Rosiglitazone Prevents Amyloid- Oligomer-Induced ... - IOS Press

Journal of Alzheimer’s Disease 39 (2014) 239–251 DOI 10.3233/JAD-130680 IOS Press

239

Rosiglitazone Prevents Amyloid-␤ Oligomer-Induced Impairment of Synapse Formation and Plasticity via Increasing Dendrite and Spine Mitochondrial Number Shujun Xua,∗ , Guilan Liua , Xiaoming Baob , Jie Wuc , Shaomin Lid , Bangxu Zhenga , Roger Anwyle and Qinwen Wanga,∗ a Zhejiang Provincial Key Laboratory of Pathophysiology, Department of Physiology and Pharmacology, School of

Medicine, Ningbo University, Ningbo, Zhejiang, China b The No. 2 Hospital of Ningbo, Ningbo, China c Division of Neurology, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, AZ, USA d Center for Neurologic Diseases, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA, USA e Trinity College Institute of Neuroscience, Department of Physiology, Trinity College, Dublin, Ireland Handling Associate Editor: Ling-Qiang Zhu Accepted 6 September 2013

Abstract. Rosiglitazone has been known to attenuate neurodegeneration in Alzheimer’s disease (AD), but the underlying mechanisms remain to be fully elucidated. In this study, living-cell image, immunocytochemistry, and electrophysiology were used to examine the effects of soluble amyloid-␤ protein (A␤) oligomers and rosiglitazone on the synapse formation, plasticity, and mitochondrial distribution in cultured neurons. Incubation of hippocampal cultures with amyloid-␤ (A␤)42 oligomers (0.5 ␮M) for 3 h significantly decreased dendritic filopodium and synapse density. Pretreatment with rosiglitazone (0.5–5 ␮M) for 24 h prevented the A␤42 -induced loss of dendritic filopodium and synapse in a dose-dependent manner. However, neither A␤42 oligomer nor rosiglitazone has a significant effect on the velocity and length of dendritic filopodia. Electrophysiological recording showed that acute exposure of slices with 0.5 ␮M A␤42 oligomers impaired hippocampal long-term potentiation (LTP). Pre-incubation of hippocampal slices with rosiglitazone significantly attenuated the A␤42 -induced LTP deficit, which depended on rosiglitazone concentrations (1–5 ␮M) and pretreatment period (1–5 h). The beneficial effects of rosiglitazone were abolished by the peroxisome proliferator-activated receptor gamma (PPAR␥) specific antagonist, GW9662. Moreover, the mitochondrial numbers in the dendrite and spine were decreased by A␤42 oligomers, which can be prevented by rosiglitazone. In conclusion, our data suggested that rosiglitazone prevents A␤42 oligomers-induced impairment via increasing mitochondrial numbers in the dendrite and spine, improving synapse formation and plasticity. This process is most likely through the PPAR␥-dependent pathway and in concentration and time dependent manners. The study provides novel insights into the mechanisms for the protective effects of rosiglitzone on AD. Keywords: Alzheimer’s disease, amyloid-␤, mitochondria, PPAR␥, rosiglitazone, synapse formation, synapse plasticity ∗ Correspondence to: Shujun Xu and Qinwen Wang, Zhejiang Provincial Key Laboratory of Pathophysiology, Department of Physiology and Pharmacology, School of Medicine, Ningbo University, Ningbo 315211, Zhejiang, China. Tel.: +86 574 8760 9594; Fax: +86 574 8760 8638; E-mails: [email protected]; wangqingwen @nbu.edu.cn.

ISSN 1387-2877/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved

240

S. Xu et al. / Rosiglitazone Prevents Synaptic Impairment

INTRODUCTION Alzheimer’s disease (AD) is a common age-related neurodegenerative disorder characterized by multiple cognitive deficits including worsening of memory, judgments, and comprehension, and deterioration in global function [1]. Alarmingly, there is currently no effectively therapeutic treatment for AD. Thus, AD is a severe threat for human health as the global elderly population increases. It is well known that the progressive accumulation of amyloid-␤ protein (A␤) plays a central role in the genesis of AD. A␤ amyloidogenesis is a complex process which involves the sequential formation of different forms of amyloid species, such as monomer, oligomer, protofibrils, and mature fibrils [2]. The oligomers, but not the A␤ monomers, are the key factor to determine their toxicity as self-association is thought to be required [3]. The role of amyloid fibrils in the pathogenesis of AD remains controversial. Severity of dementia has shown a weak correlation with the fibrillar amyloid plaque burden in AD patients [4]. A␤ developed synaptotoxicity in the absence of plaques in amyloid-␤ protein precursor (A␤PP) transgenic mice [5]. Recent evidence has suggested that the proximate effectors of neurotoxicity are oligomeric A␤ assemblies [6]. Assemblies ranging from dimers to 24-mers are recognized as A␤ oligomers [7]. Protein structure study has revealed a cylindrical barrel structure of the oligomer, which is formed by several (e.g., six) antiparallel strands, and a steric zipper structure of the fibril, which is formed by much more (e.g., several hundreds) anti-parallel strands [8]. The transformation of oligomers to fibrils involves crossing a high free energy barrier. It has been reported that soluble A␤ oligomers maintain their original structure in the cultured media during incubation (up to 6 h). They do not change to large molecules such as protofibrils or to small molecules such as monomers [9]. The A␤ oligomer count in cerebrospinal fluid is a biomarker for AD [10]. Soluble A␤42 oligomers have been shown to inhibit long-term potentiation (LTP) in vivo and in vitro, causing neuronal death and synapse deterioration [11, 12]. Soluble A␤42 oligomers also have been implicated in the behavioral deficits in rats injected with A␤ oligomers-containing culture media or in A␤PP transgenic mice [13, 14]. Taken together, preventing soluble A␤42 oligomer-induced neurotoxicity is proposed to be an attractive mechanism for the treatment for AD. The nuclear receptor peroxisome proliferatoractivated receptor gamma (PPAR␥) is a ligand-

