Microglia CREB-Phosphorylation Mediates Amyloid-β-Induced Neuronal Toxicity

Abstract

Extracellular accumulation of amyloid-β (Aβ) forming senile plaques is one of the hallmark pathologies in Alzheimer’s disease (AD), while the mechanisms underlying the neuronal toxic effect of Aβ are not fully understood. Here, we found that intracerebroventricular infusion of the aged Aβ₄₂ in mice only induces memory deficit at 24 h but not at 7 days. Interestingly, a remarkably increased CREB (cAMP response element-binding protein) Ser133-phosphorylation (pS133-CREB) with microglial activation was detected at 24 h but not at 7 days after Aβ infusion. Aβ treatment for 24 h increased pS133-CREB level in microglia of the hippocampal non-granular cell layers with remarkably decreased pS133-CREB immunoreactivity in neurons of the hippocampal granular cell layers, including CA1, CA3, and DG subsets. Inhibition of microglia activation by minocycline or CREB phosphorylation by H89, an inhibitor of protein kinase A (PKA), abolished Aβ-induced microglia CREB hyperphosphorylation with restoration of neuronal function and attenuation of inflammatory response, i.e., reduced levels of interleukin-6 (IL6) and pCREB binding of matrix metalloproteinase-9 (MMP9) DNA. Finally, treatment of the primary hippocampal neurons with Aβ-potentiated microglia media decreased neuronal GluN1 and GluA2 levels, while simultaneous inhibition of PKA restored the levels. These novel findings reveal that intracerebroventricular infusion of Aβ only induces transient memory deficit in mice and the molecular mechanisms involve a stimulated microglial CREB phosphorylation.

Keywords

Alzheimer’s disease amyloid-β cAMP response element-binding protein microglia protein kinase A

INTRODUCTION

Alzheimer’s disease (AD) is the most common form of dementia in the elderly. Pathologically, AD is characterized by extracellular aggregates of the amyloid-β (Aβ) forming senile plaques and intraneuronal accumulation of the hyperphosphorylated microtubule-associated protein tau, forming neurofibrillary tangles [1 –3]. Aβ is the main constituent of brain parenchymal and vascular amyloid, and it contributes to cerebrovascular lesions and is neurotoxic [4, 5]. In many different Aβ species, Aβ₄₂ is much more toxic than Aβ₄₀, possibly because Aβ₄₂ has stronger tendency to aggregate [6]. Early studies suggest that Aβ can induce oxidative stress [7], mitochondrial dysfunction [8], glial responses [9], and lipid dysregulation [10] in AD.

As the resident macrophage cells, microglia act as the first and main form of active immune defense in the central nervous system. They are constantly scavenging plaques, damaged or unnecessary neurons and synapses, and infectious agents [11]. At the same time, microglia activation also damages the central nervous system in pathological condition. For instance, the pro-inflammatory microglia can produce and secret reactive oxygen species and cytokines in response to various environmental stimuli, and causes neuronal injury and death if this response is prolonged. In the early stages of neurodegeneration disease like AD, microglia are activated to the inflammatory responses [12].

As a ubiquitous and constitutively expressed transcription factor, cAMP response element-binding protein (CREB) has an important role in synaptic plasticity and long-term memory formation in the brain and has been shown to be integral in the formation of spatial memory [13], and as well as in neuronal survival [14]. CREB is downregulated in the patients of AD and increasing the expression of CREB is being considered as a possible therapeutic target for AD. The CREB protein has various functions in different brain areas [15], and its signaling pathway has been studied in both affective [16] and neurodegenerative [17, 18] disorders. To express a target gene, CREB requires phosphorylation at the critical serine-133, representing a point at which several major signal pathways converge [19]. Protein kinase A (PKA) is the upstream kinase of CREB and regulates the phosphorylation of CREB at Ser133. In neurons, the PKA/CREB signal pathway is a key regulator of synaptic plasticity and memory capacities [20 –24].

To explore the mechanisms underlying the neuronal toxic effects of Aβ₄₂ in the current study, we found unexpectedly that the brain ventricular infusion of aged Aβ₄₂ only induced a transient learning and memory deficit. by increasing microglia CREB phosphorylation, which in term caused neuronal impairments.

MATERIALS AND METHODS

Animals and treatments

Wildtype C57BL mice (2 months old, male) were purchased from Beijing Huafukang, and the mice were housed under a 12-h light/dark cycle and kept with accessible food and water at 25°C. Aβ₄₂ (Qiangyao Biotechnology) was aged according to the protocol described previously [25]. The Aβ₄₂ (2μg/μl) was dissolved in dimethyl sulfoxide (DMSO), and then was incubated at 37°C for 7 days before use. The mice were anesthetized and injected through brain unilateral ventricle with the aged Aβ₄₂ (5μl), and the control group were injected with same volume of sterile normal saline containing DMSO (1%). The coordinates for the injection were as follows: anterior-posterior: –0.3 mm; mediolateral: –1 mm; dorsoventral: –2.3 mm (from bregma and dura, flat skull). The behavior tests were carried out at 24 h or 7 days after the treatment, and then the mice were sacrificed for the following biochemical measurements. To inhibit microglia, the mice were received intraperitoneal injection of minocycline (50 mg/kg/day, Sigma) for 10 days before stereotactic brain infusion of Aβ₄₂ and the other groups were received intraperitoneal injection with normal saline.

