Enhanced ROS Generation Mediated by Alzheimer's Disease Presenilin Regulation of InsP₃R Ca²⁺ Signaling

Abstract

Familial Alzheimer's disease (FAD) is caused by mutations in amyloid precursor protein and presenilins (PS1, PS2). Many FAD-linked PS mutations affect intracellular calcium (Ca²⁺) homeostasis by proximal mechanisms independent of amyloid production by dramatically enhancing gating of the inositol trisphosphate receptor (InsP₃R) intracellular Ca²⁺ release channel by a gain-of-function effect that mirrors genetics of FAD and is independent of secretase activity. Electrophysiological recordings of InsP₃R in FAD patient B cells, cortical neurons of asymptomatic PS1-AD mice, and other cells revealed they have higher occupancy in a high open probability burst mode, resulting in enhanced Ca²⁺ signaling. Exaggerated Ca²⁺ signaling through this mechanism results in enhanced generation of reactive oxygen species, believed to be an important component in AD pathogenesis. Exaggerated Ca²⁺ signaling through InsP₃R–PS interaction is a disease specific and robust proximal mechanism in AD that may contribute to the pathology of AD by enhanced generation of reactive oxygen species. Antioxid. Redox Signal. 14, 1225–1235.

Introduction

Alzheimer's disease (AD) is a common form of dementia involving slowly developing and ultimately fatal neurodegeneration. Age is the main risk factor but the major molecular mechanisms are still unclear. A hallmark feature of AD is accumulation of extracellular β amyloid (Aβ) plaques, intracellular neurofibrillary tangles, and neuronal loss (19). Mutations in presenilins (PS1 and PS2) and amyloid precursor protein (APP) cause most early-onset, autosomal dominant familial cases of the disease (FAD) (73). Presenilins are transmembrane proteins that are synthesized and localized in the endoplasmic reticulum (ER) membrane (1). Together with nicastrin, APH-1, and PEN-2, PS forms a protein complex that is transported to the cell surface and endosomes, where it functions as a γ-secretase that cleaves several type 1 transmembrane proteins, including APP (14, 45). Mutant PS are believed to affect APP processing by either enhancing the total production of Aβ or the relative proportion of the more amyloidogenic Aβ-42 form (77). In the amyloid hypothesis of AD, accumulation of amyloidogenic Aβ aggregates or oligomers due to defective processing and clearance is a proximal feature that causes neural toxicity leading to brain pathology (27, 29). Identification of three components in FAD: PS1, PS2, and APP that are linked in a biochemical pathway that impinges on Aβ production has strongly influenced acceptance of the amyloid hypothesis (29).

Disrupted intracellular Ca²⁺ homeostasis has also been implicated in AD. Extracellular Aβ influences intracellular Ca²⁺ homeostasis in vitro (33, 53) and in vivo (6, 7, 38, 39). In addition to disrupting APP processing, many FAD-linked PS mutations have been shown to affect Ca²⁺ homeostasis by Aβ independent mechanisms (40, 64). Ca²⁺ signaling disruptions associated with expression of FAD PS have been manifested as attenuated capacitive Ca²⁺ entry (30, 41, 78), but most usually as exaggerated Ca²⁺ release from the ER (30, 43, 65, 70, 72), the major intracellular Ca²⁺ storage organelle. The mechanisms underlying exaggerated ER Ca²⁺ release have been ascribed to enhanced loading of the ER lumen (65) due to disruption of a putative Ca²⁺ channel function of wild-type PS (54, 75) or to enhanced activity of the SERCA Ca²⁺ pump (23). Exaggerated Ca²⁺ release has also been accounted for by enhanced Ca²⁺ liberation from normal stores through ryanodine receptor (RyR) (10, 64, 71) and inositol trisphosphate receptor (InsP₃R) (43, 68) Ca²⁺ release channels. The latter phenotypes have been observed both in vivo (10, 65, 70, 71) and in vitro (17, 31, 35, 42). Enhanced release from normal stores has been attributed either to enhanced Ca²⁺ release channel expression (10, 11, 36, 72) or, in the case of the InsP₃R, to enhanced activity in response to its ligand InsP₃ (12, 13, 35). Regarding the latter, it is notable that enhanced agonist-induced InsP₃R-mediated Ca²⁺ signals have been used diagnostically to identify FAD patient cells (31, 35). A biochemical interaction of the InsP₃R with both wild-type (WT) as well as FAD mutant PS1 and PS2 has been demonstrated (13). Furthermore, single channel recordings of InsP₃R have shown that FAD mutant PS1 and PS2 expression is associated with an apparent sensitization of the InsP₃R channel to InsP₃, resulting in enhanced InsP₃R Ca²⁺ release channel gating (13). The single channel studies were performed in the absence of Aβ or cellular pathology, suggesting that modulation of InsP₃R channel gating is a fundamental mechanism that contributes to exaggerated Ca²⁺ signaling in FAD PS-expressing cells.

