ARTICLE IN PRESS

Phytomedicine 16 (2009) 830838

www.elsevier.de/phymed

The effects of Rhodiola rosea extract on 5-HT level, cell proliferation and
quantity of neurons at cerebral hippocampus of depressive rats
Q.G. Chena, Y.S. Zenga,, Z.Q. Qua, J.Y. Tanga, Y.J. Qina, P. Chungb,
R. Wongc, U. Haggc
a
Department of Histology and Embryology, Zhongshan School of Medicine, Sun Yat-sen University, 74# Zhongshan Road 2,
Guangzhou 510080, China
b
Private Practice, Hong Kong, China
c
Biomedical and Tissue Engineering, Faculty of Dentistry, The University of Hong Kong, Hong Kong, China

Abstract
The purpose of this study was to investigate the effects of Rhodiola rosea extract and depression on the serotonin
(5-HT) level, cell proliferation and quantity of neurons at cerebral hippocampus of depressive rats induced by Chronic
Mild Stress (CMS). Seventy male Sprague-Dawley rats were divided into seven groups (10 per group): normal control
group, untreated depressive rat model group, negative control group, positive control group, low dosage Rhodiola
rosea extract (1.5 g/kg) group, medium dosage Rhodiola rosea extract (3 g/kg) group and high dosage Rhodiola rosea
extract (6 g/kg) group. After the depressive rats induced by CMS had received Rhodiola rosea extract for 3 weeks, the
5-HT levels at cerebral hippocampus were detected by high performance liquid chromatography. Bromodeoxyuridine
(BrdU) was injected in vivo to label the proliferating cells at hippocampus, and morphometry was used to count the
hippocampal neurons. The results showed that the 5-HT level of the three experimental groups had recovered to
normal status. The immunohistochemistry of hippocampus BrdU positive cells had returned to the normal level in the
group of depressive rats with low dosage Rhodiola rosea extract. In conclusion the results demonstrated that Rhodiola
rosea extract could improve 5-HT level in hippocampus in depressive rats, and low dosage Rhodiola rosea could
induce neural stem cell proliferation at hippocampus to return to normal level, repairing the injured neurons at
hippocampus.
r 2009 Elsevier GmbH. All rights reserved.

Keywords: Chronic mild stress; Depression; Rhodiola rosea extract; Hippocampus; Serotonin; Cell proliferation

Introduction ism of depression shows that depression results from the

Depression is a group of syndromes characterized by
notable and persistent mood disorders, and is one of the
most prevalent psychiatric disorders faced by humans,
belonging to the category of gloomy disease in Chinese
medicine (Cui and Liu 2003). Research on the mechan-
Corresponding author. Tel./fax: +86 20 87331452.
E-mail address: zengysh@mail.sysu.edu.cn (Y.S. Zeng).
interaction of nerve-endocrine-immunology system. But
the etiopathogenisis and pathophysiological mechan-
isms are complicated and have not been fully explained.
Current research has mainly targeted neurobiochemistry
and cerebral hippocampus injury.
Current research presumes that the serotonin (5-HT)
is the neurotransmitter most associated with depression.
Central and peripheral 5-HT levels decrease in depres-
sion patients (Stockmeier 1997). The 5-HT acts through

