p38 MAPK Is a Major Regulator of Amyloid Beta-Induced IL-6 Expression in Human Microglia

Abstract

The accumulation of amyloid beta (Aβ) plaques in the brain is a hallmark of Alzheimer’s disease (AD) pathology. Microglial activation-mediated neuroinflammation has been implicated in the pathogenesis of AD and the expression levels of interleukin-6 (IL-6) were increased in the brains of AD patients. However, the mechanisms by which IL-6 expression is regulated in human microglia are incompletely understood. Here, we show that Aβ_1–40 oligomers (Aβ₄₀) dose-dependently stimulate IL-6 expression in HMC3 human microglial cells. Treatment with Aβ₄₀ promotes the transcription of IL-6 and tumor necrosis factor α (TNFα) mRNAs in both HMC3 and THP-1 cells. Mechanistic studies reveal that Aβ₄₀-induced increase of IL-6 secretion is associated with the activation of p38 mitogen-activated protein kinase (p38 MAPK). Inhibition of p38 MAPK by BIRB 796 or SB202190 abrogates Aβ₄₀-induced increase of IL-6 production. Through analyzing brain specimens, we found that the immunoreactivity for IL-6 and phosphorylated (the activated form) p38 MAPK were markedly higher in microglia of AD patients than in age-matched control subjects. Moreover, our studies identified the co-localization of IL-6 with phosphorylated p38 MAPK in microglia in the cortices of AD patients. Taken together, these results indicate that p38 MAPK is a major regulator of Aβ-induced IL-6 production in human microglia, which suggests that targeting p38 MAPK may represent a new approach to ameliorate Aβ accumulation-induced neuroinflammation in AD.

Introduction

Alzheimer’s disease (AD), the most common cause of dementia, is characterized by the early accumulation of extracellular amyloid-beta peptide (Aβ) in neuritic plaques and followed by hyperphosphorylated tau-containing intracellular neurofibrillary tangles, along with the progressive loss of memory and cognitive functions [1, 2]. In response to Aβ accumulation, microglia are activated to facilitate the clearance of Aβ aggregates, but unfortunately this process can trigger a neuroinflammatory response that could be detrimental to neurons [3, 4]. The neuroinflammatory responses include the elevated production of pro-inflammatory cytokines, such as interleukin (IL)-1, IL-6 and tumor necrosis factor alpha (TNFα), which are associated with various neurodegenerative disorders, including AD, Parkinson’s disease, multiple sclerosis, amyotrophic lateral sclerosis and HIV-associated dementia [4–9]. Increased IL-6 expression levels have been observed in brain specimens from AD patients [10] and are linked to diffuse plaques representing the early stage of plaque formation [11]. Moreover, it has been shown that human cognitive functions adversely correlate with serum IL-6 levels [12]. An epidemiologic meta-analysis indicated that IL-6 levels may be a useful biological marker to correlate with the severity of cognitive impairments in AD patients [13]. Preclinical models provided further support and showed that cerebral overexpression of IL-6 resulted in transgene dose- and animal age-dependent deficits in avoidance learning as well as progressive neurodegenerative pathologies [14]. These results suggest that the elevated production of IL-6 may play a critical role in neurodegenerative disorders. Nevertheless, the molecular mechanisms by which IL-6 expression is regulated in human microglial cells remain poorly understood.

Aβ peptides consist of 38 to 43 amino acids and are generated from a sequential proteolytic cleavage of the amyloid precursor protein (APP) by β- and γ-secretase [15]. Aβ monomers can bind to one another to form Aβ oligomers, polymers, insoluble fibrils, and eventually senile plaques. Aβ oligomers are soluble and exhibit the strongest toxicity to neurons [16], which can cause neuronal cell damage, impairment of synaptic plasticity and memory loss [16–18]. The Aβ_1–40 (Aβ₄₀) and Aβ_1–42 (Aβ₄₂) peptides are two of the most abundant amyloid species in senile plaques [19]. Aβ₄₀ oligomers can be delivered to the parenchyma via the glymphatic transport system and co-localize with senile plaques, suggesting an important role of Aβ₄₀ in plaque formation [20]. Although the impact of Aβ on microglia-mediated neuroinflammation has been reported in previous studies using animal models or rodent cell lines [21–25], the neuroinflammatory response of human microglia to Aβ stimulation has not been investigated previously. Moreover, a recent study reported that treatment with a selective p38 inhibitor alleviated neuropathology and cognitive impairments in a mouse model of AD [26]. However, the molecular mechanisms by which p38 inhibition is beneficial for AD are not completely understood. In the present study, we show that Aβ₄₀ selectively induces the activation of p38 MAPK (p38 hereafter), but not the Erk and JNK MAPKs, in HMC3 human microglial cells. Inhibition of p38 by BIRB 796 (BIRB) or SB202190 (SB) is sufficient to block Aβ₄₀-induced increase in IL-6 production. Moreover, our studies have confirmed the colocalization of IL-6 with phosphorylated p38 (p-p38) in microglia in the cortices of AD patients. Collectively, these results pinpoint a major role of the p38 pathway in modulating Aβ-induced IL-6 production in human microglia, suggesting that targeted inhibition of p38 may represent a new therapeutic approach to curb Aβ accumulation-induced neuroinflammation in AD.