Here, we review recent data that suggest that enhanced InsP₃R channel gating and resulting exaggerated Ca²⁺ signaling caused by biochemical and functional interactions of FAD PS and InsP₃R is a disease specific and robust proximal mechanism in AD. In addition, new data are presented that show that exaggerated Ca²⁺ signaling through this mechanism results in enhanced generation of reactive oxygen species (ROS), believed to be an important component in AD pathogenesis (21, 49, 59, 66).

Materials and Methods

Cell culture

DT40 cells stably expressing PS1-WT or PS1-M146L were created and maintained as described (13). B-lymphoblast lines derived from human FAD patients and normal individuals (Coriell Institute, Camden, NJ) were maintained at 37°C (95/5% air/CO₂) in RPMI 1640 (Invitrogen, Carlsbad, CA) supplemented with 15% fetal bovine serum (Hyclone, Logan, UT), 2 mM L-glutamine, 100 units/ml penicillin, and 100 μg/ml streptomycin.

Calcium imaging

Human B-lymphocytes (Coriell Institute) were plated onto a CellTek-(BD Biosciences, San Jose, CA) coated glass-bottom perfusion chamber mounted on the stage of an inverted microscope (Eclipse TE2000; Nikon, Melville, NY) and incubated with fura-2 AM (2 μM; Invitrogen) for 30 min at room temperature in Hanks' balanced salt solution (HBSS, Sigma, St. Louis, MO) containing 1% BSA. Cells were then continuously perfused with HBSS containing 1.8 mM CaCl₂ and 0.8 mM MgCl₂ (pH 7.4). Ca²⁺ signals were elicited by cross-linking the B cell receptor (BCR) with 50 ng/ml anti-human IgM antibody (Southern Biotech, Birmingham, AL). In some experiments, cells were perfused with complete culture medium containing 10% FBS. Fura-2 was alternately excited at 340 and 380 nm, and the emitted fluorescence filtered at 510 nm was collected and recorded (13, 76) using a CCD-based imaging system running Ultraview software (Perkin Elmer, Waltham, MA). Dye calibration was achieved by applying experimentally determined constants to the standard equation [Ca²⁺] = K_d · β · (R − R _min)/(R _max − R).

Electrophysiology

Preparation of isolated nuclei from cells was performed as described (13, 46, 76). In brief, cells were washed twice with PBS and suspended in nuclear isolation solution containing (in mM): 150 KCl, 250 sucrose, 1.5 β-mercapoethanol, 10 Tris-HCl, 0.05 phenylmethylsulphonyl fluoride, and protease inhibitor cocktail (Complete, Roche Diagnosis, Indianapolis, IN), pH 7.3. Nuclei were isolated using a Dounce glass homogenizer and plated onto a 1-ml glass-bottomed dish containing standard bath solution (in mM): 140 KCl, 10 HEPES, 0.5 BAPTA, and 0.192 CaCl₂ (free [Ca²⁺] = 90 nM). The pipette solution contained (in mM): 140 KCl, 10 HEPES, 0.5 dibromo-BAPTA, and 0.001 free Ca²⁺, pH 7.3. Free [Ca²⁺] in solutions was adjusted by Ca²⁺ chelators with appropriate affinities and confirmed by fluorometry as described (13). Data were recorded at room temperature and acquired using an Axopatch 200A amplifier (Molecular Devices, Sunnyvale, CA), filtered at 1 kHz, and digitized at 5 kHz with an ITC-16 interface (Instrutech) and Pulse software (HEKA Instruments Inc., Bellmore, NY).

