Return to Abstracts Index

8.0 Towards an animal model of Alzheimer's disease: Can phorbol esters fan the flames? N.R. Smalheiser. Dept. of Psychiatry, University of Illinois,M/C 912, 1601 W. Taylor Street, Chicago, IL 60612. E-mail: smalheiser@psych.uic.edu

Keywords: animal model, phorbol esters, kinase C, tumor promoters, carcinogenesis

Abstract: Initiator-promoter protocols have been classically employed for the study of experimental carcinogenesis in mouse skin. Initiators cause specific changes in critical genes; in contrast, tumor promoting agents stimulate signal transduction and alter gene expression within tissues, nonspecifically but effectively "fanning the flames" of a latent cancer process. Alzheimer's disease (AD) is another chronic degenerative disease of aging which has been generally conceptualized in terms of initiator-promoter-progression scenarios. The etiologies and primary pathogenetic pathways of cancer and AD are presumably distinct, yet in both diseases, a common set of molecules and cellular responses have been strongly implicated as contributing to progression -- for example, cytokines, free radicals, proteases/ inhibitors, abnormal calcium homeostasis, altered phosphorylation, altered cytoskeletal organization, inflammatory responses and accelerated extracellular matrix turnover. Phorbol esters which bind and activate protein kinase C induce this spectrum of changes not only in skin (where they largely account for tumor promoting activity) but in neural tissue as well. These findings suggest the hypothesis that repeated application of phorbol esters may accelerate the development of AD-type lesions in animals predisposed by a specific prior initiating stimulus. Initiation-promotion protocols of this type are novel and have not previously been tried in AD, but can be readily adapted to existing animal models (e.g., overproducing amyloid) which so far have produced only limited AD-type pathology.

8.1 The cause of neuronal pathology in Alzheimer's disease. J.C. Vickers, C.E. King, T.C. Dickson, P.A. Adlard, I. Jacobs and H.L. Saunders. Division of Pathology, Clinical School, University of Tasmania, 43 Collins Street, Hobart, Tasmania 7000, Australia. E-mail: James.Vickers@path.utas.edu.au

Keywords: Cellular and Animal Models, neurofilament, tau, ß-amyloid, neurofibrillary, physical damage

Abstract: Our recent hypothesis on the cause of neuronal pathology in Alzheimer's disease (AD) proposes that the masses of ß-amyloid filaments that form plaques cause prolonged physical damage to axons, setting off a sprouting reaction in the neuron that involves the cytoskeletal changes that lead to neurofibrillary pathology. We used double labelling immunohistochemical techniques to examine the earliest forms of neuronal pathology associated with ß-amyloid deposition in the neocortex of elderly individuals likely to be in the initial stages of AD. In such cases, abnormal neurites were found within ß-amyloid plaques, particularly those located in supragranular layers of the neocortex. These early forms of dystrophic neurites were characterised by their neurofilament immunoreactivity and ring- or club-like morphology (1). We found identical structures in the rodent cortex following experimentally induced physical damage (2). In Alzheimer's disease cases, the immunocytochemical profile of abnormally sprouting dystrophic neurites suggested an evolution from axons containing accumulations of neurofilaments to more end-stage structures labelled for tau and stained with thioflavine S (3). In addition, the initial stage of neurofibrillary tangle formation appears to intimately involve alterations in neurofilaments. Cumulatively, this data suggests that prolonged stimulation of the neuronal reaction to injury by plaque formation would lead to profound cytoskeletal changes and abnormal axonal sprouting, as well as similar cytoskeletal alterations in the cell body of origin as part of the "retrograde reaction". This may ultimately result in the spectrum of neurofilament and tau abnormalities that constitute the neurofibrillary pathology of AD.

1. Vickers et al (1996) Experimental Neurology 141, 1-11.
2. King et al. (1997) Neuroreport 8, 1663-1665.
3. Vickers et al. (1994) Neuroscience 62, 1-13.