Novel object recognition

The test equipment is a white, square chamber with a bottom of 50×50 cm. Before each behavioral experiment was performed on each mouse, the test bench was sprayed with 75% alcohol and wiped clean. For 24 h groups, the mice were placed in the empty chamber to explore for 5 min before operation (habituation). At 6 h after the surgery, the mice were put back to the central area with two objects placed at the two corners of the chamber for 5 min (familiarization). After 24 h, one of the objects was replaced with the new one and the mice were put back in the central area of the chamber to explore for 5 min (test), and the times of the mice explored for the new object was recorded. For 7 days groups, the task started with a habituation trial on day 5 (5 days after the surgery) followed by a familiarization trial (day 6) in which two objects were presented to the animals and the test trial (day 7). To ensure the rigorous experiments, the order in which behavioral experiments were performed was kept as the same order as the familiarization on the first day and the test on the second day.

Western blotting

Western blotting was performed per the methods well established in our laboratory [26]. The hippocampi of the mice were dissected and homogenized with RIPA buffer (Beyotime) on ice. The proteins were separated by 10% SDS-PAGE, transferred onto nitrocellulose membranes and then blocked by 5% BSA for 1 h in room temperature. After washings by Tween-TBS, the membranes were incubated in primary antibodies overnight, and then incubated with secondary antibodies for 1 h. The blots were visualized by using Odyssey Infrared Imaging System (Licor Biosciences, Lincoln, NE, USA). The primary antibodies used in the present study include CREB (cell signaling, 1:1000), pS133-CREB (cell signaling, 1:1000), GluN2A (cell signaling, 1:1000), GluN2B (abcam, 1:1000), GluN1 (Millipore, 1:500), GluA1 (abcam, 1:1000), GluA2(Millipore, mab397), PSD93(abcam,ab2930), PSD95 (cell signaling, 1:1000), CaMKII (cell signaling,3362), synaptotagmin (abcam, ab13259), and β-actin (abcam, ab6276).

Golgi staining

After anesthetized, the mice were perfused with 0.9% NaCl and then the brains were immersed in the Golgi solution for 14 days. After dehydrated in 30% sucrose-PBS for 48 h, the brains were cut 100μm with a vibrating microtome (VT1000S, Leica, Germany). Then the slices were moved into CXA solution for staining.

Immunohistochemistry and immunofluorescence

After anesthetized, the mice were perfused by normal saline and 4% paraformaldehyde (PFA). The brains were removed and post-fixed in PFA for 24 h and then dehydrated in 30% sucrose-PBS for 48 h. After embedding, brains were cut into 30μm coronal sections. For immunohistochemistry (IHC) and immunofluorescence (IF), the experiments were carried out according to the established methods [27]. The brain slices were incubated with the same primary antibodies (Iba1, abcam, 1:200; pS133-CREB, cell signaling, 1:100; NeuN, Millipore, 1:50; GFAP, cell signaling, 1:200). For IHC, the images were observed using a microscope (Nikon, Tokyo, Japan). For IF, the images were captured using a Carl Zeiss LSM710 confocal microscope.

BV2 culture

Bv2 mouse microglia cells were grown in high glucose Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum and they were maintained in 95% humidified and 5 % CO₂. After 6 h treated with the aged Aβ₄₂ (2.5μM) or H89 2HCl (15μM), the cell extracts were used for western blotting and other tests. For studying the effects of microglia on the neurons, the bv2 cells were treated with the aged Aβ₄₂ (2.5μM) or simultaneously with H89 (15μM) for 1 h, then the media were replaced with fresh DMEM without Aβ₄₂ and H89 for 5 h. Then culture media were collected and added (1:9, v/v) into the primary hippocampal neurons cultured for 7 days in vitro (7 div). After 5 h, the neuronal extracts were prepared for the biochemical measurements.

Primary neuron culture

The hippocampal neurons were isolated from rats at embryonic day 15–18 (E15-18). The brain tissues were cut into small pieces and digested with 0.125% trypsin solution for 15 min. Neurons were cultured in dishes coated with 100μg/ml poly-D-lysine. The medium contains 97% neurobasal medium (Gibco), 2% B-27 (Gibco) and 1× GlutaMAX (Gibco). After cultured for 7 div, the bv2 media were added (1:9, v/v) at day 8 and the neurons were cultured for another 5 h.

Chromatin immunoprecipitation assay

Chromatin immunoprecipitation (ChIP) was analyzed using a kit (#17–371, Merk Millipore) by following the manufacturer’s instructions. To quantify CREB-bound level of MMP-9 DNA, qPCR was performed with following primers: MMP-9- ChIP-F: GGATTCAGGAGTCAGAGCAC, and MMP-9-ChIP-R: CATTGCTTCAAAGCACAGAC.

Statistical analyses

Data were analyzed by Student’s t–test or one-way ANOVA (GraphPad Software) and presented as mean±SEM. The statistically significance was set at p < 0.05.

RESULTS

Intracerebroventricular infusion of Aβ₄₂ only induces transient memory deficit with a remarkably increased CREB phosphorylation in hippocampus

