A novel anticancer agent, icaritin, induced cell growth inhibition, G1 arrest and mitochondrial transmembrane potential drop in human prostate carcinoma PC-3 cells

Article abstract of DOI:10.1016/j.ejphar.2007.02.039

Icariin and icaritin with prenyl group have been demonstrated for their selective estrogen receptor modulating activities. We screened their effects on cell growth in human prostate carcinoma PC-3 cell line (estrogen receptor positive) in vitro. PC-3 cell line was used for the measurement of anti-carcinoma activities of 0-100 μmol/l icaritin and 30 μmol/l icariin. 1 μmol/l 17-β estradiol (E₂) served as the estrogen positive control, and 1 μmol/l ICI 182,780 [7 α-[9 (4,4,5,5,5-pentafluoropentyl) sulfinyl] nonyl]-estra-1,3,5(10)-triene-3,17h-diol]] served as the specific estrogen receptor antagonist. Primary cultured rat prostate basal cells used as cell growth selective control. The growth-inhibitory effects were analyzed using MTT assay, and fluorochrome staining, flow cytometry, and immunoblotting were employed to illustrate the possible mechanisms. When treated with icaritin for 24 to 72 h, cell growth was strongly inhibited (at 48 h IC₅₀ was 10.74 ± 1.59 μmol/l, P < 0.001) companied with a mitochondrial transmembrane potential (_Ψm) drop. Meanwhile, few changes in IC₅₀ could be observed when co-incubated with ICI 182,780. Icaritin-induced growth inhibition was associated with G₁ arrest (P < 0.05), and G₂-M arrest depending upon doses. Consistently with G₁ arrest, icaritin increased protein expressions of pRb, p27^Kip1 and p16^Ink4a, while showed decrease in phosphorylated pRb, Cyclin D1 and CDK4. Comparatively, icariin has much lower effects on PC-3 cells and showed only weak G₁ arrest, suggesting a possible structure-activity relationship. These findings suggested a novel anticancer efficacy of icaritin mediated selectively via induction of cell cycle arrest but not associated with estrogen receptors in PC-3 cells.

Full text of DOI:10.1016/j.ejphar.2007.02.039

European Journal of Pharmacology 564 (2007) 26–36
www.elsevier.com/locate/ejphar
A novel anticancer agent, icaritin, induced cell growth inhibition,
G1 arrest and mitochondrial transmembrane potential drop
in human prostate carcinoma PC-3 cells☆
1
1
⁎
Xin Huang , Danyan Zhu , Yijia Lou
Institute of Pharmacology & Toxicology and Biochemical Pharmaceutics, College of Pharmaceutical Sciences,
Zhejiang University, 353, Yan'an Road, Hangzhou, 310031, China
Received 2 June 2006; received in revised form 26 January 2007; accepted 1 February 2007
Available online 27 February 2007
Abstract
Icariin and icaritin with prenyl group have been demonstrated for their selective estrogen receptor modulating activities. We screened their
effects on cell growth in human prostate carcinoma PC-3 cell line (estrogen receptor positive) in vitro. PC-3 cell line was used for the
measurement of anti-carcinoma activities of 0–100 μmol/l icaritin and 30 μmol/l icariin. 1 μmol/l 17-β estradiol (E2) served as the estrogen
positive control, and 1 μmol/l ICI 182,780 [7 α-[9 (4,4,5,5,5-pentafluoropentyl) sulfinyl] nonyl]-estra-1,3,5(10)-triene-3,17h-diol]] served as the
specific estrogen receptor antagonist. Primary cultured rat prostate basal cells used as cell growth selective control. The growth-inhibitory effects
were analyzed using MTT assay, and fluorochrome staining, flow cytometry, and immunoblotting were employed to illustrate the possible
mechanisms. When treated with icaritin for 24 to 72 h, cell growth was strongly inhibited (at 48 h IC50 was 10.74 1.59 μmol/l, Pb0.001)
companied with a mitochondrial transmembrane potential (_Ψm) drop. Meanwhile, few changes in IC50 could be observed when co-incubated
with ICI 182,780. Icaritin-induced growth inhibition was associated with G1 arrest (Pb0.05), and G2-M arrest depending upon doses. Consistently
with G1 arrest, icaritin increased protein expressions of pRb, p27Kip1 and p16Ink4a, while showed decrease in phosphorylated pRb, Cyclin D1 and
CDK4. Comparatively, icariin has much lower effects on PC-3 cells and showed only weak G1 arrest, suggesting a possible structure–activity
relationship. These findings suggested a novel anticancer efficacy of icaritin mediated selectively via induction of cell cycle arrest but not
associated with estrogen receptors in PC-3 cells.