Materials and Methods

Materials

Aβ₄₀ and scrambled Aβ₄₀ (Scr-Aβ₄₀) peptides were purchased from rPeptide (Athens, GA). Brefeldin A was purchased from Biolegend (San Diego, CA). BIRB 796 and SB202190 were purchased from Selleckchem (Houston, TX). LPS was purchased for Sigma (E coli O55:B5, cat# L2880, St. Louis, MO). Rabbit anti-human p-p38, p38, p-p44 (pErk), p44 (Erk), IL-6, IL-10, TGF-β and JNK1 monoclonal antibodies, and mouse anti-human p-JNK1 monoclonal antibodies were purchased from Cell Signaling (Danvers, MA). Mouse anti-human CD68 monoclonal antibody was obtained from ThermoFisher (Cat# 14-0688-82, clone: KP1). Mouse anti-human TMEM119 monoclonal antibody (#400111) was obtained from Synaptic Systems (Gottingen, Germany). Mouse anti-human α-tubulin monoclonal antibody was obtained from Santa Cruz Biotechnology (Santa cruz, CA). Human IL-6 DuoSet ELISA (DY206) and human TNF-α DuoSet ELISA (DY210) kits were purchased from R&D Systems (Minneapolis, MN). EMEM medium was purchased from American Type Culture Collection (ATCC, cat# 30–2003). RPMI-1640 medium was obtained from Gibco (Gibco, cat# 11875093).

Preparation of Aβ oligomers

The soluble Aβ₄₀ oligomers were prepared as previously described [27–29] with minor modifications. Briefly, lyophilized Aβ₄₀ peptide powder was dissolved in 1% (w/v) NH₄OH to prepare a stock solution of 231 mM. The stock was diluted to 100 μM solution using serum-free RPMI-1640 medium and was aliquoted and stored in a −80°C freezer. Aβ₄₀ solution was incubated at 4°C for 24 h prior to use.

Human Brain Specimens

A total of 11 formalin-fixed and paraffin-embedded human brain tissue blocks were obtained from the Carroll A. Campbell, Jr. Brain Bank at the Medical University of South Carolina (MUSC). The specimens included 6 brain tissue blocks isolated from AD patients and another 5 brain samples isolated from age-matched non-demented control subjects (Table 1). Tissue sections (6-μm-thick) were prepared by the Biorepository Core facility of Hollings Cancer Center at MUSC.

Table 1.

Demographic data and clinical characteristics of the brain specimens used for this study

Subject ID	Category	Age	Gender	ApoE Genotype	Clinical Diagnosis	Braak stage
HB82	Control	85	F	E3/E3	Stroke	0
HB76	Control	63	F	E3/E3	HTN	0
HB7	Control	80	F	E3/E3	Cancer	0
HB80	Control	81	M	E3/E3	Cancer	0
HB77	Control	81	F	E3/E3	Stroke	0
HB135	AD	87	F	Unknown	AD	IV/VI
HB140	AD	90	M	Unknown	AD	V/VI
HB138	AD	77	M	Unknown	AD	V/VI
HB186	AD	76	M	Unknown	AD	V/VI
HB122	AD	78	F	E3/E3	AD	V/VI
HB112	AD	59	F	E3/E4	AD	II/III

Confocal microscopy

Immunofluorescent staining and confocal microscopic assays were performed to determine IL-6 and p-p38 levels in microglia in the brain tissues of AD patients. Briefly, tissue sections were deparaffinized using xylene and rehydrated in gradient ethanol. Antigen retrieval was achieved by boiling with 10 mM Sodium Citrate buffer (pH 6.0). Then tissue sections were permeabilized with 0.3% Triton X-100 for 15 min and blocked with 5% goat serum in 0.1% Triton X-100/PBS at room temperature (RT) for 1 h. Primary antibodies chicken anti-IBA1 (1:1000, SYSY, #234009), rabbit anti-human IL-6 (1:200, Cell Signaling, #12153) and mouse anti-phosphorylated p38 (1:200, Cell Signaling, #58970) were incubated overnight at 4°C. Secondary antibodies Alexa Fluor-594 labeled goat anti-chicken IgG (1:500, Invitrogen, #A10680), Alexa Fluor-488 labeled goat anti-rabbit IgG (1:500, Invitrogen, #A10680) and Alexa Fluor-647 labeled goat anti-mouse IgG (1:500, Invitrogen, #A21244) were incubated at RT for 2 h. Nuclei were counterstained with DAPI and slides were mounted with anti-fade reagent (Invitrogen, P36930). Images were captured and analyzed using an LSM 880 Airyscan super-resolution confocal microscope (Carl Zeiss, Oberkochen, Germany). Mean fluorescence intensity (MFI) was quantified using ImageJ software (NIH). Area of IL-6 or p-p38 positive staining was measured using the Analyze Particles plugin of ImageJ.

Cell Culture

The HMC3 human microglia, Raw 264.7 mouse macrophage and THP-1 human monocyte cell lines were obtained from ATCC. HMC3 cells were cultured using EMEM medium containing 10% FBS, 2 mM L-glutamine and 100 μg/mL of penicillin-streptomycin (Invitrogen). The BV-2 microglial cell line was generously provided by Dr. Sanjay Maggirwar (University of Rochester Medical Center, Rochester, NY, USA). THP-1, BV-2 and Raw 264.7 cells were cultured with RPMI-1640 medium containing 10% FBS, 2 mM L-glutamine and 100 μg/mL of penicillin-streptomycin (Invitrogen). All the cells were cultured in a humidified incubator with 5% CO2 and 95% air at 37°C.

Cell viability assay

The MTT assays were performed to determine cell viability. Briefly, HMC3 cells were seeded into 96-well plates at a density of 3,000 cells per well. One day after cell plating, cells were treated with various concentrations of Aβ₄₀ or Scr- Aβ₄₀ for 48 h. Then MTT reagent was added to the medium (0.5 mg/mL, final concentration) and incubated at 37°C, 5% CO₂ for 4 h. The medium was removed and 100 μl DMSO was added to each well to dissolve formazan precipitate thoroughly for 15 min. The absorbance at 570 nm and 650 nm was read using a Synergy H1 microplate reader (BioTek).

