- Email: [email protected]
671 Mutation
Neurobiology of Aging, Vol. 18, No. 6, pp. 573–580, 1997 Copyright © 1997 Elsevier Science Inc. Printed in the USA. All rights reserved 0197-4580/97 $17.00 1 .00
PII:S0197-4580(97)00149-8
Abnormalities in Alzheimer’s Disease Fibroblasts Bearing the APP670/671 Mutation G. E. GIBSON,1* M. VESTLING,† H. ZHANG,* S. SZOLOSI,* D. ALKON,‡ L. LANNFELT,† S. GANDY§ AND R. F. COWBURN† *Cornell University Medical College at Burke Medical Research Institute, White Plains, NY 10605, †Karolinska Institute, Department of Clinical Neuroscience & Family Medicine, Section for Geriatric Medicine, Novum, KFC. S-141 86, Huddinge, Sweden, ‡National Institute of Neurological Diseases and Stroke, Bethesda, MD, 20892 §Cornell University Medical College, New York, NY 10021 Received 26 February 1997; Revised 2 October 1997; Accepted 6 October 1997 GIBSON, G., M. VESTLING, H. ZHANG, S. SZOLOSI, D. ALKON, L. LANNFELT, S. GANDY, AND R. F. COWBURN. Abnormalities in Alzheimer’s disease fibroblasts bearing the APP670/671 mutation. NEUROBIOL AGING 18(6) 573–580, 1997.—Abnormalities in cultured fibroblasts from familial Alzheimer’s Disease (FAD) cases uniquely enable the determination of how gene defects alter cell biology in living tissue from affected individuals. The current study focused on measures of calcium regulation and oxidative metabolism in fibroblast lines from controls and FAD individuals with the Swedish APP670/671 mutation. Bombesin-induced elevations in calcium in APP670/671 mutation-bearing lines were reduced by 40% (p , 0.05), a striking contrast to the 100% increase seen in sporadic AD and presenilin-1 (PS1) mutation-bearing cells in previously published studies. The APP670/671 mutation-bearing lines did not exhibit the exaggerated 4-bromo-A23187 releasable pool of calcium following 10 nM bradykinin, the enhanced sensitivity of calcium stores to low concentrations of bradykinin, nor the reduced activity of a-ketoglutarate dehydrogenase previously reported in cells from sporadic AD and mutant PS1 FAD. Thus, an altered regulation of internal calcium stores is common to all AD lines, but the calcium pool affected and the polarity of the alteration varies, apparently in association with particular gene mutations. Comparison of signal transduction in cell lines from multiple, genetically characterized AD families will allow testing of the hypothesis that these various pathogenic FAD abnormalities that lead to AD converge at the level of abnormal signal transduction. © 1997 Elsevier Science Inc. Alzheimer’s disease Calcium Fibroblasts Bombesin Bradykinin Signal transduction
Amyloid precursor protein
a-ketoglutarate dehydrogenase
to examine fibroblasts from individuals with the APP670/671 mutation for the possible signaling abnormalities that have been reported in fibroblasts from other AD populations. Several abnormalities have been reported in cultured fibroblasts from sporadic and presenilin-1 (PS1) mutation-bearing AD subjects (for reviews see 9,23). These include alterations in calcium (24,35), oxidative metabolism (36), cyclic AMP (21), the phosphatidylinositide cascade including protein kinase C (3,12,22, 28), G-proteins (27), K1 channels (6,7), and amyloid-b-peptide production (4,25,32,41). The current studies compare these previously reported AD-related differences in measures of calcium regulation and energy metabolism to these same variables in APP670/671 mutation-bearing cells. Calcium can both activate (17) and inactivate (29) key enzymes of energy metabolism. For example, calcium inactivates a-ketoglutarate dehydrogenase (KGDHC), an enzyme that is strikingly reduced in AD brains (10). Oxidative processes are altered in AD fibroblasts (36), and mitochondria isolated from fibroblasts from Alzheimer patients are more sensitive to free radicals than those isolated from control
INTRODUCTION
ABNORMAL signal transduction systems have been implicated in the pathophysiology of Alzheimer’s disease (AD), but their precise role has been difficult to establish. Cultured fibroblasts offer many advantages for the study of signal transduction in AD. The signals are more robust in these living cells than in postmortem material (such as brain) and the results are not complicated by agonal or postmortem events. Also, any effects of the patients’ diets or drugs are diluted millions of times. Fibroblasts also allow comparison of cell biological systems and pathways, such as signal transduction, in cells from AD patients with specific known genetic abnormalities. One of the most extensively investigated families with a genetic defect leading to AD is the Swedish AD family with the amyloid precursor protein (APP) 670/671 mutation (30,34). All individuals with this defect invariably develop AD (1,31). Fibroblasts from these patients produce excess amyloid b-peptide compared to cells from control relatives (4,15,25,40) and this excess amyloid-b-peptide production presumably promotes cerebral deposition of amyloid. The purpose of the current study was
1 Address correspondence to: G. E. Gibson, Cornell University Medical College, Burke Medical Research Institute, 785 Mamaroneck Avenue, White Plains, NY 10605, email: [email protected]
573
574
GIBSON ET AL.
fibroblasts (28). Amyloid-b-peptide produces free radicals in vitro (33), so it is possible that the enhanced amyloid-b-peptide production in patients with the APP670/671 mutation may alter mitochondria and the activities of the key mitochondrial enzyme KGDHC, which, in turn, would alter mitochondrial calcium regulation. Thus, the activity of KGDHC was also examined in cells from patients with the APP670/671 mutation. MATERIALS AND METHODS
Materials Fura-2 AM and 4-bromo-A23187 were from Molecular Probes (Eugene, OR). Dulbecco’s Modified Eagle’s Medium (GIBCO, Grand Island, NY) and fetal bovine serum, HEPES, EGTA, a-ketoglutaric acid, leupeptin, Coenzyme A, and nicotinamide adenine dinucleotide (Sigma Chemical Co., St. Louis, MO) were from the indicated companies. Bovine albumin fraction V was from United States Biochemical Corp. (Cleveland, OH). The reagents for the Bradford assay were from Bio-Rad (Hercules, CA). Fibroblasts Fibroblast cell lines were established at the Department of Geriatric Medicine of the Karolinska Institute from skin biopsies taken from affected and nonaffected individuals in the Swedish family with the APP670/671 mutation. Six lines were established from individuals with the APP670/671 mutation, of which four lines were from symptomatic carriers and two lines from asymptomatic individuals. For the analysis in this paper, the six lines with the mutation were combined because of the high penetrance of AD in individuals with this mutation. The average age (mean 6 SD) of the individuals with the mutation was 53 6 8 years. Six lines were also established from nonaffected individuals who did not have the APP670/671 mutation who were within the same family. The average age of these control donors was 60 6 11 years. Cells were maintained under strictly controlled conditions, according to the procedures of Cristofalo and Charpentier (5). Antibiotics were not used because these are known to alter signal transduction systems including those mediated by cyclic adenosine monophosphate (AMP) (19,20), calcium (38), and protein kinase C (16). In brief, cultures were grown in Dulbecco’s minimal essential medium supplemented with 10% fetal bovine serum at 37°C in a humidified atmosphere with 5% CO2 in air. Cells were routinely seeded at a density of 104/cm2 and subcultured every 7 days upon reaching confluence. The cells for each assay were treated exactly as in the original reports that showed AD-related changes for that variable, since our goal was to test whether previously reported abnormalities in PS1 mutant cells and sporadic AD cells could be documented in cells from the APP670/671 mutation family. Thus, in those cases in which measurements had not been made in this laboratory, previously reported abnormalities were first replicated in PS1 mutant cells before commencing studies in cells with the APP670/ 671 mutation. The PS1 cell lines that were examined have the Glu246 mutation. This mutation may lead to different changes than other PS1 mutations (46). Enzyme Activity Measurements Following subculture, cells were incubated for 3 weeks to ensure that all cells reached the G0 stationary phase. Cells were collected by scraping in 10 mM sodium phosphate (pH 7.2), 150 mM NaCl, and 1 mM EGTA, and centrifuged at 800 3 g for 10 min. After washing with the same buffer twice, cell pellets were
suspended by manual homogenization in 50 mM Tris-HCl (pH 7.2), 1 mM dithiothreitol, 0.2 mM EGTA, 0.4% Triton X-100, and 50 mM leupeptin. The freshly prepared lysate was spun for 3 s in a microcentrifuge, and the supernatant was used for spectrophotometric assays of KGDHC at 30 6 1°C (42). No activities were detectable in the pellet, but both supernatant and pellet were used to assess total protein in each flask by the Bradford assay. Every cell line was tested on five different days with triplicate measures each day. Response to Low Concentrations of Bradykinin Confluent fibroblasts were subcultured on 25-mm round glass cover slips at a density of 500/cm2 and allowed to grow for 7 days. Cells were incubated in balanced salt solution (BSS) (140 mM NaCl, 5 mM KCl, 1.5 mM MgCl2, 5 mM glucose, 10 mM HEPES, and 2.5 mM CaCl2) with 2 mM Fura-2 AM at room temperature for 1 h and then rinsed three times with calcium-free BSS. The coverslip was transferred to a Teflon Leiden coverslip dish and mounted on the microscope stage of an Olympus IMT-2 inverted microscope. The dual excitation wavelengths were achieved by passing the light generated with a Xenon 75-watt arc lamp through dual monochromaters at wavelengths of 350 and 378 nm (bandpass of 2 nm). Images were taken with a Hamamatsu 2400C SIT camera and converted to 8-bit images with specialized computer boards from Photon Technology International. The ratio of images after excitation at 350 and 378 nm was converted to [Ca21]i with Photon Technology International Imaging software (Princeton, NJ) using a viscosity constant of 0.7 (13,37). Bradykinin (BK) (0.2 nM) was made separately for each measurement by adding 10 mL of 20 nM stock (in 0.2% dimethyl sulfoxide) into 1 mL of BSS and sonicating in a cleanser bath for 40 s immediately before the measurement. Experiments were performed at room temperature. Fibroblasts were maintained on the microscope stage in 1 mL of BSS and basal [Ca21]i was measured for 1 min. One mL of 0.2 nM BK was then added to the cell dish to make the final BK 0.1 nM. The [Ca21]i was then measured for another 3 min. Cells with increases of [Ca21]i more than 100% of their basal values were regarded as responding cells (18,24). Response to Bombesin These images were collected and the analysis was done as described above except that the agonist was bombesin (1 mM) and measurements were done in calcium-free BSS (BSS with 0.1 mM CaCl2 and 1 mM EGTA to give 7.18 nM calcium) (24). Evaluation of the 4-bromo-A23187 Releasable Pool of Calcium Following 10 nM BK Confluent cells were subcultured on 25-mm round glass cover slips at a density of 1 3 104 cells/cm2. Fibroblasts were loaded with 2 mM fura-2AM in incubation media for 1 h at room temperature, which prevents sequestration of the dye in the organelles. The incubation media was Dulbecco’s Modified Eagles Medium without phenol red with 20 mM HEPES and 1% bovine serum albumin (pH 7.4 at room temperature). Following this incubation, the coverslip was rinsed three times with testing media without fura-2 (37°C). The calcium-free testing media was similar to the incubation media but with the addition of 2.5 mM EGTA (59 nM calcium; pH 7.4 at 37°C). The cover slips were secured in a Leiden coverslip dish and placed on a temperature-controlled micro-incubator with 2 mL testing media. The micro-incubator was mounted onto the stage of an Olympus inverted microscope equipped with a Nikon 40 3 UV fluor objective. To minimize the temperature gradient between different parts of the coverslip, the
ALZHEIMER FIBROBLASTS
575 Response to Low Concentrations of BK
FIG. 1. KGDHC activity in controls and cells with the APP670/671 mutation. Values for each cell line are the mean of each cell line measured three times per day on 4 –5 different days. X and error bars represent mean 6 SEM of all six lines.
