Introduction
Alzheimer’s disease (AD) is the most prevalent neurodegenerative illness in older adults and has a characteristic neuropathology that includes plaques comprised of β amyloid (Aβ) and tangles of hyperphosphorylated tau [
1,
2]. Aβ42 is formed by the sequential proteolytic cleavage of the amyloid precursor protein (APP), by β- and γ-secretases, via what is referred to as the amyloidogenic pathway and are commonly found as soluble oligomers and fibrillar plaques [
3]. A considerable body of evidence suggests that soluble Aβ oligomers are the predominant neurotoxic species of Aβ, with the Aβ42 fragment being found to be particularly potent [
4,
5]. Aβ42 oligomers exert their toxic effects by binding to neuronal synapses, causing disruption to normal synaptic signaling, which often leads to neuronal death [
6,
7]. However, the precise receptors to which Aβ42 binds remain very controversial.
Glutamate is the primary excitatory neurotransmitter in the brain and activates both ionotropic glutamate receptors and G protein-coupled metabotropic glutamate receptors (mGluRs) [
8]. Alterations in normal mGluR5 signaling is associated with the autism spectrum disorder, fragile X syndrome, and a number of neurodegenerative diseases that include: Huntington’s disease, Parkinson’s disease and AD [
9‐
17]. mGluR5 couples to the heterotrimeric G protein Gα
q/11, which activates phospholipase C resulting in increased inositol-1,4,5-triphosphate formation and the release of Ca
2+ from intracellular stores [
3,
12]. Recently, it has been demonstrated that mGluR5 also serves as the receptor for both cellular prion protein (PrP
c) and Aβ42, which activate mGluR5 to release Ca
2+ from intracellular stores [
17‐
20].
It is now well recognized that mGluR5 regulates the activity of the fragile X mental retardation protein (FMRP), a RNA binding protein that functions to repress protein synthesis at synapses [
21‐
25]. The activation of mGluR5 results in FMRP-dependent increases in both APP and FMRP expression [
16,
26]. This increase in APP expression results in the augmented secretion of both toxic Aβ42 oligomers and the non-toxic soluble APP (sAPPα) fragment [
16,
24]. Thus, although the precise mechanism remains unclear, this data suggests that, in addition to an established role in fragile X syndrome, mGluR5 and FMRP likely contribute directly to AD pathogenesis [
16,
27].
In the present study, we assess the role for mGluR5 in AD pathogenesis by crossing the APPswe/PS1∆E9 double transgenic mouse model of AD with mGluR5 null mice. We find that the genetic deletion of mGluR5 reverses the spatial memory deficits observed in APPswe/PS1∆E9 mice. FMRP protein expression is also increased APPswe/PS1∆E9 mice and following the genetic deletion of mGluR5 FMRP expression is reduced to wild-type levels and results in the significant decrease of aggregated β-amyloid levels and β-amyloid plaque number in APPswe/PS1∆E9 mice. These studies provide evidence for a central role of mGluR5 in AD pathogenic mechanisms that are associated with increased Aβ.
Discussion
The neurodegeneration characteristic of AD is associated with Aβ oligomers, which bind to neuronal synapses to both alter synaptic signaling and promote synaptic loss [
35‐
37]
. We show here, that mGluR5 signaling in double transgenic APPswe/PS1∆E9 mice not only contributes to impaired spatial learning, but also increases the formation of Aβ oligomers and β-amyloid plaques which are the pathophysiological hallmarks of AD. In addition, spatial learning deficits observed in double transgenic APPswe/PS1∆E9 mice are ameliorated by the genetic deletion of mGluR5, which is consistent with the recent report that pharmacological blockade of mGluR5 with the mGluR5-specific antagonist MTEP improved behavioural deficits in APPswe/PS1∆E9 mice [
17]. However, we unexpectedly found that mGluR5 signaling may also contribute to the etiology of AD pathology by promoting the formation of Aβ oligomers in APPswe/PS1∆E9 mice, which are considered to be important mediators of neurotoxic signaling at glutamatergic synapses. It is also now recognized that mGluR5 functions as a receptor for both Aβ oligomers [
17‐
19], and PrP
c to mediate Aβ oligomer-dependent release of Ca
2+ from intracellular stores, an event which may be neurotoxic [
17]. Aβ oligomers also mediate clustering of mGluR5 at synapses, the consequence of which may contribute to the functional impairment of synaptic activity [
18]. Our data suggest that Aβ oligomer interactions with mGluR5 may also function to accelerate Aβ production via the mGluR5- and FMRP-dependent activation of the amyloidogenic pathway. Thus, it appears that in the double transgenic APPswe/PS1∆E9 mouse model of AD, Aβ oligomers initiate a neurodegenerative positive feedback loop that increases mGluR5 activity leading to an increase in Aβ oligomer formation that culminates in AD-like pathology and impaired spatial learning. The genetic deletion of mGluR5 in APPswe/PS1∆E9 mice effectively severs this feedback loop, reducing Aβ oligomer and plaque formation resulting in improved cognitive performance in APPswe/PS1∆E9 mice.
