UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Cellular signaling of human microglia in response to [beta]-amyloid 1-40 Helm, Jeffrey 2001

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

Item Metadata


831-ubc_2001-0399.pdf [ 3.98MB ]
JSON: 831-1.0089979.json
JSON-LD: 831-1.0089979-ld.json
RDF/XML (Pretty): 831-1.0089979-rdf.xml
RDF/JSON: 831-1.0089979-rdf.json
Turtle: 831-1.0089979-turtle.txt
N-Triples: 831-1.0089979-rdf-ntriples.txt
Original Record: 831-1.0089979-source.json
Full Text

Full Text

C E L L U L A R S I G N A L I N G O F H U M A N M I C R O G L I A I N R E S P O N S E T O (3AMYLOID  1-40  by JEFFREY H E L M B . S c , Washington State University, (U.S.A.) 1998 A THESIS S U B M I T T E D IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF  SCIENCE in  T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Pharmacology and Therapeutics, Graduate Program in Neuroscience) We accept this thesis as conforming to the required standard  T H E UNIVERSITY O F BRITISH C O L U M B I A September 2001 © J e f f r e y Helm, 2001  In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the r e q u i r e m e n t s f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g of t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the head of my department or by h i s or her r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n .  Department  of  The U n i v e r s i t y o f B r i t i s h Columbia Vancouver, Canada  Abstract Microglia are resident immune cells of the brain that are activated in response to trauma and inflammation. Activated microglia exhibit characteristics similar to peripheral macrophages, such as the expression of irnmunomolecules, secretion of proinflammatory substances and phagocytic activity. Like macrophages, microglia exhibit these characteristics in order to defend the brainfrominfection and aid in the repair of damaged tissue. However, in Alzheimer's Disease (AD) microglia can become overactivated resulting in the release of substances that escalate Marnmation and ultimately cause neuronal death. A protein implicated in the progression of AD is P-amyloid (AP). AP production is increased in AD and deposits of AP form throughout the brain, which are correlated to the activation of microglia and neuronal death. Studies have shown that Ap can activate microglia and cause changes in the cellular functions of these cells. For example, in microglia Ap has been shown to cause increases in the production of pro-inflarnrnatory cytokines and reactive oxygen species. The objective of this work was to characterize the actions of Ap40, a commonly expressed form of Ap, on the mobilization of intracellular calcium ([Ca ]0 in human 2+  microglia. The rational was that subsequent pharmacological modulation of the calcium signals induced by Ap40 could then be used to alter the cellular functions of microglia, such as the secretion of neurotoxic factors. Thefirststudy used calcium sensitive microfluorescence to examine AP40 actions on [Ca ], mediated signaling pathways. AP40 2+  application (4 and 10 uM) to microglia caused a rise in [Ca  2+  ]j to a plateau  level which  was sustained following the removal of the peptide. Calcium-free external solution (Ca-  ii  free PSS) was used to show that the primary contribution to the [Ca ]; rise came from the 2+  influx o f extracellular calcium. A small amount o f intracellular calcium release is also possible since Ca-free PSS did not totally inhibit the Ap40-induced  [Ca ]i 2 +  response. The  Ap40 mediated calcium influx was sensitive to depolarization since low chloride solution applied extracellularly inhibited the influx o f calcium. Additional experiments suggested that a store-operated channel ( S O C ) did not mediate the Ap40-induced calcium influx since an inhibitor of this pathway, SKF96365, had no effect on the [Ca ]jrise. A t 2+  present, a specific modifier o f the Ap40-induced influx pathway has not been identified. The next study examined the actions o f Ap40 on C O X - 2 expression. C O X - 2 is an enzyme responsible for prostaglandin synthesis and free radical formation that is overexpressed in A D . Microglia cultures were treated with Ap40 for 24 hrs and the expression o f C O X - 2 was determined through R T - P C R analysis. The results show that  Ap40 significantly increased C O X - 2 expression. However, it is not know if the enhancement o f C O X - 2 is linked to the Ap40-induced increase in [Ca ]j. 2+  The third study examined the potential of Ap40 to induce the production o f neurotoxic substances by human microglia. Neuroblastoma cells were treated with supernatant from microglia exposed to AP40 and the neurotoxic effects were evaluated by assessing cell viability. The results indicate that supernatant from AP40 treated microglia decreased neuroblastoma viability, however the decrease was not significantly different from AP40 applied directly to neuroblastoma cells. This result suggests that a larger number o f human microglia are required to record the effects of neurotoxic substances in the cell viability assays used.  iii  The final study was based on preliminary data indicating that [Ca ]; signaling mediated by adenosine triphosphate ( A T P ) and platelet activating factor (PAF) in microglia from A D brain is abnormal. This study examined the actions o f chronic A p 4 0 treatment (one or two days) on the [Ca ]j responses o f microglia to A T P and P A F . 2+  The results show that chronic A p 4 0 treatment o f human microglia diminished P A F (but no A T P ) responses and also elevated basal [Ca ]i levels, suggesting that the peptide 2+  perturbs calcium signaling. These results provide a viable in vitro model since A p 4 0 treatment o f human fetal microglia mimics the properties o f adult microglia in A D . In summary, A p 4 0 in human fetal microglia induces a rise in [ C a ] j d u e t o a S O C 2 +  independent pathway and also increases the expression o f neurotoxic factors such as C O X - 2 . Furthermore, it appears that A P 4 0 treatment perturbs calcium signaling in human fetal microglia in a manner similar to what is observed in A D microglia.  iv  Table of Contents Abstract Table of Contents List of Figures and Tables List of Abbreviations Acknowledgements  1. Introduction  1.1.  Microglia  1.2.  Microglia and Disease  1.3.  p-Amyloid  1.4.  AP and Microglia  1.5.  Which Ap to use?  1.6.  Study Aims 1.6.1. Aim 1: The Ap40 [Ca ]i signal 2+  1.6.2. Aim 2: Ap40 in COX-2 expression 1.6.3. Aim 3: AP40 in neurotoxin induction 1.6.4. Aim 4: AP40 alteration of [Ca ] signaling 2+  (  1.6.5. Summary of Aims  2. Materials and Methods  2.1.  Preparation of human microglia  2.2.  Study 1: The A04O  [Ca ]i response 2+  in microglia  2.2.1. Calcium spectrofluorometry 2.2.2. Solutions and chemicals 2.3.  Study 2: The effects of Ap40 on COX-2 expression  2.4.  Study 3: AP40 induced neurotoxic potential 2.4.1. Reagents 2.4.2. Cell Culture 2.4.3. Neurotoxicity of microglia supernatants 2.4.4. Cell death and viability assays  2.5  Study 4: The effects of Ap40 on ATP & PAF responses  3. Results  3.1  Study 1: The Ap40 [Ca ] response in microglia  3.2.  Study 2: The effects of Ap40 on COX-2 expression  .3.3. 3.4.  ;  Study 3: Ap40-induced neurotoxic potential Study 4: The effects of Ap40 on ATP & PAF responses  vi  4. Discussion  62  4.1.  The use of primary cell cultures of human microglia  63  4.2.  The use of Ap in vitro as stimulus  64  4.3.  Ap40 induced [Ca ]j signaling pathway  65  4.4  Ap40 induced COX-2 expression in microglia  69  4.5.  AP40 induction of neurotoxic substances from microglia  71  4.6.  The actions of Ap40 on ATP and PAF [Ca ] responses  73  4.7.  Final Summary  75  2+  2+  ;  References  77  vii  List of Figures and Tables Figure 1: The cycle of microglia activation in Alzheimer's Disease  7  Table 1: Approximate conversion of UV ratios to [Ca2+]I  21  Figure 2: The effects of Ap40 in human microglia with Ca-PSS  32  ' Table 2: The n values and statistics of experimentsfromStudy 1  34  Figure 3: The effects of AP40 in human microglia with Ca-free PSS  35  Figure 4: Ca-free PSS inhibits Ap40-induced influx  37  Figure 5: Low Cl' PSS inhibits Ap40-induced influx  40  Figure 6: The effects of SKF on Ap40-induced influx  41  Figure 7: The effects of AP40 on COX-2 expression in human microglia  45  Figure 8: The neurotoxic effects of Ap40 treated human microglia  49  Figure 9: Normal [Ca ]i responses to ATP and PAF in human microglia  54  Figure 10: AP40 effects on basal [Ca ]i in human microglia  56  Figure 11: Ap40 effects on PAF amplitudes in human microglia  58  Figure 12: Ap40 effects on PAF influx response in human microglia  60  2+  2+  viii  List of Abbreviations AD  Alzheimer's Disease  ATP  adenosine tri-phosphate  APP  amyloid precursor protein  AP  P-amyloid  AP40  P-amyloid 1-40  Ap42  p-amyloid 1-42  BBB  blood brain barrier  Ca  ionic calcium  2 +  cDNA  complementary-deoxyribonucleic acid  CNS  central nervous system  COX-2  cyclooxygenase-2  Ca-free PSS  calcium free physiological saline  Ca-PSS  normal physiological saline  [Ca ]i  intracellular calcium  2+  °C  degrees Celsius  ECM  extracellular matrix  ER  endoplasmic reticulum  Fura 2 A M  Fura-2 acetoxymethylester  GAPDH  glyceraldehydes-3-phosphate dehydrogenase  hrs  hours  low Cl" PSS  low chloride physiological saline  ix  mg  milligrams  min  minutes  ml  milliliter  mM  millimolar  mRNA  messenger ribonucleic acid  pg  microgram  uM  micromolar  nm  nanometers  PAF  platelet activating factor  RNA  ribonucleic acid  RT-PCR  reverse transcriptase polymerase chain reaction  SEM  standard error o f the mean  SOC  store operated channel  SKF963651  SKF  UV  ultra-violet light  x  Acknowledgements  I would first like to thank Dr. McLarnon for his guidance and instruction. I would also like to thank all the other members o f the McLarnon laboratory past and present; Sonia Franciosi, Choi Hyun Beom, Clarence Khoo, and Vikram Goghari.  I would also like to thank Dr. Seung U . K i m and all the members o f his laboratory for their assistance.  Special thanks to Df. Nagai and Choi for their help with the P C R work, and Andis Klegeris for his help on the neurotoxicity work, I couldn't have done it without you guys!  Cosmic thanks goes out to my father, the greatest scientist and man that I know.  xi  1.  Introduction 1.1.  Microglia  Rio-Hortega's early work in 1932 described the morphology and three functional states o f microglia that are known today; the resting or ramified state, an activated phagocytic state and an intermediate state [Haerter-Gebicke 1996, Kreutzberg 1996, Stoll 1999]. These three functional states o f microglia are their most striking characteristic and allude to their functional role within the brain.  The function o f the ramified form o f microglia has been overshadowed by the activated  morphologies, largely because it is much easier to study the activated states since resting microglia are very sensitive to changes in the extracellular environment, as a consequence they become activated very easily. In other words, due to the cell preparation process needed for the study o f microglia in vitro the cells are, for all practical purposes, in a state o f low activation and are not in a fully resting state [Haerter-Gebicke 1996, Kato 2000]. However, this low level o f activation does not prevent the study of microglial activation or the examination of activated microglia. The activated phagocytic state o f microglia closely resembles circulating monocytes; indeed, microglia express many o f the same receptors as macrophages in the blood stream [Gehrmann 1995, Kato 2000, Moore 1996]. This macrophagic behavior o f microglia, along with observations o f activated microglia at sites o f brain injury, first implicated microglia as irnmunoreactive cells o f the central nervous system ( C N S ) [Kreutzberg 1996, McGeer 1995]. U p o n activation microglia exhibit a number of common cellular responses: increased proliferation, localization to the site o f injury/stimulation, upregulation o f immunomolecules and morphological changes to a more phagocytic phenotype [Gehrmann 1995, Stoll 1999]. These cellular responses o f activated microglia bring into question the lineage o f these cells. O n one hand they appear to be very similar to macrophages o f the blood stream; however, microglia also have a ramified  1  morphology irnlike macrophages and they appear to be able to proliferate independent of bone marrow cells [Gehrmann 1995, Haerter-Gebicke 1996]. Regardless o f the developmental origin o f microglia it is important to stress that microglia are a unique cell type and are not simply peripheral macrophages that cross the blood-brain barrier ( B B B ) and enter the C N S . Microglia have a unique role in the brain and are very sensitive to changes in the C N S . A s a consequence, a major focus o f microglia research centers on signaling pathways o f microglia activation. There is a large body of evidence that suggests that increases in the activation state o f microglia are associated with neurodegenerative diseases o f the brain. A common factor in cellular signaling, and consequently cellular activity, is modulations o f intracellular calcium ( [ C a ] i ) . 2+  Calcium does much more than modulate the electrophysiological properties o f the cell membrane, proper regulation o f calcium is also important for cell signaling and survival [Felder 1994, Verkhratsky 1996]. Across all cell types calcium has been shown to signal modulations in channel and receptor properties, peptide function, and gene transcription [Carafoli, 1999. Ginty 1997]. Thus it makes sense that cells do not rely entirely on membrane channels and pumps for [Ca ]; 2+  regulation. A number of organelles have been shown to contain significant stores o f calcium for uptake and release, the two primary ones being the endoplasmic reticulum and mitochondria [Bode 1996]. The calcium stores provided by these organelles are referred to collectively as  intracellular  stores and they have been shown to play a large role in the maintenance o f [Ca ]i levels. 2+  Microglia are no different in regards to other cell types in their use o f calcium in cellular signaling. Indeed, changes in microglia calcium are o f particular interest in the study o f microglia function since it is thought that the transformation from a resting to an activated state in microglia is governed by changes in gene expression [Bader 1994]. Since the majority o f transcription modulators are sensitive to calcium, it would stand to reason that a signal for microglia activation  2  could be carried by modulations in intracellular calcium levels [Ghosh 1995, Sola 1999]. Thus research has been directed towards studying the calcium signaling responses induced by substances that are thought to activate microglia. The obvious candidates are substances that initiate inflammatory responses in the brain or are released as a result o f cell damage since activated microglia have been observed in these conditions. Inflammatory initiators such as bacterial lipopolysaccharide ( L P S ) have been shown to cause changes in basal [Ca ]j. Work by Bader et al. (1994) illustrated that L P S causes a transient 2+  increase in [ C a ] i in cultured rat microglia. Furthermore, work from the McLarnon laboratory has 2+  shown that inflammatory cytokines such as T N F - a and I L - i p , also cause sustained rises o f basal [ C a ] i in human microglia [Goghari 2000, McLarnon 2001]. Thus it appears that pro2+  inflammatory substances that activate microglia also cause changes in the [Ca ]; levels o f 2+  microglia. Proteins found in neurodegenerative pathologies such as complement and p-amyloid also cause changes in [ C a ] i , as do molecules such as adenosine tri-phosphate ( A T P ) and platelet 2+  activating factor (PAF), which are often released in conditions o f cell damage and ischemia [Wang 2000, Wang 1999]. These activators of microglia will be discussed in more detail in later sections, their brief mentioning at this point is only to illustrate that changes in microglia [Ca ]j can be 2+  caused by a diverse group o f activating agents. Current research suggests that [Ca ];and the level o f microglia activation are closely linked 2+  since substances that modulate [Ca ]; also appear to activate microglia. The next section will 2+  discuss how the level o f microglia activation may be cause for concern in inflammatory diseases o f the brain.  3  1.2.  