CD68 Immunofluorescence

CD68 expression levels were assessed using immunofluorescence and confocal microscopy. Briefly, HMC3 cells were cultured in 4-well CELLview^™ dishes (Greiner Bio-One North America) and treated with Aβ₄₀ or Scr-Aβ₄₀ for 48 h. Cells were washed with PBS and fixed with 4% paraformaldehyde for 15 min. Next, cells were permeabilized with 0.2% Triton X-100 (Sigma-Aldrich) for 5 min and then incubated with blocking buffer (5% goat serum and 0.1% Triton X-100 in PBS) at RT for 1 h. Cells were incubated with mouse anti-human CD68 (1:100, ThermoFisher Cat# 14-0688-82, clone: KP1) and rabbit anti-Iba1 (1:500, FUJIFILM, #019–19741) antibodies overnight at 4°C. Then, cells were incubated with Alexa Fluor-488-labeled goat anti-rabbit (1:500, Invitrogen, #A11008) and Alexa Fluor-647-labeled goat anti-mouse (1:500, Invitrogen, #A32728) secondary antibodies at RT for 2 h. Nuclei were counterstained with 1 μg/mL DAPI, and cells were mounted with ProLong Gold anti-fade reagent (Invitrogen, Cat# 36930). Images were captured and analyzed using an LSM 880 Airyscan super-resolution confocal microscope (Carl Zeiss, Oberkochen, Germany). Mean fluorescence intensity (MFI) was quantified using ImageJ (NIH).

Real-time reverse transcriptase-PCR (RT-PCR)

Total RNA was extracted using TRIzol reagent (Invitrogen). Quality and quantity of RNA samples were assessed using a Synergy H1 microplate reader (BioTek). First-strand cDNA was synthesized from 1 μg of total RNA using an iScript cDNA Synthesis Kit (Bio-Rad) according to the manufacturer’s instructions. The PCR primers used for this study are listed in Table 2. Expression levels of human IL-6, IL-1β and TNF-α mRNA were measured using iTaq Universal SYBR® Green Supermix (Bio-Rad) on a LightCycler® 480 System (Roche). GAPDH expression levels were used as an internal reference to calculate the changes in target mRNA expression levels using the 2^−ΔΔCT method as described previously [30].

Table 2.

PCR Primers used in this study

Genes	Forward primers	Reverse primers
IL-6	5′-GAAGGCAGCAGGCAACAC-3′	5′-TGAACTCCTTCTCCACAAGCG-3′
IL-1β	5′-AACCTCTTCGAGGCACAAGG-3′	5′-AGCCATCATTTCACTGGCGA-3’
TNF-α	5′-CTTCTGCCTGCTGCACTTTG-3’	5′-GTCACTCGGGGTTCGAGAAG-3′
GAPDH	5′-GACAGTCAGCCGCATCTTCT-3′	5′-GCGCCCAATACGACCAAATC-3′

Western blotting

Western blot analyses were performed as previously described [31]. Briefly, protein samples were prepared using cell lysis buffer (Cell Signaling) supplemented with a cocktail of proteinase inhibitors (Sigma). Protein concentrations were quantified using a Bio-Rad Dc protein assay kit (Bio-Rad Laboratories, Hercules, CA). Proteins (30 μg) were resolved on 4 – 20% Mini-Protean TGX gels (Bio-Rad) and transferred onto 0.2 μm PVDF membrane (Millipore). Blots were blocked with 5% non-fat milk at RT for 1 h before incubating with primary antibodies at 4°C overnight. After extensive washing with TBS-T, blots were incubated with appropriate HRP-conjugated secondary antibodies at RT for 1.5 h. Protein bands were detected using an ECL Plus Western Blotting Detection System (GE Healthcare Life Science) and an Odyssey Fc Imaging System (LI-COR Biosciences).

Cytokine measurements

Enzyme-linked immunosorbent assay (ELISA) was employed to measure pro-inflammatory cytokines in supernatants of HMC3 and THP-1 cultures. HMC3 cells were seeded in 12-well plates at 60,000 cells per well, while THP-1 cells were seeded in 12-well plates at 300,000 cells per well. One day after cell plating, cells were treated with Aβ₄₀ in serum-free medium with or without p38 inhibitors. Cell culture supernatants were collected, and cells debris were removed by centrifuging at 4000 rpm for 5 min. IL-6 and TNFα concentrations were measured using a human IL-6 DuoSet ELISA kit (DY206) and a human TNF-α DuoSet ELISA kit (DY210), respectively, according to the manufacturer’s instructions (R&D Systems).

Mitochondrial ROS Analysis

Mitochondrial ROS (mROS) levels were determined using a published protocol [32]. Briefly, HMC3 cells were cultured in 4-well CELLview^™ cell culture dishes (Greiner Bio-One, North America). One day after cell plating, cells were treated with Aβ₄₀ (2 μM) for 6 h. Then, cells were incubated with 2.5 μM MitoSox Red probe in serum-free EMEM medium at 37°C for 30 minutes. Nuclei were counterstained with DAPI. Images were captured using an LSM 880 laser scanning confocal microscope (Carl Zeiss, Oberkochen, Germany). The levels of mROS florescence intensity were analyzed using ImageJ software (NIH).

Flow cytometric analysis

MitoSox Red staining and flow cytometric assays were performed to detect mitochondrial superoxide (mROS) levels using a published protocol [33]. Briefly, THP-1 cells were treated with various doses of Aβ₄₀ for a duration as indicated in the figure legends. Then, cells were incubated with serum-free RPMI-1640 medium containing 2.5 μM MitoSox Red at 37°C for 30 min. The levels of mROS were analyzed by measuring the mean fluorescence intensity (MFI) of MitoSox using a CytoFLEX Flow Cytometer (Beckman Coulter, Brea, CA).