micro-incubator was continuously gassed with 5% CO2/95% air at a flow rate of three standard cubic feet per hour (11). These studies utilized a photomultiplier tube to record the response of a whole field with three to six cells. This permitted complete definition of the temporal response. Following a 3-min equilibration period, calcium measurement was initiated by alternately exciting the cells with light at 350 and 380 nm five times per second using a Delta scan, and light emission was measured at 510 nm (Photon Technology International; S. Brunswick, NJ). The cells were treated with BK and 4-bromo-A23187 by removing about 90% of the testing medium and adding it to a cup containing BK or 4-bromo-A23187. The solution was mixed by repeated pipetting and then reapplied to the cells. These manipulations took less than 10 s. The cells were tested on three separate days, with three samples tested on each day. [Ca21]i was calculated by the method of Grynkewicz et al. (13) after correction for background fluorescence.
The goal of the current studies was to test if we could confirm previously reported changes in the presenilin-1 lines in a limited number of lines (18) and to then compare these results with cells from APP670/671 mutation-bearing lines. Initial studies confirmed the enhanced calcium response of fibroblasts from familial AD (FAD) individuals with PS1 mutations to stimulation by low BK concentrations (Fig. 2). Thirty-one of 367 (8.4%) fibroblasts measured in three lines from AD patients with PS1 mutations (AG6840, AG6848, AG8170) responded to low concentrations of BK. Only 1 of 306 fibroblasts (0.3%) measured in three agematched control lines (AG7657-Canadian escapee, AG9878, GM4260) responded to 0.1 nM BK. The response to low BK concentrations did not distinguish individual FAD APP670/671 cell lines from controls. Previous results showed that a response of greater than 1.5% of the cells to low concentrations of BK distinguished sporadic AD and PS1 FAD cell lines from controls (18). In cells with the APP670/671 mutation, 11 of 588 (1.87%) tested cells from six lines responded to 0.1 nM BK, whereas only 2 of 612 (0.33%) control cells responded. Although examination of the combined data suggests that the 1.5% criteria applies to these APP670/671 cell lines, this standard is not applicable on an individual cell line basis. Only two of the APP670/671 mutationbearing cell lines met this criteria while four of the six lines did not (Fig. 2). Five of the control lines had no response, whereas in one line, 1.7% of the cells responded. Thus, the differences that occurred in fibroblasts from the sporadic AD and the PS1 mutant FAD individuals (previous studies) were not apparent in cells bearing the APP670/671 mutation. Response to Bombesin The response of [Ca21]i to bombesin (1 mM) was altered in fibroblasts with the APP670/671 mutation (Fig. 3). Studies in three
Statistics Analysis of variance was done with SPSS (Chicago, IL). Given the high penetrance in this family (1), data from the symptomatic and asymptomatic mutation-bearers was grouped and compared to the non-mutation-bearing cell lines. RESULTS
KGDHC Activity in Fibroblasts with the APP670/671 Mutation Diminished KGDHC activities have been reported in fibroblasts (2,10,42) and brain (10) from AD patients. The mean activity (mean 6 SEM) of KGDHC for the control group was 10.06 6 2.26 nmol/min per mg protein (6 lines). The mean activity of KGDHC for the APP670/671 mutation group was 10.23 6 2.53 nmol/min per mg protein (6 lines). The individual values for each line are presented in Figure 1. The results demonstrate that KGDHC activities did not differ between the control and mutationbearing lines.
FIG. 2. Bradykinin-sensitive calcium stores in fibroblasts from control and AD individuals with APP670/671 or PS1 (GLU246) mutations. Values for each line are the results from .100 cells per line measured on 3 dishes a day on 2–3 days. X and error bars in each group represent mean 6 SEM of all the lines in that group. Circles are control values and squares are values of AD with gene mutations. Control 1 (n 5 6) and APP670/671 mutation-bearing lines (n 5 6) are from the Swedish kindred. Control 2 (n 5 3) and Presenilin-1 (GLU246) mutation-bearing lines (n 5 3) are from Coriell Institute for Medical Research in Camden, NJ.
576
GIBSON ET AL.
FIG. 3. Bombesin-induced changes in [Ca21]i in fibroblasts from controls and individuals with the APP670/671 mutation. Values are means 6 SEM (n 5 6 for each). The two groups vary significantly from each other (p 5 0.002). Three dishes/day were examined on two different days for each cell line. More than 100 cells were evaluated for each cell line.
AD (AG6840, AG6848, AG8170) and three control (AG7657, AG9878, GM4260) fibroblast lines confirmed that the responses to 1 mM bombesin were exaggerated in AD fibroblasts. The line-toline variation generally agreed with the results from the group that initially reported the bombesin-induced change with AD (24). Mean basal [Ca21]i (mean 6 SEM) was identical in the control and APP670/671 mutation-bearing cells (control 100.8 6 19.9 vs. mutation 99.9 6 10.2 nM). However, the peak response to bombesin was reduced by 40% in fibroblasts with the APP670/671 mutation (control 731.2 6 82.9 vs. mutation 530.7 6 85.8 nM; p 5 0.002). The mean integrated area over basal of the bombesin response was reduced by 36% in the APP670/671 mutationbearing fibroblasts (control 27088 6 10250 vs. mutation 17228 6 7531; p 5 0.044). Some cells did not respond to bombesin. The percentage of cells responding was reduced by 29% in the APP670/671 mutation-bearing fibroblasts (control 69 6 18% vs. mutation 49 6 17%; p 5 0.041). When only responders were analyzed, no differences were apparent between the fibroblasts with or without the APP670/671 mutation. 4-Bromo-A23187 Releasable Pool of Calcium Following 10 nM BK The 4-bromo-A23187 releasable pool of calcium following 10 nM bradykinin did not vary between the control and mutation bearing lines (Fig. 4). Treatment with 10 nM BK increases [Ca21]i by release of calcium from internal stores and blocks any changes in [Ca21]i due to subsequent stimulation with 10 nM BK, thapsigargin, or bombesin (1 mM) (11). However, subsequent addition of 4-bromo-A23187 still increases [Ca21]i. A typical temporal response is shown in Figure 4a. The majority of cellular calcium was released by BK, but subsequent addition of 4-bromoA23187 released additional calcium stores. This calcium store did not vary between the control lines and those with the APP670/671
mutation at either the peak or at 1, 2, or 3 min after addition of 4-bromo-A23187 (Fig. 4b). Nor did the integrated area above basal vary between the mutation-bearing cells and controls. DISCUSSION
Studies that preceded the discovery of the presenilins reported decreases in KGDHC activities that ranged from 23 (nonsignificant) (10) to 40% (42) in fibroblasts from patients now known to have presenilin-1 mutations, other familial mutations, and sporadic AD. A recent study with a larger number of subjects showed normal activity of KGDHC in fibroblasts from a presenilin-1 family, and a 40% reduction in fibroblasts from sporadic AD patients (2). The results of the present study using APP670/671 mutation-bearing cell lines, when taken together with data from the above studies on sporadic AD and PS-1 mutation-bearing AD fibroblasts, would suggest that reduced KGDHC activity is a feature of sporadic but not familial AD. The fact that APP670/671 and PS-1 mutation-bearing cell lines have an over production of the amyloid-b-peptide in common (40) would suggest further that reduced KGDHC activities are not a consequence of over exposure to excess levels of Ab peptide. This notion is further supported by previous studies showing that reduced KGDHC in AD brain occurs both in affected regions with a high density of amyloidcontaining plaques, as well as in relatively spared regions (10). Following depletion of internal calcium stores by addition of 10 nM BK, neither BK nor bombesin increases [Ca21]i, whereas calcium can still be released from internal stores by 4-bromoA23187. The size of this 4-bromo-A23187 releasable pool of calcium following 10 nM BK is constant over BK concentrations ranging from 1–1000 nM. Previous studies suggest this may be a mitochondrial pool of calcium (11) and that this store of calcium is increased in the lines bearing the presenilin GLU246 mutation and in fibroblasts from sporadic Alzheimer patients (11). Whether
ALZHEIMER FIBROBLASTS
577
FIG. 4. A) Characterization of the 4-bromo-A23187 releasable pool of calcium following 10 nM BK. All incubations were performed in calcium-free media. A basal reading was obtained for 1 min. BK (10 nM) was added at 1 min (60 s: left panel). 4-bromo-A23187 (20 mM) was added at 4 min (240 s: right panel). Note: The 4-bromo-A23187 y axis scale was enlarged for clarity. B) Internal calcium stores in fibroblasts from AD patients with the APP670/671 mutation. Cells were first stimulated with BK and, 3 min later, 4-bromo-A23187 was added. The vertical axis is the increase over [Ca21]i at 3 min after BK stimulation. Values are means 6 SEM of six control cell lines and six mutation-bearing lines tested three times per day on three different days.
