Abstract
There is a significantly elevated incidence of epilepsy in Alzheimer's disease (AD). Moreover, there is neural hyperexcitation/synchronization in transgenic mice expressing abnormal levels or forms of amyloid precursor protein and its presumed, etiopathogenic product, amyloid-β1–42 (Aβ). However, the underlying mechanisms of how Aβ causes neuronal hyperexcitation remain unclear. Here, we report that exposure to pathologically relevant levels of Aβ induces Aβ form-dependent, concentration-dependent, and time-dependent neuronal hyperexcitation in primary cultures of mouse hippocampal neurons. Similarly, Aβ exposure increases levels of nicotinic acetylcholine receptor (nAChR) α7 subunit protein on the cell surface and α7-nAChR function, but not α7 subunit mRNA, suggesting post-translational upregulation of functional α7-nAChRs. These effects are prevented upon coexposure to brefeldin A, an inhibitor of endoplasmic reticulum-to-Golgi protein transport, consistent with an effect on trafficking of α7 subunits and assembled α7-nAChRs to the cell surface. Aβ exposure-induced α7-nAChR functional upregulation occurs before there is expression of neuronal hyperexcitation. Pharmacological inhibition using an α7-nAChR antagonist or genetic deletion of nAChR α7 subunits prevents induction and expression of neuronal hyperexcitation. Collectively, these results, confirmed in studies using slice cultures, indicate that functional activity and perhaps functional upregulation of α7-nAChRs are necessary for production of Aβ-induced neuronal hyperexcitation and possibly AD pathogenesis. This novel mechanism involving α7-nAChRs in mediation of Aβ effects provides potentially new therapeutic targets for treatment of AD.
Introduction
One hallmark of Alzheimer's disease (AD) is the deposition of amyloid-β (Aβ) plaques, which is considered to be a result of aberrant amyloid precursor protein (APP) processing and elevated Aβ production (Goedert and Spillantini, 2006). It still is unclear how Aβ deposition contributes to neuronal damage, but exposure to pathologically relevant levels of Aβ induces hyperexcitation in individual neurons and neural circuits (Del Vecchio et al., 2004). In AD patients, the incidence of epilepsy (10–22%) is significantly higher than that in age-matched, non-AD controls (<1%) (Amatniek et al., 2006; Hommet et al., 2008). In animals that overexpress Aβ at pathologically relevant levels, epileptiform activity has been noted in the entorhinal–hippocampal circuit (Palop et al., 2007), and there is a reduction in seizure threshold (Westmark et al., 2008). There are increased Ca2+ transients in neurons near Aβ plaques (Busche et al., 2008). These findings link Aβ deposition to neuronal hyperexcitation and aberrant epileptiform activity, which, when combined with responses to this activity, could cause synaptic impairment and cognitive deficits relevant to AD pathogenesis (Leonard and McNamara, 2007; Palop and Mucke, 2009). However, entities that mediate Aβ-induced neuronal hyperexcitation are largely unknown.
Nicotinic acetylcholine receptors containing α7 subunits (α7-nAChRs) have been linked to Aβ deposition and AD pathogenesis. Acute exposure to Aβ alters α7-nAChR function (Liu et al., 2001, 2009; Pettit et al., 2001; Dineley et al., 2002; Lamb et al., 2005). However, there is enhanced expression of α7-nAChRs in both AD patients and AD model animals (Hellström-Lindahl et al., 1999; Counts et al., 2007; Ikonomovic et al., 2009), even in glial cells (Xiu et al., 2005; Yu et al., 2005), and α7-nAChR function is not reduced in adult (7-month-old) APP transgenic mice (Spencer et al., 2006). Thus, longer-term exposure to Aβ may enhance rather than reduce α7-nAChR expression and function in both AD patients and animal models, which is possibly relevant to AD progression.
α7-nAChRs play important roles in regulating neuronal excitability (Dani, 2000; McKay et al., 2007). α7-nAChRs have high Ca2+ permeability (Bertrand et al., 1992), and their activation elevates presynaptic glutamate release (Dani, 2000; McKay et al., 2007), both of which could contribute to neuronal hyperexcitation. In addition, animals with gain-of-function α7-nAChR mutations exhibit significantly higher susceptibility to seizures (Broide et al., 2002). Disrupting interactions of Aβ with α7-nAChRs is reported to diminish Aβ toxicity (Wang et al., 2009), and genetic deletion of nAChR α7 subunits significantly rescues synaptic impairment and learning/memory deficits in APP mice (Dziewczapolski et al., 2009). Collectively, this evidence suggests the hypothesis that chronic exposure to pathologically relevant levels of Aβ upregulates α7-nAChR function, which then contributes to neuronal hyperexcitation.
In this study, we test this hypothesis using multiple approaches with mouse primary neuronal and slice cultures. We find that exposure to nanomolar concentrations of Aβ first induces increased expression and function of α7-nAChRs, and then induces neuronal hyperexcitation.
Materials and Methods
Primary cell culture.
The protocol for preparation of primary neuronal cultures was approved by the Institutional Animal Care and Use Committee of the Barrow Neurological Institute and St. Joseph's Hospital and Medical Center. Wild-type (WT) C57/BL, nAChR α7 subunit knock-out (KO), and glutamic acid decarboxylase-green fluorescent protein (GAD-GFP) mice (provided by Dr. Scott C. Steffensen, Brigham Young University) were used in this study. As we previously described (He et al., 2012), the day before culture, poly-d-lysine 0.02% solution was added to culture dishes. Dishes were swirled to make sure the entire bottom was coated, and then dishes were left in a 37°C, 5% CO2 incubator overnight. The next day, dishes were washed three times with sterile water and left in the incubator after the final wash. One-to-three-day-old mice (of either gender) were killed, and hippocampal CA1-CA3 regions were dissected under a stereo microscope. Tissue was minced with scissors in ice-cold Neurobasal medium (Invitrogen) and then digested with Papain (20 U/mg; Worthington) at 30°C for 20 min in tubes shaken at 120 rpm in a water bath shaker. After enzyme digestion, the reaction was stopped by adding inactivated fetal bovine serum into the medium. Then, digested tissue was filtered and transferred into 15 ml tubes. After trituration, tissue was centrifuged at 1500 rpm for 3 min to form pellets containing dissociated cells, and the supernatant was removed and replaced with Neurobasal medium, which was used to resuspend the pellets. This process was repeated three times. After the final centrifugation, the supernatant was replaced with Neurobasal medium containing 0.03 mm choline chloride and supplemented with 0.5% (w/v) l-glutamine and 2% B27 serum-free supplement. Cells were resuspended, counted based on Trypan blue exclusion, and plated at a density of 4.5–5.0 × 104 cells per cm2 in 35 mm culture dishes. Cells were kept within the 37°C, 5% CO2 incubator for future use.
Organotypic slice cultures.
