Altered processing of amyloid precursor protein ... - Wiley Online Library

Journal of Neurochemistry, 2002, 83, 1262–1271

Altered processing of amyloid precursor protein in the human neuroblastoma SH-SY5Y by chronic hypoxia Nicola J. Webster, Kim N. Green, Chris Peers and Peter F. T. Vaughan Institute for Cardiovascular Research, University of Leeds, Leeds, UK

Abstract Alzheimer’s disease (AD) is more prevalent following an ischemic or hypoxic episode, such as stroke. Indeed, brain levels of amyloid precursor protein (APP) and the cytotoxic amyloid b peptide (Ab) fragment are enhanced in these patients and in animal models following experimental ischaemia. We have investigated the effect of chronic hypoxia (CH; 2.5% O2, 24 h) on processing of APP in the human neuroblastoma, SH-SY5Y. We demonstrate that constitutive and muscarinic-receptor-enhanced secretion of the a-secretase cleaved fragment of APP, sAPPa, was reduced by 60% in CH cells. The caspase inhibitor BOC-D(Ome)FMK did not reverse this effect of CH, and CH cells were as viable as controls, based on MTT assays. Thus, loss of sAPPa is not

related to cell death or caspase processing of APP. Preincubation with antioxidants did not reverse the effect of CH, and the effect could not be mimicked by H2O2, discounting the involvement of reactive oxygen species in hypoxic loss of sAPPa. CH did not affect muscarinic activation of extracellular-signal regulated kinase. However, expression of ADAM 10 (widely believed to be a-secretase) was decreased approximately 50% following CH. Thus, CH selectively decreases processing of APP by the a-secretase pathway, most likely by decreasing levels of ADAM 10. Keywords: ADAM 10, amyloid precursor protein, antioxidants, chronic hypoxia, western blot. J. Neurochem. (2002) 83, 1262–1271.

A characteristic feature of Alzheimer’s disease (AD) is the accumulation, in neuronal senile plaques, of an aggregation of a 40–42 amino acid peptide, the amyloid b-peptide (Ab), derived from the amyloid precursor protein (APP) by the sequential action of b- and c-secretases (De Strooper and Annaert 2000; Selkoe 2001). In an alternate pathway, APP is cleaved within the Ab domain by a-secretase to form a large (100-kDa) N-terminal fragment (sAPPa), which is secreted into the extracellular medium. Processing of APP by the a-secretase pathway not only precludes the formation of Ab (Mattson 1997; De Strooper and Annaert 2000; Selkoe 2001) but also gives rise to sAPPa, which has been reported to have neuroprotective properties (Mattson et al. 1993; Smith-Swintosky et al. 1994; Furukawa et al. 1996). One hypothesis to account for neuronal damage associated with AD is that processing of APP by the b- and c-secretase pathway puts neurones at risk due to a decrease in the production of this neuroprotective sAPPa (Mattson et al. 1993). Thus, any change in the regulation of the processing of APP which decreases the a-secretase pathway could lead to the development of AD by either an increased production of Ab or increased neurotoxicity due to a decreased sAPPa formation.

Ischemic episodes represent one example of cellular stress that can modify APP processing. For example, APP and Ab levels have been shown to be elevated following mild and severe brain ischemia in post-mortem samples of human AD brain and in animal models using experimental ischaemia (Abe et al. 1991; Kalaria et al. 1993; Kogure and Kato 1993; Tomimoto et al. 1994; Koistinaho et al. 1996; Yokota et al. 1996; Jendroska et al. 1997). In addition, cerebral ischemia and cerebrovascular disease, such as stroke, have been reported to increase the risk of dementia, in particular AD, and stroke has been reported to enhance cognitive decline in patients with early symptoms of AD (Kalaria 2000). The mechanism underlying such alterations in APP processing following a period of ischemia, however, is unknown.

1262

Received August 12, 2002; accepted August 20, 2002. Address correspondence and reprint requests to Dr P. F. T. Vaughan, Wolfson Centre for Age-related Diseases, Guy’s, King’s and St Thomas’s School of Biomedical Sciences, 3rd Floor Hodgkin Building, King’s College, St Thomas’s Street, London SE1 9RT. E-mail: [email protected] Abbreviations used: AD, Alzheimer’s disease; APP, amyloid precursor protein; CH, chronic hypoxia/chronically hypoxic; ROS, reactive oxygen species; sAPPa, soluble a-secretase-derived N-terminal fragment of APP.

2002 International Society for Neurochemistry, Journal of Neurochemistry, 83, 1262–1271

Hypoxic modulation of APP processing 1263

Cerebral ischemia can compromise neuronal function as a result of decreased substrate delivery, accumulation of metabolic products, acidosis or oxygen deprivation (hypoxia). Our previous studies have shown that a period of chronic hypoxia (CH) enhances depolarisation-evoked catecholamine secretion from PC12 cells (Taylor et al. 1999) and depolarisation- and muscarine-evoked noradrenaline release from the human neuroblastoma SH-SY5Y (Webster et al. 2001). Interestingly, part of the mechanism by which CH enhances catecholamine secretion from PC12 cells is via the formation of a novel Ca2+ channel associated with insertion of Ab in the plasma membrane (Taylor et al. 1999; Green and Peers 2001). This observation suggested that one effect of CH is to alter the balance of APP processing in favour of the b- and c-secretase pathway, thus leading to an increase in Ab formation and a decrease in sAPPa secretion. Prolonged periods of hypoxia are associated with apoptosis (Tanaka et al. 1994; Choi 1996; Bossenmeyer et al. 1998) and recent studies have suggested a link between apoptosis and the regulation of APP processing. For example, LeBlanc 1 (1995) and Galli et al. (1998) have reported that APP and Ab formation are enhanced during neuronal apoptosis and that the nonamyloidogenic (a-secretase) pathway decreases. Furthermore, there is evidence that APP and the AD-associated presenilin molecules (PS-1 and PS-2) are substrates for caspases, the activation of which is a prerequisite for the initiation of apoptosis. Cleavage of APP by caspases leads either to production of Ab (Barnes et al. 1998; Gervais et al. 1999), or to the formation of fragments distinct from those produced by secretases (Pellegrini et al. 1999; Weidemann et al. 1999). Both PS-1 and PS-2 are cleaved by caspase-3 during apoptosis (Kim et al. 1997; Loetscher et al. 1997; Vito et al. 1997) which, given the role of these proteins in APP processing (De Strooper et al. 1998), suggests an additional mechanism by which apoptosis could regulate production of Ab. Thus, during apoptosis APP can be cleaved by alternate processing pathways, in particular cleavage of APP by caspases decreases the formation of the a-secretase product, sAPPa. The present study has examined directly the effect of CH on the regulation of APP processing. Our results show that CH decreases sAPPa secretion in both SH-SY5Y and PC12 cells by a mechanism not involving apoptosis but very likely involves a decrease in the level of ADAM 10, a candidate protein for the a-secretase (Lammich et al. 1999).