8.2 Herpes simplex type 1 virus in brain is a risk factor for Alzheimer's disease. R. F. Itzhaki*, G. K. Wilcock+, W-R. Lin*. *Molecular Neurobiology Lab. UMIST, Manchester. +Dept. of Care of the Elderly, Frenchay Hospital, Bristol. E-mail: ruth.itzhaki@umist.ac.uk

Keywords: Epidemiology and Risk Factors, Alzheimer's disease, Apolipoprotein E, Herpes Simplex Type 1 virus, Etiology, Herpes Labialis

Abstract: The E4 allele of the gene for apolipoprotein E (apoE) is a risk factor in sporadic Alzheimer's disease (AD) but is neither essential nor sufficient for development of the disease. Environmental factors must presumably be involved also. One possible factor is herpes simplex type 1 virus (HSV1). We previously detected latent HSV1 in post mortem brain from many aged normal people and AD patients, using polymerase chain reaction (PCR). We proposed that limited virus reactivation, resulting from stress or immunosuppression, causes more damage in those destined to develop AD than in normals because of specific host or virus characteristics [1-3].

We have since examined the apoE genotypes of 46 AD patients (mean age, 79, range, 54-96) and 44 age-matched normals (mean age 79, range, 58-95), using PCR for HSV1 detection, and for genotyping (after restriction nuclease digestion). We found the apoE4 allele frequency of AD patients to be 53% in those who are HSV1-positive in brain and 10% in those who are HSV1-negative. In normals, the frequencies were 4% and 6% in HSV1-positives and HSV1-negatives, respectively. Multiple logistic regression methods show that the value for the HSV1-positive AD group differs significantly from that of the other groups: the relative risk of AD (apoE4 plus HSV1-positivity) versus these groups equals 19.0 (95% C.I: 7.2-49.7). Thus, possession of an E4 allele and HSV1 in brain confers a strong risk of developing the disease whereas either apoE4 or HSV1 alone does not do so [4-6]. (The fact that a high proportion of normals are HSV1-positive and most do not possess an E4 allele argues against the possibility that AD patients, or those possessing an E4 allele, are more susceptible to HSV1 infection of brain.)

We have searched by PCR also for another virus, varicella zoster (VZV), which has a predilection for neuronal latency, but have not found it in any of our brain specimens [7]. The presence of one neurotropic virus - HSV1 - in many aged human brains, and the absence of another - VZV - is consistent with our hypothesis that HSV1 is specifically involved in the aetiology of AD.

In the PNS, most people harbour latent HSV1 but on reactivation it causes cold sores only in some. We have found that apoE is involved here: in 40 cold sore sufferers and 33 non-sufferers, the apoE4 allele frequencies were 36% and 9% respectively (p<0.0001) [6]. Thus, our data on AD patients and on cold sore sufferers indicate that the combination of HSV1 and apoE4 is particularly damaging in the nervous system. HSV1 in brain is the first environmental agent to be definitely implicated in AD. Our findings point to the future possibility of prevention of the disease by immunisation, and of retardation of its progression in early stage patients, by use of anti-viral agents.

We have examined also apoE genotypes in 60 patients with another disorder caused by HSV1 - but in the cornea - herpes simplex keratitis. Our results show that apoE4 allele frequency is not a risk factor in this disease (Lin et al, submitted). It thus seems that the damaging combination of apoE4 and HSV1 is specific for the nervous system.

1. Jamieson GA, Itzhaki RF et al. J Med Virol 1991; 33; 224-227.

2. Jamieson GA, Itzhaki RF et al. J Pathol 1992; 167: 365-368.

3. Itzhaki RF, Jamieson GA et al. In: Corain B eds. Alzheimer's Disease: Advances in Clinical and Basic Research. New York: Wiley & Sons, 1993: 97-102.