We found that there was no change in swim speed the mouse groups tested suggesting that altered locomotion or anxiety did not account for changes in MWM performance, in the data presented to illustrate learning we showed both path length and escape latency. While an expedient escape latency could be attributed to hyperactivity, the fact that there was no difference in swim speed, and no significant difference in path length, this indicates that the mGluR5 knockout mice learn the task, as well as the wild type animals. We also found no changes in anxiety behaviour in the elevated plus maze. Our results were completely consistent with recent data showing mGluR5 antagonism improved APPswe/PS1∆E9 mouse behaviour in MWM [
17]. The previously mGluR5 knockout deficit in the MWM reported by Lu et al. [
38] was in mice of undocumented age and did not involve the comparison of aged mice.
The activation of mGluR5 triggers the mRNA binding protein FMRP to release APP repression, promoting its translation [
20,
23,
24]. In fragile X syndrome, this interaction leads to increased cleavage of APP [
19,
22], suggesting that similar processes may be occurring in AD brain. APP cleavage occurs via a non-amyloidogenic pathway and an amyloidogenic pathway, which is considered to function as an “overflow” pathway activated as a consequence of excessive APP production resulting in increased APP cleavage and the formation of toxic Aβ oligomers [
3,
39]. Thus, Aβ mediated activation of mGluR5 may result in both increased FMRP expression and APP processing. We propose that the observed increase in FMRP protein expression in APPswe/PS1∆E9 mice represents a compensatory mechanism that is engaged to potentially reduce ectopic/transgenic translational control of APP translation. However, since mTOR activity is also increased in APPswe/PS1∆E9 mice increased FMRP expression levels may be ineffective in reducing APP translation. There is evidence that G protein-coupled receptors regulate γ-secretase activity in a β-arrestin-dependent manner to alter Aβ production [
40]. However, the link between mGluR5 signaling and β-arrestin activation is controversial [
41]. As a consequence, Aβ oligomer-stimulated mGluR5 activity may function to promote the cleavage of APP via the amyloidogenic pathway.
The degeneration of glutamatergic synapses is associated with disrupted neuronal signaling and synaptic loss has traditionally been associated with alterations in ionotropic NMDA and AMPA receptor signaling [
3,
42‐
44]. However, the pharmacological blockade of the NMDAR has not met with success for the treatment of neurodegenerative disease [
38]. The recent identification of mGluR5 as a potential disease modifier in AD provides a potential for novel mGluR5 antagonist-based pharmacological interventions for the treatment of AD. mGluR5 blockade was effective in improving spatial learning in APPswe/PS1∆E9 mice and we reported that mutant huntingtin protein aggregates are significantly reduced in a mutant Huntingtin knock-in model that lacks mGluR5 expression [
12]. Taken together these observations suggest a generalized potential for mGluR5 antagonists in the treatment of neurodegenerative disease.
In summary, we show that mGluR5, which functions as a membrane bound receptor scaffolds for both PrPc and Aβ oligomers [
17‐
19], functions to transduce Aβ oligomer-mediated signals as alterations in intracellular Ca
2+ signaling, as well as FMRP/Aβ mediated pathological signaling in the APPswe/PS1∆E9 mouse model of AD. Although these studies provide additional evidence establishing mGluR5 as a potentially effective drug target to antagonize AD progression, additional pharmacological studies will be required in non-human primates to validate the utility of mGluR5 antagonists for the treatment of AD patients.