Microglia and Disease  One might assume that since microglia aid in the repair of damaged tissue and the prevention of infection in the CNS, that the activation of microglia is a beneficial event in all cases. However, microglia have been implicated in actively contributing to the development of pathology in a number of neuronal diseases such as multiple sclerosis, HIV dementia, and Alzheimer's disease (AD) [Haerter-Gebicke 1996, Moore 1996]. In these diseases microglia become over-activated and start producing substances that either inhibit proper healing or are directly neurotoxic. It appears that a complex network of activation and inhibition controls the actions of microglia. Thus, chronic activation can trigger a lethal potentiality in microglia, that is usually only reserved for foreign antigens or diseased cells, to spill over onto healthy bystander neurons and glia that surround the area of insult [McGeer 2000]. A primarily example of a breakdown in the normal control mechanisms of microglia is illustrated in the pathology of A D . The dominant theory of A D pathogenesis holds that microglia are not the initiators of the disease but that they contribute to the pathology. Research has shown that one of the initial events in the development of A D is the abnormal processing of the amyloid precursor protein (APP), which leads to the production of abnormally high levels of the P-amyloid protein (AP) [Selkoe 1999, Wilson 1999]. APP and A p are both constitutively expressed in glia and neurons throughout the brain and are present in normal healthy individuals [Fukumoto 1999, Neve 1998]. Although the function of APP is not clear, knockout mice studies suggest that APP is necessary for development [Seabrook 1999]. The increase in APP processing leads to high concentrations of A p that then aggregate into insoluble plaques that form extracellularly throughout the brain [Busciglio 1993, Koo 1999]. Naturally microglia attempt to clear the excess A p debris as  would be expected from any resident macrophage; however, the plaques are dense and difficult to metabolize thus it is not possible to clear them quickly enough. Microglia react to the AP plaques on two different levels, one that regards the plaque as extracellular debris, and another level o f activation that is specific to the AP peptide itself. The premise that microglia react to AP as debris is supported by observations that show that microglia react to changes in the extracellular matrix ( E C M ) [Monning 1995] and also internalize A p fibrils [Paresce 1996]. Indeed, the main theory explaining the presence o f microglia at amyloid plaques is that they are attempting to clear the lesion [Wisniewski 1998]. In addition, AP has been shown to activate microglia directly through cells surface receptors such as the scavenger receptor (SR) and a Clq complement protein receptor [Jiang 1994, Khoury 1996]. The complement-signaling pathway that leads to microglia activation has also been shown to be upregulated in the A D brain [Eikelenboom 1996, Yasojima 1999]. AP can activate this pathway because the 1-16 region o f AP has a high binding affinity to the Clq complement protein; consequently, the A p i - 1 6 region o f AP can bind to the Clq receptors present on microglia and activate the complement-signaling pathway [Jiang 1994]. This may be a major pathway o f microglia activation in A D since it has been shown that Clq is one o f the most over expressed complement proteins found in A D brain [Yasojima 1999]. Further studies have also suggested that the 10-16 region o f AP preferentially binds to microglia and induces the induction o f a neurotoxin and an increase in phagocytosis [Giulian 1996]. These two pathways, along with the generalized response o f microglia to debris would be enough to implicate microglia in A D , but there are still other potentially harmful microglia interactions that occur.  U p o n activation microglia secrete a number o f substances that amplify the inflammatory response; cytokines increase the general level o f activity and recruit other microglia and astrocytes  5  into the inflammatory response [Kato 2000, Mattson 1997, McGeer 1995] and free oxygen species and other neurotoxic agents kill nearby neurons [Haerter-Gebicke 1996, Gehrmann 1995, Giulian 1995, McGeer 1995]. Neuron death contributes to the iriflarnmatory cascade because dying neurons can release substances such as A T P that have been shown to activate microglia [Moller 2000, Wang 1999]. There is also evidence that healthy neurons secrete substances that inhibit the inflammatory activity o f astocytes and microglia [Galoyan 2000]; thus the more neuronal death, the more inflammation, which subsequently leads to increased rates o f neuronal death. In addition, the elevation o f inflammation and activation o f microglia with pro-inflammatory cytokines has been shown to increase the production o f cyclooxygenase-2 ( C O X - 2 ) and subsequent prostanoids, which have the potential to cause neuronal damage through vasoconstriction and the production o f oxidative free radical species [Minghetti 1998]. C O X - 2 is the rate-limiting enzyme in prostaglandin synthesis and it is upregulated in brain inflammation, and in cases o f ischemia and A D in humans [Pasinetti 1998, Sairanen 1998]. Microglia also initiate a positive-feedback loop in regards to A p since activated microglia have been shown to secrete A p as well as iron which facilitates A p aggregation and plaque formation [Busciglio 1993, Chung 1999, Halliday 2000]. Thus a cycle o f A p stimulation, microglia activation, and inflammation develops which is visualized in Figure 1. In addition to the web o f microglia activators, preliminary studies from this laboratory using post-mortem adult microglia from cases o f A D suggest that normal microglia responses are impaired in A D . A T P and P A F were used to induce [Ca ]; responses in microglia from confirmed 2+  cases o f A D and control; in the latter case adult microglia were obtained from individuals with no diagnosis o f A D . Preliminary data gained thus far from this study suggest that responses to A T P and P A F in A D microglia are attenuated in comparison to control [unpublished].  6  Figure 1  Abnormal APP Processing  Ap40 & A042 production  Amyloid Plaque Formation  activators such as ATP and PAF  N E U R O N A L D E A T H  Figure 1: The network o f interactions in the formation o f an autotoxic loop in Alzheimer's Disease. A l l inputs in this illustration are excitatory in nature.  7  ATP and PAF were used as stimuli since previous work has established that rapid transient increases in intracellular calcium occur in response to both agonists and these responses have been well characterized and are highly reproducible [McLarnon 2000, Wang 2000]. These data suggest some component of AD causes microglia to alter their normal response patterns to stimuli. The cause of this change in microglia responsiveness as well as its relationship to the development of AD is currently unknown. By considering all the inflammatory interactions involved in the pathology of AD it is evident that a relatively small change in Ap processing can lead to an exponential increase in pathological symptoms and microglia autotoxicity. Therefore it is reasonable to conclude that any effort to control the activity and subsequent inflammation that microglia express in the brain would be of great therapeutic use in treating AD. 1.3.  p-Amyloid  As the previous discussion of microglia in disease would indicate, research on the Ap peptide is currently one of the major lines of study in AD. A pathological feature of the AD brain is the presence of amyloid plaques throughout the cortical and hippocampal region that are often found in the presence of activated microglia and neuronal cell death [Koo 1999, Wilson 1999, Wisniewski 1998]. The primary component of the amyloid plaques is p-amyloid, which is primarily expressed as Ap40 and Ap42, which are two different splice variants derived from the amyloid precursor protein (APP) [Koo 1999, Sinha 1999, Wilson 1999]. As previously mentioned, although the function of APP remains unknown, it is clear that APP is necessary in normal development since genetic knockouts of the APP and APP-like gene are lethal [Seabrook 1999, Slunt 1994]. Ap is produced by differential processing of the APP protein by three different enzymes a, p, and y secretase [Sinha 1999, Wilson 1999]. Normally a and p secretase cleave the  8  APP protein and then y secretase cleaves the fragments once again to produce Ap40 and AP42 respectively [Citron 1996]. Both Ap variants are produced in healthy individuals and are thought to play a role in normal development and cellular functioning [Busciglio 1993] and both neurons and glia have been shown to produce APP and A p [Fukumoto 1999]. Current work implicates that a dysfunction in APP processing is the causative factor in the development of A D . Mutations of the APP gene have been linked to cases of A D and Down syndrome, which displays plaques and cognitive deficits similar to A D [Koo 1999, Lansbury 1999, Lamer 1999]. Mutations in the other two major genetic markers for A D , PS-1 and PS-2 have also enhanced A p production in both transfected cells and transgenic animals [Duff 1999, Seabrook 1999, Sinha 1999]. Thus there are many lines of evidence that link A p to the development of A D like pathological symptoms. It is also clear from the earlier discussion of microglia that A p is capable of initiating a signaling cascade that ultimately leads to a cycle of microglia activation, neuronal death, and further microglia activation. In order to understand the involvement of Ap in the escalating cycle of microglia activity and inflammation a brief review of some of the research in the area of Ap signaling in microglia follows. 1.4.  A p and Microglia  From the previous discussion of microglia in disease, it is clear that the Ap peptide is capable of causing changes in microglia function. A p has been shown to modulate the secretion of cytokines, receptor expression, properties of ion channels, calcium levels, morphology, gene expression, production of neurotoxic factors, and even the secretion of A p by the microglia itself [Araujo 1992, Bader 1994, Bitting 1996, Giulian 1996, Silei 1999, Tan 1999]. In the scope of relating microglia to neurodegeneration most of the work can be put into two general categories;  9  studies dealing with upstream signaling elements that could lead to microglia activation, and studies that deal with downstream products o f activated microglia. Studies o f the upstream signaling events o f AP in microglia are exclusively centered on [Ca ]j levels. This can be linked to two separate causal factors. First, the availability o f calcium 2+  sensitive dyes such as Fura-2 and Indo has made the study o f [ C a ] i a relatively simple and 2+  straightforward procedure. Secondly, calcium modulation is the most likely candidate for carrying the activation signal since many activators o f microglia cause changes in [ C a ] i levels as well (see 2+  section 1.1).  Thus, it is likely that the initial signaling actions o f AP will be observable in  modulations o f [ C a ] i in microglia. 2+  One o f the first publications to show that Ap modulates [Ca ]j in microglia was from the 2+  Korotzer laboratory. Incubation o f an active fragment o f the Ap peptide, AP25-35 at 25 u M , in rat microglia cultures caused a 50% increase in basal [ C a ] i levels over a period o f one hour when 2+  compared to control conditions without Ap present [Korotzer 1995]. The same study also observed that a longer incubation time o f 3 and 6 hours with the full-length  AP peptide (Ap42) resulted in a  significant change in the basal [Ca ]j. However this work did not attempt to determine the 2+  characteristics o f the [Ca ]j response to Ap beyond measuring an overall increase when compared 2+  to untreated control cultures. Work by Silei and colleagues (1999) went a step further and attempted to examine the source o f the [Ca ]; response. Silei et al. used human fetal microglia 2+  cultures incubated with Ap25-35 at 4 0 u M for 90 minutes. The results show that Ap caused an increase in [Ca ]j that was blocked in calcium-free solution suggesting that the Ap response in 2+  microglia was governed by the influx o f calcium from the extracellular solution and not from intracellular stores o f calcium. Furthermore, work from the McLarnon laboratory has shown that a low concentration o f the full-length  Ap peptide (AP42) in human microglia at 1 u M is sufficient to  10  cause large increases in the [Ca ]j [unpublished]. These studies provide evidence that both full 2+  length and active fragments o f AP have the potential to modulate [ C a ] levels in microglia with 2+  the expectation that this could signal changes in the activation and function o f microglia downstream o f the initial [Ca ] response. Evidence that A p exerts actions downstream o f the initial [Ca ] signal is provided by 2+  Araujo & Cotman (1992), who showed that pro-inflammatory cytokines are released as a result o f A p treatment. In this study 10 pg/ml o f full-length A p 4 2 was sufficient to induce a significant increase in IL-1 release in rodent microglia. Furthermore, a number o f papers out o f the Landreth laboratory provide evidence that A p activation o f pro-inflammatory cytokines and chemokines causes the initiation o f tyrosine kinase and M A P kinase dependent pathways. Work by MacDonald (1998 & 1997) has found that A p 4 0 and Ap25-35 at 50-60 u M stimulates tyrosine phosphorylation, E R K , and p38 M A P K expression. Studies from the same laboratory (Combs 1999 & 2000) have also shown that A p acting through the same kinase dependent pathways stimulates the production o f pro-inflammatory and reactive oxygen species in monocytes and rat microglia that are potentially neurotoxic. Work from the Giulian laboratory suggests that AP signals the production o f a unique neurotoxic agent that is secreted by microglia. The results from in vitro experiments have shown that 1 u M o f A p 4 2 or A p 4 0 was sufficient to induce the production o f a neurotoxin by microglia [Giulian 1996], This, as o f yet unidentified, neurotoxin induced roughly 75-80% neuronal death. The same concentrations o f A p , applied without the presence o f microglia, resulted in no significant neuronal cell loss. Furthermore, the neurotoxin in question was determined, through a number o f molecular assays, to be neither a reactive oxygen species nor a known cytokine [Giulian  ll  1996], which are the two most common secretory products o f microglia thought to contribute to neuronal damage. The studies cited previously provide evidence that A p can signal changes in microglia that can account for a number o f pathological symptoms that are observed in A D . Ap has the capacity to activate and induce possible pathological events in microglia through signals carried by changes in [Ca ]j. Once activated by A p microglia can then enter a seemingly never ending and ever 2+  escalating cycle o f inflammation and neuronal death with the production o f pro-inflammatory molecules, neurotoxic substances, and increasing amounts o f A p . If more can be done to understand the signaling pathways that Ap initiates in microglia then therapeutic approaches can be developed that can treat A D and other inflammatory diseases o f the brain. T o that end the focus o f this thesis is to examine the signaling characteristics o f A p in human microglia cells.  1.5.  Which AP to use?  The purpose of this work is to characterize the effects of AP on signaling pathways in microglia, however, there are many different A p peptides that have been used, as noted from the previous discussion o f earlier work in this area. Active fragments o f A p such as AP25-35 are popular since they are relatively easy to use compared with the longer full-length A p peptides AP40 and Ap42. However, active fragments o f Ap are not found in the brain o f A D inflicted individuals [Neve 1998]. In addition, small active fragments of A p do not have the same activating potential as do the full-length peptide because they usually have only one active site. For example, the binding area o f the A p peptide for both the C l q and the induction o f the neurotoxin described by Giulian (1996) is in the 1-16 region so the popular Ap25-35 fragment would be unable to activate either o f these two pathways. Thus it seems premature to limit the activating potential o f  12  AP through the use o f Ap fragments until all the pathways o f Ap signaling have been characterized.  Therefore, the more common full length  Ap40 and Ap42 were the  only  Ap  peptides considered for this thesis since they are the most relevant Ap peptides concerning the pathology o f AD and they have the greatest activating potential. The majority o f  AD research, with microglia or otherwise,  has focused on  Ap42 since  it is  thought to be the main AP variant found in AD plaques and the more neurotoxic and insoluble o f the two  Ap isoforms  [Sinha  1999,  Wilson  1999].  However, there is some dispute as to whether  Ap  is directly neurotoxic and the role o f solubility in the activity o f the Ap peptide is also debated. There are a large number o f studies that have shown that the Ap peptide is not directly toxic to neurons since neurons can be grown in the presence o f  Ap peptides  [Wujek  peptides directly added to neurons do not cause cell damage [Giulian peptides into the brain does not cause neuronal damage [Games  1996],  1996],  1992].  