inducible transcription factor that increases insulin sensitivity, regulates lipid and energy metabolism as well as suppress inflammatory gene expression [15]. The thiazolidinediones (TZD), which act as agonists of the PPAR␥, have been widely used in the treatment of type 2 diabetes [16]. Epidemiological studies indicated that type 2 diabetes increase the risk of AD [17]. Rosiglitazone belongs to TZD and is a high-affinity PPAR␥ agonist. It has been proposed as a possible drug for treatment of AD. In animal models of AD, treatment with rosiglitazone reduced the A␤ burden and inflammatory responses, attenuated the disease-related behavioral impairments [18–21]. In addition, clinical studies also suggest beneficial effects of rosiglitazone in AD patients [1, 22]. However, the mechanisms for the potential role of rosiglitazone in AD have not been elucidated yet. In particular, the protective mechanisms underlying A␤ oligomer-induced neurotoxicity are nearly unknown. In this study, with the methods of electrophysiology, immunocytochemistry, and livingcell image, we investigated the effects of soluble A␤42 oligomers and rosiglitazone on the synaptic formation, plasticity, and mitochondria numbers in the dendrite and spines in rats to elucidate the underlying mechanisms of the protective effects of rosiglitazone on A␤-induced synapse impairment. MATERIALS AND METHODS Aβ oligomer preparation Soluble A␤42 oligomers were prepared as described previously [23]. Briefly, synthetic A␤42 (Bachem) was dissolved in ice-cold 1, 1, 1, 3, 3, 3-hexafluoro2-propanol (HFIP) (Sigma-Aldrich), thoroughly vortexed, and aliquoted to be frozen until use. A␤ was spin-vacuumed just prior to the experiment, dissolved in HFIP solution (final concentration: 10% (v/v) HFIP) and incubated at room temperature for 20 min. The solution was centrifuged at 14,000 g for 15 min at 4◦ C and the supernatant was collected. The HFIP was completely evaporated to obtain a 50 ␮M A␤ solution and kept at room temperature under constant stirring for 48 h. And then the tube was transferred to refrigerator and maintained at 4◦ C. Primary hippocampal neuronal cultures and transfection Use and care of animals followed the guidelines of the Ningbo University Animal Research Advisory


Committee. Primary neuronal cultures from postnatal 1-day-old Wistar rats were prepared as previous reported [24]. Briefly, the hippocampi from diencephalic structures were dissected and digested in 0.25% trypsin (Invitrogen) for 15 min at 37◦ C. Dissociated cells were plated on 35 mm culture dishes previously coated with poly-D-lysine (100 ␮g/ml) at density of 7 × 105 cells/cm2 . Cultures were maintained in a humidified incubator with 5% CO2 at 37◦ C. The plating medium was Dulbecco’s Modified Eagle Media (Invitrogen) supplemented with 10% FBS, 10% F-12 (Invitrogen). The medium was changed to Neurobasal medium (Invitrogen) supplemented with 2% B27, 1% glutamine after 24 h. At 5 day in vitro (DIV 5), cells were treated with 5 ␮M cytosine arabinofuranoside (Invitrogen) to reduce glial cell growth. Thereafter, half of the medium was replaced twice a week with Neurobasal medium containing 2% B27, 1% glutamine. The neurons were transfected with farnesylated enhanced green fluorescent protein (F-GFP)/GFP-actin or together with Mito-DsRed by Lipofectamine 2000 (Invitrogen) at DIV 5. Immunocytochemisty Primary cultured hippocampal neurons were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) for 10 min at room temperature, and permeabilized with 0.01% Triton in PBS for 10 min before treatment with 10% BSA for 1 h at room temperature. Cells were then incubated in primary antibodies in PBS containing 10% BSA overnight at 4◦ C. Rabbit anti-synaptotagmin (Chemicon) was used at 1:400. Mouse anti-postsynaptic density 95 (PSD-95) (Chemicon) was used at 1:200. After washing with PBS three times, cells were incubated with secondary antibodies for 1 h at room temperature. Both Alexa-546 antirabbit and Alexa-488 anti-mouse secondary antibodies (Invitrogen) were used at 1:2000. Imaging of distal neuronal dendrites was performed with a Fluoview 1000 confocal microscope (Olympus, Tokyo, Japan). The background of images was subtracted, and a single threshold was chosen manually to define clusters so that clusters corresponded to puncta at least twofold greater intensity than the diffuse fluorescence on the dendrite shaft. Hippocampal slices preparation Slices of the rat hippocampus (Wistar rats, males; age 3-4 weeks) were prepared as described previously

241

[12]. Rat brain was rapidly removed after decapitation and placed in ice-cold oxygenated (95% O2 /5% CO2 ) physiological media. Several transverse slices (350 ␮m) were cut using Vibratome Intracell Plus 1000 (UK) and the slices were placed in a storage container containing oxygenated medium at room temperature (20–22◦ C) for 1 h. The slices were then transferred to a submersion recording chamber and continuously superfused at a rate of 5-6 ml/min at 30–32◦ C. The control media containing (in mM): NaCl 120, KCl 2.5, NaH2 PO4 1.25, NaHCO3 26, MgSO4 2.0, CaCl2 2.0, D-glucose 10. All solutions contained 100 ␮M picrotoxin (Sigma, St Louis, MO) to block GABAA mediated activity. Aβ oligomers and rosiglitazone treatment At DIV 7, the hippocampal neurons develop rapidly and become activation [25]. The filopodia actively move and retract into a more stable spine-like shape [26]. Whereas at DIV 15, neurons reach a “mature” stage, and most synapses are formed and are sensitive to the effectors [26, 27]. Thus, for the filopodia study, cultured hippocampal neurons were incubated with 0.5 ␮M soluble A␤42 oligomers for 3 h at DIV 7. For the spine, synapse, and mitochondrial research, soluble A␤42 oligomers (0.5 ␮M) were applied to cultured neurons for 3 h at DIV 15. Rosiglitazone was added with the dose of 0.1, 0.5, or 5 ␮M for 24 h, respectively, prior to A␤42 oligomers incubation. GW9662 (5 ␮M) was added alone or in the presence of 5 ␮M rosiglitazone. For in vitro electrophysiological recording, 0.5 ␮M soluble A␤42 oligomers were perfused for 40 min before high frequency stimulation (HFS). Rosiglitazone was pre-incubated to slices prior to HFS for 1 h at the doses of 1 or 5 ␮M, or for 5 h at the doses of 1, 2, or 5 ␮M, GW9662 (5 ␮M) was added at 5 h before HFS. Confocal imaging and analysis After drug treatments, the neurons were maintained in a recording chamber with extracellular solution (148.00 mM NaCl, 3.00 mM KCl, 3.00 mM CaCl2 , 10.00 mM HEPES, and 8.00 mM glucose, pH 7.3.) at room temperature. Digital images of GFP were collected on an Fluoview 1000 confocal microscope (Olympus) using a 60 × oil objective lens without optical zoom at an excitation wavelength of 488 nm.