Materials and methods
Materials
Sulfo-NHS-SS-biotin and Neutravidin were purchased from Thermo Scientific (Waltham, MA, USA. Mouse anti-FMRP antibody was purchased from Abcam (Cambridge, UK). DHPG was purchased from Tocris Bioscience (Bristol, UK). Myo-[3H]-inositol was purchased from Perkin Elmer (Waltham, MA, USA). Dowex I-X8 resin, rabbit anti APP antibody, rabbit anti-phospho mTOR (pSer2481) antibody and rabbit anti-mTOR antibody were purchased from Sigma-Aldrich (St. Louis, MO, USA).Goat anti-rabbit secondary antibody was purchased from BioRad (Hercules, CA, USA). Aggregated Aβ human ELISA kit and rabbit Aβ peptide antibody were purchased from Invitrogen (Carlsbad, CA, USA). Rabbit anti-mGluR5 was purchased from Millipore (Billerica, MA, USA). Finally Vector Elite ABC kit rabbit and mouse, and Vector SG substrate was purchased from Vector laboratories. All other biochemical reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA).
Mouse model
STOCK B6C3-Tg (APPswe/PSEN1∆E9)85Dbo/J mice that carry the human APP with Swedish mutation and the DeltaE9 mutation of the human presenilin 1 gene and mGluR5 knockout mice B6;129-Grm5
tm1Rod/J (
mGluR5
-/-
) were purchased from Jackson Laboratory (Bar Harbor, ME) [
38,
45]. APPswe/PS1∆E9/mGluR5
-/- mice were generated by crossing APPswe/PS1∆E9 mice with a C57/Bl6 background, with mGluR5-/- C57/Bl6 females. Offspring were tail snipped and genotyped using PCR with primers specific for the APP sequence and primers specific for mGluR5. The following genotypes were generated and used for experiments APPswe/PS1ΔE9/mGluR5
-/-, APPswe/PS1ΔE9/mGluR5
+/+, APPswe/PS1ΔE9
-/-/mGluR5
-/- and wild type littermate controls. Animals were housed in an animal care facility in cages of 2 or more animals, and were maintained on a 12 hour light/12 hour dark cycle at 24°C. Mice received food and water
ad libitum. Mice were aged to 9 and 12 months before use in a battery of behavioural and biochemical experiments. All animal experiments were conducted in accordance with the University of Western Ontario animal care committee.
Morris water maze
Any Maze software connected to a video camera was used to track the activity of animals within the maze. The maze was a white opaque plastic pool (120 cm in diameter), filled with water and maintained at 25°C to prevent hypothermia. A clear escape platform (10 cm diameter) was placed 25 cm from the perimeter, hidden 1 cm beneath the surface of the water. Visual cues were placed on the walls in the room with the maze. Acquisition phase (day 1–4): Mice were trained over 4 consecutive days, with 4 trials per day with 15 min intervals between trials. Mice were randomly started from 4 equally spaced points around the pool, across each of the 4 daily trials. Animals were given 90 s to find the escape platform, if they failed to do so they were guided to the platform where they remained for 30 s, before being returned to their home cage. Swim speed and escape latency were recorded and statistical analysis performed using Graph Pad Prism. Probe trial (day 5): the probe trial is performed to assess spatial memory. This is a singular 60 s trial, in which the platform is removed and mice are allowed to swim freely in the pool. Time spent in the target quadrant was recorded and statistical analysis performed using Graph Pad Prism software.
Open field
VersaMax animal activity monitors (AccuScan Instruments Inc, Columbus, OH, USA) were used to measure locomotor activity. Experiments were performed during the light cycle of the mice. Mice were allowed to explore open field boxes (20 × 20 cm Plexi-glass boxes) for 120 min during which time movement was measured at 5 min intervals using beam breaks converted to cm. Measurements of total distance travelled, center time and number of rearings were calculated and statistical analysis performed using Graph Pad Prism software.
Elevated plus maze
Mice were individually placed in one of the enclosed arms of a plus-maze and the behavior of the animals was recorded over a 5 min period by a ceiling-mounted video camera. The amount of time spent in each of the arms, the number of arm entries (an arm entry was defined as all four of the paws being placed in an arm of the plus-maze). The elevated plus-maze had two arms enclosed by 21 cm high walls; whereas the remaining two arms were open (arms were 24.8 cm long × 7.7 cm wide). The maze was situated in a dimly lit room, such that the closed arms were darkened, whereas open arms were somewhat illuminated. All behavioural experiments were blinded.