Organotypic slice cultures were prepared from newborn WT C57/BL mouse pups as previously described (Stoppini et al., 1991; De Simoni and Yu, 2006). Briefly, brains were removed from killed, 0–1-d-old mice and placed in minimal essential medium (Invitrogen). Brain tissues, including cortex and hippocampus, were dissected, and coronal slices (300 μm) were generated using a vibratome (Vibratome), placed onto Millicell-CM culture membranes (Millipore), and transferred to six-well plates filled with 1.2 ml of culture medium [per 100 ml: 50 ml of Basal Medium Eagle, 25 ml of EBSS, 23 ml of horse serum, 0.5 ml of Glutamax (200 mm stock), 0.1 ml of gentamicin (10 mg/ml stock), 0.4 ml of fungizone (250 μg/ml), and 1.45 ml of d-glucose (45%; 42 mm total)]. Cultures were then incubated in humidified, 5% CO2 at 36–37°C, with medium changes every 2 d. After 7 d in vitro, media was replaced with 1.2 ml of serum-free media [per ∼100 ml: 97 ml of Neurobasal, 2 ml of B27, 0.5 ml of Glutamax (200 mm), 0.1 ml of ascorbic acid (0.5 m), and 0.68 ml of d-glucose (45%; 42 mm total)].
Electrophysiology.
The standard external solution contained 140 mm NaCl, 3 mm KCl, 2 mm MgCl2, 2 mm CaCl2, 10 mm d-glucose, 10 mm HEPES, pH 7.4 with Tris base. The pipette solution for current-clamp recordings used to measure intrinsic excitability contained 130 mm KMeSO4, 10 mm KCl, 10 mm HEPES/K-HEPES, 2 mm MgSO4, 0.5 mm EGTA, and 3 mm ATP, pH 7.3 with KOH. For miniature EPSC (mEPSC) recordings, 0.3 μm TTX, 50 μm picrotoxin, and 50 μm APV were added to the external solution to block voltage-gated sodium channels, GABAA receptors, and glutamate receptors. For whole-cell recording of ligand-gated ion channel activity and for mEPSC recordings, the pipette solution contained 140 mm potassium gluconate, 5 mm KCl, 10 mm HEPES, 0.2 mm EGTA, 2 mm MgCl2, 4 mm MgATP, 0.3 mm Na2GTP, and 10 mm Na2-phosphocreatine, pH 7.3 with KOH. Current-clamp recordings to measure neuronal excitability were performed using subtle modifications of a previously described protocol (Desai et al., 1999; Karmarkar and Buonomano, 2006). Recordings were considered acceptable only if cells exhibited overshooting action potentials and their resting membrane potential did not vary by >5 mV. The input resistance was required to be >80 MΩ and to vary <15% during recordings. Neuronal intrinsic excitability was measured as the number of spikes in response to a series of fixed, 2 or 4 s ms current injection steps (0, 30, 60, 90, 120, 150, or 180 pA). In whole-cell recordings, peak currents and cell membrane capacitance as an indication of membrane area were used to calculate current densities. Data were acquired at 2 kHz using an Axopatch 200B amplifier and pClamp 9.2 software (Molecular Devices) and analyzed using clampfit 9.2 (Molecular Devices) and Mini Analysis (Synaptosoft) software as previously described (Liu et al., 2009).
Drugs, Aβ preparation, and Aβ treatment.
Drugs used in this study were choline (Sigma-Aldrich), (−)nicotine, methyllycaconitine (MLA), dihydro-β-erythroidine (DHβE) (Sigma-Aldrich), AMPA (Tocris Bioscience), NMDA (Tocris Bioscience), GABA (Sigma-Aldrich), TTX (Tocris Bioscience), brefeldin A (BFA; Calbiochem), scrambled Aβ1-42, and authentic Aβ1-42 (Aβ; rPeptide). Whenever the abbreviation Aβ is used in the text (except for the introduction) we refer to Aβ1-42, and we specify the form when relevant. Aβ peptides were dissolved in 1,1,1,3,3,3-hexafluoro-2-propanol at a concentration of 1 mg/ml in 250 μl of aliquots, air dried, and stored at −20°C. Aliquots were resuspended in dimethyl-sulfoxide and further diluted in PBS, pH 7.4, to a final concentration of 100 μm and incubated at 37°C without shaking. Oligomeric aggregates of Aβ were obtained after incubation for 6 h. For fibril formation, Aβ was incubated at 37°C for at least 96 h. Both oligomeric and fibrillar forms were verified by atomic force microscopy (AFM) before being added into culture medium. Primary cultured neurons were maintained at 37°C in a humidified atmosphere of 5% CO2/95% air for 14 d before Aβ treatments, which involved renewing Aβ-containing culture medium (100 nm) daily for 0–10 d. The same renewal procedure with medium containing scrambled Aβ1-42 (containing the same amino acids, but in random order) was performed as a control. For pharmacological studies, nAChR antagonists (MLA or DHβE) were applied chronically along with Aβ for 7 d, or there was subchronic (10 min) exposure to MLA or DHβE after a prescribed period of Aβ exposure before testing for neuronal hyperexcitation or mEPSC studies.
Electron microscopy.
The aggregation forms of Aβ used in this study were determined in part by negative-staining electron microscopy (EM). Immediately before use in slice or cell culture studies, 2 μl of aliquots from 100 μm stock solutions were adsorbed for 1 min to the surface of plastic-coated, carbon-backed, 300-mesh copper grids, which were blotted to remove excess liquid and covered with 5 μl of 2% aqueous uranyl acetate for 2 min. Then, excess liquid was removed, and samples were allowed to air dry and then stored. Subsequently, samples were viewed at 34,000–175,000× magnification using a Philips CM12 electron microscope (Fig. 1C), and images were collected with an AMT XR40 digital camera (Advanced Microscopy Techniques). Sizes were calibrated by comparison to sizes of catalase crystal standards (Ted Pella).
AFM imaging.
AFM also was used to monitor the morphology of Aβ aggregates before experiments. Aliquots were removed from Aβ stock samples and immediately spotted on freshly cleaved mica. After 2 min, the mica was washed with 1 ml of deionized water, and then dried with compressed nitrogen for storage. Subsequently, topographic AFM images were obtained in air at room temperature using a tapping mode AFM with a Nanoscope IIIa controller (Veeco). Images were acquired using oxide-sharpened Si3N4 AFM tips (k = 40 N/m; fo, ∼300 kHz) (OTESPA, Veeco) at scan rates of 2–3 Hz and at scan resolution of 512 samples per line (Fig. 1D). Images were subjected to second-order polynomial flattening as needed to reduce the effects of image bowing and tilt. AFM images were analyzed with Scanning Probe Imaging Processor software (Image Metrology, www.imagemet.com) to generate height distribution histograms for each sample.
Immunostaining.