4. Lin W-R, Shang D, Itzhaki. Biochem Soc Trans 1995; 23: 594S.

5. Lin W-R, Shang D, Itzhaki RF. Molec Chem Neuropath 1996; 28: 135-141.

6. Itzhaki RF, Lin W-R, Shang D, Wilcock GK, Faragher B, Jamieson GA. The Lancet 1997; 349: 241-244.

7. Lin W-R, Casas I, Wilcock GK, Itzhaki, RF. J Neurol Neurosurg & Psych. 1997; 62: 586-589.

8.3 Do Beta-App And A Presenilin Fragment Form A Multimeric Septahelical Receptor?S.W. Barger. Departments of Geriatrics and Anatomy, University of Arkansas for Medical Sciences; Geriatric Research Education and Clinical Center, J.L. McClellan Veterans Affairs Medical Center; Little Rock AR 72205, USA. E-mail: swbarger@life.uams.edu

Keywords: Presenilins, Amyloid precursor protein (APP), G-protein, Topology

Abstract: The genes for beta-APP and presenilins (PS) have been linked to familial Alzheimer's disease. Interactions between the two proteins have been implied by the effects of PS and their mutation on beta-APP processing. More definitive evidence of a physical interaction between beta-APP and PS was reported by Dewji & Singer (1) and Weidemann et al. (2). Elegant topology studies indicate that PS has an enigmatic octahelical membrane-spanning configuration (3). Previous studies also have demonstrated proteolysis of PS between the sixth and seventh putative transmembrane (TM) regions (4). I propose that the N-terminal fragment of PS (containing TMs 1-6) forms a complex with beta-APP to generate a heterodimeric G-protein-coupled receptor or non-receptor G-protein modulator. The interaction of the APP/PS complex with a G-protein may influence apoptotic pathways or play a role in beta-APP trafficking.

In support of this model, the cytoplasmic tail of beta-APP interacts with G-alpha(sub)o (5), and PS-mediated apoptosis is exacerbated by beta-APP expression but attenuated by pertussis toxin (6). In addition, cell death resulting from expression of the C-terminal 100 residues of beta-APP is dependent upon a G-protein-associated kinase (7). Finally, the beta-APP TM and PS1 TM5 bear homology to TM1 and TM6, respectively, of G-protein-coupled septahelical receptors. The proposed model is being explored through measurements of GTPase activity in membrane preparations of cells co-expressing beta-APP and PS1, their AD-associated mutations, or an APP/PS1 fusion protein. Insight also may be gained through use of such cells to study biological responses that are modified by pharmacological manipulation of G-proteins. Cross-linking, coprecipitation, and other measures of interaction between the APP/PS1 complex and G-proteins also would be illustrative. (Supported by the Inglewood Foundation for Alzheimer's Research)

1. Dewji NN, Singer SJ (1996) Proc. Natl. Acad. Sci. USA 93: 12575.

2. Weidemann A, et al. (1997) Nature Med. 3: 328.

3. Doan A, et al. (1996) Neuron 17: 1023.

4. Thinakaran G, et al. (1996) Neuron 17: 181.

5. Yamatsuji T, et al. (1996) Science 272: 1349.

6. Wolozin B, et al. (1996) Science 274: 1710.

7. McPhie DL, et al. (1996) Soc. Neurosci. Abstr. 22: 204.4.

8.4 The Role Of Apoe Neurotoxicity In Alzheimer's Neuropathology. K.A. Crutcher, M. Tolar, M. A. Marques. University of Cincinnati and ApoLogic, Inc.E-mail: crutchka@email.uc.edu

Keywords: Molecular Pathology, apolipoprotein E (ApoE), neurotoxicity, fragment, proteolysis, inflammation

Abstract: Apolipoprotein E (apoE) exhibits an isoform-specific association with Alzheimerís disease (AD) such that the E4 isoform co-segregates with a higher risk of the disease. We propose that this association arises from neurotoxicity of apoE and/or fragments of apoE metabolism. This hypothesis is based on the following lines of evidence obtained from in vitro studies: 1) Synthetic peptides derived from the receptor binding domain of apoE are toxic to neurons; 2) The N-terminal 22 kDa thrombin-cleavage fragment of apoE is neurotoxic with the E4-derived fragment being more toxic than the E3-derived fragment; 3) Full-length apoE exhibits neurotoxicity, again with the E4 isoform being more toxic than the E3 isoform, and this toxicity is correlated with the appearance of apoE fragments in the media; 4) Protease inhibitors provide significant protection against the neurotoxic effects of apoE and inhibit the production of apoE fragments. A likely possibility for the source of apoE neurotoxicity in AD are microglial cells, which produce apoE and are associated with senile plaques. The presence of microglial cells and other markers of inflammation in areas of neuropathology, as well as the epidemiological evidence for protective effects of NSAIDs, have led to the suggestion that AD may represent a chronic inflammatory condition. ApoE may, therefore, be a participant in this inflammatory response and contribute to the neurite degeneration and neuronal death that occurs in the disease. This hypothesis predicts that apoE fragments should be produced in, and associated with, areas of neuropathology in AD. A further prediction is that inheritance of the E4 allele might increase the risk of other conditions in which chronic inflammation occurs."

