Serum Starvation Induces BACE1 Processing and Secretion.

BACKGROUND: ß-secretase (BACE1) is a type 1 transmembrane protein implicated in Alzheimer's Disease (AD) pathogenesis. Cleavage of Amyloid Precursor Protein (APP), initiated by BACE1 and followed by γ-secretase, leads to the formation of toxic Aß peptides. Increased levels of BACE1 have been detected in the CSF of AD patients compared to age-matched healthy controls indicating that neurodegenerative conditions induce shedding of BACE1. OBJECTIVE: To mimic such conditions, we examined whether serum deprivation stimulates proteolysis-dependent secretion of BACE1. METHOD: Detection of BACE1 secretion in BACE1 overexpressing cells or ADAM10/ADAM17 knockout fibroblasts cultured under serum deprivation conditions, using western blot analysis. RESULTS: We found that serum deprivation of U251 neuroblastoma or HEK293T cells overexpressing BACE1 stimulated secretion of BACE1. Using ADAM10/ADAM17 knockout fibroblasts and inhibitors of both ADAM10 and ADAM17, we obtained data indicating that these proteases are involved in serum-starvation induced shedding of BACE1. This is unexpected since BACE1 is localized mainly in lipid rafts while ADAM10 is localized mainly in nonlipid raft domains. We hypothesized that serum deprivation results in alterations in the lipid composition of the membrane which can alter the localization of ADAM10 and BACE1. In support, we obtained results indicating that extraction of membrane cholesterol following incubation with methyl ß cyclodextrin potentiated the effect of serum deprivation. Secreted BACE1 was also found to be enzymatically active towards immunoprecipiated full length APP. CONCLUSION: Serum starvation induces ADAM10-mediated BACE1 secretion.

Ageing research in Greece.

Ageing research in Greece is well established. Research groups located in universities, research institutes or public hospitals are studying various and complementary aspects of ageing. These research activities include (a) functional analysis of Clusterin/Apolipoprotein J, studies in healthy centenarians and work on protein degradation and the role of proteasome during senescence at the National Hellenic Research Foundation; (b) regulation of cell proliferation and tissue formation, a nationwide study of determinants and markers of successful ageing in Greek centenarians and studies of histone gene expression and acetylation at the National Center for Scientific Research, Demokritos; (c) work on amyloid precursor protein and Presenilin 1 at the University of Athens; (d) oxidative stress-induced DNA damage and the role of oncogenes in senescence at the University of Ioannina; (e) studies in the connective tissue at the University of Patras; (f) proteomic studies at the Biomedical Sciences Research Center Alexander Fleming; (g) work on Caenorhabditis elegans at the Foundation for Research and Technology; (h) the role of ultraviolet radiation in skin ageing at Andreas Sygros Hospital; (i) follow-up studies in healthy elderly at the Athens Home for the Aged; and (j) socio-cultural aspects of ageing at the National School of Public Health. These research activities are well recognized by the international scientific community as it is evident by the group's very good publication records as well as by their direct funding from both European Union and USA. This article summarizes these research activities and discuss future directions and efforts towards the further development of the ageing field in Greece.

Presenilin-1 binds cytoplasmic epithelial cadherin, inhibits cadherin/p120 association, and regulates stability and function of the cadherin/catenin adhesion complex.

Here we show that presenilin-1 (PS1), a protein involved in Alzheimer's disease, binds directly to epithelial cadherin (E-cadherin). This binding is mediated by the large cytoplasmic loop of PS1 and requires the membrane-proximal cytoplasmic sequence 604-615 of mature E-cadherin. This sequence is also required for E-cadherin binding of protein p120, a known regulator of cadherin-mediated cell adhesion. Using wild-type and PS1 knockout cells, we found that increasing PS1 levels suppresses p120/E-cadherin binding, and increasing p120 levels suppresses PS1/E-cadherin binding. Thus PS1 and p120 bind to and mutually compete for cellular E-cadherin. Furthermore, PS1 stimulates E-cadherin binding to beta- and gamma-catenin, promotes cytoskeletal association of the cadherin/catenin complexes, and increases Ca(2+)-dependent cell-cell aggregation. Remarkably, PS1 familial Alzheimer disease mutant DeltaE9 increased neither the levels of cadherin/catenin complexes nor cell aggregation, suggesting that this familial Alzheimer disease mutation interferes with cadherin-based cell-cell adhesion. These data identify PS1 as an E-cadherin-binding protein and a regulator of E-cadherin function in vivo.