After whole-cell current recording sessions had ended, cells from hippocampal neuronal cultures were injected with biocytin (5 mg/ml, included in recording electrodes and infused into the intracellular space by converting to true, whole-cell recording mode) while still in their 35 mm culture dishes. Cells then were fixed in 4% paraformaldehyde for 10 min and washed at room temperature with PBS three times for 5 min each. Then, a PBS-based blocking solution containing 5% normal goat serum and 0.3% Triton X-100 was applied for 1 h. After incubation at 4°C overnight with primary, goat anti-neurogranin antibody (1:100 dilution; Santa Cruz Biotechnology), cultures were washed with PBS three times for 5 min each. Thereafter, Avidin-AF488 (1:1000 dilution; Invitrogen) was added to fluorescently label biocytin-containing cells subject to recording. Alexa 555-conjugated anti-goat secondary IgG (Invitrogen) was applied in blocking solution for 2 h at room temperature to interact with immobilized, anti-neurogranin antibodies to confirm identity of pyramidal neurons (Fig. 1B). The cultures were then finally washed three times for 5 min each with PBS and prepared for fluorescence microscopy.
Affinity isolation of α7-nAChRs.
To capture α7-nAChRs expressed on cell surfaces of cultured, primary hippocampal cells, we used a modification of a previously described protocol (Charpantier et al., 2005). Cells were maintained for 30 min in the culture incubator in media containing or in the absence of 10 μm nonbiotinylated α-bungarotoxin (α-Bgt). Thereafter, 300 nm biotinylated α-Bgt was added to the medium for 45 min. Cells were then washed three times with PBS and lysed as described previously (Charpantier et al., 2005) to solubilize membrane-associated proteins. Streptavidin-agarose beads were added to precipitate surface α7-nAChRs bound to biotinylated α-Bgt. Total (from the cell surface and from intracellular pools) α7-nAChR assemblies were isolated by adding 300 nm biotinylated α-Bgt alone or in the presence of 10 μm unlabeled α-Bgt to separately processed cell lysates, followed by reaction with streptavidin-agarose beads. Beads were pelleted by centrifugation at 14,000 × g, and the supernatants from samples were transferred to new tubes. The BCA protein assay (Pierce) was used to determine protein concentrations in cell lysates. Affinity isolated samples were resolved according to mass by SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose membranes (Bio-Rad), and immunoblotted using a goat anti-α7 subunit antibody (1:100; H302, Santa Cruz Biotechnology) overnight at 4°C. After membranes were washed, horse radish peroxidase-conjugated anti-goat secondary antibody (Santa Cruz Biotechnology) was added and incubated for 1 h at room temperature. α7 Subunit-like immunoreactive material was then visualized after reaction using an enhanced chemiluminescence kit (GE Healthcare BioSciences) and quantified using ImageJ software 1.45s (National Institutes of Health, Bethesda, MD).
Quantitative real-time PCR.
Total RNA from primary neuronal cultures was extracted using column purification (Qiagen). First-strand cDNA was generated with random primers using iScript cDNA synthesis kits (Bio-Rad). Real-time analysis was performed on a Bio-Rad ICycler system (Bio-Rad) using the following conditions: an initial denaturation for 10 min at 95°C, followed by 40 cycles of 15 s of denaturation at 95°C, 30 s of annealing at 60°C, and 1 min extension at 72°C. Analysis was performed according to the ΔΔCt method using GAPDH as the housekeeping gene standard. The primer sequences are as follows: ACHA7, 5′-TTGATGCCACGTTCCACACCAATG-3′ and 5′-AGGGAACCAGCGAACGTCAATGTA-3′; ACHB2, 5′-ACAATGCTGACGGCATGTACGA-3′ and 5′-TGCTGGTCAAATGGGAAGTGCT-3′; and GAPDH, 5′-TGACTCTACCCACGGCAAGTTCAA-3′ and 5′-ACGACATACTCAGCACCAGCATCA-3′.
Data analysis and statistics.
Data are given as mean ± SEM with numbers of determinations (n). A probability level of p < 0.05 was considered to be statistically significant based on two-tailed Student's t tests, one-way or two-way ANOVA followed by appropriate tests.
Results
Identification of cell phenotype
Pyramidal cells in mouse primary hippocampal cultures were initially and tentatively identified based on their morphology, and their identities were then validated by characterization based on their electrophysiological features [Fig. 1A; slower action potential (AP) firing rates (Fig. 1Aa) than those seen for nonpyramidal neurons (Fig. 1Ab) upon injection of 60 pA currents] (Alkondon and Albuquerque, 1995). Recorded neurons were positively identified by postrecording injection of biocytin and reaction with fluorescent avidin (Fig. 1Ba,Bb) and had identities subsequently confirmed by immunostaining following patch-clamp recording as either neurogranin-positive pyramidal neurons (Fig. 1Ba′) or neurogranin-negative nonpyramidal neurons (Fig. 1Bb′).
Identification of neuronal phenotypes of cultured hippocampal neurons and determination of Aβ forms. A, AP firing rates and their responses to injection of 60 pA of current were determined using current-clamp recording and used initially to distinguish pyramidal-like (Aa) from nonpyramidal-like (Ab) neurons in primary hippocampal cultures. The scale bars indicate a 0.5 s interval (horizontal) or a 10 mV membrane potential change (vertical). B, The recorded neurons in A were further labeled by postrecording injection of biocytin and subsequent reaction with streptoavidin (Ba, Bb) but distinguished based on neurogranin (Ba', Bb') immunostaining (calibration bar, 20 μm). C, D, EM (C) or AFM (D) were used to monitor aggregation forms of Aβ made as described in Materials and Methods and then used in this study. Preparations of Aβ in the form of oligomers (at a concentration of 100 nm equivalent of Aβ monomer) contained short, irregular rods, globules, and some rings as evident in EM (Ca) or AFM (Db) images (C: scale bar, 100 nm; D: particle size is color-coded as on the right-hand panel). Fibril preparations (100 nm equivalent of Aβ monomer) are enriched for long fibrils as seen on EM (Cb) or AFM (Dc) images. These results indicate our ability to prepare and define specific forms of Aβ.
Chronic Aβ fibril treatment induces neuronal hyperexcitation
In identified pyramidal neurons, we compared effects of exposure to fibrillar Aβ or scrambled Aβ (a peptide made with the same amino acid composition as Aβ, but in random sequence and used as a nonaggregating control) on neuronal excitability. Throughout, we determined the aggregation state or form of Aβ using EM (Fig. 1C) and AFM (Fig. 1D), confirming that preparations were either oligomeric (Fig. 1Ca,Db) or fibrillar (Fig. 1Cb,Dc).
