Background
Alzheimer's disease (AD), a progressive neurodegenerative disease, is the most common cause of dementia [
1]. The typical symptoms of AD include learning and memory loss and cognitive impairment [
1,
2]. Amyloid-β (Aβ) plaque deposition is one of the major pathological hallmarks in AD [
2‐
4]. Aβ is generated from amyloid precursor protein (APP) by β-site APP cleavage enzyme 1 (BACE1) and presenilin 1 (PS1) [
3,
5]. The toxic Aβ activates astrocytes and microglia to trigger chronic inflammation, thereby releasing cytokines, pro-inflammatory mediators and reactive oxygen species (ROS) and leading to progression of AD [
6‐
10]. Several lines of evidence suggest that Aβ-mobilized calcium (Ca
2+) flow is closely associated with the inflammatory response in astrocytes [
11‐
13]. On Aβ stimulation, elevated intracellular Ca
2+ level promotes cytokine secretion by activating protein phosphatase 2B (PP2B, also called calcineurin) and its downstream NF-κB and nuclear factor of activated T cells (NFAT) [
14‐
16]. However, which type of cation channel is responsible for the Aβ-activated Ca
2+ signaling and inflammation remains for further investigation.
Transient receptor potential ankyrin 1 (TRPA1) channel is a type of nonselective transmembrane cation channel with multiple ankyrin repeats on its N-terminal [
17‐
20]. TRPA1 channel is expressed in primary sensory neurons and non-neuronal cells and may be a sensor for detecting ROS, cold temperature and cannabinoids [
21‐
24]. Upon detection of these signals, the TRPA1 channel is activated, which results in increased intracellular Ca
2+ levels and activated downstream signaling cascades [
17‐
20]. In the brain, the TRPA1 channel plays an important role in regulating brain development and the physiological function of astrocytes [
24,
25]. In addition, it may be a key gatekeeper in regulating the inflammatory response with stimuli including bacterial endotoxin, environmental irritants or inflammatory mediators [
26‐
28]. Recently, AD research has focused on the emerging role of Ca
2+-related signaling pathways in the pathogenesis of AD [
29,
30]. However, the role and underlying molecular mechanism(s) of the TRPA1–Ca
2+ signaling cascade in AD pathogenesis are still elusive.
In this study, we aimed to investigate the role of TRPA1 channels in AD pathogenesis and the possible molecular mechanisms in a mouse model. We determined the expression of TRPA1 channels in vivo by using wild-type (WT) and APP/PS1 transgenic (Tg) mice, then investigated whether the TRPA1 channel plays a role in the development of AD by using APP/PS1 Tg and APP/PS1 Tg/TRPA1−/− mice. Finally, we assessed the importance of TRPA1 and the potential mechanism underlying the regulation of Aβ-mediated inflammation in mice and astrocytes.
Methods
Reagents
Rabbit antibody for TRPA1 (NB110-40763) was from Novus (Littleton, CO, USA). Goat antibody for Akt (sc-1619), rabbit antibodies for PP2B (sc-9070), mouse antibodies for glial fibrillary acidic protein (GFAP, sc-166481), goat anti-rabbit FITC-conjugated (sc-2012), goat anti- mouse Texas red-conjugated (sc-2781) and FITC-conjugated (sc-2010) antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Mouse antibodies for GFAP (MAB360), NeuN (MAB3770), ionized calcium-binding adapter molecule 1 (IBA-1, MABN92), von Willebrand factor (vWF, MAB7356), and the cellular PP2B activity kit were from Millipore (Darmstadt, Germany). Rabbit LDLR-related protein 1 (LRP-1, L2170), mouse antibody for α-tubulin (T-9026), bovine serum albumin (BSA), phosphatase inhibitor cocktails 1 and 2, HC030031, allyl isothiocyanate (AITC), ethylene glycol tetraaceticacid (EGTA), ethylenediaminetetraacetic acid (EDTA), cyclosporine (CsA) and fenvalerate (Fen) were from Sigma-Aldrich (St. Louis, MO, USA). Mouse antibody for Aβ (SIG-39320-200) was from Covance (Dedham, MA, USA). Rabbit antibody for β-APP C-terminal fragment (βCTF, 802801) was from BioLegend (San Diego, CA, USA). Mouse antibody for ATP-binding cassette transporter A1 (ABCA1, ab18180), IL-4 (ab9622) and IL-10 (ab9969) were from Abcam (Cambridge, MA, UK). Rabbit antibody for apolipoprotein E (apoE, 1930-5) was from Epitomics (Burlingame, CA, USA). Mouse anti-phosphor-Akt (587 F11) was from Cell Signaling (Danvers, MA, USA). Retrieval buffer was from Biocare Medical (Concord, CA, USA). The mounting medium with DAPI was from Vector Laboratories (Burlingame, CA, USA). TurboFect was from Fermentas (Glen Burnie, MD, USA). The ELISA kit for NF-κB activity was from Cayman Chemical (Ann Arbor, MI, USA) and for NFAT activity was from Active Motif (Carlsbad, CA, USA). ELISA kits for IL-1β, IL-4, IL-6 and IL-10 and mouse antibodies for IL-1β (AF-401-NA) and IL-10 (AF-406-NA) were from R&D systems (Minneapolis, MN, USA). Quest™ Fluo-8 NW calcium assay kit was from AAT Bioquest (Sunnyvale, CA, USA).
