• Alzheimer's disease;
  • Amyloid-beta;
  • Neuroinflammation;
  • NF-kappaB;
  • Obovatol;
  • Tg2576


  1. Top of page
  2. Abstract
  3. Methods and materials
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

J. Neurochem. (2012) 120, 1048–1059.


Etiology of Alzheimer’s disease (AD) is obscure, but neuroinflammation and accumulation of β-amyloid (Aβ) are implicated in pathogenesis of AD. We have shown anti-inflammatory and neurotrophic properties of obovatol, a biphenolic compound isolated from Magnolia obovata. In this study, we examined the effect of obovatol on cognitive deficits in two separate AD models: (i) mice that received intracerebroventricular (i.c.v.) infusion of Aβ1–42 (2.0 μg/mouse) and (ii) Tg2576 mice-expressing mutant human amyloid precursor protein (K670N, M671L). Injection of Aβ1–42 into lateral ventricle caused memory impairments in the Morris water maze and passive avoidance tasks, being associated with neuroinflammation. Aβ1–42-induced abnormality was significantly attenuated by administration of obovatol. When we analyzed with Tg2576 mice, long-term treatment of obovatol (1 mg/kg/day for 3 months) significantly improved cognitive function. In parallel with the improvement, treatment suppressed astroglial activation, BACE1 expression and NF-κB activity in the transgenic mice. Furthermore, obovatol potently inhibited fibrillation of Aβin vitro in a dose-dependent manner, as determined by Thioflavin T fluorescence and electron microscopic analysis. In conclusion, our data demonstrated that obovatol prevented memory impairments in experimental AD models, which could be attributable to amelioration of neuroinflammation and amyloidogenesis by inhibition of NF-κB signaling pathway and anti-fibrillogenic activity of obovatol.

Abbreviations used



Alzheimer’s disease


amyloid precursor protein


β-site APP-cleaving enzyme




dimethyl sulfoxide


electrophoretic mobility shift assay


glial fibrillary acidic protein


inducible nitric oxide synthase


nuclear factor


non-steroidal anti-inflammatory drug


phosphate-buffered saline


thioflavin T

Amyloid plaque, composed of beta-amyloid (Aβ) is a pathological hallmark of Alzheimer’s disease (AD). Aβ1–42 is a highly neurotoxic and fibrillogenic form of Aβ, and considered as a prime trigger of AD pathogenesis (Glabe 2008). Elevated Aβ1–42 or increase in ratio of Aβ1–42:Aβ1–40 was found in pre-symptomatic familial cases of AD patients (Citron et al. 1992; Cai et al. 1993; Suzuki et al. 1994). Additional copies of the amyloid precursor protein gene cause overproduction of total Aβ, developing early-onset AD (Rovelet-Lecrux et al. 2006). These data indicate Aβ may underlie early neuronal lesions in AD brains.

Sequential cleavage of amyloid precursor protein (APP) by β- and γ-secretase is involved in Aβ. First step for Aβ production is that β-site APP-cleaving enzyme (BACE) 1 cleaves APP to form Aβ N terminus, APPβ and a C-terminal fragment, C99. Next, γ-secretase cut the BACE 1 product to generate Aβs with two variants, Aβ1–40 and Aβ1–42 (Vassar and Citron 2000). In contrast, α-secretase cleaves APP within the Aβ domain to produce APPα and C83, precluding formation of Aβ by competing with BACE1 (Cole and Vassar 2008). Importantly, there is rise in levels of BACE1 and its product (C-terminal fragment of APP) in the sporadic AD brains (Holsinger et al. 2002). Consequently, BACE1 has been considered as a prime therapeutic target for intervention of AD pathogenesis (Fukumoto et al. 2010).

Neuroinflammation has been described as the culprit of AD or, alternatively, as an attempt by the immune system to contain accumulation of Aβ plaques in the brain (Wyss-Coray 2006). Although the role of inflammation in AD is still on debate, accumulating evidence indicates that neuroinflammatory process significantly contributes to pathogenesis of AD (Hwang et al. 2002). The significance of the inflammatory process in AD pathogenesis has been highlighted by epidemiological, retrospective studies reporting a lower incidence of AD in populations receiving long-term treatment with non-steroidal anti-inflammatory drugs (NSAIDs) (Stewart et al. 1997; Walker and Lue 2007; Trepanier and Milgram 2010).

As putative therapeutic tools for AD, various modulators for nuclear factor (NF)-κB signaling pathway have been challenged to modify the neuroinflammatory process in AD models, because NF-κB is a positive regulator for expression of inflammatory molecules including cytokines, inducible nitric oxide synthase (iNOS) and cyclo-oxygenase (COX)-2 (Sastre et al. 2006; Kim et al. 2009; Lee et al. 2009a). Anti-inflammatory drug treatment reduces not only neuroinflammatory reactions, but also amyloid deposition in animal models for AD (Yan et al. 2003; Heneka et al. 2005). Significantly, it has been identified that the BACE1 gene promoter region contains NF-κB binding site (Sambamurti et al. 2004). In parallel with the identification, a study demonstrated that BACE1 mRNA and protein levels are increased by proinflammatory mediators and down-regulated by NSAIDs (Sastre et al. 2006).

Magnolia extract contains at least 255 different ingredients such as alkaloids, coumarins, flavonoids, lignans, neolignans, phenylpropanoids and terpenoids (Lee et al. 2011b). Multiple researchers have focused on pharmacological effects of biphenol-structured neolignans including magnolol, honokiol, 4-O-methylhonokiol and obovatol. For instance, compounds isolated from Magnolia family have been shown to own anti-inflammatory (Munroe et al. 2007), neuroprotective (Lin et al. 2006), and antioxidant properties (Fujita and Taira 1994). Furthermore, a potent anxiolytic property of magnolol and honokiol was demonstrated in several studies (Kuribara et al. 1998; Maruyama et al. 1998). Importantly, Ock et al. (2010)demonstrated that obovatol suppressed lipopolysaccharide-induced microglial activation in vitro and in vivo, and exerted neuroprotective effects against neuroinflammation-induced neurotoxicity. We also showed anti-inflammatory properties of obovatol (Kim et al. 2008; Lee et al. 2009b, 2011b).

We employed two different AD animal models including Tg2576 expressing a human APP variant linked to AD and mice that received intracerebroventricular (i.c.v.) inoculation of Aβ1–42, to examine whether obovatol ameliorates cognitive impairments. It has been described that cognitive impairments in both animal models we used in the current study are associated with neuroinflammatory responses (Yan et al. 2001; Kukar et al. 2007; Kotilinek et al. 2008). Therefore, we hypothesize that the natural compound may attenuate Aβ1–42 formation and decline in memory function through blocking NF-κB activity. Here, we show that the compound stabilizes cognitive functions in both animal models for AD.

Methods and materials

  1. Top of page
  2. Abstract
  3. Methods and materials
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Isolation of obovatol

We isolated obovatol from the leaves of Magnolia obovata and purified (purity: ≥ 95%) as described previously (Kwon et al. 1997). Figure 1 shows the chemical structure of the compound.


Figure 1.  Chemical structure of obovatol.

Download figure to PowerPoint

Animals and treatment

All of the experimental procedures were approved by IACUC of Chungbuk National University (approval number: CBNUA-144-1001-01), and animals were maintained and handled with the humane animal care and use guidelines of NIH.