0944-7113/$ - see front matter r 2009 Elsevier GmbH. All rights reserved.
doi:10.1016/j.phymed.2009.03.011
 ARTICLE IN PRESS
Q.G. Chen et al. / Phytomedicine 16 (2009) 830838
5-HT receptor which is closely related to depression
(Kan and Ming 2005). Decrease of 5-HT receptor
expression levels in cerebrum may be the mechanism of
depression. In recent years, many studies (Amat et al.
1998) have presumed that injury in cerebral hippocam-
pus exerts a signicant inuence upon the incidence of
depression. Cerebral hippocampus not only highly
expresses subtypes of 5-HT receptor, and receptors of
glucocorticoid and mineralocorticoid, but also receives
5-HT nerve bers innervation from rapheal nuclei
(Chalmers and Watson 1991). Not only does the content
of 5-HT at hippocampus decrease, but also the function
and level of 5-HT receptor are reduced during long-term
stress state (Amat et al. 1998). Since 5-HT can bind
with 5-HT receptor to protect hippocampus, making its
function normally, the inuence that stress exerts upon
the 5-HT level and receptor at hippocampus might
render it more vulnerable to detrimental factors. 5-HT
derived from rapheal nuclei can modulate the function
of hippocampus directly.
Rhodiola rosea extract in the present study belongs
to crassulaceae integripetal Rhodiola herb genus, a
perennial herb, which is in 96 species globally worldwide
and 73 in China. Rhodiola rosea contains over 40
chemical components, of which salidroside, p-tyrosol,
rosavins (includes rosavin, rosin, and rosarin), rhodioni-
side, rhodiolin, rosiridin are thought to be the major
pharmacological active components (Kelly 2001). Most
multi-phenolic compounds and avonoids are anti-
oxidative and protects of cardiovascular system and
organs. The major impact of polysaccharide compounds
includes anti-viral, immunity improvement, down-
hyperglycemia and anti-tumor effects. In summary,
the pharmacological effects of Rhodiola rosea include
adaptogenic and anti-stress effects, anti-anoxia, anti-
fatigue, immunity improvement, protection for central
nervous system and cardiovascular system (Brown et al.
2002).
Recently, studies related to effects of Rhodiola rosea
in both depressive rat test (Panossian et al. 2008;
Mattioli et al. 2009) and human treatment (Darbinyan
et al. 2007) have been carried out and demonstrated the
antidepressive effects of the medicinal plant. However,
current literature has only focused on behavioral
changes in depression, and few researchers have referred
to the effects of Rhodiola rosea on neurotransmitter and
neural pathology abnormities (e.g. 5-HT, cell prolifera-
tion, etc.) in hippocampus of depressive rats (Lin et al.
2005). There have been a growing number of publica-
tions on the effect of Rhodiola rosea on cerebrum. Song
et al. (2005) showed that Rhodiola rosea had a protective
effect on the hippocampus and dentate gyrus of cerebral
ischemia-reperfusion in injured rats. Zhang et al. (2005)
investigated the effect of medicated serum containing
salidroside on the differentiation of hippocampus neural
stem cells into neurons in neonate rats, and the results
831
showed medicated serum containing salidroside in vitro
could stimulate neural stem cells to differentiate into
neurons. It is assumed that Rhodiola rosea extract could
be applied to alleviate pathological changes in depres-
sive hippocampal neurons, which may promote survival
of the injured hippocampal neurons, facilitate prolifera-
tion of the hippocampus cells, and would be benecial to
the treatment of depression. The aim of this study was
to investigate the effect of Rhodiola rosea extract on
Chronic Mild Depression on an experimental model.
Meterials and methods
Rhodiola rosea extract
An edible alcohol extract of Rhodiola rosea root was
used in this study. The root of the plant was dried and
ground coarsely with a grinder. For the preparation
of the alcohol extract, the coarse powder of the plant
was extracted with 70% alcohol twice, 2 hours at a time.
The extract was condensed by vacuum concentration,
and then collected with ethanol precipitation, nally
spray dried to yield a reddish-brown powder. The yield
of Rhodiola rosea extract was about 3-5% (w/w).
The concentration of the known active component
salidroside in the extract was determined by HPLC
and was found to be 4% w/w.
Preparation for experimental animals and
experiment
Seventy male Sprague-Dawley rats body weight
180-200 g, were normally fed for 1 week to adjust (Tang
et al. 2008). Room temperature was kept 18-22 1C, the
rats had sufcient water and food supply, and controlled
noise interference, 12/12 h day and night alternation.
The rats were prepared by feeding them 1% sucrose
during adaptation period. After 1 week of adaptation,
no food or water was supplied for 20 h, and controlled
the intake of tap water and 1% sucrose of each rat
within one hour by measuring the weight of bottle.
Thereafter the rats were divided into normal control
group (10 rats) and depression model group (60 rats)
based to the approximate intake of 1% sucrose
including (Tang et al. 2008): i) normal control group:
no treatment for 4 weeks (weeks 0-4). Then received oral
0.5% sodium carboxymethycellulose (solvent) 2 mL
twice daily via intragastric administration for three
weeks (weeks 5-7). ii) untreated depressive rat model
group: prepared depressive rats, received chronic mild
stress modied from Grippo et al. (2005) and Lin et al.
