Background
Neuroinflammation, characterized by excessive glial activation and overproduction of proinflammatory cytokine and chemokines, plays a critical role in the pathogenesis of neurodegeneration in Alzheimer’s disease (AD) and related neurodegenerative disorders [
1]. Inhibition of glial activation is shown to improve synaptic dysfunction and behavioral deficits in AD animal models [
2].
Amyloid-β (Aβ) peptides are key molecules in AD pathology as they aggregate to form amyloid plaques, the primary neuropathological hallmarks of AD [
3]. Aggregated Aβ peptides, in the forms of fibrils and oligomers, are prominent triggers for microglial activation [
4‐
6]. Accumulating evidence demonstrates that the redox-active Cu(II) ions can be coordinated to the histidine (His6, His13, and His14) or Tyr10 of monomeric Aβ. This coordination leads to formation of the Cu(II)-Aβ complex (predominantly with 1:1 stoichiometry), an aggregation process that involves a conformational change [
7‐
9]. Copper homeostasis is profoundly affected in AD. The concentration of extracellular Cu(II), especially within the amyloid plaques, is markedly elevated in AD brains [
10,
11]. An elevated Cu(II) concentration in the cerebrospinal fluid from patients with AD has also been reported [
12]. Micromolar concentrations of Cu(II) induce dramatic Aβ aggregation at a mildly acidic environment (pH 6.6), a condition that mimics the cerebral acidosis in AD [
13]. Therefore, it is possible that excess extracellular Cu(II), via driving Aβ aggregation, participates in AD pathology. The interactions between Cu(II) and monomeric Aβ are suggested to lead to enhancement of Aβ neurotoxicity [
14,
15]. It is unknown whether these interactions modulate the effect of Aβ on microglial activation.
The transcription factor nuclear factor (NF)-κB has been identified as a key regulator of cellular immune responses, including microglial activation [
16]. NF-κB activation can be directly driven by reactive oxygen species (ROS) [
17,
18]. ROS, derived from NADPH oxidase (NOX) or mitochondria, play important roles in regulating microglial responses to various stimuli. ROS from NOX may be involved in microglial activation stimulated by fibrillar Aβ, lipopolysaccharide (LPS), or Zn(II) ions [
5,
19,
20]. On the other hand, LPS and Cu(II) are reported to increase the microglial production of mitochondrial ROS (mtROS), which can drive the production of proinflammatory cytokines and chemokines [
21‐
24].
Therefore, the present study was aimed to determine (1) whether Cu(II) coordination enhances the effect of Aβ on microglial activation and the subsequent neurotoxicity, (2) whether this effect involves NF-κB and ROS, and (3) where these ROS are generated.
Materials and methods
Chemicals
Human Aβ1–40 or Aβ1–42, dihydroethidium (DHE), MitoTracker Green, and MitoSOX Red were purchased from Invitrogen (Carlsbad, CA, USA). Mouse anti-A11 or 6E10 antibodies were from Covance (San Diego, CA, USA). CuCl2, ZnCl2, cytosine arabinoside, pentoxifylline, aminoguanidine, N-acetyl-cysteine (NAC), apocynin, diphenylene iodonium (DPI), and LPS were from Sigma-Aldrich (St. Louis, MO, USA). BAY11-7082 and SC-514 were purchased from Merck (Darmstadt, Germany). Rabbit anti-inducible nitric oxide synthase (iNOS), IκB-α, phospho-IκB-α, p65, phospho-p65, β-actin, or microtubule-associated protein 2 (MAP2) antibodies were acquired from Cell Signaling Technology (Beverly, MA, USA). Mouse anti-CD11b antibody was purchased from Abcam (Cambridge, MA, USA).
Preparation of Cu(II)-Aβ complex, fibrillar Aβ, and oligomeric Aβ
Lyophilized Aβ1–40 or Aβ1–42 (1.0 mg) was dissolved in 1 ml hexafluoroisopropanol (HFIP) for 2 h at room temperature. Then, HFIP was removed under a gentle stream of nitrogen gas. The lyophilized peptide was dissolved in dimethyl sulfoxide (DMSO) to obtain a 500 μM stock solution. To generate the Cu(II)-Aβ complex, Aβ stock solution was brought to the required concentrations in 20 mM Hepes buffer (containing 153 mM NaCl; pH 6.6) in the presence or absence of CuCl2 (with an equimolar Aβ:Cu(II) ratio). The reaction mixtures were incubated for 24 h at 37 °C. Similarly, the Zn(II)-Aβ complex was prepared with Aβ1–40 and ZnCl2 (with an equimolar Aβ:Zn(II) ratio). In the cellular experiments, the Cu(II)-Aβ or Zn(II)-Aβ complex was diluted with the culture media to appropriate concentrations.