Inhibition of neurite outgrowth by familial Alzheimer's disease-linked presenilin-1 mutations.

Two (P117L; M146L) familial Alzheimer's disease (FAD)-causing presenilin-1 (PS1) mutations have been tested fortheir effect in stably transfected mouse neuroblastoma (N2a) cell lines. The P117L mutation is associated with the earliest onset of AD reported so far (24 years), while the M146L is less pathogenic with the onset at about 43 years. Overexpression of wild-type (wt) PS1 gene was associated with the marked increase in the number and the length of neuritic outgrowths accompanied by accumulation of PS1 immunoreactivity in neurites. The highly pathogenic P117L mutation completely suppressed this effect and the pattern of PS1 immunolabeling resembled a cup structure with all immunoreactivity gathered at one pole of the cell. The effect of less pathogenic M146L mutation was similar, but not as pronounced. These findings suggest that one of the normal functions of PS1 may be the control of neurite outgrowth, and the inhibitory effect of two FAD-linked mutations stresses its importance in the cellular mechanism that leads to the development of Alzheimer's disease (AD).

Distinct secretases, a cysteine protease and a serine protease, generate the C termini of amyloid beta-proteins Abeta1-40 and Abeta1-42, respectively.

The carboxy-terminal ends of the 40- and 42-amino acids amyloid beta-protein (Abeta) may be generated by the action of at least two different proteases termed gamma(40)- and gamma(42)-secretase, respectively. To examine the cleavage specificity of the two proteases, we treated amyloid precursor protein (APP)-transfected cell cultures with several dipeptidyl aldehydes including N-benzyloxycarbonyl-Leu-leucinal (Z-LL-CHO) and the newly synthesized N-benzyloxycarbonyl-Val-leucinal (Z-VL-CHO). All dipeptidyl aldehydes tested inhibited production of both Abeta1-40 and Abeta1-42. Changes in the P1 and P2 residues of these aldehydes, however, indicated that the amino acids occupying these positions are important for the efficient inhibition of gamma-secretases. Peptidyl aldehydes inhibit both cysteine and serine proteases, suggesting that the two gamma-secretases belong to one of these mechanistic classes. To differentiate between the two classes of proteases, we treated our cultures with the specific cysteine protease inhibitor E-64d. This agent inhibited production of secreted Abeta1-40, with a concomitant accumulation of its cellular precursor indicating that gamma(40)-secretase is a cysteine protease. In contrast, this treatment increased production of secreted Abeta1-42. No inhibition of Abeta production was observed with the potent calpain inhibitor I (acetyl-Leu-Leu-norleucinal), suggesting that calpain is not involved. Together, these results indicate that gamma(40)-secretase is a cysteine protease distinct from calpain, whereas gamma(42)-secretase may be a serine protease. In addition, the two secretases may compete for the same substrate. Dipeptidyl aldehyde treatment of cultures transfected with APP carrying the Swedish mutation resulted in the accumulation of the beta-secretase C-terminal APP fragment and a decrease of the alpha-secretase C-terminal APP fragment, indicating that this mutation shifts APP cleavage from the alpha-secretase site to the beta-secretase site.

The kunitz protease inhibitor form of the amyloid precursor protein (KPI/APP) inhibits the proneuropeptide processing enzyme prohormone thiol protease (PTP). Colocalization of KPI/APP and PTP in secretory vesicles.

Proteolytic processing of proenkephalin and proneuropeptides is required for the production of active neurotransmitters and peptide hormones. Variations in the extent of proenkephalin processing in vivo suggest involvement of endogenous protease inhibitors. This study demonstrates that "protease nexin 2 (PN2)," the secreted form of the kunitz protease inhibitor (KPI) of the amyloid precursor protein (APP), potently inhibited the proenkephalin processing enzyme known as prohormone thiol protease (PTP), with a Ki,app of 400 nM. Moreover, PTP and PN2 formed SDS-stable complexes that are typical of kunitz protease inhibitor interactions with target proteases. In vivo, KPI/APP (120 kDa), as well as a truncated form of KPI/APP that resembles PN2 in apparent molecular mass (110 kDa), were colocalized with PTP and (Met)enkephalin in secretory vesicles of adrenal medulla (chromaffin granules). KPI/APP (110-120 kDa) was also detected in pituitary secretory vesicles that contain PTP. In chromaffin cells, calcium-dependent secretion of KPI/APP with PTP and (Met)enkephalin demonstrated the colocalization of these components in functional secretory vesicles. These results suggest a role for KPI/APP inhibition of PTP in regulated secretory vesicles. In addition, these results are the first to identify an endogenous protease target of KPI/APP, which is developmentally regulated in aging and Alzheimer's disease.