(2005) for 4 weeks (weeks 0-4). After chronic mild
stress, extent of weight increasing of prepared depressive
rats was lower than that of normal control group above.
 ARTICLE IN PRESS
832 Q.G. Chen et al. / Phytomedicine 16 (2009) 830838
Tap water intake of two groups did not change with the
time passing. 1% sucrose intake of prepared depressive
rats was decreased, but 1% sucrose intake of normal
control group was increased (Table 1, weeks 04). Then
no treatment for 3 weeks (weeks 57). 1% sucrose intake
was used as an indicator for depression. iii) negative
control group: prepared depressive rats as in (ii) (weeks
04). Then received oral 0.5% sodium carboxymethy-
cellulose as in (i) (weeks 57). iv) positive control group:
prepared depressive rats as in (ii) (weeks 04). Then
received oral 0.5% sodium carboxymethycellulose as in
(i) (weeks 57), dissolved with uoxetine (Eli Lilly,
USA), an antidepressive drug at dosage 2.2 mg/kg
(Wang et al. 2005). v) low dosage of Rhodiola rosea
extract model group: prepared depressive rats as in
(ii) (weeks 0-4). Then received oral 0.5% sodium
carboxymethycellulose as in (i) (weeks 5-7), dissolved
with Rhodiola rosea extract (1.5 g/kg) (Holistal Interna-
tional Ltd., Hong Kong, China). vi) medium dosage
of Rhodiola rosea extract model group: prepared
depressive rats as in (ii) (weeks 0-4). Then received oral
0.5% sodium carboxymethycellulose as in (i) (weeks
5-7), dissolved with Rhodiola rosea extract (3.0 g/kg). vii)
high dosage of Rhodiola rosea extract model group:
prepared depressive rats as in (ii) (weeks 0-4). Then
received oral 0.5% sodium carboxymethycellulose as in
(i) (weeks 5-7), dissolved with Rhodiola rosea extract
(6.0 g/kg).
During this part of the experiment the rats were kept
with room temperature 18-22 1C, had adequate water
and standard rat feeds, controlled noise interference,
12h/12h day and night alternation.
All groups were examined by weight, tap water intake
and 1% sucrose intake once a week. The rats were killed
after week 7. The hippocampal tissue of brain for half of
the groups (n 5) were removed for high-performance
liquid chromatography (HPLC) (Waters 1525, USA)
detection of serotonin (5-HT) (Banasr et al. 2004). In
other half of the groups (n 5), the proliferative cells in
hippocampus were detected by immunohistochemistry
using bromodeoxyuridine (BrdU) (Sigma, USA) label-
ling (Santarelli et al. 2003).
Statistical testing of difference between groups was
performed using one way ANOVA with the level of
signicance set at po0.05. Data were analyzed with a
statistical analysis computer software (SPSS v12.0,
Chicago, USA).
Administration of drugs
During the experimental period of three weeks the
drugs were prepared and administrated as follows: 0.5%
CMC solution (0.5 gram of sodium carboxymethycellu-
lose dissolved in 100 ml of double-steamed water) was
the solvent; the Rhodiola rosea extract was dissolved in
0.5% CMC and administrated in different dosage by
6.0 g/kg, 3.0 g/kg, and 1.5 g/kg respectively; uoxetine
was disolved in 0.5% CMC and administrated by
2.2 mg/kg. The total daily drug dosage was 4 ml, 2 ml
administrated twice a day.
Specimens taken and section
The rats were sacriced, infused with 200 ml of 0.1 M
PB, then infused and xed with 400 ml of 4%
polyoxymethylene (0.1 M PB preparation). The rat
brains were postxed with fresh 4% polyoxymethylene
for 4-6 hours, and then soaked in 20% sucrose (0.1 M
PB preparation) and 30% sucrose until the specimens
sank to the bottom. Coronal frozen sections began from
hippocampus beaks and slices were obtained every
300 mm with 40 mm and 15 mm thickness, respectively.
Six slices were taken from each brain biopsy, and in
four sets.
Preparation and HPLC for 5-HT detection
The rats were anesthetized with 1% sodium pento-
barbital, decapitated, hippocampi was taken out in an
icy environment, weighed, put in EP tube, and measured
the 5-HT by HPLC with electrochemical detection as
previously described (Vasconcelos et al. 2004; Roma-
niuk et al. 2001). Preparation of the material: added
0.5 ml perchlorate extract (0.1 M perchloric containing
0.1% cysteine, internal standard 64 ng/ml) to each
specimen, ground into homogenate in the icy bath,
centrifuged by 18,000 roll/min, 4 1C for 20 min, put
supernatant in another centrifugal tube, 20 1C for
preservation, and 20 ml for analysis. Chromatographic
conditions: 1525 high-pressure pump from WATERS
Company, HP1049A electrochemical detector, test
voltage 0.7 Mv, BDS C18 chromato bar; (Dalian Yilite
Company), 1.6  250 mm, particle diameter 5 mm. Buffer
solution: 3 mM sodium heptanesulfonate, 1003 mM
sodium acetate, 853 mM citric acid, 0.2 M medta,
pH 4.0. Mobile phase: buffer solution: methanol 92:8
(V/V). Flow: 1.0 ml/min. N2000 chromatography data
workstations (Zhejiang Intelligent Information Engi-
neering Institute). Qualitative analysis: comparison
between samples peak retention time and standard
peak retention time. Quantitative Methods: DHBA as
the internal standard, drew standard curve with ratio of
5-HT standard substance peak area to internal standard
peak area, calculated sample content by applying ratio
of sample peak area to internal standard peak area to
the curve. The concentrations of 5-HT in samples were
expressed as ng/g wet tissue.
Table 1. Comparison of the body weight, tap water intake, and 1% sucrose intake in seven groups of rats.

