Biological and Pharmaceutical Bulletin
Online ISSN : 1347-5215
Print ISSN : 0918-6158
ISSN-L : 0918-6158
Regular Articles
Nobiletin Reduces Intracellular and Extracellular β-Amyloid in iPS Cell-Derived Alzheimer’s Disease Model Neurons
Junko KimuraKosuke ShimizuKoji KajimaAkihito YokosukaYoshihiro MimakiNaoto OkuYasushi Ohizumi
Author information
Supplementary material

2018 Volume 41 Issue 4 Pages 451-457


Alzheimer’s disease (AD) is the most common cause of dementia, with progressive memory impairment. Recently, neprilysin, a β-amyloid (Aβ)-degrading enzyme has become featured as a drug target for AD. Previously, we identified nobiletin from citrus peels as a natural compound possessing anti-dementia activity. In addition, we demonstrated that nobiletin improved memory in memory-impaired animals and, further, that Aβ levels were markedly decreased in the brains of these animals. We demonstrated in vitro that nobiletin up-regulates neprilysin expression and activity in human neuroblastoma cells. However, the action of nobiletin with regard to Aβ degradation under in vitro AD pathological conditions remains unclear. In this study, we examined whether nobiletin could enhance the degradation of intra- and extracellular Aβ using human induced pluripotent stem cell-derived AD model neurons, which generate an excess of Aβ1–42 due to the familial AD presenilin-1 mutation. The neurons were treated in the presence or absence of nobiletin. The results of real-time quantitative RT-PCR indicated that neprilysin mRNA levels were significantly up-regulated by nobiletin. Furthermore, immunostaining with an anti-Aβ antibody revealed that nobiletin substantially reduced the intraneuronal content of Aβ. Interestingly, the results of Aβ1–42 immunoassays confirmed that nobiletin also significantly decreased the levels of Aβ1–42 released into the cellular medium. These results suggest that nobiletin enhanced the reduction of intra- and that extracellular Aβ levels under AD pathologic conditions, and this is associated with the up-regulation of neprilysin expression. Collectively, nobiletin appears to be a promising novel prophylactic seed drug or functional food for AD.

Alzheimer’s disease (AD) is an age-related neurodegenerative disorder with progressive learning and memory impairments, and is responsible for the majority of dementia cases.1) However, fundamental prophylactic and therapeutic drugs for AD have yet to be developed. One pathological hallmark of AD is β-amyloid (Aβ) deposits in the brain, namely, senile plaques, which are caused by the aggregation of insoluble Aβ fibrils.2) These plaques robustly trigger neurodegeneration; the accumulation of intracellular soluble Aβ and, in particular, Aβ1–42, evokes synaptic dysfunction.3,4) In the brain, physiological Aβ peptides are constantly being generated from the amyloid precursor protein (APP) by β- and γ-secretases; they are concurrently degraded by a group of Aβ-degrading enzymes, including neprilysin.5,6) If the balance between Aβ generation and degradation is disrupted over a long period, it is possible that pathogenic Aβ forms can appear.5,7,8) An increase in Aβ1–42 levels in the brain is strongly correlated with aging.9)

Neprilysin is an enzyme that degrades Aβ soluble monomers and oligomers,10) and it is widely distributed in mammalian brains5,8,11) where it is expressed primarily in neurons.12) In neprilysin-knockout mice, human Aβ1–42 peptide injected into the brain remains intact without being catabolized.13) In addition, it has been reported that a marked decrease in the levels of neprilysin is observed in the hippocampus region of aged mice compared with young mice.14) In the human brain, neprilysin levels are known to decline with age.8) Thus, neprilysin is a potential key target for AD prophylactic and/or therapeutic agents.15)

Previously, we successfully isolated nobiletin, a polymethoxylated flavone from the peels of Citrus depressa HAYATA (Rutaceae), which is mostly cultivated in Okinawa, Japan, as a natural compound having anti-dementia activity. Our previous in vivo studies showed that nobiletin ameliorates memory impairment in animals,1619) including in intracerebroventricular infusion of Aβ1–40 in rats,20) APP-transgenic (APP-SL 7–5 Tg) mice,21) senescence-accelerated mice,17) and triple transgenic AD model (3XTg-AD) mice.22) Of note, nobiletin significantly reduces Aβ deposition in the hippocampus of APP-SL 7–5 Tg mice21) and soluble Aβ1–40 levels in the brain of 3XTg-AD mice.22) However, the mechanism of action of nobiletin for Aβ degradation under AD pathological conditions remains unclear.