plaque and  AP  and the infusion o f  Ap  Thus it appears that at the  level o f direct neurotoxic activity there is no preference for either AP variant. Many researchers also question the relevance o f the solubility factor o f the Ap peptide. Giulian's work on the microglia neurotoxin has shown  Ap40 and Ap42  causing the production o f a microglia-derived neurotoxin [Giulian  to be equally effective in  1996].  Further work by Giulian  illustrated that the beta-pleated sheet structure, which is formed by aggregated Ap, is unnecessary in inducing the microglia neurotoxin. These studies support the view that microglia respond to specific binding domains o f the AP peptide that are, at least in the case o f the neurotoxin and complement pathway, present in both the  Ap40 and Ap42 peptide.  likely a candidate in signaling pathological changes in microglia as  13  Thus it appears that  Ap42.  Ap40 is  Overall the choice  as  between using either o f the two peptides comes down to the issues o f solubility, specificity, and practicalities o f use. The goal o f this thesis is to examine the effects o f A P signaling in microglia. T o truly look at A p signaling an effort must be made to ensure that responses elicited are specific to the A p peptide. Since A p 4 2 is more insoluble and more likely to form aggregates in solution there is a chance that any result obtained could be clouded by more generalized responses o f microglia in their attempt at clearing the amyloid plaques. Furthermore, the relative insolubility o f A p 4 2 also introduces solvents that are not present in the brain that could confound any results obtained. This could result in a lack o f reproducibility in data obtained using the Ap42 peptide variant. In addition, communications with researchers at Astra Zeneca, in preparation for a possible collaborative project, advised the use o f A p 4 0 based on a greater degree o f reproducibility with its use in their research protocols. Thus, due to the insolubility of A p 4 2 and the possible complications that working with the peptide could present, it was decided that A p 4 0 would be the peptide studied in this thesis work. In summary, A p 4 0 was used because it is physiologically relevant in the study o f A D and has been shown to be a potent activator o f microglia. Furthermore, the relative solubility o f A P 4 0 has the potential to produce more reproducible results then A P 4 2 since there is less chance o f nonspecific effects due to peptide aggregation. The goal o f this thesis Was to characterize the actions o f A p 4 0 over a wide spectrum o f human microglia properties that address both signaling pathways and cellular functions. The specific goals o f this thesis are outlined in four separate aims.  14  1.6.  Study Aims 1.6.1. Aim 1 /Study 1: To determine if acute application of A/340 induces changes in [Ca Ji  in human microglia and to determine the characteristics of the  2+  A/340 induced [C^Ji  response.  The focus o f this Aim is the characterization o f changes in [ C a ] j that occur as a result o f 2+  A(340 mediated  signaling in microglia. A s mentioned previously, changes in [ C a ] i have the 2+  potential to induce microglia activation. Although previous studies have shown that active Ap fragments can cause changes in [ C a ] i in rodent and human microglia [Araujo 2+  Giulian  1996, Korotzer 1995], no  1992, Combs 2000,  previous published work has examined the effects o f  Ap40  on  [ C a ] in human microglia. Moreover, it is important to note that previous work examining 2+  changes in [ C a ] j due to Ap have used very high concentrations o f Ap. 2+  AP40 in the 40-60 u M range  were routinely used [Combs  For example, levels o f  2000, MacDonald 1998] when  other  studies have illustrated that full-length Ap at l p M i s sufficient in altering microglia function [Gulian  1996].  shown that  Furthermore, preliminary work with  Ap42 applied acutely  microglia. Due to this,  Ap42 in the  McLarnon laboratory has also  at l u M is sufficient in inducing a rise in [ C a ] i in human 2+  AP40 at 4 and 10 u M concentrations  were used and specific protocols were  applied in order to characterize the mechanisms underlying the response induced by the peptide. The examination o f the  Ap40 induced calcium response is important because  the  characteristics and properties o f the [ C a ] i response can contribute to understanding the 2+  mechanisms by which Ap signals functional changes in microglia. This work is also important because no published research has yet examined the effect o f  Ap40 on  [ C a ] i levels in human 2+  microglia. Furthermore, deteirnining how microglia respond to Ap will contribute to the  15  understanding of microglia activation and the role of Ap and microglia in inflammatory diseases such as A D .  /. 6.2.  Aim 2 / Study 2: To determine if AfMO treatment of human microglia results in the upregulation of the COX-2 enzyme.  In the first A i m the focus was on initial signaling events caused by human microglia. This Aim's focus is on functional actions of  AP40 application to  Ap40 downstream of  the [Ca ]j 2+  response. A s outlined in the discussion of microglia and disease in section 1.2., C O X - 2 has been implicated in inflammation and has also been shown to be upregulated in A D . These studies suggest that C O X - 2 expression may be one pathway in which an autotoxic inflammatory response, as seen in neurodegenerative diseases such as A D , is perpetuated. Furthermore, although it has been shown that C O X - 2 is upregulated in neurons in A D and AP40 treatment of neuroblastoma [Pasinetti 1998], no study has examined  Ap40 induced C O X - 2  the goal is to determine if human microglia treated with  expression in microglia cells. Thus  Ap40 upregulate  C O X - 2 expression. If a  specific modulator of the Ap40-induced calcium response is identified in the first study, then the effects of that modulator on Ap40-induced C O X - 2 expression will also be studied. In this way, the relationship between the  AP40 calcium signal and C O X - 2  expression would be established.  1.6.3. Aim 3 / Study 3: To determine if incubation of A/340 in human microglia can induce the production of neurotoxic agents. Once again the focus of this A i m is on the functional effects of  Ap40.  A s mentioned  previously, Ap has been shown to induce the production o f neurotoxic substances in rodent and human microglia. However, neurotoxin production induced by  Ap40 in human microglia has not  yet been studied. The purpose of this study is to determine whether concentrations of  16  Ap40 at 4  and 10 u M can induce the production of a neurotoxin in human microglia cultures. If modulators o f the Ap40-induced calcium signal are available, then experiments examining the relationship between the AP40 calcium signal and neurotoxin production will be examined. Once again, the goal is to try to understand the potential mechanisms through which Ap40 activates microglia into a pathological state that contributes to the inflammatory diseases such as A D .  1.6.4. Aim 4 / Study 4: To determine if the treatment of human microglia with A/340 alters [Ca ] signaling pathways. 2+  i  The rationale for this A i m follows preliminary work mentioned in section 1.2. which suggests that human microglia from confirmed cases o f A D respond differently to A T P and P A F when compared to human microglia not diagnosed with A D . The purpose o f this study is to examine whether AP40 can alter microglia responses to A T P and P A F in a manner similar to the altered responses in A D microglia. One o f the reasons for the use o f A T P and P A F is that the calcium responses for both agonists are different. Both stimuli cause the release o f calcium from internal stores in the endoplasmic reticulum (ER), but the A T P response consists primarily o f E R release. O n the other hand P A F induces a small initial depletion o f internal E R stores followed by a sustained influx o f calcium through store-operated calcium channels ( S O C ) [McLarnon 2000, Wang 2000, Wang 1999]. Thus the actions o f AP40 on calcium release from E R stores and influx through the S O C can be studied in human microglia. This is the first study to use methodology o f this nature to examine the effects o f Ap40 treatment on the normal cellular responses o f human microglia. If long-term incubation o f AP40 does produce changes in microglia responses to A T P and P A F it will provide evidence that AP40 can alter normal [Ca ]; signaling pathways in microglia. In addition, such a result would validate 2+  that this in vitro model mimics responses observed in A D microglia.  17  1.6.5. Summary ofAims Although the four aims of this thesis center around different characteristics of microglia, they all address the same general question; what are the possible mechanisms and actions of Ap40 that induces microglia into a state of activation that is potentially pathogenic? Although much work has been done in the area of Ap and microglia, the majority of previous studies have used either rodent microglia or cell lines. Such cells do not represent a model of disease in the human brain as well as human microglia. For example, rodent microglia exhibit properties such as the expression of inducible nitric oxide synthase (NOS), which is a substance not produced in human microglia [Lee 1993]. Furthermore, most of the previous work also used active fragments of AP that are not normally present in the human brain, or they have used concentrations of Ap that were much higher than what is needed to induce responses in human microglia. All of the studies proposed for this thesis will provide new insight in the study of microglia and Ap. The combination of human microglia and the use of the full length Ap40 peptide makes the study of the AP40 induced calcium response in Aim 1, and the expression of a neurotoxin in Aim 3, novel. The examination of Ap40 induced COX-2 expression in human microglia, as outlined in Aim 2, is also unique. Although COX-2 expression has been shown to be upregulated in the AD brain and in Ap treated neurons, no study has yet looked at COX-2 expression in human microglia treated with Ap. Furthermore, the study of Ap40 treatment on microglia responses in Aim 4 is a completely new approach in the study of microglia. No previous laboratory has attempted to examine the integrity of calcium signaling pathways in microglia after AP treatment. The intention of thefirstthree studies is to examine the actions of Ap40 at the initiation of the  [Ca ]i 2+  signal on through to the functional effects that are produced in microglia as a result of  that signal. The first study's purpose is to characterize the Ap40-induced [Ca  18  2+  ]i  response and the  expression of COX-2 and neurotoxic substances examines the effects of AP40 further downstream. The final study examines whether the diminished responsiveness observed in preliminary work with AD microglia can be modeled in human microglia treated with Ap40. All of these experiments are novel and will provide insight into the effects of Ap40 on human microglia. Although there is no single unifying hypothesis in this thesis, this work represents a diverse characterization of the actions of Ap40 on human microglia. The intention of this characterization is to provide the initial framework needed to progress the study of AP40, microglia and disease.  2.  Materials and Methods 2.1.  Preparation o f h u m a n microglia  Human microglia cultures were isolatedfromhuman fetal brain tissue following legalized therapeutic abortions. Approval for this work was grantedfromthe Ethics Committee of the University of British Columbia. Brain tissue was incubated in phosphate buffered saline (PBS) solution containing 0.25 % trypsin and DNase (40 pg/ml) for 30 minutes at 37°C and then dissociated into single cells. The dissociated cells were then cultured in T75 flasks in medium consisting of Dulbecco's modified Eagle's medium (DMEM) containing 5% horse serum, 5mg/ml glucose and 20 pg/ml gentamicin. After 2-4 weeks of growth in flasks,free-floatingmicroglia were harvested and plated on glass coverslips or in culture dishes depending on the protocol to be used. Immunostaining of cells following this procedure was used to verify that the overall purity of the cultures exceeded 98%.  19  2.2.  Study 1; the examination of a A04O induced calcium response in human microglia 2.2.1. Calcium spectrofluorometry  Glass coverslips plated with microglia were incubated with the calcium dye Fura-2 acetoxymethylester (Fura-2 AM) for 25 minutes. Methods for Fura-2 usage are reviewed in Hirst 1999. Fura-2 is a radiometric dye with an excitation spectrum that shifts when it binds to calcium. When bound to calcium maximum Fura-2 fluorescence is observed at a 340 nm wavelength and in calciumfreeconditions optimum fluorescence is emitted at 380 nm. The concentration of free [Ca ]i can then be calculated by using the ratio of bound calcium at 340 nm overfreecalcium at 2+  380 nm (340/380) which is proportional to [Ca ]i levels within the cell. Acetoxymethylester (AM) 2+  is applied in combination with Fura-2 in order to facilitate the loading of Fura-2 into the cell as well as the accurate recording of [Ca ]j levels. The addition of A M on Fura-2 renders the molecule 2+  calcium insensitive and permeable to the cellular membrane. Once inside the cell esterases cleave off the A M group allowing the Fura-2 to bind to calcium and also preventing the molecule from leaving the cell. Thus the addition of A M prevents Fura-2frombinding and fluorescing calcium in the extracellular solution as well as keeping the dye inside the cell. Fura-2 A M was applied with pluronic acid at equal concentrations (1 pM) in a normal physiological saline solution (Ca-PSS) at room temperature in order to facilitate solubilization of the Fura-2 AM. After loading the dye the coverslip was transferred to a dye-free solution of PSS to wash out excess Fura-2 for 5-7 min. The coverslip was then mounted on a Zeiss Axiovert inverted microscope with a x40 quartz objective lens. Alternating wavelengths of 340 and 380 nm of UV light was applied at 8 second intervals and fluorescence signals were obtained at an emission wavelength of 510 nm Fluorescent emissions were captured by a digital camera (DVC-1310, DVC Co. Austin, TX) and were  20  processed using an Empix imaging software package by recording and calculating the 340/380 UV ratio which is used as a quantitative measure of [Ca ], levels. Table 1 shows 340/380 UV ratios 2+  for calcium and the approximate corresponding [Ca ]jthat is represented by the specific ratio 2+  value.  Table  1:  Approximate conversions of  F 3 4 0 / F380  ratios to [Ca  ]i  F340 / F o  [Ca^MriM)  0.8  550  0.6  365  0.4  200  0.2  45  38  2.2.2. Solutions and chemicals Human Ap 1-40 was purchasedfromCalifornia Peptides (Napa, CA) and prepared in a 1% ammonium hydroxide (NH4OH) solution as recommended by Astra Zeneca or solublized in sterile distilled water or Ca-PSS in a stock solution of ImM. Application of the peptide followed closely after its preparation, however for short-term storage the stock solution was kept at 4°C. The physiological calcium-containing saline solution (Ca-PSS) contained (in mM): NaCl (126), KC1 (5), MgCl (1.2), CaCl (1), D-glucose (10) HEPES (10), and was set to pH 7.4. In calcium-free 2  2  physiological saline solution (Ca-free PSS), 1 mM of EGTA was used instead of CaCk, otherwise the ingredients and concentrations matched the Ca-PSS solution. A low chloride solution was also applied in some experiments and it had two changesfromthe regular Ca-PSS mixture; the NaCl was replaced with Na-gluconate (126 mM), and KSO4 and MgSC»4 replaced the Cl" salt so that the 21  total Cl' concentration was at 2 mM. SKF963651 (SKF) was solubilized in double distilled H2O and stored at a stock concentration of 50 mM. Fura 2 A M was purchasedfromMolecular Probes (Eugene. OR) and ionomycion was obtainedfromSigma (St. Louis, MO). Experiments were performed by applying A(340 directly to the coverslips at the concentration of 4 or 10 uM. Two general types of experiments were performed, Ap40 in Ca-free PSS or in normal Ca-PSS solution. Experiments in Ca-free PSS were done to determine the contribution of internal stores on the AP40 induced calcium response. Ca-free PSS, low chloride solution, and SKF (50 uM) were applied directly in other experiments in order to determine the characteristics of the calcium response induced by AP40. 2.3.  Study 2; the examination Ap40 treatment of human microglia on C O X - 2 expression  AP40 was solubilized in sterile distilled water for all the experiments in this study and the microglia cells were isolated and prepared the same as in Study 1. The basic RT-PCR procedures used in this study are described previously by Nagai (2001). RT-PCR was performed with two different sense and antisense olignucleotide primers of COX-2 Primer 1:  sense; 5 '-TTC-AAA-TGA-GAT-TGT-GGG-AAA-ATT-GCT-3' antisense; 5'-AGA-TCA-TCT-CTG-CCT-GAG-TAT-CTT-3'; 304 base pair (bp) product.  Primer2:  sense; 5'-CAC-AAT-GTG-GCT-GAG-GGA-ACA-CAA-CA-3' antisense; 5'-GAC-TGG-TAT-TTC-ATC-TGC-CTG-CTC-TGG-3'; 489 bp product.  