242


A series of imaging pictures of filopodia from living neurons were collected at DIV 7 in time series mode. The images were captured every 30 s for 5 min. The length of an individual filopodium was defined as the linear distance between its tip and base, and the lengths of all filopodia were traced and measured using Image J software. The filopodium densities were determined by counting all filopodia observed on the microscopic imaging session and dividing by dendritic segment length. In filopodium motility analysis, frames were aligned in Image J software, filopodium protrusion tips throughout the series were tracked in Image-Pro Plus 5.1 software using manual tracking plug-in and the velocities were measured. To measure spine density, images were acquired at DIV 15 in 2-D stack. Spine densities were analyzed using Fluoview-1000 software. All lengths of the primary and secondary dendritic branches were measured by tracing their extension and the spines were counted manually. To measure synapse density, synaptotagmin containing puncta and PSD-95 containing puncta were labeled by red and green fluorescent, respectively. Then, the co-localization of puncta was counted. For mitochondrial analysis, two-channel confocal imaging was performed on neurons expressing EGFP and MitoDsRed at DIV 14-15. For all analysis, images were analyzed blind to treatments and data were collected from at least three independent experiments. Electrophysiological recordings Standard electrophysiological techniques were used to record field potentials. The medial perforant pathway of the dentate gyrus was given a presynaptic stimulation using a bipolar insulated tungsten wire electrode, and field excitatory postsynaptic potentials (EPSPs) were recorded from the middle one-third of the molecular layer of the dentate gyrus with a glass microelectrode, at a control test frequency of 0.033 Hz. LTP was induced by HFS consisting of eight trains, eight of each stimuli at a frequency of 200 Hz, intertrain interval 2 s to evoke an initial EPSP of the train of double the normal test EPSP amplitude. Recordings were analyzed using p-CLAMP (Axon Instruments, CA, USA). Statistical analyses Data are presented as mean ± SEM. One-way ANOVA with post hoc comparisons was used for statistical analyses. p < 0.05 was considered statistically significant.

RESULTS Aβ oligomers decrease dendritic filopodium density in hippocampal neurons To examine the effect of A␤ oligomers on development of dendritic filopodia, primary cultured hippocampal neurons were incubated with 0.5 ␮M soluble A␤42 oligomers for 3 h at DIV 7, and the density of dendritic filopodia was analyzed. The filopodium density was markedly decreased in neurons treated with soluble A␤42 oligomers compared to that in the control group (n = 30, p < 0.01, Fig. 1A-B, F). These results suggested that treatment of A␤ oligomers for 3 h induces filopodia loss in cultured hippocampal neurons. Rosiglitazone prevents dendritic filopodium loss caused by Aβ oligomers in hippocampal neurons in a dose-dependent manner To determine whether rosiglitazone prevents the filopodium loss caused by A␤ oligomers, cultured neurons were treated with rosiglitazone (0.1, 0.5, or 5 ␮M) for 24 h before 0.5 ␮M soluble A␤42 oligomers incubation. Results showed that rosiglitazone prevented the A␤42 oligomer-induced filopodium loss in a dosedependent manner (Fig. 1C-F). The filopodium density in 0.5 ␮M or 5 ␮M rosiglitazone treated group was not significantly different compared to that in the control group (p > 0.05), but was significantly increased compared to that in the group treated with A␤42 oligomers alone (p < 0.01). However, treatment with 0.1 ␮M rosiglitazone for 24 h did not prevent the filopodium loss induced by A␤42 oligomers. The filopodium density in the 0.1 ␮M rosiglitazone-treated group was not significantly altered compared to that in A␤42 oligomers-treated group (p > 0.05). Neither Aβ oligomers nor rosiglitazone alters the filopodium length and motility Additionally, we assessed the alteration of other parameters of filopodium development including filopodium length and filopodium motility. Neither the length (Fig. 2A) nor the motility (Fig. 2B) of filopodia was altered in the presence of soluble A␤42 oligomers (0.5 ␮M, 3 h) alone or treated with rosiglitazone (0.1, 0.5, and 5 ␮M) 24 h before A␤42 oligomer incubation. Our results suggested that both A␤ oligomers and rosiglitazone selectively affect filopodium density, without altering filopodium length and motility.


243

Fig. 1. Rosiglitazone prevented the filopodium loss caused by A␤42 oligomers at DIV 7 in a dose-dependent manner. Filopodium morphology was revealed by co-transfection of F-GFP with GFP-actin. Representative images of neurons treated with vehicle (A), 0.5 ␮M soluble A␤42 oligomers (B), 0.1 ␮M rosiglitazone + 0.5 ␮M soluble A␤42 oligomers (C), 0.5 ␮M rosiglitazone + 0.5 ␮M soluble A␤42 oligomers (D), and 5 ␮M rosiglitazone + 0.5 ␮M soluble A␤42 oligomers (E). Scale bar: 20 and 5 ␮m. F) Quantitative comparison of the density of dendritic filopodia. Rosig., rosiglitazone. **p < 0.01 versus vehicle-treated group, ## p < 0.01 versus A␤42 -treated group.

Fig. 2. Neither A␤42 oligomers nor rosiglitazone altered the filopodium length and motility at DIV 7. Primary hippocampal neuron cultures were incubated with either soluble A␤42 oligomers for 3 h or co-incubated with rosiglitazone (pretreated for 24 h). A) Representative a quantitative comparison of length of dendritic filopodia. No significant difference was found among these groups. B) Representative a quantitative comparison of motility of filopodia. No significant difference was found among these groups. Rosig., rosiglitazone.

244


Fig. 3. Dose-dependent effects of rosiglitazone on preventing the A␤42 oligomer-induced spine loss at DIV 15. Spine morphology was revealed by co-transfection of F-GFP with GFP-actin. Representative images of neurons treated with vehicle (A), 0.5 ␮M soluble A␤42 oligomers (B), 0.1 ␮M rosiglitazone + 0.5 ␮M soluble A␤42 oligomers (C), 0.5 ␮M rosiglitazone + 0.5 ␮M soluble A␤42 oligomers (D), and 5 ␮M rosiglitazone + 0.5 ␮M soluble A␤42 oligomers (E). Scale bar: 20 and 5 ␮m. F) Quantification of dendritic spine density. Rosig., rosiglitazone. **p < 0.01 versus vehicle-treated group, ## p < 0.01 versus A␤42 -treated group.