To study the effect of Aβ on learning and memory in mice, we injected stereotaxically the aged Aβ₄₂ into the lateral ventricle of 2-month-old wildtype mice. The control group was injected with the same volume of normal saline (containing 1% DMSO, v/v). The novel object recognition test was performed at 24 h and 7 days after the surgery. We observed that the Aβ₄₂ injected mice showed significantly decreased exploring times to the novel object compared with the control group at 24 h after infusion (Fig. 1A). However, no significant difference was detected between Aβ₄₂ and the control groups at 7 days after the infusion (Fig. 1B). To explore the molecular bases underlying the observed phenomenon, we examined the activity or expression level of CREB and synapse-associated proteins at 24 h and 7 days after the infusion. Compared with the control group, PSD93 was significantly increased and GluA2 was decreased in Aβ₄₂ mice at 24 h after the infusion, while no difference of GluN2A, GluN2B, GluN1, GluA1, and PSD95 was detected between Aβ₄₂ and the control group (Fig. 1C, E). Interestingly, we observed that the phosphorylation level of CREB at Ser133 (pS133-CREB) was significantly increased in Aβ₄₂ mice at 24 h after the infusion (Fig. 1C, E). No significant difference of pS133-CREB and the synapse-associated proteins was found between the two groups at 7 days after the infusion (Fig. 1D, F). We also observed that number of dendritic spines in mouse brain slices treated with Aβ₄₂ for 24 h was decreased compared with the control group (Fig. 1G, H), but this difference was not shown in 7 days group (Fig. 1I, J). These data suggest that Aβ₄₂ can cause acute damage to the brain leading to learning and memory impairments in mice, while this damage may be recovered naturally after 7 days.

Fig. 1.

Intracerebroventricular infusion of aged Aβ only induces transient memory deficit with remarkable elevation of pCREB. The mice (2 months old) received unilateral ventricular infusion of the aged Aβ. After 24 h or 7 days, the memory ability was tested by novel object recognition (NOR). Then, the mice were sacrificed and the hippocampi were collected for measurement of the memory-associated proteins and spine densities. A) The exploring times for novel object was decreased in mice exposed to Aβ for 24 h with no difference in movement speed (n = 10 per group). B) No difference of exploring times between Aβ and control mice at 7 days (n = 7 per group). C, E) Levels of Ser133-phosphorylated CREB (pCREB) and PSD93 were significantly increased and GluA2 decreased, while no changes in total CREB and other synapse-associated proteins were detected in 24 h Aβ group measured by western blotting (n = 6 for each group). D, F) No difference of the synapse-associated proteins and CREB were detected between Aβ and Ctrl groups at 7 days (n = 5). G, H) The spine density was decreased in 24 h Aβ group (Scale bar = 2μm). I, J) No difference of spine densities was detected between two groups at 7 days. Data were presented as mean±SEM (unpaired t test, ^*p < 0.05, ^**p < 0.01).

Intracerebroventricular infusion of Aβ only transiently activates microglia

Using microglia marker Iba-1 for immunohistochemical staining, we observed that microglia were extensively activated shown by the increased number and enlarged diameter of microglia cells in the hippocampus of 24 h Aβ₄₂ treated mice (Fig. 2A). Quantitative analyses showed that the average diameter of microglia was increased to 1.5-fold of the control group (Fig. 2C) with a remarkably increased number of microglia at 24 h after Aβ₄₂ treatment (Fig. 2D), while these abnormalities were not detected at 7 days (Fig. 2B, E, F). These data suggest that Aβ₄₂ may damage neurons through activation of microglia in the acute phase (24 h) after Aβ₄₂ treatment. Immunohistochemical staining with the astrocyte marker GFAP showed that the number and morphology of astrocytes in the Aβ₄₂ group was not changed compared with the control group at 24 h (Fig. 2G, H), indicating that Aβ does not significantly affect astrocytes at the acute phase. Furthermore, the number of neurons in the hippocampus of the Aβ₄₂ group was also not significantly changed compared with the control group (Fig. 2I, J), suggesting that Aβ₄₂ treatment does not cause neuron death in acute phase.

Fig. 2.

Intracerebroventricular infusion of aged Aβ only transiently activates microglia. A, C, D) Aβ exposure activates microglia demonstrated by the increased Iba1 immunoreactivity and the enlarged diameter of the microglia in hippocampus of the 24 h group. B, E, F) No activation of microglia cells was detected at 7 days after Aβ administration. G, H) No difference of GFAP staining (astrocytes marker) in hippocampus of 24 h group. I, J) No difference of NeuN staining (neuron marker) in hippocampus of 24 h group. Scale bar = 100μm, n = 4 for each group. Data were presented as mean±SEM (unpaired t test).

Intracerebroventricular infusion of Aβ increases microglial pS133-CREB level

The aforementioned results showed that level of pS133-CREB in hippocampus of the 24 h Aβ₄₂ group increased significantly (Fig. 1C, E). However, the results of immunohistochemical staining showed that the expression level of pS133-CREB in the hippocampal granule cell layer and pyramidal layer (including DG, CA3, CA1) was significantly lower than that of the control group at 24 h after Aβ₄₂ infusion; however, in other non-granular cell layer areas of the hippocampus, the expression of pS133-CREB was significantly increased compared with the control group (Fig. 3A), suggesting that Aβ₄₂ has different activation states of CREB in different cell types. To further explore the enhanced expression of pS133-CREB cell types, we performed immunofluorescence co-localization staining. The immunoreactivity of pS133-CREB was significantly reduced in the hippocampal granule cell layer, while extensive co-localization of pS133-CREB with Iba-1, a marker of microglia, was detected in the non-granular cell layer at 24 h after Aβ₄₂ treatment (Fig. 3B). By co-immunofluorescence staining of NeuN and pS133-CREB, we found that the expression of pS133-CREB in the NeuN-positive cells of the Aβ₄₂ group was significantly lower than that of the control group in the hippocampus (Fig. 3C). At day 7, no significant change of pS133-CREB in the granule cell layer was detected in Aβ₄₂ group (Fig. 3D). These results reveal that Aβ₄₂ may activate microglia by increasing the phosphorylation of CREB, by which it causes neuronal damages in acute phase.

Fig. 3.