Fibrillar Aβ was prepared by dissolving Aβ
1–40 in ddH
2O, which was followed by incubation at 37 °C for 7 days [
5]. To prepare Aβ oligomers, Aβ
1–40 was dissolved to 500 μM in HFIP and was diluted to 50 μM with ddH
2O. After a 10- to 20-min incubation at room temperature, the samples were centrifuged at 14,000×
g for 15 min. The supernatant was transferred to a new siliconized tube and subjected to a gentle stream of nitrogen gas for 5–10 min to evaporate the HFIP. The samples were then stirred at 500 rpm using a Teflon-coated micro stir bar for 48 h at 22 °C [
25].
Aggregation assay
To determine the Cu(II)-induced Aβ aggregation, the reaction mixtures containing 50 μM Aβ
1–40 with or without 50 μM Cu(II) were incubated for 24 h at 37 °C; then, they were centrifuged at 13,000×
g for 15 min to sediment aggregated proteins. The soluble peptide concentration in the supernatant was determined using Coomassie Plus (Bradford) Assay Reagent (Thermo, Rockford, IL, USA) [
13,
26]. To quantify the amyloid fibrils, the thioflavin T fluorescence assay was applied [
26]. After incubation, the samples containing Cu(II)-Aβ
1–40, Aβ
1–40, or fibrillar Aβ
1–40 at 50 μM peptide concentration were diluted with 50 mM glycine-NaOH buffer (pH 8.5) containing 10 μM thioflavin T (Sigma-Aldrich) to a final volume of 100 μl. Fluorescence was monitored using a Varioskan Flash multimode reader (Thermo), with an excitation at 446 nm and emission at 490 nm.
Dot blot
Five microliters of samples containing Cu(II)-Aβ (with equimolar or subequimolar Aβ1–40:Cu(II) ratios) or oligomeric Aβ1–40 at a 10 μM peptide concentration were applied to a nitrocellulose membrane (Millipore, Bedford, MA, USA). The membrane was air dried and then blocked with 10 % nonfat milk in Tris-buffered saline with Tween 20 (TBST) overnight at 4 °C. Following three 5-min washes, the membrane was incubated with A11 antibody (1:1000) for 1 h at room temperature. After washing, the blots were incubated with horseradish peroxidase-conjugated anti-rabbit IgG (1:10,000, Santa Cruz, Dallas, TX, USA) for 1 h at room temperature. The blots were detected with a chemiluminescence kit (Pierce, Rockford, IL, USA). Then, the same membrane was stripped and immunoblotted with 6E10 (1:1000), as described above.
Microglial cell culture
The immortalized mouse microglial cell line BV-2 was routinely grown in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Grand Island, NY, USA) supplemented with 10 % fetal calf serum (FCS; Gibco), 100 U/ml penicillin, and 100 μg/ml streptomycin in a humidified atmosphere containing 5 % CO
2 at 37 °C. Primary microglial cultures from newborn (24 h) Sprague-Dawley pups (Shanghai Laboratory Animal Center, Chinese Academy of Science, Shanghai, China) were prepared as previously described [
21]. Briefly, cerebral cortex was dissected out and minced in cold DMEM. After a 20-min trypsinization (0.25 %), the cell suspension was centrifuged at 900×
g for 10 min. The pelleted cells were resuspended and seeded in 75-cm
2 tissue culture flasks containing DMEM supplemented with 10 % FCS. The mixed glial cultures were maintained at 37 °C in a humidified 5 % CO
2 atmosphere, and the medium was changed every 48 h. Once confluent (12–14 days), the flasks were shaken (37 °C, 72 rpm) for 6 h. Microglial cells were harvested, centrifuged, and seeded onto 96-well plates at a density of 5 × 10
4 cells/well.
To obtain conditioned media for treating neurons, primary microglia were first cultured in 96-well plates for 24 h in DMEM supplemented with 10 % FCS. Then, they were washed and changed to the Neurobasal medium with 2 % B27 supplement (Gibco) and cultured for 24 h in the presence or absence of Cu(II)-Aβ1–40, Aβ1–40 (preincubated at 37 °C for 24 h; if not mentioned, the preincubated Aβ1–40 was used in the following experiments), or Cu(II). The microglia-conditioned media were collected and briefly centrifuged; then, they were diluted with Neurobasal medium containing B27 to treat neurons.
