Refine
Document Type
- Doctoral Thesis (7)
- Article (1)
- Part of Periodical (1)
Has Fulltext
- yes (9)
Is part of the Bibliography
- no (9)
Keywords
Institute
- Pharmazie (4)
- Biochemie und Chemie (3)
Das Ziel der vorliegenden Arbeit bestand darin, die Wechselwirkung des AD-assoziierten Proteins PS1 mit dem Cytoskelett in verschiedenen biochemischen und immuncytochemischen Modellen zu untersuchen. Auf diese Weise sollten Hinweise für das Verständnis der bisher nur in Grundzügen bekannten physiologischen Funktion von PS1 erhalten werden. Ausgehend vom Nachweis der Anreicherung endoproteolytischer PSI-Fragmente in Cytoskelett-Fraktionen aus Hirngewebe wurde die spezifische Assoziation des Proteins mit Mikrofilamenten, Mikrotubuli und Neurofilamenten untersucht. Alle durchgeführten biochemischen und immuncytochemischen Experimente zeigen übereinstimmend die spezifische Assoziation von PS1 mit Mikrofilamenten. So konnten PSI-Fragmente über zwei Depolymerisations/ Polymerisations-Zyklen zusammen mit Actin aus Rattenhirn aufgereinigt werden. Die spezifische Spaltung von Actin-Filamenten durch Gelsolin in cytoskelettreichen Fraktionen aus Hirngewebe führte zur Freisetzung kürzerer Filamente, die nach Zentrifugation zusammen mit PS1- Fragmenten im Überstand erschienen. In immuncytochemischen Färbungen kolokalisierte PS1 mit Actin-Filamenten in primären Gliazellen und der neuronalen Zelinie NT2. Eine vor kurzem veröffentlichte Arbeit zeigte außerdem die Kolokalisation von PS1 und Mikrofdamenten in primären Hippocampus-Neuronen. Die fehlende Kosedimentation von in-vitro translatiertem PS1 und Deletionskonstrukten von PS1 mit aufgereinigten und in-vitro assemblierten Actin-Filamenten, das Fehlen konservierter Actin-Bindungsmotive in der Aminosäuresequenz von PS1 sowie bekannte Interaktionen von PS1 mit Actin-bindenden Vertretern der Filamin- und Armadillo-Proteinfamilien lassen auf eine indirekte Assoziation von PS1 mit Mikrofdamenten schließen. Die vorliegende Arbeit konnte die Frage, welche Rolle Mikrotuli oder Neurofilamente bei der Cytoskelett-Assoziation von PS1 spielen, nicht abschließend Mären. PS1-Fragemente wurden im ersten, nicht jedoch im zweiten Depolymerisations/Polymerisations-Zyklus zusammen mit Tubulin aus Rattenhirn aufgereinigt. In dem gleichen Versuchsansatz kosedimentierte in-vitro translatiertes PS1 nicht mit Mikrotubuli. Immuncytochemische Färbungen zeigten dagegen die Kolokalisation von PS1 mit Mikrotubuli in primären Hippocampus-Neuronen, die auch von zwei aktuellen Arbeiten in primären Hippocampus-Neuronen und murinen Embryonen bestätigt wurde. Die spezifische Depolymerisation von Mikrotubuli in cytoskelettreichen Fraktionen war unter den in der vorliegenden Arbeit verwendeten Bedingungen nicht erfolgreich, so daß aus diesem Experiment keine Aussagen zur Assoziation von PS1 mit Mikrotubuli abgeleitet werden konnten. Es Iäßt sich die Hypothese aufstellen, daß sehr kleine PSI-haltige Vesikel über andere Proteine als PS1 mit Mikrotubuli verbunden sind, so daß PS1 mit Mikrotubuli scheinbar kolokalisiert ist. Hochauflösenden lichtmikroskopische bzw. elektronenmikroskopische Aufnahmen oder biochemische Versuche, die nicht zur Zerstörung der Vesikel durch Detergens führen, könnten für die Überprüfung dieser Hypothese herangezogen werden. In-vitro translatiertes PS1 sowie Deletionskonstrukte von PS1 kosedimentierten nicht mit aufgereinigten und in-vitro assemblierten Neurofilamenten, und in primären Hippocampus-Neuronen kolokalisierte PS1 nicht mit Neurofilamenten. Es wäre jedoch wünschenswert, die Experimente mit geeigneten Positiv-Kontrollen für die Kosedimentation und geeigneteren Antikörpern, die bei einem besseren Signal/Hintergrund-Verhältnis eine stärkere Vergrößerung der Immunfluoreszenzen erlauben, zu wiederholen. Im transgenen Tiermodell beeinträchtigten pathogene Mutationen von PS1 nicht, zumindest nicht quantitativ, die Fähigkeit von PS1, an Cytoskelett-Elemente zu binden. Es wäre jedoch möglich, daß es Unterschiede in der Cytoskelett-Assoziation von wt PS1 und PS1 mit pathogenen Mutationen gibt, die nur im Gewebe von FAD-Patienten zu detektieren sind, nicht aber im transgenen Tiermodell, da die bisher bekannten AD-Tiermodelle nicht alle Aspekte der Erkrankung reproduzieren. Die vorliegenden Ergebnisse können die Basis bilden für weitergehende Untersuchungen zur Funktion von PS1 und der Bedeutung von PS1-Cytoskelettinteraktionen in der Alzheimer Demenz. Aktuelle Arbeiten fokussieren dabei stark auf die Interaktion von PS1 mit Mikrofilament-bindenden Vertretern der Armadillo-Proteinfamilie, die bei Zell-Zell-Adhäsionsprozessen eine wichtige Rolle spielen. Im Diskussionsteil wurde ausgeführt, daß aber, gerade im Hinblick auf die Neurodegeneration bei der Alzheimer Demenz, ein möglicher Einfluß von PS1 auf den schnellen axonalen Transport ein interessanter Ansatz für weitere Versuche sein kann.