Rosiglitazone prevents Aβ42 oligomers-induced loss of dendritic spines and synapses of hippocampal neurons in a dose-dependent manner To further study the role of A␤ oligomers and rosiglitazone on synapse formation, culture hippocampal neurons were treated either with soluble A␤42 oligomers alone for 3 h or together with rosiglitazone which was pretreated for 24 h, and then spine densities were analyzed at DIV 15. We found that treatment of 0.5 ␮M soluble A␤42 oligomers for 3 h significantly reduced spine density in hippocampal neurons (n = 26, p < 0.01, Fig. 3A-B, F). Pre-treatment with rosiglitazone for 24 h prevented the A␤42 oligomers-induced spine loss in a dose-dependent manner (Fig. 3C-F). The spine density in the 0.5 ␮M or 5 ␮M rosiglitazonetreated group was significantly larger than that in the

group treated with A␤42 oligomers alone (p < 0.01) and was not significantly different from that in controls (p > 0.05). However, 0.1 ␮M rosiglitazone did not prevent the spine loss caused by soluble A␤42 oligomers. The spine density in the 0.1 ␮M rosiglitazone-treated group was not significantly different compared to that in the group treated with A␤42 oligomers alone. To further verify the results, we also analyzed the role of soluble A␤42 oligomers and rosiglitazone on synapse density by immunocytochemistry. Anti-presynaptic marker synaptotagmine and anti-postsynaptic marker PSD-95 specific antibodies were used and the colocalization puncta were analyzed. We found that the synapse numbers were also significantly decreased by soluble A␤42 oligomers treatment and rosiglitazone prevented the A␤42 oligomer-induced synapse loss in a dose-dependent manner (Table 1, n = 16).


245

Fig. 4. GW9662 abolished the preventive effect of rosiglitazone on A␤42 oligomer-induced filopodium and synapse loss. A) Quantitative comparison of the synapse densities among the groups treated with soluble A␤42 oligomers, soluble A␤42 oligomers plus rosiglitazone and soluble A␤42 oligomers, rosiglitazone plus GW9662, respectively. B) Quantitative comparison of the spine densities among the groups treated with soluble A␤42 oligomers, soluble A␤42 oligomers plus rosiglitazone, and soluble A␤42 oligomers, rosiglitazone plus GW9662, respectively. C) Quantitative comparison of the filopodium densities among the groups treated with soluble A␤42 oligomers (0.5 ␮M), soluble A␤42 oligomers plus rosiglitazone (5 ␮M), and soluble A␤42 oligomers, rosiglitazone plus GW9662 (5 ␮M), respectively. **p < 0.01 versus A␤42 -treated group. ++ p < 0.01 versus rosiglitazone-treated group, p > 0.05 versus A␤ -treated group. 42

Table 1 Rosiglitazone prevented A␤42 oligomer-induced loss of dendritic synapses of hippocampal neurons in a dose-dependent manner Group Control A␤42 5 ␮M rosiglitazone A␤42 + 0.1 ␮M rosiglitazone A␤42 + 0.5 ␮M rosiglitazone A␤42 + 5 ␮M rosiglitazone

Synapse number/100 ␮m 41.12 ± 2.11 26.56 ± 1.52** 42.52 ± 2.85 28.28 ± 2.49 39.78 ± 1.77# 41.3 ± 2.18##

**p < 0.01 versus control group, ## p < 0.01 versus soluble oligomeric A␤42 -treated group, # p < 0.05 versus soluble oligomeric A␤42 treated group, p > 0.05 versus control group, n = 16.

Rosiglitazone protects neurons against Aβ oligomers-induced loss of dendritic synapses and filopodia by activating PPARγ pathway Next, we determined whether rosiglitazone exerts its protective effects on the A␤42 oligomer-induced synapse and filopodium loss through the PPAR␥ pathway. Cultured neurons (DIV 6, DIV 14) were treated with rosiglitazone (5 ␮M) in the presence or absence of PPAR␥-specific antagonist GW9662 (5 ␮M) for 24 h before soluble A␤42 oligomers (0.5 ␮M, 3 h) incubation. GW9662 treatment alone did not alter the synapse density (103.62 ± 6.33 % of control, p > 0.05, n = 16). However, GW9662 abolished the preventive effects of rosiglitazone on the soluble A␤42 oligomerinduced synapse loss. The synapse density in the group treated with GW9662 plus rosiglitazone was significantly lower than that in rosiglitazone-treated group (p < 0.01, n = 20) and the value was not signif-

icantly different from that in the group treated with soluble A␤42 oligomers alone (p > 0.05, Fig. 4A). Similarly, GW9662 also abolished the preventive effects of rosiglitazone on the soluble A␤42 oligomers-induced spine loss (Fig. 4B). The spine density in the GW9662 and rosiglitazone co-treated group was significantly lower than that in rosiglitazone-treated group (p < 0.01) and was not significantly different from that in the group treated with A␤42 oligomers alone (p > 0.05, n = 20). The role of GW9662 on filopodium was also tested and we found that GW9662 completely abolished the protective effects of rosiglitazone on soluble A␤42 oligomer-induced filopodium loss (Fig. 4C). The filopodium density in the group treated with GW9662 plus rosiglitazone was significantly lower than that in rosiglitazone-treated group (p < 0.01, n = 20), and it was not significantly different from that in the group treated with soluble A␤42 oligomers alone (p > 0.05, n = 20). Rosiglitazone prevents Aβ42 oligomers-induced LTP impairment in hippocampal slices We further addressed the functional consequences of rosiglitazone on synaptic plasticity. 0.5 ␮M soluble A␤42 oligomers were incubated to hippocampal slices for 40 min prior to HFS. As shown in Fig. 5, HFS of medial perforant pathway induced a sustained enhancement of field EPSPs in the dentate gyrus presented as the LTP in the control slices (n = 8). However, in the soluble A␤42 oligomer-treated slices, the same HFS failed to induce the LTP; the EPSP