this latter change reflects a uniform response or a few cells that have greatly exaggerated responses cannot be estimated by these methods. The current results show that this compartment is not altered in cells bearing the APP670/671 mutation. This further supports our interpretation of the KGDHC activity data that suggest mitochondria are unaffected in APP670/671 mutantbearing fibroblasts. AD-related changes in the phosphatidylinositide cascade have been shown in fibroblasts from sporadic AD and PS1 mutant FAD individuals by several methods. Multiple investigators report diminished protein kinase C levels and activities in AD fibroblasts (3,12,44), although not in cells bearing the APP670/671 mutation (45). AD fibroblasts also have a heightened sensitivity to BK, an activator of the phosphatidylinositide cascade, as shown by an exaggerated inositol 1,4,5-trisphosphate (IP3) response (22) and an
enhanced response of [Ca21]i to low concentrations of BK (18). The latter study suggests that the response induced by 0.1 nM BK has a higher degree of specificity for identifying AD cells than tetraethylammonium- or bombesin-induced [Ca21]i responses (ref. 7 and 24, respectively). In 13 out of 14 AD cell lines, greater than 1.5% of the cells responded, whereas none of 11 controls showed this response (18). The differences in the response of sporadic AD cells to BK does not occur at higher BK concentrations (100 nM). In two other FAD families, the response to 100 nM BK was depressed, variable, and cell cycle dependent (43). In view of this confusion in the literature, we began the present studies by first confirming the heightened sensitivity of cells from sporadic AD cases and PS1 mutant FAD individuals in order to exclude the possibility that any technical differences existed between this study and the original report (18). Although the methodology and
578
GIBSON ET AL.
results were identical to earlier studies, cells from only two of six lines with the APP670/671 mutation have an enhanced calcium sensitivity to BK. Thus, the sensitivity of internal BK responsive calcium stores is not a general property of Alzheimer cells. Of the various calcium pools that were examined, the only one that showed a mutation-related difference was the calcium compartment that was sensitive to bombesin. Bombesin activates membrane-bound phospholipase C to release IP3, which subsequently leads to release of calcium from the endoplasmic reticulum. In contrast to results with other AD fibroblast lines in which this pool was increased (24), this compartment of calcium was lower in the APP670/671 mutation-bearing cells. The current results suggest an interaction between amyloid-b-peptide and the bombesin-sensitive calcium pools. A cell line by cell line comparison of the amyloid-b-peptide production in these same cells (4,25) reveals a significant negative correlation to the percent of cells responding to bombesin in the current study (r 5 20.74; p 5 0.022). Although direct experiments with amyloid-bpeptide suggest that external addition of synthetic amyloid-bpeptide does not alter bombesin-sensitive calcium stores (7), endogenous production over long times may lead to a very different response. Abnormalities in the receptor-mediated release of internal calcium pools are a consistent finding in cells from Alzheimer patients, but the precise nature of how these pools are altered varies between the familial and sporadic AD populations. BK and bombesin act at the cell membrane receptors to release IP3. Both agonists are generally regarded to release calcium from the same IP3-sensitive internal store. However, the measured calcium response depends upon the bombesin/BK receptors, phosphatidylinositide turnover, G proteins, Ca21-ATPase activities, the IP3 receptors, and the size of the calcium pool. The reduced calcium response after high agonist concentrations (1 mM bombesin) may be due to a diminished activity of any one of these steps or a smaller internal calcium pool in the cell lines bearing the APP670/ 671 mutation. Distinguishing which of these is involved requires further experiments. This diminished response contrasts with cells from presenilin-1 families or sporadic AD patients in which the calcium response is increased. Interestingly, transfection of the PS-1 mutation into PC12 cells also increases the size of this calcium store as estimated by the response to high agonist concentration (100 nM bradykinin) (14). The sensitivity of IP3 calcium store to low concentrations of BK was similar in fibroblasts from control and the APP670/671 mutation-bearing cells. By comparison, cells from sporadic AD subjects or presenilin-1 AD patients have increased sensitivity to low BK concentrations. This suggests that the fibroblasts from the APP670/671 mutation-bearing cells do not have increased BK receptors as has been shown for these other cell lines (22). These results demonstrate that the APP670/671 mutation does not cause changes in the signal transduction variables similar to those reported to be altered in fibroblasts from other AD kindreds or in sporadic AD cases. Although controversy surrounds many
studies of AD fibroblasts, appropriate attention to detail has lead to a consensus about many of the changes (9). The results in this paper reveal that changes in signal transduction persist in cultured fibroblasts, but that the AD-related differences vary with different gene defects. The only abnormality that has been previously reported in the APP670/671 mutation-bearing cells is that of altered amyloid-b-peptide production (4,25). The additional change reported in the current studies is that of a reduction in the bomesin-induced calcium increases. In addition to the lack of changes reported in this manuscript, the activity of protein kinase C (45) and the b-adrenergic-stimulated formation of cyclic AMP (46) are also normal in fibroblasts with the APP670/671 mutation. To date, no change nor group of changes has been consistently reported in fibroblasts from all of the different AD families or in sporadic AD cases. These results and those in the parallel study on b-adrenergic-stimulated cyclic AMP production (46) and protein kinase C levels and activity (45) demonstrate that fibroblast lines bearing the APP670/671 mutation show changes in signal transduction that are quite different from those seen in sporadic AD cases or in PS-1 mutant FAD individuals. It will now be important to investigate the molecular events that link PS1 and PS2 mutations to abnormal calcium and oxidation, and to determine whether presenilin/calcium/oxidative interactions are “upstream,” “downstream,” or “in parallel” to key features of AD: namely amyloidosis and neurodegeneration. Multiple factors, including various gene mutations, can cause AD. At some level, these different abnormalities converge to produce the symptoms and pathology that are common to AD regardless of etiology. The robust changes reported in fibroblasts from AD subjects without APP mutations that were not present in the APP670/671 mutation-bearing lines (reduced KGDHC activity and abnormal calcium regulation) would appear to precede and perhaps initiate alterations in amyloid-b-peptide in those patients. Impaired mitochondrial function leads to production of amyloidgenic fragments (8) and brains of patients that died with mitochondrial gene mutations have plaques (26). Increasing cellular calcium by addition of 4-bromo-A23187 also elevates production of amyloid-b-peptide (39). The differences in calcium regulation in the APP670/671 mutation-bearing cells may be caused by the exaggerated amyloid b-peptide production that is known to occur in these fibroblasts (25) because amyloid-b-peptide is known to lead to calcium dysregulation (31). ACKNOWLEDGMENTS
The authors are grateful to Craig Grossman for making many of the a-ketoglutarate dehydrogenase activity measurements, to Karen Scheffold for her excellent help with the tissue culture, and to Dr. Rene Etcheberrigaray (Institute for Cognitive and Computational Sciences, Georgetown University Medical Center) for helping us to establish the methods for the experiments with low concentrations of BK and bombesin. The work was supported by National Institute of Aging Grants AG11921 (GG), AG13780 (SG), and AG09464 (SG). The Åke Wiberg and Gamla Tja¨narinnor Foundations are also thanked for financial support.
REFERENCES 1. Axelman, K.; Basun, H.; Winblad, B.; Lannfelt, L. A large Swedish family with Alzheimer’s disease with a codon 670/671 amyloid precursor protein mutation: A clinical and genealogical investigation. Arch. Neurol. 51:1193–1197; 1994. 2. Blass, J. P.; Sheu, K. -F. R.; Piaciantini, S.; Sorbi, S. Inherent abnormalities in oxidative metabolism in AD: Interaction with vascular abnormalities. In: de la Torre, J. C.; Hachinski, V. eds. Cerebro-
vasular Pathology in Alzheimer’s Disease. New York: New York Academy of Sciences; 1997 (In press). 3. Bruel, A. G.; Cherqui, G.; Columelii, S.; Margelin, D.; Roudier, M.; Sinet, P.; Prieur, M.; Perignon, J.; Delarab, J. Reduced protein kinase C activity in sporadic Alzheimer’s disease fibroblasts. Neurosci. Lett. 133:89 –92; 1991. 4. Citron, M.; Vigo-Pelfrey, C.; Teplow, D. B.; Miller, C.; Schenk, D.;
ALZHEIMER FIBROBLASTS
5. 6.
7. 8.
9.
10.
11. 12.
13. 14.
15.
16. 17. 18.
19.
20. 21. 22.
23.