The hypothesis that oxidative stress plays a role in the pathogenesis of Alzheimer’s disease (AD) was tested by studying oxidative damage, acitvities of antioxidant enzymes and levels of reactive oxygen species (ROS) in several models. To this end, mouse models transgenic for mutant presenilin (PS1M146L) as well as mutant amyloid precursor protein (APP) and human post mortem brain tissue from sporadic AD patients and age-matched controls were studied. Aging leads to an upregulation of antioxidant enzyme activities of Cu/Zn-superoxide dismutase (Cu/Zn-SOD), glutathione peroxidase (GPx) and glutathione reductase (GR) in brains from C57BL/6J mice. Simultaneously, levels of lipid peroxidation products malondialdehyde MDA and 4-hydroxynonenal HNE were reduced. Additionally, pronounced gender effects were observed, as female mice display better protection against oxidative damage due to higher activity of GPx. Hence, antioxidant enzymes provide an important contribution to the protection against oxidative damage. In PS1M146L transgenic mice oxidative damage was only detectable in 19-22 months old mice, arguing for an additive effect of aging and the PS1 mutation. Both HNE levels in brain tissue as well as mitochondrial and cytosolic levels of ROS in splenic lymphocytes were increased in PS1M146L mice. Antioxidant defences were unaltered. In PDGF-APP and PDGF-APP/PS1 trangenic mice no changes in any of the parameters studied were observed in any age group. In contrast, Thy1-APP transgenic mice display oxidative damage as assessed by increased HNE levels. Reduced activity of Cu/Zn-SOD may explain this observation. Additionally, gender modified this effect, as female APP transgenic mice display higher b-secretase cleavage of APP and simultaneously increased HNE levels and reduced Cu/Zn-SOD activity earlier than male mice, i.e. from an age of 3 months and before the formation of Ab plaques. Reduced Cu/Zn-SOD activity was also found in another APP transgenic mouse model, in APP23 mice. In post mortem brain tissue from sporadic AD patients activities of Cu/Zn-SOD and GPx were however increased, and changes were most pronounced in temporal cortex. Simultaneously, levels of HNE but not MDA were elevated. Additionally, in vitro stimulation of lipid peroxidation led to increased MDA formation in samples from AD patients, indicating that increased activity of Cu/Zn-SOD and GPx are insufficient to protect against oxidative damage. Furthermore, the observed changes were subject to a gender effect, as samples from female AD patients showed increased activities of Cu/Zn-SOD and GPx as well as increased HNE levels, indicating that brain tissue from females is more sensitive towards oxidative damage. Levels of soluble Ab1-40 were positively correlated with with MDA levels and activities of Cu/Zn-SOD and GPx. Additionally, levels of lipid peroxidation products MDA and HNE are gene-dose-dependently modulated by the Apolipoprotein E4 allele, the most important genetic risk factor for AD known so far. While MDA levels were negatively correlated with MMSE scores, a measure for cognitive function, HNE levels were highest in AD patients with moderate cognitive impairment. Hence, increased HNE levels may play an important role in neurodegenerative events at an early disease stage. In summary, oxidative damage, as assessed by increased HNE levels, could be detected in sporadic AD patients and in different transgenic mouse models. The results of this thesis therefore support the further research of pharmacological targets aiming at augmentation of antioxidant defences for therapy or prophylaxis of Alzheimer’s disease.