Ap40 at 4 and 10 uM was incubated in microglia cultures for ~24 hrs. Total RNA was extracted using TRIzol reagent (GIBCO-BRL, Gaithersburg, MD). Complimentary cDNA templatesfromeach sample was preparedfrom2 pg of total RNA primed with random hexamers  22  (Pharmacia, Gaithersburg, MD) using 200 units of MMLV reverse transcriptase (GIBCO-BRL) followed by 40 PCR amplification cycles (94°C for 30 sec, annealing at 55°C for 60 sec, and extension at 72°C for 90 sec). Gtyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a reaction standard [Ercolani 1988]. Ten pi of each PCR product was analyzed by 1.5% agarose gel electrophoresis. Authentic bands were determined by selective enzyme digestion. 2.4.  Study 3; the examination of A04O induced neurotoxic potential in human microglia 2.4.1. Reagents  The following reagents were used in this study and were purchasedfromSigma (St. Louis, MO): bacterial lipopolysaccharide (LPS,fromEscherichia coli 055.B5); diaphorase (EC, from Clostridium kluyveri, 5.8 U/mg solid); p-iodonitrotetrazolium violet; nicotinamide adenine dinucleotide (NAD+); MTT [3-(4,5-dimethylthiazol2-gl)-2,5-diphenyl tetrazolium bromide]. Human recombinant interferon-g (IFN-y) was purchasedfromBachem (Torrance, CA), and both calcein/acetoxy methylester (AM), and ethidium homodimer (EthD-1) were purchased from Molecular Probes (Eugene, OR). 2.4.2. Cell Culture The human microglia cells were obtained and prepared the same as in the previous two studies. Instead of being plated on glass coverslips the microglia were seeded in 12-well plates at the concentration of 6-8 x 10 cells per well in 0.8 ml of Dulbecco's modified Eagle's medium 4  nutrient mixture F12 ham (DMEM-F12) containing 5% fetal bovine serum (FBS). Undifferentiated human neuroblastoma SH-SY5Y were also plated in 12-well plates at the concentration of 2 x 10 /ml in 0.5 ml of DMEM-F12, with 5% FBS. 5  23  2.4.3. Neurotoxicity of microglia supernatants The procedure is similar to those outlined in Klegeris (2000). A|540 at 4 or 10 uM was solubilized in sterile distilled water was added to the plated microglia cultures and the wells were then incubated for 24 hrs in a humidified 5% CO2, 95% air atmosphere at 37°C. For a negative control, plastic wells with Ap40 but no microglia were used as well as microglia containing wells without any Ap40 stimulus. A positive control stimulus of LPS (0.5 pg/ml) & IFN-y (333U/ml) on microglia cultures was also used since this mixture has been used as a positive control before in neuroblastoma. All control wells were incubated for the same amount of time under the same conditions as the treated microglia wells. At the end of the 24 hrs incubation period the supernatant from the SH-SY5Y neuroblastoma wells was removed and then replaced with 0.5 ml of supernatant from the treated microglia and control wells. The neuroblastoma were then incubated for 72 hrs and the supernatant was analyzed for lactate dehydrogenase (LDH) as a measure of cell death and the MTT cell viability assay for formazan production was also performed. 2.4.4. Cell death and viability assays The level of neuronal cell death was determined by measuring the release of LDH from dead neuroblastoma cells. The enzymatic assay for measuring LDH activity is described by Decker (1988) and Klegeris (2000). After the 72 hrs incubation of the neuroblastoma with supernatantfromthe treated microglia cultures 100 pi of supernatant was transferred to another set of wells with the addition of 15 pi of lactate solution (36mg/ml in phosphate-buffered saline (PBS)) and 15 pi of p-iodonitrotetrazolium violet solution (2 mg/ml in PBS). A reaction was initiated by 15 ul of NAD+/disaphorase solution (3mg/ml NAD+, 2.3 mg solid/ml diaphorase). After a 15 min incubation period the reaction was halted with the addition of 15 pi oxamate (16.6 mg/ml). Optical densities were measured by a Model 450 microplate reader (Bio-Rad Laboratories, Richmond, CA)  24  with a 490-nm filter. The amount of LDH released was expressed as a percentage of the value obtained in comparison to wells where 100% of the cells were lysed with 1% Triton X-100. Cell viability was determined by the MTT assay described by Mosmann (1983) and Klegeris (2000). This assay is based on the ability of viable cells to convert tetraxolium salt (MTT) to colored formazan. The viability of the neuroblastoma cells was determined by adding 1 mg/ml of MTT to the cell cultures. After a 2 hrs incubation the formazan was extracted by adding an equal volume of extraction buffer (20% sodium dodecyl sulfate and 50% N,N-dimethyl formamide, pH 4.7). The well plates were then incubated for 16-20 hrs at 37°C. The amount of formazan formed was determined by measuring 100 pi aliquots at 570 nm with the plate reader. A percentage value was obtained by comparing treated cells to cells incubated infreshmedium only. 2.5.  Study 4; the examination of A04O treatment on human microglia responses to ATP and PAF  Ap40 was solubilized in sterile double distilled water for all the experiments in this study. All other procedures and solutions used for the cell culture and calcium spectrofluorometry were identical to the procedures outlined in thefirststudy. AP40 was incubated with plated coverslips of human microglia in DMEM at 37°C for 1 or 2-day treatment (~20 or 48 hours). After the incubation period spectrofluorescence of the [Ca ]j was examined in these cells in response to 2+  ATP and PAF. ATP was solubilized in distilled water at a stock concentration of 10 mM and applied at a concentration of 100 uM. PAF was also solubilized in distilled water at a stock of 100 pM and was applied at a concentration of 100 nM. Responsesfromthe treatment group were compared to responsesfromuntreated cells that were tested on the same day.  25  3.  Results 3.1.  Study 1; the examination of a A(340 induced calcium response in human microglia  Initial experiments investigated the effects of acute application of Ap40 (4 and 10 uM) on [Ca ]i levels in cultured human microglia. Control experiments were performed where [Ca ]j 2+  2+  levels were recorded without the application of  AP40, a representative  experiment is shown in  Figure 2A. The trace shown is of a mean of all the [Ca ]i responses (mean of 22 cells), note the 2+  lack of any noticeable change in [Ca ]j for up to 10 minutes. In addition, the application of fresh 2+  Ca-PSS at approximately 300 sec did not cause any observable changes in [Ca ]j. AP studies 2+  employed the application of AP40 in calcium containing physiological saline solution (Ca-PSS). A representative experiment is shown in Figure 2B which illustrates that Ap40 application at 10 pM cases a slow progressive linear rise in [Ca ]j (mean of 26 cells). The rise in [Ca ]j was not 2+  2+  immediate but occurred within two minutes after Ap40 application. A plateau level of [Ca ]i was 2+  reached after approximately 8 min. of treatment and did not change following the removal of Ap (data not shown). A similar rise in [Ca ]i due to Ap40 application was replicated for both 4 and 10 2+  uM concentrations in a total of 14 experiments (see Table 2). The change in [Ca ]i before AP40 2+  application was compared to the change in [Ca ]j with Ap40 application in all experiments with a 2+  standard f-Test and the results were highly significant (see Table 2). In order to establish whether the [Ca ]i induced response to AP40 is mediated by external 2+  or internal sources similar experiments were carried out in calcium-free physiological saline solution (Ca-free PSS). If a component of the [Ca ]j response is due to internal stores then Ap40 2+  application in Ca-free PSS will still illicit a response, on the other hand, if the [Ca ]i rise is 2+  26  primarily due to the influx of calcium from the extracellular solution then no rise in [Ca ]j would 2+  be seen. A representative control experiment where [Ca ]j levels were recorded in Ca-free PSS 2+  without Ap40 is shown in Figure 3 A (mean of 17 cells). No evident changes in [Ca ]; occurred 2+  over the course of the experiment (10 min). A typical response of A[540 application in Ca-free PSS is shown in Figure 3B (mean of 22 cells), and shows that no significant rise in [Ca ]i was observed 2+  with the application of AP40 in Ca-free PSS. The introduction of Ca-free PSS caused a slight reduction in basal [Ca ]j and when calcium was reintroduced into the extracellular solution a rise 2+  in [Ca ], was observed. A lowering of basal [Ca ]i in response to the removal of extracellular 2+  2+  calcium is commonly observed in these microglia cultures thus it is unlikely that the rise in [Ca ]i 2+  following the re-introduction of Ca-PSS is due to AP40 treatment. The lack of a AP40 [Ca ]j 2+  response in Ca-free PSS for both Ap40 concentrations (4 & 10 uM) was replicated in a total of 9 separate experiments (Table 2). Once again the change in [Ca ]i before AP40 application in Ca2+  free PSS was compared to the change in [Ca ]i with AP40 application in all experiments using a 2+  standard t-Test (see Table 2). The resultsfromthese experiments indicate that Ap40 applied in calcium-free solution does not cause a significant increase in [Ca ]i, which suggests that Ap40 2+  induces an influx of calciumfromthe extracellular solution with no depletion of internal stores of calcium. A series of experiments were next carried out using Ca-free PSS solution applied following the response induced by AP40 in Ca-PSS. These studies were used to investigate if introduction of Ca-free PSS subsequent to the Ap induced [Ca ]irise inhibited the increase in [Ca ]j. If calcium 2+  2+  influx was the cause of the [Ca ]j response then removal of calciumfromthe external solution at 2+  the height of the [Ca ], rise should eliminate the influx of calcium and the [Ca ]j concentration 2+  2+  should drop back down to basal levels. To test this hypothesis AP40 was applied in Ca-PSS and a  27  rise in [Ca ]j was induced (Figure 4 A). Upon the removal of extracellular calcium, near the peak 2+  of the response, [Ca ]j dropped quickly, however the calcium influx was not fully inhibited. A 2+  second application of Ca-free PSS did inhibit the Ap40-induced influx even further, but not all the way to initial basal [Ca ]j levels, suggesting a small internal stores component to the Ap40 2+  calcium response (mean of 9 cells). A second example of this protocol is presented in Figure 4B (mean of 26 cells) where the application of Ca-free PSS at the height of the Ap40 [Ca ]j response 2+  caused a sharp drop in [Ca ]i, yet once again the [Ca ]i rise was not completely inhibited. The re2+  2+  introduction of Ca-PSS after Ca-free PSS reversed the fall in [Ca ]i (Figure 4B). Inhibition of the 2+  Ap40-calcium response was also observed when A P 4 0 was maintained during the application of Ca-free PSS (Figure 4C, mean of 23 cells). Both of the experiments shown in Figure 4A & B suggest that while the main component of the Ap40-induced [Ca ] response in microglia is due to calcium influx, there may still be a small 2+  f  contribution of internal calcium stores since the removal of calciumfromthe extracellular solution did not fully inhibit the [Ca ]i rise. Inhibition of the [Ca ] j rise induced by Ap40 with the removal 2+  2+  of extracellular calcium was replicated in a total of 11 experiments (Table 2). However, in only 3 of the 11 experiments was the Ap40-induced rise in [Ca ]j inhibited fully. These experiments 2+  suggest that although calcium influx constitutes the major component of the calcium response of human microglia to A P 4 0 , there also appears to be evidence that A P 4 0 causes limited mobilization of calciumfromintracellular stores. In order to better elucidate the nature of the influx pathway activated by Ap40 a series of experiments were performed that attempted to modulate this pathway. A high conductance anion channel has previously been described in human microglia that is activated with both depolarizing and hyperpolarizing stepsfromholding potential, suggesting a role in the maintenance of cell  28  potential [McLarnon 1997]. More recent work has shown that a reduction in the expression of these anion channels with a low chloride physiological saline solution (low Cl" PSS) inhibits calcium influx through store-operated calcium channels (SOC) in human microglia [McLarnon 2000]. It was further suggested that the inhibition of SOC was secondary to cell depolarization induced by a decrease in the expression of Cl" channels [McLarnon 2000]. Furthermore, although detailed properties of the anion channels are unknown, the data suggest that these ion channels are active under basal conditions. As shown in Figure 5 (mean of 32 cells) low Cl" PSS was applied at the peak of the Ap40-mediate [Ca ]i response. The application of low chloride solution caused an 2+  immediate and sharp drop in [Ca ]i that approached baseline levels. The inhibition of the AP402+  mediated [Ca ]j response was replicated in a total of 5 separate experiments (Table 2). This result 2+  suggests that the influx pathway activated by AP40 is sensitive to changes in membrane potential modulated by altering the activity of anion channels in microglia. The major calcium influx pathway in microglia is the SOC (store-operated or capacitative) pathway that is sensitive to anion channel modulation [Wang 2000]. Since Ap40-mediated influx is also sensitive to anion channel modulation it is possible that the Ap40 effect is mediated through the SOC pathway. However, there was no change in [Ca ]i with Ap40 application in Ca-free PSS 2+  (Figure 3B), indicating that there was no depletion of intracellular stores, thus no SOC activation would be possible. On the other hand, experiments which attempted to block the [Ca ]i rise with 2+  the removal of extracellular calcium do suggest that a small amount of internal release of calcium is occurring which may come from ER stores. Thus the ability of Ap40 to deplete ER stores in order to activate the SOC pathway remains uncertain. However, the SOC is the major calcium influx pathway in microglia because microglia, like other unexcitable tissue, do not express voltage-gated calcium channels [Eder 1998, McLarnon 1997]. Therefore Ap40 could be activating the SOC  29  pathway directly without signaling depletion from ER stores or a slow ER depletion may be causing SOC activation. A pharmacological maneuver was next used to investigate the dependence of the  Ap40-  induced rise in [Ca ], on the SOC pathway. SKF963651 (SKF), a known inhibitor of the SOC 2+  pathways [Li 1999], was applied at the peak of the Ap40-induced [Ca ]j rise. If AP40 was 2+  activating SOC pathways directly then SKF application should inhibit calcium influx and the [Ca ]i level should drop. On the other hand, if AP40 is causing an influx of calcium through some 2+  other pathway, SKF application should have little or no effect on [Ca ]i levels. Figure 6A 2+  illustrates that SKF does inhibit the SOC pathway in microglia since it causes a sharp drop in [Ca ]i levels when applied to the SOC-dependent plateau phase of a [Ca ]j response to platelet 2+  2+  activating factor (PAF) [Wang 2000] (mean of 10 cells).  Figure 6B illustrates a representative  experiment and shows that SKF application had no effect on the Ap40-induced influx of calcium (mean of 12 cells). The inability of SKF to modulate the Ap40-induced calcium influx was replicated in 3 separate experiments (Table 2). These experiments suggest that Ap40 is causing an influx of calcium that is independent of the store-operated influx pathway. The goal of this study was to characterize the [Ca ]i response initiated by Ap40 in human 2+  microglia. Initial experiments with AP40 in Ca-PSS and Ca-free PSS indicated that AP40 induced a rise in [Ca ] due to calcium influxfromthe extracellular solution. Further experiments then 2+  showed that the Ap40 induced calcium influx was blocked by the removal of calciumfromthe extracellular solution and the depolarization of the cell membrane with low Cl- PSS. However, complete inhibition of the Ap40-induced rise in [Ca ]i did not occur with Ca-free PSS, which 2+  suggests that Ap40 may also cause a small release of calciumfrominternal stores. At this point the Ap40-induced calcium response appeared to be mainly due to calcium influx and have the  30  properties of the SOC influx pathway. Thus it was necessary to determine the possible role of the SOC pathway in the AP40 induced [Ca ]i response. The SOC inhibitor SKF was applied to the 2+  Ap40 induced influx and had no observable effect. Therefore the results of this study indicate that AP40 induces a rise in [Ca ]j that is mainly due to an influx pathway in human microglia that is 2+  not mediated by the SOC pathway. The ability to block the mechanism by which Ap40 alters [Ca ]j would allow for the 2+  modulation of cell functions, such as the secretion of inflammatory factors, that contribute to the role of microglia in pathologies such as AD. At present, however, no pharmacological agent has been identified which would specifically inhibit the Ap40-induced calcium pathway. The use of low Cl" PSS would not be feasible as a modifier since this procedure would result in sustained cell depolarization that would presumably alter many cellular functions. Nevertheless, this work has identified a putative target for pharmacological manipulation in order to modulate the actions of AP40 that are mediated through the mobilization of [Ca ]j. Further studies are needed to test for 2+  agents with the capacity to block the Ap40-induced [Ca ]i response and thus the downstream 2+  actions of Ap40 as well. The next study addressed a cellular function that is modulated as a result of AP40 treatment of human microglia. The specific function investigated was the expression of COX-2 in microglia, an enzyme that is over expressed in AD.  31  Figure 2: The effects of Ap40 on [Ca ] in human microglia with Ca-PSS. Graphs are f  representative experiments showing the mean of the responses measured. The error bars presented illustrate the SEM at points along the trace. (A) [Ca ]i activity recorded in human microglia cells 2+  in Ca-PSS solution with no Ap40 present (22 cells). Note that no rise or drop in [Ca ]i is 2+  observed. A Ca-PSS solution change was also applied at approximately 300 sec with no apparent effect on [Ca ] . (B) A rise in [Ca ] induced by Ap40 in Ca-PSS (26 cells). Ap40 application 2+  2+  f  f  (10 uM) induces a rise in [Ca ]j within 2 minutes that increases in a linear manner. Analysis of 2+  [Ca ]i levels with a standard /-Test indicates that the rise is significant ( ***, P < 0.0001) when 2+  compared to initial baseline levels. Washout of the Ap40fromthe extracellular solution does not slow or diminish the rise induced by Ap40.  