UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Chronic Inflammation as the Mechanistic Link between Type 2 Diabetes Mellitus and Alzheimer's Disease Bahniwal, Manpreet Kaur 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_september_Bahniwal_Manpreet.pdf [ 2.19MB ]
Metadata
JSON: 24-1.0074350.json
JSON-LD: 24-1.0074350-ld.json
RDF/XML (Pretty): 24-1.0074350-rdf.xml
RDF/JSON: 24-1.0074350-rdf.json
Turtle: 24-1.0074350-turtle.txt
N-Triples: 24-1.0074350-rdf-ntriples.txt
Original Record: 24-1.0074350-source.json
Full Text
24-1.0074350-fulltext.txt
Citation
24-1.0074350.ris

Full Text

       Chronic Inflammation as the Mechanistic Link between Type 2 Diabetes Mellitus and Alzheimer’s Disease  by  Manpreet Kaur Bahniwal  B.Sc. (Hons.), The University of British Columbia, 2010      A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   MASTER OF SCIENCE  in  THE COLLEGE OF GRADUATE STUDIES  (Biology)   THE UNIVERSITY OF BRITISH COLUMBIA (Okanagan)    July 2014       © Manpreet Kaur Bahniwal, 2014	   ii	  Abstract     Recent research has identified type 2 diabetes mellitus (T2DM) as a risk factor for neurodegenerative disorders such as Alzheimer’s disease (AD). However, the mechanisms involved in this interaction still remain unclear. Chronic neuroinflammation caused by activation of glial cells in the brain contributes to neuronal loss and disease progression in AD. This thesis investigated if chronic inflammation could be a mechanistic link responsible for increasing the risk of AD in individuals with T2DM. Specifically, I hypothesized that high levels of glucose and the phenomenon of insulin resistance observed in T2DM could accelerate neuronal loss and eventually lead to AD by increasing glial cell activation and/or enhancing neuronal injury caused by disease-specific agents. Since astrocytes are the most abundant glial cell type in the brain, in vitro experiments were conducted using primary human astrocytes and U-118 MG human astrocytic cells as models of primary astrocytes. I found that supernatants of astrocytic cells incubated in high glucose (30.5 mM) were more toxic to neuronal cells pre-treated with high glucose compared to supernatants of astrocytic cells incubated in low glucose (5.5 mM). High glucose increased mRNA expression of the pro-inflammatory cytokine interleukin (IL)-6 and enhanced secretion of both IL-6 and IL-8 by all astrocytic cells. Data obtained indicated that increased activation of the p38 mitogen activated protein kinase (MAPK) may be mediating the effects of high glucose in astrocytic cells. In addition, high glucose increased the susceptibility of undifferentiated human SH-SY5Y neuronal cells and retinoic-acid differentiated SH-SY5Y cells to injury by hydrogen peroxide and fibrillar amyloid beta-42 protein (Aβ42), respectively. Human primary glial 	   iii	  cells and astroglial cell lines expressed the insulin receptor (INSR) and its signaling components. Moreover, insulin was found to modulate secretion of pro-inflammatory cytokines by primary human astrocytes with 1nM insulin concentration causing maximum enhancement of IL-6 and IL-8 secretion. Therefore, high glucose and insulin levels in T2DM could contribute to an early appearance of AD-like symptoms by increasing glial cell-mediated inflammation. This research highlights novel mechanisms responsible for AD progression and could help identify new preventative and treatment strategies for AD.                           	   iv	  Preface    The data described in this thesis have not been published anywhere yet. Select findings were however presented at the Experimental Biology Conference in Boston, USA and UBC Okanagan Biochemistry and Molecular Biology Departmental seminars. I obtained appropriate certification and training to work with Biosafety Level 1 and Level 2 materials including various human cell lines and primary human astrocytes. I am responsible for all the experimental data presented in this thesis, except for the primer optimization experiments (Fig. 3) that were performed by Jocelyn Madeira. Figure 23B was kindly provided by Lindsay Spielman.                            	   v	  Table of Contents  Abstract .............................................................................................................................. ii Preface ............................................................................................................................... iv Table of Contents .............................................................................................................. v List of Tables ................................................................................................................... viii List of Figures ................................................................................................................... ix List of Abbreviations ........................................................................................................ xi Acknowledgements .......................................................................................................... xv Dedication ....................................................................................................................... xvi Chapter 1. Introduction .................................................................................................... 1      1.1 Alzheimer’s disease .................................................................................................. 1      1.2 Inflammatory hypothesis of neurodegeneration ....................................................... 2            1.2.1 Cells of the central nervous system ................................................................. 2            1.2.2 Neuroinflammation in AD .............................................................................. 3       1.3 Type 2 diabetes mellitus (T2DM) as a risk for AD .................................................. 6      1.4 Chronic low-grade inflammation in T2DM ............................................................. 8      1.5 Chronic inflammation: underlying link between metabolic disorders             and AD ................................................................................................................... 10      1.6 Hyperglycemia ....................................................................................................... 10      1.7 Insulin in the periphery and the brain ..................................................................... 14            1.7.1 Insulin signaling pathways ............................................................................ 15            1.7.2 Hyperinsulinemia .......................................................................................... 16            1.7.3 Anti-inflammatory effects of insulin in the periphery .................................. 18            1.7.4 Pro-inflammatory effects of insulin .............................................................. 19      1.8 Cell culture models ................................................................................................. 23      1.9 Research overview and hypotheses ........................................................................ 25 Chapter 2. Materials and Methods ................................................................................ 28      2.1 Chemicals and reagents .......................................................................................... 28      2.2 Equipment and supplies ......................................................................................... 30      2.3 Cell culture models ................................................................................................. 31      2.4 Establishing toxicity of stimulated U-118 MG cell supernatants towards             SH-SY5Y neuronal cells ........................................................................................ 32      2.5 Preparation of U-118MG supernatants with different glucose             concentrations ......................................................................................................... 34      2.6 LDH release assay .................................................................................................. 35      2.7 MIT assay ............................................................................................................... 36      2.8 Effects of high glucose on mRNA expression of pro-inflammatory cytokines            by astrocytic cells ................................................................................................... 37            2.8.1 RNA extraction ............................................................................................. 37            2.8.2 RNA spin column protocol ........................................................................... 38            2.8.3 cDNA synthesis ............................................................................................. 40            2.8.4 Real-time quantitative polymerase chain reaction (qPCR) ........................... 40      2.9 Effects of high glucose on release of pro-inflammatory cytokines by astrocytic            cells ......................................................................................................................... 42            2.9.1 Collection of supernatants from U-118 MG cell, U-373 MG cell and 	   vi	                      primary human astrocytes incubated in media containing different                     glucose concentrations .................................................................................. 42             2.9.2 ELISA ........................................................................................................... 43     2.10 Western blotting .................................................................................................... 46              2.10.1 Protein extraction ...................................................................................... 46              2.10.2 Sodium dodecyl sulfate-polyacrylamide gel electrophoresis                         (SDS-PAGE) .............................................................................................. 47              2.10.3 Electrophoretic transfer and protein detection .......................................... 49     2.11 Toxicity of hydrogen peroxide and Aβ42 towards SH-SY5Y cells in the             presence of different glucose concentrations ........................................................ 51              2.11.1 Undifferentiated SH-SY5Y neuronal cells ................................................ 51              2.11.2 Differentiated SH-SY5Y neuronal cells .................................................... 52     2.12 Expression of insulin signaling components by glial cells ................................... 53              2.12.1 RNA extraction and cDNA synthesis ........................................................ 53              2.12.2 Primers ...................................................................................................... 53              2.12.3 Polymerase chain reaction (PCR) ............................................................. 55              2.12.4 Polyacrylamide gel electrophoresis (PAGE) ............................................. 56     2.13 Effects of varying insulin concentrations on the release of pro-inflammatory              cytokines by primary human astrocytes ................................................................ 57     2.14 Statistical analysis ................................................................................................. 58 Chapter 3. Results ........................................................................................................... 59      3.1 Toxicity of stimulated U-118 MG astrocytic cells towards SH-SY5Y             neuronal cells .......................................................................................................... 59      3.2 Toxicity of U-118 MG cells incubated in different glucose concentrations            towards SH-SY5Ycells .......................................................................................... 63      3.3 Effects of high glucose on pro-inflammatory cytokines gene expression and            secretion by astrocytic cells .................................................................................... 69             3.3.1 Gene expression of pro-inflammatory cytokines ......................................... 69             3.3.2 Secretion of pro-inflammatory cytokines ..................................................... 73      3.4 Intracellular signaling pathways activated by high glucose in astrocytic             cells ......................................................................................................................... 78      3.5 Effect of high glucose on susceptibility of neurons to a cytotoxic insult by            hydrogen peroxide or Aβ42 .................................................................................... 81             3.5.1 Undifferentiated SH-SY5Y cells .................................................................. 81             3.5.2 Differentiated SH-SY5Y cells ...................................................................... 84      3.6 Expression of insulin signaling components in glia-like cell lines and primary             glial cells ................................................................................................................ 88      3.7 Effect of insulin on secretion of pro-inflammatory cytokines by primary              human astrocytes .................................................................................................... 91 Chapter 4. Discussion ...................................................................................................... 94      4.1 Effects of high glucose on toxicity of astrocytic cells towards neuronal             cells ......................................................................................................................... 94      4.2 Effects of high glucose on astrocytic cell-mediated inflammation and             intracellular signaling pathways ........................................................................... 100             4.2.1 High glucose enhances gene expression and secretion of pro-inflammatory                      cytokines by astrocytic cells ....................................................................... 100 	   vii	              4.2.2 Effects of high glucose on activation of intracellular signaling pathway in                      astrocytic cells ............................................................................................ 102      4.3 Direct effects of high glucose on neuronal cells .................................................. 104      4.4 Insulin signaling components are present in glia-like cell lines and primary             glial cells .............................................................................................................. 110      4.5 Insulin modulates secretion of pro-inflammatory cytokines by primary             human astrocytes .................................................................................................. 112 Chapter 5. Conclusions and Future Work .................................................................. 114      5.1 Limitations of Research ....................................................................................... 114      5.2 Future Work ......................................................................................................... 115      5.3 Significance of findings ....................................................................................... 116 References ...................................................................................................................... 118 Appendix A .................................................................................................................... 136    	   	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	  	   	   viii	  List of Tables  Table 1: Glucose concentration in various media used in experiments ........................... 29  Table 2: Primer sequences for qPCR experiments ........................................................... 41 Table 3: Sequences of primers used in the experiments ................................................... 55                                       	   ix	  List of Figures  Figure 1: Mechanisms of hyperglycemia/ hyperlipidemia –induced inflammation,                 oxidative stress and insulin resistance in adipocytes and macrophages ............ 13  Figure 2: Hyperinsulinemia and insulin resistance in T2DM contribute to AD                  pathology ........................................................................................................... 17  Figure 3: Primer-efficiency data for IL-6: (A) Melt curve; (B) Amplification curve;                  and (C) Standard curve obtained during optimization experiments by                  Madeira, J .......................................................................................................... 42  Figure 4: Primers for INSRA and INSRB target two different isoforms of INSR .......... 54  Figure 5: Effect of 48 h stimulation on U-118 MG astrocytic cell viability .................... 61  Figure 6: Viability of SH-SY5Y cells treated with supernatants from U-118 MG                  cells stimulated for 48 h with increasing concentrations of IFN-γ in the                          presence or absence of  100 U/ml of IL-1β ....................................................... 62  Figure 7: Viability of SH-SY5Y cells treated with supernatants from U-118 MG                  cells that were left unstimulated or were stimulated with combination                  of IFN-γ (150 U/ml) and IL-1β (100 U/ml) in the presence of different                  glucose concentrations (5.5 and 30.5 mM) ....................................................... 64  Figure 8: Time-dependent glucose consumption by U-118 MG cells (A) and                  SH-SY5Y cells (B) placed in different glucose concentrations ........................ 66  Figure 9: Viability of SH-SY5Y cells treated with supernatants from U-118 MG                  cells that were left unstimulated or were stimulated with a combination                  of IFN-γ (150 U/ml) and IL-1β (100 U/ml) in the presence of different                  glucose concentrations (5.5, 10.5, 20.5 and 30.5 mM) for 48 h ........................ 67  Figure 10: Viability of SH-SY5Y cells treated with supernatants from U-118 MG                    cells that were left unstimulated or were stimulated with a combination                    of IFN-γ (150 U/ml) and IL-1β (100 U/ml) in the presence of different                    glucose concentrations (5.5, 10.5, 20.5 and 30.5 mM) for 48 h ...................... 68  Figure 11: Effect of high glucose on IL-6 (A) and IL-8 (B) mRNA expression by                   unstimulated and stimulated U-118 MG astrocytic cells ................................ 71  Figure 12: Effect of high glucose on IL-6 (A) and IL-8 (B) mRNA expression by                      unstimulated and stimulated human astrocytes ............................................... 72  Figure 13: Effect of high glucose on secretion of pro-inflammatory cytokines                    IL-6 (A) and IL-8 (B) by U-118 MG astrocytic cells ..................................... 74 	   x	   Figure 14: Effect of high glucose on secretion of pro-inflammatory cytokines                    IL-6 (A) and IL-8 (B) by U-373 MG astrocytic cells ..................................... 76  Figure 15: Effect of high glucose on secretion of pro-inflammatory cytokines                    IL-6 (A) and IL-8 (B) by primary human astrocytes ...................................... 77  Figure 16: Western blots for total STAT-3 (A) and JNK (B) obtained by separating                    U-118 MG cell lysates on 8% SDS-PAGE and immunoblotting                    using specific antibodies ................................................................................. 79  Figure 17: Western blots for phospho- and total p38 MAPK (A) and p44/42                    MAPK obtained by separating U-118 MG cell lysates on 8%                    SDS-PAGE and immunoblotting using specific antibodies ............................ 80  Figure 18: Viability of SH-SY5Y neuronal cells treated with hydrogen peroxide                    in the presence of different glucose concentrations ........................................ 83  Figure 19: Phase contrast images of SH-SY5Y neuronal cells at day 1 (A),                    day 4 (B) and day 7 (C) of differentiation with 10 µM retinoic acid .............. 85  Figure 20: Viability of retinoic acid-differentiated SH-SY5Y neuronal cells                    treated with hydrogen peroxide in the presence of different glucose                    concentrations .................................................................................................. 86  Figure 21: Viability of retinoic acid-differentiated SH-SY5Y neuronal cells treated                    with Aβ42 in the presence of different glucose concentrations ...................... 87  Figure 22: DNA polyacrylamide gels showing gene expression of different insulin                   receptors and receptor signaling molecules by SH-SY5Y neuroblastoma                    cells (left) and HepG2 hepatocytoma cells (right) .......................................... 89  Figure 23: DNA polyacrylamide gels showing gene expression of different insulin                   receptors and receptor signaling molecules by U-118 MG astrocytic                    cells (A), primary human astrocytes (B; left) and primary human                    microglia (B; right) .......................................................................................... 90  Figure 24: IL-6 secretion (A) by primary human astrocytes stimulated with IFN-γ                   (150 U/ml) and IL-1β (100 U/ml) in the presence of different insulin                      concentrations (1 pM to 1 µM) for 48 h .......................................................... 92  Figure 25: IL-8 secretion (A) by primary human astrocytes stimulated with IFN-γ                   (150 U/ml) and IL-1β (100 U/ml) in the presence of different insulin                   concentrations (1 pM to 1 µM) for 48 h .......................................................... 93  Figure 26: Mechanisms of hyperglycemia–induced neuronal toxicity .......................... 106 	   xi	   List of Abbreviations  Aβ- Amyloid beta  AD- Alzheimer’s disease AGEs- Advanced glycation end products  ANOVA- Analysis of variance BBB- Blood brain barrier BLAST- Basic alignment search tool  β-NAD- Beta-nicotinamide adenine dinucleotide bp- Base pair  BSA- Bovine serum albumin cDNA- Complementary DNA CI- Confidence interval CNS- Central nervous system CSF- Cerebrospinal fluid Ct- Cycle threshold DAG- Diacylglycerol DHAP- Dihydroxyacetone phosphate DMEM-F12- Dulbecco’s modified Eagle medium nutrient mixture F-12 Ham DMF- N-dimethylformamide DMSO- Dimethyl sulfoxide EDTA- Ethylenediaminetetraacetic acid ELISA- Enzyme linked immunosorbent assay ER- Endoplasmic reticulum 	   xii	  FBS- Fetal bovine serum GFAP- Glial fibrillary acidic protein GLUT- Glucose transporters GSK- Glycogen synthase kinase GSH- Glutathione GSSG- Glutathione disulphide H2O- Water H2O2-  Hydrogen peroxide HSD- Honestly significant difference ICAM- Intracellular adhesion molecule IDE- Insulin degrading enzyme IDT- Integrated DNA technologies IGF- Insulin-like growth factor IkB - Inhibitor of NFkB IkKβ- Inhibitor of NFkB kinase subunit beta IL- Interleukin INSR- Insulin receptor INSRA- INSR isoform A INSRB- INSR isoform B INT- Iodonitrotetrazolium chloride IRS- Insulin receptor substrate JNK- Janus kinase LAL- Limulus amebocyte lysate 	   xiii	  LDH- Lactate dehydrogenase LSD- Least-significant difference MAPKs- Mitogen-activated protein kinases MCP- Monocyte chemotactic protein MIF- Migration-inhibitory factor MIQE- Minimum information for publication of quantitative real-time PCR experiments MMP- Matrix metalloproteinase  MPTP- 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine MTT- 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide NADP- Nicotinamide adenine dinucleotide phosphate NADPH- Reduced nicotinamide adenine dinucleotide phosphate NCBI- National centre for biotechnology information NFkB- Nuclear factor kappa B NFTs- Neurofibrillary tangles  NRT- No reverse-transcriptase NSAIDs- Non-steroidal anti-inflammatory drugs NTC- No template control O2.- - Superoxide anion O.D.- Optical density .OH- Hydroxyl radical PAGE- Polyacrylamide gel electrophoresis PBMCs- Peripheral blood mononuclear cells PBS- Phosphate-buffered saline 	   xiv	  PCR- Polymerase chain reaction PD- Parkinson’s disease  PI3K- Phosphatidylinositol 3-kinase PKC- Protein kinase C qPCR- Quantitative polymerase chain reaction  RIPA- Radio immune precipitation assay ROS- Reactive oxygen species RR- Relative risk RT-PCR- Reverse transcription-polymerase chain reaction S.E.M.- Standard error of the mean SDS- Sodium dodecyl sulphate STAT- Signal transducers and activators of transcription T2DM- Type 2 diabetes mellitus TBS- Tris-buffered saline TBS-T- TBS-0.1% tween TBE- Tris-borate-EDTA TCA- Trichloroacetic acid TEMED- Tetramethylethylenediamine TGF- Transforming growth factor TNF- Tumor necrosis factor  TREM- Triggering receptor expressed on myeloid cells UDP-GlcNAc- Uridine diphosphate-N-acetylglucosamine   	   xv	  Acknowledgements    I extend my sincerest gratitude to my supervisor, Dr. Andis Klegeris, who taught me the qualities of patience and perseverance and guided me through every step of the way. Thank you so much for making sure that I was happy, safe and busy during my program. I learnt a great deal about teamwork and professionalism working with you.    Thank you to Drs. Sanjoy Ghosh, Deanna Gibson and Mark Rheault and their laboratory personnel for allowing me to use their laboratory equipment. Special thanks to Amy Botta for providing me with guidance about my western blotting experiments. Thanks to my committee members Dr. Bruce Mathieson and Dr. Sanjoy Ghosh for their valuable comments and suggestions. A very special thanks to Dr. Jonathan P. Little, my co-supervisor, who taught me various techniques; helped me troubleshoot and interpret my data; and allowed me to use antibodies for western blotting experiments. A very special thank you to Dr. Joyce Boon for her continued support and encouragement. Most importantly, I thank my colleagues at the Laboratory of Cellular and Molecular Pharmacology for helping me to carry on through all circumstances and always being there for me.    Finally, I thank my family and friends for supporting all my decisions and providing me a fresh perspective on difficult situations in my life. A special thanks to my husband, Sameer for his unconditional love and support.    	   xvi	  Dedicated to my dearest husband and his family                                     	   1	  Chapter 1. Introduction   1.1 Alzheimer’s disease (AD)  Neurodegenerative disorders are incurable, debilitating conditions of the central nervous system (CNS) that lead to cognitive impairment and inability to independently carry out daily tasks, thereby severely compromising the quality of life. Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by progressive decline in the ability to form new memories and access existing ones due to extensive neuronal loss in the hippocampus and cerebral cortex of the brain (Block et al., 2007; Honjo et al., 2012). AD affects more than 34 million people in the world and is the most common cause of dementia (60-70% of all cases) (Barnes & Yaffe, 2011). The incidence of AD increases with age and the decline in cognitive function is more than what would be expected in normal aging (Cummings, 2004). The prevalence of AD is expected to triple over the next forty years due to longer life expectancies and demographic changes, placing an enormous burden on healthcare systems (Cummings, 2004; Barnes & Yaffe, 2011). Therefore, research targeting neurodegeneration is necessary for designing therapeutic and preventative strategies to meet the needs of the aging population and reduce overall health care costs.  The main histopathological features of AD include intracellular neurofibrillary tangles (NFTs) composed of hyper-phosphorylated tau protein and extracellular amyloid plaques composed mainly of beta-amyloid protein (Aβ) deposits produced by the proteolytic cleavage of the amyloid precursor protein (Cummings, 2004; LaFeria et al., 2007; Bird, 2008; De Felice et al., 2008; Iqbal & Grundke-Iqbal, 2008; Mondragon-	   2	  Rodriguez et al., 2012). NFTs and Aβ plaques are also found in healthy non-AD brains. However, the density of NFTs and plaques is significantly higher in the brains of individuals with AD (Cummings, 2004). The accumulation of hyper-phosphorylated tau protein deposits and amyloid plaques is known to be a central event in the degeneration of neurons.   1.2 Inflammatory hypothesis of neurodegeneration   An important hallmark of a number of neurodegenerative disorders including AD, Parkinson’s disease (PD) and amyotrophic lateral sclerosis is the presence of chronic neuroinflammation in the brain (Sastre et al., 2006; Frank-Cannon et al., 2009; Krause & Muller, 2010). Although inflammation is likely not the initiating event in AD pathology, it plays a critical role in disease progression.   1.2.1 Cells of the central nervous system   Neurons are the basic functional unit of the nervous system. As excitable cells, neurons process and transmit information in the form of electrochemical signals by forming connections with other neurons. Different types of glial cells in the CNS provide support and protection to neurons.  Glial cells are classified into two main categories: microglia and macroglia. Microglia, the resident tissue macrophages in the CNS, belong to the mononuclear phagocyte system and represent the primary immune effector cells of the brain 	   3	  (Tambuyzer et al., 2009). They are derived from progenitor cells in the bone marrow and comprise 5-20% of all glial cells in the CNS (Lawson et al., 1990; Lawson et al., 1992). Microglia are capable of both pro-inflammatory and anti-inflammatory functioning to protect the brain from foreign pathogens (Tambuyzer et al., 2009; Kettenmann & Verkhratsky, 2011). They scavenge cellular debris; secrete neurotrophic substances essential for neuronal growth and survival; and function in repair and remodeling of the CNS (Hanisch & Kettenmann, 2007; Napoli & Neumann, 2009).  