Aging and age-related diseases are becoming more and more important for our society and our health care system. Alzheimer's disease (AD) is a disorder that destroys some parts of the brain and is characterized by global cognitive decline including a progressive irreversible loss of memory, orientation, and reasoning. “Healthy aging”, therefore, is one of the major aims for modern medicine. Apoptosis, or programmed cell death, plays an important role for example in fetal development, as well as for learning processes. T-lymphocytes usually undergo apoptosis in order to terminate an acute inflammation. The aim of this thesis was to explore the changes in the apoptotic mechanism of peripheral lymphocytes from Alzheimer’s disease (AD) patients in contrast to physiological aging. The experiments were conducted with lymphocytes of healthy volunteers of different ages, AD patients and young and aged mice. Moreover, transgenic mice carrying familiar AD-related mutations were examined. The aging study of peripheral cells of ‘healthy’-aged volunteers revealed an age-related increase of basal apoptosis. In addition, spontaneous apoptosis as well as apoptosis induced by oxidative stress (ROS) or by Fas engagement were enhanced in aging. A closer look at the subcellular basis of the lymphocytes (e.g. B-, NK-, CD4+-, and CD8+-T cells) determined that all lymphocyte subsets were affected by aging. Therefore, it could be concluded that the regulation of apoptosis is generally impaired in lymphocytes of aged persons. The increased susceptibility to oxidative stress supports the ‘Free radical theory of aging’ that claims the radicals to be the cause for the aging-process. In mice an increase of basal, spontaneous and ROS-induced apoptosis was detected in T cells from the spleen, as well. An oral treatment over two weeks with the Ginkgo biloba extract EGb761 showed a clear reduction of ROS-induced apoptosis in the treated group. Interestingly, basal and spontaneous apoptosis, e.g. physiological apoptosis, were not effected by the plant extract. This is an important benefit for therapy since physiological apoptosis has a great relevance in the elimination of cancer-cells for example. In conclusion, the antidementive drug EGb761 reduces specifically ROS-induced apoptosis that a plays an important role in aging as shown in this thesis. Based on the data found in healthy aging, lymphocytes from AD patients were assessed for apoptosis. The cells show enhanced levels of basal, spontaneous, and Fas-induced apoptosis. In subsequent experiments it was demonstrated that mainly the T cells were responsible for the findings. However, the NK-cells provided an important impact as well. In concordance with AD-affected neurons, peripheral lymphocytes of AD patients show clear signs of apoptotic cell death. In addition, basal apoptosis of T cells and the CD4/CD8-ratio showed a correlation with the severity of the dementia. Therefore, it could be speculated that apoptosis is due to activation-induced cell death (AICD) that occurs in acute and chronic activation of adaptive immunity. In AD there is a chronic neuroinflammation in the CNS triggering degeneration of neural tissue. In order to explore this, the experimental model of lymphocyte’s activation was established in healthy aging first. The study included the detection of various events of lymphocyte’s activation on the basis of the T cell subsets (CD4+ and CD8+). The inducibility to mitogenic stimulation clearly decreased in both subsets in aging. In contrast, T lymphocytes from AD patients showed an enhanced activation subsequent to mitogenic stimulation compared with age-matched nondemented persons. Only proliferation of CD8+ T cells was clearly reduced in AD. This data could be clues that an increased generation of memory T cells due to chronic neuroinflammation might be evident in AD. Memory T lymphocytes show increased inducibility upon mitogenic activation. Interestingly, CD8+ memory T cells display decreased prolifertive capacity. Due to activation, cells die by apoptosis later on. It could be concluded that AD patients display an increased amount of memory T cells compared to controls. The data implicate that there could be a cross talk between inflammatory within the brain and inflammatory cells of the periphery. This is an interesting point since the brain used to be assumed as immune-privileged zone. According to the experiment, the information of the diseased brain is transferred to white blood cells. The connection of those two compartments might raise the opportunity to observe and probably to influence easily not-accessible regions like the brain. Transgenic mice carrying mutations in familiar AD-relevant genes (Amyloid-Precursor-Protein, Presenilin-1, respectively) displayed enhanced levels of apoptotic T cells from the spleen, as well. It seems that those mutated proteins influence the regulation of apoptosis. Probably, they are involved in the increased cell death of T- and NK-cells, as well. Animals overexpressing Presenilin-1 showed reduced levels of apoptotic cell death. It was demonstrated with molecuar biology tools that Presenilin-1, processed during apoptosis, has an anti-apoptotic effect.