1–42–infused mice

Two month-old male ICR mice were purchased from Samtako (Osan, Gyeonggi, South Korea). Treatment of obovatol was conducted as described in Fig. 2. Obovatol was dissolved in 100% ethanol, diluted with 20 time–volume of water and administered to ICR mice (p.o.) with three different doses (0.2, 0.5, and 1.0 mg/kg/day for 14 days) prior to intracerebroventricular (i.c.v.) infusion of Aβ1–42. Mice that received vehicle served as control group. In order to prepare for fibrillized form of Aβ1–42 before infusion, Aβ1–42 (purity 97.1%; Sigma-Aldrich, St Louis, MO, USA) was dissolved in sterile saline (200 mg/L) and incubated for 5 days at 37°C. Animals were anesthetized with sodium pentobarbital (50 mg/kg) and placed in a stereotaxic instrument. The prepared Aβ1–42 was injected into the third ventricle with the following coordinates: −1.0 mm anterior/posterior, +1.0 mm medial/lateral and −2.5 mm dorsal/ventral from Bregma (2.0 μg/mouse). Obovatol treatment was continued until animals were killed.


Figure 2.  Timeline depicting treatment of obovatol and Aβ1–42, and assessments of cognitive functions in Aβ1–42 injected (a) and Tg2576 mice (b).

Download figure to PowerPoint

Human APP mutant transgenic mice (Tg2576)

Human AD mutant APP (K670N, M671L) transgenic mice (Tg2576) with C57BL/6/SJL background were obtained from Taconic (Hudson, NY, USA). They have been shown that amyloid plaques and cognitive dysfunction appears when they are about 1-year-old (Hsiao et al. 1996). Nine-month-old Tg2576 mice and non-transgenic littermates were treated with obovatol (1 mg/kg) or vehicle (5% ethanol in water) for 3 months before behavioral tests. Treatment was continued until the mice were killed.

Behavioral Tests

Morris water maze test

The Morris water maze test was performed as previously described with minor modifications (Vorhees and Williams 2006). We used a white circular pool (diameter: 100 cm and height: 35 cm) with a featureless inner surface to perform the Morris water maze examination. The circular pool was filled with non-toxic black-dyed water and kept at 22∼25°C. The pool was divided into four quadrants with equal area. A black platform (diameter: 8 cm, height: 10 cm) was submerged 1 cm below the water surface of one of the quadrants, so that it was invisible at water level. The pool was located in a test room, which contained various prominent visual cues. The swimming time and distance of individual mouse, from the start position to the platform, was monitored and analyzed by a video tracking system (SMART-LD program; Panlab, Barcelona, Spain). Each animal was allowed to learn the location of the hidden platform through daily acquisition tests (three times/day). During the acquisition sessions, the mouse were released at one of three randomly determined locations and allowed to find the hidden platform. The escape latency was defined as the time the animals spent until they found and climbed onto the platform. The maximum trial length was 240 s. When the animals were not able to find the platform within 240 s, the examiner guided the mouse by hand to the platform and an escape latency of 240 s was recorded. We were afraid that fatigue or hyperthermia may compromise the behavioral tests, because of untypically long trial time. Thus, we had them dry with warm towel after they reached to the platform. And we extended the inter-trial time to 60 s. During the interval, the mouse was kept on the escape platform before starting the next trial allowing them to take a rest. When they waited for next trial on the platform, some mice jumped into water pool. There was no significant difference between animal groups as to whether they stayed on the platform during the interval. After then, the animals were placed in the pool again, but at a different location. The mouse was returned to its cage, after the third trial was finished.

A probe trial was conducted 24 h after the last acquisition test to assess the retention of spatial memory. For the test, each mouse was allowed to swim and search the pool for 60 s after the platform was removed from the maze. Consolidated spatial memory was estimated by the time spent in the target quadrant area.

Passive avoidance test

The step-through passive avoidance test was performed as described previously with slight modification (Fujiwara et al. 2009). The step-through passive avoidance apparatus is divided into bright (12 × 10 × 12 cm) and dark chamber (12 × 10 × 12 cm) by a guillotine door. The bright chamber was illuminated by fluorescence light during the tests. Individual animal was placed into the bright compartment and allowed to explore and get used at the new environment prior to training trial. The animal moved freely into the dark compartment through the opened door. The pre-acquisition trial was followed by a training trial in 15 min. For the test, individual mouse was placed in the bright compartment. When mice entered the dark compartment, the door closed and an electrical foot shock (0.4 mA, for 2 s) was delivered. The test trial was conducted 24 h after the training trial. The time each mouse spent in the illuminated compartment before entering the dark compartment was defined as latency for both training and test trials. Maximal latency for the tasks was 300 s.

Biochemical assays

After behavioral tests were finished, animals were killed by carbon dioxide gas. For the biochemical assays, the brains were immediately removed from skull. The cortex and hippocampus were dissected out on ice, and then stored at −80°C until use.

Western blotting

Western blot was performed as depicted in the previous study (Lee et al. 2011a). Briefly, brain tissues were homogenized with lysis buffer and the lysate was centrifuged at 15 000 g for 15 min. Proteins (40 μg) were loaded and electphoresed on a 10 or 15% sodium dodecyl sulfate–polyacrylamide gel, and then transferred to a polyvinylidene difluoride membrane (Hybond ECL; Amersham Pharmacia Biotech Inc., Piscataway, NJ, USA). After blocking in 5% (w/v) non-fat dried milk dissolved in Tris-buffered saline (10 mM Tris, 150 mM NaCl and 0.05% tween-20, pH 7.8), the membrane was incubated with specific primary antibody against APP (1 : 500; Santa Cruz Biotechnology, Santa Cruz, CA, USA), BACE1 (1 : 500; Santa Cruz Biotechnology), C99 (1 : 500; Santa Cruz Biotechnology), Aβ (1 : 1000; Santa Cruz Biotechnology), iNOS (1 : 2000; Upstate), COX-2 (1 : 500; Santa Cruz Biotechnology) or β-actin (1 : 5000; Sigma-Aldrich) for 1 h at 23°C. The blots were then incubated in the corresponding horseradish peroxidase-conjugated anti-rabbit, anti-mouse and anti-goat immunoglobulin G (Santa Cruz Biotechnology). The immunocomplex was detected using the ECL western blotting detection system. The relative density of the protein bands was quantified by densitometry using Electrophoresis Documentation and Analysis System 120 (Eastman Kodak Com., Rochester, NY, USA).

Electrophoretic mobility shift assay

The DNA-binding activity of NF-κB was determined by using electrophoretic mobility shift assay (EMSA). The assay was performed according to the manufacturer’s instruction (Promega, Madison, WI, USA). Briefly, nuclear fraction was obtained from brain tissue lysates by sequential centrifugation with solution A (10 mM HEPES, pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dithiothreitol, 0.2 mM phenyl methyl sulfonyl fluoride) and solution C (solution A, 420 mM NaCl and 20% glycerol). The nuclear extract (2 μg) was hybridized with 32P end-labeled NF-κB oligonucleotide at room temperature for 20 min and the resultant DNA-protein complex was separated on 6% non-denaturing acrylamide gel. The radioactivity was visualized by exposing to film at −70°C. The autoradiograph was analyzed by Image J (NIH, Bethesda, MD, USA).

α-, β- and γ-secretases activity

Activity of each secretase was measured by sercretase assay kits (α-secretase activity kit: R&D systems, Minneapolis, MN, USA; β-secretase assay kit: PANVERA, Madison, WI, USA; and γ-secretase activity kit: R&D systems). For α- or γ-secretase activity, 50 μL of tissue lysate was mixed with 50 μL of reaction buffer and incubated with substrate for 1 h at 37°C avoiding exposure to light. The resultant fluorescence was measured using a fluorometer (excitation at 355 nm and emission at 510 nm) equipped with Felix software (BMG Labtechnologies, Offenburg, Germany). The enzymatic activity was proportional to fluorescence intensity which was expressed as fluorescence units.