Paisley, Scotland, UK), supplemented with 10% (v/v) fetal calf serum, 1% (v/v) non-essential amino acids and 0.1% (v/v) gentamicin. Cells were incubated at 37C in a humidified incubator gassed with air and 5% CO2, passaged every 7 days and used for up to 20 passages. Cells used for the study of sAPPa secretion were harvested in phosphate-buffered saline without Ca2+ or Mg2+ (PBSCa2+) and subcultured into 12-well tissue culture plates at a seeding density of 1 · 105 cells/mL. For immunocytochemistry, cells were subcultured into 6-well plates containing sterile 22 · 22 mm glass coverslips at a final concentration of 1 · 104 cells/mL. PC12 cells were originally obtained from the American Tissue Type Cell Collection, Manassas, VA, USA. Cells were cultured in RPMI-1640 2 medium containing L-glutamine (Invitrogen, Paisley, UK) with 20% FCS and 1% pencillin-streptomycin (Invitrogen). Cells were incubated in a 37C humidified incubator gassed with air and 5% CO2 and passaged every 7 days by centrifuging at 70 g for 5 min and resuspending in fresh media in 75 cm2 flasks. Cells used for the study of sAPPa secretion were centrifuged at 70 g for 5 min, resuspended 1 : 4 in fresh media in 25 cm2 flasks and cultured for 3–4 days. SH-SY5Y and PC12 cells maintained under chronically hypoxic conditions (CH) were subcultured in the same way but for 24 h prior to experiments were transferred to an environment continuously gassed with 2.5–10% O2, 5% CO2, balanced with N2 at 37C. Corresponding control cells were maintained in a 95% air, 5% CO2 incubator for the same period.

Materials and methods

Determining sAPPa secretion In order to determine sAPPa secretion, confluent layers of SH-SY5Y cells, in 12-well plates, were rinsed, twice, with PBS + Ca2+ (800 lL) followed by incubation with either muscarine (50 lM) or 12-O-tetradecanoylphorbol-13-acetate (TPA; 100 nM) in Hepes-buffered saline (HBS; composition in mM: NaCl 35, KCl 5, MgCl2 0.6, CaCl2 2.5, HEPES 10, glucose 6, pH 7.4; 400 lL) for 1 h. Constitutive sAPPa secretion was determined by incubating cell layers in HBS alone. PC12 cells from one 25 cm2 flask were collected by centrifuging at 70 g for 5min, washed by resuspending in PBS + Ca2+ and then incubated in HBS (1 mL) for 1 h. After 1-h incubation, the HBS from both SH-SY5Y and PC12 cells was collected and centrifuged at 15 000 g for 2 min to remove cell debris. Supernatants were lyophilised, re-suspended in SDS sample buffer (200 lL; 125 mM Tris/HCl, 2% (w/v) SDS, 20% (v/v) glycerol, 2 mM EGTA, 2 mM EDTA, 1% (v/v) bromophenol blue, 10% (v/v) 2-mercaptoethanol) and boiled for 10 min. Samples (20 lL, adjusted for the different protein content of each cell layer) were either loaded onto 10% (w/v) polyacrylamide SDS gels or stored at ) 70C until processed. The protein content of each well was determined in SDS sample buffer cell extracts (without bromophenol blue and 2-mercaptoethanol) using the bicinchoninic acid assay (Sigma, Poole, UK) following the manufacturer’s instructions. Inhibitors were added directly to culture media in each well for 30 min prior to and during the period of CH, as indicated in the relevant figure legends.

Cell culture and chronic hypoxia The SH-SY5Y cell line was kindly provided by Dr J. L. Biedler of the Sloan-Kettering Institute for Cancer Research (Rye, NY, USA). Cells were cultured in a 1 : 1 mixture of Ham’s F12 medium and Eagle’s minimal essential medium (MEM; Gibco Life Sciences,

Detection of ERK1/2, APP and ADAM 10 For detection of ERK1/2, confluent layers of SH-SY5Y cells, which had been cultured either under hypoxic or normoxic conditions for the final 24 h, in 12-well plates were rinsed twice with PBS + Ca2+(800 lL) and incubated with HBS (400 lL) in the