Hydrogen peroxide generation measurements

DT40 cells stably expressing PS1-WT or PS1-M146L were washed twice with HBSS and suspended in 2 ml HBSS. Amplex Red reagent and HRP were added and the fluorescence intensity was monitored continuously for 20 min (PTI QuantaMaster fluorimeter; Photon Technology Intl, Birmingham, NJ; ex: 530 nm, em: 590 nm). The increase in fluorescence intensity over 20 min in control DT40 cells was used to normalize signals from PS-expressing cells. PC12 cells were transiently transfected with pIRES-EGFP, pIRES-GFP-PS1WT, or pIRES-GFP-M146L and grown in 6-well plates for 48 h. Cells were washed twice in HBSS and suspended in 2 ml HBSS. Amplex Red reagent and HRP were added to the incubation medium and the fluorescence intensity of the supernatant was measured after 30 min. Fluorescence intensity was calculated as (F - F_o)/([protein])/F_GFP, where F is the fluorescence intensity of the cell supernatant, F_o is the fluorescence intensity of the supernatant without cells, and F_GFP is the fluorescence intensity of the cell supernatant in pIRES-EGFP empty vector transfected cells. In some experiments, PC12 cells were treated with 10 μM BAPTA-AM for 30 min to chelate the intracellular Ca²⁺ stores prior to addition of Amplex Red reagents.

Analysis and Statistics

Data were summarized as the mean ± S.E.M. and the statistical significance of differences between means was assessed using unpaired t-tests or analysis of variance (ANOVA) for repeated measures. Differences between means were accepted as statistically significant at the 95% level (p < 0.05).

Results

FAD PS enhancement of InsP₃R gating is a robust phenotype observed in many cells

By co-immunoprecipitation from transfected cells and mouse brain lysates, biochemical interactions of WT and FAD mutant PS1 and 2 with the InsP₃R were identified (12, 13). Functionally, FAD PS had gain-of-function consequences for the activity of the channel that were not observed for WT PS, and which in turn were associated with exaggerated [Ca²⁺]_i signaling in intact cells (12, 13). FAD and WT PS1 were stably expressed in chicken DT40 cell pre-B lymphocytes, and patch clamp experiments were performed on isolated nuclei to record InsP₃R ion channels in their native membrane (46). No novel ion channel activities were observed in nuclei from the wild-type (WT) or mutant PS1-expressing native DT40 cells (13). In saturating [InsP₃], PS1-WT was without effect on single InsP₃R channel activity channel (channel open probability P_o = 0.53 in control versus 0.57 in PS1-WT expressing cells), whereas PS1-M146L expression increased channel activity (P_o = 0.83; p < 0.01; Figs. 1A and 1C) by locking the channel open for long periods (Fig. 1D). In suboptimal [InsP₃] (100 nM), mutant PS1 stimulated the channel activity by four fold (P_o = 0.63 versus 0.16 for control cells; Figs. 1B and 1C) to levels similar to those observed for control channels in saturating [InsP₃] (13). Similar results were obtained with an FAD PS2 mutant (N141I) (12, 13).

FIG. 1.

Effects of PS1 expression on InsP₃R single channel activity in DT40 cells. Representative InsP₃R single-channel current recordings in presence of saturating (10 μM; A) or sub-saturating (100 nM; B) InsP₃ in DT40 cells stably transfected with PS1 WT or M146L. Pipette [Ca²⁺] was 1 μM, optimal for channel activity; arrows: zero current level. Summary of effects of PS1 expression on InsP₃R P_o (C), mean open time (t_o) (D), and mean closed time (t_c) (E). Asterisks: p < 0.01, unpaired t-test. From Cheung et al. (13) with permission from Elsevier. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article at www.liebertonline.com/ars).

