Potent activity of nobiletin-rich Citrus reticulata peel extract to facilitate cAMP/PKA/ERK/CREB...

TRANSLATIONAL NEUROSCIENCES - ORIGINAL ARTICLE

Potent activity of nobiletin-rich Citrus reticulata peel extractto facilitate cAMP/PKA/ERK/CREB signaling associatedwith learning and memory in cultured hippocampal neurons:identification of the substances responsiblefor the pharmacological action

Ichiro Kawahata • Masaaki Yoshida • Wen Sun • Akira Nakajima • Yanxin Lai • Naoya Osaka •

Kentaro Matsuzaki • Akihito Yokosuka • Yoshihiro Mimaki • Akira Naganuma • Yoshihisa Tomioka •

Tohru Yamakuni

Received: 10 October 2012 / Accepted: 4 April 2013

� Springer-Verlag Wien 2013

Abstract cAMP/PKA/ERK/CREB signaling linked to

CRE-mediated transcription is crucial for learning and

memory. We originally found nobiletin as a natural com-

pound that stimulates this intracellular signaling and

exhibits anti-dementia action in animals. Citrus reticulata

or C.unshiu peels are employed as ‘‘chinpi’’ and include a

small amount of nobiletin. We here provide the first evi-

dence for beneficial pharmacological actions on the cAMP/

PKA/ERK/CREB cascade of extracts from nobiletin-rich

C.reticulata peels designated as Nchinpi, the nobiletin

content of which was 0.83 ± 0.13 % of the dry weight or

16-fold higher than that of standard chinpi extracts. Nchinpi

extracts potently facilitated CRE-mediated transcription in

cultured hippocampal neurons, whereas the standard chinpi

extracts showed no such activity. Also, the Nchinpi extract,

but not the standard chinpi extract, stimulated PKA/ERK/

CREB signaling. Interestingly, treatment with the Nchinpi

extract at the concentration corresponding to approximately

5 lM nobiletin more potently facilitated CRE-mediated

transcriptional activity than did 30 lM nobiletin alone.

Consistently, sinensetin, tangeretin, 6-demethoxynobiletin,

and 6-demethoxytangeretin were also identified as bioactive

substances in Nchinpi that facilitated the CRE-mediated

transcription. Purified sinensetin enhanced the transcription

to a greater degree than nobiletin. Furthermore, samples

reconstituted with the four purified compounds and nobi-

letin in the ratio of each constituent’s content in the extract

showed activity almost equal to that of the Nchinpi extract

to stimulate CRE-mediated transcription. These findingsI. Kawahata and M. Yoshida equally contributed to this work.

I. Kawahata � W. Sun � A. Nakajima � Y. Lai � N. Osaka �T. Yamakuni (&)

Department of Pharmacotherapy, Graduate School of

Pharmaceutical Sciences, Tohoku University, 6-3 Aoba,

Aramaki Aoba-ku, Sendai 980-8578, Japan

e-mail: [email protected]

M. Yoshida

Research Laboratory, Kotaro Pharmaceutical Co., Ltd, Ro 96-1

Kashima-machi, Hakusan, Ishikawa 920-0201, Japan

Present Address:A. Nakajima

Department of Neuropsychopharmacology and Hospital

Pharmacy, School of Medicine, Nagoya University, Tsuruma-

cho, Showa-ku, Nagoya 466-8560, Japan

K. Matsuzaki

Department of Environmental Physiology, Faculty of Medicine,

Shimane University, 89-1 Shioya, Izumo-shi,

Shimane 693-8501, Japan

A. Yokosuka � Y. Mimaki

Laboratory of Medicinal Plant Science, School of Pharmacy,

Tokyo University of Pharmacy and Life Science, Hachioji,

Tokyo 192-0392, Japan

A. Naganuma

Department of Molecular and Biochemical Toxicology,

Graduate School of Pharmaceutical Sciences, Tohoku

University, 6-3 Aoba, Aramaki, Aoba-ku, Sendai 980-8578,

Japan

Y. Tomioka

Department of Oncology, Pharmacy Practice and Sciences,

Graduate School of Pharmaceutical Sciences, Tohoku

University, 6-3 Aoba, Aramaki, Aoba-ku, Sendai 980-8578,

Japan

123

J Neural Transm

DOI 10.1007/s00702-013-1025-x

suggest that above four compounds and nobiletin in the

Nchinpi extract mainly cooperated to facilitate potently

CRE-mediated transcription linked to the upstream cAMP/

PKA/ERK/CREB pathway in hippocampal neurons.

Keywords PKA/ERK/CREB signaling � CRE-mediated

transcription � Hippocampal neurons � Learning and

memory � Nobiletin-rich Citrus reticulata peel

Introduction

N-Methyl-D-aspartate (NMDA) receptors (NMDARs) are

recognized as being placed in the center of the learning

process (Riedel et al. 2003). Activation of extracellular

signal-regulated kinase (ERK) via NMDARs is necessary

for learning (Atkins et al. 1998). Moreover, adenosine

30,50-cyclic monophosphate (cAMP)-dependent protein

kinase (PKA) has been strongly implicated in hippocampal

long-term potentiation (LTP) associated with learning and

memory (Frey et al. 1993; Abel et al. 1997). Stimulation of

NMDARs leads to activation of cAMP-dependent signal-

ing pathway, which eventually converges upon ERK

(Adams and Sweatt 2002). ERK performs several functions

relevant for establishing short- and long-term memory

(Sweatt 2004), and its activation leads to a number of

cellular changes associated with the development of LTP,

such as alterations in gene expression and protein synthe-

sis, and stabilization of dendritic spines. Among the direct

downstream targets of activated ERK is cAMP-response

element-binding protein (CREB; Atkins et al. 1998; Tully

et al. 2003; Impey et al. 1998; Cammarota et al. 2000;

Athos et al. 2002; Levenson et al. 2004; Chwang et al.

2006). Conversely, in vivo blockade or inhibition of ERK

signaling or CRE-mediated gene expression results in

impaired associative learning (Atkins et al. 1998; Athos

et al. 2002; Kim et al. 1991). Taken together, the available

data suggest that the regulation of cAMP/PKA/ERK/CREB

signaling linked to CRE-mediated transcription is a

potential critical step in the clinical control of neurological

disorders involving cognitive dysfunction.

Nobiletin, a citrus polymethoxylated flavone, was orig-

inally identified by our group as a substance that potenti-

ates cAMP/PKA/ERK/CREB signaling pathway coupled

with CRE-mediated transcription in cultured hippocampal

neurons and exerts its neurotrophic action by facilitating

this intracellular signaling cascade in PC12D cells (Nagase

et al. 2005a). Our recent study showed that nobiletin can

pass through the blood–brain-barrier in mice (Saigusa et al.

2011). Compellingly, it was shown by us that this com-

pound indeed improves memory deficits in amyloid pre-

cursor protein (APP) transgenic mice overexpressing

human APP695 harboring the double Swedish and London

mutations (Onozuka et al. 2008) and that nobiletin reverses

the NMDA receptor antagonist MK801-impaired memory

by activating ERK signaling in the hippocampus of mice

(Nakajima et al. 2007), suggesting that this compound has

a potential therapeutic benefit for dementia, including

Alzheimer’s disease (AD).

Citrus peels are a rich source of flavonoids, including

nobiletin and its structurally related compounds; but their

contents vary in a species-dependent manner (Nagata et al.

2006). The dried peels of Citrus reticulata Blanco and C.

unshiu Markovich, designated as AURANTII NOBILIS

PERICAPRIUM (Toriizuka 2003), have been widely

employed from ancient times as a crude drug in Kampo

medicine, and are referred to as ‘‘chinpi’’, especially in

Japan and China. Chinpi is employed as stomachic and a

crude drug to improve bronchial and asthmatic conditions,

etc. in traditional medicine. Although our recent pilot study

showed that some batches of chinpi tested had little effect

on the PKA/ERK/CREB signaling cascade in cultured

hippocampal neurons, the use of these citrus peels in

Kampo medicine prompted us to ascertain whether nobi-

letin-rich chinpi could have a potential therapeutic effect

on dementia. For this purpose, we screened over 250 kinds

of peels from C. unshiu or C. reticulata cultivated in Japan

and China, by performing a reporter gene assay with a

firefly luciferase reporter plasmid containing CRE, and

thereby successfully found four batches of nobiletin-rich C.

reticulata peels the extracts of which facilitated CRE-

mediated transcription in cultured hippocampal neurons.