Mice
The investigation conformed to the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, eighth edition, 2011), and all animal experiments were performed in accordance with the approved guidelines by the Animal Care and Utilization Committee of the National Yang-Ming University (#1031269). B6.Cg-Tg(APPswe, PSEN1dE9)85Dbo/J (APP/PS1 Tg) mice and TRPA1−/− mice were purchased from Jackson Laboratory (Bar Harbor, ME, USA) and were backcrossed to C57BL mice for at least 10 generations to ensure genetic homogeneity. For APP/PS1 Tg/TRPA1−/− mice, TRPA1−/− mice were crossed with the APP/PS1 Tg background, and the genotypes were confirmed by PCR of genomic DNA. Mice were housed in barrier facilities, maintained on a 12-h/12-h dark cycle. Temperature (22 °C) and humidity (40-60 %) of the vivarium were tightly controlled. Mice were group-housed 3–4 per cage and fed a regular chow diet, which contained 4.5 % fat by weight (0.02 % cholesterol) (Newco Distributors, Redwood, CA, USA). At the end of the experiment, mice were euthanized with CO2, then brains were harvested for histological analysis and stored at −80 °C. The isolated brains were homogenized and lysates were subjected to western blot analysis.
Western blot analysis
Cells and brain tissues were lysed in immunoprecipitation lysis buffer (50 mmol/L Tris pH 7.5, 5 mmol/L EDTA, 300 mmol/L NaCl, 1 % Triton X-100, 1 mmol/L phenylmethylsulfonyl fluoride, 10 μg/mL leupeptin and 10 μg/mL aprotinin). Aliquots of brain lysates or cell lysates were separated on SDS-PAGE, transferred to membranes and immunoblotted with primary antibodies (1:1000), then horseradish peroxidase-conjugated secondary antibody (1:1000). Bands were revealed by use of an enzyme-linked chemiluminescence detection kit (PerkimElmer, Waltham, MA) and density was quantified by use of Imagequant 5.2 (Healthcare Bio-Sciences, Philadelphia, PA).
Immunohistochemistry staining
The brain sections were fixed in 4 % paraformaldehyde and 15-μm cross sections were prepared. Sections were incubated with retrieval buffer for 10 min, blocked with 2 % BSA for 60 min and incubated with primary antibody (1:100) overnight at 4 °C, then FITC- or Texas red-conjugated secondary antibody (1:400) for 1 h at 37 °C. Antigenic sites were visualized under a Nikon TE2000-U microscope (Tokyo) with QCapture Pro 6.0 software (QImaging, BC, Canada).
Fibrilization of Aβ1–42
Aβ
1–42 was purchased from American Peptide Co. (Sunnyvale, CA, USA), solubilized in sterile water (1 mg/mL) and incubated 1 to 7 days at 37 °C for fibrillation as described [
31‐
33]. The level of Aβ fibrilization on a 16.5 % tricine gel was examined by western blot analysis (Additional file
1: Figure S1).