For β-secretase assay, 10 μL of brain tissue lysate was mixed with 10 μL of BACE1 substrate (Rh-EVNLDAEFK-Quencher) and incubated for 1 h at room temperature. After the reaction was halted by addition of 10 μL of stop buffer (2.5 M sodium acetate), fluorescence intensity was determined using a fluorometer (excitation at 545 nm and emission at 590 nm) equipped with Felix software (BMG Labtechnologies). Triplicates were run for all controls, blanks and samples.

Measurement of Aβ levels

In order to determine Aβ1–42 and Aβ1–40 levels in the cortex and hippocampus, we used ELISA kits (IBL; Immuno-Biological Co., Ltd, Fujioka, Japan). ELISA was conducted as described in manufacturer’s manual. Briefly, 100 μL of sample was pipetted into the primary coated-plate and was incubated overnight at 4°C. After rinse with washing buffer three times, 100 μL of secondary antibody was added to each well and the mixture was incubated for 1 h at 4°C. After rinse, color was developed by addition of chromogen in the dark. The reaction was stopped and optical density was determined at 450 nm using a microplate absorbance reader (Sunrise™; TECAN, Mannedorf, Switzerland).


After the behavioral tests, animals were perfused with 4% paraformaldehyde [in 0.1 M phosphate-buffered saline (PBS)] under anesthetization of pentobarbital (100 mg/kg). The brains were taken out from skull and post-fixed in 4% paraformaldehyde for 24 h at 4°C. The brains were transferred to 30% sucrose solutions. Subsequently, brains were cut into 40 μm sections by using cryostat microtome (Leica CM1850; Leica Microsystems, Seoul, Korea). After multiple washing in PBS, the sections were incubated with primary antibody to BACE1 (1 : 1000; Sigma-Aldrich) or glial fibrillary acidic protein (GFAP, 1 : 2000; Covance, Berkeley, CA, USA). After washing in PBS, the sections were incubated in biotinylated goat anti rabbit IgG (1 : 2000 dilution; Vector Laboratories, Burlingame, CA, USA) for 1 h at room temperature. Next, the sections were incubated with avidin-conjugated peroxidase complex (ABC kit, 1 : 200; Vector Laboratories) for 60 min followed by PBS washing. The immunocomplex was visualized by using 3, 3′-diaminobenzidine tetrahydrochloride (0.02%) as a chromogen. Finally, the sections were rinsed, mounted on poly-glycine-coated slides and microscopical analysis was performed.

Aβ fibril formation

Inhibitory effect of obovatol on Aβ fibril formation was determined following the method described in the previous study (Lee et al. 2010). Briefly, Aβ1–42 was dissolved in water and centrifuged at 150 000 g with an ultracentrifuge (SCP 85H2; Hitachi, Tokyo, Japan) to remove any aggregates. The concentration of Aβ1–42 solution was 25 mM. 5 mM stock solution of thioflavin T (ThT; Sigma) was prepared by dissolving the chemical in dimethyl sulfoxide (DMSO) and the solution was stored at −20°C. ThT stock solution was diluted in 50 mM glycine–NaOH buffer (pH 8.5) to make 5 μM working solution. Obovatol was dissolved in DMSO to make a stock solution at a concentration of 1 mg/mL. To measure inhibitory effect of obovatol on Aβ fibril formation, solution of Aβ1–42 was incubated in the presence of obovatol (0, 0.05, 0.5, 5 and 10 μg/mL with a final 1% DMSO concentration). The mixture was constantly stirred for 24 h at 37°C. The resultant was mixed with ThT working solution and fluorescence was monitored by a TECAN spectrofluorometer with 1 cm path length quartz cell (TECAN, Grodig, Austria) at 25°C (excitation: 450 nm, emission: 485 nm). The relative fluorescence intensity at 482 nm was determined and used as a measure of the amount of fibrillar aggregates formed in solution. The following equation provides an estimate of Aβ aggregation based on the fluorescence intensities: Aβ aggregation (% of control) = (F2 − F0)/F1 − F0 × 100 (F0, dye alone fluorescence; F1, Aβ fluorescence; F2, Aβ + obovatol fluorescence).

Next, we monitored whether obovatol inhibited the growth of Aβ1–42 fibrils using a LEO 1530 electron microscopy with a 5000–20 000× magnification (Zeiss, Thornwood, NY, USA), as described in the previous study (Lee et al. 2009a). Briefly, Aβ1–42 solution in the presence or absence of obovatol was placed on carbon/formvar-coated grids. 1% phosphotungstic acid was used to stain the grids, and excessive fluid was immediately removed allowing for air-dry. Aβ fibril formation was assessed at 0, 24 and 48 h.

Statistical analysis

All statistical analysis was performed with GraphPad Prism 4 software (Version 4.03; GraphPad software, Inc., San Diego, CA, USA). Group differences in the escape distance, latency, velocity in the Morris water maze task were analyzed using two-way anova with repeated measures, the factors being treatment and testing day. Otherwise were analyzed by one-way anova followed by Dunnette’s post hoc test. All values are presented as mean ± SEM. Significance was set at p < 0.05 for all tests.


  1. Top of page
  2. Abstract
  3. Methods and materials
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Aβ-infused animal model

Effects of obovatol on cognitive impairments

Spatial memory function was determined by the Morris water maze test. The animals were trained for 4 days (3 times/day) prior to Aβ infusion, and their spatial memory was examined for four consecutive days (Fig. 2a). The cognitive function was rated by distance and time to locate the platform. The escape latency was gradually decreased with training sessions without difference among groups (Fig. 3a and b). However, Aβ1–42 infusion significantly increased the latency compared with vehicle injection, while obovatol treatment (0.5 and 1.0 mg/kg) significantly inhibited the effect of Aβ1–42 in a dose-dependent manner (Fig. 3a and b). The impaired memory function was recovered to control level, 3 days after Aβ1–42 infusion (Fig. 3a and b). Treatment of Aβ1–42 or obovatol did not show any effects on the swimming speed during the test (Fig. 3c). Retention of the spatial memory was determined by probe test. The results showed that control animals spent more time in target quadrant than the other quadrants indicating the animals remembered the location of platform (Fig. 3d). In contrast, Aβ1–42-received mice did not have a quadrant preference. In addition, the probe trial clearly showed that Aβ1–42-mediated memory deficit was restored by treatment of obovatol.


Figure 3.  Inhibitory effect of obovatol on Aβ1–42-induced memory impairments. Mice were treated with i.c.v. infusion of Aβ1–42 (2 μg/mouse) after 2 week-long treatment of obovatol. Swimming time (a), swimming distance (b) and average speed (c) to locate on the platform were recorded. (d) The spatial memory was determined by probe test. Quadrant 1 is the target quadrant. 0.5 and 1.0 mg/kg of obovatol show preventative effects 1 day after Aβ1–42 infusion. Only 1.0 mg/kg of obovatol rescues from memory impairment 2 days after Aβ1–42 injection. Values are presented as mean ± SEM from seven mice. *p < 0.05 versus control, **p < 0.01 versus control, #p < 0.05 versus Aβ1–42, ##p < 0.01 versus Aβ1–42, @p < 0.01 versus the other quadrants.