Interestingly, our recent pilot clinical study suggested the

possibility that 1-year oral administration of decocted

nobiletin-rich C. reticulata peel could be of benefit for

improving the cognition of patients with AD, with no

adverse side effects, such as digestive symptoms (Seki

et al. 2013).

In the present study, using cultured hippocampal neu-

rons we demonstrated for the first time that an extract

from the same batch of nobiletin-rich C. reticulata peel

stimulated PKA/ERK/CREB signaling and CRE-mediated

transcription associated with learning and memory. Fur-

thermore, in this batch of nobiletin-rich citrus peel, we

identified four other natural compounds that cooperated

with nobiletin and thereby contributed to the activity of the

nobiletin-rich extract to stimulate CRE-mediated tran-

scription in hippocampal neurons in primary culture.

Materials and methods

Materials

Nobiletin was extracted and isolated from C. depressa

peels through a series of chromatographic separations, as

I. Kawahata et al.

123

described previously (Nagase et al. 2005a; Onozuka et al.

2008). The isolated nobiletin was further recrystallized

from acetone. Consequently, the purity of the purified

sample was more than 99 %, as confirmed by high-per-

formance liquid chromatography (HPLC) analysis. Purified

nobiletin was dissolved in dimethyl sulfoxide (DMSO) to

make a 30-mM stock solution. Aliquots of the stock solu-

tion were stored at -25 �C to avoid loss of pharmacolog-

ical activities owing to freeze–thaw cycles. The stock

solutions were then diluted to appropriate concentrations

prior to use. Narirutin, tangeretin, and hesperidin with

purities of more than 99 % were purchased from Wako

Pure Chemicals Industries, Ltd. Other chemicals and drugs

were of reagent grade or of the highest quality available.

Preparation of AURANTII NOBILIS PERICAPRIUM

extract

Different batches of dried peels of C. reticulata Blanco and

C. unshiu Markovich (listed in Table 1) were imported

from China or obtained in Japan and employed as AUR-

ANTII NOBILIS PERICAPRIUM for the present study.

The original plant sources of the crude drugs tested here

were identified by Dr. Tokurou Shimizu of the National

Institute of Fruit Tree Science, National Agriculture and

Food Research Organization (NARO), Japan. Dried

extracts made from these dried peels were prepared by

Kotaro Pharmaceutical Co., Ltd (Osaka, Japan) according

to the following procedures: in brief, after all the peels

were oven-dried at 60 �C for 2 h and finely cut, about 10 g

of the dried peels were immersed in 400 mL of distilled

water, and the water was heated to boiling. After a 1-h hot-

water extraction at 100 �C followed by cooling to room

temperature, the extracts were filtered and lyophilized to

obtain the dried extracts. These dried extracts were sub-

sequently examined for contents of flavonoids and also

used for in vitro and in vivo studies. For in vitro studies, the

extracts were dissolved in DMSO and stored as stock

solutions at -25 �C, and diluted prior to use for the in vitro

studies. The concentration of nobiletin in 300 lg/mL of the

Nchinpi #1 extract employed for treatment of cultured

neurons was approximately 5 lM, when assayed by using

the HPLC system described below. For in vivo studies in

mice, 11.1 g of the dried extract of Nchinpi #1 was sus-

pended in 30 mL of saline (0.9 % NaCl) and sonicated

with an ultrasonic bath for 30 min, and aliquots of the

suspension were stored at -25 �C because of avoiding loss

of the pharmacological activities due to freeze–thaw

cycles. Also, to confirm the concentration of nobiletin of

the Nchinpi #1 extract suspension tested in this study, a

small amount of the suspension was subjected to determi-

nation using an HPLC system prior to use for animal

experiments.

Determination of flavonoid contents

The contents of the polymethoxylated flavones nobiletin

and tangeretin and those of the flavanones narirutin and

hesperidin were determined as follows: the dried extracts

(about 0.5 g) were extracted with 50 mL of 70 % methanol

by a 20-min ultrasonication, filtered, and analyzed by

HPLC (Hitachi L-7000 series). Purified nobiletin and tan-

geretin were also dissolved in 70 % methanol to make the

standard solutions of 0.05 and 0.03 mg/mL, respectively,

while purified narirutin and hesperidin were each dissolved

in 50 % methanol to make a standard solution of 0.1 mg/mL.

HPLC analysis was carried out under the following con-

ditions: for determination of nobiletin and tangeretin

Table 1 Content of four flavonoids and nobiletin content/nariutin content rations in the extracts from crude drugs tested in the present study

Sample no. Species Content (% of dry weight) Ratio of nobiletin

content to narirutin one

Nobiletin Tangeretin Nariruin Hesperidin

NChinpi #1 Citrus reticulata Blanco 0.68 0.20 0.31 1.93 2.19




NChinpi (mean ± SD) 0.83 ± 0.13*** 0.25 ± 0.06*** 0.32 ± 0.06 2.48 ± 0.37

Chinpi #1 Citrus reticulata Blanco 0.060 0.034 1.32 2.83 0.05

Chinpi #2 Citrus unshiu Markovich 0.047 0.019 1.54 2.54 0.03



Chinpi (mean ± SD) 0.05 ± 0.01 0.02 ± 0.01 1.93 ± 0.59** 2.76 ± 0.17

The content of flavonoids was expessed as mean (n = 3). The coefficient of variation of each content was within 1 %

** p \ 0.001 vs. Nchinpi, *** p \ 0.0001 vs. chinpi (n = 4), unpaired t test

Potent activity of nobiletin-rich Citrus reticulata peel

123

contents, an ODS column (Mightysil RP-18 GPS,

4.6 9 150 mm, 5 lm; Kanto Chemical Co., INC., Japan),

equipped with a detector (an ultraviolet absorption pho-

tometer, wavelength: 338 nm) and operated at a column

temperature of 40 �C, mobile phase of 40 % acetonitrile in

H2O, and a flow rate of 0.8 ml/min was used. Narirutin and

hesperidin contents were determined by using an ODS

column (L-column2 ODS, 4.6 9 250 mm, Chemicals

Evaluation and Research Institute, Japan) equipped with a

detector (an ultraviolet absorption photometer, wavelength:

285 nm) and operated at column temperature of 30 �C,

mobile phase of H2O/acetonitrile/acetic acid (40:10:1), and

a flow rate of 0.8 mL/min. An injection volume of 10 lL

was used for all standard solutions and sample solutions

tested. To simplify the description, we designated dried

nobiletin-rich C. reticulata peels and dried standard

C. reticulata or C. unshiu ones as Nchinpi and chinpi (or

standard chinpi), respectively.

Three-dimensional HPLC

The dried extracts (1 g) from dried nobiletin-rich C. re-

ticulata peels (Nchinpi #1) or dried peels of the same

species with a standard nobletin content (chinpi #1) listed

in Table 1 were extracted with 50 mL of 50 % methanol

by a 20-min ultrasonication, filtered, and analyzed by

HPLC (Hitachi L-7000 series) under the following condi-

tions: 10 lL of each sample was applied to a ODS column

(Mightysil RP-18 GP S, 4.6 9 150 mm, 5 lm, Kanto

Chemical Co., INC.). The elution sequence was 15 %

CH3CN in H2O for the first 5 min, a linear gradient of

15–40 % CH3CN in H2O over the subsequent 40 min, and

40 % CH3CN in H2O for the final 15 min. The flow rate

was 0.8 mL/min and the column temperature was set at

40 �C. The UV spectrum ranging from 200 to 350 nm was

collected with a PDA detector (Hitachi L-7450).