Johnston, J.; Winblad, B.; Venizelos, N.; Lannfelt, L.; Selkoe, D. J. Excessive production of amyloid beta-protein by peripheral cells of symptomatic and presymptomatic patients carrying the Swedish familial Alzheimer disease mutation. Proc. Natl. Acad. Sci. USA 91:11993– 11997; 1994. Cristofalo, V. J.; Charpentier, R. A standard procedure for cultivating human diploid fibroblast like cells to study cellular aging. J. Tiss. Cult. Meth. 3&4:117–121; 1980. Etcheberrigaray, R.; Ito, E.; Oka, K.; Tofel-Grehl, B.; Gibson, G. E.; Alkon, D. L. Potassium channel dysfunction in fibroblasts identifies patients with Alzheimer’s disease. Proc. Natl. Acad. Sci. USA 90: 8209 – 8213; 1993. Etcheberrigaray, R.; Ito, E.; Oka, K.; Kim, C. S.; Alkon, D. L. Soluble beta-amyloid induction of Alzheimer’s phenotype for human fibroblast K1 channels. Science 264:276 –279; 1994. Gabuzda, D.; Busciglio, J.; Chen, L. B.; Matsudaira, P.; Yankner, B. A. Inhibition of energy metabolism alters the processing of amyloid precursor protein and induces a potentially amyloidogenic derivative. J. Biol. Chem. 269:13623–13628; 1994. Gibson, G.; Martins, R.; Blass, J.; Gandy, S. Altered oxidation and signal transduction systems in fibroblasts from Alzheimer patients. In: Mailer. D.; Gispen, W., eds. The Current Status of the Calcium Hypothesis of Brain Aging and Alzheimer’s Disease. Life Science 59:477– 490; 1996. Gibson, G. E.; Sheu, K-F. R.; Blass, J. P.; Baker, A.; Carlson, K. C.; Harding, B.; Perrino, P. Reduced activities of thiamine-dependent enzymes in brains and peripheral tissues of Alzheimer patients. Arch. Neurol. 45:836 – 840; 1988. Gibson, G. E.; Toral-Barza, L.; Tofel-Grehl, B.; Szolosi, S.; Zhang, H. Calcium stores in cultured fibroblasts and their changes with Alzheimer’s disease. Biochim. Biophys. Acta 1316:71–77; 1996. Govoni, S.; Bergamaschi, S.; Racchi, M.; Battaini, F.; Binetti, A.; Bianchetti, A.; Trabucchi, M. Cytosol protein kinase C down regulation in fibroblasts Alzheimer’s disease patients. Neurol. 43:1195– 1199; 1993. Grynkiewicz, G.; Poenie, M.; Tsien, R. Y. A new generation of Ca21 indicators with greatly improved fluorescence properties. J. Biol. Chem. 260:3440 –3450; 1985. Guo, Q.; Katsutosh, F.; Sopher, B. L.; Pham, D. G.; Robinson, N.; Martin, G. M.; Mattson, M. P. Alzheimer’ PS-1 mutation perturbs calcium homeostasis and sensitizes PC12 cells to death induced by amyloid b-peptide. NeuroReport 8:379 –383; 1996. Haass, C.; Lemere, C. A.; Capell, A.; Citron, M.; Seubert, P.; Schenk, D.; Lannfelt, L.; Selkoe, D. J. The Swedish mutation causes earlyonset Alzheimer’s disease by beta-secretase cleavage within the secretory pathway. Nature Med. 1:1291–1296; 1995. Hagiwara, M.; Inagaki, M.; Kanamura, K.; Ohta, H.; Hidaka, H. Inhibitory effects of aminoglycosides on renal protein phosphorylation by protein kinase C. J. Pharmacol. Exp. Ther. 244:355–360; 1988. Hansford, R. G.; Castro, F. Role of Ca21 in pyruvate dehydrogenase interconversion in brain mitochondria and synaptosomes. Biochem. J. 227:129 –136; 1985. Hirashima, N.; Etcheberrigaray, R.; Battaini, R.; Bergamaschi, S.; Racchi, M.; Battaini, F.; Binetti, G.; Govoni, S.; Alkon, D. L. Calcium responses in human fibroblasts: a diagnostic molecular profile for Alzheimer’s disease. Neurobiol. Aging 17:549 –555; 1996. Hori, R.; Yamamoto, K.; Saito, H.; Kohno, M.; Inui, K. Effect of aminoglycoside antibiotics on cellular functions of kidney epithelial cell line (LLC-PK1): A model system for aminoglycoside nephrotoxicity. J. Pharmacol. Exp. Ther. 230:724 –728; 1984. Howard, M.; Frizzell, R. A.; Bedwell, D. M. Aminoglycoside antibiotics restore CFTR function by overcoming premature stop mutations. Nature Med. 2:467– 469; 1996. Huang, H. -M.; Gibson, G. E. Altered b-adrenergic receptor stimulated cAMP formation in cultured skin fibroblasts from Alzheimer donors. J. Biol. Chem. 268:14616 –1462; 1993. Huang, H. -M.; Lin, T. A.; Sun, G. Y.; Gibson, G. E. Increased insitol(1,4,5)trisphosphate accumulation correlates with an upregulation of bradykinin receptors in Alzheimer’s disease. J. Neurochem. 64:761–766; 1995. Huang, H. -M.; Martins, R.; Gandy, S.; Etcheberrigaray, R.; Ito, E.; Alkon, D. L.; Blass, J. P. Gibson, G. E. The use of cultured fibroblasts
579
24.
25.
26.
27.
28. 29. 30.
31.
32.
33.
34.
35. 36. 37. 38. 39. 40.
41.
in elucidating the pathophysiology and diagnosis of Alzheimer’s disease. In: Disterhoft, J. F.; Gispen, W. H.; Traber, J.; Khachaturian, Z. S., eds. Calcium Hypothesis of Aging and Dementia. New York: New York Academy of Science; 1994:225–244. Ito, I.; Oka, K.; Etcheberrigaray, R.; Nelson, T. J.; Mc Phie, D.; Tofel-Grehl, B.; Gibson, G. E.; Alkon, D. L. Internal Ca21 mobilization is altered in fibroblasts with Alzheimer disease. Proc. Natl. Acad. Sci. USA. 91:534 –538; 1994. Johnston, J. A.; Cowburn, R. F.; Norgren, S.; Wiehager, B.; Venizelos, N.; Winblad, B.; Vigo-Pelfrey, C.; Schenk, D.; Lannfelt, L.; O’Neill, C. Increased beta-amyloid release and levels of amyloid precursor protein (APP) in fibroblast cell lines from family members with the Swedish Alzheimer’s disease APP670/671 mutation. FEBS Letters 354:274 –278; 1994. Kaido, M.; Fugimura, H.; Soga, F.; Toyooka, S. K.; Yoshikawa, H.; Nishimura, T.; Higashi, T.; Inui, K.; Imanishi, H.; Yorifuji, S.; Yanagihara, T. Alzheimer-type pathology in a patient with mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes (MELAS). Acta Neuropathol. 92:312–318; 1996. Kim, C. S.; Han, Y. F.; Etcheberrigaray, R.; Nelson, T. J.; Olds, J. L.; Yoshioka, T.; Alkon, D. L. Alzheimer and beta-amyloid-treated fibroblasts demonstrate a decrease in a memory-associated GTPbinding protein, Cp20. Proc. Natl. Acad. Sci. USA. 92:3060 –3064; 1995. Kumar, U.; Dunlop, D. M.; Richardson, J. S. Mitochondria from Alzheimer’s fibroblasts show decreased uptake of calcium and increased sensitivity to free radicals. Life Science 54:1855–1860; 1994. Lai, J. C.; Cooper, A. J. Brain alpha-ketoglutarate dehydrogenase complex: Kinetic properties, regional distribution, and effects of inhibitors. J. Neurochem. 47:11376 –1386; 1986. Lannfelt, L.; Bogdanovic, N.; Appelgren, H.; Axelman, K.; Lilius, L.; Hansson, G.; Schenk, D.; Hardy, J.; Winblad, B. Amyloid precursor protein mutation causes Alzheimer’s disease in a Swedish family. Neurosci. Letts. 168:254 –256; 1994. Mark, R. J.; Hensley, K.; Butterfield, D. A.; Mattson, M. P. Amyloid beta-peptide impairs ion-motive ATPase activities: Evidence for a role in loss of neuronal Ca21 homeostasis and cell death. J. Neurosci. 15:6239 – 6249; 1995. Martins, R. N.; Turner, B. A.; Carroll, R. T.; Sweeney, D.; Kim, K. S.; Wisniewski, H. M.; Blass, J. P.; Gibson, G. E.; Gandy, S. High levels of amyloid-b-protein from S182 (GLU246) familial Alzheimer cells. NeuroReport 7:217–220; 1995. Mattson, M. P. Calcium and neuronal injury in Alzheimer’s disease Contributions of beta-amyloid precursor protein mismetabolism, free radicals, and metabolic compromise. Ann. NY Acad. Sci. 747:50 –76; 1994. Mullan, M.; Crawford, F.; Axelman, K.; Houlden, H.; Lilius, L.; Winblad, B.; Lannfelt, L. A pathogenic mutation for probable Alzheimer’s disease in the APP gene at the N-terminus of beta-amyloid. Nature Genetics 1:345–347; 1992. Peterson, C.; Gibson, G. E.; Blass, J. P. Altered calcium uptake in cultured skin fibroblasts from patients with Alzheimer’s disease. N. Engl. J. Med. 312:1063–1064; 1985. Peterson, C.; Goldman, J. Alterations in calcium content and biochemical processes in cultured skin fibroblasts from aged and Alzheimer donors. Proc. Natl. Acad. Sci. USA. 83:2758 –2762; 1986. Poenie, M. Alteration of intracellular fura-2 fluorescence by viscosity: A simple correction. Cell Calcium 11:85–91; 1990. Redman, R. S.; Silinsky, E. M. Decrease in calcium currents induced by aminoglycoside antibiotics in frog motor nerve endings. Br. J. Pharm. 113:375–378; 1994. Querfurth, H. W.; Selkoe, D J. Calcium ionophore increases amyloid beta peptide production by cultured cells. Biochem. 33:4550 – 4561; 1994. Querfurth, H. W.; Wijsman, E. M; St. George-Hyslop, P. H.; Selkoe, D. J. Beta APP mRNA transcription is increased in cultured fibroblasts from the familial Alzheimer’s disease-1 family. Mol. Brain Res. 28:319 –337; 1995. Scheuner, D.; Eckman, C.; Jensen, M.; Song, X.; Citron, M.; Suzuki, N.; Bird, T. D.; Hardy, J.; Hutton, M.; Kukull, W.; Larsson, E.; Levy-Lahad, E.; Viitanen, M.; Peskind, E.; Poorkaj, P.; Schellenberg, G.; Tanzi, R.; Wasco, W.; Lannfelt, L.; Selkoe, D.; Younkin, S.
580 Secreted amyloid b-protein similar to that in the senile plaques of Alzheimer’s disease is increased in vivo by presenilin 1 and 2 and APP mutations linked to familial Alzheimer’s disease. Nature Med. 2:864 – 869; 1996. 42. Sheu, K. -F. R.; Cooper, A. J. L.; Koike, K.; Koike, M.; Linsday, J. G.; Blass, J. P. Abnormality of the a-ketaglutarate dehydrogenase complex in fibroblasts from familial Alzheimer’s disease. Ann. Neurol. 35:312–318; 1994. 43. Tatebayashi, Y.; Massatoshi, T.; Kashiwagi, T.; Masayasu, O.; Kurumadani, T.; Sekiyama, A.; Kanayama, G.; Hariguchi, S.; Nishimura, T. Cell-cycle dependent abnormal calcium responses in fibroblasts from patients with familial Alzheimer’s disease. Dementia 6:9 –16; 1995.
GIBSON ET AL. 44. VanHuynh, T.; Cole, G.; Katzman, R.; Huang, K. P.; Saitoh, T. Reduced protein kinase C immunoreactivity and altered protein phosphorylationin Alzheimer’s disease fibroblasts. Arch. Neurol. 46: 1195–1199; 1989. 45. Vestling, M.; Adem, A.; Lannfelt, L.; Cowburn, R. F. Protein kinase C levels and activity in cultured skin fibroblasts from affected and nonaffected of the Swedish family with the amyloid precursor protein 670/671 mutation. Soc. Neurosci. Abst. 25:744.14; 1995. 46. Vestling, M.; Adem, A.; Racchi, M.; Govoni, S.; Lannfelt, L; Gibson, G. E.; Cowburn, R. F. Differential regulation of adenylyl cyclase activity in fibroblasts from sporadic and familial Alzheimer’s disease cases with PS-1 and Swedish APP mutations. Neuroreport 8:2031– 2035; 1997.