Download figure to PowerPoint

Anti-neuroinflammatory effects of obovatol

1–42 is able to cause neuroinflammation and cognitive impairment because inflammatory molecules such as TNFα and IL-6 can suppress the hippocampal long term potentiation (Tancredi et al. 1992; Braida et al. 2004; Salminen et al. 2009). We speculated that anti-inflammatory property of obovatol might be related to memory improvement in Aβ1–42-infused mice. Therefore, we attempted to detect neuroinflammatory reactions in the mice brains by conducting EMSA for NF-κB and immunostaining for GFAP, a marker for astrocytes. We found that inoculation of Aβ1–42 induced astroglial activation, as their processes became thicker and shorter in comparison to the control (Fig. 4a). In addition, EMSA showed that the amyloid peptide caused significant rise in DNA binding activity of NF-κB (Fig. 4b). In contrast, obovatol treatment alleviated the neuroinflammatory responses caused by Aβ1–42 infusion.


Figure 4.  Anti-neuroinflammatory effects of obovatol in Aβ1–42-infused mice brains. (a) Immunostainings for GFAP, a marker for astrocytes reveal that systemic i.c.v. infusion Aβ1–42 induces astroglial activation in the cortex and hippocampus and the gliosis is relieved by obovatol treatment. (b) DNA-binding activity of NF-κB is measured by EMSA. The assay indicates that i.c.v. injection of Aβ1–42 increases NF-κB activity and obovatol attenuates the activity in dose-dependent fashion. OB, obovatol; scale bar: 25 μm.

Download figure to PowerPoint

Transgenic mice expressing human mutant APP

Effect of obovatol on memory deficits in Tg2576 mice

Tg2576 and non-transgenic littermates were treated with obovatol (1 mg/kg) for 3 months and subject to the Morris water maze and passive avoidance tests. Tests were performed as depicted in Fig. 2b and Methods section. For the Morris water maze test, mice were trained to learn the location of a submerged platform (3 times/day). By the second test, performance of all groups was statistically identical (Fig. 5a and b). However, transgenic mice started to show retardation in obtaining spatial memory from the third test, as escape latency was not shortened in transgenic mice group. Notably, treatment of obovatol almost restored memory function of Tg2576 mice to that of non-transgenic mice (Fig. 5a and b). Twenty hours after the last acquisition session, the probe tests were carried out to reflect memory consolidation. Before the test, the hidden platform was removed and the amount of time animals spent in target quadrant was measured. We observed that vehicle-received Tg2576 mice did not have quadrant preference whereas non-transgenic mice spent more time in target quadrant than the other three quadrants. Importantly, obovatol-treated mice significantly improved retention of spatial memory in the transgenic mice, as the time spent in target quadrant was increased (Fig. 5c).


Figure 5.  Inhibitory effects of obovatol on memory impairments in Tg2576 mice. Tg2576 mice and their wild-type littermates were treated with obovatol (1 mg/kg) or vehicle for 3 months, and then subject to memory tests. The Morris water maze tests and passive avoidance tests were performed as described in Method section. (a) Tg2576 mice take longer time than non-transgenic mice to find the hidden platform from the third trial. (b) Tg2576 mice swim longer distance than non-transgenic mice to locate on the platform. (c) Spatial memory was determined by the probe test. Quadrant 1 is the target quadrant. (d) Passive avoidance test shows Tg2576 mice have decreased the latency to enter the dark compartment in comparison to non-transgenic mice. The memory deficits in Tg2576 mice were attenuated by obovatol treatment (1 mg/kg). Values are presented as mean ± SE from 10 mice. OB, obovatol, *p < 0.05 versus control, **p < 0.01 versus control, #p < 0.05 versus Tg2576, ##p < 0.01 versus Tg2576, @p < 0.05 versus the other quadrants.

Download figure to PowerPoint

To determine whether treatment of obovatol improve the contextual memory in Tg2576, we performed step-through passive-avoidance tests. Animals showed statistically identical performance at the training sessions regardless of genotypes and treatments. In contrast, Tg2576 mice showed a significant impairment in memory retention compared with non-transgenic control at the test session (Fig. 5d). Importantly, the reduction in Tg2576 mice memory retention was relieved by obovatol treatment.

Effects of obovatol on Aβ generation

In an attempt to determine the effect of obovatol treatment on Aβ formation in Tg2576 mouse brain, we measured the level of Aβ1–42 in the brain by using a sensitive ELISA. In the hippocampus and cortex of transgenic mice, significant elevated concentration of Aβ1–42 was detected as compared with non-transgenic littermates (Fig. 6a and b). Long-term oral treatment of obovatol ameliorated the rise in Aβ1–42 concentration in the cortex and hippocampus of transgenic mice. In consistent with the observations, we found that β- and γ-secretase activities are higher in the transgenic cortex and hippocampus than wild type, and administration of obovatol significantly diminished the increase in β- and γ-secretase activities (Fig. 7b and c). However, there was no significant alteration in α-secretase activity (Fig. 7a). Next, we immunostained for BACE1 and results showed that increased expression of BACE1 in Tg2576 was ameliorated by oral treatment of obovatol (Fig. 8a). We also measured levels of BACE1 and C99 by employing western blot analysis. BACE1 or C99 expression was greater in the hippocampus and cortex of Tg2576 mice than wild-type mice (Fig. 8b). Treatment of obovatol significantly attenuated the elevation in Tg2576.


Figure 6.  Inhibitory effects of obovatol on Aβ1–42 formation in Tg2576 mice. The level of Aβ1–42 in the hippocampus and cortex of Tg2576 mice was measured by ELISA. (a) Aβ1–42 concentration is markedly higher in Tg2576 hippocampus than in non-transgenic hippocampus, which is significantly lowered by administration of obovatol. (b) Tg2576 cortex contains greater concentration of Aβ1–42 than non-transgenic cortex, which is significantly decreased by administration of obovatol. OB, obovatol, *p < 0.05 versus control, **p < 0.01 versus control, #p < 0.05 versus Tg2576, ##p < 0.01 versus Tg2576.

Download figure to PowerPoint


Figure 7.  Effects of obovatol on secretase activity. Obovatol (1 mg/kg) was treated with Tg2576 mice and their non-transgenic littermates for 3 months. α-, β- and γ-secretases activities were measured by using enzyme assay kit for each secretase. (a) There is no significant difference in α-secretase activity between Tg2576 and non-transgenic mice. (b) β-Secretase activity is elevated in both cortex and hippocampus of Tg2576 mice compared with wild type mice, and the elevated activity was significantly attenuated by obovatol. (c) γ-Secretase activity is up-regulated in both cortex and hippocampus in comparison to non-transgenic mice, and it is ameliorated by obovatol. *p < 0.05 versus control, **p < 0.01 versus control, #p < 0.05 versus Tg2576, ##p < 0.01 versus Tg2576.

Download figure to PowerPoint


Figure 8.  Inhibitory effects of obovatol on BACE1 and C99 expression in Tg2576 mice. (a) Immunostainings show the increased immunoreactivity for BACE1 in Tg2576 brain, and obovatol relieves the Immunoreactivity. (b) Elevated expressions of BACE1 and C99 are detected in Tg2576 brains. Obovatol attenuates the elevated production of BACE1 and C99. OB, obovatol; *p < 0.05 versus control, ***p < 0.001 versus control, #p < 0.05 versus Tg2576, ##p < 0.01 versus Tg2576. Scale bar = 50 μm.

