|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
The Influence of Amyloid-Beta, a Major Pathological Marker in Alzheimer's Disease, on Molecular Cognitive Processes of APP+PS1 Transgenic Mice by Chad Anthony Dickey A dissertation submitted in partial fulfillment of the requirement s for the degree of Doctor of Philosophy Department of Pharmacology and Therapeutics College of Medicine University of South Florida Major Professor: David Morgan, Ph.D. Marcia Gordon, Ph.D. Keith Pennypacker, Ph.D. Kenneth Ugen, Ph.D. Javier Cuevas, Ph.D. Date of Approval: May 25th, 2004 Keywords: gene expression, amyloid pla ques, immediate early genes, synaptic plasticity, memory Copyright 2004, Chad Anthony Dickey
i TABLE OF CONTENTS LIST OF TABLES ii LIST OF FIGURES iii ABSTRACT v INTRODUCTION 1 Alzheimers disease 1 Synaptic plasticity and LTP 5 Behavior 8 Constitutive Protein Expression 11 IEG Expression 14 The Role of Na+/K+ ATPase in Cognitive Function 22 PAPER I: SELECTIVELY REDUCE D EXPRESSION OF SYNAPTIC PLASTICITY-RELATED GENES IN APP+PS1 TRANSGENIC MICE 25 PAPER II: AMYLOID SUPPRESSES IN DUCTION OF GENES CRITICAL FOR MEMORY CONSOLIDATION IN APP+PS1 TRANSGENIC MICE 60 PAPER III: DYSREGULATION OF THE NA+/K+ ATPASE BY AMYLOID: IMPLICATIONS FOR NEURITIC D YSTROPHIA IN ALZHEIMERS DISEASE 92 CONCLUSIONS 128 REFERENCES 145 APPENDICES 160 Appendix A. Quantitative Real Ti me PCR (qRT-PCR) Protocol 161 Appendix B. Na+/K+ ATPase Activity Assay Protocol 170 Appendix C. Ce ll Culture and A Preparation 173 ABOUT THE AUTHOR End Page
ii LIST OF TABLES INTRODUCTION TABLE 1. Overview of gene function 24 PAPER I TABLE 1. Expression of neuronal genes in transgenic mice compared to nontransgenic mice (array and qRT-PCR data) 51 TABLE 2. Genes that are up-regulated in amyloid-containing areas of transgenic mice brains compared to the same areas of nontransgenic mice brains (microarray data) 53 TABLE 3. Genes that are up-regulated in both amyloid-containing brain regions and amyloid-free brain regions of transgenic mice (array and qRT-PCR data) 55 PAPER II TABLE 1. Expression of non-inducible genes in APP+PS1 mice stimulated by environmental novelty compared to nontransgenic littermates by qRT-PCR 91 CONCLUSIONS TABLE 1. Genes that are selectively and significantly down-regulated in the amyloid-containing hi ppocampi of APP+PS1 mice (qRT-PCR data) 144
iii LIST OF FIGURES PAPER I FIGURE 1. Gene expression profile of tr ansgenic mice in amyloidcontaining and amyloid-free br ain regions by qRT-PCR 57 FIGURE 2. Gene expression profile of Alzheimer disease tissue in amyloid-containing and amyloi d-free brain regions by qRTPCR 59 PAPER II FIGURE 1. Time course of gene expressi on in the hippocampus of transgenic mice by qRT-PCR 84 FIGURE 2. Gene expression in hippocampi of APP only, PS1 only and APP+PS1 transgenic mice at 18 months of age by qRT-PCR 86 FIGURE 3. Gene expression profile in hi ppocampus, posterior cortex and caudate nucleus of APP+PS1 transgenic mice at 18 months of age by qRT-PCR 88 FIGURE 4. I EG expression measured by qR T-PCR in hippocampi of APP+PS1 transgenic mice and non-transgenic littermates following Induction by environmental novelty 90 PAPER III FIGURE 1. Expression of Na+/K+ ATPase mRNA in APP+PS1 mice by qRT-PCR 117 FIGURE 2. Activity of the Na+/K+ ATPase enzyme in APP+PS1 mice 119 FIGURE 3. Verification of Na+/K+ ATPase III antibody specificity by immunohistochemistry folowi ng pre-incubation with purified protein and Western blotting 121
iv FIGURE 4. Hippocampal and cortical immu nohistochemistry for Na+/K+ ATPase in APP+PS1 mice and non-transgenic littermates 123 FIGURE 5. Dual immunostaining of Na+/K+ ATPase III and dystrophic neurites 125 FIGURE 6. Amyloid-beta 1-42 inhibits Na+/K+ ATPase activity at high concentrations 127 APPENDIX C FIGURE 1. Induced expression of IEGs in differentiated N2A cells by nerve growth factor 178
v The Influence of Amyloid-Beta, a Major Pathological Marker in Alzheimer's Disease, on Molecular Cognitive Processes of APP+PS1 Transgenic Mice Chad Anthony Dickey ABSTRACT Alzheimers disease (AD) is characterized by anterograde amnesia followed by a progressive decline in cogniti ve function. Post mortem examination of forebrain tissue from sufferers reveal s the presence of extra-cellular amyloidbeta plaques, intracellular neurofibrillary tangles, activation of glial cells and massive neuron loss. Transgenic mice expr essing mutated forms of the amyloid precursor protein (APP) and presenilin-1 (PS1) genes develop neuritic amyloid plaques, glial cell activation and memory deficits, without the formation of intracellular tangles and neurodegeneration. The mechanisms by which these transgenic mice develop mnemonic deficiencies are unclear. Gene expre ssion of aged memory-deficient APP+PS1 mice compared with non-transgenic littermates measured by microarray and subsequent quantitative real-t ime PCR (qRT-PCR) analysis revealed 6 inducible immediate-early genes (IEGs) and 5 other more stably expressed plasticityrelated genes (PRGs) that were si gnificantly down-regulated in amyloidcontaining hippocampus, but not down-regulated in amyl oid-free cerebellum. Other genes linked to memory remained stably expressed in both regions. Analysis of forebrain AD tissue rev ealed that all genes m easured were down-
vi regulated presumably due to neurodegeneration, while the amyloid-free region maintained stable expression. IEG expre ssion in APP+PS1 mice was sensitive to lower levels of amyloid. However, only in the presence of a substantially larger amyloid burden, when memory deficit s reliably persist, were both PRGs and IEGs down-regulated. Importantly, we found that IEG expression was decreased in APP+PS1 mice following exposure to a novel environment, indicating that the induction of these IEGs was impaired, ra ther than the basal ex pression of resting mice. Na+/K+ ATPase, an enzyme critical for t he maintenance of membrane potential, was identified as a down-regulated PRG. We found that activity of this enzyme was both impaired in the hippocampi of APP+PS1 mice and specifically inhibited by high concentrations of amyloid-bet a. Na+/K+ ATPase immunostaining revealed decreased protein in the ar ea surrounding the amyloid plaque, where dystrophic neurites were visible, indicati ng amyloid may disrupt ionic gradients resulting in neuritic dystrophia. These findings suggest that amyloid accumulation may result in the impairment of IEG induction and disrupti on of the Na+/K+ ATPase, possibly eliciting the memory loss developed in APP+PS1 mice.
1 INTRODUCTION ALZHEIMERS DISEASE. Memory loss is the primary and mo st deleterious symptom of AD. Sufferers initially lose their cognitive abilities followed slowly by autonomic deterioration. AD pathology is characteriz ed by both the extra-cellular cerebral accumulation of A 1-42 (Hardy, 1997) and the aggr egation of intracellular tangles of a hyperphosphorylated form of the tubulin-associat ed protein, tau ( ) (Mattson, 1997). The AD brain is not only characterized by these plaques and tangles, but also by neuronal and synaptic loss, as evidenced by the downregulation of cytoskeletaland exocytotic-associated proteins. The severity of AD can then be determined based on this co llective pathology (Masliah et al., 2001;Mukaetova-Ladinska et al., 2000;Beno witz and Routtenberg, 1997;Yao and Coleman, 1998;Callahan et al., 2002). Amyloid precursor protein (APP) is processed post-translationally by several proteases known as -, and -secretase. Agonists for the secretase and inhibitors of both the and have shown promise for use in AD therapy. The -secretase cleaves membrane-bound APP within the A 1-42 amino acid site, yielding peptide products that lack any aggregat ive properties. -secretase (BACE1) acts closer to t he N-terminal of APP leaving a longer
2 fragment for -secretase to cleave which result s in production of the deleterious amyloidogenic A 1-42 mole cule. The activity of the -secretase on the remaining membrane-bound portion of APP following -se cretase cleavage can yield either the 1-42 A fragment with fibrillo genic properties or a much less sticky 1-40 peptide product (Selkoe, 2001;De wachter and Van Leuven, 2002). Whether the 1-40 or 1-42 peptide is produced can be largely dependent upon polymorphisms in the presenilin-1 and -2 (PS1 & PS2, respectively) genes which greatly influence the activity of the -secretase. In fact, mutations in both the APP and presenilin genes (denoted as mAPP and mPS1 or mPS2 in transgenic models) can influence amyloi d plaque formation causing dramatic effects on the proclivity of individuals to develop an early-onset form of AD. APP mutations can lead to increased amyloid pathology by favoring cleavage of the APP N-terminal fragment by or -secretase rather than the -secretase (Citron et al., 1997) or by increasing the aggregative properties of the 142 peptide itself (Wisniewski et al., 1991) Mutations in PS1 or PS2 leading to modified secretase activity can result in more r apid production of A 1-42 (Selkoe, 2001). In addition to the neurodegenerative effe cts of amyloid a ccretion, other adverse consequences of the cerebral deposition of A peptide include dysregulation of Ca2+ signaling, mitochondrial da mage, free radical production (Mattson, 1997;Mattson, 2002) and recrui tment of both pro-inflammatory components and glial cells (Roses, 2000). Accumulating evidence suggests that A precipitates the cascade of pathogeni c events leading to the massive neurological dysfunction and dem entia characteristic of AD (for review see Hardy
3 and Selkoe, 2002). One example of this evidence is that A through interaction with 7 nicotinic receptors (Dineley et al., 2001), initiates the mitogen activated protein kinase (MAPK) cascade leading to its over-activation and eventual dysregulation. Downstream targets of the MAPK cascade include tau (Arendt, 2001) and several immediate early genes im portant for long-te rm potentiation and memory formation. As more evi dence becomes available, a clearer understanding of the relations hip between the various pat hological features of AD and the best strategy for its pr evention will become more apparent. Aside from the involvement of the APP and presenilin genes in AD progression, the normal role that these genes play in brain function has only been clearly defined for presenilin. PS1 prot eins are associated with the synaptic membrane and weave across it like stitches through fabric. This protein is then activated by proteolysis into two subunits which leads to its ability to cleave other membrane-associated proteins, includi ng APP and notch (neurite outgrowthpromoting cell surface glycoprotein), a cell adhesion molecule involved in neuronal development and the fate dete rmination of other stem cell types throughout life (Milner and Bigas, 1999;Ar tavanis-Tsakonas et al., 1995). The only progress made so far towards identif ying the function of APP has been that it is a ubiquitous membrane-bound phospho protein. Several hypotheses have been presented for its role, but none of these have been generally accepted (da Cruz e Silva EF and da Cruz e Silva OA, 2003). Another gene that is critically linked to the onset of AD is ApoE, a protein that is produced by astrocytes and esco rts low-density lipoprotein (LDL) through
4 aqueous environments, making it an essent ial component of the cholesterol transport system. There are th ree alleles of this gene ( 2, 3, and 4), which is located on chromosome 19, and it has been demonstrated that ApoE 4 is the strongest predictor of late-onset AD (after 60 years of age). Epidemiologic studies have revealed that i ndividuals who express the 4 allele have a greater occurrence of late-onset AD com pared to those of the more common 3 allele, while those carrying the 2 allele are protected from AD (Wahrle and Holtzman, 2003). Many putative explanations for the action of ApoE on the neuropathology of AD exist (for review see Laws et al ., 2003). For example, there is evidence that 4 out-competes A for the brains protective clearance machinery, allowing more plaque-forming peptide to aggregate (S t George-Hyslop, 2000). In addition, it has been demonstrated that ApoE 4 has poor membrane repair properties (Yang et al., 1997), perhaps indicative of why hippocampal atrophy in AD patients with the ApoE 4 allele is more severe (Lehtovirta et al., 1995). The development of transgenic and knocko ut mice has proven critical to the understanding of AD and recent develop ments have lead to the identification of several interesting therapeutic target s, including BACE1 (Ohno et al., 2004) and the secretase (Eriksen et al., 2003). Howe ver, to identify these targets, mice with pathological similarities to that of AD patients had to be generated. Mice transgenic for the mut ant forms of the human AP P (Hsiao et al., 1996) and PS1 (Duff et al., 1996) genes were generated and our group developed a viable doubly transgenic mouse model that ex presses both. These APP+PS1 mice exhibit accelerated amyloid pathology microglial activation, astrocyte
5 proliferation, neuritic d ystrophia and an impaired cogni tive phenotype (Holcomb et al., 1998;Gordon et al., 2002).Over the pa st four years, we have demonstrated that these mice develop the hippocampal and cortical A plaques similar to those found in the AD brain, and despite the lack of the intracellular fibrillary tangles and neuronal degeneration, they do exhibit memory deficits, thereby providing a good model for the study of amyloid-a ssociated memory loss (Gordon et al., 2001;Holcomb et al., 1999;Morgan et al., 2000;Dickey et al., 2001;Jantzen et al., 2002). SYNAPTIC PLASTICITY & LTP. The memory deficits exhibited by the APP+PS1 mice provide a potential utility in uncovering mechanisms of memory dysfunction. Following the application of a tetanic burst of stimulation in the hippocampus, synaptic efficacy is greatly increased for as long as several weeks, a phenomenon called longterm potentiation (LTP). LTP demonstrates that synaptic connectivity of hippocampal neurons of the perforan t, mossy fiber and Schaffer collateral pathways in the cornu ammonis layers 1 and 3 (CA1 and CA3) can be modified structurally and electrically followin g a burst of high frequency stimulation (Winder and Schramm, 2001;Zola-Morgan et al., 1986;Tsien et al., 1996;Malenka and Nicoll, 1999;Shapiro, 2001). Th is event results in the activation of N-methylD-aspartate receptors (NMDAR), a llowing the second messenger ion, Ca2+, into the cell. This is an associative proce ss requiring depolarization of the synaptic membrane via activation of aminomethylphosphonic acid receptors (AMPAR)
6 which dislodges the Mg2+ ion blocking the NMDAR por e and the binding of extracellular glutamate to open the lig and-gated NMDAR calcium channel. Modification of the pre-synaptic cell is also required for LTP and this process seems to be induced by a retrograde me ssenger released from the post-synaptic neuron, such as nitric oxi de (NO; Kandel et al., 2000). The LTP phenomenon is thought to result from changes in the morphology and receptor properties of sy napses that are dependent on Ca2+ activated mechanisms in the cell. One mechani sm of the modulat ory nature of Ca2+ may be attributed to the activation of Ca2+/calmodulin kinase II (CaMKII the catalytic subunit of CaMKII), as demonstrat ed both by the inability of neurons to induce LTP when CaMKII is specifically inhibited (Perkel et al., 1990) and the failure of targeted CaMKII knockouts to elicit LTP (Silva et al., 1992). This enzyme seems essential for the initiation (Otmakhov et al., 1997) of LTP, which can persist from hours to weeks. The di screpancy in the duration of LTP is not understood, although the transit ion from induction to maintenance seems critical. One hypothesis for the ability of LTP to be maintained is that of the silent synapse which proposes that post-synapt ically silent neuronal junctions can become activated either by recruitment of new AMPARs to the membrane or by the spillover of glutamate onto nearby NMDAR synapses (Winder and Schramm, 2001). Currently there are two accepted phases of LTP, aptly named the early (within the first 30 minutes of stimulat ion) and late (beginning 30 minutes after stimulation) LTP phases. What distingui shes the two from each other is the requirement of de novo protein synthesis in the late phase (Bolshakov et al.,
7 1997;Ma et al., 1999). A proposed synt hesis-independent intermediate phase has also been described that requires t he activation of protein kinase A (PKA) and its modulation of several key signal transduction cascades (Nayak et al., 1998). Early phase LTP, which has been functionally linked to short-term memory, requires only NMDAR and protei n kinase C (PKC) activation (Huang et al., 1996), two effects well upstream of gene transcription. When an intense late phase stimulating current such as maxi mal electro-convulsive seizure (MECS, stimulates neurons in a similar way as LTP, but more reproducibly) is administered, rapid induction of a gr oup of genes known as immediate early genes (IEGs, so named because they do not require new protein synthesis for transcription) ensues (Cole et al., 1990; Lyford et al., 1995). The regulated expression of these genes pr ovides modulatory effects on neuronal structure and function through manipulation of a network of constitutively expressed proteins, implicating them in enduring forms of LTP, and therefore long-term memory formation. An initial shortcoming of LTP was t hat it required a very intense induction stimulus (several short successive applic ations of ~100Hz) well outside of the physiological range (Tsukada et al., 1994) However, in 1999, Winder et al. demonstrated that oscillations emi tted by hippocampal neurons during exploratory behavior called the theta ( ) frequency (3-13Hz) can induce LTP, but activation of the ERK/MAPK cascade is required (Winder and Schramm, 2001;Bach et al., 1995). Another mechanism that induces synaptic plasticity is the application of low frequency stimulati on (1Hz) for a prolonged period, causing
8 long-term depression (LTD; Dudek and Bear 1992), which results in lowered excitability of the cell for prolonged periods. Ther efore, there are two mechanisms that can induce synaptic plasticity within the hippocampus, LTP and LTD, both with opposite effects. Intere stingly, a phenomenon known as synaptic history has been described in which differential phosphorylation of AMPARs results in different enzymatic specific ity, with LTP being correlated with CaMKII specificity and LTD with PKA specificit y (Mayford et al., 1995;Lee et al., 2000). Still, other forms of NM DAR-independent LTP are f ound at the CA3 mossy fiber pathway and also in the ce rebellum at the Purkinje-par allel fiber synapse (Winder et al., 1999). These mechanisms may involve the mitogen activated protein kinase (MAPK) cascade and other unk nown components. In summary, the activation of post-synaptic neurons can elicit signal transducing mechanisms (Ca2+, MAPK, etc) that event ually lead to changes in synaptic efficacy through altered gene expression. BEHAVIOR. Recent evidence reveals that seve ral activity-inducible IEGs, along with other constitutively expressed proteins, are up-regulated in animals exposed to novel environmental stimuli and tasks that require learning, further supporting the notion of gene induction mediating memory st orage. Different aspects of learning and memory, such as spatial navigation, location learning, consolidation, and retrieval, seem to require different ex pressional patterns of these genes. Animal models generated with select ive expression or disrupt ion of some of the
9 constitutively expressed LTP-associ ated genes have had varying success at linking LTP to specific types of memory (Tsien et al., 1996;Tang et al., 1999;Silva et al., 1992;Mansuy et al., 1998;Huang et al., 1996). The ability to determine defects or improvements in various forms of memory in these transgenics requires the employment of several well-accepted training batteries. The Morris water maze is most often used to assess memory retention. It places a mouse in a pool of opaque water with a small flagged platform just beneath the surface (Morr is, 1984). The mouses aversion to swimming provides the impetus to search for the platform, and once it is found, a normal mouse should be able to locate it quickly in subsequent trials by recognizing the flag as a cue. A hi ppocampal-dependent versi on of the trial incorporates the use of visual cues outside of the pool, making this a spatial navigation task. Another spatial test t hat removes water as the aversion and replaces it with an obtrusive noise and brig ht light is the Barnes maze (Barnes, 1979). Here, animals are placed in an open field with holes surrounding the perimeter, one of which exits to an escape tunnel. Cognitive abilities are measured in a similar manner as that of the Morris wate r maze, either by cuing the animal with a flag near the escape tunnel opening and measuring errors made in trials following learni ng of the flag location or by placing exterior visual cues around the field without the fl ag for hippocampal-dependent learning assessment. Contextual fear conditioni ng is another means of measuring memory that uses an aversive stimulus (normally el ectric shock) coupled to a sound. The
10 animal eventually learns to freeze in response to the sound, preventing it from receiving an additional shock. Social transmission of food preference (Kogan et al., 1997) utilizes a tester mouse that is exposed to a food with a particular scent. This animal is then readmitted to the gener al population and thes e other mice are exposed to food with the same scent used earlier and a new scent. These animals exhibit learning if they only eat the food that t he tester had eaten because they know that it is safe. One other example is novel object recognition (Mansuy et al., 1998;Mumby et al., 1996; Myhrer, 1988), in which a mouse is exposed to two objects for some time. Then, with the addition of a third object, learning is determined by increased latency with this new object. Animals such as the GluR1 knockout (KO) mouse (Zamanillo et al., 1999), the CaMKII KO mouse (Silva et al., 1992), and the calcineurin KO mouse (Mansuy et al., 1998) have demonstrat ed impaired performance in behavioral tasks and reduction in LTP phenotype. Ho wever, one particular animal overexpressing the NMDAR subunit NR2B exhi bited marked increases in both LTP and behavioral task performance, edifying the possibility of memory being influenced by NMDAR-dependent LTP (Tang et al., 1999).
11 CONSTITUTIVE PROTEIN EXPRESSION. The properties of several constitutively expressed genes make them excellent targets for future memory-relat ed studies. Aside from proteins essential for the basic functions of neurons such as the Na+/K+ ATPase, other key components of the LTP system of me mory formation are neurotransmitter receptors. The NMDAR is an ionotropic glutamate rece ptor that also requires glycine binding as a co-agonist for ac tivation, and is composed of several heteromeric transmembrane subunits. Ther e are three types of subunits (NR1, NR2 and NR3), each with a variable number of isoforms designated by a letter (for example, four isoforms of NR2 include NR2A, NR2B, NR2C and NR2D; CullCandy et al., 2001). Each receptor requi res the incorporation of several NR1 subunits and at least one NR2 subunit. NR3 can co-assemble with any NR1/NR2 complex. The NR2 subunit determines t he rate of current decay through the channel and the affinity of the receptor for glutamate, while NR1 determines the affinity for glycine. NR2A and NR2B elicit a high conductance coupled with a high affinity for extra-cellular Mg2+, while NR2C and NR2D exhibi t the opposite effects (Cull-Candy et al., 2001). Interestingl y, during developm ent NR2B and NR2D are highly expressed, but as maturation pr ogresses they are r eplaced with NR2A or NR2C resulting in a faster rate of cu rrent decay. This illustrates a possible explanation as to how c onstitutive over-expression of NR2B may result in memory enhancement (Tang et al., 1999). As previously noted, AMPARs are al so fundamental to the induction of LTP, providing enough depolarization upon ligand-binding to release the Mg2+ ion
12 that blocks the NMDAR channels. These re ceptors have several isoforms, each with varying affinities for glutamate and conductance properties. AMPARs have even been directly linked to NMDARs via modulatory proteins, allowing crosstalk between the two channels (Soderling and De rkach, 2000). In AD, AMPARs have an altered expression, and this may c ontribute to the associated dementia. Modulation of plasticity and LTP also involves the presence of sensors on the post-synaptic membrane that can mo dify synaptic connectivity. One such protein is a recently described gene pr oduct called calsyntenin-1 (CSTN-1), a transmembrane protein that is specific to post-synaptic dendrites (Vogt et al., 2001). The cytoskeletal domain binds Ca2+ and undergoes a conformational change allowing serine proteases in the synaptic cleft to cleave the extra-cellular portion. The inner subunit may then serve as a Ca2+-signaling molecule that associates with the spine apparatus at the base of the dendritic shaft, a structure that has been postulated to be responsib le for acquisition of extra membrane material during synaptogenesis. Another non-IEG protein that seems to be associated with post-synaptic plasticity is synaptopodin, a molecule that associates with actin filaments specific to the dendritic tree in the hippocampus, cortex, and striatum (Mundel et al., 1997) This protein has protein kinase C (PKC) phosphorylation site s along with a C-terminal 90 amino acid repeat domain that is common to the PDZ family of proteins (Mundel et al., 1997). Upon induction following LTP, synatopodins linear motif allows side-to-side arrangement with actin, possibly serving as a mediator betw een the cytoskeleton
13 and other associated proteins, facilitating the subsequent filopodia extension and retraction properties that PSDs exhibit. Calmodulin is an intracellular adapter molecule that binds Ca2+ ions following synaptic transmission. CaMK II is an extremely important neuronspecific serine/threonine kinase that auto-phosphorylates upon binding of the Ca2+/calmodulin complex. This autophosphorylation maintains CaMKII in an active state even after Ca2+ levels have dissipated. The subunit of CaMKII is the catalytic domain of the enzyme and t he mRNA for this gene is present at dendrites, suggesting dendritic translation by the polyribosomes located there (Steward and Levy, 1982;Steward, 1987). Transgenic mice expressing a constitutively phosphorylated, and therefore active form of CaMKII are incapable of expressing frequency induced LTP, in fact LTP is converted to LTD in these animals (Bach et al., 1995;Mayford et al., 1995), rendering them unable to perform spatial memory dependent tasks. However, these mice are able to learn from a contextual conditioning trial, l ending further credence to the notion of the significant role intermediate frequencies pl ay in fine-tuning cognitive functions. Therefore, CaMKII is e ssential to the induction and maintenance of LTP, probably attributed to its specificity for countless substrates. Among these substrates are component s of the vesicular exocytosis system involved in neurotransmitter re lease (Turner et al., 1999). Briefly, synapsin 1 attaches vesicles to actin f ilaments of the axonal cytoskeleton until phosphorylated. Once released, two protei ns on the plasma membrane, syntaxin and SNAP-25, bind synaptobrevin on the vesicular membrane. The Ca2+ sensor,
14 synaptotagmin, is also located on the ve sicle and allows release only if Ca2+ is bound (Brunger, 2000). Not only are all of thes e pre-synaptic proteins a substrate for CaMKII, reiterating its powerful c apability to modulate all aspects of neurotransmitter release and therefore LTP induction, but many post-synaptic substrates exist as well, such as AMPAR s and other receptor s, explaining its presence at the PSDs, and demonstrating its role in LTP maintenance. IEG EXPRESSION. In 1990, Cole et al. created a phage library from MECS stimulated hippocampal cells, which subsequently lead to the identification of several genes that seemed to be maximally expressed in response to this LTP-like stimulation (Cole et al., 1989). Previously, it was dem onstrated that ubiquit ous transcription factors such as c-fos (Morgan et al., 1987) and zif 268 (Cole et al., 1989) are induced within a short time of seizur e activity and a progressively stronger stimulus increases the abundan ce of these IEGs mRNAs. The library created by Cole et al. revealed 15 activity-inducible IEGs, half being transcription factors, the other half being previously undescribed pr otein effector genes able to influence processes such as cell and neurite growth, signal transduction and cytoskeletal rearrangement (Lanahan and Worley, 1998). The transcription factors from this gr oup of 15 include mostly ubiquitous transactivators, such as cfos, c-jun (which together form AP-1), Nur77 (NGFI-B), Egr3 and Zif268 (Egr1/NGFI-A), both of the latter being zinc-finger transcriptional regulators. Both Nur77 and Zif268 are of interest because their mRNA can be
15 induced rapidly by nerve growth factor (N GF). Zif268 is of particular interest because its mRNA is rapidly upregulated bo th in the hippocampus of rats after exposure to a novel stimulus (Hall et al., 2000) and in the temporal gyrus of monkeys following associative learning tasks (Miyashita et al., 1998). Additionally, viable heterozygous Zif268 knockout mice ex hibit both long-term memory deficits and more rapid decay of LTP than wildtype littermates (Jones et al., 2001). The Zif268 promoter contains 2 cA MP responsive elements and therefore it is assumed that Zif268 transcription is modulated by CREB (Changelian et al., 1989;Sakamoto et al., 1991). In addition, the MAPK cascade appears to regulate Zif268 transcription following LTP induc tion (Davis et al., 2000). Zif268 expression is also tempered by the negativ e regulator NAB-2, a Zif268-specific transcriptional repressor protein that is ve ry similar to Stat6 (Abdulkadir et al., 2001). Therefore, there is both a constitutive (CREB) and inducible (MAPK cascade) mechanism controlling Zif268 expr ession, neither of which require any de novo protein expression, while phosphor ylation or dephosphorylation of NAB2 may well be the switch that determines when an increase in Zif268 mRNA is necessary. Zif268 binds the consensus GCG(T/G)GGGCG and can induce the expression of many cellular factors, incl uding growth factors, cytokines and cell adhesion molecules, that are essential fo r the specificity involved with synaptic plasticity (McCaffrey et al., 2000;Khac higian et al., 1996;Kramer et al., 1994;Maltzman et al., 1996), pr operties that warrant the tight regulation of Zif268 expression.