246


Fig. 5. Rosiglitazone prevented the LTP inhibition by A␤42 oligomers in concentration and time- dependent manners. A) In soluble A␤42 oligomers treated slices (exposure for 40 min prior to HFS), the amplitudes of evoked EPSPs at 60 min after HFS (open circles) were significantly smaller than those in the control group (filled circles) (p < 0.05 versus vehicle-treated group). B) Pre-incubation of slices with rosiglitazone for 5 h prior to HFS did not alter LTP production (p > 0.05 versus vehicle-treated group), but prevented the soluble A␤42 oligomer-induced inhibition of LTP production (p < 0.01 versus A␤42 -treated group). C) Rosiglitazone prevented soluble A␤42 oligomers-induced LTP impairment at 2 ␮M for 5 h (filled circles) and 5 ␮M for 1 h (open circles) (p < 0.01 versus A␤42 -treated group). D) Rosiglitazone (1 ␮M) did not prevent soluble A␤42 oligomer-induced LTP impairment at a short term incubation (1 h, open circles), only a partially prevention at a longer incubation (5 h, filled circles). E) Pre-incubation slice with 5 ␮M GW9662 for 5 h prior to high frequency stimulation did not alter LTP production (p > 0.05 versus vehicle-treated group), but abolished the preventive effects of rosiglitazone on LTP production (p < 0.01 versus rosiglitazone and A␤42 -treated group). The traces in panel A and B in the inset of (A) illustrate the EPSPs prior to and 60 min after HFS, respectively. Rosig.: rosiglitazone.


amplitudes at 60 min after HFS were significantly decreased compared with those in the control group (n = 8, p < 0.05, One-way ANOVA, Fig. 5A). Preincubation of slices with rosiglitazone alone for 5 h prior to HFS did not alter LTP production (p > 0.05), but prevented the soluble A␤42 oligomer-induced inhibition of LTP production in concentration and time-dependent manners (Fig. 5B-D). The LTP inhibition was completely restored after pre-incubated the slices with 5 ␮M rosiglitazone for 5 h, 5 ␮M rosiglitazone for 1 h, or 2 ␮M rosiglitazone for 5 h (n = 5). The EPSP amplitudes at 60 min after HFS in above three groups were significantly increased compared with those in the group treated with soluble A␤42 oligomers alone (p < 0.01) and were not significantly different from those in controls (p > 0.05, Fig. 5B, C). Pre-incubation with 1 ␮M rosiglitazone for 5 h (n = 5) partially restored the inhibition of LTP by soluble A␤42 oligomers; the EPSP amplitudes at 60 min after HFS were significantly increased compared with those in the group treated with soluble A␤42 oligomers alone (p < 0.01), but were significantly decreased compared with those in controls (p < 0.05). While pre-incubated with 1 ␮M rosiglitazone for 1 h (n = 5) did not prevent the inhibition of LTP by soluble A␤42 oligomers, the EPSP amplitudes at 60 min after HFS were not significantly different from those in the group treated with A␤42 oligomers alone (p > 0.05, Fig. 5D). Preincubation slice with 5 ␮M GW9662 for 5 h prior to high frequency stimulation did not alter LTP production (p > 0.05), but abolished the preventive effects of rosiglitazone on LTP production (Fig. 5E). The EPSP amplitudes at 60 min after HFS in the GW9662 and rosiglitazone co-treated group were significantly lower than those in the rosiglitaone-treated group (p < 0.01) and were not significantly different from those in the group treated with soluble A␤42 oligomers alone (p > 0.05, Fig. 5E). Rosiglitazone prevents Aβ42 oligomers-induced loss of mitochondria in dendrites and spines in hippocampal neurons Disruption of mitochondrial localization may exert adverse effects on synaptic functions [28]. To find out whether rosiglitazone prevents soluble A␤42 oligomer-induced impairment of synapse function through mitochondrial redistribution, the mean dendritic mitochondrial index (ratio of total mitochondrial length in the dendritic shaft to dendritic length in a given dendritic segment) and the ratio of spines with mitochondria were measured. We found that the

247

mitochondrial indexes were significantly decreased by soluble A␤42 -oligomer treatment (n = 15, p < 0.01). The decrease of mitochondrial indexes was prevented by 5 ␮M rosiglitazone treatment (p < 0.01 versus the soluble A␤42 oligomer-treated group, Fig. 6A-J). To find out whether the decrease mitochondrial indexes in the soluble A␤42 oligomer-treated group was caused by decreased number or length of mitochondria, we analyzed the number and length of discrete mitochondria. The number of mitochondria per 100 ␮M was decreased in the soluble A␤42 oligomer-treated group (p < 0.05). 5 ␮M rosiglitazone prevented the decrease of mitochondrial number caused by A␤42 oligomers (p < 0.05 versus the soluble A␤42 oligomertreated group, Fig. 6K). However, the length of mitochondria was not changed either in the group treated with soluble A␤42 oligomers alone or A␤42 oligomers plus rosiglitazone (Fig. 6L). Moreover, we also found that the percentage of spines with mitochondria was also decreased in the A␤42 oligomer-treated group (p < 0.05). 5 ␮M rosiglitazone prevented soluble A␤42 oligomer-induced decrease in percentage of spines with mitochondria (p < 0.05 versus A␤42 oligomer-treated group, Fig. 6M). This protective effect of rosiglitazone on the percentage of mitochondria in spines was also prevented by 5 ␮M GW9662 (15.93 ± 1.64% in the GW9662-treated group, p < 0.05 versus rosiglitazone-treated group).