Primary hippocampal neuron culture
Hippocampi dissected from pups at postnatal day 1 were used to prepare neuronal cultures as previously described [
27]. Briefly, pooled hippocampi were washed, triturated, and then treated with papain for 20 min. The mixture was centrifuged at 900×
g for 5 min, and the precipitate was resuspended with MEM (Gibco) containing 10 % FCS and 5 % horse serum. The cells were seeded onto a 96-well plate at a density of 1 × 10
4/well. The medium was changed to Neurobasal medium with B27 supplement 4 h after plating. To prevent the proliferation of nonneuronal cells, cytosine arabinoside (10 μM) was added to the cultures at the second day. Half of the medium was removed every 3 days and replaced with fresh medium. Neurons were cultured for 10–14 days for experiments.
Cell viability assay
Microglial cells and hippocampal neurons were seeded onto 96-well plates and allowed to grow. When they reached the required confluence, the medium was changed to serum-free medium, and the cells were treated with Cu(II)-Aβ1–40, Aβ1–40, or Cu(II) for 24 h. The survival of cells was determined by incubating the cells with a cell counting kit-8 (CCK-8; 5 mg/ml; Dojindo, Kumamoto, Japan) at 37 °C for 2 h. The absorbance was measured at a wavelength of 450 nm. The cell viability of each treatment group was calculated relative to the control group that was treated with the medium. In another experiment, hippocampal neurons were treated with conditioned media from Cu(II)-Aβ1–40-stimulated microglia. The cell viability was determined 24 h later.
Measurement of tumor necrosis factor-α (TNF-α) and nitric oxide production
The TNF-α concentration in the culture media of BV-2 or primary microglial cells was determined using an enzyme-linked immunosorbent assay kit (Pierce) according to the manufacturer’s instructions.
Griess reagent (Sigma-Aldrich) was used to measure the nitric oxide production. Briefly, 50 μl of medium from each culture well was placed in triplicate wells of a 96-well plate. An equal volume of Griess reagent was added to each well for 10 min. The optical densities were measured at 540 nm. The nitrite concentration was determined from a standard curve prepared by adding NaNO2 to DMEM.
Hydrogen peroxide (H2O2) assay
The production of H2O2 in BV-2 cells was quantified by a fluorometric assay with Amplex Red reagent (Molecular Probes, Eugene, OR, USA) according to the manufacturer’s instructions. Briefly, the reaction mixtures in Krebs-Ringer phosphate buffer with glucose (KRPG; 145 mM NaCl, 5.7 mM Na3PO4, 4.86 mM KCl, 0.54 mM CaCl2, 1.22 mM MgSO4, 5.5 mM glucose, pH 7.35) containing 50 μM Amplex Red and 0.1 U/ml horseradish peroxidase were prewarmed at 37 °C for 10 min; then, 1 × 105 cells resuspended in KRPG, and Cu(II)-Aβ1–40, Aβ1–40, Cu(II), or medium (for control group), were added. The absorbance at 560 nm was measured at different time points using a Varioskan Flash multimode reader (Thermo).
Cytosolic and mitochondrial ROS detection
BV-2 cells were seeded in 3.5-cm cell culture dishes at a density of 1.5 × 105/well. When the cells reached the required confluence, they were stimulated with Cu(II)-Aβ1–40, Aβ1–40, or Cu(II) for 1 h. Then, the cells were washed twice with PBS and labeled at 37 °C for 30 min with the following fluorescent probes: 0.1 μM DHE (oxidized by cytosolic ROS to fluorescent ethidium that binds nuclear DNA, Ex518 nm/Em605 nm), 200 nM MitoTracker Green (specifically labels mitochondria, Ex490 nm/Em516 nm), and 5 μM MitoSOX Red (specifically detects mitochondrial superoxide, Ex510 nm/Em580 nm). The fluorescence intensity was detected on a Nikon Eclipse 80i microscope (Melville, NY, USA).
Immunofluorescence staining
Primary microglial cells or hippocampal neurons were fixed with 4 % paraformaldehyde for 1 h and blocked in 10 % BSA for 1 h at room temperature. Then, they were stained overnight at 4 °C with mouse anti-CD11b (1:200) or rabbit anti-MAP2 (1:50) antibodies. After incubation with an Alexa Fluor 488 donkey anti-mouse or anti-rabbit IgG secondary antibody (Invitrogen), cells were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) and imaged by a Leica TCS SP2 AOBS confocal microscope (Leica, Wetzlar, Germany).