1264 N. J. Webster et al.

presence of muscarine (50 lL) at 37C for the times indicated in Fig. 6. After incubation the HBS was removed, cell layers rinsed twice with ice cold HBS and extracted with SDS sample buffer (200 lL) containing Na3VO4 (1 mM), PMSF (0.2 mM), aprotinin (20 lg/mL), leupeptin (20 lg/mL). For detection of APP and ADAM 10, cell layers were rinsed twice in PBS + Ca2+ (800 lL) and extracted immediately following lysis with SDS sample buffer. The cell lysates, for ERK1/2, APP and ADAM 10, were boiled and samples removed for determination of protein content followed by addition of 2-mercaptoethanol (10% v/v) and bromophenol blue (1% v/v). The content of ERK1/2, APP and ADAM 10 were determined in 40-lg protein samples by western blotting techniques. Western blotting After separation on 10% polyacrylamide, sodium dodecyl sulfate gels, proteins were transferred to polyvinylidene difluoride (PVDF) membrane. Membranes were blocked for 1 h with 5% (w/v) non-fat powdered milk in PBS, containing 0.05% (v/v) Tween 20 (PBST) and sAPPa detected by incubation with the monoclonal antibody 3 6E10 (1: 2000 dilution; Signet Pathology, Denham, MA, USA), which recognizes residues 1–17 of the Ab domain of APP, or monoclonal antibody 22C11 (1 : 100 dilution; Chemicon International Ltd, Harrow, UK), which recognises residues 60–100 of the N-terminus of APP. Full length, cell-associated APP was detected with the monoclonal antibody 22C11 (1 : 100 dilution). Activated ERK1/2 was detected with a monoclonal-antibody, specific for the phosphorylated (activated) form of ERK1/2 (1 : 1000; New England Biolabs; Hitchin, UK). Total ERK1/2 was detected with polyclonal anti-p44/42 MAP kinase antibody (1 : 1000; New England Biolabs). ADAM 10 was detected with B421 polyclonal antibody (a kind gift from Prof. B. De Strooper, Leuven, Belgium). PVDF membranes were incubated with primary antibodies for 3 h, washed extensively with PBST and incubated with anti-mouse (or anti-rabbit for polyclonal primary antibodies) peroxidase-linked secondary antibody (1 : 1000 dilution; Amersham International, Amersham, Bucks., UK) for 1 h. Membranes were washed for 30 min with PBST, and sAPPa or APP was visualized using the enhanced chemi-luminescence method (Amersham Pharmacia Biotech. Ltd, Bucks., UK). Approximate molecular masses were estimated using calibrated prestained standards (Bio-Rad Laboratories, Hemel Hempstead, UK). Quantification of the bands was performed by laser scanning densitometry under conditions in which there was a linear response (Scion Image freeware, Scion Corporation, www.scioncorp.com). Data are presented as example western blots and mean ± SEM obtained from densitometric analysis. Cell viability Cell viability was measured by determining the reduction of 3-(4,5dimethylthiazole-2yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma). SH-SY5Y cells were cultured in 12-well plates for 4–5 days. MTT (200 lL of a stock solution 5 mg/mL in PBS, filter sterilised to remove insoluble material) was added to each well containing 2 mL F12/MEM and FCS followed by incubation for 3 h at 37C in a humidified incubator. Acid propanol (2 mL 0.04 M HCl in propan-2-ol; Sigma) was added to each well and the resulting solution triturated 20 times to dissolve the blue formazan crystals. The absorbance of the solution was read at 570 nm against a 1 : 1 mixture of acid propanol and media as a blank. Viability was

expressed as a mean percentage of absorbance of control cells ± SEM. Immunofluorescence Cells used for immunofluorescence microscopy were grown on 22 mm2 glass coverslips in 6-well plates for 4–5 days. Media was removed and the cell layer rinsed three times with PBS + Ca2+ and then fixed with 10% (v/v) formalin (Sigma) in PBS + Ca2+ for 15 min. Cells were then washed three times in PBS before blocking with 5% goat serum in PBS for 30 min at room temperature (20C). The coverslips were incubated with 22 Cll antibody (diluted 1 : 100 in PBS + Ca2+) or 3D6 antibody (1 lg/mL) for 18 h at 4C. Each coverslip was then washed thoroughly in PBS + Ca2+ and incubated for 2 h at room temperature in a 1 : 1000 dilution of anti-mouse FITC-conjugated secondary antibody (Sigma). After washing thoroughly in PBS + Ca2+, coverslips were mounted with Vectasheild (Vectar Laboratories Ltd, Peterborough, UK) onto glass 5 microscope slides (Merk, Lutterworth, UK). Images were captured from a fluorescence microscope (Axioskop, Carl Zeis Ltd, Welwyn Garden City, UK) at · 20 objective using a JVC KY-F55BE 3-chip CCD camera and an Aquis digital imaging system (Synoptics Ltd, Cambridge, UK). Statistical analysis Where statistical comparisons were made between only two data sets, a paired or unpaired, two-tailed Student’s t-test was used, as appropriate. One-way analysis of variance (ANOVA) was used to examine significance between more than two data sets, with a Tukey post-test examination. Two-way ANOVA was used to examine significance where there were two independent variables, and significance determined by a Tukey post test. p < 0.05 was considered significant in each test.

Results

Previous studies have shown that secretion of sAPPa is constitutive and treatment with phorbol esters or muscarine favours this a-secretase pathway (Mills and Reiner 1999; Canet-Aviles et al. 2002). Figure 1(a–c) shows representative western blots of the effect of decreasing O2 concentrations on the constitutive and muscarine-enhanced sAPPa secretion. sAPPa secretion from CH cells cultured at 10% O2 was very similar to that observed from normoxic cells (Figs 1a and d–e). When cells were cultured for 24 h in an environment of 5% O2, constitutive release resembled that of control cells but muscarine-stimulated release, although not significantly less (Students t-test), was decreased by 39 ± 19.9% (Figs 1b and d–e). When cells were incubated at an even lower O2 level of 2.5% O2, both constitutive (Fig. 1d) and muscarine-evoked (Fig. 1e) sAPPa secretion were significantly decreased by 66.2 ± 9.6% and 63.3 ± 5.4% of controls, respectively (p < 0.01). Exposure to CH for 24 h was required to obtain this decrease, as little inhibition was observed after 12 h CH and no further effect was observed in cells treated with CH for 48 h (data not



(a)

(d)

(b)

(e) (c)

Fig. 1 Effect of reduced O2 concentration on sAPPa secretion from SH-SY5Y cells. SH-SY5Y cells were grown for 4–5 days under normoxic conditions (N) or exposed to lower O2 concentrations of 10, 5 or 2.5% for the final 24 h (CH). Cell layers were then incubated for 1 h in HBS (constitutive) or in HBS containing muscarine (50 lM; muscarine). HBS was collected and anaylsed for sAPPa as described under experimental procedures. Representative western blots of sAPPa

secreted following 24-h treatment with 10% (a), 5% (b) or 2.5% (c) O2 are shown. Densitometric analysis of western blots for constitutive (d) and muscarine-enhanced (e) sAPPa-secretion is shown graphically. Secretion of sAPPa from CH cells is expressed as a percentage of the corresponding normoxic controls and data represent mean ± SEM for the number of experiments shown in parentheses above each column. Western blots were obtained using the monoclonal antibody 6E10.