Enhanced InsP₃R channel activity appears to be a conserved feature of FAD PS-expressing cells, since similar results were observed in cells expressing other FAD mutant PS (12). γ-secretase-dead mutants also significantly enhanced InsP₃R channel activity, although to a lesser extent than the FAD mutants, indicating that the secretase activity of PS is not required for its effects on InsP₃R gating (12). Importantly, P_o of channels recorded from cells infected with frontotemporal dementia-associated mutant PS1 was not different from controls (12). Thus, several FAD-mutant PS have similar effects on InsP₃R gating, and these effects appear not to be recapitulated in PS mutants associated with a different neurological disease.

These effects of FAD PS on InsP₃R gating appear to be remarkably robust, since they have now been observed in five different cellular systems: Sf9 cells, chicken DT40 B cells, human FAD patient B lymphoblasts and fibroblasts, and neurons from FAD-PS AD mice, in the absence of and preceding disease pathology (12, 13). The strikingly similar effects of mutant PS expression of InsP₃R channels in such diverse cellular systems from different species suggest that they are robust in all cell types and for all InsP₃R isoforms. The electrophysiological studies suggest that FAD PS stimulate InsP₃R gating by a mechanism that involves PS-mediated effective sensitization of the channel to InsP₃, most likely through an allosteric mechanism. Modal gating analysis suggested that FAD mutant PS regulate channel activity by impinging upon the normal ligand activation mechanisms (12). Ligand regulation of InsP₃R gating is largely mediated by altering propensity of the channel to gate in particular modes (34). Strongly activated channels gate in a high-P_o H mode characterized by long bursting activities; an intermediate-P_o I mode is characterized by fast channel openings and closings; and a low-P_o L mode characterized by long closings with brief openings. FAD PS appear to impinge on modal gating. InsP₃R gating was examined in lymphoblasts from three human FAD patients harboring PS1-M146L, PS1-A246E, or PS2-N141I, as well as from two different age-matched unaffected individuals. InsP₃R from the two unaffected individuals had similar low channel P_o (0.18 and 0.23, respectively; Figs. 2A and 2B) with channel activities characterized by brief openings with relatively long closings (Figs. 2A and 2C). Strikingly, InsP₃R P_o recorded from all three FAD patients were profoundly elevated by 2–3-fold when compared with the unaffected individuals (PS1-M146L: 0.62; PS1-A246E: 0.67; PS2-N141I: 0.50; Figs. 2A and 2B), mainly due to a marked decrease in the mean channel closed time t_c (Fig. 2C), with many channels observed to be bursting for extended periods (Fig. 2E). In cells from unaffected individuals, the L and I gating modes dominated channel kinetics, whereas InsP₃R recorded from FAD patient lymphoblasts spent 50%–75% of the time in the high P_o H mode (Figs. 2D and 2E). Analogous results were obtained with low (100 nM) InsP₃. InsP₃R P_o was 0.04 in an unaffected individual, whereas P_o was increased by five-fold to 0.22 in cells from the PS1-A246E patient (Figs. 2F and 2G).

FIG. 2.

Effects of FAD PS on InsP₃R gating in human FAD patient B lymphoblasts. (A) Representative human InsP₃R current traces at +20 mV in isolated nuclei from human FAD lymphoblasts and control lymphobalsts from age-matched individuals without FAD activated with 10 μM InsP₃ and 1 μM Ca²⁺ in pipette solution. Summary of InsP₃R channel P_o (B), τ_o (open circle) and τ_c ( filled circle) (C) and modal gating analysis (D) from different human lymphoblasts. Asterisks: p < 0.05 by ANOVA compared with CTL1. (E) Modal gating analyses of InsP₃R channel from human lymphoblasts. Distinct human InsP₃R gating behaviors in unaffected normal individual vs FAD lymphoblasts with PS1 mutations. Each section shows continuous recording with gating mode assignment in color code below it. In cells from normal individuals, low channel P_o is associated with switching between L and I modes. In cells from all three individuals with FAD, enhanced InsP₃R gating is manifested by increased occupancy of the H mode at the expense of L mode. (F–H) Single InsP₃R channel current traces from human lymphoblasts activated by suboptimal InsP₃. (F) Representative current traces (+20 mV) in isolated nuclei from human FAD lymphoblasts and age-matched control B lymphoblasts activated by suboptimal 100 nM InsP₃ and 1 μM Ca²⁺. Summary of InsP₃R channel P_o (G), and τ_o (open circle) and τ_c ( filled circle) (H) from aged-matched control and FAD human B-lymphocytes. Asterisks: p < 0.05 by Student's t-test. From Cheung et al. (12) with permission from the American Association for the Advancement of Science. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article at www.liebertonline.com/ars).