Nitric oxide assay

Levels of nitric oxide (NO) were evaluated using a Griess nitrite measurement kit (Thermofisher, MA). Briefly, HMC3 and Raw 264.7 cells were seeded in 12-well plates at 60,000 cells and 200,000 cells per well, respectively. Cells were then stimulated with LPS (0.1 μg/mL) or Aβ₄₀ (5 μM) for 24 h. Equal volumes of supernatants and Griess reagent were mixed and incubated in 96-well plates at RT for 30 min. Then the absorbance was measured at 548 nm using a Synergy H1 microplate reader (BioTek). NO concentrations in cell culture supernatants were calculated using the sodium nitrite standard curve.

Phagocytosis assay

HMC3 and BV-2 cells were plated in 4-well CELLview^™ cell culture dishes (Greiner Bio-One North America) at 1 × 10⁴ and 3 × 10⁴ cells/well, respectively. One day after cell plating, cells were pretreated with 2 μM BIRB for 1 h prior to incubation with 1 μM fluorescence-labeled FAM-488-Aβ₄₂ at 37°C for 2 h. Cells were then washed with PBS to remove the non-phagocytized Aβ peptides and fixed with 4% paraformaldehyde. Cells were then incubated with blocking buffer (5% goat serum, 0.1% Triton X-100 in PBS) for 1 h and followed with rabbit anti-Iba1 (1:500, FUJIFILM, #019–19741) at 4°C overnight. Alexa Fluor-647-labeled goat anti-rabbit (1:500, Invitrogen, #A21244) secondary antibody was incubated for 2 h to visualize Iba1 staining. Slides were mounted with anti-fade reagent (Invitrogen, P36930). Images were captured and processed using a Zeiss LSM 880 laser scanning confocal microscope. Imaging data were analyzed using ImageJ software (NIH).

Statistical analysis

Data is presented as mean ± SEM. Comparisons between two groups were carried out using Student’s t-test. Multiple comparisons were performed using one-way analysis of variance (ANOVA). Differences were considered statistically significant at p < 0.05. All analyses were carried out using GraphPad Prism software (GraphPad Software, Inc. San Diego, CA).

Results

Aβ₄₀ stimulates IL-6 production in HMC3 human microglial cells

We investigated the effects of Aβ₄₀ on IL-6 expression in HMC3 cells. The data showed that Aβ₄₀ treatment resulted in a dose-dependent increase of IL-6 production in HMC3 cells (Fig. 1A, Supplemental Fig. S1A). In contrast, treatment with an equal amount of Scr-Aβ₄₀ did not significantly affect IL-6 production in HMC3 cells (Fig. 1B). We also determined the effects of Aβ₄₀ treatment on the expression of other inflammatory mediators, including several M1 and M2 markers, in HMC3 cells. Our data indicate that treatment with Aβ₄₀ increases iNOS expression but has no significant effects on TGF-β expression. In contrast, LPS treatment led to a marked increase in expression levels of iNOS and TGF-β. Moreover, our data showed that IL-10 levels were not detectable in HMC3 cells (Supplemental Fig. S1B). Additionally, we also examined the impact of Aβ on human microglial cell proliferation. The results indicated that treatment with Aβ₄₀ slightly inhibited the proliferation of HMC3 cells, whereas Scr-Aβ₄₀ did not show any such effects (Supplemental Fig. S1C).

Fig. 1. Aβ₄₀ dose-dependently stimulates IL-6 production in human microglia.

(A, B) HMC3 cells were treated with various doses of Aβ₄₀ or Scr-Aβ₄₀, brefeldin A (0.5 μg/mL) was added to the medium during the last 4 h of culture. IL-6 levels were assessed by immunoblotting at 8 h after treatment. Tubulin was probed as loading control. (C) IL-6 Levels in supernatants of HMC3 cells were determined using ELISA at 8 h after Aβ₄₀ treatment. Data indicates that Aβ₄₀ dose-dependently stimulates IL-6 secretion. (D) Time course studies depict that Aβ₄₀ (2 μM) promotes IL-6 secretion in a time-dependent manner. (E) HMC3 cells were treated with 2 μM Aβ₄₀ or Scr-Aβ₄₀. Immunofluorescence and confocal microscopic assays were performed to assess CD68 expression levels. Representative images of confocal microscopic analysis of CD68 expression are shown. (F) Mean fluorescence intensity (MFI) of CD68 immunostaining was quantified using ImageJ software and the results show that Aβ₄₀ increases CD68 expression in HMC3 cells. Data are presented as mean +/− SEM of three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001.

To further determine if Aβ₄₀ affects IL-6 secretion from human microglia, we measured IL-6 levels in cell culture supernatants using ELISA assays. The results showed that Aβ₄₀ stimulated the secretion of IL-6 by HMC3 cells in a dose- and time-dependent fashion (Fig. 1C & D). Given that the increased production of inflammatory cytokines is a well-documented feature of microglial activation [3, 4], these new findings indicate that Aβ₄₀ stimulation may activate HMC3 human microglia. CD68 is thought to be a marker of activated microglia [34, 35], which prompted us to determine the influence of Aβ₄₀ on CD68 expression. As shown in Fig. 1E & F, immunofluorescence and confocal microscopic assays unveil that treatment with Aβ₄₀, but not Scr-Aβ₄₀, significantly increases CD68 expression levels in HMC3 cells.