Download figure to PowerPoint

Effects of obovatol on neuroinflammatory reactions

It has been described that glial cells-mediated neuroinflammation occurs in Tg2576 mice (Benzing et al. 1999). GFAP, an astrocyte-specific filament protein is up-regulated when astrocytes are activated. We immunostained for astrocytes and examined whether obovatol treatment suppresses activation of astrocytes in Tg2576 mice. Activation of astrocytes was observed in the cortex and hippocampus of Tg2576, as shown by GFAP-immunoreactive cells with thick and short processes in the brain regions (Fig. 9a). The number of activated astroglial cells was lower in obovatol-treated Tg2576 brains than vehicle-treated Tg2576.


Figure 9.  Anti-neuroinflammatory effects of obovatol in Tg2576 mice brains. (a) Immunostainings for GFAP show astroglial activation in the cortex and hippocampus of Tg2576 mice and the gliosis is relieved by obovatol treatment. (b) Western blot analysis reveals up-regulation of inflammatory proteins including iNOS and COX-2 in the transgenic mice brains, which is attenuated by obovatol treatment. (c) EMSA indicates that NF-κB activity is increased in transgenic mice brains and obovatol attenuates the activity. OB, obovatol, *p < 0.05 versus control, **p < 0.01 versus control, #p < 0.05 versus Tg2576. Scale bar: 25 μm.

Download figure to PowerPoint

In agreement with the immunostainings, elevated expression of proinflammatory proteins including iNOS and COX2 was detected in the transgenic mice brains compared with wild-type mice, as determined by immunoblottings (Fig. 9b). Significantly, EMSA showed that DNA binding activity of NF-κB was higher in the cortex and hippocampus of Tg2576 mice than in non-transgenic mice (Fig. 9c). This increased level of inflammatory molecules was attenuated by administration of obovatol.

Anti-fibrillation effect of obovatol

We assumed that stabilization of cognitive function by obovatol might be partially achieved by its anti-fibrillation property. Thus, we examined whether obovatol inhibits formation of Aβ1–42 fibril in vitro. The fibril formation was monitored by ThT fluorescence associated with fibrils. When we measured ThT fluorescence intensity in the absence of obovatol, Aβ1–42 fibrils were readily generated by the incubation of Aβ1–42 (20 μM) at pH 7.5 and 37°C (Fig. 10a). Low concentration of obovatol (0.05 and 0.5 μg/mL) did not show significant inhibition of the fibril generation. However, 5.0 and 50 μg/mL of obovatol significantly suppressed Aβ1–42 fibrillogenesis (Fig. 10a). The fibrillation was decreased up to 25% of the control level by 50 μg/mL of obovatol.


Figure 10.  Inhibitory effect of obovatol on fibrillation of Aβ1–42. Aβ1–42 was incubated at 37°C for 24 h in the presence or absence of obovatol. (a) Fibril formation is detected by the ThT fluorescence. (b) Formation of fibrillar material is confirmed by electron microscopy. OB = obovatol, *p < 0.05 versus control, **p < 0.01 versus control. Scale bar = 1 μm.

Download figure to PowerPoint

In an attempt to confirm that obovatol inhibits fibrillation of Aβ1–42, we obtained electron microscope images of fibrillar Aβ1–42 in the absence and presence of obovatol. Typical fibrillar structures were detected as shown in Fig. 10b. In the absence obovatol, fibrils of Aβ1–42 grew markedly. In contrast, very little fibrillar material was formed in the presence of obovatol (Fig. 10b).


  1. Top of page
  2. Abstract
  3. Methods and materials
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Alzheimer’s disease is the most common cause of dementia, but etiology of the neurodegenerative disease remains unclear. There is no effective way to cure the disorder or stop the neurodegenerative processes. Thus, much research has been focused on discovering how to intervene AD neurodegeneration. In this study, we examined whether obovatol is able to mitigate memory deficits in AD animal models. This study showed that obovatol improved cognitive function of the animals in the cued and contextual memory tests. Importantly, obovatol ameliorated neuroinflammatory responses and Aβ formation in the cortex and hippocampus of the animal models, and such pharmacological effects might be related to the memory stabilization. Our data also showed that treatment of obovatol suppressed activation of NF-κB, which might contribute to anti-inflammatory and anti-amyloidogenic effects of obovatol, because the transcription factor is a positive regulator for inflammation as well as BACE1 expression (Chen et al. 2011).

Chronic inflammation is associated with a broad spectrum of neurodegenerative diseases including AD (Glass et al. 2010). Increased markers for neuroinflammation such as activated glial cells, proinflammatory cytokines and chemokines are found in or near the pathologic lesions of AD (Wyss-Coray 2006). Furthermore, neuroinflammatory reaction is detected in AD animal models such as Tg2576 and Aβ-infused mice (Benzing et al. 1999; Dong et al. 2011). Critical role of neuroinflammation for AD pathogenesis was highlighted by an epidemiological study where NSAIDs reduce AD incidence by an average of 58% (Szekely et al. 2004). NSAIDs also have been shown to reduce levels of highly amyloidogenic Aβ1–42 peptide and Aβ deposition in a mouse model of AD (Lim et al. 2000). In agreement with the notions, obovatol treatment significantly attenuated neuroinflammation and this effect was accompanied by the improved cognitive performance. Thus, anti-inflammatory compounds including obovatol might be potential therapeutic tools for AD neurodegeneration.

There was rise in escape distance and escape latency 24 and 48 h after Aβ1–42 peptide infusion. Somebody may argue that the impairments come from sensorimotor deficit caused by acute effects of surgery. However, we did not observe any sensorimotor deficits in the animals during the tests. Furthermore, Aβ1–42-infused animals exhibited impairments in spatial memory as determined by probe test (Fig. 3). We assume that the Aβ1–42-induced increase in escape latency is related to neuroinflammation. Neuroinflammatory reactivity was shown by immunostaining for GFAP and increased DNA-binding activity of NF-κB (Fig. 4). Ameliorated neuroinflammation by obovatol treatment was associated with restoration of cognitive performance. In supportive to this study, multiple other studies demonstrated positive correlation between Aβ-mediated neuroinflammation and memory deficit (Craft et al. 2006; Ralay Ranaivo et al. 2006; Medeiros et al. 2010). It is not clear which inflammatory pathway is responsible for the deterioration as of now, but Medeiros et al. demonstrated that TNF-alpha signalling is related to cognitive decline induced by Aβ. Thus, further study is required to elucidate which specific pathway causes Aβ-induced cognitive impairments.

