Current Neurovascular Research, 2005, 2, 3-12
3
Amyloid Beta Peptide 1-40 Stimulates the Na+/Ca2+ Exchange Activity of SNCX Menjor Tino Unlap*, Corey Williams, Darryl Morin, Brian Siroky, Attila Fintha, Amanda Fuson, Layla Dodgen, Gergely Kovacs, Peter Komlosi, William Ferguson and Phillip Darwin Bell Nephrology Research and Training Center, Departments of Medicine and Physiology, Division of Nephrology, University of Alabama at Birmingham, Birmingham, AL 35294, USA Abstract: The Na+/Ca2+ exchangers, RNCX and SNCX, were cloned from mesangial cells of salt sensitive and salt resistant Dahl/Rapp rats, respectively, and differ at amino acid 218 (RNCXi/SNCXf) and in the exons expressed at the alternative splice site (RNCXB, D /SNCXB, D, F). These isoforms are also expressed in myocytes, neurons, and astrocytes where they maintain cytosolic calcium homeostasis. We demonstrated that cells expressing SNCX were more susceptible to oxidative stress than cells expressing RNCX. Others demonstrated that amyloid β peptide (Aβ) augments the adverse effects of oxidative stress on calcium homeostasis. Therefore, we sought to assess the effect of Aβ 1-40 on the abilities of OK-PTH cells stably expressing RNCX and SNCX and human glioma cells, SKMG1, to regulate cytosolic calcium homeostasis. Our studies showed that Aβ 1-40 (1 µM) did not affect RNCX activity, as assessed by changes in [Ca2+]i (∆[Ca2+]i, 260 ± 10 nM to 267 ± 8 nM), while stimulating exchange activity 2.4 and 3 fold in cells expressing SNCX (100 ± 8 to 244 ± 12 nM) and in SKMG1 cells (90 ± 11 nM to 270 ± 18 nM), respectively. Our results also showed that Aβ 140, while not affecting the rate of Mn2+ influx in cells expressing RNCX, stimulated the rate of Mn2+ influx 2.8 and 2.9 fold in cells expressing SNCX and in SKMG1 cells. Thus, our studies demonstrate that Aβ-induced cytosolic calcium increase is mediated through certain isoforms of the Na+/Ca2+ exchanger and reveals a possible mechanism by which Aβ 1-40 can alter cytosolic calcium homeostasis.
Key Words: Na+/Ca2+ exchanger, amyloid beta peptide, neurotoxicity, neuronal degeneration, cytosolic calcium. INTRODUCTION Alzheimer’s disease (AD) is characterized by neuronal degeneration, the basis of which is not fully known. Studies have shown that, in AD, cytotoxic effects consisting of dysregulation of cytosolic calcium homeostasis which leads to sustained elevation in cytosolic calcium, exacerbation of oxidative stress, and apoptosis are evident and may play a role in the neuronal degeneration that is a hallmark of AD (Barger et al. 1993; Ho et al. 2001a; Chan et al. 2002). These cytotoxic effects are attributed to proteolytic products of the amyloid precursor protein (APP), including amyloid beta peptide or Aβ (Aβ 1-40, Aβ 1-42 and Aβ 25-35) and CT105 (Ho et al. 2001a; Huang et al. 2000a; Kim et al. 1999). One of the early signs of AD is the appearance of amyloid plaques whose development is closely linked to activated astrocytes and microglia (Nagele et al. 2004). Astrocytes serve as debris clearing cells in neurodegeration and, when overburdened by excessive levels of debris including amyloid beta peptides, undergo lysis and release these peptides which can accumulate and form astrocytic amyloid plaques (Nagele et al. 2003; Nagele et al. 2004). Excessive levels of amyloid peptides can also induce elevation of cytosolic calcium and reactive oxygen species
*Address correspondence to this author at the University of Alabama at Birmingham, UAB Station, 865 Sparks Center, Birmingham, AL 35294, USA; Tel: (205) 934-3806; Fax: (205) 934-1147; E-mail:
[email protected] Received: June 15, 2004; Revised: October 1, 2004; Accepted: October 4, 2004
1567-2026/05 $50.00+.00