We have previously investigated the regulatory effects of nobiletin on intracellular signaling in vitro. These earlier studies demonstrated that nobiletin induces the activation of the protein A (PKA)/extracellular signal-regulated kinase (ERK)/cAMP response element (CRE)-binding protein (CREB) intracellular signaling pathway, and subsequently up-regulates CRE-dependent transcription.16,23) These findings were observed in subclone PC12D of the rat pheochromocytoma cell line PC12. Furthermore, nobiletin elevates the expression of memory- and transcription-related genes in PC12 and PC12D cells.24,25) Interestingly, we recently found that nobiletin also enhances the expression of the neprilysin gene and protein, as well as the activity of the enzyme, in the human neuroblastoma cell line, SK-N-SH.26)

In the present study, we examined whether nobiletin promotes Aβ degradation in vitro. We used two types of induced pluripotent stem (iPS) cell-derived neurons as in vitro models of AD, alongside normal healthy control cells. The AD-type iPS cell harbors the Polish familial AD (FAD) presenilin-1 mutation of proline 117 to leucine (PS1P117L)27) and, consequently, exhibits heightened Aβ1–42 generation after neurodifferentiation. In contrast, the normal-type iPS cell has no mutation in this gene (i.e., PS1WT) and was used as a healthy control. Two types of iPS cells were produced from genetically modified healthy human somatic cells.28,29) Here, for the first time, we demonstrate that nobiletin reduced the levels of both intra- and extracellular Aβ, likely via the up-regulation of neprilysin gene expression, in human iPS cell-derived AD model neurons. These results provide important clues for understanding the pharmacological actions of nobiletin. Collectively, our findings suggest that nobiletin plays a key role in Aβ degradation and may be efficacious in ameliorating AD as a novel seed drug or functional food.


Preparation of Nobiletin

Nobiletin (purity >99%, Fig. 1) was obtained from the peels of C. depressa. The fruits were kindly provided by Dr. Masamichi Yano of the National Institute of Fruit Tree Science in Shizuoka, Japan. The extraction, isolation, and purification of nobiletin were performed as previously described.16,25) The purity of purified nobiletin was confirmed by HPLC and 1H- and 13C-NMR.16) The obtained nobiletin was dissolved in dimethyl sulfoxide (DMSO) (Sigma-Aldrich Co., LLC, St. Louis, MO, U.S.A.) at a concentration of 100 mM and stored at −20°C.

Fig. 1. Chemical Structure of Nobiletin Isolated from the Peels of C. depressa

Cultures of Human iPS Cell-Derived Neurons

Human iPS-PS1P117L and -PS1WT cells were purchased from ReproCELL, Inc. (Tokyo, Japan) and cultured for neurodifferentiation following the manufacturer’s protocol. The cells were seeded at a suitable density for each experimental system in culture dishes or multiwell plates with surfaces coated with 0.002% Poly-L-lysine (PLL; Sigma-Aldrich) and coating solution (ReproCELL). Cells were induced to differentiate into AD or normal model neurons, respectively, by incubation in maturation medium (ReproCELL) containing additive A (ReproCELL), 100 U/mL penicillin, and 100 µg/mL streptomycin (Life Technologies Corp., Carlsbad, CA, U.S.A.) in 5% CO2 at 37°C for 14–21 d.