A variety of Aβ aggregates including dimers, soluble oligomers, protofibrils, diffuse plaques, and fibrillar deposits can be seen in the senile plaques (Caughey and Lansbury 2003). It is believed that the fibrillar form of Aβ peptide is generated in the early stages of AD (Drouet et al. 2000), which causes neuronal dysfunction and death (Yankner 1996). In support of the suggestion, neurotoxicity of Aβ peptide has been demonstrated (Emre et al. 1992). In this study, we showed that i.c.v. infusion of Aβ1–42 fibril causes memory impairments. Moreover, we detected that treatment of Tg2576 mice with obovatol significantly lowered cognitive deterioration and this effect was coincided with 35% and 25% reduction in Aβ1–42 concentration in the hippocampus and cortex, respectively. Aβ1–42 is derived from APP by sequential actions of BACE1 and γ-secretase. BACE1 is essential for initiation of Aβ1–42 generation and the enzyme expression is up-regulated in AD brains (Holsinger et al. 2002; Sastre et al. 2006; Cole and Vassar 2008). It is likely that BACE1 expression is regulated by neuroinflammatory events (Sastre et al. 2008). Importantly, NF-κB controls expression of BACE1, and activation of the inflammatory transcription factor enhances Aβ formation (Sambamurti et al. 2004; Chen et al. 2011). Therefore, therapeutic tools targeting NF-κB would likely be of benefit in the treatment of AD (Sastre et al. 2006). There is abundant evidence that NF-κB activity is associated with amyloidogenesis. For instance, (−)-epigallocatechin-3-gallate, a compound from green tea rescues cognitive function and reduces β-secretase activity through inhibition of NF-κB pathway in preseniline 2 mutant mice (Lee et al. 2009a). Further support comes from a report that the mutation on the NF-κB binding site of BACE1 promoter decreases the promoter activity resulting in reduced expression of BACE1 (Bourne et al. 2007). In this investigation, we observed that obovatol suppressed DNA-binding activity of NF-κB and Aβ contents in the brains of Tg2576 mice (Figs 6 and 9), suggesting that obovatol may reduce amyloidogenesis by way of blunting NF-κB signaling pathway.

This study showed that obovatol (0.5–50 μg/mL) can inhibit Aβ aggregation or promote its destabilization. Moreover, electron microscopy scanning also strongly supports this notion, as suppression of Aβ fibril generation by obovatol was directly visualized. We and others have shown that natural compounds such as (−)-epigallocatechin-3-gallate and polyphenols had an anti-fibrillogenic property (Ehrnhoefer et al. 2008; Riviere et al. 2008; Lee et al. 2009a). Obovatol seems to have a comparable inhibitory effect with such reported compounds. It has been described that memory deficits in middle-aged Tg2576 mice are caused by extracellular deposition of 56 kDa soluble Aβ assembly (Lesne et al. 2006). Oral intake of obovatol reduced memory impairments of Tg2576 mice. Thus, it can be assumed that attenuation of Aβ production and aggregation by obovatol rescued from cognitive deterioration in Tg2576 mice.

We observed robust ameliorative effects of obovatol in the passive avoidance tests (Fig. 5d). Anti-inflammatory property of obovatol might contribute to the results at least in part. However, it is possible that potential effects of obovatol on anxiety-like behavior have contributed to the observed cognitive performance. Compounds isolated from Magnolia extract such as 4-O-methylhonokiol, honokiol and magnolol were shown to have anxiolytic properties (Kuribara et al. 1998; Maruyama et al. 1998; Han et al. 2011). Interestingly, obovatol was also demonstrated to possess anxiolytic properties (Seo et al. 2007). We could not find any studies that Magnolia extract or obovatol increases anxiety-like behavior. But we cannot rule out the possibility that potential effects of obovatol treatment on anxiety-like behavior contribute to their delayed entrance into the dark.

Magnolol, honokiol and obovatol are major compounds contained in Magnolia officinalis (Shen et al. 2009). They share neolignan structures, and are assumed to have similar pharmacokinetic behaviors. Without pharmacokinetic data, it is hard to predict whether obovatol reaches to the brain to exert its pharmacological effects, but it is noteworthy that concentration of magnolol is fourfold higher in the mouse brain than in plasma when the compound (5 mg/kg) is injected intravenously, indicating that magnolol could pass through the blood–brain barrier, and act on the brain (Tsai et al. 1996). Similar dose of obovatol to that in current study elicits anti-neuroinflammatory (10 mg/kg × 4 days) and anxiolytic effects (0.2, 0.5 and 1.0 mg/kg), supporting that obovatol is able to reach to the brain and act on the central nervous system (Ma et al. 2009; Ock et al. 2010). In the previous study, we have shown that extract of Magnolia officinalis (5 and 10 mg/kg for 1 week) protects against scopolamine-induced memory impairments (Lee et al. 2009c). It is assumed that the extract contains about 0.1–0.3% of obovatol (Lee et al. 2011b). Based on the data, we tried 0.03 mg/kg dose of obovatol in preliminary studies. However, we did not detect significant protective effect on memory deficits in animal models. Thus, we increased the dose up to 1 mg/kg for further studies. It was described that magnolia bark extract did not cause any toxic effects in rats treated with two different doses of 480 mg/kg for 21 days and 240 mg/kg for 90 days (Liu et al. 2007). Furthermore, the oral LD50 for the extract is higher than 50 g/kg in mice, and treatment mice with magnolia extract (2500 mg/kg) did not cause any clinical signs of toxicity or mortality (Li et al. 2007). These data indicate that obovatol could be safe and effective to apply to clinical studies.

Overall, these results showed that obovatol isolated from Magnolia officinalis unequivocally attenuates AD-like abnormalities including cognitive impairments and neuroinflammatory reactions in AD animal models suggesting this compound could be eligible for intervening neuronal dysfunction in AD brains.


  1. Top of page
  2. Abstract
  3. Methods and materials
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This work was supported by the National Research Foundation Grant (MRC, 2010-0029480), Priority Research Center Program (2011-0031403), and funded by the Ministry of Education, Science and Technology of Korea. We disclose that there is no conflict of interest.