Group Parameter (X7sx, g) Time (week)

0 4 5 6 7

normal control body weight 262.00710.33 313.00 76.32 325.50 77.96a 344.5079.26h 358.0076.32o
tap water intake 0.9670.27 2.1370.49 2.0570.06 2.4670.39 2.8770.21
1% sucrose intake 11.0671.34 12.3070.87 12.5370.91aa 13.8270.96hh 15.4871.56oo

untreated depressive body weight 261.0077.75 303.50 712.92 314.00 714.49b 332.50 712.30i 348.00711.83p

Q.G. Chen et al. / Phytomedicine 16 (2009) 830838

rat model tap water intake 0.9470.29 2.3270.38 2.2770.22 2.4470.21 2.4670.13
1% sucrose intake 11.1871.27 7.4170.91 9.1470.94bb 10.7771.29ii 12.1370.89pp

negative control body weight 259.5077.25 300.5078.16 309.0078.10c 328.5076.69j 348.5076.39q

ARTICLE IN PRESS
tap water intake 0.9770.36 2.3170.49 2.2870.25 2.2470.37 2.6470.26
1% sucrose intake 11.2871.27 8.2471.06 9.7571.22cc 10.8071.00jj 11.9570.81qq

positive control body weight 261.5079.44 303.0076.75 319.5077.98d 341.50710.01k 361.50710.01r

tap water intake 1.0270.30 2.3670.36 2.7770.28 2.5370.24 2.6870.38
1% sucrose intake 11.0471.10 7.7470.80 11.5970.86dd 13.4270.73kk 14.8271.15rr
low dosage of body weight 260.0079.13 303.0078.43 324.0076.99e 339.0076.58l 366.5077.84s
Rhodiola rosea extract tap water intake 0.9970.33 2.3970.22 2.6770.33 2.8970.18 2.2170.30
model 1% sucrose intake 11.1071.22 7.4470.72 9.4270.64ee 12.3170.96ll 15.4471.14ss
medium dosage of body weight 262.00712.06 304.5078.58 324.5074.38f 348.5076.26m 363.5079.44t
Rhodiola rosea extract tap water intake 0.9770.27 2.2870.38 2.0670.37 2.0770.29 2.1670.25
model 1% sucrose intake 10.8171.36 7.2470.66 9.9271.29ff 12.3370.96mm 13.3070.97tt

high dosage of body weight 263.00710.85 305.0077.45 328.0077.15g 343.5076.26n 360.0075.77u

Rhodiola rosea extract tap water intake 0.9870.19 2.1670.21 2.0870.28 2.3570.22 2.1170.28
model 1% sucrose intake 11.2671.39 7.2270.81 8.4071.28gg 9.3070.60nn 12.2771.05uu

For body weight: Oneway ANOVA; aVSc, aVSb, aVSf, cVSd, cVSg, bVSc, bVSf, dVSf, dVSg, hVSi, hVSj, hVSl, iVSk, iVSl, iVSm, iVSn, jVSk, JVSl, jVSm, jVSn, kVSl, kVSm, kVSn, lVSm, oVSp,
oVSq, oVSr, pVSr, qVSr, rVSt, rVSn, tVSn, rVSs, Po0.05; aVSd, aVSe, aVSg, bVSd, bVSe, bVSg, cVSe, cVSf, dVSe, eVSf, eVSg, fVSg, hVSk, hVSm, hVSn, iVSj, IVSn, mVSn , oVSt, oVSs, oVSu,
pVSq, pVSs, pVSt, pVSu, qVSs, qVSt, qVSu, qVSt, sVSt, sVSu, tVSu, P40.05. For tap water intake: Oneway ANOVA, P40.05. For 1% sucrose intake: Oneway ANOVA aaVSbb, aaVScc,
aaVSdd, aaVSee, aaVSff, aaVSgg, bbVSdd, ccVSdd, ccVSgg, ddVSee, ddVSff, ddVSgg, eeVSgg, ffVSgg, hhVSii, hhVSjj, hhVSll, hhVSmm, hhVSnn, iiVSkk, iiVSll, iiVSmm, jjVSkk, jjVSll, jjVSmm,
jjVSnn, kkVSll, kkVSmm, kkVSnn, IIVSnn, ooVSpp, ooVSqq, ooVStt, ooVSuu, ppVSrr, ppVSss, ppVStt, qqVSrr, qqVSss, qqVStt, qqVSun, qqVStt, rrVStt, rrVSnn, ssVStt, ssVSnn, ttVSnn,
ttVSuu, Po0.05; bbVScc, bbVSee, bbVSff, bbVSgg, ccVSee, ccVSff, eeVSff, hhVSkk, iiVSjj, iiVSnn, IIVSmm, ooVSrr, ooVSss, ppVSqq, ppVSuu, rrVSss, P40.05.