In hippocampal pyramidal neurons treated with 100 nm scrambled Aβ for 7 d, few spontaneous APs were observed (Fig. 2Aa), whereas in neurons treated with 100 nm fibrillar Aβ (concentration indicated is in monomeric Aβ equivalents) for 7 d, there was spontaneous bursting of APs (Fig. 2Ac). To define changes in neuronal excitability after Aβ treatment, we assessed effects of injection of depolarizing current into cells on neuronal activity (Desai et al., 1999; Karmarkar and Buonomano, 2006). Treatment for 7 d with 100 nm Aβ fibrils significantly increased neuronal AP spiking activity in response to injected currents (90 pA; Fig. 2Ad) when compared with results from cells treated with 100 nm scrambled Aβ (Fig. 2Ab). In mouse organotypic cultured hippocampal slices, 100 nm Aβ fibril treatment for 7 d also increased CA1 pyramidal neuronal excitability (Fig. 2Bb) relative to effects of scrambled Aβ (Fig. 2Ba). To quantify changes in neuronal excitability after Aβ treatment, we measured input–output relationships that derived from plots of spike numbers occurring each 4 s as a function of injected, depolarizing current amplitude (Desai et al., 1999; Karmarkar and Buonomano, 2006). For injected currents of 30–180 pA, numbers of APs were significantly increased above control levels by Aβ fibril exposure (Fig. 2C; in cultured neurons, n = 24 for Aβ fibril treatment; n = 16 for scrambled Aβ treatment; Fig. 2D; in cultured slices, n = 8 for Aβ fibril treatment; n = 8 for scrambled Aβ treatment; *p < 0.05; **p < 0.01, t test). We also measured changes relative to resting potential in membrane potential in response to depolarizing current injection (Fig. 2Ab, Ad, Ba, and Bb, traces). In 16 neurons treated with scrambled Aβ, 90 pA current injection depolarized the membrane potential by 9.4 ± 0.8 mV, whereas in 24 neurons treated with fibrillar Aβ, the membrane potential depolarized by 13.6 ± 1.7 mV (Fig. 2E; t test, **p < 0.01). In addition, exposure for 7 d to 100 nm Aβ fibrils caused an increase in resting input resistance in cultured hippocampal pyramidal neurons compared with effects in scrambled Aβ-treated controls (Aβ scrambled: 580 ± 67 MΩ, n = 8; Aβ fibril: 860 ± 88 MΩ, n = 8; t test, **p < 0.01; data summarized, but figure not shown), consistent with an increase in neuronal excitability (Fan et al., 2008; Mongiat et al., 2009). In hippocampal primary cultures and cultured slices, we also found significantly larger increase in membrane potential responses to injected current after fibrillar Aβ exposure relative to effects in the presence of control, scrambled Aβ (Fig. 2E). These results indicate that exposure of mouse hippocampal primary or slice cultures to 100 nm Aβ fibrils for 7 d induces pyramidal neuron hyperexcitation.
Chronic Aβ treatment induces neuronal hyperexcitation. A, Cultured hippocampal pyramidal neurons were treated with 100 nm scrambled (Scr) or fibrillar (Fib) Aβ for 7 d starting at 14 d in vitro (DIV). After exposure to Aβ fibrils for 7 d, cultured hippocampal pyramidal neurons showed an increase in spontaneous AP firing (Ac) compared with scrambled Aβ-treated controls (Aa). Typical traces also are shown of 90 pA current injection-induced AP firing after scrambled (Ab) or fibrillar (Ad) Aβ exposure (100 nm for 7 d). B, Representative current-clamp recordings are illustrated for pyramidal neurons in hippocampal slice cultures responding to 120 pA current injections after treatment for 7 d with 100 nm scrambled (Ba) or fibrillar (Bb) Aβ exposure. C, Input–output relationships between current injection and number of spikes per 4 s recorded from cultured pyramidal neurons treated with Aβ fibrils (Fib; ■) or scrambled Aβ (Scr; ●; *p < 0.05, **p < 0.01, 2-way ANOVA). D, Input–output relationships between current injection and number of spikes per 4 s recorded from pyramidal neurons in hippocampal slice cultures treated with Aβ fibrils (Fib; ●) or scrambled Aβ (Scr; ▴; **p < 0.01, 2-way ANOVA). E, Aβ fibril treatment elevated membrane potential changes [membrane potential (MP) minus resting potential (RP); ordinate, mV] elicited by 90 pA current injection (**p < 0.01, t test).
Time-dependent, concentration-dependent, and form-dependent effects of Aβ exposure on neuronal excitability
To define temporal patterns of Aβ exposure on intrinsic excitability, we treated mouse primary hippocampal cultures with 100 nm Aβ fibrils for 0, 1, 4, 7, or 10 d and monitored pyramidal neuronal responses to 90 pA current injection. Time-dependent effects were evident from current-clamp recordings (Fig. 3Aa). Effects of Aβ fibril treatment for 7 or 10 d on AP spike numbers (Fig. 3Ab; t test, **p < 0.01) and membrane potential depolarization (Fig. 3Ac; **p < 0.01, t test) in response to 90 pA current injection were significantly higher than for scrambled peptide-treated controls.
Time-dependent, concentration-dependent, and form-dependent effects of Aβ on neuronal excitability. A, Sample traces (Aa) of effects of 100 nm (equivalent of Aβ monomers) fibrillar Aβ exposure for the indicated periods on membrane excitability recorded during 90 pA current injection from hippocampal pyramidal neurons. Note the membrane potential change and time scale calibration bars. Bar graph summarizing the data indicates that 100 nm Aβ fibril exposure (filled bars) significantly increased spike number after 7 or 10 d of treatment relative to scrambled Aβ-treated controls (open bars) (Ab, **p < 0.01, t test), and membrane potential changes relative to resting potential (Ac, **p < 0.01, t test). B, Bar graphs summarize results as numbers of AP spikes counted (Ba) or changes in membrane potential (Bb, *p < 0.05, **p < 0.01, 1-way ANOVA) following exposure for 7 d to 100 nm of the indicated Aβ forms (Scr, scrambled; Mon, monomeric; Oli, oligomeric; Fib, fibrillar). C, Bar graphs summarizing concentration studies illustrate that 100 nm or 1 μm Aβ fibril exposure significantly increased spike number (Ca) or changes in membrane potential after 7 d treatment (Cb, **p < 0.01, 1-way ANOVA).
Aβ exists in a variety of different forms, these being soluble monomers, small oligomeric aggregates (e.g., dimers to dodecamers), or large, plaque-forming, fibrillar aggregates. These different forms of Aβ exhibit different effects on neurons, neural circuits, and neural networks (Bucciantini et al., 2002; Tsai et al., 2004; Cleary et al., 2005; Shankar et al., 2008). Here, we found that 7 d of exposure of cultured primary hippocampal pyramidal neurons to 100 nm fibrillar Aβ produced significantly (**p < 0.01, 1-way ANOVA) increased neuronal excitability whether measured as spike numbers (Fig. 3Ba) or by the extent of membrane potential depolarization (Fig. 3Bb) induced by 90 pA current injection. Oligomeric Aβ treatment induced a more modest but still significant increase in spike numbers (Fig. 3Ba; *p < 0.05, 1-way ANOVA).
Studies done to define concentration-dependent effects of treatment with Aβ fibrils on primary hippocampal pyramidal neuronal excitability showed that Aβ fibril exposure at 100 or 1000 nm was required to produce statistically significant increases in spike numbers (Fig. 3Ca) and the extent of membrane potential depolarization (Fig. 3Cb) in response to injection of a 90 pA current, relative to effects of exposure to 1 or 10 nm fibrillar Aβ or scrambled Aβ controls (n = 8 for each group, **p < 0.01, 1-way ANOVA).