shown). In addition, exposure of SH-SY5Y cells to 2.5% O2, 24 h significantly decreased TPA-evoked sAPPa secretion by 67.6 ± 4.9% (p < 0.03; n ¼ 4). In all subsequent studies reported here, responses in control (normoxic) cells were therefore compared with cells cultured under 2.5% O2 for the final 24 h before commencement of the experiment (CH cells). The effect of CH on constitutive and muscarine-evoked sAPPa secretion was not a consequence of reoxygenation as, when CH cells were incubated in HBS in air for 30 min and 1 h before the addition of muscarine (50 lM), sAPPa secretion was still inhibited by 70.4 ± 9.2% and 67.4 ± 4.3% (n ¼ 3) of the corresponding normoxic controls, respectively. Importantly, levels of APP species, detected by western blotting whole cell lysates with the antibody 22C11, only decreased to 70 ± 6.1% (n ¼ 6) in cell layers treated with CH for 24 h compared with normoxic controls. In addition, immunocytochemical studies with 22C11 showed little difference in surface labelling of APP in unpermeabilised cells (Figs 2a and b) and permeabilised cells following CH (data not shown). Thus, our findings indicate that hypoxia decreases secretion of sAPPa by over 60%, an effect which cannot be attributed simply to a decrease in total APP levels. Our previous studies (Webster et al. 2001) suggest that under these conditions of CH, muscarine-evoked release of

noradrenaline in SH-SY5Y cells is enhanced by over 100%. In parallel studies, exposure of PC12 cells to 10% O2 for 24 h, conditions previously shown by us (Taylor et al. 1999) to induce the formation of a Ca2+-permeable Ab channel in the plasma membrane, also resulted in a significant decrease in constitutive sAPPa secretion by 50 ± 17.1% (p < 0.02; data not shown). No evidence could be obtained in this study for the generation of Ab following exposure of SH-SY5Y cells to CH. Thus, immunochemical studies using the antibody 3D6 specific for the N-terminal region of Ab (Figs 2c and d) failed to detect an increase in faint staining observed in normoxic cells following CH. This observation was confirmed using the antibody 6E10, which also failed to detect enhanced surface labelling of Ab following CH (not shown). In addition, exposure of SH-SY5Y cells to 0.1 or 1 lM Ab1–40 for 24 h had no effect on either constitutive or muscarine-evoked sAPPa secretion (data not shown). Thus, it appears unlikely that CH leads to an increase in Ab production to compensate for the decrease in sAPPa secretion in SH-SY5Y cells. Cell viability was measured by employing the MTT assay. No inhibition of cell viability was observed following a 24-h period of CH when compared with normoxic controls (Fig. 3a). In contrast, treatment of SH-SY5Y cells with 100 nM and 1 lM staurosporine for 24 h, which induces apoptosis (Boix et al. 1997), significantly decreased cell



Fig. 2 Immunofluorescent labelling of APP and Ab in non-permeabilised SH-SY5Y cells. SH-SY5Y cells were grown for 4–5 days under normoxic conditions (a and c) or exposed to CH conditions (2.5% O2) for the final 24 h (b and d). Surface labelling for APP in non-permeabilised cells was then detected with antibody 22C11 (a, b) and surface labelling for Ab in non-permeabilised cells detected with antibody 3D6 (c and d), as described under experimental procedures. Representative fluorescent images of three (22C11) and seven (3D6) separate experiments are shown in the left hand panels (a–d) with the corresponding phase-contrast images shown in the right hand panels (a¢–d¢). Scale bar ¼ 35 lm and applies to all the images.

viability to 43.8 ± 1.4% and 10.9 ± 1.7% of control values, respectively (p < 0.05). Thus, SH-SY5Y cells treated with 100 nM staurosporine were used as a positive control for apoptosis. Pre-treatment of SH-SY5Y cells with 20 lM BOCD(Ome)-FMK (BOC-FMK; a general inhibitor of caspases) reversed 100 nM staurosporine inhibition of cell viability from 66.7 ± 2.4% to 75.6 ± 1.9% (Fig. 3b; p < 0.05).

Increasing BOC-D(Ome)-FMK concentration to 100 lM had no additional effect on inhibition of cell viability by staurosporine (Fig. 3b). In contrast, pre-treatment with BOCD(Ome)-FMK 2 and 20 lM did not reverse the inhibitory effect of CH on secretion of sAPPa (Figs 3c and d). Thus, concentrations of BOC-D(Ome)-FMK which partly reversed the effect of staurosporine on cell viability had no effect on inhibition of sAPPa secretion caused by CH, indicating that the effect of CH on sAPPa secretion does not appear to be mediated by caspases. Recently it has been reported that reactive oxygen species 6 (ROS) levels are increased during prolonged hypoxia (Chandel et al. 1998, Hohler et al. 1999; Chandel et al. 2000; Chandel and Schumacker 2000). An increase in ROS is well known to be involved in the pathology of AD (Behl 1997; Miranda et al. 2000; Varadarajan et al. 2000). No effect on the inhibition of either constitutive (Fig. 4a) or muscarine-enhanced sAPPa secretion (Fig. 4b) by CH was observed following pre-treatment of SH-SY5Y cells with a mixture of the antioxidants superoxide dismutase (SOD, 50 U/mL), ascorbic acid (100 lM) and melatonin (150 lM). Similarly, incubation of cells with a concentration of H2O2 that did not affect cell viability (40 lM; Fig. 5a), resulted in no inhibition of sAPPa secretion (Figs 5b and c). Our previous studies (Webster et al. 2001) have shown that CH does not affect the regulation of [Ca2+]i by muscarine. Thus the effect of CH on sAPPa-secretion is unlikely to be due to changes in Ca2+ homeostasis. We have also shown that muscarine enhances sAPPa-secretion by a pathway involving PKCa and ERK1/2 (Canet-Aviles et al. 2002). However, no change in the activation of ERK1/2 by muscarine was detected following exposure of SH-SY5Y cells to CH (Figs 6a and b), and hypoxia did not alter total ERK levels (Fig. 6c) indicating that candidate pathways coupling hypoxia to a reduction of sAPPa secretion are unlikely to include ERK1/2. APP is cleaved to form sAPPa by the action of a secretase (see Introduction). Whilst the molecular identity of this enzyme has yet to be established, much evidence suggests that one or more of the ADAM proteins may serve this role (Lammich et al. 1999). To this end, we investigated the effects of hypoxia on levels of ADAM 10 in SH-SY5Y cells. As shown in Fig. 7, ADAM 10 levels were reduced by approximately 50% (p < 0.04, n ¼ 3 experiments). Thus, the reduction in sAPPa caused by chronic hypoxia may simply be due to a reduction in the non-amyloidogenic processing of APP. Discussion