Treatment with Aβ₄₀ promotes the transcription of IL-6 and TNF mRNAs in HMC3 cells

To further determine whether Aβ₄₀ affects the expression of other pro-inflammatory cytokines, we analyzed the expression levels of IL-6, TNFα and IL-1β mRNAs in HMC3 cells at various time points after Aβ₄₀ treatment. Our data show that Aβ₄₀ treatment increases the transcription of both IL-6 and TNFα mRNAs in a time-dependent fashion (Fig. 2A & B). The peak for Aβ-induced up-regulation of IL-6 and TNFα mRNA expression was at 3 h after treatment. Although there was an increase in TNFα transcription following Aβ₄₀ treatment, we had difficulty in measuring TNFα cytokine in supernatant using ELISA, which is consistent with a recent report showing that HMC3 cells are not efficient at secreting TNFα [36]. Furthermore, we did not observe any significant changes in IL-1β mRNA expression levels in response to Aβ₄₀ treatment (Fig. 2C). This result is consistent with a previous report indicating that IL-1β expression was not detectable in HMC3 cells [37].

Fig. 2. Aβ₄₀ promotes the transcription of IL-6 and TNF-α in HMC3 cells.

HMC3 cells were treated with Aβ₄₀ (2 μM) and RNA samples were prepared at different time points after treatment. mRNA expression levels for IL-6 (A), TNF-α (B) and IL-1β (C) were determined using real-time RT-PCR assays and were normalized to fold change (FC) relative to that in control cells. Data are presented as mean +/− SEM of three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001.

Aβ₄₀ treatment stimulates the transcription of IL-6, TNFα and IL-1β in THP-1 cells

Since microglia are the resident monocytes/macrophage of the central nervous system [38], we decided to examine the effects of Aβ₄₀ on the transcription of IL-6, TNFα and IL-1β mRNAs in THP-1 human monocytes. Quantitative RT-PCR analyses revealed that treatment with either LPS or Aβ₄₀ leads to a significant increase in IL-1β, IL-6, and TNFα mRNA levels in THP-1 cells (Fig. 3A–C). Furthermore, we assessed the protein levels of IL-6 and TNFα in supernatants of THP-1 cell cultures. The results demonstrate that Aβ₄₀ dose-dependently stimulates the secretion of IL-6 and TNFα by THP-1 cells (Fig. 3D & E).

Fig. 3. Aβ₄₀ increases expression levels of IL-6, TNF-α and IL-1 mRNAs in THP-1 cells.

THP-1 cells were treated with Aβ₄₀ (2 μM) or LPS (0.1 μg/mL) and mRNA expression levels for IL-6 (A), TNF-α (B) and IL-1β (C) were determined using real-time RT-PCR. (D, E) THP-1 cells were treated with indicated doses of Aβ₄₀. Protein levels of IL-6 (D) and TNF-α (E) in supernatants of THP-1 cell cultures were analyzed using ELISA. Data are presented as mean +/− SEM of three independent experiments. *** p < 0.001.

Aβ₄₀ selectively induces p38 activation in HMC3 and THP-1 cells

The p38 signaling pathway has been implicated in modulating stress and inflammatory responses in various model systems [39–41]. However, the role of p38 in regulating the production of IL-6 in human microglia has yet to be determined. Our data indicated that Aβ₄₀ treatment selectively activated p38, but not the ErK and JNK, MAPK signaling pathways, in HMC3 cells (Fig. 4A). The inability of Aβ₄₀ to induce JNK activation was further confirmed by the observation that UV exposure strongly induced JNK phosphorylation in 293T cells, whereas Aβ₄₀ treatment had no significant effects on JNK phosphorylation in HMC3 cells (Fig. 4B). Interestingly, we found that exposure to UV significantly increased the levels of phosphorylated JNK (Supplemental Fig. S2), suggesting that UV exposure may activate the JNK pathway in HMC3 cells.

Fig. 4. Aβ₄₀ selectively induces p38 activation in HMC3 and THP-1 cells.

(A) HMC3 cells were treated with indicated doses of Aβ₄₀ for 4 h. Expression levels of p-p38, p38, p-Erk, Erk, p-JNK, and JNK were assessed by immunoblotting. (B) UV irradiated 293T cells were used as a positive control for p-JNK immunoblotting. The data indicate that Aβ₄₀ selectively induces p38, but not Erk and JNK, activation in HMC3 cells. (C) Immunoblotting assays indicate that Aβ₄₀ (5 μM) time-dependently stimulates p38 activation in HMC3 cells. (D) Immunoblotting results show that Aβ₄₀ induces p38 activation in THP-1 cells in a dose- and time-dependent fashion. Data represent two independent experiments with similar results.

In addition, our data showed that Aβ₄₀ induced p38 activation in a dose- and time-dependent manner and the peak for Aβ₄₀-induced p38 activation was about 2 h after treatment (Fig. 4C). Notably, levels of phosphorylated (the activated form) p38 remain significantly higher even 24 h after Aβ₄₀ treatment (Fig. 4C), suggesting that Aβ₄₀ treatment may result in a sustained p38 activation in human microglia. In contrast, phosphorylated p65 (p-p65) levels were only transiently and slightly increased at 2 h after Aβ treatment (Fig. 4C). These findings suggest that p38 signaling, but not the NF-κB signaling pathway may play a major role in modulating Aβ-induced increases in the production of pro-inflammatory cytokines such as IL-6 and TNFα.

Next, we investigated whether Aβ₄₀ influences the activation of p38, Erk and JNK kinases in macrophages. Our data showed that Aβ40 dose-dependently stimulated the activation of the p38 signaling pathway, but not the Erk and JNK pathways in THP-1 cells (Fig. 4D). Furthermore, a time course study indicated that Aβ₄₀ stimulated p38 activation within 30 min after treatment, the peak of Aβ₄₀-induced p38 activation was approximately at 1 h after treatment and this activation was persistent for more than 24 h (Fig. 4D). Taken together, our data demonstrate for the first time that Aβ₄₀ dose-dependently stimulates p38 activation in both HMC3 and THP-1 cells.

