Address correspondence and reprint requests to Hubert Vaudry, INSERM U413, European Institute for Peptide Research (IFRMP 23), Laboratory of Cellular and Molecular Neuroendocrinology, UA CNRS, University of Rouen, 76821 Mont-Saint-Aignan, France. E-mail:firstname.lastname@example.org
Accumulation of β-amyloid peptide (Aβ), which is a landmark of Alzheimer's disease, may alter astrocyte functions before any visible symptoms of the disease occur. Here, we examined the effects of Aβ on biosynthesis and release of diazepam-binding inhibitor (DBI), a polypeptide primarily expressed by astroglial cells in the CNS. Quantitative RT–PCR and specific radioimmunoassay demonstrated that aggregated Aβ25−35, at concentrations up to 10−4m, induced a dose-dependent increase in DBI mRNA expression and DBI-related peptide release from cultured rat astrocytes. These effects were totally suppressed when aggregation of Aβ25−35 was prevented by Congo red. Measurement of the number of living cells revealed that Aβ25−35 induced a trophic rather than a toxic effect on astrocytes. Administration of cycloheximide blocked Aβ25−35-induced increase of DBI gene expression and endozepine accumulation in astrocytes, indicating that protein synthesis is required for DBI gene expression. Altogether, the present data suggest that Aβ-induced activation of endozepine biosynthesis and release may contribute to astrocyte proliferation associated with Alzheimer's disease.
The term endozepines designates a family of regulatory peptides that have been initially isolated from the rat brain on the basis of their ability to displace the binding of benzodiazepines from their receptors (Guidotti et al. 1983). All endozepines characterized so far derive from an 86-amino acid precursor polypeptide called diazepam-binding inhibitor (DBI), which generates, through proteolytic cleavage, several biologically active peptides including the triakontatetraneuropeptide DBI[17–50] (TTN) and the octadecaneuropeptide DBI[33–50] (ODN) (Ferrero et al. 1986; Slobodyansky et al. 1989).
Alzheimer's disease is a neurogenerative disorder characterized by the presence of senile plaques in the brain associated with intense astrogliosis. Senile plaques are primarily comprised of 39- to 43-amino acid peptides called beta-amyloid peptides (Aβ) (Kang et al. 1987). The Aβ25−35 fragment, which forms aggregates in vitro (Pike et al. 1993), is considered to bear the neurotoxic activity of Aβ (Yankner et al. 1990; Wei et al. 2000). During the prenatal period, a period which corresponds to intense astrogliosis (Kadhim et al. 1988; Rakic 1991), high concentrations of DBI mRNA and DBI-derived peptides occur in the rat brain (Malagon et al. 1993; Bürgi et al. 1999). These observations, together with the fact that endozepines increase [3H]thymidine incorporation in rat astrocytes (Gandolfo et al. 1999, 2000), suggest that endozepines may act as neurotrophic factors regulating proliferation and/or survival of astroglial cells.
It has been previously reported that the concentration of endozepines is higher in the cerebrospinal fluid of patients with Alzheimer's disease as compared to control subjects (Ferrarese et al. 1990). This observation prompted us to examine the effect of Aβ on endozepine production. The present study provides evidence that Aβ25−35 increases both transcription of the DBI gene and release of endozepines in cultured rat astrocytes, supporting the view that endozepines may play a role in glial cell proliferation in Alzheimer patients.
Materials and methods
Dulbecco's modified Eagle's medium, F12 culture medium, d(+)-glucose, bovine serum albumin, trifluoroacetic acid, beta-amyloid 1–40 and 25–35 peptides (Aβ1−40, Aβ25−35), cycloheximide, Tri reagent and Congo red were purchased from Sigma (St. Louis, MO, USA). Interleukin-1β (IL-1β) was obtained from Eurobio (Les Ulis, France). l-Glutamine, N-2-hydroxyethylpiperazine-N-2-ethane sulphonic acid (HEPES) and the antibiotic–antimycotic solution were from Biowhittaker (Gagny, France). Fetal bovine serum was from Dutscher (Brumath, France). Trypsin-EDTA was from Gibco (Invitrogen, Cergy Pontoise, France). Acetonitrile was from Prolabo (Fontenay-sous-Bois, France). Calcein-AM and propidium iodide were from Molecular Probes (Eugene, OR, USA). The scrambled Aβ25−35 peptide and [Tyr0]-ODN were from Neosystem (Strasbourg, France). Rat ODN was synthesized by using the standard Fmoc procedure as previously described (Leprince et al. 1998).