16 Nur77 mRNA, in addition to NGF responsiveness, has also been shown to be induced by electroconvulsive seizures (Watson and Milbrandt, 1989), amphetamine administration (Werme et al., 2000;St Hilaire et al., 2003) and other stress-inducing stimuli (Kovacs and Sawchenko, 1996). It is linked to inhibitory processes in the dopaminer gic system assisting in molecular mechanisms responsible for fine control of motor function (Drolet et al., 2002). Nur77 is an orphan thyroid hormone re ceptor gene that undergoes rapid posttranslational modification and phosphorylati on (Fahrner et al., 1990). Additionally, it binds to DNA consensus sequences sim ilar to that of thyr oid hormone receptor, however, unlike other known members of the thyroid hormone receptor superfamily, it does not require dimerizati on in order to bind as a transactivator (Wilson et al., 1992). The role of Nur77 in memory consolidation has not been well established, and it remains unclear as to whether this gene is linked to the learning process or simply the stre ss elicited by exposure to behavioral paradigms. The group of effector IEGs elicited by MECS includes several proteins with extremely novel characteristics. activin is a transforming growth factor implicated in neuronal survival that is predominately involved in developmental processes (Andreasson and Worley, 1995). Rheb, a Ras homologue, is a small GTP-binding protein that interacts with Ra f kinase and seems to integrate growth factor signals with PKA-signaling cascades specifically in PSDs where both of these components are activated (Yamagata et al., 1994). Therefore, Rheb may utilize the hierarchical Ras/Raf cascade to provide highly localized responses to
17 many growth factors, since signaling would be restricted to those molecular layers with activated PKA. Another e ffector IEG with uni que properties is a member of the pentraxin family called Narp. This protein is thought to be secreted into the synaptic cleft and prom ote clustering of preand post-synaptic AMPA receptors (O'Brien et al., 2002), ther efore altering plasticity and possibly establishing the lasting cellular c hanges necessary for long term memory formation. Perhaps the two most intriguing mem bers from this group of IEGs are an inducible form of a metabotropic glutamat e receptor binding protein called Homer that is found only in activated synaps es and the activity-regulated cytoskeletal associated protein (Arc). Three Home r genes encode Homer-1, 2, and 3 proteins (Bottai et al., 2002), all modulators of intracellular Ca2+ release. The three Homer proteins and their isoforms all have an EVH1 domain specific for a proline rich motif present in metabotropic glutam ate receptors (mGluR1 and mGluR5), ryanodine receptors, IP3 receptors and S hank proteins. This domain links mGluRs and NMDARs with intracellular Ca2+ stores. Other Homer functions include regulation of 1) spontaneous acti vity innate to mGluR1, mGluR5 and K+ channels, 2) directed axonal outgrowth, and 3) dendritic spine morphology. In particular, the Homer-1 gene gives rise to two IEG products following neuronal activity, Homer-1a and Ania3, by a novel conversion of an intronic sequence to a coding exon (Brakem an et al., 1997). The inducible IEG promoter contains a CRE-site, once again implicating CREB and the MAPK signaling cascade in LTP-mediated gene induction. These
18 IEG products are much smaller than t he constitutive Homer-1 gene isoforms, Homer-1b and 1c, which express a bulky oligomerization dom ain that seems critical in preventing their migration acro ss the axon hillock (Bo ttai et al., 2002). LTP requires a precise ratio of mGluRs in both the postand pre-synapse, and crosstalk between pre-synaptic mGluR5 subunits and post-synaptic NMDARs via Shank proteins has been demonstrated (Ang o et al., 2000). Therefore Homer-1a, lacking the restrictive mult imerization domain, compet es with the constitutive isoforms for binding sites on the C-termi nus of newly synthesized metabotropic glutamate receptors and serves as a hom ing signal for the mGluRs to the presynapse. In fact, the more LTP-induci ng stimulus applied to the neuron, the larger the induction of Home r-1a protein compared with -1b and -1c, resulting in more mGluRs being targeted to the dendrit es. Once the receptors are delivered, Homer appears to dissociate and be targeted for degradation, attributed to an ubiquination domain within the IEG produc t. The presence of specific mGluR subunits at the pre-synaptic membrane may indicate both their functional significance as synaptic glutamate sens ors and their ability to potentiate glutamate release in an extra-cellul ar concentration-dependent manner. These features, coupled with the ability of mG luR5 to crosstal k with post-synaptic NMDARs, may demonstrate that LTP is maintained through direct interaction between the preand postsynaptic neuron, underscoring the importance of the inducible IEG Homer-1a in mediati ng this positive feedback mechanism. Arc exhibits the most unique traits of any of the IEGs elucidated from the MECS library. Following an LTP induci ng stimulus, Arc mRNA is rapidly
19 upregulated in a matter of 15 minutes wit hin the cell body layer of neurons. By 30 minutes, the mRNA is detected near the dendr itic spines. By 1 hour Arc protein is detected in distal branches of the dendr itic trees, demonstrating that Arc is translated locally on the polyribosomes pr esent at the dendrites (Lyford et al., 1995). These polyribosomes are dormant except at synapses with activated mGluRs, therefore local translation at the dendrites is only possible when those synapses are active (Weiler and Greenough, 1993;Steward and Worley, 2001;Steward, 1995), prov iding selective expression of transcripts such as Arc. Other mRNAs that ar e translated by dendritic ribosom al rosettes are the catalytic domain of CaMKII and microtubule-associ ated protein 2 (MAP2), both of which are derived from constitutively express ed genes (Wallace et al., 1998). This leaves Arc as the only identif ied inducible IEG product that is locally translated at activated synapses. Recent studies have provided additi onal information about Arc that makes it an intriguing candidate for memory-related studies. Guzowski et al. injected Arc antisense oligonucleoti des into the brains of rats and using a spatial water maze task demonstrated that long-term memory consolidation wa s significantly impaired, while short-term cognition was not (Guzowski et al., 2000). Also, in concordance with the idea that early phase LTP corresponds to short-term memory, and late phase LTP (which requires new protein synthesis) with longterm memory, animals injected with Arc antisense oligos were able to undergo induction of LTP, but were unable to maintain it. Arc induction also seems dependent upon NMDAR activation, while AMPAR stimulation alone will not
20 produce a rise in expression (Steward and Worley, 2001). Additionally, animals exposed to an enriched environment have increased hippocampal levels of Arc expression, and animals most proficient at spatial tasks have had correlated elevations in Arc mRNA levels (Mont ag-Sallaz et al., 1999;Pinaud et al., 2001;Guzowski et al., 2001). The expr ession of Arc appears to be highest following a novel task, as demonstrated by a decrease in expression in mice over-trained for the same trial. Likewise, by simply changing the location of the platform in the water maze, Arc levels become elevated (Kelly and Deadwyler, 2002). All of these data correlate well wit h the notion that some sort of enduring synaptic rearrangement is essential for the consolidation of long-term memory. Taken together with the proper ties of the Arc protein and its expression patterns, these findings belie the possible mechanist ic role of Arc in memory formation. Arc co-sediments with F-acti n macroaggregates and bears some homology with -spectrin, further implicating it in mediation of synaptic plasticity. The protein is a 45kD hydrophilic prot ein with several phosphorylation sites including sites for PKC and CaMKII. Arc is expressed basally at low levels, with the thought being that this protein is induced during any NMDA-mediated synaptic activity. This relationship was es tablished by findings of a reduction in the basal mRNA levels following NM DAR antagonist, MK-801, administration (Lyford et al., 1995), accompanied by a robust increase when animals receive methamphetamine (Kodama et al., 1998). The expression of Arc mRNA is almost exclusive to the brain, and predominates in the hippocampus and cortex. Cortical levels are actually higher than those of the hippocampus, fu rther implicating a
21 functional relationship with long-term me mory storage through LTP maintenance and lasting synaptic changes in the PSDs of cortical neurons. Following an LTP inducing stimulus, the rapid induction of mRNA reaches a maximal level within 1 hour and this peak is maintained for 8 hours, followed by a return to basal expression levels within 24 hours. A signal within the mRNA itself s eems the mechanism for the selective transport of particular mRNAs to the dendr ites. This trafficking is a two-step process consisting of two types of signals within the molecule; a routing signal, for selective nuclear retention or mole cular layer transport, and a localization signal, that results in the selective accumulation of mRNA at a particular intracellular site (Stewa rd et al., 1998). Both Ca2+ and cAMP induce Arc transcription, and this inducibility is dependent upon the activation of PKA and the ERK/MAPK cascade (Waltereit et al., 2001). Interestingly, the Arc promoter lacks a complete putative CRE site, tent atively eliminating CREB transcriptional activation, a common downstream target of the MAPK pathw ay. The promoter does however contain an AP-1 binding site, a transcription factor composed of the two IEGs, c-fos and c-jun, possi bly indicating a potential regulatory mechanism that provides activitydependent modulation of transcriptional activation. Growth factors have also been implicat ed in LTP-promoted Arc induction, namely insulin (Kremerskothen et al ., 2002) and brain-derived neurotrophin factor (BDNF;Yin et al., 2002). Insulin s eems to mediate Arc expression through the ERK/MAPK cascade, and BDNF, know n to activate CREB and the MAPK
22 cascade, has also been linked to an incr eased expression of Arc. Therefore, Ca++ influx that activates PKA and subsequently the MAPK cascade, may mediate the CREB-independent activation of Arc transcription, emphasizing the possibly significant roles that effector IEGs and the ERK/MAPK cascade play in cognitive retention. THE ROLE OF NA+/K+ ATPASE IN COGNITIVE FUNCTION. The role of the membrane-bound Na+/K+ ATPase is to transport Na+ ions out of the cytosol and K+ ions in to the cell against their concentration gradient thereby establishing a me mbrane potential to allow subsequent depolarizations. This process requires energy input that is derived from ATP hydrolysis. The importance of this enzyme in cognitive pr ocesses is belied by its consumption of ~40% of total cerebral ATP. It has been previously establis hed that specific inhibition of Na+/K+ ATPase by oubain, prevents the consol idation of memory (Mark and Watts, 1971;Watts and Mark, 1971), and therefore it is postulated that the ability of neurons to conduct action potent ials or facilitate LTP is critically dependent on the integrity and func tionality of this enzyme. In addition to 3 tissue-specific subtypes of the Na+/K+ ATPase, there are two major subunits that comprise each molecule; a catalytic subunit and a membrane anchoring subunit. Very recent evidence indicates the presence of a cerebral subunit, a member of the FXYD fam ily, which positively regulates the activity of Na+/K+ ATPase (Crambert et al., 2003). This subunit is a single transmembrane protein, however much of its function has yet to be determined.
23 The activity of Na+/K+ ATPase can be increased or decreased by pharmacologic intervention with the anti-conv ulsant phenytoin or the cardiac glycoside digitalis, respectively. Phenytoin decreases the intr acellular concentration of Na+ ions by two possible mechanisms; inhibition of voltage gated Na+ channels by locking them in an inactivated state (Worley and Baraban 1987;Kuo and Bean, 1994) and/or the hyper-activation of Na+/K+ ATPase (Gutman and Boonyaviroj, 1977) by preventing the phos phorylation of the subunit and thereby blocking its internalization (Guillaume et al., 1989).
24 TABLE 1. Overview of gene fuction Gene Name IEG Vesicle Release LTP Induced Synaptic Plasticity Memory Induced Known Dysregula tion in AD PreSynaptic PostSynaptic ApoE Homer-1a Calsyntenin-1 Zif268 Arc Nur77 Narp Na+/K+ ATPase AMPA1 CaMKII Gap43 GAPDH GFAP NMDA2B Synapsin-1 Syanptopodin Synaptotagmin NAB2 Syntaxin Vimentin Neurofilament M Synaptophysin
25 PAPER I: SELECTIVELY REDUCED EXPRESSION OF SYNAPTIC PLASTICITYRELATED GENES IN APP+PS1 TRANSGENIC MICE Chad A. Dickey1, Jeanne F. Loring2, Julia Montgomery2, Marcia N. Gordon1, P. Scott Eastman2 & Dave Morgan1 1 Alzheimers Disease Research Laborat ory, Department of Pharmacology, University of South Florida, College of Medicine 12901 Bruce B. Downs Blvd, MDC 9, Tampa, Fl 33612 2 Department of Life Sciences, Incyte G enomics, Inc., 3160 Porter Drive, Palo Alto, CA 94304, USA These authors contributed equally to this manuscript Published in Journal of Neuroscience 2003 Jun 15;23(12):5219-26. ACKNOWLEDGEMENTS: Dr. Tom Beac h generously provided human tissue from the Brain Donation Pr ogram at Sun Health Resear ch Institute in Sun City, Arizona. This work was support ed by NIA grants AG15490 and AG18478.
26 ABSTRACT A critical question in Alzheimer's disease (AD) research is the cause of memory loss that leads to dementia. The APP+ PS1 transgenic mouse is a model for amyloid deposition, and like AD, the mi ce develop memory deficits as amyloid deposits accumulate. We profiled gene expression in these transgenic mice by microarray and quantitative RT-PCR (qRT-P CR). At the age when these animals developed cognitive dysfunction, they had reduced mRNA expression of several genes essential for LTP and memory formation (Arc, Zif268, NR2B, GluR1, Homer-1a, Nur77/TR3). These changes appeared to be related to amyloid deposition, as mRNA expression was unchanged in the regions that did not accumulate amyloid. Transgene expr ession was similar in both amyloidcontaining and amyloid-free r egions of the brain. Interestingly, these changes occurred without apparent changes in synapt ic structure, as a number of presynaptic marker proteins (GAP-43, synapsin, synaptophysin, synaptopodin, synaptotagmin, syntaxin), remained stabl e. Additionally, a number of genes related to inflammation were elevated in transgenic mice, primarily in the regions containing amyloid. In AD cortical ti ssue the same memory-associated genes were down-regulated. However, all sy naptic and neuronal transcripts were reduced, implying that the loss of neur ons and synapses contributed to these changes. We conclude that reduced expr ession of selected genes associated with memory consolidation are linked to memory loss in both circumstances. This suggests that the memory loss in APP+PS 1 transgenic mice may model the early memory dysfunction in AD prior to the degeneration of synapses and neurons.
27 INTRODUCTION Memory loss is an early and progressive symptom of Alzheimers disease (AD). During the period of cognitive decline, pathological hallmarks such as amyloid plaques and neurofibrillary tangles become evident (Selkoe and Hardy, 2002). In the later stages of AD, ther e is a profound loss of synaptic markers, but early in the disease this loss is modest, and in some instances not discernable (Mukaetova-Ladinska et al., 2000;Tir aboschi et al., 2000;Masliah et al., 2001;Minger et al., 2001). A clue to under standing the basis of early memory loss in AD may come from transgenic mice that develop impaired memory function associated with amyloid deposition, but never show the extensive loss of synapses or neurons typical of AD. The APP+PS1 mouse model of amyloi d deposition develops a memory loss that is consistently observed by 15 months of age (Morgan et al., 2000;Arendash et al., 2001) and is corre lated with the extent of amyloid deposition. Like AD, these mice deposit am yloid primarily in the cerebral cortex and hippocampus, leaving the brains tem and cerebellum largely unaffected (Holcomb et al., 1998;Holcomb et al., 1999;Gordon et al., 2002). To assess changes that occur in the brain in conj unction with amyloidassociated memory loss we used microarray analysis and qRT-P CR to survey the genes that were upor down-regulated in amyloid-cont aining regions of APP+PS-1 transgenic mouse brain. We found that the amyloid-containing r egions of the transgenic mouse brain had high levels of expre ssion of inflammation-associated genes, a characteristic that is also typical of AD brain. However, the most intriguing part of
28 the gene expression profile was a selectiv e decrease in transgenic mice of genes known to be important in long-term potent iation (LTP) and memory consolidation, restricted to amyloid-containing brain regi ons. Expression of genes involved in synaptic and neuronal structure was severe ly diminished in the human AD brain, but the transgenic animals did not show dec reases in any of these genes. This suggests that the memory deficits in the early stages of human AD may be the result of dysfunctional changes that precede the frank loss of synapses. METHODS Materials Mice were bred in our facility and genotyped using previously described methods (Gordon et al., 2001). All mice were 17-18 months of age at time of death. Mice were deeply anesthetiz ed with pentobarbital (100 mg/kg) and perfused transcardially with phosphate bu ffered saline. Brains were quickly removed and amyloid-containing (cor tex and hippocampus) and amyloid-free areas (cerebellum, striatum, and brai nstem) were immediately dissected and frozen on dry ice. For microarray analys is, cortex and hippocampus were pooled and dealt with as amyloid bearing sample s, while cerebellum, striatum and brainstem were combined and treated as amyloid-free tissues. For qRT-PCR, hippocampal RNA alone was isolated and r egarded as amyloid-containing while cerebellum alone was used for amyloi d-free analysis. Transgenic and nontransgenic animals were sacrificed in r andom order to minimize unintentional bias in handling of the samples. T he working memory performance of the
29 APP+PS1 mice used in these studies was si gnificantly different from their nontransgenic littermates used as cont rols (Austin et al., 2003). Fresh-frozen human cerebellar and medial temporal gyrus (MTG) specimens were obtained fr om the Brain Donation Pr ogram at Sun Health Research Institute in Sun City, Arizona. Subjects with a clinical history of dementia were diagnosed as AD usi ng neuropathologic consensus criteria, including those published by CERAD (Mi rra et al., 1991) and the NIA-Reagan Institute (1997). Control subj ects did not have a clinical history of dementia and did not meet neuropathologic cr iteria for AD or other neurologic disorders. The average postmortem intervals were 2.6 h for AD and 2.3 h for control specimens (6 Female, 2 Male for AD; 6 Female 2 Male for age matched control). The average age for the AD specimens wa s 87 years, and the average age for the controls was 88 years. Microarray analysis Brain specimens from four 17-18 mo old APP+PS1 tr ansgenic mice or four non-transgenic littermates were grouped into two categories; amyloid-containing (cortex and hippocampus) and amyloid-free areas (cerebellum, striatum, and brainstem). Messenger RNA was isolat ed from individual tissue samples, labeled with Cy3 or Cy5, and used for competitive hybridization to cDNA microarrays as described (Loring et al., 2001). Samples were analyzed pair wise (matched areas from transgenic and non-tr ansgenic animals) on a total of 16 microarrays. The microarrays (8 Rat LifeArray 1 and 8 Rat LifeArray 2; 4 each
30 for the amyloid-containing region samples and the amyloid-free regions) used for this study were constructed from a collection of 15,981 rodent cDNAs representing 10,060 different genes (Incyte Genomics, In c). All of the sequences differentially expressed in the present study are mo re than 90% identical in mouse and rat and competitive hybridization of rat and m ouse brain transcripts to these microarrays showed that 89% of t he cDNA clones hybridized mouse and rat samples equally (data not shown). Microarrays constructed from the same library are currently available commercially from Agilent Technologies (Palo Alto, CA). Data collected from the microarray hybridizations were subjected to two low-frequency data correction algorithms to compensate for systematic variations in data quality as described pr eviously (Yue et al., 2001). To minimize the detection of fals e positive signals (Type I statistical errors), the data were subjected to a tw o-stage criterion to identify genes of interest. The limit of detection of differ ential expression (LDDE) was calculated to be 1.4 fold by multiple quality control hybridizations as previously described (Yue et al., 2001). This LDDE indicates t hat in a single hybridization of an RNA sample against itself a 1.4 fold differ ence occurred with a P < 0.01. To further limit the number of false pos itive genes identified, we r equired that in 3 of the 4 individual hybridizations the differential expression value had to exceed 1.4 for a given gene to be analyzed further. For the second criterion, the statistical significance of the average differentia l expression value for any gene was calculated using Z-scores and a NORMSD IST function, and was required to exceed a value of 1.4 (not simply 1.0) with a P < 0.05. In cases where a given
31 gene was represented by multiple sequenc es on the microarray, all data were combined for statistical analysis. Deta iled analysis indicates that some sequences on microarrays are more variabl e in self-self hybridizations than others, leading to multiple false positi ves if one simply uses a differential expression value cutoff without sufficient repl icates for statistical determinations. Moreover, because the variance estimate s are sequence-specific, the use of Zscores based upon the variance estimate of all spots on the array is not always appropriate, and should not be used as the only criterion in determining differentially expressed genes. One caveat of using these stringent criteria is an increased probability of Type II statistical errors (false negatives). Still we believe this approach will identify those s equences which are most robustly and consistently modified. For samples exhibiting significant differential ex pression in the amyloidcontaining and amyloid-fr ee brain areas of trans genic animals, a one-way analysis of variance was performed A planned post-hoc comparison of means (least significant difference) was used to determine if the expression was significantly different in the amyloid-c ontaining vs. the amyl oid-free brain areas for individual genes. Each of the mi croarray clones that showed differential hybridization was unambiguously annotat ed by comparing its sequence (200-700 bp) to GenBank using the BLAST 2 algorithm.
32 qRT-PCR Total RNA was prepared from diss ected cerebellar and hippocampal tissue of 8 APP+PS1 mice and 8 non-tr ansgenic littermates that were 18-19 months old. Memory deficits and amyl oid burden were established in the transgenic mice by radial arm water maze testing and immunohistochemical methods, respectively (Austin et al., 2003). Human temporal cortex (medial temporal gyrus; Brodmans area 21) and cerebellar tissue samples from 8 AD patients and 8 age-matched controls were pu lverized by mortar and pestle on dry ice and 10-30mg of this powdered tissue was used to extract RNA. The homogenate from rotor-stator emulsification (Tissuemizer) of all tissues used were applied to RNeasy mini-spin co lumns (Qiagen) with on-column DNase treatment followed by el ution with RNase-free water, according to the manufacturers specifications. All total RNA samples were then reverse transcribed with a 1:1 mixt ure of oligo dT (25ng/l Invitrogen) and random hexamers (2.5ng/l, Invitrogen) to provi de ample cDNA synthesis from both 18S rRNA (Schmittgen and Zakrajsek, 2000) and poly-adenylated mRNA. The final reverse transcription reaction included temp late (described below), 1M betaine (Sigma-Aldrich), providing more heat labi lity to nucleic acid and protein thermostabilization, 1x cDNA first-strand synt hesis buffer (Invitrogen), 7mM MgCl, 1mM dNTPs (Each A,G,C,T, Invitrogen), 40 uni ts of RNaseOut (Invitrogen), 3mM DTT (Invitrogen), and 25 units of recombinant Su perscript II reverse transcriptase (RT, Invitrogen). The reaction was brought to 20l with water and then incubated for 15 minutes at 25 C, followed by 30 minutes at 42 C and then 30 minutes at
33 60 C. The reaction was then heated at 95 C to denature the enzymes and stop the reaction. A standard curve was estab lished within the reverse transcription reaction by adding total RNA (template) from an intra-experimental mouse or human RNA pool covering 3 logs (50, 20, 10, 5, 2, 1, 0.5 & 0.2ng) to separate wells. Two mass quantities (10 & 2ng) of total RNA from all samples being investigated were added to individual wells within the re verse transcription reaction for comparison to the standard cu rve. The RNA standard curve verifies the linearity of the RT-PCR r eaction and controls for slight inefficiencies in the transcription and amplification steps of the procedure (below). Primer pairs for qRT-PCR were generated to amplify ~100bp fragments of the gene of interest using the web-bas ed applications Prim er3 and the Oligo Toolkit at http://www.operon.com/. These ol igos were tailored according to the species being analyzed. Prim ers were initially optimiz ed using a PCR reaction followed by agarose gel analysis. If ethidi um bromide staining revealed a single band, the primer concentrations were opt imized by comparing at least three concentrations of each primer and noting wh ich combination was most efficient at generating a PCR product (using SYBR gr een detection, see below), without producing signal in control wells lacking te mplate. Experimental wells containing cDNA 25l PCR reactions were run in triplicate consisting of a master mix containing 12.5l of 2x SYBR Green Mast er Mix (Applied Biosystems), 0.25-4.5l of forward and reverse primers in varied co mbinations, either 2l of cDNA from an RT reaction or 2l of water, to control for non-specific amplif ication due to selfpriming or contamination. The remainder of the 25l volume was achieved by
34 adding water. 96-well plates were mixed by pipetting and then centrifuged. Twostep PCR was run on the MJ Research Op ticon (Boston, MA) as follows; 1 cycle of 95 C for 15 minutes followed by 40 cycles of 95 C for 15 seconds and 60-65 C for 1 minute (primer annealing temperatur e is decided according to the Operon website calculations). This was followed by melt curve analysis beginning at 55 C and increasing by 1 C to 100 C every 10 seconds, with fluorescence measured at every interval. None of the primer pairs demonstrat ed more than 1 peak of fluorescence as derived from the Opti con software, indicating a single gene product without primer-dimer formation. The target gene primers dem onstrated similar amplific ation efficiencies as compared with 18S ribosomal RNA, allo wing for quantitation of fold-change using 18S signals to normalize the results for the quantity of starti ng RNA. The slope of the regression line for the standard curve determined efficiency, which varied by less than 5% for all amplicons. Samples we re run in triplicate, with both the 10ng and 2ng RT reactions represented for each. Standard curve samples were included on all plates to avoid errors due to minor plate-to-pla te variations in amplification efficiency. The standard curve was calculated by plotting the threshold cycle (Ct, the point at which rela tive fluorescence exceeds a fixed value above background against the log nanogram quantity of RNA added to the RT reactions). A linear regression was perfo rmed and the slope-relating Ct to log RNA was calculated. T he average Ct for each sample was then used to determine the corresponding log ng of standard RNA using the slope of the standard curve. These logarithmic va lues were then converted to a mass
35 quantity of standard RNA acco rding to Ct. These mass values for the genes of interest were then divided by the 18S va lues of the same sample to determine fold-change in expression relative to the standard RNA pool. For each region these fold change values for samples in the experimental (transgenic or AD) and control (non-transgenic or age-matched non -diseased) groups were analyzed for significance using 1-way ANOVAs. The differential expression value between regions was similarly com pared using one-way ANOVA. RESULTS The mice used for analysis were 17-18 monthold APP+PS1 transgenic mice that demonstrated impaired wo rking memory performance compared to non-transgenic littermates (Morgan et al., 2 000;Austin et al., 2003). Different mice were used for microarray and qRT-PCR studies. We first used a competitive hybridization approach to co mpare transcripts from transgenic mouse brain with matched non-transgenic tiss ues on cDNA microarrays. Instead of assigning an arbitrary cut-off value to detect changes in regulation, the microarray analysis was based on multip le arrays with independent samples and statistical comparisons to determine those genes that were differentially expressed (see Methods). Using the rigor ous criteria for the microarray analysis set forth above (see Methods), only fi ve genes demonstrated significantly decreased expression in the amyloid-containing areas (h ippocampus plus cortex for the array studies) of the transgenic animals (Table 1). This group of genes included three genes that are essential for normal memory function (Arc, Zif268,
36 Nur77/TR3) and two associated with neuronal /synaptic activity (Na+, K+ ATPase III, calsyntenin). Except for Zif268 (s ee below) these reductions were restricted to amyloid-containing brain regions, and di d not occur in regions from the same mice that were amyloid-free (cerebellum, st riatum, brainstem in the array studies; Table 1), even though transgene expressi on was high in these regions (see below). Using qRT-PCR, we saw a similar patte rn of specific amyloid-associated down-regulation for most of the genes in this group (Table 1). Because of the 50% reduction in the memory-associa ted genes Zif268 and Arc found with the microarray, we expanded t he qRT-PCR analysis to include other genes related to synaptic function and memory processes (Fi gure 1). This analysis showed that the transcript for the ionotropic glutam ate receptor 1 (GluR1/AMPA1) was downregulated by 30% and the NMDA receptor subunit 2B (NR2B) RNA, known to be critical for cognitive processes (Tang et al., 1999), was decreased by 18% in amyloid-containing tissue (hippocampus alone for qRT-PCR studies). Expression of Homer-1a, which is a member of the Homer family of metabotropic glutamate receptor binding proteins (Brakeman et al., 1997) was reduced by 40%. Not only were these genes significantly down regul ated in comparison to non-transgenic mice, they were also decreased signific antly compared to t he expression in amyloid-free tissue (cerebellum only for qRT-PCR studies) for the same animals (bracket above bars in Figure 1). Only one of the synaptic plasticity-related genes was significantly down-regulated in the amyloid-free region; Zif268 expression
37 measured by qRT-PCR was 18% lower in the cerebellum, which was still significantly less reduced than in the hippocampus (Figure 1). Not all of the synaptic plasticity g enes we examined were changed in the transgenic mice. Expression of synaptopod in (Mundel et al., 1997), which is argued to be involved in synaptic plasticity, was not changed, nor was GAP-43, also involved in synaptic plasticity and sprouting (Routtenberg et al., 2000), suggesting that not all synaptic-plasticit y genes were modified in the transgenic animals. Other synaptic/neuronal markers were also expressed at normal levels in the transgenic animals, including synaptophysin, neu rofilament-M, synapsin-1 and synaptotagmin V. The st ability of these largely presynaptic markers was also observed in the results from the microarray study (Table 1). The microarray analyses revealed an up-regulation of a group of genes associated with inflammation and the acute phase response (Table 2). We quantified the up-regulation of three of these genes by qRT-PCR (GFAP, ApoE, and vimentin). Expression of glial fibrillary acidic pr otein (GFAP), an astrocytespecific intermediate filament, was increas ed by 6.3 fold in the APP+PS1 mice compared with non-transgenic lit termates (Table 3). This reflects the astrocyte proliferation known to occur in the APP+PS1 animals (Holcomb et al., 1998) and is consistent with earlier immunoassay resu lts measuring this protein (Gordon et al., 2002). Additionally, another non-neuronal intermedi ate filament, vimentin, was up-regulated by more than two-fold and Apolipoprotein E (ApoE), another acute phase protein, was up-regulated to a lesser extent. These data are consistent with earlier reports describing reproducible glial activation and acute
38 phase reactivity in the APP+PS1 mice (H olcomb et al., 1998;Matsuoka et al., 2001; Gordon et al., 2002). Most of these genes were unaffected in the areas of the transgenic brain that were relatively free of plaques. However, a few of the genes were also upregulated in non-plaq ue-bearing regions (Table 3). Two of these genes were highly expressed bec ause they contained sequence that was homologous to the sequence of the APP transgene. The APP transcripts on the microarray and the array sequences corre sponding to the prion protein 5 untranslated region (part of the transgene promoter constr uct (Hsiao et al., 1996) were elevated to the same extent in both the amyloid-cont aining and amyloidfree regions. Sequences on the microarra y that encoded other regions of the prion gene were unaffected in transgenic an imals (not shown). Notably, except for the transgenes and two ot hers, the elevation in the amyloid containing regions was significantly greater t han in the amyloid-free region s of the transgenic brain (Table 3, asterisks). Coup led with the absence of upregulation in amyloid-free regions for most of the genes in Table 2, it seems likely that the modest upregulation of several inflammation rela ted genes in these amyloid-free regions reflects a lower level of generalized inflammation in these areas. We considered it critical to compare the expression of the synaptic marker genes in the mouse model with specimens from AD patients. A previous microarray analysis (Loring et al., 2001) indicated that many markers of synapses were decreased in the amyloidcontaining areas of AD brain. We extended this analysis by qRT-PCR m easurements of a group of synaptic markers in the medial temporal gyrus of the cerebral co rtex (MTG; Brodman's
39 area 21) and cerebellum of 8 AD pati ents and 8 cognitively normal age-matched controls. As shown in Figure 2, all of the postsynaptic plasticity associated transcripts we analyzed were significant ly under-expressed by more than 50% in the amyloid-containing cortical region, while the amyloid-free cerebellar tissue lacked any significant change in expressi on of the same genes, consistent with the relative sparing of t he cerebellum in AD. In this manner the pattern of changes in transgenic mice and AD cases was similar. However, in contrast to the observations we made for the mouse, in samples of AD MTG all of the other synaptic markers (synaptophysin, synaps in, synaptopodin, synaptotagmin, GAP43) were also reduced, as was neur ofilament M. Simila r to the transgenic mouse, GFAP was elevated in the amyloidcontaining region of the AD brain. Also similar to the mouse, there were no significant changes in the amyloid-free human cerebellum, coupling amyloid to t he apparent loss of mRNA in the AD samples. These data are consistent wit h the considerable evidence showing neurodegeneration and synaptic lo ss in amyloid-containing regions of late stage AD. The results also highlight the simila rities and differences between the severe loss of most cognitive function in AD and the relatively selective memory loss in amyloid-depositing transgenic mice wh ich do not undergo neurodegeneration (Matsuoka et al., 2001). DISCUSSION The results of our microarray and qRT-PC R study show that the expression of several postsynaptic genes known to be in tegral for the establishment of LTP
40 and long-term memory, was selectively down-regulated in the amyloidcontaining regions of memory-defici ent APP+PS1 mice. The amyloid-free regions of these transgenic mice do not exhibit the same down-regulation. Additionally, several presynaptic genes that are often used as synaptic or neuronal markers are unaltered in these ani mals, similar to recent microarray findings in APP transgenic mice at an age preceding amyloid deposition (Stein and Johnson, 2002). These findings indicate that the failure of APP+PS1 mice to consolidate information for future recall may be precipitated by the amyloid dependent down-regulation of genes known to be critical for cognitive function. Late phase LTP is thought to corr espond to some forms of long-term memory consolidation because of the requi rement for de novo protein synthesis for both processes (Morris, 1998). In APP transgenic mice, the formation of LTP is impaired in some, but not all studi es (Chapman et al., 1999;Larson et al., 1999;Fitzjohn et al., 2001). Recently it was found that oligomeric forms of the A peptide were particularly effective at impairing LTP when injected into the hippocampus in vivo (Walsh et al., 2002). Of particular relevance are previous observations that the plasticity-related genes we examined are essential for late phase LTP and long-term memory formation (Bolshakov et al., 1997;Ma et al., 1999). Antisense oligonucleotides against Ar c mRNA intracranially injected into rats eliminated both late phase LTP and long-term memory formation without affecting short-term forms of both processes(Guzowski et al., 2000). Mice with a targeted inactivation of Zif268 also lack the ability to express long-term synaptic and behavioral plasticity, although short-term forms of plasticity remain intact
41 (Jones et al., 2001). It has been report ed that over-expression of the NR2B subunit leads to improved memory functi on (Tang et al., 1999), while intracranial injection of antisense oligo nucleotides directed against NR2B mRNA inhibits LTP and cognitive function (Clayton et al., 2002). Additionally, there is a large amount of evidence suggesting an aggregation of AMPA receptors (GluRs) and metabotropic glutamate rec eptors (mGluRs) with NMDA receptors via proteins such as Homer and PSD95, further linking LTP with memory storage (Ango et al., 2000;Xiao et al., 1998). Homer-1a is elevated after synaptic activation and plays a role in targeting mGluRs to sy napses (Ango et al., 2000) Therefore all of these genes are known to play a role in cognitive ability, and their selective down-regulation in a mouse model of amyloid-associated memory deficits establishes them as possible candidat es for pharmacotherapeutic approaches to AD. Although many of the genes known to influence memory were diminished in the APP+PS1 mice, there were severa l in our study that did not show downregulated expression. GAP43, one of the unchanged genes, has been shown to improve cognitive function when over-expressed in transgenic mice (Routtenberg et al., 2000). Synaptopodin, another protein lin ked to plasticity (Mundel et al., 1997;Deller et al., 2000;Yamazaki et al., 2001), is also stable in the amyloidcontaining regions of the APP+PS1 mouse br ain. These data imply that amyloid is selective in down-regulating only cert ain memory-associated genes, which are mostly postsynaptic. There are several mechanisms which might account for these changes. One possibility is direct interaction of A with one or more signal
42 transduction pathways. Alternatively, t he regulation could be via an indirect pathway involving the acute phase respons e (e.g. cytokine effects secondary to amyloid-induced microglia activation). The first idea is supported by the fact that several of the down-regulated genes ar e immediate early genes that can be induced by activating the ERK signali ng cascade (Davis et al., 2000;Mazzucchelli et al., 2002). A itself appears capable of modifying ERK signaling (Dineley et al., 2001) potentially interfering with expression of these genes. Alternatively, as the induction of these IEGs is dependent upon synaptic activity, a generalized decrease in neuronal activity caused by A might also explain the reductions in expression. There is also evidence suppor ting the indirect mode of regulation, since the glial reactions leading to t he activation of inflammatory mediators correspond temporally to the period when memory loss is occurring (Gordon et al., 2002). Dissociating between these mec hanisms and identifying the specific components involved will also benefit development of ra tional pharmacotherapies for AD. We observed that many of the memo ry-associated genes were similarly deficient in the amyloid-c ontaining regions of the human AD brain, confirming earlier reports from our gr oup and others (Ginsberg et al., 2000;Loring et al., 2001;Bi and Sze, 2002). However, in terpretation of this observation is complicated by the fact that multiple neuronal and synaptic marker genes were also under-expressed in the AD cortex we examined. But our samples were from relatively late-stage AD, when amyloid deposition and neuronal dysfunction are extensive, and several reports indicate that widespread loss of synapses appears
43 to follow, not precede, loss of memory f unction. In a detailed study of synaptic markers during the course of AD, it wa s reported that loss of synaptic markers only occurred in Braak stages 5 & 6, late in the disease when the pathology is most widespread (Mukaetova-Ladinska et al., 2000). Another study (Tiraboschi et al., 2000) failed to detect changes in the synaptic markers synaptophysin or choline acetyl transferase in mild AD (MM SE = 20) and a third report found that mildly demented individuals (CDR 0.5-1) had no change in synaptotagmin or GAP43, although a slight reduction in synaptophysin was found which worsened as the disease progressed (Masliah et al., 2001). A study by Minger et al found that only in AD cases that were severe (MMSE < 4) was there a significant reduction in choline acetyltransferase synaptophysin, syntaxin or SNAP-25. None of these studies examined the synaptic plasticity markers that we observed to be decreased in the transgenic mice. In light of this evidence and our observations, we propose that the APP+ PS1 mouse that 1) develops both forebrain-specific amyloid deposits and me mory deficits, 2) suffers acute phase reactions in the brain, and 3) has decreased memory-associated gene expression without the loss of synaptic integrity markers, is likely to be an appropriate model for ant erograde amnesias found in early stage AD. Transgenic models for human disease hav e become a critical tool in the progression of therapeutic strategies to clin ical trials. It is cr ucial, therefore, to validate the models for specific characteri stics of the human disease. This report describes further characterization of a transgenic mouse model that constitutively expresses both the human APP and PS1 genes containing mutations that are
44 known to accelerate amyloid deposit ion and dementia in AD cases. The transcripts of several postsynaptic genes t hat are thought to be essential for the retention of memory and maintenance of LTP are specifically down-regulated in the cognitive areas of APP+PS1 mouse br ain without the loss of expression of genes involved in synaptic function. When these genes were analyzed in late stage human AD temporal cortex, we found that the same memory-associated genes were down-regulated, but unlike the mouse, all markers for synaptic integrity and neuronal stab ility were also under-expressed. Along with the activation of glial cells and induction of ac ute phase reactants, the fact that the APP+PS1 mice deposit amyloid in the fo rebrain, develop anterograde amnesia and have altered expression of these me mory-associated genes specifically in the cognitive domains of the brain suggests that some form of amyloid is facilitating the memory loss in these animals. These data also contribute to the growi ng pool of evidence that specific genes such as Arc, Zif268, NR2B, and Ho mer-1a are critical to normal memory function, and that their dysregulation may under lie the early phase of memory loss that occurs in AD. Therefore it will be of great interest to evaluate methods to up-regulate these genes as potential t herapeutic targets for use in early-stage AD and possibly other disorders involving memory deficiency.