Inhibition of p38 abrogates the Aβ₄₀-induced increase in IL-6 secretion

Based on the observation that Aβ₄₀ selectively stimulates p38 activation (Fig. 4), we hypothesized that inhibition of p38 activation could attenuate Aβ-induced neuroinflammatory response. To test this hypothesis, we investigated if pharmacologic inhibition of p38 by BIRB 796 (BIRB), a potent p38 inhibitor [42], could ameliorate Aβ₄₀-induced increase of IL-6 production. The results indicate that BIRB dose-dependently inhibits Aβ₄₀-induced p38 activation in HMC3 cells (Fig. 5A). Our data also show that pre-incubation with BIRB prevents Aβ₄₀-induced p38 activation in THP-1 cells (Fig. 5B). Furthermore, immunoblotting assays demonstrate that inhibition of p38 by BIRB abrogates Aβ-induced up-regulation of IL-6 in HMC3 cells (Fig. 5C).

Fig. 5. Inhibition of p38 abolishes Aβ₄₀-induced IL-6 production in HMC3 and THP-1 cells.

(A) HMC3 cells were pre-incubated with BIRB or DMSO as vehicle control (drug concentration was marked as 0) 30 min prior to Aβ₄₀ (2 μM) treatment. Expression levels of p-p38 and total p38 were determined by Western blotting at 4 h after Aβ₄₀ treatment. (B) THP-1 cells were pre-incubated with BIRB 30 min prior to Aβ40 (2 μM) treatment or DMSO. Expression levels of p-p38 and total p38 were determined by immunoblotting at 2 h after Aβ treatment. (C) HMC3 cells were incubated with BIRB or DMSO as vehicle control 30 min prior to Aβ₄₀ (2 μM) treatment. Brefeldin A (0.5 μg/mL) was added to the medium during the last 4 h of culture. IL-6 expression levels were assessed by immunoblotting at 8 h after Aβ₄₀ treatment. (D) ELISA data indicates that BIRB dose-dependently attenuates Aβ-induced increase in IL-6 secretion. (E) ELISA data depicts that SB202190 (SB) dose-dependently inhibits Aβ-induced IL-6 secretion by HMC3 cells. (F) ELISA reveals that BIRB dose-dependently attenuates Aβ-induced increase in IL-6 secretion by THP-1 cells. Data are presented as mean +/− SEM of three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. control.

Next, we investigated the effects of p38 inhibition on Aβ₄₀-induced IL-6 secretion. Levels of IL-6 in cell culture supernatants were analyzed using ELISA. The data reveal that inhibition of p38 by BIRB inhibits the Aβ₄₀-induced increase of IL-6 secretion by HMC3 cells (Fig. 5D). The ability of p38 inhibition to block Aβ₄₀-induced IL-6 secretion was further confirmed by using another structurally and mechanistically different p38 inhibitor, SB202190 (SB) (Fig. 5E). Additionally, we show that inhibition of p38 by BIRB dose-dependently inhibits Aβ₄₀-induced secretion of IL-6 by THP-1 cells (Fig. 5F). Collectively, these results demonstrate that inhibition of p38 by SB or BIRB is sufficient to abolish Aβ₄₀-induced increase in IL-6 secretion.

Aβ₄₀-induced p38 activation is associated with increased mROS production

It has been shown that mitochondria and the generation of mitochondrial ROS (mROS) are critical to T cell activation and LPS-induced pro-inflammatory response in mouse microglia [43, 44]. However, the role of mROS in Aβ₄₀-induced p38 activation in human microglia remains largely unknown. We thus examined the effects of Aβ₄₀ treatment on mROS production. Our data indicate that Aβ₄₀ treatment markedly increases the generation of mROS in THP-1 cells in a time- and dose-dependent manner (Fig. 6A–C). Furthermore, using confocal microscopy, we found that Aβ₄₀ treatment significantly increases the levels of mROS in HMC3 microglial cells (Fig. 6D & E). Together with the observation that Aβ₄₀ treatment selectively induces p38 activation in both HMC3 and THP-1 cells (Fig. 4), our studies demonstrate that Aβ₄₀-induced p38 activation is associated with an increase in mROS production. These findings are consistent with previous reports showing that oxidative stress is a potent inducer of p38 activation, and that oxidative stress-induced p38 activation has been implicated in Aβ- and tau-induced toxicities in neurons [45–49]. Furthermore, it was reported that Aβ can stimulate the production of nitrogen oxide (NO) in rodent glial cells [50–52]. However, our data show that Aβ₄₀ has no significant effects on NO production in HMC3 cells, whereas LPS markedly increases NO production in Raw 264.7 cells (Supplemental Fig. S3).

Fig. 6. Aβ₄₀ stimulates the production of mROS in HMC3 and THP-1 cells.

(A) MitoSox staining and flow cytometric assays were performed to measure mROS levels in THP-1 cells at 6 h after Aβ₄₀ (2 μM) treatment. Representative flow cytometric graphs are shown, and the results indicate that Aβ₄₀ increases mROS production in THP-1 cells. (B) Treatment with Aβ₄₀ (2 μM) increases mROS production in THP-1 cells in a time-dependent manner. (C) Aβ₄₀ dose-dependently promotes mROS production in THP-1 cells. (D) MitoSox staining and confocal microscopy were employed to measure mROS in HMC3 cells at 6 h after Aβ₄₀ (2 μM) treatment. (E) Mean fluorescence intensity (MFI) of MitoSOX staining was quantified using ImageJ software and the quantified results are graphed. Data are presented as mean +/− SEM of three independent experiments. ** p < 0.01, *** p < 0.001 vs. control.