© 2007 Published by Elsevier B.V.
Keywords: Icaritin; Icariin; PC-3; G1 arrest; Mitochondrial transmembrane potential (_Ψm); Apoptosis
1. Introduction
the increase is not comparable to North American and European
populations, the incidence ratio in many Asian centers is similar
to that of the high-risk countries (Sim and Cheng, 2005). At
present, there is no effective therapy available for the treatment
of androgen-independent stage of prostate cancer (advanced
prostate cancer), which usually arises after hormonal depriva-
tion/ablation therapy.
Naturally occurring icariin and icaritin present in Epimedium
brevicornum Maxim. (Berberidaceae), which has been tradi-
tionally used in Chinese folk medicine that has a wide range of
pharmacological and biological activities, including regulating
cardiovascular, circulatory, genital, and bone marrow systems,
stimulating neurite growth and possessing estrogenic activity
(Wang and Lou, 2004; Liu and Lou, 2004; Ye and Lou, 2005).
In vitro studies have demonstrated that icariin possesses
Prostate cancer is the most common cancer as well as the
second leading cause of cancer-related deaths in men in Western
countries (Greenlee et al., 2001). However, there has been a
recent trend in Asia towards increasing incidence of prostate
cancer, such as Japan and Singapore, reporting a more rapid
increase than high-risk countries. Whilst the absolute value of
☆
Grant sponsor: National Natural Science Foundation of China; Grant
number: No. 30070904, 30171121, 30472112, 30672564.
⁎
Corresponding author. Tel./fax: +86 571 87217206.
E-mail addresses: hx06@yahoo.com.cn (X. Huang), zdyzxb@zju.edu.cn
(D. Zhu), yijialou@zju.edu.cn (Y. Lou).
1
Tel./fax: +86 571 87217206.
0014-2999/$ - see front matter © 2007 Published by Elsevier B.V.
doi:10.1016/j.ejphar.2007.02.039

X. Huang et al. / European Journal of Pharmacology 564 (2007) 26–36
27
antitumor activity (Zhao et al., 1997), while icaritin exhibits its
2. Materials and methods
selective estrogen receptor modulating (SERM) effect mostly
via estrogen receptor-related mediating pathway, acting as a
phytoestrogen (Wang and Lou, 2004; Wang et al., 2006).
Limited in vitro studies indicated that several prenylated
flavonoids present in the hop plant (Humulus lupulus) possess
anticarcinogenic properties (Colgate et al., 2006). Here, a
prenylflavonoid named icaritin (C20H20O6; molecular weight
356.38) and its glycoside icariin (C33H40O15; molecular weight
676.65; which substituted Rha- and Glc-side chain for 3, 7-OH
of icariin]) (Fig. 1) were employed to find their anti-prostate
carcinoma activities.
The phytoestrogens (flavonoids coumestans and lignans)
may display estrogenic or anti-estrogenic activity by binding to
the various forms of the estrogen receptor or by competing for
the binding sites of estrogen biosynthesizing or metabolizing
enzymes (Santti et al., 1998). It has been suggested that estro-
gens and their receptors may be involved in the development
and progression of prostate cancer. Both estrogen α receptor
and estrogen β receptor could be detected in the development of
prostate cancer cell lines, such as LNCaP, DU145, and PC-3,
and it appears unlikely that alterations in the expression of either
estrogen receptor are commonly involved in the progression of
prostate cancer. There is a higher affinity of phytoestrogens
typically for estrogen β receptor than estrogen α receptor, and
estrogen β receptor is expressed in the PC-3 prostate epithelial
cell lines in vitro (Linja et al., 2003).