PII:S0197-4580(97)00149-8
Abnormalities in Alzheimer’s Disease Fibroblasts Bearing the APP670/671 Mutation G. E. GIBSON,1* M. VESTLING,† H. ZHANG,* S. SZOLOSI,* D. ALKON,‡ L. LANNFELT,† S. GANDY§ AND R. F. COWBURN† *Cornell University Medical College at Burke Medical Research Institute, White Plains, NY 10605, †Karolinska Institute, Department of Clinical Neuroscience & Family Medicine, Section for Geriatric Medicine, Novum, KFC. S-141 86, Huddinge, Sweden, ‡National Institute of Neurological Diseases and Stroke, Bethesda, MD, 20892 §Cornell University Medical College, New York, NY 10021 Received 26 February 1997; Revised 2 October 1997; Accepted 6 October 1997 GIBSON, G., M. VESTLING, H. ZHANG, S. SZOLOSI, D. ALKON, L. LANNFELT, S. GANDY, AND R. F. COWBURN. Abnormalities in Alzheimer’s disease fibroblasts bearing the APP670/671 mutation. NEUROBIOL AGING 18(6) 573–580, 1997.—Abnormalities in cultured fibroblasts from familial Alzheimer’s Disease (FAD) cases uniquely enable the determination of how gene defects alter cell biology in living tissue from affected individuals. The current study focused on measures of calcium regulation and oxidative metabolism in fibroblast lines from controls and FAD individuals with the Swedish APP670/671 mutation. Bombesin-induced elevations in calcium in APP670/671 mutation-bearing lines were reduced by 40% (p , 0.05), a striking contrast to the 100% increase seen in sporadic AD and presenilin-1 (PS1) mutation-bearing cells in previously published studies. The APP670/671 mutation-bearing lines did not exhibit the exaggerated 4-bromo-A23187 releasable pool of calcium following 10 nM bradykinin, the enhanced sensitivity of calcium stores to low concentrations of bradykinin, nor the reduced activity of a-ketoglutarate dehydrogenase previously reported in cells from sporadic AD and mutant PS1 FAD. Thus, an altered regulation of internal calcium stores is common to all AD lines, but the calcium pool affected and the polarity of the alteration varies, apparently in association with particular gene mutations. Comparison of signal transduction in cell lines from multiple, genetically characterized AD families will allow testing of the hypothesis that these various pathogenic FAD abnormalities that lead to AD converge at the level of abnormal signal transduction. © 1997 Elsevier Science Inc. Alzheimer’s disease Calcium Fibroblasts Bombesin Bradykinin Signal transduction
Amyloid precursor protein
a-ketoglutarate dehydrogenase
to examine fibroblasts from individuals with the APP670/671 mutation for the possible signaling abnormalities that have been reported in fibroblasts from other AD populations. Several abnormalities have been reported in cultured fibroblasts from sporadic and presenilin-1 (PS1) mutation-bearing AD subjects (for reviews see 9,23). These include alterations in calcium (24,35), oxidative metabolism (36), cyclic AMP (21), the phosphatidylinositide cascade including protein kinase C (3,12,22, 28), G-proteins (27), K1 channels (6,7), and amyloid-b-peptide production (4,25,32,41). The current studies compare these previously reported AD-related differences in measures of calcium regulation and energy metabolism to these same variables in APP670/671 mutation-bearing cells. Calcium can both activate (17) and inactivate (29) key enzymes of energy metabolism. For example, calcium inactivates a-ketoglutarate dehydrogenase (KGDHC), an enzyme that is strikingly reduced in AD brains (10). Oxidative processes are altered in AD fibroblasts (36), and mitochondria isolated from fibroblasts from Alzheimer patients are more sensitive to free radicals than those isolated from control
INTRODUCTION
ABNORMAL signal transduction systems have been implicated in the pathophysiology of Alzheimer’s disease (AD), but their precise role has been difficult to establish. Cultured fibroblasts offer many advantages for the study of signal transduction in AD. The signals are more robust in these living cells than in postmortem material (such as brain) and the results are not complicated by agonal or postmortem events. Also, any effects of the patients’ diets or drugs are diluted millions of times. Fibroblasts also allow comparison of cell biological systems and pathways, such as signal transduction, in cells from AD patients with specific known genetic abnormalities. One of the most extensively investigated families with a genetic defect leading to AD is the Swedish AD family with the amyloid precursor protein (APP) 670/671 mutation (30,34). All individuals with this defect invariably develop AD (1,31). Fibroblasts from these patients produce excess amyloid b-peptide compared to cells from control relatives (4,15,25,40) and this excess amyloid-b-peptide production presumably promotes cerebral deposition of amyloid. The purpose of the current study was
1 Address correspondence to: G. E. Gibson, Cornell University Medical College, Burke Medical Research Institute, 785 Mamaroneck Avenue, White Plains, NY 10605, email: [email protected]
573
574
GIBSON ET AL.
fibroblasts (28). Amyloid-b-peptide produces free radicals in vitro (33), so it is possible that the enhanced amyloid-b-peptide production in patients with the APP670/671 mutation may alter mitochondria and the activities of the key mitochondrial enzyme KGDHC, which, in turn, would alter mitochondrial calcium regulation. Thus, the activity of KGDHC was also examined in cells from patients with the APP670/671 mutation. MATERIALS AND METHODS
Materials Fura-2 AM and 4-bromo-A23187 were from Molecular Probes (Eugene, OR). Dulbecco’s Modified Eagle’s Medium (GIBCO, Grand Island, NY) and fetal bovine serum, HEPES, EGTA, a-ketoglutaric acid, leupeptin, Coenzyme A, and nicotinamide adenine dinucleotide (Sigma Chemical Co., St. Louis, MO) were from the indicated companies. Bovine albumin fraction V was from United States Biochemical Corp. (Cleveland, OH). The reagents for the Bradford assay were from Bio-Rad (Hercules, CA). Fibroblasts Fibroblast cell lines were established at the Department of Geriatric Medicine of the Karolinska Institute from skin biopsies taken from affected and nonaffected individuals in the Swedish family with the APP670/671 mutation. Six lines were established from individuals with the APP670/671 mutation, of which four lines were from symptomatic carriers and two lines from asymptomatic individuals. For the analysis in this paper, the six lines with the mutation were combined because of the high penetrance of AD in individuals with this mutation. The average age (mean 6 SD) of the individuals with the mutation was 53 6 8 years. Six lines were also established from nonaffected individuals who did not have the APP670/671 mutation who were within the same family. The average age of these control donors was 60 6 11 years. Cells were maintained under strictly controlled conditions, according to the procedures of Cristofalo and Charpentier (5). Antibiotics were not used because these are known to alter signal transduction systems including those mediated by cyclic adenosine monophosphate (AMP) (19,20), calcium (38), and protein kinase C (16). In brief, cultures were grown in Dulbecco’s minimal essential medium supplemented with 10% fetal bovine serum at 37°C in a humidified atmosphere with 5% CO2 in air. Cells were routinely seeded at a density of 104/cm2 and subcultured every 7 days upon reaching confluence. The cells for each assay were treated exactly as in the original reports that showed AD-related changes for that variable, since our goal was to test whether previously reported abnormalities in PS1 mutant cells and sporadic AD cells could be documented in cells from the APP670/671 mutation family. Thus, in those cases in which measurements had not been made in this laboratory, previously reported abnormalities were first replicated in PS1 mutant cells before commencing studies in cells with the APP670/ 671 mutation. The PS1 cell lines that were examined have the Glu246 mutation. This mutation may lead to different changes than other PS1 mutations (46). Enzyme Activity Measurements Following subculture, cells were incubated for 3 weeks to ensure that all cells reached the G0 stationary phase. Cells were collected by scraping in 10 mM sodium phosphate (pH 7.2), 150 mM NaCl, and 1 mM EGTA, and centrifuged at 800 3 g for 10 min. After washing with the same buffer twice, cell pellets were
suspended by manual homogenization in 50 mM Tris-HCl (pH 7.2), 1 mM dithiothreitol, 0.2 mM EGTA, 0.4% Triton X-100, and 50 mM leupeptin. The freshly prepared lysate was spun for 3 s in a microcentrifuge, and the supernatant was used for spectrophotometric assays of KGDHC at 30 6 1°C (42). No activities were detectable in the pellet, but both supernatant and pellet were used to assess total protein in each flask by the Bradford assay. Every cell line was tested on five different days with triplicate measures each day. Response to Low Concentrations of Bradykinin Confluent fibroblasts were subcultured on 25-mm round glass cover slips at a density of 500/cm2 and allowed to grow for 7 days. Cells were incubated in balanced salt solution (BSS) (140 mM NaCl, 5 mM KCl, 1.5 mM MgCl2, 5 mM glucose, 10 mM HEPES, and 2.5 mM CaCl2) with 2 mM Fura-2 AM at room temperature for 1 h and then rinsed three times with calcium-free BSS. The coverslip was transferred to a Teflon Leiden coverslip dish and mounted on the microscope stage of an Olympus IMT-2 inverted microscope. The dual excitation wavelengths were achieved by passing the light generated with a Xenon 75-watt arc lamp through dual monochromaters at wavelengths of 350 and 378 nm (bandpass of 2 nm). Images were taken with a Hamamatsu 2400C SIT camera and converted to 8-bit images with specialized computer boards from Photon Technology International. The ratio of images after excitation at 350 and 378 nm was converted to [Ca21]i with Photon Technology International Imaging software (Princeton, NJ) using a viscosity constant of 0.7 (13,37). Bradykinin (BK) (0.2 nM) was made separately for each measurement by adding 10 mL of 20 nM stock (in 0.2% dimethyl sulfoxide) into 1 mL of BSS and sonicating in a cleanser bath for 40 s immediately before the measurement. Experiments were performed at room temperature. Fibroblasts were maintained on the microscope stage in 1 mL of BSS and basal [Ca21]i was measured for 1 min. One mL of 0.2 nM BK was then added to the cell dish to make the final BK 0.1 nM. The [Ca21]i was then measured for another 3 min. Cells with increases of [Ca21]i more than 100% of their basal values were regarded as responding cells (18,24). Response to Bombesin These images were collected and the analysis was done as described above except that the agonist was bombesin (1 mM) and measurements were done in calcium-free BSS (BSS with 0.1 mM CaCl2 and 1 mM EGTA to give 7.18 nM calcium) (24). Evaluation of the 4-bromo-A23187 Releasable Pool of Calcium Following 10 nM BK Confluent cells were subcultured on 25-mm round glass cover slips at a density of 1 3 104 cells/cm2. Fibroblasts were loaded with 2 mM fura-2AM in incubation media for 1 h at room temperature, which prevents sequestration of the dye in the organelles. The incubation media was Dulbecco’s Modified Eagles Medium without phenol red with 20 mM HEPES and 1% bovine serum albumin (pH 7.4 at room temperature). Following this incubation, the coverslip was rinsed three times with testing media without fura-2 (37°C). The calcium-free testing media was similar to the incubation media but with the addition of 2.5 mM EGTA (59 nM calcium; pH 7.4 at 37°C). The cover slips were secured in a Leiden coverslip dish and placed on a temperature-controlled micro-incubator with 2 mL testing media. The micro-incubator was mounted onto the stage of an Olympus inverted microscope equipped with a Nikon 40 3 UV fluor objective. To minimize the temperature gradient between different parts of the coverslip, the
ALZHEIMER FIBROBLASTS
575 Response to Low Concentrations of BK
FIG. 1. KGDHC activity in controls and cells with the APP670/671 mutation. Values for each cell line are the mean of each cell line measured three times per day on 4 –5 different days. X and error bars represent mean 6 SEM of all six lines.
