Increased expression of BIN1 mediates Alzheimer genetic risk by modulating tau pathology.

Genome-wide association studies (GWAS) have identified a region upstream the BIN1 gene as the most important genetic susceptibility locus in Alzheimer's disease (AD) after APOE. We report that BIN1 transcript levels were increased in AD brains and identified a novel 3 bp insertion allele ∼28 kb upstream of BIN1, which increased (i) transcriptional activity in vitro, (ii) BIN1 expression levels in human brain and (iii) AD risk in three independent case-control cohorts (Meta-analysed Odds ratio of 1.20 (1.14-1.26) (P=3.8 × 10(-11))). Interestingly, decreased expression of the Drosophila BIN1 ortholog Amph suppressed Tau-mediated neurotoxicity in three different assays. Accordingly, Tau and BIN1 colocalized and interacted in human neuroblastoma cells and in mouse brain. Finally, the 3 bp insertion was associated with Tau but not Amyloid loads in AD brains. We propose that BIN1 mediates AD risk by modulating Tau pathology.

Contribution of Kunitz protease inhibitor and transmembrane domains to amyloid precursor protein homodimerization.

BACKGROUND: The two major isoforms of the human amyloid precursor protein (APP) are APP695 and APP751. They differ by the insertion of a Kunitz-type protease inhibitor (KPI) sequence in the extracellular domain of APP751. APP-KPI isoforms are increased in Alzheimer's disease brains, and they could be associated with disease progression. Recent studies have shown that APP processing to Aß is regulated by homodimerization, which involves both extracellular and juxtamembrane/transmembrane (JM/TM) regions. OBJECTIVE: Our aim is to understand the mechanisms controlling APP dimerization and the contribution of the ectodomain and JM/TM regions to this process. METHODS: We used bimolecular fluorescence complementation approaches coupled to fluorescence-activated cell sorting analysis to measure the dimerization level of different APP isoforms and APP C-terminal fragments (C99) mutated in their JM/TM region. RESULTS: APP751 was found to form significantly more homodimers than APP695. Mutation of dimerization motifs in the TM domain of APP or C99 did not significantly affect fluorescence complementation. CONCLUSION: These findings indicate that the KPI domain plays a major role in APP dimerization. They set the basis for further investigation of the relation between dimerization, metabolism and function of APP.

[The precursor of amyloid peptide in Alzheimer disease: a protein with multiple functions]. / Le précurseur du peptide amyloïde de la maladie d'Alzheimer: une protéine aux multiples fonctions.

Cellular metabolism of the amyloid precursor protein (APP) has been widely studied, but the function of the protein remains elusive. APP knock out mice do not show any phenotype, due to in vivo compensation by APLP genes, encoding proteins similar to APP. In order to study the neuronal metabolism of APP, human APP has been expressed in rat cortical neurons in culture. Following differentiation in culture, rat cortical neurons are organized into networks of connected cells, which show neuronal activity in the form of spontaneous and synchronous calcium oscillations. Expression of human APP in these neuronal networks inhibits calcium oscillations, while downregulation of endogenous APP expression increases the frequency and decreases the amplitude of oscillations. Therefore, APP controls neuronal calcium homeostasis and excitability. In the same experimental model, APP is also able to control the neuronal synthesis of cholesterol. Finally, the APP carboxy terminal domain is involved in the epigenetic control of gene expression. Modulation of neuronal expression of APP allows to identify several important functions of the precursor of the amyloid peptide found in senile plaques of Alzheimer disease.

[Alzheimer disease: cellular and molecular aspects]. / Maladie d'Alzheimer: aspects cellulaires et moléculaires.