Hippocampal neuronal cultures

Hippocampal neuronal cultures were prepared as described

previously (Nagase et al. 2005a; Nakajima et al. 2007;

Matsuzaki et al. 2008). In brief, under ether anesthesia,

embryos of an 18-day pregnant Sprague–Dawley rat (Japan

SLC) were removed and immediately decapitated. Hippo-

campal neurons were dissociated according to the standard

methods used for a SUMITOMO Nerve Cell Culture Sys-

tem (Sumitomo Bakelite), plated, and cultured for trans-

fection and Western blot analysis, as described below.

Transient transfection and reporter gene assay

Neurons were plated on 48-well plates at a density of

8 9 104 cells/well in Neurobasal Medium without Phenol

Red (Gibco Co., Ltd.) and containing B-27 Supplement

(GIBCO CO., LTD.), 500 lM L-glutamine, and penicillin–

streptomycin. At 4 days in vitro (D.I.V.), one-half of

the medium was replaced with the medium containing a

mitotic inhibitor, cytosine arabinofuranoside (Ara-C), to

minimize glial proliferation; and the cells were then cul-

tured for 10 days, as described previously (Nagase et al.

2005a; Nakajima et al. 2007; Matsuzaki et al. 2008).

Transfection and reporter gene assays were conducted as

described previously (Nagase et al. 2005a). Neurons were

co-transfected with a firefly luciferase reporter plasmid

containing CRE (Clontech) and a Renilla luciferase control

vector, phRG-TK (Promega), which was used as an internal

control to normalize the difference in transfection effi-

ciency. Lipofectamine 2000 (Invitrogen) was employed for

the transfection. Sixteen hours after transfection, the cul-

tured neurons were treated with extracts of different bat-

ches of Nchinpi, chinpi, nobiletin, or other samples tested

at the concentrations indicated in the figure legends or with

0.1 % DMSO as vehicle, in the Ara-C-containing medium

supplemented with B-27 supplement and L-glutamine.

Following an 8-h incubation, the cultured neurons were

washed twice with ice-cold Dulbecco’s PBS. Thereafter,

the cells were lysed with 65 ll of Passive Lysis Buffer

(Promega), and the dual-luciferase reporter assay was

performed, as described previously (Nagase et al. 2005a).

Regarding treatment experiments with H-89, a PKA-

selective inhibitor (Calbiochem), and U0126, a MEK

inhibitor (Calbiochem), each of both protein kinase inhib-

itors or a vehicle was added to the culture medium at

30 min prior to treatment with the Nchinpi #1 extract, and

thereafter cells were incubated with 300 lg/mL of the

Nchinpi extract for 8 h to be subjected to reporter gene

assay.

Western blot analysis

For Western blot analysis, hippocampal neurons were

dissociated, plated at a density of 1 9 106 cells/dish on

35-mm culture dishes coated with poly-L-lysine, and cul-

tured at 37 �C in the medium described above. At 10–14

D.I.V., the cultured neurons were treated for 10 or 15 min

with the Nchinpi #1 or chinpi #1 extract, or with NMDA or

forskolin (as positive controls) at the concentrations indi-

cated in the figure legends, or with 0.1 % DMSO as a

vehicle, in the Ara-C-containing medium without B-27

supplement and L-glutamine. They were then subsequently

analyzed by Western blotting. Western blot analyses of cell

extracts from hippocampal neuronal cultures were con-

ducted as described previously (Nagase et al. 2005a;

Nakajima et al. 2007; Matsuzaki et al. 2008). Blots

were incubated with anti-phospho-PKA substrate, anti-

phospho-ERK1/2 (Thr202/Tyr204), anti-total ERK1/2, and

I. Kawahata et al.

123

anti-phospho-CREB (Ser133) antibodies (Cell Signaling

Technology), and anti-14-3-3-b antibody (Santa Cruz

Biotechnology). For Western blot analysis of the level for

nuclear ERK phosphorylation which contributes to tran-

scription associated with learning and memory in the hip-

pocampus of mice, nuclear extracts were prepared from the

hippocampi of untrained animals or trained ones that

underwent 7-day oral administration of the Nchinpi #1

extract or saline followed by an injection of MK-801 or a

vehicle as described below. Preparation of the nuclear

extracts and Western blot analysis of the level for nuclear

ERK phosphorylation with anti-phospho-ERK1/2 antibody

were conducted as described previously (Nakajima et al.

2007). Immunoreactivities were visualized with Super-

Signal West Pico Chemiluminescent Substrate (Pierce).

The band intensities were quantitatively analyzed by

SCION image software.

Extraction and identification of the compounds

that contributed to the pharmacological action

of Nchinpi extract toward cultured hippocampal

neurons

Nchinpi #1 extract (81.11 g) was suspended in 30 % ace-

tonitrile (360 mL). The obtained suspension was centri-

fuged, and the supernatant was loaded onto a Varian Bond

Elut C18 column and eluted with 30 % acetonitrile to give

Fr.1 (79.95 g), and then with MeOH to give Fr. 2 (poly-

methoxyflavonoid fraction, 1.16 g). Fr. 2 (505 mg) was

subjected to ODS column chromatography, and the column

was eluted with 40 % acetonitrile—H2O and then purified

by repeated silica gel column chromatography eluted with

acetone—hexane (1:1) or ethyl acetate—hexane (4:1) to

give nobiletin (224 mg) and Fr.2–1 (non-nobiletin sub-

fraction, 314 mg). Fr.2–1 (200 mg) was subjected to

repeated ODS column chromatography and crystallization

to yield compounds 1 (20 mg), 2 (11 mg), 3 (14 mg), and 4

(12 mg) described below:

Tangeretin (1). Colorless needles, mp 150–151 �C. IR

mmax (KBr) cm-1: 1,650, 1,608, 1,514, 1,465, 1,408, 1,364,

1,266, 1,182, 1,110, 1,074, 1,019, 970, 830. UV kmax

(MeOH) nm (log e): 323 (4.46), 270 (4.31). Positive-ion

ESI–MS m/z 373 ([M ? H]?).

6-Demethoxynobiletin (2). Colorless needles, mp

199–200 �C. IR mmax (KBr) cm-1: 1,640, e): 338 (4.31),

269 (4.29), 248 (4.27). 1H NMR (CDCl3): d 7.57 (1H, dd,

J = 10.7, 2.8 Hz), 7.40 (1H, d, J = 2.8 Hz), 6.97 (1H, d,

J = 10.7 Hz), 6.60 (1H, s), 6.42 (1H, s), 3.99 (3H, s), 3.97

(3H, s), 3.95 (3H, s), 3.94 (3H 9 2, s). 13C NMR (CDCl3):

d 177.8, 160.5, 156.4, 156.3, 151.9, 151.7, 149.2, 130.7,

124.0, 119.5, 111.1, 108.9, 108.5, 107.1, 92.4, 61.5, 56.5,

56.3, 56.0, 55.9. Positive-ion ESI–MS m/z 373

([M ? H]?).

Sinensetin (3). Colorless needles, mp 175–176 �C. IR

mmax (KBr) cm-1: 1,636, 1,601, 1,516, 1,421, 1,326, 1,253,

1,121, 1,022, 843. UV kmax (MeOH) nm (log e): 329 (4.39),

239 (4.30). 1H NMR (CDCl3): d 7.50 (1H, dd, J = 10.7,

2.5 Hz), 7.31 (1H, d, J = 2.5 Hz), 6.95 (1H, d,

J = 10.7 Hz), 6.78 (1H, s), 6.59 (1H, s), 3.98 (3H, s), 3.97

(3H, s), 3.96 (3H, s), 3.94 (3H, s), 3.90 (3H, s). 13C NMR

(CDCl3): d 177.2, 161.1, 157.7, 154.5, 152.6, 151.8, 149.3,

140.4, 124.1, 119.6, 112.9, 111.1, 108.7, 107.4, 96.2, 62.2,

61.5, 56.3, 56.1, 56.0. Positive-ion ESI–MS m/z 373

([M ? H]?).