Das Tau-Protein bildet im Verlauf von zahlreichen Demenzen, mit Morbus Alzheimer als die bekannteste unter ihnen, Aggregate, die sogenannten „neurofibrillären Geflechte“, die aus Fibrillen, den „paired helical filaments“ (PHFs) bestehen. Außerdem ist das Tau-Protein ein essentielles „microtubule-associated protein“, welches für die Aufrechterhaltung des neuronalen Zellmetabolismus notwendig ist. Dies veranlasste uns dazu, das Tau-Protein als Monomer, in der Mikrotubuli-gebundenen Form und als Fibrille zu charakterisieren. Die Technik, die wir hierzu verwendeten war die NMR-Spektroskopie, die als einzige strukturaufklärende Technik mit atomarer Auflösung dazu in der Lage ist, intrinsisch ungeordnete Proteine, wie das Tau-Protein, zu charakterisieren. Zwar war die Signalzuordnung der NMR-Spektren eine große Herausforderung, dennoch war es möglich praktisch alle Rückgratresonanzen sogar für die längste Tau-Isoform htau40 mit 441 Resten erfolgreich eindeutig zu identifizieren. Mit Hilfe der Zuordnung war es möglich das Tau-Protein bezüglich residualer Strukturelemente, Rückgratdynamik und Bindungsverhalten zu untersuchen. Wir konnten zeigen, dass in der C-terminalen Hälfte des Tau-Proteins, in welcher eine charakteristische Domäne vorliegt, die durch vier imperfekte „repeat“-Regionen (Länge ist jeweils ca. 31 Reste) gekennzeichnet ist, partieller beta-Strukturcharakter vorliegt. Ebenfalls weist diese Region eine verhältnismäßig hohe Rigidität auf. Aus diesem Grund betrachten wird diesen sequentiellen Bereich als den Aggregationskeim, was auch durch die Beobachtung verstärkt wird, das genau diese Zone den rigiden Teil der Fibrillen bildet. Diese aggregationsanfällige Region bindet gleichzeit stark an Mikrotubuli, wodurch ihre Pathogenität im gesunden biologischen System blockiert sein sollte. Mutationen oder Instabilität in den Mikrotubuli können jedoch dazu führen, dass immer höhere Mengen an Tau in freier gelöster Form vorliegen, sich Dimere ausbilden, welche dann weiter aggregieren und schließlich PHFs bilden, die eine starke cross-beta-Struktur aufweisen. Die übrigen Bereiche, wie die N-terminale Hälfe oder die äußersten 50 C-terminalen Reste weisen hingegen einen partiellen alpha-helikalen Charakter auf und eine höhere Peptidrückgratflexibilität. Deshalb kann man diese Elemente als aggregationsinhibierend betrachten. Das genauere Zusammenspiel dieser Elemente muss noch aufbauend auf der vorliegenden Dissertation verstärkt im Detail untersucht werden.