The most important, novel observation of this study is that exposing SH-SY5Y and PC12 cells, derived from human sympathetic neurones and rat adrenal glands, respectively, to CH inhibits sAPPa secretion. Previous studies, by our group,



(a)

(c)

(b)

(d)

Fig. 3 Effect of CH, staurosporine and caspase inhibition on viability of SH-SY5Y cells and sAPPa secretion. (a) SH-SY5Y cells were cultured for 4–5 days under normoxic conditions (N) or exposed to 2.5% O2 for the final 24 h (CH). In parallel experiments, SH-SY5Y cells grown under normoxic conditions were exposed to staurosporine (10 nM, 100 nM, 1 lM staurosporine) for the final 24 h. Cell viability was determined using the MTT assay, as described in Materials and methods, and cell viability expressed as a percentage of absorbance of control cells. Data represent mean ± SEM for 3–5 experiments (performed in duplicate). (b) SH-SY5Y cells were cultured for 4–5 days under normoxic conditions and treated with staurosporine (100 nM; hatched bars) for 24 h before measuring cell viability using the MTT assay. BOC-FMK, at the concentrations indicated, was applied to culture media 30 min before the addition of

staurosporine and reapplied 12 h later. Data represent mean ± SEM of cell viability [expressed as in (a)] for 4–7 experiments, each performed in duplicate. In (c) and (d) cells used to measure muscarinestimulated sAPPa secretion were pre-incubated with BOC-FMK (2 lM and 20 lM) for 30 min before the final 24-h incubation period in CH or normoxia. Inhibitor was also reapplied 12 h later. sAPPa secretion was then stimulated with muscarine in HBS (50 lM) as described in Materials and methods. (c) A representative western blot for muscarine-enhanced sAPPa secretion for normoxic (N) and CH cells obtained using the antibody 6E10. (d) Densitometric analysis of muscarine-enhanced sAPPPa secretion from normoxic (open bars) and CH (hatched bars) cells expressed as a percentage of normoxic cells in the absence of BOC-FMK. Data represent mean ± SEM of five separate experiments.

have shown that exposing SH-SY5Y and PC12 cells to CH leads to an increase in catecholamine secretion. In the case of PC12 cells, this is partly due to an increase in Ab formation giving rise to a novel calcium channel in the plasma membrane (Taylor et al. 1999; Green and Peers 2001). However, the enhanced depolarisation-evoked noradrenaline secretion from chronically hypoxic SH-SY5Y cells was due exclusively to an enhancement of voltage-gated calcium channel activity and not to the formation of novel calcium channels subsequent to an increased production of Ab (Webster et al. 2001). The present study failed to detect enhanced cell surface immunofluorescence with an antibody specific for the N-terminal region of Ab, under conditions when it was possible to measure cell surface Ab in PC12 cells exposed to CH (Taylor et al. 1999). In addition, we have been unable to detect Ab in CH SH-SY5Y cells using either the method of western blotting or a more sensitive ELISA assay. These observations provide further evidence against a hypothesis that the decrease in sAPPa following exposure of SH-SY5Y cells to CH is accompanied by an increase in Ab formation.

This paper focused on two aspects of CH that might account for the change in APP processing away from sAPPa secretion in the SH-SY5Y cell line. Firstly, the idea that CH can induce apoptosis and, secondly, the potential involvement of ROS in mediating this effect on sAPPa secretion. Ischaemia and hypoxia have been shown to induce apoptosis 7 (Tanaka et al. 1994; Choi 1996; Bossenmeyer et al. 1998), and an alteration of APP processing during apoptosis has recently been reported (Barnes et al. 1998; Gervais et al. 1999; Pellegrini et al. 1999; Weidemann et al. 1999). In particular, caspases have been found to cleave APP in the C-terminal region leading to retention of the N-terminal 8 fragment rather than its secretion (Pellegrini et al. 1999; Weidemann et al. 1999). Thus, one potential mechanism was that CH leads to abnormal APP processing as a consequence of caspase activation. However, in the present study a general inhibitor of caspases, BOC-D(Ome)-FMK, did not reverse CH inhibition of sAPPa secretion under conditions when this inhibitor did partially reverse staurosporine-induced decreases in cell viability. This effect of staurosporine has been shown to be due to apoptosis in SH-SY5Y due to activation



(b)

(a)

Fig. 4 Effect of antioxidants on the inhibition of sAPPa secretion by CH from SH-SY5Y cells. SH-SY5Y cells were cultured for 4–5 days under normoxic conditions (N, open columns) or exposed to 2.5% O2 for the final 24 h (CH, hatched columns). Cell layers were exposed to superoxide dismutase (50 U/mL), ascorbic acid (100 lM) and melatonin (150 lM) for 30 min prior to and during the final 24 h growth. The antioxidants were also present during the assay procedure. Constitu-

tive (a) or muscarine-enhanced (50 lM; b) sAPPa secretion was assayed as described in Materials and methods. Densitometric analysis of western blots (top panel) for constitutive (a) and muscarineenhanced (b) sAPPa secretion are expressed as mean ± SEM of percentage of normoxic controls for three constitutive and five muscarine-enhanced separate experiments (lower panel). Western blots were obtained using the antibody 6E10.