6-Demethoxytangeretin (4). Colorless needles, mp

213–214 �C. IR mmax (KBr) cm-1: 1,638, 1,598, 1,510,

1,342, 1,249, 1,211, 1,184, 1,112, 1,048, 841, 802. UV kmax

(MeOH) nm (log e): 310 (4.29), 269 (4.37). 1H NMR

(CDCl3): d 7.86 (2H, d, J = 11.3 Hz), 7.00 (2H, d,

J = 11.3 Hz), 6.58 (1H, s), 6.41 (1H, s), 3.98 (3H, s), 3.96

(3H, s), 3.93 (3H, s), 3.86 (3H, s). 13C NMR (CDCl3): d177.8, 162.1, 160.6, 156.4, 156.2, 151.9, 130.7, 127.6,

123.8, 114.4, 109.0, 106.9, 92.5, 61.5, 56.6, 56.2, 55.4.

Positive-ion ESI–MS m/z 343 ([M ? H]?).

Also, each fraction and purified tangeretin, 6-demeth-

oxynobiletin, 6-demethoxytangeretin, and sinensetin (with

purities of more than 98 %) was tested in the reporter gene

assay.

Animal experiments

The animal experimental procedures used in this study

were approved by the Committee on the Care and Use of

Experimental Animals, Tohoku University, in accordance

with the Guide for the Care and Use of Laboratory Animals

published by the US National Institute of Health.

Animals

7-week-old male ddY mice were obtained from Nippon

SLC (Hamamatsu, Japan). Animals were housed in cages

with free access to food and water under the condition

of constant temperature (23 ± 1 �C) and humidity

(55 ± 5 %) and adapted to a standard 12-h light/12-h dark

cycle (light cycle: 9:00–21:00).

Contextual fear conditioning

Animals were placed in the training chamber and allowed

to explore for 2 min, and thereafter they received an

electric shock (2 s, 0.7 mA). The 2 min/2 s shock para-

digm was repeated for a total of three shocks. After the last

shock, animals were allowed to explore the context for an

additional 1 min prior to removal from the training

chamber. During training, freezing behavior was measured

during 1 min after each shock. To assess contextual


123

learning and memory, the animals were placed back into

the training context 24 h after fear conditioning, and scored

for freezing for 5 min. Freezing behavior, defined as ces-

sation of all but respiratory movement, was measured by

observing the animals every 5 s. To examine the effects of

7-day treatment with the Nchinpi #1 extract at a dose of

3.67 g/kg/day, on day 7 the Nchinpi extract or vehicle was

orally administered 90 min prior to treatment with MK-801

(0.08 mg/kg, i.p.), which was given 30 min before training

of contextual fear conditioning paradigm. This dose of

MK-801 was chosen on the basis of the previous available

data showing that the same dose of MK-801 is capable of

impairing both contextual associative learning and learn-

ing-associated ERK activation during fear conditioning

(Atkins et al. 1998). Also, the dose of 3.67 g/kg/day of

Nchinpi #1 extract was chosen and orally administered for

7 days, since this dose corresponds to that of nobiletin of

25 mg/kg/day, which is capable of improving learning and

memory impairment in our previous behavioral pharma-

cological studies (Matsuzaki et al. 2006; Nakajima et al.

2007; Onozuka et al. 2008). Saline (0.9 % NaCl) was

orally given as a vehicle for 7 days. Additionally, for

Western blot analysis of the level for nuclear ERK phos-

phorylation coupled with transcription associated with

learning and memory in the hippocampus of mice as

described previously (Nakajima et al. 2007), animals were

trained after administration of the Nchinpi extract or a

vehicle followed by an injection of MK-801 or a vehicle, or

were untrained after an injection of a vehicle on day 7. The

Nchinpi #1 extract or a vehicle was orally administered

90 min prior to treatment with MK-801. Animals were

killed by decapitation immediately after training, and

thereafter the brains were quickly removed for collection of

the hippocampi to prepare the nuclear extracts.

Statistical analyses

The data for the freezing during training of contextual fear

conditioning paradigm were analyzed by two-way repeated

measures ANOVA. Other data were analyzed by one-way

ANOVA, followed by the Student-Neumann-Keuls test. A

level of p \ 0.05 was considered statistically significant.

Results

Contents of flavonoids in Nchinpi and standard chinpi

extracts

As shown in Fig. 1 and Table 1, the content of nobiletin

was much higher in the extracts from the four batches of

Nchinpi than in those from four batches of standard chinpi.

In the extracts from the Nchinpi batches the content of

nobiletin was 0.83 ± 0.13 % , whereas for the standard

chinpi it was 0.05 ± 0.01 % (Table 1). Similarly, the

content of tangeretin, a compound structurally related to

nobiletin, in the Nchinpi extracts was far higher than that in

the extracts from the standard chinpi ones: tangeretin

contents in the Nchinpi and standard chinpi extracts were

0.25 ± 0.06 and 0.02 ± 0.01 %, respectively (Fig. 1;

Table 1). In contrast, the content of narirutin in the stan-

dard chinpi extracts was much higher than that in the

Nchinpi ones, being 1.93 ± 0.59 and 0.32 ± 0.06 %,

respectively (Fig. 1; Table 1). With respect to the ratio of

the content of nobiletin to that of narirutin, the extracts

from Nchinpi and chinpi gave values of 2.19–3.57 and

0.02–0.05, respectively, possibly indicating that a high

value of the ratio might be useful as an index of the quality

of the nobiletin-rich extract. As for the content of hesper-

idin, a flavonoid glycoside, Nchinpi and standard chinpi

extracts showed similar amounts (2.48 ± 0.37 and

2.76 ± 0.17 %, respectively; Fig. 1, Table 1), which fitted

NobiletinHesperidinNarirutin

Tangeretin

Nchinpi #1A

HesperidinNarirutin

NobiletinTangeretin

chinpi #1B

Fig. 1 Three-dimensional HPLC chromatograms of extracts of

nobiletin-rich chinpi and standard chinpi. Analyses of the extracts

from both Nchinpi #1 (a) or chinpi #1 (b) were performed by using

three-dimensional HPLC as described in the ‘‘Materials and methods’’

section

I. Kawahata et al.

123

well with the description of AURANTII NOBILIS PERI-

CAPRIUM defined in the revised 16th Edition Japanese

Pharmacopeia. Similarly, the morphological data con-

cerning Nchinpi were in agreement with the description in

the Japanese Pharmacopeia (Yoshida et al., data not

shown).

Effects of Nchinpi and standard chinpi extracts

on CRE-mediated transcription in cultured

hippocampal neurons

Next we ascertained whether the extracts of Nchinpi could

potentiate the CRE-mediated transcription crucial for the

formation of long-term memory (Tully et al. 2003; Impey

et al. 1998; Cammarota et al. 2000; Athos et al. 2002), by

performing a by reporter gene assay using cultured hip-

pocampal neurons. All Nchinpi extracts #1 to #4, tested at a

concentration of 300 lg/mL, facilitated CRE-mediated

transcription, which was sixfold that of the vehicle control,

whereas the chinpi extracts #1 and #2 did not significantly

potentiate the transcription as compared with the vehicle

control at the same concentration (Fig. 2a).

Concentration-dependent stimulatory effects of Nchinpi

extract on CRE-mediated transcription

and phosphorylation of ERK, CREB, and PKA

substrates, in cultured hippocampal neurons

Next we evaluated in more detail the effects of the Nchinpi

#1 extract on CRE-mediated transcription and ERK and

CREB phosphorylation in cultured hippocampal neurons.

The Nchinpi #1 extract concentration-dependently

enhanced CRE-mediated transcription (Fig. 2b) and stim-

ulated the phosphorylation of ERK and CREB in a con-

centration-dependent manner (Fig. 3a, b, respectively). As

shown in Fig. 3c, the Nchinpi extract also concentration-

dependently increased the phosphorylation level of PKA

substrates. Additionally, the Nchinpi #1 extract exhibited

no cytotoxic effects at any concentrations tested, as eval-

uated by use of the MTT assay (data not shown). Consis-

tent with the results of our comparative study on the effects

of the Nchinpi #1 extract and chinpi #1 one on CRE-

mediated transcription, Western blot analysis showed that

the Nchinpi extract stimulated the phosphorylation of ERK

as well as that of PKA substrates at the concentration of

300 lg/mL (Fig. 4a). On the other hand, chinpi #1 extract

little influenced the levels of phosphorylation of these

proteins at the same concentration (Fig. 4a).