Die Alzheimer-Demenz (AD) ist gekennzeichnet durch extrazelluläre Ablagerungen des Amyloid-beta-Peptids (Aß), durch neurofibrilläre Bündel bestehend aus dem Tau-Protein, massiven Neuronenverlust und synaptische Dysfunktion. Weiterhin ist bekannt, dass mitochondriale Dysfunktion sowie ein gestörter NO-Stoffwechsel eine entscheidende Rolle bei der AD spielen. Um genauere Informationen über die Ursache der mitochondrialen Dysfunktion zu erhalten, wurden akute, chronische und dosisabhängige Effekte von Aß auf die NO-Produktion und die mitochondriale Funktion untersucht. Als Zellkulturmodelle standen PC12- und HEK-Zellen zur Verfügung, die entweder mit humanem Wildtyp-APP (APPwt) oder mit der schwedischen Doppelmutation im APP-Gen (APPsw) stabil transfiziert waren. APPsw-PC12-Zellen wiesen Aß-Spiegel im pikomolaren Bereich auf. Im Vergleich dazu hatten APPsw-HEK-Zellen ca. 20fach erhöhte Aß-Spiegel im niedrig-nanomolaren Bereich. Interessanterweise wiesen sowohl APPsw-PC12- als auch APPsw-HEK-Zellen im Vergleich zu den jeweiligen Kontrollzellen signifikant erhöhte NO-Spiegel auf. Dies ging in beiden Zellsystemen mit signifikant erniedrigten ATP-Spiegeln einher. Die Inkubation untransfizierter Zellen mit extrazellulärem Aß1-42 führte nur zu einem schwachen Anstieg der NO-Spiegel und zu einem leichten Abfall der ATP-Spiegel. Dies weist darauf hin, dass in erster Linie intrazelluläre Aß-Effekte den NO-Anstieg und die ATP-Reduktion bewirken. Die 48-stündige Inkubation mit dem gamma-Sekretasehemmstoff DAPT führte zur beinahe vollständigen Normalisierung der NO- und ATP-Spiegel in APP-transfizierten PC12- und HEK-Zellen. Das stützt die Hypothese, dass der gestörte NO-Stoffwechsel und die mitochondriale Dysfunktion durch Aß-Anreicherungen hervorgerufen werden und nicht durch eine Überexpression von APP. Passend zu den reduzierten ATP-Spiegeln zeigten APPsw-PC12-Zellen eine signifikant erniedrigte Cytochrom-C-Oxidase-Aktivität. Des Weiteren konnte APP in Mitochondrien von APPsw-PC12-Zellen nachgewiesen werden. Die Reduktion der ATP-Spiegel und die verminderte Cytochrom-C-Oxidase-Aktivität können also zum einen durch die Aß-bedingten erhöhten NO-Spiegel und zum anderen durch die Anwesenheit von APP bzw. Aß im Mitochondrium hervorgerufen werden. Auf der Ebene des mitochondrialen Membranpotentials wiesen die beiden Zelllinien stark unterschiedliche Ergebnisse auf. APPsw-PC12-Zellen zeigten unter basalen Verhältnissen ein leicht hyperpolarisiertes mitochondriales Membranpotential, was auf einen Gegenregulationsmechanismus hinweist. APPsw-HEK-Zellen wiesen bereits basal ein signifikant erniedrigtes mitochondriales Membranpotential auf. Nach Inkubation mit dem gamma-Sekretasehemmstoff DAPT normalisierte sich sowohl die Hyperpolarisation des mitochondrialen Membranpotentials in APPsw-PC12-Zellen als auch die Depolarisation in APPsw-HEK-Zellen. Anhand der in dieser Arbeit gewonnenen Daten konnte ein Modell sowohl für die sporadische als auch für die familiäre AD entwickelt werden. APPsw-PC12-Zellen spiegeln hierbei die pathogenen Mechanismen in Patienten mit sporadischer AD wider, wohingegen APPsw-HEK-Zellen die initialen Veränderungen bei Patienten mit familiärer AD aufzeigen. Mitochondriale Fehlfunktion und ein gestörter NO-Stoffwechsel stellen entscheidende initiale Pathomechanismen bei AD dar. Innerhalb der Gruppe der Antidementiva konnte gezeigt werden, dass sowohl Ginkgo-biloba-Extrakt als auch Piracetam schützende Effekte auf die mitochondriale Funktion ausüben. Aufgrund der wichtigen Rolle von mitochondrialer Fehlfunktion in der Pathogenese der Alzheimer Demenz stellen Ginkgo-biloba-Extrakt und Piracetam zwei sehr interessante Präventions- und Therapieoptionen bei Patienten mit leichten kognitiven Störungen bzw. bei Patienten mit AD dar.