(a)

(c)

(b)

Fig. 5 Effect of H2O2 on viability of SH-SY5Y cells and secretion of sAPPa. SH-SY5Y cells were cultured for 4–5 days under normoxic conditions and incubated with the indicated concentration of H2O2 for the final 24 h before measuring cell viability using the MTT assay (a). Data represent mean ± SEM of cell viability (expressed as in Fig. 3) for three experiments each carried out in duplicate. Cells used for measuring constitutive (b) or muscarine-enhanced (c) sAPPa secre-

tion were cultured under normoxic conditions for 4–5 days (N) or incubated in 40 lM H2O2 (H2O2) or 2.5% O2 (CH) for the final 24 h, as described in Materials and methods. Densitometric analysis of western blots (top panel) for constitutive (a) and muscarine-enhanced (b) sAPPa secretion are expressed as mean ± SEM. sAPPa secretion as percentage of normoxic controls for three separate experiments (lower panel). Western blots were obtained using the antibody 6E10.



(a)

(a)

(b)

(b)

(c)

Fig. 6 Effect of CH on muscarine-enhanced p42/44 ERK phosphorylation in SH-SY5Y cells. (a) SH-SY5Y cells were cultured for 4–5 days under normoxic conditions (N) or exposed to 2.5% O2 for the final 24 h (CH). Cell layers were then incubated in HBS containing 50 lM muscarine for the time points indicated. After incubation, cell layers were extracted with SDS sample buffer containing protease and phosphatase inhibitors and the levels of phosphorylated p42/44 ERK determined by western blotting as described in Materials and methods. (b) Densitometric analysis of muscarine-enhanced levels of phosphorylated p42/44 ERK is shown graphically. Phosphorylation in N (s) and CH (d) cells is expressed as a percentage of the level of phosphorylated p42/44 ERK detected at 0 time in normoxic cells. Data points represent mean ± SEM for four separate experiments. A representative western blot of total p42/44 ERK protein levels is shown in (c).

9 of caspase 2 and caspase 3 (Lopez and Ferrer 2000). Interestingly, in another study on oxidative stress-induced cell death, caspase 3 inhibition with BOC-D(Ome)-FMK delayed, but did not ultimately prevent, cell death in SH-SY5Y (Krishnamurthy et al. 2000). Furthermore, in the present study, assays with MTT provided no evidence for loss of cell viability following CH compared with up to 90% decrease in viability with staurosporine (Fig. 4). Thus, a role for caspases and apoptosis in mediating the effect of CH on decreased sAPPa secretion in SH-SY5Y cells is unlikely. One effect of CH is generation of ROS (Chandel et al. 1998; Hohler et al. 1999; Chandel et al. 2000; Chandel and Schumacker 2000). Thus, the possibility that ROS mediate the effect of CH on APP processing in these cell lines was examined. In this context it is of interest that previous studies have shown that exposure of SH-SY5Y cells to

Fig. 7 Hypoxia decreases ADAM 10 expression. (a) Western blot indicating total levels of ADAM 10 (probed with B421 polyclonal antibody) from cells cultured under normoxic conditions (N) and cells exposed to 2.5% O2 (CH) for the final 24 h (b) Densitometric analysis of three separate experiments is expressed as mean ± SEM normalized to total ADAM 10 detected in normoxic cells.

peroxynitrite and H2O2 impaired muscarinic-receptor linked phosphoinositide hydrolysis coupled to G-proteins (Li et al. 1996, 1998) and inhibits carbachol-stimulated tyrosine phosphorylation of proteins (Jope et al. 1999). H2O2 has been used by several groups to induce oxidative stress and affect APP processing in SH-SY5Y cells. Thus, Olivieri et al. (2001) reported that exposure of SH-SY5Y cells to 50 lM H2O2 for 30 min was cytotoxic and increased Ab1–40 and Ab1–42 release. In another study Misonou et al. (2000) found that H2O2 (100–250 lM) increased intracellular Ab whereas Zhang et al. (1997) used millimolar H2O2 to induce apoptosis and increase production of amyloidogenic fragments. However, these effects occur at concentrations of H2O2 above 50 lM and are accompanied by loss of cell viability. In the present study 40 lM H2O2 did not alter cell viability nor sAPPa secretion in SH-SY5Y cells (Fig. 5). Thus, there is no correlation between the action of H2O2 and CH on SH-SY5Y cells, which suggests that the generation of H2O2 is not a mechanism by which CH alters APP processing. This conclusion is further supported by our observation that the presence of antioxidants during CH and following removal of cells from CH failed to reverse the decrease in sAPPa secretion from CH cells (Fig. 4). These latter observations also suggest that the effect of CH on sAPPa-secretion is not related to an Ôoxygen burstÕ when the cells are returned to ÔnormalÕ levels of O2. In our previous study (Webster et al. 2001) muscarineevoked noradrenaline release was increased by over 100% following exposure of SH-SY5Y cells to CH. This was not accompanied by a change in the release of calcium from intracellular stores or Ôstore-depletedÕ calcium influx across the plasma membrane. Thus, the decrease in muscarine-



enhanced sAPPa secretion observed in this study is unlikely to be the consequence of a general inhibitory effect of CH on energy stores, muscarinic receptors and their coupling to changes in intracellular calcium stores or other gross effects on calcium homeostasis. Furthermore the lack of effect of CH on the activation of ERK 1/2 by muscarine suggests that the decrease in sAPPa secretion following CH is not a consequence of disruption of the signalling pathway coupling muscarinic receptor activation to secretion of sAPPa upstream of ERK1/2. It thus appears that the effect of CH on APP processing in SH-SY5Y cells is selective and differs, in mechanism, from that observed for PC12 cells. In conclusion, the current study has shown that exposing cell lines of neuronal origin to CH leads to a decrease in sAPPa secretion by a mechanism which does not appear to be related to the induction of apoptosis by either activation of caspases or generation of ROS. The most likely possibility is that CH inhibits sAPPa secretion by decreasing the level of protein of the disintegrin metalloprotease, ADAM 10, which is a strong candidate for the a-secretase (Lammich et al. 1999). Studies are in progress to determine whether the decrease in ADAM 10 protein observed in this study is due to an effect of CH on either expression of ADAM 10 mRNA or activation of proteolytic degradation of ADAM 10 protein. Acknowledgements The support of The Alzheimer’s Society for this work is acknowledged with thanks. NJW and KNG were in receipt of MRC Studentships.