Presenilin-1 forms complexes with the cadherin/catenin cell-cell adhesion system and is recruited to intercellular and synaptic contacts.

In MDCK cells, presenilin-1 (PS1) accumulates at intercellular contacts where it colocalizes with components of the cadherin-based adherens junctions. PS1 fragments form complexes with E-cadherin, beta-catenin, and alpha-catenin, all components of adherens junctions. In confluent MDCK cells, PS1 forms complexes with cell surface E-cadherin; disruption of Ca(2+)-dependent cell-cell contacts reduces surface PS1 and the levels of PS1-E-cadherin complexes. PS1 overexpression in human kidney cells enhances cell-cell adhesion. Together, these data show that PS1 incorporates into the cadherin/catenin adhesion system and regulates cell-cell adhesion. PS1 concentrates at intercellular contacts in epithelial tissue; in brain, it forms complexes with both E- and N-cadherin and concentrates at synaptic adhesions. That PS1 is a constituent of the cadherin/catenin complex makes that complex a potential target for PS1 FAD mutations.

Enrichment of presenilin 1 peptides in neuronal large dense-core and somatodendritic clathrin-coated vesicles.

Presenilin 1 is an integral membrane protein specifically cleaved to yield an N-terminal and a C-terminal fragment, both membrane-associated. More than 40 presenilin 1 mutations have been linked to early-onset familial Alzheimer disease, although the mechanism by which these mutations induce the Alzheimer disease neuropathology is not clear. Presenilin 1 is expressed predominantly in neurons, suggesting that the familial Alzheimer disease mutants may compromise or change the neuronal function (s) of the wild-type protein. To elucidate the function of this protein, we studied its expression in neuronal vesicular systems using as models the chromaffin granules of the neuroendocrine chromaffin cells and the major categories of brain neuronal vesicles, including the small clear-core synaptic vesicles, the large dense-core vesicles, and the somatodendritic and nerve terminal clathrin-coated vesicles. Both the N- and C-terminal presenilin 1 proteolytic fragments were greatly enriched in chromaffin granule and neuronal large dense-core vesicle membranes, indicating that these fragments are targeted to these vesicles and may regulate the large dense-core vesicle-mediated secretion of neuropeptides and neurotransmitters at synaptic sites. The presenilin 1 fragments were also enriched in the somatodendritic clathrin-coated vesicle membranes, suggesting that they are targeted to the somatodendritic membrane, where they may regulate constitutive secretion and endocytosis. In contrast, these fragments were not enriched in the small clear-core synaptic vesicle or in the nerve terminal clathrin-coated vesicle membranes. Taken together, our data indicate that presenilin 1 proteolytic fragments are targeted to specific populations of neuronal vesicles where they may regulate vesicular function. Although full-length presenilin 1 was present in crude homogenates, it was not detected in any of the vesicles studied, indicating that, unlike the presenilin fragments, full-length protein may not have a vesicular function.

Release of nontransmembrane full-length Alzheimer's amyloid precursor protein from the lumenar surface of chromaffin granule membranes.

We previously demonstrated the presence of a soluble form of full-length Alzheimer's amyloid precursor protein (APP) in the lumen of adrenal medullary chromaffin granules (CG). Furthermore, full-length APP is released from CG membranes in vitro at pH 9.0 by an enzymatic mechanism, sensitive to protease inhibitors [Vassilacopoulou et al. (1995) J. Neurochem. 64, 2140-2146]. In this study, we found that when intact CG were subjected to exogenous trypsin, a fraction of APP was not digested, consistent with an intragranular population of APP. To examine the substrate-product relationship between membrane and soluble full-length APP, we labeled CG transmembrane APP with 3-(trifluoromethyl)-3-(m-[125I]iodophenyl)diazirine ([125I]TID), a lipophilic probe, specific for membrane-spanning domains of proteins. APP released from the membranes at pH 9.0 was not labeled with [125I]TID. In addition, this APP was not biotinylated in intact CG. Combined, the results indicate that APP released from CG membranes derives from a unique nontransmembrane population of membrane-associated APP, located in the lumenal side of CG membranes. Dithiobis(succinimidylpropionate) (DSP) cross-linking indicated that APP in CG is situated in close proximity with other proteins, possibly with APP itself. APP complexes were also detected under nonreducing conditions, without DSP cross-linking. These results, combined with our previous studies, indicate that full-length APP within CG exists as three different populations: (I) transmembrane, (II) membrane-associated/nontransmembrane, and (III) soluble. The existence of nontransmembrane populations suggests that putative gamma-secretase cleavage sites of APP, assumed to be buried within the lipid bilayer, could be accessible to proteolysis in a soluble intravesicular environment.