8.5 Pseudodementia Can Be Treated By Modulation Of Stress Reactions Even In Dementia: Report Of Three Cases. R. Cocchi. CASA DI CUR VILLA SILVIA, VIA A. GARIBALDI, 64, 60019 SENIGALLIA ITALY. E-mail: mc7057@mclink.it

Keywords: Therapeutics, DEMENTIA, PSEUDODEMENTIA, STRESS, GABA/GLUTAMATE

Abstract: There is not any drug therapy of dementias, because we cannot revive dead neurons. We can instead act on the share of pseudodementia due to the dysfunction of alive but sick neurons, which always escort any type of dementia. On the other hand, stress reactions play a major rol in intellectual impairments of old people. When it happens so, they can get out pseudodementia. Stress reactions can come out from external stressors, or internal stressors or both. The internal stressor most likely is an illness itself, but this one could also be true dementia (eg. DAT). These facts can account for some consequences: i. We can found a share of pseudodementia in true dementia too; ii. When pseudodementia is a share of true dementia the only therapeutical success we can obtain now comes from the cut down of the pseudodementia share; iii. We can modulate stress reactions by drugs, by this way reducing pseudodementia; iv. Modulation of stress reactions affects even the course of true dementia. According to this perspective too, the best approach cannot be the monotherapy therapy because it acts on only one of many altered neurochemical mechanisms. Moreover it obliges to high doses to which the body will answer with stress reactions (to the drug itself) and consequent neurochemical resetting leading to lower efficiency of the drug used. Multiple drug therapy let acting in a synergistical way on many neurochemical altered circuits, by allowing low doses. Drug modulation of stress responses mainly acts on GABA and related brain mechanisms (GLU, ACH, 5HT, DA, NA, ATP etc.). It starts by increasing GABA A inhibition and applies to backward biochemical steps in order to reduce monoamines GABA B inhibition and the excess glutamate, by restoring its trans-formation into GABA. The results of this approach have a direct effect on the EEG. Quantitative EEG analysis seems a tool to point the time course out. Three cases are presented according to these premises: A female aged 53, with DAT; A 62 old male with depressive pseudodementia, without significant EEG findings; A male aged 52 with se-vere EEG alterations and pseudodementia. Therapies lasted more than two years each.

References
1) Cocchi R.: Drug therapy of pseudodementia as modulation of stress reactions: 3 cases. It. J. Intellect. Impair. 1996, 9: 173-180.
2) Cocchi R.: Possibilities and limits of drug therapies in dementias. It. J. Intellect. Impair. 1997, 10: 29-33

8.6 Ferritin is secreted into the cerebrospinal fluid in Alzheimer's disease: Implications for plaque formation. S.R. Robinson1, A. McRae2 and W. Zhao3. 1: Vision, Touch & Hearing Research Centre, University of QLD, QLD, Australia. 4072. 2: Department of Psychology, Monash University, Clayton, VIC, Australia, 3168. 3: Department of Anatomy & Cell Biology, University of Göteborg, S-413 90, Göteborg, Sweden. E-mail: s.robinson@vthrc.uq.edu.au

Keywords: Histopathological Course and Diagnosis, Ferritin, Iron, Microglia, Cerebrospinal fluid, Amyloid

Abstract: In vitro studies have shown that soluble beta-amyloid is readily precipitated by micromolar concentrations of iron. An involvement of iron in plaque formation is further suggested by the high concentration of iron within many beta-amyloid plaques. Since ferritin is the protein that binds most iron in the brain, we have proposed (Robinson et al., 1995; Batton et al., 1997) that the deposition of beta-amyloid in Alzheimer's disease is initiated by the extracellular release of ferritin. A prediction of this hypothesis is that the cerebrospinal fluid of Alzheimer's patients should contain elevated titres of secreted ferritin.