Secondary cultures of rat cortical astrocytes were prepared as previously described (Brown 1999) with minor modifications. Briefly, cerebral hemispheres from newborn Wistar rats were collected in Dulbecco's modified Eagle's medium/F12 (2 : 1; v/v) culture medium supplemented with 2 mm glutamine, 1% insulin, 5 mm HEPES, 0.4% glucose and 1% of the antibiotic–antimycotic solution. The tissues were dissociated mechanically with a syringe equipped with a 1-mm gauge needle, and filtered through a 100-µm sieve (Falcon, Franklin Lakes, NJ, USA). Dissociated cells were re-suspended in culture medium supplemented with 10% fetal bovine serum and seeded at a density of 0.6 × 106 cells/mL on 150-cm2 flasks (Techno Plastique Products, Trasandinger, Switzeland). When cultures were confluent, astrocytes were isolated from mixed glial cultures by shaking the flasks with an orbital agitator (KS 15, Bühler, Germany). Adhesive cells were detached by trypsination and pre-plated for 10 min to discard contaminating oligodendrocytes and microglial cells. Then, the non-adhering astrocytes were harvested and plated on Petri dishes (Dutscher) at a density of 0.2 × 106 cells/mL.
Cells were seeded on 60-mm dishes (Dutscher) for measurement of endozepine secretion and content, or on 35-mm dishes for measurement of cell survival and DBI gene expression. The cells were incubated at 37°C in a humid atmosphere (5% CO2). After 4 days of culture, 100% of the cells were labelled with antibodies against glial fibrillary acidic protein and no labelling was observed with antibodies against galactocerebroside or isolectine B4, markers of oligodendrocytes or microglia, respectively (data not shown). All experiments were performed on 5- to 7-day-old secondary cultures.
Measurement of endozepines
Cultured astrocytes were incubated at 37°C with fresh serum-free medium in the absence or presence of test substances. At the end of the incubation, culture media were collected and the cells were homogenized in 2 m ice-cold acetic acid. Peptides contained in culture media were concentrated on Sep-Pak C18 cartridges (Alltech Europe, Lokeren Belgium). Bound material was eluted with 50% (v/v) acetonitrile/water containing 0.1% trifluoroacetic acid (v/v), the solvent was evaporated by vacuum centrifugation (Speed Vac concentrator, Savant, Hicksville, NY, USA) and the samples were kept dry until radioimmunoassay.
The concentrations of ODN-like immunoreactivity (ODN-LI), were measured by radioimmunoassay using an antiserum raised against synthetic rat ODN (Tonon et al. 1990). [Tyr0]-ODN was iodinated by using the chloramine-T procedure (Vaudry et al. 1978) and purified on a Sep-Pak C18 cartridge. Dried samples of cultured media and cell extracts were resuspended in phosphate buffer (100 mm; pH 8) containing 0.1% Triton X-100. The final dilution of the ODN antiserum was 1 : 30 000 and the total amount of tracer was 6000 cpm/tube. After a 2-day incubation at 4°C, the antibody-bound ODN fraction was precipitated by addition of 100 µL bovine γ-globulin (1%, w/v) and 2 mL of polyethylene glycol (20%, w/v). After centrifugation (5000 g, 4°C, 30 min) the supernatant was removed and the pellet containing the bound fraction was counted in a gamma counter (LKB Wallac, Rockville, MI, USA).
Measurement of interleukin-1β release
Cultured astrocytes were incubated at 37°C for 6 h with fresh serum-free medium in the absence or presence of test substances. At the end of the incubation period, culture media were collected, concentrated by vacuum centrifugation and the samples were kept dry until assay. The dry extracts were reconstituted in 250 µL of assay buffer and IL-1β content was measured by using a commercial ELISA kit (Biosource Europe, Nivelles, Belgium).