32  Figure 2 A  0.4 0.35  0.3 o co CO  -  0.25  0.2  0.15  0.1 100  200  300 Time (sec)  Figure 2 B 0.5  33  400  500  600  Table 2: Study 1; Ap40 induced calcium responses in human microglia concentration  n value  total # of cells  P value  (uM)  A|340 induced calcium  4  9  157  < 0.0001  rise in Ca-PSS  10  5  72  < 0.0001  No A04O induced calcium  4  6  129  0.0672  rise in Ca-free PSS  10  3  70  Ca-free PSS effects  4  5  68  on AP40 induced influx  10  3  38  Low chloride effects  4  5  95  10  3  14  on Ap40 induced influx SKF effects on AP40 induced influx  34  Figure 3; The effects of Ap40 on [Ca ]; in human microglia with Ca-free PSS. Graphs are representative experiments showing the mean of the responses measured. The error bars presented illustrate the SEM at points along the trace. A standard f-Test was used in all statistical analysis. (A) [Ca ]i activity recorded in human microglia cells in Ca-free PSS solution with no AP40 2+  present (17 cells). Note that no rise or drop in [Ca ]j is observed. A Ca-free PSS solution change 2+  was also applied at approximately 480 sec with no apparent effect on [Ca ]i. (B) Application of 2+  AP40 (4 uM) in Ca-free PSS resulted in no significant increase in [Ca ]j levels (22 cells). Note 2+  that the drop in basal [Ca ]; seen with the application of Ca-free PSS is commonly observed in 2+  these cells. Thus the rise in [Ca ]i upon re-introduction of extracellular calcium should not be 2+  construed as the actions of Ap40.  35  Figure 3A 0.4 Ca-free P S S  Ca-PSS 0.35  0.3  J  0.25  i  0.2  0.15  0.1 600  500  400  300  200  100  Time (sec)  Figure 3B 0.5 Ca-PSS  Ca-free P S S  Ca-PSS i  0.45 Ap40 ( 4 M ) M  0.4  1 0.35 0.3  0.25  0.2 100  200  300  400 Time (sec)  36  500  600  700  Figure  4: Removal of extracellular calcium inhibits A(340 induced [Ca ]i responses. Graphs are 2+  representative experiments showing the mean of the responses measured. The error bars presented illustrate the SEM at points along the trace. A standard f-Test was used in all statistical analysis. (A) AP40 caused a significant rise in [Ca ]j (**, P = 0.0027). Removal of extracellular calcium 2+  after the Ap40 (4uM) induced [Ca ]j response resulted in a quick drop in [Ca ]; (9 cells). A 2+  2+  second application of Ca-free PSS caused a further drop in [Ca ]i, however full inhibition of the 2+  [Ca ]i response was not obtained. (B) Ap40 caused a significant rise in [Ca ]i (***, P = 0.0001). 2+  2+  Removal of extracellular calcium after the AP40 (lOpM) induced [Ca ]j response resulted in a 2+  sharp drop in [Ca ]; (26 cells). Once again full inhibition of the Ap40-induced [Ca ]i response 2+  2+  did not occur. Replacement of extracellular calcium resulted in a reversal of the [Ca ]i drop 2+  suggesting that the calcium influx pathway remains activated. PAF application (lOOnM) produced a robust [Ca ], response illustrating that the microglia remain viable. (C) In this experiment Ap40 2+  (4pM) was maintained with the application of Ca-free PSS (23 cells). Ap40 caused a significant rise in [Ca ]j (***, P = 0.0001). Even in the presence of Ap40, Ca-free PSS inhibited the influx of 2+  calcium into the cell.  37  Figure 4A 0.5  Ca-free PSS  Ca-PSS  Ca-free PSS  Ca-PSS  AP40 (4uM)  0.45  0.4  Vi  -  **  0.35 3  u. 0.3  0.25  200  100  400  300  Time (sec)  100  200  300  400  500 Time (sec)  38  500  600  700  800  Figure 4 C 0.4 .  Ca-PSS  i  39  Ca-free PSS  Figure 5: Ap40 (4uM) induced calcium influx in Ca-PSS followed by low chloride PSS (32 cells). The graph shown is a representative experiment showing the mean of the responses measured. The error bars presented illustrate the SEM at points along the trace. A standard t-Test was used in all statistical analysis. AP40 caused a significant rise in [Ca ]j (***, P = 0.0001). Low chloride PSS 2+  inhibited Ap40 induced influx in a manner similar to Ca-free PSS.  40  Figure 6: The effects of SKF on Ap40-induced influx. The graphs shown are representative experiments showing the mean of the responses measured. The error bars presented illustrate the SEM at points along the trace. A standard f-Test was used in all statistical analysis. (A) This experiment illustrates that SKF (50 uM) does inhibit the SOC pathway in microglia since the SOCinflux phase of the PAF response is inhibited by SKF application (10 cells). (B) Ap40 caused a significant rise in [Ca ] (**, P = 0.0037). Ap40-induced (lOuM) calciuminflux in Ca-PSS was 2+  f  not inhibited by the SOC inhibitor SKF (50uM) (16 cells).  41  Figure 6 A 0.6 i  Ca-free PSS  Ca-PSS  SKF  PAF  0.5  S  0.4  <•> Ik O  u-  0.3  0.2  0.1 100  200  300  500  400  600  Time (sec)  Figure 6B  Ca-PSS  0.5  AP40 (10uM)  0.45  S  Ca-free PSS  SKF  0.4  n  U.  -  0.35 0.3  0.25  0.2 100  200  300 Time (sec)  42  400  500  600  3.2.  Study 2; the examination Ap40 treatment of human microglia on C O X - 2 expression  The goal of this study was to determine if AP40 treatment of human microglia can modulate cellular functions downstream of the initial Ap40-induced increase in [Ca ];; the functional 2+  process examined was COX-2 expression. The initial hope of this study was that a method for modulating the Ap40-induced [Ca ]i response would be found that could be used as a maneuver 2+  for altering the functional actions of Ap40, such as COX-2 expression or the production of neurotoxic substances studied later on. However, as the earlier section indicated, no modulatory agent for the  [Ca ]i 2+  response mediated by Ap40 was found. Therefore, the action of Ap40 on  COX-2 expression was examined without any modulation of the Ap40 signal. In order to address the concern that AP40 treatment over 24 hrs diminished the viability of cultured human microglia, experiments were first performed that were similar to those outlined in the fourth study where the responsiveness of the microglia was assessed with ATP and PAF after treatment with Ap40. It was determined that after 24 hrs treatment with AP40 (4 & 10 uM) microglia still retained their responsiveness and showed no indication that their viability was compromised (data not shown). The expression of COX-2 RNA in human microglia following Ap40 treatment was then characterized by RT-PCR. In thefirstof the three experiments (Figure 7A) a large increase in COX-2 expression was seen for both doses of Ap40 (4 & 10 uM) and a dose response is evident. Primer2 was used for this PCR amplification. The results of the second experiment are shown in Figure 7B. In the top gel, low expression of COX-2 is observed even in control where no Ap40 was added. Ap40 treatment of the microglia at 4 and 10 u.M caused a dose dependent increase in COX-2 expression. Replication of the PCR  43  amplification produced an observable increase in COX-2 expression only at the high dose of AP40 (lOuM). The difference in expression and molecular weight of both PCR products can be attributed to different primers that were used. In the top band Primer 1 was used and Primer2 was used for the bottom gel (see section 2.3. in Material and Methods). The resultsfromthe third experiment (Figure 7C) also show an increase in COX-2 expression with Ap40 treatment and the use of Primer 1. No band was observable in both control or with 4uM of Ap40 treatment; however, AP40 treatment at 10 pM did produce faint bands in both PCR amplifications. GAPDH bands demonstrate that the amount of mRNA for each group was equal since GAPDH is a constitutively expressed enzyme and acts as a positive control. Datafromthree separate experiments show that Ap40 treatment of human microglia over a 24 hrs period enhances COX-2 expression. Although these results are non-quantitative, a clear increase in COX-2 expression is observable with Ap40 treatment. When taken together with the findings of thefirststudy which focused on [Ca  2+  ]i  signaling of Ap40 it is clear that the levels of  Ap40 found to cause a [Ca ]j response in human microglia are also capable enhancing COX-2 2+  expression. Thus the resultsfromthesefirsttwo studies suggest that the [Ca ]; response initiated 2+  by Ap40 could signal the enhanced expression of COX-2. Further work should examine this possible connection and investigate whether modulators of the Ap40-induced [Ca  2+  ]i  response can  also modulate COX-2 expression. The third study investigated Ap40 actions on another pathogenic action of microglia, the production of neurotoxic substances.  44  Figure 7: The effects of AP40 on COX-2 expression in human microglia. The gels shown in each section are PCR amplifications of cDNAfroma single experiment where human microglia was exposed to AP40 at 0, 4, and 10 uM. The expression of GAPDH, a constituitively expressed enzyme, was used as a positive control to ensure equal levels of mRNA in all of the wells. Molecular markers were also routinely run on the gels to determine the approximate molecular weight of the PCR products. The brightest marker for all gels corresponds to the molecular weight of 600 bp. (A) Primer2 was used for this experiment (see section 2.3.). Increased expression of COX-2 due to Ap40 treatment is observable with both 4 and 10 uM concentrations and no basal COX-2 expression was observed in control.  ( B ) In thefirstPCR run primerl was used and a  small amount of basal COX-2 expression is observable and Ap40 treatment resulted in a dosedependant increase in COX-2 expression. However, the second PCR amplification with primer2 produced no observable bands in control or 4 uM, faint COX-2 expression was only visible with AP40 treatment at 10 pM. The difference in molecular weights seen between both PCR runs is because of the different COX-2 primers used. Primer2 produces slightly larger COX-2 PCR products of 489 bp. (C) Primerl was used for both PCR runs. Both control and Ap40 treatment at 4 uM caused no observable COX-2 expression while AP40 at 10 uM caused visible expression of COX-2.  45  Figure 7 A AP40 (uM)  Figure 7 C  hp  AJ340 (uM) 0 4 10  GAPDH  5004N>  300 -  bp  0  Ap40(uM) 4 10  46  GAPDH  3.3.  Study 3; the examination of AF340 induced neurotoxic potential in human microglia  The objective of this study was to determine whether A[340 treatment of human microglia could alter cellular functions and induce the production of neurotoxic substances. AP40 (4 and 10 pM) was incubated for 24 hrs with or without microglia cells, after which the media was transferred to neuroblastoma cells that were incubated for an additional 72 hrs. Ap40 incubated without the presence of microglia served as a control for any possible neurotoxicity of the AP40 peptide alone. The viability of the neuroblastoma cultures after incubation was determined using the MTT assay and cell death was determined with the LDH assay as outlined in the materials and methods section. According to the LDH assay, supernatant from Ap40-treated microglia caused very low levels of cell death, less than 10% (Figure 8 A). This result was not significantly different from control conditions where the supernatantfromuntreated microglia was used. A further comparison of supernatantfromAp40-treated microglia with Ap40 application directly to neuroblastoma cultures did not yield any significant differences. Analysis of cell viability with the MTT assay yielded some unexpected results. Supernatant from Ap40 treated microglia did decrease cell viability of neuroblastoma cultures when compared to untreated microglia; however, direct application of Ap40 to neuroblastoma cultures also decreased cell viability (Figure 8B). A comparison of the level of cell viability between Ap40 treated and microglia & Ap40 supernatant treated neuroblastoma did not show any significant differences. The degree in the reduction of cell viability with Ap40 treatment alone is somewhat surprising because previous work with solubilized Ap40 in this neuroblastoma cell line did not 47  cause a decrease in cell viability at concentrations twice as high as what was used in the present work [Lambert 1994]. However, aggregated forms of Ap have been shown to cause degeneration of neuroblastoma cells [Cedazo-Minguez 2001] and it is possible that some of the Ap40 aggregated into more potentially neurotoxic forms during the course of treatment in these cells. Regardless, the initial hypothesis was that Ap40 treatment of microglia would cause a greater degree of neurotoxicity than Ap40 alone since other studies have shown this to be the case with Ap treatment of rodent microglia and monocyte cell lines [Combs 1999, Giulian 1996]. If microglia do in fact produce neurotoxic agents, as numerous studies would suggest, then the most obvious recourse is to activate microglia to the point where their neurotoxin production surpasses the direct neurotoxic action of Ap40 in neuroblastoma cells. The two most obvious changes in the current protocol that could accomplish this would be to increase the density of microglia per treatment well and/or increase the length of Ap40 incubation in the microglia cultures. Because both Ap40 treated microglia and Ap40 directly both cause decreases in neuroblastoma viability at levels that are not significantly differentfromeach other, no conclusion can be made at this time as to whether or not AP40 treated microglia produce neurotoxic substances.  48  Figure 8: The neurotoxic effects of supernatant from Ap40 treated microglia or AP40 alone on neuroblastoma. A standard t-Test was used in all statistical analysis. (A) The level of cell death in neuroblastoma cells as determined by the LDH assay is shown. No significant differences were calculated between any of the treatment groups. (B) The level of cell viability in neuroblastoma as determined by the MTT assay is shown. Significance was only found between AP40 free and AP40 treated groups. Within the AP40 treated microglia group both 10 uM (*, P = 0.0399) and 4 uM (**, P = 0.0043) treatment significantly decreased neuroblastoma viability. With AP40 treatment alone on neuroblastoma significant reduction was seen at 10 uM (#, P = 0.0343). Ap40 treated microglia did not exhibit any significant neurotoxic effect greater than Ap40 treatment alone on neuroblastoma cultures.  49  Figure 8A  LDH  50-,  40-  © Q  o O vP  304.  20  10-]  A(J40 (10uM)  Afi40 (4uM)  no Ap40  50  AP40 (10uM) & uglia  Ap40 (4uM) & uglia  \ no Ap40 & uglia  3.4.  Study 4; the examination of Ap40 treatment on human microglia responses to ATP and PAF  Thefirstthree studies were designed to characterize [Ca  2+  ]i  signaling pathways induced by  Ap40 in human microglia and to determine Ap40 actions on the cellular functions of microglia. The specific functions examined were the expression of COX-2 and neurotoxic substances. The objective of this final study is not directly related to the previous three and addresses whether treatment of human microglia with Ap40 alters calcium-mediated signaling pathways. As outlined in section 1.2., preliminary data from this laboratory suggests that microglia responses to the agonists ATP and PAF are attenuated in adult human microglia from confirmed cases of AD relative to adult microglia from non-AD subjects. In this study, AP40 was incubated for either a one or two-days in cultures of human microglia. Following treatment, responses to ATP and PAF were recorded using calcium-imaging procedures similar to those described in study 1. A number of different measurements were recorded and responses were compared between treated (Ap40 at 10 uM) and untreated microglia. The variables that were studied included basal levels of calcium, amplitude of agonist response, and the sustained influx component of agonist response. Figure 9A and B illustrates normal ATP and PAF responses and shows how the characteristics of the responses were analyzed. As noted, the ATP response is primarily due to the mobilization of calciumfromER stores. The PAF response also has an initial ER store component followed by an additional sustained influx of calcium mediated by the SOC pathway. There were a total of 4 control experiments to 8 treated for each incubation time point. A standard student f-test was then used between treated and control groups to determine the level of significance in all analysis.  