Over the last years there has been an increasing interest in the involvement of the MVA-pathway and of members of the small GTPases, in the development and progression of AD. Earlier investigations mainly focused on the role of cholesterol in disease pathology. This research was supported by retrospective cohort studies, initially showing beneficial effects of the long-term intake of cholesterol lowering statins, on the incidence of the development of sporadic AD. However, in more recent literature increasing attention has been paid to the isoprenoids, FPP and GGPP, due to their crucial role in the post-translational modifications of members of the superfamily of small GTPases. In AD, these proteins were amongst others shown to be involved in mechanisms affecting APP processing, ROS generation and synaptic plasticity. A major factor impeding the clarification of the role of the MVA-pathway intermediates in these mechanisms was the lack of a sensitive and accurate method to determine FPP and GGPP levels in brain tissue. Hence, a state of the art HPLC-FLD method for the quantification of the isoprenoids FPP and GGPP in brain tissue was successfully developed. After the introduction of a double clean-up step from complex brain matrix samples and the synthesis of an appropriate IS (DNP), the method was fully validated according to the latest FDA guideline for bioanalytical method validation. Furthermore, this method was transferred to a faster and more sensitive, state of the art UHPLC-MS/MS application. Additionally, the method was shown to be applicable for mouse brain tissue and data was generated from an in vivo mouse simvastatin study and for different mouse models. According to the aims of the thesis, the current work describes for the first time absolute isoprenoid concentrations in human frontal cortex white and grey matter. Furthermore, this is the first report of isoprenoid levels in the frontal cortex of human AD brains. Further results were shown from mouse brains originating from different mouse models, including the Thy-1 APP mouse model mimicking AD pathology in terms of Aβ formation or C57Bl/6 mice at different ages. AD prevalence can be clearly correlated with increasing age. Therefore, three different generations of mice were investigated. The study demonstrated constant isoprenoid and cholesterol levels in the first half of their life followed by a significant increase of FPP and GGPP in the second half (between 12 and 24 month of age). Cholesterol levels were also elevated in the aged group, but again the effect was less pronounced than shown for the isoprenoids. These results lead to the tentative conclusion that cerebral isoprenoid levels are elevated during aging and that this accumulation is amplified during AD leading to accelerated neuronal dysfunction. In a different mouse study, using the C57Bl/6 mice, in vivo drug intervention with the HMG-CoA reductase inhibitor simvastatin revealed strong inhibition of the rate limiting step of the mevalonate/isoprenoid/cholesterol pathway and resulted in the first report of significantly reduced FPP and GGPP levels in brain tissue of statin treated mice. These results open for the first time the possibility to monitor drug effects on cerebral isoprenoid levels and correlate these data with a modulation of APP processing, which was shown by our group in previous studies. Interestingly, apart from the isoprenoid reduction following statin treatment the reduction of brain cholesterol was also significant but to a lesser extent. These findings support the notion that isoprenoid levels are more susceptible to statin treatment than cholesterol levels. Furthermore, this suggests a strong cellular dependence on FPP and GGPP, as the pool seems to be easily depleted, which finally could lead to cell death. The first investigations of farnesylated Ras and geranylgeranylated Rac protein levels by means of immuno-blotting, substantiated the notion of a decreased abundance of prenylated small GTPases under statin influence as a consequence of reduced isoprenoid levels. These findings demonstrate for the first time a correlation of FPP and GGPP levels with the abundance of small GTPases. These findings together with the results from the AD study prove that isoprenoid levels are not strictly subject to the same regulation as cholesterol levels. To further understand the physiological regulation in the cell, in vitro experiments with different inhibitors of the mevalonate/isoprenoid/cholesterol pathway were conducted. These results confirmed the isoprenoid and cholesterol reducing effects of statin treatment as observed in the aforementioned in vivo mouse study. Interestingly, cholesterol synthesis inhibition targeted after FPP as the branch point, led to significantly elevated FPP levels. FTase inhibition led to significantly reduced FPP levels, whereas inhibition of the GGTase I did not show a significant change of either isoprenoid levels.