Macroglia comprise a heterogeneous population of cells and are further subdivided into four different cell types including: a) ependymal cells, which form the lining of the ventricles in the CNS; b) oligodendrocytes and c) Schwann cells, which are involved in myelination of axonal processes in the CNS and peripheral nervous system respectively; and d) astrocytes, which are star-shaped glial cells that provide biochemical (lactate, growth factors etc.) and structural support to neurons (Maragakis & Rothstein, 2006). Among all glial cells, astrocytes are the most abundant and versatile and are involved in an array of different functions including glutamate recycling, glycogen storage, secretion of neurotrophic factors, maintenance of the blood brain barrier, regulation of extracellular ion concentrations and modulation of synaptic transmission (Araque et al., 1999; Kurosinski & Gotz, 2002; Caudle, 2006; Ransom & Ransom, 2012).  1.2.2  Neuroinflammation in AD  Neuroinflammation in AD is primarily mediated by glial cells that become activated in response to pathological formations (e.g., Aβ plaques in AD) and release a 	   4	  host of toxic and pro-inflammatory substances in an attempt to dispose of the formations (Block et al., 2007; Hashioka et al., 2009; Parpura et al., 2012; Smith et al., 2012; Steele & Robinson, 2012). As discussed earlier, under normal circumstances, glial cells function in providing support and protection to neurons. However, due to the constant presence of pathological formations in AD, the uncontrolled activation of glial cells contributes to amplification of the immune response causing a state of chronic neuroinflammation in the brain and substantial damage to the surrounding neurons. Activated microglia initiate several inflammatory events that may cause accelerated neuronal death including: a) production of pro-inflammatory cytokines tumor necrosis factor alpha (TNF- α), interleukin (IL)-6 and IL-8; b) release of neurotoxins such as glutamate and proteases; c) production of free radicals including reactive oxygen and nitrogen species; and d) complement activation leading to assembly of the membrane attack complex (Block et al., 2007). Similar to microglia, astrocytes can become reactive and contribute to inflammation and neuronal death (Fuller et al., 2009; Hashioka et al., 2009; Garwood et al., 2011). The resulting neuronal damage can, in turn, lead to activation of more glial cells, leading to a vicious cycle of neuroinflammation and cell death even after the removal of the initial trigger.  Several studies support the presence of chronic neuroinflammation in AD and suggest that neuroinflammation not only contributes to neuronal death but also causes an increase in the number of pathological formations, which in turn perpetuates inflammation (Sastre et al., 2006; Frank-Cannon et al., 2009; Krause & Muller, 2010). Various lines of epidemiological and clinical evidence have implicated inflammation in the progression of AD. Long-term consumption of non-steroidal anti-inflammatory drugs 	   5	  (NSAIDs) to treat arthritis or other symptoms has been reported to reduce risk of AD (Lindsay et al., 2002; Zandi et al., 2002). Increased levels of pro-inflammatory cytokines including IL-1, IL-6 and TNF-α (Wyss-Coray, 2006; Pellicano et al., 2010) as well as activation of the classical complement pathway response (Eikelenboom et al., 1989; Fonseca et al., 2004) have been demonstrated in the brains of AD patients as well as rodent models of the disease. Amyloid plaques in the post-mortem AD brains are surrounded by activated microglia and astrocytes (Lue et al., 2001). Moreover, in vivo imaging studies have demonstrated inverse correlations between activation of microglia in AD patients and scores on the Mini Mental State Examination (Edison et al., 2008). According to recent genetic studies, immune/microglia module is strongly associated with late-onset AD and the triggering receptor expressed on myeloid cells (TREM)-2 (microglial receptor) has been identified as one of the most potent risk modifiers (Guerreiro et al., 2013; Jonsson et al., 2013; Zhang et al., 2013).  Increased levels of C-reactive protein (a marker of acute inflammation) have been correlated with the formation of senile plaques in AD (Strang et al., 2012). Moreover, there is much evidence to suggest the involvement of systemic inflammation in AD (Holmes, 2013). Increases in the serum levels of TNF-α (Bonotis et al., 2008) and altered peripheral immune responses, including increased secretion of IL-1β and IL-6 by peripheral blood mononuclear cells (PBMCs) (Reale et al., 2005) have been reported in AD. Therefore, it is believed that channels of communication exist between peripheral and central inflammation in AD.  The multifaceted nature of AD and the complex mechanisms of neuroinflammation and neurodegeneration have made it extremely difficult to find a cure. 	   6	  Several promising new drugs have recently failed in phase 3 clinical trials (Green et al., 2009; Quinn et al., 2010). Therefore, in addition to treatment strategies, new research initiatives are focusing on identification of risk factors and preventative measures for AD.   1.3 Type 2 diabetes mellitus (T2DM) as a risk factor for AD 	  A number of environmental and genetic factors have been shown to increase the risk of AD (Lambert et al., 2009; Barnes & Yaffe, 2011). Recently, T2DM has emerged as a significant risk factor for AD (Biessels et al., 2006; Lu et al., 2009). T2DM is a complex metabolic disorder characterized by high circulating levels of blood glucose due to loss of insulin sensitivity (Lin & Holscher, 2007). T2DM is often a consequence of obesity in people who are genetically predisposed to the disease and is associated with microvascular and macrovascular complications including cardiovascular disease, blindness, kidney failure and neuropathy (Lin & Holscher, 2007). Approximately 347 million people in the world and 20 million people in the US have diabetes; T2DM comprises 90% of all diabetes cases throughout the world (WHO, 2012). The prevalence of diabetes is increasing at an alarming rate of 5% every year (WHO, 2012). Research in the past decade has associated diabetes with an increased risk of AD and vascular dementia (Biessels et al., 2006; Lu et al., 2009). Based on an analysis of nine high quality studies, it was determined that individuals with probable T2DM present a nearly two fold higher risk of developing AD compared to individuals without diabetes (Biessels et al., 2006). A meta-analysis of eight population-based studies found a significant increase in the risk of AD and all-cause dementia in patients with diabetes: relative risk (RR) was 	   7	  estimated to be 1.39 with a confidence interval (CI) 1.17-1.66 for AD and 1.47 (CI: 1.25-1.73) for all-cause dementia (Lu et al., 2009). A recent meta-analysis of fifteen epidemiological studies concluded that T2DM is a risk factor for probable AD (since definite diagnosis of AD is only possible upon postmortem neuropathological examination) with a pooled RR of 1.57 (CI: 1.41-1.75) (Vagelatos & Eslick, 2013). Moreover, adjustment for the cardiovascular risk factors that could contribute to the association of T2DM and AD led to an increased risk estimate, thereby indicating that T2DM is involved in the pathogenesis of AD independently of vascular risk factors. However, neuropathological studies do not seem to support the epidemiological evidence, leading some researchers to argue that T2DM is more likely contributing to cerebral infarcts rather than AD-type pathology (Peila et al., 2002; Moroz et al., 2008; Ahtiluoto et al., 2010). Moreover, randomized controlled trials testing the effect of T2DM treatments on cognitive outcomes have shown inconclusive results or no evidence of a beneficial effect (Launer et al., 2011). Differences in study designs and study participants as well as lack of objective measures of cognitive functions could be responsible for such a disparity between the results. Nevertheless, the association between T2DM and neurodegeneration warrants extensive prevention efforts to effectively control the projected increase of dementia in the growing number of elderly individuals who have, or at risk for, T2DM.  Interestingly, many characteristics/consequences of T2DM including cardiovascular disease, obesity, hyperlipidemia and hypertension have also been identified as risk factors for AD (Martins et al., 2006). High-fat feeding models used for induction of insulin resistance in rodents are known to impair neurocognitive function 	   8	  (Murray et al., 2009). The underlying mechanisms that increase the risk of AD in individuals with T2DM are not fully understood but may be related to impaired glucose homeostasis, insulin resistance and high levels of circulating fatty acids (Lin & Holscher, 2007; Sims-Robinson et al., 2010). The increased prevalence of diabetes combined with the current obesity epidemic in developed countries, aging population and increased risk of AD in diabetic patients is bound to negatively impact the well being of human populations in the coming years. A review based on population attributable risk scores suggests that lowering prevalence of diabetes by 25% could potentially prevent 203,000 cases of AD worldwide and 42,000 cases in the US (Barnes & Yaffe, 2011).   1.4 Chronic low-grade inflammation in T2DM   Chronic inflammation is associated with metabolic disorders of obesity, insulin resistance and T2DM. Professor Ebstein first reported the involvement of inflammation in T2DM in 1876 by showing that the anti-inflammatory drug sodium salicylate reduced glycosuria in diabetic patients (Ebstein, 1876). It is believed that inflammatory markers are increased before the onset of T2DM; indicating that inflammation occurs early during the period of impaired glucose tolerance. Moreover, chronic low-grade inflammation is known to contribute to a state of insulin resistance in obesity and T2DM (Lin & Holscher, 2007; Granic et al., 2009).  Inflammation in obesity and T2DM is characterized by monocyte/macrophage infiltration into adipose tissue and release of pro-inflammatory cytokines including IL-1, IL-6 and TNF-α by macrophages and adipocytes (Wellen & Hotamisligil, 2005; King, 	   9	  2008). Among these cytokines, TNF-α is known to be a key player in the induction of insulin resistance in the adipose tissue through activation of mitogen-activated protein kinases (MAPKs) and impairment of insulin signaling at the level of the insulin receptor substrates (IRS) (King, 2008; Nieto-Vazquez et al., 2008). T2DM is associated with systemic low-grade elevation of circulating markers and mediators of inflammation including C-reactive protein, IL-6, IL-1β, TNF-α and fibrinogen (Barzilay et al., 2001; Granic et al., 2009). Moreover, such elevations of circulating pro-inflammatory mediators are associated with cardiovascular complications of T2DM as well as an increased risk of gestational diabetes in pregnant women (Tracy et al., 1997; Retnakaran et al., 2003). Acute phase markers of inflammation including intracellular adhesion molecule (ICAM)-1, monocyte chemotactic protein (MCP)-1 and TNF-α have been implicated in the development of diabetic nephropathy (Fornoni et al., 2008). It has also been suggested that a number of drugs targeting diabetic nephropathy including angiotensin receptor blockers and angiotensin converting enzyme inhibitors could exert anti-inflammatory effects (Wolf et al., 2002). Chronic inflammatory conditions such as rheumatoid arthritis may increase the risk for T2DM (Sattar et al., 2003). Drugs targeting inflammatory pathways such as the nuclear factor kappa B (NFkB) pathway in humans improve insulin sensitivity (Kim et al., 2001; Hundal et al., 2002). In addition, inhibition of inflammatory mediators such as TNF-α and Janus kinase (JNK) pathways protects against insulin resistance in animal models of obesity (Uysal et al., 1997; Hirosumi et al., 2002).    	   10	  1.5 Chronic inflammation: underlying link between metabolic disorders and AD  Progression of T2DM is associated with high circulating levels of glucose, insulin and fatty acids, all of which contribute to inflammation in the periphery. As indicated earlier, there is evidence to suggest cross talk between peripheral and central inflammation. In fact, levels of IL-6 are increased in the brains of patients with T2DM, indicating increased central inflammation in T2DM (Sonnen et al., 2009). Therefore, chronic low-grade peripheral inflammation in T2DM may contribute to earlier appearance of AD-related pathology by enhancing neuroinflammatory processes in the CNS. However, it is also possible that the high levels of glucose and the phenomenon of insulin resistance in the brain could directly cause activation of glial cells and enhance neuroimmune responses, thereby accelerating neuronal death.    1.6 Hyperglycemia   The human brain is a metabolically active organ and uses glucose as its primary fuel. Metabolism of glucose in the brain is primarily oxidative (McCall, 2004). Various isoforms of glucose transporters (GLUT) allow entry of glucose into the brain across the blood brain barrier (BBB) and subsequently into the individual brain cells including glia and neurons (McCall, 2004). Brain glucose concentrations are approximately 15 – 20% of blood glucose concentrations (Gruetter et al., 1998). Hence, during severe uncompensated diabetes, where glucose concentrations in the blood can reach 20 mM, brain glucose concentrations can rise to 4.5 mM (Powers, 1981).  	   11	  Hyperglycemia (elevated levels of blood glucose) in T2DM is a manifestation of insulin resistance (Sims-Robinson et al., 2010). Hyperglycemia is a characteristic feature of T2DM and is one of the first symptoms to manifest before diagnosis of the disease. It is reported that several complications of diabetes including diabetic nephropathy, neuropathy and retinopathy are due to the inability of the respective cells to decrease glucose transport into the cell under the conditions of hyperglycemia (Fornoni et al., 2008; King, 2008). High levels of intracellular glucose increase flux of glucose through various metabolic pathways (e.g., sorbitol-aldose reductase pathway) since glucose levels are more than what can be used in glycolysis. This leads to increased formation of advanced glycation end products (AGEs); increased production of reactive oxygen species (ROS); activation of protein kinase C (PKC); and upregulation of transforming growth factor (TGF)-β (Fornoni et al., 2008; King, 2008; Sims-Robinson et al., 2010). Such pathways initiated by hyperglycemia contribute to inflammation, oxidative stress, mitochondrial dysfunction, and accumulation of extracellular matrix leading to tissue scarring and damage (Fornoni et al., 2008; Sharma et al., 2010).   Similar mechanisms are responsible for the negative effects of hyperglycemia on neuronal function (Sims-Robinson et al., 2010). AGEs are heterogeneous molecules formed by irreversible, non-enzymatic reactions between ketones or aldehydes and amino groups of proteins (Sato et al., 2006). Hyperglycemia is known to cause enhanced release of free oxygen radicals that act as intermediates in the AGE pathway (Omori et al., 2008). AGEs cause aggregation of Aβ and formation of NFTs thus possibly contributing to AD progression (Sato et al., 2006). As discussed earlier, hyperglycemia leads to mitochondrial dysfunction and increased release of reactive oxygen species resulting in 	   12	  oxidative stress, which may further contribute to neuronal death and AD pathology (Omori et al., 2008).  Inflammation in obesity and T2DM is characterized by infiltration of monocytes/macrophages into the adipose tissue and release of various pro-inflammatory mediators by macrophages and adipocytes. Macrophages and adipocytes share many similar functions and their co-localization in the inflamed adipose tissue reflects an overlap of metabolic and inflammatory signaling pathways in the periphery (Wellen & Hotamisligil, 2005). In both macrophages and adipocytes, hyperglycemia - and/or hyperlipidemia -induced mitochondrial and endoplasmic reticulum stress causes activation of JNK and PKC-9 (Ozcan et al., 2004; Lin et al., 2005) (Fig. 1). JNKs are tyrosine kinases belonging to the MAPK family. When activated, both JNK and PKC-9 phosphorylate IRS-1 on serine residue-307 leading to impairment of insulin signaling (Ozcan et al., 2004). JNK increases expression of various inflammatory genes through activation of the transcription factor AP-1 and signal transducers and activators of transcription (STAT) (Davis, 2000). On the other hand, PKC-9 activates IkKβ (inhibitor of NFkB kinase subunit beta), which phosphorylates IkB (inhibitor of NFkB). This causes activation of NFkB and consequent upregulation of multiple inflammatory mediators including IL-6 and TNF-α (Wellen & Hotamisligil, 2005; King, 2008). This onset of inflammation in the adipose tissue is a critical contributor to the state of insulin resistance in obesity and T2DM. In conclusion, hyperglycemia may exacerbate inflammation through activation of inflammatory pathways in cells capable of an immune response.   	   13	   Figure 1: Mechanisms of hyperglycemia/ hyperlipidemia –induced inflammation, oxidative stress and insulin resistance in adipocytes and macrophages. AGEs (advanced glycation end products); ROS (reactive oxygen species); ER (endoplasmic reticulum); INSR (insulin receptor); IRS (insulin receptor substrate); JNK (janus kinase); PKC (protein kinase C); NFkB (nuclear factor kappa B); STAT (signal transducers and activators of transcription).   Recent research efforts have focused on the effects of hyperglycemia on inflammatory signaling in peripheral monocytes. In vitro hyperglycemia (15-35 mM) enhances production of pro-inflammatory cytokines such as IL-1β and induces expression of toll like receptors in human monocytes via multiple mechanisms including the activation of PKC, NFkB and nicotinamide adenine dinucleotide phosphate (NADPH)- dependent oxidase pathways (Dasu et al., 2007; Dasu et al., 2008). Similar pathways are involved in hyperglycemia-induced release of pro-inflammatory cytokines TNF-α, IL-6, IL-8 and MCP-1 from rat microglia (Quan et al., 2007; Quan et al., 2011). The effect of  	   14	  hyperglycemia on activation and cytokine release of astrocytes has not been studied widely. However, Acheampong et al., 2009 demonstrated that hyperglycemia caused an increase in complement factor C3 expression in the U87 astrocytic cell line and total nitrate production in primary human fetal astrocytes (Acheampong et al., 2009).  1.7 Insulin in the periphery and the brain   Insulin, produced by the pancreas, is a hormone that normalizes rising blood glucose levels in the periphery. It belongs to the insulin family of structurally related proteins that includes insulin-like growth factor (IGF)-1 and IGF-2. Insulin signals the classical insulin-sensitive tissues such as liver, muscle and adipose to take up glucose and store it as glycogen (Hallschmid & Schultes, 2009). Traditionally, the brain was assumed to be insensitive to the effects of insulin. However, recent research has established that peripheral insulin crosses the BBB via an active receptor-mediated transport system (Banks, 2004). In response to intravenous insulin infusions (plasma insulin increase from a mean basal level of ~77 pM to a mean level of ~1750 pM), the cerebrospinal fluid insulin concentration increases from a mean basal level of ~6 pM to a mean of ~18 pM (Genuth, 1973; Wallum et al., 1987). It has not been established conclusively whether the brain produces insulin (Banks, 2004). In the brain, insulin improves cognition, enhances memory formation and acts as an adiposity signal by reducing food intake and promoting energy expenditure (Hallschmid & Schultes, 2009). Insulin is also believed to regulate synaptic plasticity and circuit development (Hallschmid & Schultes, 2009).  	   15	  1.7.1 Insulin signaling pathways    Insulin signals via the insulin receptor (INSR) to activate two different signaling pathways: MAPK pathway and the phosphatidylinositol 3-kinase (PI3K) pathway (Kim et al., 2006). The MAPK pathway is responsible for the mitogenic effects of insulin, including protein synthesis and cell growth. On the other hand, the PI3K pathway is involved in the metabolic effects of insulin by inducing translocation of GLUT-4 to the cell membrane, which facilitates glucose uptake into cells (Neumann et al., 2008; Duarte et al., 2012).  INSR is a heterotetrameric glycoprotein belonging to the tyrosine kinase family of growth factor receptors. It is composed of two extracellular ligand-binding α subunits (120-135 kDa), linked by disulphide bonds to two membrane-spanning β subunits (92-97 kDa) (Lee & Pilch, 1994). Two isoforms of the INSR exist due to alternative splicing of the transcript: INSR isoform A (INSRA) and INSR isoform B (INSRB) (Seino & Bell, 1989). These isoforms differ by the presence of exon 11 in INSRB. The adipose tissue has the highest expression level of the INSR, followed by about 30% relative expression in the liver, heart and lungs, and about 10% in the brain, muscle and kidney (Bailyes et al., 1997). INSRB accounts for 80% of the total INSR expression in the liver and 60% of the total INSR expression in the muscle and adipose tissue (Benecke et al., 1992). In general, INSRB is primarily involved in metabolic insulin signaling in differentiated tissues (Belfiore et al., 2009). Unlike INSRB, INSRA is a high affinity receptor for IGF-2; upregulation of INSRA is associated with increased IGF-2 signaling and decreased metabolic insulin signaling (Belfiore et al., 2009). 	   16	  INSR is distributed in a widespread but selective pattern in the brain and is more enriched in neurons compared to glia (Laron, 2009). Insulin and/or IGF-1/2 stimulation lead to activation of the INSR and downstream insulin signaling pathways in the neurons of the CNS (Duarte et al., 2012). However, specific insulin signaling in glia has not been studied extensively. A recent study demonstrated that insulin treatment in normal human astrocytes derived from the fetal brain led to a dose-dependent increase in Akt phosphorylation and PI3K activity (Heni et al., 2011). Therefore, the insulin-signaling cascade is functional in human astrocytes. Moreover, insulin stimulated cell proliferation and glycogen storage in astrocytes (Heni et al., 2011). To the best of our knowledge, specific insulin signaling components in microglial cells have not been reported.   1.7.2  Hyperinsulinemia   Hyperinsulinemia and insulin resistance in T2DM is believed to contribute to AD pathology by two different mechanisms: 1) downregulation of the INSR and its signaling molecules leading to increased activation of the glycogen synthase kinase (GSK)-3β which promotes tau phosphorylation and formation of NFTs (Lesort & Johnson, 2000; Freude et al., 2005) and 2) accumulation of Aβ caused by high levels of insulin competing with Aβ for the insulin degrading enzyme (IDE)  (Gasparini & Xu, 2003) (Fig. 2). Insulin infusions in healthy adults to reach plasma insulin levels typical of insulin resistance, while maintaining euglycemia, led to increased plasma and CSF levels of pro-inflammatory cytokines such as IL-1α, IL-1β, TNF-α and IL-6 as well as increased plasma and cerebrospinal fluid (CSF) levels of Aβ42 (Fishel et al., 2005). Accumulation 	   17	  of Aβ42 in the brain contributes to neurotoxicity in AD (de la Monte, 2012). Therefore, peripheral hyperinsulinemia may not only contribute to peripheral and central inflammation but may also lead to accumulation of Aβ42, thus contributing to AD pathogenesis.    Figure 2: Hyperinsulinemia and insulin resistance in T2DM contribute to AD pathology. Aβ (amyloid β protein); PI3K (phosphatidylinositol 3-kinase); GSK-3β (glycogen synthase kinase-3β); NFTs (neurofibrillary tangles).   Interestingly, patients with moderate to severe AD might have increased fasting plasma insulin levels and decreased CSF insulin levels compared to healthy controls resulting in a relative CNS insulin deficit (Craft et al., 1998). A similar scenario has also been noted in obesity (Kern et al., 2006). Impaired trans-endothelial insulin transport across the BBB is thought to be responsible for the reduced CNS/plasma insulin ratio (Hallschmid & Schultes, 2009). Intranasal insulin administration could override the 	   18	  impaired BBB transport and has been shown to improve cognition in patients with mild to moderate AD (Reger et al., 2008; Craft et al., 2012). However, resistance of various brain regions to the actions of insulin could still contribute to disease progression since several components of the PI3K insulin signaling pathway are deficient in AD and T2DM (Jolivalt et al., 2008; Liu et al., 2011). Therefore, defective insulin signaling seems to be a common denominator of metabolic and neurodegenerative disorders.   1.7.3 Anti-inflammatory effects of insulin in the periphery   The widely accepted anti-inflammatory effects of insulin were first observed in human umbilical vein endothelial cells, in which insulin caused a dose-dependent increase in the expression of the endothelial nitric oxide synthase (Aljada & Dandona, 2000) and release of nitric oxide (Zeng & Quon, 1996). Physiological concentrations of insulin are known to decrease expression of various pro-inflammatory mediators including ICAM-1, MCP-1 and the pro-inflammatory transcription factor, NFkB in human aortic endothelial cells (Aljada et al., 2000; Aljada et al., 2001). Animal studies have demonstrated suppression of endotoxin-induced pro-inflammatory transcription factors and gene expression by insulin (Brix-Christensen et al., 2004; Jeschke et al., 2004). In addition, insulin suppresses macrophage migration-inhibitory factor (MIF) expression by adipocytes (Sakaue et al., 1999), TNF-α generation by peritoneal exudate cells (Satomi et al., 1985) and carrageenan-induced inflammation in rats (Ottlecz et al., 1977). In vivo experiments involving low-dose insulin infusions in obese subjects provided solid evidence for an anti-inflammatory effect of insulin. Insulin was shown to 	   19	  reduce ROS generation as well as p47phox (cytosolic protein involved in NADPH oxidase activation) and NFkB expression in mononuclear cells (Dandona et al., 2001). Moreover, insulin infusions, while maintaining normoglycemia, decreased plasma concentrations of ICAM-1, MCP-1, matrix metalloproteinase (MMP)-2, MMP-9 and plasminogen activator inhibitor-1 (Dandona et al., 2001; Aljada et al., 2002). Insulin caused a 40% reduction in plasma C-reactive protein and serum amyloid A concentrations in patients with acute myocardial infarction (Chaudhuri et al., 2004). Moreover, anti-inflammatory, anti-oxidant and anti-thrombotic effects of low-dose insulin infusions, which are independent of decreases in glucose concentrations, have also been observed in intensive care unit patients and patients undergoing coronary artery bypass (Dandona et al., 2007a; Dandona et al., 2007b).   1.7.4 Pro-inflammatory effects of insulin   Recent studies investigating the effect of exogenous and endogenous insulin on atherosclerotic processes including foam cell formation, smooth muscle cell proliferation and coagulation have challenged the well-established anti-inflammatory role of insulin. It has been proposed that hyperinsulinemia can act as a pro-inflammatory and pro-atherogenic factor, thus contributing to atherosclerosis in diabetes and other insulin-resistant states (DeFronzo, 2010; Rensing et al., 2011). Conversely, others believe that such cardiovascular complications are a result of the insulin resistant state and not hyperinsulinemia per se (Garg et al., 2003; Dandona et al., 2007a). High insulin concentrations are compensatory to insulin resistance (Rensing et al., 2011). Disruptions 	   20	  in the PI3K/Akt insulin signaling pathway due to insulin resistance lead to impaired glucose uptake by metabolic tissues and high glucose concentrations in the blood. Consequently, more insulin is produced by the pancreas leading to a state of hyperinsulinemia.  Literature concerning insulin-mediated regulation of immune responses of peripheral monocytes and macrophages derives from the research investigating the effects of hyperinsulinemia on atherosclerosis and mortalities related to adverse cardiovascular events. Transformation of macrophages to foam cells contributes to inflammation and plays a key role in atherosclerosis (Takahashi et al., 2002; Moore & Tabas, 2011). Macrophages from obese and type 2 diabetic patients are insulin resistant and show increased expression of various inflammatory markers (Liang et al., 2004). Inhibition of macrophage apoptosis by insulin contributes to instability of atherosclerotic plaques in vivo (Iida et al., 2002). Moreover, high concentrations of insulin suppress high-density lipoprotein-mediated cholesterol efflux from macrophages by inhibiting various enzymes involved in this process, thereby contributing to foam cell formation (Yamashita et al., 2010).  The effects of insulin on the inflammatory status of monocytes and macrophages could delineate some of the possible effects of insulin on microglial immune responses. High insulin concentrations in vitro (100 nM) cause a dose-dependent increase in gene expression and secretion of pro-inflammatory cytokines TNF-α (Iida et al., 2001) and IL-8 (Wurm et al., 2008) in THP-1 cell-derived macrophages and primary human monocytes, respectively. Similar insulin concentrations lead to augmentation of MMP-9 expression in THP-1 monocytic cells (Fischoeder et al., 2007). MMP-9 induction 	   21	  facilitates accelerated chemotaxis of monocytes and macrophages (Kappert et al., 2008). These effects of insulin at high concentrations on inflammatory mediators in monocytes and macrophages are mediated by the MAPK insulin signaling pathway and are not suppressed by inhibition of the PI3K pathway. Hyperinsulinemia is also known to induce NADPH oxidase-dependent superoxide production in human monocytes and murine macrophages via insulin receptor-mediated activation of PI3K and PKC (San Jose et al., 2009). Activation of the NADPH oxidase turns on p38 MAPK and NFkB signaling pathways leading to proliferation of the cells and activation of MMP-9 (San Jose et al., 2009).  