The CCAAT/enhancer binding protein (C/EBP) δ is differently regulated by fibrillar and oligomeric forms of the Alzheimer amyloid-β peptide

Primary mixed glial cultures consisting of 5-10% microglia and 90-95% astrocytes (hereafter referred to as astro-microglial cultures) were obtained from male Sprague Dawley pups less than 24 h of age as previously described [22]. The experimental procedures were approved by Stockholm North Local Committee on Ethics of Animal Experiments and performed in accordance with international standards on animal welfare. Cells were seeded in 60 mm culture dishes and maintained in DMEM glutamax containing 10% fetal bovine serum (FBS) and 0.1% penicillin-streptomycin (PEST) for 20-22 days (all from GIBCO). Medium was exchanged every 3-4 day. 18 h prior to experiment, medium was exchanged to DMEM glutamax containing 0.2% FBS and 0.1% PEST. Confluent cells were exposed to 10 μM Aβ(1-42), 10 μM Aβ(42-1) (American peptide), 10 ng/ml rat recombinant IL-1β (Biosource) and/or 1 μg/ml lipopolysaccharide (LPS, Sigma) for 3 h before analysis. Aβ preparations enriched in oligomeric or fibrillar forms were prepared as previously described [23, 24]. Briefly, Aβ was dissolved in hexafluoroisopropanol (HFIP) to a final concentration of 1 mM. HFIP was then removed by centrifugation under vacuum, and Aβ was then resuspended in dimethyl sulfoxide (DMSO, Sigma) to a final concentration of 5 mM. To obtain oligomers, culture medium was added to the peptide to a final concentration of 100 μM followed by an incubation at 4°C for 24 h. For a fibril-enriched preparation, 10 mM HCl was added to the peptide to a final concentration of 100 μM, followed by incubation for 24 h at 37°C. The different Aβ preparations were analyzed using western blot (see western blot) and Thioflavin T assay (T3516, Sigma). In the Thioflavin T assay, Aβ samples were analyzed in phenol red-free medium and according to the manufacturer's instructions.

Nuclear extracts were prepared as follows: cells were washed in ice-cold PBS, scraped and subsequently centrifuged for 2 min. Thereafter the pellet was washed again in PBS, centrifuged for 25 s and resuspended in HB buffer (10 mM Tris HCl pH 7.4, 10 mM KCl, 1.5 mM MgCl₂, 0.5 mM PMSF, 0.5 mM β-mercaptoethanol), centrifuged for 25 s and resuspended in lysis buffer (HB buffer, 0.4% Nonidet-40). Samples were then lysed on ice for 10 min. Nuclear fractions were pelleted by centrifugation for 5 min, and then resuspended in buffer C (10 mM HEPES pH 7.9, 0.4 M NaCl, 1 mM EDTA, 1 mM DTT, 1 mM PMSF) and thereafter placed on a shaking platform for 30 min and centrifuged for 5 min. The supernatant containing nuclear extract was collected and stored at -80°C for approximately 24 h before use. All steps were performed at 4°C and all centrifugations were performed at 16,000 g.

Transgenic mice, tg-ArcSwe, carrying both the Artic (E693G) and the Swedish (KM670/671NL) APP double mutations were originally of C57Bl/6-CBA-F1 background [25], but then bred on a C57Bl/6 background for more than six generations (incipient congenics). The experiments were approved by an ethical committee (Uppsala djurförsöksetiska nämnd) and performed in compliance with national and local animal care and use guidelines (protocol #C223/8). Animals were 15.5-17 month of age when sacrificed. At this age tg-ArcSwe mice have previously been shown to possess a plaque burden of approximately 4%, as measured by Aβ immunostaining of tissue sections followed by quantitative image analysis [26, 27]. Mice were anesthetized with 0.3 ml Avertin (25 mg/ml) and intracardially perfused with 0.9% saline solution. The brains were removed and dissected into amyloid-containing forebrain (including cortex and hippocampus) and amyloid-sparse hindbrain (cerebellum and brainstem) (cf. [25]). After dissection the samples were snap-frozen and stored in liquid nitrogen until use. As a control for changes related to age and Aβ deposition in the animals, 4 month-old tg-ArcSwe and non-transgenic animals were also investigated. Protein lysates were prepared as follows: forebrain and hindbrain were homogenized in ice-cold RIPA buffer (50 mM Tris HCl pH 8, 150 mM NaCl, 1% Nonidet-40, 0.5% sodium deoxycholate, 1% SDS, 1 mM PMSF) to a concentration of 25 mg/ml and placed on a shaking platform for 5 h at 4°C. Thereafter the samples were centrifuged for 20 min at 9,500 g and the protein containing supernatant was collected and stored at -80°C until used.