A conclusive diagnosis of Alzheimer's disease (AD) can be made only by correlating clinical findings and neuropathological studies of post-mortem tissues. Two leading neuropathological changes correlate with the diagnosis of AD: first, the neurofibrillary tangles (NFTs) which accumulate in neuronal perikarya and are made of paired helical filaments (PHFs) containing the microtubule-associated protein tau; second, extracellular amyloid deposits in the form of diffuse or neuritic senile plaques which contain the amyloid peptide. In AD, NFTs can be easily visualized using antibodies recognizing the microtubule associated protein tau and are composed of bundles of PHFs. In the autopsy-derived AD brain, tau is hyperphosphorylated and more than 30 phosphorylation sites have been identified in PHF-tau proteins. The formation of NFTs is thought to be associated with a collapse of the microtubule network, disturbances of axoplasmic transports, synapse loss, neuritic atrophy, and neuronal death. Senile plaques are extracellular lesions which have been shown by electron micro-scopic studies to contain amyloid fibrils. Fibrils were isolated and a small 4.2 kDa poly-peptide was purified from this material. The amyloid peptide found in amyloid deposits of AD is designated Abeta. Since the Abeta peptide is small and unlikely to be a primary translational product, it was predicted to arise from a larger precursor. In 1987, this amyloid peptide precursor (APP) was characterised from the analysis of a full-length cDNA encoding a primary translational product of 695 residues. This protein is synthetized by neurons as a 100-kDa glycosylated transmembrane protein with a single membrane spanning domain. The use of cellular models has clearly identified two catabolic pathways for APP. A non amyloidogenic pathway, in which APP is cleaved by beta-secretase within the sequence of the amyloid peptide. This cleavage precludes the formation of the full-length Abeta found in the amyloid core of senile plaques. A second catabolic pathway of APP leads to the production of Abeta from its precursor. In this amyloidogenic pathway, APP is cleaved by beta-secretase at the N-terminus of Abeta. The C-terminal fragment of APP thus formed is in turn cleaved by beta-secretase to release the full-length amyloid peptide. In primary cultures of neurons over-expressing APP, the production of intraneuronal Abeta induces neuronal apoptosis. This neurotoxicity, which is not observed in epithelial cells, seems to be related to the formation of intraneuronal aggregates of Abeta 1-42. In AD, the specific inhibition of beta- or beta-secretase activities would decrease the production of Abeta from its precursor, in such a way that its relative concentration could be low enough to avoid the formation of aggregates. Molecules which can interact with Abeta in order to inhibit its aggregation are also being developed. Immunization against Abeta has also been tested in both animal models and clinical studies. Although these clinical studies had to be interrupted due to the development of T-lymphocyte meningoencephalitis in some patients, very preliminary results indicate that antibodies against Abeta slow cognitive decline in AD, and generate areas of neocortex devoid of senile plaques.

Adenylosuccinate lyase deficiency: study of physiopathologic mechanism(s).

Nucleotide concentrations were normal in adenylosuccinate lyase-deficient fibroblasts, and the succinylpurines were not toxic for cultured neuronal cells.

Mutant presenilin 1 proteins induce cell death and reduce tau-dependent processes outgrowth.

The expression of familial Alzheimer's disease mutants of presenilin-1 (PS1) proteins has been observed to induce cell death in cellular systems. To investigate how this phenomenon might be associated to alterations of the microtubule network, we have studied the effect of wild-type and mutant (C263R, P264L and delta9) PS1 proteins expression on the formation of microtubule-dependent processes outgrowth and the association of PS1 to the insoluble cytoskeletal fraction in a cell line expressing the tau microtubule-associated protein. Expression of wild-type and mutant PS1 was associated with increased cell death, most marked for the P264L and delta9 mutants. The three PS1 mutants induced a significant reduction of the length of cell processes. These effects were not associated to a change in tau phosphorylation. However, the mutant PS1 proteins increased the proportion of insoluble tau in the cytoskeletal fraction and they were concentrated in the same fraction. These results suggest that PS1 proteins interact with the microtubule network, affect its organization and that this phenomenon, more marked for the PS1 mutants, might play a role in the cell dysfunction induced by mutant PS1 proteins.