DISCUSSION Perturbations in synapse formation and synaptic plasticity by pathological levels and forms of A␤ might be directly linked to the memory deficits in AD [29, 30]. In this study, we found that exposure of soluble A␤42 oligomers to cultured hippocampal neurons significantly decreased the density of dendritic filopodia. This result suggests that soluble A␤42 oligomers may disturb the initial stage of synapse formation since the dendritic filopodia are most abundant and dynamic during the periods of rapid synaptogenesis. It has been known that dendritic filopodia play vital roles in initiating synaptogenic contacts and dendritic spines formation [31]. This observation provides further insights into the neurotoxicity of soluble A␤42 oligomers in AD pathology. Rosiglitazone prevented soluble A␤42 oligomer-induced filopodia loss in a dose-dependent manner. These results suggested a potential protective effect of rosiglitazone in AD pathology during initial synapse formation. Nevertheless, neither soluble A␤42 oligomers nor rosiglitazone

248


Fig. 6. Rosiglitazone prevented A␤42 oligomer-induced loss of mitochondria in dendrites and spines of hippocampal neurons. Mitochondrial and spine morphology was revealed by co-transfection of F-GFP, GFP-actin, and Mito-DsRed. Representative images of neurons treated with vehicle (A-C), 0.5 ␮M soluble A␤42 oligomers (D-F), and 5 ␮M rosiglitazone + 0.5 ␮M soluble A␤42 oligomers (G-I). Scale bar: 30 and 10 ␮m. J) Quantitative comparison of the dendritic mitochondrial index, **p < 0.01 versus vehicle-treated group, ## p < 0.01 versus soluble A␤42 oligomers-treated group. K) Quantitative comparison of mitochondrial number per 100 ␮m dendrite, *p < 0.05 versus vehicle-treated group, # p < 0.05 versus soluble A␤42 oligomer-treated group. L) Quantitative comparison of dendritic mitochondrial length. M) Quantitative comparison of the percentage of spines with mitochondria. Rosig. : rosiglitazone.


altered the filopodium motility and length in the study, suggesting that both soluble A␤42 oligomers and rosiglitazone affect specially on the density of dendritic filopodia. Consistent with the role of soluble A␤42 oligomers and rosiglitazone in filopodium density, A␤42 oligomers also reduced spine and synapse densities and rosiglitazone pre-treatment prevented soluble A␤42 oligomer-induced spine and synapse loss in a dose-dependent manner. Previous study has demonstrated that rosiglitazone rescues the apolipoprotein E4-mediated spine loss in primary cortical neurons [32]. Our result suggested that rosiglitazone also has a neuroprotective effect on the soluble A␤42 oligomerinduced loss of synapses in AD. In addition to its protective role in synapse formation, rosiglitazone also prevented A␤42 oligomerinduced impairment in synaptic plasticity. Our experiments demonstrated that A␤42 oligomers exhibit a strong inhibitory effect on the induction of hippocampal LTP, and this inhibition can be prevented by rosiglitazone. Interestingly, the effects of rosiglitazone on A␤42 oligomer-induced LTP impairment were dependent on both concentrations and treatment period in the study. At low concentration (1 ␮M), rosiglitazone did not exhibit a protective effect when neurons were pretreated for 1 h but showed partial protection by pretreatment for 5 h. Whereas at middle concentration (2 ␮M), rosiglitazone completely prevented soluble A␤42 oligomer-induced LTP impairment when neurons were pretreated for 5 h and at high concentration (5 ␮M), rosiglitazone completely prevented soluble A␤42 oligomer-induced LTP impairment when neurons were pretreated for 1 h. Therefore, rosiglitazone may exert its protective effects not only in a dose-dependent manner, but also in a time-dependent manner. Soluble A␤ oligomers bind to synapses and trigger activation of downstream signal pathways, and cause neuronal tau hyperphosphorylation, oxidative stress, and synapse deterioration and loss [18, 33, 34]. The mechanism of the protective effects of rosiglitazone might be by interfering with the pathway triggered by A␤ oligomers either via a nuclear receptor PPAR␥dependent or PPAR␥-independent signaling pathway [15, 35, 36]. Our results indicated that GW9662 (a PPAR␥ specific antagonist) in the concentration we used has no effect on synapse formation and plasticity, but it abolished the protective role of rosiglitazone. It is possible that the protective effects of rosiglitzone on A␤ oligomer-induced synapse loss and synaptic plasticity impairment are most likely mediated by acti-

249

vating PPAR␥ pathway. Moreover, a previous study demonstrated that PPAR␥ stimulation might promote mitochondrial biogenesis [37]. Our results also indicated that rosiglitazone prevents the soluble A␤42 oligomer-induced decrease in spine mitochondrial number in the primary cultured hippocampal neurons. It is generally believed that mitochondrial distribution is adapted to cellular physiology, thus the mitochondria will be concentrated in subcellular regions with high metabolic requirement [28, 38]. Disruption of mitochondrial localization to synapses may exert adverse effects on synaptic functions [28]. Mitochondrial activity is a sensitive way to modulate synaptic plasticity in hippocampus [39]. Besides the role in the mitochondrial biogenesis, PPAR␥ pathway also has been reported to have anti-inflammatory and anti-oxidative stress function by increasing two mitochondrial regulators: adenosine monophosphate-activated protein kinase (AMPK) and Sirtuin 1 (SIRT1) [40]. Thus, rosiglitazone may prevent soluble A␤42 oligomerinduced impairment of synaptic function through the PPAR␥ pathway. Increasing dendrite and spine mitochondria numbers is also as part of the mechanism of the prevention of synapse formation and plasticity impairment. Results from clinical trials of rosiglitazone on early AD are contradictory. Previous studies demonstrate that rosiglitazone preserves cognition in patents with early disease and mild cognitive impairment [1, 22]. However recent studies indicate that rosiglitazone has no effect on mild-to-moderate AD either in a monotherapy study or used as adjunctive therapy to acetylcholinesterase inhibitors [41, 42]. One possible reason is that the Watson study included a mixture of subjects with dementia and mild cognition impairment leading to a disparity in baseline severity [42]. Our results have shown that rosiglitazone may exert its protective effects not only in a dose-dependent manner, but also in a time-dependent manner. Thus, it is possible that rosiglitazone may exert its beneficial effect in patients for longer periods and at even earlier stages in the disease [43]. There would be a disparity in the neuroprotective effects of rosiglitazone between this drug in foundational research in vitro, and the clinical research in vivo due to the blood-brain barrier and other factors. Further studies are needed to elucidate whether the concentrations of rosiglitazone that we used in this study would exert the similar protective effects on individuals in clinical trials. However, our observations provide a potential cellular mechanism for rosiglitazone in protection of hippocampal synapses against A␤-induced cognitive deficits in AD.

250


ACKNOWLEDGMENTS

[12]

This work was supported by the National Natural Science Foundation of China (30900430, 81271209, and 30970932), the Project-sponsored by SRF for ROCS, SEM (No. 20101561), Zhejiang Provincial Natural Science Foundation (LY12H09001), Ningbo Key Science and Technology Project (2011C51006), Ningbo Talent Project, Innovative Research Team of Ningbo (2009B21002), and the K.C. Wong Magna Fund in Ningbo University. Authors’ disclosures available online (http://www.jalz.com/disclosures/view.php?id=1943).