micro-incubator was continuously gassed with 5% CO2/95% air at a flow rate of three standard cubic feet per hour (11). These studies utilized a photomultiplier tube to record the response of a whole field with three to six cells. This permitted complete definition of the temporal response. Following a 3-min equilibration period, calcium measurement was initiated by alternately exciting the cells with light at 350 and 380 nm five times per second using a Delta scan, and light emission was measured at 510 nm (Photon Technology International; S. Brunswick, NJ). The cells were treated with BK and 4-bromo-A23187 by removing about 90% of the testing medium and adding it to a cup containing BK or 4-bromo-A23187. The solution was mixed by repeated pipetting and then reapplied to the cells. These manipulations took less than 10 s. The cells were tested on three separate days, with three samples tested on each day. [Ca21]i was calculated by the method of Grynkewicz et al. (13) after correction for background fluorescence.
The goal of the current studies was to test if we could confirm previously reported changes in the presenilin-1 lines in a limited number of lines (18) and to then compare these results with cells from APP670/671 mutation-bearing lines. Initial studies confirmed the enhanced calcium response of fibroblasts from familial AD (FAD) individuals with PS1 mutations to stimulation by low BK concentrations (Fig. 2). Thirty-one of 367 (8.4%) fibroblasts measured in three lines from AD patients with PS1 mutations (AG6840, AG6848, AG8170) responded to low concentrations of BK. Only 1 of 306 fibroblasts (0.3%) measured in three agematched control lines (AG7657-Canadian escapee, AG9878, GM4260) responded to 0.1 nM BK. The response to low BK concentrations did not distinguish individual FAD APP670/671 cell lines from controls. Previous results showed that a response of greater than 1.5% of the cells to low concentrations of BK distinguished sporadic AD and PS1 FAD cell lines from controls (18). In cells with the APP670/671 mutation, 11 of 588 (1.87%) tested cells from six lines responded to 0.1 nM BK, whereas only 2 of 612 (0.33%) control cells responded. Although examination of the combined data suggests that the 1.5% criteria applies to these APP670/671 cell lines, this standard is not applicable on an individual cell line basis. Only two of the APP670/671 mutationbearing cell lines met this criteria while four of the six lines did not (Fig. 2). Five of the control lines had no response, whereas in one line, 1.7% of the cells responded. Thus, the differences that occurred in fibroblasts from the sporadic AD and the PS1 mutant FAD individuals (previous studies) were not apparent in cells bearing the APP670/671 mutation. Response to Bombesin The response of [Ca21]i to bombesin (1 mM) was altered in fibroblasts with the APP670/671 mutation (Fig. 3). Studies in three
Statistics Analysis of variance was done with SPSS (Chicago, IL). Given the high penetrance in this family (1), data from the symptomatic and asymptomatic mutation-bearers was grouped and compared to the non-mutation-bearing cell lines. RESULTS
KGDHC Activity in Fibroblasts with the APP670/671 Mutation Diminished KGDHC activities have been reported in fibroblasts (2,10,42) and brain (10) from AD patients. The mean activity (mean 6 SEM) of KGDHC for the control group was 10.06 6 2.26 nmol/min per mg protein (6 lines). The mean activity of KGDHC for the APP670/671 mutation group was 10.23 6 2.53 nmol/min per mg protein (6 lines). The individual values for each line are presented in Figure 1. The results demonstrate that KGDHC activities did not differ between the control and mutationbearing lines.
FIG. 2. Bradykinin-sensitive calcium stores in fibroblasts from control and AD individuals with APP670/671 or PS1 (GLU246) mutations. Values for each line are the results from .100 cells per line measured on 3 dishes a day on 2–3 days. X and error bars in each group represent mean 6 SEM of all the lines in that group. Circles are control values and squares are values of AD with gene mutations. Control 1 (n 5 6) and APP670/671 mutation-bearing lines (n 5 6) are from the Swedish kindred. Control 2 (n 5 3) and Presenilin-1 (GLU246) mutation-bearing lines (n 5 3) are from Coriell Institute for Medical Research in Camden, NJ.
576
GIBSON ET AL.
FIG. 3. Bombesin-induced changes in [Ca21]i in fibroblasts from controls and individuals with the APP670/671 mutation. Values are means 6 SEM (n 5 6 for each). The two groups vary significantly from each other (p 5 0.002). Three dishes/day were examined on two different days for each cell line. More than 100 cells were evaluated for each cell line.
AD (AG6840, AG6848, AG8170) and three control (AG7657, AG9878, GM4260) fibroblast lines confirmed that the responses to 1 mM bombesin were exaggerated in AD fibroblasts. The line-toline variation generally agreed with the results from the group that initially reported the bombesin-induced change with AD (24). Mean basal [Ca21]i (mean 6 SEM) was identical in the control and APP670/671 mutation-bearing cells (control 100.8 6 19.9 vs. mutation 99.9 6 10.2 nM). However, the peak response to bombesin was reduced by 40% in fibroblasts with the APP670/671 mutation (control 731.2 6 82.9 vs. mutation 530.7 6 85.8 nM; p 5 0.002). The mean integrated area over basal of the bombesin response was reduced by 36% in the APP670/671 mutationbearing fibroblasts (control 27088 6 10250 vs. mutation 17228 6 7531; p 5 0.044). Some cells did not respond to bombesin. The percentage of cells responding was reduced by 29% in the APP670/671 mutation-bearing fibroblasts (control 69 6 18% vs. mutation 49 6 17%; p 5 0.041). When only responders were analyzed, no differences were apparent between the fibroblasts with or without the APP670/671 mutation. 4-Bromo-A23187 Releasable Pool of Calcium Following 10 nM BK The 4-bromo-A23187 releasable pool of calcium following 10 nM bradykinin did not vary between the control and mutation bearing lines (Fig. 4). Treatment with 10 nM BK increases [Ca21]i by release of calcium from internal stores and blocks any changes in [Ca21]i due to subsequent stimulation with 10 nM BK, thapsigargin, or bombesin (1 mM) (11). However, subsequent addition of 4-bromo-A23187 still increases [Ca21]i. A typical temporal response is shown in Figure 4a. The majority of cellular calcium was released by BK, but subsequent addition of 4-bromoA23187 released additional calcium stores. This calcium store did not vary between the control lines and those with the APP670/671
mutation at either the peak or at 1, 2, or 3 min after addition of 4-bromo-A23187 (Fig. 4b). Nor did the integrated area above basal vary between the mutation-bearing cells and controls. DISCUSSION
Studies that preceded the discovery of the presenilins reported decreases in KGDHC activities that ranged from 23 (nonsignificant) (10) to 40% (42) in fibroblasts from patients now known to have presenilin-1 mutations, other familial mutations, and sporadic AD. A recent study with a larger number of subjects showed normal activity of KGDHC in fibroblasts from a presenilin-1 family, and a 40% reduction in fibroblasts from sporadic AD patients (2). The results of the present study using APP670/671 mutation-bearing cell lines, when taken together with data from the above studies on sporadic AD and PS-1 mutation-bearing AD fibroblasts, would suggest that reduced KGDHC activity is a feature of sporadic but not familial AD. The fact that APP670/671 and PS-1 mutation-bearing cell lines have an over production of the amyloid-b-peptide in common (40) would suggest further that reduced KGDHC activities are not a consequence of over exposure to excess levels of Ab peptide. This notion is further supported by previous studies showing that reduced KGDHC in AD brain occurs both in affected regions with a high density of amyloidcontaining plaques, as well as in relatively spared regions (10). Following depletion of internal calcium stores by addition of 10 nM BK, neither BK nor bombesin increases [Ca21]i, whereas calcium can still be released from internal stores by 4-bromoA23187. The size of this 4-bromo-A23187 releasable pool of calcium following 10 nM BK is constant over BK concentrations ranging from 1–1000 nM. Previous studies suggest this may be a mitochondrial pool of calcium (11) and that this store of calcium is increased in the lines bearing the presenilin GLU246 mutation and in fibroblasts from sporadic Alzheimer patients (11). Whether
ALZHEIMER FIBROBLASTS
577
FIG. 4. A) Characterization of the 4-bromo-A23187 releasable pool of calcium following 10 nM BK. All incubations were performed in calcium-free media. A basal reading was obtained for 1 min. BK (10 nM) was added at 1 min (60 s: left panel). 4-bromo-A23187 (20 mM) was added at 4 min (240 s: right panel). Note: The 4-bromo-A23187 y axis scale was enlarged for clarity. B) Internal calcium stores in fibroblasts from AD patients with the APP670/671 mutation. Cells were first stimulated with BK and, 3 min later, 4-bromo-A23187 was added. The vertical axis is the increase over [Ca21]i at 3 min after BK stimulation. Values are means 6 SEM of six control cell lines and six mutation-bearing lines tested three times per day on three different days.