A novel Polish presenilin-1 mutation (P117L) is associated with familial Alzheimer's disease and leads to death as early as the age of 28 years.

The majority of early-onset familial Alzheimer's disease (FAD) is associated with mutations in the presenilin-1 (PS1) gene. We describe a novel Polish PS1 mutation of Pro117Leu, associated with the earliest average age of onset and death so far reported in a PS-linked, FAD kindred. Human kidney 293 and mouse neuroblastoma N2a cells were stably transfected with wild-type and PS1 P117L. There was a significant increase in the amyloid beta42/40 ratio in the N2a P117L PS1 transfected cells compared with N2a transfected with wild-type PS1. What role PS has in the pathogenesis of AD remains to be determined, however, the severity of the clinical picture associated with this PS1 mutation stresses the importance of presenilin.

Appican expression induces morphological changes in C6 glioma cells and promotes adhesion of neural cells to the extracellular matrix.

Appicans are secreted or cell-associated brain chondroitin sulfate proteoglycans produced by glia cells and containing Alzheimer amyloid precursor protein (APP) as a core protein. Here, we report that rat C6 glioma cells transfected with appican displayed a dramatic change in their phenotypic appearance compared with untransfected cells or cells transfected with APP. Appican-transfected cells lost the round appearance of the untransfected control C6 cells, acquired a flat morphology, and elaborated more processes than control cells. Untransfected, or APP-transfected C6, cells were completely dissociated from their substrate after 40 min of treatment with cell dissociation solution. Under the same conditions, however, <20% of the appican-transfected C6 cells were dissociated from their substrate, suggesting that the appican-transfected glia cells attach more avidly to their substrate than do untransfected or APP transfected control cells. In contrast, appican-transfected fibroblast cells showed no morphological changes and dissociated from their substrate similarly to untransfected fibroblast cells. Extracellular matrix (ECM) prepared from appican-transfected C6 cell cultures contained high levels of appican and was a significantly better substrate for the attachment of C6 cells than ECM from either untransfected or APP-transfected cultures. Furthermore, cell adhesion to ECM was independent of the level of appican expression of the plated cells. ECM from appican-transfected C6 cultures stimulated adhesion of other neural cells including primary astrocytes, Neuro2a neuroblastoma, and PC12 pheochromocytoma, but not fibroblast cells. Conditioned media from appican-transfected C6 cultures failed to promote cell adhesion. Together, these data suggest that secreted appican incorporates into ECM and promotes adhesion of neural cells. Furthermore, our data suggest that the chondroitin sulfate chain engenders APP with novel biological functions.

Melatonin prevents death of neuroblastoma cells exposed to the Alzheimer amyloid peptide.

Studies from several laboratories have generated evidence suggesting that oxidative stress is involved in the pathogenesis of Alzheimer's disease (AD). The finding that the amyloid beta protein (Abeta) has neurotoxic properties and that such effects are, in part, mediated by free radicals has provided insights into mechanisms of cell death in AD and an avenue to explore new therapeutic approaches. In this study we demonstrate that melatonin, a pineal hormone with recently established antioxidant properties, is remarkably effective in preventing death of cultured neuroblastoma cells as well as oxidative damage and intracellular Ca2+ increases induced by a cytotoxic fragment of Abeta. The effects of melatonin were extremely reproducible and corroborated by multiple quantitative methods, including cell viability studies by confocal laser microscopy, electron microscopy, and measurements of intracellular calcium levels. The importance of this finding is that, in contrast to conventional antioxidants, melatonin has a proposed physiological role in the aging process. Secretion levels of this hormone are decreased in aging and more severely reduced in AD. The reported phenomenon may be of therapeutic relevance in AD.