These observations in human B-lymphoblasts with endogenous levels of PS and InsP₃R are remarkably similar to those observed in single InsP₃R channel recordings in nuclei isolated from insect Sf9 cells, DT40 cells, and embryonic cortical neurons from transgenic mice with human FAD PS1 knocked into the mouse PS1 locus (12, 13). FAD PS enhance InsP₃R gating by mode switching, causing the channel to spend more time in the H mode. Modal gating regulation may have important functional consequences. The channel open time in the L mode is ∼10 ms, which may not be long enough for Ca²⁺ flux through the channel to increase local [Ca²⁺] sufficiently to recruit additional InsP₃R- or RyR-mediated Ca²⁺ release by Ca²⁺-induced Ca²⁺ release (CICR). In contrast, the much longer openings in the H mode (>200 ms) can provide a large Ca²⁺ flux to enable a normally local Ca²⁺ signal to be amplified and propagated by CICR (20). It was suggested (12) that because InsP₃R and RyR are clustered and spatially organized to provide local [Ca²⁺]_i signals as a critical element of physiological specificity, mode-shifting by FAD PS may result not only in exaggerated local Ca²⁺ signaling, but also disruption of spatial specificity by enabling CICR to transmit signals more globally (20, 34). Mode switching by FAD PS of InsP₃R gating may account for observations InsP₃ dependent exaggerated RyR mediated Ca²⁺ signals in neurons (e.g., (71, 72)).

The effects of FAD PS on InsP₃R gating result in altered [Ca²⁺]_i signaling

The effects of FAD PS expression on the behavior of the InsP₃R measured at the single channel level are consistent with the observations of exaggerated InsP₃-mediated [Ca²⁺]_i signals in FAD patient fibroblasts (35) and other cells with mutant PS expressed (68, 69). To address whether the effects observed at the single-channel level were associated with altered [Ca²⁺]_i signaling, InsP₃R-mediated Ca²⁺ signals were measured in the same human FAD patient B cell-derived lymphoblasts that were used for single-channel studies. InsP₃R-mediated Ca²⁺ signals were elicited by cross-linking the B cell receptor (BCR) with IgM antibody. With strong stimulation (5 μg/ml), 20% of cells from both control individual and an FAD patient responded with similar [Ca²⁺]_i oscillations and spiking (Figs. 3B and 3D), whereas a further 27% of FAD patient cells responded with exaggerated high-level transient responses (Figs. 3A–3C). With weak stimulation (50 ng/ml), [Ca²⁺]_i oscillations and spiking were triggered in ∼20% of control cells (Figs. 3E and 3G), due to periodic Ca²⁺ release through the InsP₃R, whereas in cells isolated from FAD patients, both the number of responding cells as well as the oscillation/spiking frequency were increased (Figs. 3E, 3G, and 3H). Perfusion with normal culture medium containing serum that generates low-level InsP₃ production (76) induced spontaneous [Ca²⁺]_i oscillations/spiking in ∼25% of control patient cells (Figs. 3F and 3G). In contrast, the number of PS1 FAD patient B lymphoblasts displaying spontaneous Ca²⁺ oscillations was increased by 100% and the oscillation/spiking frequency doubled (Figs. 3F–3H). These responses are consistent with an enhanced sensitivity and activity of InsP₃-mediated Ca²⁺ release in human FAD patient cells, in concordance with enhanced InsP₃R channel activity recorded in the same cells.

FIG. 3.