this latter change reflects a uniform response or a few cells that have greatly exaggerated responses cannot be estimated by these methods. The current results show that this compartment is not altered in cells bearing the APP670/671 mutation. This further supports our interpretation of the KGDHC activity data that suggest mitochondria are unaffected in APP670/671 mutantbearing fibroblasts. AD-related changes in the phosphatidylinositide cascade have been shown in fibroblasts from sporadic AD and PS1 mutant FAD individuals by several methods. Multiple investigators report diminished protein kinase C levels and activities in AD fibroblasts (3,12,44), although not in cells bearing the APP670/671 mutation (45). AD fibroblasts also have a heightened sensitivity to BK, an activator of the phosphatidylinositide cascade, as shown by an exaggerated inositol 1,4,5-trisphosphate (IP3) response (22) and an
enhanced response of [Ca21]i to low concentrations of BK (18). The latter study suggests that the response induced by 0.1 nM BK has a higher degree of specificity for identifying AD cells than tetraethylammonium- or bombesin-induced [Ca21]i responses (ref. 7 and 24, respectively). In 13 out of 14 AD cell lines, greater than 1.5% of the cells responded, whereas none of 11 controls showed this response (18). The differences in the response of sporadic AD cells to BK does not occur at higher BK concentrations (100 nM). In two other FAD families, the response to 100 nM BK was depressed, variable, and cell cycle dependent (43). In view of this confusion in the literature, we began the present studies by first confirming the heightened sensitivity of cells from sporadic AD cases and PS1 mutant FAD individuals in order to exclude the possibility that any technical differences existed between this study and the original report (18). Although the methodology and
578
GIBSON ET AL.
results were identical to earlier studies, cells from only two of six lines with the APP670/671 mutation have an enhanced calcium sensitivity to BK. Thus, the sensitivity of internal BK responsive calcium stores is not a general property of Alzheimer cells. Of the various calcium pools that were examined, the only one that showed a mutation-related difference was the calcium compartment that was sensitive to bombesin. Bombesin activates membrane-bound phospholipase C to release IP3, which subsequently leads to release of calcium from the endoplasmic reticulum. In contrast to results with other AD fibroblast lines in which this pool was increased (24), this compartment of calcium was lower in the APP670/671 mutation-bearing cells. The current results suggest an interaction between amyloid-b-peptide and the bombesin-sensitive calcium pools. A cell line by cell line comparison of the amyloid-b-peptide production in these same cells (4,25) reveals a significant negative correlation to the percent of cells responding to bombesin in the current study (r 5 20.74; p 5 0.022). Although direct experiments with amyloid-bpeptide suggest that external addition of synthetic amyloid-bpeptide does not alter bombesin-sensitive calcium stores (7), endogenous production over long times may lead to a very different response. Abnormalities in the receptor-mediated release of internal calcium pools are a consistent finding in cells from Alzheimer patients, but the precise nature of how these pools are altered varies between the familial and sporadic AD populations. BK and bombesin act at the cell membrane receptors to release IP3. Both agonists are generally regarded to release calcium from the same IP3-sensitive internal store. However, the measured calcium response depends upon the bombesin/BK receptors, phosphatidylinositide turnover, G proteins, Ca21-ATPase activities, the IP3 receptors, and the size of the calcium pool. The reduced calcium response after high agonist concentrations (1 mM bombesin) may be due to a diminished activity of any one of these steps or a smaller internal calcium pool in the cell lines bearing the APP670/ 671 mutation. Distinguishing which of these is involved requires further experiments. This diminished response contrasts with cells from presenilin-1 families or sporadic AD patients in which the calcium response is increased. Interestingly, transfection of the PS-1 mutation into PC12 cells also increases the size of this calcium store as estimated by the response to high agonist concentration (100 nM bradykinin) (14). The sensitivity of IP3 calcium store to low concentrations of BK was similar in fibroblasts from control and the APP670/671 mutation-bearing cells. By comparison, cells from sporadic AD subjects or presenilin-1 AD patients have increased sensitivity to low BK concentrations. This suggests that the fibroblasts from the APP670/671 mutation-bearing cells do not have increased BK receptors as has been shown for these other cell lines (22). These results demonstrate that the APP670/671 mutation does not cause changes in the signal transduction variables similar to those reported to be altered in fibroblasts from other AD kindreds or in sporadic AD cases. Although controversy surrounds many
studies of AD fibroblasts, appropriate attention to detail has lead to a consensus about many of the changes (9). The results in this paper reveal that changes in signal transduction persist in cultured fibroblasts, but that the AD-related differences vary with different gene defects. The only abnormality that has been previously reported in the APP670/671 mutation-bearing cells is that of altered amyloid-b-peptide production (4,25). The additional change reported in the current studies is that of a reduction in the bomesin-induced calcium increases. In addition to the lack of changes reported in this manuscript, the activity of protein kinase C (45) and the b-adrenergic-stimulated formation of cyclic AMP (46) are also normal in fibroblasts with the APP670/671 mutation. To date, no change nor group of changes has been consistently reported in fibroblasts from all of the different AD families or in sporadic AD cases. These results and those in the parallel study on b-adrenergic-stimulated cyclic AMP production (46) and protein kinase C levels and activity (45) demonstrate that fibroblast lines bearing the APP670/671 mutation show changes in signal transduction that are quite different from those seen in sporadic AD cases or in PS-1 mutant FAD individuals. It will now be important to investigate the molecular events that link PS1 and PS2 mutations to abnormal calcium and oxidation, and to determine whether presenilin/calcium/oxidative interactions are “upstream,” “downstream,” or “in parallel” to key features of AD: namely amyloidosis and neurodegeneration. Multiple factors, including various gene mutations, can cause AD. At some level, these different abnormalities converge to produce the symptoms and pathology that are common to AD regardless of etiology. The robust changes reported in fibroblasts from AD subjects without APP mutations that were not present in the APP670/671 mutation-bearing lines (reduced KGDHC activity and abnormal calcium regulation) would appear to precede and perhaps initiate alterations in amyloid-b-peptide in those patients. Impaired mitochondrial function leads to production of amyloidgenic fragments (8) and brains of patients that died with mitochondrial gene mutations have plaques (26). Increasing cellular calcium by addition of 4-bromo-A23187 also elevates production of amyloid-b-peptide (39). The differences in calcium regulation in the APP670/671 mutation-bearing cells may be caused by the exaggerated amyloid b-peptide production that is known to occur in these fibroblasts (25) because amyloid-b-peptide is known to lead to calcium dysregulation (31). ACKNOWLEDGMENTS
The authors are grateful to Craig Grossman for making many of the a-ketoglutarate dehydrogenase activity measurements, to Karen Scheffold for her excellent help with the tissue culture, and to Dr. Rene Etcheberrigaray (Institute for Cognitive and Computational Sciences, Georgetown University Medical Center) for helping us to establish the methods for the experiments with low concentrations of BK and bombesin. The work was supported by National Institute of Aging Grants AG11921 (GG), AG13780 (SG), and AG09464 (SG). The Åke Wiberg and Gamla Tja¨narinnor Foundations are also thanked for financial support.
REFERENCES 1. Axelman, K.; Basun, H.; Winblad, B.; Lannfelt, L. A large Swedish family with Alzheimer’s disease with a codon 670/671 amyloid precursor protein mutation: A clinical and genealogical investigation. Arch. Neurol. 51:1193–1197; 1994. 2. Blass, J. P.; Sheu, K. -F. R.; Piaciantini, S.; Sorbi, S. Inherent abnormalities in oxidative metabolism in AD: Interaction with vascular abnormalities. In: de la Torre, J. C.; Hachinski, V. eds. Cerebro-
vasular Pathology in Alzheimer’s Disease. New York: New York Academy of Sciences; 1997 (In press). 3. Bruel, A. G.; Cherqui, G.; Columelii, S.; Margelin, D.; Roudier, M.; Sinet, P.; Prieur, M.; Perignon, J.; Delarab, J. Reduced protein kinase C activity in sporadic Alzheimer’s disease fibroblasts. Neurosci. Lett. 133:89 –92; 1991. 4. Citron, M.; Vigo-Pelfrey, C.; Teplow, D. B.; Miller, C.; Schenk, D.;
ALZHEIMER FIBROBLASTS
5. 6.
7. 8.
9.
10.
11. 12.
13. 14.
15.
16. 17. 18.
19.
20. 21. 22.
23.