45 ACKNOWLEDGEMENTS Dr. Tom Beach generously provided hum an tissue from the Brain Donation Program at Sun Health Research Institut e in Sun City, Arizona. This work was supported by NIA grants AG15490 and AG18478.
46 REFERENCES (1997) Consensus recommendations for the postmortem diagnosis of Alzheimer's disease. The National In stitute on Aging, and Reagan Institute Working Group on Diagnostic Criteria for the Neuropathological Assessment of Alzheimer's Disease. Neur obiol Aging 18: S1-S2. Ango F, Pin JP, Tu JC, Xiao B, Worley PF, Bockaert J, Fagni L (2000) Dendritic and axonal targeting of type 5 metabotropic glutamate receptor is regulated by homer1 protei ns and neuronal excitation J Neurosci 20: 8710-8716. Arendash GW, King DL, Gordon MN, Morgan D, Hatcher JM, Hope CE, Diamond DM (2001) Progressive, age -related behavioral impairments in transgenic mice carrying both mutant amyl oid precursor protein and presenilin-1 transgenes. Brain Res 891: 42-53. Austin LA, Arendash GW, Gordon G, Diamond DM, DiCarlo G, Dickey C, Ugen KE, Morgan D (2003) Short-term Abeta vaccinations do not improve cognitive performance in cognitively-impaired APP+PS1 mice. Behavioral Neuroscience. Bi H, Sze CI (2002) N-methyl-D-a spartate receptor subunit NR2A and NR2B messenger RNA levels are alter ed in the hippocampus and entorhinal cortex in Alzheimer's diseas e. J Neurol Sci 200: 11-18. Bolshakov VY, Golan H, Kandel ER Siegelbaum SA (1997) Recruitment of new sites of synaptic transmission during the cAMPdependent late phase of LTP at CA3-CA1 synapses in t he hippocampus. Neuron 19: 635-651. Brakeman PR, Lanahan AA, O'Brien R, Ro che K, Barnes CA, Huganir RL, Worley PF (1997) Homer: a protein that selectively binds metabotropic glutamate receptors. Nature 386: 284-288. Chapman PF, White GL, Jones MW, Cooper-Blacketer D, Marshall VJ, Irizarry M, Younkin L, Good MA, Bli ss TV, Hyman BT, Younkin SG, Hsiao KK (1999) Impaired synaptic plasticity and learning in aged amyloid precursor protein transgenic mice. Nat Neurosci 2: 271-276. Clayton DA, Mesches MH, Alvarez E, Bickford PC, Browning MD (2002) A hippocampal NR2B deficit can mimic age-related changes in long-term potentiation and spatial learning in the Fischer 344 rat. J Neurosci 22: 36283637. Davis S, Vanhoutte P, Pages C, Caboche J, Laroche S (2000) The MAPK/ERK cascade targets both Elk-1 and cAMP response elementbinding
47 protein to control long-term potentia tion-dependent gene expression in the dentate gyrus in vivo. J Neurosci 20: 4563-4572. Deller T, Merten T, Roth SU, Mun del P, Frotscher M (2000) Actinassociated protein synaptopodin in the rat hippocampal forma tion: localization in the spine neck and close association wit h the spine apparatus of principal neurons. J Comp Neurol 418: 164-181. Dineley KT, Westerman M, Bui D, Be ll K, Ashe KH, Sweatt JD (2001) Beta-amyloid activates the mitogenactivated protein kinase cascade via hippocampal alpha7 nicotinic acetylcholi ne receptors: In vitro and in vivo mechanisms related to Alzheimer's di sease. J Neurosci 21: 4125-4133. Fitzjohn SM, Morton RA, Kuenzi F, Ro sahl TW, Shearman M, Lewis H, Smith D, Reynolds DS, Davies CH, Co llingridge GL, Seabrook GR (2001) Agerelated impairment of synaptic transmissi on but normal long-term potentiation in transgenic mice that overexpress t he human APP695SWE mutant form of amyloid precursor protein. J Neurosci 21: 4691-4698. Ginsberg SD, Hemby SE, Lee VM, Eber wine JH, Trojanowski JQ (2000) Expression profile of transcripts in Alzheimer's disease tangle-bearing CA1 neurons. Ann Neurol 48: 77-87. Gordon MN, Holcomb LA, Jantzen PT, DiCarlo G, Wilcock D, Boyett KW, Connor K, Melachrino J, O'Callaghan JP, Morgan D (2002) Time course of the development of Alzheimer-like pathol ogy in the doubly transgenic PS1+APP mouse. Exp Neurol 173: 183-195. Gordon MN, King DL, Diamond DM, Jantzen PT, Boyett KV, Hope CE, Hatcher JM DiCarlo G, Gottschall WP, Morgan D, Arendash GW (2001) Correlation between cognitive deficit s and Abeta deposits in transgenic APP+PS1 mice. Neurobiol Aging 22: 377-385. Guzowski JF, Lyford GL, Stevens on GD, Houston FP, McGaugh JL, Worley PF, Barnes CA (2000) Inhibi tion of activity-dependent arc protein expression in the rat hippocampus impai rs the maintenance of long-term potentiation and the consolidation of long-term memory. J Neurosci 20: 39934001. Holcomb L, Gordon MN, McGowan E, Yu X, Benkovic S, Jantzen P, Wright K, Saad I, Mueller R, Morgan D, S anders S, Zehr C, O'Campo K, Hardy J, Prada CM, Eckman C, Younkin S, Hsiao K, Duff K (1998) Accelerated Alzheimertype phenotype in transgenic mice carry ing both mutant amyloid precursor protein and presenilin 1 tr ansgenes. Nat Med 4: 97-100. Holcomb LA, Gordon MN, Jantzen P, Hsiao K, Duff K, Morgan D (1999) Behavioral changes in transgenic mice expressing both amyloid precursor
48 protein and presenilin-1 mutations: lack of association with amyloid deposits. Behav Genet 29: 177-185. Hsiao K, Chapman P, Nilsen S, Eckman C, Harigaya Y, Younkin S, Yang F, Cole G (1996) Correlative memory def icits, Abeta elevation, and amyloid plaques in transgenic mice. Science 274: 99-102. Jones MW, Errington ML, French PJ, Fine A, Bliss TV, Garel S, Charnay P Bozon B, Laroche S, Davis S (2001) A requirement for the immediate early gene Zif268 in the expression of late LTP and long-term memories. Nat Neurosci 4: 289-296. Larson J, Lynch G, Games D, Seubert P (1999) Alterations in synaptic transmission and long-term potentiation in hippocampal slices from young and aged PDAPP mice. Brain Res 840: 23-35. Loring JF, Wen X, Lee JM, Seilham er J, Somogyi R (2001) A gene expression profile of Alzheimer's di sease. DNA Cell Biol 20: 683-695. Ma L, Zablow L, Kandel ER, Sieg elbaum SA (1999) Cyclic AMP induces functional presynaptic boutons in hippoc ampal CA3CA1 neuronal cultures. Nat Neurosci 2: 24-30. Masliah E, Mallory M, Alford M, DeTeresa R, Hansen LA, McKeel DW, Jr., Morris JC (2001) Altered expression of synaptic proteins occurs early during progression of Alzheimer's di sease. Neurology 56: 127-129. Matsuoka Y, Picciano M, Malester B, LaFrancois J, Zehr C, Daeschner JM, Olschowka JA, Fonseca MI, O'Bani on MK, Tenner AJ, Lemere CA, Duff K (2001) Inflammatory responses to amyl oidosis in a transgenic mouse model of Alzheimer's disease. Am J Pathol 158: 1345-1354. Mazzucchelli C, Vantaggiato C, Ciamei A, Fasano S, Pakhotin P, Krezel W, Welzl H, Wolfer DP, Pages G, Valv erde O, Marowsky A, Porrazzo A, Orban PC, Maldonado R, Ehrengruber MU, Cestari V, Lipp HP, Chapman PF, Pouyssegur J, Brambilla R (2002) K nockout of ERK1 MAP kinase enhances synaptic plasticity in the striatum and fa cilitates striatal-m ediated learning and memory. Neuron 34: 807-820. Minger SL, Honer WG, Esiri MM, McD onald B, Keene J, Nicoll JA, Carter J, Hope T, Francis PT (2001) Synaptic pathology in prefr ontal cortex is present only with severe dementia in Alzheimer disease. J Neuropathol Exp Neurol 60: 929-936. Mirra SS, Heyman A, McKeel D, Sumi SM, Crain BJ, Brownlee LM, Vogel FS, Hughes JP, van Belle G, Berg L ( 1991) The Consortium to Establish a
49 Registry for Alzheimer's Disease (CERAD ). Part II. Standar dization of the neuropathologic assessment of Alzheime r's disease. Neurology 41: 479-486. Morgan D, Diamond DM, Gottschall PE, Ugen KE, Dickey C, Hardy J, Duff K, Jantzen P, DiCarlo G, Wilcock D, Connor K, Hatcher J, Hope C, Gordon M, Arendash GW (2000) A beta peptide vaccination prevents memory loss in an animal model of Alzheimer's disease. Nature 408: 982-985. Morris RG (1998) Synaptic plasticity. Down with novelty. Nature 394: 834835. Mukaetova-Ladinska EB, Garcia-Siera F, Hurt J, Gertz HJ, Xuereb JH, Hills R, Brayne C, Huppert FA, Paykel ES, McGee M, Jakes R, Honer WG, Harrington CR, Wischik CM (2000) Stagi ng of cytoskeletal and beta-amyloid changes in human isocortex reveals biphasic synaptic protein response during progression of Alzheimer's dis ease. Am J Pathol 157: 623-636. Mundel P, Heid HW, Mundel TM, Kruger M, Reiser J, Kriz W (1997) Synaptopodin: an actin-associated protein in telencephalic dendrites and renal podocytes. J Cell Biol 139: 193-204. Routtenberg A, Cantallops I, Zaffut o S, Serrano P, Namgung U (2000) Enhanced learning after genetic overexpressi on of a brain growth protein. Proc Natl Acad Sci U S A 97: 7657-7662. Schmittgen TD, Zakrajsek BA (2000) Effe ct of experimental treatment on housekeeping gene expression: validation by real-time, quantitative RT-PCR. J Biochem Biophys Methods 46: 69-81. Schuff N, Capizzano AA, Du AT, Am end DL, O'Neill J, Norman D, Kramer J, Jagust W, Miller B, Wolkowitz OM, Yaffe K, Weiner MW (2002) Selective reduction of N-acetylaspartate in medial temporal and parietal lobes in AD. Neurology 58: 928-935. Stein TD, Johnson JA (2002) Lack of neurodegeneration in transgenic mice overexpressing mutant amyloid precursor protein is associated with increased levels of transthyretin and the activation of cell survival pathways. J Neurosci 22: 7380-7388. Tang YP, Shimizu E, Dube GR, Rampon C, Kerchner GA, Zhuo M, Liu G, Tsien JZ (1999) Genetic enhancement of le arning and memory in mice. Nature 401: 63-69.
50 Tiraboschi P, Hansen LA, Alford M, Masliah E, Thal LJ, Corey-Bloom J (2000) The decline in synapses and cholinergic activity is asynchronous in Alzheimer's disease. Walsh DM, Klyubin I, Fadeeva JV, Cullen WK, Anwyl R, Wolfe MS, Rowan MJ, Selkoe DJ (2002) Naturally secret ed oligomers of am yloid beta protein potently inhibit hippocampal long-term potent iation in vivo. Nature 416: 535-539. Xiao B, Tu JC, Petralia RS, Yuan JP, Doan A, Breder CD, Ruggiero A, Lanahan AA, Wenthold RJ, Worley PF (1998) Homer regulates the association of group 1 metabotropic glutamat e receptors with multival ent complexes of homerrelated, synaptic proteins. Neuron 21: 707-716. Yamazaki M, Matsuo R, Fukazawa Y, Ozawa F, Inokuchi K (2001) Regulated expression of an actin-associ ated protein, synapt opodin, during longterm potentiation. J Neurochem 79: 192-199. Yue H, Eastman PS, Wang BB, Minor J, Doctoler o MH, Nuttall RL, Stack R, Becker JW, Montgomery JR, Vainer M, Johnston R (2001) An evaluation of the performance of cDNA microarrays for detecting changes in global mRNA expression. Nucleic Acids Res 29: E41.
51 TABLE 1. Expression of neuronal genes in transgeni c mice compared to nontransgenic mice (array and qRT-PCR data). Array Data Percentage of nontransgenic mean SEM. (z-score in parentheses) qRT-PCR Data Percentage of nontransgenic mean SEM Marker mRNA Protein Function AmyloidContaining Area Amyloid-Free Area AmyloidContaining Area AmyloidFree Area Arc LTP-associated structural 64 3* (2.8) 108 6 (0.4) 48 10* 97 5 Calsyntenin Postsynaptic Ca(2+) signaling 64 3* (2.8) 79 5 (1.33) 99 9 nd Gap43 Neuritic growth, plasticity 115 7 (0.75) 87 7 (0.75) 104 10 94 4 GAPDH Energy metabolism 91 4 (0.49) 97 1 (0.15) 82 7 nd GluR1 Postsynaptic receptor 102 2 (0.1) 105 3 (0.25) 70 10* 95 6 Homer-1a LTP-associated, regulatory nd nd 60 7* 95 9 Na, K ATPase III Neuronal ion gradient, transmission 66 4* (2.6) 77 3 (1.5) 57 4* 99 5 NAB2 LTP-related, regulatory nd nd 115 7 nd Neurofilament M Neuronal structur al 103 1 (0.15) 104 1 (0.20) 82 7 nd * significantly different from nontransgenic mice at P<0.05
52 TABLE 1 (CONTINUED) * significantly different from nontransgenic mice at P<0.05 Array Data Percentage of nontransgenic mean SEM. (z-score in parentheses) qRT-PCR Data Percentage of nontransgenic mean SEM Marker mRNA Protein Function AmyloidContaining Area Amyloid-Free Area AmyloidContaining Area AmyloidFree Area NR2B Receptor implicated in memory consolidation nd nd 82 5* 99 5 Nur-77 LTP-associated, regulatory 61 4* (3.2) 77 1 (1.5) nd nd Synapsin Presynaptic vesicle-associated 71 8* (2.0) 92 2 (0.4) 103 7 82 18 Synaptophysin Presynaptic vesicle-associated 126 8 (1.3) 91 4 (0.5) 101 5 98 5 Synaptopodin LTP-associated structural nd nd 91 5 nd Synaptotagmin 5 Presynaptic vesicle-associated 98 4 (0.1) 106 2 (0.3) 106 19 nd Syntaxin Presynaptic vesicle-associated nd nd 94 7 nd Zif268 LTP-associated, regulatory 60 5* (3.3) 88 8 (0.68) 45 8* 82 3*
53 TABLE 2. Genes that are up-regulated in amyl oid-containing area s of transgenic mice brains compared to the same areas of nontransgenic mice brains (microarray data). Function and Location Gene Name Percentage of nontransgenic mean SEM (z-score in parentheses) Alpha-1 type III collagen 162 7* (3.3) Alpha-fibrinogen Fga 368 46* (13.6) Beta-2-microglobulin 203 9* (5.4) Beta-5 integrin 190 11* (4.7) Beta-galactoside-binding lectin 177 14* (4.1) Cd63 193 16* (4.9) Complement component C1 q alpha chain 289 27* (9.6) Complement component C1q beta chain 200 16* (5.2) Complement component C4 297 22* (10.1) Histidine-rich glycoprotein 172 13* (3.8) Ig heavy chain VDJh2 region 160 10* (3.2) Lysozyme 183 8* (4.4) MHC class I 166 6* (3.5) MHC class I A1(f) alpha chain RT1.A1(f) 182 8* (4.3) Transferrin 178 10* (4.1) Secreted/ Cell Surface Tyrosine kinase binding protein 254 6* (7.9) Glial fibrillary acidic protein 483 64* (19.4) Cytoskeletal Vimentin 181 14* (4.3) Cathepsin D 253 32* (7.9) Cathepsin S 248 21* (7.6) Inflammation Lysosomal Cathepsin Y 187 22* (4.6) * All entries are significantly different from amyloid-containing regions of nontransgenic mice at P < 0.05
54 TABLE 2 (CONTINUED). Function and Location Gene Name Percentage of nontransgenic mean SEM (z-score in parentheses) Acyl-CoA binding protein 166 6* (3.5) Apolipoprotein D 159 10* (3.2) Apolipoprotein E 160 4* (3.2) Lipid binding Niemann Pick type C2 180 12* (4.2) Alpha-D-mannosidase 162 7* (3.3) Cystathionine beta-synthase 377 22* (14.1) Kelch-like 1 189 15* (4.7) Mitochondrial Epsilon-trimethyllysine 2oxoglutaratedioxygenase 175 21* (4.0) Intracellular Serine protease inhibitor 161 21* (3.3) SPI-2 serine protease inhibitor 176 14* (4.0) Acute-phase reaction Secreted/ Cell Surface Thyroid hormone receptor alpha 618 73* (26.1) * All entries are significantly different from amyloid-containing regions of nontransgenic mice at P < 0.05
55 TABLE 3. Genes that are up-regulated in both amyloidcontaining brain regions and amyloid-free brain regions of transgenic mi ce (array and qRT-PCR data). Array Data Percentage of nontransgenic mean SEM qRT-PCR Data Percentage of nontransgenic mean SEM Marker RNA AmyloidContaining Area Amyloid-Free Area AmyloidContaining Area AmyloidFree Area APP 192 7 (4.8) 209 7 (5.7) nd nd Transgene Prion protein (UTR) 223 15 (6.4) 242 14 (7.3) nd nd Apolipoprotein E 160 4 (3.2) 171 10 (3.8) 156 16 nd Alpha-1 type III collagen 162 7 (3.3) 167 12 (3.6) nd nd Complement component C1q alpha chain 289 27* (9.7) 182 18* (4.3) nd nd Complement component C4 297 22* (10.1) 154 8* (2.9) nd nd Cystathionine beta-synthase 377 22* (14.1) 194 16* (4.9) nd nd GFAP 483 64 *(19.4) 182 13* (4.3) 629 115 nd Thyroid hormone receptor alpha 618 73* (26.1) 234 24* (6.9) nd nd Inflammation Tyrosine kinase binding protein 254 6* (7.9) 161 7* (3.3) nd nd * Data in BOLD are significantly more highly expressed in amyl oid-containing regions than amyloid-free regions of tr ansgenic mice. nd: not determined
56 FIGURE 1. Gene expression profile of transgeni c mice in amyloid-containing and amyloid-free brain regions by qRT-PCR. The differential expression (transgeni c expression relative nontransgenic expression) is presented for genes that are down-regul ated in transgenic animals and primarily postsynaptic (left hal f of figure) and those that are stably expressed and largely presynaptic (right half of figure). The relati ve expression in the amyloid-containing r egion (hippocampus; black bars) and amyloid-free region (cerebellum; gray bars) are shown when both regions were analyzed. These regions are also designated in the line to the right of "A plaques" with a "+" beneath the bar indicating amyloid-containi ng region or a "-" beneath the bar indicating an amyloid-free region. Follo wing qRT-PCR, each gene transcript level for each sample is first normalized to the 18S RNA level measured from the same reverse transcription. For each r egion, relative expression was determined by dividing each transgenic value by t he average of the nontransgenic values. The relative expression values for the amyloid-containing and amyloid-free regions could also be compar ed statistically to determine if the reduction in expression was different in the two struct ures. The values represented in this figure are the mean + SEM We did not determine cerebellum measurements for several genes that were unaffected in t he hippocampus of the transgenics, as indicated by "N.D." in the pl ace of the bar on the figure. * indicates significant differences between APP+PS1 mice and non-transgenic littermates (p < 0.05). "" above the bracket indicates trans gene associated down-regulation to a greater extent (P < 0.05) in the amyloidcontaining than the amyloid-free regions.
57 0 100A rc G luR 1 Ho m er -1 a N a +, K+ ATPa s e NR2B Zif2 68 G A P-4 3 Synaptophysin N eu rof ilam e nt M Synap sin 1 S ynaptopo d in Sy na pto ta gm in VRelative Expression SEM (% of Non-Tg Control)*150 50 A Plaques+ + + + + + + + * * * ND+ -+ -+ -+ -NDNDND*
58 FIGURE 2. Gene expression profile of Alzheimer disease tissue in amyloidcontaining and amyloid-free br ain regions by qRT-PCR. The differential expression (Alzheimer relative to age-matched normals) is presented for genes that are primarily post synaptic (left half of figure) and those that are primarily presynaptic (right half of figure). The relative expression in the amyloid-containing region (temporal co rtex; area 21; black bars) and amyloidfree region (cerebellum; gray bars) are shown when both regions were analyzed. These regions are also designated in the line to the right of "A plaques" with a "+" beneath the bar indicating amyloid cont aining region or a "-" beneath the bar indicating an amyloid-free region. Followi ng qRT-PCR, each gene transcript level for each sample is first normalized to 18S RNA measurements from the same reverse transcription. For each region, relative expression was determined by dividing each Alzheimer sample value by the average of the age-matched normal values. The relative expression values for the amyloid-containing and amyloidfree regions could also be compared statisti cally to determine if the reduction in expression was different in the two struct ures. The values represented in this figure are the mean + SEM. We did not determine cerebellum measurements for several genes as indicated by "N.D." in the place of the bar on the figure. * indicates significant differences betw een Alzheimer and age-matched normals (p < 0.05). "" above the bracket indicate s disease associated down-regulation to a greater extent (P < 0.05) in the amyloid-containi ng than the amyloid-free regions.
59 0 100Glu R1 H om e r1b Na +, K+ A T Pas e N R2 B Zif 26 8 G A P-4 3 Sy n apt op hy s in Ne ur ofila m en t M Synapsin 1 S y na pt opod in S yn a ptot a gmi n VRelative Expression SEM (% of Non-AD Control)*150 50 A Plaques + + -+ + + + + * * * * * *+ -+ -+ -NDNDND ND+ -
60 PAPER II: AMYLOID SUPPRESSES INDUCTION OF GENES CRITICAL FOR MEMORY CONSOLIDATION IN APP+PS1 TRANSGENIC MICE Chad A. Dickey1, Marcia N. Gordon1, Jerimiah E. Mason1, Nedda J. Wilson1, David M. Diamond2, John F. Guzowski3 & Dave Morgan1 1 Alzheimers Disease Research Laborat ory, Department of Pharmacology, University of South Florida, College of Medicine 12901 Bruce B. Downs Blvd, MDC 9, Tampa, Fl 33612 2 Departments of Psychology and Pharmaco logy, College of Arts and Sciences, University of South Florida, and Medica l Research, Veterans Hospital, 4202 E. Fowler Ave. Tampa, FL 33620 3 Department of Neurosciences, Basic M edical Sciences Building, Room 145, University of New Mexico, Health Sciences Center, Albuquerque, NM 87131 Published in Journal of Neurochemistry 2004 Jan;88(2):434-42.
61 ABSTRACT Mice transgenic for mutated forms of the amyloid precur sor protein (APP) plus presenilin-1 (PS1) genes (APP+ PS1 mice) gradually develop memory deficits which correlate with the extent of amyloid deposition. The expression of several immediate-early genes (IEGs: Arc, Nur77 and Zif268) and several other plasticity-related genes (GluR1, CaMKII and Na-KATPase III) critical for learning and memory was normal in young APP+PS1 mice preceding amyloid deposition, but declines as mice grow older and amyloid deposits accumulate. Gene repression was less in APP+PS1 mous e brain regions that contain less A and in APP mice compared to APP+PS1 mi ce, further linking the extent of amyloid deposition and t he extent of gene repre ssion. Critically, we demonstrated that amyloid depos ition led specifically to impaired induction of the IEGs with no effects on basal expression us ing exposure to a novel environment 30 minutes prior to sacrifice to i nduce IEGs. These data imply that A deposition can selectively reduce expression of mult iple genes linked to synaptic plasticity and provides a molecular basis for memo ry deficiencies f ound in transgenic APP mice and, most likely, in early stage Alzheimer's disease (AD). Presumably, pharmacological agents blocking the A -related inhibition of gene expression will have benefit in AD. Keywords: Alzheimers, Induction, Imm ediate-early genes, Tr ansgenic, Amyloid, Plasticity Running Title: Amyloid Suppre sses Memory-Related Gene Induction
62 INTRODUCTION Mutations in the amyloid precursor protein (APP) and presenilin-1 (PS1) genes accelerate A deposition and lead to an aggres sive form of early-onset Alzheimers disease (AD; Selkoe 2001) Mi ce transgenic for the mutant forms of the human APP (Hsiao et al. 1996) and PS1 (Duff et al. 1996) genes have been bred to generate a viable doubly trans genic mouse presenting accelerated amyloid pathology (APP+PS1 mice; Holcomb et al. 1998). Despite the lack of intracellular fibrillary tangles and neuronal degeneration, these mice do exhibit anterograde amnesia and incr eased neuro-inflammation with age, providing a good model for the study of amyloid-a ssociated memory loss at a stage of pathology similar to that in early AD (Mukaetova-Ladinska et al. 2000;Morgan et al. 2000;Gordon et al. 2002;Arendash et al. 2001). Based on work involving DNA microarray comparisons, we previously described the down-regulation of several genes in 18+ month-old APP+PS1 mice using quantitative real-time RT-PCR (qRT -PCR; Dickey et al. 2003). Of particular interest was a panel of down-regulated i mmediate early genes (IEGs) implicated in memory function, such as Zif268 (H all et al. 2000;Miyashita et al. 1998) and activity-regulated cytoskeletal-associated pr otein (Arc; Guzowski et al. 2001). We demonstrated that several more synaptic genes were also selectively downregulated in the amyloid-containing hi ppocampus of APP+PS1 mice (e.g. the ionotropic glutamate receptor 1 (GluR1) and Na, K ATPase III), while other vital synaptic mRNAs remained unchanged (e.g. synaptophysin and growthassociated protein 43 (Gap43)).