Exaggerated Ca²⁺ signaling in human FAD B cells. (A) Representative single cell Ca²⁺ responses to strong anti-human IgM antibody stimulation (5 μg/ml, added at arrow) in control human B-lymphoblasts (CTL) or FAD lymphoblasts carrying PS1-A246E mutation. Dark lines below and to the left of each trace indicate zero Ca²⁺. (B) Percentage of cells responding to anti-IgM stimulation. Red: percentage responding with Ca²⁺ oscillations; blue: percentage responding with large amplitude Ca²⁺ transients. (C) Summary of peak amplitudes of high-amplitude transient Ca²⁺ responses triggered by 5 μg/ml anti-IgM. (D) Ca²⁺ oscillation frequency in response to anti-IgM. Data summarized from 3 experiments with 30 cells in each experiment. Asterisk: p < 0.05 by Student's t test. (E) Representative single cell Ca²⁺ responses to weak anti-human IgM antibody stimulation (50 ng/ml, added at arrow) and (F) spontaneous oscillations during perfusion with serum-containing medium in B lymphoblasts from unaffected individuals (CTL) and FAD lymphoblasts with PS1-M146L, PS1-A246E, or PS2-N141I. Dark lines indicate zero Ca²⁺ level. (G) Summary of percentage of cells responding to anti-IgM stimulation (black bars) or undergoing spontaneous Ca²⁺ oscillations in complete medium with serum (blue bars). (H) Summary of Ca²⁺ oscillation frequency in response to anti-IgM (black bars) or spontaneous Ca²⁺ oscillations observed in presence of complete medium (blue bars). Data in each group summarized from four experiments with 30 cells in each experiment. Asterisks or #: p < 0.05 by ANOVA as compared with respective controls. From Cheung et al. (12) with permission from the American Association for the Advancement of Science. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article at www.liebertonline.com/ars).

FAD PS-mediated enhanced [Ca²⁺]_i signaling causes enhanced ROS production

InsP₃-mediated [Ca²⁺]_i signals regulate many cell physiological processes. A major question is whether the effects of FAD mutant PS on InsP₃R-mediated Ca²⁺ signaling have relevance for disease pathogenesis. It is possible that exaggerated Ca²⁺ signals in AD may influence mitochondrial function, gene transcription, and Aβ production (57), features associated with AD. Ca²⁺ released from the ER through the InsP₃R can be rapidly taken up by mitochondria located in close proximity (60). Periodic release of Ca²⁺ from the ER stimulates mitochondrial enzymes of the TCA cycle, leading to the maintenance of efficient oxidative phosphorylation and ATP production (15, 28). However, in response to excess Ca²⁺ release, the Ca²⁺-induced increase in mitochondrial NADH can exceed that needed to sustain oxidative phosphorylation, and lead to an electron-saturated respiratory chain that results in increased mitochondrial ROS production. Oxidative stress has been invoked in AD and as a link between aging and neurodegeneration (22, 48, 55, 56). The identification of the InsP₃R as a molecular target of PS regulation suggests a mechanism that may link mutations in PS to enhanced oxidative stress.

Because elevated rates of mitochondrial electron transport result in enhanced generation of superoxide that is rapidly dismutated to H₂O₂ that leaks through plasma membranes to the extracellular solution, we measured ROS production by recording the amount of H₂O₂ released from cells under basal conditions and in response to agonist stimulation. Stable expression of either PS1-WT or PS1-M146L did not affect the basal production of H₂O₂ by DT40 cells (Figs. 4A and 4B). In contrast, BCR stimulation with a low dose of IgM (0.05 μg/ml) significantly enhanced H₂O₂ production in PS-expressing cells, most markedly in the PS1-M146L-expressing cells (Figs. 4A and 4B). Enhanced H₂O₂ production observed in PS1-expressing cells was entirely dependent on the expression of InsP₃R, because it was absent in InsP₃R-deficient DT40-KO cells (Fig. 4C). These results indicate that the agonist-stimulated production of ROS and, importantly, the potentiation by mutant PS1, are caused by InsP₃R-mediated Ca²⁺ release.

FIG. 4.