Intracerebroventricular infusion of aged Aβ remarkably increases microglial pS133-CREB with reduced neuronal immunoreactivity at 24 h. A) Aβ remarkably increased pS133-CREB immunoreactivity in hippocampal non-granular cell layers and non-pyramidal layer with significantly decreased staining in granule cell layers (DG) and pyramidal layer (including CA1, CA3), measured at 24 h by immunohistochemical staining (Scale bar = 100μm), or (B) by co-immunofluorescent staining of Iba1 and pS133-CREB (Scale bar = 50μm). C) Aβ significantly decreased neuronal pS133-CREB immunoreactivity measured by co-immunofluorescent staining of NeuN and pS133-CREB in hippocampal of 24 h group (Scale bar = 50μm). D) Aβ does not significantly change glial pS133-CREB immunoreactivity by co-immunofluorescent staining of Iba1 and pS133-CREB in hippocampal CA1, CA3, DG of the 7 days group (Scale bar = 50μm).

Inhibition of microglial activity attenuates Aβ-induced microglia elevation of pCREB with improved memory

Minocycline is a potent microglial inhibitor. After intraperitoneal infusion of minocycline for 10 days, the aged Aβ₄₂ was infused into the lateral ventricle (Fig. 4A). The novel object recognition test showed that minocycline restored the exploring times to the novel object (Fig. 4C) without changing the movement speed of the mice (Fig. 4B), suggesting improvement of cognitive function by minocycline. Immunohistochemistry data showed that minocycline inhibited Aβ₄₂-induced microglia activation (Fig. 4D), shown by decreasing number and diameter of the microglia (Fig. 4E, F). By co-immunofluorescence staining of pS133-CREB and Iba-1, we observed that inhibition of microglia by minocycline attenuated Aβ₄₂-induced elevation of pS133-CREB co-stained with Iba-1 (Fig. 4G) with reduction of IL-6 (Fig. 4H). Among several synapse-associated proteins, we observed that GluA2 was remarkably decreased in Aβ₄₂ treated mice at 24 h after the infusion, while minocycline restored the level of GluA2 (Fig. 4I, J). These data suggest that inhibiting microglial activation attenuates its CREB phosphorylation with attenuation inflammation and synapse impairment and restoration of learning and memory.

Fig. 4.

Inhibition of microglia prevents Aβ-induced microglial pS133-CREB expression and impairments of memory. A) The mice were received intraperitoneal injection of minocycline (Mino) for 10 days before Aβ infusion. B) There was no difference in movement speed between the four groups of mice (n = 10 per group). C) The exploring time for novel object was decreased in mice exposed to Aβ for 24 h while Mino reversed the reduction. D-F) Mino inhibits microglia activation demonstrated by reduced Iba1 immunoreactivity and the diameter of the microglial cells in hippocampus (Scale bar = 25μm, n = 4). G) Mino attenuates Aβ-induced pCREB expression in microglia measured by co-immunofluorescent staining of Iba1 and pS133-CREB in the hippocampus at 24 h after Aβ infusion (Scale bar = 10μm). H) Mino attenuates Aβ-induced elevation of IL-6 in hippocampus measured by ELISA (n = 5). I, J) Mino restored Aβ-induced reduction of GluA2 in hippocampal extracts measured by western blotting (n = 6). Data were presented as mean±SEM (one-way ANOVA, ^*p < 0.05, ^**p < 0.01, ^***p < 0.001).

Aβ activates microglia through CREB phosphorylation

To further verify the effect of Aβ on microglia, we treated the mouse microglial cell line bv2 with aged Aβ₄₂ for 6 h. The results showed that the level of pS133-CREB in Aβ₄₂-treated bv2 cells was significantly increased compared with the control group with a simultaneous upregulation of PKA, an important upstream kinase for CREB (Fig. 5A, B). The level of IL-6 in the bv2 culture medium was also significantly increased after Aβ treatment (Fig. 5C). Matrix metalloproteinase-9 (MMP-9) is an important protease secreted by microglia that participates in inflammatory reactions. Chromatin immunoprecipitation (ChIP) results showed that the binding of MMP-9 DNA to pCREB was significantly increased in Aβ₄₂ treated bv2 cells (Fig. 5D). After treatment of bv2 with PKA inhibitor (H89 2HCl) and inhibition of CREB phosphorylation (Fig. 5A, B), the level of IL-6 in bv2 culture medium was decreased (Fig. 5C), and the binding of MMP-9 DNA and pCREB was also decreased (Fig. 5D). These results suggest that Aβ₄₂ activates microglia PKA-CREB signaling pathway, and thus increases the expression of MMP-9 and promotes IL-6 secretion by increasing CREB phosphorylation.

Fig. 5.

Aged Aβ activates microglia and induces inflammation through CREB phosphorylation in Bv2 cells. Bv2 cells were treated with Aβ or simultaneously Aβ and H89 for 6 h. A, B) Aβ treatment increased CREB phosphorylation at Ser133 (pCREB) with activation of PKA measured by using PKA-substrate assay and western blotting, while inhibition of PKA by H89 attenuated Aβ-induced CREB phosphorylation (n = 6). C) Aβ increased IL-6, while inhibition of PKA-associated CREB phosphorylation by H89 restored IL-6 level in the culture medium measured by using an ELISA kit (n = 6 per group). D) Aβ increased binding level of MMP-9 DNA and pCREB detected by using a ChIP kit (n = 6 per group). Data were presented as mean±SEM (one-way ANOVA, ^*p < 0.05, ^**p < 0.01, ^***p < 0.001).