Measurement of cell survival
Cultured cells were incubated at 37°C with fresh serum-free culture medium in the absence or presence of aggregated Aβ25−35 for 6–36 h. Cells were then incubated for 10 min at 37°C with 0.3 µg/mL calcein-AM (producing green fluorescence in living cells) and 2 µg/mL of propidium iodide (producing red fluorescence in dead cells), rinsed twice with culture medium without probe and examined on an inverted microscope (Leica, Heidelberg, Germany) equipped with a double pass filter. For quantification of surviving cells, astrocytes were incubated with calcein-AM, rinsed twice with phosphate buffer (100 mm; pH 7.4) and lysed with 1% sodium dodecyl sulphate. Fluorescence intensity was measured (λ excitation = 485 nm and λ emission = 530 nm) with a FL600 fluorescence microplate reader (Bio-Tek Instruments, Winooski, VT, USA). Measurement of the number of living cells was performed after 36 h of culture by means of a cell counter (Z2 coulter counter model; Beckman, Villepinte, France).
Cultured astrocytes were incubated at 37°C with fresh serum-free culture medium in the absence or presence of test substances. At the end of the incubation, culture media were removed and the cells were rinsed twice with RNase-free saline phosphate buffer (1 m, pH 7.4; Invitrogen, Cergy Pontoise, France).Total RNA was extracted by the guanidine thiocyanate–phenol–chloroform method (Chomczynski and Sacchi 1987) using the Tri reagent. Approximately 1 µg of total RNA was reverse transcribed by ImPROM-IITM reverse transcriptase (Promega, Charbonnières-Les-Bains, France) using random hexanucleotides as primers. Reverse transcription–polymerase chain reaction (RT–PCR) was performed on 15 ng of total cDNA with 1 × SYBR Green universal PCR Master mix (Applied Biosystem, Courtaboeuf, France) containing dNTPs, MgCl2, AmpliTaq Gold DNA polymerase, forward (5′-TGCTCCCGCGCTTTCA-3′) and reverse (5′-CTGAGTCTTGAGGCGCTTCAC-3′) DBI primers (300 nm, each; Proligo, Paris, France). DBI cDNA was first subjected to 50°C for 2 min and 95°C for 10 min, followed by 40 reaction cycles of 95°C for 15 s, 60°C for 1 min, using the ABI Prism 7000 sequence detection system (Applied Biosystem).
The amount of DBI cDNA in each sample was calculated by the comparative threshold cycle (Ct) method and expressed as 2exp(–ΔΔCt) using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an internal control.
All values presented in the figures are means SEM. Student's t-test and anova followed by Bonferroni's test were applied to determine statistical differences between values.
Effects of β-amyloid peptides on endozepine production
Exposure of astrocytes to graded concentrations of aggregated Aβ1−40 (10−7 to 10−4m) for 1 h resulted in a dose-dependent increase of ODN-LI in the culture medium (Fig. 1a). Pre-incubation of the β-amyloid peptide with Congo red (2 mm, 12 h) suppressed its effect on ODN-LI release (Fig. 1b). Incubation of cultured astrocytes with aggregated Aβ25−35 (10−7 to 10−4m) also induced a dose-dependent increase of ODN-LI in the incubation medium (Fig. 2a). The half-maximum effect was observed at a concentration of 2 × 10−6m and the maximum stimulation of endozepine release (180% over control; p < 0.001) was obtained at a concentration of 3 × 10−5m Aβ25−35. Pre-incubation of Aβ25−35 with Congo red (2 mm, 12 h) totally blocked the effect of the peptide (10−5m, 1 h) on ODN-LI release (Fig. 2b). The scrambled Aβ25−35 peptide (10−5m, 1 h) had no effect on ODN-LI release (Fig. 2c).
Time-course experiments over a period of 36 h revealed that incubation of astrocytes with aggregated Aβ25−35 (10−5m) provoked a time-dependent increase of ODN-LI in both culture media and cellular extracts (Fig. 3). The maximum effect of Aβ25−35 on ODN-LI release reached a plateau at 1 h. After 8 h, the stimulatory effect of Aβ25−35 on ODN-LI secretion gradually declined to reach basal level after 12 h of incubation.