63 Our earlier work found that in mo st instances, gene expression in APP+PS1 mice was normal in the amyloi d free cerebellum. Here, we describe additional studies to test the lin kage between amyloid deposition and suppression of these genes, and address w hether the diminished IEG expression is due to reduced basal expre ssion or impaired induction. MATERIALS & METHODS Mice were bred in our facility and genotyped using previously described methods (Gordon et al. 2002). For mo st experiments, each mouse was individually transported out of the vivarium to a hol ding room 30 minutes before sacrifice. Mice were deeply anesth etized with pentobarbitol (100 mg/kg) and perfused transcardially with 0.9% sali ne. Brains were quickly removed and regions dissected and frozen on dry ice. We have shown previously that the behavior and amyloid burden of 18 month-ol d APP+PS1 mice were significantly different from the non-transgenic littermates used for these studies (Austin et al. 2003). IEGs were induced by exposure to a novel env ironment. Fourteen mice (six 18 month-old APP+PS1 and eight age-matched non-transgenic littermates) were handled and weighed for three consecutive days to acclimate them to the holding room and the experiment er. On the fourth day, t he mice were placed into a 0.6m x 0.6m Plexiglas open field containi ng objects for the mice to investigate or crawl through for 5 minutes. All mice ac tively explored this environment and no differences between transgenic and non-transgenic mice could be discerned in
64 the notes taken by the experimenter. Afte r 5 minutes, the mice were placed into a new cage located in a different room from their home cage for 30 minutes to avoid disturbing other mice. Following this 30 minute period, mice were killed as described above. The novel environment was cleaned with 30% ethanol between mice. To ascertain basal expression levels of IEGs, 14 other mice (six 18 monthold APP+PS1 and eight age-matched non-trans genic littermates) from the same cohort were removed from their home cage and killed with no delay. Total RNA was prepared from dissect ed brain tissue of APP only, PS1 only and APP+PS1 mice along with non-tr ansgenic littermates as previously described (Dickey et al. 2003). Briefl y, total RNA samples were reverse transcribed (RTed) with MMLV reverse trans criptase and 1M betaine. A standard curve was established by transcribing in creasing amounts of total RNA (covering 3 logs) from an RNA pool compiled from all samples used in that experiment. Two mass quantities (10 & 2ng) or, in la ter experiments, a single mass quantity of 5ng of total RNA from each sample we re RTed for comparison to the standard curve. Primer pairs for qRT-PCR were generated from the web-based applications Primer3 and the Oligo T oolkit (Operon). Experimental wells containing 25l PCR reactions were run in triplicate in 96-we ll plates. Two-step PCR was run on the MJ Research Opticon (Boston, MA) as follows; 1 cycle of 95 C for 15 minutes followed by 40 cycles of 95 C for 15 seconds and 60-65 C for 1 minute. All primer pairs amplifi ed a single peak of fl uorescence by melt curve analysis. The standard curve was calc ulated by plotting the threshold cycle
65 (Ct) against the log nanogram (ng) quantity of RNA added to the RT reactions. PCR efficiency varied by less than 5% for all amplicons. A linear regression was performed and the slope, relating Ct to log ng RNA, was calculated, and converted to a mass quantit y of standard RNA. These mass values for the genes of interest were then divided by 18S ri bosomal RNA mass values of the same RT reaction to determine fold-change in expr ession relative to the standard RNA pool. We chose 18S rRNA as the endo genous control gene based on empirical data of our own and results from others (Schmittgen and Zakrajsek 2000) indicating that other commonly used house-keeping genes such as -actin and GAPDH have more variable expression in tissue, particularly brain, while 18S remained steady. These fold change values for samples in the experimental and control groups were analyze d for significance using 1-wa y ANOVAs (Dickey et al. 2003). For the mice in the behaviora l induction study a 2-way ANOVA was performed with the main variables being genotype and induction state. RESULTS Quantitative RT-PCR analys is of hippocampal tiss ue from APP+PS1 mice at 2, 6 and 18 months of age revealed decreases in specific mRNA transcripts with increased amyloid accu mulation. At 2 months of age, prior to development of cerebral amyloid, no loss of mRNA expression was seen for any marker (Figs 1A & B). However, by 6 months of age, the IEGs Nur 77 (a nuclear orphan receptor), Arc and Zif268 were signific antly under-expressed by 35, 40, and 22%, respectively, compared to non-transgenic li ttermates (Fig 1A). By 18 months, in
66 addition to further reduced IEG expression, GluR1, ca lcium/calmodulin kinase II (CaMKII ) and Na, K ATPase III mRNA expression was significantly reduced by 20-30% (Fig 1B). Synaptophysin and G ap43 remained unchanged throughout the lifespan of the tr ansgenic mice. The findings herein also revealed two distinct patterns in gene regulation. The early decr ease in IEG expression at 6 months in the APP+PS1 mice (Fig 1A) is contrast ed by the later reduction in the more constitutively expressed pl asticity genes GluR1, CaMKII and Na, K ATPase III seen at 18 months (Fig 1B). Figure 1C was generated by averaging the individual sample percent reduction values for each gene, then averaging these according to IEG, plasticity-related or nonchanging synaptic category and doing one-way ANOVA on these values to determine signi ficance. This figure demonstrates that a significant reduction at 6 months is only seen with the IEGs when analyzed in this manner, while at 18 months both IE G and plasticity-related gene expression was significantly reduced. To further understand the relations hip between amyloid deposition and mRNA expression, we investigated t he gene expression of 18 month-old singly transgenic APP or PS1 mice compared to mice expressing both APP+PS1 transgenes. The amyloid-free PS1 transgenic mice had no changes in expression of any genes (Fig 2). Howe ver, the singly transgenic APP mice exhibited significant 30-40% loss of the IEGs Nur77, Ar c and Zif268, without significant loss of the more constitutive ly expressed synaptic proteins GluR1, CaMKII and Na, K ATPase III (Fig 2).
67 A regional analysis was then perfo rmed on 18 month-old APP+PS1 mice, comparing gene expression levels in the hippocampus, posterior cortex and caudate nucleus (striatum). In the pos terior cortex, APP+PS1 mice had a significant reduction in the IEGs, howe ver the other plasticity-related genes, GluR1 and Na, K ATPase III remained unchanged (Fig 3). CaMKII was also significantly reduced, but not to the same extent as in the hippocampus (Fig 3). Caudate nucleus analysis revealed a similar pattern of expression; significantly reduced IEGs (25-35%) without reductions in the other plasticity-related genes (Fig 3). For all genotypes and brai n regions, Gap43 and synaptophysin expression was unaltered (Figs 2& 3). To determine whether the reduction of the IEGs in the memory-deficient APP+PS1 mice was a result of lower basal expression level or decreased experience-dependent induction, we exposed 18 month-old APP+PS1 mice and their non-transgenic littermates to a novel environment for 5 minutes followed by 30 minutes in a new cage. This manipulat ion provided the means to increase IEG expression, resulting in increased repr esentative mRNA levels (see methods above). For the non-induced (basal) cont rol groups, additional 18 month-old APP+PS1 mice and their non-transgenic litterma tes were taken from their home cage and rapidly killed. Upon analyzing t he hippocampi of these mice by qRTPCR, we found that there was a signific ant 2.5 fold induction of IEGs in our behavioral paradigm in non-transgenic mi ce (Fig 4). This induction was significantly blunted by half for Arc and Nu r77 in the APP+PS1 mice compared to induced non-transgenic littermates (Fig 4). The basal expression of the IEGs was
68 nearly identical in the two genotypes (Fig 4). Zif268 expression was also induced significantly by the environmental novel ty, but suppression of its induction in APP+PS1 mice was not statistically sign ificant. The other plasticity-related genes, GluR1, CaMKII and Na, K ATPase III were not induced by exposure to the novel environment. However these genes did have significantly reduced expression when analyzed by genoty pe, while Gap43 and synaptophysin remained unaffected by genotype or exposur e to a novel environment (Table 1). Glial fibrillary acidic protein (GFAP) mRNA increased in the hippocampi of APP+PS1 mice. At 2 months, GFAP was unchanged, by 6 months it was increased by 80% and at 18 months levels increased to over 300% that of nontransgenic littermates (data not shown), demons trating the potential role for glial cells in the pathogenesis of amyloid-asso ciated dementias and confirming earlier ELISA and histochemical results in APP+PS1 mice (Morgan et al. 1991;Gordon et al. 2002). We have also increased the breadth of our invest igations into gene expression by analyzing many more genes finding that several more remain unaltered in the 18 month-old A PP+PS1 hippocampus compared to nontransgenic littermates. These inclu de Ha-Ras, Rheb, Shank3, synapsin, microtubule-associated protein 2 (MAP2), the Na, K ATPase subunit and the two metabotropic glutamate receptors, mGluR1 and mGluR5. These findings reiterate the selectivity of gene dysregul ation associated with either amyloid deposition itself or its subsequent deleterious effects on the brain.
69 DISCUSSION Post-mortem brain tissue from AD pat ients is riddled with amyloid plaques comprised of A peptide polymers and intracellula r neuro-fibrillary tangles of hyperphosphorylated tau filam ents. Neuron loss and gliosis are also evident. The cognitive impairments of the disease coul d be attributed to any combination of these pathologies, and recent evidence su pports each of t hese components. The development and subsequent usage of tran sgenic animal models expressing genes known to produce AD-like pathology have provided a way to elucidate the contributions of these indi vidual pathological features Here, we have utilized the memory-deficient APP+PS1 mouse model to analyze expression of genes that appear to be essential for the formation of memories. Using qRT-PCR, we have found that as amyloid becomes reliably det ectable in the brains of the APP+PS1 mice (6 mos.), IEGs begi n to realize decreased expression. As amyloid burden increases, resulting in further gliosis and onset of memory loss, several other more constitutively expressed genes crit ical for plasticity and memory function are significantly reduced. Perhaps of great est significance here is the finding that the induction of the IEGs was significantly impaired, indicating that the triggering mechanism for their rapidly increased expression is being inhibited somehow in these mice that develop amyloid. T hese data provide strong support for the hypothesis that amyloid, itself, impairs the induction and expression of genes thought to be essential for retain ing newly acquired information.
70 Our first indication that amyloid might repress genes essential for memory came via a microarray study (Dickey et al 2003). Intrigued by the novelty of the idea that amyloid may suppress molecu lar mechanisms for memory, independent of neuron or synapse loss, we examined other memory associated genes and as of this writing have found 11 that are down-regulated. For these studies, we chose a representative panel of thes e down-regulated genes associated with LTP and memory consolidation, 3 of whic h are immediate early (inducible) genes (Arc, Nur77 & Zif268) and 3 with more constitutive expression (Na, K ATPase III, GluR1 and CaMKII ). Mice receiving antisens e oligonucleotides targeting the mRNA of Arc, an IEG t hat is translated at the syn apses (Lyford et al. 1995), have disrupted LTP maintenance and dem onstrate impaired long-term memory (Guzowski et al. 2000). Nur77 mRNA, also an IEG product, is the transcript for an orphan thyroid hormone receptor that is dramatically increased in response to several excitotoxic stimuli (Watson and Milbrandt 1989;St Hilaire et al. 2003). Zif268 is a zinc finger transcription fact or that is rapidly up-regulated in the hippocampus of rats after exposure to a nov el stimulus (Hall et al. 2000) and by water maze training (12). Additionally, homozygous Zif268 knockout mice exhibit long term-term memory deficits (Jones et al. 2001) underscoring the critical nature of this gene to proper cognitive function. The Na, K ATPase III subunit, when inhibited by oubain, is known to di srupt memory consolidation (Watts and Mark 1971;Mark and Watts 1971). Recent r eports have implicated the ionotropic glutamate receptor GluR1 (AMPA1) as e ssential for various forms of synaptic plasticity and memory retention (Schmi tt et al. 2003;Lee et al. 2003). CaMKII is
71 also thought to be critical for memory retention, as evidenced by mice transgenic for activated forms of the protein having impaired memo ry retention (Mayford et al. 1995;Mayford et al. 1996). Expression of both the growth associated protein, Gap43, a gene important for growth c one formation as well as memory (Routtenberg et al. 2000), and synaptophysin, a protein widely used as a marker for nerve terminals, remains unchanged in the APP+PS1 mouse brain. Thus, not all genes involved in synaptic function or plasticity are down-regulated by A deposits. To understand the contribution of amyl oid to the dysregulation of gene expression in the APP+PS1 mice we analyzed hippocampal tissue from 2 monthold mice, an age prior to amyloid deposit ion (Gordon et al. 2002), 6 month-old mice, when amyloid deposits are just beginning to form, and 18 month-old mice, when there is considerable amyloid in the forebrain (Gordon et al. 2002). Critically, the stable expression of all genes analyzed at 2 months (Fig 1), an age when APP and PS1 over-expression is present, but amyloid deposits have not yet appeared, argues that transgene expressi on per se is not the cause of the reduced expression. It is the accumu lation of amyloid that appears to be regulating expression in such a way t hat some genes, predominately the IEGs, are diminished when amyloid deposits are still low (6 mos.) while others (e.g. GluR1 and Na, K ATPase III) are only affected when amyloid deposition is substantial (18 mos.). CaMKII expression is significantly decreased at 6 months although not to the same degree as the IEGs This could possibly indicate that CaMKII is one of the first genes being a ffected by the presence of excess
72 amyloid in the brain, which is signific ant due to its involvement in the calcium signaling cascade thought to mediate recept or activated transcription (Blanquet et al. 2003). As summarized in figure 1C these findings provide evidence that slightly reduced IEG expression is not su fficient to cause memory impairments, but, assuming that protein expression of these genes follows t hat of their mRNA, the additional down-regulation of these more constitutively expressed plasticityrelated genes, as detected at 18 mos., may be necessary to disrupt memory. This emphasizes the robustness of mnemonic processes (or the lack of sensitivity of behavioral tests), as only when partial inhibition of the expression of multiple genes are present in aggregate are memory deficits readily apparent. The APP only mice have roughly one third of the A deposition as APP+PS1 mice (Jaffar et al. 2001), while the PS1 mice resemble non-transgenic mouse brain when analyzed for amyl oid burden and amyloid associated pathogenesis. Our findings in figure 2 rev eal that the singly transgenic APP mice also have reduced IEG expression without reductions in the other plasticityrelated genes, while the PS1 singly tr ansgenic mice do not have reductions in either set of genes. This is consistent with the selective impairment of IEG expression found at 6 mos. We had previously shown that gene expression in the cerebella of APP+PS1 mice, a region without detectabl e amyloid levels, remained unchanged when compared with non-transgenic litte rmates (Dickey et al. 2003). Subsequently, we analyzed two additional r egions of the forebrain in 18 monthold APP+PS1 mice; posterior cortex, a region with fewer amyloid plaques than
73 the hippocampus, and the caudate nucl eus, a brain region that does not accumulate fibrillar amyloid pl aques, yet has abundant diffuse A (Gordon et al. 2002). Interestingly, upon analysis of both of these regions, we found that like the APP singly transgenic mice and the 6 m onth-old APP+PS1 mice, only the IEGs were significantly under-expressed. Sim ilar to the 6 month-old APP+PS1 mice, CaMKII was slightly down-regulated in t he posterior cortex but not as dramatically as in the hippocampus. Again, this is consistent with the relative amyloid burden associated wit h these two conditions. Perhaps most significant is the finding that the caudate nucleus which only contains diffuse A still has reduced IEG expression. This further implicates A itself, rather than a mediator of neuro-inflammation, as the precipitator of this mRNA reduction. These findings again argue that the IEGs ar e very sensitive to the pr esence of amyloid, possibly indicating a direct interaction of th e amyloid peptide with some mechanism regulating the expression of these genes. One final critical issue in this invest igation was the natur e of the reductions seen in IEG expression. These genes are ex pressed at low levels when mice are in a resting state. However, upon inducti on by some stimulus (e.g. maximal electroconvulsive shock (MECS; Cole et al. 1990) or a novel environment (Guzowski et al. 2001) there is a rapid and robust increase in their expression, at which point they have dram atic effects on synapse structure and properties. This increase is transient so that within minutes to hours, some IEGs return to their basal expression level. By using a novel environment, in which mice were allowed to investigate several new obj ects for 5 minutes, we were able to
74 demonstrate a significant induction of the IEGs Arc, Nur77 and Zif268. We selected a 30 minute duration as the time point because previous studies have shown that Arc and Nur77 reach a maxima l level of expression by 30 minutes and begin to return to basal levels afte r 1 hour (Lyford et al. 1995;Tang et al. 1997). We felt that genes that were i nduced maximally would provide us the greatest opportunity to see any signific ant reductions in their expression. Previous studies have indicated that zif 268 mRNA has a slightly greater half-life than other IEG RNAs (Guzowski et al. 2000;Guzowski et al. 2001) which would effectively increase basal levels, making it more difficult to detect differences due to a single behavioral event, possibly ex plaining the reason we were unable to see significant reductions in its expressi on. We had suspected that our standard transport and handling procedures prior to killing the mice were sufficient to induce the IEGs. Anticipating this, our standard protocol is to retrieve each mouse from the vivarium 30 minutes bef ore pentobarbitol injection. The finding that only the induction of these genes wa s impaired in APP+PS1 mice confirms these suspicions. These observations emphasize the need to control the conditions leading up to euthanasia in st udies of gene expression in brain. These data, when taken together with the findings that A is precipitating reductions in gene expression, indicate that in amyloid-depositing APP+PS1 mice, the ability to induce IEG expressi on is suppressed. This suppression may be a direct effect of A on a receptor altering signal transduction cascades (possibly the Ras/MAPK pathway or the Ca++ signaling cascade (Dineley et al. 2001)). Alternatively, A deposition may impair increases in general synaptic
75 activity presumed to result from expo sure to environmental novelty, leading to less induction of these activity dependent IEGs. The diminution of Na, K ATPase activity may permit normal re sting activity in neurons, but lead to failure when demands for ion pumping increase with elevated synaptic activity. Another possibility remains that these events ar e secondary to increased inflammation associated with glial activation, particularl y given the considerable effects that cytokines can have on gene expression. Ho wever, at least for the IEGs this seems unlikely, as microglial activati on and cytokine production are low in 6 month APP+PS1 mice, when t he IEGs are already realizing significant decreases (Fig 1A & C; (Gordon et al. 2001;Benzi ng et al. 1999;Mehlhorn et al. 2000). Ultimately, the data pres ented here point to A or some direct effect of its presence in the brain, as the causal agent for suppressed induction of IEG expression. The failure to identify reduced expression of rheb in APP+PS1 mouse hippocampus provides an example of one IEG that does not seem to be influenced by the presence of A Shank3, involved in formation of the postsynaptic density (Boeckers et al. 2002), along with calsyntenin-1 and synaptopodin (Dickey et al. 2003), also remain unchanged in the APP+PS1 mouse, eliminating the possibility of a generalized but exclusive loss of postsynaptic mRNAs, with axon terminals rema ining intact and functional. Also, the two dendritic metabotropic glutamate rec eptors, mGluR1 and mG luR5, that have been linked to memory fo rmation and LTP-maintenance (Riedel et al. 2000), do not exhibit significant decreases (P etersen et al. 2002). Other genes with
76 unchanged expression are M AP-2, thought to propagate growth cone formation of the axon (Gordon-Weeks 1991) and the Na, K ATPase subunit, which modulates the activity of the ATPase subunit (Crambert and Geering 2003), demonstrating a specific reduction in the catalytic subunit expression level. The stability of these other markers emphasizes the significance of those genes that do suffer from amyloid -related repression. The recent finding by Hock et al. t hat vaccination with t he amyloid peptide stabilized and, in some cases, improved cognitive ability in humans suffering from AD (Hock et al. 2003), argues that A itself likely plays in the development of memory dysfunction in AD beyond t he effects on neuronal and synapse loss. The data presented here provide new ev idence to support the idea that A impedes the induction of genes critical for synaptic plasticity. The APP+PS1 transgenic mouse model, therefore, prov ides a means with which to understand how amyloid diminishes the functioning of essential memory-related molecular systems within the neuron. There is a dynamic nature of specific neuronal transcripts that lend to the ability of the brain to reta in information and it seems that amyloid disrupts this syst em in multiple ways. It is possible that as the IEG induction decreases, it reaches a point at which the strengt h of the subsequent memory trace is inadequate to modify t he mouses behavior. It has already been shown in multiple mouse models that am yloid deposition is associated with the decline of cognitive function (Chen et al. 2000;Gordon et al. 2001;Westerman et al. ;Heikkinen et al. 2002). These data provid e the first hints of a molecular basis for this problem. The results herein suggest potential targets for therapy and
77 prophylaxis of amyloid-associated dementias and possibly other forms of cognitive impairments that are to this poi nt poorly understood at the level of gene expression. ACKNOWLEDGEMENTS This work was supported by AG15490 and AG18478 from NIH.
78 REFERENCES Arendash G. W., King D. L., Gordon M. N., Morgan D., Hatcher J. M., Hope C. E., and Diamond D. M. (2001) Progressive, age-related behavioral impairments in transgenic mice carrying bot h mutant amyloid precursor protein and presenilin-1 transgenes. Brain Res 891, 42-53. Austin L. A., Arendash G. W., Go rdon G., Diamond D. M., DiCarlo G., Dickey C., Ugen K. E., and Morgan D. (2 003) Short-term A-beta vaccinations do not improve cognitive performance in cognitively-impair ed APP+PS1 mice. Behavioral Neuroscience Benzing W. C., Wujek J. R., Ward E. K., Shaffer D., As he K. H., Younkin S. G., and Brunden K. R. ( 1999) Evidence for glial-mediated inflammation in aged APP(SW) transgenic mice. Neurobiol Aging 20, 581-589. Blanquet P. R., Mariani J., and Dere r P. (2003) A calcium/calmodulin kinase pathway connects brain-derived ne urotrophic factor to the cyclic AMPresponsive transcription factor in the rat hippocampus. Neuroscience 118, 477490. Boeckers T. M., Bockmann J., Kreut z M. R., and Gundelfinger E. D. (2002) ProSAP/Shank proteins a family of higher order organizing molecules of the postsynaptic density wit h an emerging role in hum an neurological disease. J Neurochem 81, 903-910. Chen G., Chen K. S., Knox J., Inglis J ., Bernard A., Martin S. J., Justice A., McConlogue L., Games D., Freedman S. B., and Morris R. G. (2000) A learning deficit related to age and bet a-amyloid plaques in a mouse model of Alzheimer's disease. Nature 408, 975-979. Cole A. J., Abu-Shakra S., Saffen D. W., Baraban J. M., and Worley P. F. (1990) Rapid rise in transcription factor mRNAs in rat brai n after electroshockinduced seizures. J Neurochem 55, 1920-1927. Crambert G. and Geering K. (2003) FXYD proteins: new tissue-specific regulators of the ubiquitous Na,KATPase. Sci STKE 2003, RE1. Dickey C. A., Loring J. F., Eastman P. S., Montgomery J. R., Gordon M., and Morgan D. G. (2003) Selectively reduc ed expression of synaptic plasticityrelated genes in APP+PS1 transgenic mice. J Neurosci 23, 5219-5226. Dineley K. T., Westerman M., Bui D., Bell K., Ashe K. H., and Sweatt J. D. (2001) Beta-amyloid activates the mitogen -activated protein kinase cascade via hippocampal alpha7 nicotinic acetylcholine receptors: In vitro and in vivo mechanisms related to Alzheimer's disease. J Neurosci 21, 4125-4133.
79 Duff K., Eckman C., Zehr C., Yu X., Prada C. M., Perez-tur J., Hutton M., Buee L., Harigaya Y., Yager D., Morgan D., Gordon M. N ., Holcomb L., Refolo L., Zenk B., Hardy J., and Younkin S. (1996) Increased amyloid-beta42(43) in brains of mice expressing mutant presenilin 1. Nature 383, 710-713. Gordon-Weeks P. R. (1991) Control of microtubule assembly in growth cones. J Cell Sci Suppl 15, 45-49. Gordon M. N., Holcomb L. A., Jant zen P. T., DiCarlo G., Wilcock D., Boyett K. W., Connor K., Melachrino J., O'Callaghan J. P., and Morgan D. (2002) Time course of the development of Alzheimer-like pathology in the doubly transgenic PS1+APP mouse. Exp Neurol 173, 183-195. Gordon M. N., King D. L., Diamond D. M., Jantzen P. T., Boyett K. V., Hope C. E., Hatcher J. M., DiCarlo G., Gottschall W. P., Morgan D., and Arendash G. W. (2001) Correlation between cognitive deficits and Abeta deposits in transgenic APP+PS1 mice. Neurobiol Aging 22, 377-385. Guzowski J. F., Lyford G. L., Stevens on G. D., Houston F. P., McGaugh J. L., Worley P. F., and Barnes C. A. (200 0) Inhibition of activity-dependent arc protein expression in the rat hippocampus impairs the maintenan ce of long-term potentiation and the consoli dation of long-term memory. J Neurosci 20, 39934001. Guzowski J. F., Setl ow B., Wagner E. K., and McGaugh J. L. (2001) Experience-dependent gene expression in t he rat hippocampus after spatial learning: a comparison of the immediat e-early genes Arc, cfos, and zif268. J Neurosci 21, 5089-5098. Hall J., Thomas K. L., and Everitt B. J. (2000) Rapid and selective induction of BDNF expression in the hi ppocampus during cont extual learning. Nat Neurosci 3, 533-535. Heikkinen T., Puolivali J., Liu L., Rissa nen A., and Tanila H. (2002) Effects of ovariectomy and estrogen treatm ent on learning and hippocampal neurotransmitters in mice. Horm Behav 41, 22-32. Hock C., Konietzko U., Streffer J. R., Tracy J., Signorell A., MullerTillmanns B., Lemke U., Henk e K., Moritz E., Garcia E ., Wollmer M. A., Umbricht D., de Quervain D. J., Hofmann M., Maddalena A., Papassotiropoulos A., and Nitsch R. M. (2003) Antibodies against beta -amyloid slow cognitive decline in Alzheimer's disease. Neuron 38, 547-554. Holcomb L., Gordon M. N., McGowan E ., Yu X., Benkovic S., Jantzen P., Wright K., Saad I., Mueller R., Morgan D., Sanders S., Zehr C., O'Campo K., Hardy J., Prada C. M., Eckman C., Youn kin S., Hsiao K., and Duff K. (1998)
80 Accelerated Alzheimer-type phenotype in transgenic mice carrying both mutant amyloid precursor protei n and presenilin 1 transgenes. Nat Med 4, 97-100. Hsiao K., Chapman P., Nilsen S., E ckman C., Harigaya Y., Younkin S., Yang F., and Cole G. (1996) Correlative me mory deficits, Abeta elevation, and amyloid plaques in transgenic mice. Science 274, 99-102. Jaffar S., Counts S. E., Ma S. Y., D adko E., Gordon M. N., Morgan D., and Mufson E. J. (2001) Neuropathology of mice carrying mutant APP(swe) and/or PS1(M146L) transgenes: alterations in the p75(NTR) cholinergic basal forebrain septohippocampal pathway. Exp Neurol 170, 227-243. Jones M. W., Errington M. L., French P. J., Fine A., Bliss T. V., Garel S., Charnay P., Bozon B., Laroche S., and Davi s S. (2001) A requirement for the immediate early gene Zif268 in the ex pression of late LTP and long-term memories. Nat Neurosci 4, 289-296. Lee H. K., Takamiya K., Han J. S., Man H., Kim C. H., Rumbaugh G., Yu S., Ding L., He C., Petralia R. S., W enthold R. J., Gallagher M., and Huganir R. L. (2003) Phosphorylation of the AMPA Receptor GluR1 Subunit Is Required for Synaptic Plasticity and Retention of Spatial Memory. Cell 112, 631-643. Lyford G. L., Yamagata K., Kaufmann W. E., Barnes C. A., Sanders L. K., Copeland N. G., Gilbert D. J., Jenkins N. A., Lanahan A. A., and Worley P. F. (1995) Arc, a growth factor and acti vity-regulated gene, encodes a novel cytoskeleton-associated protein that is enriched in neuronal dendrites Neuron 14, 433-445. Mark R. F. and Watts M. E. (1971) Drug inhibition of memory formation in chickens. I. Long-term memory. Proc R Soc Lond B Biol Sci 178, 439-454. Mayford M., Bach M. E., Huang Y. Y., Wang L., Hawkins R. D., and Kandel E. R. (1996) Control of memory formation through regulated expression of a CaMKII transgene. Science 274, 1678-1683. Mayford M., Wang J., Kandel E. R., an d O'Dell T. J. (1995) CaMKII regulates the frequency-response functi on of hippocampal synapses for the production of both LTD and LTP. Cell 81, 891-904. Mehlhorn G., Hollborn M., and Schliebs R. (2000) Induction of cytokines in glial cells surrounding cortical beta-a myloid plaques in transgenic Tg2576 mice with Alzheimer pathology. Int J Dev Neurosci 18, 423-431. Miyashita Y., Kameyama M., Has egawa I., I, and Fukushima T. (1998) Consolidation of Vi sual Associative Long-Term Memo ry in the Temporal Cortex of Primates. Neurobiol Learn Mem 70, 197-211.
81 Morgan D., Diamond D. M., Gottschall P. E., Ugen K. E., Dickey C., Hardy J., Duff K., Jantzen P., DiCarlo G., Wilcock D., Connor K., Hatcher J., Hope C., Gordon M., and Arendash G. W. (2000) A beta peptide vaccination prevents memory loss in an animal model of Alzheimer's disease. Nature 408, 982-985. Morgan D. G., C.E.Finch, and S.A.J ohnson (1991) RNA metabolism in Alzheimer's disease: Selective increase in GFAP RNA, in In Alzheimer's Disease:Basic Mechanisms, Diagnos is and Therapeut ic Strategies (K.Iqbal, D.R.C.McLachlan, B.Winblad, and H.M. Wisniewski, eds.), pp. 407-416. John Wiley and Sons, New York. Mukaetova-Ladinska E. B., Garcia-Siera F., Hurt J., Gertz H. J., Xuereb J. H., Hills R., Brayne C., Huppert F. A., Pa ykel E. S., McGee M., Jakes R., Honer W. G., Harrington C. R., and Wischik C. M. (2000) Staging of cytoskeletal and beta-amyloid changes in human isocorte x reveals biphasic synaptic protein response during progression of Alzheimer's disease. Am J Pathol 157, 623-636. Petersen S., Bomme C., Baastrup C., Kemp A., and Christoffersen G. R. (2002) Differential effects of mGluR1 and mGlur5 antagonism on spatial learning in rats. Pharmacol Biochem Behav 73, 381-389. Riedel G., Casabona G., Platt B., Macpha il E. M., and Nicoletti F. (2000) Fear conditioning-induced timeand subregionspecific increase in expression of mGlu5 receptor protein in rat hippocampus. Neuropharmacology 39, 1943-1951. Routtenberg A., Cantallops I., Zaffu to S., Serrano P., and Namgung U. (2000) Enhanced learning after genetic overex pression of a brai n growth protein. Proc Natl Acad Sci U S A 97, 7657-7662. Schmitt W. B., Deacon R. M., Seeburg P. H., Rawlins J. N., and Bannerman D. M. (2003) A wit hin-subjects, within-task demonstration of intact spatial reference memory and impaired s patial working memory in glutamate receptor-A-deficient mice. J Neurosci 23, 3953-3959. Schmittgen T. D. and Za krajsek B. A. (2000) E ffect of experimental treatment on housekeeping gene expression : validation by real -time, quantitative RT-PCR. J Biochem Biophys Methods 46, 69-81. Selkoe D. J. (2001) Clearing the brain's amyloid cobwebs. Neuron 32, 177-180. St Hilaire M., Tremblay P. O., Levesque D., Barden N., and Rouillard C. (2003) Effects of cocaine on c-fos and NG FI-B mRNA expression in transgenic mice underexpressing glucocorticoid receptors. Neuropsychopharmacology 28, 478-489.
82 Tang Y. P., Murata Y., Nagaya T., Noda Y., Seo H., and Nabeshima T. (1997) NGFI-B, c-fos, and c-jun mRNA ex pression in mouse brain after acute carbon monoxide intoxication. J Cereb Blood Flow Metab 17, 771-780. Watson M. A. and Milbrandt J. (1989) The NGFI-B gene, a transcriptionally inducible member of t he steroid receptor gene superfamily: genomic structure and expression in ra t brain after seizure induction. Mol Cell Biol 9, 4213-4219. Watts M. E. and Mark R. F. (1971) Separate actions of ouabain and cycloheximide on memory. Brain Res 25, 420-423. Westerman M. A., Cooper-Blacketer D., Mariash A., Kotilinek L., Kawarabayashi T., Younkin L. H. Carl son G. A., Younkin S. G., and Ashe K. H. (2002) The relationship between Abeta an d memory in the Tg2576 mouse model of Alzheimer's disease. J Neurosci 22(5):1858-1867.
83 FIGURE 1. Time course of gene expression in the hippocampus of transgenic mice by qRT-PCR. Panel A shows the differential expressi on (transgenic expression (n=7-8 APP+PS1) relative to non-transgenic expres sion (n=7-8 Non-Tg; set as 100% for each gene) for three IEGs (Arc, Zif268 and Nu r77) that were 1) unaffected at 2 months, prior to amyloid development, 2) significantly down-regulated in transgenic animals by 6-months of age, and 3) further down-regulated by 18 months. Synaptophysin remained equiva lent to non-transgenic animals throughout their lifespan. Panel B shows the differential expression for a set of plasticity-related genes (CaMKII GluR1, and Na, K ATPase III) that were unchanged at both the 2 and 6 month time points, wit h the exception of CaMKII at 6 months. However, by 18 months t hese genes were all significantly down regulated. Gap43 remained equivalent to non-transgenic animals throughout their lifespan. Panel C summarizes thes e findings by averaging the values for each gene into the represent ative category and analyzin g these for significance by one-way ANOVA. The values repres ented in this figure are the mean + SEM. * indicates significant differences between APP+PS1 mice and non-transgenic littermates (p < 0.05).