Interestingly, activation of THP-1 cell-derived macrophages by lipopolysaccharide (LPS) (6 h), followed by insulin treatment (100 nM) for 1 h and 24 h leads to significantly higher production of TNF-α and IL-6 by insulin-treated cells compared to cells treated with LPS alone (Brundage et al., 2008). On the other hand, 24 h pre-treatment of THP-1 cell-derived macrophages with insulin (100 nM) attenuates the production of TNF-α and IL-8, when the cells are subsequently challenged with LPS (Cuschieri et al., 2008). Another study showed that insulin alone did not affect production of IL-6 and TNF-α in THP-1 monocytes (Bunn et al., 2010). However, insulin combined with the saturated non-esterified fatty acid palmitate led to more IL-6 and TNF-α production compared to palmitate alone (Bunn et al., 2010). This synergistic action of insulin on the release of pro-inflammatory cytokines was mediated by the MAPK signaling pathway. Using HuH7 hepatocytes, Iwasaki et al., 2009 showed that insulin may exert short-term anti-inflammatory and long-term pro-inflammatory effects. Insulin alone (1 nM) did not affect NFkB-mediated transcription of pro-inflammatory 	   22	  genes. However, simultaneous treatment of HuH7 hepatocytes with insulin and TNF-α led to a dual response: short-term (6 h) suppression and long-term (36 h) stimulation of NFkB mediated transcription (Iwasaki et al., 2009). The former effect was mediated by the PI3K signaling pathway, while the latter was mediated by the MAPK signaling and led to a seven-fold increase in TNF-α-stimulated NFkB-dependent transcription by insulin.  As discussed earlier, insulin at low concentrations exerts anti-inflammatory effects on monocyte and macrophage-mediated inflammation. Conversely, high insulin concentrations seem to act in a pro-inflammatory manner. The effects of different insulin concentrations on immune responses of microglia and astrocytes have not been studied. However, we can hypothesize that similar to monocytes and macrophages, an insulin concentration threshold exists that may determine the pro-inflammatory vs. anti-inflammatory effects of insulin on microglial immune responses. The same pattern may also be characteristic of insulin-induced responses of astrocytes. Moreover, based on the evidence presented above, the effects of insulin on monocytes and macrophages not only depend on the concentration and time of administration of insulin, but also on the presence of other stimulatory agents. Therefore, insulin may exacerbate glial cell mediated neuroinflammation in chronic inflammatory conditions such as AD due to the presence of several other inflammatory triggers and the ongoing cycle of inflammation.  The anti-inflammatory vs. pro-inflammatory effects of insulin in the CNS have not been investigated. If the PI3K insulin signaling pathway is deficient in AD, insulin sensitizers could increase insulin signaling and help sustain cognition, reduce tau phosphorylation and decrease levels of insulin available to compete with Aβ degradation. 	   23	  However, if insulin acts on the glial cells via the unaffected MAPK pathway to increase inflammation in the brain, the neuroprotective effects of insulin could be overridden by the neuronal damage caused by increased inflammation. Therefore, the effects of insulin on the glial cells must be studied in detail before any further conclusions can be drawn with regard to the benefits of insulin therapy in AD.    1.8 Cell culture models   Immortalized cell lines were used in this thesis work to model cells of the CNS. Derived from cells of a particular tissue, such cell lines represent a population of cells that have undergone some form of mutation to allow them to grow for prolonged periods of time in vitro. A number of characteristics make these cell lines invaluable for research in biochemistry and cell biology: 1) They are inexpensive; 2) readily available; 3) relatively easy to use and maintain in the laboratory; and 4) represent a simple model system where the effect of one or multiple variables can be systematically manipulated. However, due to mutations, cell lines may differ from the original cell type in terms of morphology, functions and expression of different receptors and signaling molecules. Therefore, important discoveries/observations made by using cell lines were replicated in primary cells and/or different models that retain more features of the original cell type. As opposed to cell lines, primary cells are obtained directly from the human tissues. Primary human glial cells are difficult to obtain, they proliferate slowly and have a limited life span. However, they are considered better models of glial cells compared to immortalized cell lines.  	   24	  The experiments in this thesis involved in vitro cell culture studies using the U-118 MG and/or the U-373 MG human astrocytoma cell lines to model human astrocytes. Primary human astrocytes were also used to replicate important experiments. Both U-118 MG and U-373 MG cells have different characteristics in terms of their growth rates and morphologies; however they produce similar neurotoxic responses when stimulated and possess properties similar to long-term cultured astrocytes. The U-373 MG cells and primary human astrocytes can be stimulated using interferon (IFN)-γ to induce their activation and cytotoxicity towards neuronal cells (Hashioka et al., 2009). While stimulants such as LPS and IFN-γ are not known to be present in most CNS neuroinflammatory conditions, the cytotoxic response and glial cell phenotype induced by such stimuli is similar to that induced by Aβ in the brain (Klegeris et al., 1999; Hashioka et al., 2009). As demonstrated in this thesis, U-118 MG cells require co-stimulation with IFN-γ and IL-1β to induce maximal cytotoxic response.  The SH-SY5Y human neuroblastoma cell line was used to model neurons. The SH-SY5Y cells are neuroblastic cells with multiple fine and short cell processes. A number of studies have used SH-SY5Y cells and U-373 MG cells to observe glia-neuronal interactions (Hashioka et al., 2009; Hashioka et al., 2011a; Little et al., 2012). Another model of neurons involved differentiation of SH-SY5Y cells with retinoic acid. Such treatment has been shown to give rise to fully differentiated neurons that possess many of the characteristics of primary neuronal cultures (Lopes et al., 2010).    	   25	  1.9 Research overview and hypotheses   Neuroinflammation mediated by the glial cells of the brain, including microglia and astrocytes, plays a critical role in the progression of AD. Both these glial types contribute to homeostasis and neuroprotection under physiological conditions. However, glial cells are activated by specific molecules present in AD (such as Aβ) leading to secretion of a host of pro-inflammatory and toxic substances designed to destroy foreign agents. Due to the non-specific nature of innate immune responses, such pathological activation of glial cells results in collateral damage to neurons, which in turn leads to further activation of glial cells and a state of chronic neuroinflammation in the brain.  T2DM, a metabolic disorder characterized by the presence of systemic low-grade inflammation, insulin resistance and elevated circulating levels of glucose, has been linked to an increased risk of developing AD. However, the mechanisms responsible for this link are still under investigation. I focused on chronic inflammation as the possible mechanistic link between T2DM and AD. The central hypothesis of this thesis is that insulin resistance and hyperglycemia, observed in metabolic disorders, increase glial cell-mediated neuroinflammation and enhance neuronal injury caused by disease-specific agents, which eventually leads to accelerated neuronal loss observed in AD.  Since recent studies have already investigated the effects of high glucose on monocytes and microglial cells, this thesis focused on the effects of high glucose on astrocyte-mediated neuroinflammation and neurotoxicity. Moreover, astrocytes are the most abundant glial type in the brain and play important functions in both normal and disease states.  	   26	  In vitro cell culture studies were conducted to investigate the effect of high glucose concentrations on neurotoxicity of stimulated astrocytic cells. Supernatants of stimulated astrocytic cells incubated in the presence of various glucose concentrations were applied to neuronal cells to determine if high glucose increased toxicity of astrocytes towards neuronal cells. Moreover, the effect of different glucose concentrations on the survival of neurons exposed directly to cytotoxic agents was investigated. Real-time quantitative polymerase chain reaction (qPCR) and enzyme linked immunosorbent assay (ELISA) techniques were used to study the mRNA expression and release of pro-inflammatory cytokines by astrocytic cells under conditions of varying glucose concentrations. Phosphorylation of several signaling molecules in astrocytes was investigated using western blotting techniques to determine the intracellular pathways involved in high glucose-induced inflammation.   Since there is little to no literature available on the effects of insulin on glial cells, the effects of different insulin concentrations on inflammatory properties of astrocytes were also studied. Specifically, the release of pro-inflammatory cytokines by stimulated primary human astrocytes incubated in the presence of different insulin concentrations was determined. In addition, expression of the INSR and insulin signaling components was investigated in different glial cells lines and primary glial cells using reverse transcription-polymerase chain reaction (RT-PCR).  Research on glial-mediated neuroinflammation and neurotoxicity in the presence of different glucose and insulin concentrations may provide valuable information toward understanding the link between T2DM and AD.  Outcomes of this project may guide 	   27	  future research on detailed mechanisms responsible for AD progression and help design preventative and therapeutic strategies for AD.  The following are the five main objectives of this thesis work:  1) To explore the effects of high glucose concentrations on astrocyte-mediated neurotoxicity. 2) To determine if high glucose concentrations enhance mRNA expression and secretion of pro-inflammatory cytokines by astrocytic cells and investigate the intracellular pathways involved.  3) To determine if high glucose concentrations increase susceptibility of neurons to such cytotoxic insults as hydrogen peroxide and Aβ42. 4) To investigate the expression of the INSR and insulin signaling components by different cell types. (Directed studies) 5) To examine the effect of different insulin concentrations on the release of pro-inflammatory cytokines by astrocytes.                  	   28	  Chapter 2. Materials and Methods   2.1 Chemicals and reagents  Aβ 1-42 (Aβ42) was purchased from EZBiolab Inc. (Carmel, IN, USA). Human recombinant insulin (expressed in yeast), D-(+)-glucose, L-(-)-glucose, D-mannitol, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT), beta-nicotinamide adenine dinucleotide (β-NAD), diaphorase (from Clostridium kluyveri), dimethyl sulfoxide (DMSO), ExtrAvidin alkaline phosphatase, iodonitrotetrazolium chloride (INT), phosphatase substrate tablets, sodium L-lactate (lactate), sodium orthovanadate, acetonitrile, poly-L-lysine, retinoic acid, ammonium persulphate, 10 mM tris-HCl, tetramethylethylenediamine (TEMED) and triton X-100 were obtained from Sigma Aldrich (Oakville, ON, Canada).  The following reagents were obtained from ThermoFisher Scientific (Ottawa, ON, Canada): acrylamide/bis (29:1, 30% solution), bovine serum albumin (BSA), fetal bovine serum (FBS), diethanolamine, Dulbecco’s modified Eagle medium nutrient mixture F-12 Ham (DMEM-F12), DMEM low glucose media, penicillin/streptomycin, ethylenediaminetetraacetic acid (EDTA) sodium salt, glycine, hydrogen peroxide, hydrochloric acid, N,N-dimethylformamide (DMF), EZ-run pre-stained protein ladder, Pierce BCA protein assay kit, sodium borate, sodium dodecyl sulphate (SDS), sodium chloride, sodium carbonate, sodium bicarbonate, monobasic sodium phosphate, dibasic sodium phosphate, sodium tris (hydroxymethyl)aminomethane (Tris), trichloroacetic acid (TCA), SuperSignal West Pico chemiluminescent substrate, GeneRuler DNA ladder, 0.05%  and 0.25% trypsin/EDTA solutions. Table 1 lists the glucose concentrations of the 	   29	  different media used in the experiments. Note that the glucose concentrations listed in the thesis experiments do not take into account the amount of glucose in FBS that was added to all media. However, addition of FBS did not change glucose concentration in any media by more than 7%.  Table 1: Glucose concentration in various media used in experiments. All media were supplemented with penicillin (100 U/ml) and streptomycin (100 µg/ml). The average glucose concentration in FBS was 6.8 ± 0.2 mM.  Media  Media used for preparation Glucose conc. in the media (mM)  FBS conc. (% v/v) Glucose conc. after addition of FBS (mM) F10 media DMEM-F12 17.5   10 16.4 F5 media DMEM-F12 17.5  5  17.0 F10 low glucose media DMEM low glucose 5.5  10  5.6 F5 low glucose media  DMEM low glucose 5.5   5 5.6  Bromophenol blue, Gelgreen stain, magnesium chloride pentahydrate and magnesium chloride hexahydrate were obtained from VWR International (Mississauga, ON, Canada). Human recombinant IFN-γ and IL-1β, as well as ELISA kits for MCP-1, IL-6 and IL-8 were purchased from Peprotech (Rocky Hill, NJ, USA). Ssofast qPCR reaction mix, Aurum RNA extraction kit and iScript complementary DNA (cDNA) synthesis kit were purchased from Bio-Rad (Mississauga, ON, Canada). Phosphate-buffered saline (PBS) tablets were purchased from Takara Bio Incorporated (Madison, WI, USA). Sodium fluoride was purchased from Anachemia (Montreal, QC, Canada). Nitrocellulose membrane was obtained from Pall Corporation (Washington, NY, USA). β-mercaptoethanol was purchased from Promega Corporation (Madison, WI, USA). 	   30	  Glycine and glycerol were obtained from MP Biomedicals (Santa Ana, CA, USA). All antibodies used in western blotting experiments were purchased from Cell Signaling Technology (Beverly, MA, USA). Milli-Q water, a type of ultrapure water purified using ion exchange and activated charcoal cartridges and dispensed through a 0.22 µm membrane filter, was obtained from the Milli-Q Direct water purification system (Cat#ZR0Q00800, Merck Millipore, Darmstadt, Germany). Endotoxin-free water for preparation of Aβ solvent was available in the limulus amebocyte lysate (LAL) chromogenic endotoxin quantification kit (Thermo Scientific).   2.2 Equipment and supplies  Cell culture experiments were mostly conducted in 24-well plastic cell culture plates (Corning Inc., Corning, NY, USA). 96-well sterile plastic plates and 10 cm tissue culture dishes (Corning) were used in experiments involving treatments in small volumes and collection of large volumes of supernatants, respectively. T-75 flasks (Sarstedt, Montreal, QC, Canada) were used for culturing cells in a Steri-Cycle HEPA Class 100 CO2 incubator (Model#370, Thermo Scientific). A VistaVision phase contrast inverted microscope was used to visualize cell morphology (Model#82026-630, VWR International) and a Motic inverted microscope (Model AE31) equipped with a Moticam 3000 camera attachment (Motic, Richmond, BC, Canada) was used for taking phase contrast digital microscopy pictures. A hemocytometer (ChangBioscience, Castro Valley, CA, USA) was used for cell counting and the Sorvall RT1 Centrifuge (Cat#75002384, Thermo Scientific) was used to centrifuge samples for harvesting cells and collecting 	   31	  supernatants. The FLUOstar Omega microplate reader (BMG Labtech, Nepean, ON, Canada) was used for measuring absorbance of samples in different colorimetric assays.   Six-well plastic cell culture plates (Corning) were used for experiments involving protein and RNA extraction. RNA was quantified using Eppendorf Biophotometer plus (Cat# 952000006, Thermo Scientific) and C1000 Thermal Cycler (Model#185-1048, Bio-Rad) was used for reverse transcription of RNA to cDNA. qPCR was performed in white 96-well plates (Bio-Rad) using the CFX96 Real Time System (Model# 185-5201, Bio-Rad). The Mini-Protean Tetra Cell system with HC Power supply (Cat#165-8001 and 164-5070, respectively, Bio-Rad) was used for casting and running protein samples on polyacrylamide gels and the Mini Trans-Blot Electrophoretic Transfer Cell (Cat#170-3930, Bio-Rad) was used for western blotting transfer experiments. One Touch Ultra Mini blood glucose monitoring system and One Touch Ultra blue test strips were obtained from a local pharmacy.   2.3 Cell culture models   All cell lines and primary cell cultures used in the experiments were obtained from the Kinsmen Laboratory of Neurological Research at the University of British Columbia, Vancouver, Canada. U-118 MG and/or U-373 MG astrocytoma cells (astrocyte cell model) and SH-SY5Y neuroblastoma cells (neuronal model) were used in cell culture studies to model astrocytes and neurons respectively. Primary human astrocytes were isolated from human surgical tissues by Dr. Sadayuki Hashioka at the Kinsmen Laboratory of Neurological Research according to previously published 	   32	  procedures (Hashioka et al., 2009). All cell cultures were stored in liquid nitrogen and grown in DMEM-F12 media supplemented with 10% FBS, penicillin (100 U/ml) and streptomycin (100 µg/ml) (F10 media). The cells were cultured in T-75 flasks and incubated at 37 °C in a CO2 incubator (humidified 5% CO2 and 95% air atmosphere). Cells frozen in liquid nitrogen were thawed by removing vials from liquid nitrogen followed by immersion in water at 37 °C. Cells were then removed from the vial, suspended in 10 ml of F10 medium and centrifuged for 10 min at 250 g. After centrifugation, medium was discarded and cells were re-suspended in 10 ml of fresh F10 medium. The cells were transferred into a T-25 flask and allowed to grow to confluence in the 37 °C CO2 incubator. The cells were eventually transferred to a T-75 flask and used in experiments.   2.4. Establishing toxicity of stimulated U-118 MG cell supernatants towards SH-SY5Y neuronal cells   To establish toxicity of U-118 MG astrocytic cells towards SH-SY5Y neuronal cells, supernatants of U-118 MG cells activated with different concentrations of IFN-γ and IL-1β were transferred onto SH-SY5Y cells. Experiments using U-118 MG cells were performed on at least three independently grown batches of cells. For use in experiments, U-118 MG cells, which grow adherent to plastic, were harvested from T-75 flasks by removing the supernatant and incubating the cells with 1.5 ml of 0.25% trypsin/EDTA solution for 2 min at 37 °C. This allowed cell detachment from the bottom of the T-75 flasks. Trypsin was inactivated by adding 10 ml of F10 	   33	  medium. The cells were counted using the hemocytometer, centrifuged and re-suspended in DMEM-F12 media supplemented with 5% FBS and antibiotics (F5 media) to achieve a final concentration of 0.2 million cells/ml. For cell supernatant transfer experiments, 0.8 ml of U-118 MG cell culture was added to each well of the 24-well plate and cells were allowed to adhere to the bottom of the wells for 24 h at 37 °C. Following 24 h incubation, the medium was replaced with 0.8 ml of fresh F5 medium and cells were allowed to recover for 15 min in the incubator. Subsequently, the cells were stimulated with a combination of different concentrations of IFN-γ plus IL-1β (100 U/ml) and incubated for 48 h at 37 °C. 0.4 ml of the U-118 MG cell supernatants were transferred onto SH-SY5Y cells that had been plated 24 h earlier.  Adherent SH-SY5Y cells were harvested from T-75 flasks in a fashion similar to experiments with the U-118 MG cells except 0.05% trypsin/EDTA solution was used instead of 0.25% trypsin/EDTA solution. SH-SY5Y cells were counted, centrifuged at 450 g for 7 min, and re-suspended in F5 media to a final concentration of 0.2 million cells/ml. 0.4 ml of SH-SY5Y cell culture was plated in sterile 24-well plates and incubated for 24 h at 37 °C to allow cells to adhere to the plate prior to transfer of U-118 MG supernatants.  For astrocytic cell supernatant transfer, SH-SY5Y supernatants were aspirated and replaced with 0.4 ml of supernatants from unstimulated and stimulated U-118 MG cells. SH-SY5Y cells were then incubated in a 37 °C CO2 incubator for 72 h followed by assessment of cell viability using the lactate dehydrogenase (LDH) release and MTT assays.   	   34	  2.5 Preparation of U-118 MG supernatants with different glucose concentrations   U-118 MG cells were harvested from T-75 flasks as described in section 2.4.  Cells were re-suspended in DMEM low glucose medium (5.5 mM glucose) supplemented with 5% FBS and antibiotics (F5 low glucose) to achieve a final concentration of 0.2 million cells/ml. Two ml of U-118 MG cell suspensions were seeded onto sterile 6-well plates and the cells were allowed to adhere to the bottom of the wells at 37 °C. Following 24 h incubation, the media were replaced with 2 ml of fresh F5 low glucose media and cells were allowed to recover for 15 min in the incubator. Subsequently, equal volume aliquots containing different glucose concentrations were added to the respective wells to achieve different glucose concentrations in the wells (5.5, 10.5, 20.5 and 30.5 mM). U-118 MG cells were then stimulated with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) and incubated at 37 °C for 48 h. Multiple sets of supernatants (0.4 ml each) were collected from each sample treatment and frozen at -20 °C.  SH-SY5Y cells were harvested as described in section 2.4.1. and were re-suspended in F5 low glucose media to achieve a final concentration of 0.2 million cells/ml. 0.4 ml of SH-SY5Y cell suspensions were plated in sterile 24-well plates and incubated for 24 h at 37 °C. After incubation, the SH-SY5Y cells were treated with supernatants from U-118 MG cells stimulated in the presence of different glucose concentrations. Alternatively, SH-SY5Y cells were first pre-treated with 0.4 ml of F5 high glucose media (F5 media with a final glucose concentration of 30.5 mM) for 24 h prior to the treatment with glial cell supernatants. This was done to prime the SH-SY5Y 	   35	  cells using high glucose concentrations and to determine whether incubation of SH-SY5Y cells in high glucose would increase their susceptibility to toxic action of supernatants obtained from astrocytic cells stimulated in the presence of different glucose concentrations. In both cases, SH-SY5Y cell viability was assessed by the MTT assay after 72 h incubation of cells with glial supernatants at 37 °C. Supernatant transfer experiments to study the effect of high glucose were technically complex since both the astrocytic and neuronal cells utilize glucose in their metabolism. Therefore, it was challenging to control the glucose concentrations under the different treatment conditions. Based on the possibility that SH-SY5Y cells cultured for 72 h in the presence of 5.5 mM glucose may show decreased viability due to glucose starvation, a different experimental design was implemented. In a sub-set of experiments, astrocytic glial supernatants on SH-SY5Y cells were replaced every 24 h during the 72 h incubation period by thawing out and pre-warming batches of fresh glial supernatants.   2.6 LDH release assay   LDH is a stable cytoplasmic enzyme present in all cells. Upon damage to the plasma membrane of the cells, this enzyme is rapidly released into the cell culture supernatant.  The LDH release assay quantifies cell death by measuring LDH activity in the cell-free supernatant. LDH activity is determined in a coupled enzymatic reaction involving the reduction of the tetrazolium salt INT to formazan dye, which has a broad absorption maximum at 492 nm and its formation is measured spectrophotometrically.  For the LDH release assay, INT was added to 0.1 ml of supernatants collected in 	   36	  96-well plates to reach the final concentration of 2 mg/ml and the optical density (O.D.) at 492 nm (O.D.initial) was measured. Subsequently, 30 µl of working solution containing lactate (750 µg/ml), β-NAD (60 µg/ml) and diaphorase (55 µg/ml) in PBS was added and the plates were incubated on a bench-top rocker at room temperature for 15-30 min. The O.D. at 492 nm (O.D.final) was measured again once the 100% lysis control well appeared dark red.   The percent cell death was calculated by comparing the change in absorbance in each well to the 100% lysis control well in which untreated cells were lysed with 1% Triton X-100 and the control well containing only media. The following equations were used: Change in absorbance of each sample = O.D.final - O.D.initial Percent cell death = ((Change in the absorbance of the sample – Change in the absorbance of the media only well)/(Change in the absorbance of the lysis well – Change in the absorbance of the media only well))*100  2.7 MTT assay   MTT assay is a colorimetric assay that measures the reduction of the yellow tetrazolium dye MTT by the mitochondrial enzyme succinate dehydrogenase to an insoluble, dark purple product. The formazan product is solubilized by an organic solvent and the absorbance is measured at 570 nm. Since the reduction of MTT can only occur in metabolically active cells, this assay provides a measure of cell viability.  	   37	  30 µl of MTT dye solution (5 mg/ml) was added to the 0.3 ml of cell culture media remaining in each well of the 24-well plates after removal of 100 µl of cell supernatants for the LDH release assay. Plates were then placed in the 37 °C CO2 incubator for 1 h. Subsequently, 0.4 ml of 20% SDS/50% DMF in milli-Q water was added to each well to solubilize the reaction product. The plates were then placed in a wet box and incubated for 3-4 h in the 37 ⁰C dry incubator. 0.1 ml of each sample was then transferred to a 96-well plate by cutting the end off the pipette tip and mixing the well contents before transfer. O.D. at 570 nm was measured and the percent cell viability was calculated using the following formula:  Percent cell viability = ((Absorbance of the sample - Absorbance of the well containing media without cells)/(Absorbance of the sample treated with fresh media only - Absorbance of the well containing media without cells))*100  2.8 Effects of high glucose on mRNA expression of pro-inflammatory cytokines by astrocytic cells   2.8.1 RNA extraction   RNA from U-118 MG astrocytic cells and primary human astrocytes was extracted using the Aurum mini-kit from Bio-Rad by following the extraction protocol provided by the manufacturer. Experiments using U-118 MG cells were performed on at least three independently grown batches of cells. Human primary astrocytes isolated from two different human surgical cases were used. 	   38	  Both cell types were seeded onto sterile 6-well plates at a final concentration of 0.5 million cells/ml in 2 ml of F10 low glucose media. The cells were then incubated in the 37 °C, 5% CO2 incubator for 24 h to allow them to adhere to the bottom of the plates. Following incubation, the media were replaced with 2 ml F5 low glucose media and the cells were incubated at 37 °C for 15 min. Subsequently, equal volume aliquots containing different glucose concentrations were added to the respective wells to achieve glucose concentrations that are stated in the results section. The cells were incubated at 37 °C for additional 24 h. Next day, media were replaced with F5 low glucose media containing different glucose concentrations followed by stimulation of cells with a combination of IFN-γ (150 U/ml) and IL-1β (100 U/ml). After 24 h incubation of the cells at 37 °C, total cellular RNA was extracted.   2.8.2 RNA spin column protocol  All reagents listed in this section were provided in the Aurum mini-kit (Bio-Rad) with the following exceptions: ice-cold sterile PBS (see Appendix A) and 70% molecular grade ethanol. RNA extraction protocol involved removing all the media from the adherent cells followed by washing the cells 2X with ice-cold sterile PBS. Cells were lysed by adding 350 µl of RNA lysis solution. The cell lysate was transferred into 1.5 ml Eppendorf tubes and 350 µl of 70% molecular grade ethanol was added. The solutions were subsequently transferred into the spin columns placed in 2 ml capped tubes provided in the kit. Throughout the extraction protocol, the spin columns were kept covered using the lids of the 2 ml capped tubes. The spin columns were centrifuged for 30 s at 12,000 	   39	  xg to force the lysis/ethanol solution through the columns. The flow-through was discarded and 750 µl of low-stringency wash solution was added to the column. Once again, the columns were centrifuged for 30 s at 12,000 xg and flow-through discarded. Next, 80 µl of the DNase I solution (5 µl of DNase I re-constituted in 10mM Tris-HCl combined with 75 µl of DNase dilution solution) was added to each column followed by a 15 min incubation at room temperature. This step allowed removal of possible contaminating genomic DNA. Following incubation, the DNase was inactivated by adding 700 µl of high-stringency wash solution. The columns were centrifuged for 30 s at 12,000 xg followed by addition of 700 µl of low-stringency wash solution and centrifugation for 1 min at 12,000 xg. The flow-through was discarded after each centrifugation step. The dry columns were then centrifuged for additional 2 min at 12,000 xg to remove any residual wash solution. Following the addition of 80 µl of RNA elution solution, the columns were incubated at room temperature for 1 min and then centrifuged for 2 min at 12,000 xg to elute RNA. For RNA elution from primary human astrocyte samples, 40 µl of RNA elution solution was used instead of 80 µl. The total RNA was collected in 1.5 ml Eppendorf tubes provided with the kit. RNA was quantified spectrophotometrically using the Eppendorf Biophotometer plus (Thermo Scientific). The RNA purity values were within the acceptable OD260/280 range of 1.8 to 2.0. All RNA samples were stored at -80 °C.     	   40	  2.8.3 cDNA synthesis   One µg of each RNA sample was converted to cDNA using the iScript cDNA synthesis kit from Bio-Rad by following the protocol provided by the manufacturer. The RT reactions were conducted in 0.2 ml Eppendorf tubes. 4 µl of RT reaction mix and 1 µl of reverse transcriptase enzyme were combined with appropriate volumes of the RNA solution and nuclease-free water to a total reaction volume of 20 µl. RT reaction mix contained a mix of oligo(dT) and random hexamer primers. The tubes were finger-vortexed, briefly centrifuged (6,000 xg) and then arranged in the C-1000 thermal cycler (Bio-Rad) for the RT reaction, which consisted of three steps: 5 min at 25 °C, 30 min at 42 °C, 5 min at 85 °C. The cDNA samples obtained were diluted 1:5 in nuclease-free water and stored at -20 °C.   2.8.4 Real-time quantitative polymerase chain reaction (qPCR)  The primer sequences for IL-6, IL-8 and MCP-1 were available from previous studies (Madeira et al., 2013) (Table 2). Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) guidelines (Bustin et al., 2009) were followed to test for optimal annealing temperature, primer-dimer formation and primer efficiency of each primer pair. An amplification curve was used to determine the optimal temperature for each primer pair. Amplification specificity for the target was determined by the presence of a single peak in the melting curve analysis. A standard curve of cDNA template dilutions as a function of cycle threshold (Ct) values provided the primer 	   41	  efficiency data (Efficiency = 90-110%; R2 > 0.98) for each primer pair. Primer efficiency data for IL-6 are shown in Figure 3; these data are representative of other primers used in these experiments.  