Aβ affects neuronal synaptic protein expression and development through microglial PKA-dependent manner

To further verify whether Aβ₄₂ causes neuron damages through microglia, we treated bv2 cells with Aβ₄₂ for 1 h, and then replaced the old medium with fresh DMEM and cultured for 5 h. The media were collected and added to the primary cultured hippocampal neurons at a ratio of 1:9 (v/v, 10% bv2 medium) for 5 h. Culture of the neurons with Aβ-potentiated media decreased expression level of synaptic proteins GluN1, GluA2, and synaptotagmin (Syt) without affecting GluN2A and GluN2B, while simultaneous inhibition of PKA in microglia by H89 attenuated Aβ-induced reduction of GluN1 and GluA2 (Fig. 6A, B). Immunofluorescent staining data also showed Aβ-potentiated media remarkably inhibited development of neuronal processes, while simultaneous inhibition of PKA in microglia attenuated neuronal impairments induced by Aβ-potentiated media (Fig. 6C). These data demonstrate that Aβ induces neuronal damage through microglia PKA-mediated CREB phosphorylation.

Fig. 6.

Aβ-potentiated microglial media affect neuronal development and synaptic protein expression via PKA-dependent manner. The bv2 cells were treated with aged Aβ (2.5μM), or Aβ plus H89 (15μM) or vehicle (DMSO, 1:1000, v/v) for 1 h, then the culture media were removed. After thorough washing, the bv2 cells were cultured with new medium without Aβ and H89 for 5 h. Then the culture media were collected and added (1:9, v/v) into the primary hippocampal neurons. After 5 h, the neuronal extracts were used for measurement of synapse-associated proteins by western blotting (n = 6). C) The neuronal morphology changes measured by immunofluorescent staining of anti-MAP2 antibody (scale bar = 25μm). Data were presented as mean±SEM (one-way ANOVA, ^*p < 0.05, ^**p < 0).

DISCUSSION

Alzheimer’s disease (AD) is the most common neurodegenerative disorder. As one of the two major pathological hallmarks of AD, Aβ is toxic to the neurons. However, due to the morphological diversity and the diversity of targets, the mechanisms underlying the toxic effects of Aβ on neurons become very complicated. Tremendous studies show that Aβ can induce neuroinflammation, oxidative stress, and mitochondrial dysregulation [28]. There are also increasing reports demonstrated that Aβ can enhance neuronal activity and synaptic plasticity [29], improve hippocampal LTP with positive regulation of neurotransmission and memory [30], and maintain the balance of lipid metabolism by binding and transporting cholesterol [31]. Aβ is generated by β-secretase [32 –34], γ-secretase [35] and δ-secretase [36] cleavage of amyloid-β protein precursor (AβPP), while the mice with AβPP of β-site APP-cleaving enzyme 1 (BACE1) knockout exhibit memory impairment [37, 38]. We speculate that the time window (i.e., acute or chronic) we choose to do the measurement may at least contribute the observed protective or detrimental role of Aβ in the nervous system. For instance, we found in the present study that intracerebroventricular infusion of Aβ only caused neuronal damages and memory deficit at 24 h. By 7 days after the treatment, the Aβ-induced damages were no longer detected. These data suggest that the short-term exposure to Aβ is transient and reversible, and neuronal function may be recovered by breaking Aβ overproduction in early stages of the AD progression. Although we observed elevation of PSD93 protein level at 24 h after Aβ infusion, we currently do not understand why only PSD93, but not others, was selectively increased. We speculate that expression of PSD93 may be more actively regulated by pCREB, or PSD93 expression may be more predominant in microglia than the neurons, or it may be just a non-specific stress response because the elevation of PSD93 was not detected in 7 days group or in the brains.

Regarding the effects of the exogenously infused Aβ₄₂ on learning and memory, diverging observations have been reported. By using novel object recognition test as used in the current study, some reported that oligomer Aβ₄₂ could cause learning and memory impairments in mice after being injected for 24 h but the memory deficit was no longer detectable at 10 days if Aβ₄₂ was not injected again [39]; rats had no social memory impairment at 7 days after being injected with aged Aβ₄₂ [40], which were in agreement with the observations in the current paper. Other studies using Morris water maze and Y-maze tests showed that the mice injected with aged Aβ₄₂ had learning and memory damages and loss of neurons at 14 days after the injection [41, 42]; the cognitive impairments of the mice were even detected at 21 days after single injection of the aged Aβ₄₂ [43]. In our current study, we could also see reduction of some synapse-associated proteins at 14 days after Aβ₄₂ exposure, but we did not detect memory deficits at 7 days and 14 days. The reason causing these discrepancies is currently not fully understood. We speculate that the paradigms applied for behavioral tests may be one of them.

Microglial cell is the first line of defense of the nervous system by secreting many inflammatory mediators, including complement proteins, cytokines, and chemokines [44 –46]. In general, microglia can sense and monitor the alterations in surrounding environment, and then they change the morphology into amoebic-like and migrate to the damaged sites bring protective or meanwhile toxic effects [47]. Aβ can activate the microglia, making microglia to phagocytose and clear Aβ, and by which it brings protective effects. When microglia attempt to remove toxic substances from nervous systems, it also secretes a large number of inflammatory factors that can damage the neurons. For example, incubating mouse microglia with Aβ promotes mitogenesis and secretion IL-1 and TNF-α [48]. We also observed in the current study that Aβ treatment for 6 h significantly increased levels of IL-6 and MPP-9 in bv2 cells. In the brain of AD patients, microglia cells have been detected to be largely accumulated in areas where amyloid plaques are deposited. Aβ can bind to the receptor complex containing the pattern recognition receptor CD36 and a heterodimer composed of the Toll-like receptors (TLR) to promote nuclear transport of nuclear factor κB (NFκB) [49], and thus activate the cAMP/PKA/CREB signaling pathway [50], resulting in expression of inflammatory factors [51].