51  Significant differences in basal [Ca ]j were seen in both the one and two-day treated 2+  microglia as compared to control (Figure 10). Differences in basal [Ca ]j were determined by 2+  pooling the 340/380 calcium ratios at time zero for all the control and treated cells for each treatment group. This result is consistent with other studies that show elevated [Ca ]j levels in 2+  rodent and human microglia after incubation with activefragmentsof Ap [Korotzer 1995, Silei 1999]. Amplitudes of PAF responses following Ap40 treatment were significantly diminished for both one and two-day treatments (Figure 11). The influx component of the PAF response due to the SOC pathway was also examined in Ap40 treated microglia by comparing the [Ca ]j level 1 2+  min after the initial PAF peak. A significant decrease in the SOC mediated [Ca ]i plateau of PAF 2+  was seen for both treatment groups (Figure 12). These results show that AP40 treatment may be altering normal calcium-mediated signaling in response to PAF in human microglia. Furthermore, this data also suggests that the actions of AP40 in human fetal microglia can model the functional responses found in microgliafromAD brain. An analysis of the ATP responses resulted in no significant differences in the ATP amplitude or level of influx between the treated and control groups. Both PAF and ATP utilize the same internal calcium stores in their initial response peaks, however the signaling pathways for both agonists are not identical since they utilize different receptors and second messenger systems [McLarnon 2000]. Therefore, Ap40 could be only acting on the PAF signaling pathways in the alteration of [Ca ]j signaling. 2+  The purpose of this study was to examine if Ap40 treatment in cultured human microglia mimics conditions found in vivofromAD patients where calcium-mediated cellular responses to the common agonists ATP and PAF were altered. Significant elevation of basal [Ca ]j was 2+  52  observed in all AP40 treated cells over both one and two-day treatments. Furthermore, significant attenuation of the PAF response was also seen in both the one and two-day Ap40 treated microglia in regards to both internal stores release and the level of calcium influx through SOC. These results mimic the preliminary datafromAD microglia and thus suggest that the use of fetal human microglia cultures is a viable model in studying conditions that occur in the adult human brain. Furthermore, this data also shows that Ap40 can effectively alter normal [Ca ]i responses to 2+  common signaling agents of microglia.  53  Figure 9: Normal PAF and ATP [Ca ], responses in human microglia. Graphs are representative 2+  experiments showing the mean of the responses measuredfromcontrol groups that received no Ap40 treatment. The green line indicates what was determined as the amplitude of the response and the blue line indicates where the time point for determining the level of influx was taken. (A) A characteristic response to PAF in human microgliafroma control experiment that was treated for one day with AP40freemedium (24 cells). Note the large initial peak in [Ca ]i followed by a 2+  sustained elevated [Ca ]i level. The initial peak in [Ca ]i is primarily reliant on internal stores of 2+  2+  calcium and the subsequent sustained elevation in [Ca ]i is reliant on the influx of extracellular 2+  calcium through the SOC pathway. ( B ) A characteristic response to ATP in human microglia from a control experiment that was treated for one day with Ap40freemedium (55 cells). Note that the major observable difference when compared to the PAF response is that there is no elevated shoulder after the initial [Ca ]ipeak. 2+  54  Figure 9A  PAF  0.7  0.6  Ca-free PSS  Ca-PSS  0.8_  lonomycin  PAF amplitude  0.5  -  0.4  0.3  0.2  50  100  150  200  250  300  350  400  450  500  Time (sec)  Figure 9B  Ca-free PSS  Ca-PSS  0.7.  lonomycin ATP  0.6  0.5  0.4 5  u.  ATP amplitude  0.3  0.2  influx level time point 0.1  50  100  150  200  250  Time (sec)  55  300  350  400  450  500  Figure 10: Ap40 incubation (10 uM) in human microglia cultures for one and two-day treatment significantly increased basal levels of [Ca ]. The level of basal [Ca 2+  [Ca ]i 2+  2+  ]i  was taken as the first  measurement at time zero for all experiments. A standard /-Test was used in all statistical  analysis. (A) This graph represents the basal [Ca ]j levels after one-day treatment of media 2+  without AP40 (control) or with AP40 (10 uM). Control represents 4 experiments with a total of 233 cells. AP40 treatment represents 8 experiments with a total of 166 cells (*** = p < 0.0001). ( B ) This graph represents the basal [Ca ]; levels after two days of Ap40 treatment (10 uM) or 2+  control. Control represents 4 experiments with a total of 127 cells. AP40 treatment represents 8 experiments with a total of 208 cells (* = p < 0.05)  56  Figure 10A  0.3-  g> co > co S. o  0.2  .2 f? o  S  0.1  Ap40 Treatment  Control  Figure 10B 0.3-  o •  .2 CO  o  u  0.2  Control  /p 40 Treatment  57  Figure 11: A P 4 0 incubation (lOpM) in human microglia cultures for one and two-day treatment significantly attenuated the amplitude of the PAF induced calcium response. A standard /-Test was used in all statistical analysis. (A) This graph represents the amplitude of the PAF [Ca ]i response 2+  after one-day treatment of media without ApMO (control) or with A p 4 0 (10 uM). Control represents 2 experiments with a total of 79 cells. A P 4 0 treatment represents 4 experiments with a total of 85 cells (*** = p < 0.0001). (B) This graph represents the PAF amplitudes after two days of A p 4 0 treatment (10 pM) or control. Control represents 2 experiments with a total of 52 cells. A p 4 0 treatment represents 4 experiments with a total of 112 cells (*** = p < 0.0001).  58  Figure 1 1 A  Control  Ap 40 Treatment  Figure 11B  0.44  Control  AP 40 Treatment  59  Figure 12: A p 4 0 incubation (lOuM) in human microglia cultures for one and two-day treatment significantly decreased the level of C a influx through the SOC as observed in the calcium plateau 2+  that is characteristic of a normal PAF response (see Figure 9). A time point one minute after the peak of the PAF response was used as a measure of the level of influx. A standard /-Test was used in all statistical analysis. (A) This graph represents the level of Ca influx in response to PAF after 2+  one-day treatment of media without A(J40 (control) or with A|340 (10 pM). Control represents 2 experiments with a total of 79 cells. A p 4 0 treatment represents 4 experiments with a total of 85 cells (*** = p < 0.0001). (B) This graph represents the PAF amplitudes after two days of A p 4 0 treatment (10 uM) or control. Control represents 2 experiments with a total of 52 cells. A p 4 0 treatment represents 4 experiments with a total of 112 cells (**= p<0.001). :  60  Figure 12A 0.5-.  •5  o.4^  i  •o  0.34 © o • co  < 0. 0.2H  0.1  Control  > /P 40 Treatment  Figure 12B 0.3-  X 3  •D  o  0.2-  3  < OL  0.1  AP40 Treatment  Control  61  4.  Discussion This thesis has characterized some of the effects of Af340 on signaling pathways in human  fetal microglia cells. The relevance of such work lies in a growing body of research implicating that the actions of Ap on microglia contribute to the progression of Alzheimer's Disease (AD). The current hypothesis regarding the role of microglia in AD is that the microglia become chronically over-activated which leads them to damage diseased and healthy tissue alike [McGeer 2000]. Ap is a peptide that activates microglia directly through specific binding domains as well as through amyloid plaques, which are difficult for the microglia to metabolize and subsequently cause chronic activation of microglia. The initial causative event in AD development is thought to be abnormal processing of APP that leads to disproportionably large amounts of AP forming and being deposited in the brain [Selkoe 1999, Koo 1999]. Thus, it is theorized that excess Ap is made which accumulates into large Ap plaque deposits. High concentrations of Ap, in and around these plaques, activate microglia and cause them to produce substances that ultimately kill neurons, increase AP production, and activate microglia even further [Eikelenboom 1996, Jiang 1994, McGeer 1995] (see Figure 1). Although AP40 is one of the major forms of Ap that is found in the brain, little work has been done to characterize the actions of this amyloid peptide on human microglia. This lack of research, along with the fact that Ap40 is more soluble and potentially provides more reproducible results than Ap42, determined that Ap40 would be the Ap peptide studied in this thesis. The individual studies of this thesis project were designed to examine different potential effects of Ap40 on microglia with the first study exarnining signaling characteristics induced by Ap40 and the other three studies looking at downstream effects of Ap40 treatment on human microglia.  62  4.1.  The use of primary cell cultures of human microglia  One major challenge in the execution of these studies was the use of primary cultures of human microglia from fetal brain tissue. This was the one critical factor in these experiments that was beyond the direct control of our laboratory and the availability and quality of tissue was not consistent. At times new cells were not available for long periods sincefreshbrain tissue was not available. In other instances the tissue samples themselves were not of good quality, the likely result of extra long storage and small sample sizes. As a consequence, cells isolatedfromthese tissues showed abnormalities including high and unstable [Ca ]i baselines, low cell density, 2+  inadequate Fura-2 uptake, and little or no responsiveness to the common agonists ATP and PAF. Even when good quality cultures were available the viability of the cultures would become compromised over a period of time and, after approximately 2 weeks on average, the cells would fail to respond normally and take on some of the abnormal characteristics previously mentioned. Thus, the availability and quality of the primary human microglia cultures was a limiting factor in all of these studies. Nevertheless, cultured human microglia comprise a novel in vitro preparation since very few studies have reported resultsfromsuch cells. Given the difficulties in obtaining consistent primary cultures of human microglia, it may be prudent to examine if microglia cell lines could be used instead. Such cell lines would have to be human in origin since rodent microglia have different cellular properties in comparison to human cells. At present no human microglia cell line is available. THP cells derivedfromblood monocytes are commonly used instead of human microglia, however monocytes, while being similar to microglia, do not always respond the same as human microglia. However, if a monocyte cell line were shown to respond to Ap40 in a manner similar to the human microglia used in this study then perhaps initial screening and pilot studies in a monocyte cell line would be feasible.  63  Such use of an appropriate cell line could narrow down areas of study that would allow for the more efficient use of primary human microglia cultures. In addition, there are a number of laboratories that are developing a functional human microglia cell line and there is the possibility that one will become available in the near future. Consequently, if a viable microglia cell line was established, future experiments would befreeof the limitations imposed by the use of primary microglia cultures and studies that require large numbers of cells, such as the neurotoxicity study, could be undertaken more easily. The use of a human microglia cell line would also allow for the more efficient screening of possible agents that could target the Ap40-induced [Ca ]j signaling 2+  pathway characterized in this work. Therefore future research expanding thefindingsof this thesis would benefit greatlyfromthe development and use of a human microglia cell line. 4.2.  The use of A p in vitro as a stimulus  Another central issue in the methods surrounding this thesis is the use of Ap40. A p is a difficult molecule to work with, and as mentioned in the introduction, issues of solubility and aggregation constantly surface. The problem with much of the research around A p is that very little effort is made to characterize the aggregation and solubility level of the peptide in solution. The degree of aggregation is often described in vague terms with solutions of A p either assumed to be mostly aggregated or mostly solubilized. Although it is understandable that most laboratories do not have the resources to characterize the rates of A p aggregation in a variety of solvents, it is surprising that none of the well-financed laboratories in thisfield(of which there are many) have thought to undertake this project. The activity of Ap cannot be accurately determined if the functional state of the Ap is not known and controlled for. Furthermore, it would be prudent to be able to characterize the level of solubility and the conformational state of Ap in vivo in order to perform in vitro studies where the physical characteristics of the A p peptide are controlled for in a  64  manner that mirrors the in vivo conditions. Therefore, further studies that progress the findings of this thesis should consider characterizing the solubility and aggregation rate of Ap over a number of solvents, time courses, and temperatures. Then accurate comparisons between levels of Ap aggregation and Ap activity can be made. 4.3.  Ap40-induced [Ca ]i signaling pathway 2+  The first study of this thesis work examined the [Ca ], signaling pathway induced by AP40 2+  in human microglia. The results have shown that the application of AP40 (4 and 10 uM) to microglia in Ca-PSS causes a rise in [Ca ]j that reaches a plateau level that is rnaintained even 2+  when the peptide is washed out of the extracellular solution. When this experiment was repeated in Ca-free PSS, AP40 caused no evident alteration of [Ca ]j levels. The removal of extracellular 2+  calcium at the peak of the Ap40-induced [Ca ]i rise caused a decrease in [Ca ]j levels, however 2+  2+  there was not complete inhibition of the [Ca ]jrise suggesting a small intracellular stores 2+  component to the response. This data indicates that AP40 induces a change in [Ca ]i that is 2+  mediated primarily through the influx of extracellular calcium into the cell. The Ap40-induced influx response was examined further with the application of low Cl" PSS, which has been shown previously to inhibit influx through the SOC pathway in microglia [McLarnon 2000]. The application of low Cl" PSS at the peak of the Ap40-induced rise in [Ca ]; 2+  caused a decrease in [Ca ]i, thereby suggesting that the actions of Ap40 in human microglia are 2+  sensitive to changes in anion channel expression. It is possible that depolarization caused by low Cl" PSS is responsible for inhibiting the calcium influx response. These results show that Ap40induced influx of calcium in microglia responds to the removal of extracellular calcium and the modulation of anion channels in a manner similar to the SOC pathway. However, it is possible that Ap40 is causing the influx of calcium through a pathway independent of the SOC since the  65  contribution of intracellular stores to the Ap40 response appears to be very small. This possibility was addressed by using SKF96365 (SKF), a known inhibitor of this pathway [Li 1999]. SKF caused no observable effect on the Ap40-induced calcium influx in human microglia indicating that the SOC pathway did not mediate the calcium influx response. Furthermore, it is unlikely that a voltage-gated calcium channel contributes to the influx since microglia, like other unexcitable cells, do not express such channels [McLarnon 1997]. Therefore it appears that Ap40 initiates a calcium influx pathway, which is not mediated by SOC. Ap40 seems to be activating an influx pathway independent of SOC, thus a review of some possible routes for calcium entry into microglia that would explain the results of this study is needed. One possibility is that the Ap molecule itself forms channels in the cell membrane. Work by Arispe (1993) has shown that Ap40 at a very high concentration of 0.46 mM is capable of forming calcium channels in synthetic bilayer membranes and the resulting current and channel properties were similar to other know calcium channels. More recently, work by Lin (1999) has also shown that Ap40 forms calcium channels in lipid vesicles at a concentration of 0.12 mM. However, the concentrations of Ap40 used in both of these studies were much higher than the concentrations used in this thesis work; the Arispe study, for example, used Ap40 concentrations more than 40 times larger. Other work by Fukuyama (1994) that examined [Ca ]j increases of 2+  neuronal cells with Ap40 application at 46 pM ruled out the possibility of a large effect of APchannels because the magnitude of the Ca  2+  current recorded was too small to account for the level  of calcium-uptake observed. If 46 pM of Ap40 is too small an amount to cause the formation of Ap-calcium channels then it would stand to reason that concentrations of AP40 at 4 and 10 pM would be even less likely to form channels. Furthermore, it is unlikely that the formation of Apcalcium channels would result in the slow increase in [Ca ]j that is observed in these results. A 2+  66  calcium pore in the cell membrane should cause a much faster and greater rise in [Ca ]j than what z+  is observed in these cells. Thus it seems unlikely that the formation of Ap40 calcium channels is the explanation for the Ap40-induced [Ca ], responses observed in this study. 