Neurofibrillary tangles and tau phosphorylation.

Neurofibrillary tangles (NFTs) are a characteristic neuropathological lesion of Alzheimer's disease (AD). They are composed of a highly-phosphorylated form of the microtubule-associated protein tau. We are investigating the relationship between NFTs and microtubule stability and how tau phosphorylation and function is affected in transgenic models and by co-expression with beta-amyloid precursor protein and presenilins. In most NFT-bearing neurons, we observed a strong reduction in acetylated alpha-tubulin immunoreactivity (a marker of stable microtubules) and a reduction of the in situ hybridization signal for tubulin mRNA. In transfected cells, mutated tau forms (corresponding to tau mutations identified in familial forms of frontotemporal dementias linked to chromosome 17) were less efficient in their ability to sustain microtubule growth. These observations are consistent with the hypothesis that destabilization of the microtubule network is an important mechanism of cell dysfunction in Alzheimer's disease. The glycogen synthase kinase-3 beta (GSK-3 beta) generates many phosphorylated sites on tau. We performed a neuroanatomical study of GSK-3 beta distribution showing that developmental evolution of GSK-3 beta compartmentalization in neurons paralleled that of phosphorylated tau. Studies on transfected cells and on cultured neurons showed that GSK-3 beta activity controls tau phosphorylation and tau functional interaction with microtubules. Tau phosphorylation was not affected in neurons overexpressing beta-amyloid precursor protein. Transgenic mice expressing a human tau isoform and double transgenic animals for tau and mutated presenilin 1 have been generated; a somatodendritic accumulation of phosphorylated transgenic tau proteins, as observed in the pretangle stage in AD, has been observed but NFTs were not found, suggesting that additional factors might be necessary to induce their formation.

Is aggregation of beta-amyloid peptides a mis-functioning of a current interaction process?

In a previous study, Hughes et al. [Proc. Natl. Acad. Sci. USA 93 (1996) 2065-2070] demonstrated that the amyloid peptide is able to interact with itself in a two-hybrid system and that interaction is specific. They further supported that the method could be used to define the sequences that might be important in nucleation-dependent aggregation. The sequence of the amyloid peptide can be split into four clusters, two hydrophilic (1-16 and 22-28) and two hydrophobic (17-21 and 29-42). We designed by molecular modeling and tested by the two-hybrid approach, series of mutations spread all over the sequence and changing the distribution of hydrophobicity and/or the spatial hindrance. In the two-hybrid assay, interaction of native Abeta is reproduced. Screening of mutations demonstrates that the C-domain (residues 29-40 (42)), the median domain (residues 17-22) and the N-domain (1-16) are all crucial for interaction. This demonstrates that almost all fragments of the amyloid peptide but a loop (residues 23-28) and the C-term amino acid are important for the native interaction. We support that the folded three-dimensional (3D) structure is the Abeta-Abeta interacting species, that the whole sequence is involved in that 3D fold which has a low secondary structure propensity and a high susceptibility to mutations and thus should have a low stability. The native fold of Abeta could be stabilized in Abeta-Abeta complexes which could in other circumstances facilitate the nucleation event of aggregation that leads to the formation of stable senile plaques.

Yersinia enterocolitica can deliver Yop proteins into a wide range of cell types: development of a delivery system for heterologous proteins.