However, there is no report on the evaluation of anticancer
activity of the selective estrogen receptor modulators, icaritin
and icariin against prostate cancers. In the present investigation,
human prostate carcinoma PC-3 cells were employed to evaluate
the anticancer activity of icariin and icaritin, and for the first time
tested their efficacy against human prostate carcinoma PC-3
cells. We hypothesized there exists a structure–activity relation-
ship inside these compounds by showing their efficacy in
inhibiting growth and causing death of PC-3 cells. Furthermore,
we investigated the mechanistic rationale for the observed
efficacy of these compounds and hope those could be the most
likely targets for clinical therapy in prostate cancer cells.
2.1. Preparation for icariin and icaritin
Icariin was purchased from Zhejiang institute for the control
of pharmaceutical and biological products. Its hydrolysate,
icaritin, was prepared from icariin by treatment with cellulase
enzyme and purified using High Performance Liquid Chroma-
tography (HPLC) method. The structure of icaritin was
determined by mass spectrometry (esquire300plus_01073)
analysis. These compounds were dissolved in dimethyl
sulphoxide (DMSO) as stock solutions, and used directly in
the cell culture treatment.
2.2. Cell lines and cell growth-inhibition studies
The human prostate cancer cell line PC-3 was obtained from
Institute of Biochemistry and Cell Biology, SIBS, CAS and
cultured in RPMI 1640 medium (Gibco, Rockvile, MD, USA),
supplemented with 10% bovine serum albumin (BSA), 1%
penicillin/streptomycin in a 5% CO2 atmosphere at 37 °C. The
cells were seeded at a density of 5×103 cells/well in a 96 well
culture plate. After 24 h, the cells were treated with 5, 10, 30,
50, and 100 μmol/l icaritin and 30 μmol/l icariin dissolved in
DMSO (final concentration 0.1%) and control cells were treated
with 0.1% DMSO and 1 μmol/l estradiol. Cells treated with
icaritin, icariin or DMSO for 24, 48, and 72 h were incubated
with 20 μl of 5 mg/ml 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyl-tetrazolium bromile (MTT) solution for 4 h. Follow-
ing this, the medium was discarded, and 500 μl DMSO was
added. Absorbance at 570 nm was determined with an ELx800
universal microplate reader (Bio-Rad, USA). By MTT method,
cell numbers were obtained as absorbance values. The results
are expressed as viability compared with that of control cells.
Each treatment and time point had three independent plates. The
representative data shown in this study were reproducible in
three independent experiments.
2.3. The primary culture of rat prostate basal cells and cell
growth-inhibition studies
Minced rat ventral prostate was dissociated by collagenase II
digestion, suspended in RPMI-1640 containing 10% BSA, and
subjected to Percoll centrifugation to separate the epithelial cells
from stromal cells. The cells were cultured in RPMI 1640
medium, supplemented with 10% BSA, 1% penicillin/strepto-
mycin in a 5% CO2 atmosphere at 37 °C. The cells were seeded
at a density of 1×104 cells/well in a 96 well culture plate. After
24 h, the cells were treated with 5, 10, 30, and 50 μmol/l icaritin
dissolved in DMSO (final concentration 0.1%) and control cells
were treated with 0.1% DMSO, respectively. MTT method as
described above was used to assess cell viability.
2.4. Interaction with ICI 182,780
Some phytoestrogens, such as genistein and daidzein, act
both as agonists and antagonists at estrogen receptors. In the
Fig. 1. Chemical structures of icariin, icaritin, 17-β estradiol, and ICI 182,780.

28
X. Huang et al. / European Journal of Pharmacology 564 (2007) 26–36
present study, we examined the effects of icariin and icaritin as
well as their interaction with ICI 182,780, a pure estrogen
receptor antagonist. The concentration of icariin and icaritin
was set as detailed above, and the effect of cotreatment with
1 μmol/l ICI 182,780 on cell growth was assessed by the MTT
assay method after the desired treatment time as described above.
Rockford, USA) and exposed to film. Digital images of the films
were captured and quantified using the bio-imaging system (Bio-
Rad, USA). The expression levels of aimed proteins in treated
cultures were compared with those of untreated control cultures.