Inhibition of p38 diminishes Aβ phagocytosis by microglia

Microglia were thought to be a major neural cell type involved in clearing of Aβ deposition in the brain [53, 54]. However, whether p38 is involved in regulating the phagocytosis of Aβ in human microglia has not been studied previously. Our confocal microscopic data indicate that blocking of p38 by BIRB markedly inhibits the phagocytosis of Aβ by HMC3 cells (Fig. 7A & B). Similar results were observed using BV-2 mouse microglial cells (Fig. 7C & D). These results demonstrate a role of p38 in modulating the phagocytosis of Aβ by microglia, suggesting that BIRB-mediated blockade of Aβ uptake may be a possible mechanism by which p38 inhibition attenuates Aβ-induced IL-6 production in HMC cells.

Fig. 7. Inhibition of p38 attenuates the phagocytosis of Aβ by microglia.

(A) HMC3 cells were treated with BIRB (2 μM) at 1 h before incubation with 1 μM FAM-488 labeled-Aβ₄₂ for 2h. Cells were fixed with 4% paraformaldehyde and stained with an IBA1-specific antibody to label microglia. Nuclei were counterstained with DAPI. Levels of Aβ₄₂ uptake were assessed using confocal microscopy. (B) Confocal imaging data were analyzed and quantified using ImageJ. Levels of Aβ phagocytosis were presented as MFI of FAM-488 labeled-Aβ₄₂. (C) BV-2 cells were treated and analyzed using the same protocol as described in A. Shown are representative confocal microscopic images of Aβ phagocytosis. (D) Confocal imaging data were analyzed and quantified using ImageJ. Levels of Aβ phagocytosis were presented as MFI of FAM-488 labeled-Aβ₄₂. Data are presented as mean +/− SEM of three independent experiments. * p < 0.05, ** p < 0.01 vs. DMSO control.

IL-6 colocalized with phosphorylated p38 in human microglia of AD patients

It was reported that IL-6 expression levels were elevated in the cortices of AD patients [10, 11]. However, the cellular sources for the increased IL-6 in AD remain to be determined. Using immunofluorescence and Airyscan super-resolution confocal microscopy, we investigated if IL-6 expression levels were altered in microglia of AD patients. Consistent with the observation of a recently published study [55], we confirmed that the morphologic characteristics of human microglia were markedly altered in the cortices of AD patients, which displayed the pathological feature of shortened and de-ramified processes along with discontinuity and puncta IBA1 immunostaining (Fig. 8A). Quantification of the confocal microscopic imaging data demonstrated that both IL-6 and phosphorylated p38 (p-p38) expression levels were dramatically increased in microglia of AD patients compared with those in age-matched control subjects (Fig. 8B). Transmembrane protein 119 (TMEM119) is a cell-surface protein highly expressed in both mouse and human microglia and is thought to be a microglia-specific marker [56]. Using a TMEM-specific antibody, we further confirmed that most microglia, especially those with de-ramified and amoeboid characteristics of activated microglia, expressed high levels of IL-6. In contrast, a few ramified microglia with smaller size did not show immunoreactivity for IL-6 (Supplemental Fig. S4). Our studies also included the secondary antibody only controls to test the specificity of the antibodies (Supplemental Fig. S5).

Fig. 8. Increased expression and co-localization of IL-6 with p-p38 in microglia in the cortices of AD patients.

(A) IL-6 and IBA1 co-labeling and confocal microscopy were performed to assess IL-6 expression levels in microglia in human brain tissues isolated from AD patients and age-matched control subjects. (B) Quantification of confocal imaging data unveiled that the fluorescent intensity and positively stained area for IL-6 and p-p38 were substantially higher in microglia from AD patients than those from age-matched control subjects. (C) Representative confocal microscopic images depict the elevated expression and co-localization of IL-6 with p-p38 in human AD microglia. Scale bar = 20 μm. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. control subjects (Ctrl).

Although activated microglia are considered as the major source of inflammatory cytokines, it is worth noting that astrocytes and microglia can work collaboratively to regulate neuroinflammation in AD [3]. Moreover, there are conflicting reports regarding whether astrocytes produce IL-6 or not [57, 58]. To examine if reactive astrocytes express IL-6 in AD brains, we performed a confocal microscopic analysis of AD patient samples. Consistent with a recently published observation [58], we found that a portion of the astrocytes showed immunoreactivity for IL-6 in the brain tissues of AD patients (Supplemental Fig. S6).

To determine the role of the p38 pathway in regulating IL-6 expression in AD microglial cells, we investigated if there is a link between increased IL-6 levels and the activated form of p38 protein in postmortem brain specimens isolated from AD patients. Our data have confirmed the colocalization of IL-6 with p-p38 in microglia in the cortices of AD patients (Fig. 8C). Taken together, the results support the hypothesis that p38 may play a major role in modulating IL-6 production in microglia of AD patients.