Y. enterocolitica translocates virulence proteins, called Yop effectors, into the cytosol of eukaryotic cells. Here we investigated whether Y. enterocolitica could translocate Yops into a range of eukaryotic cells including neurons and insect cells. Y. enterocolitica translocated the hybrid reporter protein YopE-Cya into each of the eukaryotic cell types tested. In addition, Y. enterocolitica was cytotoxic for each of the adherent cell types. Thus we detected no limit to the range of eukaryotic cells into which Y. enterocolitica can translocate Yops. The Yop effectors YopE, YopH and YopT were each cytotoxic for the adherent cell types tested, showing that not only is Y. enterocolitica not selective in its translocation of particular Yop effectors into each cell type, but also that the action of these Yop effectors is not cell type specific. Invasin and/or YadA, two powerful adhesins were required for translocation of Yop into non-phagocytic cells but not for translocation into macrophages. To use the Yersinia translocation system for broad applications, a Y. enterocolitica translocation strain and vector for the delivery of heterologous proteins into eukaryotic cells was constructed. This strain + vector combination lacks the translocated Yop effectors and allows delivery into eukaryotic cells of heterologous proteins fused to the minimal N-terminal secretion/translocation signal of YopE. Using this strategy translocation of a YopE-Diphtheria toxin subunit A hybrid protein into several cell types has been shown.

The processing and biological function of the human amyloid precursor protein (APP): lessons from different cellular models.

One of the major neuropathological hallmarks of Alzheimer's disease is the presence of senile plaques in vulnerable regions of CNS. These plaques are formed of aggregated amyloid peptide. Amyloid peptide is released by the cleavage of its precursor (APP). The establishment of cell lines expressing human APP allowed to characterize both amyloidogenic and non-amyloidogneic pathways of APP catabolism and to identify some of the proteins involved in this processing (known as secretases). This led to a better comprehension of amyloid peptide production, which needs to be further characterized since gamma-secretase is as yet not identified; moreover, we still lack a clear overview of the interactions between APP and other proteins promoting Alzheimer's disease (tau, presinilinsellipsis). An important limitation of these cell lines for studying the mechanisms involved in Alzheimer's disease is supported by the observation that human APP expression does not modify transfected cells survival. The infection of primary neuronal cultures with full-length human APP indicates that APP expression induces neuronal apoptosis by itself; this neurotoxicity does not rely on extracellular production of APP derivatives (secreted APP, amyloid peptide). It is now essential to understand, in neuronal models, the production, localization and involvement of amyloid peptide in neurodegenerative processes.

A GG nucleotide sequence of the 3' untranslated region of amyloid precursor protein mRNA plays a key role in the regulation of translation and the binding of proteins.

The alternative polyadenylation of the mRNA encoding the amyloid precursor protein (APP) involved in Alzheimer's disease generates two molecules, with the first of these containing 258 additional nucleotides in the 3' untranslated region (3'UTR). We have previously shown that these 258 nucleotides increase the translation of APP mRNA injected in Xenopus oocytes (5). Here, we demonstrate that this mechanism occurs in CHO cells as well. We also present evidence that the 3'UTR containing 8 nucleotides more than the short 3'UTR allows the recovery of an efficiency of translation similar to that of the long 3'UTR. Moreover, the two guanine residues located at the 3' ends of these 8 nucleotides play a key role in the translational control. Using gel retardation mobility shift assay, we show that proteins from Xenopus oocytes, CHO cells, and human brain specifically bind to the short 3'UTR but not to the long one. The two guanine residues involved in the translational control inhibit this specific binding by 65%. These results indicate that there is a correlation between the binding of proteins to the 3'UTR of APP mRNA and the efficiency of mRNA translation, and that a GG motif controls both binding of proteins and translation.

The role of presenilin-1 in the gamma-secretase cleavage of the amyloid precursor protein of Alzheimer's disease.

Presenilin-1 (PS1) is required for the release of the intracellular domain of Notch from the plasma membrane as well as for the cleavage of the amyloid precursor protein (APP) at the gamma-secretase cleavage site. It remains to be demonstrated whether PS1 acts by facilitating the activity of the protease concerned or is the protease itself. PS1 could have a gamma-secretase activity by itself or could traffic APP and Notch to the appropriate cellular compartment for processing. Human APP 695 and PS1 were coexpressed in Sf9 insect cells, in which endogenous gamma-secretase activity is not detected. In baculovirus-infected Sf9 cells, PS1 undergoes endoproteolysis and interacts with APP. However, PS1 does not cleave APP in Sf9 cells. In CHO cells, endocytosis of APP is required for Abeta secretion. Deletion of the cytoplasmic sequence of APP (APPDeltaC) inhibits both APP endocytosis and Abeta production. When APPDeltaC and PS1 are coexpressed in CHO cells, Abeta is secreted without endocytosis of APP. Taken together, these results conclusively show that, although PS1 does not cleave APP in Sf9 cells, PS1 allows the secretion of Abeta without endocytosis of APP by CHO cells.