Nchinpi #1 extract facilitates CRE-mediated

transcription via a PKA- and ERK-dependent

mechanism in cultured hippocampal neurons

As described above, the Nchinpi #1 extract not only

facilitated CRE-mediated transcription but also stimulated

phosphorylation of ERK, CREB, and PKA substrates in

hippocampal neurons in primary culture. PKA- and ERK-

dependent signaling pathways work upstream of CRE-

mediated transcription as being associated with learning

and memory (Kim et al. 1991; Adams and Sweatt 2002;

Atkins et al. 1998; Athos et al. 2002; Tully et al. 2003;

Impey et al. 1998; Cammarota et al. 2000; Levenson et al.

Fig. 2 Effects of the extracts of distinct batches of Nchinpi and

chinpi on CRE-mediated transcription in hippocampal neurons in

primary culture. Neurons were plated on 48-well plate, cultured for

14 days, and then transfected with a luciferase reporter construct for

CRE-mediated transcription for 16 h. a Effects of extracts of Nchinpi

and chinpi on CRE-mediated transcription. Following transfection,

neurons were treated with 300 lg/mL of the extracts from Nchinpi #1

to #4 or chinpi #1 to #2 for 8 h. Value are mean ± SEM (n = 4).

***p \ 0.001 vs. control. b Concentration-dependent effects of

Nchinpi #1 extract on CRE-mediated transcription. Following trans-

fection, neurons were treated with 10–600 lg/mL of the extract from

Nchinpi #1 for 8 h. Values are mean ± SEM (n = 4). **p \ 0.01,

***p \ 0.001 vs. control


123

2004; Chwang et al. 2006). It was thus examined whether

the facilitated CRE-mediated transcription by the Nchinpi

#1 extract links to the upstream PKA- and ERK-dependent

signaling pathways in cultured hippocampal neurons, by

the use of H-89, a PKA-selective inhibitor, or U0126, an

inhibitor of MEK that specifically phosphorylates ERK.

When the neuronal cells were treated with each of both

inhibitors at a concentration of 10 lM prior to treatment

with the Nchinpi extract, this facilitation of CRE-mediated

transcription was abolished (Fig. 5), suggesting that the

Nchinpi #1 extract facilitates CRE-mediated transcription

via both upstream signaling pathways.

Improvement of MK-801-induced learning

and memory impairment by Nchinpi #1 extract in mice

cAMP/PKA/ERK/CREB signaling linked to CRE-medi-

ated transcription is crucial for learning and memory (Tully

et al. 2003; Impey et al. 1998; Cammarota et al. 2000;

Athos et al. 2002). To test whether the Nchinpi #1 extract

could have a beneficial effect on impaired learning and

memory in an animal model of dementia, we investigated

action of the Nchinpi #1 extract on a non-selective

NMDAR antagonist MK-801-induced learning and mem-

ory impairment in the contextual fear conditioning para-

digm, in which animals learn to associate a normally

innocuous context with an aversive stimulus (Atkins et al.

1998; Kim et al. 1991). The Nchinpi #1 extract was orally

given to animals at a dose of 3.67 g/kg/day for 7 days to

examine the effects of the Nchinpi extract on MK-801-

induced impairment of contextual fear learning and mem-

ory. A repeated measures ANOVA was used to assess

freezing responses in the training session, and a significant

effect of treatment (F(2,26) = 33.986, p \ 0.0001), shock

event (F(2,52) = 220.664, p \ 0.0001) and treatment 9

shock event interaction (F(4,52) = 18.907, p \ 0.0001)

(Fig. 6a) was found. After the second and third shock, mice

treated with MK-801 showed less freezing behavior com-

pared with control group (second: F(2,26) = 16.6334,

p \ 0.0001 by one-way ANOVA, p \ 0.001 by post hoc;

third: F(2,26) = 45.940, p \ 0.0001 by one-way ANOVA,

p \ 0.001 by post hoc). Treatment of mice with the

Nchinpi #1 extract at the dose of 3.67 g/kg/day for 7 days

reversed the MK-801-induced retarded acquisition of

contextual fear memory (p \ 0.001 by post hoc) (Fig. 6a).

In the test session performed 24 h after training, mice

injected with MK-801 showed a less freezing behavior than

did control mice (F(2,26) = 49.772, p \ 0.0001 by one-way

ANOVA; p \ 0.001 by post hoc) (Fig. 6b). The 7-day

administration of the Nchinpi #1 extract at this dose

A Phospho-ERK1/2

Total-ERK1/2

*

Ph

osp

ho

-ER

K

1

3

2

Nchinpi #1 extract (μg/ml)

Control 30 60 120 240 4800

Phospho-CREB

14-3-3-β

Ph

osp

ho

-CR

EB

1

3

2

Control 30 60 120 240 480

B

***

*

0

Nchinpi #1 extract (μg/ml)

45

9766

116

14-3-3-β

Phospho-PKA substrates

C

Control 30 60 120 240480

Nchinpi #1extract (μg/ml)

kDa

Fig. 3 Nchinpi #1 extract stimulates phosphorylation of ERK,

CREB, and PKA substrates in a concentration-dependent mode in

cultured hippocampal neurons. Concentration-dependent effects of

Nchinpi #1 extract on phosphorylation of ERK (a) and CREB (b) in

cultured hippocampal neurons. Cells were treated with the extract

(30–480 lg/mL) for 10 min. Western blot analyses were performed

using antibodies specific for phospho-ERK1/2 (a) or phospho-CREB

(b). Values are mean ± SEM (n = 3). *p \ 0.05, ***p \ 0.001 vs.

control. c Effects of Nchinpi #1 extract on phosphorylation of PKA

substrates in the cultured neurons. Western blot analyses were

performed using antibody against substrates phosphorylated by PKA.

Similar results were obtained from three independent experiments.

All blots were stripped and reprobed with antibodies specific for total

ERK1/2 or 14-3-3-b to verify that equal amounts of proteins had been

electrophoresed in each lane

I. Kawahata et al.

123

reversed MK-801-induced fear memory impairment by

approximately 70 % (p \ 0.001 by post hoc) (Fig. 6b).

Open-field test showed that this repeated administration of

the Nchinpi extract had no effects on the horizontal activity

of mice (data not shown). Additionally, the 7-day oral

administration of the Nchinpi extract restored MK-801-

induced inhibition of learning-associated activation of

ERK in the hippocampus of mice (Fig. 6c).

Screening of a non-nobiletin subfraction

of the polymethoxyflavonoid fraction of the Nchinpi

extract to identify substances that contributed

to the extract-stimulated CRE-mediated transcription

in cultured hippocampal neurons

To search for the substances responsible for the stimulatory

action of the Nchinpi #1 extract on CRE-mediated tran-

scription in cultured hippocampal neurons, we prepared

the following two fractions of this extract according to

the fractionation procedure shown in Fig. 7a: the poly-

methoxyflavonoid fraction with a high nobiletin content of

42.9 % (Fr. 2) and the non-polymethoxyflavonoid fraction

(Fr. 1). Fr. 2 was further divided into the nobiletin subfrac-

tion and the non-nobiletin one (Fr. 2–1, Fig. 7a) to identify

the substances other than nobiletin that might have been

involved in the transcription-stimulating activity of the

Nchinpi #1 extract. Fr. 2–1 was first subjected to reporter

gene assay, and thereby the stimulatory activity was detec-

ted in Fr. 2–1 as it was in the nobiletin subfraction (Fig. 8a).