Together, these results indicate that Aβ-induced neuronal hyperexcitation is dependent on Aβ form (oligomeric or fibrillar), Aβ exposure duration (≥7 d), and Aβ concentration (≥100 nm).
Chronic Aβ exposure upregulates α7-nAChR function on pyramidal neurons before onset of hyperactivity
Given indications for expression in the hippocampus of nAChRs, including those that contain α7 subunits, and for interactions between those receptors and Aβ, we next asked about the status of α7-nAChR function in cultured, primary pyramidal neurons under control conditions and following Aβ exposure. Several studies have indicated that nAChRs, including those containing α7 subunits, are sensitive to acute or subchronic (exposures on the order of seconds to minutes) effects of Aβ (see Introduction), but more work is needed to define effects of more chronic (days of exposure) effects of Aβ exposure on α7-nAChRs. Most pyramidal neurons (43 of 48 tested) in primary hippocampal cultures exhibited α7-nAChR-mediated patch-clamp, whole-cell current recording responses to a selective α7-nAChR agonist (choline, 10 mm; Fig. 4A). Interestingly, raw magnitudes (Fig. 4A) and current densities (normalized to conductance; Fig. 4B) of responses increased when cells were treated for ≥4 d with 100 nm Aβ fibrils relative to effects of scrambled Aβ exposure. Importantly, the upregulation in α7-nAChR function occurs before onset of Aβ exposure-induced neuronal hyperactivity. Interestingly, there were no significant effects of Aβ exposure on levels of functional α7-nAChR in GAD-GFP-tagged GABAergic neurons from cultures prepared from GAD-GFP mice (data not shown), indicating that upregulation seems to occur only on pyramidal neurons.
Chronic Aβ treatment-mediated upregulation of α7-nAChR function and increased AMPA receptor-mediated mEPSC frequency in cultured hippocampal neurons. A, Representative traces of 10 mm choline-induced, whole-cell current responses in cultured hippocampal neurons treated with 100 nm (Aβ monomer equivalent) fibrillar Aβ for the indicated periods. Note the calibration bars for duration of agonist exposure and for current amplitude. B, Bar graph summarizing effects on 10 mm choline-induced peak current density (ordinate, pA/pF) from neurons treated with 100 nm fibrillar Aβ (Fib; filled bars) for the indicated periods relative to effects of scrambled (Scr; open bars) Aβ. Effects on heightened sensitivity to Aβ emerge after 4 d of treatment. C, D, Mature, cultured hippocampal neurons (DIV 14) were treated with 100 nm (Aβ monomer equivalent) scrambled (C, Scr) or fibrillar (D, Fib) Aβ for 7 d starting on 14 DIV. Sample traces are shown of AMPA receptor-mediated mEPSCs recorded in the presence of 0.5 μm TTX, 50 μm d-APV, or 100 μm picrotoxin to block voltage-gated sodium channels, NMDA receptors, or GABA receptors, respectively, over longer (a) or shorter (b) time courses (note the time or current calibration bars). E, F, Population and (in insets) bar graphs summarizing results show no effect of fibrillar Aβ (Fib; filled bars) relative to scrambled Aβ (Scr; open bars) on mESPC amplitude (E), but there is an increase in mEPSC frequency (F; shorter interevent interval) in pyramidal neurons after treatment for 7 d with Aβ fibrils compared with scrambled Aβ-treated controls (**p < 0.01, t test).
Increased mEPSC activity in Aβ-treated hippocampal cultures
Because α7-nAChRs disposed on presynaptic nerve terminals play important roles in regulation of glutamate release (Gray et al., 1996), and because excessive glutamatergic activity could account for Aβ exposure-induced hyperexcitation, we measured effects of chronic Aβ fibril or scrambled Aβ treatment on AMPA receptor-mediated mEPSCs that occur spontaneously in cultured pyramidal neurons (in the presence of 0.5 μm TTX, 50 μm d-APV, and 100 μm picrotoxin to block sodium channel, NMDA receptor, and GABAA receptor activity, respectively; Fig. 4C,D). We found a significant increase in mEPSC frequency, but not amplitude, after 100 nm fibrillar Aβ treatment for 7 d, but not after treatment with scrambled Aβ (Fig. 4E,F). These results suggest that fibrillar Aβ exposure could increase presynaptic α7-nAChR function controlling vesicular glutamate release, perhaps thus contributing to Aβ fibril-induced neuronal hyperexcitation.
Chronic Aβ selectively upregulates α7-nAChR function
To assess specificity of chronic Aβ-mediated upregulation of α7-nAChR function, and to see whether Aβ affected some of the many other potential mediators of hyperexcitation, we examined effects of chronic Aβ on a variety of ligand-gated ion channels. As shown in Figure 5A,B, whole-cell current responses to 10 mm choline of cultured pyramidal neurons were increased, relative to effects after exposure to scrambled Aβ, slightly, significantly, or dramatically, respectively, following 4 d of exposure to Aβ monomers, oligomers, or fibrils. However, the whole-cell current responses to GABA, AMPA, or NMDA receptor agonist were unchanged after exposure to 100 nm Aβ, regardless of Aβ form, when parallel, quantitative assessments revealed significant increases in choline-induced currents in neurons exposed to oligomeric or fibrillar Aβ (Fig. 5B). These results indicate that exposure for ≥4 d to Aβ aggregates, especially fibrillar Aβ, selectively upregulates α7-nAChR function on hippocampal pyramidal cells.
Specific upregulation of α7- nAChR function by Aβ in primary cultured hippocampal neurons. A, Typical traces (−60 mV holding potential; see current amplitude and duration of ligand application calibration scales) are shown for whole-cell current recording responses to choline (10 mm) after 4 d of treatment with different forms of 100 nm Aβ. Significant potentiation of α7-nAChR-mediated currents was observed after exposure to 100 nm Aβ oligomers (Oli) or fibrils (Fib) but not monomers (Mon), when compared with scrambled (Scr) Aβ (current densities ± SE in pA/pF: Scr, 9.8 ± 1.4, n = 22; Mon, 10.4 ± 1.5 pA/pF, n = 14; Oli, 15.8 ± 2.3, n = 21; Fib, 20.8 ± 2.1, n = 26; Oli vs Scr, **p < 0.01; Fib vs Scr, **p < 0.01, 1-way ANOVA). B, Results are summarized for whole-cell recording peak current density in pA/pF for responses to GABA (0.1 mm), AMPA (0.1 mm), NMDA (0.1 mm), or choline (10 mm) after 4 d of exposure to the indicated form of Aβ (100 nm). Whole-cell currents mediated by GABAA, AMPA, or NMDA receptors in the same cells were not affected by treatment with different forms of Aβ when choline responses were altered. These results indicate that exposure to Aβ large aggregates selectively upregulates α7-nAChR function but not function of other ligand-gated ion channels.