Western blot
BV-2 cells were plated at a density of 1.5×105 cells/ml into 6-well plates and allowed to grow. At different time points of Cu(II)-Aβ1–40 treatment, the cells were collected and lysed with a radioimmunoprecipitation assay (RIPA) buffer (50 mM Tris [pH 7.4], 150 mM NaCl, 1 % Triton X-100, 1 % sodium deoxycholate, 0.1 % sodium dodecyl sulfate [SDS], EDTA, sodium orthovanadate, sodium fluoride, and leupeptin) containing a protease inhibitor cocktail (Sigma-Aldrich). Equal amounts (30 μg) of proteins were loaded on a 10 % SDS-PAGE gel and transferred to polyvinylidene fluoride membranes. To avoid nonspecific binding, the membrane was blocked with 5 % nonfat milk in TBST for 1 h at room temperature and then incubated overnight at 4 °C with primary antibodies specific for iNOS, phospho-IκB-α, IκB-α, phospho-p65, p65, or β-actin at a 1:1000 dilution. The membrane was washed three times with TBST and incubated for 1 h with a horseradish peroxidase-conjugated secondary antibody (1:5000; Abcam). The bands were visualized using enhanced chemiluminescence (Pierce).
Data analysis and statistics
Data are represented as the means ± SEM. Statistical analyses were performed using Student’s t-test and one-way or two-way analysis of variance (ANOVA) followed by Bonferroni post-test comparisons. Differences of P < 0.05 were considered statistically significant.
Discussion
Our data show for the first time that Cu(II) enhances the effect of Aβ on microglial activation and the subsequent neurotoxicity. This activating effect is NF-κB-dependent and involves NOX-independent, mitochondria-derived ROS.
Cu(II) is suggested to bind monomeric Aβ peptides (predominantly with 1:1 stoichiometry) with high affinity, leading to the formation of Cu(II)-Aβ complex [
7‐
9]. In the present study, the results of an assay of Aβ aggregation by sedimentation, thioflavin T fluorescence assay, and dot blot analysis indicate the formation of Cu(II)-Aβ complex, whose conformation or nature differs from fibrillar or oligomeric Aβ and monomeric Aβ.
In the following experiments, we found that Cu(II)-Aβ induced an activating morphological phenotype of microglia and microglial release of TNF-α and nitric oxide. The increase in nitric oxide production was accompanied by an increase in iNOS expression. Moreover, a subneurotoxic concentration of Cu(II)-Aβ produced an indirect, microglia-mediated neurotoxicity, as demonstrated by the impaired neuronal viability and dendritic damage. Importantly, Aβ or Cu(II) alone induced neither the microglial release of soluble neurotoxic factors nor the microglia-mediated neurotoxicity. Therefore, it is likely that the interactions between Cu(II) and Aβ lead to enhancement of the Aβ effect on microglial activation. In contrast to the observations that either neutralization of TNF-α or inhibition of iNOS is sufficient to prevent neurotoxicity produced by aggregated Aβ-activated microglia [
29,
33], we found that repression of microglial release of both TNF-α and nitric oxide, but not of TNF-α or nitric oxide alone, was protective for hippocampal neurons. These data indicate that TNF-α and nitric oxide generated by Cu(II)-Aβ-stimulated microglia need to act synergistically to elicit neuronal damage.
NF-κB is an important regulator of the gene expression of proinflammatory mediators. The inhibitory IκB-α can be phosphorylated by activated IKK and then degraded, leading to phosphorylation and translocation of p65 and activation of gene transcription. In the present study, the inhibitors of IκB-α phosphorylation and IKK-2 prevented the Cu(II)-Aβ-induced microglial release of TNF-α and nitric oxide, suggesting that the actions of Cu(II)-Aβ depend on NF-кB. Cu(II)-Aβ stimulation also induced phosphorylation and degradation of IкB-α as well as phosphorylation of p65, further confirming the involvement of NF-кB.
ROS are critical for microglial activation [
20,
34]. We found that Cu(II)-Aβ (but not Aβ or Cu(II) alone) triggered extracellular H
2O
2 accumulation. NOX, an enzyme that catalyzes the production of superoxide, is a producer of extracellular and intracellular ROS. We found that the NOX inhibitors did not affect Cu(II)-Aβ-induced H
2O
2 production. Furthermore, Cu(II)-Aβ was unable to evoke superoxide release. These data indicate that Cu(II)-Aβ-elicited extracellular H
2O
2 is not from NOX.