In the present study we have measured ferritin levels in the cerebrospinal fluid of 8 non-demented control patients, 33 presumed Alzheimer's patients, 8 presumed vascular dementia patients and 10 Parkinson's disease patients. All samples were obtained with informed consent from living donors at the University of Göteborg (ethical approval #298-86), and stored at -70oC until use. Purified horse spleen apoferritin (ICN) was used as a positive control. Proteins in each sample were separated on SDS-PAGE, stained with Coomassie blue then subjected to a computerized densitometry scan.

On SDS-PAGE, purified spleen apoferritin yielded several protein bands. The strongest band was about 20kDa, but other weak bands could be seen, including one at 53kDa. Cerebrospinal fluid samples from Alzheimer's and vascular dementia patients displayed a prominent protein band at 53kDa. Antisera to L-chain ferritin (ICN; 1:10,000) recognized a 53kDa band on Western blots of the apoferritin. This band was absent from most control samples, but was strongly expressed in most Alzheimer's and vascular dementia samples.

Secreted (serum) ferritin can be strongly immunolabelled with antisera to L-chain ferritin, and unlike cellular ferritin, it is composed of 50-58kDa subunits (Linder et al., 1996). These facts strongly suggest that the protein we have identified in cerebrospinal fluid is a subunit of secreted ferritin. Its concentration was almost 10-fold higher in the Alzheimer's and vascular dementia groups than in the control group (p<0.001; ANOVA and Newman-Keuls multiple comparisons test). The concentrations of secreted ferritin in the Parkinson's disease group were mildly elevated but they did not differ significantly from those in the control group.

Evidence from hepatic research indicates that ferritin is secreted by cells as part of an inflammatory response, and that secreted ferritin may contribute to the transfer of iron from iron-rich cells to iron depleted cells. We propose that in Alzheimer's disease, iron-rich ferritin is secreted by activated microglial cells. Factors in the extracellular environment, such as ascorbate, catacholamines or acidic pH, release iron from the ferritin, which subsequently binds and precipitates beta-amyloid peptide. In Alzheimer's disease this process may be aided by the presence of isoforms of beta-amyloid that are readily precipitated by iron, and by the presence of molecular chaperones, such as alpha 1-antichymotrypsin. The neural degeneration caused by the aggregated beta-amyloid will trigger an inflammatory response in local microglia, thereby perpetuating the cycle. If our hypothesis is correct, the pathogenesis of Alzheimer's disease may be slowed by treatments that: 1) reduce the secretion of ferritin by microglia; 2) chelate extracellular iron; 3) scavenge iron-dependent free radicals.

References:
1) Batton, C.I. et al. (1997) Ferritin-rich microglia are concentrated within beta-amyloid plaques. Alzheim. Res. 3: 23-28.
2) Linder, M.C. et al. (1996) Serum ferritin: Does it differ from tissue ferritin? J. Gastroenterol. Hepatol., 11: 1033-1036.
3) Robinson, S.R. et al. (1995) Most amyloid plaques contain ferritin-rich cells. Alzheim. Res. 1: 191-196.

8.7 A Proposed Neurotoxic Lesion Model for the Elucidation of the Role of APP in Neurite Outgrowth and Synapse Maintenance In Vivo. M. Paradis, R, Lee, R, J. Wurtman. Massachusetts Institute of Technology, Dept. of Brain & Cognitive Sciences, Cambridge MA 02139. E-mail: mdparadi@athena.mit.edu

Keywords: APP, Lesion, Neurotoxins, Neurite Outgrowth, Synapse Maintenance, Animal model

Abstract: Several researchers have shown that Amyloid Precursor Protein (APP) mRNA and immunoreactivity in experimental animals are rapidly and greatly increased in regions of ischemic, truamatic or neurotoxic brain injury. Impaired neuronal function and loss of synapses in APP knock-out (KO) mice, as well as work with the RERMS peptide sequence of APP, suggest to us that the post-lesion increase in APP is an adaptive response indicative of a functional role for APP, or its cleavage products, in synapse maintenance and/or formation. Our cell culture work has demonstrated that intracellular APP751, APP695 and Abeta regulate levels of tau mRNA and protein, as well as its phosphorylation state. This finding links the precursors of the two pathological hallmarks of AD, amyloid plaques and neurfibrillary tangles, and implicates APP and/or its cleavage products in the promotion of neurite outgrowth.