Cell culture
The primary culture of astrocytes and neurons was prepared as described [
24,
34]. Briefly, the cortex and hippocampus were isolated from pups on postnatal 1 day and loosely homogenized by use of a sterile razor blade in DMEM/F12 (HyClone, Logan, UT). Tissues were digested with 0.01 % trypsin and incubated at 37 °C for 25 min; the cell suspension was titrated by use of a 70-μm nylon mesh. Isolated cells were seeded onto 75-mm flasks and incubated for 7 days in DMEM/F12 supplemented with 10 % FBS, 100 U/mL penicillin and 100 μg/mL streptomycin at 37 °C. Cells were re-suspended, followed by orbital shaking at 180 rpm for 24 h to remove microglia and oligodendrocytes. The purified astrocytes that tightly adhered to the bottom of the flasks were then detached with trypsin and seeded onto culture dishes and incubated for an additional 7 days to return to a resting state. The primary neuron culture was prepared as follows: the cortex was isolated from pups on postnatal 1 day and loosely homogenized by use of a sterile razor blade in DMEM (HyClone, Logan, UT). Isolated cells were seeded onto 3.5-cm dishes and incubated in Neurobasal media supplemented with 2 % B-27 supplement, 0.25 % GlutaMAX, 10 % FBS, 100 U/mL penicillin and 100 μg/mL streptomycin (Thermo Fisher Scientific, Waltham, MA, USA) at 37 °C. Cells were treated with cytosine-1-β-D-arabinofuranoside (10 μM) from day 2 and half the medium was changed every 3 days for 14 days’ culture. Human embryonic kidney 293 (HEK293) cells and mouse brain microvascular endothelial cells (BMECs), bEnd.3 cells, were cultured in DMEM supplemented with 10 % FBS, 100 U/mL penicillin and 100 μg/mL streptomycin. The growth media was replaced every other day.
Plasmid construction and transient transfection
The coding region for the human TRPA1 DNA fragment was cloned into a pCMV5 N-Flag vector with MluI and HindIII restriction sites. The sequence of isolated DNA fragments was confirmed by sequence analysis. TurboFect was used for transient transfection experiments according to the manufacturer’s instructions. Briefly, 1 μg of vector or TRPA1 plasmid was transfected into HEK293 cells. Transfected cells were used in further experiments.
Detection of Ca2+ influx
Primary astrocytes or HEK293 cells were pretreated with Fluo-8NW dye for 1 h, then medium was replaced with fresh medium containing test compounds. The intensity of fluorescence was evaluated by fluorometry (Molecular Devices, Sunnyvale, CA, USA) with 490-nm excitation and 525-nm emission. Images were captured under a TE2000-U fluorescence microscope and quantified with use of QCapture Pro 6.0.
Measurement of PP2B activity
The activity of PP2B in primary astrocytes or fresh brain lysates was measured by use of a cellular PP2B activity kit.
Measurement of inflammatory cytokines
The concentrations of inflammatory cytokines including IL-1β, IL-4, IL-6 and IL-10 in culture medium or brain lysates were measured by use of ELISA kits.
Measurement of DNA-binding activity on NF-κB and NFAT
The DNA-binding activity of NF-κB and NFAT in primary astrocytes and brain was measured by use of ELISA kits.
Immunocytochemical staining
Primary astrocytes were fixed with 4 % paraformaldehyde for 30 min, blocked with 2 % BSA for 30 min and incubated with primary antibodies (1:100), then FITC- or Texas red-conjugated secondary antibodies (1:400). Cellular images were viewed under a TE2000-U fluorescence microscope and quantified with use of QCapture Pro 6.0.
Open field activity
The locomotor activity of mice was assessed in a cage (length × width × height: 28.5 × 28.5 × 30 cm). Mice were placed in the central of the cage and allowed to explore the open field for 5 min. The behavior was recorded by video, and the movement distance, percentage of resting time in the zone and trajectory were calculated for each mouse by use of Smart v3.0 software with the Panlab Harvard apparatus (Cornellà, Barcelona, Spain). The floor and internal walls were cleaned with ethanol between each trial.