84 0 50 100 150 05101520 Months of AgeRelative Expression SEM (% of Non-Tg Control)Gap43 Zif268 Nur77* *A* *Arc 0 50 100 150 05101520 Months of AgeRelative Expression SEM (% of Non-Tg Control)Synaptophysin Na, K ATPase GluR1* *B* *CaMKII 0 50 100 150 05101520 Months of AgeRelative Expression SEM (% of Non-Tg Control)Non-Changing Synaptic Genes Plasticity-Related Genes Immediate-Early Genes* *C
85 FIGURE 2. Gene expression in hippocampi of APP only, PS1 only and APP+PS1 transgenic mice at 18 months of age by qRT-PCR. This figure compares expression of doubly transgenic APP+PS1 mice and age-matched single transgenics with non-tr ansgenic littermates. Black bars indicate PS1 only mice, gray bars indica te APP only mice and white bars indicate APP+PS1 mice. The far left panel descr ibes the differential expression (transgenic expression (n=7-8 APP+PS1) re lative to non-transgenic expression (n=7-8 Non-Tg; set as 100% for each gene) for Arc, Zif268 and Nur77. These genes were down-regulated in both APP only and APP+PS1 transgenic animals, however the PS1 only transgenics had no def icits in expression. The center panel reveals that the plasticity-related genes CaMKII GluR1, and Na, K ATPase III were down-regulated at 18 months only in the APP+PS1 transgenic mice, while APP only and PS1 only expression remained equivalent to that of non-transgenics. The right panel demonstrates the specificity of this effect by showing that other synaptic genes were not influenced by genotype in any of the 3 transgenic models. Expression va lues for the APP only and APP+PS1 transgenics could also be compared statis tically to determine if the differing levels of expression were significant between the two models. The values represented in this figure are the mean + SEM. * indicates significant differences between APP+PS1 mice and non-transgenic littermates (p < 0.05). "" above the bracket indicates differences (P < 0.05) in expression between APP+PS1 mice and APP only mice.
86 0 100Nur77Zif268ArcGluR1GAP43SynaptophysinRelative Expression SEM (% of Non-Tg Control)* * * * * Na, K ATPase CaMKII Non-Changing Synaptic Genes Plasticity-Related Genes Immediate-Early Genes150 50
87 FIGURE 3. Gene expression profile in hip pocampus, posterior cortex and caudate nucleus of APP+PS1 transgenic mice at 18 m onths of age by qRT-PCR. This figure compares gene expression in three different brain regions of doubly transgenic APP+PS1 mice with non-transgenic littermates. Black bars indicate caudate nucleus (CN), gray bars i ndicate posterior cortex (CX) and white bars indicate hippocampus (HC) in APP+PS 1 mice. The far left panel describes the differential expression in each brain region (transgenic expression (n=7-8 APP+PS1) relative to non-transgenic expres sion (n=7-8 Non-Tg; set as 100% for each gene) for Arc, Zif268 and Nur77. T hese genes were significantly downregulated in all three regions. The center panel reveals that the plasticity-related genes CaMKII GluR1, and Na, K ATPase III were down-regulated at 18 months only in the HC of APP+PS1 tr ansgenic mice, with the exception of CaMKII which is significantly down-regul ated in the CX. The right panel demonstrates the specificity of this effe ct by showing that other synaptic genes were not influenced in any of the brain r egions. Expression values for the HC and CX could also be compared statistically to determine if the differing levels of expression were significant between the two regions. CN was not compared, as these values were even closer to nontransgenic than those of CX. The values represented in this figure are the mean + SEM. * indicates significant differences between APP+PS1 brain regions and those of non-transgenic littermates (p < 0.05). "" above the bracke t indicates differences (P < 0.05) in expression between PCX and HPC of APP+PS1 mice.
88 0 100Nur77Zif268ArcGluR1GAP43SynaptophysinRelative Expression SEM (% of Non-Tg Control)* * * * * Na, K ATPase CaMKII Non-Changing Synaptic Genes Plasticity-Related Genes Immediate-Early Genes* * *150 50
89 FIGURE 4. IEG expression measured by qRTPCR in hippocampi of APP+PS1 transgenic mice and non-transgenic littermates following Induction by environmental novelty. This figure compares expression of IEGs between APP+PS1 mice and non-transgenic littermates either following in duction by a 5 minute exposure to a novel environment with a subsequent 30 minute period in a new cage, or immediate euthanitazation after removal from their home environment. White bars indicate those mice that were euthanatized quickly to measure basal expression. Black bars indicate thos e animals that were exposed to environmental novelty to measure i nduced expression. ANOVA indicated a significant effect of induction for Arc (F= 38.8; P < 0.0001), Nur77 (F= 26.1; P < 0.0001), and Zif268 (F= 27.992; P < 0.0001). There was also a significant interaction between genotype and induction for Arc (F= 4.4; P < 0.05) and Nur77 (F= 4.3; P < 0.05), but not Zif268. The va lues represented in this figure are the mean + SEM. * indicates a significant interaction effect between genotype and treatment (p < 0.05) as determined by two-way ANOVA.
90 0 100 200 300 ArcNur77Zif268Relative Expression SEM (% of Non-Tg Control)* Non-TgNon-TgNon-Tg APP+PS1APP+PS1APP+PS1
91 TABLE 1. Expression of non-inducible genes in APP+PS1 mice stimulated by environmental novelty compared to non-transgenic littermates by qRT-PCR. Values indicate percent reduction in APP+PS1 mouse hippocampus RNA expression compared to non-transgenic littermates SEM. No induction occurred for these genes, therefore mice were analyzed by genotype only using one-way ANOVA. * indicates significantly differ ent from non-transgenic mice at P<0.05. Gene Properties Marker mRNA Non-Induced CaMKII 87.3 5.2* GluR1 83.8 6.6* PlasticityRelated Genes Na, K ATPase III 83.3 5.9* Gap43 95.6 4 Non-Changing Synaptic Genes Synaptophysin 95.4 5.1
92 PAPER III: DYSREGULATION OF THE NA+/K+ ATPASE BY AMYLOID: IMPLICATIONS FOR NEURITIC DYSTROPHIA IN ALZHEIMERS DISEASE Chad A. Dickey, Marcia N. Gordon, Donna M. Wilcock, Donna L. Herber, Melissa J. Freeman, Menchu Barcenas & Dave Morgan Alzheimers Disease Research Laborat ory, Department of Pharmacology, University of South Florida, College of Medicine 12901 Bruce B. Downs Blvd, MDC 9, Tampa, Fl 33612
93 ABSTRACT The pathology of Alzheimers disease (A D) is comprised of extracellular amyloid plaques, intracellular tau tangles, dystrophic neurites and neurodegeneration. The mechanisms by which these various pathological features arise are under intense inve stigation. Here, based on pilot gene expression studies, we have further anal yzed the relationship between Na+/K+ ATPase and amyloid using the APP+PS1 transgenic mouse model. We report that in addition to decreased mRNA expr ession, there is decreased overall activity of the enzyme specifically in the amyloid-containing hippocampi of the APP+PS1 mice. In addition, dual immunolabeling reveals an absence of Na+/K+ ATPase protein in a circumferentia l zone near congophilic plaques that is occupied by dystrophic neurites. We al so demonstrate that cerebral Na+/K+ ATPase activity can be directly inhibited by high concentrations of A In AD, the local concentration of A within the immediate vicinity of plaques may inhibit Na+/K+ ATPase activity. Loss of ionic hom eostasis may result in synaptic edema and the appearance of dystrophic neurites. It might also be expected that the electrotonic properties of processes woul d be modified by loss of ionic balance, leading to disruption of synaptic int egration and, concei vably, learning and memory. Therapies aimed at maintainin g osmotic balance by enhancing Na+/K+ ATPase activity in early stage AD may delay the diseases progression.
94 INTRODUCTION Alzheimers disease (AD) has severa l well characterized post-mortem pathological markers that include both gliosis and dystrophic neurites surrounding extracellular am yloid plaques. In addition, intracellular tangles of hyper-phosphorylated tau and massive neuro degeneration are seen later in the disease process. Mutated forms of bot h amyloid precursor protein (APP) and presenilin 1 (PS1) leads to an increased rate of amyloid deposition and therefore an earlier onset of the dementia associ ated with AD (Selkoe, 2001). Doubly transgenic mice expressing these human mut ants of the APP (Hsiao et al., 1996) and PS1 (Duff et al., 1996) genes (APP+PS1 mice; Holcomb et al., 1998) exhibit a tremendous amount of amyl oid deposition and gliosis without the formation of tangles or neuron loss, and yet they still develop anterograde amnesia as they age, similar to what is seen in the ear ly stages of AD (Morgan, 2003). Memory deficits without the loss of neurons indicate that amyloid-associ ated disruption of some step in neural processing can result in memory deficits. Previously we have described decreased expression of genes critical for learning and memory and impaired inducti on of several immediate early genes (IEGs) in aged, memory deficient APP+PS 1 mice (Dickey et al., 2003;Dickey et al., 2004). Increased neural activity dur ing learning is argued to be a primary inducing stimulus for these IEGs (D ragunow, 1996). One possible mechanism to describe this phenomenon would be that amyloid is diminishing the ability of neurons to facilitate sufficient electrical signaling to induce changes in synaptic plasticity essential for memory consolidat ion. Here we present evidence that in
95 addition to a decreased level of Na+/K+ ATPase mRNA, the activity of this enzyme is significantly decreased in t he APP+PS1 hippocampus but not in the amyloid-free cerebellum, an important finding as ouabain administration has been shown to impair memory consoli dation (Watts and Ma rk, 1971; Mark and Watts, 1971). We decided to investigat e further the interactions of A and Na+/K+ ATPase activity to understand bette r the role of this enzyme in AD and memory dysfunction. MATERIALS AND METHODS Tissue Preparation Mice were bred in our facility and genotyped using previously described methods (Gordon et al., 2001). The wo rking memory performance of the APP+PS1 mice used in these studie s was impaired when compared to nontransgenic littermates as published previ ously (Austin et al., 2003; untreated groups were studied here). Briefly, 17-18 month old mice were deeply anesthetized with pentobarbital (100 mg/k g) and perfused transcardially with phosphate buffered saline. Brains were removed and halved into right and left hemispheres. The right hemisphere was imm ediately dissected into regions and frozen on dry ice, while the left he misphere was post-fixed in 4% paraformaldehyde for 24 hours and subsequently processed through a cryoprotection schedule of 10, 20 and 30% sucros e. Frozen brains were sectioned horizontally on a slid ing microtome at 25 m and stored in Dulbeccos phosphate buffered saline plus azide at 4 C.
96 Real time RT-PCR Analysis As previously described, real time RT-PCR was used to measure expression levels of mouse Na+/K+ ATPase III (NAK3) mRNA from total RNA extraction of both hippocampus and ce rebellum tissue (Dickey et al., 2003;Dickey et al., 2004). 18S ribosomal RNA was used as the stably expressed denominator across tissue samples. The following primers for NAK3 were designed using the mouse sequence from the NCBI accession number NM_144921: Forward primer 5 CAA GAA GAG CAA GGC CAA AG 3 and reverse primer 5 TTG TAT TTC CGG CA G ACC TC 3. The sequences for the 18S rRNA primers were acquired from Schmittgen and Zakrajsek (2000). Na+/K+ ATPase Activity Assay & Amyloid Preparation An assay to detect specific activity of Na+/K+ ATPase by measuring the release of phosphate was developed using a variation of the method described by Ellis et al. (2000). Freshly frozen dissected hippocampi and cerebella (2030mg tissue weight) from APP+PS1 and non-transgenic littermates were homogenized using a rotor-stator homogenizer in 1 ml of cold suspension buffer containing 85mM NaCl, 20mM KCl, 4mM MgCl, 0.2mM EGTA and 30mM histidine pH 7.2. Saponin was added to t he samples to a final concentration of 20g/ml. They were then incubated at 37 C for 15 minutes and immediately returned to ice. Protein concentrati on was measured by Bradford assay and concentrations were adjusted to 10mg/ml tissue weight.
97 In a 96-well plate, 60l of ATP buffer containing 140mM NaCl, 20mM KCl, 3mM MgCl, 30mM histidine and 3mM ATP were added to wells. Two sets of samples were setup, one with ATP buffe r only and the other with 100M of the Na+/K+ ATPase selective inhibitor ouabain added to the ATP buffer. Subsequently, 10l of pr otein homogenates were added to the ATP buffer ouabain, which were then mixed by pipetting and incubated at 37 C for 30 minutes. The reaction was stopped by adding 120l of an acid molybdate solution consisting of 0.5g ammonium moly bdate (Sigma, St. Louis, MO) in 0.5M sulfuric acid. After mixing, 10l of Fi ske Subbarow Reducer (Sigma, St. Louis, MO) was added and wells were mixed again. The plate was allowed to incubate covered at room temperature for 10 minutes and then measured spectrophotometrically at 660nm. Specific activity was measured by taking the difference between samples ouabain and, using a standard curve of serially diluted 4mM phosphoric acid, converting this OD value to mols of inorganic phosphate (Pi) liberated/mg protein/ hour. All reactions were performed in triplicate, which were then averaged to produce the single value for the sample. Differences between APP+PS1 and non-transgenics were analyzed for significance using one-way ANOVA. For analysis of activi ty inhibition by A purified active cortical Na+/K+ ATPase (Sigma, St. Louis, MO) was pre-incubated for 2 hours in a 37 C orbital shaker separately with 112. 5, 225 and 450 g/ml of A 1-42 or vehicle only. Activity was then measured using the same activity assay as above and
98 significance was measured using one-way ANOVA comparing activity between vehicle treated and A treated. A used for these studies was gen erated by resuspending 1mg of commercially available recombinant A 1-42 peptide (rPeptide, Athens, GA) in 221l of 1,1,1,3,3,3-Hexafl uoro-2-propanol (HFIP, Sigm a, St. Louis, MO) to generate 45g A 1-42 films. These films were resuspended in 2l of dry DMSO, followed by agitation and subsequent addition of 48l of cold water and overnight incubation at 4 C. This yielded approximatel y a 900g/ml suspension of A Quality of the peptide wa s confirmed using SDS-PA GE analysis. 10l of the overnight preparation was mixed with 2x sample buffer (62.5mM Tris-HCl, 10% glycerol, 10% SDS, and 0.1% bromophenol blue), without -mercaptoethanol or heat, and loaded onto a 4-20% gradient Tris-g lycine gel (Bio-Rad, Hercules, CA) and run at 100mV for 60-90 minutes. This was followed by an overnight Comassie Blue stain of the gel and a prol onged destain in water the following day. Histology Immunohistochemical and immunofluor escence analyses for Na+/K+ ATPase III and phosphorylated neurofilament, respectively, were performed on the same 25m free-floating hippocampal sections. Sections were treated 15 minutes with 10% methanol, 10% hydr ogen peroxide and 80% PBS to block endogenous peroxidase activity, and then washed 3 times with PBS. Sections were subsequently treated with sodium borohydride for 15 minutes to reduce
99 background auto-fluorescence (Clancy and Cauller, 1998) followed by washing with PBS. Sections were then permeabiliz ed for 30 minutes with 100mM lysine, 0.2% triton x-100, 2% goat serum and 2% horse serum in PBS, and washed 3 times with PBS. Sections were then incubated overnight in a 1:300 dilution of goat F(ab)2 anti-mouse IgG (Protos Imm unoresearch, Burlingame, CA) to block endogenous mouse IgG. Sections were washed the following day and coincubated with a 1:5000 dilution of a rabbit anti-rat Na+/K+ ATPase III IgG (Upstate Biotech, Lake Placid, NY) and a 1:10,000 dilution of a mouse monoclonal IgG1 ascites pool specific fo r phosphorylated forms of neurofilament (SMI-312; Sternberger Monocl onals, Lutherville, MD) in 2% goat serum and 2% horse serum in PBS. The following day, sections were washed, and then incubated in both a 1:3000 dilution of ant i-rabbit biotinylated secondary antibody (Vector Labs, Burlingame, CA) and a 1: 100 dilution of anti-mouse fluorescein conjugated secondary antibody (Vector Labs, Burlingame, CA) for 2 hours. After washing, the tissue was incubated with Vectastain Elite ABC kit (Vector Labs, Burlingame, CA). The tissue wa s then washed and stained with a diaminobenzidine:peroxide system plus nickel enhancement (DAB/Ni2+), followed by final washes. Compact amyloid plaques were visualized using Congo red staining after sections were slide mounted and dried (Gordon et al., 2001). Briefly, slides were incubated in alkali ne alcoholic saturated sodium chloride (AASSC), followed by 0.2% Congo r ed in AASSC, then dehydrated and coverslipped with xylene-free Vectamount (Vector Labs, Burlingame, CA; other mounting media were tested, but onl y Vectamount was suited for both
100 immunohistochemistry, immunofluorescenc e and Congo red staining). The extent of nonspecific binding was assessed in the absence of primary antibodies for all assays. Specificity of the Na+/K+ AT Pase antibody was confirmed by reduced staining following a 2 hour pre-incubation of the antibody with purified cerebral ATPase at a 1 to 4 molecule ratio (1 antibody to 4 protein molecules). Both immunostains were characterized individua lly before the co-incubation procedure was implemented. Western Blot Brain homogenate from the activity assay above was equilibrated to 10g of total protein. This homogenate, along with 1 mg of purified cerebral Na+/K+ ATPase, was diluted 1:1 with loadi ng buffer containing 4% SDS and 5% mercaptoethanol, heated to boiling for 5 minutes and loaded onto a 7.5% Trisglycine gel which was then electrophor esed at 100mV for one hour in the presence of SDS. The protein was subs equently transferred onto an Immobilon membrane (Millipore, Billerica, MA) for an hour at 100mV. The blot was rinsed with borate saline + 0.05% Tween-20 (BST) and blocked overnight at 4 C in 5% non-fat dry milk (NFDM). The following day a 1:2000 dilution of anti-rat Na+/K+ ATPase III antibody (Upstate Biotech, Lake Placid, NY) in 0.5% NFDM+BST was applied to the blot for 1 hour, follo wed by washing and a subsequent 1 hour incubation with a 1:5000 diluti on of hrp-labeled anti-ra t IgG (Sigma, St. Louis, MO) in 0.5% NFDM+BST. After wash ing, the blot was developed for
101 chemiluminescence using a luminol substr ate kit (Santa Cruz Biotech, Santa Cruz, CA) and film exposure. RESULTS The pathology underlying AD is vast and complex, but through the use of transgenic models that develop various aspects of the disease, our understanding of how each patho logical feature contributes to the progression of AD can be improved. Here we have used the APP+PS1 transgenic mouse model to better understand how the deposition of amyloid and the subsequent deleterious effects of its presence are c ontributing to the consistent memory loss seen in these animals (Morgan et al ., 2000;Gordon et al., 2002;Arendash et al., 2001;Austin et al., 2003). Specifically we have found that Na+/K+ ATPase, a protein critical for not only brain function, but survival, is adversely affected by the presence of amyloid and this interacti on may be driving a major pathological event in the progression of amyloid-associated dementia. Initial quantitative real-time PCR (qRT-P CR) investigations revealed that the expression of the Na+/K+ ATPase brain-specific catalytic subunit ( III) mRNA is significantly (25-30%) decreased in the amyloid rich hippocampus of memory deficient APP+PS1 mice w hen compared to both the non-transgenic littermate hippocampus and the amyloid-fr ee cerebellum of t hese same doubly transgenic mice (Figure 1). This supported t he notion that amyloid is specifically altering expression of genes, thereby contri buting the inability of these mice to perform well in a radial arm water maze task. To determine whether the Na+/K+
102 ATPase protein itself was being modi fied in agreement wit h the reduced mRNA levels, we initially performed Western bl ot analysis; however, we found that this method was not sufficiently sensitiv e to detect the modest reduction in expression (data not shown). We turned to a more quantitative approach, using a colorimetric assay to measure activi ty of the enzyme. Using ouabain as a selective inhibitor for Na+/K+ ATPase we demonstrated that the enzymatic activity as measured by mols of Pi liberated/ mg of protein/ hour was significantly reduced by ~25% in the hippocampi of aged, memory-deficient APP+PS1 mice compared with non-transgeni c littermates (Figure 2). These values are consistent with those found 3 decades ago by Stefanovic et al. (1974). Upon analysis of the cerebellum of thes e mice, we showed that the specific enzyme activity in APP+PS1 transgenics re mained equivalent to that in nontransgenic littermates; however the overall activity of Na+/K+ ATPase in this region was only one fifth of that in the hippocampus (Figure 2). The specificity of the Na+/K+ ATPase subunit antibody used for immunochemistry was determined by severa l means. Figure 3 de monstrates that following a pre-incubation of the antibody with purified Na+/K+ ATPase protein, staining with DAB+Ni2+ was dramatically reduced (Figure 3B) compared with normal staining (Figure 3A; the bright r ed spots are amyloid plaques stained with Congo red dye). Western blot analysis of the purified enzyme revealed two major bands the largest corresponding to t he 140kD whole Na+/K+ ATPase enzyme consisting of both the and subunits and the smaller 110kD band corresponding specifically to the dissociated subunit only (Figure 3C). The
103 Western banding pattern of mouse hippoc ampal homogenate (signified by the H above the right lane) was virtually ident ical with that of the purified Na+/K+ ATPase (signified by the P above the le ft lane; Figure 3C). These data provided sufficient evidence of the reliability of ant ibody specificity to continue with more comprehensive anatomical analyses of the APP+PS1 cerebral Na+/K+ ATPase distribution. Immunohistochemical staining with DAB+Ni2+ of cerebral sections from non-transgenic and APP+PS1 mice for the Na+/K+ ATPase III subunit revealed specific membrane-associated staining as evidenced by the high magnification images in Figure 4 of CA3 pyramidal ce ll bodies (Figure 4A) and cell bodies of the insular cortex (Figure 4B) with reticu lar staining in the neuropil. Lower power images of the hippocampus in non-transgeni c mice (Figure 4C) demonstrate an absence of staining along the granular layer of the blades of the dentate gyrus. Additionally, staining is less intense within the hilus and along the mossy fiber projections to CA3, demonstrating a lower expression of this protein subunit in the inner molecular layer of CA3, but not in the pyra midal layer (Figure 4A). White matter lacked staining consis tent with the need for high Na+/K+ ATPase activity only at the nodes of Ranvier in myelinated tracts. Cortical staining (Figure 4D) revealed a fairly uni form pattern, with the exclusion of white matter. There appears to be slightly less in tense staining in cortical layers 1 and 2. The glaring difference in Na+/K+ ATPase staining between non-transgenic and APP+PS1 forebrain areas is the absence of staining where amyloid plaques are presumably present in the APP+PS1 mice (Figure 4E & F). Subsequent Congo
104 red staining of these am yloid plaques confirmed t hat the absence of Na+/K+ ATPase staining was not only seen in the center of the plaque, but also in a penumbral zone immediately surrounding the congophilic pl aque, resulting in a halo (Figure 4G & H). Higher power magnification clearly demonstrated the lack of Na+/K+ ATPase staining surroundi ng the congophilic plaques (Figure 5A; 100x: Figure 5B; 400x). These findings led us to postulate t hat the osmotic im balance brought on by an absence of Na+/K+ ATPase may result in the neuritic dystrophia surrounding plaques, a pathological featur e consistently seen in APP+PS1 mice. We designed a dual immunostain utiliz ing immunohistochemical methods for Na+/K+ ATPase and immunofluore scence methods for phosphorylated neurofilament, a dystrophic neurite marker we evaluated earlier (Gordon et al., 2002). Figure 5B is representative of our findings for Na+/K+ ATPase with Congo red and Figure 5C is likewise represent ative of our findings for phosphorylated neurofilament (dystrophic neurites) on the sa me plaque. Using this assay, we were able to demonstrate that the dystr ophic neurites were almost exclusively present within the zone surrounding the congophilic plaque that lacked Na+/K+ ATPase staining, as represented by an ov erlay of the fluorescent neurites from Figure 5C onto the bright fi eld image of Na+/K+ ATPase and Congo red staining from Figure 5B (Figure 5D). Finally, to determine whether amyloid could inactivate Na+/K+ ATPase activity, we measured the activity of purified cerebral Na+/K+ ATPase after exposure to various concentrations of A 1-42 peptide. Figure 6A shows that the
105 amyloid preparation contained peptide fo rms corresponding to ~15, ~11, and ~4kD (right lane), when compared to the mo lecular weight standards in the left lane. Figure 6B demonstrates that in creasing concentrations of A dosedependently reduced Na+/K+ ATPase activity. There were significant reductions in Na+/K+ ATPase activity at the lower concentrations of 112 and 225 g/ml compared to vehicle, but maximal reductions were at the highest concentration of amyloid (450g/ml). DISCUSSION Over the past 6 years, our group has characterized various aspects of the APP+PS1 transgenic mice including their pathology, behavior, and gene expression. With age, these mice progre ssively develop more amyloid plaques surrounded by dystrophic neurites, activat ed microglia and astr ocytes (Gordon et al., 2002). With increasing amyloid burden, aged animals consistently develop memory deficits in the radial arm water maze (RAWM; Holcomb et al., 1999;Arendash et al., 2001;Gordon et al., 2001). We have also demonstrated that several genes critical for synaptic plasticity and memory consolidation are down regulated in these mice in regions accumulating amyloid (Dickey et al., 2003) and the induction of a subset of immediate-early genes is impaired when the transgenic mice are introduced to a novel environment (Dickey et al., 2004). Here, we have shown that Na+/K+ ATPase has decreased mRNA expression and enzyme activity in the amyloid containing hippocampus of APP+PS1 transgenic mice. We have also de monstrated by immunohistology that
106 Na+/K+ ATPase protein expression is reduced in the immediate vicinity of congophilic plaques, a zone where dystr ophic neurites are most prevalent, suggesting that disrupted ionic homeostasi s may contribute to their formation. Additionally, high concentrations of A 1-42 directly inhibit the activity of Na+/K+ ATPase. This suggests that in the ar ea surrounding amyloid plaques, where the local A concentration is likely high, Na+/K+ ATPase activity is inhibited, precipitating edema in neurites close to the plaques. From previous gene expression studies we found that the mRNA for the Na+/K+ ATPase III subunit was significantly dow n-regulated in the hippocampi of APP+PS1 mice compared to non-trans genic littermates and to the amyloidfree cerebella (Dickey et al., 2003). These reductions were also demonstrated in human Alzheimers disease samples, c onsistent with data from previous investigations (Chauhan et al., 1997; Dickey et al., 2003). Here, we have replicated these findings in a s eparate cohort of aged APP+PS1 mice, demonstrating a consistent deficiency (F igure 1). Two major advantages of the qRT-PCR method are sensitivity and r obustness, as discrete changes in expression can be measured and sample si ze can be increased to allow for more reliable statistical analyses, respecti vely. One problem is that current conventional methods used to measure protein expression lack the same sensitivity now available for RNA analyses, and therefore may not discern similar phenomena at the protein leve l. Using Western blot analysis, we were able to see slight differences in Na+/K+ ATPase protein level in transgenic mice; however this approach lacked statistical significance and was less than
107 convincing to our eyes (a representative blot is shown in Figure 3C). Using a sensitive colorimetric assay to measure activity of Na+/K+ ATPase developed by Ellis et al. (2000), we were able to demonstrate that in the APP+PS1 hippocampus, the activity of ouabain-sensit ive ATPase was significantly impaired (Figure 2) while Na+/K+ ATPase activity in the amyloid-free cerebellum remained unperturbed with respect to genotype. Cerebel lar activity was substantially lower than that seen in the non-tr ansgenic hippocampal tissue, perhaps indicative of the abundant white matter found in this region, where Na+/K+ ATPase would predominantly be located at the nodes of Ranvier. These data demonstrate that the function of Na+/K+ ATPase is pertur bed in a brain region that contains high overall concentrations of A Previous investigations ha ve shown that in cultured neurons, Na+/K+ ATPase activity can be blocked directly by various A peptide fragments (Mark et al., 1995) and other studies have shown that Na+/K+ ATPase protein levels are decreased in AD tissue but not in normal aged tissue (Harik et al., 1989;Liguri et al., 1990) But, this is the first time that reduced activity has been demonstrated in homogenates from an animal model for amyloid deposition indicating that in vitro effects of A retain relevance in vivo It remains uncertain if the A in the homogenate is direct ly inhibiting the enzyme in vitro as the homogenate is diluted 20 fold in the assay. However, even if this reduced activity results from homogenate-derived A inhibition in vitro it clearly indicates there is sufficient A in vivo to block Na+/K+ ATPase activity. We then immunostained APP+PS1 mouse sections for Na+/K+ ATPase to better understand where the reductions in enzyme level were occurring. Initial
108 examination of hippocampal staining rev ealed that in both the non-transgenic and APP+PS1 mice, Na+/K+ ATPase staining was found to be membrane specific (Figures 3A & B). Analysi s of staining in both APP+PS1 and nontransgenic tissue also revealed reductions in Na+/K+ ATPase along the mossy fiber pathway from dentate gyrus to CA3 and in the hilus and slight reductions in layers 1 and 2 of the cortex (Figure 4A-F ). Of particular interest, in areas where congophilic plaque staining wa s apparent, Na+/K+ ATPase staining was absent, and more specifically, there appeared to be no or little Na+/K+ ATPase staining in a penumbral zone surrounding the plaque s stained with Congo red (Figures 4E-H and 5A-B). From previous studies, we knew that dystrophic neurites could be visualized in the APP+PS1 mice dire ctly adjacent to congophilic amyloid plaques (Gordon et al., 2002) and the finding of reduced Na+/K+ ATPase staining in this same area suggests t hat dysregulation of the enzyme may in some way be linked to the disr uption of neurite integrity. Subsequently, we stained sections for Na+/K+ ATPase and dystrophic neurites together, along with amyloid plaqu es using Congo red. Because of the ubiquitous distribution of both Na+/K+ ATPase and pho sphorylated neurofilament (used to label dystrophic neurites), we found empirically that staining the ATPase immunohistochemically and the dystrophic neurites with a fluorescein labeled secondary antibody was the most effectiv e way to see both markers on the same section along with the Congo red stained plaques. Imagi ng of these sections did reveal that dystrophic neur ites are in the circumferential area surrounding the congophilic amyloid plaques wh ere Na+/K+ ATPase stai ning is absent (Figure
109 5D). One possible explanation for this would be that amyloid binding to the Na+/K+ ATPase and inhibiting its activi ty would cause the neurites to begin swelling, resulting in dysregulation of neuronal si gnaling. Another possible mechanism for the reductions in Na+/K+ ATPase would be that interactions of amyloid with surface proteins, such as integrins (Sabo et al., 1995;Goodwin et al., 1997) and focal adhesion proteins (Willia mson et al., 2002) lead to activated signal transduction cascades that medi ate tyrosine phosphorylation (Grace and Busciglio, 2003) thereby pr omoting the removal of ATPase from the neuronal membrane and its subsequent down -regulation. A recent st udy suggests that the tyrosine kinase, Lyn, can phosphorylate Na +/K+ ATPase resulting in its removal from the membrane (Bozulic et al ., 2004). When taken t ogether, these data suggest that either direct inhibition of Na+/K+ ATPase by amyloid or its removal due to amyloid-mediated activation of a signaling cascade, could result in the formation of dystrophic neurites due to osmotic and ionic imbalances. A final investigation analyzed the e ffects of amyloid peptide on the activity of a purified cerebral preparation of ac tive Na+/K+ ATPase. Amyloid has been shown to bind various cell surface protei ns (Mark et al., 1995;Dineley et al., 2001;Verdier and Penke, 2004) and i nduce neuro-toxicity (Caughey and Lansbury, 2003;Stine, Jr. et al ., 2003;Gong et al., 2003). To determine the effect of A 1-42 on Na+/K+ ATPase activity, we pre-incubated purified Na+/K+ ATPase with varying concentrations of am yloid then colorimetrically measured its activity. Although amyloid did suppress activity at 112 g/ml (~10M) and 225 g/ml (~50M), it was the highest c oncentration (450g/ml or ~100M) that
110 precipitated the largest reduction in activi ty compared to vehicle, nearly rendering it completely inactive (Figure 6B). This suggests that the A 1-42 peptide can directly binding to the Na+/K+ ATPase and affecting its activity. Mark et al. suggests that the 25-35 amino acid region of the A peptide induces oxidative stress thereby impairing Na+/K+ ATPase activity (Mark et al., 1995), and the findings presented herein strengthen the ar gument for the role of amyloid in Na+/K+ ATPase dysregulation. Addition ally, the presence of short 11 and 15 kD polymers of amyloid in the DMSO/water preparation suggests that these might be what is driving this reduced activity. These data indicate that reduced activity in Na+/K+ ATPase may be elicited by high concentrations of A which may then lead to local neuritic edema and the appearance of dystrophic neurites in APP+PS1 mice. One area in the AD brain that might harbor such a high concentration of A as needed to suppress the activity of Na+/K+ ATPase woul d be the microdomain near and immediately around the plaques. This is the area demonstrated to have reduced immunostaining for Na+/K+ ATPase. A previous study has suggested that when Na+/K+ ATPase activity is specifically inhibited by ouabain, mice are unable to consolidate new memories (Watts and Mark, 1971) and in addition cells exposed to ouabain have reduced Na+/K+ ATPase III subunit mRNA expression (Huang et al., 1997), confirming our findings that direct inhibition of Na+/K+ ATPase activity can result in reduced mRNA expression, possibly through a yet unidentified signaling cascade. These data suggest that drugs targeted at activating Na+/K+ ATPase and maintainin g ionic balance in these neurons may
111 benefit Alzheimers patients by delaying bot h the onset of neurit ic dystrophia and memory dysfunction. ACKNOWLEDGEMENTS This work was supported by AG15490 and AG18478 from NIH.