Thus, we propose construction of a model for AD based on selective chemical lesioning of brain structures (i.e. cell bodies, fiber tracts or terminal fields of specific neurotransmitter pathways) using known neurotoxins such as 6-Hydroxydopamine, 5,7-Dihydroxytryptamine, Kainic Acid, 1921gG-Saporin (a novel cholinotoxin), or RMP-7 (a permeabilizer of the neurovascular endothelium). Such lesions can be performed in wild-type, KO (e.g. APP, APLP1, APLP2) or transgenic (APP overexpression) mice. Molecular, pharmacological, and behavioral investigations of our model will allow us to 1) elucidate the _in_vivo_ functions of APP, its isoforms, its cleavage products and its gene family members (collectively referred to as the APP Protein Family) in synapse maintenance and regeneration; 2) assess the relationships between mechanisms of neurovascular integrity and those of AD pathogenesis, including possible roles for inflammation and inflammatory mediators; and 3) provide a screening procedure for compounds with proposed therapeutic value. Unlike previous chemolesion experiments which only investigated effects at the site of the lesion, our model will explicity address the systems consequences of lesioning by exploring effects in cells both pre- and postsynaptic to the lesion.

One of our principal hypotheses is that the APP Protein Family is involved in neurite outgrowth and regeneration; hence we would expect to see increased mRNA expression, and increased levels of specific members of the APP Protein Family, prior to or coincident with the onset of regeneration. Cellular and biochemical studies in our chemolesion model will illuminate in vivo interactions between APP Protein Family members and tau, the presenilins or synaptophysin; the cellular origins of these protein will be determined by their co-localization with marker proteins (e.g. GFAP and MAP2). The results of these studies may then be coupled with behavioral/cognitive measures, such as performance on Morris Water and Radial Arm Mazes, in pre- and post-lesioned animals. Furthermore, these animals may be treated with established or novel therapeutics to probe their mechanisms of action and to determine their treatment potential.

8.8 Oxidative Stress Is Central To The Pathogenesis Of Alzheimer Disease. M.A. Smith, G. Perry. Case Western Reserve University, Institute of Pathology, 2085 Adelbert Road, Cleveland, Ohio 44106, USA. E-mail: mas21@po.cwru.edu

Keywords: oxidative stress, free radicals, antioxidants, therapeutics

Oxidative damage not only occurs to the proteins comprising the lesions of Alzheimer disease, neurofibrillary tangles and senile plaques, but also precedes lesion formation in neurons at risk of death during the disease. While oxidative stress may not be the primary etiology of the disease, it precedes specific cellular and tissue damage that underlies the onset of dementia. Our assertion links Alzheimer disease to normal aging where oxidative stress has been implicated. Further, our idea is supported by epidemiological data linking the prevalence of Alzheimer disease to diet, an established link to longevity. Additionally, the proteins known from genetics to play a role in AD such as beta protein precursor, apolipoprotein E or presenilins either regulate apoptosis or bind transition metals - both important mechanisms related to oxidative stress. The clinical efficacy of reducing oxidative stress in Alzheimer disease is highlighted by the reduction in prevalence and progression by agents such as nonsteroidal anti-inflammatories, estrogen, and vitamin E which all share the one common feature of reducing oxidative stress.

Therefore, we suggest that oxidative stress is a central feature of the pathogenesis of Alzheimer disease and that treatments that specifically target sources of oxygen radicals may have particular therapeutic efficacy. Specifically, we suggest two broad based treatment strategies, based on preventing the generation of or ameliorating the effects of oxidative stress will be particularly effective.