Characterization of appican, the chondroitin sulfate proteoglycan form of the Alzheimer amyloid precursor protein.

In this report we focus on the characterization of appican, the chondroitin sulfate proteoglycan form of amyloid precursor protein (APP), and the role that it and other proteoglycans may play in AD. Appican is expressed by certain transformed cell lines of neural origin, namely C6 cells and N2a neuroblastomas. It is detected in both human and rat brain and in primary cultures is expressed by astrocytes, but not neurons. The core protein of appican has been shown to be an alternatively spliced isoform of APP, lacking exon 15 of the APP gene, originally identified in leukocytes (L-APP). Splicing out of exon 15 results in the joining of exons 14 and 16, and formation of an Asp-Xaa-Ser-Gly consensus sequence for chondroitin sulfate chain attachment to serine 619 of L-APP, which lies 16 amino acids upstream of the A beta peptide sequence. Mutation of this serine residue to an alanine prevented chondroitin sulfate chain addition to the core protein. Levels of appican expression could be regulated by growth conditions independently of APP, suggesting that these molecules may serve distinct physiological roles within the cell. Morphological changes were also observed in both astrocytic and transformed cell cultures, that appeared to reflect changes in levels of appican expression. Preliminary data suggest that appican may be a strong cell adhesion molecule. Transfected C6 glioma cells overexpressing appican remained attached to tissue culture dishes markedly better than either C6 cells over-expressing exon-15 containing APP or WT C6 cells. Appican-enriched extracellular matrix (ECM) was also observed to serve as a much better substrate for attachment of N2a neuroblastomas, pheocromocytoma PC12 cells and primary astrocytes compared to APP enriched ECM.

Mechanism of amyloid beta-peptide (1-42) toxicity in PC12 cells.

Amyloid beta-peptide (A beta) deposition and loss of cholinergic neurons are characteristics of Alzheimer's disease. There is evidence that A beta is neurotoxic. The role of signal transduction pathways on A beta-induced toxicity in PC12 cells was investigated. Our results revealed that A beta-induced arachidonic acid was released in a time-dependent manner. Inhibitors of cyclooxygenase (1 microM indomethacin) and lipooxygenase (100 microM nordihydroguairetic acid) protected PC12 cells against A beta-induced toxicity. These data suggest that A beta toxicity is mediated by activation of the arachidonic acid cascade. Furthermore, protein kinase C activators (phorbol ester and 1-oleyl-2-acetyl-glycerol) and tacrine reversed A beta-induced toxicity. These results suggest that A beta toxicity can be modulated by manipulating signal transduction pathways and may provide the basis for novel therapeutic interventions.

Intracellular cyclic AMP inhibits constitutive and phorbol ester-stimulated secretory cleavage of amyloid precursor protein.

alpha-Secretase cleaves the full-length Alzheimer's amyloid precursor protein (APP) within the amyloid beta peptide sequence, thus precluding amyloid formation. The resultant soluble truncated APP is constitutively secreted. This nonamyloidogenic processing of APP is increased on stimulation of the phospholipase C/protein kinase C pathway by phorbol esters. Here we used C6 cells transfected with APP751 to examine whether the alpha-secretase cleavage is regulated by the adenylate cyclase signal transduction pathway. Forskolin, an activator of adenylate cyclase, inhibited both the constitutive and phorbol ester-stimulated secretion of nexin II (NXII), the secreted product of the alpha-secretase cleavage of APP751. At 1 microM, forskolin inhibited secretion of NXII by approximately 50% without affecting either the intracellular levels of total APP or the secretion of secretory alkaline phosphatase. In contrast, 1,9-dideoxyforskolin, an inactive analogue of forskolin, did not affect secretion of NXII. These results indicated that forskolin specifically inhibited the alpha-secretase cleavage of APP751. Forskolin treatment increased the intracellular concentration of cyclic AMP (cAMP), suggesting that the forskolin effects on APP cleavage may be mediated by cAMP. In support of this suggestion, both dibutyryl cAMP, a cAMP analogue, and isoproterenol, an activator of adenylate cyclase, also inhibited secretion of NXII. These data indicate that forskolin inhibition of the nonamyloidogenic cleavage of APP is mediated by the second messenger cAMP, which together with the protein kinase C signal transduction pathway modulates the secretory cleavage of APP.

Identification and neuron specific expression of the S182/presenilin I protein in human and rodent brains.