2.7. Mitochondrial transmembrane potential (_Ψm)
examination
2.5. Cell cycle analysis by flow cytometry
To assess the mitochondrial ΔΨ_Ψm, 5,5′,6,6′-tetrachloro-
1,1′,3,3′-tetraethylbenzimidazolocarbocyanine iodide (JC-1,
Sigma, USA) staining was used. JC-1 is a cationic dye that
exhibits potential-dependent accumulation in mitochondria, indi-
cated by a fluorescence emission shift from green (525 10 nm) to
red (610 10 nm). Mitochondria depolarization is specifically
indicated by a decrease in the red to green fluorescence intensity
ratio (Zuliani et al., 2003).
PC-3 cells were grown in 10% serum condition in the culture
medium as mentioned above. At ∼30% confluency, cells were
treated with DMSO control or 5, 10, 30, and 50 μmol/l icaritin
and 30 μmol/l icariin and at the end of desired treatment time
(24 and 48 h), cells were harvested and 1×106 cells were placed
into a polypropylene tube and centrifuged. The supernatant was
removed and 1 ml 4 °C 70% ethanol was added dropwise to the
cell pellet during vortexing. The cells were kept at 4 °C until
DNA staining. Fixed cells were treated with 100 μg/ml RNase
A in phosphate-buffered saline solution (PBS) for 1 h, followed
by staining with 50 μg/ml propidium iodide in PBS. Flow
cytometric analysis of cell cycle distribution and apoptosis was
performed with a BD FACSCalibur with a 488-nm (blue) argon
laser (Becton Dickinson, San Jose, CA). Data acquisition was
performed with CellQuest 3.1 software and data were analyzed
with ModFit LT 3.0 software (Variety Software House,
Topsham, ME).
2.8. Fluorescence morphological examination
Cell morphology of icaritin-induced apoptosis was investi-
gated by staining the cells with a combination of the fluorescent
DNA-binding dyes acridine orange (AO) and ethidium bromide
(EB). Cells were sorted into four groups with dual staining: (a)
viable, non-apoptotic (VNA), (b) viable, apoptotic (VA), (c)
non-viable, apoptotic (NVA), and (d) non-viable, non-apoptotic
(NVNA). Briefly, cells were harvested and washed with PBS
after being exposed to different concentrations of icaritin
and icariin or DMSO for 48 h and were stained with 100 μg/ml
AO/EB for 5 min. Then cells were observed under a fluores-
cence microscope (Leica) according to their color and structure
(Pitrak et al., 1996).
2.6. Protein extraction and western blot analysis
The prostate cancer cells were plated and cultured in com-
plete medium and allowed to attach for 24 h followed by the
addition of 5, 10, 30, and 50 μmol/l icaritin and 30 μmol/l icariin.
Incubation was carried out for 24 and 48 h. Control cells were
incubated in the medium with 0.1% DMSO for the same time
period. After the indicated period, cells were harvested at 4 °C
and lysed on ice in extraction buffer containing Tris–HCl
(20 mM, pH 7.5), NaCl (150 mM), EDTA (1 mM), Triton X-100
(1%), sodium deoxycholate (0.5%) plus PMSF (1 mM),
leupeptin (10 μg/ml), and aprotinin (30 μg/ml). Lysates were
cleared by centrifugation (4 °C, 14000 g, 30 min). Total protein
present in each lysate was quantified by using a modified
Lowry assay (DC protein assay; Bio-Rad, Hercules, USA).
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis
(SDS-PAGE), Western blotting, and densitometric measurement
were performed by standard protocols. Total protein content
(40 μg lysate per lane) was analyzed, and the proteins were
transferred to a nitrocellulose membrane. Transferred mem-
branes were blocked for 1 h in 5% nonfat milk in Tris-buffered
saline containing 0.1% Tween-20 (TBS/T). Primary antibodies
were added in 5% milk, and the blot was incubated overnight at
4 °C. The blots were washed three times for 5 min each with
10 ml TBS/T and incubated with secondary antibody (anti-
mouse HRP or anti-rabbit HRP, 1:3000; Santa Cruz Biotech-
nology, USA) in 10 ml TBS/T with gentle agitation for 1 h at
room temperature. Then the blots were washed three times for
5 min each with TBS/T, exposed to a chemiluminescent detection
system using the SuperSignal West Pico Substrate (Pierce,
2.9. Statistical analysis
All experiments were repeated thrice. Western blot results
were presented from a representative experiment. The number
of viable/dying cells or cell colonies in the control group or the
initial time point was assigned a relative value of 100%. All data
are expressed as means standard deviation (S.D.), and the level
of significance between two groups was assessed with Student's
t-test. P values of less than 0.05 were considered to be statisti-
cally significant.