The reconstituted sample with both subfractions in the ratio

of each component’s content in Fr. 2 also facilitated the

transcription as potently as the Fr. 2, as also shown in

Fig. 8a. Notably, tangeretin, 6-demethoxynobiletin, 6-de-

methoxytangeretin, and sinensetin were found in the Fr. 2–1

(Fig. 7b). We further tested whether these four compounds

were part of the activity of the Nchinpi extract to facilitate

CRE-mediated transcription. All four compounds exhibited

such activity. Among these compounds tested at a concen-

tration of 30 lM, sinensetin was the most potent in facili-

tating the transcription (Fig. 8b). In addition, as determined

by using HPLC, the contents of tangeretin, 6-demethoxyn-

obiletin, 6-demethoxytangeretin, and sinensetin were 0.20,

Fig. 4 The Nchinpi #1 extracts, but not the chinpi #1 one, facilitates

phosphorylation of ERK and PKA substrates in cultured hippocampal

neurons. As described under ‘‘Materials and methods’’, cells were

treated with extracts of Nchinpi #1 and with chinpi #1 or with

forskolin or NMDA (as positive controls) at the concentrations

indicated, or with vehicle for 15 min, and then subjected to Western

blot analysis with antibodies specific for phospho-ERK1/2 or

phospho-PKA substrates. Blots were then stripped and reprobed with

anti-total ERK1/2 antibody. a Shown are blots of representative

results obtained from at least three independent experiments. b Values

obtained by densitometry of the blots in ‘‘a’’ are mean ± SEM

(n = 3). ***p \ 0.001 vs. control

Fig. 5 The Nchinpi #1 extracts facilitate CRE-mediated transcription

via PKA- and ERK-dependent signaling pathway in cultured hippo-

campal neurons. Neurons were plated on 48-well plate, cultured

for 14 days, and then transfected with a luciferase reporter construct

for CRE-mediated transcription for 16 h. Neuronal cells were

then treated with H-89 or U0126 at a concentration of 10 lM for

30 min and subsequently incubated with 300 lg/mL of Nchinpi

#1 extract for 8 h. C control, Veh vehicle, Nb 30 lM nobiletin. Values

are mean ± SEM (n = 4). *p \ 0.05, ***p \ 0.001 vs. control,###p \ 0.001 vs. vehicle


123

0.059, 0.045, and 0.074 %, respectively, as calculated on the

basis of the dried Nchinpi #1 extract (Yoshida, data not

shown). Notably, in our reconstitution experiment where the

four compounds were reconstituted with nobiletin at the

ratio of each constituent’s content in the extract (see the

content of nobiletin shown in Table 1), this reconstituted

sample facilitated the CRE-mediated transcription as

potently as Fr. 2, whose CRE-mediated transcription-facil-

itating activity was almost equal to that of the Nchinpi #1

extract (Fig. 8a). Also, our further comparative pharmaco-

logical study of sinensetin and nobiletin revealed that sin-

ensetin facilitated the transcription appreciably more

potently than nobiletin at any concentration tested on hip-

pocampal neurons in primary culture (Fig. 8c).

Discussion

In the present study, we found that extracts of four batches

of nobiletin-rich C.reticulata peels, designated as Nchinpi,

had a nobiletin content that was 16-fold higher than that of

standard chinpi extracts. We confirmed that all four

Nchinpi extracts facilitated CRE-mediated transcription in

cultured hippocampal neurons, whereas the standard chinpi

extracts had no enhancing effects on the transcription.

Consistent with these results, the Nchinpi #1 extract, but

not the chinpi #1 extract, facilitated PKA/ERK/CREB

signaling in culture. Moreover, we also found that tan-

geretin, 6-demethoxynobiletin, 6-demethoxytangeretin,

and sinensetin, which were substances contained in the

extract, actually contributed to the CRE-mediated tran-

scription-facilitating activity of the Nchinpi #1 extract

toward hippocampal neurons in primary culture. Addi-

tionally, the Nchinpi #1 extract restored MK-801-induced

learning and memory impairment by activation of ERK

signaling in animal.

Importantly, in the Nchinpi #1 extract of nobiletin-rich

C.reticulata peels, tangeretin, 6-demethoxynobiletin,

6-demethoxytangeretin, and sinensetin were shown to

cooperate with nobiletin to facilitate this CRE-mediated

transcription. By contrast, the standard chinpi #1 extract

could potentiate neither CRE-mediated transcription nor

PKA/ERK signaling in the cultured hippocampal neurons.

The four identified compounds were more abundant in the

Nchinpi #1 extract than in the chinpi #1 one, as the con-

tents of nobiletin and the four compounds (as % of dried

weight) in the chinpi #1 extract were 0.06 % for nobiletin,

0.034 % for tangeretin, 0.006 % for 6-demethoxynobiletin,

0.005 % for 6-demethoxytangeretin, and 0.007 % for sin-

ensetin (Yoshida, unpublished data), as opposed to the

respective values of 0.68, 0.20, 0.059, 0.045, and 0.074 %

for Nchinpi #1 extract. Accordingly, it might be likely that

such differences in the contents of these four flavonoids

and nobiletin may account for the difference in pharma-

cological activities between Nchinpi and standard chinpi;

although further chemical characterization of these natural

compounds in Nchinpi and standard chinpi is necessary.

1 2 30

10

20

30

40

50

60

Shock event

Control n = 10

MK-801 n = 11Nchinpi#1 extract + MK-801

3.67 g/kg, p.o., n = 8

TrainingA

Fre

ezin

g (

%)

***

****

###

###

MK-801 (0.08 mg/kg)

3.67Nchinpi #1

extract (g/kg)

Control

***

###

Fre

ezin

g (

%)

TestB

0

10

20

30

40

50

60

70

Control Vehicle MK-801

Nchinpi+

MK-801

Trained

Phospho-ERK1/2

Total-ERK1/2

Control Vehicle MK-801 Nchinpi+

MK-801

Trained

Ph

osp

ho

-ER

K1/

2

0

20

40

60

80

100

C

Fig. 6 Effects of 7-day oral administration of the Nchinpi #1 extract

on MK-801-induced learning and memory impairment in animal.

a The Nchinpi #1 extract (3.67 g/kg, p.o.) or a vehicle was orally

given to mice once daily for 7 consecutive days (i.e. day 1–7). On day

7, the Nchinpi extract or a vehicle was orally administered 120 min

before training, and MK-801 (0.08 mg/kg) or a vehicle was intraper-

itoneally given 30 min before training. b The test session was per-

formed 24 h after training. Values are mean ± SEM ***p \ 0.001

vs. control, ###p \ 0.001 vs. MK-801. c Restoration of MK-801-

induced impairment of learning-associated activation of ERK in the

hippocampus of mice by 7-day oral administration of Nchinpi #1

extract. The experimental procedures are described in ‘‘Materials and

methods’’. Values are mean ± SEM (n = 3). *p \ 0.05, **p \ 0.01

vs. control, ###p \ 0.001 vs. a vehicle (trained), $$$p \ 0.001 vs. MK-

801 (trained)

I. Kawahata et al.

123

Also, hesperidin, which is a major flavonoid detectable

commonly in both the extracts of standard chinpi and

Nchinpi, was found in the Fr. 1 of the Nchinpi #1 extract

(Yoshida, unpublished data): 300 lg/mL of Nchinpi #1

extract corresponded to approximately 9.5 lM hesperidin

on the basis of its content listed in Table 1. However, this

purified compound did not affect the transcription in the

cultured hippocampal neurons even at the concentration of

100 lM (Kawahata et al., unpublished data). Accordingly,

hesperidin is unlikely to have contributed to the tran-

scription-enhancing activity of the Nchinpi #1 extract.

Thus, the Nchinpi extract can be characterized by its robust

activity of stimulating CRE-mediated transcription in hip-

pocampal neurons in primary culture, in which nobiletin as

well as the four other natural compounds participate. We,

thus, can conclude that the Nchinpi extract was pharma-

cologically distinguishable from the standard chinpi

extract.

It is also noteworthy that sinensetin had a more potent

activity in terms of facilitating CRE-mediated transcription

in the cultured neurons than did nobiletin, although we

previously reported that nobiletin was the most potent in

facilitating CRE-mediated transcription among the five

compounds tested in PC12D cells (Nagase et al. 2005b).