[15]

REFERENCES

[16]

[1]

[2]

[3]

[4] [5]

[6] [7] [8]

[9]

[10]

[11]

Risner ME, Saunders AM, Altman JF, Ormandy GC, Craft S, Foley IM, Zvartau-Hind ME, Hosford DA, Roses AD, Rosiglitazone in Alzheimer’s Disease Study Group (2006) Efficacy of rosiglitazone in a genetically defined population with mild-to-moderate Alzheimer’s disease. Pharmacogenomics J 6, 246-254. Ahmed M, Davis J, Aucoin D, Sato T, Ahuja S, Aimoto S, Elliott JI, Van Nostrand WE, Smith SO (2010) Structural conversion of neurotoxic amyloid-beta(1-42) oligomers to fibrils. Nat Struct Mol Biol 17, 561-567. Walsh DM, Townsend M, Podlisny MB, Shankar GM, Fadeeva JV, El Agnaf O, Hartley DM, Selkoe DJ (2005) Certain inhibitors of synthetic amyloid beta-peptide (Abeta) fibrillogenesis block oligomerization of natural Abeta and thereby rescue long-term potentiation. J Neurosci 25, 24552462. Tam JH, Pasternak SH (2012) Amyloid and Alzheimer’s disease: Inside and out. Can J Neurol Sci 39, 286-298. Mucke L, Masliah E, Yu GQ, Mallory M, Rockenstein EM, Tatsuno G, Hu K, Kholodenko D, Johnson-Wood K, McConlogue L (2000) High-level neuronal expression of abeta 1-42 in wild-type human amyloid protein precursor transgenic mice: Synaptotoxicity without plaque formation. J Neurosci 20, 4050-4058. Larson ME, Lesne SE (2012) Soluble Abeta oligomer production and toxicity. J Neurochem 120(Suppl 1), 125-139. Sakono M, Zako T (2010) Amyloid oligomers: Formation and toxicity of Abeta oligomers. FEBS J 277, 1348-1358. Laganowsky A, Liu C, Sawaya MR, Whitelegge JP, Park J, Zhao M, Pensalfini A, Soriaga AB, Landau M, Teng PK, Cascio D, Glabe C, Eisenberg D (2012) Atomic view of a toxic amyloid small oligomer. Science 335, 1228-1231. Lacor PN, Buniel MC, Chang L, Fernandez SJ, Gong Y, Viola KL, Lambert MP, Velasco PT, Bigio EH, Finch CE, Krafft GA, Klein WL (2004) Synaptic targeting by Alzheimer’s-related amyloid beta oligomers. J Neurosci 24, 10191-10200. Wang-Dietrich L, Funke SA, Kuhbach K, Wang K, Besmehn A, Willbold S, Cinar Y, Bannach O, Birkmann E, Willbold D (2013) The smyloid-beta oligomer count in cerebrospinal fluid is a biomarker for Alzheimer’s disease. J Alzheimers Dis 34, 985-994. Selkoe DJ (2008) Soluble oligomers of the amyloid betaprotein impair synaptic plasticity and behavior. Behav Brain Res 192, 106-113.

[13]

[14]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

Wang Q, Walsh DM, Rowan MJ, Selkoe DJ, Anwyl R (2004) Block of long-term potentiation by naturally secreted and synthetic amyloid beta-peptide in hippocampal slices is mediated via activation of the kinases c-Jun N-terminal kinase, cyclindependent kinase 5, and p38 mitogen-activated protein kinase as well as metabotropic glutamate receptor type 5. J Neurosci 24, 3370-3378. Cleary JP, Walsh DM, Hofmeister JJ, Shankar GM, Kuskowski MA, Selkoe DJ, Ashe KH (2005) Natural oligomers of the amyloid-beta protein specifically disrupt cognitive function. Nat Neurosci 8, 79-84. Kotilinek LA, Bacskai B, Westerman M, Kawarabayashi T, Younkin L, Hyman BT, Younkin S, Ashe KH (2002) Reversible memory loss in a mouse transgenic model of Alzheimer’s disease. J Neurosci 22, 6331-6335. Nicolakakis N, Hamel E (2010) The nuclear receptor PPARgamma as a therapeutic target for cerebrovascular and brain dysfunction in Alzheimer’s disease. Front Aging Neurosci 2, pii: 21. doi: 10.3389/fnagi.2010.00021 Luconi M, Cantini G, Serio M (2010) Peroxisome proliferator-activated receptor gamma (PPARgamma): Is the genomic activity the only answer? Steroids 75, 585-594. Biessels GJ, Kappelle LJ, Utrecht Diabetic Encephalopathy Study G (2005) Increased risk of Alzheimer’s disease in Type II diabetes: Insulin resistance of the brain or insulin-induced amyloid pathology? Biochem Soc Trans 33, 1041-1044. Escribano L, Simon AM, Gimeno E, Cuadrado-Tejedor M, Lopez de Maturana R, Garcia-Osta A, Ricobaraza A, PerezMediavilla A, Del Rio J, Frechilla D (2010) Rosiglitazone rescues memory impairment in Alzheimer’s transgenic mice: Mechanisms involving a reduced amyloid and tau pathology. Neuropsychopharmacology 35, 1593-1604. Cowley TR, O’Sullivan J, Blau C, Deighan BF, Jones R, Kerskens C, Richardson JC, Virley D, Upton N, Lynch MA (2012) Rosiglitazone attenuates the age-related changes in astrocytosis and the deficit in LTP. Neurobiol Aging 33, 162-175. Toledo EM, Inestrosa NC (2010) Activation of Wnt signaling by lithium and rosiglitazone reduced spatial memory impairment and neurodegeneration in brains of an APPswe/PSEN1DeltaE9 mouse model of Alzheimer’s disease. Mol Psychiatry 15, 272-285. Rodriguez-Rivera J, Denner L, Dineley KT (2011) Rosiglitazone reversal of Tg2576 cognitive deficits is independent of peripheral gluco-regulatory status. Behav Brain Res 216, 255-261. Watson GS, Cholerton BA, Reger MA, Baker LD, Plymate SR, Asthana S, Fishel MA, Kulstad JJ, Green PS, Cook DG, Kahn SE, Keeling ML, Craft S (2005) Preserved cognition in patients with early Alzheimer disease and amnestic mild cognitive impairment during treatment with rosiglitazone: A preliminary study. Am J Geriatr Psychiatry 13, 950-958. Kayed R, Head E, Thompson JL, McIntire TM, Milton SC, Cotman CW, Glabe CG (2003) Common structure of soluble amyloid oligomers implies common mechanism of pathogenesis. Science 300, 486-489. Xu S, Zhou S, Xia D, Xia J, Chen G, Duan S, Luo J (2010) Defects of synaptic vesicle turnover at excitatory and inhibitory synapses in Niemann-Pick C1-deficient neurons. Neuroscience 167, 608-620. Geissler M, Faissner A (2012) A new indirect co-culture set up of mouse hippocampal neurons and cortical astrocytes on microelectrode arrays. J Neurosci Methods 204, 262-272.