3. Results
3.1. Effects of icariin and icaritin on growth of PC-3 and rat
prostate basal cells
To assess the biological activity of these compounds in terms
of cell growth and death, PC-3 cells were treated with 5, 10, 30,
50, and 100 μmol/l doses of icaritin for 24, 48, and 72 h. Icaritin
(30–100 μmol/l) showed a strong dose- and time-dependent
inhibition of cell growth, accounting for 66.10% to 86.27%
(Pb0.001), 72.13% to 89.90% (Pb0.001), and 73.09% to
88.67% (Pb0.001) growth inhibition after 24, 48, and 72 h of
treatment, respectively (Fig. 2A and B). However, similar
treatment with icariin (30 μmol/l for 24, 48, and 72 h) showed

X. Huang et al. / European Journal of Pharmacology 564 (2007) 26–36
29
weak efficacy, accounting for 5.95% (PN0.05), 7.73% (PN0.05),
and 2.48% (PN0.05) growth inhibition, respectively (Fig. 2A).
Meanwhile, primary cultured rat prostate basal cells treated with
icaritin (30–50 μmol/l) showed a comparatively weak dose- and
time-dependent inhibition of cell growth, accounting for 64.87%
to 59.48% (PN0.05), 55.79% to 49.98% (PN0.05), and 50.98%
to 50.82% (PN0.05) growth inhibition after 24, 48, and 72 h of
treatment, respectively. These data suggest that icaritin is
much potent in inhibiting the growth of PC-3 (at 48 h IC50 was
10.74 1.59 μmol/l, Pb0.001) as compared with icariin.
Meanwhile, icaritin induced a more differentiated, fibroblast-
like phenotype in PC-3 cells after longer incubation (48–72 h,
Fig. 2C). This effect might be related to the growth-inhibitory
effects of transforming growth factor-beta (TGF-beta), which
could generate the occurrence of a more fibroblast-like
phenotype. Since transforming growth factor-beta modulates
cell cycle progression in different cell types, further examinations
will detect how long would PC-3 cells survive in the presence of
30 or 50 μmol/l icaritin, if they would change their phenotype
further, and the possible mechanism. Also it would be examined
whether withdrawal of icaritin could allow the cells to regain their
previous phenotype or whether the changes are permanent. The
differences in biological activity of icaritin and icariin could be
attributed to the difference in their chemical structures (Fig. 1).
PC-3 cancer cell line was more sensitive to icaritin than rat
prostate basal cells (Fig. 2D and E).
Fig. 2. Inhibitory effects of icaritin and icariin on PC-3 cells growth. (A, B) Selective effects of icaritin and icariin on exponentially growing PC-3 cells, 5×103 cell/
well were plated in 96 well plates and the next day, cells were treated with either DMSO vehicle control or 5–100 μmol/l icaritin and 30 μmol/l icariin in complete
⁎⁎⁎
Pb0.001 versus control. (C) Cell morphology in DMSO,
medium. Results were averaged from three independent experiments and presented as means S.D.,
30 μmol/l ICA and 5–50 μmol/l ICT treated PC-3 cells (magnification 100×). (D, E) Effect of icaritin on exponentially growing primary cultured rat prostate basal
cells, 1×104 cell/well were plated in 96 well plates and the next day, cells were treated with either DMSO vehicle control or 5–50 μmol/l icaritin in complete medium
(a–e: treated with DMSO, and 5–50 μmol/l ICT, respectively; magnification 100×). Results were averaged from three independent experiments and presented as
⁎
⁎⁎
means S.D., Pb0.05, Pb0.01 versus viability of PC-3 cells in the same treatment group. After 24, 48, and 72 h exposure, total cells were treated with MTT for 4 h
and cell numbers were harvested as absorbance values. ICA, icariin; ICT, icaritin. PBC, prostate basal cells.