The discrepancy between the results of the present study

and those of our previous one might be due to the cell-type-

dependent difference in the response to these natural

compounds. cAMP/PKA/ERK/CREB signaling is pivotal

for learning and memory (Atkins et al. 1998; Tully et al.

2003; Impey et al. 1998; Cammarota et al. 2000; Athos

et al. 2002; Levenson et al. 2004; Chwang et al. 2006; Kim

et al. 1991), and Ab inhibits this signaling pathway in

cultured hippocampal neurons (Vitolo et al. 2002).

Previous studies reported impaired CREB phosphorylation

in the brains of AD patients as well as in APP/PS1 trans-

genic mice (Yamamoto-Sasaki et al. 1999; Gong et al.

2004). As described above, our recent pilot clinical study

suggested the possibility that nobiletin-rich C. reticulata

peel #1 could be of benefit to improve the cognition of

patients with AD, with no adverse events (Seki et al. 2013).

Also, our current animal behavior test evidently showed

that oral administration of the Nchinpi #1 extract indeed

reversed MK801-impaired fear learning and memory in

mice, without any effects on animal shock sensitivity

(Nakajima et al., unpublished observations). Intriguingly,

the Nchinpi extract could improve not only MK-801-

impaired fear memory formation to a greater degree than

nobiletin, but also restore even MK-801-retarded memory

acquisition which nobiletin had no effects on in mice (Sun

et al., unpublished observations). In addition, our present

in vitro study demonstrated that the extract from the same

batch of Nchinpi facilitated PKA/ERK/CREB signaling

pathway-dependently the downstream CRE-mediated

transcription in neurons in primary culture. With regard to

CREB phosphorylation, we previously reported restoration

of Ab-impaired phosphorylation of CREB by nobiletin in

hippocampal neurons (Matsuzaki et al. 2006). Together

with these findings reported by other groups and us, the

results of the present study raise the possibility that these 4

bioactive compounds responsible for the potent CRE-

mediated transcription-stimulating activity of the Nchinpi

extract in vitro might antagonize Ab action in concert with

nobiletin, whereby they might have a potential therapeutic

benefit in AD patients.

How does the Nchinpi extract effectively facilitate PKA/

ERK/CREB signaling coupled with CRE-mediated

Nchinpi extract(100% as dry weight)

Fr. 1(98.57% )

Fr. 2 Polymethoxyflavonoid fraction

(1.43%)

Nobiletin(0.63%)

Fr. 2-1(0.89%)

ODS column chromatography40% acetonitrile

Silica gel column chromatographyAcetone / hexane (1 : 1) Silica gel column chromatography

Ethyl acetate / hexane (4 : 1)

Varian Bond Elut C18

30% acetonitrile MeOH

A

R1 R2 R3 R4 R5

Nobiletin OCH3 OCH3 OCH3 OCH3 OCH3

Tangeretin (1) H OCH3 OCH3 OCH3 OCH3

6-Demethoxynobiletin (2) OCH3 OCH3 H OCH3 OCH3

Sinensetin (3) OCH3 OCH3 OCH3 OCH3 H

6-Demethoxytangeretin (4) H OCH3 H OCH3 OCH3

B

O

R1

OCH3

OR2

R3

R4

R5

Fig. 7 Fractionation procedure used on Nchinpi extract #1 (a) and

the chemical structures of nobiletin and the four compounds identified

in subfraction 2–1 of the extract (b). Detailed experimental

procedures for fractionation of the Nchinpi #1 extract are described

in ‘‘Materials and methods’’


123

transcription in hippocampal neurons? An increase in the

intracellular concentration of cAMP causes PKA- or

exchange protein activated by cAMP- or Epac-dependent

activation of Ras or Rap1, a member of the Ras family,

thereby leading to ERK activation (Ye and Carew 2010).

The activation of ERK is required for activation of nuclear

RSK2, which in turn phosphorylates CREB to stimulate

CRE-dependent transcription (Xing et al. 1996). Our

present findings that the Nchinpi extract stimulated the

phosphorylation of PKA substrates in the neuronal cultures

thus possibly implicate cAMP/PKA-mediated signaling in

the action of the Nchinpi #1 extract. It is, therefore,

plausible that cAMP/PKA-dependent activation of Ras or

Rap1, which induces eventually the onset of ERK signal-

ing, may mechanistically in part be implicated in Nchinpi

extract-mediated facilitation of CRE-mediated transcrip-

tion associated with learning and memory in hippocampal

neurons.

In summary, the present study demonstrated that

extracts of nobiletin-rich C. reticulata peels, Nchinpi’s,

facilitated PKA/ERK/CREB signaling in hippocampal

neurons and potently enhanced CRE-mediated transcrip-

tion, which is known to be associated with learning and

memory (Atkins et al. 1998; Athos et al. 2002; Kim et al.

1991). Consistently, the Nchinpi extract did improve MK-

801-impaired learning and memory impairment by acti-

vation of ERK signaling in an animal model of dementia.

Moreover, tangeretin, 6-demethoxynobiletin, 6-demeth-

oxytangeretin, and sinensetin were identified as other

substances responsible, in cooperation with nobiletin, for

the facilitative action of the Nchinpi extracts on CRE-

mediated transcription in these neurons. Collectively, these

findings suggest that the Nchinpi extract has pharmaco-

logical advantages over standard chinpi extract in terms of

stimulatory effects on the intracellular signaling cascade

associated with learning and memory in hippocampal neu-

rons. Elucidation of the molecular mechanism underlying

Fig. 8 Effects of Nchinpi #1 extract, fraction, and subfractions

prepared from the extract, reconstituted samples, nobiletin, and

purified four compounds identified on CRE-mediated transcription in

cultured hippocampal neurons. a Effects of Nchinpi #1 extract,

fraction, and subfractions prepared from the extract, and reconstituted

samples on CRE-mediated transcription in cultured neurons. Neuro-

nal culture, transfection, and treatment with tested samples were

conducted as described under ‘‘Materials and methods’’. NC1:

300 lg/mL of Nchinpi extract #1; Fr. 2: 4.29 lg/mL of Fr. 2

(1.43 % of the extract); Fr. 2–1: 2.67 lg/mL of subfraction 2–1

(0.89 % of the extract); N: 1.89 lg/mL (or 5.0 lM) of nobiletin

isolated from Fr. 2 (0.63 % of the extract); Fr. 2–1 ? N: a

reconstituted sample prepared by combination of Fr. 2–1 and N;

REC: another reconstituted sample containing 0.585 lg/mL of

tangeretin, 0.177 lg/mL of 6-demethoxynobiletin, 0.135 lg/mL of

6-demethoxytangeretin, and 0.222 lg/mL of sinensetin found in Fr.

2–1 and 5.0 lM nobiletin in the ratio of each constituent’s content in

the extract. Nb 30 lM nobiletin; C control. Values are mean ± SEM

(n = 4). **p \ 0.01, ***p \ 0.001 vs. control. b Effects of the four

compounds purified from Fr. 2–1, each at the concentration of 30 lM,

on CRE-mediated transcription in cultured hippocampal neurons.

Culture and transfection were conducted as indicated in ‘‘Materials

and methods’’. C control, S1 tangeretin, S2 6-demethoxynobiletin, S36-demethoxytangeretin, S4 sinensetin, Nb nobiletin. Values are

mean ± SEM (n = 4). *p \ 0.05, **p \ 0.01, ***p \ 0.001 vs.

control. §§p \ 0.01 vs. nobiletin. c Concentration-dependent effects

of sinensetin and nobiletin on CRE-mediated transcription in cultured

hippocampal neurons. Cells were treated with sinensetin or nobiletin

at the concentrations of 1-30 lM. CNT control, Nob nobiletin, Sinsinensetin. Values are mean ± SEM (n = 4). ***p \ 0.001,##p \ 0.01, and ###p \ 0.001 vs. control. $$$p \ 0.001 vs. nobiletin

at each concentration tested

b

I. Kawahata et al.

123

the in vitro and in vivo beneficial actions of the extract of

nobiletin-rich C. reticulata peels might enable us to

develop new therapeutic drugs for dementia.