833
 ARTICLE IN PRESS
834 Q.G. Chen et al. / Phytomedicine 16 (2009) 830838
Nissl staining for slices and survival neurons counting
in hippocampus
The sample was washed in 0.01 M PBS for 5 min,
soaked in 1% neutral red staining solution for 20 min,
rinsed three times with tap water, 5 s for each time,
gradient alcohol dehydration with 70%, 80%, 95% and
100%, with 3 seconds for each gradient, cleared twice
using xylene, rst time for 5 seconds, second time for 5
minutes, and mounted by neutral gum. The stained
brains were sliced, images of 5 districts (CA1-1, CA1-1,
CA3, DG-1, DG-2) in unilateral hippocampus of every
experimental group obtained by using Leica microscope
with the digital camera system under 400 times
magnication, counted neurons in these ve districts
with Photoshop software. Used the mean value of
neuron quantity in the 5 districts of right and left lateral
hippocampi to indicate the neuron quantity.
BrdU label, BrdU immunohistochemistry and BrdU-
positive cell count
Five rats in each group were selected at random, and
labeled the proliferated cells with intraperitoneal injec-
tion of BrdU (dissolved in the saline 7 M NaOH), with
75 mg/kg, every two hours. The rats had four injections
and sacriced 24 hours after the last injection. BrdU
immunohistochemical staining was done according to
Kuhn et al. (1996): the slices washed in 0.01 M PBS
5 min  3 times followed by treatment with 3% H2O2 at
room temperature for 30 min, washed with 0.01 M PBS
5 min  3 times, 50% formamide/2  SSC incubator
treatment at 65 1C for 2 hours, washed with 2  SSC
5 min  3 times, 2 M HCl incubator treatment at 37 1C
for 30 min, washed with 0.1 M boric acid buffer
(pH 8.0) 5 min  3 times, blocked with 1:50 normal
goat serum for 30 min, incubated with 1:400 BrdU
antibody, 4 1C overnight, washed with 0.01 M PBS
5 min  3 times, reacted with 1:200 biotin labeled sheep-
anti-rat IgG at room temperature for 30 min, washed
with 0.01 M PBS 5 min  3 times, 1:100 SABC at room
temperature for 30 min, DAB coloration, lightly dyed
using hematoxylin, conventionally dehydrated and
mounted. The hippocampal BrdU positive cells of the
granular cell underlayer of dentate gyrus on the 6 brain
slices of each rat was counted under light microscope,
using the total cell quantity of bilateral hippocampal
subgranular zones (SGZ) on the 6 slices to reect the
proliferation status of cells in SGZ.
Statistical analysis
The EXCEL and SPSS13.0 software were used to
calculate Mean and standard deviations, and applied
ANOVA or t-test were used to analyze weight, tap water
intake and 1% sucrose intake. Single-factor analysis
of variance was used to assess neurons count, BrdU-
positive cells count and 5-HT content in hippocampus
with. The level of signicance was set at po0.05.
Results
Weight, tape water and sucrose intake
After the application of sodium carboxymethycellu-
lose, uoxetine and Rhodiola rosea extract, extent of
weight increasing of prepared depressive rat model
groups was lower than that of normal control group.
Tap water intake of seven groups did not change with
the time passing. 1% sucrose intake in low dosage of
Rhodiola rosea extract model group rats increased and
recovered to normal level (Table 1, weeks 57).
5-HT level in rats hippocampus
After the application of uoxetine, low-dose, med-
ium-dose and high-dose of Rhodiola rosea extract to
the groups of depression model rats for three weeks,
the 5-HT content of the untreated model group and
negative control group were markedly lower than
normal control group (po0.05, Table 2). After the
administration of uoxetine, 5-HT level in hippocampus
of depressive rats increased and become signicantly
higher (po0.05, Table 2) than that of normal control
group. The 5-HT level in low dosage of Rhodiola rosea
extract model group was also signicantly higher
(po0.05, Table 2) than that of the normal control
group. In the medium dosage group and high dosage
group, 5-HT level had returned to the normal level, a
signicant increase (po0.05, Table 2).
Number of BrdU-positive cells in hippocampus
Three weeks of administrated the depression model
rats with uoxetine, low-dose, medium-dose and high-
dose of Rhodiola rosea extract resulted in that the
number of BrdU positive cells at hippocampi in
untreated model group and negative control model
group (Fig. 1G) had decreased markedly (Fig. 1F)
(po0.05, Table 2). After administration of uoxetine
(Fig. 1H), the number of BrdU positive cells in
hippocampus increased signicantly (p40.05, Table 2)
and had returned to normal level. Low-dose adminis-
tration of Rhodiola rosea extract (Fig. 1I) increased the
number of BrdU positive cells to normal status (p40.05,
Table 2), whereas both medium-dose and high-dose
administration of Rhodiola rosea extract (Fig. 1J) did
not increase the number of BrdU-positive cells in
hippocampus.
 ARTICLE IN PRESS
Q.G. Chen et al. / Phytomedicine 16 (2009) 830838 835

Table 2. Comparison of the contents of 5-HT at hippocampus, the number of BrdU positive cells at hippocampal DG and the
number of neuron at hippocampus of brain in seven groups of rats.