Table 2: Primer sequences for qPCR experiments.  Target gene  Forward primer (5’ à 3’)  Reverse primer (5’ à 3’)  IL-6 GAC CCA ACC ACA AAT GCC A GTC ATG TCC TGC AGC CAC TG IL-8 CTG GCC GTG GCT CTC TTG CCT TGG CAA AAC TGC ACC TT MCP-1 CTC TGC CGC CCT TCT GTG TGC ATC TGG CTG AGC GAG  qPCR experiments were carried out in duplicate on 3-5 independent biological samples and at least two different surgical cases of primary cells in 96-well plates (Bio-Rad). Each 10 µl reaction volume comprised of 1 ng of cDNA sample, 0.8 µl of each primer (5 µM), 5µl Ssofast Evagreen Supermix (Bio-Rad) and 3.2 µl nuclease-free water was pipetted into each well of the 96-well plate. To ensure there was no DNA contamination in the original samples, no template control (NTC) and no reverse-transcriptase (NRT) control were included in the experiments. The PCR program of the CFX96 Real Time System (Bio-Rad) consisted of the following steps: 95°C for 5 min, followed by 40 cycles of 95°C for 5 s, 59°C for 5 s, followed by a dissociation stage of 95°C for 10 s, 65°C for 5 s, and 95°C for 5 s. Gene expression in each sample was normalized to the expression of β-actin reference gene. Gene expression in treated samples was expressed relative to the control sample where cells were left unstimulated in 5.5 mM glucose F5 low glucose media. Relative expression values were calculated using GeneExMacro OM 3.0 software (Bio-Rad). 	   42	   Figure 3: Primer-efficiency data for IL-6: (A) Melt curve; (B) Amplification curve; and (C) Standard curve obtained during optimization experiments by Madeira, J. All results are within the range of MIQE accepted values.    2.9 Effects of high glucose on release of pro-inflammatory cytokines by astrocytic cells   2.9.1 Collection of supernatants from U-118 MG cell, U-373 MG cell and primary human astrocytes incubated in media containing different glucose concentrations   Primary human astrocytes were harvested from T-75 flasks as described for U-118 MG cells in section 2.4.1. To collect supernatants for ELISA experiments, U-118 MG cells, U-373 MG cells and primary human astrocytes were harvested and re-suspended in F5 low glucose media to achieve a final concentration of 0.2 million 	   43	  cells/ml. 0.4 ml of U-118 MG and U-373 MG cell cultures were seeded onto sterile 24-well plates and 0.1 ml of primary human astrocyte cell culture was seeded onto a sterile 96-well plate. All cell types were incubated for 24 h at 37 °C. Following 24 h incubation, the media on U-118 MG/U-373 MG cells and primary human astrocytes were replaced with 0.4 ml and 0.2 ml of F5 low glucose media respectively. The cells were allowed to recover for 15 min in the incubator. Subsequently, equal volume aliquots containing different glucose concentrations were added to the respective wells to achieve different glucose concentrations in the wells (5.5, 10.5, 20.5 and 30.5 mM). The cells were incubated for 24 h followed by stimulation of U-118 MG cells and primary human astrocytes with IFN-γ (150 U/ml plus IL-1β (100 U/ml) and U-373 MG cells with IFN-γ (150 U/ml) only for another 24 h. Following incubation, collected supernatants were frozen at -20 °C and used in IL-6 and IL-8 ELISA experiments.   2.9.2 ELISA  ELISA was used to detect the presence of specific proteins in tissue culture medium. It is based on highly specific antigen-antibody interactions. In this assay, the primary antibody directed against the antigen is fixed to the well of a 96-well plate and the supernatant containing the desired antigen is then added to the well. The antigen, if present, binds to the primary antibody. A secondary enzyme-conjugated antibody is then added. This antibody binds to the original antigen-antibody complex. When the substrate of the enzyme is added, the enzyme breaks down the substrate into a colored compound, the formation of which can be measured spectrophotometrically.  	   44	  Cell-free supernatants frozen at -20 °C were thawed for 24 h at 4 °C for use in ELISA experiments. ELISA assays for IL-6 and IL-8 were performed using the respective ELISA development kit, according to the manufacturer’s (Peprotech) instructions. Different ELISA reagents and their compositions are listed in Appendix A. On day 1, the polyclonal rat anti-human primary antibodies for IL-6 and IL-8 were diluted 1:100 and 1:200 respectively in the coating buffer (Na2CO3-NaHCO3/H2O solution, pH=9.6). 50 µl of the primary antibody solution was added to each well of a 96-well plate. The plate was covered with Parafilm and incubated overnight at 4 ⁰C.  On day 2, the coating buffer was discarded and 180 µl of blocking solution (1% BSA and 1% skim milk powder in PBS) was added to each well. The plate was placed in a wet-box and incubated at 37 ⁰C for 1 h. To calibrate the ELISA signal, a series of standard solutions ranging from a concentration of 0.0032 – 10 ng/ml was prepared for each cytokine by making 1:5 serial dilutions in F5 medium. The plate was then washed twice with PBS-Tween (0.05% Tween in PBS), leaving the solution in the wells after the second wash. Using the suction system, PBS-Tween was removed and 0.1 ml of the cell-free supernatant was immediately pipetted into each well. The standard cytokine solutions were added in duplicate wells, and four wells in each plate received F5 media only. These were referred to as the blank wells. The plate was covered with Parafilm and incubated overnight at 4 ⁰C.  On day 3, samples were discarded and the plate was washed three times with PBS-Tween, followed by addition of 0.1 ml of biotinylated rat anti-human secondary antibody at a dilution of 1:400 in blocking solution. The plate was placed in a wet-box 	   45	  and incubated at 37 ⁰C for 45 min. The secondary antibody solution was discarded and the plate was washed four times with PBS-Tween. 0.1 ml of ExtrAvidin-alkaline phosphatase was added at a dilution of 1:10,000 in the blocking solution. The plate was incubated in a wet-box at 37 ⁰C for an additional 45 min period. Finally, the plate was washed five times with PBS-Tween and 0.1 ml of detection solution (1 mg/ml phosphatase substrate in substrate solution) was pipetted into each well. O.D. at 405 nm was immediately measured and the measurement repeated after 3 h incubation of the plate at 37 ⁰C or when a light yellow color appeared. Cytokine concentration in the cell-free supernatants was determined by constructing a calibration curve using the O.D. values obtained from wells containing standard dilutions with known cytokine concentrations. Change in absorbance for each sample (dA) was calculated by subtracting the initial O.D. measurement from the final reading. Change in absorbance of the blank wells (dAblank) was subtracted from the change in absorbance for each sample (dA) to calculate the corrected change in absorbance for each sample (dAcorr). A linear trendline (equation y = mx, where m is the slope of the line) was added to the calibration curve (dAcorr for the cytokine standards as a function of the cytokine concentration). The following equations were used to determine cytokine concentrations and the detection limit of the assay:  Cytokine concentration (ng/ml) = dAcorr /m  Detection limit of the assay (ng/ml) = (Average of dAblank + 2 x standard deviation of dAblank)/m  	   46	  2.10 Western blotting   2.10.1 Protein extraction   Compositions of all buffers and reagents used in western blotting are listed in Appendix A. U-118 MG astrocytic cells were seeded onto sterile 6-well plates at a final concentration of 1 million cells/ml in 3 ml of F10 media. The cells were then incubated in the 37 °C, 5% CO2 incubator for 24 h to allow them to adhere to the bottom of the plates. Following incubation, the media were replaced with 3 ml F5 low glucose media and the cells were incubated at 37 °C for 15 min. Subsequently, equal volume aliquots at different glucose concentrations were added to the respective wells to achieve different glucose concentrations in the wells. The cells were incubated for another 24 h at 37 °C. Following incubation, the cells were stimulated with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for 30 min and the whole cell protein was extracted.  For protein extraction, all media were removed from the adherent cells and the cells were washed once with 2 ml of ice-cold sterile PBS. Subsequently, 1 ml of 10% TCA was added and the cells were incubated at 4 °C for 30 min. The cells were washed three times with 2 ml, 1 ml and 0.5 ml of ice-cold sterile PBS. Using a rubber policeman, cells were scraped into 0.1 ml of radio immune precipitation assay (RIPA) buffer and collected in 1.5 ml tubes kept on ice. The cells in RIPA buffer were sonicated (Cat# CL-188, Qsonica, Newtown, CT, USA) at 20% amplitude for 1 min (on/off to avoid excess heating) on ice, followed by centrifugation of the lysate for 10 min at 15,870 xg and 4 °C. If a pellet was seen after the first sonication, the sonication and centrifugation process 	   47	  was repeated and the supernatants were finally collected and stored at -20 °C.  Protein concentration in all samples was determined using the BCA Protein Assay Kit (Thermo Scientific). A series of BSA standards ranging from 0.025 to 2 mg/ml were prepared by diluting the BSA stock solution in RIPA buffer. Protein samples extracted form cells were diluted 10X in RIPA buffer before use in the assay. 10 µl of protein samples was pipetted into duplicate wells of a 96-well plate followed by addition of 0.2 ml of working solution (50:1, Reagent A:B provided in the kit). Contents of the well were mixed thoroughly on a plate shaker for 30 s. The plate was incubated at room temperature for 20 min followed by measurement of O.D. at 570 nm. A standard curve of the average O.D. of the BSA standards as a function of the concentration was constructed and used to calculate protein concentration in the unknown samples.    2.10.2 Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)   SDS-PAGE was carried out to separate proteins on an 8% polyacrylamide gel using the Mini-PROTEAN Electrophoresis Cell (Bio-Rad). The resolving gel was prepared by combining 5 ml of 4X resolving gel buffer (see Appendix A), 5.4 ml of 29.2% acrylamide: 0.8% bis acrylamide solution and 9.6 ml of milli-Q water in a 250 ml filter flask. The solution was then exposed to vacuum for 15 min to remove air bubbles. Subsequently, 150 µl of 10% ammonium persulphate and 10 µl of TEMED (0.775 g/ml) were added and the solution was immediately poured into the space between the gel-forming plates (up to the level of the green shutter of the gel cassette) using a 10 ml pipette. The gel was allowed to polymerize for 30 - 45 min. Following polymerization, 	   48	  the empty space above the gel was rinsed with distilled water to remove any sticking gel. The plates were then dried with filter paper. Next, the stacking gel was prepared by combining 2.5 ml of 4X stacking gel buffer (see Appendix A), 1.5 ml of 29.2% acrylamide: 0.8% bis acrylamide solution, 6 ml of milli-Q water, 100 µl of 10% ammonium persulphate and 10 µl of TEMED in a 50 ml tube. The solution was immediately pipetted between the glass plates to about 2 mm below the top of the short glass plate. A 10-well gel comb was inserted and the stacking gel was allowed to polymerize for 30 min.   While waiting for the stacking gel to polymerize, different protein samples were diluted in PBS (based on the determined protein concentrations) to achieve 30 µg of total protein for all samples. 5X loading buffer (see Appendix A) was added to each protein sample to reach a final concentration of 1X. Total sample volume was 30 µl. All protein samples in 1.5 ml microtubes were denatured by heating them at 100 °C for 5 min followed by a brief centrifugation. The denatured protein samples were now ready for loading.  Once the stacking gel had finished polymerizing, the gel cassette was removed from the casting stand. The shutters of the gel cassette were released to allow the gel-forming plates to be removed. Two gels were then placed into the electrode assembly apparatus with the short glass plate facing inward. The inner chamber between the two gels and the outer chamber of the electrophoresis apparatus was filled with 1X electrophoresis running buffer (see Appendix A). Following removal of the gel comb, 5 µl of the pre-stained protein ladder and 30 µl of protein samples were loaded into the respective wells using gel-loading tips. The electrophoresis apparatus was then connected 	   49	  to the power supply and the gel was run for 3 h at 60 V.   2.10.3 Electrophoretic transfer and protein detection   Following separation, the proteins were transferred onto the nitrocellulose membrane using the Mini Trans-Blot Electrophoresis Transfer Cell (Bio-Rad). The electrophoretic transfer was performed at 60 V overnight at 4 °C in 1X transfer buffer (see Appendix A). Next day, the membrane was washed with 1X tris-buffered saline-0.1% tween (TBS-T) and blocked in 5% BSA in TBS-T for 1 h at room temperature on the bench-top rocker. The membranes were then washed once with TBS-T followed by three 5 min washes in TBS-T. For every wash, enough fresh TBS-T was first added to cover the membranes and the membranes in TBS-T were then incubated at room temperature on the bench-top rocker for 5 min. Finally, membranes were incubated on the bench-top rocker at 4 °C overnight with the respective primary antibodies diluted in 5% BSA in TBS-T. A 1:1,000 dilution was used for anti-phospho-p38 MAPK antibody (Cell signaling #9211) and anti-phospho-JNK antibody (Cell signaling #9251). 1:2,000 dilution was used for anti-phospho-STAT-3 antibody (Cell signaling #9131) and anti-phospho-p44/42 MAPK antibody (Cell signaling #9101).  Next day, antibody solutions were removed and the membranes were washed once with TBS-T followed by three 5 min washes in TBS-T. Subsequently, membranes were incubated with horseradish peroxidase-conjugated anti-rabbit IgG secondary antibody (1:2,000 dilution in 5% skim milk in TBS-T) on the bench-top rocker for 1 h at room temperature. Following incubation, the membranes were again washed once with 	   50	  TBS-T followed by four 5 min washes in TBS-T. Western blots were developed using the SuperSignal West Pico chemiluminescent substrate solution comprised of a stable peroxide buffer and luminol solution (Thermo Scientific). Each blot was incubated with the substrate solution (0.25 ml each of the stable peroxide buffer and luminol solution) for 5 min and developed in the Fluorchem Q image analysis system (Cell Biosciences). Images were captured using the AlphaView Q 3.0 gel acquisition software (Cell Biosciences).  Following image acquisition, each blot was washed with TBS-T to remove the substrate solution and subsequently antibodies were stripped off using the stripping buffer (see Appendix A). The blot was incubated in the stripping buffer at room temperature for 10 min on the bench-top rocker. The buffer was discarded and the 10 min incubation step was repeated with fresh stripping buffer. Once again, the buffer was discarded and two 10 min incubations were performed in PBS followed by two 5 min incubations in TBS-T. The blots were now ready for the blocking stage and re-probing with different antibodies. The blocking and washing steps were preformed as described above in this section and the membranes were incubated on the bench-top rocker at 4 °C overnight with the respective primary antibodies diluted 1:1,000 in 5% BSA in TBS-T. Antibodies directed against the following epitopes were used: total-p38 MAPK (Cell signaling #9212), total-JNK (Cell signaling #9252), total-STAT-3 (Cell signaling #9132) and total-p44/42 MAPK (Cell signaling #9102). The blots were developed the following day after incubation with the secondary antibody as described above.   	   51	  2.11 Toxicity of hydrogen peroxide and Aβ42 towards SH-SY5Y cells in the presence of different glucose concentrations   2.11.1 Undifferentiated SH-SY5Y neuronal cells   SH-SY5Y cells were harvested and re-suspended in F5 low glucose medium to achieve a final concentration of 0.2 million cells/ml. 0.4 ml of cell culture media were seeded onto sterile 24-well plates and incubated at 37 °C for 24 h. Following incubation, media were replaced with 0.4 ml of fresh F5 low glucose media. The cells were allowed to recover for 15 min in the incubator. Subsequently, equal volume of glucose solution from aliquots at different concentrations was added to the respective wells to achieve different glucose concentrations in the wells (5.5, 10.5, 20.5 and 30.5 mM) and the cells incubated for 24 h. After incubation, cells were once again treated with fresh media containing different glucose concentrations followed by treatment with hydrogen peroxide (0.25 and 0.5 mM) for 24 h or 5 µM Aβ42 dissolved in Aβ solvent (35 acetonitrile: 113 endotoxin-free water: 295 sterile PBS) for 72 h (Shen & Murphy, 1995). Cell viability was assessed by the LDH release and MTT assays. Aβ42 stock solution (500 µM) was prepared by dissolving the peptide in the Aβ solvent, vortexing the solution vigorously till it turned opaque, and finally leaving the solution in the 37 °C incubator for 18 h to allow formation of Aβ fibrils (Shen & Murphy, 1995). Aliquots of the stock solution were stored at -20 °C.    	   52	  2.11.2 Differentiated SH-SY5Y neuronal cells   Retinoic acid-induced differentiation of SH-SY5Y neuroblastoma cells results in cells with morphological and biochemical characteristics similar to that of mature neurons (Lopes et al., 2010). Such differentiated cells have long, extensively branched neurites and they express several neurospecific markers including nuclear markers and synapse proteins.   SH-SY5Y cells were plated at a concentration of 0.2 million cells/ml in 0.3 ml of F5 media in sterile 24-well plates coated with poly-L-lysine (Day 0). The stock solution of poly-L-lysine was diluted to 0.1 mg/ml in sterile milli-Q water and 0.25 ml of the working solution was added to each well of the 24-well plate. The plates were incubated for 1 h at room temperature. Subsequently, the poly-L-lysine solution was removed and the wells were washed with sterile PBS solution three times. The plates were then air dried at room temperature and exposed to UV light in the biosafety cabinet for 1 hr. On Day 1, media on SH-SY5Y cells were replaced with 0.4 ml of DMEM-F12 media containing 1% FBS (F1 media) and 10 µM retinoic acid. SH-SY5Y cells were incubated for 72 h at 37 °C. On Day 4, media were once again replaced with 0.4 ml of fresh F1 media containing 10 µM retinoic acid and the cells were incubated for another 72 h. On Day 7, media were switched to 0.4 ml of F1 low glucose media containing 10 µM retinoic acid. Finally, on day 8, glucose solution was added to some wells to achieve a concentration of 30.5 mM glucose in respective wells. Differentiated SH-SY5Y cells were allowed to recover for 15 min in the 37 °C incubator followed by treatment with either 5 µM Aβ42 dissolved in Aβ solvent for 72 h or 0.25 mM and 0.5 mM hydrogen 	   53	  peroxide for 24 h. Cell viability was assessed by the MTT assay.   2.12 Expression of insulin signaling components by glial cells   2.12.1 RNA extraction and cDNA synthesis  RNA from U-118 MG astrocytic cells, primary human astrocytes, SH-SY5Y neuroblastoma and HepG2 hepatocytoma cells was extracted using the Aurum mini-kit from Bio-Rad by following the spin column protocol provided by the manufacturer (see section 2.8.2). Briefly, cells were seeded onto sterile 6-well plates at a final concentration of 0.5 million cells/ml in 2 ml F10 media. The cells were incubated in the 37 °C, 5% CO2 incubator for 24 h to allow them to adhere to the bottom of the plates. Following incubation, the media were replaced with 2 ml of F5 media. The cells were incubated at 37 °C for additional 24 h and total cellular RNA extracted. One µg of each RNA sample was converted to cDNA using the iScript cDNA synthesis kit from Bio-Rad by following the protocol provided by the manufacturer (see section 2.8.3). cDNA samples from primary human microglia were available from previous experiments.   2.12.2 Primers   Primer sequences for five different target genes including INSR, INSRA, INSR INSRB, IRS-1 and IRS-2 were obtained from a previous publication (Heni et al., 2011). All primers were purchased from Integrated DNA Technologies (IDT) and checked for 	   54	  specificity for the target gene using national centre for biotechnology information (NCBI) primer basic alignment search tool (BLAST). A single gene encodes for INSR, which is a receptor tyrosine kinase composed of two extracellular ligand-binding α subunits, linked by disulphide bonds to two membrane-spanning β subunits (Duarte et al., 2012). INSRA and INSRB represent the two isoforms of the INSR that are generated by alternative splicing. The two isoforms differ by the presence of exon 11 (36 base pair (bp) in length) in the INSRB compared to INSRA, which does not have exon 11. IRS-1 and -2 are signaling molecules involved in transmitting the signal from the exterior of the cell to intracellular signaling pathways once insulin binds to its receptor. Figure 4 depicts schematically the specificity of INSRA and INSRB primers for their respective targets.    Figure 4: Primers for INSRA and INSRB target two different isoforms of INSR. INSR primer pair targets both INSRA and INSRB since it anneals to the region of the INSR gene that is shared by both isoforms. Since INSRA lacks exon 11, specific primers designed for exon 11 in INSRB would not anneal to INSRA gene. INSR gene is shown in light blue; first few bp sequences of exon 11 are shown in pink; and primer pairs are shown in light grey (lengths of primers are not to scale).  	   55	  Table 3 lists the five different target genes studied, the respective primer sequences and expected product sizes. All primers were optimized for annealing temperature using a gradient between 50 °C and 65 °C. Glyceraldehyde-3-phosphate dehydrogenase (G3PDH) is expressed in all cells due to its role in basic cellular metabolism and was hence used as the housekeeping gene (Villanueva et al., 2012). Table 3: Sequences of primers used in the experiments. Target gene  Forward primer (5’ à 3’)   Reverse primer (5’ à 3’)  Product size (bp) INSR GCT GGA TTA TTG CCT CAA AGG  TGA GAA TCT TCA GAC TCG AAT GG 75 INSRA TTT TCG TCC CCA GGC CAT CCA CCG TCA CAT TCC CAA C  63 INSRB TTT CGT CCC CAG AAA AAC CTC T CCA CCG TCA CAT TCC CAA C  98 IRS-1 GCC TAT GCC AGC ATC AGT TT  TTG CTG AGG TCA TTT AGG TCT TC 95 IRS-2 TGA CTT CTT GTC CCA CCA CTT  CAT CCT GGT GAT AAA GCC AGA 77 G3PDH  CCA TGT TCG TCA TGG GTG TGA ACC A GCC AGT AGA GGC AGG GAT GAT GTT C 215   2.12.3 Polymerase chain reaction (PCR)   GoTaq Green Master Mix (400 µM each of deoxyadenosine triphosphate, deoxyguanosine triphosphate, deoxycytidine triphosphate and deoxythymidine triphosphate, as well as 3 mM MgCl2 in Green GoTaq reaction buffer at pH 8.5) from 	   56	  Promega was used to amplify cDNA. The PCR reactions were conducted in 0.5 ml thin-wall PCR tubes with the total reaction volume of 25 µl comprised of 5 µl of cDNA, 5µl of the primer mix (forward + reverse primer, 5 µM each), 12.5 µl of GoTaq Green Master Mix and 2.5 µl of nuclease-free water. The tubes were finger-vortexed, briefly centrifuged (6,000 xg) and placed in the C-1000 thermal cycler (Bio-Rad). The PCR amplification protocol consisted of the following steps: Denaturation at 95 °C for 3 min followed by 30 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s and extension at 72 °C for 1 min, with the final extension step lasting 5 min. The amplified PCR products were stored at -20 °C.   2.12.4 Polyacrylamide gel electrophoresis (PAGE)  Following RT-PCR, the amplified PCR products were separated and visualized using PAGE. Briefly, 6 ml of 29.2% acrylamide: 0.8% bis acrylamide solution, 6 ml of milli-Q water and 3 ml of 5X tris-borate-EDTA (TBE) buffer (see Appendix A) were combined in a 250 ml filter flask. The solution was then degassed under vacuum for 15 min. Subsequently, 100 µl of 10% ammonium persulphate and 10 µl of TEMED (0.775 g/ml) were added and the solution was immediately poured into the space between the gel-forming plates using a 10 ml pipette. The gel was allowed to polymerize for 30 min. The electrophoresis apparatus was filled with 1X TBE buffer. Each PCR product was mixed with the 6X DNA loading dye supplied with the GeneRuler DNA ladder (ThermoFisher Scientific). 10 µl of each PCR product sample or the 25-700 bp GeneRuler DNA ladder was loaded into the wells of the 12% polyacrylamide gel. The 	   57	  electrophoresis apparatus was then connected to the power source and the gel was run for 1.5 h at 100 V followed by staining with Gelgreen stain (diluted from 10,000X stock solution to 3X staining solution in milli-Q water) for 30 min. Subsequently, the gel was transferred to the Fluorchem Q image analysis cabinet (Cell Biosciences) and exposed to UV light. The image of the gel was captured using the AlphaView Q 3.0 gel acquisition software (Cell Biosciences).   2.13 Effects of varying insulin concentrations on the release of pro-inflammatory cytokines by primary human astrocytes   Primary human astrocytes were harvested and re-suspended in F10 medium to achieve a final concentration of 0.2 million cells/ml.  0.1 ml of primary human astrocyte cell culture was seeded into each well of a sterile 96-well plate and incubated for 24 h at 37 °C. Following 24 h incubation, the media were replaced with 0.2 ml of F5 media and the cells were allowed to recover for 15 min in the incubator. Insulin stock (1 mM) was diluted in F5 media to make a series of aliquots of different insulin concentrations. Moreover, a separate aliquot series of different insulin concentrations was prepared using a denatured insulin stock solution (heated at 100 °C for 5 min). Subsequently, equal volume of insulin solution at different concentrations was added to the respective wells to achieve different insulin concentrations in the wells (1 pM – 1 µM) followed by astrocyte stimulation with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml). Following incubation of cells at 37 °C for 48 h, collected supernatants were frozen at -20 °C. IL-6 and IL-8 concentrations in supernatants were measured by ELISA after thawing the 	   58	  supernatants.   2.14 Statistical analysis   Statistical analysis was performed using the IBM SPSS software (version 16.0). Randomized blocks design one-way or two-way analysis of variance (ANOVA) was used to determine the significance of findings in different experiments. This method was selected to minimize the influence of day-to-day variation in the absolute values of various parameters measured. To determine which of the treatments were significant within each variable, one-way and two-way ANOVA were followed by Fisher’s least-significant difference (LSD) post hoc test and Tukey’s honestly-significant difference (HSD) multiple comparison tests, respectively. qPCR data were tested for normality using the Shapiro-Wilks test in SPSS and one-way ANOVA was performed on the 2ΔCt values. Data are presented as means ± standard error of the mean (S.E.M.). A P-value less than 0.05 was considered statistically significant.          	   59	  Chapter 3. Results   3.1 Toxicity of stimulated U-118 MG astrocytic cells towards SH-SY5Y neuronal cells   Supernatant transfer experiments were conducted (as described in section 2.4) to establish toxicity of U118-MG astrocytic cells towards SH-SY5Y neuronal cells. U-118 MG cells were cultured in F5 medium (17.5 mM glucose) for 24 h. The cells were then stimulated with a combination of different concentrations of IFN-γ (1.2, 6, 30 and 150 U/ml) plus IL-1β (100 U/ml) and incubated for 48 h at 37 °C. Control samples included unstimulated U-118 MG cells (unstimulated control). Stimulation of U-118 MG cells with IL-1β (100 U/ml) only was also performed. At the end of 48 h incubation, the viability of U-118 MG cells was assessed using the LDH release assay (Fig. 5A) and the MTT assay (Fig. 5B). There were no significant effects of different stimulants on U-118 MG cell viability.  Supernatants from the U-118 MG cells were transferred onto the SH-SY5Y neuronal cells; which were subsequently incubated for 72 h at 37 °C. The LDH release (Fig. 6A) and MTT assays (Fig. 6B) were performed on SH-SY5Y cell cultures to determine neuronal cell viability at the end of the 72 h period. Compared to SH-SY5Y cells treated with supernatants from unstimulated astrocytic cells, the LDH release assay showed a trend towards increased cell death in SH-SY5Y cells treated with supernatants from U-118 MG cells stimulated with the highest concentration of IFN-γ (150 U/ml) plus IL-1β (100 U/ml). However, this result did not achieve statistical significance according to the LDH release assay. On the other hand, the MTT assay showed a statistically 	   60	  significant decrease in SH-SY5Y cell viability after their treatment with supernatants from U-118 MG cells stimulated with the highest concentration of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) (P< 0.05). Such a decrease in SH-SY5Y cell viability was not seen when the cells were treated with supernatants from U-118 MG cells stimulated with increasing concentrations of IFN-γ only. Similarly, supernatants of U-118 MG cells stimulated with IL-1β only did not show any toxicity towards SH-SY5Y neuronal cells.          	   61	   Figure 5: Effect of 48 h stimulation on U-118 MG astrocytic cell viability. U-118 MG cells were stimulated with increasing concentrations of IFN-γ in the presence or absence of 100 U/ml of IL-1β. Following 48 h incubation at 37 °C, cell viability was assessed by the LDH release (A) and MTT (B) assays. Data from five independent experiments are presented. The concentration-dependent effects of the stimulants were assessed by the randomized blocks design one-way ANOVA, followed by Fisher’s LSD post-hoc test.  No statistically significant differences were observed.    IFN-γ conc. (U/ml)% Cell death01.2 630150010203040IFN-γ only IFN-γ + IL-1βA.  U-118 MG cell death: LDH AssayIFN-γ conc. (U/ml)% Cell viability01.2 630150050100150200IFN-γ only IFN-γ + IL-1βB.  U-118 MG cell viability: MTT Assay	   62	  Figure 6: Viability of SH-SY5Y cells treated with supernatants from U-118 MG cells stimulated for 48 h with increasing concentrations of IFN-γ in the presence or absence of 100 U/ml of IL-1β. Supernatants of U-118 MG astrocytic cells stimulated with various concentrations of stimulants were transferred onto SH-SY5Y neuronal cells. The viability of SH-SY5Y cells was assessed using the LDH release (A) and MTT (B) assays after 72 h incubation at 37 °C. Data from five independent experiments are presented. The concentration-dependent effects of the stimulants were assessed by the randomized blocks design one-way ANOVA, followed by Fisher’s LSD post hoc test. *P<0.05, significantly different from cells treated with supernatants from unstimulated U-118 MG cells. IFN-γ conc. (U/ml)% Cell death01.2 630150010203040IFN-γ only IFN-γ + IL-1βA.  