S. Xu et al. / Rosiglitazone Prevents Synaptic Impairment [26]

[27]

[28]

[29]

[30]

[31] [32]

[33]

[34] [35]

[36]

Fiala JC, Feinberg M, Popov V, Harris KM (1998) Synaptogenesis via dendritic filopodia in developing hippocampal area CA1. J Neurosci 18, 8900-8911. Di Loreto S, Balestrino M (1997) Development of vulnerability to hypoxic damage in in vitro hippocampal neurons. Int J Dev Neurosci 15, 225-230. Rui Y, Gu J, Yu K, Hartzell HC, Zheng JQ (2010) Inhibition of AMPA receptor trafficking at hippocampal synapses by beta-amyloid oligomers: The mitochondrial contribution. Mol Brain 3, 10. Sanchez PE, Zhu L, Verret L, Vossel KA, Orr AG, Cirrito JR, Devidze N, Ho K, Yu GQ, Palop JJ, Mucke L (2012) Levetiracetam suppresses neuronal network dysfunction and reverses synaptic and cognitive deficits in an Alzheimer’s disease model. Proc Natl Acad Sci U S A 109, E2895-E2903. Ardiles AO, Tapia-Rojas CC, Mandal M, Alexandre F, Kirkwood A, Inestrosa NC, Palacios AG (2012) Postsynaptic dysfunction is associated with spatial and object recognition memory loss in a natural model of Alzheimer’s disease. Proc Natl Acad Sci U S A 109, 13835-13840. Jontes JD, Smith SJ (2000) Filopodia, spines, and the generation of synaptic diversity. Neuron 27, 11-14. Brodbeck J, Balestra ME, Saunders AM, Roses AD, Mahley RW, Huang Y (2008) Rosiglitazone increases dendritic spine density and rescues spine loss caused by apolipoprotein E4 in primary cortical neurons. Proc Natl Acad Sci U S A 105, 1343-1346. De Felice FG, Vieira MN, Bomfim TR, Decker H, Velasco PT, Lambert MP, Viola KL, Zhao WQ, Ferreira ST, Klein WL (2009) Protection of synapses against Alzheimer’s-linked toxins: Insulin signaling prevents the pathogenic binding of Abeta oligomers. Proc Natl Acad Sci U S A 106, 1971-1976. Selkoe DJ (2001) Alzheimer’s disease: Genes, proteins, and therapy. Physiol Rev 81, 741-766. Feinstein DL, Spagnolo A, Akar C, Weinberg G, Murphy P, Gavrilyuk V, Dello Russo C (2005) Receptor-independent actions of PPAR thiazolidinedione agonists: Is mitochondrial function the key? Biochem Pharmacol 70, 177-188. Pardo R, Enguix N, Lasheras J, Feliu JE, Kralli A, Villena JA (2011) Rosiglitazone-induced mitochondrial biogenesis in white adipose tissue is independent of peroxisome

[37]

[38]

[39]

[40]

[41]

[42]

[43]

251

proliferator-activated receptor gamma coactivator-1alpha. PLoS One 6, e26989. Miglio G, Rosa AC, Rattazzi L, Collino M, Lombardi G, Fantozzi R (2009) PPARgamma stimulation promotes mitochondrial biogenesis and prevents glucose deprivationinduced neuronal cell loss. Neurochem Int 55, 496-504. Wang X, Su B, Zheng L, Perry G, Smith MA, Zhu X (2009) The role of abnormal mitochondrial dynamics in the pathogenesis of Alzheimer’s disease. J Neurochem 109(Suppl 1), 153-159. Kimura R, Ma LY, Wu C, Turner D, Shen JX, Ellsworth K, Wakui M, Maalouf M, Wu J (2012) Acute exposure to the mitochondrial complex I toxin rotenone impairs synaptic long-term potentiation in rat hippocampal slices. CNS Neurosci Ther 18, 641-646. Chiang MC, Cheng YC, Lin KH, Yen CH (2013) PPAR␥ regulates the mitochondrial dysfunction in human neural stem cells with tumor necrosis factor alpha. Neuroscience 229, 118129. Gold M, Alderton C, Zvartau-Hind M, Egginton S, Saunders AM, Irizarry M, Craft S, Landreth G, Linnamägi U, Sawchak S (2010) Rosiglitazone monotherapy in mild-to-moderate Alzheimer’s disease: Results from a randomized, doubleblind, placebo-controlled phase III study. Dement Geriatr Cogn Disord 30, 131-146. Harrington C, Sawchak S, Chiang C, Davies J, Donovan C, Saunders AM, Irizarry M, Jeter B, Zvartau-Hind M, van Dyck CH, Gold M (2011) Rosiglitazone does not improve cognition or global function when used as adjunctive therapy to AChE inhibitors in mild-to-moderate Alzheimer’s disease: Two phase 3 studies. Curr Alzheimer Res 8, 592-606. Tzimopoulou S, Cunningham VJ, Nichols TE, Searle G, Bird NP, Mistry P, Dixon IJ, Hallett WA, Whitcher B, Brown AP, Zvartau-Hind M, Lotay N, Lai RY, Castiglia M, Jeter B, Matthews JC, Chen K, Bandy D, Reiman EM, Gold M, Rabiner EA, Matthews PM (2010) A multi-center randomized proof-of-concept clinical trial applying [18 F]FDG-PET for evaluation of metabolic therapy with rosiglitazone XR in mild to moderate Alzheimer’s disease. J Alzheimers Dis 22, 1241-1256.