Transgenic expression of the shortest human tau affects its compartmentalization and its phosphorylation as in the pretangle stage of Alzheimer's disease.

We have generated transgenic mice expressing the shortest human tau protein, the microtubule-associated protein that composes paired helical filaments in Alzheimer's disease. Transgenic tau transcripts and proteins were strongly expressed in neurons in the developing and adult brain. In contrast to the endogenous tau that progressively disappeared from neuronal cell bodies during development, the human transgenic tau remained abundant in cell bodies and dendrites of a subset of neurons in the adult. This somatodendritic transgenic tau was immunoreactive with antibodies to tau phosphorylated on Thr181 and Thr231 and with the conformation-dependent Alz50 antibody. A few astrocytes expressing the transgenic tau were strongly immunoreactive with antibodies to additional tau phosphorylation sites, ie, at Ser262/ 356 and Ser396/404. All of these phosphorylation sites have been identified in paired helical filaments-tau proteins. In electron microscopy, the transgenic tau was detected into microtubules in axons and in dendrites but not in cell bodies. Neurofibrillary tangles were not detected in transgenic animals examined up to the age of 19 months. These results indicate that transgenic manipulation of tau expression and intracellular targeting is sufficient per se to affect tau compartmentalization, phosphorylation, and conformation partly as it is observed at the pretangle stage in Alzheimer's disease.

The long term adenoviral expression of the human amyloid precursor protein shows different secretase activities in rat cortical neurons and astrocytes.

Recombinant adenoviruses were used for the expression of human amyloid precursor protein (APP) of Alzheimer's disease in primary cultures of rat cortical neurons and astrocytes. The catabolic pathways of human APP were studied 3 to 4 days after infection, when the equilibrium of APP production was reached. Although the expression of human wild type APP (WtAPP) by rat neurons induced the production of both extracellular and intraneuronal amyloid peptide (Abeta), Abeta was not detected in the culture medium of rat astrocytes producing human WtAPP. Because a low beta-secretase activity was previously reported in rodent astrocytes, we wondered whether modifications of the APP amino acid sequence at the beta-secretase clipping site would modify the astrocytic production of Abeta. Interestingly, rat astrocytes produced high amounts of Abeta after expression of human APP carrying a double amino acid substitution responsible for Alzheimer's disease in a large Swedish family (SwAPP). In both rat cortical neurons and astrocytes, the beta-secretase cleavage of the human SwAPP occurred very early in the secretion process in a cellular compartment in which a different sorting of SwAPP and WtAPP seems unlikely. These results suggest that human WtAPP and SwAPP could be processed by different beta-secretase activities.

Proteolytical processing of mutated human amyloid precursor protein in transgenic mice.

The evidence that betaA4 is central to the pathology of Alzheimer's disease (AD) came from the identification of several missense mutations in the amyloid precursor protein (APP) gene co-segregating with familial AD (FAD). In an attempt to study the proteolytical processing of mutated human APP in vivo, we have created transgenic mice expressing the human APP695 isoform with four FAD-linked mutations. Expression of the transgene was controlled by the promoter of the HMG-CR gene. Human APP is expressed in the brain of transgenic mice as shown by Western blot and immunohistology. The proteolytic processing of human APP in the transgenic mice leads to the generation of C-terminal APP fragments as well as to the release of betaA4. Despite substantial amounts of betaA4 detected in the brain of the transgenic mice, neither signs of Alzheimer's disease-related pathology nor related behavioural deficits could be demonstrated.

