDNA synthesis kit (Affymetrix).
B Second-Strand Master Mix (Per sample) RNase-free water 91 μl 5× Second-strand reaction mix 30 μl 10 mM dNTP 3 μl Escherichia coli DNA ligase 1 μl Escherichia coli DNA polymerase I 4 μl RNase H 1 μl All materials were supplied in the one-cycle cDNA synthesis kit (Affymetrix).
C IVT Reaction Master Mix (Per sample) Template cDNA 6 μl RNase-free water 14 μl 10× IVT labeling buffer 4 μl IVT labeling NTP mix 12 μl IVT labeling enzyme mix 4 μl All unspecified materials were supplied in the IVT labeling kit (Affymetrix).
D 12× 2-(N-morpholino)ethanesulfonic acid (MES) MES hydrate (Sigma, #M5287) 64.61 g MES sodium salt (Sigma, #M5057) 193.3 g DEPC-treated water (Ambion, #9920) 800 ml Adjust pH to 6.5-6.7 and top to 1 l with DEPC-treated water Filter and store at 4oC in dark
E 2× Hybridization buffer (50 ml) 12× MES 8.3 ml 5 M NaCl (Ambion, #9760G) 17.7 ml 0.5 M EDTA (Sigma, #E7889) 4 ml 10% (v/v) Tween® 20 (Pierce Biotechnology, #28320) 100 μl DEPC-treated water (Ambion, #9920) 19.9 ml Store at 4oC in dark
293 F 2× Stain buffer (250 ml) 12× MES 41.7 ml 5 M NaCl (Ambion) 92.5 ml 10% (v/v) Tween® 20 (Pierce Biotechnology) 2.5 ml DEPC-treated water (Ambion) 113.3 ml Filter and store at 4oC in dark
G SAPE solution mix (Stains 1 and 3) (Per probe array) 2× Stain buffer 600 μl 50 mg/ml BSA (Invitrogen, #15561-020) 48 μl 1 mg/ml SAPE (Molecular Probes, #S866) 12 μl DEPC-treated water (Ambion) 540 μl
H Antibody solution mix (Stain 2) (Per probe array) 2× Stain buffer 300 μl 50 mg/ml BSA (Invitrogen) 24 μl 10 mg/ml goat IgG (Sigma, #I5256) 6 μl 0.5 mg/ml biotinylated antibody 3.6 μl (Vector Laboratories, Burlingame, CA, USA, #BA0500) DEPC-treated water (Ambion) 266.4 μl
I Wash buffer A (Non-stringent wash buffer) 20× SSPE (Invitrogen, #15591-043) 300 ml 10% (v/v) Tween® 20 (Pierce Biotechnology) 1 ml DEPC-treated water (Ambion) 699 ml Filter and store at room temperature
J Wash buffer B (Stringent wash buffer) 12× MES 83.3 ml 5 M NaCl (Ambion) 5.2 ml 10% (v/v) Tween® 20 (Pierce Biotechnology) 1 ml DEPC-treated water (Ambion) 910.5 ml Filter and store at 4oC in dark
K Composition of 20× SSPE (Invitrogen) 3 M NaCl 0.2 M NaH2PO4 0.02 M EDTA
294 Appendix IV Gel formulations and solutions for proteomics analysis
A Polyacrylamide gel formulations (Per 1.0 mm thick gel) 8% 12% Milli-Q water 2.4 ml 13.2 ml 30% Acrylamide/Bis solution (Bio-Rad Laboratories) 1.3 ml 16 ml Resolving gel buffer (refer to Appendix IIE) 1.25 ml 10.4 ml 10% (w/v) APS (Bio-Rad Laboratories) 25 μl 200 μl TEMED (Fluka) 5 μl 40 μl Note: 8% gel formulation is based on the Mini-PROTEAN® 3 cell electrophoresis system; 12% gel formulation is based on the PROTEAN® II xi 2D cell electrophoresis system
B Fixative solution Methanol (Merck, #1.06009) 500 ml Glacial acetic acid (Merck, #1.00063) 50 ml Milli-Q water 450 ml
C Developing solution Sodium carbonate (Merck, #1.06392) 20 g 37% (w/v) formaldehyde (Sigma, #F1635) 400 μl Milli-Q water 999.6 ml
D Stop solution EDTA (Bio-Rad Laboratories, #161-0729) 14.6 g Milli-Q water 1 l
295