  1. Top of page
  2. Abstract
  3. Methods and materials
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Benzing W. C., Wujek J. R., Ward E. K., Shaffer D., Ashe K. H., Younkin S. G. and Brunden K. R. (1999) Evidence for glial-mediated inflammation in aged APP(SW) transgenic mice. Neurobiol. Aging 20, 581589.
  • Bourne K. Z., Ferrari D. C., Lange-Dohna C., Rossner S., Wood T. G. and Perez-Polo J. R. (2007) Differential regulation of BACE1 promoter activity by nuclear factor-kappaB in neurons and glia upon exposure to beta-amyloid peptides. J. Neurosci. Res. 85, 11941204.
  • Braida D., Sacerdote P., Panerai A. E., Bianchi M., Aloisi A. M., Iosue S. and Sala M. (2004) Cognitive function in young and adult IL (interleukin)-6 deficient mice. Behav. Brain Res. 153, 423429.
  • Cai X. D., Golde T. E. and Younkin S. G. (1993) Release of excess amyloid beta protein from a mutant amyloid beta protein precursor. Science 259, 514516.
  • Caughey B. and Lansbury P. T. (2003) Protofibrils, pores, fibrils, and neurodegeneration: separating the responsible protein aggregates from the innocent bystanders. Annu. Rev. Neurosci. 26, 267298.
  • Chen C. H., Zhou W., Liu S., Deng Y., Cai F., Tone M., Tone Y., Tong Y. and Song W. (2011) Increased NF-kappaB signalling up-regulates BACE1 expression and its therapeutic potential in Alzheimer’s disease. Int. J. Neuropsychopharmacol. 15, 7790.
  • Citron M., Oltersdorf T., Haass C., McConlogue L., Hung A. Y., Seubert P., Vigo-Pelfrey C., Lieberburg I. and Selkoe D. J. (1992) Mutation of the beta-amyloid precursor protein in familial Alzheimer’s disease increases beta-protein production. Nature 360, 672674.
  • Cole S. L. and Vassar R. (2008) The role of amyloid precursor protein processing by BACE1, the beta-secretase, in Alzheimer disease pathophysiology. J. Biol. Chem. 283, 2962129625.
  • Craft J. M., Watterson D. M. and Van Eldik L. J. (2006) Human amyloid beta-induced neuroinflammation is an early event in neurodegeneration. Glia 53, 484490.
  • Dong Y. F., Kataoka K., Tokutomi Y. et al. (2011) Perindopril, a centrally active angiotensin-converting enzyme inhibitor, prevents cognitive impairment in mouse models of Alzheimer’s disease. FASEB J. 25, 29112920.
  • Drouet B., Pincon-Raymond M., Chambaz J. and Pillot T. (2000) Molecular basis of Alzheimer’s disease. Cell. Mol. Life Sci. 57, 705715.
  • Ehrnhoefer D. E., Bieschke J., Boeddrich A., Herbst M., Masino L., Lurz R., Engemann S., Pastore A. and Wanker E. E. (2008) EGCG redirects amyloidogenic polypeptides into unstructured, off-pathway oligomers. Nat. Struct. Mol. Biol. 15, 558566.
  • Emre M., Geula C., Ransil B. J. and Mesulam M. M. (1992) The acute neurotoxicity and effects upon cholinergic axons of intracerebrally injected beta-amyloid in the rat brain. Neurobiol. Aging 13, 553559.
  • Fujita S. and Taira J. (1994) Biphenyl compounds are hydroxyl radical scavengers: their effective inhibition for UV-induced mutation in Salmonella typhimurium TA102. Free Radic. Biol. Med. 17, 273277.
  • Fujiwara H., Tabuchi M., Yamaguchi T. et al. (2009) A traditional medicinal herb Paeonia suffruticosa and its active constituent 1,2,3,4,6-penta-O-galloyl-beta-D-glucopyranose have potent anti-aggregation effects on Alzheimer’s amyloid beta proteins in vitro and in vivo. J. Neurochem. 109, 16481657.
  • Fukumoto H., Takahashi H., Tarui N. et al. (2010) A noncompetitive BACE1 inhibitor TAK-070 ameliorates Abeta pathology and behavioral deficits in a mouse model of Alzheimer’s disease. J. Neurosci. 30, 1115711166.
  • Glabe C. G. (2008) Structural classification of toxic amyloid oligomers. J. Biol. Chem. 283, 2963929643.
  • Glass C. K., Saijo K., Winner B., Marchetto M. C. and Gage F. H. (2010) Mechanisms underlying inflammation in neurodegeneration. Cell 140, 918934.
  • Han H., Jung J. K., Han S. B., Nam S. Y., Oh K. W. and Hong J. T. (2011) Anxiolytic-like effects of 4-O-methylhonokiol isolated from Magnolia officinalis through enhancement of GABAergic transmission and chloride influx. J. Med. Food 14, 724731.
  • Heneka M. T., Sastre M., Dumitrescu-Ozimek L. et al. (2005) Acute treatment with the PPARgamma agonist pioglitazone and ibuprofen reduces glial inflammation and Abeta1-42 levels in APPV717I transgenic mice. Brain 128, 14421453.
  • Holsinger R. M., McLean C. A., Beyreuther K., Masters C. L. and Evin G. (2002) Increased expression of the amyloid precursor beta-secretase in Alzheimer’s disease. Ann. Neurol. 51, 783786.
  • Hsiao K., Chapman P., Nilsen S., Eckman C., Harigaya Y., Younkin S., Yang F. and Cole G. (1996) Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice. Science 274, 99102.
  • Hwang D. Y., Chae K. R., Kang T. S. et al. (2002) Alterations in behavior, amyloid beta-42, caspase-3, and Cox-2 in mutant PS2 transgenic mouse model of Alzheimer’s disease. FASEB J. 16, 805813.
  • Kim J. C., Lee S. Y., Kim S. Y. et al. (2008) HSV-1 ICP27 suppresses NF-kappaB activity by stabilizing IkappaBalpha. FEBS Lett. 582, 23712376.
  • Kim T. I., Lee Y. K., Park S. G. et al. (2009) l-Theanine, an amino acid in green tea, attenuates beta-amyloid-induced cognitive dysfunction and neurotoxicity: reduction in oxidative damage and inactivation of ERK/p38 kinase and NF-kappaB pathways. Free Radic. Biol. Med. 47, 16011610.
  • Kotilinek L. A., Westerman M. A., Wang Q. et al. (2008) Cyclooxygenase-2 inhibition improves amyloid-beta-mediated suppression of memory and synaptic plasticity. Brain 131, 651664.
  • Kukar T., Prescott S., Eriksen J. L., Holloway V., Murphy M. P., Koo E. H., Golde T. E. and Nicolle M. M. (2007) Chronic administration of R-flurbiprofen attenuates learning impairments in transgenic amyloid precursor protein mice. BMC Neurosci. 8, 54.
  • Kuribara H., Stavinoha W. B. and Maruyama Y. (1998) Behavioural pharmacological characteristics of honokiol, an anxiolytic agent present in extracts of Magnolia bark, evaluated by an elevated plus-maze test in mice. J. Pharm. Pharmacol. 50, 819826.
  • Kwon B. M., Ro S. H., Kim M. K., Nam J. Y., Jung H. J., Lee I. R., Kim Y. K. and Bok S. H. (1997) Polyacetylene analogs, isolated from hairy roots of Panax ginseng, inhibit Acyl-CoA : cholesterol acyltransferase. Planta Med. 63, 552553.
  • Lee J. W., Lee Y. K., Ban J. O., Ha T. Y., Yun Y. P., Han S. B., Oh K. W. and Hong J. T. (2009a) Green tea (-)-epigallocatechin-3-gallate inhibits beta-amyloid-induced cognitive dysfunction through modification of secretase activity via inhibition of ERK and NF-kappaB pathways in mice. J. Nutr. 139, 19871993.
  • Lee S. Y., Cho J. S., Yuk D. Y. et al. (2009b) Obovatol enhances docetaxel-induced prostate and colon cancer cell death through inactivation of nuclear transcription factor-kappaB. J. Pharmacol. Sci. 111, 124136.
  • Lee Y. K., Yuk D. Y., Kim T. I., Kim Y. H., Kim K. T., Kim K. H., Lee B. J., Nam S. Y. and Hong J. T. (2009c) Protective effect of the ethanol extract of Magnolia officinalis and 4-O-methylhonokiol on scopolamine-induced memory impairment and the inhibition of acetylcholinesterase activity. J. Nat. Med. 63, 274282.