Nest-building test
The nest-building test was performed as described [
35]. Each mouse was housed in single cages containing two pieces of cotton (5 × 5 cm). The presence and quality of nests built were recorded by the nesting score, measured on a 5-point scale: 1 = cotton not noticeably touched, 2 = cotton partially torn up, 3 = mostly shredded cotton but often no identifiable nest location, 4 = a markedly nesting site but flat nest, and 5 = a (near) perfect nest. Nesting score was recorded manually at 72-h intervals.
Y-maze test
The Y-maze test was performed as described [
36]. The Y-maze apparatus consists of three arms of channels made of stainless steel joined in the middle to form a “Y” shape. The mice were placed into one of the arms (start arm) and allowed to explore the maze with only one of the arms closed for 10 min (training trial). After 3 h, mice were placed back in the start arm of the Y maze. Then, mice were allowed to explore all three arms freely for 5 min (test trial). The number of entries into each arm, the distance of movement and the first choice of entry were assessed in video recordings.
Morris water maze (MWM)
MWM was performed as described [
36]. A large circular tank (0.8 m diameter, 0.4 m depth) was filled with water (25 ± 1 °C, 20 cm depth), and the escape platform (8 × 4 cm) was submerged 1 cm below the surface. The training section was monitored by a video system. The escape latency and trajectory of swimming were recorded for each mouse. The hidden platform was located at the center of one of the four quadrants in the tank. The location of the platform was fixed throughout the testing. Mice had to navigate using extra-maze cues that were placed on the walls of the maze. From days 1 to 4, mice went through three trials with an inter-trial interval of 5 min. The mouse was placed into the tank facing the side wall randomly at one of four start locations and allowed to swim until it found the platform or for a maximum of 120 sec. Mice that failed to find the platform within 120 sec were guided toward the platform. The animal then remained on the platform for 20 sec before being removed from the pool. The day after the hidden platform training, a probe trial was conducted to determine whether the mouse used a spatial strategy to find the platform. On day 5, the platform was removed from the pool and the mouse was allowed to swim freely for 120 sec. The proportion of time spent in each quadrant of the pool and the number of times the mouse crossed the former position of the hidden platform were recorded.
Statistical analysis
Results are presented as mean ± SEM. Data from cell studies were evaluated by non-parametric tests. Mann-Whitney U test was used to compare 2 independent groups. Kruskal-Wallis followed by Bonferroni post-hoc analyses was used to account for multiple testing. Data from animal studies were evaluated by parametric tests. Two-way ANOVA followed by LSD test was used for multiple comparisons. The reagent effect and genotypic effect were two independent factors for this analysis. SPSS v20.0 (SPSS Inc, Chicago, IL) was used for analysis. Differences were considered statistically significant at P < 0.05.
Discussion
Our study demonstrated the novel role of the TRPA1 channel in astrocytes under an Aβ-elicited inflammatory environment and thus its potential involvement in AD pathogenesis. The protein level of the TRPA1 channel was increased in astrocytes of APP/PS1 Tg mice at 8 months old as compared with WT mice at the same age. As well, loss of function of the TRPA1 channel impeded AD progression, as evidenced by improved nest building ability, spatial learning and memory and decreased Aβ plaque deposition and cytokine production in APP/PS1 Tg mice.
In vitro, TRPA1 channels mediated the Aβ-triggered Ca
2+ influx and inflammation in astrocytes. Gain of function of TRPA1 in TRP channel–deficient HEK293 cells further supports that TRPA1 activation is crucial in Aβ-mediated Ca
2+ influx. Aβ evoked TRPA1-Ca
2+ signaling, which in turn activated PP2B, NF-κB and NFAT. The activation of these proteins increased the production of inflammatory cytokines. This inflammatory cascade in our model agreed with findings by Fernandez et al. and Furman et al., who established that increased Ca
2+ influx is a key event for activation of PP2B signaling and inflammation in astrocytes [
14‐
16]. Finally, behavioral analysis demonstrated that functional loss of TRPA1 ameliorated AD progression and improved neuropsychiatric signs, positively affecting cognition and spatial learning and memory. Collectively, our results suggest that the TRPA1 channel is involved in Aβ-triggered inflammatory responses of astrocytes and the development of AD.