and apoptosis (Ho et al. 2001a; Huang et al. 2000a; Kim et al. 1999). The Aβ-induced cytosolic calcium increase is mediated, in part, through oxidative stress. In human neuroblastoma cells, SH-SY-5Y, Aβ induced elevation of cytosolic calcium, exacerbation of oxidative stress, and apoptosis (Ho et al. 2001b). These cytotoxic effects were augmented in the presence of the neurotoxic amino acid, homocysteine, but abrogated by pre-incubation with the antioxidant, vitamin E (Ho et al. 2001b). Also, the H2O-soluble hydroxyl and superoxide radical scavenger, fullerenol-1, dose-dependently (100 nM-100 µM) attenuated the Aβ- or CT105-induced cytosolic calcium increase in pheochromocytoma cells (PC12) (Huang et al. 2000b) and in primary cortical neurons (Huang et al. 2000a) and abolished apoptosis. Thus, the cytotoxic effects of Aβ are mediated, in part, through increased oxidative stress which elevates cytosolic calcium that results in apoptosis. The Aβ-induced cytosolic calcium increase is also mediated, in part, through dysregulation of Na+/Ca2+ exchange. In the human neuroblastoma cell line, SK-N-SH, one of the proteolytic products of APP, CT105, attenuated Na+/Ca2+ exchanger activity as assessed in the reverse mode (Kim et al. 1999). In rat and human brain vesicles, Aβ 25-35 inhibited Na-gradient dependent 45Ca2+ uptake by 50%, an effect which was blocked by Congo Red but unaffected by vitamin E (antioxidant) and N-tert-butyl-alpha-phenylnitrone (spin trapping agent) (Wu et al. 1997) which appears to indicate a direct interaction between Aβ 25-35 and the ©2005 Bentham Science Publishers Ltd.
4
Current Neurovascular Research, 2005, Vol. 2, No. 1
Na+/Ca2+ exchanger. In PC12 cells, the Aβ-induced cytosolic calcium elevation was abrogated in the absence of extracellular calcium but was not affected by preincubation with the Ca-ATPase inhibitor, thapsigargin, which empties intracellular calcium storage (Huang et al. 2000b). In addition, the Aβ-induced cytosolic calcium increase was not affected by ligand-gated or voltage-gated calcium channel blockers, the NMDA receptor antagonist MK801, cholera toxin, or by pertusis toxin (Huang et al. 2000b). These results demonstrate that Aβ attenuate Na+-gradient dependent Ca2+ entry that is likely to occur through the Na+/Ca2+ exchanger. We previously demonstrated that oxidative stress attenuated the activities of the two exchanger isoforms that we cloned from mesangial cells of salt sensitive (SNCX) and salt resistant (RNCX) Dahl/Rapp rats (Unlap et al. 2000; Unlap et al. 2002; Unlap et al. 2003). Our studies demonstrated that SNCX showed diminished capacity to reduce agonist induced cytosolic calcium increase (Hwang et al. 2003), enhanced sensitivity to changes in pH (Williams et al. 2004), and enhanced susceptibility to oxidative stress than RNCX (Unlap et al. 2002; Unlap et al. 2003). These two isoforms are also expressed in neurons and astrocytes where they play a critical role in regulating cytosolic calcium (Blaustein et al. 1991; Li and Lytton 2002). Therefore, this study was carried out to assess the effects of Aβ (1-40) on these two exchanger isoforms in order to study the potential role that these exchangers might play in mediating the cytotoxic effects of Aβ in neuronal degeneration. METHODS Cell Culture Opossum proximal tubule kidney cells (OK-PTH, ATCC) and human astrocyte derived glioblastoma cells (SKMG1) (Hatano et al. 2004) were grown in MEM and DMEM/F12 media (Life Technologies, Inc., Rockville, MD) supplemented with 10% and 20% Fetalclone III (Mediatech, Inc., Herndon, VA), respectively, containing 240 µg/ml Lglutamine, 82 units/ml penicillin, and 82 µg/ml streptomycin (Life Technologies, Inc.) in a humidified atmosphere under 95% air/5% CO 2 at 37 C. Media was changed twice a week and cells were routinely passaged 72 hrs after seeding.