Group 5-HT (X7sx, ng/g) Hippocampal DG BrdU hippocampus neuron

positive cells (X7sx) number (X7sx)
n 5 5 10

normal control left side 1237743aaa

right side 0.268070.03a 33.0074.42aa 1231739hhh

untreated depressive rat left side 1126755bbb

model right side 0.182070.03b 18.0071.52bb 1124731iii

negative control left side 1142740ccc

right side 0.214070.03c 17.2071.53cc 1139734jjj

positive control left side 1188725ddd

right side 0.396070.06d 31.6078.11dd 1193724kkk

low dosage of Rhodiola left side 1208724eee

rosea extract model right side 0.322070.04e 30.2073.52ee 1213729lll

medium dosage of left side 1139730fff

Rhodiola rosea extract right side 0.276070.03 f
19.2073.70 ff 1134750mmm
model

high dosage of Rhodiola left side 1121756ggg

rosea extract model right side 0.254070.02g 18.0071.87gg 1120720nnn

For 5-HT: Oneway ANOVA; aVSb, aVSc, aVSd, aVSe, bVSd, bVSe, bVSf, bVSg, cVSd, cVSe, cVSf, cVSg, dVSe, dVSf, dVSg, eVSf, eVSg, po0.05;
aVSf, aVSg, bVSc, fVSg, p40.05. For BrdU positive cells: Oneway ANOVA; aaVSbb, aaVScc, aaVSff, aaVSgg, bbVSdd, bbVSee, ccVSdd, ccVSee,
ccVSff, ddVSff, ddVSgg, eeVSff, eeVSgg, po0.05; aaVSdd, aaVSee, bbVScc, bbVSff, bbVSgg, ccVSgg, ddVSee, ffVSgg, p40.05. For hippocampus
neuron number: Oneway ANOVA; aaaVSbbb, aaaVSccc, aaaVSfff, aaaVSggg, bbbVSddd, bbbVSeee, cccVSddd, cccVSeee, dddVSfff, dddVSggg,
eeeVSfff, eeeVSggg, hhhVSiii, hhhVSjjj, hhhVSmmm, hhhVSnnn, iiiVSkkk, iiiVSlll, jjjVSkkk, jjjVSlll, kkkVSmmm, kkkVSnnn, lllVSmmm,
lllVSnnn, po0.05; aaaVSddd, aaaVSeee, bbbVSccc, bbbVSfff, bbbVSggg, cccVSfff, cccVSggg, dddVSeee, fffVSggg, hhhVSkkk, hhhVSlll, iiiVSjjj,
iiiVSmmm, iiiVSnnn, jjjVSmmm, jjjVSnnn, kkkVSlll, mmmVSnnn, p40.05.

Number of neurons in hippocampus
The total number of hippocampal neurons in positive
control model (Fig. 1C) and low-dose Rhodiola rosea
extract model group (Fig. 1D) were restored to normal
levels (p40.05, Table 2) (Fig. 1A). In untreated model
group, negative control model (Fig. 1B), medium-dose
Rhodiola rosea extract model group, high-dose Rhodiola
rosea extract group (Fig. 1E), the total number of
neurons in the hippocampus were signicantly lower
than normal level (po0.05, Table 2).
Discussion
Chronic medium intensity stimulation model is
currently the most widely applied model of depression,
making animals receive long-term mild stress to simulate
various negative events faced by people in their daily life
(Willner 1997).
In the present study, with lasting chronic medium
intensity stimulation, the amount of body weight gain of
the groups for depression models was less than that of
normal control group. The results indicated that, after
four weeks of chronic medium intensity stimulation, the
intake of 1% sucrose decreased in the depression model
groups, illustrating that those rats had clinical symptom
of anhedonia and thus suffered from depression.
Consequently it could be presumed that chronic medium
intensity stimulation successfully induced depression in
the rats.
Recent research showed that depression was asso-
ciated with suppressed proliferation and apoptotic
changes in hippocampal tissue (Heine et al. 2004).
Suppressed proliferation and apoptotic changes in the
rat dentate gyrus after acute and chronic stress are
reversible. Hippocampal neurogenesis is required for the
behavioral effects of antidepressants (Santarelli et al.
2003). Therefore, a measurement of BrdU positive cells
and neurons in the hippocampus can be an indicator
to measure the effectiveness of an antidepressive agent.
 ARTICLE IN PRESS
836 Q.G. Chen et al. / Phytomedicine 16 (2009) 830838