For measurement of total Aβ levels, ninety-six-well plates were coated with 50 ng/well of the N-terminal Aβ-specific antibody 82E1 (IBL-Hamburg, Hamburg, Germany) in PBS and blocked with 1% BSA in PBS with 0.15% Kathon (Rohm & Haas, Philadelphia, PA). Brain tissues homogenized in RIPA-buffer were supplemented with formic acid (to a final concentration of 70%), neutralized with 1 M Tris (pH 10) and diluted in ELISA incubation buffer (PBS with 0.1% BSA and 0.05% Tween-20). Samples or standard (recombinant Arctic Aβ1-40), were treated similarly, and added to the plates in duplicates and incubated for 2 h at room temperature with shaking. Biotinylated mAb27 (1 μg/ml), with a conformation-dependent epitope in the mid-domain of Arctic Aβ [28], was used to measure total Aβ. The biotinylated detection antibody as well as streptavidin-horseradish peroxidase (SA-HRP; diluted 1:2000; Mabtech, Nacka, Sweden) was allowed to incubate for 1 h in successive steps. K-blue aqueous (ANL-Produkter, Älvsjö, Sweden) was used as HRP-substrate, the reaction was stopped with 1 M H₂SO₄ and the optical density was measured at 450 nm with a SpectraMax 190 (Molecular Devices, Sunnyvale, CA). Wells were washed three times between each step in ELISA washing buffer (PBS with 0.1% Tween-20 and 0.15% Kathon).

Protein concentrations were estimated using Bicinchoninic acid assay (BCA, Pierce). Protein lysates (20 μg/lane from primary astro-microglial cultures and 40 μg/lane from transgenic mice tissues) were mixed with Laemmli sample buffer and boiled for 5 min. Proteins were separated on 12% SDS-polyacrylamide gels and thereafter blotted onto polyvinylidene diflouride (PVDF) membranes (Amersham Pharmacia). For analysis of oligomer- and fibril-enriched Aβ preparations, samples were not preheated and they were run on a 12% SDS-polyacrylamide gel without stacking gel. The membranes were incubated in PBS containing 5% milk, 0.1% Tween-20 (Bio-Rad) and 0.1% BSA (Sigma) for 2 h at room temperature to block unspecific binding, and thereafter incubated in the above-mentioned PBS solution with primary antibody overnight at 4°C. Antibody concentrations were as follows: C/EBPα ((14AA)X) and C/EBPδ ((C-22)X, Santa Cruz Biotechnologies) 1:4000 for primary astro-microglial cultures and 1:1000 for transgenic samples, β-actin (A-2066, Sigma) 1:5000 and 6E10 (ab49682, Abcam) raised against Aβ(1-17) 1:2000. Next, membranes were washed in PBS containing 2.5% milk and 0.1% Tween-20 for 4 × 15 min and thereafter incubated with secondary antibody HRP coupled anti-rabbit IgG or anti-mouse IgG (GE Healthcare) 1:5000, diluted in the same solution, for 1 h at room temperature. Finally, membranes were washed 4 × 15 min in PBS containing 0.1% Tween-20. Blots were incubated in ECL Plus reagents (GE Healthcare) for 5 min and exposed to Hyper film (GE Healthcare). Membranes exposed to anti-C/EBPβ ((C19)X, Santa Cruz Biotechnologies) were first washed 5 min in TBS and rinsed 10 s in methanol [29]. Thereafter the membranes were allowed to dry before overnight incubation at 4°C with primary antibody, 1:500, diluted in TBS containing 5% milk and 0.05% Tween-20. Membranes were then washed 2 × 15 s in TBS containing 0.05% Tween-20 and incubated for 1 h at RT with secondary antibody (HRP coupled anti-rabbit IgG) 1:5000, diluted in TBS containing 5% milk and 0.05% Tween-20. Finally, membranes were washed 2 × 30 min in TBS containing 0.05% Tween-20, incubated in ECL Plus reagents for 5 min and exposed to Hyper film.