2+  Another possible mechanism for the effects of Ap40 would be that it activates specific receptors that modulate calcium entry into the cell such as the Clq receptor that has been shown to be present on microglia and be activated by A p peptides [Jiang 1994]. However, Clq activation is not linked to an influx pathway, activation of the Clq receptor has been shown to modulate [Ca ]i 2+  levels by mobilizing ER stores of calcium [Kishore 2000, Lovik 2001]. This activation of the Clq receptor could explain the small intracellular store component of the Ap40-induced response that is suggested in those experiments where the [Ca ]irise is not completely inhibited by the removal of 2+  extracellular calcium. However, Clq activation cannot explain the major component of the Ap40 response that is due to calcium influx, although it does illustrate the possibility that AP40 could be inducing a [Ca ]j response through more than one signaling pathway. 2+  Yet another explanation for the actions of AP40 on [Ca ]i in microglia is possible. If the 2+  peptide acted as an inhibitor of the ER calcium pump (SERCA inhibitor) then a leak of calcium from the ER could result in a slowly developing influx of calcium. Thus rather than an all or nothing signalfromthe ER to activate the SOC pathway, an enhancement of a calcium leak channel would occur. Such channels have been proposed previously and may even be active in regulating normal basal [Ca ]j levels [Bode 1996]. However, this mechanism does not explain the 2+  lack of a [Ca ]i increase with Ap40 application in Ca-free PSS experiments since a small leak of 2+  calciumfromthe ER should be observable. It is quite possible that such a calcium leakfromER stores may be too small to be detected with the imaging apparatus used in these experiments.  67  The resultsfromthis study are consistent with previous studies that have shown that Ap peptides cause an elevation of [Ca ]i in rodent and human microglia as well as monocyte cell lines 2+  [Combs 1999, Korotzer 1995, Silei 1999]. Previous work by Silei (1999) has also shown that the actions of AP25-35 on [Ca  2+  ]i  in human microglia are mainly due to the influx of extracellular  calcium which they attribute to voltage sensitive calcium channels. However, studies have shown that voltage gated calcium channels are not found in human microglia [Elda 1998] and the channel inhibitors used in the Silei study could be having non-specific effects. Nevertheless, the finding that Ap induced a [Ca  2+  ]i  response which is primarily mediated by a calcium influx pathway is  consistent with the results of this thesis. Recent workfromCombs et al. (1999) is also relevant to the discussion of the present results. They showed that aggregated AP40 (40 pM) causes a [Ca  2+  ]j increase in THP  cells which  was attributed to the mobilization of calciumfrominternal stores. However, examination of their data indicates that calcium influx is the primary contributor of the Ap40.  Specifically, the  [Ca ]j 2+  [Ca ]j increase induced 2+  by  rise in Ca-PSS is roughly three times greater than the rise induced  in Ca-free PSS. Thus it appears that the primary component of the Ap40 response in the Combs study is actually due to influx, which is consistent with the results with human microglia reported in this thesis. Although the experimentsfromthisfirststudy have not pinpointed the precise mechanism of calcium entry into microglia as a result of Ap40 treatment, they have narrowed the focus of the search and have characterized the features of the  [Ca ]j 2+  response. The major action of Ap40 is an  influx of extracellular calcium into microglia independent of the SOC pathway that is maintained even when Ap40 is removedfromthe extracellular solution. Intracellular stores of calcium may also contribute a small part to the  [Ca ]i 2+  rise. A future set of experiments that could determine the  68  pathway of calcium-influx would be to screen inhibitors of the Clq receptor, as well as A(3channels, and modulators of the SERCA pump in order to look for modulatory actions on the Apinduced influx. If the actions of the Ap40 induced rise in [Ca ]i can be narrowed down to a 2+  precise receptor pathway or influx mechanism then it is possible that the over-activation of microglia in cases of neurodegeneration can be controlled and the progression of diseases such as A D can be slowed or even halted. The next two studies move on from the initial signaling actions of Ap40 to functional consequences of Ap40 signaling.  4.4.  Ap40 induced COX-2  expression in microglia  This study moves its focus downstream from the Ap40 [Ca ]i signal to the examination of 2+  Ap40 actions on COX-2 expression in human microglia. It was undertaken because of a large body of evidence showing that COX-2 is upregulated in conditions of inflammation, ischemia, and neurodegeneration such as in cases of A D . It is also thought that inhibition of COX-2 is how NSAID type drugs exert their anti-inflammatory effects. Furthermore, pro-inflammatory cytokines and activators of microglia also have been shown to upregulate COX-2 expression [Minghetti 1998]. Therefore it appears that COX-2 expression is enhanced in response to inflammation thus AP40, a further activator of microglia, has the potential to also upregulate COX-2 expression. The results indicate that application of AP40 (4 and 10 uM) is able to enhance COX-2 expression in human microglia. As mentioned in the discussion, COX-2 can exert neurotoxic actions through vasoconstriction and the production of reactive oxygen species [Minghetti 1998]. Thus Ap40 has the potential to cause neurotoxic effects in microglia by enhancing the expression of COX-2. Furthermore, since COX-2 is over-expressed in the pathology of A D , AP can account for yet another physiological symptom of A D since this data shows that Ap40 also enhances C O X 2 expression. Although increasing the expression of COX-2 in microglia has the potential to cause  69  neuronal damage COX-2 is an enzyme in prostaglandin synthesis that has the capability of catalyzing the production of a number of different prostaglandin products that have a diverse range of functions [Aloisi 1999, Minghetti 1998]. Therefore the end result of increased COX-2 expression in microglia, be it neurotoxic or not, remains unclear. The rationale for this study was to link the imaging studies on Ap40 mediated  [Ca ]i 2+  signaling with changes in the cellular function of microglia. Ideally a direct link between the AP40 [Ca ]i 2+  signal and COX-2 expression would be established with a procedure that could be applied to  both experimental situations which would modulate both the calcium signal and COX-2 expression. However, since the precise mechanism of the Ap40-induced  [Ca ]i 2+  response was not determined, a  pharmacological agent that could be applied to block the actions of AP40 on [Ca ]i was not 2+  identified. Although Ca-free PSS and low Cl" PSS did inhibit the Ap40-induced calcium response the use of such solutions in an attempt to modulate Ap40-induced COX-2 expression would not be feasible since long-term exposure to such solutions would cause many changes in microglia independent of Ap40 effects. Thus, it can only be hypothesized at this time that the  [Ca ]i 2+  response to Ap40 could contribute to the increase in COX-2 expression observed in this study. Future experiments that examined the effects of agents which blocked the Ap40-induced [Ca ]j 2+  response on COX-2 expression would be useful in deterrnining the characteristics of AP40 signaling in microglia. Since the effects of COX-2 expression in microglia remains unclear another set of future studies that assay for prostaglandin production and characterize the inflammatory effects of the prostanoid products produced would be useful in deterrnining the implications of COX-2 expression as a result of Ap40 treatment. In this way the specific downstream products that are produced due to COX-2 upregulation can be determined and therapeutic approaches that target  70  these end products can be developed. Drugs that are designed to control these end products of COX-2 expression would potentially have a greater therapeutic impact then targeting the COX-2 enzyme with NSAID drugs since the inhibition of COX-2 causes the inhibition of prostaglandins that are both inflammatory and anti-inflammatory in nature [Minghetti 1998]. The third study addresses another possible action of Ap40 on microglia function, the production of neurotoxic substances, of which enhanced COX-2 production may be a contributing factor. 4.5.  Ap40 induction of neurotoxic substances from microglia  The production of a AP-induced neurotoxin in microglia is an important concept in relating microglia to the pathogenesis of AD and it has been documented in rodent microglia, macrophage cell lines, and postmortem human microglia [Giulian 1996]. However, no previously published work has examined the potential of Ap40 in inducing the production of neurotoxic substance in fetal human microglia. If neurotoxin production was found in Ap40 treated microglia then it would provide direct evidence that Ap40 treatment can be linked to a dominant pathological symptom of AD. The other three studies, while being related to the pathology of AD, are not end point functions of neurodegeneration that can be linked to the cognitive deficits seen in AD. [Ca ]i iT  signaling is an initial event that can lead to many different outcomes, COX-2 is elevated in the AD brain but its role in pathogenesis is still not fully understood, and the role that diminished microglia responsiveness has in neurodegeneration is at this time purely conjecture. All of the other studies have the potential to illuminate the pathogenesis of AD but neurotoxicity is an endpoint symptom that can easily explain the behavioral outcome of AD. Thus this was an important study not only from the view that Ap40 induced neurotoxicity in human microglia is novel, but it was also examining the possibility that Ap40 can cause a direct pathological symptom that is seen in AD.  71  Two standard molecular assays; MTT, which determined cell viability, and LDH, which determined cell death, measured the level of neurotoxicity. The LDH assay proved ineffective since all the treatment groups had similar low levels of cell death. The MTT assay on the other hand showed that both A(}40 treated microglia and Ap40 alone caused a decrease in neuroblastoma viability without microglia involvement. Furthermore, the decreased level of neuroblastoma viability was not significantly differentfromneuroblastoma treated with supernatant from microglia and AP40. Thus the resultsfromthe MTT assay indicate that no conclusions as to the ability of microglia treated with AP40 to produce neurotoxic substances can be made. The possibility remains that AP40 treatment of microglia does cause the production of neurotoxic substances but the methods used were unable to show this effect. Assuming that human microglia do react to Ap40 and produce neurotoxins then it becomes a matter of stimulating the microglia enough so that the neurotoxin production of microglia in response to Ap40 becomes greater than the neurotoxic effects of Ap40 alone. This issue was briefly touched upon in the results section, but to paraphrase it once again, there are two obvious courses of action that could be taken. Thefirstmethod would be to increase the incubation time of AP40 with microglia cells, with a greater timeframeof AP40 treatment it is possible that the microglia will produce more neurotoxic substances. Another possible approach would be to increase the density of microglia per treatment well. The reasoning being that a greater number of microglia cells could increase the concentration of neurotoxin in the supernatant. Thus it is possible that a slight alteration in the procedure would be able to document the induction of neurotoxic substances in microglia as a result of Ap40 application. However, one of the greatest difficulties in performing these experiments is the amount of microglia cells that are needed. In order to have a single well of microglia approximately 6 x 10  72  4  cells are needed. Thus one run of three concentrations of AP40 (0, 4, and 10 uM) would use 1.8 x 10 microglia cells. This large amount of microglia is often difficult to obtain due to factors 5  discussed earlier on of the use of microglia cultures in section 4.1. Thus, in addition to the protocol changes suggested previously, it would be useful to determine the neurotoxin inducing potential of AP40 in an appropriate microglia cell line. A cell line could be used in this work to optimize the experimental conditions needed for AP40 induction of neurotoxic substances. Once the neurotoxic potential of AP40 was characterized in a cell line then experiments using the same optimized methods with primary human microglia cultures could be preformed. In this way valuable primary cultures of human microglia would be used more efficiently and would not be needed for the assessment of experimental protocols.  4.6.  The actions of Ap40 on ATP and PAF [Ca ]j responses 2+  This study examined the effects of Ap40 exposure on the reactions of human microglia to ATP and PAF. Preliminary datafromthis laboratory suggests that responses to both ATP and PAF are attenuated and basal [Ca ]j levels are elevated in human microgliafromAD cases. These 2+  preliminary results suggest that normal [Ca ]i signaling is altered in AD microglia. 2+  The effect of Ap40 on human microglia was studied in an effort to determine if Ap has a role in causing this alteration of [Ca ]j signaling that is observed in AD microglia. The results 2+  indicate that Ap40, over both a 20 and 48 hrs timeframe(approximately one and two day treatment), was able to cause elevated [Ca ]i levels and attenuated PAF responses in both ER store 2+  depletion (amplitude peak) and SOC amplitude ([Ca ], plateau). However, ATP responses in 2+  AP40 treated microglia were not diminished. Although both ATP and PAF utilize similar stores of internal calcium the signaling pathways for these agents are different. Thus it appears that Ap40 is able to disrupt [Ca ]j signaling in response to PAF but not ATP. These data provide evidence that 2+  73  Ap40 is able to alter normal [Ca ]i signaling responses in microglia. The results also support the 2+  theory that Ap40 actions on microglia cause some of the alterations in cellular signaling observed in AD since diminished PAF responses were also observed in preliminary work with AD microglia. One explanation for the elevated [Ca ]j levels observed in the Ap40 treated cells can be 2+  drawn from the results obtainedfromthefirststudy (section 3.1.). A clear resultfromthe first study is that AP40 induces a SOC independent influx pathway in microglia. Therefore, it would be reasonable to hypothesize that long-term activation of this Ap40-influx pathway could cause increased basal [Ca ]j levels. 2+  Raised [Ca ] levels may also explain a possible reason for the decreased levels of SOC 2+  influx seen in the PAF response of Ap40 treated cells. Since the purpose of the SOC is to replenish stores after depletion, if another influx pathway was activated previously then there would be less need for a large influx through the SOC since there would be more available calcium already present in the cell. However this possibility still does not explain the attenuated PAF amplitudes and there is no current evidence to support this theory. Furthermore, the use of calcium for signaling within the cell is often very localized to specific domains of activation [Verkhratsky 1996]. Therefore an influx of calcium through one pathway may not necessarily supply another calcium signaling pathway. The clear result of this study is that Ap40 is capable of altering normal [Ca ]j responses of 2+  microglia. Furthermore, since attenuated [Ca ]i responses are also seen in AD microglia, this data 2+  provides evidence that Ap40 can account for yet another physiological symptom of AD. The similarity in the responsiveness of adult human AD microglia and fetal human microglia also supports the use of primary fetal microgha cultures as a viable model in the study of disease in the adult human brain.  