Determination of β amyloid oligomer concentration by sandwich ELISA
Brains were dissected into right and left hemisphere, with one hemisphere used to analyse oligomeric Aβ. Quantification of Aβ oligomers, in 12 month old fresh mouse brains, was performed using a sandwich ELISA kit (KHB3491, Invitrogen) according to manufacturer’s instructions. Briefly, wild-type, APPswe/PS1ΔE9/mGluR5-/-, APPswe/PS1ΔE9/mGluR5+/+, APPswe/PS1ΔE9-/-/mGluR5-/- mouse brains were homogenized in Tris Buffered Saline (25 mM Tris–HCl, pH 7.4; 150 mM NaCl) supplemented with protease inhibitor cocktail (1 mM AEBSF, and10 μg/ml aprotinin of both leupeptin). Brain homogenates were centrifuged at 100,000 × g at 4°C for 1 hour. The supernatant was then diluted 1:10 before carrying out the ELISA, which was performed in triplicate, measuring only Aβ oligomers, as detailed in the manufacturer's protocol. Protein was quantified using the Bradford protein assay (BioRad). The final Aβ values were determined following normalization to total protein levels.
β amyloid immunohistochemistry
Brains were dissected into left and right hemisphere, with one hemisphere used for histology. Brains were coronally sectioned to include both the cortex and hippocampus. Immunohistochemistry was performed on 40 μm free floating sections using a peroxidase based immunostaining protocol. In brief, endogenous peroxidase activity was quenched using 0.1% hydrogen peroxide, after which the membranes were permeabilized using 1% triton X100 in TBS. Non-specific binding was blocked using 1.5% normal goat serum, followed by incubation in primary antibody for Aβ (1:200, rabbit polyclonal, Invitrogen) overnight at 4°C. Sections were washed in 1xTBS, and then incubated in secondary antibody (biotinylated goat anti-rabbit, 1:400, Vector Elite ABC kit rabbit, Vector Laboratories) for 90 min at 4°C. Finally sections were incubated in an avidin biotin enzyme reagent (Vector Elite ABC kit rabbit, Vector Laboratories). Immunostaining was visualised using a chromogen (Vector SG substrate, Vector Laboratories). Sections were mounted on slides and visualized using a Zeiss LSM-510 META multiphoton laser scanning microscope with a Zeiss 10× lens, representative 900 μm × 900 μm areas of cortex (4 regions of interest) and hippocampus (2 regions of interest) were imaged for analysis. The number of Aβ positive puncta per image was counted using the cell counter tool in Image J (NIH, USA).
Cell surface biotinylation
Cell surface biotinylation was performed as previously described11. 350 μm coronal brain slices from wild-type, APPswe/PS1ΔE9/mGluR5-/-, APPswe/PS1ΔE9/mGluR5+/+, APPswe/PS1ΔE9-/-/mGluR5-/-mice were prepared using a vibratome system (Leica). Slices were recovered in KREBS buffer (127 mM NaCl, 2 mM KCl, 10 mM glucose, 1.2 mM KH2PO4, 26 mM NaH2CO3, 1 mM MgSO4, 1 mM CaCl2, pH 7.4) continuously gassed with 95%O2/5%CO2 for 30 min at 37°C. Slices were transferred to tubes and biotinylated for 1 hr in 1.5 mg/ml sulfo-NHS-SS-biotin (Thermo Scientific) for 1 h on ice. Slices were washed and biotinylation quenched with 100 μM glycine in HBSS for 30 min on ice. Following washes in HBSS, tissue was lysed in RIPA buffer (0.15 M NaCl, 0.05 M Tris–HCl, pH 7.2, 0.05 M EDTA, 1% Nonidet P40, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS) containing protease inhibitors (1 mM AEBSF and 10 μg/ml of both leupeptin and aprotinin), lysates were then polytroned until homogenous. Biotinylated proteins were then precipitated on NeutrAvidin beads using equivalent amounts of cellular protein for each sample. Biotinylated proteins were subjected to SDS-PAGE and immunoblotted with Rabbit polyclonal mGluR5 antibody (1:1000, dilution), as described below.