Neuronal Marker Immunostaining

The iPS-PSP117L and -PSWT cells were seeded at 5.0×104 cells/well in the coated-wells of a 16-well glass chamber slide system (Thermo Fisher Scientific, Inc., Waltham, MA, U.S.A.) and cultured for 14 d for neurodifferentiation. Subsequently, the cultured cells were washed with Dulbecco’s phosphate-buffered saline (DPBS; Life Technologies), fixed with 4% paraformaldehyde (Merck KGaA, Darmstadt, Germany) and rendered permeable with 0.2% Triton X-100 (Sigma-Aldrich). The cells were then blocked with 1% bovine serum albumin (BSA; Sigma-Aldrich) in DPBS for 1 h at room temperature. The cells were then stained for 2 h at room temperature with Alexa 488-conjugated Milli-Mark™ FluoroPan Neuronal Marker (Merck Millipore Corp., Billerica, MA, U.S.A.), which is composed of anti-neuronal nuclei (NeuN), microtubule associated protein 2, βIII-tubulin, and neurofilament heavy (NF-H) antibodies in 1% BSA/DPBS. The nuclei were then counterstained with 1 µg/mL Hoechst33342 (Dojindo Laboratories, Kumamoto, Japan) for 10 min at room temperature. The stained cells were observed by confocal laser microscopy (Nikon Instech Co., Ltd., Tokyo, Japan). The number of the neuronal marker-immunoreactive cells was counted in ten randomly chosen areas (above 600 cells in total). The proportion of differentiated cells was calculated on the basis of the number of Hoechst33342-stained nuclei.

Real-Time Quantitative RT-PCR (qRT-PCR)

For qRT-PCR, the iPS cell-derived AD or normal model neurons were treated in maturation medium contained 3, 10, or 30 µM nobiletin, or 0.1% DMSO as a vehicle control, at a cell density of 3.0×104 cells/well in the coated-wells of a 96-well plate (Becton, Dickinson and Company, Franklin Lakes, NJ, U.S.A.) in a 5% CO2 incubator at 37°C for 24 h. Expression level of each gene was determined by using SYBR® Green Cells-to-CT™ Kits (Applied Biosystems, Foster City, CA, U.S.A.) according to the manufacturer’s protocol. Reverse transcription proceeded at 37°C for 60 min and then at 95°C for 5 min. qRT-PCR was performed under the conditions of 95°C for 10 min followed by 50–60 cycles at 95°C for 15 s and 60°C for 60 s, and then 1 cycle each at 95°C for 15 s, 60°C for 30 s, and 95°C for 15 s in order to generate a dissociation curve. Gene expression was confirmed by using the following purchased primers (TaKaRa Bio, Inc., Shiga, Japan), whose sequences were optimally designed for qRT-PCR by TaKaRa Bio: human neprilysin (MME, NCBI Reference Sequences: NM_000902.3, position: 5037) and human β-actin (Actb, NCBI Reference Sequences: NM_001101.3, position: 1135) as an internal control. Detection was conducted by using a Thermal Cycler Dice® Real-Time System and software (TaKaRa Bio). All data were normalized to β-actin mRNA expression. The relative amount of gene expression was calculated using the 2−ΔΔCt method.30,31)

Immunocytochemical Analysis

The neurons on coated-glass chamber slide were treated with 3, 10, and 30 µM nobiletin, or with 0.1% DMSO, for 24 h, followed by washing with DPBS, fixation with 4% paraformaldehyde, and permeabilization with 0.2% Triton X-100. They were then blocked with 3% BSA/DPBS for 1 h at room temperature. The neurons were assessed by performing an immunocytochemical analysis using the anti-Aβ1–16 (6E10) monoclonal antibody in 1% BSA/DPBS as the primary antibody (1 : 500; Covance, Inc., Princeton, NJ, U.S.A.) and then a fluorescein isothiocyanate (FITC)-conjugated anti-mouse immunoglobulin G (IgG) antibody in 1% BSA/DPBS as the secondary antibody (1 : 100; Millipore). Nuclei were stained with 1 µg/mL Hoechst33342 and neurons were observed by confocal laser microscopy (Carl Zeiss AG, Oberkochen, Germany).

Enzyme-Linked Immunosorbent Assay (ELISA)

The iPS-derived AD or normal model neurons were treated with 3, 10, or 30 µM nobiletin, or 0.1% DMSO as a vehicle control. After 24 h, the neurons (at 7.5×104 cells/well) were lysed by cOmplete™ Lysis-M containing cOmplete™ protease inhibitor cocktail (Roche Diagnostics GmbH, Mannheim, Germany) and their medium supernatants (at 3.6×104 cells/well) were collected, respectively. The levels of Aβ1–42 in the neurons or medium supernatants were measured by using a high sensitive sandwich ELISA with human antibodies [BAN50: anti-human Aβ1–16 and BC05 (Fab’ fragment): anti-human Aβ35–43; Wako Pure Chemical Industries, Ltd., Osaka, Japan]. Absorbance was measured using an absorption spectrometer (Tecan Group, Ltd., Männedorf, Switzerland).