Many individuals with familial Alzheimer disease (FAD) have mutations in a gene termed S182 or presenilin I (PS-I). Currently, the PS-I gene product has not been identified and its function remains unknown. Here we report that affinity purified antibodies against the predicted amino acid sequence of the PS-I gene product detected in homogenates of human, mouse, and rat brains a single antigen of approximately 48 kDa. This antigen was also present in immortalized human and mouse neuronal cell cultures. Brain tissue fractionation showed that all PS-I antigen was found in the membrane fraction. In stained tissue sections of mouse central nervous system (CNS), PS-I antigen was found only in neurons throughout brain and spinal cord and was located within cell bodies, axons, and dendrites. Remarkably the relative partition among these three compartments varied dramatically. A striking feature of PS-I expression was its intense concentration in some (but not all) dendrites, at levels substantially above those in the parent perikarya. In most of the cerebrum, PS-I staining in axons was very weak or undetectable. By contrast, many axons in portions of the brainstem and in the spinal cord showed marked PS-I immunoreactivity. Similarly, staining of sections from human temporal cortex showed that PS-I was present mainly in neuronal cell bodies and dendrites. These data show that in the CNS, PS-I is expressed mainly in neurons and suggests that this protein may perform a neuron specific function. The pattern of PS-I expression in the CNS would suggest that the premature neurodegeneration associated with PS-I mutations involves a primary neuronal process rather than a secondary effect of PS-I produced in non-neuronal cells.

Cholinergic agonists stimulate secretion of soluble full-length amyloid precursor protein in neuroendocrine cells.

The Abeta peptide of Alzheimer disease is derived from the proteolytic processing of the amyloid precursor proteins (APP), which are considered type I transmembrane glycoproteins. Recently, however, soluble forms of full-length APP were also detected in several systems including chromaffin granules. In this report we used antisera specific for the cytoplasmic sequence of APP to show that primary bovine chromaffin cells secrete a soluble APP, termed solAPPcyt, of an apparent molecular mass of 130 kDa. This APP was oversecreted from Chinese hamster ovary cells transfected with a full-length APP cDNA indicating that solAPPcyt contained both the transmembrane and Abeta sequence. Deglycosylation of solAPPcyt showed that it contained both N- and O-linked sugars, suggesting that this APP was transported through the endoplasmic reticulum-Golgi pathway. Secretion of solAPPcyt from primary chromatin cells was temperature-, time-, and energy-dependent and was stimulated by cell depolarization in a Ca2+-dependent manner. Cholinergic receptor agonists, including acetylcholine, nicotine, or carbachol, stimulated the rapid secretion of solAPPcyt, a process that was inhibited by cholinergic antagonists. Stimulation of solAPPcyt secretion was paralleled by a stimulation of secretion in catecholamines and chromogranin A, indicating that secretion of solAPPcyt was mediated by chromaffin granule vesicles. Taken together, our results show that release of the potentially amyloidogenic solAPPcyt is an active cellular process mediated by both the constitutive and regulated pathways. solAPPcyt was also detected in human cerebrospinal fluid. Combined with the neuronal physiology of chromaffin cells, our data suggest that cholinergic agonists may stimulate the release of this APP in neuronal synapses where it may exert its biological functions. Moreover, vesicular or secreted solAPPcyt may serve as a soluble precursor of Abeta.

Heat-shock induces abnormalities in the cellular distribution of amyloid precursor protein (APP) and APP fusion proteins.

The heat shock or stress response may play a role in the pathogenesis of Alzheimer's disease. We conducted experiments to visualize microscopically the distribution of wild type amyloid precursor protein (APP) and the behavior of an APP deletion mutant under stress. This was achieved by heat-shock treatment of cells expressing fusion recombinant APP proteins tagged with secreted placental alkaline phosphatase (SEAP). The fusion proteins were cleaved and secreted in a manner similar to wild type APP in unstressed control cells. SEAP activity was detected by cytochemical methods within the cytoplasm in less than 10% of transfected unstressed cells. Heat shocked cells showed a striking difference from the control cells in that over 90% of the stressed cells displayed strong intracytoplasmic SEAP activity occurring with Golgi-like pattern and/or membranous distribution. The effects of heat shock were not due to a peculiar behavior of the clones and depended on the APP portion of the constructs. This study shows miscompartmentalization of APP under stress. Such cellular changes may bear important implications in the processing of APP.