Mitochondria serve as another major source of ROS in most cells [
31,
32]. Mitochondrion-generated superoxide, which does not cross the inner membrane, can be converted to membrane-permeable H
2O
2 by matrix Mn-superoxide dismutase, and H
2O
2 can diffuse to the cytosol and extracellular space (thus can be detected in the culture media) [
35]. We observed an accumulation of mtROS in BV-2 cells stimulated with Cu(II)-Aβ. Given the finding that Cu(II)-Aβ-elicited extracellular H
2O
2 is not from NOX, it is likely that H
2O
2 originates in the mitochondria. In the meantime, cytosolic ROS accumulated in cells stimulated with Cu(II)-Aβ, indicating that some mtROS were diffused to cytosol. There is evidence that mtROS can diffuse to cytosol to signal NF-кB activation and proinflammatory cytokine expression [
17,
22]. Therefore, our results strongly suggest that mtROS are involved in Cu(II)-Aβ-triggered microglial activation. Our findings that H
2O
2 release due to Cu(II)-Aβ stimulation is NOX-independent seem to be in contrast to those that fibrillar Aβ, LPS, or Zn(II) evokes NOX-dependent ROS production [
5,
19,
20]. However, we noticed that, similar to the recent findings of Park et al. [
23] and Voloboueva et al. [
24], LPS elicited robust mtROS production. Therefore, microglial activation upon inflammatory stimulations may involve both NOX- and mitochondrion-derived ROS.
Noteworthy, we found that NAC (a ROS scavenger that readily passes into cells) inhibited the Cu(II)-Aβ-elicited microglial release of TNF-α and nitric oxide as well as microglia-mediated neuronal death. These results suggest that mtROS production precedes the production of soluble neurotoxic factors. Oxidative stress, an early event in AD pathogenesis, plays a critical role in neuronal dysfunction and death in this disease [
36,
37]. Studies with murine models provide supportive evidence that administration of NAC blocks oxidative damage in AD [
38,
39]. Clinical trials also show that NAC treatment is potentially beneficial to AD patients [
38‐
41]. Our findings that NAC ameliorates neuronal injury mediated by Cu(II)-Aβ-activated microglia may further warrant the use of NAC in AD therapy.
In our work, preincubated (37 °C for 24 h) Aβ alone (up to 5 μM) had little effect on microglial activation. We noticed that the structure of these Aβ peptides is neither oligomeric nor fibrillar, which may be one explanation for this observation. Aβ can be taken up by microglia [
42]. So, it is possible that Cu(II)-Aβ is taken up by microglial cells to induce mtROS production and that the difference in potency for microglial activation between Cu(II)-Aβ and Aβ is due to uptake difference.
The role of microglia in the pathogenesis of AD might be switched from a beneficially phagocytic effect to a proinflammatory cytotoxic role (with the classic M1 phenotype) with the age-dependent accumulation of extracellular Aβ aggregates [
43]. At the early stage of AD, microglia might play a modulatory role via phagocytosis of Aβ plaques. In addition, microglia may exert protective functions through intracellular Cu(II) sequestration, which may prevent further plaque formation [
44].
Interestingly, recent evidence shows that Cu(I), which also promotes Aβ aggregation [
45], may polarize inflamed microglial populations from the neurotoxic (M1) phenotype to the neuroprotective (M2) phenotype via inhibition of nitric oxide production [
46]. This observation implicates that the valence state of copper may be important in the regulation of microglial phenotype. The microglial responses to Cu(I)-Aβ and Cu(II)-Aβ and the subsequent neurotoxicity might be different as a result of their different redox capability.
Taken together, the present study identifies Cu(II) as a cofactor that promotes the effect of Aβ on microglial activation and the subsequent neurotoxicity. Moreover, the Cu(II)-Aβ-triggered microglial activation may involve NF-кB activation and mtROS production. These findings support the notion that metal ion-induced Aβ aggregation might be important in microglia-mediated neuroinflammation and might provide a novel anti-inflammatory strategy for AD.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
FY, PG, and ZH performed the experiments; JL and XG analyzed the data and wrote the manuscript; and JL, YC, and HC designed the study. All authors read and approved the final version of the manuscript.