Mechanisms of Aβ-induced upregulation of α7-nAChR cell-surface expression and function
To elucidate possible mechanisms of Aβ fibril-induced upregulation of α7-nAChR function, we determined effects of peptide treatments on levels of nAChR α7 subunit mRNA and protein in primary hippocampal cultures. We found that, after treatment with 100 nm fibrillar Aβ for 4 d, there was a significant elevation in numbers of surface α7-nAChRs as assessed by levels of α-Bgt binding complex-derived α7 subunit protein (Fig. 6B). By contrast, there was no change in total (cell surface plus intracellular pools from cell lysates) α7 subunit protein levels in toxin binding-derived complexes (Fig. 6B) or in α7 subunit mRNA levels (Fig. 6A; there also was no effect on nAChR β2 subunit mRNA). Other studies indicated that there was no change in levels of functional α7-nAChRs in cultures treated for 7 d with the α7-nAChR-selective antagonist MLA (data not shown), indicating that simple inhibition of α7-nAChR function, as occurs in our hands upon acute-to-subchronic exposure to Aβ (Liu et al., 2001), does not account for Aβ-induced upregulation of α7-nAChRs.
Mechanisms of Aβ treatment-induced α7-nAChR upregulation in cultured hippocampal neurons. A, Real-time RT-PCR analysis shows that Aβ fibril (Fib; filled bars) exposure for 4 d did not alter nAChR α7 subunit message levels in primary hippocampal pyramidal neuron cultures relative to levels after treatment with scrambled (Scr; open bars) Aβ-treated neurons. Data in each group were internally normalized to GAPDH mRNA levels, and the lack of an effect on nAChR β2 subunit mRNA levels is shown. B, Affinity purification-Western blot experiments (inset) showed that chronic exposure to fibrillar Aβ, relative to effects of scrambled Aβ, increased apparent levels of complexes on the cell surface containing nAChR α7 subunits capable of binding biotinylated α-Bgt, subsequently isolated by adherence to streptavidin beads, and then reactive with anti-nAChR α7 subunit antibody (H302, Santa Cruz Biotechnology), but not the total amount of such complexes isolated from lysed cells. The bar graph summarizes results (fibrillar Aβ vs scrambled Aβ; **p < 0.01, t test). C, Typical whole-cell current response traces for functional activation of α7-nAChR by 10 mm choline shows that treating hippocampal pyramidal neuron cultures with BFA (10 μg/ml) inhibited Aβ fibril exposure-induced increases in α7-nAChR function. D, Bar graph summaries the results of measurements of current density (ordinate, pA/pF) induced by 10 mm choline in cultured hippocampal neurons treated as indicated with scrambled (Scr) or fibrillar (Fib) Aβ alone in the presence of BFA (10 μg/ml) and indicates that BFA exposure inhibits the Aβ fibril exposure-induced increase in α7-nAChR function (fibril: 23.6 ± 3.2, n = 8; fibril plus BFA: 12.2 ± 2.1, n = 11; **p < 0.01, 1-way ANOVA). E, Treating the cultures with BFA (10 μg/ml) also inhibited Aβ fibril exposure-induced increases in surface nAChR α7 subunit protein expression as revealed based on affinity purification-Western analyses. F, Bar graph summaries the effects on surface nAChR α7 subunit protein levels in cultured hippocampal neurons treated as indicated. BFA (10 μg/ml) exposure inhibited Aβ fibril exposure-induced increase in nAChR α7 subunit levels on the cell surface (fibril: 176 ± 23%, n = 4; fibril plus BFA: 106 ± 8.9%, n = 4; **p < 0.01, 1-way ANOVA).
To further elucidate bases for Aβ fibril-induced upregulation of α7-nAChR function and cell-surface nAChR α7 protein levels, we inhibited protein trafficking and monitored changes in function and surface expression of α7-nAChRs. We used BFA, which has been reported to inhibit transport of proteins from the endoplasmic reticulum (ER) to the Golgi (Oda et al., 1987; Broutman and Baudry, 2001). Coincubation of primary hippocampal neuronal cultures with 10 μg/ml BFA completely prevented the increase in α7-nAChR function (Fig. 6C,D) and in surface α7 subunit protein levels (Fig. 6E,F) induced by 4 d of 100 nm Aβ fibril treatment. These results suggest that chronic Aβ fibril treatment increases the trafficking of functionally capable α7-nAChRs from the ER to the Golgi and then on to the cell surface.
α7-nAChR function is necessary for chronic Aβ-induced neuronal hyperexcitation
Studies described above showed that the onset of increase in α7-nAChR function occurred earlier than the appearance of neuronal hyperexcitation, suggesting a causal relationship between Aβ-induced upregulation of α7-nAChRs and neuronal hyperexcitation. To test the roles played by α7-nAChR upregulation in neuronal hyperexcitation after Aβ chronic exposure, we measured neuronal excitability in hippocampal cultures prepared from nAChR α7 subunit KO mice or from WT mice. Interestingly, in nAChR α7 subunit KO mouse hippocampal cultures, Aβ treatment (fibrils, 100 nm for 7 d) failed to increase neuronal excitability compared with activities in WT mouse cultures (Fig. 7A). Similarly, fibrillar Aβ treatment for 7 d in the continuous presence of 10 nm MLA also failed to increase neuronal excitability (Fig. 7A). Statistical analyses further confirmed these observations whether measuring AP frequency or changes in membrane potential in response to current injection (Fig. 7B,C). To exclude the possible roles of other nAChR subtypes (e.g., α4β2-nAChRs) in Aβ-induced hyperexcitation, we cotreated neurons with 100 nm fibrillar Aβ and an α4β2-nAChR antagonist (3 μm DHβE) for 7 d. We found that DHβE did not prevent Aβ-induced hyperexcitation (Fig. 7A–C). These results suggest that α7-nAChR function is necessary for the initiation of chronic Aβ-induced neuronal hyperexcitation.
Roles of α7-nAChR in chronic Aβ exposure-induced neuronal hyperexcitation. A, Typical traces of AP generation in response to 90 pA current injection (note the −60 mV holding potential and the time course and membrane potential calibration bars) show that genetic deletion of nAChR α7 subunits (α7−/−) or pharmacological block of α7-nAChRs (cochronic treatment for 7 d of neuronal cultures with 100 nm fibrillar Aβ1–42 and 10 nm MLA), but not pharmacological block of β2*-nAChR (via cochronic treatment with Aβ and 3 μm DHβE for 7 d), prevents neuronal hyperexcitation typically seen after 7 d of 100 nm fibrillar Aβ (Fib) exposure and cultures from WT mice and more closely resembles effects of control exposure to scrambled Aβ (Scr). B, Calculated input–output relationships for AP spike activity (ordinate) as a function of current injected (90 pA) in cultured hippocampal neurons prepared from WT or α7 KO mice show that genetic deletion of α7-nAChR or pharmacological block of α7-nAChRs, but not inactivation of β2*-nAChR, prevents the increase in AP generation (number per 4 s; ordinate) (**p < 0.01, WT Aβ fibril vs α7 KO Aβ fibril, or WT Aβ fibril vs WT Aβ fibril plus MLA, 2-way ANOVA). C, Genetic deletion or pharmacological block of α7-nAChRs also prevents the increase in changes in membrane potential relative to resting potential during current injection as typically induced by chronic fibrillar Aβ exposure. Coexposure to a β2*-nAChR antagonist along with Aβ for 7 d did not prevent neuronal hyperexcitation (**p < 0.01, WT Aβ fibril vs α7 KO Aβ fibril, or WT Aβ fibril vs WT Aβ fibril plus MLA, 1-way ANOVA). D, Sample traces show that genetic deletion of α7-nAChR (α7−/−) or pharmacological block of α7-nAChRs prevents 7 d fibrillar Aβ (Fib) exposure-induced increases in mEPSCs recorded from cultured hippocampal pyramidal neurons and relative to control effects of exposure to scrambled Aβ (Scr). E, Cumulative fraction plots for mEPSC interevent intervals show that genetic deletion of α7-nAChR or pharmacological block of α7-nAChRs, but not pharmacological block of β2*-nAChR, prevents the increase of mEPSC frequency. F, Genetic deletion or pharmacological block of α7-nAChRs with MLA, but not DHβE, prevents the increase of mEPSC frequency induced by chronic Aβ Fibril (Fib) exposure (**p < 0.01, WT Aβ fibril vs α7 KO Aβ fibril, or WT Aβ fibril vs WT Aβ fibril plus MLA, 1-way ANOVA).