30
X. Huang et al. / European Journal of Pharmacology 564 (2007) 26–36
Co-incubation with 1 μmol/l ICI 182,780 caused negligible
alteration to treatment of 1 μmol/l estradiol, 30, 50 μmol/l icaritin
and 30 μmol/l icariin for 24, 48, and 72 h (Fig. 3). Thus, inhibition
of cell proliferation could not be associated with estrogen
receptors, but might be the result of the induction of apoptosis
or cell cycle growth arrest. We hypothesized that icaritin-induced
inhibition of cell proliferation was due to alterations in cell cycle
control and programmed cell death.
3.3. Icaritin induced a strong G1 arrest in cell cycle progression
of prostate cancer cells
Fig. 3. Effect of cotreatment with pure antiestrogen ICI 182,780 on cell growth
inhibition induced by icaritin and icariin in PC-3 cells. 5×103 cell/well were plated
in 96 well plates and the next day, cells were treated with either DMSO vehicle
control or 1 μmol/l estradiol, 30, 50 μmol/l icaritin and 30 μmol/l icariin for 24, 48,
and 72 h. Incubation with DMSO alone was performed as control and the final
concentration of vehicle, DMSO, in the medium never exceeded 0.1%. After
incubation for desired days, the MTTassay was performed to measure cell viability.
Results were averaged from three independent experiments and presented as
means S.D.; ICA, icariin; ICT, icaritin; E2, estradiol; ICI, ICI 182,780.
Inhibition of deregulated cell cycle progression in cancer
cells is an effective strategy to halt tumor growth (Noble et al.,
2005). Because we observed a strong growth-inhibitory effect
of icaritin, we then analyzed its possible inhibitory effect on cell
cycle progression following 5, 10, 30, and 50 μmol/l doses of
icaritin treatment for 48 h. We observed that icaritin (5, 10, and
30 μmol/l) induced G1 arrest after the treatment in PC-3 cells
(Fig. 4A). An optimum dose-dependent effect was observed
at 48 h of treatment where icaritin caused 59.35 6.80%,
63.42 6.18%, and 77.05 8.08% (Pb0.05) PC-3 cells in G1
phase as compared with 55.81% in control (Fig. 4B). An increase
in G1 cell population was mostly at the expense of S phase
cells (Pb0.01) with a minimal decrease in G2-M cell population
3.2. Icaritin-induced growth inhibition with no association with
estrogen receptors
Then, an experiment was designed to determine whether the
growth-inhibiting effect induced by icaritin and icariin could be
blocked by ICI 182,780, a specific estrogen receptor antagonist.
Fig. 4. Effect of icariin and icaritin on cell cycle progression in PC-3 cells. Cells were cultured in complete medium, and treated with either DMSO vehicle control or
30 μmol/l icariin and 5–50 μmol/l icaritin. After 48 h of these treatments, cells were collected, washed with PBS, digested with RNase, and then cellular DNA was
stained with propidium iodide as detailed in Materials and methods. Flow cytometric analysis was then performed for cell cycle distribution. (A) Propidium iodide
fluorescence pattern for cell cycle distribution in different treatment. (B) The percentage of cell cycle distribution for each treatment group. Mean S.D. of three
⁎
⁎⁎
independent samples. ICA, icariin; ICT, icaritin; Pb0.05; Pb0.01 versus control.

X. Huang et al. / European Journal of Pharmacology 564 (2007) 26–36
31
(Fig. 4B). 50 μmol/l icaritin induced a medium increase in G2-M
cell population (Fig. 4A and B). 48 h treatment of 30 μmol/l icariin
showed only weak G1 arrest 65.37 7.08% (PN0.05). We then
investigated whether G1 cell cycle regulatory molecules were
altered in icaritin treated PC-3 cells.
time-dependent increase in the expression of p27Kip1 and
p16Ink4a from 0 to 30 μmol/l icaritin (Fig. 5A), and no alteration
of these expression between 0 and 50 μmol/l icaritin (data not
shown). Reprobing of membranes for β-actin confirmed equal
protein loading (Fig. 5A and B, bottom). The induction both of
p27Kip1 and p16Ink4a protein expression was generally associ-
ated with the inhibition of cell growth, as they have been
demonstrated to be cyclin-dependent kinase inhibitors (Ricardo
et al., 1999; Sharpless and DePinho, 1999).