The association of deregulated cellular Ca
2+ homeostasis and AD pathogenesis has been established
in vitro and
in vivo [
41‐
43]. Emerging evidence further supported the causal relationship between the impaired Ca
2+ homeostasis and synaptic dysfunction and neuronal degeneration of AD [
1]. Specifically, exposure of Aβ to neurons disrupts Ca
2+ homeostasis and causes oxidative damage, thereby leading to neuronal death and synaptic dysfunction via multiple receptor-dependent pathways including NMDA receptor or voltage-gated Ca
2+ channels [
1]. However, research into Ca
2+ homeostasis of astrocytes and its role in AD pathogenesis is still in its infancy. Several lines of evidence indicated that TRP channels such as canonical (TRPC), melastatin (TRPM), and vanilloid (TRPV) channels also play a role in AD pathogenesis [
44,
45]. Especially, astrocyte-related inflammation is a key factor in AD progression [
8,
9,
46,
47]. Aβ-induced Ca
2+ influx is closely associated with the inflammatory responses in astrocytes [
11‐
13]. In astrocytes, Aβ elicits the production of ROS and nitric oxide (NO), which can activate TRPM2, TRPM7, TRPC5 and TRPV1 and increase the intracellular level of Ca
2+, thereby leading to AD-related events in the brain including neurodegeneration and inflammation [
30].
However, how TRPA1 activation, particularly in astrocytes, contributes to AD pathogenesis has not been investigated. Our results demonstrated that Aβ triggered TRPA1-dependent Ca
2+ influx and subsequently increased PP2B activity, which promoted activation of NF-κB and NFAT, thereby leading to production of pro-inflammatory cytokines from astrocytes. Most importantly, we observed these key events in our
in vivo model. Of note, in astrocytes, both pretreatment with HC030031 and genetic loss of function of TRPA1 channels could only partially obstruct the Aβ-elicited Ca
2+ influx. In contrast, pretreatment with EGTA and EDTA completely abolished Ca
2+ influx induced by Aβ. These results imply the existence of other routes of Ca
2+ influx triggered by Aβ. Indeed, several proteins regulating Ca
2+ flow on Aβ challenge in astrocytes have been identified [
42,
43]. In response to ligands, TRP channels can form homo- or heterotetramers or activate other TRP channels, thereby increasing intracellular Ca
2+ level and activating downstream Ca
2+-mediated signaling cascades [
48,
49]. However, whether other TRP channels participate in the Aβ-induced TRPA1 channel activation and if so, how they activate the TRPA1 channel remains unknown.
PP2B activity of astrocytes is highly associated with neurodegenerative diseases [
50,
51]. Increased numbers of PP2B-positive astrocytes were found in the immediate vicinity of extracellular Aβ deposition in patients with dementia and in an AD mouse model [
50,
51]. In addition, disruption of Ca
2+ homeostasis causes hyperactivity of PP2B signaling cascades, which can amplify the offset effect of Ca
2+ dysregulation [
15]. This notion was further supported by our findings that TRPA1-derived Ca
2+ mobilization was critical in Aβ-activated PP2B, as evidenced by reduced PP2B activity in TRPA1 antagonist–treated or TRPA1-deficient astrocytes or brain tissues of APP/PS1 Tg/TRPA1
−/− mice. PP2B can lessen Akt activity its downstream signaling pathways, which are involved in inflammation and AD [
52‐
54]. In fact, deletion of TRPA1 decreased PP2B activity but increased the phosphorylation of Akt in brain tissues of APP/PS1 Tg mice (Additional file
1: Figure S5). Thus, TRPA1 − Ca
2+ − PP2B signaling may play an important role in AD progression. However, the detailed molecular mechanism by which PP2B regulates AD progression needs further investigation.