In addition, we used similar genetic and pharmacological manipulations to assess whether or not α7-nAChRs also contribute to chronic Aβ-induced increases in mEPSC frequency. As demonstrated by typical recording traces shown in Figure 7D, treatment of hippocampal cultures in 100 nm Aβ fibrils for 7 d, prepared either from nAChR α7 subunit KO mice, or from WT mice but incubated in the presence of 10 nm MLA for those 7 d, failed to induce a significant increase in mEPSC frequency (Fig. 7E) compared with effects seen for Aβ-treated cultures from WT mice. Cotreatment with DHβE (3 μm) for 7 d failed to prevent the Aβ-induced increase in mEPSC frequency (Fig. 7D,E). Current-clamp recording from the same neurons confirmed that ligand inhibition and gene knock-out effects on neuronal excitability measured as changes in membrane potential in response to current injection parallel effects on mEPSC frequency (Fig. 7F). These data indicate that upregulation of α7-nAChR expression and function, on pyramidal neuronal soma and proximal dendrites and/or on presynaptic glutamatergic terminals, plays an important role in Aβ-induced neuronal hyperexcitation.
Effects of short-term pharmacological blockade of α7-nAChRs on Aβ-induced neuronal hyperexcitation and on glutamatergic hyperactivity
Mindful of the potential therapeutic or neuroprotective use of nicotinic ligands in AD or its in vitro models, we next sought to determine effects of short-term nAChR manipulation on Aβ exposure-induced neuronal excitability of cultured pyramidal neurons. Strikingly, simple treatment with 10 nm MLA for 10 min before and continuing through and beyond recording of responses to depolarizing current injection (60 pA for 80 ms) inhibited neuronal hyperexcitation otherwise seen following exposure of primary pyramidal neurons in cultures to 100 nm fibrillar Aβ for 7 d (Fig. 8A). The inhibition of Aβ-induced hyperexcitation was reversed following removal of MLA (Fig. 8A). Similarly, 10 min pretreatment with MLA (10 nm), but not with DHβE (3 μm), reduced mEPSC activity in cultures exposed for 7 d to 100 nm fibrillar Aβ (Fig. 8B). This effect was due to a decrease in mEPSC frequency (increased interevent interval; Fig. 8Cb) but not amplitude (Fig. 8Ca). These results again indicate that functional α7-nAChRs play a role in maintaining hyperexcitation induced by chronic Aβ exposure.
Effects of nAChR antagonist exposure on Aβ-induced neuronal excitability and mEPSCs activity. A, Typical current-clamp recording traces show that subchronic treatment with MLA (10 nm; 10 min pretreatment continuing during and beyond current injection; see time and voltage calibration bars and the timeline for current injection) abolishes (trace 2) neuronal hyperexcitation otherwise induced by 7 d of 100 nm fibrillar Aβ exposure (trace 1). After washout of MLA not just for 2 min (trace 3), but for 5 min (trace 4), neuronal hyperexcitability returns. B, Representative traces show that, relative to activity in cultures treated with Aβ fibrils for 7 d (Ba), 10 nm MLA exposure for 10 min before and continuing throughout the recording session (Bb), but not 3 μm DHβE exposure (Bc), reduces mEPSC activity. C, Cumulative fraction plots of the data from B quantify the reduction in the presence of MLA of mEPSC frequency (increased interevent interval) but not amplitude (**p < 0.01, WT Aβ fibril vs WT Aβ fibril plus MLA, 1-way ANOVA).
Discussion
Principal findings of this study
We provide direct evidence for the first time that chronic exposure to pathologically relevant concentrations of Aβ induces cultured hippocampal pyramidal neuron hyperexcitation. This is manifest in increases in the extent of membrane potential depolarization and in spontaneous bursting of APs in response to current injection, and it is accompanied by increased AMPA receptor-mediated mEPSC frequency but not amplitude, indicative of a higher probability of vesicular glutamate release. The effects are most pronounced for fibrillar Aβ exposure, but they also occur in the presence of Aβ oligomers, and are first significant after 7 d of peptide exposure. Importantly, we also demonstrate that chronic exposure to Aβ upregulates α7-nAChR cell-surface expression and function via a BFA-sensitive process and before the onset of neuronal hyperexcitation. Further, we show, using pharmacological and genetic manipulations, that α7-nAChR function is required for manifestation of Aβ-induced neuronal hyperexcitation. Together, our findings suggest that chronic Aβ-induced hyperexcitation is a pathophysiological phenotype and that α7-nAChR function is required for Aβ-induced neural hyperexcitation.
Time course, forms, and concentrations of Aβ involved in neuronal hyperexcitation
Studies of Aβ effects on neuronal function and excitability have been seemingly contradictory (Puzzo et al., 2008; Palop and Mucke, 2010), perhaps due to differences in experimental protocols used in Aβ studies and variables, such as peptide concentration and form. For instance, exposure to low Aβ concentrations enhances, but exposure to high Aβ concentrations depresses, hippocampal synaptic plasticity (Puzzo et al., 2008). Oligomeric Aβ modulates neuronal function more dramatically than monomeric Aβ (Shankar et al., 2008) and has more toxic effects (Hardy and Selkoe, 2002). Acute Aβ exposure depresses, but chronic Aβ exposure upregulates, α7-nAChR function (Liu et al., 2001; Yu et al., 2005). Nanomolar Aβ concentrations (1–100 nm) are thought to be most pathologically relevant based on levels found in AD patients and in animal models of disease (Cirrito and Holtzman, 2003). Our results covering a range of Aβ forms, concentrations, and times of exposure found the most dramatic effects occurring over a period of 4–7 d of treatment with oligomeric and, particularly, fibrillar Aβ at 100 nm, which is at the high end of pathologically relevant Aβ concentrations, assuming that all concentrations are provided as monomeric Aβ equivalents.