°
Reverse and Forward Mode Na+/Ca2+ Exchange Activity Measurements Vector-transfected OK-PTH cells (VOK), transfected OK-PTH cells expressing RNCX (ROK) or SNCX (SOK), and SKMG1 cells were grown to 80% confluency on cover slips in 60 mm cell culture dishes with MEM (OK-PTH cells) or DMEM/F12 media (SKMG1) supplemented with 10% Fetalclone III and 82 µg/ml penicillin/streptomycin. The growth media was removed and cover slips were incubated in 3 mls of serum free MEM containing 24 µM fura 2-AM (TEF Labs, Austin, TX) for 1 hour at 37°C to allow loading of the cells with dye. For reverse-mode (Na+ efflux/Ca2+ influx) Na +/Ca2+ exchange activity measurement, Na+-dependent Ca2+-uptake was assessed in the presence and absence of Aβ 1-40 (1 µM) by replacing the fura-2 solution with 3 mls of 150-Ringer solution (150 mM NaCl, 5 mM KCl, 1 mM MgSO 4, 1.6 mM Na 2HPO4, 0.4 mM NaH2PO4, 5
Unlap et al.
mM D-glucose, 1.5 mM CaCl2, 10 mM HEPES) with 3 mls of 150-Ringer solution containing 10 µM monensin (SigmaAldrich, St Louis, MO) and 400 µM of ouabain (SigmaAldrich) to Na+-load the cells. Aβ 1-40 was obtained from rPeptide (Athens, GA), reconstituted in 100 mM Tris-HCl, pH 7.4, at 1 µM, incubated at 4oC for 12 hrs and used at a final concentration of 1 µM. Cover slips were placed in cuvettes and Na+-dependent Ca2+ uptake was assessed by perfusing cells with 150-Ringer solution to establish a baseline followed by perfusion with 0-Ringer solution (150 mM N-methyl-D-glucamine, 5 mM KCl, 1 mM MgSO4, 1.6 mM Na2HPO4, 0.4 mM NaH 2PO4, 5 mM D-glucose, 1.5 mM CaCl2, 10 mM HEPES) at a rate of 1.7 ml/min to elevate cytosolic Ca2+ ([Ca2+]i). The forward (Na+ influx/Ca 2+ efflux) mode was assessed by examining the abilities of the fura-2 loaded cells to regulate basal and ATP-induced [Ca2+]i increase in the presence and absence of Aβ 1-40 (1 µM) in non Na+-loaded cells. The effect of Aβ 1-40 on basal cytosolic [Ca2+]i was examined by perfusing cells with 150-Ringer solution in the presence and absence of 1 µM of Aβ 1-40. The effects of Aβ 1-40 on the abilities of the cells to reduce ATP-induced [Ca2+]i increase was assessed by perfusing cells with 150Ringer solution followed by the addition of 1mM ATP to elevate [Ca2+]i in presence or absence of 1 µM Aβ 1-40. [Ca2+]i measurements were performed using dualexcitation wavelength fluorescence spectroscopy using a DeltaScan (Photon Technologies, Princeton, NJ). Excitation wavelengths were set at 340 nm and 380 nm and alternated at 25 Hz. Emission wavelength was set at 510 nm, with data collection at a rate of 1 point/sec using PTI software. Background corrections were made prior to the experimental measurements. Baseline fura-2 ratios were measured for at least 100 sec in cells that were perfused with 150-Ringer solution at a rate of 1.7 ml/min. Cells were discarded if the baseline drifted either up or down. After obtaining a stable baseline reading, reverse mode or forward mode exchanger activity was induced as previously described. Fura-2 ratio was monitored continuously before, during, and after inducing exchanger activity until the ratio returned to a stable baseline. All solutions had pH of 7.4 with temperature maintained at 37°C. In addition, there was no evidence of dye leakage throughout the experiment. Calibration of [Ca2+]i Calibrations were performed to convert fura 2 ratios into [Ca2+]i values. [Ca2+]i was calculated using the equation described by Grynkiewicz et al.: (Grynkiewicz et al. 1985). [Ca2+]i = Kd × (S f2/Sb2) × (R-R min)/(Rmax-R), where Kd is the effective dissociation constant of fura 2, and has a value of 224 nM, R is the fluorescence ratio obtained at 340 nm/380 nm, Rmin and Rmax are the ratios in absence and presence of Ca2+ respectively, and Sf2 and Sb2 are the emissions at 380 nm in the absence and presence of Ca2+, respectively. SKMG1 cells and OK-PTH cells expressing the vector, RNCX, or SNCX were loaded with 24 µM fura 2 for 1 hr, followed by resuspension in 150-Ringer solution. Calibration was accomplished after permeabilizing the cells with 5 µM ionomycin and measuring fluorescence at both wavelengths, 340 and 380 nm, under Ca2+-free (in 2 mM
Aβ Stimulates Exchanger Activity
Current Neurovascular Research, 2005, Vol. 2, No. 1 5
EGTA) or Ca2+-saturated (in 0.25 M CaCl2) conditions to obtain Rmin, Rmax, Sf2, and Sb2.