References Abe K., Tanzi R. E. and Kogure K. (1991) Selective induction of Kunitztype protease inhibitor domain-containing amyloid precursor protein mRNA after persistent focal ischemia in rat cerebral cortex. Neurosci. Lett. 125, 172–174. Barnes N. Y., Li L., Yoshikawa K., Schwartz L. M., Oppenheim R. W. and Milligan C. E. (1998) Increased production of amyloid precursor protein provides a substrate for caspase-3 in dying motoneurons. J. Neurosci. 18, 5869–5880. Behl C. (1997) Amyloid b-protein toxicity and oxidative stress in Alzheimer’s disease. Cell Tissue Res. 290, 471–480. Boix J., Llecha N., Yuste V. J. and Comella J. X. (1997) Characterisation of the cell death process induced by staurosporine in human neuroblastoma cell lines. Neuropharmacology 36, 811–821. Bossenmeyer C., Chihab R., Muller S., Schroeder H. and Daval J. L. (1998) Hypoxia/reoxygenation induces apoptosis through biphasic induction of protein synthesis in cultured rat brain neurons. Brain Res. 787, 107–116. Canet-Aviles R., Anderton M., Hooper N. M., Turner A. J. and Vaughan P. F. T. (2002) Muscarine enhances soluble amyloid precursor protein secretion in human neuroblastoma SH-SY5Y by a pathway dependent on protein kinase Ca, src-tyrosine kinase and extracellular signal-regulated kinase but not phospholipase C. Mol. Brain Res. 102, 62–72.

Chandel N. S. and Schumacker P. T. (2000) Cellular oxygen sensing by mitochondria: old questions, new insight. J. Appl. Physiol. 88, 1880–1889. Chandel N. S., Maltepe E., Goldwasser E., Mathieu C. E., Simon M. C. and Schumacker P. T. (1998) Mitochondrial reactive oxygen species trigger hypoxia-induced transcription. Proc. Natl Acad. Sci. USA 95, 11715–11720. Chandel N. S., McClintock D. S., Feliciano C. E., Wood T. M., Melendez J. A., Rodriguez A. M. and Schumacker P. T. (2000) Reactive oxygen species generated at mitochondrial complex III stabilize hypoxia-inducible factor-1a during hypoxia: a mechanism of O2 sensing. J. Biol. Chem. 275, 25130–25138. Choi D. W. (1996) Ischemia-induced neuronal apoptosis. Curr. Opin. Neurobiol. 6, 667–672. De Strooper B. and Annaert W. (2000) Proteolytic processing and cell biological functions of the amyloid precursor protein. J. Cell Sci. 113, 1857–1870. De Strooper B., Saftig P., Craessaerts K., Vanderstichele H., Guhde G., Annaert W., Von Figura K. and Van Leuven F. (1998) Deficiency of presenilin-1 inhibits the normal cleavage of amyloid precursor protein. Nature 391, 387–390. Furukawa K., Sopher B. L., Rydel R. E., Begley J. G., Pham D. G., Martin G. M., Fox M. and Mattson M. P. (1996) Increased activityregulating and neuroprotective efficacy of a-secretase-derived secreted amyloid precursor protein conferred by a C-terminal heparin-binding domain. J. Neurochem. 67, 1882–1896. Galli C., Piccini A., Ciotti M. T., Castellani L., Calissano P., Zaccheo D. and Tabaton M. (1998) Increased amyloidogenic secretion in cerebellar granule cells undergoing apoptosis. Proc. Natl Acad. Sci. USA 95, 1247–1252. Gervais F. G., Xu D., Robertson G. S., Vaillancourt J. P., Zhu Y., Huang J., LeBlanc A., Smith D., Rigby M., Shearman M. S., Clarke E. E., Zheng H., Van Der Ploeg L. H., Ruffolo S. C., Thornberry N. A., Xanthoudakis S., Zamboni R. J., Roy S. and Nicholson D. W. (1999) Involvement of caspases in proteolytic cleavage of Alzheimer’s amyloid-b precursor protein and amyloidogenic Ab peptide formation. Cell 97, 395–406. Green K. N. and Peers C. (2001) Amyloid b peptides mediate hypoxic augmentation of Ca2+ channels. J. Neurochem. 77, 953–956. Hohler B., Lange B., Holzapfel B., Goldenberg A., Hanze J., Sell A., Testan H., Moller W. and Kummer W. (1999) Hypoxic upregulation of tyrosine hydroxylase gene expression is paralleled, but not induced, by increased generation of reactive oxygen species in PC12 cells. FEBS Lett. 457, 53–56. Jendroska K., Hoffmann O. M. and Patt S. (1997) Amyloid b peptide and precursor protein (APP) in mild and severe brain ischemia. Ann. NY Acad. Sci. 826, 401–405. Jope R. S., Song L., Grimes C. A. and Zhang L. (1999) Oxidative stress oppositely modulates protein tyrosine phosphorylation stimulated by muscarinic G protein-coupled and epidermal growth factor receptors. J. Neurosci. Res. 55, 329–340. Kalaria R. N. (2000) The role of cerebral ischemia in Alzheimer’s disease. Neurobiol. Aging 21, 321–330. Kalaria R. N., Bhatti S. U., Lust W. D. and Perry G. (1993) The amyloid precursor protein in ischemic brain injury and chronic hypoperfusion. Ann. NY Acad. Sci. 695, 190–193. Kim T. W., Pettingell W. H., Jung Y. K., Kovacs D. M. and Tanzi R. E. (1997) Alternative cleavage of Alzheimer-associated presenilins during apoptosis by a caspase-3 family protease. Science 277, 373–376. Kogure K. and Kato H. (1993) Altered gene expression in cerebral ischemia. Stroke 24, 2121–2127. Koistinaho J., Pyykonen I., Keinanen R. and Hokfelt T. (1996) Expression of b-amyloid precursor protein mRNAs following transient focal ischaemia. Neuroreport 7, 2727–2731.