Discussion

AD and AD-related dementias (ADRD) affect more than 6 million senior Americans aged 65 or older, but effective therapies have yet to be developed [1, 2]. One of the major roadblocks toward the success of AD prevention and treatment is that we still do not fully understand the pathogenic mechanisms of this devastating disease. The chronic deposition of Aβ peptides in senile plaques stimulates a sustained inflammatory response, leading to a persistent increase in the production of pro-inflammatory cytokines in the AD brain, which is referred to as neuroinflammation [3]. Since pro-inflammatory cytokines have been shown to affect the neuropathology and cognitive function in neurodegenerative disease animal models, Aβ-induced neuroinflammation is thought to be an important contributing factor to the pathogenesis of AD [10–13]. In agreement with this concept, it has been shown that increased levels of inflammatory cytokine production can cause synaptic dysfunction, neuronal death, and neurodegenerative disorders [6, 7, 10, 14]. Notably, IL-6 expression levels are increased in the brains of AD patients [10, 11]. Moreover, a growing body of evidence suggests that increased levels of IL-6 production may play a causal role in mediating inflammatory stress-induced neurodegenerative pathologies and cognitive declines [6, 11, 14]. However, the mechanisms by which IL-6 expression is regulated in brain tissues of AD patients have not been clearly elucidated. In the present study, we show that Aβ₄₀ dose-dependently stimulates IL-6 production in HMC3 human microglial cells. We also show that Aβ treatment promotes the transcription of IL-6 and TNFα mRNAs in both HMC3 and THP-1 cells. Furthermore, our data reveal that Aβ selectively induces p38 activation and pharmacological inhibition of p38 by BIRB or SB abrogates the Aβ-induced increase in IL-6 secretion. These results suggest that activation of the p38 signaling pathway may play a critical role in Aβ-induced IL-6 production in human microglia.

Microglial activation has been found in the pre-plaque stage in transgenic animal models of AD [59, 60]. Reactive microglia were also observed in the brains of AD patients and were colocalized with Aβ plaques [61, 62]. The sequential emergence of amyloid accumulation and microglial activation and release of cytokines suggests that microglia may play a significant role in mediating Aβ-induced neural cell death, neuroinflammation and subsequent tau pathology, all of which are involved in AD pathogenesis. Mitochondrial ROS (mROS) can function as signaling intermediates and are required for T cell activation [43]. Moreover, the production of mROS is required for LPS-induced inflammatory response in mouse microglia [44]. Nevertheless, the role of mROS in Aβ-induced p38 activation and IL-6 production in human microglia has not been explored previously. Our data reveal that Aβ₄₀ treatment increases the levels of mROS in both THP-1 and HMC3 cells. These results demonstrate a link between Aβ₄₀-induced p38 activation and the increased production of mROS in human microglial cells. Consistent with these findings, previous studies indicated that oxidative stress could induce p38 activation [45, 46].

The accumulation of Aβ plaques in the brain is thought to be a critical and initial event in the pathogenesis of AD, which is the main concept of the Aβ hypothesis of AD [63]. However, therapeutics that block Aβ-induced neuroinflammation have yet to be developed. Recently, the anti-human Aβ antibody, Aduhelm (aducanumab), was granted a fast-track approval by the U.S. Food and Drug Administration (FDA), for AD treatment. AD patients who received aducanumab treatment displayed a dose- and time-dependent reduction of Aβ plaques, while patients in the control arm did not show such changes [64, 65]. These results not only support the Aβ hypothesis of AD, but also suggest that targeting Aβ by Aduhelm is a viable therapeutic approach for AD treatment. Although p38 has been found to be a key mediator of various stress responses [39–41], the role of p38 in modulating IL-6 production in human microglia was largely unknown prior to our present study. Our data demonstrate that activation of p38 is critical for Aβ₄₀-induced increase of IL-6 production in human microglia, suggesting that blocking of p38 may represent an alternative therapeutic approach for AD via targeting the downstream effector of the Aβ-induced pathogenic process. In agreement with this, there is evidence that p38 is involved in regulating Aβ-induced synaptic dysfunction and that inhibition of p38 ameliorates tau pathology, leading to decreases in hyperphosphorylated tau levels in microglia [66–69]. Moreover, it has been shown that genetic depletion of p38 attenuates AD pathology in a mouse model of AD [69].

Although previous studies have shown that IL-6 levels were elevated in the brains of AD patients [10, 11], it remains unclear about the exact brain cell types where the elevated levels of IL-6 come from. More importantly, a link between the elevated IL-6 expression levels and increased immunostaining for the activated form of p38 (p-p38) in microglia in the brains of AD patients has yet to be established. In the present study, we found that the immunoreactivities for both IL-6 and p-p38 were markedly higher in microglia from AD patients than those from age-matched control subjects. More importantly, our studies have demonstrated for the first time that IL-6 colocalize with p-p38 in microglia in the cortices of AD patients.

The role of p38 signaling in Aβ-induced alterations in neuroinflammatory cytokines such IL-6 in human microglia was largely unknown prior to the present study. This is an important knowledge gap because it has been clearly shown that there are species-specific differences in gene expression patterns between mouse and human microglia [70]. Our data show that treatment with Aβ₄₀ leads to persistent activation of p38 in HMC3 human microglia. In contrast, only a transient increase of phosphorylated p65 was observed in HMC3 cells following Aβ₄₀ treatment. These results suggest that p38 is likely a key regulator of Aβ-induced neuroinflammatory response in human microglia. In agreement with this idea, our data have confirmed that pharmacological inhibition of p38 by SB or BIRB is sufficient to abrogate Aβ-induced increase in IL-6 production in HMC3 cells.

In summary, our studies offer the first experimental evidence that Aβ₄₀ stimulates IL-6 production in human microglia in a dose- and time-dependent fashion. The Aβ₄₀-induced increase in IL-6 production is associated with the activation of the p38 pathway. Inhibition of p38 abrogates Aβ₄₀-induced increase in IL-6 production. An analysis of clinical brain specimens revealed that the immunoreactivities for IL-6 and p-p38 were markedly higher in microglia of AD patients than in age-matched control subjects. Moreover, our studies have confirmed the co-localization of IL-6 with p-p38 in microglia in the cortices of AD patients. These results demonstrate a critical role for p38 in mediating Aβ-induced neuroinflammatory response such as increased IL-6 production from microglia in AD.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.