SH-SY5Y cell death: LDH AssayIFN-γ conc. (U/ml)% Cell viability01.2 630150050100150200IFN-γ only IFN-γ + IL-1βB.  SH-SY5Y cell viability: MTT Assay*	   63	  3.2 Toxicity of U-118 MG cells incubated in different glucose concentrations towards SH-SY5Y cells  The effect of high glucose concentrations on toxicity of U-118 MG astrocytic cells towards SH-SY5Y neuronal cells was determined using supernatant transfer experiments (as described in section 2.5). These experiments were conducted in three different ways.  In the first set of experiments, U-118 MG cells were cultured in F5 low glucose medium for 24 h. The cells were then stimulated with IFN-γ (150 U/ml) plus IL-1β (100 U/ml) in different glucose concentrations (5.5 and 30.5 mM) followed by 48 h incubation at 37 °C.  The control samples included unstimulated U-118 MG cells in different glucose concentrations (5.5 and 30.5 mM) and osmolarity controls (5.5 mM glucose and 25.0 mM D-mannitol). Supernatants of U-118 MG cells were transferred onto SH-SY5Y cells grown in F5 low glucose medium (5.5 mM glucose). At the time of transfer, some wells containing SH-SY5Y received fresh medium with different glucose concentrations (5.5 and 30.5 mM) instead of astrocytic cell supernatants. After 72 h incubation at 37 °C, SH-SY5Y cell viability was measured using the MTT assay (Fig. 7). The results were analyzed using randomized blocks design two-way ANOVA; followed by Tukey’s HSD test for multiple comparisons. Supernatants from both unstimulated and stimulated U-118 MG cells decreased the viability of SH-SY5Y cells to a similar extent compared to treatment with fresh medium only (P< 0.01). Comparing the 5.5 mM and 30.5 mM glucose concentrations, SH-SY5Y cells showed a decreased viability when treated with U-118 MG cells incubated in low glucose concentration (5.5 mM) compared to high 	   64	  glucose concentration (30.5 mM) (P< 0.05). As expected, the viability of cells stimulated under conditions of osmolarity control was similar to that of SH-SY5Y cells incubated in low glucose supernatants.   Figure 7: Viability of SH-SY5Y cells treated with supernatants from U-118 MG cells that were left unstimulated or were stimulated with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) in the presence of two different glucose concentrations (5.5 and 30.5 mM). SH-SY5Y cells were cultured in F5 low glucose medium. At the time of supernatant transfer, some cells were treated with fresh medium containing different glucose concentrations (see ‘Medium’ 5.5 and 30.5 mM). The viability of SH-SY5Y cells was assessed using the MTT assay after 72 h incubation at 37 °C. Data from five independent experiments are presented. Randomized block design two-way ANOVA followed by Tukey’s HSD multiple comparison test was used to determine the effect of different glucose concentrations and different treatments on SH-SY5Y viability. Unstimulated and stimulated treatments were significantly different from medium treatment (##P<0.01); however there was no difference between unstimulated and stimulated treatments. In terms of glucose concentrations, SH-SY5Y cell viability in the presence of 30.5 mM glucose was significantly higher than cell viability in the presence of 5.5 mM glucose concentration (*P<0.05).   The decreased viability of SH-SY5Y cells when treated with U-118 MG cells stimulated in low glucose concentration (5.5 mM) compared to high glucose 	   65	  concentration (30.5 mM) could be due to starvation of cells. Since these cells are continually using glucose from their growth medium/supernatants for metabolism, it is impossible to precisely control the glucose concentrations during the incubation period of the cells. To estimate the actual glucose concentrations in the cell culture medium over the incubation period, U-118 MG and SH-SY5Y cells were incubated in media with different glucose concentrations, and glucose concentrations in cell supernatants were monitored using the One Touch Ultra Mini blood glucose monitoring system. As seen in Figure 8A, when U-118 MG cells were incubated in different glucose concentrations for 48 h, the glucose levels in the 5.5 mM sample fell below the detection limit (1.1 mM) of the One Touch Ultra Mini blood glucose monitoring system after 36 and 48 h of incubation. Glucose levels were low (~1.6 mM), but still detectable in SH-SY5Y cell cultures (Fig. 8B) at the end of the 72 h incubation period. Since supernatant transfer experiments involve incubation of U-118 MG cells for 48 h, glucose concentrations in supernatants from U-118 MG cells incubated in 5.5 mM glucose medium would be below 1.1 mM. Therefore, SH-SY5Y cells treated with these supernatants would run out of glucose in 24-48 h.  To overcome this technical issue, several sets of U-118 MG supernatants stimulated in the presence of different glucose concentrations (5.5, 10.5, 20.5 and 30.5 mM) were frozen at -20 °C. During the 72 h incubation of SH-SY5Y cells, supernatants were replaced every 24 h with freshly thawed aliquots of U-118 MG supernatants. The SH-SY5Y cell viability was assessed at the end of the 72 h incubation period using the MTT assay (Fig. 9). No significant difference was seen in the cell viability of SH-SY5Y 	   66	  Figure 8: Time-dependent glucose consumption by U-118 MG cells (A) and SH-SY5Y cells (B) initially placed in different glucose concentrations. Data from three independent experiments are presented. Based on one-way ANOVA, there was a statistically significant time-dependent decrease in glucose concentration in all samples except the U-118 MG and SH-SY5Y cells incubated in 30.5 mM glucose.   0 12 24 36 48 0102030Time (h)Glucose conc. (mM)5.5 mM Glc 10.5 mM Glc 20.5 mM Glc 30.5 mM Glc A.  U-118 MG cells Time (h)Glucose conc. (mM)0 12 24 36 48 60 720102030B. SH-SY5Y cells 	   67	   Figure 9: Viability of SH-SY5Y cells treated with supernatants from U-118 MG cells that were left unstimulated or were stimulated with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) in the presence of different glucose concentrations (5.5, 10.5, 20.5 and 30.5 mM) for 48 h. SH-SY5Y cells were cultured in F5 low glucose medium. The viability of SH-SY5Y cells was assessed using the MTT assay after 72 h incubation at 37 °C with replacement of fresh supernatants every 24 h. Data from five independent experiments are presented. Randomized block design two-way ANOVA followed by Tukey’s HSD multiple comparison test was used to assess the effect of different glucose concentrations and different treatments on SH-SY5Y viability. There were no significant differences in viability of SH-SY5Y cells exposed to different treatments.  cells exposed to the various treatments. Viability of SH-SY5Y cells was above 65% for all treatments.   The last set of supernatant transfer experiments involved similar treatment of U-118 MG cells as discussed above, except that SH-SY5Y cells were pre-treated with a high glucose concentration of 30.5 mM for 24 h prior to treatment with U-118 MG supernatants. SH-SY5Y cell viability was measured after 72 h incubation at 37 °C. The MTT assay (Fig. 10) showed a decreased viability of SH-SY5Y treated with U-118 MG supernatants compared to cells treated with fresh medium (P< 0.01). However, there was 5.5 30.5 5.5 30.5 5.5 +25.0 mMMannitol5.5 10.5 20.5 30.5 5.5 +25.0 mMMannitol050100150% Cell viabilityGlucose conc. (mM) MediumUnstimulatedStimulated	   68	  no difference between the viability of SH-SY5Y cells treated with supernatants from stimulated and unstimulated U-118 MG cells. Within the samples exposed to unstimulated and stimulated U-118 MG cell supernatants, SH-SY5Y cells treated with 30.5 mM glucose concentration show a decreased viability compared to the cells treated with 5.5 mM glucose concentration (P< 0.05).  Figure 10: Viability of SH-SY5Y cells treated with supernatants from U-118 MG cells that were left unstimulated or were stimulated with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) in the presence of different glucose concentrations (5.5, 10.5, 20.5 and 30.5 mM) for 48 h. SH-SY5Y cells were pre-treated with F5 media containing 30.5 mM glucose for 24 h prior to transfer of U-118 MG cell supernatants. At the time of supernatant transfer, some cells were treated with fresh medium containing different glucose concentrations (5.5 and 30.5 mM). The viability of SH-SY5Y cells was assessed by the MTT assay after 72 h incubation at 37 °C. Data from eight independent experiments are presented. Randomized block design two-way ANOVA followed by Tukey’s HSD multiple comparison test was used to determine the effect of different glucose concentrations and different treatments on SH-SY5Y viability. Unstimulated and stimulated treatments were significantly different from medium treatment (##P<0.01); however there was no difference between unstimulated and stimulated treatments. In terms of glucose concentrations, SH-SY5Y cell viability in the presence of 30.5 mM glucose concentration was significantly lower when compared to 5.5 mM glucose concentration (*P<0.05). 	   69	  3.3 Effects of high glucose on pro-inflammatory cytokines mRNA expression and secretion by astrocytic cells   3.3.1 Gene expression of pro-inflammatory cytokines  qPCR was used to determine the effects of high glucose on mRNA expression of the pro-inflammatory cytokines IL-6, IL-8 and MCP-1. Figures 11 and 12 illustrate the expression of IL-6 and IL-8 by U-118 MG cells and primary human astrocytes respectively. Cells were pre-treated with different concentrations of glucose for 24 h followed by stimulation with IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for additional 24 h. Values obtained were normalized to the reference gene β-actin and expressed as a fold increase compared to the mRNA level in unstimulated cells incubated in F5 low glucose (expression level = 1). In both cells types, high glucose (30.5 mM) alone did not affect the gene expression of IL-6 and IL-8. Stimulation in the presence of 5.5 mM glucose led to a 240- and 175-fold increase in IL-6 expression by U-118 MG cells (Fig. 11A) and primary human astrocytes (Fig. 12A) respectively. Moreover, stimulation in the presence of 30.5 mM glucose led to a statistically significant increase in expression of IL-6 compared to stimulation in the presence of 5.5 mM glucose: 420- and 250-fold increase in U-118 MG cells (Fig. 11A) and primary human astrocytes (Fig. 12A) respectively (P< 0.05).  There was a 35- and 220-fold increase in IL-8 gene expression by stimulated U-118 MG cells (Fig. 11B) and primary human astrocytes (Fig. 12B) respectively when compared to unstimulated cells. However, stimulation in the presence of high glucose did 	   70	  not affect the IL-8 expression in both U-118 MG cells (Fig. 11B) and primary human astrocytes (Fig. 12B). MCP-1 data (not shown) obtained from U-118 MG cells and primary human astrocytes followed similar trends as the IL-8 data. All statistical analyses were performed on 2ΔCt values and data were tested for normality using the Shapiro-Wilk test (P> 0.05) in SPSS.            	   71	   Figure 11: Effect of high glucose on IL-6 (A) and IL-8 (B) mRNA expression by unstimulated and stimulated U-118 MG astrocytic cells. U-118 MG cells were incubated in the presence of two different glucose concentrations (5.5 and 30.5 mM) for 24 h followed by stimulation with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for additional 24 h. Total RNA was extracted and mRNA expression was analyzed using qPCR. Results were normalized to the reference gene β-actin and are presented as a fold increase compared to the mRNA level in unstimulated U-118 MG cells incubated in F5 low glucose (expression level = 1). Data from five independent experiments are presented. Differences in expression levels were assessed by the randomized blocks design one-way ANOVA, followed by Fisher’s LSD post-hoc test. *P<0.05; significantly different from cells stimulated in F5 low glucose (5.5 mM).  30.5 5.5 + 25.0 mM Mannitol5.5 30.5 5.5 +25.0 mM Mannitol012345200400600800Fold change over unstimulated U-118 MG cells*UnstimulatedStimulated Glucose conc. (mM)A.  U-118 MG IL-6 gene expression30.5 5.5  + 25.0 mM Mannitol5.5 30.5 5.5 + 25.0 mM Mannitol01234520304050Fold change over unstimulated U-118 MG cellsGlucose conc. (mM)UnstimulatedStimulated B.  U-118 MG IL-8 gene expression	   72	    Figure 12: Effect of high glucose on IL-6 (A) and IL-8 (B) mRNA expression by unstimulated and stimulated human astrocytes. Primary human astrocytes were incubated in two different glucose concentrations (5.5 and 30.5 mM) for 24 h followed by stimulation with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for additional 24 h. Total RNA was extracted and mRNA expression was analyzed using qPCR. Results were normalized to the reference gene β-actin and are presented as a fold increase compared to the mRNA level in unstimulated primary human astrocytes incubated in F5 low glucose (expression level = 1). Data from five independent experiments are presented. Human astrocytes isolated from two different human surgical cases were used. Differences in expression levels were assessed by the randomized blocks design one-way ANOVA, followed by Fisher’s LSD post-hoc test. *P<0.05; significantly different from cells stimulated in F5 low glucose (5.5 mM). 30.5 5.5 + 25.0 mM Mannitol5.5 30.5 5.5 +25.0 mM Mannitol012345100200300400Fold change over unstimulated primary human astrocytes*UnstimulatedStimulated Glucose conc. (mM)A.  Primary human astrocyte IL-6 gene expressionFold change over unstimulated primary human astrocytes30.5 5.5 + 25.0 mM Mannitol5.5 30.5 5.5 +25.0 mM Mannitol012345100200300400Glucose conc. (mM)UnstimulatedStimulated B.  Primary human astrocyte IL-8 gene expression	   73	  3.3.2 Secretion of pro-inflammatory cytokines  The effect of glucose on the secretion of IL-6 and IL-8 by U-118 MG and U-373 MG astrocytic cells as well as primary human astrocytes was investigated using ELISA. Osmolarity controls containing 5.5 mM glucose + 25.0 mM D-mannitol or L-glucose were incorporated. In all ELISA experiments, the osmolarity controls showed similar results to the 5.5 mM glucose sample. As illustrated in Figure 13A, unstimulated U-118 MG cells did not secrete significant amounts of IL-6 (values were below the detection limit of 0.5 ng/ml). Based on randomized blocks design two-way ANOVA, stimulated U-118 MG cells showed a significant increase in IL-6 secretion compared to unstimulated cells (P< 0.01). Within the data obtained from stimulated cells, increasing glucose concentrations (10.5, 20.5 and 30.5 mM) further enhanced IL-6 secretion compared to the 5.5 mM glucose concentration (P< 0.05).  Secretion of IL-8 was measured in both unstimulated and stimulated U-118 MG cells (Fig. 13B).  Using randomized blocks design two-way ANOVA, it was found that IL-8 secretion was significantly higher in stimulated cells compared to unstimulated cells (P< 0.05). Within each data set, increasing glucose concentrations (20.5 and 30.5 mM in unstimulated cells and 30.5 mM in stimulated cells) were associated with enhanced IL-8 secretion (P< 0.01 for unstimulated data and P< 0.05 for stimulated data).      	   74	   Figure 13: Effect of high glucose on secretion of pro-inflammatory cytokines IL-6 (A) and IL-8 (B) by U-118 MG astrocytic cells. U-118 MG cells were incubated in the presence of different glucose concentrations (5.5, 10.5, 20.5 and 30.5 mM) for 24 h followed by stimulation with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for additional 24 h. Supernatants were collected and analyzed for levels of the cytokines using specific ELISAs. Data from nine independent experiments are presented. Differences in cytokine secretion in each data set (unstimulated and stimulated) were assessed by the randomized blocks design one-way ANOVA, followed by Fisher’s LSD post hoc test. *P<0.05, **P<0.01; significantly different from cells incubated in F5 low glucose (5.5 mM).  5.5 10.5 20.5 30.5 5.5 + 25.0 mM Mannitol5.5 10.5 20.5 30.5 5.5 + 25.0 mM Mannitol   L-glucose012345IL-6 secretion (ng/ml)Stimulated Unstimulated***Glucoseconc. (mM) A.  U-118 MG IL-6 secretion5.5 10.5 20.5 30.5 5.5 + 25.0 mM Mannitol5.5 10.5 20.5 30.5 5.5 + 25.0 mM Mannitol   L-glucose05101520IL-8 secretion (ng/ml)Unstimulated Stimulated *****Glucoseconc. (mM) B.  U-118 MG IL-8 secretion	   75	  In contrast to U-118 MG cells, U-373 MG cells showed significant secretion of both IL-6 and IL-8 irrespective of stimulation. Based on randomized blocks design two-way ANOVA, stimulated U-373 MG cells showed a statistically significant increase in IL-6 secretion compared to unstimulated cells (P< 0.05) (Fig. 14A). Within each data set, increasing glucose concentrations (20.5 and 30.5 mM) were associated with enhanced IL-6 secretion (P< 0.01). In contrast, there were no statistically significant differences in IL-8 secretion between unstimulated and stimulated U-373 MG cells or cells exposed to different glucose concentrations (Fig. 14B).  Similar to U-118 MG cells, IL-6 secretion by unstimulated primary human astrocytes was below the detection limit of the ELISA (IL-6 detection limit: 0.2 ng/ml) (Fig. 15A). According to the randomized block design two-way ANOVA, stimulated primary human astrocytes showed a statistically significant increase in IL-6 secretion compared to unstimulated cells (P< 0.01). Increasing glucose concentrations (20.5 and 30.5 mM) in stimulated primary human astrocytes led to a statistically significant increase in IL-6 secretion compared to cells stimulated in 5.5 mM glucose (P< 0.05). Stimulation led to a statistically significant increase in IL-8 secretion compared to unstimulated cells (Fig. 15B) (P< 0.01). Within the stimulated data, there was an increased IL-8 secretion by cells incubated in 30.5 mM glucose compared to 5.5 mM glucose (P< 0.05).     	   76	   Figure 14: Effect of high glucose on secretion of pro-inflammatory cytokines IL-6 (A) and IL-8 (B) by U-373 MG astrocytic cells. U-373 MG cells were incubated in the presence of different glucose concentrations (5.5, 10.5, 20.5 and 30.5 mM) for 24 h followed by stimulation with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for additional 24 h. Supernatants were collected and analyzed for levels of the cytokines using specific ELISAs. Data from five independent experiments are presented. Differences in cytokine secretion in each data set (unstimulated and stimulated) were assessed by the randomized blocks design one-way ANOVA, followed by Fisher’s LSD post hoc test. **P<0.01; compared to cells incubated in F5 low glucose (5.5 mM).  IL-6 secretion (ng/ml)5.5 10.5 20.5 30.5 5.5 + 25.0 mM Mannitol5.5 10.5 20.5 30.5 5.5 + 25.0 mM Mannitol0246Unstimulated Stimulated ** ******Glucoseconc. (mM) A.  U-373 MG IL-6 secretionIL-8 secretion (ng/ml)5.5 10.5 20.5 30.5 5.5 + 25.0 mM Mannitol5.5 10.5 20.5 30.5 5.5 + 25.0 mM Mannitol05101520Unstimulated Stimulated Glucoseconc. (mM) B.  U-373 MG IL-8 secretion	   77	    Figure 15: Effect of high glucose on secretion of pro-inflammatory cytokines IL-6 (A) and IL-8 (B) by primary human astrocytes. Primary human astrocytes were incubated in the presence of different glucose concentrations (5.5, 10.5, 20.5 and 30.5 mM) and left unstimulated or stimulated with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for 48 h. Supernatants were collected and analyzed for levels of the cytokines using specific ELISAs. Data from five independent experiments are presented. Human astrocytes isolated from two different human surgical cases were used. Differences in cytokine secretion in each data set (unstimulated and stimulated) were assessed by the randomized blocks design one-way ANOVA, followed by Fisher’s LSD post hoc test. *P<0.05; significantly different from cells stimulated in F5 low glucose (5.5 mM).    IL-6 secretion (ng/ml)5.5 10.5 20.5 30.5 5.5 10.5 20.5 30.5 5.5 + 25.0 mM Mannitol   L-glucose0.00.51.01.52.0UnstimulatedStimulated **Glucoseconc. (mM) A.  Primary human astrocyte IL-6 secretionIL-8 secretion (ng/ml)5.5 10.5 20.5 30.5 5.5 10.5 20.5 30.5 5.5 + 25.0 mM Mannitol   L-glucose05101520Unstimulated Stimulated *Glucoseconc. (mM) B.  Primary human astrocyte IL-8 secretion	   78	  3.4 Intracellular signaling pathways activated by high glucose in astrocytic cells   U-118 MG astrocytic cells were incubated in different glucose concentrations for 24 h at 37 °C. Following incubation, the cells were either left unstimulated or stimulated with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for 30 min and the whole cell protein was extracted. 30 µg of extracted proteins was separated on 8% SDS-PAGE and transferred onto nitrocellulose membrane. For both STAT-3 and JNK, no phosphorylation was observed in any of the samples. Total STAT-3 and total JNK were observed in all samples as shown in Figure 16. For both p38 MAPK and p44/42 MAPK, quantification of protein bands was done using densitometry analysis on the Image J software (National Institutes of Health, Maryland, USA). Band densities of phosphorylated proteins were divided by the band densities of the total protein to provide a relative band density. Next, fold activation was expressed relative to the unstimulated sample at 5.5 mM glucose concentration (set at 1). Randomized blocks design one-way ANOVA was performed on the relative band density values. As illustrated in Figure 17A, there was a statistically significant increase in phosphorylation of p38 MAPK in the stimulated samples compared to the unstimulated samples. Within the unstimulated data, there was a trend towards increased p38 MAPK phosphorylation in 30.5 mM glucose compared to the 5.5 mM glucose. However, this difference was not statistically significant. No statistically significant differences were observed for p44/42 MAPK activation between different samples (Fig. 17B).  	   79	   	  	    Figure 16: Western blots for total STAT-3 (A) and JNK (B) obtained by separating U-118 MG cell lysates on 8% SDS-PAGE and immunoblotting using specific antibodies. Whole cell protein was extracted from U-118 MG cells incubated in different glucose concentrations for 24 h followed by stimulation with IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for 30 min. Blot patterns are representative of three independent experiments.        	   80	     Figure 17: Western blots for phospho- and total p38 MAPK (A) and p44/42 MAPK obtained by separating U-118 MG cell lysates on 8% SDS-PAGE and immunoblotting using specific antibodies. Whole cell protein was extracted from U-118 MG cells incubated in different glucose concentrations for 24 h followed by stimulation with IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for 30 min. Blot patterns are representative of three independent experiments. Densitometry analysis was performed using the Image J software to quantify protein bands. Relative band densities were calculated by dividing band densities of phosphorylated proteins by the band densities of the total. Fold activation was expressed relative to the 5.5 mM glucose unstimulated sample (set at 1). Randomized blocks design one-way ANOVA was performed on the relative band density values, followed by Fisher’s LSD post-hoc test. *P<0.05; significantly different from unstimulated cells incubated in the same glucose concentration. 	   81	  3.5 Effect of high glucose on susceptibility of neurons to a cytotoxic insult by hydrogen peroxide or Aβ42  3.5.1 Undifferentiated SH-SY5Y cells   To investigate whether high glucose concentrations increase susceptibility of neuronal cells to an injurious stimulus, SH-SY5Y cells were incubated in the presence of increasing glucose concentrations (5.5, 7.5, 10.5, 20.5 and 30.5 mM) for 24 h followed by treatment with two different concentrations of hydrogen peroxide (0.25 and 0.5 mM) for additional 24 h. The viability of SH-SY5Y cells was assessed at the end of the experiment by the LDH release (Fig. 18A) and MTT (Fig. 18B) assays. The percentage death of SH-SY5Y cells treated with 0.25 mM hydrogen peroxide was not statistically different from the percentage cell death of untreated SH-SY5Y cells (Fig. 18A). However, 0.5 mM hydrogen peroxide caused increased death of SH-SY5Y cells compared to the untreated controls (P<0.001). Moreover, within the samples treated with 0.5 mM hydrogen peroxide, incubation in the presence of 10.5, 20.5 and 30.5 mM glucose led to a statistically significant increase in cell death compared to the cells treated with 0.5 mM hydrogen peroxide in the presence of 5.5 mM glucose (P< 0.01). Results of the MTT assay (Fig. 18B) were in agreement with the LDH release assay except that the percentage viability of SH-SY5Y cells treated with both 0.25 mM and 0.5 mM hydrogen peroxide concentrations was significantly decreased compared to the untreated cells (P<0.05). Similar to the results obtained with the LDH release assay, the MTT data showed that SH-SY5Y cells treated with 0.5 mM hydrogen peroxide in the presence of 	   82	  10.5, 20.5 and 30.5 mM glucose showed a significant decrease in viability compared to the cells treated with the same hydrogen peroxide concentration in the presence of 5.5 mM glucose (P<0.001).      	   83	  Figure 18: Viability of SH-SY5Y neuronal cells treated with hydrogen peroxide in the presence of different glucose concentrations. SH-SY5Y cells were pre-treated with varying glucose concentrations (5.5, 7.5, 10.5, 20.5 and 30.5 mM) for 24 h. The viability of SH-SY5Y cells was assessed using the LDH release (A) and MTT (B) assays after 24 h treatment with two different concentrations of hydrogen peroxide (0.25 and 0.5 mM). Data from six to twelve independent experiments are presented. Randomized blocks design two-way ANOVA, followed by Tukey’s HSD multiple comparison tests was performed. **P<0.01 and ***P<0.001; significantly different from cells treated with 0.5 mM hydrogen peroxide in 5.5 mM glucose. The viability of SH-SY5Y cells treated with 0.25 mM (B) and 0.5 mM (A,B) hydrogen peroxide concentrations was significantly decreased compared to the untreated cells across all glucose concentrations (P<0.05). % Cell death5.5 7.5 10.5 20.5 30.5 5.5 +25.0 mM Mannitol0204060Untreated0.25 mM H2O20.5 mM H2O2******Glucose conc. (mM) A.  SH-SY5Y cell death: LDH Assay% Cell viability5.5 7.5 10.5 20.5 30.5 5.5 +25.0 mM Mannitol050100150Untreated0.25 mM H2O20.5 mM H2O2*********Glucose conc. (mM) B.  SH-SY5Y cell viability: MTT Assay	   84	  Since Aβ deposits are considered to play a critical role in neuronal death in the brains of AD patients, toxicity of Aβ42 (5 and 10 µM) towards SH-SY5Y neuronal cells was investigated in the presence of different glucose concentrations. However, Aβ42 was not toxic to undifferentiated SH-SY5Y cells at concentrations tested (data not shown).   3.5.2 Differentiated SH-SY5Y cells   SH-SY5Y neuronal cells differentiated by 10 µM retinoic acid were used as another model of human neurons. Phase contrast microscopy images of SH-SY5Y cells were taken at day 1, 4 and 7 of the differentiation period. Figure 19 shows the morphological changes in SH-SY5Y neuronal cells over the course of their differentiation. Differentiated SH-SY5Y neuronal cells show long and extensively branched processes. Viability of differentiated SH-SY5Y cells was assessed using the MTT assay after treatment with 0.25 mM and 0.5 mM hydrogen peroxide for 24 h (Fig. 20) or 5 µM Aβ42 for 72 h (Fig. 21). As illustrated in Figure 20, in untreated samples, incubation in the presence of 30.5 mM glucose and the mannitol control led to a small but statistically significant decrease in cell viability compared to cells incubated in the presence of 5.5 mM glucose (P< 0.05). Moreover, when treated with 0.25 mM hydrogen peroxide, the cells incubated in the presence of mannitol showed a decreased viability compared to the cells incubated in the presence of 5.5 mM glucose (P< 0.05). The decreased viability of cells from mannitol control samples compared to the samples containing 5.5 mM glucose in both untreated cells and cells treated with 0.25 mM hydrogen peroxide was unexpected. There was a statistically significant decrease in the 	   85	  viability of differentiated SH-SY5Y cells treated with 0.5 mM hydrogen peroxide compared to the untreated controls (P<0.01). However, no differences in viability were observed between the cells treated with hydrogen peroxide and incubated in the presence of 5.5 vs. 30.5 mM glucose concentrations.   Figure 19: Phase contrast images of SH-SY5Y neuronal cells at day 1 (A), day 4 (B) and day 7 (C) of differentiation with 10 µM retinoic acid. Differentiation was done as described in section 2.11.2. Photos are representative of three independent experiments. Magnification bars in A-C = 100 µm.    	   86	  Figure 20: Viability of retinoic acid-differentiated SH-SY5Y neuronal cells treated with hydrogen peroxide in the presence of different glucose concentrations. Retinoic acid differentiated SH-SY5Y cells were treated with two different hydrogen peroxide concentrations (0.25 and 0.5 mM) in the presence of two different glucose concentrations (5.5 and 30.5 mM). The viability of SH-SY5Y cells was assessed using the MTT assay after 24 h incubation at 37 °C. Data from six independent experiments are presented. Randomized blocks design two-way ANOVA, followed by Tukey’s HSD multiple comparison tests was performed. **P<0.01; cells treated with 0.5 mM hydrogen peroxide showed a statistically significant decline in viability compared to untreated cells in the presence of the same glucose concentration suggesting toxicity of hydrogen peroxide. #P<0.05; incubation in 30.5 mM glucose or the mannitol control led to a small decline in viability compared to incubation in 5.5 mM glucose even in the absence of any treatment. ξP<0.05; toxicity of 0.25 mM hydrogen peroxide was surprisingly more pronounced in the mannitol control compared to cells incubated in the presence of 5.5 mM glucose.  Similar to untreated samples in Figure 20, incubation of differentiated SH-SY5Y cells for 72 h led to a decrease in viability in the presence of 30.5 mM glucose compared to the 5.5 mM glucose concentration (Fig. 21). However, as expected, viability of cells incubated in the presence of mannitol control was similar to the 5.5 mM glucose sample. Treatment with 5 µM Aβ42 led to a slight decrease in viability in the presence of 5.5 mM % Cell viability5.5 30.5 5.5 + 25.0 mM Mannitol050100150Untreated0.25 mM H2O20.5 mM H2O2Glucose conc. (mM) ******##ξ	   87	  glucose. However, in the presence of 30.5 mM glucose, Aβ42 was more toxic by ~40% compared to its activity in the presence of 5.5 mM glucose. Viability of cells treated with Aβ42 in the presence of 30.5 mM glucose was statistically lower when compared to untreated samples at 30.5 mM glucose. Toxicity was not observed when cells were treated with Aβ solvent in the presence of 30.5 mM glucose (data not shown).  