112 REFERENCES Arendash GW, King DL, Gordon MN, Morgan D, Hatcher JM, Hope CE, Diamond DM (2001) Progressive, age -related behavioral impairments in transgenic mice carrying both mutant amyl oid precursor protein and presenilin-1 transgenes. Brain Res 891: 42-53. Austin LA, Arendash GW, Gordon G, Diamond DM, DiCarlo G, Dickey C, Ugen KE, Morgan D (2003) Short-term Abeta vaccinations do not improve cognitive performance in cognitively-impaired APP+PS1 mice. Behavioral Neuroscience. Bozulic LD, Dean WL, Delamere NA ( 2004) The influence of Lyn kinase on Na,K-ATPase in porcine lens epitheliu m. Am J Physiol Cell Physiol 286: C90C96. Caughey B, Lansbury PT (2003) Prot ofibrils, pores, fibrils, and neurodegeneration: separating the responsible protein aggregates from the innocent bystanders. Annu Rev Neurosci 26: 267-298. Chauhan NB, Lee JM, Siegel GJ (1997) Na,K-ATPase mRNA levels and plaque load in Alzheimer's disease. J Mol Neurosci 9: 151-166. Clancy B, Cauller LJ (1998) Reducti on of background autofluorescence in brain sections following immersion in sodium borohydride. J Neurosci Methods 83: 97-102. Dickey CA, Gordon MN, Mason JE, Wilson NJ, Diamond DM, Guzowski JF, Morgan D (2004) Amyloid suppresses in duction of genes critical for memory consolidation in APP + PS1 transgeni c mice. J Neurochem 88: 434-442. Dickey CA, Loring JF, Eastman PS, Montgomery JR, Gordon M, Morgan DG (2003) Selectively reduced expression of synaptic plasticity-related genes in APP+PS1 transgenic mice. J Neurosci 23: 5219-5226. Dineley KT, Westerman M, Bui D, Be ll K, Ashe KH, Sweatt JD (2001) Beta-amyloid activates the mitogenactivated protein kinase cascade via hippocampal alpha7 nicotinic acetylcholi ne receptors: In vitro and in vivo mechanisms related to Alzheimer's di sease. J Neurosci 21: 4125-4133. Dragunow M (1996) A role for immediat e-early transcription factors in learning and memory. Behav Genet 26: 293-299. Duff K, Eckman C, Zehr C, Yu X, Pr ada CM, Perez-tur J, Hutton M, Buee L, Harigaya Y, Yager D, Morgan D, Gor don MN, Holcomb L, Refolo L, Zenk B,
113 Hardy J, Younkin S (1996) Increased am yloid-beta42(43) in brains of mice expressing mutant presenili n 1. Nature 383: 710-713. Ellis DZ, Nathanson JA, Sweadner KJ (2000) Carbachol inhibits Na(+)K(+)-ATPase activity in choroid plexus via stimulation of the NO/cGMP pathway. Am J Physiol Cell Physiol 279: C1685-C1693. Gong Y, Chang L, Viola KL, Lacor PN Lambert MP, Finch CE, Krafft GA, Klein WL (2003) Alzheimer's disease-affe cted brain: presence of oligomeric A beta ligands (ADDLs) suggests a molecula r basis for reversible memory loss. Proc Natl Acad Sci U S A 100: 10417-10422. Goodwin JL, Kehrli ME, Jr., Uemura E (1997) Integrin Mac-1 and betaamyloid in microglial release of ni tric oxide. Brain Res 768: 279-286. Gordon MN, Holcomb LA, Jantzen PT, DiCarlo G, Wilcock D, Boyett KW, Connor K, Melachrino J, O'Callaghan JP, Morgan D (2002) Time course of the development of Alzheimer-like pathol ogy in the doubly transgenic PS1+APP mouse. Exp Neurol 173: 183-195. Gordon MN, King DL, Diamond DM, Jantzen PT, Boyett KV, Hope CE, Hatcher JM DiCarlo G, Gottschall WP, Morgan D, Arendash GW (2001) Correlation between cognitive deficit s and Abeta deposits in transgenic APP+PS1 mice. Neurobiol Aging 22: 377-385. Grace EA, Busciglio J (2003) Abe rrant activation of focal adhesion proteins mediates fibrillar amyloid bet a-induced neuronal dystrophy. J Neurosci 23: 493-502. Harik SI, Mitchell MJ, Kalaria RN ( 1989) Ouabain binding in the human brain. Effects of Alzheimer's dis ease and aging. Arch Neurol 46: 951-954. Holcomb L, Gordon MN, McGowan E, Yu X, Benkovic S, Jantzen P, Wright K, Saad I, Mueller R, Morgan D, S anders S, Zehr C, O'Campo K, Hardy J, Prada CM, Eckman C, Younkin S, Hsiao K, Duff K (1998) Accelerated Alzheimertype phenotype in transgenic mice carry ing both mutant amyloid precursor protein and presenilin 1 tr ansgenes. Nat Med 4: 97-100. Holcomb LA, Gordon MN, Jantzen P, Hsiao K, Duff K, Morgan D (1999) Behavioral changes in transgenic mice expressing both amyloid precursor protein and presenilin-1 mutations: lack of association with amyloid deposits. Behav Genet 29: 177-185. Hsiao K, Chapman P, Nilsen S, Eckman C, Harigaya Y, Younkin S, Yang F, Cole G (1996) Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice. Science 274: 99-102.
114 Huang L, Kometiani P, Xie Z (1997) Differential regulation of Na/KATPase alpha-subunit isoform gene expres sions in cardiac myocytes by ouabain and other hypertrophic stimuli. J Mol Cell Cardiol 29: 3157-3167. Liguri G, Taddei N, Nassi P, Latorra ca S, Nediani C, Sorbi S (1990) Changes in Na+,K(+)-ATPase, Ca2( +)-ATPase and some soluble enzymes related to energy metabolism in brains of patients with Alzheimer's disease. Neurosci Lett 112: 338-342. Mark RF, Watts ME (1971) Drug i nhibition of memo ry formation in chickens. I. Long-term memory. Proc R Soc Lond B Biol Sci 178: 439-454. Mark RJ, Hensley K, Butterfield DA Mattson MP (1995) Amyloid betapeptide impairs ion-motive ATPase activi ties: evidence for a role in loss of neuronal Ca2+ homeostasis and cell death. J Neurosci 15: 6239-6249. Morgan D (2003) Learning and memory deficits in APP transgenic mouse models of amyloid deposition. Neurochem Res 28: 1029-1034. Morgan D, Diamond DM, Gottschall PE, Ugen KE, Dickey C, Hardy J, Duff K, Jantzen P, DiCarlo G, Wilcock D, Connor K, Hatcher J, Hope C, Gordon M, Arendash GW (2000) A beta peptide vaccination prevents memory loss in an animal model of Alzheimer's disease. Nature 408: 982-985. Sabo S, Lambert MP, Kessey K, Wade W, Krafft G, Klein WL (1995) Interaction of beta-amyloi d peptides with integrins in a human nerve cell line. Neurosci Lett 184: 25-28. Schmittgen TD, Zakrajsek BA (2000) E ffect of experimental treatment on housekeeping gene expression: validation by real-time, quantitative RT-PCR. J Biochem Biophys Methods 46: 69-81. Selkoe DJ (2001) Clearing the brain's amyloid cobwebs. Neuron 32: 177180. Stefanovic V, Ebel A, Hermetet JC, Mandel P (1974) Na+-K+-ATPase activity in brain regions of C57 and DBA mice. J Neurochem 22: 1139-1141. Stine WB, Jr., Dahlgren KN, Kra fft GA, LaDu MJ (2003) In vitro characterization of conditions for am yloid-beta peptide oligomerization and fibrillogenesis. J Biol Chem 278: 11612-11622. Verdier Y, Penke B (2004) Binding Site s of Amyloid beta-Peptide in Cell Plasma Membrane and Implications for Al zheimer's Disease. Curr Protein Pept Sci 5: 19-31.
115 Watts ME, Mark RF (1971) S eparate actions of ouabain and cycloheximide on memory. Brain Res 25: 420-423. Williamson R, Scales T, Clark BR, Gibb G, Reynolds CH, Kellie S, Bird IN, Varndell IM, Sheppard PW, Everall I, Anderton BH (2002) Rapid tyrosine phosphorylation of neuronal proteins in cluding tau and focal adhesion kinase in response to amyloid-beta peptide exposure: involvement of Src family protein kinases. J Neurosci 22: 10-20.
116 FIGURE 1. Expression of Na+/K+ ATPase mRNA in APP+PS1 mice by qRTPCR. This figure shows APP+PS1 transgenic mouse expression (n=7-8 APP+PS1, open bars) as a percentage of the average of non-transgenic expression (n=7-8 Non-Tg, solid bars) for the Na+/K+ ATPase III subunit in the hippocampus and the cerebellum. In 18 m onth old APP+PS1 mice, the amyloid free cerebellum maintains expression equiv alent to that of non-transgenic mice. However, hippocampal expression of Na+/K+ ATPase is significantly decreased in APP+PS1 mice compared with age matched littermates, indicating that the amyloid contained within this region ma y be contributing to the decrease in mRNA expression. The values repres ented in this figure are the mean + SEM. indicates significant differences between APP+PS1 mi ce and non-transgenic littermates (p < 0.05) when measur ed by one-way ANOVA. indicates differences (p < 0.05) in expressi on between hippocampus and cerebellum.
117 0 25 50 75 100 125 HippocampusCerebellumChange in Expression as Percentage of Non-Tg SEM*
118 FIGURE 2. Activity of the Na+/K+ ATPase enzyme in APP+PS1 mice Activity of ouabain-sensitive ATPase was assayed colorimetrically in APP+PS1 mice tissue (n = 8, open bars) and non-transgenic littermate tissue (n = 8, solid bars) and is presented here as mols of phosphate liberated by Na+/K+ ATPase per mg of protein in one hour. Both amyloid-free cerebellar tissue and amyloid containing hippocam pal tissue were analyzed. There was no decrease in APP+PS1 cerebellar Na+/K+ ATPase activity compared to nontransgenic mice. However, in the amyloi d-containing hippoc ampus, there was a significant decrease in the activity of the enzyme. Cerebellar activity was only 20% of that seen in non-transgenic hi ppocampus. The free-phosphate/molybdate complex was measured at 660nm following a 30 minute incubation. Specific activity was calculated as the difference between samples ouabain. A phosphoric acid standard curve was used to calculate the molar concentration of phosphate liberated by Na+/K+ ATPase/m g total protein/hour. The values represented in this figure are the mean + SEM. indicates significant differences between APP+PS1 mice and non-trans genic littermates (p < 0.05) when measured by one-way ANOVA.
119 0 5 10 15 20 HippocampusCerebellummols of Pi liberated / mg protein/ hour SEM *
120 FIGURE 3. Verification of Na+/K+ ATPase III antibody specificity by immunohistochemistry follo wing pre-incubation with purif ied protein and Western blotting. Horizontal sections were i mmunostained for Na+/K+ ATPase III with (Panel B) and without (Panel A) pre-incubation with 70 units of purified enzyme followed by DAB with nickel intensific ation and Congo red staining. The preincubation significantly decreased staining, confirming antibody specificity. Panel C is a representative western blot for Na+/K+ ATPase III using both purified Na+/K+ ATPase (P lane) and hippocampal homogenate (H lane). Both lanes have two predominant bands at 140 and 110 kD, corresponding to the whole enzyme and the subunit, respectively, as indicat ed by the arrows. Scale bar for panels A and B = 50m.
122 FIGURE 4. Hippocampal and cortical immunohist ochemistry for Na+/K+ ATPase in APP+PS1 mice and non-transgenic littermates. Twenty five m horizontal hippocam pal and cortical sections were immunostained for Na+/K+ ATPase III followed by DAB with Ni2+ intensification. Left panels are hippocampus, right panel s are cortical sections. High power magnification (Panels A & B; Scale bar = 8.33m) revealed membrane-specific staining of neuronal ce ll bodies for Na+/K+ ATPase in the insular cortex (Panel B) and CA3 of t he hippocampus with less intense staining of the mossy fiber pathway (Panel A). Hippocampal staining of non-transgenic mice showed a ubiquitous distribution of Na+/K+ ATPase throughout the gray matter, while less staining was apparent along the pyramidal layer of Ammons horn, the hilus and the granular layer of the dentate gyrus (Panel C). Cortical staining was also substantial with slight faintness in layers 1 and 2 (Panel D). White matter in both structures remai ned unstained. A similar staining pattern was observed in the APP+PS1 mice wit h the exception of large non-stained holes in the gray matter, presumably w here amyloid plaques are located (Panels E & F). Immunostaining followed by Congo red histochemistry co nfirmed that the Na+/K+ ATPase staining was absent not only immediately where plaques are located, but also in a zone surrounding the plaque (Panels G & H). Scale bar for panels C-H = 120m.
124 FIGURE 5. Dual Immunostaining of Na+/K+ ATPase III and dystrophic neurites. Horizontal sections were immunosta ined for Na+/K+ ATPase using DAB with nickel intensification followe d by Congo red staining for A plaques after which a discernable circumferential void in ATPase staining surrounding the plaque was observed (Panel A 100x magnifi cation; Panel B 400 x magnification). Immunofluorescent staining of dystrophic neurites usin g the anti-phosphorylated neurofilament antibody SMI312 followed by Congo red st aining demonstrate a close relationship between amyloid pla ques and dystrophic neurites (Panel C; Congo red yellow fluorescence was digitall y suppressed to more clearly reveal the green-stained neurites). This double immunostaining procedure is shown as a digital overlay of the fluorescein-l abeled dystrophic neurite image onto the bright field peroxidase-labeled Na+/ K+ ATPase + Congo red image in panel D which demonstrates that dystrophic neur ites are present predominantly in the zone devoid of Na+/K+ ATPase staining surrounding the congophilic plaques (C). Scale bar for Panel A = 50m; scal e bar for Panels B-D = 16.67m.
126 FIGURE 6. Amyloid-beta 1-42 inhibits Na+/ K+ ATPase activity at high concentrations. SDS-PAGE analysis followed by comassi e blue staining revealed; 15 kD tetramers, 11kD trimers and 4kD monomers of A were present in the DMSO/water suspension (Panel A; left). Purified cerebral Na+/K+ ATPase was pre-incubated with three concentrations of the A 1-42 suspension. Panel B describes the effects of A on Na+/K+ ATPase activity as a percentage of the vehicle. The activity measured in t he presence of each concentration of A 1-42 is indicated by the diamond symbol connected by a line. These findings demonstrate significant effects of all three concentrations of A on activity, with 75% ablation of activity at the highest concentration of 450 g/ml. The assay was measured spectrophotometrically at 660nm following a 2 hour pre-incubation with the A preparations and a subsequent 30 minute reaction time. Values are presented here as a percentage of the ac tivity of purified Na+/K+ ATPase incubated vehicle. The effect of the 2 hour pre-incubation on activity is negligible and indicated by the 0 g/ml concentration. A phosphoric acid standard curve was used to calculate the molar concent ration of phosphate liberated by Na+/K+ ATPase. The values represented in this figure are the mean + SEM. (p < 0.05) and *** (p < 0.001) indicates significant differences between vehicle + A and vehicle alone when meas ured by one-way ANOVA.
128 CONCLUSIONS Since the creation of the first trans genic mouse in 1982 by Palmiter, Chen and Brinster (Palmiter et al., 1982) and t he subsequent use of embryonic stem cells to knockout genes in mice (Bradl ey et al., 1984), the ability to alter molecular expression has lead to many ground breaking discoveries in medical sciences. Through the use of transgenic and knockout mice, scientists have been able to identify the function of many hum an genes as well as the causes for a host of diseases plaguing mankin d. Over the past 2 decades, many manipulations of these two systems hav e been reported allowing for expression of multiple genes in a single animal or the ability to induce these systems with a dietary supplement (Mortensen, 1993;S aam and Gordon, 1999;Shockett et al., 1995;Bockamp et al., 2002). Alzheimers disease (AD) relies heavily on these technologies as the pathology associated with AD is an effe ct of the gradual accumulation of abnormal proteins in the brains of patient s that eventually leads to the clinical manifestation of severe cognitive damage. Only a model that can duplicate this slow deleterious progression of pathology can be used to develop treatments for AD. Pathological indicators for AD inclu de forebrain amyloid deposition resulting in extra-cellular plaques of -pleated sheets and intracellular accumulation of
129 hyperphosphorylated tau filam ents. A proper model for AD would also realize severe neuron loss and glial activation, as well as dystrophic neurites. But perhaps the most important f eature for a model of AD to have is the inability to consolidate newly acquired information into lasting memories. Mutant forms of the human APP and PS1 genes, which have been linked to a more aggressive form of AD pat hological progression, were used to generate transgenic mice. Singly trans genic APP mice were demonstrated to develop amyloid pathology and glial activa tion (Hsiao et al., 1996), while singly transgenic PS1 mice lacked any significant plaque pathology (Duff et al., 1996); however, neither model developed consistent memory deficits. These mice were then crossed to breed viable doubly trans genic progeny (Holcomb et al., 1998). The over-expression of the two transgenes led to an enhanced forebrain amyloid pathology and glial activation. In addition, these mice developed cognitive deficits that were correlated with amyloid depositi on (Holcomb et al., 1999). Since that time, the APP+PS1 doubly transgenic m ouse has provided a good model for therapeutic development against amyloid-as sociated dementias. Over the past 6 years, our group has thor oughly characterized the pathology and behavior of these mice, and more recently we have begun to analyze the molecular changes in the APP+PS1 brain that may present other therapeu tic targets for future benefit to AD sufferers. Here we present evidence describ ing altered gene expression in the APP+PS1 mice compared to non-transgeni c littermates. Initial microarray analysis revealed that only 5 genes were down-regulated in the amyloid-
130 containing forebrain of the APP+PS1 mi ce, whereas the amyloid-free areas lacked any significant down-regulation; Dic key et al., 2003). These 5 genes have been linked to synaptic plasticity and me mory function, demonstrating their potentially critical role in the memory deficits of APP+PS1 mice. Additionally, expression of genes associated with infl ammation and the acute phase reaction were up-regulated selectively in the amyloi d-containing regions (Paper I; Table 2) while other genes, such as APP, were upregulated in both amyloid-containing and amyloid free regions compared to non-transgenic mice due to transgene over-expression (Paper I; Table 3). Using quantitative real time PCR (q RT-PCR), we were able to confirm most of the microarray results and add several others genes in the same categories as those found by the microa rray (Paper I; Figure 1). One important feature of these data was the finding that multiple genes also associated with synaptic plasticity and mnemonic processes, such as Gap43, Rheb and several synaptic vesicle related markers, we re not down-regulated in the amyloidcontaining region of APP+PS1 mice, sugges ting that specific mechanisms are being interfered with in the APP+PS1 forebr ain in response to the presence of amyloid. As of this writing, we hav e found a total of 11 genes that are downregulated in the amyloid containing hi ppocampi of 17-18 month old APP+PS1 mice, and each of these is involved in synaptic plasticity to some degree. Of these 11 genes, 6 belong to the inducible immediate-early gene category while the other 5 are more stably expressed, but are still linked to lasting changes in synaptic morphology and function (Table 1). LTPor LTD-induced synaptic
131 plasticity and the resultant gene expres sion have been shown to be essential for the long-term storage of memories in the cortex (for review see Steward, 2002 and Lynch, 2004). These findings emphas ize that the amyloid-associated disruption of synaptic plasticity-relat ed mRNA expression in the APP+PS1 mice may be responsible for their inability to consolidate memories. We also analyzed expression of these genes in both AD and nondiseased human tissue which revealed that the amyloid-containing region had decreased expression of all mRNAs anal yzed, while the amyloid-free tissue maintained normal expression (Paper I; Figur e 2). We attribut ed this non-specific loss of all mRNAs to the neurodegeneration seen with AD which is absent in the APP+PS1 mice. Together, these findings dem onstrate that in amyloid-containing brain regions of memory-deficient APP+PS 1 mice, there is a selective reduction of mRNAs associated with learning and memory, and that this impaired expression may be contributing to their memory dysfunction. To further analyze the dynamics of t hese reductions in mRNA expression, we examined tissue from APP+PS1 mice at three time points. Due to the expense associated with qRT-PCR, we chose three representative genes from both the IEG (Arc, Zif268 and Nur77) and other plasticity-related (CaMKII AMPA1 and Na+/K+ ATPase III) categories, as well as two non-changing genes (synaptophysin and Gap43). Mice at 2 mont hs of age lacked any reductions in expression of the genes we analyzed, a si gnificant finding because it ensured that the reductions seen in the older mice were not due simply to APP and PS1 over-expression, but rather the pathological accumula tion of amyloid elicited by
132 their expression (Paper II; Figure 1). Anot her interesting outco me of this time course analysis was that at 6 months, an age when fibrillar amyloid plaques are just beginning to appear, the IEGs were already significantly down-regulated, while the other plasticity-related genes had not yet demonstrated significant reductions (Paper II; Figure 1); however by 18-months, when memory deficits and high amyloid burden are consistently detectable (Gordon et al., 2002;Gordon et al., 2001), both sets of genes were down-regulated (Paper II; Figure 1), suggesting that amyloid must be present at a sufficient ly high concentration to block expression of both IE G and other plasticity-related mRNAs to initiate cognitive decline. Additional evidence for the enhanced s ensitivity of IEG expression to lower amounts of A was found when we analyzed APP and PS1 singly transgenic mice. The expressi on of IEGs was significantly decreased in APP only mice whereas the other plasticity re lated genes were not down-regulated, a similar phenomenon to that seen in the 6-month old APP+PS1 mice (Paper II; Figure 2). Interestingly, 18-month ol d APP singly transgenic mice and 6-month old doubly transgenic mice have a very similar amyloid load and both lack detectable memory deficits. The amyloidfree PS1 singly transgenic mice did not exhibit any significant loss in expre ssion of the genes we analyzed (Paper II; Figure 2). This emphasizes the signi ficance of the massive amyloid burden found in aged APP+PS1 mice and the necessi ty for the down-regulation of not just the IEGs, but also the other plasti city related genes for memory deficits to occur.
133 Previously, we described that gene expression in the amyloid-free cerebellum remained consistent with nontransgenic littermates, demonstrating specificity for the down-regulated genes (Paper I; Figure 1). Subsequent expression analyses of the posterior cort ex from 18-month old APP+PS1 mice, a region with a less concentra ted amount of amyloid, and the caudate nucleus (striatum), which lacks any formation of plaques but has a high level of diffuse amyloid load, revealed once again that IEG expression is dysregulated by the presence of lower plaque load than the other plasticity-related genes. IEG expression was inhibited in all three regi ons analyzed, while the other plasticityrelated genes were only down-regulated in the hippocampus (Paper II; Figure 3). The down-regulation of IEG expression in the striatum is of parti cular interest due to the lack of plaque formation, but high soluble A load. This perhaps provides the greatest evidence that amyloid itself is facilitating this disruption in some way. Together, the time course, genoty pe and anatomical analyses of expression provide thorough evidence t hat amyloid is causing the downregulation of IEG expression, but it is only when plaque formation is dramatically increased that the other plastici ty-related genes are down-regulated and detectable memory function is disrupt ed. One possible reason for the apparent necessity of both sets of genes to be down -regulated for memory deficits to occur could be due to the lack of sensitivity of current behavioral trials. It could be that memory is just beginning to fail at the time when IEG expression is first downregulated. Recent reports have descri bed impaired cognitive function when IEG expression is dramatically knocked down or ablated entirely (Guzowski et al.,
134 2000;Bozon et al., 2003), but current methods to detect discreet changes associated with marginally down-regulated expression, as is the case of the APP+PS1 mice, may not yet be su fficiently discriminating. The issue of the molecular mechani sms behind the amyloid-associated memory loss presents a unique challenge be cause of the slow development of pathology. Initially, we decided to address the issue of whether the induction of IEG expression or the low-level basal expression was impaired in the APP+PS1 mice. This was critical because these genes are massively up-regulated following transcriptional induction, at which point they are capable of e liciting their most robust effects. Following induction, IEG expression transiently returns to near undetectable levels within 24 hours. Previ ously, to control for exactly this possibility, each mouse was transported from the vivarium to the laboratory and set aside for 30 minutes, assuming t hat the environmental change would be sufficient to induce IEG expression. To test if induction was in fact bei ng sufficiently facilitated by a simple environmental change, we introduced one of two cohorts of mice (each cohort consisted of 7-8 APP+PS1 mice and 78 non-transgenic littermates) to a novel environment for 5 minutes and then returned them to a new cage for 30 minutes prior to sacrifice. The other cohort was swiftly removed from their home cage and immediately anesthetized. We designated the former as induced and the latter as basal. Quantitative RT -PCR of the hippocampi of these mice revealed that both APP+PS1 mice and non-transgenic li ttermates experienced significant induction of the IEGs Arc, Nur77 and Zif268, however the induction of Arc and
135 Nur77 in the APP+PS1 was signific antly impaired compared to the nontransgenic mice (Paper II; Figure 4). Additi onally, the expression of the basal animals was the same for both trans genic and non-transgenic. This was an essential piece of evidence, as it ensur ed that our previous data were measuring functional expression of t he IEGs. Zif268 induction was not significantly impaired in this paradigm, a finding not unexpected as the half-life of Zif268 message is longer than most IEGs, ma king it more difficult to analyze based on a single behavioral event. But, similar to the other 2 genes, the basal expression remained the same for both APP+PS1 and non-transgenic mice, further validating the method of induction us ed in our previous studies. There are several possible argum ents as to why IEG expression is decreased, including that amyloid may be binding very specifically to a cell surface protein, inhibiting a signaling ca scade essential for IEG activation. The MAPK signaling cascade, specific ally ERK, and CREB have both been implicated in the transcriptional initia tion of IEG expression (Sgambato et al., 1998;Ying et al., 2002;Murphy et al., 2004). Particularly, activation of the NMDA receptor has been shown to elicit expressi on of several IEGs, and this activation requires the MAPK cascade (Walker and Carlock, 1993;Cammarota et al., 2000; Sato et al., 2001). Amyloid and its vari ous peptide fragments have been shown to inhibit LTP and facilitate LTD in an NMDA receptor-dependent manner (Kim et al., 2001). Therefore it is possible that am yloid is disrupting LTP via a specific interaction with the NMDAR, which in tu rn prevents sufficient stimulation to induce IEG expression. To further postulate, the reduc ed expression of the NR2B
136 subunit in the APP+PS1 mice at a poi nt when amyloid-burden is high may suggest a negative feedback mechanism for the LTP system, in which perpetual ligand binding of the NMDAR eventua lly results in its decreased mRNA expression, similar to that seen wi th thyroid hormone receptor systems (Guissouma et al., 2000). However, these putative explanations are in conflict with the fact that the clinical use of low concentrations of the NMDAR antagonist, memantine (for review see Danysz and Parsons, 2003) is beneficial for the treatment of AD. Perhaps the clinical mani festation of AD is due in part to an over-activation of brain circuitry, in e ssence drowning out t he fine-tuned signaling required for mnemonic function, a sit uation that may be attenuated by mild inhibition of the NMDAR. Another possible mechanism that may be altered due to agonist/antagonist relationships of amyloi d with various receptors, is calcium signaling pathways. Evidence for amyl oid binding to both nicotinic and muscarinic cholinergic receptors has been reported (Dineley et al., 2002;Kelly et al., 1996), both of which are involved in Ca2+ signaling, either by increasing its conductance into the cell or by releasing intracellular stores, respectively. We have demonstrated that the expre ssion of the catalytic subunit of Ca2+/calmodulin-dependent kinase (CaMKII ) is down-regulated in APP+PS1 transgenic mice, and has significantly reduced expression in the presence of lower levels of amyloid, unlike the other plasticity-related genes, AMPA1 and Na+/K+ ATPase (Paper II; Figures 1 and 3), although not to the same extent as the IEGs. This suggests that calcium si gnaling may be disrupted in the presence
137 of low amyloid burden and this may be precipitating the decrease in IEG expression. As calcium signals are shorter-lived than other cascade messengers, they present good candidates for initiators of the transcripti on of the extremely transient IEGs (Greenberg et al., 1992;v an Haasteren et al., 1999). Conversely, intracellular calcium release was found to be facilitated by an AD-linked mutated form of PS1 (Stutzmann et al., 2004). As these findings suggested, AD-linked mutations in PS1 may not only result in t he over-production of A ,but also of the APP intracellular domain (AICD), which is the fragment thought to stimulate the activation of IP3 receptor s on the endoplasmic reticulu m (Leissring et al., 2002). This increased intracellular calcium re lease was demonstrated to activate Ca2+sensitive K+ channels on the cell surface, result ing in a hyperpolarization. This may explain another mechanism by which calcium is impeding normal neuronal function and decreasing signal transmission. In addition to amyloid pathology, APP+ PS1 mice develop chronic focal inflammation surrounding fibrillar plaques. Th is includes activation of microglia and astrocytes (Gordon et al., 2002), whic h can elicit cytokine release, acute phase reactants and free radical producti on. APP production is even thought to be promoted by interleukin 1 demonstrating the role that cytokines may have on disease progression (Blume and Vitek, 1989 ). Transcriptional regulation of IEG expression by cytokines has been well es tablished (Emch et al., 2001;Konsman et al., 2000) and, while this is a possible mechanism by which IEG expression is being inhibited, it is unlikely because IE G expression is significantly decreased by 6 months in the APP+PS1 mice, a point at which amyloid is accumulating with
138 little inflammatory response (Gordon et al., 2002;Apelt and Schliebs, 2001). In addition, most reported cytokine effect s on IEG expression have shown that increased cytokine production elicits an increase in IEG expression, not a decrease (Jenab and Quinones-Jenab, 2002;Sim i et al., 2002;Heinrich et al., 2003). Besides the elegant possibilities di scussed above, there is perhaps a more rudimentary explanation for impaired IEG induction; that the neurons simply cannot conduct adequate electrical impulse s any longer. One of the plasticityrelated genes with consistently down -regulated mRNA expression in amyloidcontaining APP+PS1 mice tissue was Na+/K+ ATPase III (Paper I; Figure 1: Paper II; Figures 1, 2 and 3: Paper III; Fi gure 1). Na+/K+ ATPase is more than a plasticity-related gene in that it consumes nearly 40% of the ATP in brain tissue and maintains the membrane potential th roughout the parenchyma, emphasizing its critical role in every function of the brain. Without Na +/K+ ATPase, neurons would not be able to maintain membrane pot entials resulting in the loss of their ability to conduct electrical impulses and subsequently store information (Watts and Mark, 1971). To ascertain whether th is enzyme was malfunctioning in the APP+PS1 mice, we analyzed the activity of oubain-sensitive ATPase in cerebral tissue of memory-deficient APP+PS1 mice compared with non-transgenic littermates. We found that there was si gnificantly impaired Na+/K+ ATPase activity in the hippocampi of APP+PS1 mi ce, but not in the cerebella (Paper III; Figure 2). These findings s uggested that the reduced acti vity is precipitated by the presence of amyloid.