Johnston, J.; Winblad, B.; Venizelos, N.; Lannfelt, L.; Selkoe, D. J. Excessive production of amyloid beta-protein by peripheral cells of symptomatic and presymptomatic patients carrying the Swedish familial Alzheimer disease mutation. Proc. Natl. Acad. Sci. USA 91:11993– 11997; 1994. Cristofalo, V. J.; Charpentier, R. A standard procedure for cultivating human diploid fibroblast like cells to study cellular aging. J. Tiss. Cult. Meth. 3&4:117–121; 1980. Etcheberrigaray, R.; Ito, E.; Oka, K.; Tofel-Grehl, B.; Gibson, G. E.; Alkon, D. L. Potassium channel dysfunction in fibroblasts identifies patients with Alzheimer’s disease. Proc. Natl. Acad. Sci. USA 90: 8209 – 8213; 1993. Etcheberrigaray, R.; Ito, E.; Oka, K.; Kim, C. S.; Alkon, D. L. Soluble beta-amyloid induction of Alzheimer’s phenotype for human fibroblast K1 channels. Science 264:276 –279; 1994. Gabuzda, D.; Busciglio, J.; Chen, L. B.; Matsudaira, P.; Yankner, B. A. Inhibition of energy metabolism alters the processing of amyloid precursor protein and induces a potentially amyloidogenic derivative. J. Biol. Chem. 269:13623–13628; 1994. Gibson, G.; Martins, R.; Blass, J.; Gandy, S. Altered oxidation and signal transduction systems in fibroblasts from Alzheimer patients. In: Mailer. D.; Gispen, W., eds. The Current Status of the Calcium Hypothesis of Brain Aging and Alzheimer’s Disease. Life Science 59:477– 490; 1996. Gibson, G. E.; Sheu, K-F. R.; Blass, J. P.; Baker, A.; Carlson, K. C.; Harding, B.; Perrino, P. Reduced activities of thiamine-dependent enzymes in brains and peripheral tissues of Alzheimer patients. Arch. Neurol. 45:836 – 840; 1988. Gibson, G. E.; Toral-Barza, L.; Tofel-Grehl, B.; Szolosi, S.; Zhang, H. Calcium stores in cultured fibroblasts and their changes with Alzheimer’s disease. Biochim. Biophys. Acta 1316:71–77; 1996. Govoni, S.; Bergamaschi, S.; Racchi, M.; Battaini, F.; Binetti, A.; Bianchetti, A.; Trabucchi, M. Cytosol protein kinase C down regulation in fibroblasts Alzheimer’s disease patients. Neurol. 43:1195– 1199; 1993. Grynkiewicz, G.; Poenie, M.; Tsien, R. Y. A new generation of Ca21 indicators with greatly improved fluorescence properties. J. Biol. Chem. 260:3440 –3450; 1985. Guo, Q.; Katsutosh, F.; Sopher, B. L.; Pham, D. G.; Robinson, N.; Martin, G. M.; Mattson, M. P. Alzheimer’ PS-1 mutation perturbs calcium homeostasis and sensitizes PC12 cells to death induced by amyloid b-peptide. NeuroReport 8:379 –383; 1996. Haass, C.; Lemere, C. A.; Capell, A.; Citron, M.; Seubert, P.; Schenk, D.; Lannfelt, L.; Selkoe, D. J. The Swedish mutation causes earlyonset Alzheimer’s disease by beta-secretase cleavage within the secretory pathway. Nature Med. 1:1291–1296; 1995. Hagiwara, M.; Inagaki, M.; Kanamura, K.; Ohta, H.; Hidaka, H. Inhibitory effects of aminoglycosides on renal protein phosphorylation by protein kinase C. J. Pharmacol. Exp. Ther. 244:355–360; 1988. Hansford, R. G.; Castro, F. Role of Ca21 in pyruvate dehydrogenase interconversion in brain mitochondria and synaptosomes. Biochem. J. 227:129 –136; 1985. Hirashima, N.; Etcheberrigaray, R.; Battaini, R.; Bergamaschi, S.; Racchi, M.; Battaini, F.; Binetti, G.; Govoni, S.; Alkon, D. L. Calcium responses in human fibroblasts: a diagnostic molecular profile for Alzheimer’s disease. Neurobiol. Aging 17:549 –555; 1996. Hori, R.; Yamamoto, K.; Saito, H.; Kohno, M.; Inui, K. Effect of aminoglycoside antibiotics on cellular functions of kidney epithelial cell line (LLC-PK1): A model system for aminoglycoside nephrotoxicity. J. Pharmacol. Exp. Ther. 230:724 –728; 1984. Howard, M.; Frizzell, R. A.; Bedwell, D. M. Aminoglycoside antibiotics restore CFTR function by overcoming premature stop mutations. Nature Med. 2:467– 469; 1996. Huang, H. -M.; Gibson, G. E. Altered b-adrenergic receptor stimulated cAMP formation in cultured skin fibroblasts from Alzheimer donors. J. Biol. Chem. 268:14616 –1462; 1993. Huang, H. -M.; Lin, T. A.; Sun, G. Y.; Gibson, G. E. Increased insitol(1,4,5)trisphosphate accumulation correlates with an upregulation of bradykinin receptors in Alzheimer’s disease. J. Neurochem. 64:761–766; 1995. Huang, H. -M.; Martins, R.; Gandy, S.; Etcheberrigaray, R.; Ito, E.; Alkon, D. L.; Blass, J. P. Gibson, G. E. The use of cultured fibroblasts
579
24.
25.
26.
27.
28. 29. 30.
31.
32.
33.
34.
35. 36. 37. 38. 39. 40.
41.
in elucidating the pathophysiology and diagnosis of Alzheimer’s disease. In: Disterhoft, J. F.; Gispen, W. H.; Traber, J.; Khachaturian, Z. S., eds. Calcium Hypothesis of Aging and Dementia. New York: New York Academy of Science; 1994:225–244. Ito, I.; Oka, K.; Etcheberrigaray, R.; Nelson, T. J.; Mc Phie, D.; Tofel-Grehl, B.; Gibson, G. E.; Alkon, D. L. Internal Ca21 mobilization is altered in fibroblasts with Alzheimer disease. Proc. Natl. Acad. Sci. USA. 91:534 –538; 1994. Johnston, J. A.; Cowburn, R. F.; Norgren, S.; Wiehager, B.; Venizelos, N.; Winblad, B.; Vigo-Pelfrey, C.; Schenk, D.; Lannfelt, L.; O’Neill, C. Increased beta-amyloid release and levels of amyloid precursor protein (APP) in fibroblast cell lines from family members with the Swedish Alzheimer’s disease APP670/671 mutation. FEBS Letters 354:274 –278; 1994. Kaido, M.; Fugimura, H.; Soga, F.; Toyooka, S. K.; Yoshikawa, H.; Nishimura, T.; Higashi, T.; Inui, K.; Imanishi, H.; Yorifuji, S.; Yanagihara, T. Alzheimer-type pathology in a patient with mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes (MELAS). Acta Neuropathol. 92:312–318; 1996. Kim, C. S.; Han, Y. F.; Etcheberrigaray, R.; Nelson, T. J.; Olds, J. L.; Yoshioka, T.; Alkon, D. L. Alzheimer and beta-amyloid-treated fibroblasts demonstrate a decrease in a memory-associated GTPbinding protein, Cp20. Proc. Natl. Acad. Sci. USA. 92:3060 –3064; 1995. Kumar, U.; Dunlop, D. M.; Richardson, J. S. Mitochondria from Alzheimer’s fibroblasts show decreased uptake of calcium and increased sensitivity to free radicals. Life Science 54:1855–1860; 1994. Lai, J. C.; Cooper, A. J. Brain alpha-ketoglutarate dehydrogenase complex: Kinetic properties, regional distribution, and effects of inhibitors. J. Neurochem. 47:11376 –1386; 1986. Lannfelt, L.; Bogdanovic, N.; Appelgren, H.; Axelman, K.; Lilius, L.; Hansson, G.; Schenk, D.; Hardy, J.; Winblad, B. Amyloid precursor protein mutation causes Alzheimer’s disease in a Swedish family. Neurosci. Letts. 168:254 –256; 1994. Mark, R. J.; Hensley, K.; Butterfield, D. A.; Mattson, M. P. Amyloid beta-peptide impairs ion-motive ATPase activities: Evidence for a role in loss of neuronal Ca21 homeostasis and cell death. J. Neurosci. 15:6239 – 6249; 1995. Martins, R. N.; Turner, B. A.; Carroll, R. T.; Sweeney, D.; Kim, K. S.; Wisniewski, H. M.; Blass, J. P.; Gibson, G. E.; Gandy, S. High levels of amyloid-b-protein from S182 (GLU246) familial Alzheimer cells. NeuroReport 7:217–220; 1995. Mattson, M. P. Calcium and neuronal injury in Alzheimer’s disease Contributions of beta-amyloid precursor protein mismetabolism, free radicals, and metabolic compromise. Ann. NY Acad. Sci. 747:50 –76; 1994. Mullan, M.; Crawford, F.; Axelman, K.; Houlden, H.; Lilius, L.; Winblad, B.; Lannfelt, L. A pathogenic mutation for probable Alzheimer’s disease in the APP gene at the N-terminus of beta-amyloid. Nature Genetics 1:345–347; 1992. Peterson, C.; Gibson, G. E.; Blass, J. P. Altered calcium uptake in cultured skin fibroblasts from patients with Alzheimer’s disease. N. Engl. J. Med. 312:1063–1064; 1985. Peterson, C.; Goldman, J. Alterations in calcium content and biochemical processes in cultured skin fibroblasts from aged and Alzheimer donors. Proc. Natl. Acad. Sci. USA. 83:2758 –2762; 1986. Poenie, M. Alteration of intracellular fura-2 fluorescence by viscosity: A simple correction. Cell Calcium 11:85–91; 1990. Redman, R. S.; Silinsky, E. M. Decrease in calcium currents induced by aminoglycoside antibiotics in frog motor nerve endings. Br. J. Pharm. 113:375–378; 1994. Querfurth, H. W.; Selkoe, D J. Calcium ionophore increases amyloid beta peptide production by cultured cells. Biochem. 33:4550 – 4561; 1994. Querfurth, H. W.; Wijsman, E. M; St. George-Hyslop, P. H.; Selkoe, D. J. Beta APP mRNA transcription is increased in cultured fibroblasts from the familial Alzheimer’s disease-1 family. Mol. Brain Res. 28:319 –337; 1995. Scheuner, D.; Eckman, C.; Jensen, M.; Song, X.; Citron, M.; Suzuki, N.; Bird, T. D.; Hardy, J.; Hutton, M.; Kukull, W.; Larsson, E.; Levy-Lahad, E.; Viitanen, M.; Peskind, E.; Poorkaj, P.; Schellenberg, G.; Tanzi, R.; Wasco, W.; Lannfelt, L.; Selkoe, D.; Younkin, S.
580 Secreted amyloid b-protein similar to that in the senile plaques of Alzheimer’s disease is increased in vivo by presenilin 1 and 2 and APP mutations linked to familial Alzheimer’s disease. Nature Med. 2:864 – 869; 1996. 42. Sheu, K. -F. R.; Cooper, A. J. L.; Koike, K.; Koike, M.; Linsday, J. G.; Blass, J. P. Abnormality of the a-ketaglutarate dehydrogenase complex in fibroblasts from familial Alzheimer’s disease. Ann. Neurol. 35:312–318; 1994. 43. Tatebayashi, Y.; Massatoshi, T.; Kashiwagi, T.; Masayasu, O.; Kurumadani, T.; Sekiyama, A.; Kanayama, G.; Hariguchi, S.; Nishimura, T. Cell-cycle dependent abnormal calcium responses in fibroblasts from patients with familial Alzheimer’s disease. Dementia 6:9 –16; 1995.
GIBSON ET AL. 44. VanHuynh, T.; Cole, G.; Katzman, R.; Huang, K. P.; Saitoh, T. Reduced protein kinase C immunoreactivity and altered protein phosphorylationin Alzheimer’s disease fibroblasts. Arch. Neurol. 46: 1195–1199; 1989. 45. Vestling, M.; Adem, A.; Lannfelt, L.; Cowburn, R. F. Protein kinase C levels and activity in cultured skin fibroblasts from affected and nonaffected of the Swedish family with the amyloid precursor protein 670/671 mutation. Soc. Neurosci. Abst. 25:744.14; 1995. 46. Vestling, M.; Adem, A.; Racchi, M.; Govoni, S.; Lannfelt, L; Gibson, G. E.; Cowburn, R. F. Differential regulation of adenylyl cyclase activity in fibroblasts from sporadic and familial Alzheimer’s disease cases with PS-1 and Swedish APP mutations. Neuroreport 8:2031– 2035; 1997.