Figure 21: Viability of retinoic acid-differentiated SH-SY5Y neuronal cells treated with Aβ42 in the presence of different glucose concentrations. Retinoic-acid differentiated SH-SY5Y cells were treated with 5 µM Aβ42 in the presence of two different glucose concentrations (5.5 and 30.5 mM). The viability of SH-SY5Y cells was assessed using the MTT assay after 72 h incubation at 37 °C. Data from six independent experiments are presented. Randomized blocks design two-way ANOVA, followed by Tukey’s HSD multiple comparison tests was performed. *P<0.05; cells treated with 5 µM Aβ42 showed a statistically significant decline in viability compared to untreated cells in the presence of the same glucose concentration suggesting toxicity of Aβ42. #P<0.05; incubation in 30.5 mM glucose led to a small decline in viability compared to incubation in 5.5 mM glucose even in the absence of any treatment. ξP<0.05; Aβ42 toxicity was more pronounced in 30.5 mM glucose as indicated by the significant decline in viability of cells treated with Aβ42 in the presence of 30.5 mM glucose compared to cells treated with Aβ42 in the presence of 5.5 mM glucose.  % Cell viability5.5 30.5 5.5 +25.0 mM Mannitol050100150Untreated5 µM Aβ42*#Glucose conc. (mM) * *ξ	   88	  3.6 Expression of insulin signaling components in glia-like cell lines and primary glial cells  Expression of genes for different insulin receptors and receptor signaling molecules was studied in various human cell lines and primary human glial cells. RNA was extracted from each cell type, cDNA was synthesized and subsequently amplified using RT-PCR. cDNA from primary human microglia was available from previous experiments. G3PDH was used as the housekeeping gene. mRNA for G3PDH housekeeping gene was detected in all samples. Figure 22 shows the mRNA expression of different insulin signaling components by SH-SY5Y neuroblastoma cells and HepG2 hepatocytoma cells. SH-SY5Y neuroblastoma cells expressed all insulin signaling components whereas HepG2 hepatocytoma cells did not express IRS-2. U-118 MG astrocytoma cells (Fig. 23A), primary human astrocytes (Fig. 23B) and primary human microglia (Fig. 23B) expressed all receptor-signaling components even though the DNA bands for certain gene products were faint (e.g., INSRA for primary human astrocytes and INSRB for primary human microglia).            	   89	   Figure 22: DNA polyacrylamide gel showing gene expression of different insulin receptors and receptor signaling molecules by SH-SY5Y neuroblastoma cells (left) and HepG2 hepatocytoma cells (right). Total RNA was extracted from respective cell types and amplified via RT-PCR using primers for INSR, INSRA, INSRB, IRS-1 and IRS-2. All human cell lines were used without prior differentiation. The amplified PCR products were separated on 12% polyacrylamide gels and visualized using Gelgreen dye. Expression patterns of mRNAs by both cell types are representative of three independent experiments. 25-700 bp DNA ladder can be seen in the lane marked MW (molecular weight).    	   90	    Figure 23: DNA polyacrylamide gels showing gene expression of different insulin receptors and receptor signaling molecules by U-118 MG astrocytic cells (A), primary human astrocytes (B; left) and primary human microglia (B; right). Total RNA was extracted from respective cell types and amplified via RT-PCR using primers for INSR, INSRA, INSRB, IRS-1 and IRS-2. cDNA from primary human microglia was available from previous experiments. The amplified PCR products were separated on 12% polyacrylamide gels and visualized using Gelgreen dye. Expression patterns of mRNAs by different cell types are representative of three independent experiments that were performed on cells from at least two different surgical cases for both primary human astrocytes and microglia. 25-700 bp DNA ladder can be seen in the lanes marked MW (molecular weight). Figure 23B was kindly provided by Spielman, L.   	   91	  3.7 Effect of insulin on secretion of pro-inflammatory cytokines by primary human astrocytes   The effect of a range of insulin concentrations on the secretion of IL-6 and IL-8 by primary human astrocytes was investigated using ELISA. Unstimulated primary human astrocytes showed negligible secretion of IL-6 and IL-8 (below detection limit of the respective assays: 0.2 ng/ml for IL-6 and 0.5 ng/ml for IL-8), which was not affected by insulin. Therefore, data from unstimulated primary human astrocytes are not shown.  Primary human astrocytes were also stimulated with a combination of IFN-γ (150 U/ml) and IL-1β (100 U/ml) in the presence of different insulin concentrations (1 pM – 1 µM) for 48 h. In addition, this experiment was repeated with insulin denatured by heating at 100 °C for 5 min. As illustrated in Figures 24A and 25A, IL-6 and IL-8 secretion by stimulated primary human astrocytes followed a bell shaped curve with increasing insulin concentrations. Primary human astrocytes incubated in the presence of 1 nM insulin showed a statistically significant increase in IL-6 secretion compared to the cells incubated in the absence of insulin (P< 0.01). Similarly, primary human astrocytes incubated in the presence of 1 and 10 nM insulin showed a statistically significant increase in IL-8 secretion compared to the cells incubated in the absence of insulin (P< 0.01). Heat-denatured insulin had no effect on the IL-6 and IL-8 secretion by stimulated primary human astrocytes (Figs. 24B and 25B).   	  	  	   92	  Figure 24: IL-6 secretion (A) by primary human astrocytes stimulated with IFN-γ (150 U/ml) plus IL-1β (100 U/ml) in the presence of different insulin concentrations (1 pM to 1 µM) for 48 h. The experiments were also repeated by using denatured insulin at different concentrations (B). Data from six to eleven independent experiments are presented. Human astrocytes isolated from two different human surgical cases were used. The dotted lines represent the detection limit of the assay. IL-6 secretion by unstimulated cells was below the detection limit of the assay and hence is not shown here. Differences in cytokine secretion were assessed by the randomized blocks design one-way ANOVA, followed by Fisher’s LSD post-hoc test. **P<0.01; significantly different from cells stimulated in F5 media.   IL-6 secretion (ng/ml)0 pM 1 pM 10 pM 100 pM 1 nM 10 nM 100 nM 1 µM0246**Insulin concentration  A.  Stimulation with IFN-γ + IL-1β + insulinIL-6 secretion (ng/ml)0 pM 1 pM 10 pM 100 pM 1 nM 10 nM 100 nM 1 µM01234B.  Stimulation with IFN-γ + IL-1β + heat-inactivated insulinInsulin concentration  	   93	  Figure 25: IL-8 secretion (A) by primary human astrocytes stimulated with IFN-γ (150 U/ml) plus IL-1β (100 U/ml) in the presence of different insulin concentrations (1 pM to 1 µM) for 48 h. The experiments were also repeated by using denatured insulin at different concentrations (B). Data from six to eleven independent experiments are presented. Human astrocytes isolated from two different human surgical cases were used. The dotted lines represent the detection limit of the assay. IL-8 secretion by unstimulated cells was below the detection limit of the assay and hence is not shown here.  Differences in cytokine secretion were assessed by the randomized blocks design one-way ANOVA, followed by Fisher’s LSD post-hoc test. **P<0.01; significantly different from cells stimulated in F5 media.   0 pM 1 pM 10 pM 100 pM 1 nM 10 nM 100 nM 1 µM01234IL-8 secretion (ng/ml)****A.  Stimulation with IFN-γ + IL-1β + insulinInsulin concentration  IL-8 secretion (ng/ml)0 pM 1 pM 10 pM 100 pM 1 nM 10 nM 100 nM 1 µM0123B.  Stimulation with IFN-γ + IL-1β + heat-inactivated insulinInsulin concentration  	   94	   Chapter 4. Discussion  4.1 Effects of high glucose on toxicity of astrocytic cells towards neuronal cells    Effect of different glucose concentrations on the toxicity of astrocytic supernatants towards SH-SY5Y neuronal cells was investigated in vitro by using a cell culture system that had been used previously to show that supernatants of activated astrocytes were toxic towards neuronal cells (Hashioka et al., 2009; Hashioka et al., 2012; Villanueva et al., 2012). Activation of astrocytes in the CNS occurs in response to injury and disease and is characterized by upregulation of glial fibrillary acidic protein (GFAP) expression (Fuller et al., 2009; Ton & Xiong, 2013). GFAP is an intermediate filament protein involved in cell migration and cytoskeletal changes. It is a marker of astrocyte activation and proliferation. Activated astrocytes are neurotoxic due to the production and release of several inflammatory mediators, as well as secretion of excitatory amino acids (Fuller et al., 2009; Steele & Robinson, 2012; Ton & Xiong, 2013).  Peripheral hyperglycemia in T2DM exacerbates inflammation by activating pathways involving macrophages and adipocytes (Wellen & Hotamisligil, 2005; King, 2008). Such inflammatory pathways contribute to development of insulin resistance by impairment of the insulin-signaling cascade (Nieto-Vazquez et al., 2008; Pansuria et al., 2012). Recent studies have indicated the presence of inflammation in the brains of patients with T2DM (Sonnen et al., 2009; Chen et al., 2011). Since persistent activation 	   95	  of glial cells, specifically astrocytes which are the most abundant glial cell type, is a critical contributor to neuroinflammation and neurotoxicity, I hypothesized that activation of astrocytic cells in elevated glucose concentrations causes a concentration-dependent increase in toxicity of astrocytic cell supernatants towards neuronal cells. For these experiments, U-118 MG cells were used as the astrocyte model. Therefore, it was important to first examine the toxicity of U-118 MG supernatants towards SH-SY5Y cells in normal medium.  To establish neurotoxicity, U-118 MG cells were stimulated with increasing concentrations of IFN-γ along with IL-1β for 48 h to cause their activation. Subsequently, their supernatants were transferred onto SH-SY5Y cells. It was found that stimulation of U-118 MG cells with the highest concentration of IFN-γ (150 U/ml) in combination with 100 U/ml IL-1β led to significant toxicity of their supernatants, which caused a decrease in viability of SH-SY5Y neuronal cells. These concentrations of stimulants were used for subsequent experiments.   Most in vitro cell culture studies on peripheral monocytes and brain microglia have classified glucose concentrations below 10 mM (5.5 – 10 mM) as ‘low glucose’ and above 10 mM (10.5 – 35 mM) as ‘high glucose’ (Dasu et al., 2007; Dasu et al., 2008; Pereira Tde et al., 2010; Quan et al., 2011).  To investigate effects of glucose, 5.5 mM glucose was used as the low glucose concentration in this thesis. On the other hand, 10.5, 20.5 and 30.5 mM glucose concentrations were used as high glucose concentrations. U-118 MG cells were stimulated with a combination of IFN-γ (150 U/ml) plus IL-1β (100 U/ml) in the presence of two different glucose concentrations (5.5 and 30.5 mM) for 48 h followed by application of supernatants to SH-SY5Y cells that had been cultured in F5 	   96	  low glucose medium (5.5 mM glucose) for 24 h before the transfer. SH-SY5Y cells treated with U-118 MG cell supernatants showed a significant decrease in viability compared to cells treated with fresh medium only (Fig. 7). However, no differences were observed in the viability of neuronal cells treated with supernatants from stimulated and unstimulated U-118 MG cells. Moreover, contrary to my hypothesis, SH-SY5Y cells treated with supernatants from U-118 MG cells incubated in 5.5 mM glucose showed a significant decrease in viability compared to cells treated with supernatants from U-118 MG cells incubated in 30.5 mM glucose. A statistically significant difference was also observed in viability of SH-SY5Y cells treated with fresh medium only with the cells incubated in 30.5 mM glucose showing higher viability compared to cells incubated in 5.5 mM glucose.   I reasoned that this decrease in viability in the presence of lower glucose concentrations could be caused by starvation of cells owing to the continual usage of glucose in the medium/supernatants for cell metabolism, especially when the medium had a low glucose concentration to begin with. This concept was explored by monitoring changes in glucose concentration in U-118 MG and SH-SY5Y cell cultures when incubated in media containing different initial glucose concentrations. Note that all glucose concentrations were measured relatively accurately at the beginning of the experiment except for 30.5 mM (Fig. 8). The One Touch Ultra Mini blood glucose monitoring system is able to detect glucose concentrations from 1.1 to 33.3 mM (Chang et al., 2009). Therefore, it is likely that the decrease in accuracy of measurements at high glucose concentrations was due to the system approaching its upper detection limit. Glucose concentrations fell below 1.1 mM when U-118 MG cells were incubated in 5.5 	   97	  mM glucose media for 48 h. Since these supernatants were transferred onto SH-SY5Y cells followed by further incubation of SH-SY5Y cells for 72 h, it is likely that SH-SY5Y cells ran out of glucose and therefore showed reduced viability at the end of 72 h incubation.   Consequently, in the subsequent set of experiments, supernatants on SH-SY5Y cells were replaced every 24 h with freshly thawed aliquots of U-118 MG supernatants to avoid depletion of glucose in the cell culture medium. However, at the end of 72 h incubation of SH-SY5Y cells, their viability was above 65% for all treatments indicating that 24 h replacement of supernatants led to loss of toxicity of U-118 MG supernatants towards SH-SY5Y neuronal cells (Fig. 9). The viability of SH-SY5Y cells treated with U-118 MG cells was similar to that of cells treated with fresh medium.  Finally, the last set of experiments in this series involved pre-treatment of SH-SY5Y neuronal cells with high glucose concentration of 30.5 mM for 24 h prior to their exposure to U-118 MG supernatants. This was done to mimic the hyperglycemic conditions in the brain where both glial cells and neurons are being exposed to the high glucose concentrations. When compared to cells treated with fresh medium, there was a decrease in viability of SH-SY5Y cells treated with U-118 MG supernatants. Viability of SH-SY5Y cells treated with supernatants from U-118 MG cells incubated in 5.5 mM glucose dropped to approximately 50% compared to fresh media treatment (Fig. 10). There was a further 15% decrease in viability of SH-SY5Y cells treated with supernatants from U-118 MG cells incubated in 30.5 mM glucose. This was in accordance with the central hypothesis of this thesis since treatment of U-118 MG cells with high glucose 	   98	  concentration of 30.5 mM led to a decrease in viability of SH-SY5Y cells compared to cells treated with 5.5 mM glucose.  The results obtained are consistent with a recent study that demonstrated a 40% reduction in viability of SK-N-SH neuroblastoma cells treated with hyperglycemic (30 mM) supernatants of RBA-1 cells (astrocytic cells derived from primary rat astrocytes) compared to cells treated with control culture medium (Hsieh et al., 2014). Even when incubated in fresh medium, SH-SY5Y cells showed a decreased viability in 30.5 mM glucose compared to 5.5 mM glucose. However, there was no difference between the viability of SH-SY5Y cells treated with supernatants from stimulated and unstimulated U-118 MG cells.  Essentially, Figures 7 and 10 can be seen as providing contradictory results with Figure 7 rejecting the hypothesis and Figure 10 confirming the hypothesis. For both experiments, SH-SY5Y cells were cultured in F5 low glucose medium. The only difference in the two sets of experiments that generated Figures 7 and 10 was that SH-SY5Y cells were pre-treated with high glucose (30.5 mM) for 24 h prior to treatment with U-118 MG supernatants. Therefore, the effect of high glucose on SH-SY5Y cells was more pronounced compared to its effect on cytotoxic secretions of astrocytic cells. The high glucose treatment most likely primed the SH-SY5Y cells in some way making them more susceptible to injury by astrocytic supernatants. This conclusion is also supported by the decrease in SH-SY5Y cell viability by incubation in 30.5 mM media only compared to 5.5 mM media (Fig. 10).  Interestingly, no difference was observed between the viability of SH-SY5Y cells treated with supernatants from stimulated and unstimulated U-118 MG cells in both 	   99	  Figures 7 and 10. This was unexpected since stimulation with the same concentration of stimulants led to a significant decline in viability of SH-SY5Y cells under experimental conditions described in Figure 6B. It is possible that differences in the media composition contributed to this result. To establish neurotoxicity of U-118 MG cells in Figure 6, experiments were conduced in DMEM/F12 media (17.5 mM glucose concentration). On the other hand, for testing of the effects of glucose, DMEM low glucose medium (5.5 mM glucose) had to be used. These media differ in the composition and concentration of various amino acids, vitamins and inorganic salts. Moreover, DMEM low glucose medium does not contain additional ingredients such as biotin, sodium pyruvate, linoleic acid, lipoic acid, copper, zinc and thymidine present in the DMEM/F12 medium. Such differences in composition of media may affect the growth rate and gene expression of cells (Chaudhry et al., 2008; Ahmado et al., 2011). Therefore, the data could be interpreted as loss of effect when U-118 MG cells were stimulated in DMEM low glucose media. Conversely, the results could be interpreted as gain of toxicity by unstimulated cells in the DMEM low glucose media with unstimulated cells showing toxicity similar to stimulated cells.  Either way, it would be imperative to repeat U-118 MG cell toxicity experiments in DMEM low glucose media to find the optimal stimulation concentrations. However, as discussed earlier, SH-SY5Y cells may show reduced viability in these experiments owing to starvation in the presence of abnormally low glucose concentrations. An alternative control experiment would involve using DMEM low glucose medium supplemented with glucose to 17.5 mM.  Comparing results of such an experiment with the data obtained in DMEM/F12 medium (Fig. 6) could 	   100	  reveal the contribution of the additional factors found in DMEM/F12 compared to DMEM medium.  4.2 Effects of high glucose on astrocytic cell-mediated inflammation and intracellular signaling pathways   4.2.1 High glucose enhances gene expression and secretion of pro-inflammatory cytokines by astrocytic cells    Gene expression of pro-inflammatory cytokines IL-6 and IL-8 by U-118 MG cells and primary human astrocytes was investigated in the presence of different glucose concentrations (5.5 and 30.5 mM) using qPCR. High glucose alone did not cause significant increase in gene expression of IL-6 and IL-8 by U-118 MG cells and primary human astrocytes (Figs. 11 and 12). Stimulation in the presence of 5.5 mM glucose increased IL-6 expression by both U-118 MG cells (Fig. 11A) and primary human astrocytes (Fig. 12A) to a similar extent. An enhanced increase in expression of IL-8 was observed in primary human astrocytes (Fig. 12B) (220 fold) stimulated in 5.5 mM glucose compared to U-118 MG cells (Fig. 11B) (35 fold).  Pre-treatment in the presence of high and low glucose for 24 h followed by stimulation with IFN-γ (150 U/ml) plus IL-1β (100 U/ml) for additional 24 h led to a statistically significant increase in U-118 MG (Fig. 11A) and primary human astrocyte (Fig. 12A) IL-6 expression in the presence of 30.5 mM glucose compared to 5.5 mM glucose. This increase was not observed for IL-8 	   101	  and MCP-1 gene expression in both cell types (see Figs. 11B and 12B for IL-8 data). Therefore, both cell types showed consistent results.   ELISA was used to determine the effect of glucose on secretion of IL-6 and IL-8 by U-118 MG cells, U-373 MG cells and primary human astrocytes. The three cell models showed some similar and some differing results. All stimulated cell types showed a glucose concentration-dependent increase in IL-6 secretion with the cells stimulated in 20.5 mM and 30.5 mM glucose secreting statistically higher amounts of IL-6 compared to cells stimulated in 5.5 mM (Figs. 13A, 14A and 15A). Unstimulated U-118 MG cells (Fig. 13A) and primary human astrocytes (Fig. 15A) secreted negligible amounts of IL-6; whereas unstimulated U-373 MG cells showed a glucose concentration-dependent increase in IL-6 secretion with a similar trend to stimulated cells (Fig. 14A). There was a trend towards a glucose concentration-dependent increase in IL-8 secretion by unstimulated and stimulated U-373 MG cells (Fig. 14B) and unstimulated primary human astrocytes (Fig. 15B); and a statistically significant concentration-dependent increase in IL-8 secretion by unstimulated (20.5 and 30.5 mM) and stimulated (30.5 mM) U-118 MG cells (Fig. 13B) and stimulated (30.5 mM) primary human astrocytes (Fig. 15B).    As discussed earlier, high glucose-induced activation of astrocytes and consequent increase in production of inflammatory mediators has not been widely studied. A recent study investigated effects of high glucose, which was defined as 15 mM glucose, on primary astrocytes isolated from BALB/C mice and found time-dependent upregulation in the gene expression of various pro-inflammatory cytokines including IL-6, TNF-α and IL-1β in addition to enhanced expression of GFAP indicating increased activation of astrocytes (Wang et al., 2012). Treatment of RBA-1 astrocytic cells with 	   102	  high glucose concentrations (30 mM) induced upregulation of MMP-9, an endopeptidase involved in tissue remodeling and inflammation in the CNS (Hsieh et al., 2013; Hsieh et al., 2014). Elevated levels of MMP-9 have been observed in various brain disorders including AD where several pro-inflammatory mediators can induce MMP-9 expression (Yong et al., 2001). In vitro hyperglycemia (15-35 mM) enhances production of various pro-inflammatory cytokines by human monocytes and rat microglia (Shanmugam et al., 2003; Dasu et al., 2007; Quan et al., 2007; Li et al., 2010; Li-Bo et al., 2011; Quan et al., 2011). Therefore, this thesis supports the conclusion that glial cells including astrocytes and microglia exhibit increased production and release of pro-inflammatory mediators in the presence of increased glucose levels. This may lead to increased inflammation in the brains of diabetic patients and contribute to early appearance of AD symptoms.   4.2.2 Effects of high glucose on activation of intracellular signaling pathways in astrocytic cells    To determine the intracellular signaling pathways responsible for the possible inflammatory effects of glucose, western blotting experiments were conducted to identify phosphorylation of various protein targets including JNK, STAT-3, p38 MAPK and p44/42 MAPK. Whole cell protein was extracted from U-118 MG cells that had been incubated in different glucose concentrations for 24 h followed by stimulation (or no stimulation) with a combination of IFN-γ (150 U/ml) and IL-1β (100 U/ml) for 30 min. Densitometry analysis was done on the protein bands and the phosphorylated proteins were normalized to total protein. No phosphorylation of JNK and STAT-3 was observed 	   103	  in any of the treatments; total JNK and STAT-3 were however present in all samples. This indicated that the JNK-STAT-3 pathway was not activated by glucose alone, stimulation alone or a combination of both at the 30 min timepoint. Previous studies have shown phosphorylation of STAT-3 in primary human astrocytes following 30 min stimulation with 50 U/ml IFN-γ (Hashioka et al., 2011b). Activation of JNK-STAT-3 pathway is also observed in rat brain astrocytes when treated with phorbol 12-myristate 13-acetate (PMA) (Hwang et al., 2007). Moreover, primary mice astrocytes showed ROS-dependent activation of STAT-3 and NFkB in response to 15 mM glucose treatment for 24 h (Wang et al., 2012). Differences in cell models may have contributed to this contradictory result since U-118 MG cells may have different intracellular signaling pathway activation patterns compared to primary cells. In addition, optimization of the stimulation time and glucose concentration may provide further information about the phosphorylation status of JNK and STAT-3 in U-118 MG cells.   Phosphorylation of p38 MAPK and p44/42 MAPK was observed in unstimulated and stimulated cells at all glucose concentrations. Stimulation led to increased activation of p38 MAPK compared to unstimulated cells irrespective of the glucose concentration. There was a trend towards increased phosphorylation of p44/42 MAPK in stimulated cells compared to unstimulated cells. Moreover, there was also a trend towards increased p38 MAPK phosphorylation in 30.5 mM glucose unstimulated cells and increased p44/42 MAPK phosphorylation in 30.5 mM glucose stimulated cells when compared to 5.5 mM glucose. However, these trends did not reach statistical significance, which could be attributed to lack of statistical power. Nevertheless, increased phosphorylation of the 	   104	  components of the MAPK pathway may contribute to the inflammatory effects of glucose in astrocytic cells similar to that seen in human monocytes (Dasu et al., 2007).  Induction of MMP-9 in RBA-1 cells treated with 30 mM glucose was mediated by NADPH oxidase-dependent ROS generation, activation of transcription factor AP-1 and activation of several MAPKs including ERK1/2, p38 MAPK and JNK1/2 (Hsieh et al., 2014). Astrocytic NADPH oxidase has a critical role in AD neurotoxicity by contributing to increased ROS production (Abramov & Duchen, 2005). Moreover, as discussed previously, induction of MMP-9 led to neuronal apoptosis when SK-N-SH neuroblastoma cells were treated with hyperglycemic supernatants of RBA-1 cells (Hsieh et al., 2014). Similar mechanisms, such as NADPH oxidase activation, ROS production and NFkB activation, are involved in the pro-inflammatory effects of glucose in human monocytes and microglial cells (Dasu et al., 2007; Quan et al., 2007; Quan et al., 2011). Moreover, high glucose stimulation of placental rat macrophages led to activation of TLR-dependent inflammatory pathways and switch in the inflammatory profile from anti-inflammatory to pro-inflammatory type (Sisino et al., 2013). Such a change in inflammatory profile was also observed in rat model of maternal diabetes (Sisino et al., 2013).   4.3 Direct effects of high glucose on neuronal cells  Neurons are more vulnerable to high glucose-induced toxicity compared to cells of the liver, muscle and adipose tissue (Anitha et al., 2006; Sharma et al., 2010). This is because the insulin-sensitive glucose transporter GLUT-4 regulates glucose intake in metabolic cells. On the other hand, neurons express the insulin-independent glucose 	   105	  transporters GLUT-1 and -3 (Choeiri et al., 2002). However, certain neuronal populations in the hippocampus, cerebellum and olfactory bulb express GLUT-4 in addition to GLUT-1 and -3 and hence insulin-dependent glucose transport may occur to some extent in these regions (Apelt et al., 1999; Choeiri et al., 2002). Overall, it is difficult for neurons to regulate intracellular glucose concentrations once glucose levels in the blood become high. A multitude of different pathways have been proposed to explain the mechanism of glucose-induced neurotoxicity including: 1) glucose-driven oxidative stress and downstream activation of MAPKs; 2) sorbitol-aldose reductase (polyol) pathway; 3) hexosamine pathway; 4) PKC pathway; and 5) formation of AGEs (Brownlee, 2005; Tomlinson & Gardiner, 2008; Sims-Robinson et al., 2010) (Fig. 25).   High levels of intracellular glucose increase flux through the electron transport chain complexes of the mitochondria leading to ROS production (including superoxide anion free radicals) and oxidative stress (Sharma et al., 2010). The mitochondrial superoxide dismutase enzyme converts superoxide anions to hydrogen peroxide that can be further converted to water and oxygen by glutathione peroxidases (Tomlinson & Gardiner, 2008). However, the ability of glutathione peroxidases to carry out this reaction is impaired owing to the sorbitol-aldose reductase pathway (Stavniichuk et al., 2012). High glucose levels saturate the glycolysis pathway leading to excess glucose being shunted to the sorbitol-aldose reductase pathway. The aldose reductase enzyme reduces glucose to sorbitol; which has low plasma membrane permeability and its accumulation in the cell can lead to tissue swelling and toxicity (Brownlee, 2005). Moreover, this reaction uses the cofactor NADPH; which is essential for regeneration of reduced glutathione from glutathione disulphide and consequent conversion of hydrogen peroxide 	   106	  to water by glutathione peroxidases. In the scenario of low levels of reduced glutathione, hydrogen peroxide causes production of hydroxyl radicals (Tomlinson & Gardiner, 2008). Therefore, by decreasing the levels of reduced glutathione, high glucose increases susceptibility to intracellular oxidative stress.     Figure 26: Mechanisms of hyperglycemia–induced neuronal toxicity. NADP (nicotinamide adenine dinucleotide phosphate); NADPH (reduced NADP); GSSG (glutathione disulphide); GSH (glutathione); O2.- (superoxide anion); H2O2 (hydrogen peroxide); .OH (hydroxyl radical); H2O (water); UDP-GlcNAc (uridine diphosphate-N-	   107	  acetylglucosamine); AGEs (advanced glycation end products); DHAP (dihydroxyacetone phosphate); DAG (diacylglycerol); PKC  (protein kinase C). Adapted from Brownlee, 2005 and Tomlinson & Gardiner, 2008.    