3.4. Icaritin-mediated alterations in expression of phosphory-
lated pRb, pRb, Cyclin D1 and CDK4 complexes, p27Kip1, and
p16Ink4a
Furthermore, we investigated the protein expression level of
a cyclin-dependent protein kinase, CDK4, which interacts with
D type cyclins and facilitates the progression of cells through G1
cell cycle phase (Marcos and Mariano, 2005). In the studies
analyzing its effect on the level of CDK4 and cyclin D1
associated with G1 phase, icaritin treatment for 24 and 48 h
showed a dose-dependent decrease in the expression of these
proteins (Fig. 5A and B) except an obvious increase of Cyclin
D1 at 50 μmol/l treatment (Fig. 5B).
Since, we observed optimum G1 cell cycle arrest by icaritin
at 48 h in PC-3 cells, and a relatively similar effect at 24 h (data
not shown); we investigated whether G1 cell cycle regulatory
molecules are altered in icaritin treated PC-3 cells. Total cell
lysates were prepared following icaritin treatment of cells at 5,
10, 30 doses and icariin treatment of cells at 30 μmol/l dose for
24 and 48 h, and 50 μmol/l icaritin at 48 h. Western blot analysis
of total cell lysates (40 μg/sample) showed a strong dose- and
Fig. 5. Effect of icaritin and icariin on G1 cell cycle regulators in PC-3 cells. (A) Cells were cultured in complete medium, and treated with either with DMSO vehicle
control or 30 μmol/l icariin and 5, 10, 30 μmol/l icaritin as described in Materials and methods. (B) Cells were cultured in complete medium, and treated with 50 μmol/
l icaritin for 48 h. At the end of treatments, total cell lysates were prepared and subjected to SDS-PAGE followed by Western immunoblotting. Membranes were probed
with anti-phosphorylated pRb (p-pRb), pRb, Cyclin D1, CDK4, p27Kip1, p16Ink4a and β-actin antibodies followed by peroxidase-conjugated appropriate secondary
antibodies, and visualized by enhanced chemiluminescence detection system. Experiments were repeated three times, and share the similar results, a representative blot
is shown for each protein. ICA, icariin; ICT, icaritin.

32
X. Huang et al. / European Journal of Pharmacology 564 (2007) 26–36
Fig. 6. Effect of icaritin and icariin on mitochondrial transmembrane potential, sun-G1 cell subpopulation and morphology in PC-3 cells. (A) After cultured in complete
medium, and treated with either DMSO vehicle control or 30 μmol/l icariin and 5–50 μmol/l icaritin as described in Materials and methods, PC-3 cells were incubated
with JC-1 (0.3 μg/ml) for 15 min at 37 °C. Pictures were taken under a fluorescent microscope (magnification 100×). (B) Propidium iodide fluorescence pattern for
sun-G1 cell subpopulation in different treatment. The experiments were repeated thrice with similar results and a representative illustration is shown for each treatment.
Apo, apoptosis. (C) Identification of apoptotic cells using acridine orange-ethidium bromide staining in different treatment (magnification 100×).

X. Huang et al. / European Journal of Pharmacology 564 (2007) 26–36
33
Collectively, these results indicated a novel mechanism of
icaritin-induced G1 cell cycle arrest in PC-3 cells. We further
sought to investigate whether additional molecular mechanism
(s) exists by which icaritin inhibits G1 cell cycle arrest and
studied the Retinoblastoma protein (pRb), which is a central
molecule in G1 cell cycle control, and accordingly, its product
(p-pRb). During the progression of G1 phase and entering into S
phase, the pRb is hyperphosphorylated by early and late G1
phase specific cyclin-dependent kinases and releases E2F
transcription factors which, in turn, activates S phase specific
genes and help progress cells enter into S phase (Liang, 2005).