Krishnamurthy P. K., Mays J. L., Bijur G. N. and Johnson G. V. (2000) Transient oxidative stress in SH-SY5Y human neuroblastoma cells results in caspase dependent and independent cell death and tau proteolysis. J. Neurosci. Res. 61, 515–523. Lammich S., Kojiro E., Postina R., Gilbert S., Pfeiffer R., Jasionowski M., Haass C. and Fahrenholz. F. (1999) Constitutive and regulated alpha-secretase cleavage of Alzheimer’s amyloid precursor protein by a disintegrin metalloprotease. Proc. Natl Acad. Sci. USA 96, 3922–3927. LeBlanc A. (1995) Increased production of 4 kDa amyloid b peptide in serum deprived human primary neuron cultures: possible involvement of apoptosis. J. Neurosci. 15, 7837–7846. Li X., Song L. and Jope R. S. (1996) Cholinergic stimulation of AP-1 and NFjB transcription factors is differentially sensitive to oxidative stress in SH-SY5Y neuroblastoma: relationship to phosphoinositide hydrolysis. J. Neurosci. 16, 5914–5922. Li X., De Sarno P., Song L., Beckman J. S. and Jope R. S. (1998) Peroxynitrite modulates tyrosine phosphorylation and phosphoinositide signalling in human neuroblastoma SH-SY5Y cells: attenuated effects in human 1321N1 astrocytoma cells. Biochem. J. 331, 599–606. Loetscher H., Deuschle U., Brockhaus M., Reinhardt D., Nelboeck P., Mous J., Grunberg J., Haass C. and Jacobsen H. (1997) Presenilins are processed by caspase-type proteases. J. Biol. Chem. 272, 20655–20659. Lopez E. and Ferrer I. (2000) Staurosporine- and H-7-induced cell death in SH-SY5Y neuroblastoma cells is associated with caspase-2 and caspase-3 activation, but not with activation of the FAS/FAS-1caspase-8 signaling pathway. Brain Res. Mol. Brain Res. 85, 61–67. Mattson M. P. (1997) Cellular actions of b-amyloid precursor protein and its soluble and fibrillogenic derivatives. Physiol. Rev. 77, 1081–1132. Mattson M. P., Cheng B., Culwell A. R., Esch F. S., Lieberburg I. and Rydel R. E. (1993) Evidence for excitoprotective and intraneuronal calcium-regulating roles for secreted forms of the b-amyloid precursor protein. Neuron 10, 243–254. Mills J. and Reiner P. B. (1999) Regulation of amyloid precursor protein cleavage. J. Neurochem. 72, 443–460. Miranda S., Opazo C., Larrondo L. F., Munoz F. J., Ruiz F., Leighton F. and Inestrosa N. C. (2000) The role of oxidative stress in the toxicity induced by amyloid b-peptide in Alzheimer’s disease. Prog. Neurobiol. 62, 633–648. Misonou H., Morishima-Kawashima M. and Ihara Y. (2000) Oxidative stress induces intracellular accumulation of amyloid b-protein (Ab) in human neuroblastoma cells. Biochemistry 39, 6951–6959. Olivieri G., Baysang G., Meier F., Muller-Spahn F., Stahelin H. B., Brockhaus M. and Brack C. (2001) N-acetyl-1-cysteine protects

SHSY5Y neuroblastoma cells from oxidative stress and cell cytotoxicity: effects on b-amyloid secretion and tau phosphorylation. J. Neurochem. 76, 224–233. Pellegrini L., Passer B. J., Tabaton M., Ganjei J. K. and D’Adamio L. (1999) Alternative, non-secretase processing of Alzheimer’s b-amyloid precursor protein during apoptosis by caspase-6 and -8. J. Biol. Chem. 274, 21011–21016. Selkoe D. J. (2001) Alzheimer’s disease: genes, proteins, and therapy. Physiol. Rev. 81, 741–766. Smith-Swintosky V. L., Pettigrew L. C., Craddock S. D., Culwell A. R., Rydel R. E. and Mattson M. P. (1994) Secreted forms of b-amyloid precursor protein protect against ischemic brain injury. J. Neurochem. 63, 781–784. Tanaka M., Ito H., Adachi S., Akimoto H., Nishikawa T., Kasajima T., Marumo F. and Hiroe M. (1994) Hypoxia induces apoptosis with enhanced expression of Fas antigen messenger RNA in cultured neonatal rat cardiomyocytes. Circ. Res. 75, 426–433. Taylor S. C., Batten T. F. and Peers C. (1999) Hypoxic enhancement of quantal catecholamine secretion. Evidence for the involvement of amyloid b-peptides. J. Biol. Chem. 274, 31217–31222. Tomimoto H., Wakita H., Akiguchi I., Nakamura S. and Kimura J. (1994) Temporal profiles of accumulation of amyloid beta/A4 protein precursor in the gerbil after graded ischemic stress. J. Cereb. Blood Flow Metab. 14, 565–573. Varadarajan S., Yatin S., Aksenova M. and Butterfield D. A. (2000) Review: Alzheimer’s amyloid b-peptide-associated free radical oxidative stress and neurotoxicity. J. Struct. Biol. 130, 184–208. Vito P., Ghayur T. and D’Adamio L. (1997) Generation of antiapoptotic presenilin-2 polypeptides by alternative transcription, proteolysis, and caspase-3 cleavage. J. Biol. Chem. 272, 28315– 28320. Webster N. J., Vaughan P. F. T. and Peers C. (2001) Hypoxic enhancement of evoked noradrenaline release from the human neuroblastoma SH-SY5Y. Mol. Brain Res. 89, 50–57. Weidemann A., Paliga K., Durrwang U., Reinhard F. B., Schuckert O., Evin G. and Masters C. L. (1999) Proteolytic processing of the Alzheimer’s disease amyloid precursor protein within its cytoplasmic domain by caspase-like proteases. J. Biol. Chem. 274, 5823–5829. Yokota M., Saido T. C., Tani E., Yamaura I. and Minami N. (1996) Cytotoxic fragment of amyloid precursor protein accumulates in hippocampus after global forebrain ischemia. J. Cereb. Blood Flow Metab. 16, 1219–1223. Zhang L., Zhao B., Yew D. T., Kusiak J. W. and Roth G. S. (1997) Processing of Alzheimer’s amyloid precursor protein during H2O2induced apoptosis in human neuronal cells. Biochem. Biophys. Res. Commun. 235, 845–848.