Downstream activation of MAPKs including JNK and p38 MAPK caused by glucose-induced oxidative stress leads to activation of various transcription factors and consequent impairment of neuronal cytoskeleton, reduced conduction velocity and apoptosis (Sharma et al., 2010). Further down in the glycolytic pathway, increased production of fructose-6-phosphate leads to formation of uridine diphosphate-N-acetylglucosamine (UDP-GlcNAc) via the hexosamine pathway; which impairs function of intracellular proteins and transcription factors by adding itself onto serine and threonine residues (Sayeski & Kudlow, 1996; Tomlinson & Gardiner, 2008). In addition, high levels of diacylglycerol produced during increased flux through glycolysis contributes to PKC activation and consequent modulation of gene expression including increased ROS production and upregulation of TGF-β (Brownlee, 2005). In hyperglycemic conditions, glyceraldehyde-3-phosphate of the glycolytic pathway can be converted to a highly reactive intermediate known as methylglyoxal; which contributes to formation of AGEs (Brownlee, 2005; Tomlinson & Gardiner, 2008; Sims-Robinson et al., 2010). As discussed earlier, formation of heterogeneous AGEs contributes to neuronal damage by causing aggregation of Aβ and formation of NFTs; thus possibly contributing to AD progression (Sato et al., 2006). Since high glucose concentrations contribute to neurotoxicity, I investigated if high glucose would increase susceptibility of neuronal cells to an injurious stimulus. Incubation in 30.5 mM glucose alone led to a small but statistically significant decline in viability of SH-SY5Y cells as indicated by the MTT assay (Fig. 18B). Moreover, as 	   108	  expected, the viability of SH-SY5Y cells treated with 0.25 or 0.5 mM hydrogen peroxide was significantly reduced compared to untreated cells. However, the LDH release assay showed no differences between the viability of untreated cells and cells treated with 0.25 mM hydrogen peroxide (Fig. 18A). Two possibilities could explain the observed differences between the MTT and LDH release assay results; either hydrogen peroxide denatures the LDH enzyme or the cell death induced by hydrogen peroxide is mostly apoptotic and hence does not involve leakage of the intact LDH enzyme through the cell membranes. It is believed that neuronal death induced by hydrogen peroxide is both necrotic and apoptotic; with multiple factors including the concentration of hydrogen peroxide and duration of treatment influencing the outcomes (Jiang et al., 2001; Cole & Perez-Polo, 2002; Lim et al., 2002).  Incubation of SH-SY5Y neuronal cells in high glucose concentrations (10.5 – 30.5 mM) for 24 h led to increased cell death compared to cells incubated in 5.5 mM glucose when treated with 0.5 mM hydrogen peroxide as indicated by the LDH release (Fig. 18A) and MTT assays (Fig. 18B).  SH-SY5Y neuronal cells treated with 0.25 mM hydrogen peroxide showed a similar trend in the MTT assay without reaching statistical significance. Therefore, high glucose treatment increased susceptibility of SH-SY5Y neuronal cells to injury by hydrogen peroxide. Interestingly, a similar result was observed in Figure 10 where incubation of SH-SY5Y cells in 30.5 mM glucose for 24 h increased their susceptibility to injury by hyperglycemic astrocytic supernatants. On the other hand, Aβ42 (5 and 10 µM) was not found to be toxic to undifferentiated SH-SY5Y neuronal cells and hence high glucose experiments were not continued with this injurious agent. Several studies have demonstrated toxicity of Aβ40 and Aβ25-35 fragments towards SH-	   109	  SY5Y cells indicated by nick-end DNA labeling, morphological changes and marginal release of LDH enzyme (Li et al., 1996). However, Aβ42 toxicity observed in vivo is difficult to replicate in vitro using undifferentiated SH-SY5Y neuronal cells.  Differentiation of SH-SY5Y neuronal cells using retinoic acid has been used in various disease models (Christensen et al., 2011; Korecka et al., 2013). Retinoic acid-induced differentiation of SH-SY5Y neuroblastoma cells, in addition to biochemical and gene expression changes, results in cells developing long, extensively branched neurites (Lopes et al., 2010). This is evident in the phase contrast microscopy pictures of SH-SY5Y cells differentiated using retinoic acid (Fig. 19). Moreover, SH-SY5Y cells must be differentiated before they become sensitive to Aβ42 toxicity (Lambert et al., 1994). As seen in Figures 20 and 21, treatment with high glucose alone led to a small but statistically significant decline in viability of differentiated SH-SY5Y neuronal cells. A similar result was also observed in undifferentiated SH-SY5Y cells. Unexpectedly, cells incubated under the conditions of mannitol control showed a decrease in viability compared to the 5.5 mM glucose sample in both untreated cells and cells treated with 0.25 mM hydrogen peroxide (Fig. 20). Hydrogen peroxide treatment (0.5 mM) decreased viability of differentiated SH-SY5Y cells but no differences in viability were seen between the cells exposed to different glucose concentrations (Fig. 20). The unexpected results in this assay could be attributed to the technical difficulties in measuring viability of differentiated SH-SY5Y cells since, unlike undifferentiated SH-SY5Y neuronal cells, differentiated cells become very vulnerable and even slight disturbances to the culture plate or pressure from aspiration of media could lead to their detachment from the plate. This significantly impacts the measurements of cell viability. In the presence of 5.5 mM 	   110	  glucose, treatment with 5 µM Aβ42 led only to a slight decline in viability of differentiated SH-SY5Y cells; however, toxicity of Aβ42 increased in the presence of 30.5 mM glucose (Fig. 21). This further validates the hypothesis that high glucose concentrations increase susceptibility of neuronal cells to injury by disease-specific injurious agent such as Aβ42 found in AD.  Glucose-induced neuronal damage discussed earlier may contribute to the effects of glucose in increasing susceptibility of neuronal cells to injurious stimuli. However, further experiments are required to elucidate the specific mechanisms involved in this observed effect.   4.4 Insulin signaling components are present in glia-like cell lines and primary glial cells   Although INSR is encoded by a single gene, alternative splicing generates two isoforms of the INSR: INSRA and INSRB, which differ by the absence of exon 11 in INSRA. Using RT-PCR, it was found that neuronal cells and hepatocytes expressed all components of insulin receptor signaling (except IRS-2 for HepG2 cells) (Fig. 22). HepG2 hepatocytoma cells were included in the study as a positive control. However, IRS-2 expression was absent in these cells. SH-SY5Y neuroblastoma cells expressed all insulin signaling components; which is in accordance with current research stating the vital role of insulin in regulating neuronal survival, synaptic plasticity and memory formation (Banks et al., 2012; Duarte et al., 2012). Neurons in the brain express mainly the INSRA isoform of INSR (Chiu & Cline, 2010); however, the SH-SY5Y cell INSRA DNA band was faint in this study. Differences between neuroblastoma cell lines and 	   111	  primary human neurons could be the reason for this discrepancy in isoform expression. Moreover, in addition to alternative splicing, post-translational glycosylation and formation of heterodimers with homologous insulin-like growth factor (IGF) receptors leads to further modification of INSRs in different cells and tissues (Chiu & Cline, 2010). Therefore, the structure, binding affinities and function of INSR isoforms are cell and tissue specific.  INSR expression level is developmentally regulated and it is elevated/higher in neurons compared to glial cells (Unger et al., 1991). Studies using cultured neuronal and glial cells from neonatal rats established that a brain-specific INSR is expressed by neurons (Lowe et al., 1986). On the other hand, the structural and binding characteristics of INSRs expressed by glial cells of the brain as well as peripheral immune cells (e.g., monocytes and macrophages) are similar to that of peripheral metabolic tissues (Lowe et al., 1986). Therefore, two types of INSRs are expressed in the mammalian brain – peripheral-type INSR expressed by glial cells and the neuronal-type INSR expressed by neurons. The neuronal-type INSR shows some significant differences from peripheral INSRs including a lower molecular weight of the α subunit that lacks neuraminidase sensitivity (115 kDa instead of 130 kDa), a 2-5 times higher affinity for IGF-1 and IGF-2, and the absence of negative cooperativity (Heidenreich et al., 1983; Gammeltoft et al., 1985; Lowe et al., 1986). The relationship between the glial (peripheral-type)/neuronal INSRs and the splice variants (INSRA and INSRB) has not been conclusively established. However, neurons primarily express the INSRA whereas the peripheral metabolic tissues primarily express INSRB (Belfiore et al., 2009).  U-118 MG cells (astrocyte model) (Fig. 23A) and primary human astrocytes (Fig. 	   112	  23B) were found to express all INSR signaling components suggesting that astrocytes contribute to the effects of insulin in the brain. Most importantly, primary human microglia also express all INSR signaling components (Fig. 23B). The results also indicated that primary human astrocytes might have higher expression of INSRB whereas primary human microglia may express more of INSRA. As discussed earlier, INSRB is primarily involved in insulin signaling in peripheral metabolic tissues and may also be the dominant INSR expressed by glial cells of the brain. On the other hand, INSRA is a high affinity receptor for IGF-2 mainly expressed by neurons.  Heni et al. (2011) found that fetal human astrocytes express all components of insulin signaling including the INSR, IRS-1 and IRS-2. They also established that INSRA comprised 2/3 of the INSR expression by human astrocytes. Therefore, it is possible that different glial cells of the brain express both INSRA and INSRB isoforms to varying extents. Moreover, these findings indicate that the structural and functional aspects of INSRs in the brain are more complex than just distinguishing between INSRA and INSRB expression by various cells in vitro and may involve formation of dimers with IGF receptors as well as specific post-translational modifications.   4.5 Insulin modulates secretion of pro-inflammatory cytokines by primary human astrocytes   There is limited research investigating the effects of insulin on glial cells of the brain. Insulin therapies have gained much attention in the field of AD treatment options owing to the effects of insulin on neuronal survival and synaptic plasticity (Reger et al., 	   113	  2008; Craft et al., 2012). Insulin is no longer considered just a peripheral hormone and research establishing its critical role in the CNS is accumulating rapidly (de la Monte, 2012; Duarte et al., 2012). Glial cells that are known to exacerbate neuroinflammation in AD express the INSR and other important components of the insulin-signaling cascade. Moreover, the effects of insulin on the inflammatory status of glial cells are unknown. Therefore, I investigated whether insulin would modulate secretion of pro-inflammatory cytokines by primary human astrocytes stimulated with a combination of IFN-γ (150 U/ml) and IL-1β (100 U/ml).  Treatment of primary human astrocytes with increasing concentrations of insulin led to a bell-shaped response curve of both IL-6 (Fig. 24A) and IL-8 (Fig. 25A) secretion. Insulin increased IL-6 (maximum at 1 nM) and IL-8 (maximum at 1 and 10 nM) secretion by stimulated primary human astrocytes. Heat-denatured insulin did not affect IL-6 (Fig. 24B) and IL-8 (Fig. 25B) secretion. Such bell-shaped response curves have been observed previously for mitogenic effects of IGF-1 and insulin on fibroblasts (Lamothe et al., 1998) and insulin-mediated regulation of muscarinic receptor function in neurons (Coulson et al., 2004). High insulin concentrations saturate the insulin binding sites on the INSR and stabilize the binding of pre-bound insulin; which explains the bell-shaped response curves of insulin (De Meyts & Whittaker, 2002). Therefore, a threshold for insulin concentrations exists above which insulin may act as a contributor to inflammation by increasing secretion of pro-inflammatory cytokines by astrocytes. Such pro-inflammatory effects of high insulin concentrations have been observed in peripheral monocytes and macrophages and are known to be mediated by the MAPK pathway (Fischoeder et al., 2007; Iwasaki et al., 2009). The signaling pathways employed by 	   114	  insulin in mediating inflammation in astrocytes still remain to be elucidated. Moreover, it would be interesting to see if insulin affects microglial cells in a similar manner.    Chapter 5. Conclusions and Future Work   5.1 Limitations of Research    Several limitations of this thesis research exist including but not limited to the use of cell lines and the cell culture models of neuroinflammation. The cell lines used to model astrocytes and neurons are derived from tumors of respective cell types and hence may exhibit properties and gene expression patterns that differ from their primary counterparts. To overcome this limitation, key experiments were repeated with primary human astrocytes and differentiated neuronal cells. In addition, the cell culture model of neuroinflammation involves stimulation of astrocytic cells with different cytokines that may not be involved in causing activation of astrocytes in vivo. However, the signaling pathways and neurotoxic patterns of astrocytic cells stimulated with cytokines in vitro are similar to those exhibited by reactive astrocytes in vivo. Therefore, although the stimuli used in vitro may not fully reproduce the brain microenvironment, the objective of inducing astrocytic activation to make them toxic to neurons is still achieved.   Another limitation of this research is the use of the supernatant transfer experiments to identify effects of glucose on astrocyte toxicity towards neuronal cells. As is evident from the data, this model is more useful when employed for drug screening 	   115	  where the effects on cell viability are more pronounced. The subtle differences in cell viability between 5.5 and 30.5 mM glucose concentrations are difficult to measure using the cell viability assays. Moreover, the clinical significance of such differences cannot be established in an in vitro model. Therefore, on one hand, in vitro cell culture studies allow us to tease out the effects of a single variable, but on the other hand, the role of other factors are oversimplified and not measured.   The glucose concentrations used in the thesis experiments are not achieved in the human brain even under conditions of uncompensated diabetes. Brain glucose concentrations are only 15-20% of the blood glucose concentrations (Gruetter et al., 1998). Therefore, the lowest glucose concentration (5.5 mM) used in these experiments is in fact the highest glucose concentration that could be achieved in the brain. The model cells however would not survive in culture with concentrations below 5.5 mM glucose. Moreover, the adverse effects of high glucose observed in culture still apply theoretically to the cells in the brain.   5.2 Future Work    This thesis work supports the hypothesis that high glucose concentrations enhance expression and secretion of pro-inflammatory cytokines by astrocytic cells. However, the upstream mechanisms involved in this effect are still unclear. It will be important to repeat the western blotting experiments in primary human astrocytes with different incubation times in glucose to investigate the signaling pathways involved in the pro-inflammatory effects of glucose. Similarly, the mechanism of increased susceptibility of 	   116	  neuronal cells to injury in the presence of high glucose and the effects of insulin on the inflammatory status of glial cells remain to be elucidated. It is possible the pathways that exacerbate inflammation in the presence of high glucose and high insulin concentrations are similar. In terms of the toxicity of astrocytic supernatants towards neuronal cells in the presence of different glucose concentrations, it would be useful to perform co-culture experiments in which both cell types are incubated in the same glucose concentration; a scenario more realistic in vivo. Although my in vitro experiments indicate that exposure to high glucose can exacerbate inflammation caused by glial cells, only pre-clinical studies in animal models of diabetes and AD can help establish if this increase in inflammation leads to early appearance of AD-like symptoms and pathology.   5.3 Significance of findings  This thesis demonstrates the exacerbation of astrocyte-mediated inflammation under high glucose conditions seen in metabolic disorders such as T2DM. High glucose (30.5 mM) increased the gene expression of IL-6 and secretion of IL-6 and IL-8 by several human astrocyte models. Activation of p38 MAPK could be involved in mediating the inflammatory effects of high glucose in astrocytes. Moreover, incubation in high glucose concentrations (10.5 – 30.5 mM) enhanced susceptibility of neuronal cells to injury by astrocytic supernatants, hydrogen peroxide and Aβ42. Therefore, hyperglycemia in T2DM may contribute to the increased risk for AD by exacerbating glial-mediated inflammation and enhancing neuronal injury caused by disease-specific agents.  	   117	  Peripheral hyperinsulinemia in T2DM is a consequence of peripheral insulin resistance and may lead to periods of high insulin concentrations in the brain (Neumann et al., 2008). In later stages of T2DM, pancreatic insulin production and secretion is reduced leading to low levels of insulin in the blood. AD is being characterized as Type 3 diabetes owing to the recent research that indicates reduced levels of insulin in the brain (de la Monte & Wands, 2008; Duarte et al., 2012). Moreover, it has been suggested that deficiency of the INSR and impairment of the insulin-signaling cascade in the brain are the underlying factors for appearance of AD pathology (Neumann et al., 2008; Liu et al., 2011). As illustrated in this thesis, INSR and its signaling components are expressed in primary glial cells and insulin modulates secretion of pro-inflammatory cytokines by primary human astrocytes.  This study, as well as the epidemiological studies indicating an increased risk of AD in diabetic patients, warrant further research investigating the role of chronic inflammation as the link between metabolic and neurodegenerative disorders. A recent study demonstrated that metabolic inflammation and activation of glial cells in response to acute MPTP (1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine) treatment in diabetic mice lead to increased loss of dopaminergic neurons, thus contributing to PD (Wang et al., 2014). Therefore, metabolic inflammation in T2DM is being recognized for its role in increasing risk for neurodegenerative disorders and the mechanisms involved are being investigated. The global prevalence of T2DM has already increased to epidemic levels. An added burden of neurodegenerative disorders in the diabetic population would have a huge impact on health care systems worldwide. Results of this research may help highlight novel factors that contribute to the progression of AD and help guide future 	   118	  research on preventative or treatment interventions.      References   Abramov, A.Y. & Duchen, M.R. (2005) The role of an astrocytic NADPH oxidase in the neurotoxicity of amyloid beta peptides. Philos Trans R Soc Lond B Biol Sci, 360, 2309-2314.  Acheampong, E.A., Roschel, C., Mukhtar, M., Srinivasan, A., Rafi, M., Pomerantz, R.J. & Parveen, Z. (2009) Combined effects of hyperglycemic conditions and HIV-1 Nef: a potential model for induced HIV neuropathogenesis. Virol J, 6, 183.  Ahmado, A., Carr, A.J., Vugler, A.A., Semo, M., Gias, C., Lawrence, J.M., Chen, L.L., Chen, F.K., Turowski, P., da Cruz, L. & Coffey, P.J. (2011) Induction of differentiation by pyruvate and DMEM in the human retinal pigment epithelium cell line ARPE-19. Invest Ophthalmol Vis Sci, 52, 7148-7159.  Ahtiluoto, S., Polvikoski, T., Peltonen, M., Solomon, A., Tuomilehto, J., Winblad, B., Sulkava, R. & Kivipelto, M. (2010) Diabetes, Alzheimer disease, and vascular dementia: a population-based neuropathologic study. Neurology, 75, 1195-1202.  Aljada, A. & Dandona, P. (2000) Effect of insulin on human aortic endothelial nitric oxide synthase. Metabolism, 49, 147-150.  Aljada, A., Ghanim, H., Mohanty, P., Kapur, N. & Dandona, P. (2002) Insulin inhibits the pro-inflammatory transcription factor early growth response gene-1 (Egr)-1 expression in mononuclear cells (MNC) and reduces plasma tissue factor (TF) and plasminogen activator inhibitor-1 (PAI-1) concentrations. J Clin Endocrinol Metab, 87, 1419-1422.  Aljada, A., Ghanim, H., Saadeh, R. & Dandona, P. (2001) Insulin inhibits NFkappaB and MCP-1 expression in human aortic endothelial cells. J Clin Endocrinol Metab, 86, 450-453.  Aljada, A., Saadeh, R., Assian, E., Ghanim, H. & Dandona, P. (2000) Insulin inhibits the expression of intercellular adhesion molecule-1 by human aortic endothelial cells through stimulation of nitric oxide. J Clin Endocrinol Metab, 85, 2572-2575. 	   119	   Anitha, M., Gondha, C., Sutliff, R., Parsadanian, A., Mwangi, S., Sitaraman, S.V. & Srinivasan, S. (2006) GDNF rescues hyperglycemia-induced diabetic enteric neuropathy through activation of the PI3K/Akt pathway. J Clin Invest, 116, 344-356.  Apelt, J., Mehlhorn, G. & Schliebs, R. (1999) Insulin-sensitive GLUT4 glucose transporters are colocalized with GLUT3-expressing cells and demonstrate a chemically distinct neuron-specific localization in rat brain. J Neurosci Res, 57, 693-705.  Araque, A., Parpura, V., Sanzgiri, R.P. & Haydon, P.G. (1999) Tripartite synapses: glia, the unacknowledged partner. Trends Neurosci, 22, 208-215.  Bailyes, E.M., Nave, B.T., Soos, M.A., Orr, S.R., Hayward, A.C. & Siddle, K. (1997) Insulin receptor/IGF-I receptor hybrids are widely distributed in mammalian tissues: quantification of individual receptor species by selective immunoprecipitation and immunoblotting. Biochem J, 327 ( Pt 1), 209-215.  Banks, W.A. (2004) The source of cerebral insulin. Eur J Pharmacol, 490, 5-12.  Banks, W.A., Owen, J.B. & Erickson, M.A. (2012) Insulin in the brain: There and back again. Pharmacol Ther.  Barnes, D.E. & Yaffe, K. (2011) The projected effect of risk factor reduction on Alzheimer's disease prevalence. Lancet Neurol, 10, 819-828.  Barzilay, J.I., Abraham, L., Heckbert, S.R., Cushman, M., Kuller, L.H., Resnick, H.E. & Tracy, R.P. (2001) The relation of markers of inflammation to the development of glucose disorders in the elderly: the Cardiovascular Health Study. Diabetes, 50, 2384-2389.  Belfiore, A., Frasca, F., Pandini, G., Sciacca, L. & Vigneri, R. (2009) Insulin receptor isoforms and insulin receptor/insulin-like growth factor receptor hybrids in physiology and disease. Endocr Rev, 30, 586-623.  Benecke, H., Flier, J.S. & Moller, D.E. (1992) Alternatively spliced variants of the insulin receptor protein. Expression in normal and diabetic human tissues. J Clin Invest, 89, 2066-2070.  Biessels, G.J., Staekenborg, S., Brunner, E., Brayne, C. & Scheltens, P. (2006) Risk of dementia in diabetes mellitus: a systematic review. Lancet Neurol, 5, 64-74.  Bird, T.D. (2008) Genetic aspects of Alzheimer disease. Genet Med, 10, 231-239.  	   120	  Block, M.L., Zecca, L. & Hong, J.S. (2007) Microglia-mediated neurotoxicity: uncovering the molecular mechanisms. Nat Rev Neurosci, 8, 57-69.  Bonotis, K., Krikki, E., Holeva, V., Aggouridaki, C., Costa, V. & Baloyannis, S. (2008) Systemic immune aberrations in Alzheimer's disease patients. J Neuroimmunol, 193, 183-187.  Brix-Christensen, V., Andersen, S.K., Andersen, R., Mengel, A., Dyhr, T., Andersen, N.T., Larsson, A., Schmitz, O., Orskov, H. & Tonnesen, E. (2004) Acute hyperinsulinemia restrains endotoxin-induced systemic inflammatory response: an experimental study in a porcine model. Anesthesiology, 100, 861-870.  Brownlee, M. (2005) The pathobiology of diabetic complications: a unifying mechanism. Diabetes, 54, 1615-1625.  Brundage, S.I., Kirilcuk, N.N., Lam, J.C., Spain, D.A. & Zautke, N.A. (2008) Insulin increases the release of proinflammatory mediators. J Trauma, 65, 367-372.  Bunn, R.C., Cockrell, G.E., Ou, Y., Thrailkill, K.M., Lumpkin, C.K., Jr. & Fowlkes, J.L. (2010) Palmitate and insulin synergistically induce IL-6 expression in human monocytes. Cardiovasc Diabetol, 9, 73.  Bustin, S.A., Benes, V., Garson, J.A., Hellemans, J., Huggett, J., Kubista, M., Mueller, R., Nolan, T., Pfaffl, M.W., Shipley, G.L., Vandesompele, J. & Wittwer, C.T. (2009) The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem, 55, 611-622.  Caudle, R.M. (2006) Memory in astrocytes: a hypothesis. Theor Biol Med Model, 3, 2.  Chang, A., Orth, A., Le, B., Menchavez, P. & Miller, L. (2009) Performance analysis of the OneTouch UltraVue blood glucose monitoring system. J Diabetes Sci Technol, 3, 1158-1165.  Chaudhry, M.A., Vitalis, T.Z., Bowen, B.D. & Piret, J.M. (2008) Basal medium composition and serum or serum replacement concentration influences on the maintenance of murine embryonic stem cells. Cytotechnology, 58, 173-179.  Chaudhuri, A., Janicke, D., Wilson, M.F., Tripathy, D., Garg, R., Bandyopadhyay, A., Calieri, J., Hoffmeyer, D., Syed, T., Ghanim, H., Aljada, A. & Dandona, P. (2004) Anti-inflammatory and profibrinolytic effect of insulin in acute ST-segment-elevation myocardial infarction. Circulation, 109, 849-854.  Chen, J., Cui, X., Zacharek, A., Cui, Y., Roberts, C. & Chopp, M. (2011) White matter damage and the effect of matrix metalloproteinases in type 2 diabetic mice after stroke. Stroke, 42, 445-452.  	   121	  Chiu, S.L. & Cline, H.T. (2010) Insulin receptor signaling in the development of neuronal structure and function. Neural Dev, 5, 7.  Choeiri, C., Staines, W. & Messier, C. (2002) Immunohistochemical localization and quantification of glucose transporters in the mouse brain. Neuroscience, 111, 19-34.  Christensen, J., Steain, M., Slobedman, B. & Abendroth, A. (2011) Differentiated neuroblastoma cells provide a highly efficient model for studies of productive varicella-zoster virus infection of neuronal cells. J Virol, 85, 8436-8442.  Cole, K.K. & Perez-Polo, J.R. (2002) Poly(ADP-ribose) polymerase inhibition prevents both apoptotic-like delayed neuronal death and necrosis after H(2)O(2) injury. J Neurochem, 82, 19-29.  Coulson, F.R., Jacoby, D.B. & Fryer, A.D. (2004) Insulin regulates neuronal M2 muscarinic receptor function in the ileum of diabetic rats. J Pharmacol Exp Ther, 308, 760-766.  Craft, S., Baker, L.D., Montine, T.J., Minoshima, S., Watson, G.S., Claxton, A., Arbuckle, M., Callaghan, M., Tsai, E., Plymate, S.R., Green, P.S., Leverenz, J., Cross, D. & Gerton, B. (2012) Intranasal insulin therapy for Alzheimer disease and amnestic mild cognitive impairment: a pilot clinical trial. Arch Neurol, 69, 29-38.  Craft, S., Peskind, E., Schwartz, M.W., Schellenberg, G.D., Raskind, M. & Porte, D., Jr. (1998) Cerebrospinal fluid and plasma insulin levels in Alzheimer's disease: relationship to severity of dementia and apolipoprotein E genotype. Neurology, 50, 164-168.  Cummings, J.L. (2004) Alzheimer's disease. N Engl J Med, 351, 56-67.  Cuschieri, J., Bulger, E., Grinsell, R., Garcia, I. & Maier, R.V. (2008) Insulin regulates macrophage activation through activin A. Shock, 29, 285-290.  Dandona, P., Aljada, A., Mohanty, P., Ghanim, H., Hamouda, W., Assian, E. & Ahmad, S. (2001) Insulin inhibits intranuclear nuclear factor kappaB and stimulates IkappaB in mononuclear cells in obese subjects: evidence for an anti-inflammatory effect? J Clin Endocrinol Metab, 86, 3257-3265.  Dandona, P., Chaudhuri, A., Ghanim, H. & Mohanty, P. (2007a) Proinflammatory effects of glucose and anti-inflammatory effect of insulin: relevance to cardiovascular disease. Am J Cardiol, 99, 15B-26B.  Dandona, P., Chaudhuri, A., Mohanty, P. & Ghanim, H. (2007b) Anti-inflammatory effects of insulin. Curr Opin Clin Nutr Metab Care, 10, 511-517. 	   122	   Dasu, M.R., Devaraj, S. & Jialal, I. (2007) High glucose induces IL-1beta expression in human monocytes: mechanistic insights. Am J Physiol Endocrinol Metab, 293, E337-346.  Dasu, M.R., Devaraj, S., Zhao, L., Hwang, D.H. & Jialal, I. (2008) High glucose induces toll-like receptor expression in human monocytes: mechanism of activation. Diabetes, 57, 3090-3098.  Davis, R.J. (2000) Signal transduction by the JNK group of MAP kinases. Cell, 103, 239-252.  De Felice, F.G., Wu, D., Lambert, M.P., Fernandez, S.J., Velasco, P.T., Lacor, P.N., Bigio, E.H., Jerecic, J., Acton, P.J., Shughrue, P.J., Chen-Dodson, E., Kinney, G.G. & Klein, W.L. (2008) Alzheimer's disease-type neuronal tau hyperphosphorylation induced by A beta oligomers. Neurobiol Aging, 29, 1334-1347.  de la Monte, S.M. (2012) Brain insulin resistance and deficiency as therapeutic targets in Alzheimer's disease. Curr Alzheimer Res, 9, 35-66.  de la Monte, S.M. & Wands, J.R. (2008) Alzheimer's disease is type 3 diabetes-evidence reviewed. J Diabetes Sci Technol, 2, 1101-1113.  De Meyts, P. & Whittaker, J. (2002) Structural biology of insulin and IGF1 receptors: implications for drug design. Nat Rev Drug Discov, 1, 769-783.  DeFronzo, R.A. (2010) Insulin resistance, lipotoxicity, type 2 diabetes and atherosclerosis: the missing links. The Claude Bernard Lecture 2009. Diabetologia, 53, 1270-1287.  Duarte, A.I., Moreira, P.I. & Oliveira, C.R. (2012) Insulin in central nervous system: more than just a peripheral hormone. J Aging Res, 2012, 384017.  Ebstein, W. (1876) Zur Therapie des Diabetes mellitus, ins-besondere über. die Anwendung des salicylsauren Natron beidemselben Berliner Klin. Wochenschr 13, 337-340.  Edison, P., Archer, H.A., Gerhard, A., Hinz, R., Pavese, N., Turkheimer, F.E., Hammers, A., Tai, Y.F., Fox, N., Kennedy, A., Rossor, M. & Brooks, D.J. (2008) Microglia, amyloid, and cognition in Alzheimer's disease: An [11C](R)PK11195-PET and [11C]PIB-PET study. Neurobiol Dis, 32, 412-419.  Eikelenboom, P., Hack, C.E., Rozemuller, J.M. & Stam, F.C. (1989) Complement activation in amyloid plaques in Alzheimer's dementia. Virchows Arch B Cell Pathol Incl Mol Pathol, 56, 259-262. 	   123	   Fischoeder, A., Meyborg, H., Stibenz, D., Fleck, E., Graf, K. & Stawowy, P. (2007) Insulin augments matrix metalloproteinase-9 expression in monocytes. Cardiovasc Res, 73, 841-848.  Fishel, M.A., Watson, G.S., Montine, T.J., Wang, Q., Green, P.S., Kulstad, J.J., Cook, D.G., Peskind, E.R., Baker, L.D., Goldgaber, D., Nie, W., Asthana, S., Plymate, S.R., Schwartz, M.W. & Craft, S. (2005) Hyperinsulinemia provokes synchronous increases in central inflammation and beta-amyloid in normal adults. Arch Neurol, 62, 1539-1544.  Fonseca, M.I., Zhou, J., Botto, M. & Tenner, A.J. (2004) Absence of C1q leads to less neuropathology in transgenic mouse models of Alzheimer's disease. J Neurosci, 24, 6457-6465.  Fornoni, A., Ijaz, A., Tejada, T. & Lenz, O. (2008) Role of inflammation in diabetic nephropathy. Curr Diabetes Rev, 4, 10-17.  Frank-Cannon, T.C., Alto, L.T., McAlpine, F.E. & Tansey, M.G. (2009) Does neuroinflammation fan the flame in neurodegenerative diseases? Mol Neurodegener, 4, 47.  Freude, S., Plum, L., Schnitker, J., Leeser, U., Udelhoven, M., Krone, W., Bruning, J.C. & Schubert, M. (2005) Peripheral hyperinsulinemia promotes tau phosphorylation in vivo. Diabetes, 54, 3343-3348.  Fuller, S., Munch, G. & Steele, M. (2009) Activated astrocytes: a therapeutic target in Alzheimer's disease? Expert Rev Neurother, 9, 1585-1594.  Gammeltoft, S., Fehlmann, M. & Van Obberghen, E. (1985) Insulin receptors in the mammalian central nervous system: binding characteristics and subunit structure. Biochimie, 67, 1147-1153.  