As a widely expressed transcription factor in neurons, CREB is at the core of various signaling pathways for synaptic plasticity and memory formation. CREB is mainly regulated by phosphorylation, and the Ser133 site is the most critical site of the transcriptional activity regulation. The Ser133 of CREB can be phosphorylated by many kinases, including PKA, protein kinase C (PKC), Ca^{2 +}/calmodulin-dependent kinase II and IV (CaMKII and CaMKIV), mitogen/stress-activated kinase (MSK), ribosomal S6 kinase (RSK), AKT, and MAPKAP kinase-2 (MK2) [52 –58]. We found with surprising here that Aβ treatment for 24 h induced Ser133-CREB hyperphosphorylation in total hippocampal extracts of mice but with an impaired memory. Therefore, we further investigated the pS133-CREB level in neurons and microglia, respectively. We found that Aβ treatment for 24 h significantly decreased the CREB phosphorylation in the neurons with a remarkably increased pS133-CREB in microglia. These results can explain the Aβ-induced neuronal impairments and memory deficits. By co-culturing the primary neurons with Aβ-potentiated bv2 media, we further confirmed that Aβ treatment in the acute phase could induce neuronal impairments by pS133-CREB-dependent microglial secretion of pro-inflammatory cytokines.

In summary, we found in the present study that intracerebroventricular infusion of aged Aβ induces transient memory deficit with neuronal impairments, and the molecular mechanisms involve pS133-CREB-dependent microglial activation.

Footnotes

ACKNOWLEDGMENTS

This work was supported in parts by grants from Natural Science Foundation of China (31730035, 91632305, 91632111 and 81721005), and grant from the Ministry of Science and Technology of China (2016YFC1305800).

Authors’ disclosures available online ().

References

Grundke-Iqbal

, Iqbal

, Tung

, Quinlan

, Wis-niewski

, Binder

(1986) Abnormal phosphorylation of the microtubule-associated protein tau (tau) in Alzheimer cytoskeletal pathology. Proc Natl Acad Sci USA 83, 4913–4917.

Lee

, Balin

, Otvos

Jr , Trojanowski

(1991) A68: A major subunit of paired helical filaments and derivatized forms of normal Tau. Science 251, 675–678.

Wood

, Mirra

, Pollock

, Binder

(1986) Neu-rofibrillary tangles of Alzheimer disease share antigenic determinants with the axonal microtubule-associated protein tau (tau). Proc Natl Acad Sci USA 83, 4040–4043.

Ghiso

, Frangione

(2002) Amyloidosis and Alzheimer’s disease. Adv Drug Deliv Rev 54, 1539–1551.

Selkoe

(2001) Clearing the brain’s amyloid cobwebs. Neuron 32, 177–180.

Chen

, Glabe

(2006) Distinct early folding and aggregation properties of Alzheimer amyloid-beta peptides Abeta40 and Abeta42: Stable trimer or tetramer formation by Abeta42. J Biol Chem 281, 24414–24422.

Butterfield

, Swomley

, Sultana

(2013) Amyloid beta-peptide (1-42)-induced oxidative stress in Alzheimer disease: Importance in disease pathogenesis and progression. Antioxid Redox Signal 19, 823–835.

Eckert

, Hauptmann

, Scherping

, Meinhardt

, Rhein

, Drose

, Brandt

, Fandrich

, Muller

, Gotz

(2008) Oligomeric and fibrillar species of beta-amyloid (A beta 42) both impair mitochondrial function in P301L tau transgenic mice. J Mol Med (Berl) 86, 1255–1267.

Sondag

, Dhawan

, Combs

(2009) Beta amyloid oligomers and fibrils stimulate differential activation of primary microglia. J Neuroinflammation 6, 1.

10.

Serra-Batiste

, Ninot-Pedrosa

, Bayoumi

, Gairi

, Maglia

, Carulla

(2016) Abeta42 assembles into specific beta-barrel pore-forming oligomers in membranemimicking environments. Proc Natl Acad Sci U S A 113, 10866–10871.

11.

Gehrmann

, Matsumoto

, Kreutzberg

(1995) Microglia: Intrinsic immuneffector cell of the brain. Brain Res Brain Res Rev 20, 269–287.

12.

Kingwell

(2012) Neurodegenerative disease: Microglia in early disease stages. Nat Rev Neurol 8, 475.

13.

Kandel

(2012) The molecular biology of memory: cAMP, PKA, CRE, CREB-1, CREB-2, and CPEB. Mol Brain 5, 14.

14.

Walton

, Dragunow

(2000) Is CREB a key to neuronal survival? Trends Neurosci 23, 48–53.

15.

Carlezon

Jr , Duman

, Nestler

(2005) The many faces of CREB. Trends Neurosci 28, 436–445.

16.

Schmidt

, Duman

(2010) Peripheral BDNF produces antidepressant-like effects in cellular and behavioral models. Neuropsychopharmacology 35, 2378–2391.

17.

Rouaux

, Jokic

, Mbebi

, Boutillier

, Loeffler

, Boutillier

(2003) Critical loss of CBP/p300 histone acetylase activity by caspase-6 during neurodegeneration. EMBO J 22, 6537–6549.

18.

Saura

, Valero

(2011) The role of CREB signaling in Alzheimer’s disease and other cognitive disorders. Rev Neurosci 22, 153–169.

19.

Platenik

, Kuramoto

, Yoneda

(2000) Molecular mechanisms associated with long-term consolidation of the NMDA signals. Life Sci 67, 335–364.

20.