Aggregated Aβ25−35 also induced a time-dependent increase in the content of ODN-LI in cultured astrocytes which was maximum (+80%) after 36 h of incubation (p < 0.001) (Fig. 3). The stimulatory effect of Aβ25−35 (10−5m, 24 h) on the content of ODN-LI was abrogated by pre-incubation of Aβ25−35 with Congo red (2 mm) (Fig. 4a). Treatment of cells with cycloheximide (5 × 10−6m) reduced the basal level of ODN-LI in the cell extracts and totally suppressed Aβ25−35-induced increase of ODN-LI content (10−5m, 24 h) (Fig. 4b). Double staining with calcein-AM/propidium iodide showed that cycloheximide did not modify the intensity of the green fluorescence labelling, indicating that cycloheximide did not affect cell survival (data not shown).
Effects of Aβ25−35 on cell survival and proliferation
The viability of cultured astrocytes was visualized by double staining with calcein-AM/propidium iodide (Fig. 5). After 12 h in serum-free culture medium, almost all cells were alive, as revealed by the absence of red fluorescence staining (dead cells) with propidium iodide (Fig. 5a). Incubation of astrocytes with aggregated Aβ25−35 (10−5m, 12 h) did not induce any modification of the staining (Fig. 5b). Quantification of the green fluorescence signal (produced by calcein-AM) showed that incubation of astrocytes with aggregated Aβ25−35 (10−5m, 6–36 h) provoked a modest but significant increase of the fluorescence intensity with a maximum effect at 36 h (Fig. 5c).
Incubation of cells with 10−5m aggregated Aβ25−35 for 36 h increased the number of cells by 52%(Fig. 6a). During the same time of incubation, ODN (10−11m) and IL-1β (6 × 10−10m) increased the number of cells by 32% and 25%, respectively (Fig. 6a). Exposure of astrocytes to aggregated Aβ25−35 (10−5m) for 6 h stimulated the release of IL-1β in the culture medium (+95%) (Fig. 6b) and the effect of Aβ25−35 was mimicked by a 6-h incubation with 10−11m ODN (+92%).
Effect of Aβ25−35 on diazepam-binding inhibitor gene expression
Exposure of astrocytes to graded concentrations of aggregated Aβ25−35 (10−6 to 10−4m) for 24 h resulted in a dose-dependent increase in DBI mRNA level (Fig. 7a). Pre-incubation of Aβ25−35 with Congo red (2 mm, 12 h) suppressed the effect of Aβ25−35 (10−5m, 24 h) on DBI gene expression (Fig. 7b). Treatment of cells with cycloheximide (5 × 10−6m) totally abolished the increase in DBI mRMA level induced by Aβ25−35 (Fig. 7c).
Although endozepines are very abundant and widely distributed in the central nervous system (Tonon et al. 1990; Alho et al. 1991), the role of these neuropeptides still remains enigmatic. A report has shown an increase in endozepine content in the cerebrospinal fluid of patients with Alzheimer's disease (Ferrarese et al. 1990). The present study demonstrates that β-amyloid peptides, the main constituents of senile plaques in Alzheimer's disease brain, activate the biosynthesis and release of endozepines in rat astrocytes.
We first observed that both Aβ1−40 and Aβ25−35 induce a concentration- and time-dependent stimulation of endozepine secretion by cultured astrocytes. Pre-incubation of Aβ1−40 and Aβ25−35 with Congo red, a chemical compound that prevents β-amyloid fibril formation (Lorenzo and Yankner 1994; Soto et al. 1996), suppressed the stimulatory effect of the peptides, indicating that only the aggregated forms of Aβ can activate endozepine release.
Prolonged incubation of astrocytes with Aβ25−35 induced a sustained increase of endozepine content in cells while the concentration of ODN-LI in culture medium gradually declined. It has been previously reported that cultured astrocytes release proteolytic enzymes which actively degrade various neuropeptides (Mentlein et al. 1990; Mentlein and Dahms 1994; Masmoudi et al. 2005). Therefore, the decrease in ODN-LI observed in the incubation medium after 12 h of culture is likely attributable to the accumulation of proteolytic enzymes in the medium, leading to degradation of DBI-related peptides and thus to the loss of immunoreactive material in the extracellular medium.