Effects of PS1 expression on ROS generation in DT40 and PC12 cells. (A) Continuous monitoring of H₂O₂ generation by Amplex Red fluorescence intensity in control (blue), and stably transfected PS1-WT (red) and M146L (green) expressing DT40 lines. Colored lines indicate mean intensity of three experiments, and gray area shows s.e.m of traces. Arrows indicate addition of anti-IgM or vehicle. (B) Summary of H₂O₂ generated after 20 min in presence or absence of 50 ng/ml BCR antibody. Amplex Red fluorescence normalized to that observed at time 0 in unstimulated WT cells. (C) Summary of H₂O₂ generated after 20 min in presence or absence of 50 ng/ml BCR antibody in InsP₃R KO DT40 lines. Amplex Red fluorescence normalized to that observed at time 0 in unstimulated WT cells. (D) Summary of H₂O₂ generated over 30 min in the presence or absence of 100 μM ATP by PC12 cells expressing GFP, PS1-WT, or PS1-M146L. Amplex Red fluorescence normalized to that observed at time 0 in unstimulated GFP-transfected cells. (E) Summary of H₂O₂ generated over 30 min in presence or absence of 100 μM ATP by PC12 cells expressing GFP, PS1-WT, or PS1-M146L with or without pretreatment with 10 μM BAPTA-AM. Amplex Red fluorescence normalized to that observed at time 0 in unstimulated GFP-transfected cells. (F) Summary of H₂O₂ generated over 30 min in the presence or absence of 1 μM ATP by PC12 cells expressing GFP, PS1-WT, or PS1-M146L. Amplex Red fluorescence normalized to that observed at time 0 in unstimulated GFP-transfected cells. Single and double asterisks and number symbols indicate p < 0.05 or p < 0.01, respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article at www.liebertonline.com/ars).

Analogous experiments were performed in neuronal PC12 cells transfected (efficiency ∼50%) with either PS1-WT or PS1-M146L cDNAs, and stimulated with a high (100 μM) and low (1 μM) dose of ATP to activate P2Y2 receptors (74). The responses observed in PC12 cells with a high dose of ATP were similar to those in DT40 cells, except that basal production of H₂O₂ was enhanced by PS1 expression (Figs. 4D and 4F), with the effect more pronounced in the cells expressing PS1-M146L (Fig. 4F). Because FAD PS effects on InsP₃R channel gating are more pronounced at low levels of InsP₃, we also performed experiments using a sub-saturating dose of ATP (1 μM). PC12 cells expressing PS1-M146L generated significantly higher H₂O₂ than either control or PS1-WT cells (Fig. 4F). The agonist-induced elevations of H₂O₂ production were completely inhibited in cells loaded with BAPTA to buffer changes in [Ca²⁺]_i (Fig. 4E). These results indicate that FAD-associated PS1-M146L promotes ROS production, and the production is dependent on InsP₃R-mediated Ca²⁺ release.

Discussion

Whereas observations of disrupted Ca²⁺ signaling in AD have been well documented, the physiological implications of altered Ca²⁺ signaling and the roles that these changes play in AD pathogenesis are unknown. The identification of the InsP₃R as a molecular mechanism of Ca²⁺ disruption associated with expression of FAD PS may provide new insights. In both DT40 and PC12 cells, expression of PS1-M146L caused an exaggerated ROS production in response to agonists that mobilized Ca²⁺ through the InsP₃R. This effect of mutant PS1 was mediated by Ca²⁺ release through the InsP₃R because it was abolished both in cells that lacked InsP₃R expression and by blocking InsP₃R-mediated elevations of [Ca²⁺]_i. These results therefore demonstrate that exaggerated Ca²⁺ signals caused by mutant PS1 stimulation of InsP₃R gating result in enhanced ROS generation. Accordingly, the sensitization by mutant PS of the InsP₃R to low [InsP₃] that may exist in unstimulated cells might cause chronically enhanced ROS generation. In agreement, enhanced ROS generation was observed in unstimulated PS1-expressing neuronal PC12 cells. These results therefore suggest a molecular mechanism by which mutant PS could exacerbate oxidative stress, independent of effects of Aβ.