Esteban

, Shi

, Wilson

, Nuriya

, Huganir

, Malinow

(2003) PKA phosphorylation of AMPA receptor subunits controls synaptic trafficking underlying plasticity. Nat Neurosci 6, 136–143.

21.

Lonart

, Schoch

, Kaeser

, Larkin

, Sudhof

, Linden

(2003) Phosphorylation of RIM1alpha by PKA triggers presynaptic long-term potentiation at cerebellar parallel fiber synapses. Cell 115, 49–60.

22.

Nguyen

, Woo

(2003) Regulation of hippocam-pal synaptic plasticity by cyclic AMP-dependent protein kinases. Prog Neurobiol 71, 401–437.

23.

Powell

(2006) Gene targeting of presynaptic proteins in synaptic plasticity and memory: Across the great divide. Neurobiol Learn Mem 85, 2–15.

24.

Zheng

, Keifer

(2009) PKA has a critical role in synaptic delivery of GluR1- and GluR4-containing AMPARs during initial stages of acquisition of in vitro classical conditioning. J Neurophysiol 101, 2539–2549.

25.

Zeng

, Jiang

, Hu

, Ma

, Chai

, Sun

, Xu

, Li

, Bao

, Feng

, Hu

, Chu

, Chai

, Hong

, Wang

, Liu

(2016) Spatial training promotes short-term survival and neuron-like differentiation of newborn cells in Abeta1-42-injectedrats. Neurobiol Aging 45,64–75.

26.

, Wang

, Liu

, Deng

, Zhang

, Tian

, Wang

, Chen

, Yang

, Zhang

, Wang

, Xu

, Liao

, Wang

(2007) Phosphorylation of tau antagonizes apoptosis by stabilizing beta-catenin, a mechanism involved in Alzheimer’s neurodegeneration. Proc Natl Acad SciUSA 104,3591–3596.

27.

Zhang

, Song

, Liu

, Kang

, Kwon

, Duong

, Seyfried

, Hu

, Liu

, Wang

, Cheng

, Sun

, Yu

, Levey

, Ye

(2014) Cleavage of tau by asparagine endopeptidase mediates the neurofibrillary pathology in Alzheimer’s disease. Nat Med 20, 1254–1262.

28.

Polanco

, Li

, Bodea

, Martinez-Marmol

, Meunier

, Gotz

(2018) Amyloid-beta and tau complexity - towards improved biomarkers and targeted therapies. Nat Rev Neurol 14, 22–39.

29.

, Uemura

, Hashimoto

, Nasser-Ghodsi

, Arimon

, Lill

, Palazzolo

, Krainc

, Hyman

, Berezovska

(2013) Neuronal activity and secreted amyloid beta lead to altered amyloid beta precursor protein and presenilin 1 interactions. Neurobiol Dis 50, 127–134.

30.

Puzzo

, Privitera

, Leznik

, Fa

, Staniszewski

, Palmeri

, Arancio

(2008) Picomolar amyloid-beta positively modulates synaptic plasticity and memory in hippocampus. J Neurosci 28, 14537–14545.

31.

Umeda

, Mori

, Zheng

, Tomiyama

(2010) Regulation of cholesterol efflux by amyloid beta secretion. J Neurosci Res 88, 1985–1994.

32.

Cai

, Wang

, McCarthy

, Wen

, Borchelt

, Price

, Wong

(2001) BACE1 is the major beta-secretase for generation of Abeta peptides by neurons. Nat Neurosci 4, 233–234.

33.

Vassar

, Bennett

, Babu-Khan

, Kahn

, Mendiaz

, Denis

, Teplow

, Ross

, Amarante

, Loeloff

, Luo

, Fisher

, Fuller

, Edenson

, Lile

, Jarosinski

, Biere

, Curran

, Burgess

, Louis

, Collins

, Tre-anor

, Rogers

, Citron

(1999) Beta-secretase cleavage of Alzheimer’s amyloid precursor protein by the transmembrane aspartic protease BACE. Science 286, 735–741.

34.

Bao

, Qin

, Mahaman

YAR

, Zhang

, Huang

, Zeng

, Xia

, Ke

, Wang

, Liu

, Wang

, Ye

, Wang

(2018) BACE1 SUMOylation increases its stability and escalates the protease activity in Alzheimer’s disease. Proc Natl Acad Sci USA 115, 3954–3959.

35.

Takami

, Nagashima

, Sano

, Ishihara

, Morishima-Kawashima

, Funamoto

, Ihara

(2009) gamma-Secretase: Successive tripeptide and tetrapeptide release from the transmembrane domain of beta-carboxyl terminal fragment. J Neurosci 29, 13042–13052.

36.

Zhang

, Obianyo

, Dall

, Du

, Fu

, Liu

, Kang

, Song

, Yu

, Cabrele

, Schubert

, Li

, Wang

, Brandstetter

, Ye

(2017) Inhibition of delta-secretase improves cognitive functions in mouse models of Alzheimer’s disease. Nat Commun 8, 14740.

37.

Kobayashi

, Zeller

, Cole

, Buttini

, McConlogue

, Sinha

, Freedman

, Morris

, Chen

(2008) BACE1 gene deletion: Impact on behavioral function in a model of Alzheimer’s disease. Neurobiol Aging 29, 861–873.

38.

Muller

, Deller

, Korte

(2017) Not just amyloid: Physiological functions of the amyloid precursor protein family. Nat Rev Neurosci 18, 281–298.

39.

Balducci

, Beeg

, Stravalaci

, Bastone

, Sclip

, Biasini

, Tapella

, Colombo

, Manzoni

, Borsello

, Chiesa

, Gobbi

, Salmona

, Forloni

(2010) Synthetic amyloid-beta oligomers impair long-term memory independently of cellular prion protein. Proc Natl Acad Sci USA 107, 2295–2300.