74  4.7.  Final Summary  Research undertaken for this thesis provides evidence that A(340 activates microglia primarily through a SOC-independent calcium influx pathway in human microglia. A small internal stores component may also be present in response to Ap40 since removal of extracellular calcium did not completely inhibit the [Ca ]irise. These experiments characterized the initial 2+  [Ca ]i signal in response to Ap40 and the further two studies focused on downstream functional 2+  effects of this [Ca ]i signal. However, a direct connection between the Ap40 [Ca ]i signal and 2+  2+  downstream functional effects cannot be established at this time since no pharmacological intervention is available that is capable of blocking the actions of Ap40-mediated [Ca]i signaling 2 +  without compromising cell viability. In the second study it was determined that Ap40 enhances the expression of COX-2 in human microglia. While COX-2 has been shown to produce neurotoxicfreeoxygen species, COX2 also produces many different prostaglandins with diverse functions, thus the full implications of this increase in COX-2 expression remain unclear. Future experiments in this area should focus on the expression of the prostaglandins that are produced in response to Ap40 treatment and subsequent COX-2 expression. The third study examined the potential of Ap40 to induce the production of neurotoxic substances in human microglia. The production of neurotoxic substancesfromApMO treated microglia was not found. However, this might be due to the limitations of the methods used, which relied on the availability of a large number of cultured human microglia. Thefinalstudy focused on Ap40's potential to cause alterations in normal [Ca ]i responses 2+  in microglia. This work was pursued because of preliminary data showing that AD microglia have elevated basal [Ca ]i levels and attenuated A T P and P A F responses. Ap40 treatments of fetal 2+  75  human microglia caused elevated [Ca ]i levels and attenuated PAF responses. These results show 2+  that an in vitro model of Ap application to cultured microglia can mimic the characteristics of AD microglia. This thesis focused on a continuum of Ap40 actions in microglia;fromthe initial [Ca ]i 2+  signal to the functional expression of enzymes and the putative production of neurotoxic substances. In addition, it provided evidence the Ap40 alters the normal [Ca ]; responses in fetal 2+  human microglia; a result which preliminary work suggests is also a characteristic of AD microglia. These studies providefreshinsight into the involvement and role of microglia in diseases of the brain and new directions in future research have emerged as a result.  76  References:  Aloisi F, De Simone R, Columbia-Cabezas S, Levi G. Opposite effects of interferon-y and prostaglandin E2 on tumor necrosis factor and interleukin-10 production in microglia: A regulatory loop controlling microglia pro- and anti-inflammatory activities. Journal of Neuroscience Research 1999; 56:571-580. Araujo DM, Cotman CW. Beta-amyloid stimulated glial cells in vitro to produce growth factors that accumulate in senile plaques in Alzheimer's disease. Brain Research 1992; 569:141-145. Arispe N , Rojas E, Pollard HB. Alzheimer disease amyloid P protein forms calcium channesl in bilayer membranes: Blockade by tromethamine and aluminum. Proc Natl Acad Sci USA 1993; 90:567-571. Bader M , Taupenot L, Ulrich G, Aunis D, Ciesielski-Treska J. Bacterial endotoxin induces [Ca ]; 2+  transients and changes the organization of actin in microglia. Glia 1994; 11:336-344. Bitting L, Naidu A, Cordell B, Murphy GM. Beta-amyloid peptide secretion by microglia cell line is induced by beta-amyloid (25-35) and lipopolysaccharide. Journal of Biological Chemistry 1996; 271(27):16084-16089. Bode H, Netter KJ. Agonist-releasable intracellular calcium stores and the phenomenon of storedependent calcium entry. Biochemical Pharmacology 1996; 51:993-1001. Busciglio J, Gabuzda DH, Matsudaira P, Yankner B. Generation of beta-amyloid in the secretory pathway in neuronal and nonneuronal cells. Proc Natl Acad Sci USA 1993; 90:2092-2096. Carafoli E, Genazzani A, Guerini D. Calcium controls the transcription of its own transporters and channels in developing neurons. Biochem Biophys Res Com 1999; 266:624-632.  77  Cedazo-Minguez A, Huttinger M, Cowburn RF. P-VLDL protects against AP(l-42) and apoE toxicity in human SH-SY5Y neuroblastoma cells. NeuroReport 2001; 12(2):201 -206. Chung H, Brazil MI, Soe TT, Maxfield FR. Uptake, degradation, and release offibrillarand soluble forms of Alzheimer's amyloid beta-peptide by microglia cells. Journal of Biological Chemistry 1999; 274(45):32301-32308. Citron M et al. Evidence that the 42 and 40 amino acid forms of amyloid beta protein are generated from the P-amyloid precursor protein by different protease activites. Proc Natl Acad Sci USA 1996;93:13170-13175. Combs CK, Johnson DE, Cannady SB, Lehman TM, Landreth GE. Identification of microglia signal transduction pathways mediating a neurotoxic response to amyloidogenicfragmentsof betaamyloid and prion protiens. Journal ofNeuroscience 1999; 19(3):928-939.  Combs, CK, Johnson DE, Karlo C» Cannaday STj Landreth GE. Inflammatory mechanisms in Alzheimer's disease: inhibition of p-amyloid-stimulated proinflammatory responses and neurotoxicity by PPARgama agonists. Journal ofNeuroscience 2000; 20(2):558-567.  Decker T, Lohmann-Matthes M. A quick and simple method for the quantitation of lactate dehydrogenase release in measurements of cellular cytotoxicity and tumor necrosis factor (TNF) activity. Journal ofImmunological Methods 1988; 15:61-69.  Duff K. Transgenic mouse models of Alzheimer's disease. Handbook of Molecular-Genetic Techniques for Brain and Behavior Research vol. 13. Elsevier Science 1999:638-651.  Eder C. Ion channels in microglia (brain macrophages). American Physiological Society 1998; 275:C327-C342.  Eikelenboom P, Veerhuis R. The role of complement and activated microglia in the pathogenesis of Alzheimer's disease. Neurobiology ofAging 1996; 17(5):673-380.  78  Ercolani L, Florence B, Denaro M, Alexander M. Isolation and complete sequence of a functional human glyceraldehayde-3 -phosphate dehydrogenase gene. Journal of Biological Chemistry 1998; 263(30): 15335-41.  Felder CC, Singer-Lahat D, Mathes C. Voltage-independant calcium channels. Biochemical Pharmacology 1994; 48(11):1997-2004.  Fukumoto H, Tomita T, Matsunaga H, Ishibashi Y, Saido T, Iwatsubo T. Primary cultures of neuronal and non-neuronal rat brain cells secrete similar proportions of amyloid beta peptides ending at Ab40 and Ab42. NeuroReport 1999; 10:2965-2969.  Fukuyama R, Wadhwani KC, Galdzicki Z, Rapoport SI, Ehrenstein G. P-amyloid polypeptide increases calcium-uptake in PC 12 cells: a possible mechanism for its cellular toxicity in Alzheimer's disease. Brain Research 1994; 667':269-272.  Galoyan A. Neurochemistry of brain neuroendocrine immune system: signal molecules. Neurochemical Research 2000; 25(9/10):1343-1355.  Games D, Khan KM, Soriano FG, Keim PS, Davis DL, Bryant K, Lieberburg I. Lack of Alzheimer pathology after p-amyloid protein injections in rat brain. Neurobiology ofAging 1992; 13:569-576.  Gehrmann J, Matsumoto Y, Kreutzberg GW. Microglia: intrinsic immuneffector cell of the brain. Brain Research Reviews 1995; 20:269-287.  Ghosh A, Greenberg ME. Calcium signaling in neurons: molecular mechanisms and cellular consequences. Science 1995; 268:239-247.  Ginty DG. Calcium regulation of gene expression: Isn't that spatial?. Neuron 1997; 18:183-186.  79  Giulian D, Haverkamp LJ, Li J, Karshin WL, Yu J, Tom D, Li X, Kirkpatrick JB. Senile plaques stimulate microglia to release a neurotoxin found in Alzheimer brain. Neurochem Int 1995; 27:119-137.  Giulian D, Haverkamp LJ, Karshin W, Tom D, Li J, Kirkpatrick J, Kuo YM, Roher AE. Specific domains of beta-amyloid from Alzheimer plaque elicit neuron killing in human microglia. Journal of Neuroscience 1996; 16(19):6021-6037.  Goghari V, Franciosi S, Kim SU, Lee YB, McLarnon JG. Acute applications of interleukin-ip induces Ca responses in human microgha. Neuroscience Letters 2000; 281:83-86. 2+  Haerter-Gebicke PJ, Van Calker D, Norenberg W, Illes P. Molecular mechanisms of microglial activation; implications for regeneration and neurodegenerative diseases. Neurochem Int 1996; 29(1):1-12.  Halliday G, Robinson SR, Shepherd C, Kril J. Alzheimer's disease and inflammation: a review of cellular and therapeutic mechanisms. Clinical and Experimental Pharmacology and Physiology 2000; 27:1-8.  Hirst RA, Harrison C, Hirota K, Lambert DG. Measurement of [Ca ]i in whole cell suspensions 2+  using Fura-2. Methods in Molecular Biology 1999 ; 114:31-39.  Jiang H, Burdick D, Glabe CG, Cotman CW, Tenner AJ. Beta-amyloid activates complement by binding to a specific region of the collagen-like domain of the Clq A chain. Journal of Immunology 1994:5050-5059.  Kato H, Walz W. The initiation of the microglia response. Brain Pathology 2000; 10:137-143.  Khoury JE, Hickman SE, Thomas C A Cao L, Silverstein SC, Loike JD. Scavenger receptormediated adhesion of microglia to p-amyloid fibrils. Nature 1996; 382:716-719.  80  Kishore U , Reid K B M . Clq: structure, function, and receptors. Immunopharmacology 2000; 49:159-170.  Klegeris A , McGeer PL. Interaction of various intracellular signaling mechanisms involved in mononuclear phagocyte toxicity toward neuronal cells. Journal of Leukocyte Biology 2000; 67:127-133.  Koo E H , Lansbury PT, Kelly JW. Amyloid diseases: abnormal protein aggregation in neurodegeneration. Proc Natl Acad Sci USA 1999;96:9989-9990.  Korotzer AR, Whittemore ER, Cotman CW. Differential regulation by p-amyloid peptides of intracellular free C a  2 +  concentration in cultured rat microglia. European Journal of Pharmacology  1995; 288:125-130.  Kreutzberg GW. Microglia: a sensor for pathological events in the CNS. Trends in Neuroscience 1996;19:312-318.  Lambert MP, Stevens G , Sabo S, Barber K, Wang G, Wade W, Krafft G , Snyder S, Holzman TF, Klein WL. B/A4-ecoked degeneration of differentiated SH-SY5 Y human neuroblastoma cells. Journal of Neuroscience Research 1994; 39:377-385.  Lansbury PT. Evolution of amyloid: what normal protein folding may tell us about fibrillogenesis and disease. Proc Natl Acad Sci USA 1999;96:3342-3344.  Lamer AJ. Hypothesis: amyloid beta-peptides truncated at the N-terrninus contribute to the pathogenesis of Alzheimer's disease. Neurobiology of Aging 1999; 20:65-69.  Li L , Bressler B, Prameya R, Dorovini-Zis K, Van Breemen C. Agonist-stimulated calcium entry in primary cultures of human cerebral microvascular endothelial cells. Microvascular Research 1999; 57:211-226.  81  Lin H, Zhu YJ, Lai R. Amyloid P protein (1-40) forms calcium-permeable, Zn -sensitive channel z+  in reconstituted lipid vesicles. Biochemistry 1999; 38:11189-11196. Lovik G, Larsen Sand K, Iversen JG, Rolstad B. ClqRp Elicits a Ca response in rat NK cells but 2+  does not influence NK-mediated cytotoxicity. Scandinavian Journal of Immunology 2001; 53:410415.  Mattson MP, Barger SW, Furukawa K, Bruce AJ, Wyss-Coray T, Mark RJ, Mucke L. Cellular signaling roles of TGF beta, TNF alpha and beta APP in brain injury responses and Alzheimer's disease. Brain Research Reviews 1997; 23:47-61.  McDonald DR, Brunden KR, Landreth GE. Amyloidfibrilsactivate tyrosine lcinase-dependent signaling and superoxide production in microglia. Journal of Neuroscience 1997; 17(7):22842294.  McDonald DR, Bamberger ME, Combs CK, Landreth GE. Beta-amyloidfibrilsactivate parallel mitogen-activated protein kinase pathways in microglia and THP1 monocytes. Journal of Neuroscience 1998; 18(12):4451-4460.  McGeer PL, McGeer EG. The inflammatory response system of brain: implications for therapy of Alzheimer and other neurodegenerative diseases. Brain Research Reviews 1995; 21:195-218.  McGeer PL, McGeer EG. Autotoxicity and Alzheimer disease. Archives of Neurology 2000; 57:789-790.  McLarnon JG, Franciosi S, Wang X, Goghari V, Bae JH, Kim SU. Acute actions of TNF-oc on intracellular Ca and K currents in human microglia. Neuroscience, In Press. 2+  +  McLarnon JG, Helm J, Goghari V, Franciosi S, Choi HB, Nagai A, Kim SU. Anion channels modulate store-operated calcium influx in human microglia. Cell Calcium 2000; 28(4):261-268.  82  McLarnon JG, Xu R, Lee YB, Kim SU. Ion channels of human microglia in culture. Neuroscience 1997; 78(4):1217-1228.  Minghetti L, Levi G. Microglia as effector cells in brain damage and repair: focus on prostanoids and nitric oxide. Progess in Neurobiology 1998; 54:99-125.  Moller T, Kann O, Verkhratsky A, Kettenmann H. Activation of mouse microglia cells affects P2 receptor signaling. Brain Research 2000; 853:49-59.  Moore S, Thanos S. The concept of microglia in relation to central nervous system disease and regeneration. Progress in Neurobiology 1996; 48:441-460.  Monning U, Sandbrink R, Weidemann A, Banati RB, Masters CL, Beyreuther K. Extracellular matrix influences the biogenesis of amyloid precursor protein in microglia cells. Journal of Biological Chemistry 1995; 270(13):7104-7110.  Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Journal of Immunological Methods 1983; 65:55-63.  Nagai A, Nakagawa E, Choi HB, Hatori K, Kobayashi S, Kim SUK. Erythropoietin and Erythropoietin receptors in human CNS neurons, astrocytes, microglia, and oligodendrocytes grown in culture. Journal of Neuropathology and Experimental Neurology 2001; 60(4):386-392.  Neve RL, Robakis NK. Alzheimer's disease: a re-examination of the amyloid hypothesis. Perspectives on Disease. Trends in Neuroscience 1998; 21(l):15-9.  Paresce DM, Ghosh RN, Maxfield FR. Microglia cells internalize aggregates of the Alzheimer's Disease amyloid p-protein via a scavenger receptor. Neuron 1996; 17:553-565.  Pasinetti GM, Aisen PS. Cyclooxygenase-2 expression is increased in frontal cortex of Alzheimer's disease brain. Neuroscience 1998; 87(2):319-324.  83  Rio-Hortega, P. Microglia. In: Cytology and Cellular Pathology of the Nervous System, pp. 482534. Ed. W. Penfield. Hober: New York 1932.  Sairanen T, Ristirnaki A, Karjalainen-Lindsberg M, Paetau A, Kaste M, Lindsberg PJ. Cyclooxygenase-2 is induced globally in infarcted human brain. American Neurological Association 1998; 738-747.  Seabrook GR, Rosahl TW. Transgenic animals relevant to Alzheimer's disease. Neuropharmacology 1999;38:1-17.  Selkoe DJ. Translating cell biology into therapeutic advances in Alzheimer's disease. Nature 1999; 399:A23-31.  Silei C, Fabrizi C, Venturini G, Salmona M, Bugiani O, Tagliavini F, Lauro GM. Activation of microglia cells by PrP and P-amyloid fragments raises intracellular calcium through L-type voltage sensitive calcium channels. Brain Research 1999; 818:168-170.  Sinha S, Lieberburg I. Cellular mechanisms of Ap production and secretion. Proc Natl Acad Sci USA 1999;96:11049-11053.  Slunt HH, Thinakaran G, Von Koch C, Lo HC, Tanzi RE, Sisodia SS. Expression of a ubiquitous, cross-reactive homologue of the mouse P-amyloid precursor protein (APP). Journal of Biological Chemistry 1994;269:2637-2644.  Sola C, Barron S, Tusell JP, Serratosa J. The Ca Vcalmoduliri signaling system in the neural 2  resonse to excitability. Involvement of neuronal and glial cells. Progress in Neurobiology 1999; 58:207-232.  Stoll G, Jander S. the role of microglia and macrophages in the pathophysiology of the CNS. Progress in Neurobiology 1999; 58:233-247.  84  Tan J, Town T, Paris D, Mori T, Suo Z, Crawford F, Mattson MP, Flavell RA, Mullan M. Microglia activation resultingfromCD40-CD40L interaction after beta-amyloid stimulation. Science 1999; 286:2352-2355.  Verkhratsky A, Kettenmann H. Calcium signalling in glial cells. Trends in Neuroscience 1996; 19(8):346-352.  Wang X, Bae JH, Kim SU, McLarnon JG. Platelet-activating factor induced Ca signaling in 2+  human microglia. Brain Research 1999;842:159-165.  Wang X, Kim SU, Van Breemen C, McLarnon JG. Activation of purinergic P2X receptors inhibits P2Y-mediated Ca influx in human microglia. Cell Calcium 2000; 27(4):205-212. 2+  Wilson CA et al. Intracellular APP processing and Ap production in Alzheimer disease. J Neuropath Exp Neurol 1999; 58(8):787-794.  Wisniewski H M et al. Pathogenesis of amyloid beta plaques. Neurosci News 1998; l(6):30-34.  Wujek JR, Dority MD, Frederickson RCA, Brudent KR. Deposits of Ap fibrils are not toxic to cortical and hippocampal neurons in vitro. Neurobiology ofAging 1996; 17:107-113.  Yasojima K, Schwab C, McGeer EG, McGeer PL. Up-regulated production and activation of the complement system in Alzheimer's disease brain. American Journal ofPathology 1999; 154(3):927-936.  85  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items