Statistical Analysis

Statistical significance of differences was evaluated by one-way ANOVA, followed by Tukey’s test using the GraphPad Prism 5.0 software (GraphPad Software, Inc., La Jolla, CA, U.S.A.). A level of p<0.05 was considered statistically significant.


Confirmation of iPS Cell Neurodifferentiation by Detection of Neuronal Markers

To examine whether cultured iPS-PSP117L and -PSWT cells differentiated into neurons, we performed immunostaining using a whole neuronal marker cocktail. Confocal laser microscopy revealed that the axons, cell bodies, and nuclei were immunopositive for neuronal markers (Fig. 2A). Two types of iPS cells almost completely differentiated into neurons (Fig. 2B).

Fig. 2. Confirmation of iPS-PS1P117L and -PS1WT Cell Differentiation into Neurons

After 14 d of culture, both types of cells were fixed, made permeable, and then immunostained with a cocktail of antibodies against neuronal markers (green). Nuclei were stained with Hoechst33342 (blue). Neurodifferentiation was observed by using confocal laser microscopy. Scale bar: 100 µm.

Neprilysin Gene Expression Was Up-Regulated by Nobiletin in AD Model Neurons

To determine whether nobiletin could induce the up-regulation of neprilysin gene expression, AD and normal model neurons were treated with 3, 10, or 30 µM nobiletin, or with 0.1% DMSO as a vehicle control, for 24 h. Neprilysin mRNA levels were evaluated using qRT-PCR. As shown in Fig. 3, the basal expression levels of neprilysin gene in the AD model neurons were approximately 2-fold higher than those in the normal model neurons.

Fig. 3. Effect of Nobiletin on the Expression of the Neprilysin Gene in Neurons

The iPS cell-derived AD and normal model neurons were stimulated with nobiletin at 3, 10, or 30 µM, or with 0.1% DMSO as a vehicle control, for 24 h. Then, neprilysin mRNA levels were evaluated by qRT-PCR. The values are presented as the mean±standard deviation (S.D., n=3–4). * p<0.05 vs. vehicle-treated AD model, #p<0.05 and ###p<0.001 vs. vehicle-treated normal model neurons.

In the AD model neurons treated with 10 µM nobiletin, the neprilysin mRNA levels were significantly up-regulated by approximately 1.7-fold (p<0.05) as compared with the vehicle-treated AD model neurons (Fig. 3). Moreover, although not significantly different, neprilysin mRNA levels became approximately 1.3-fold by treatment with 30 µM nobiletin in the AD model neurons. On the other hand, in the normal model neurons, treatment with 10 µM nobiletin caused an increase in neprilysin expression approximately 1.4-fold as compared with that in vehicle-treated them, though the difference was not significant.

Nobiletin Reduced Intracellular Aβ Levels in AD Model Neurons

We performed an immunocytochemical analysis using the anti-Aβ1–16 (6E10) antibody to examine whether nobiletin could reduce the intracellular levels of Aβ. Intense fluorescence signals of intracellular Aβ were detected in the vehicle-treated AD model neurons (Fig. 4A). The fluorescence signal intensity was moderately decreased by 3 µM nobiletin and markedly reduced by 10 and 30 µM nobiletin in the AD model neurons in comparison with that of the vehicle-treated AD model neurons. In contrast, the fluorescence intensity was only minimally detected in the normal model neurons (Fig. 4B).