Acknowledgments We thank Dr. T. Shimizu for identification of

the origins of the crude drugs employed for the present study, and Dr.

L.D. Frye for English editing of the manuscript.

References

Abel T, Nguyen PV, Barad M, Deuel TAS, Kandel ER (1997) Genetic

demonstration of a role for PKA in the late phase of LTP and in

hippocampus-based long-term memory. Cell 88:615–626

Adams JP, Sweatt JD (2002) Molecular psychology: roles for the

ERK MAP kinase cascade in memory. Annu Rev Pharmacol

Toxicol 42:135–163

Athos J, Impey S, Pineda VV, Chen X, Storm DR (2002)

Hippocampal CRE-mediated gene expression is required for

contextual memory formation. Nat Neurosci 5:1119–1120

Atkins CM, Selcher JC, Petraitis JJ, Trzaskos JM, Sweatt JD (1998)

The MAPK cascade is required for mammalian associative

learning. Nat Neurosci 1:602–609

Cammarota M, Bevilaqua LR, Ardenghi P, Paratcha G, Levi de Stein

M, Izquierdo I, Medina JH (2000) Learning-associated activation

of nuclear MAPK, CREB and Elk-1, along with Fos production,

in the rat hippocampus after a one-trial avoidance learning:

abolition by NMDA receptor blockade. Brain Res Mol Brain Res

76:36–46

Chwang WB, O’riordan KJ, Levenson JM, Sweatt JD (2006) ERK/

MAPK regulates hippocampal histone phosphorylation follow-

ing contextual fear conditioning. Learn Mem 13:322–328

Frey U, Huang YY, Kandel ER (1993) Effects of cAMP stimulate a

late stage LTP in hippocampal CA1 neurons. Science

260:1661–1664

Gong B, Vitolo OV, Trinchese F, Liu S, Shelanski M, Arancio O

(2004) Persistent improvement in synaptic and cognitive func-

tions in an Alzheimer mouse model after rolipram treatment.

J Clin Invest 114:11624–11634

Impey S, Smith DM, Obrietan K, Donahue R, Wade C, Storm DR

(1998) Stimulation of cAMP response element (CRE)-mediated

transcription during contextual learning. Nat Neurosci

1:595–601

Kim JJ, DeCola JP, Landeria-Fernandez J, Fanselow MS (1991)

N-methyl-D-aspartate receptor antagonist APV blocks acquisi-

tion but not expression of fear conditioning. Behav Neurosci

105:26–33

Levenson JM, O’Riordan KJ, Brown KD, Trinh MA, Molfese DL,

Sweatt JD (2004) Regulation of histone acetylation during

memory formation in the hippocampus. J Biol Chem

279:40545–40559

Matsuzaki K, Yamakuni T, Hashimoto M, Haque AM, Shido O,

Mimaki Y, Sashida Y, Ohizumi Y (2006) Nobiletin restoring

beta amyloid-impaired CREB phosphorylation rescues memory

deterioration in Alzheimer’s disease model rats. Neurosci Lett

400:230–234

Matsuzaki K, Miyazaki K, Sakai S, Yawo H, Nakata N, Moriguchi S,

Fukunaga K, Yokosuka A, Sashida Y, Mimaki Y, Yamakuni T,

Ohizumi Y (2008) Nobiletin, a citrus flavonoid with neurotro-

phic action, augments protein kinase A-mediated phosphoryla-

tion of the AMPA receptor subunit, GluR1, and the postsynaptic

receptor response to glutamate in murine hippocampus. Eur J

Pharmacol 578:194–200

Nagase H, Yamakuni T, Matsuzaki K, Maruyama Y, Kasahara J,

Hinohara Y, Kondo S, Mimaki Y, Sashida Y, Tank WA,

Fukunaga K, Ohizumi Y (2005a) Mechanism of neurotrophic

action of nobiletin in PC12D cells. Biochemistry

44:13683–13691

Nagase H, Omae N, Omori A, Nakagawasai O, Tadano T, Yokosuka

A, Sashida Y, Mimaki Y, Yamakuni T, Ohizumi Y (2005b)

Nobiletin and its related flavonoids with CRE-mediated

transcription-stimulating and neuritegenic activities. Biochem

Biophys Res Commun 337:1330–1336

Nagata Y, Sakamoto K, Shiratsuchi H, Ishi T, Yano M, Ohta H (2006)

Flavonoid composition of fruit tissues of Citrus species. Biosci

Biotechnol Biochem 70:178–192

Nakajima A, Yamakuni T, Matsuzaki K, Nakata N, Onozuka H,

Yokosuka A, Sashida Y, Mimaki Y, Ohizumi Y (2007)

Nobiletin, a citrus flavonoid, reverses learning impairment

associated with NMDA receptor antagonism by activation of

ERK signaling. J Pharmacol Exp Ther 321:784–790

Onozuka H, Nakajima A, Matsuzaki K, Shin RW, Ogino K, Saigusa

D, Tetsu N, Yokosuka A, Sashida Y, Mimaki Y, Yamakuni T,

Ohizumi Y (2008) Nobiletin, a citrus flavonoid, improves

memory impairment and Ab pathology in a transgenic mouse

model of Alzheimer’s disease. J Pharmacol Exp Ther

326:739–744

Riedel G, Platt B, Micheau J (2003) Glutamate receptor function in

learning and memory. Behav Brain Res 140:1–47

Saigusa D, Shibuya M, Jinno D, Yamakoshi H, Iwabuchi Y,

Yokosuka A, Mimaki Y, Naganuma A, Ohizumi Y, Tomioka

Y, Yamakuni T (2011) High-performance liquid chromatogra-

phy with photo-diode array detection for determination of

nobiletin content in the brain and serum of mice administrated

the natural compound. Anal Bioanal Chem 400:3635–3641

Seki T, Kamiya T, Furukawa K, Azumi M, Ishizuka S, Takayama S,

Nagase S, Arai H, Yamakuni T, Yaegashi N (2013) Nobiletin-

rich Citrus reticulata peels, a kampo medicine for Alzheimer’s

disease: a case series. Geriatr Gerontol Int 13:236–238

Sweatt JD (2004) Mitogen-activated protein kinases in synaptic

plasticity and memory. Curr Opin Neurobiol 14:311–317

Toriizuka K (2003) Aurantii nobilis pericarpium. In: Toriizuka K (ed)

Monographs of pharmacological research on traditional herbal

medicines. Ishiyaku Publishers, Inc., Tokyo, pp 327–334

Tully T, Bourtchouladze R, Scott R, Tallman J (2003) Targeting the

CREB pathway for memory enhancers. Nat Rev Drug Discov

2:267–277

Vitolo OV, Sant’Angelo A, Costanzo V, Battaglia F, Arancio O,

Shelanski M (2002) Amyloid b-peptide inhibition of the PKA/

CREB pathway and long-term potentiation: reversibility by

drugs that enhance cAMP signaling. Proc Nat Acad Sci USA

99:13217–13221

Xing J, Ginty DD, Greenberg ME (1996) Coupling of the RAS-

MAPK pathway to gene activation by RSK2, a growth factor-

regulated CREB kinase. Science 273:959–963

Yamamoto-Sasaki M, Ozawa H, Saito T, Rosler M, Riederer P (1999)

Impaired phosphorylation of cyclic AMP response element

binding protein in the hippocampus of dementia of the Alzhei-

mer type. Brain Res 824:300–303

Ye X, Carew TJ (2010) Small G protein signaling in neuronal

plasticity and memory formation: the specific role of Ras family

protein. Neuron 68:340–361


123

Potent activity of nobiletin-rich Citrus reticulata peel extract to facilitate cAMP/PKA/ERK/CREB...

Documents

Transcript of Potent activity of nobiletin-rich Citrus reticulata peel extract to facilitate cAMP/PKA/ERK/CREB...