Protein concentrations were established using BCA assay. EMSA was performed as previously described [21, 30] with minor modifications. Briefly, C/EBP sense oligonucleotide 5'-TGCAGATTGCGCAATCTGCA-3 (MWG) was labeled with γ-³²P dATP (Perkin Elmer) using T4 polynucleotide kinase (Fermentas). Nuclear extracts (5 μg) from primary astro-microglial cultures were incubated with anti-C/EBPδ ((C-22) X, Santa Cruz Biotechnologies) for 30 min at room temperature prior to binding reaction with ³²P-labeled probe. Samples were separated on a 5% polyacrylamide gel at 150 V for 1 h. Gels were exposed to a phosphorimager screen overnight.

C/EBPδ binding to the κB DNA element was analyzed by streptavidin-agarose pull-down [31, 32]. Single stranded biotinylated κB sense oligonucleotide 5'-AGTTGAGGGGACTTTCCCAGGC-3' and antisense oligonucleotide 5'-GCCTGGGAAAGTCCCCTCAACT-3' (MWG), were hybridized by incubation at 100°C for 1 h and then allowed to cool down slowly (for approx. 30 min) to room temperature. Nuclear extracts (150 μg) from primary astro-microglial cultures were mixed with streptavidin-agarose bead suspension (according to the manufacturer's recommendation; S1638, Sigma) and 4 μg of double stranded biotinylated oligonucleotides in 500 μl of buffer B2 (PBS, 1 mM EDTA, 1 mM DTT, protease inhibitor cocktail Complete (Roche Diagnostics)). The mixture was then placed on a shaking platform for 2 h at room temperature. Thereafter the samples were centrifuged at 550 g for 1 min and the pellet washed 3 times in buffer B2. Finally the pellet was dissolved in 35 μl 2x Laemmli sample buffer, incubated at 100°C for 5 min and centrifuged at 7000 g for 30 s and the supernatant was collected. Samples were loaded on a 12% SDS-polyacrylamide gel and analyzed by western blot. As a control for unspecific binding to the agarose beads, identical κB oligonucleotides lacking biotin were used.

Data was analyzed using student's t-test or one-way ANOVA followed by Tukey's post hoc test.

The effect of Aβ on expression levels of C/EBPα and δ in primary astro-microglial cells was analyzed by western blot. Two different preparations of Aβ1-42 were used: one containing high levels of oligomers and one containing high levels of fibrils. The different Aβ preparations were analyzed using western blot and Thioflavin T assay (Figure 1A,B). Cells were treated with Aβ enriched in oligomers or fibrils in the presence or absence of IL-1β (10 ng/ml) for 3 h prior to harvesting and preparing nuclear extracts. IL-1β, which is known to be induced in AD brains, was used to activate the cells. For comparison, cells were also treated with LPS (1 μg/ml). Two different isoforms of C/EBPα; p30 and p42, which were both significantly down-regulated by LPS treatment, were detected (Figure 1C,D and 2A,B). This is consistent with earlier findings in primary glial cultures showing that C/EBPα is mainly regulated through toll-like receptors [33]. Neither Aβ nor IL-1β alone had any effect on the levels of the two isoforms of C/EBPα. However, it should be noted that the levels of p30 and p42 appeared to be reduced when cell cultures were stimulated with fibril-enriched Aβ preparations together with IL-1β although this effect was not statistically significant.

Both IL-1β and LPS induced up-regulation of C/EBPδ (Figure 1C,E and 2A,C). Aβ enriched in oligomers (Figure 1C,E) or fibrils (Figure 2A,C) did not have any significant effect on basal C/EBPδ levels. However, treatment with Aβ containing high levels of fibrils did significantly decrease IL-1β-induced C/EBPδ levels by ~75% (Figure 2A,C). Neither Aβ preparations devoid of fibrils nor the reversed peptide Aβ42-1 (prepared as the Aβ1-42), used as a control, had any effect on IL-1β-stimulated up-regulation of C/EBPδ (Figure 1C,E and 2A,C).