139 Subsequent histological analyses of APP+PS1 cerebral sections showed that although Na+/K+ ATPase stained the sections ubiquitously, the area immediately surrounding Congo red stai ned plaques lacked Na+/K+ ATPase almost entirely (Paper III; Figures 4 and 5). The absence of Na+/K+ ATPase from this circumferential zone around plaques would disrupt the membrane potential of any cellular components in that area, wh ich is the exact area where dystrophic neurites are normally seen. Triple stai ning of Na+/K+ ATPase, dystrophic neurites and congophilic plaques revealed that the dystrophic neurites were in fact within the Na+/K+ ATPase-free z one surrounding the fibrillar plaques (Paper III; Figure 5). Soluble A was subsequently prepared in accor dance with Stein et al. (2002) and incubated with purified ce rebral Na+/K+ ATPase to assess whether or not amyloid could directly i nhibit enzyme activity. Interestingly, high concentrations of this A preparation did dramatically inhibit activity (Paper III; Figure 6), suggesting that in a local area of high amyloid concentration, such as the area immediately surr ounding an amyloid plaque, A may bind to Na+/K+ ATPase leading to its inactivation and s ubsequent removal from the membrane. It has also been reported that specific i nhibition of Na+/K+ ATPase leads to reduced mRNA expression of the III subunit (Huang et al., 1997), providing an explanation as to why the mRNA expression is also down-regulated. One of the hallmark pathological f eatures of the APP+PS1 mouse is the presence of dystrophic neurites without ac tual synaptic or neuronal loss (Gordon et al., 2002;Jantzen et al. In Prepar ation). These dystrophic neurites are synapses that are presumably no longer functioning properly and appear swollen
140 and disjointed. The exact mechanism by which these neurites form is largely unknown, although it has been postulated t hat amyloid interacts with various surface proteins which in turn stim ulate signaling ca scades and consequently lead to the accumulation of hype r-phosphorylated neurofilament and the formation of dystrophic neurites (Grace and Busciglio, 2003). Here we have presented evidence that in models of am yloid deposition, the AD-like dystrophia may occur due to amyloid dire ctly inhibiting Na+/K+ ATPa se activity, resulting in ionic imbalance and the resultant osmo tic swelling (Cooke, 1978). Presumably, synapses that have become dystrophic are no longer able to conduct action potentials to the post-synapse. Possibl y, neurons in an amyloid-induced dystrophic state increase expression of several anti-apoptotic genes (Chaudhury et al., 2003;Tortosa et al., 1998) leading to the down-regulated expression of genes involved in neurotransmitter release. This would suggest that neuronal function turns from the collective parenc hymal purpose of information transfer to the more immediate necessity of surv ival. Another possibility may be that dendritic swelling brought on by osmotic im balance via Na+/K+ ATPase inhibition could actually hamper post-synaptic conduc tion simply by increasing the area (intracellular volume) that the current has to traverse. This may impede the ability of gray matter to propagate electrical signals along the entire length of the dendritic spine, preventing sufficient depolarization at the axon hillock for subsequent synaptic transmissions. Based on the results described here, it is feasible that the blockade of Na+/K+ ATPase activity by amyloid resu lts in the dysregulation of the neuronal
141 membrane potential which in turn l eads to the appearance of swollen, dysfunctional neurites. The inability of these neurons then to conduct action potentials and elicit specif ic stimulatory signals lik e LTP may result in the formation of improper synaptic connec tions, damaged plasticity, and a general failure to consolidate information receiv ed in the form of discrete patterns of electrical frequencies into lasting memories. Late phase LTP, which has been linked to long-term memory consolidation (Nadel and Moscovitch, 1998;Abraham et al., 2002), is reported to be a critical component of molecular plasticity events, namely that late-phase LTP blockade resu lts in inhibition of IEG expression (Richardson et al., 1992;Abraham et al., 1993;Barnes et al., 1994). The amyloidinduced neuritic dystrophia described abov e may be disrupting the connectivity of circuits critical for memo ry formation, leadi ng to the inability of neurons to conduct late-phase LTP. This may prev ent the expression of IEGs and the subsequent effects on necessary persistent plasticity changes elicited by their induction. Circuits essential for learned beh avior have been well established and generally initiate in the hippocampus, with projections to the entorhinal cortex and parietal cortex (Izqui erdo and Medina, 1997). Afte r repeated training, only the parietal cortex is required for recall of this specific task, while the hippocampus and entorhinal cortex are no longer involved. It has also been established that LTP-induced protein expr ession is required for these circuitous network changes to occur and be maintained (Guzowski et al., 2000;Frankland et al., 2001). This suggests that di sruption of the ability of these particular circuits to
142 relay electrical signals would result in the inability to form lasting connections essential for memory formation, as experie nced by Alzheimers diseased patients and amyloid mouse models. Additionally, it may not be the sheer volume of forebrain amyloid that is driving thes e memory deficits, but rather where specifically the amyloid is depositing, possibly explaini ng why behavioral deficits do not always correlate with IEG expressi on or amyloid load (Holcomb et al., 1999). In other words, as amyloid load increases, the ch ances of a plaque depositing in the middle of a long-term memo ry circuit are significantly increased. Therefore, the critical issue in amyloid-associated dementia may not be how much amyloid is present, but rather w here the amyloid is deposited and what neuronal networks are being disrupted. Here we have presented evidence that amyloid plaques locally inhibiting Na+/K+ ATPase activity in APP+PS1 mice, may result in disruption of electrical signaling in brain regions essential for the establishment of lasting forms of memory by promoting neuritic dystrophia. This disruption would likely result in reduced IEG expression due to the resultant insufficiency of electrical signaling. The reduced IEG expression at 6 months in APP+PS1 mice (Paper II; Figure 1) is possibly a result of inactivated Na+/ K+ ATPase, however sufficient disruption of the circuits essential for memory consolidation are yet to be realized as amyloid burden is still relatively low. Only when plaque load increases are networks adequately impaired to cause det ectable memory deficits. These findings present several candidates for pha rmacological intervention in amyloidassociated dementia, including several IEGs as well as other plasticity-related
143 genes (e.g. CaMKII and AMPA1). The exact mechanisms by which these genes are dysregulated and their link to memory formation have yet to be understood. However, we have demonstrated a cl ose relationship between amyloid and impaired expression of several memory-a ssociated genes, as well as a direct inhibition of Na+/K+ ATPase activity by amyloid, suggesti ng that therapeutic intervention targeting the well char acterized enzyme may be warranted. Maintaining the ionic balance of crit ical mnemonic circuits through the administration of Na+/K+ ATPase agoni sts like phenytoin, may prove highly therapeutic to AD sufferers, possibly delaying the emergence of cognitive symptoms for several years and reducing medical and long-term care costs by billions of dollars.
144 TABLE 1. Genes that are selectively and significantly down-regulated in the amyloid-containing hippocampi of APP+PS1 mice (qRT-PCR data). Category Gene Name NCBI Genbank Accession # qRT-PCR Data Percentage of nontransgenic mean SEM Arc (Arg3.1) NM_018790 48+ 10 Zif268 (Egr-1) M20157 45+ 8 Nur77 (Nurr1) J04113 60+ 8 Homer1a (Vesl-1) AF093257 60+ 7 Narp (Nptx2) AF049124 62+ 8 Immediate-Early Genes -A Activin D83213 69+ 7 CaMKII NM_009792 65+ 6 AMPA1(GluR 1) NM_008165 70+ 10 NR2B NM_008171 82+ 5 Na+/K+ ATPase BC020177 57+ 4 Other PlasticityRelated Genes PSD-95 D50621 82+ 5
145 REFERENCES Abdulkadir SA, Carbone JM, Naughton CK, Humphrey PA, Catalona WJ, Milbrandt J (2001) Frequent and early loss of the EGR1 corepressor NAB2 in human prostate carcinoma. Hum Pathol 32: 935-939. Abraham WC, Logan B, Greenwood JM Dragunow M (2002) Induction and experience-dependent conso lidation of stable long-t erm potentiation lasting months in the hippocampus. J Neurosci 22: 9626-9634. Abraham WC, Mason SE, Demmer J, Williams JM, Richardson CL, Tate WP, Lawlor PA, Dragunow M (1993) Correla tions between immediate early gene induction and the persistence of long-te rm potentiation. Neuroscience 56: 717727. Andreasson K, Worley PF (1995) Inducti on of beta-A activin expression by synaptic activity and during neocortical development. Neuroscience 69: 781-796. Ango F, Pin JP, Tu JC, Xiao B, Worley PF, Bockaert J, Fagni L (2000) Dendritic and axonal targeting of type 5 metabotropic glutamate receptor is regulated by homer1 protei ns and neuronal excitation J Neurosci 20: 8710-8716. Apelt J, Schliebs R (2001) Beta-amylo id-induced glial expression of both proand anti-inflammatory cytokines in cerebral cortex of aged transgenic Tg2576 mice with Alzheimer plaque pathology. Brain Res 894: 21-30. Arendt T (2001) Disturbance of neuronal pl asticity is a critical pathogenetic event in Alzheimer's disease. Int J Dev Neurosci 19: 231-245. Artavanis-Tsakonas S, Matsuno K, Fo rtini ME (1995) Notch signaling Science 268: 225-232. Bach ME, Hawkins RD, Osman M, Kandel ER, Mayford M (1995) Impairment of spatial but not contextual memory in CaMKII mutant mice with a selective loss of hippocampal LTP in the range of the theta frequency. Cell 81: 905-915. Barnes CA (1979) Memory deficit s associated with senescence: a neurophysiological and behavioral study in the rat. J Comp Physiol Psychol 93: 74-104.
146 Barnes CA, Jung MW, McNaughton BL, Korol DL, Andreasson K, Worley PF (1994) LTP saturation and spatial learning disruption: effects of task variables and saturation levels. J Neurosci 14 : 5793-5806. Benowitz LI, Routtenberg A (1997) G AP-43: an intrinsic determinant of neuronal development and plasticity. Trends Neurosci 20: 84-91. Blume AJ, Vitek MP (1989) Focusing on IL-1-promotion of beta-amyloid precursor protein synthesis as an early event in Alzheimer's disease. Neurobiol Aging 10: 406-408. Bockamp E, Maringer M, Spangenberg C, Fees S, Fraser S, Eshkind L, Oesch F, Zabel B (2002) Of mice and models: improved animal models for biomedical research. Physiol Genomics 11: 115-132. Bolshakov VY, Golan H, Kandel ER Siegelbaum SA (1997) Recruitment of new sites of synaptic transmission during the cAMPdependent late phase of LTP at CA3-CA1 synapses in t he hippocampus. Neuron 19: 635-651. Bottai D, Guzowski JF, Schwarz MK, Kang SH, Xiao B, Lanahan A, Worley PF, Seeburg PH (2002) Synaptic activi ty-induced conversion of intronic to exonic sequence in Homer 1 immediate early gene expression. J Neurosci 22: 167-175. Bozon B, Davis S, Laroche S (2003) A requirement for the immediate early gene zif268 in reconsoli dation of recognition memory after retrieval. Neuron 40: 695-701. Bradley A, Evans M, Kaufman MH Robertson E (1984) Formation of germ-line chimaeras from embryo-derived te ratocarcinoma cell lines. Nature 309: 255-256. Brakeman PR, Lanahan AA, O'Brien R, Ro che K, Barnes CA, Huganir RL, Worley PF (1997) Homer: a protein that selectively binds metabotropic glutamate receptors. Nature 386: 284-288. Brunger AT (2000) Structur al insights into the molecular mechanism of Ca(2+)-dependent exocytosis. Curr Opin Neurobiol 10: 293-302. Callahan LM, Vaules WA, Coleman PD (2002) Progressive reduction of synaptophysin message in single neurons in Alzheimer disease. J Neuropathol Exp Neurol 61: 384-395. Cammarota M, Bevilaqua LR, Ardenghi P, Paratcha G, Levi dS, Izquierdo I, Medina JH (2000) Learning-associated activation of nuclear MAPK, CREB and Elk-1, along with Fos production, in the rat hippocampus after a one-trial
147 avoidance learning: abolitio n by NMDA receptor blockade. Brain Res Mol Brain Res 76: 36-46. Changelian PS, Feng P, King TC, Milb randt J (1989) Structure of the NGFI-A gene and detection of upstream sequences responsible for its transcriptional induction by nerve growth factor. Proc Natl Acad Sci U S A 86: 377-381. Chaudhury AR, Gerecke KM, Wyss JM Morgan DG, Gordon MN, Carroll SL (2003) Neuregulin-1 and er bB4 immunoreactivity is associated with neuritic plaques in Alzheimer disease brain and in a transgenic model of Alzheimer disease. J Neuropathol Exp Neurol 62: 42-54. Citron M, Westaway D, Xia W, Carlson G, Diehl T, Levesque G, JohnsonWood K, Lee M, Seubert P, Davis A, K holodenko D, Motter R, Sherrington R, Perry B, Yao H, Strome R, Lieberburg I, Rommens J, Kim S, Schenk D, Fraser P, St George HP, Selkoe DJ (1997) Mutant presenilins of Alzheimer's disease increase production of 42-residue amyloid be ta-protein in both transfected cells and transgenic mice. Nat Med 3: 67-72. Cole AJ, Abu-Shakra S, Saffen DW, Baraban JM, Worley PF (1990) Rapid rise in transcription factor mRNAs in rat brain after electroshock-induced seizures. J Neurochem 55: 1920-1927. Cole AJ, Saffen DW, Baraban JM, Worley PF (1989) Rapid increase of an immediate early gene messenger RNA in hippocampal neurons by synaptic NMDA receptor activation. Nature 340: 474-476. Cooke KR (1978) Ouabain and regulation of cellular volume in freshly prepared slices of rabbit renal cortex. J Physiol 279: 361-374. Crambert G, Beguin P, Uldry M, Monnet-Tschudi F, Horisberger JD, Garty H, Geering K (2003) F XYD7, the first brainand isoform-specific regulator of Na,K-ATPase: biosynthesis and function of its posttranslatio nal modifications. Ann N Y Acad Sci 986: 444-448. Cull-Candy S, Brickley S, Farrant M (2001) NMDA receptor subunits: diversity, development and disease. Curr Opin Neurobiol 11: 327-335. da Cruz e Silva EF, da Cruz e Silv a OA (2003) Protein phosphorylation and APP metabolism. Neurochem Res 28: 1553-1561. Danysz W, Parsons CG (2003) The NMDA receptor antagonist memantine as a symptomatological and neuroprotective treatment for Alzheimer's disease: preclinical evidence. Int J Geriatr Psychiatry 18: S23-S32.
148 Davis S, Vanhoutte P, Pages C, Caboche J, Laroche S (2000) The MAPK/ERK cascade targets both Elk-1 and cAMP response elementbinding protein to control long-term potentia tion-dependent gene expression in the dentate gyrus in vivo. J Neurosci 20: 4563-4572. Dewachter I, Van Leuven F (2002) Secretas es as targets for the treatment of Alzheimer's disease: the pr ospects. Lancet Neurol 1: 409-416. Dickey CA, Loring JF, Eastman PS, Montgomery JR, Gordon M, Morgan DG (2003) Selectively reduced expression of synaptic plasticity-related genes in APP+PS1 transgenic mice. J Neurosci 23: 5219-5226. Dickey CA, Morgan DG, Kudchodkar S, Weiner DB, Bai Y, Cao C, Gordon MN, Ugen KE (2001) Duration and specificity of humoral immune responses in mice vaccinated with the Alzheimer's disease-associated betaamyloid 1-42 peptide. DNA Cell Biol 20: 723-729. Dineley KT, Bell KA, Bui D, Sweatt JD (2002) beta -Amyloid peptide activates alpha 7 nicotinic acetylchol ine receptors expressed in Xenopus oocytes. J Biol C hem 277: 25056-25061. Dineley KT, Westerman M, Bui D, Be ll K, Ashe KH, Sweatt JD (2001) Beta-amyloid activates the mitogenactivated protein kinase cascade via hippocampal alpha7 nicotinic acetylcholi ne receptors: In vitro and in vivo mechanisms related to Alzheimer's di sease. J Neurosci 21: 4125-4133. Drolet G, Proulx K, Pearson D, Ro chford J, Deschepper CF (2002) Comparisons of behavioral and neurochemic al characteristics between WKY, WKHA, and Wistar rat strains. Neuropsychopharmacology 27: 400-409. Dudek SM, Bear MF (1992) Homosy naptic long-term depression in area CA1 of hippocampus and effects of N-me thyl-D-aspartate receptor blockade. Proc Natl Acad Sci U S A 89: 4363-4367. Duff K, Eckman C, Zehr C, Yu X, Pr ada CM, Perez-tur J, Hutton M, Buee L, Harigaya Y, Yager D, Morgan D, Gor don MN, Holcomb L, Refolo L, Zenk B, Hardy J, Younkin S (1996) Increased am yloid-beta42(43) in brains of mice expressing mutant presenili n 1. Nature 383: 710-713. Emch GS, Hermann GE, Rogers RC ( 2001) TNF-alpha-induced c-Fos generation in the nucleus of the solitary tract is blocked by NBQX and MK-801. Am J Physiol Regul Integr Co mp Physiol 281: R1394-R1400. Eriksen JL, Sagi SA, Smith TE, Weggen S, Das P, McLendon DC, Ozols VV, Jessing KW, Zavitz KH, Koo EH, Go lde TE (2003) NSAIDs and enantiomers of flurbiprofen target gamma-secretase and lower Abeta 42 in vivo. J Clin Invest 112: 440-449.
149 Fahrner TJ, Carroll SL, Milbrandt J (1990) The NGFI-B protein, an inducible member of the th yroid/steroid receptor fam ily, is rapidly modified posttranslationally. Mol Cell Biol 10: 6454-6459. Frankland PW, O'Brien C, Ohno M, Ki rkwood A, Silva AJ (2001) AlphaCaMKII-dependent plasticity in the cortex is required for permanent memory. Nature 411: 309-313. Gordon MN, Holcomb LA, Jantzen PT, DiCarlo G, Wilcock D, Boyett KW, Connor K, Melachrino J, O'Callaghan JP, Morgan D (2002) Time course of the development of Alzheimer-like pathol ogy in the doubly transgenic PS1+APP mouse. Exp Neurol 173: 183-195. Gordon MN, King DL, Diamond DM, Jantzen PT, Boyett KV, Hope CE, Hatcher JM DiCarlo G, Gottschall WP, Morgan D, Arendash GW (2001) Correlation between cognitive defic its and Abeta deposits in transgenic APP+PS1 mice. Neurobiol Aging 22: 377-385. Grace EA, Busciglio J (2003) Abe rrant activation of focal adhesion proteins mediates fibrillar amyloid bet a-induced neuronal dystrophy. J Neurosci 23: 493-502. Greenberg ME, Thompson MA, Sheng M (1992) Calcium regulation of immediate early gene transcription. J Physiol Paris 86: 99-108. Guillaume D, Grisar T, Delgado-Escuet a AV, Minet A, Vergniolle-Burette M, Bureau-Heeren M (1989) Phenytoin dephosphorylates the alpha(-) catalytic subunit of (Na+, K+)-ATPase. A study in mouse, cat and human brain. Biochem Pharmacol 38: 3933-3939. Guissouma H, Becker N, Seugnet I, Demeneix BA (2000) Transcriptional repression of TRH promoter function by T3: analysis by in vivo gene transfer. Biochem Cell Biol 78: 155-163. Gutman Y, Boonyaviroj P ( 1977) Mechanism of inhibi tion of catecholamine release from adrenal medulla by diphenylh ydantoin and by low concentration of ouabain (10(-10) M). Nauny n Schmiedebergs Arch Pharmacol 296: 293-296. Guzowski JF, Lyford GL, Stevens on GD, Houston FP, McGaugh JL, Worley PF, Barnes CA (2000) Inhibi tion of activity-dependent arc protein expression in the rat hippocampus impai rs the maintenance of long-term potentiation and the consolidation of long-term memory. J Neurosci 20: 39934001. Guzowski JF, Setlow B, Wagner EK McGaugh JL (2001) Experiencedependent gene expression in the rat hippoc ampus after spatial learning: a
150 comparison of the immediate-early genes Ar c, c-fos, and zif268. J Neurosci 21: 5089-5098. Hall J, Thomas KL, Everitt BJ (20 00) Rapid and selective induction of BDNF expression in the hippocampus during contextual learning. Nat Neurosci 3: 533-535. Hardy J (1997) Amyloid, the presenili ns and Alzheimer's disease. Trends Neurosci 20: 154-159. Hardy J, Selkoe DJ (2002) The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to therapeutics. Science 297: 353356. Heinrich PC, Behrmann I, Haan S, Hermanns HM, Muller-Newen G, Schaper F (2003) Principles of interleukin (IL)-6-type cytokine signalling and its regulation. Bioc hem J 374: 1-20. Holcomb L, Gordon MN, McGowan E, Yu X, Benkovic S, Jantzen P, Wright K, Saad I, Mueller R, Morgan D, S anders S, Zehr C, O'Campo K, Hardy J, Prada CM, Eckman C, Younkin S, Hsiao K, Duff K (1998) Accelerated Alzheimertype phenotype in transgenic mice carry ing both mutant amyloid precursor protein and presenilin 1 tr ansgenes. Nat Med 4: 97-100. Holcomb LA, Gordon MN, Jantzen P, Hsiao K, Duff K, Morgan D (1999) Behavioral changes in transgenic mice expressing both amyloid precursor protein and presenilin-1 mutations: lack of association with amyloid deposits. Behav Genet 29: 177-185. Hsiao K, Chapman P, Nilsen S, Eckman C, Harigaya Y, Younkin S, Yang F, Cole G (1996) Correlative memory def icits, Abeta elevation, and amyloid plaques in transgenic mice. Science 274: 99-102. Huang L, Kometiani P, Xie Z (1997) Differential regulation of Na/KATPase alpha-subunit isoform gene expres sions in cardiac myocytes by ouabain and other hypertrophic stimuli. J Mol Cell Cardiol 29: 3157-3167. Huang YY, Nguyen PV, Abel T, Kandel ER (1996) Long-lasting forms of synaptic potentiation in the mammalia n hippocampus. Learn Mem 3: 74-85. Izquierdo I, Medina JH (1997) Memo ry formation: t he sequence of biochemical events in the hippocampus and its connection to activity in other brain structures. Neurobi ol Learn Mem 68: 285-316. Jantzen PT, Connor KE, DiCarlo G, Wenk GL, Wallace JL, Rojiani AM, Coppola D, Morgan D, Gordon MN (2002) Microglial activation and beta -amyloid deposit reduction caused by a nitric oxide-releasing nonsteroidal anti-
151 inflammatory drug in amyloid precursor protein plus presenilin-1 transgenic mice. J Neurosci 22: 2246-2254. Jenab S, Quinones-Jenab V (2002) The effe cts of interleukin-6, leukemia inhibitory factor and interferon-gamma on STAT DNA binding and c-fos mRNA levels in cortical astrocytes and C6 glioma cells. Neuroendocrinol Lett 23: 325328. Jones MW, Errington ML, French PJ, Fine A, Bliss TV, Garel S, Charnay P Bozon B, Laroche S, Davis S (2001) A requirement for the immediate early gene Zif268 in the expression of late LTP and long-term memories. Nat Neurosci 4: 289-296. Kandel E, Schwartz J, Jessell T (2000) Principles of Neural Science. New York: McGraw-Hill. Kelly JF, Furukawa K, Barger SW Rengen MR, Mark RJ, Blanc EM, Roth GS, Mattson MP (1996) Amyloid bet a-peptide disrupts carbachol-induced muscarinic cholinergic signal transduction in cortical neurons. Proc Natl Acad Sci U S A 93: 6753-6758. Kelly MP, Deadwyler SA (2002) Acquisition of a novel behavior induces higher levels of Arc mRNA than does overtrained performance. Neuroscience 110: 617-626. Khachigian LM, Lindner V, Williams AJ, Collins T (1996) Egr-1-induced endothelial gene expression: a common theme in vascular injury. Science 271: 1427-1431. Kim JH, Anwyl R, Suh YH, Djamgoz MB, Rowan MJ (2001) Usedependent effects of amyloi dogenic fragments of (bet a)-amyloid precursor protein on synaptic plasticity in rat hippocampus in vivo. J Neurosci 21: 13271333. Kodama M, Akiyama K, Ujike H, Shim izu Y, Tanaka Y, Kuroda S (1998) A robust increase in expression of arc gene, an effector immediate early gene, in the rat brain after acute and chronic me thamphetamine administration. Brain Res 796: 273-283. Kogan JH, Frankland PW, Blendy JA, Cobl entz J, Marowitz Z, Schutz G, Silva AJ (1997) Spaced training induc es normal long-term memory in CREB mutant mice. Curr Biol 7: 1-11. Konsman JP, Tridon V, Dantze r R (2000) Diffusion and action of intracerebroventricularly injected interl eukin-1 in the CNS. Neuroscience 101: 957-967.
152 Kovacs KJ, Sawchenko PE (1996) R egulation of stress-induced transcriptional changes in the hypothal amic neurosecretory neurons. J Mol Neurosci 7: 125-133. Kramer B, Meichle A, Hensel G, Charnay P, Kronke M (1994) Characterization of an Krox-24/Egr-1-re sponsive element in the human tumor necrosis factor promoter. Biochim Biophys Acta 1219: 413-421. Kremerskothen J, Wendholt D, Teber I, Barnekow A (2002) Insulininduced expression of the activity-r egulated cytoskeletonassociated gene (ARC) in human neuroblastoma cells requires p21(ras), mitogen-activated protein kinase/extracellular regulated kinase and src tyrosine kinases but is protein kinase C-independent. Neurosci Lett 321: 153-156. Kuo CC, Bean BP (1994) Slow binding of phenytoin to inactivated sodium channels in rat hippocampal neurons. Mol Pharmacol 46: 716-725. Lanahan A, Worley P (1998) Imm ediate-Early Genes and Synaptic Function. Neurobiol Learn Mem 70: 37-43. Laws SM, Hone E, Gandy S, Martins RN (2003) Expanding the association between the APOE gene and t he risk of Alzheimer's disease: possible roles for APOE promoter pol ymorphisms and alterations in APOE transcription. J Neurochem 84: 1215-1236. Lee HK, Barbarosie M, Kameyama K, Bear MF, Huganir RL (2000) Regulation of distinct AM PA receptor phosphorylation si tes during bidirectional synaptic plasticity. Nature 405: 955-959. Lehtovirta M, Laakso MP, Soininen H, Helisalmi S, Mannermaa A, Helkala EL, Partanen K, Ryynanen M, Vainio P, Hartikainen P, (1995) Volumes of hippocampus, amygdala and frontal lobe in Alzheimer patients with different apolipoprotein E genotypes. Neuroscience 67: 65-72. Leissring MA, Murphy MP, Mead TR, Akbari Y, Sugarman MC, Jannatipour M, Anliker B, Muller U, Safti g P, De Strooper B, Wolfe MS, Golde TE, LaFerla FM (2002) A physiologic sign aling role for the gamma -secretasederived intracellular fragment of APP. Proc Natl Ac ad Sci U S A 99: 4697-4702. Lyford GL, Yamagata K, Kaufm ann WE, Barnes CA, Sanders LK, Copeland NG, Gilbert DJ, Jenkins NA Lanahan AA, Worley PF (1995) Arc, a growth factor and activity-regulated gene, encodes a novel cytoskeletonassociated protein that is enriched in neuronal dendrit es. Neuron 14: 433-445. Lynch MA (2004) Long-term potentiation and memory. Physiol Rev 84: 87136.
153 Ma L, Zablow L, Kandel ER, Sieg elbaum SA (1999) Cyclic AMP induces functional presynaptic boutons in hippoc ampal CA3CA1 neuronal cultures. Nat Neurosci 2: 24-30. Malenka RC, Nicoll RA (1999) Long-term potentiation--a decade of progress? Science 285: 1870-1874. Maltzman JS, Carmen JA, Monroe JG ( 1996) Transcriptional regulation of the Icam-1 gene in antigen recept orand phorbol ester-stimulated B lymphocytes: role for transcription factor EGR1. J Exp Med 183: 1747-1759. Mansuy IM, Winder DG, Moallem TM Osman M, Mayford M, Hawkins RD, Kandel ER (1998) Inducible and reve rsible gene expression with the rtTA system for the study of memory. Neuron 21: 257-265. Mark RF, Watts ME (1971) Drug i nhibition of memo ry formation in chickens. I. Long-term memory. Proc R Soc Lond B Biol Sci 178: 439-454. Masliah E, Mallory M, Alford M, DeTeresa R, Hansen LA, McKeel DW, Jr., Morris JC (2001) Altered expression of synaptic proteins occurs early during progression of Alzheimer's di sease. Neurology 56: 127-129. Mattson MP (1997) Cellular actions of beta-amyloid precursor protein and its soluble and fibrillog enic derivatives Physiol Rev 77: 1081-1132. Mattson MP (2002) Oxidative stress, perturbed calcium homeostasis, and immune dysfunction in Alzheimer's di sease. J Neurovirol 8: 539-550. Mayford M, Wang J, Kandel ER, O'Dell TJ (1995) CaMKII regulates the frequency-response function of hippocampal synapses for the production of both LTD and LTP. Cell 81: 891-904. McCaffrey TA, Fu C, Du B, Eksinar S, Kent KC, Bush H, Jr., Kreiger K, Rosengart T, Cybulsky MI, Silverman ES, Collins T (2000) High-level expression of Egr-1 and Egr-1-inducible genes in mouse and human atherosclerosis. J Clin Invest 105: 653-662. Milner LA, Bigas A (1999) Notch as a m ediator of cell fate determination in hematopoiesis: evidence and spec ulation. Blood 93: 2431-2448. Miyashita Y, Kameyama M, Ha segawa I, I, Fukushima T (1998) Consolidation of Visual Associative LongTerm Memory in the Temporal Cortex of Primates. Neurobiol Learn Mem 70: 197-211. Montag-Sallaz M, Welzl H, Kuhl D, Montag D, Schachner M (1999) Novelty-induced increased expression of immediate-early genes c-fos and arg 3.1 in the mouse brain. J Neurobiol 38: 234-246.
154 Morgan D, Diamond DM, Gottschall PE, Ugen KE, Dickey C, Hardy J, Duff K, Jantzen P, DiCarlo G, Wilcock D, Connor K, Hatcher J, Hope C, Gordon M, Arendash GW (2000) A beta peptide vaccination prevents memory loss in an animal model of Alzheimer's disease. Nature 408: 982-985. Morgan JI, Cohen DR, Hempstead JL, Curran T (1987) Mapping patterns of c-fos expression in t he central nervous system af ter seizure. Science 237: 192-197. Morris R (1984) Developments of a water-maze procedure for studying spatial learning in the rat. J Neurosci Methods 11: 47-60. Mortensen RM (1993) Double knockouts. Production of mutant cell lines in cardiovascular research. Hypertension 22: 646-651. Mukaetova-Ladinska EB, Garcia-Siera F, Hurt J, Gertz HJ, Xuereb JH, Hills R, Brayne C, Huppert FA, Paykel ES, McGee M, Jakes R, Honer WG, Harrington CR, Wischik CM (2000) Stagi ng of cytoskeletal and beta-amyloid changes in human isocortex reveals biphasic synaptic protein response during progression of Alzheimer's dis ease. Am J Pathol 157: 623-636. Mumby DG, Wood ER, Duva CA, Kor necook TJ, Pinel JP, Phillips AG (1996) Ischemia-induced object-recognition deficits in rats are attenuated by hippocampal ablation before or soon afte r ischemia. Behav Neurosci 110: 266281. Mundel P, Heid HW, Mundel TM, Kruger M, Reiser J, Kriz W (1997) Synaptopodin: an actin-associated protein in telencephalic dendrites and renal podocytes. J Cell Biol 139: 193-204. Murphy LO, MacKeigan JP, Blenis J ( 2004) A network of immediate early gene products propagates subtle differences in mitogen-activated protein kinase signal amplitude and duration. Mol Cell Biol 24: 144-153. Myhrer T (1988) The role of medial and lateral hippocampal perforant path lesions and object distinctiveness in rats reaction to novelty. Physiol Behav 42: 371-377. Nadel L, Moscovitch M (1998) Hippocam pal contributions to cortical plasticity. Neuropharmacology 37: 431-439. Nayak A, Zastrow DJ, Lickteig R, Zahniser NR, Browning MD (1998) Maintenance of late-phase LTP is a ccompanied by PKA-dependent increase in AMPA receptor synthesis. Nature 394: 680-683.