Fig. 1. The neurons in the CA3 of the hippocampus, stained by neutral red, one of the Nissl staining (A-E) and the nuclei of BrdU-
positive cells in subgranular zones (SGZ) of hippocampus (F-J). (A) The quantity of neurons was intens in the rats without
treatment. (B) 4 weeks chronic mild stress could destroy the neuron. (C) Fluoxetine could reverse the decreasing of neuronal
quantity. (D) The quantity of neurons was increased in the depressive rats fed with low dosage Rhodiola rosea for 3 weeks. (E)
Addition of high dosage Rhodiola rosea couldt promote the number of neuron of depressive rats in hippocampal CA3. Scale
bar 40 mm (A-E). (F) The proliferative cells in hippocampal SGZ were stained with antibodies against BrdU, a marker of
proliferative cell. BrdU immunoreactivity was normal in normal rats. (G) BrdU immunoreactivity was weak in the rats suffered
from depression. (H) BrdU immunoreactivity became intense in the depressive rats fed with uoxetine for 3 weeks. (I) The number
of BrdU-positive cells in hippocampal SGZ increased in the pressent of low dosage Rhodiola rosea. (J) Addition of high dosage
Rhodiola rosea decreased the number of BrdU-positive cells in hippocampal SGZ. Scale bar 40 mm (F-J).
 ARTICLE IN PRESS
Q.G. Chen et al. / Phytomedicine 16 (2009) 830838
In addition, Banasr et al. (2004) showed 5-HT was
related to hippocampal neurogenesis and this were
mediated through different and common 5-HT receptor
subtypes. Grippo et al. (2005) showed chronic mild
stress induced behavioral and physiological changes,
and may alter serotonin 1A receptor function. An
increase in 5-HT level can increase the neurogenesis
in hippocampus and thus causing an improvement in
depression, causing a change in depressive behavior to
normal status.
In the present study was found that low-dose
administration of Rhodiola rosea extract could raise
the number of hippocampal neurons, to the level of
normal rats, but not in medium-dose and high-dose
groups. This indicates that Rhodiola rosea extract has
the potential to affect the cerebral hippocampal neurons
in depressive rats but the effect is dose sensitive.
Previous studies showed that Rhodiola rosea extract
had a protective effect on the hippocampal region and
dentate gyrus in rats suffered from ischemia-reperfusion
(Song et al. 2005). Rhodiola rosea extract seems to
protect the hippocampal neurons of the rats in several
ways. Furthermore, salidroside can protect the mito-
chondria in neurons, stabilize mitochondrial membrane
potential, improve mitochondrial oxidative phosphor-
ylation process, regulate cell energy metabolism, and
thereby inhibit apoptosis of the neurons (Pang et al.
2005). The reason why medium-dose and high-dose
of Rhodiola rosea extract does not, and low-dose of
Rhodiola rosea extract does affect the hippocampal
neurons may due to that the Rhodiola rosea extract use
was the natural plant, which contains a variety of active
components. One component provided the dosage is
kept below a certain level may have a protective effect
on one organ while having an adverse effect on other
organs (Shen 1995).
In a previous study, Rhodiola rosea extract has been
found to regulate the content of 5-HT in the brain of rat
(Stancheva and Mosharrof 1987). This implies Rhodiola
rosea extract may improve the 5-HT deciency in
hippocampus in depressive rats. Our present results
indicate Rhodiola rosea extract independent of the dose
could raise the 5-HT level in the cerebral hippocampus
of depressive rats to normal levels, but low-dose could
raise the 5-HT level to a even higher level, and which
is general agreement with some previous studies (Guan
and Wu 2004).
The results of the present study indicate that low-dose
administration of Rhodiola rosea extract to depressive
rats will restore the quantity of proliferative cells (BrdU-
positive cells) in dentate gyrus. BrdU, 5-bromodeoxyur-
idine, the analog to deoxyuradine, integrates into DNA
in the S-phase of cell mitosis, so the BrdU-positive
cells appearing in hippocampal dentate gyrus are
nascent neural stem cells. It is possible that Rhodiola
rosea extract improves the hippocampal 5-HT content in
837
depressive rats, which affects the neurogenesis in dentate
gyrus and results in increase of neural stem cells.
Acknowledgements
We would like to thank Hong Kong Holistal
International Ltd for the donation of Rhodiola rosea
extract, and acknowledge purchase of sucrose was made
from Guangzhou Chemical Agents Factory, uoxetine
from Eli Lilly and Company Limited, CMC from the
Chemical Agents and Glasswares Wholesale Depart-
ment of Guangzhou Pharmaceutical Company Ltd.
References
Amat, J., Matus-Amat, P., Watkins, L.R., Maier, S.F., 1998.
Escapable and inescapable stress differentially and selec-
tively change extracelluar levels of 5-HT in the ventral
hippocampus and dorsal periaqueductal gray of the rat.
Brain Res. 797, 1222.
Banasr, M., Hery, M., Printemps, R., Daszuta, A., 2004.
Serotonin induced increases in adult cell proliferation