Fig. 4. A Decrease in Intracellular Aβ Levels in AD Model Neurons after Treatment of Nobiletin

Immunocytochemical analysis of AD (A) and normal (B) model neurons was performed by immunostaining with the Aβ antibody. Both types of neurons were treated with 3, 10, or 30 µM nobiletin for 24 h, and then fixed and permeabilized. The Aβ was visualized using anti-Aβ1–16 (6E10) 1st antibody and FITC-conjugated anti-mouse IgG 2nd antibody (green). Nuclei were stained with Hoechst33342 (blue). The fluorescent signals of Aβ and nuclei were observed by confocal laser microscopy. Scale bar: 20 µm. The levels of Aβ in nobiletin-treated AD model neurons were measured by using Aβ1–42 ELISA system (C). The values are presented as the mean±S.D. (n=3). ** p<0.01 vs. vehicle-treated normal model neurons.

Then, we quantitatively analyzed the reduction levels of intracellular Aβ by using an Aβ1–42 ELISA system. As a result, the levels of Aβ in 10 and 30 µM nobiletin-treated AD model neurons were decreased 0.8-fold and 0.6-fold, respectively, but did not significantly different between the nobiletin-treated and the vehicle-treated neurons (Fig. 4C).

Nobiletin Reduced Extracellular Levels of Aβ1–42 Released from AD Model Neurons

To examine the effect of nobiletin on extracellular Aβ1–42, we used an Aβ1–42 ELISA system to measure the Aβ1–42 released into the medium. The vehicle-treated AD model neurons released approximately two-fold more Aβ1–42 than the vehicle-treated normal model neurons (p<0.001; Fig. 5). In the 30 µM nobiletin-treated AD model neurons, Aβ1–42 levels were significantly decreased (0.7-fold, p<0.001; Fig. 5) compared with those for the vehicle-treated group.

Fig. 5. Measurement of Extraneuronal Aβ1–42 by Immunoassay

The AD and normal neurons were treated with 3, 10, or 30 µM nobiletin, or with 0.1% DMSO as a vehicle control, for 24 h. The levels of Aβ1–42 in the medium supernatants were measured by use of the Aβ1–42 ELISA system. The values are presented as the mean±S.D. (n=3–4). *** p<0.001 vs. vehicle-treated AD model neurons, ###p<0.001 vs. vehicle-treated normal model neurons.


Although Aβ1–42 appears at a high level in the brain of AD patients in comparison with normal aged brain samples,9) and soluble Aβ levels correlate with dementia severity in humans,32) the expression of neprilysin mRNA and protein is poor in senile plaque areas in the hippocampus and mid temporal gyrus of AD brains.33) A decrease in neprilysin suggests limited catabolism of Aβ oligomers, which could lead to the accumulation of senile plaques. Therefore, the up-regulation of neprilysin appears to be important for the prevention of AD.

Our previous studies have demonstrated that nobiletin improves memory in several memory-impaired dementia model animals1621,34). Recently, we showed that nobiletin up-regulates the mRNA levels of neprilysin and subsequently increases its protein expression and enzyme activity in a neuroblastoma cell line.26) The purpose of this present study was to clarify the effects of nobiletin on the degradation of intra- and extraneuronal Aβ under AD pathological conditions in vitro. Remarkably, as was shown in Fig. 4A, our immunocytochemical analysis indicated that intracellular Aβ (green) levels in AD model neurons was clearly abolished by 10 or 30 µM nobiletin treatments for 24 h. In fact, as a result of quantitatively analysis by ELISA, 10 and 30 µM nobiletin decreased in the intracellular Aβ levels 0.8-fold and 0.6-fold as compared with the vehicle control (Fig. 4C).

Thus, these results show the ability of nobiletin to reduce intracellular Aβ1–42 because the neurons interacted with the anti-Aβ1–16 (6E10) antibody after permeabilization of their plasma membranes. This antibody can recognize not only Aβ1–42 but also Aβ1–40 and/or C-terminal fragments of APP cleaved by β-secretase.35) However, the AD model neurons have the PS1 FAD mutation, which results in an excess of Aβ1–42 peptide generated in the neurons.27) Hence, we consider the observed FITC signals as specific to Aβ1–42 in the AD model neurons.

On the other hand, the cleaved Aβ1–42 is released into the extracellular space from the membrane, and part of it is returned to its intracellular location by vesicular trafficking.36) We determined the amount of Aβ1–42 released into the medium by using the Aβ1–42 ELISA system. Our results indicated that extracellular Aβ1–42 levels in the AD model neurons were significantly decreased in neurons treated with 30 µM nobiletin for 24 h (Fig. 5). Thus, extracellular Aβ1–42 could be also abolished by nobiletin.