It has been previously reported that, in fetal bovine serum-containing medium, Aβ25−35 exerts a toxic effect on cultured cortical astrocytes (Brera et al. 2000). In order to determine whether the Aβ25−35-evoked increase in endozepine concentration in the conditioned medium could be ascribed to astrocyte death, we examined the effect of Aβ25−35 on cell viability. In fact, visualization of living cells by calcein-AM staining and direct cell counting both revealed that Aβ25−35 induces an increase in the number of astrocytes, indicating that Aβ does not affect the viability of cultured astrocytes, and may even stimulate cell proliferation. In agreement with this observation, it has already been shown that Aβ exhibits a modest mitogenic activity on C6 glioma cell line (Pena et al. 1995) and that endozepines, in very much the same way as IL-1β, stimulate proliferation of cultured astrocytes (Gandolfo et al. 1999, 2000; this study). The discrepancy between our data and those reported by Brera et al. (2000) can be accounted for by the different culture conditions used, i.e. the absence (this study) or the presence of serum in the culture medium. Consistent with this hypothesis, it has been previously shown that in the absence of serum, β-amyloid fragments, including Aβ25−35, do not affect survival of cultured hippocampal astrocytes (Meske et al. 1998).
A major histological feature observed in the brain of patients with Alzheimer's disease is the accumulation of activated astrocytes around senile plaques (Arelin et al. 2002; Nagele et al. 2004). It has been proposed that Aβ may activate the surrounding astrocytes to secrete proinflammatory factors, such as IL-1, IL-6 and prostaglandins, that elicit a cascade of cellular events leading to neurodegeneration (Landolfi et al. 1998; Hu and Van Eldik 1999). Indeed, it has been found that IL-1 levels are higher in the brain of Alzheimer patients than in controls (Griffin et al. 1989). In addition, in vitro experiments have shown that treatment of cultured rat cortical astrocytes with β-amyloid peptides induce up-regulation of IL-1β (Eriksson et al. 1998; Hu et al. 1998) and that IL-1β is involved in glia proliferation (Ait-Ikhlef et al. 1999; Parish et al. 2002). On the other hand, recent studies have shown that endozepines can modulate the immune response (Cosentino et al. 2000; Marino et al. 2003), suggesting that Aβ-induced stimulation of endozepine secretion may participate to the propagation of the inflammatory process and thus to the progression of Alzheimer's disease. In support of this hypothesis, it has been found that endozepines potentiate lipopolysaccharide-induced release of tumour necrosis factor α and IL-1β from macrophages (Taupin et al. 1993) and IL-6 from monocytes (Stepien et al. 1993). The observation that the endozepine ODN, as well as Aβ25−35, stimulates the release of IL-1β from cultured astrocytes provides additional evidence that endozepines may contribute to Aβ-evoked astrogliosis.
Besides their role in cell-to-cell signalling, endozepines also act as intracrine factors through activation of peripheral-type benzodiazepine receptors (PBR) located on the outer mitochondrial membrane (Papadopoulos 1993; Galiègue et al. 2003; Lacapère and Papadopoulos 2003). The observation that Aβ provokes an increase in DBI mRNA and accumulation of endozepines in cultured astrocytes thus suggests that the effect of Aβ on astrocyte activation can be ascribed, at least in part, to stimulation of PBR. Consistent with this notion, an increase in PBR ligand binding has been observed in the brain of Alzheimer patients (Diorio et al. 1991; Cagnin et al. 2001).
The mechanism by which Aβ stimulates the biosynthesis and release of endozepines in astrocytes is currently unknown. The present study has shown that the effect Aβ25−35 on DBI mRNA level and endozepine content is suppressed by cycloheximide, indicating that de novo protein synthesis is required for basal and Aβ-induced DBI gene. On the other hand, it has been previously reported that Aβ25−35 provokes an increase in intracellular calcium concentration in rat cortical astrocytes (Jalonen et al. 1997; Stix and Reiser 1998), suggesting that calcium mobilization may trigger DBI gene expression and/or endozepine secretion. Finally, it has been shown that Aβ25−35 induces the release of IL-1β from monocytes through activation of formyl peptide receptors (FPR) (Lorton et al. 2000). Whether FRP can mediate the effects of Aβ on DBI gene expression and endozepine release deserves further investigation.
The authors wish to thank Mrs Hugette Lemonnier and Mr Gérard Cauchois for skillful technical assistance. OM was the recipient of a fellowship from the French and Tunisian Ministry of Research. TT was the recipient of a fellowship from the Chinese Ministry of Education. This study was supported by grants from INSERM (U413), the European Institute for Peptide Research (IFRMP 23) and the Conseil Régional de Haute-Normandie.