Immunoblotting
Acute brain slices prepared, as described above, and recovered for 90 min at 37°C in KREBS continuously gassed with 95%O2/5%CO2. Slices were then transferred to tubes, gassed with 95%O2/5%CO2 before being capped tightly and left to further recover at 37°C for 30 min. Slices were flash frozen with 95% ethanol in a bath of dry ice and lysed in RIPA buffer. 100 μg of total cellular protein for each sample were subjected to SDS-PAGE, followed by electroblotting onto nitrocellulose membranes. Membranes were blocked with 10% milk in wash buffer (150 mM NaCl, 10 mM Tris–HCl, pH 7.0, and 0.05% Tween 20) for 1 h and then incubated with rabbit anti-APP (1:4000), rabbit anti-p-mTOR (pSer2481, 1:1000), rabbit anti-mTOR (1:1000), mouse anti-FMRP (1:1000) or rabbit anti-actin (1:10000) antibodies in wash buffer containing 3% milk overnight. Membranes were rinsed three times with wash buffer and then incubated with secondary horseradish peroxidase-conjugated goat anti-rabbit or anti-mouse IgG diluted 1:10000 in wash buffer containing 3% skim milk for 1 h.
[3H]Inositol phosphate formation in cortical slices was performed, as previously described11. Dissected cortices from age-matched wild-type, APPswe/PS1ΔE9/mGluR5-/-, APPswe/PS1ΔE9/mGluR5+/+, APPswe/PS1ΔE9-/-/mGluR5-/- mouse brains were cross-chopped using a McIlwain tissue chopper. The cortical prisms were recovered by incubation in KREBS buffer after being gassed with 95%O2/5%CO2 in a shaking water bath at 37°C for 30 min. 40 μl of gravity-packed prisms were transferred in duplicate to tubes, gassed, and incubated with 1 μCi/ml myo-[3H]inositol in a shaking bath at 37°C for 90 min. The prisms were gassed again and incubated in a shaking water bath at 37°C for 15 min in 10 mM LiCl followed by stimulation with 5 μM or 50 μM DHPG for 20 min. The reaction was stopped by adding 900 μl of a 2:1 mix of chloroform:methanol incubated for 15 min at room temperature, followed by adding 300 μl of chloroform. The total [3H]inositol incorporated into slices was determined by counting the radioactivity present in the hydrophobic layer. 700 μl of aqueous layer was added to Dowex 1-X8 (formate form) 200–400 mesh anion exchange resin in columns. Columns were washed three times with water followed by 2 washes with 60 mM ammonium formate. Samples were eluted with 200 mM ammonium formate and 0.1 M formic acid into scintillation vials containing scintillation fluid. [3H]Inositol phosphate formation was determined by liquid scintillation using a Beckman LS 6500 scintillation system.
p-TOR and mTOR immunohistochemistry
Brains were coronally sectioned to include both the cortex and hippocampus. Immunohistochemistry was performed on 40 μm free floating sections using a peroxidase based immunostaining protocol. In brief; endogenous peroxidase activity was quenched using 0.1% hydrogen peroxide, after which the membranes were permeabilized using 1% triton X100 in TBS. Non-specific binding was blocked using 1.5% normal goat serum, followed by incubation in primary antibody for p-mTOR (1:200 diltion, rabbit polyclonal pSer2481 antibody) and mTOR (1:200, rabbit polyclonal antibody) overnight at 4°C. Sections were washed in 1xTBS, and then incubated in secondary antibody (biotinylated goat anti-rabbit, 1:400, Vector Elite ABC kit rabbit, Vector Laboratories) for 90 min at 4°C. Finally sections were incubated in an avidin biotin enzyme reagent (Vector Elite ABC kit rabbit, Vector Laboratories). Immunostaining was visualised using a chromogen (Vector SG substrate, Vector Laboratories). Sections were mounted on slides and visualized using a Zeiss LSM-510 META multiphoton laser scanning microscope with a Zeiss 10× lens, representative 900 μm × 900 μm areas of cortex (4 regions of interest) and hippocampus (2 regions of interest) were imaged for analysis. The number of p-mTOR and mTOR positive cells per image was counted using the cell counter tool in Image J (NIH, USA).
Data analysis
Means ± SEM are shown for the number of independent experiments indicated in Figure Legends. GraphPad Prism software was used to analyze data for statistical significance and for curve fitting. Statistical significance was determined by analysis of variance (ANOVA) followed by post-hoc testing.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contribution
AH: performed the behavioural experiments, beta amyloid analysis, immunohistochemistry, and biochemistry, data analysis and writing of the manuscript. JLE: performed cell surface biotinylation and IP experiments. RAD: assisted with a number of the behaviour experiments. SPC: contributed to the writing of the manuscript. SSGF: Principle investigator, contributed to the experimental design, data analysis and writing of the manuscript. All authors read and approved the final manuscript.