  • Lee J. W., Lee Y. K., Lee B. J., Nam S. Y., Lee S. I., Kim Y. H., Kim K. H., Oh K. W. and Hong J. T. (2010) Inhibitory effect of ethanol extract of Magnolia officinalis and 4-O-methylhonokiol on memory impairment and neuronal toxicity induced by beta-amyloid. Pharmacol. Biochem. Behav. 95, 3140.
  • Lee Y. J., Choi D. Y., Choi I. S., Han J. Y., Jeong H. S., Han S. B., Oh K. W. and Hong J. T. (2011a) Inhibitory effect of a tyrosine-fructose Maillard reaction product, 2,4-bis(p-hydroxyphenyl)-2-butenal on amyloid-beta generation and inflammatory reactions via inhibition of NF-kappaB and STAT3 activation in cultured astrocytes and microglial BV-2 cells. J. Neuroinflammation 8, 132.
  • Lee Y. J., Lee Y. M., Lee C. K., Jung J. K., Han S. B. and Hong J. T. (2011b) Therapeutic applications of compounds in the Magnolia family. Pharmacol. Ther. 130, 157176.
  • Lesne S., Koh M. T., Kotilinek L., Kayed R., Glabe C. G., Yang A., Gallagher M. and Ashe K. H. (2006) A specific amyloid-beta protein assembly in the brain impairs memory. Nature 440, 352357.
  • Li N., Song Y., Zhang W., Wang W., Chen J., Wong A. W. and Roberts A. (2007) Evaluation of the in vitro and in vivo genotoxicity of magnolia bark extract. Regul. Toxicol. Pharmacol. 49, 154159.
  • Lim G. P., Yang F., Chu T. et al. (2000) Ibuprofen suppresses plaque pathology and inflammation in a mouse model for Alzheimer’s disease. J. Neurosci. 20, 57095714.
  • Lin Y. R., Chen H. H., Ko C. H. and Chan M. H. (2006) Neuroprotective activity of honokiol and magnolol in cerebellar granule cell damage. Eur. J. Pharmacol. 537, 6469.
  • Liu Z., Zhang X., Cui W., Li N., Chen J., Wong A. W. and Roberts A. (2007) Evaluation of short-term and subchronic toxicity of magnolia bark extract in rats. Regul. Toxicol. Pharmacol. 49, 160171.
  • Ma H., Jo Y. J., Ma Y., Hong J. T., Kwon B. M. and Oh K. W. (2009) Obovatol isolated from Magnolia obovata enhances pentobarbital-induced sleeping time: possible involvement of GABAA receptors/chloride channel activation. Phytomedicine 16, 308313.
  • Maruyama Y., Kuribara H., Morita M., Yuzurihara M. and Weintraub S. T. (1998) Identification of magnolol and honokiol as anxiolytic agents in extracts of saiboku-to, an oriental herbal medicine. J. Nat. Prod. 61, 135138.
  • Medeiros R., Figueiredo C. P., Pandolfo P., Duarte F. S., Prediger R. D., Passos G. F. and Calixto J. B. (2010) The role of TNF-alpha signaling pathway on COX-2 upregulation and cognitive decline induced by beta-amyloid peptide. Behav. Brain Res. 209, 165173.
  • Munroe M. E., Arbiser J. L. and Bishop G. A. (2007) Honokiol, a natural plant product, inhibits inflammatory signals and alleviates inflammatory arthritis. J. Immunol. 179, 753763.
  • Ock J., Han H. S., Hong S. H., Lee S. Y., Han Y. M., Kwon B. M. and Suk K. (2010) Obovatol attenuates microglia-mediated neuroinflammation by modulating redox regulation. Br. J. Pharmacol. 159, 16461662.
  • Ralay Ranaivo H., Craft J. M., Hu W., Guo L., Wing L. K., Van Eldik L. J. and Watterson D. M. (2006) Glia as a therapeutic target: selective suppression of human amyloid-beta-induced upregulation of brain proinflammatory cytokine production attenuates neurodegeneration. J. Neurosci. 26, 662670.
  • Riviere C., Richard T., Vitrac X., Merillon J. M., Valls J. and Monti J. P. (2008) New polyphenols active on beta-amyloid aggregation. Bioorg. Med. Chem. Lett. 18, 828831.
  • Rovelet-Lecrux A., Hannequin D., Raux G. et al. (2006) APP locus duplication causes autosomal dominant early-onset Alzheimer disease with cerebral amyloid angiopathy. Nat. Genet. 38, 2426.
  • Salminen A., Ojala J., Kauppinen A., Kaarniranta K. and Suuronen T. (2009) Inflammation in Alzheimer’s disease: amyloid-beta oligomers trigger innate immunity defence via pattern recognition receptors. Prog. Neurobiol. 87, 181194.
  • Sambamurti K., Kinsey R., Maloney B., Ge Y. W. and Lahiri D. K. (2004) Gene structure and organization of the human beta-secretase (BACE) promoter. FASEB J. 18, 10341036.
  • Sastre M., Dewachter I., Rossner S. et al. (2006) Nonsteroidal anti-inflammatory drugs repress beta-secretase gene promoter activity by the activation of PPARgamma. Proc. Natl. Acad. Sci. USA 103, 443448.
  • Sastre M., Walter J. and Gentleman S. M. (2008) Interactions between APP secretases and inflammatory mediators. J. Neuroinflammation 5, 25.
  • Seo J. J., Lee S. H., Lee Y. S., Kwon B. M., Ma Y., Hwang B. Y., Hong J. T. and Oh K. W. (2007) Anxiolytic-like effects of obovatol isolated from Magnolia obovata: involvement of GABA/benzodiazepine receptors complex. Prog. Neuropsychopharmacol. Biol. Psychiatry 31, 13631369.
  • Shen C. C., Ni C. L., Shen Y. C., Huang Y. L., Kuo C. H., Wu T. S. and Chen C. C. (2009) Phenolic constituents from the stem bark of Magnolia officinalis. J. Nat. Prod. 72, 168171.
  • Stewart W. F., Kawas C., Corrada M. and Metter E. J. (1997) Risk of Alzheimer’s disease and duration of NSAID use. Neurology 48, 626632.
  • Suzuki N., Cheung T. T., Cai X. D., Odaka A., Otvos Jr L., Eckman C., Golde T. E. and Younkin S. G. (1994) An increased percentage of long amyloid beta protein secreted by familial amyloid beta protein precursor (beta APP717) mutants. Science 264, 13361340.
  • Szekely C. A., Thorne J. E., Zandi P. P., Ek M., Messias E., Breitner J. C. and Goodman S. N. (2004) Nonsteroidal anti-inflammatory drugs for the prevention of Alzheimer’s disease: a systematic review. Neuroepidemiology 23, 159169.
  • Tancredi V., D’Arcangelo G., Grassi F., Tarroni P., Palmieri G., Santoni A. and Eusebi F. (1992) Tumor necrosis factor alters synaptic transmission in rat hippocampal slices. Neurosci. Lett. 146, 176178.
  • Trepanier C. H. and Milgram N. W. (2010) Neuroinflammation in Alzheimer’s disease: are NSAIDs and selective COX-2 inhibitors the next line of therapy? J. Alzheimers Dis. 21, 10891099.
  • Tsai T. H., Chou C. J. and Chen C. F. (1996) Pharmacokinetics and brain distribution of magnolol in the rat after intravenous bolus injection. J. Pharm. Pharmacol. 48, 5759.
  • Vassar R. and Citron M. (2000) Abeta-generating enzymes: recent advances in beta- and gamma-secretase research. Neuron 27, 419422.
  • Vorhees C. V. and Williams M. T. (2006) Morris water maze: procedures for assessing spatial and related forms of learning and memory. Nat. Protoc. 1, 848858.
  • Walker D. and Lue L. F. (2007) Anti-inflammatory and immune therapy for Alzheimer’s disease: current status and future directions. Curr. Neuropharmacol. 5, 232243.
  • Wyss-Coray T. (2006) Inflammation in Alzheimer disease: driving force, bystander or beneficial response? Nat. Med. 12, 10051015.
  • Yan J. J., Cho J. Y., Kim H. S., Kim K. L., Jung J. S., Huh S. O., Suh H. W., Kim Y. H. and Song D. K. (2001) Protection against beta-amyloid peptide toxicity in vivo with long-term administration of ferulic acid. Br. J. Pharmacol. 133, 8996.
  • Yan Q., Zhang J., Liu H., Babu-Khan S., Vassar R., Biere A. L., Citron M. and Landreth G. (2003) Anti-inflammatory drug therapy alters beta-amyloid processing and deposition in an animal model of Alzheimer’s disease. J. Neurosci. 23, 75047509.
  • Yankner B. A. (1996) Mechanisms of neuronal degeneration in Alzheimer’s disease. Neuron 16, 921932.