Based on these results, the effects of Aβ fibrils in vivo were investigated. Tg-ArcSwe mice (15.5-17-months-old), with both the Artic (E693G) and Swedish (KM670/671NL) APP double mutations, were chosen since aged animals exhibit an abundance of Aβ fibrils [26]. A significant down-regulation of C/EBPα was found in both forebrain and hindbrain of aged tg-ArcSwe mice. C/EBPα p30 levels were ~31-33% lower compared with age-matched non-transgenic littermates (Figure 3A,B). Unfortunately the effect on p42 could not be determined due to co-migration of strong nonspecific bands on the western blots. In contrast to a previous report of findings in human AD patient tissue [20], the levels of C/EBPδ were decreased ~62% in forebrain of aged tg-ArcSwe mice compared with age-matched non-transgenic mice (Figure 4A). Total Aβ levels in forebrain from these aged tg-ArcSwe mice were 227 ± 48 ng Aβ/mg tissue (mean ± SEM, N = 6) and were below the detection limit (i.e., <20 pg Aβ/mg tissue) in non-transgenic mice. Our findings are consistent with the results from IL-1β-activated astro-microglial cultures treated with fibrilar Aβ (Figure 2A,C). Analysis of the relatively Aβ plaque-free hindbrain showed a down-regulation of C/EBPα p30 levels similar to that found in forebrain (Figure 3B) indicating that this is not a local Aβ plaque-dependent phenomenon. In contrast, no difference in C/EBPδ levels was observed in hindbrain when transgenic mice and non-transgenic age-matched littermates were compared (Figure 4B). To ensure that the changes in levels of C/EBPs were due to age-dependent accumulation of Aβ and formation of Aβ fibrils, and not some unrelated effect of the transgene, brain tissue from young tg-ArcSwe mice with low Aβ load were analyzed and compared with age-matched non-transgenic mice. These experiments showed that there are no significant differences between tg-ArcSwe mice and non-transgenic mice with respect to C/EBPα p30 (Figure 3C,D) and C/EBPδ levels (Figure 4C,D). However, direct comparison of non-transgenic and tg-ArcSwe mice of different ages indicated that C/EBPδ levels are down-regulated both in response to age and Aβ plaque burden (Figure 5A,B). The C/EBPδ levels in forebrain were significantly decreased in aged non-transgenic mice by ~51% and further decreased in aged tg-ArcSwe mice by ~92% compared to young animals. In contrast, the LAP isoform (~35 kDa) of C/EBPβ was significantly up-regulated (~27% higher) in the forebrain of aged tg-ArcSwe mice as compared to age-matched non-transgenic control (Figure 6A,B). C/EBPβ LAP levels were below or close to detection limit in the hindbrain of aged mice and in both forebrain and hindbrain of young animals and could therefore not be quantified.

To further investigate the effects of fibrillar forms of Aβ, EMSA was performed to analyze C/EBPδ binding activity. Primary astro-microglial cells were treated for 3 h with Aβ fibrils in the presence or absence of IL-1β prior to harvesting. The results show that C/EBPδ binding to double-stranded oligonucleotides containing a C/EBP binding site is down-regulated by ~89% in cultures concomitantly treated with IL-1β and fibril-enriched Aβ preparations compared to cells treated with IL-1β alone (Figure 7A,B). The down-regulation in DNA binding activity reflects the down-regulation in protein level.

In order to test the hypothesis that C/EBPδ can shift its preferable DNA binding site and binding partners, and to analyze the effects of Aβ, C/EBPδ binding to a κB response element was investigated. Nuclear extracts from primary astro-microglial cultures treated for 3 h with IL-1β together with Aβ were analyzed in streptavidin-agarose pull-down experiments. Cells treated with IL-1β showed increased binding to the κB element compared to untreated control cells. In the presence of fibril-enriched Aβ, IL-1β-induced binding of C/EBPδ to the κB site was completely blocked (Figure 8A,B). We also analyzed the effect of oligomer-enriched preparations of Aβ. Surprisingly, the oligomeric conformation of Aβ decreased C/EBPδ binding to the κB site to the same extent as the fibrillar form in IL-1β activated cells (Figure 8A,B).