Alzheimer’s Disease (AD) is the most common neurodegenerative disorder marked by progressive loss of memory and cognitive ability. The pathology of AD is characterised by the presence of amyloid plaques, intracellular neurofibrillary tangles and pronounced cell death. The aim of this thesis was to investigate pathways involved in the Aß cascade of neurodegeneration. Since novel findings indicate that already this Aß species exerts neurotoxic effects long before hyperphosphorylated tau, neurofibrillary tangles and extracellular Aß plaques appear, the investigations were accomplished with specific regard to the effects of intracellular Aß. The Swedish double mutation in the APP gene results in six- to eightfold increased Aß production of both Aß1-40 and Aß1-42 compared to human wildtype APP cells (APPwt). Data obtained from PC12 cells indicate that it is possible to specifically increase the Aß load without enhancing APP expression levels. On the basis of these findings, it seemed possible to investigate dose-dependent effects of Aß in multiple experimental designs. These assay designs were created in order to mimick different in-vivo situations that are discussed to occur in AD patients: APPsw PC12 cells exhibit low physiological concentrations of Aß within picomolar range in contrast to APPsw HEK cells, expressing Aß levels within the nanomolar range. Of note, the APPsw HEK cells showed a specific and highly significant increase in the intracellular accumulation of insoluble Aß1-42. Moreover, an intracellular accumulation of Aß and APP was found in the mitochondria of the HEK APPsw cells suggesting a direct impact on mitochondrial function on these cells. This effect might finally lead to disturbances in the energy metabolism of the cell or to increased cell death. Furthermore, baseline g- and ß-secretase activity was assessed since these enzymes represent promising therapeutic targets to slow or halt the disease process. As expected, ß-secretase activity was significantly elevated in all APPsw cell lines. This might be due to the proximity of the Swedish double mutation next to the N-terminus of the Aß sequence. Interestingly, g-secretase activity was similarly increased in PC12 APPsw cells. In addition, the toxicity of different Aß species was investigated in SY5Y and PC12 cells with regard to their effect on cellular viability mirrored by mitochondrial activity using MTT assay. Here, it turned out that not monomers, but already dimers are neurotoxic correlates. Fibrillar Aß species showed the highest toxicity. In the next step, SY5Y cells forming endogenous, dimeric APP and Aß were investigated. In accordance with previous findings, these cells showed a decreased MTT reduction potential in comparison to APPwt and control SY5Y cells reflecting a decrease of cellular viability. The impaired energy metabolism of the cells was even more drastically mirrored by reduced baseline ATP levels. In the second part of this thesis, the expression and intracellular distribution of Bcl-2 family proteins and pro-apoptotic mitochondrial factors under baseline conditions and during oxidative stress were analyzed in the APPwt and APPsw bearing cells. The most prominent finding was the reduction of expression levels of the anti-apoptotic factor Bcl-xL in the cytosolic fractions of APPwt and APPsw PC12 cells. This might indicate that a lack of anti-apoptotic factors or their altered intracellular distribution, rather than an increase in caspase-dependent pro-apoptotic factors, could be responsible for the increased vulnerability of APPwt- and APPsw-transfected PC12 cells against oxidative stress. Since total Bcl-xL expression was unaffected in PC12 cells, in contrast to APPwt and APPsw-expressing SY5Y and HEK cells revealing significantly decreased Bcl-xL expression levels. Thus, alterations in Bcl-xL distribution seem to be an early event in the disease process. Increasing Bcl-xL expression might potentially be one promising strategy for AD modification. PC12 and HEK cells bearing APPsw or APPwt were treated with the potent g-secretase inhibitor DAPT. Of note, DAPT did not only efficiently block Aß production, but additionally led to an elevation of the MTT reduction potential, reflecting an increase in cellular viability. As another disease-modifying strategy, several efforts are undertaken to ameliorate AD-relevant symptoms by the treatment with nerve growth factor (NGF). Generally, it is known that substituted pyrimidines have modest growth-promoting effects. Here, KP544, a novel substituted pyrimidine, was characterised. This drug increased MTT reduction potential in terminally differentiated and undifferentiated PC12 cells. Furthermore, treatment with KP544 led to a reduction in Aß1-40 secretion. Thus, one may conclude that the target of KP544, GSK-3ß, represents a connecting link between the two main pathological hallmarks of AD and might thus be a very promising therapeutic target for AD.