Although the concentration of 30 µM nobiletin was highly effective at a decrease in intra- and extraneuronal Aβ, the peak of the up-regulatory effect of nobiletin on neprilysin gene expression was at the concentration of 10 µM (1.7-fold, Fig. 3). We interpret these results as indicating that there is lag time between the elevation of gene transcription and its enzyme activation. It is conceivable that high concentrations of nobiletin increased the expression of neprilysin gene at an earlier time. Therefore, for 24 h after incubation, it is suggested that the enzyme activity was elevated by 30 µM nobiletin, but the gene transcription was reduced.

Under AD pathologic conditions, i.e., AD model neurons, nobiletin significantly increased neprilysin mRNA levels. This result suggests that nobiletin promoted the degradation of Aβ1–42 in association with the up-regulation of neprilysin expression. A senile plaque consists of amyloid fibrils formed by Aβ oligomer fibrogenesis.5) Production of Aβ under physiological conditions is the initial step in the pathogenic cascade of AD.37) In fact, nobiletin reduces the number of senile plaques in the hippocampus of APP-Tg AD model mice21) and lowers the level of soluble Aβ1–40 in the brain of 3XTg-AD mice.22) These results suggest that nobiletin may protect neurons from cell injury and death due to Aβ1–42 and may avoid the deposition of senile plaques on neurons.

Another interesting observation made in our study is that 10 µM nobiletin also had a tendency to increase neprilysin mRNA levels in the normal model neurons (Fig. 3). It seems likely that the intake of nobiletin by a healthy individual would prevent the accumulation of Aβ. Up-regulation of neprilysin is a key step for Aβ degradation, and nobiletin likely has a preventive effect on the onset of AD.

AD model neurons also showed the tendency to be up-regulated by nobiletin compared with normal model neurons (Fig. 3). It is reported that high levels of Aβ fibrils inhibit ERK signaling, whereas other forms of Aβ activate it.38) It is probable that an excess of Aβ1–42 may modulate activation of ERK in AD model neurons owing to their higher sensitivity in comparison with normal neurons.

In this study, whether nobiletin directly up-regulates neprilysin gene expression remains unclear. Previously, we showed that nobiletin evokes PKA/ERK/CREB signaling23) and CRE-dependent transcription,16) thus increasing the gene expression of the CREB-binding protein (CBP),25) c-Fos,24) c-jun, and Fra1 (data not shown) in PC12 and PC12D cells. In recent years, a number of studies have implicated epigenetic modifications in the expression of neprilysin. For example, histone deacetylase inhibitors up-regulate neprilysin expression and activity.39) Under hypoxic conditions, the expression of neprilysin is significantly down-regulated due to a decrease in histone H3 acetylation in the promoter region of neprilysin in mouse primary cortical and hippocampal neurons.40) These reports show that histone acetylation, in particular that of H3, is a key step in the activation of neprilysin gene transcription. CBP has histone acetyltransferase activity and a recent finding of ours demonstrates that nobiletin up-regulates the expression of CBP. Therefore, it is possible that nobiletin increases neprilysin activity via CBP up-regulation. Further studies are needed to further elucidate the molecular mechanisms of action of nobiletin on the promotion of Aβ degradation.


In this study, we successfully demonstrated that nobiletin reduces intra- and extraneuronal Aβ1–42 under in vitro AD pathologic conditions and that iPS cell-derived AD model neurons are a valuable resource for the in vitro evaluation of the effects of natural products on AD. In conclusion, our present findings suggest that it is possible to inhibit AD onset with nobiletin by attenuating Aβ-induced neuronal toxicity. Thus, nobiletin may be a useful seed drug or functional food for the prevention of AD.


This research was supported by grants for project research entitled ‘Development of fundamental technology for analysis and evaluation of functional agricultural products and functional foods’ from the Ministry of Agriculture, Forestry and Fisheries, Japan.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

The online version of this article contains supplementary materials.

© 2018 The Pharmaceutical Society of Japan