Garg, R., Tripathy, D. & Dandona, P. (2003) Insulin resistance as a proinflammatory state: mechanisms, mediators, and therapeutic interventions. Curr Drug Targets, 4, 487-492.  Garwood, C.J., Pooler, A.M., Atherton, J., Hanger, D.P. & Noble, W. (2011) Astrocytes are important mediators of Abeta-induced neurotoxicity and tau phosphorylation in primary culture. Cell Death Dis, 2, e167.  Gasparini, L. & Xu, H. (2003) Potential roles of insulin and IGF-1 in Alzheimer’s disease. Trends Neurosci, 26, 404-406.  Genuth, S.M. (1973) Plasma insulin and glucose profiles in normal, obese, and diabetic persons. Ann Intern Med, 79, 812-822. 	   124	   Granic, I., Dolga, A.M., Nijholt, I.M., van Dijk, G. & Eisel, U.L. (2009) Inflammation and NF-kappaB in Alzheimer's disease and diabetes. J Alzheimers Dis, 16, 809-821.  Green, R.C., Schneider, L.S., Amato, D.A., Beelen, A.P., Wilcock, G., Swabb, E.A. & Zavitz, K.H. (2009) Effect of tarenflurbil on cognitive decline and activities of daily living in patients with mild Alzheimer disease: a randomized controlled trial. JAMA, 302, 2557-2564.  Gruetter, R., Ugurbil, K. & Seaquist, E.R. (1998) Steady-state cerebral glucose concentrations and transport in the human brain. J Neurochem, 70, 397-408.  Guerreiro, R., Wojtas, A., Bras, J., Carrasquillo, M., Rogaeva, E., Majounie, E., Cruchaga, C., Sassi, C., Kauwe, J.S., Younkin, S., Hazrati, L., Collinge, J., Pocock, J., Lashley, T., Williams, J., Lambert, J.C., Amouyel, P., Goate, A., Rademakers, R., Morgan, K., Powell, J., St George-Hyslop, P., Singleton, A. & Hardy, J. (2013) TREM2 variants in Alzheimer's disease. N Engl J Med, 368, 117-127.  Hallschmid, M. & Schultes, B. (2009) Central nervous insulin resistance: a promising target in the treatment of metabolic and cognitive disorders? Diabetologia, 52, 2264-2269.  Hanisch, U.K. & Kettenmann, H. (2007) Microglia: active sensor and versatile effector cells in the normal and pathologic brain. Nat Neurosci, 10, 1387-1394.  Hashioka, S., Klegeris, A. & McGeer, P.L. (2011a) Proton pump inhibitors reduce interferon-gamma-induced neurotoxicity and STAT3 phosphorylation of human astrocytes. Glia, 59, 833-840.  Hashioka, S., Klegeris, A. & McGeer, P.L. (2012) The histone deacetylase inhibitor suberoylanilide hydroxamic acid attenuates human astrocyte neurotoxicity induced by interferon-gamma. J Neuroinflammation, 9, 113.  Hashioka, S., Klegeris, A., Qing, H. & McGeer, P.L. (2011b) STAT3 inhibitors attenuate interferon-gamma-induced neurotoxicity and inflammatory molecule production by human astrocytes. Neurobiol Dis, 41, 299-307.  Hashioka, S., Klegeris, A., Schwab, C. & McGeer, P.L. (2009) Interferon-gamma-dependent cytotoxic activation of human astrocytes and astrocytoma cells. Neurobiol Aging, 30, 1924-1935.  Heidenreich, K.A., Zahniser, N.R., Berhanu, P., Brandenburg, D. & Olefsky, J.M. (1983) Structural differences between insulin receptors in the brain and peripheral target tissues. J Biol Chem, 258, 8527-8530. 	   125	   Heni, M., Hennige, A.M., Peter, A., Siegel-Axel, D., Ordelheide, A.M., Krebs, N., Machicao, F., Fritsche, A., Haring, H.U. & Staiger, H. (2011) Insulin promotes glycogen storage and cell proliferation in primary human astrocytes. PLoS One, 6, e21594.  Hirosumi, J., Tuncman, G., Chang, L., Gorgun, C.Z., Uysal, K.T., Maeda, K., Karin, M. & Hotamisligil, G.S. (2002) A central role for JNK in obesity and insulin resistance. Nature, 420, 333-336.  Holmes, C. (2013) Review: Systemic inflammation and Alzheimer's disease. Neuropathol Appl Neurobiol, 39, 51-68.  Honjo, K., Black, S.E. & Verhoeff, N.P. (2012) Alzheimer's disease, cerebrovascular disease, and the beta-amyloid cascade. Can J Neurol Sci, 39, 712-728.  Hsieh, H.L., Chi, P.L., Lin, C.C., Yang, C.C. & Yang, C.M. (2014) Up-regulation of ROS-Dependent Matrix Metalloproteinase-9 from High-Glucose-Challenged Astrocytes Contributes to the Neuronal Apoptosis. Mol Neurobiol.  Hsieh, H.L., Lin, C.C., Hsiao, L.D. & Yang, C.M. (2013) High glucose induces reactive oxygen species-dependent matrix metalloproteinase-9 expression and cell migration in brain astrocytes. Mol Neurobiol, 48, 601-614.  Hundal, R.S., Petersen, K.F., Mayerson, A.B., Randhawa, P.S., Inzucchi, S., Shoelson, S.E. & Shulman, G.I. (2002) Mechanism by which high-dose aspirin improves glucose metabolism in type 2 diabetes. J Clin Invest, 109, 1321-1326.  Hwang, M.N., Kim, K.S., Choi, Y.W., Jou, I. & Yoon, S. (2007) PMA activates Stat3 in the Jak/Stat pathway and induces SOCS5 in rat brain astrocytes. Mol Cells, 23, 94-99.  Iida, K.T., Shimano, H., Kawakami, Y., Sone, H., Toyoshima, H., Suzuki, S., Asano, T., Okuda, Y. & Yamada, N. (2001) Insulin up-regulates tumor necrosis factor-alpha production in macrophages through an extracellular-regulated kinase-dependent pathway. J Biol Chem, 276, 32531-32537.  Iida, K.T., Suzuki, H., Sone, H., Shimano, H., Toyoshima, H., Yatoh, S., Asano, T., Okuda, Y. & Yamada, N. (2002) Insulin inhibits apoptosis of macrophage cell line, THP-1 cells, via phosphatidylinositol-3-kinase-dependent pathway. Arterioscler Thromb Vasc Biol, 22, 380-386.  Iqbal, K. & Grundke-Iqbal, I. (2008) Alzheimer neurofibrillary degeneration: significance, etiopathogenesis, therapeutics and prevention. J Cell Mol Med, 12, 38-55.  	   126	  Iwasaki, Y., Nishiyama, M., Taguchi, T., Asai, M., Yoshida, M., Kambayashi, M., Terada, Y. & Hashimoto, K. (2009) Insulin exhibits short-term anti-inflammatory but long-term proinflammatory effects in vitro. Mol Cell Endocrinol, 298, 25-32.  Jeschke, M.G., Klein, D., Bolder, U. & Einspanier, R. (2004) Insulin attenuates the systemic inflammatory response in endotoxemic rats. Endocrinology, 145, 4084-4093.  Jiang, D., Jha, N., Boonplueang, R. & Andersen, J.K. (2001) Caspase 3 inhibition attenuates hydrogen peroxide-induced DNA fragmentation but not cell death in neuronal PC12 cells. J Neurochem, 76, 1745-1755.  Jolivalt, C.G., Lee, C.A., Beiswenger, K.K., Smith, J.L., Orlov, M., Torrance, M.A. & Masliah, E. (2008) Defective insulin signaling pathway and increased glycogen synthase kinase-3 activity in the brain of diabetic mice: parallels with Alzheimer's disease and correction by insulin. J Neurosci Res, 86, 3265-3274.  Jonsson, T., Stefansson, H., Steinberg, S., Jonsdottir, I., Jonsson, P.V., Snaedal, J., Bjornsson, S., Huttenlocher, J., Levey, A.I., Lah, J.J., Rujescu, D., Hampel, H., Giegling, I., Andreassen, O.A., Engedal, K., Ulstein, I., Djurovic, S., Ibrahim-Verbaas, C., Hofman, A., Ikram, M.A., van Duijn, C.M., Thorsteinsdottir, U., Kong, A. & Stefansson, K. (2013) Variant of TREM2 associated with the risk of Alzheimer's disease. N Engl J Med, 368, 107-116.  Kappert, K., Meyborg, H., Clemenz, M., Graf, K., Fleck, E., Kintscher, U. & Stawowy, P. (2008) Insulin facilitates monocyte migration: a possible link to tissue inflammation in insulin-resistance. Biochem Biophys Res Commun, 365, 503-508.  Kern, W., Benedict, C., Schultes, B., Plohr, F., Moser, A., Born, J., Fehm, H.L. & Hallschmid, M. (2006) Low cerebrospinal fluid insulin levels in obese humans. Diabetologia, 49, 2790-2792.  Kettenmann, H. & Verkhratsky, A. (2011) [Neuroglia--living nerve glue]. Fortschr Neurol Psychiatr, 79, 588-597.  Kim, J.A., Montagnani, M., Koh, K.K. & Quon, M.J. (2006) Reciprocal relationships between insulin resistance and endothelial dysfunction: molecular and pathophysiological mechanisms. Circulation, 113, 1888-1904.  Kim, J.K., Kim, Y.J., Fillmore, J.J., Chen, Y., Moore, I., Lee, J., Yuan, M., Li, Z.W., Karin, M., Perret, P., Shoelson, S.E. & Shulman, G.I. (2001) Prevention of fat-induced insulin resistance by salicylate. J Clin Invest, 108, 437-446.  King, G.L. (2008) The role of inflammatory cytokines in diabetes and its complications. J Periodontol, 79, 1527-1534.  	   127	  Klegeris, A., Walker, D.G. & McGeer, P.L. (1999) Toxicity of human THP-1 monocytic cells towards neuron-like cells is reduced by non-steroidal anti-inflammatory drugs (NSAIDs). Neuropharmacology, 38, 1017-1025.  Korecka, J.A., van Kesteren, R.E., Blaas, E., Spitzer, S.O., Kamstra, J.H., Smit, A.B., Swaab, D.F., Verhaagen, J. & Bossers, K. (2013) Phenotypic characterization of retinoic acid differentiated SH-SY5Y cells by transcriptional profiling. PLoS One, 8, e63862.  Krause, D.L. & Muller, N. (2010) Neuroinflammation, microglia and implications for anti-inflammatory treatment in Alzheimer's disease. Int J Alzheimers Dis, 2010.  Kurosinski, P. & Gotz, J. (2002) Glial cells under physiologic and pathologic conditions. Arch Neurol, 59, 1524-1528.  LaFeria, F.M., Green, K.M. & Oddo, S. (2007) Intracellular amyloid-b in Alzheimer's disease. Nat Rev Neurosci 8, 499-509.  Lambert, J.C., Heath, S., Even, G., Campion, D., Sleegers, K., Hiltunen, M., Combarros, O., Zelenika, D., Bullido, M.J., Tavernier, B., Letenneur, L., Bettens, K., Berr, C., Pasquier, F., Fievet, N., Barberger-Gateau, P., Engelborghs, S., De Deyn, P., Mateo, I., Franck, A., Helisalmi, S., Porcellini, E., Hanon, O., de Pancorbo, M.M., Lendon, C., Dufouil, C., Jaillard, C., Leveillard, T., Alvarez, V., Bosco, P., Mancuso, M., Panza, F., Nacmias, B., Bossu, P., Piccardi, P., Annoni, G., Seripa, D., Galimberti, D., Hannequin, D., Licastro, F., Soininen, H., Ritchie, K., Blanche, H., Dartigues, J.F., Tzourio, C., Gut, I., Van Broeckhoven, C., Alperovitch, A., Lathrop, M. & Amouyel, P. (2009) Genome-wide association study identifies variants at CLU and CR1 associated with Alzheimer's disease. Nat Genet, 41, 1094-1099.  Lambert, M.P., Stevens, G., Sabo, S., Barber, K., Wang, G., Wade, W., Krafft, G., Snyder, S., Holzman, T.F. & Klein, W.L. (1994) Beta/A4-evoked degeneration of differentiated SH-SY5Y human neuroblastoma cells. J Neurosci Res, 39, 377-385.  Lamothe, B., Baudry, A., Christoffersen, C.T., De Meyts, P., Jami, J., Bucchini, D. & Joshi, R.L. (1998) Insulin receptor-deficient cells as a new tool for dissecting complex interplay in insulin and insulin-like growth factors. FEBS Lett, 426, 381-385.  Laron, Z. (2009) Insulin and the brain. Arch Physiol Biochem, 115, 112-116.  Launer, L.J., Miller, M.E., Williamson, J.D., Lazar, R.M., Gerstein, H.C., Murray, A.M., Sullivan, M., Horowitz, K.R., Ding, J., Marcovina, S., Lovato, L.C., Lovato, J., Margolis, K.L., O'Connor, P., Lipkin, E.W., Hirsch, J., Coker, L., Maldjian, J., Sunshine, J.L., Truwit, C., Davatzikos, C. & Bryan, R.N. (2011) Effects of intensive glucose lowering on brain structure and function in people with type 2 	   128	  diabetes (ACCORD MIND): a randomised open-label substudy. Lancet Neurol, 10, 969-977.  Lawson, L.J., Perry, V.H., Dri, P. & Gordon, S. (1990) Heterogeneity in the distribution and morphology of microglia in the normal adult mouse brain. Neuroscience, 39, 151-170.  Lawson, L.J., Perry, V.H. & Gordon, S. (1992) Turnover of resident microglia in the normal adult mouse brain. Neuroscience, 48, 405-415.  Lee, J. & Pilch, P.F. (1994) The insulin receptor: structure, function, and signaling Am J Physiol, 266, C319-C334.  Lesort, M. & Johnson, G.V. (2000) Insulin-like growth factor-1 and insulin mediate transient site-selective increases in tau phosphorylation in primary cortical neurons. Neuroscience, 99, 305-316.  Li, Y., Samuvel, D.J., Sundararaj, K.P., Lopes-Virella, M.F. & Huang, Y. (2010) IL-6 and high glucose synergistically upregulate MMP-1 expression by U937 mononuclear phagocytes via ERK1/2 and JNK pathways and c-Jun. J Cell Biochem, 110, 248-259.  Li, Y.P., Bushnell, A.F., Lee, C.M., Perlmutter, L.S. & Wong, S.K. (1996) Beta-amyloid induces apoptosis in human-derived neurotypic SH-SY5Y cells. Brain Res, 738, 196-204.  Li-Bo, Y., Wen-Bo, Q., Xiao-Hong, L., You-Lun, F. & Tie, Z. (2011) Intermittent high glucose promotes expression of proinflammatory cytokines in monocytes. Inflamm Res, 60, 367-370.  Liang, C.P., Han, S., Okamoto, H., Carnemolla, R., Tabas, I., Accili, D. & Tall, A.R. (2004) Increased CD36 protein as a response to defective insulin signaling in macrophages. J Clin Invest, 113, 764-773.  Lim, C.S., Lee, J.C., Kim, S.D., Chang, D.J. & Kaang, B.K. (2002) Hydrogen peroxide-induced cell death in cultured Aplysia sensory neurons. Brain Res, 941, 137-145.  Lin, L. & Holscher, C. (2007) Common pathological processes in Alzheimer disease and type 2 diabetes: a review. Brain Res Rev, 56, 384-402.  Lin, Y., Berg, A.H., Iyengar, P., Lam, T.K., Giacca, A., Combs, T.P., Rajala, M.W., Du, X., Rollman, B., Li, W., Hawkins, M., Barzilai, N., Rhodes, C.J., Fantus, I.G., Brownlee, M. & Scherer, P.E. (2005) The hyperglycemia-induced inflammatory response in adipocytes: the role of reactive oxygen species. J Biol Chem, 280, 4617-4626.  	   129	  Lindsay, J., Laurin, D., Verreault, R., Hebert, R., Helliwell, B., Hill, G.B. & McDowell, I. (2002) Risk factors for Alzheimer's disease: a prospective analysis from the Canadian Study of Health and Aging. Am J Epidemiol, 156, 445-453.  Little, J.P., Madeira, J.M. & Klegeris, A. (2012) The saturated fatty acid palmitate induces human monocytic cell toxicity toward neuronal cells: exploring a possible link between obesity-related metabolic impairments and neuroinflammation. J Alzheimers Dis, 30 Suppl 2, S179-183.  Liu, Y., Liu, F., Grundke-Iqbal, I., Iqbal, K. & Gong, C.X. (2011) Deficient brain insulin signalling pathway in Alzheimer's disease and diabetes. J Pathol, 225, 54-62.  Lopes, F.M., Schroder, R., da Frota, M.L., Jr., Zanotto-Filho, A., Muller, C.B., Pires, A.S., Meurer, R.T., Colpo, G.D., Gelain, D.P., Kapczinski, F., Moreira, J.C., Fernandes Mda, C. & Klamt, F. (2010) Comparison between proliferative and neuron-like SH-SY5Y cells as an in vitro model for Parkinson disease studies. Brain Res, 1337, 85-94.  Lowe, W.L., Jr., Boyd, F.T., Clarke, D.W., Raizada, M.K., Hart, C. & LeRoith, D. (1986) Development of brain insulin receptors: structural and functional studies of insulin receptors from whole brain and primary cell cultures. Endocrinology, 119, 25-35.  Lu, F.P., Lin, K.P. & Kuo, H.K. (2009) Diabetes and the risk of multi-system aging phenotypes: a systematic review and meta-analysis. PLoS One, 4, e4144.  Lue, L.F., Walker, D.G. & Rogers, J. (2001) Modeling microglial activation in Alzheimer's disease with human postmortem microglial cultures. Neurobiol Aging, 22, 945-956.  Madeira, J.M., Renschler, C.J., Mueller, B., Hashioka, S., Gibson, D.L. & Klegeris, A. (2013) Novel protective properties of auranofin: inhibition of human astrocyte cytotoxic secretions and direct neuroprotection. Life Sci, 92, 1072-1080.  Maragakis, N.J. & Rothstein, J.D. (2006) Mechanisms of Disease: astrocytes in neurodegenerative disease. Nat Clin Pract Neurol, 2, 679-689.  Martins, I.J., Hone, E., Foster, J.K., Sunram-Lea, S.I., Gnjec, A., Fuller, S.J., Nolan, D., Gandy, S.E. & Martins, R.N. (2006) Apolipoprotein E, cholesterol metabolism, diabetes, and the convergence of risk factors for Alzheimer's disease and cardiovascular disease. Mol Psychiatry, 11, 721-736.  McCall, A.L. (2004) Cerebral glucose metabolism in diabetes mellitus. Eur J Pharmacol, 490, 147-158.  	   130	  Mondragon-Rodriguez, S., Perry, G., Zhu, X. & Boehm, J. (2012) Amyloid Beta and tau proteins as therapeutic targets for Alzheimer's disease treatment: rethinking the current strategy. Int J Alzheimers Dis, 2012, 630182.  Moore, K.J. & Tabas, I. (2011) Macrophages in the pathogenesis of atherosclerosis. Cell, 145, 341-355.  Moroz, N., Tong, M., Longato, L., Xu, H. & de la Monte, S.M. (2008) Limited Alzheimer-type neurodegeneration in experimental obesity and type 2 diabetes mellitus. J Alzheimers Dis, 15, 29-44.  Murray, A.J., Knight, N.S., Cochlin, L.E., McAleese, S., Deacon, R.M., Rawlins, J.N. & Clarke, K. (2009) Deterioration of physical performance and cognitive function in rats with short-term high-fat feeding. FASEB J, 23, 4353-4360.  Napoli, I. & Neumann, H. (2009) Microglial clearance function in health and disease. Neuroscience, 158, 1030-1038.  Neumann, K.F., Rojo, L., Navarrete, L.P., Farias, G., Reyes, P. & Maccioni, R.B. (2008) Insulin resistance and Alzheimer's disease: molecular links & clinical implications. Curr Alzheimer Res, 5, 438-447.  Nieto-Vazquez, I., Fernandez-Veledo, S., Kramer, D.K., Vila-Bedmar, R., Garcia-Guerra, L. & Lorenzo, M. (2008) Insulin resistance associated to obesity: the link TNF-alpha. Arch Physiol Biochem, 114, 183-194.  Omori, K., Ohira, T., Uchida, Y., Ayilavarapu, S., Batista, E.L., Jr., Yagi, M., Iwata, T., Liu, H., Hasturk, H., Kantarci, A. & Van Dyke, T.E. (2008) Priming of neutrophil oxidative burst in diabetes requires preassembly of the NADPH oxidase. J Leukoc Biol, 84, 292-301.  Ottlecz, A., Gecse, A., Koltai, M. & West, G.B. (1977) Involvement of the kinin system in the insulin-induced inhibition of carrageenin oedema in rats. Monogr Allergy, 12, 131-137.  Ozcan, U., Cao, Q., Yilmaz, E., Lee, A.H., Iwakoshi, N.N., Ozdelen, E., Tuncman, G., Gorgun, C., Glimcher, L.H. & Hotamisligil, G.S. (2004) Endoplasmic reticulum stress links obesity, insulin action, and type 2 diabetes. Science, 306, 457-461.  Pansuria, M., Xi, H., Li, L., Yang, X.F. & Wang, H. (2012) Insulin resistance, metabolic stress, and atherosclerosis. Front Biosci (Schol Ed), 4, 916-931.  Parpura, V., Heneka, M.T., Montana, V., Oliet, S.H., Schousboe, A., Haydon, P.G., Stout, R.F., Jr., Spray, D.C., Reichenbach, A., Pannicke, T., Pekny, M., Pekna, M., Zorec, R. & Verkhratsky, A. (2012) Glial cells in (patho)physiology. J Neurochem, 121, 4-27. 	   131	   Peila, R., Rodriguez, B.L. & Launer, L.J. (2002) Type 2 diabetes, APOE gene, and the risk for dementia and related pathologies: The Honolulu-Asia Aging Study. Diabetes, 51, 1256-1262.  Pellicano, M., Bulati, M., Buffa, S., Barbagallo, M., Di Prima, A., Misiano, G., Picone, P., Di Carlo, M., Nuzzo, D., Candore, G., Vasto, S., Lio, D., Caruso, C. & Colonna-Romano, G. (2010) Systemic immune responses in Alzheimer's disease: in vitro mononuclear cell activation and cytokine production. J Alzheimers Dis, 21, 181-192.  Pereira Tde, O., da Costa, G.N., Santiago, A.R., Ambrosio, A.F. & dos Santos, P.F. (2010) High glucose enhances intracellular Ca2+ responses triggered by purinergic stimulation in retinal neurons and microglia. Brain Res, 1316, 129-138.  Powers, W.J. (1981) Cerebrospinal fluid to serum glucose ratios in diabetes mellitus and bacterial meningitis. Am J Med, 71, 217-220.  Quan, Y., Du, J. & Wang, X. (2007) High glucose stimulates GRO secretion from rat microglia via ROS, PKC, and NF-kappaB pathways. J Neurosci Res, 85, 3150-3159.  Quan, Y., Jiang, C.T., Xue, B., Zhu, S.G. & Wang, X. (2011) High glucose stimulates TNFalpha and MCP-1 expression in rat microglia via ROS and NF-kappaB pathways. Acta Pharmacol Sin, 32, 188-193.  Quinn, J.F., Raman, R., Thomas, R.G., Yurko-Mauro, K., Nelson, E.B., Van Dyck, C., Galvin, J.E., Emond, J., Jack, C.R., Jr., Weiner, M., Shinto, L. & Aisen, P.S. (2010) Docosahexaenoic acid supplementation and cognitive decline in Alzheimer disease: a randomized trial. JAMA, 304, 1903-1911.  Ransom, B.R. & Ransom, C.B. (2012) Astrocytes: multitalented stars of the central nervous system. Methods Mol Biol, 814, 3-7.  Reale, M., Iarlori, C., Gambi, F., Lucci, I., Salvatore, M. & Gambi, D. (2005) Acetylcholinesterase inhibitors effects on oncostatin-M, interleukin-1 beta and interleukin-6 release from lymphocytes of Alzheimer's disease patients. Exp Gerontol, 40, 165-171.  Reger, M.A., Watson, G.S., Green, P.S., Wilkinson, C.W., Baker, L.D., Cholerton, B., Fishel, M.A., Plymate, S.R., Breitner, J.C., DeGroodt, W., Mehta, P. & Craft, S. (2008) Intranasal insulin improves cognition and modulates beta-amyloid in early AD. Neurology, 70, 440-448.  Rensing, K.L., Reuwer, A.Q., Arsenault, B.J., von der Thusen, J.H., Hoekstra, J.B., Kastelein, J.J. & Twickler, T.B. (2011) Reducing cardiovascular disease risk in 	   132	  patients with type 2 diabetes and concomitant macrovascular disease: can insulin be too much of a good thing? Diabetes Obes Metab, 13, 1073-1087.  Retnakaran, R., Hanley, A.J., Raif, N., Connelly, P.W., Sermer, M. & Zinman, B. (2003) C-reactive protein and gestational diabetes: the central role of maternal obesity. J Clin Endocrinol Metab, 88, 3507-3512.  Sakaue, S., Nishihira, J., Hirokawa, J., Yoshimura, H., Honda, T., Aoki, K., Tagami, S. & Kawakami, Y. (1999) Regulation of macrophage migration inhibitory factor (MIF) expression by glucose and insulin in adipocytes in vitro. Mol Med, 5, 361-371.  San Jose, G., Bidegain, J., Robador, P.A., Diez, J., Fortuno, A. & Zalba, G. (2009) Insulin-induced NADPH oxidase activation promotes proliferation and matrix metalloproteinase activation in monocytes/macrophages. Free Radic Biol Med, 46, 1058-1067.  Sastre, M., Klockgether, T. & Heneka, M.T. (2006) Contribution of inflammatory processes to Alzheimer's disease: molecular mechanisms. Int J Dev Neurosci 24, 167-176.  Sato, T., Shimogaito, N., Wu, X., Kikuchi, S., Yamagishi, S. & Takeuchi, M. (2006) Toxic advanced glycation end products (TAGE) theory in Alzheimer's disease. Am J Alzheimers Dis Other Demen, 21, 197-208.  Satomi, N., Sakurai, A. & Haranaka, K. (1985) Relationship of hypoglycemia to tumor necrosis factor production and antitumor activity: role of glucose, insulin, and macrophages. J Natl Cancer Inst, 74, 1255-1260.  Sattar, N., McCarey, D.W., Capell, H. & McInnes, I.B. (2003) Explaining how "high-grade" systemic inflammation accelerates vascular risk in rheumatoid arthritis. Circulation, 108, 2957-2963.  Sayeski, P.P. & Kudlow, J.E. (1996) Glucose metabolism to glucosamine is necessary for glucose stimulation of transforming growth factor-alpha gene transcription. J Biol Chem, 271, 15237-15243.  Seino, S. & Bell, G.I. (1989) Alternative splicing of human insulin receptor messenger RNA. Biochem Biophys Res Commun, 159, 312-316.  Shanmugam, N., Reddy, M.A., Guha, M. & Natarajan, R. (2003) High glucose-induced expression of proinflammatory cytokine and chemokine genes in monocytic cells. Diabetes, 52, 1256-1264.  	   133	  Sharma, R., Buras, E., Terashima, T., Serrano, F., Massaad, C.A., Hu, L., Bitner, B., Inoue, T., Chan, L. & Pautler, R.G. (2010) Hyperglycemia induces oxidative stress and impairs axonal transport rates in mice. PLoS One, 5, e13463.  Shen, C.L. & Murphy, R.M. (1995) Solvent effects on self-assembly of beta-amyloid peptide. Biophys J, 69, 640-651.  Sims-Robinson, C., Kim, B., Rosko, A. & Feldman, E.L. (2010) How does diabetes accelerate Alzheimer disease pathology? Nat Rev Neurol, 6, 551-559.  Sisino, G., Bouckenooghe, T., Aurientis, S., Fontaine, P., Storme, L. & Vambergue, A. (2013) Diabetes during pregnancy influences Hofbauer cells, a subtype of placental macrophages, to acquire a pro-inflammatory phenotype. Biochim Biophys Acta, 1832, 1959-1968.  Smith, J.A., Das, A., Ray, S.K. & Banik, N.L. (2012) Role of pro-inflammatory cytokines released from microglia in neurodegenerative diseases. Brain Res Bull, 87, 10-20.  Sonnen, J.A., Larson, E.B., Brickell, K., Crane, P.K., Woltjer, R., Montine, T.J. & Craft, S. (2009) Different patterns of cerebral injury in dementia with or without diabetes. Arch Neurol, 66, 315-322.  Stavniichuk, R., Shevalye, H., Hirooka, H., Nadler, J.L. & Obrosova, I.G. (2012) Interplay of sorbitol pathway of glucose metabolism, 12/15-lipoxygenase, and mitogen-activated protein kinases in the pathogenesis of diabetic peripheral neuropathy. Biochem Pharmacol, 83, 932-940.  Steele, M.L. & Robinson, S.R. (2012) Reactive astrocytes give neurons less support: implications for Alzheimer's disease. Neurobiol Aging, 33, 423 e421-413.  Strang, F., Scheichl, A., Chen, Y.C., Wang, X., Htun, N.M., Bassler, N., Eisenhardt, S.U., Habersberger, J. & Peter, K. (2012) Amyloid plaques dissociate pentameric to monomeric C-reactive protein: a novel pathomechanism driving cortical inflammation in Alzheimer's disease? Brain Pathol, 22, 337-346.  Takahashi, K., Takeya, M. & Sakashita, N. (2002) Multifunctional roles of macrophages in the development and progression of atherosclerosis in humans and experimental animals. Med Electron Microsc, 35, 179-203.  Tambuyzer, B.R., Ponsaerts, P. & Nouwen, E.J. (2009) Microglia: gatekeepers of central nervous system immunology. J Leukoc Biol, 85, 352-370.  Tomlinson, D.R. & Gardiner, N.J. (2008) Glucose neurotoxicity. Nat Rev Neurosci, 9, 36-45.  	   134	  Ton, H. & Xiong, H. (2013) Astrocyte Dysfunctions and HIV-1 Neurotoxicity. J AIDS Clin Res, 4, 255.  Tracy, R.P., Lemaitre, R.N., Psaty, B.M., Ives, D.G., Evans, R.W., Cushman, M., Meilahn, E.N. & Kuller, L.H. (1997) Relationship of C-reactive protein to risk of cardiovascular disease in the elderly. Results from the Cardiovascular Health Study and the Rural Health Promotion Project. Arterioscler Thromb Vasc Biol, 17, 1121-1127.  Unger, J.W., Livingston, J.N. & Moss, A.M. (1991) Insulin receptors in the central nervous system: localization, signalling mechanisms and functional aspects. Prog Neurobiol, 36, 343-362.  Uysal, K.T., Wiesbrock, S.M., Marino, M.W. & Hotamisligil, G.S. (1997) Protection from obesity-induced insulin resistance in mice lacking TNF-alpha function. Nature, 389, 610-614.  Vagelatos, N.T. & Eslick, G.D. (2013) Type 2 Diabetes as a Risk Factor for Alzheimer's Disease: The Confounders, Interactions, and Neuropathology Associated With This Relationship. Epidemiol Rev.  Villanueva, E.B., Little, J.P., Lambeau, G. & Klegeris, A. (2012) Secreted phospholipase A(2) group IIA is a neurotoxin released by stimulated human glial cells. Mol Cell Neurosci, 49, 430-438.  Wallum, B.J., Taborsky, G.J., Jr., Porte, D., Jr., Figlewicz, D.P., Jacobson, L., Beard, J.C., Ward, W.K. & Dorsa, D. (1987) Cerebrospinal fluid insulin levels increase during intravenous insulin infusions in man. J Clin Endocrinol Metab, 64, 190-194.  Wang, J., Li, G., Wang, Z., Zhang, X., Yao, L., Wang, F., Liu, S., Yin, J., Ling, E.A., Wang, L. & Hao, A. (2012) High glucose-induced expression of inflammatory cytokines and reactive oxygen species in cultured astrocytes. Neuroscience, 202, 58-68.  Wang, L., Zhai, Y.Q., Xu, L.L., Qiao, C., Sun, X.L., Ding, J.H., Lu, M. & Hu, G. (2014) Metabolic inflammation exacerbates dopaminergic neuronal degeneration in response to acute MPTP challenge in type 2 diabetes mice. Exp Neurol, 251, 22-29.  Wellen, K.E. & Hotamisligil, G.S. (2005) Inflammation, stress, and diabetes. J Clin Invest, 115, 1111-1119.  WHO (2012) Diabetes fact sheet. World Health Organization.  	   135	  Wolf, G., Wenzel, U., Burns, K.D., Harris, R.C., Stahl, R.A. & Thaiss, F. (2002) Angiotensin II activates nuclear transcription factor-kappaB through AT1 and AT2 receptors. Kidney Int, 61, 1986-1995.  Wurm, S., Neumeier, M., Weigert, J., Wanninger, J., Gerl, M., Gindner, A., Schaffler, A., Aslanidis, C., Scholmerich, J. & Buechler, C. (2008) Insulin induces monocytic CXCL8 secretion by the mitogenic signalling pathway. Cytokine, 44, 185-190.  Wyss-Coray, T. (2006) Inflammation in Alzheimer disease: driving force, bystander or beneficial response? Nat Med, 12, 1005-1015.  Yamashita, M., Tamasawa, N., Matsuki, K., Tanabe, J., Murakami, H., Matsui, J. & Suda, T. (2010) Insulin suppresses HDL-mediated cholesterol efflux from macrophages through inhibition of neutral cholesteryl ester hydrolase and ATP-binding cassette transporter G1 expressions. J Atheroscler Thromb, 17, 1183-1189.  Yong, V.W., Power, C., Forsyth, P. & Edwards, D.R. (2001) Metalloproteinases in biology and pathology of the nervous system. Nat Rev Neurosci, 2, 502-511.  Zandi, P.P., Anthony, J.C., Hayden, K.M., Mehta, K., Mayer, L. & Breitner, J.C. (2002) Reduced incidence of AD with NSAID but not H2 receptor antagonists: the Cache County Study. Neurology, 59, 880-886.  Zeng, G. & Quon, M.J. (1996) Insulin-stimulated production of nitric oxide is inhibited by wortmannin. Direct measurement in vascular endothelial cells. J Clin Invest, 98, 894-898.  Zhang, B., Gaiteri, C., Bodea, L.G., Wang, Z., McElwee, J., Podtelezhnikov, A.A., Zhang, C., Xie, T., Tran, L., Dobrin, R., Fluder, E., Clurman, B., Melquist, S., Narayanan, M., Suver, C., Shah, H., Mahajan, M., Gillis, T., Mysore, J., MacDonald, M.E., Lamb, J.R., Bennett, D.A., Molony, C., Stone, D.J., Gudnason, V., Myers, A.J., Schadt, E.E., Neumann, H., Zhu, J. & Emilsson, V. (2013) Integrated systems approach identifies genetic nodes and networks in late-onset Alzheimer's disease. Cell, 153, 707-720.        	   136	   Appendix A   Buffers & Reagents:   1. ELISA Reagents  Coating buffer (0.1 M sodium bicarbonate buffer): Dissolved 0.159 g Na2CO3 and 0.293 g NaHCO3 in 0.1 l of autoclaved milli-Q water (pH adjusted to 9.6). The solution was stored at room temperature in an airtight container.  Phosphate stock solution: 3.7 g NaH2PO4.H2O and 40.3 g Na2H2PO4 were dissolved in 0.5 l of autoclaved milli-Q water and stored at room temperature in an airtight bottle.  PBS-Tween: 50 ml phosphate stock solution and 0.5 ml Tween 20 were added to 0.95 l of 0.9% w/v solution of sodium chloride. The solution was stored at room temperature in an airtight bottle.  Blocking solution: 1% w/v BSA and 1% w/v skim milk powder were dissolved in PBS, covered with Parafilm and stored at 4 °C for up to 1 week.  Diethanolamide substrate solution: Dissolved 101mg MgCl2.5H2O and 97mL diethanolamine in 1 l milli-Q water (pH adjusted to 9.8). The solution was stored in dark at 4 °C.   2. Western blotting reagents  Sterile PBS: One PBS tablet was dissolved in 100 ml of Milli-Q water. The solution was autoclaved.  	   137	  RIPA buffer: 150 mM sodium chloride, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl, pH 8.0. The solution was stored in a 50 ml tube at 4 °C.  4X Resolving gel buffer: 0.4% SDS, 1.5 M Tris base, pH 8.8. The solution was stored at room temperature in an airtight bottle.  4X Stacking gel buffer: 0.4% SDS, 0.5 M Tris base, pH 6.8. The solution was stored at room temperature in an airtight bottle.  5X Loading buffer: 10% SDS, 0.05% bromophenol blue, 50% glycerol, 0.313 M Tris-HCl, pH 6.8. Aliquots were stored at -20 °C.  1X Running buffer: 25 mM Tris base, 190 mM glycine, 0.1% SDS, pH 8.3. The solution was stored at room temperature in an airtight bottle.  1X Transfer buffer: 48 mM Tris base, 39 mM glycine, 0.04% SDS, 20% methanol.  The solution was stored at 4 °C in an airtight bottle. Stripping buffer: 10% Tween 20, 0.1% SDS, 0.2 M glycine, pH 2.2. The solution was stored at room temperature in an airtight bottle.   1X TBS-T: 50 mM Tris-HCl, 150 mM sodium chloride, 0.1% Tween 20, pH 7.4. The solution was stored at room temperature in an airtight bottle.   3. DNA PAGE buffers  10X TBE: Dissolved 108 g Tris base and 55 g sodium borate in 700 ml of milli-Q water; added 200 ml of 100 mM EDTA and adjusted pH to 8.5; added milli-Q water to 1 l. The solution was stored in an airtight container at room temperature.  5X TBE and 1X TBE: Diluted 10X TBE to 5X and 1X using milli-Q water.   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0074350/manifest

Comment

Related Items