Alpha 1-tubulin mRNA level is increased during neurite outgrowth of NG 108-15 cells but not during neurite outgrowth inhibition by CNS myelin.

alpha 1-tubulin is an isotype of alpha-tubulin, and its mRNA is expressed in the rodent nervous system. A high level of alpha 1-tubulin mRNA in neurones is associated with axonal outgrowth during development as well as with axonal regeneration after axotomy in adult animals. We quantitated alpha 1-tubulin mRNA levels in motor neurone-like NG 108-15 cells using Northern blots in order to determine whether the expression of this neurite outgrowth-associated gene is regulated in NG 108-15 cells during neurite extension and during inhibition of this process by CNS myelin. Here we report that during the acute phase of neurite outgrowth, alpha 1-tubulin mRNA level increases in NG 108-15, a maximal induction of 1.7-fold over the initial level occurring 24 h after neurite outgrowth onset. By contrast, when these cells are plated on CNS myelin alpha 1-tubulin mRNA levels show no such increase. These findings indicate that an increase of the alpha 1-tubulin mRNA level is associated with neurite outgrowth of NG 108-15 cells. More interestingly, this study also demonstrates that the inhibition of neurite outgrowth by CNS myelin may affect the expression of a gene encoding a protein involved in neurite extension.

Differential toxicity of ganciclovir for rat neurons and astrocytes in primary culture following adenovirus-mediated transfer of the HSVtk gene.

The toxicity of the suicide HSVtk gene approach is known to be targeted to DNA synthesis and, consequently, to dividing cells. This system is therefore useful for the treatment of brain tumors which contain dividing cells surrounded by a quiescent normal tissue. Adenoviruses are efficient vectors for the transfer of the HSVtk gene into the tumor but this can lead to the transduction of quiescent cells. In this study, we focused on the toxicity of the HSVtk/ganciclovir treatment for the two main cell types of the normal brain: astrocytes and neurons. Astrocytes and neurons in primary culture were infected by an adenoviral vector bearing the HSVtk gene (Ad.tk) and cells were exposed to different concentrations of ganclclovir. After 5 days of treatment, an MTT test measured a dramatic decrease in cell viability for treated astrocytes while a small decrease in cell viability was observed for neurons treated in the same experimental conditions. The differential toxicity of the HSVtk/ganciclovir treatment was also observed in cocultures of astrocytes and neurons: an immunocytochemical analysis of the treated cells showed major morphological modifications for astrocytes but not for neurons. Furthermore, our data suggest that a bystander effect is able to kill all the astrocytes while neurons from the same culture remain unaffected.

Generation of a monoclonal antibody to the carboxy-terminal domain of tau by immunization with the amino-terminal domain of the amyloid precursor protein.

A fusion protein between beta-galactosidase and the amino-terminal domain of amyloid precursor protein (APP) was used as an immunogen for the production of monoclonal antibodies. One of these antibodies, the 5D12 monoclonal antibody, labeled the neurofibrillary tangles (NFT) by immunohistochemistry, as well as isolated paired helical filaments (PHF) in electron microscopy. In immunoassay, the ascitic fluid produced by the 5D12 clone was demonstrated to contain a high titer of antibodies to heat-stable microtubule associated proteins (MAPs). By immunoblotting, the proteins recognized in heat-stable MAPs were found to correspond to tau proteins. The 5D12 antibody recognized normal tau isolated from rat and human brain homogenates, and PHF-tau isolated from the brain of patients with Alzheimer's disease (AD). By immunoblotting, the 5D12 antibody also recognized the full-length recombinant tau protein but not the fusion protein used as an immunogen. The immunoreactivity of the 5D12 antibody with tau was completely abolished when the half-carboxy domain of tau, containing the tubulin-binding repeats, was removed. This study demonstrates that the use of the amino-terminal domain of APP as an immunogen led to the generation of a monoclonal antibody to the half-carboxy domain of tau.