In the AD brain, Aβ plaques surrounding astrocytes can secrete inflammatory mediators to regulate neuroinflammation [
55]. IL-1β, IL-4, and IL-6 are involved in the initiation and progression of AD by deregulating Aβ-mediated inflammation and APP metabolism [
55‐
57]. In contrast, IL-10 can limit inflammation by reducing pro-inflammatory cytokines during AD pathogenesis [
55]. In line with these findings, our data demonstrated lower levels of IL-1β, IL-4, IL-6, but IL-10 in APP/PS1 Tg/TRPA1
−/− mice than APP/PS1 Tg mice. Moreover, functional inhibition of TRPA1 abrogated the Aβ-triggered production of IL-1β, IL-4 and IL-6 in astrocytes. The transcriptional factors NF-κB and NFAT are key regulators in gene expression of IL-1β, IL-4 and IL-6 [
37‐
40]. Consistently, we found that functional deletion of TRPA1 decreased Aβ-induced NF-κB and NFAT activity in astrocytes and mice. Nevertheless, the mechanism underlying TRPA1-mediated regulation of inflammation and AD pathogenesis needs further investigation.
Astrogliosis (referred to as reactive astrocytes) occurs prominently in response to central nervous system injury or remodeling [
9,
58]. Although the biological function of astrogliosis is not fully understood, reactive astrocytes are intimately associated with Aβ plaque and involved in the regulation of neural protection and repair, glial scarring and neuro-inflammation in the pathogenesis of AD [
58]. In addition, reactive astrocytes promote the clearance and degradation of Aβ in AD brain and thus limit the growth of Aβ plaque [
59]. Interestingly, we found greater astrogliosis after treatment with a TRPA1 antagonist or functional deletion of TRPA1
in vitro and
in vivo. In parallel, the neuro-inflammation was ameliorated with both pharmacological inhibition and genetic deletion of TRPA1 in Aβ-treated astrocytes and mouse brains. Activation of TRPA1-Ca
2+ signaling might be critical in regulating Aβ-mediated astrogliosis and neuro-inflammation.
APP can be rapidly metabolized by post-translational proteolysis via an amyloidogenic or non-amyloidogenic pathway in neurons, and its role in AD is well established [
3,
60]. In the amyloidogenic pathway, APP is cleaved by β-secretase, and N-terminal APP fragments and βCTF are produced and released. βCTF is then cleaved by γ-secretase and generates an amyloid precursor protein intracellular domain and toxic Aβ peptide (37–49 amino acids), which triggers inflammation and neuron dysfunction [
3,
5,
60]. In contrast, in the non-amyloidogenic pathway, APP is cleaved by α-secretase and yields a soluble N-terminal APPα fragment and α-C-terminal fragment (αCTF). αCTF is then cleaved by γ-secretase to generate non-toxic p83 peptide (23–25 amino acids) [
60,
61]. Disruption of Ca
2+ homeostasis may deregulate APP processing and increase Aβ generation [
61]. This notion was supported by our findings that ablation of TRPA1-Ca
2+ signaling decreased Aβ production and accumulation in the AD mouse brain. Notably, our results further demonstrated that functional loss of TRPA1 channels increased the levels of βCTF but decreased that of Aβ in the AD mouse brain. Thus, TRPA1-Ca
2+ signaling might regulate APP processing, especially at the step of cleavage of βCTF to Aβ generation.
Previous evidence linked cholesterol metabolism of the brain to Aβ clearance [
62,
63]. Astrocyte-derived apoE-containing high-density lipoprotein-like particles play a central role in neural repair during AD development [
62‐
64]. These particles can be transported by ABCA1 into the extracellular space to bind with Aβ and then are taken up by neurons, astrocytes and microglia through LRP-1 to remove the Aβ deposition [
63]. The expression of ABCA1 and LRP-1 is increased with AD progression [
65,
66]. We found that functional loss of TRPA1 channels decreased the expression of ABCA1 and LRP-1 without altering apoE expression and its co-localization with Aβ plaque in APP/PS1 Tg mice. As we described previously, Aβ accumulation was lower in the brain of APP/PS1 Tg/TRPA1
−/− than APP/PS1 Tg mice, so APP/PS1 Tg/TRPA1
−/− mice might not need such levels of ABCA1 and LRP-1 to clear Aβ. Nevertheless, we cannot conclude that TRPA1 channels are a negative regulator of Aβ clearance based on these observations.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
KIL, HTL, HCL and FCT performed experiments and analyzed the data. HJT, SKS and TSL designed the experiments and TSL wrote the paper. All authors have read and approved the submission of the manuscript.