40.

Christensen

, Marcussen

, Wortwein

, Knudsen

, Aznar

(2008) Abeta(1-42) injection causes memory impairment, lowered cortical and serum BDNF levels, and decreased hippocampal 5-HT(2A) levels. Exp Neurol 210, 164–171.

41.

Jin

, Chen

, Li

, Wang

, Zhang

, Cao

, Sha

, Wan

, Chen

(2016) Dose-dependent neuroprotection and neurotoxicity of simvastatin through reduction of farnesyl pyrophosphate in mice treated with intracerebroventricular injection of Abeta 1-42. J Alzheimers Dis 50, 501–516.

42.

Song

, Chen

, Wang

, Gao

, Zhu

(2013) Rho kinase inhibitor fasudil protects against beta-amyloid-induced hippocampal neurodegeneration in rats. CNS Neurosci Ther 19, 603–610.

43.

Ali

, Kim

(2015) Melatonin ameliorates amyloid beta-induced memory deficits, tau hyperphosphorylation and neurodegeneration via PI3/Akt/GSk3beta pathway in the mouse hippocampus. J Pineal Res 59, 47–59.

44.

Das

, Potter

(1995) Expression of the Alzheimer amyloid-promoting factor antichymotrypsin is induced in human astrocytes by IL-1. Neuron 14, 447–456.

45.

Lee

, Dickson

, Brosnan

(1995) Interleukin-1, nitric oxide and reactive astrocytes. Brain Behav Immun 9, 345–354.

46.

McGeer

, McGeer

(1995) The inflammatory response system of brain: Implications for therapy of Alzheimer and other neurodegenerative diseases. Brain Res Brain Res Rev 21, 195–218.

47.

Stence

, Waite

, Dailey

(2001) Dynamics of microglial activation: A confocal time-lapse analysis in hip-pocampal slices. Glia 33, 256–266.

48.

Araujo

, Cotman

(1992) Beta-amyloid stimulates glial cells in vitro to produce growth factors that accumulate in senile plaques in Alzheimer’s disease. Brain Res 569, 141–145.

49.

Cunningham

(2013) Microglia and neurodegeneration: The role of systemic inflammation. Glia 61, 71–90.

50.

Cho

, Kim

, Cruz

, Park

, Chu

, Song

, Joh

(2001) Repression of proinflammatory cytokine and inducible nitric oxide synthase (NOS2) gene expression in activated microglia by N-acetyl-O-methyldopamine: Protein kinase A-dependent mechanism. Glia 33, 324–333.

51.

Yates

, Burgess

, Kocsis-Angle

, Antal

, Dority

, Embury

, Piotrkowski

, Brunden

(2000) Amyloid beta and amylin fibrils induce increases in proin-flammatory cytokine and chemokine production by THP-1 cells and murine microglia. J Neurochem 74, 1017–1025.

52.

Deak

, Clifton

, Lucocq

, Alessi

(1998) Mitogen- and stress-activated protein kinase-1 (MSK1) is directly activated by MAPK and SAPK2/p38, and may mediate activation of CREB. EMBO J 17, 4426–4441.

53.

, Montminy

(1998) CREB is a regulatory target for the protein kinase Akt/PKB. J Biol Chem 273,32377–32379.

54.

Ginty

, Bonni

, Greenberg

(1994) Nerve growth factor activates a Ras-dependent protein kinase that stimulates c-fos transcription via phosphorylation of CREB. Cell 77,713–725.

55.

Gonzalez

, Montminy

(1989) Cyclic AMP stimulates somatostatin gene transcription by phosphorylation of CREB at serine 133. Cell 59, 675–680.

56.

Sun

, Enslen

, Myung

, Maurer

(1994) Differential activation of CREB by Ca2+/calmodulin-dependent protein kinases type II and type IV involves phosphorylation of a site that negatively regulates activity. Genes Dev 8, 2527–2539.

57.

Tan

, Rouse

, Zhang

, Cariati

, Cohen

, Comb

(1996) FGF and stress regulate CREB and ATF-1 viaapath-way involving p38 MAP kinase and MAPKAP kinase-2. EMBO J15, 4629–4642.

58.

Xing

, Ginty

, Greenberg

(1996) Coupling of the RAS-MAPK pathway to gene activation by RSK2, a growth factor-regulated CREB kinase. Science 273, 959–963.

Microglia CREB-Phosphorylation Mediates Amyloid-β-Induced Neuronal Toxicity

Abstract

Keywords

INTRODUCTION

MATERIALS AND METHODS

Animals and treatments

Novel object recognition

Western blotting

Golgi staining

Immunohistochemistry and immunofluorescence

BV2 culture

Primary neuron culture

Chromatin immunoprecipitation assay

Statistical analyses

RESULTS

Intracerebroventricular infusion of Aβ42 only induces transient memory deficit with a remarkably increased CREB phosphorylation in hippocampus

Intracerebroventricular infusion of Aβ only transiently activates microglia

Intracerebroventricular infusion of Aβ increases microglial pS133-CREB level

Inhibition of microglial activity attenuates Aβ-induced microglia elevation of pCREB with improved memory

Aβ activates microglia through CREB phosphorylation

Aβ affects neuronal synaptic protein expression and development through microglial PKA-dependent manner

DISCUSSION

Footnotes

ACKNOWLEDGMENTS

References

Intracerebroventricular infusion of Aβ₄₂ only induces transient memory deficit with a remarkably increased CREB phosphorylation in hippocampus