concentrations, A586 for the samples were converted to µM of MDA and compared between control and treated samples.
Mn2+-Influx Measurement
Statistics
+
2+
The Na /Ca exchanger has the same Km and Vmax for Ca2+ and Mn2+ and can transport these two divalent ions, in exchange for Na+, with the same efficiency (Frame and Milanick 1991; Milanick and Frame 1991). Therefore, an assay was set up to determine the ability of OK-PTH cells expressing the vector, RNCX or SNCX and SKMG1 cells to transport Mn2+ into the cell in the presence and absence of Aβ 1-40 (1 µM). Cells were loaded with 24 µM fura 2 for 1 hr, washed for 5 min in 150-Ringer solution and [Ca2+]i measurements were performed using a DeltaScan (Photon Technologies, Princeton, NJ). Excitation wavelengths were set at 340 nm and 359 nm and alternated at 25 Hz. Emission wavelength was set at 510 nm, with data collection at a rate of 1 point/sec using PTI software. Fura-2 fluorescence at 359 nm was measured for 400 sec in 150-Ringer solution and after obtaining a stable baseline reading, cells were perfused for 400 sec in 150-Ringer solution containing 50 µM MnCl 2 with or without Aβ 1-40 (1 µM). The slope of the fluorescence tracing at 359 nm in the presence of MnCl2 was compared between treated (with Aβ) or control (without Aβ). The effect of KB-R7943, a Na+/Ca2+ exchange inhibitor, on Aβ-induced Mn 2+ uptake was assessed by measuring Mn2+ uptake in the presence or absence of 30 µM KB-R7943 (Elias et al. 2001; Isaac et al. 2002). Cell Viability Assay To assess the viability of OK-PTH cells expressing RNCX or SNCX and SKMG1 cells in the presence of Aβ, cells were grown to confluence on cover slips and either left untreated or treated with Aβ 1-40 (1 µM) for 30 minutes, washed, and perfused with 150-Ringer solution in the presence of the fluorescence dye 2', 7'-bis-(2-carboxyethyl)5-(and-6)-carboxyfluorescein, acetoxymethyl ester (BCECF, AM). The ability of cells to take up BCECF, to hydrolyze the ester, and to retain the fluorophore intracellularly is routinely used to assess cell viability (Sellers et al. 1994). BCECF fluorescence was measured by dual-excitation-wavelength fluorescence spectroscopy (Photon Technology International). Excitation wavelengths were set at 440 and 500 nm, and emission wavelength was set at 530 nm. Data points were collected, with Photon Technology International software, at a rate of 1 point per second and expressed as counts per second. Effect of Amyloid Beta Peptide on Oxidative Stress Level To test if the effects of Aβ on SNCX activity were mediated through oxidative stress, the degree of oxidative stress in OK-PTH cells expressing RNCX or SNCX and SKMG1 cells was assessed using the BIOXYTECH® MDA586™ kit from OxisResearch (Portland, OR). This is a colorimetric analysis kit which measures the level of malondialdehyde (MDA) which reflects the degree of lipid peroxidation and thus oxidative stress in cells. Cells were either treated or not treated with 1 µM Aβ 1-40 for 30 min followed by lysis and measurement of MDA level in cell lysates at 586 nm. Using a standard curve of known MDA
Data analysis was carried out using ANOVA and an unpaired t test. The results are presented as mean ± SEM. A p value of