ROS are generated at low levels in most cells during oxidative phosphorylation in mitochondria. ROS can also be generated by NADPH oxidase and xanthine oxidase (74), but mitochondria are generally considered to be the source of most ROS in the cell, especially in the brain. Ca²⁺ released from the ER in response to agonist stimulation (61, 67), as well as by constitutive low-level InsP₃R activity (8) can be taken up by mitochondria where it enhances oxidative phosphorylation by stimulation of dehydrogenases to produce NADH, or the F1-F0-ATPase directly (3, 50). What is not clear from our studies are the functional consequences of the enhanced ROS generated as a consequent of FAD PS-mediated enhanced InsP₃R Ca²⁺ release activity that we have observed. ROS at low levels can have important signaling functions (5, 18). For example, ROS generation is important for hippocampal long-term potentiation (LTP) (63), and H₂O₂ can modify the redox state of RyR by increasing its S-glutathionylation, potentiating Ca²⁺ release and enhancing ERK and CREB phosphorylation that may impinge on gene transcription (37). Alternately, excessive ROS generation can damage proteins, lipids, and DNA. Excessive ROS generation and neurodegeneration have been linked in many studies (4, 62). Oxidative stress can trigger cell cycle activation in mature neurons, and such activation leads to cell death (32). A large body of evidence indicates that damage from ROS is a common and prominent feature in the brain in AD (21, 49, 59, 66). Aging is the major risk factor for nonfamilial forms of AD, and cumulative cellular damage caused by mitochondrial ROS generation is the basis of the “mitochondrial theory” of aging (52). Brain metabolism is reduced in AD and the activities of several mitochondrial enzymes are diminished (49, 59, 66). It has been proposed that these features may be a consequence of mitochondrial damage due to oxidative stress (22). Accordingly, excessive ROS production associated with FAD PS expression may accelerate this process. Several studies have suggested that therapeutic antioxidant strategies may be beneficial in AD (22, 55). Much of the focus on ROS in AD has been on Aβ, and its association with increased production of ROS and impairment of anti-oxidant defenses (2, 47, 58). ROS responses of cells to exogenous Aβ are enhanced in cells expressing FAD mutant PS (24–26). Nevertheless, some evidence suggests that oxidative stress in the brain may be an early event that precedes even mild cognitive decline (55). Furthermore, enhanced ROS generation has been observed in peripheral cells from AD patients and transgenic mice in the absence of Aβ (9, 16, 44, 51). Accordingly, we hypothesize that life-long enhanced InsP₃R-mediated Ca²⁺ release is a molecular mechanism, at least in FAD, that drives excessive mitochondrial ROS production with consequent pathological sequelae that result in neuronal impairment and cognitive decline. Testing this hypothesis will require the development of appropriate animals models with features of AD-associated oxidative stress and brain InsP₃R activity modified.

In summary, FAD mutant PS interact with the InsP₃R Ca²⁺ release channel, and through a gain of function, enhance its activity to low levels of InsP₃ that may exist constitutively in resting cells, causing exaggerated Ca²⁺ signaling in both the temporal and spatial domains. This Ca²⁺ signaling pathway impinges on mitochondria, enhancing the production of ROS that may, over time, contribute to the pathology of AD.

Abbreviations Used

Aβ

beta-amyloid

AD

Alzheimer's disease

APP

amyloid precursor protein

ATP

adenosine 5’-triphosphate

BAPTA

1,2-bis(o-aminophenoxy) ethane-N,N,N',N'-tetraacetic acid

BCR

B cell receptor

Ca²⁺

calcium

CICR

calcium-induced calcium release

ER

endoplasmic reticulum

FAD

familial Alzheimer's disease

H₂O₂

hydrogen peroxide

InsP₃

inositol 1,4,5-trisphosphate

InsP₃R

inositol 1,4,5-trisphosphate receptor

KO

knock-out

LTP

long-term potentiation

PS1

presenilin 1

PS2

presenilin 2

ROS

reactive oxygen species

RyR

ryanodine receptor

SERCA

sarco/endoplasmic calcium ATPase

TCA

tricarboxylic acid

WT

wild type

Acknowledgments

Acknowledgement is made to the donors of Alzheimer's Disease Research, a program of the American Health Assistance Foundation (A2008-137 to J.K.F.), NIH GM56328 and MH059937 to JKF and the Alzheimer's Disease Core Center at the University of Pennsylvania (AG 10124).

Author Disclosure Statement

No competing financial interests exist.