155 O'Brien R, Xu D, Mi R, Tang X, H opf C, Worley P (2002) Synaptically targeted narp plays an essential role in the aggregation of AMPA receptors at excitatory synapses in cultured spi nal neurons. J Neurosci 22: 4487-4498. Ohno M, Sametsky EA, Younkin LH, Oa kley H, Younkin SG, Citron M, Vassar R, Disterhoft JF (2004) BACE1 Deficiency Rescues Memory Deficits and Cholinergic Dysfunction in a Mouse Model of Alzheimer's Disease. Neuron 41: 27-33. Otmakhov N, Griffith LC, Lisman JE (1997) Postsynaptic inhibitors of calcium/calmodulin-dependent protein kinas e type II block induction but not maintenance of pairing-induced longte rm potentiation. J Neurosci 17: 53575365. Palmiter RD, Chen HY, Brinster RL (1982) Differentia l regulation of metallothionein-thymidine kinase fu sion genes in transgenic mice and their offspring. Cell 29: 701-710. Perkel DJ, Hestrin S, Sah P, Nicoll RA (1990) Excitatory synaptic currents in Purkinje cells. Proc R Soc Lond B Biol Sci 241: 116-121. Pinaud R, Penner MR, Robertson HA, Currie RW (2001) Upregulation of the immediate early gene arc in the brai ns of rats exposed to environmental enrichment: implications for molecular pl asticity. Brain Res Mol Brain Res 91: 5056. Richardson CL, Tate WP, Mason SE, Lawlor PA, Dragunow M, Abraham WC (1992) Correlation between the i nduction of an immediate early gene, zif/268, and long-term potentiation in the dentate gyrus. Brain Res 580: 147-154. Roses AD (2000) Causes or consequences of inflammation and pathological signs of Alzheimer di sease. Neurobiol Aging 21: 423-425. Saam JR, Gordon JI (1999) Inducib le gene knockouts in the small intestinal and colonic epitheliu m. J Biol Chem 274: 38071-38082. Sakamoto KM, Bardeleben C, Yates KE, Raines MA, Golde DW, Gasson JC (1991) 5' upstream sequence and genomi c structure of the human primary response gene, EGR-1/TI S8. Oncogene 6: 867-871. Sato M, Suzuki K, Nakanishi S ( 2001) NMDA receptor stimulation and brain-derived neurotrophic fa ctor upregulate homer 1a mRNA via the mitogenactivated protein kinase cascade in cult ured cerebellar granule cells. J Neurosci 21: 3797-3805. Selkoe DJ (2001) Clearing the brain's amyloid cobwebs. Neuron 32: 177180.
156 Sgambato V, Pages C, Rogard M, Besson MJ, Caboche J (1998) Extracellular signal-regulated kinase (ERK) controls immediate early gene induction on corticostriatal stimul ation. J Neurosci 18: 8814-8825. Shapiro M (2001) Plasticity, hippocampal place cells, and cognitive maps. Arch Neurol 58: 874-881. Shockett P, Difilippantonio M, Hellman N, Schatz DG (1995) A modified tetracycline-regulated system provides aut oregulatory, inducible gene expression in cultured cells and transgenic mice. Proc Natl Acad Sci U S A 92: 6522-6526. Silva AJ, Wang Y, Paylor R, Wehner JM, Stevens CF, Tonegawa S (1992) Alpha calcium/calmodulin kinase II mutant mice: deficient l ong-term potentiation and impaired spatial learning. Cold Spri ng Harb Symp Quant Biol 57: 527-539. Simi A, Porsmyr-Palmertz M, Hjer ten A, Ingelman-Su ndberg M, Tindberg N (2002) The neuroprotective agents chlo methiazole and SB203580 inhibit IL1beta signalling but not its biosynthesis in rat cortical glial cells. J Neurochem 83: 727-737. Soderling TR, Derkach VA (2000) Po stsynaptic protein phosphorylation and LTP. Trends Neurosci 23: 75-80. St George-Hyslop. Piecing Together Alz heimer's. Scientific American 7683. 2000. Ref Type: Generic St Hilaire M, Tremblay PO, Levesque D, Barden N, Rouillard C (2003) Effects of cocaine on c-fos and NGFI-B mRNA expression in transgenic mice underexpressing glucocorticoid recept ors. Neuropsychopharmacology 28: 478489. Stein TD, Johnson JA (2002) Lack of neurodegeneration in transgenic mice overexpressing mutant amyloid precursor protein is associated with increased levels of transthyretin and the activation of cell survival pathways. J Neurosci 22: 7380-7388. Steward O (1987) Regulation of synaptogenesis through the local synthesis of protein at the postsynaptic site. Prog Brain Res 71: 267-279. Steward O (1995) Targeti ng of mRNAs to subsynaptic microdomains in dendrites. Curr Opin Neurobiol 5: 55-61. Steward O (2002) mRNA at synapses synaptic plasticity, and memory consolidation. N euron 36: 338-340.
157 Steward O, Levy WB (1982) Preferent ial localization of polyribosomes under the base of dendritic spines in granule cells of the dentate gyrus. J Neurosci 2: 284-291. Steward O, Wallace CS, Lyford GL, Wo rley PF (1998) Synaptic activation causes the mRNA for the IEG Arc to localize selectively near activated postsynaptic sites on dendrites. Neuron 21: 741-751. Steward O, Worley PF (2001) Selectiv e targeting of newly synthesized Arc mRNA to active synapses requires NMDA receptor activation. Neuron 30: 227240. Stutzmann GE, Caccamo A, LaFerla FM Parker I (2004) Dysregulated IP3 signaling in cortical neurons of k nock-in mice expressing an Alzheimer'slinked mutation in presenilin1 results in exaggerated Ca2+ signals and altered membrane excitability. J Neurosci 24: 508-513. Tang YP, Shimizu E, Dube GR, Rampon C, Kerchner GA, Zhuo M, Liu G, Tsien JZ (1999) Genetic enhancement of learning and memory in mice. Nature 401: 63-69. Tortosa A, Lopez E, Ferrer I (1998) Bcl-2 and Bax protein expression in Alzheimer's disease. Acta N europathol (Berl) 95: 407-412. Tsien JZ, Huerta PT, Tonegawa S (1996) The essential role of hippocampal CA1 NMDA receptor-dependent synaptic plasticity in spatial memory. Cell 87: 1327-1338. Tsukada M, Aihara T, Mizuno M, Kato H, Ito K (1994) Temporal pattern sensitivity of long-term potentiation in hippocampal CA1 neurons. Biol Cybern 70: 495-503. Turner KM, Burgoyne RD, Morgan A ( 1999) Protein phosphorylation and the regulation of synaptic membrane tr affic. Trends Neurosci 22: 459-464. van Haasteren G, Li S, Muda M, Su sini S, Schlegel W (1999) Calcium signalling and gene expression. J Recept Signal Transduct Res 19: 481-492. Vogt L, Schrimpf SP, Me skenaite V, Frischknecht R, Kinter J, Leone DP, Ziegler U, Sonderegger P (2001) Calsynt enin-1, a proteolytically processed postsynaptic membrane protein with a cytoplasmic calcium-binding domain. Mol Cell Neurosci 17: 151-166. Wahrle SE, Holtzman DM (2003) Differential metabo lism of ApoE isoforms in plasma and CSF. Exp Neurol 183: 4-6.
158 Walker PD, Carlock LR (1993) Immedi ate early gene activation during the initial phases of the excitotoxic cascade. J Neurosci Res 36: 588-595. Wallace CS, Lyford GL, Worley PF Steward O (1998) Differential intracellular sorting of immediate ear ly gene mRNAs depends on signals in the mRNA sequence. J Neurosci 18: 26-35. Waltereit R, Dammermann B, Wulff P, Scafidi J, Staubli U, Kauselmann G, Bundman M, Kuhl D (2001) Arg3.1/Arc mRNA induction by Ca2+ and cAMP requires protein kinase A and mitogen-acti vated protein kinase/extracellular regulated kinase activation. J Neurosci 21: 5484-5493. Watson MA, Milbrandt J (1989) The NGFI-B gene, a transcriptionally inducible member of the steroid rec eptor gene superfamily: genomic structure and expression in rat brain after seizure induction. Mol Cell Biol 9: 4213-4219. Watts ME, Mark RF (1971) S eparate actions of ouabain and cycloheximide on memory. Brain Res 25: 420-423. Weiler IJ, Greenough WT (1993) Metabotro pic glutamate receptors trigger postsynaptic protein synthesis. Proc Natl Acad Sci U S A 90: 7168-7171. Werme M, Olson L, Brene S ( 2000) NGFI-B and nor1 mRNAs are upregulated in brain reward pathways by drugs of abuse: different effects in Fischer and Lewis rats. Brain Res Mol Brain Res 76: 18-24. Wilson TE, Paulsen RE, Padgett KA, M ilbrandt J (1992) Participation of non-zinc finger residues in DNA bindi ng by two nuclear orphan receptors. Science 256: 107-110. Winder DG, Martin KC, Muzzio IA, Rohr er D, Chruscinski A, Kobilka B, Kandel ER (1999) ERK plays a regulatory role in induction of LTP by theta frequency stimulation and its modulation by beta-adrenergi c receptors. Neuron 24: 715-726. Winder DG, Schramm NL (2001) Plas ticity and behavior: new genetic techniques to address multiple forms and functions. Physiol Behav 73: 763-780. Wisniewski T, Ghiso J, Frangione B (1991) Peptides homologous to the amyloid protein of Alzheimer's disease c ontaining a glutamine for glutamic acid substitution have accelerated amyloid fibril formation. Biochem Biophys Res Commun 179: 1247-1254. Worley PF, Baraban JM (1987) Site of anticonvulsant action on sodium channels: autoradiographic and electrophysiol ogical studies in rat brain. Proc Natl Acad Sci U S A 84: 3051-3055.
159 Yamagata K, Sanders LK, Kaufmann WE, Yee W, Barnes CA, Nathans D, Worley PF (1994) rheb, a growth fa ctorand synaptic activity-regulated gene, encodes a novel Ras-related protei n. J Biol Chem 269: 16333-16339. Yang DS, Smith JD, Zhou Z, Ga ndy SE, Martins RN (1997) Characterization of the binding of amyl oid-beta peptide to cell culture-derived native apolipoprotein E2, E3, and E4 isof orms and to isoforms from human plasma. J Neurochem 68: 721-725. Yao PJ, Coleman PD (1998) Reduced O-glycosylated clathrin assembly protein AP180: implication for synapt ic vesicle recycling dysfunction in Alzheimer's disease. Neurosci Lett 252: 33-36. Yin Y, Edelman GM, Vanderklish PW (2002) The brain-derived neurotrophic factor enhances synthesis of Arc in synaptoneurosomes. Proc Natl Acad Sci U S A 99: 2368-2373. Ying SW, Futter M, Rosenblum K, Webber MJ, Hunt SP, Bliss TV, Bramham CR (2002) Brainderived neurotrophic factor induces long-term potentiation in intact adult hippocampus: requirement for ERK activation coupled to CREB and upregulation of Arc syn thesis. J Neurosci 22: 1532-1540. Zamanillo D, Sprengel R, Hvalby O, Jensen V, Burnashev N, Rozov A, Kaiser KM, Koster HJ, Borchardt T, Worl ey P, Lubke J, Frot scher M, Kelly PH, Sommer B, Andersen P, Seeburg PH, Sakmann B (1999) Importance of AMPA receptors for hippocampal synaptic plasticity but not for spatia l learning. Science 284: 1805-1811. Zola-Morgan S, Squire LR, Amaral DG (1986) Human amnesia and the medial temporal region: enduring memory impairment fo llowing a bilateral lesion limited to field CA1 of the hi ppocampus. J Neurosci 6: 2950-2967.
161 APPENDIX A: QUANTITATIVE REAL TIME PCR (QRT-PCR) PROTOCOL Primer Design. 1. Using the gene of interest (G OI), find the accession number that corresponds to the proper species with cDNA or mRNA at the end of the name, and copy and paste the sequence into a word processor. 2. Delete about 50% of the 3 end of this sequence, as eukaryotic RNA degrades from that end. 3. Using the internet based prim er design program called Primer3 ( http://www-genome.wi.mit.edu/cgi-bin/primer/primer3www.cgi ), paste the shortened sequence in to the proper field. 4. Set the specification of the amp licon size to be within 75-200bps, optimally @ 100bp. Also use the rodent mispriming library available with the program. 5. Specify the Tm (annealing te mperature) to range be between 60 C and 65 C. 6. Also set the GC content between 50 and 60%. The primers should have ~55% GC content and thei r Tm should be within 1-2 C of each other. 7. Select the number to return field and type in 20. This will return 20 possible pairs rather than the standard 5 pairs the program uses for default. 8. The program returns possible primer pairs in different regions of the sequence, with the best option listed first.
162 APPENDIX A: (CONTINUED) 9. Select a primer set in which both have similar pr operties, particularly the same GC content as this determines Tm and check their self-hybridization properties on the Operon website in their Oligo Toolkit. 10. Any pair with more than 4 nucleoti des having primer-dimer potential are not used. 11. The Operon website is also where t he Tm is derived that should be use in the actual PCR reaction. 12. Then search Genbank for sequence homol ogy of the F primer. If a match of 15 or more nucleotides are found in the same species, check the reverse complimentary sequence of the R primer in a similar manner. If matches of 15 or greater are found fo r both, do not use the primers and select a new set from the Primer 3 output. 13. After finding no evidence of other potential targets, primers are ordered from IDT. RNA Extraction. 1. Frozen tissue dissections are homogenized using a tissue-mizer instrument for 30 seconds in 700ul of RLT buffer plus -mercaptoethanol (see Qiagen RNeasy protocol). The rotor-s tator is rinsed with clean water and 70% ethanol between samples and dried with a Kim-Wipe.
163 APPENDIX A: (CONTINUED) 2. Qiagen RNeasy mini-spin columns ar e used to purify RNA according to Qiagens protocol. Additionally, samp les were treated with RNase-free DNase from Qiagen. 3. 20mg of tissue/column is the rati o used, yielding approximately 16ug total RNA/20mg of brain tissue. 4. RNA was eluted in water and total RNA concentration was determined by Ribogreen Assay (Molecular Probes) using a standard curve (See Molecular Probes Protocol). 5. The concentration is adjusted to 50ng/ ul and RNA quality is checked by a 1% agarose gel with ethidium bromide. Reverse Transcription. 1. The reverse transcription (rev txn) is performed using 8 mass quantities of total RNA diluted down from a 50ng/ul standard intra-assay sample pool. This should cover 3 logs (50, 20, 10, 5, 2, 1, 0.5, 0. 2ng/ul). For single samples, make a 1:10 dilution of each with water to yield 5ng/ul. 2. Plan ahead about how many genes y ou want to analyze and the number of samples you have. 24 samples allows for a single PCR plate to be used per gene with triplicates of each plus the standards. More than 24 samples requires that two plates be used for each gene and therefore duplicate standards on each of these plates must be
164 APPENDIX A: (CONTINUED) accounted for (quadruplicates in total) to control for plate-to-plate variation. 3. Additionally, pipetting is critical fo r real time PCR, so using 8-strip PCR tubes, mini-master mixes are made for each sample and standard to draw replicates (triplicates, quadruplicates, et c.) from. Therefore, a slight excess is required in each tube, making enough for 3.5 replicates or 4.5 replicates. This adds to the tota l number of cDNA needed from the beginning. 4. Therefore, each gene will require at least 7ul (f or 3.5 triplicates) of cDNA per sample and each reverse transcription reaction is 20ul, so nearly 3 genes can be analyzed per reverse transcription reaction. e.g. 9 GOIs 7ul = 63ul. Therefore 3.5 reverse transcription reactions (70ul) would be needed for each samp le and standard to allow for complete analysis of all 9 GOIs. 5. The following equation will yield the total number of reverse transcription reactions needed: # of samples (24) + # of standards (8) number of rev txn reactions (3.5) = total # of rev txn reactions (112 round to 118 to have enough excess).
165 APPENDIX A: (CONTINUED) 6. Make a master mix for the total # of rev txn reactions using the following components (volumes are based on a singl e rev txn reaction rather than the total needed for the 118 reactions mentioned above) a. 0.5ul of 500ng/ul Oligo d(T) 15-21 b. 0.5ul of 500ng/ul Random Hexamers c. 4ul of 5x 1st Strand Buffer d. 4ul of 25mM MgCl e. 2ul of DTT f. 1ul of dNTPs g. 6ul of 5M Betaine (provides thermostabilization to RT enzyme and enhancing denaturation) h. 0.87ul RNAse free water (vortex here prior to addition of enzyme) i. 0.13 200U/ul MMLV reverse transcriptase 7. Mix by pipetting, and aliquot requi red volume minus RNA (19ul number of need RT reactions) into a 96-well hard shell PCR plate 8. Transfer 1ul the number of needed RT reactions of 5ng/ul sample RNA or varied concentration standard RNA into corresponding tubes 9. Mix by pipetting, cover with film (M J Research Microseal A) and place into the thermal cycler. 10. Create a program for 10 minutes at 25 C, 30 minutes at 42 C, 30 minutes at 60 C and 5 minutes at 95 C to kill the enzyme and stop the reaction. 11. After this is finished, spin gently and use as cDNA.
166 APPENDIX A: (CONTINUED) Primer Optimization. 1. This step is important particularly for SYBR green reactions that lack the specificity of Taqman probes and require a melt curve analysis. 2. The qRT-PCR reactions can be run at 50ul or 25ul/well, but with 25ul, extra care must be given to sealin g the plate because evaporation can occur on the edge wells. 3. Primers for GOI are diluted to 100uM stocks in water and then further diluted 1:20 to 5uM in water. 4. The cDNA from the rev txn reactions is not cleaned up or purified in any way and added straight to the PCR reac tions. Any species specific cDNA pool at 50ng/ul can be used for the primer opt to s pare that which was just made. 5. Using the primer design parameter s mentioned above, the final in-well concentration of F and R primers is usually 300nM each. To save reagents, do an initial screeni ng at these concentrati ons and look at the melt curve analysis. Therefore the r eaction mix for a single PCR reaction of a single sample is as follows (multiply by the # of PCR reactions needed, usually 3.5 per samples): a. 12.5ul of 2x SYBR green master mix b. 7.5ul water c. 1.5ul F primer
167 APPENDIX A: (CONTINUED) d. 1.5ul R primer e. 2ul cDNA 6. Aliquot enough mix for 3.5 reactions (replicates) into 8 strip PCR tubes and then mix 10 times with cDNA Ali quot 25ul from t hese 8-strip PCR tubes to corresponding wells of a 96-well hard skirted PCR plate. a. Be very careful when making the replicates during aliquoting that excess reaction does not get into the wells from one replicate to the other (the viscosity of the mix due to the presence of Tris/DMSO may cause the liquid to cling to t he outside of the tip). Be sure to use filter tips for everything and mi x the master mix very well before aliquoting into the plate. b. The plates (MJ Research) used are specially designed as well as the film (MJ Research Microseal B) that covers them to be optically clear. 7. Plates are then spun at a lo w speed and then can be stored at 4 C for 24 hours. 8. The PCR is as follows: a. 1 cycle of 95 C for 5 minutes (this acti vates the hot start Taq) b. 40 cycles of: i. 95 C for 15 seconds ii. Suggested annealing temperat ure from Operon (58-65 C) for 1 minute
168 APPENDIX A: (CONTINUED) c. Then the melt curve program runs which starts at 55 C and increases 0.5 C every 10 seconds until 100 C. It measures fluorescence at each interval and a derivative equation is used to show a peak where the fluorescenc e drops off dramatically. This uses the principle of SYBR as a double strand specific intercalating dye. 9. There should be little or no reactivity in the NTCs, precise replicates, and the melt curve analysis should show a single peak. If the melt curve shows multiple peaks, new primers must be optimized and ordered. Quantitative Real Time-PCR of Samples. 1. Once suitable primer pairs are found, use the same principles discussed above in the Primer Optimization se ction to analyze expression of the cDNA from the samples of interest that were reverse transcribed also described above in the Reverse Transcription section. Quantification and Statistical Analyses. 1. The slope of the line generated from plotting the threshold cycle (y-axis) against the log ng RNA in the standard curve indicates the efficiency of the reaction. In order to compare ex pression of 2 genes or to use a nonchanging internal control (e.g. 18S rRNA or 28S rRNA) for mathematical
169 APPENDIX A: (CONTINUED) manipulations and determination of signi ficance, the slopes must be within 0.1 of each other and have a R2-value of no less than 0.95. a. A reaction that is 100% efficient will have a slope of -3.3, however this is uncommon when using reverse transcriptase which is a very inefficient reaction. Normally the slopes are less efficient, around 4.0 (~75%). 2. Using the equation of the best fit li ne, threshold cycles from samples can be used to calculate how many log nanograms of RNA fr om the GOI are present. The anti-log of this value is then taken to determine actual nanograms of RNA. 3. The GOI value is then divided by t he internal standard control (18S or 28S) and this gives the fold change in expression.
170 APPENDIX B: NA+/K+ ATPASE ASSAY PROTOCOL Reagents. 1. Protein (Tissue) Re suspension Buffer (1L) a. 85mM NaCl (4.97g) b. 20mM KCl (1.49g) c. 4mM MgCl (0.813g) d. 0.2mM EGTA (0.076g) e. 30mM Histidine (4.66g) 2. ATPase Buffer (1L) a. 140mM NaCl (8.18g) b. 20mM KCl (1.49g) c. 3mM MgCl (0.61g) d. 30mM Histidine (4.66g) e. 3mM ATP (1.65g) f. +/100uM Oubain g. Filter Sterilize for longer shelf life h. May change color, remake 3. Acid Molybdate Solution (100mL) a. 0.5g Ammonium Molybdate in 0.5M H2SO4 b. If precipitate forms or color changes, remake 4. Saponin a. Make 2mg/ml stock in water
171 APPENDIX B: (CONTINUED) 5. Coloring Agent (Fiske-Subbarow Reducer) a. 5g in 31.5mL b. Store in amber bottle (Foil tube) @ RT c. Good for 1 month 6. Phosphate Standard Solution a. 4mM phosphoric acid Procedure. 1. Homogenize 20-30mg brain tissue in 1mL of cold protein resuspension buffer by rotor-stator for 30-45secs. Place on ice. 2. Add 10uL of 2mg/ml saponin to 1mL of homogenate (final conc. = 20ug/uL), incubate @ 37C for 15min, and place on ice. 3. Run Bradford Assay and adjust protein concentrations accordingly with protein resuspension buffer (minimum of 10mg/mL tiss ue weight or ~ 1mg/mL protein). 4. In 96-well V-bottom plate, cr eate standard curve wit h Phosphate Standard Solution, making 7 2-fold dilutions begi nning from 4mM (triplicates -final volume should be the same as protein sample wells). 5. Add 60uL of ATPase Buffer +/oubain to wells in the same V-bottom plate in step 4 for protein samples to be ana lyzed (triplicates for (-) oubain and (+) oubain, therefore, you need six wells per sample).
172 APPENDIX B: (CONTINUED) 6. Add 5-10ug protein (5 0-100ug tissue weight) to wells, mix and incubate @ 37C for 30min. 7. Add 120uL Acid Moly bdate Solution to all well s, mix, t hen add 10uL of Fiske-Subbarow Reducer, mix and incubate for 10min @ RT. 8. Centrifuge plate so that any precipitate is pellete d, and transfer 100uL to a flat-bottom optically clear plate (not polypropylene). 9. Read spectrophotometrically at 660nm. 10. Use the equation of t he line from the standard curve to convert OD to concentration of phosphate (mM). 11. Subtract the +oubain values fr om the oubain values, then use this difference to calculate how much phos phate was generated specifically by Na+/K+ ATPase.
173 APPENDIX C: CELL CULTURE AND AMYLOID PREPARATION Mouse Neuroblastoma 2A (N2A) Maintenance. 1. Adherent N2A cells frozen in 95% complete media (Earls minimum essentials medium (EMEM) + 10% heat inactivated fetal bovine serum (FBS) + 1% penicillin/streptomycin solution (PS) with 5% DMSO are rapidly thawed upon removal from nitrogen. 2. Cells are immediately placed into 20ml of warm complete medium in a vented T-75 flask and allowed to incubate at 37 C until 90% confluent. 3. Cells should be passaged every 3-4 days by treatment with 0.25% (w/v) Trypsin0.53 mM EDTA solution for 5 minutes and followed by scraping. All traces of serum that contains tr ypsin inhibitor should be removed with a PBS wash prior to adding trypsin-EDTA. 4. Cells in trypsin are then passag ed to new flasks with 20ml of fresh complete medium. Differentiation of N2A Cells. 1. N2A cells can be treated with dibutyry l-cAMP to facilitate differentiation from their oncogenic stat e to a neuronal state. APPENDIX C: (CONTINUED)
174 2. N2A cells are subcultured and added to 6 well plates at 50-60% confluency (equivalent to 1 T-75 fla sk passaged to 3 6 well plates), and are permitted to reattach over night in complete medium. 3. Complete medium is then removed and cells are rinsed with PBS. 4. Dibutyryl cAMP (db-cAMP) is added at a concentration of 1mM to serumfree EMEM + PS, which is then applied to the cells in the 6 well plates. 5. These cells are then actively differentiating into neurons which is visible under the microscope by the presence of extended neurites. This process is allowed to continue for 48-72 hours. 6. These cells can then be treated with nerve growth factor (mouse maxillary NGF-2S) by directly applying the NGF solution to a final in-well concentration of 100nM. 7. After 1 hour, remove media from wells and rinse with PBS. 8. Add 350ul of RLT buffer with -mercaptoethanol to the cells in each well, scrape, and aspirate liquid in to a Qiashredder tube. 9. Spin down at max speed for 3 minutes, remove Qiashredder from collection tube and begin RNA isolat ion as described by the RNeasy Qiagen protocol for animal cells. NGF elicits the induction of several immediate early genes in differentia ted Neuro2A cells as measured by qRT-PCR (Figure 1).
175 APPENDIX C: (CONTINUED) Preparation of 5mM A Peptide Stock Solutions. 1. Recombinant amyloid peptide (1mg ) purchased from rP eptide (Cat # A1002-2) is used to prepare monomeric, oligmeric and fibrillar forms of A 2. Using a Gastight Hamilton syringe draw 221.7ul of cold 1,1,1,3,3,3 Hexafluoroisopropanol (HFIP) and punc ture the rubber cap of the A vial to resuspend the powder. 3. Allow the solution to stand at r oom temperature for 30 minutes and sonicate for 5 additional minutes. 4. Aliquot the suspension into micr ocentrifuge tubes (VWR # 20170-293), in a 10ul (45ug) amount. 5. Allow the HFIP to evaporate overnight in a fume hood leaving only a thin film of A present at the bo ttom of the tube. 6. The next day, speedvac the tubes for 1 hour without heat to remove all traces of liquid. These can be stored at -20 C on desiccant for several months. 7. To use the A right away, resuspend the 45ug film in 2ul of freshly opened dry DMSO (Sigma # D2650) to yield a 5mM stock A solution. Vortex thoroughly and sonicate for 5 minutes.
176 APPENDIX C: (CONTINUED) Preparation of Monomeric A. 1. Dilute the 5mM A DMSO stock with 98ul of water to yield a 100uM preparation of monomeric A Then vortex and use. Preparation of Oligomeric A. 1. Dilute the 5mM A DMSO stock with 98ul of ice cold phenol-free (phenolred-free) Hams F-12 culture medium to yield a 100uM preparation of oligomeric A 2. Vortex and incubate overnight at 4 C. For optimal performance, this should be used on the following day. Preparation of Fibrillar A. 1. Dilute the 5mM A DMSO stock with 98ul of 10mM HCl to yield a 100uM preparation of fibrillar A 2. Vortex and incubate overnight at 37 C. 3. The following day, neutralize acid with equimolar amount of NaOH. For optimal performance, this shoul d be used on the following day.
177 APPENDIX C: (CONTINUED) FIGURE 1. Induced expression of IEGs in differentiated N2A cells by nerve growth factor measured by qRT-PCR. This figure shows NGF treated (g ray bars) and non-treated (black bars) db-cAMP differentiated Neuro2A expres sion (gray bars) as a percentage of the average of undifferentiated Neur o2A cells (set as 100%) for the immediate early genes (IEGs) Zif268, Arc, and Nur77. This figure demonstrates that in Neuro2A cells di fferentiated for 72 hours with db-cAMP, NGF causes a robust increase in IE G expression. Additionally, Na+/K ATPase III expression, which is selectively over-expressed in neurons compared to non-differentia ted neuro2A cells, has increased expression in the db-cAMP treated Neuro2A cells, quantitatively demonstrating that differentiation of Neuro2A cells into neurons was being facilitated by the dbcAMP treatment. The values repr esented in this figure are the mean + SEM. * indicates significant differences between db-cAMP differentiated Neuro2A expression and undifferentia ted Neuro2A (p < 0.05) when measured by oneway ANOVA. "" above the bracket indicates differences (P < 0.05) in expression between NGF-tr eated db-cAMP differentia ted Neuro2A cells and non-treated db-cAMP differ entiated Neuro2A cells.
178 APPENDIX C: (CONTINUED) 0 100 200 300 400 500 Zif268ArcNur77Na+/K+ ATPase aIIIChange in Expression as a Percent of NonDifferentiated Neuro2A Expression +/SEMIEGs * * *
ABOUT THE AUTHOR Chad A. Dickey is a neuroscientis t who researches the molecular mechanisms of Alzheimers disease as they relate to cognitive function. His other research interests include neuroimmunology and synaptic plasticity. Chad was born and reared in Tampa, Fl orida. He received his undergraduate degree in microbiology and his Masters degr ee in medical sciences in 1998 and 2002 respectively, both from the University of South Florida. He successfully defended his doctoral dissertation in 2004 at t he University of South Florida. Prior to graduate school, Chads experience in cluded HIV and Alzheimers disease immunotherapy in the laborat ories of Drs.Dave Morgan, Marcia Gordon and Ken Ugen at the University of South Fl orida. He also worked on microarray development in the laborat ory of Dr. Warren Pledger at the Moffitt Cancer Center. Chad is a member of the Phi Kappa Phi honor so ciety and the Society for Neuroscience. Chad lives in Fl orida with his wife, Adria.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001474315
007 cr mnu|||uuuuu
008 040712s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000360
Dickey, Chad Anthony.
The influence of amyloid-beta, a major pathological marker in Alzheimer's disease, on molecular cognitive processes of APP+PS1 transgenic mice
h [electronic resource] /
by Chad Anthony Dickey.
[Tampa, Fla.] :
University of South Florida,
Thesis (Ph.D.)--University of South Florida, 2004.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 186 pages.
ABSTRACT: Alzheimer's disease (AD) is characterized by anterograde amnesia followed by a progressive decline in cognitive function. Post mortem examination of forebrain tissue from sufferers reveals the presence of extracellular amyloid-beta plaques, intracellular neurofibrillary tangles, activation of glial cells and massive neuron loss. Transgenic mice expressing mutated forms of the amyloid precursor protein (APP) and presenilin-1 (PS1) genes develop neuritic amyloid plaques, glial cell activation and memory deficits, without the formation of intracellular tangles and neurodegeneration. The mechanisms by which these transgenic mice develop mnemonic deficiencies are unclear. Gene expression of aged memory-deficient APP+PS1 mice compared with non-transgenic littermates measured by microarray and subsequent quantitative real-time PCR (qRT-PCR) analysis revealed 6 inducible immediated-early genes (IEGs) and 5 other more stably expressed plasticity-related genes (PRGs) that were significantly down-regulated in amyloid-containing hippocampus, but not down-regulated in amyloid-free cerebellum. Other genes linked to memory remained stably expressed in both regions. Analysis of forebrain AD tissue revealed that all genes measured were down-regulated presumably due to neurodegeneration, while the amyloid-free region maintained stable expression. IEG expression in APP+PS1 mice was sensitive to lower levels of amyloid. However, only in the presence of a substantially larger amyloid burden, when memory deficits reliably persist, were both PRGs and IEGs down-regulated. Importantly, we found that IEG expression was decreased in APP+PS1 mice following exposure to a novel environment, indicating that the induction of these IEGs was impaired, rather than the basal expression of resting mice. Na+/K+ ATPase, an enzyme critical for the maintenance of membrane potential, was identified as a down-regulated PRG. We found that activity of this enzyme was both impaired in the hippocampi of APP+PS1 mice and specifically inhibited by high concentrations of amyloid-beta. Na+/K+ ATPase immunostaining revealed decreased protein in the area surrounding the amyloid plaque, where dystrophic neurites were visible, indicating amyloid may disrupt ionic gradients resulting in neuritic dystrophia. These findings suggest that amyloid accumulation may result in the impairment of IEG induction and disruption of the Na+/K+ ATPase, possibly eliciting the memory loss developed in APP+PS1 mice.
Adviser: David Morgan
immediate early genes.
x Pharmacology and Therapeutics
t USF Electronic Theses and Dissertations.