Specificity of effects of Aβ exposure on α7-nAChR expression and function and their post-translational basis
Our studies also sought to uncover possible mechanisms involved in hippocampal circuit hyperexcitation and epileptic seizures observed in human APP mouse models of AD (Palop et al., 2007; Minkeviciene et al., 2009). Aβ exhibits quite broad effects on a variety of receptors/channels in vitro, but most acute effects of Aβ on these targets, either on astrocytes (Abramov et al., 2004) or neurons, require much higher, micromolar concentrations of Aβ (Casley et al., 2009) than those seen in AD patient brains (1–100 nm) (Cirrito and Holtzman, 2003). Indeed, we found that there were no effects of several days of exposure to 100 nm Aβ fibrils on a variety of neurotransmitter-gated ion channels, including non-α7-nAChR subtypes or GABAA, AMPA, or NMDA receptors. Prior studies suggested that α7-nAChRs are highly sensitive to inhibition during acute exposure to Aβ (Liu et al., 2001, 2009; Pettit et al., 2001). Here, we also find that chronic Aβ exposure at nanomolar concentrations increases levels of expression and function of α7-nAChRs in hippocampal cultures. More work is required to determine whether other targets mediate Aβ's effects on neuronal excitability, and increases in functional α7-nAChR levels may not be the sole process triggering Aβ's effects, especially given our elucidation of increased input resistance and glutamatergic mESPC frequency when there is hyperexcitation. However, chronic Aβ exposure-induced effects on neuronal excitability require α7-nAChR function and occur after there is significant upregulation of α7-nAChR function.
Sensitivity to BFA of the Aβ-induced upregulation of functional and cell-surface α7-nAChRs indicates that the process has a post-translational basis, likely involving changes in protein trafficking and functional receptor assembly dependent upon enhanced transit of nAChR α7 subunits and/or assembled α7-nAChRs from the ER to the Golgi apparatus and not reflected in changes in nAChR α7 subunit mRNA levels. Interestingly, this finding is not entirely consistent with observations that there is elevated expression of α7 subunits as mRNA in both AD patients and AD animal models (Hellström-Lindahl et al., 1999; Chu et al., 2005; Counts et al., 2007), but the latter studies took a more global and long-term approach to assessing effects in brain tissues than did our studies focusing on hippocampal neuronal cultures. Studies showing no effects of chronic MLA exposure alone on levels of functional α7-nAChR indicate that upregulation cannot simply reflect adaptation to inhibition of α7-nAChRs, so other mechanisms must be involved in Aβ fibril-induced upregulation of α7-nAChR.
Roles of α7-nAChRs in AD pathogenesis and therapy
Previous work, including ours, showed inhibitory effects of acute Aβ exposure on heterologously or naturally expressed α7-nAChRs in cultured cells or brain slices (Liu et al., 2001, 2009; Pettit et al., 2001; Lamb et al., 2005). A recent study also reveals that α7-nAChR-dependent synaptic plasticity in the hippocampus can be disrupted by acute Aβ exposure (Gu and Yakel, 2011). There is not unanimity in these kinds of studies, as there are reports of activating effects of acute Aβ on α7-nAChRs heterologously expressed in oocytes (Dineley et al., 2002) or on native, non-α7-nAChRs (Fu and Jhamandas, 2003). However, reported increases in α7 subunit mRNA or protein in human AD or in mouse models of the disease (Hellström-Lindahl et al., 1999; Chu et al., 2005; Counts et al., 2007) could be taken as evidence for receptor upregulation triggered by inhibition of α7-nAChRs by chronic Aβ exposure. On the other hand, levels of α7-nAChR function have been reported to be comparable in adult, APP transgenic, or WT mice (Spencer et al., 2006). Nevertheless, given that activation of α7-nAChRs appears to significantly enhance cognitive function (Levin and Rezvani, 2002), it has been suggested that acute Aβ-mediated inhibition of α7-nAChR function could contribute to cholinergic signaling deficits and cognitive decline in AD and that enhancement of α7-nAChR function would be a viable, therapeutic strategy in treatment of AD (Buccafusco et al., 2005; Buckingham et al., 2009).
However, our data challenge this concept. By contrast to most recent studies concerning effects of acute Aβ treatment, our findings indicate that function of overexpressed α7-nAChR is required for manifestation of effects of chronic Aβ fibril exposure: effects are not seen in mice lacking α7 subunits, on subchronic inhibition of α7-nAChR function, or—by operational definition—in control cultures not exposed to Aβ and thus expressing baseline levels of α7-nAChR. We suggest that upregulation of highly calcium-permeable α7-nAChR expression and function on glutamatergic presynaptic terminals could lead to increased glutamatergic release responsible for the increased glutamate mEPSC frequency seen in Aβ-treated cultures. Other possible mechanisms cannot be ruled out (Ping and Tsunoda, 2012) and warrant further investigation, but we propose a new hypothesis that, in AD brain, aberrantly high deposition of Aβ upregulates α7-nAChR expression and function, in turn inducing neural hyperexcitation, perhaps epileptic seizures, and consequent neurotoxicity. If this hypothesis is true, then α7-nAChR antagonists should be considered as a potential therapeutic strategy for AD therapy. This idea also is consistent with recent findings that genetic deletion of nAChR α7 subunits significantly rescues synaptic impairment and learning/memory deficits in older, APP mice (Dziewczapolski et al., 2009). However, α7-nAChR-mediated effects may critically depend on animal age (and perhaps background), because one recent study (Hernandez et al., 2010), using a mouse model of AD, suggests that deletion of α7 subunits exacerbates learning deficits in young animals and alters the pattern of Aβ accumulation.
Further studies are required to distinguish possibly age-dependent roles of α7-nAChR in the pathogenesis of AD in vivo. However, our findings strongly suggest that Aβ-induced, elevated α7-nAChR function is required for neural hyperexcitation. Perhaps elevated α7-nAChR function also contributes to elevated seizure activity seen in AD and possibly excitotoxicity and ultimate cognitive deficits observed in AD. Most importantly, the current findings suggest that inhibition instead of activation of α7-nAChR function might be a novel and productive therapeutic strategy to slow or prevent progression and/or consequences of AD in older subjects.
Footnotes
This work was supported by National Institutes of Health Grants R01 DA015389, U19 DA019377, and U19 MH085193; the Barrow Neurological Foundation; Arizona Biomedical Research Commission Grants 9730 and 9615; Philip Morris USA and Philip Morris International; and a grant from the American Health Assistance Foundation (P.A.S.J.). The authors thank Dr. Michael Sierks for his assistance in atomic force microscopy imaging. The authors also thank Dr. Julie M. Miwa for her helpful comments. The contents of this report are solely the responsibility of the authors and do not necessarily represent the views of the National Institutes of Health or any of the aforementioned awarding agencies.
- Correspondence should be addressed to Jie Wu, MD, PhD, Professor, Division of Neurology, Barrow Neurological Institute, 350 West Thomas Road, Phoenix, AZ 85013-4496. Jie.Wu{at}chw.edu