and neurogenesis are mediated through different and
common 5-HT receptor subtypes in the dentate gyrus and
the subventricular zone. Neuropsychopharmacology 29,
450460.
Brown, R.P., Gerbarg, P.L., Ramazanov, Z., 2002. Rhodiola
rosea: a phytomedicinal overview. HerbalGram 56, 4052.
Chalmers, D.T., Watson, S.J., 1991. Comparative anatomical
distribution of 5-HTlA receptor mRNA and 5-HT1A
binding in the rat brain: a combined in situ hybridization/
in vitro receptor autoradiographic study. Brain Res. 561,
5160.
Cui, S.Y., Liu, X.H., 2003. The relevance between schizo-
phrenia among soldiers and social operation. J. Clin.
Psychiatry 13, 163.
Darbinyan, V., Aslanyan, G., Amroyan, E., Gabrielyan, E.,
Malmstrom, C., Panossian, A., 2007. Clinical trial of
Rhodiola rosea L. extract SHR-5 in the treatment of mild to
moderate depression. Nord. J. Psychiatry 61 (5), 343348.
Grippo, A.J., Sullivan, N.R., Damjanoska, K.J., Crane, J.W.,
Carrasco, G.A., Shi, J., Chen, Z., Garcia, F., Muma, N.A.,
Van de Kar, L.D., 2005. Chronic mild stress induces
behavioral and physiological changes, and may alter
serotonin 1A receptor function in male and cycling female
rats. Psychopharmacology 179, 769780.
Guan, J.L., Wu, Z.H., 2004. Experimental research on the
anti-aging effects of Tibet Rhodiola rosea extract. J. Chin.
Med. Mater. 27 (5), 365367.
Heine, V.M., Maslam, S., Zareno, J., Joels, M., Lucassen, P.J.,
2004. Suppressed proliferation and apoptotic changes in the
rat dentate gyrus after acute and chronic stress are
reversible. Eur. J. Neurosci. 19, 131144.
Kan, H.W., Ming, L., 2005. Effects of 5-HTR in the incidence
and treatment of depression. Shandong Med. 45, 6668.
Kelly, G.S., 2001. Rhodiola rosea: a possible plant adaptogen.
Altern. Med. Rev. 6 (3), 293302.
 ARTICLE IN PRESS
838 Q.G. Chen et al. / Phytomedicine 16 (2009) 830838
Kuhn, H.G., Dickinson-Anson, H., Gage, F.H., 1996.
Neurogenesis in the dentate gyrus of the adult rats:
age-related decrease of neuronal progenitor proliferation.
J. Neurosci. 16, 20272033.
Lin, Y.H., Liu, A.H., Xu, Y., Tie, L., Yu, H.M., Li, X.J., 2005.
Effect of chronic unpredictable mild stress on brainpan-
creas relative protein in rat brain and pancreas. Behav.
Brain Res. 165, 6371.
Mattioli, L., Funari, C., Perfumi, M., 2009. Effects of
Rhodiola rosea L. extract on behavioural and physiological
alterations induced by chronic mild stress in female rats. J.
Psychopharmacol. 23 (2), 130142.
Pang, H., Zhu, L.Q., Zhang, W.S., 2005. Effects of salidroside
on mitochondrial membrane potential and apoptosis of
neurons suffering from hypoxia and glycose deciency
injury. Chin. Med. Publ. 23, 21442146.
Panossian, A., Nikoyan, N., Ohanyan, N., Hovhannisyan, A.,
Abrahamyan, H., Gabrielyan, E., Wikman, G., 2008.
Comparative study of Rhodiola preparations on behavioral
despair of rats. Phytomedicine 15 (12), 8491.
Romaniuk, A., Koprowska, M., Krotewicz, M., Strzelczuk,
M., Wieczorek, M., 2001. Effects of 8-OHDPAT adminis-
tration into the dorsal raphe nucleus and dorsal hippo-
campus on fear behavior and regional brain monoamines
distribution in rats. Behav. Brain Res. 120 (1), 4757.
Santarelli, L., Saxe, M., Gross, C., Surget, A., Battaglia, F.,
Dulawa, S., Weisstaub, N., Lee, J., Duman, R., Arancio,
O., Belzung, C., Hen, R., 2003. Requirement of hippo-
campal neurogenesis for the behavioral effects of anti-
depressants. Science 301, 805808.
Shen, Y.J. (Ed.), 1995. Pharmacology of Chinese Medicine.
Shanghai Science and Technology Press, Shanghai, China,
pp. 1218
Song, Y.Y., Li, Y.P., Lv, Y., 2005. Protective effect of
Rhodiola rosea on cerebral ischemia-reperfusion in rat
hippocampus and dentate gyrus. Chin. Clin. Rehab. 9,
232234.
Stancheva, S.L., Mosharrof, A., 1987. Effect of the extract of
Rhodiola rosea L. on the content of the brain biogenic
monoamines. Med. Physiol. CR Acad. Bulg. Sci. 40, 8587.
Stockmeier, C.A., 1997. Neurobiology of serotonin in depres-
sion and suicide. Ann. N. Y. Acad. Sci. 836, 220232.
Tang, J.Y., Zeng, Y.S., Chen, Q.G., Qin, Y.J., Chen, S.J.,
Zhong, Z.Q., 2008. Effects of Valerian on weight and
behavior of depressive rats induced by chronic mild stress.
J. Sun Yat-sen Univ. (Med. Sci.) 29 (5), 541545.
Vasconcelos, S.M., Cavalcante, R.A., Aguiar, L.M., Sousa,
F.C., Fonteles, M.M., Viana, G.S., 2004. Effects of chronic
ethanol treatment on monoamine levels in rat hippocampus
and striatum. Braz. J. Med. Biol. Res. 37 (12), 18391846.
Wang, Z.J., Lu, B.X., Ji, Z., 2005. Effect of uoxetine
pretreatment on the expression of BDNF and bcl-2 in rat
hippocampus with cerebral ischemia reperfusion injury.
Chin. J. Minimally Invasive Neurosurg. 10, 414416.
Willner, P., 1997. Validity, reliability and utility of the chronic
mild stress model of depression: a 10-year review and
evaluation. Psychopharmacology (Berlin) 134, 319329.
Zhang, W.H., Li, Y.J., Zhong, Z.Y., 2005. Research on the
effect of salidroside on neural stem cells. Heilongjiang Med.
Sci. 28, 34.