Icaritin treatment showed a dose-dependent increase of pRb, and
that this effect was more pronounced at 10–30 μmol/l concen-
tration within 48 h of treatment (Fig. 5A) while 50 μmol/l icaritin
showed a median increase of pRb (Fig. 5B) and 30 μmol/l icariin
exhibited a few decrease (Fig. 5A). On the other hand, a dose-
dependent decrease of phosphorylated pRb has exhibited from
0–50 μmol/l icaritin, while 30 μmol/l icariin showed a weak
decrease (Fig. 5A, B).
typical for necrosis, though there is a significant decrease in cell
volume (Fig. 6C). AO-EB staining demonstrated that few
morphological alterations could be found even after 48 h icaritin
treatment. All of these results indicated that icaritin-induced
early stage dose-dependent depolarization of the mitochondrial
transmembrane followed markedly with neither apoptosis nor
necrosis after 48 h icaritin treatment (later stage).
4. Discussion
The central and novel finding in the present study is the
identification of in vitro anticancer efficacy of icaritin against
advanced human prostate carcinoma PC-3 cells. The completed
studies clearly and convincingly showed that icaritin caused a
G1 cell cycle arrest via an induction of p27Kip1 and to a lesser
extent p16Ink4a together with an inhibition in CDK4-cyclin D1
complex as an underlying mechanism in its PC-3 cell growth
inhibition. Furthermore, this agent-caused mitochondrial trans-
membrane potential (_Ψm) drop in PC-3 cells.
ICI 182,780 is a specific estrogen receptor antagonist. Here,
we observed that icaritin-induced inhibition of PC-3 growth
altered little when co-incubated with ICI 182,780, suggesting
that estrogenic activity of icaritin had no role in affecting a
growth-inhibitory response.
3.5. Icaritin-induced drop in mitochondrial transmembrane
potential at early stage treatment on PC-3 cells
In addition to cell cycle arrest, the growth inhibition induced
by icaritin treatment could also be due to programmed cell death;
hence, we investigated whether icaritin-induces apoptosis.
Since the loss of mitochondrial transmembrane potential
(_Ψm) is a hallmark for apoptosis, we investigated this potential
level in PC-3 cells using JC-1 dye. In non-apoptotic cells, JC-1
accumulates as aggregates in the mitochondria which stain red.
Whereas, in apoptotic and necrotic cells, JC-1 exists in
monomeric form and stains the cytosol green. The red/green
ratio is an indicator of _Ψm, which is independent of dye-
loading, lightscattering, and other optical factors (Szilagyi et al.,
2006). After 24 h treatment, 5–50 μmol/l icaritin lowered
mitochondrial transmembrane potential presented as green
fluorescence in a dose-dependent manner, while 0 μmol/l icaritin
and 30 μmol/l icariin treatment caused no alteration and emitted
strong red fluorescence mostly in cytosol (Fig. 6A), indicating a
dose-dependent depolarization of the mitochondrial transmem-
brane, which might be associated with cell apoptosis or
necrosis, induced by 5–50 μmol/l icaritin within 24 h treatment
(Huang et al., 2004; Bedner et al., 1999).
Then flow cytometric analysis was employed to verify the
occurrence of apoptosis. Flow cytometric analysis of cells with
sub-G1 DNA content following staining with propidium iodide
is a widely accepted technique for apoptosis detection.
However, exposure of PC-3 cells to icaritin and icariin for 48
h showed only very slight increase of sub-G1 cell subpopulation
and did not affect it significantly (Fig. 6B). This flow cytometry
analysis demonstrated that there were few cells that went
apoptosis in each treatment group.
In previous studies, prenylflavonoids have been shown to
cause alterations in cell cycle regulation in a number of cell lines
(Stevens and Page, 2004). We performed flow cytometry
analysis to confirm the alterations in cell cycle regulatory
properties of these prenylflavonoids, icaritin and icariin. Results
indicate that icaritin (5–30 μmol/l) caused a G1 arrest in PC-3
cells. The percentage of cells in G1 phase increased with
corresponding decrease in the percentage of cells in the S phase
cells. We hypothesized this cell cycle arrest might attribute to G1
cell cycle regulatory molecules and their expressive alterations.
Cyclin-dependent kinases, cyclin-dependent kinase inhibi-
tors, and cyclins play essential roles in the regulation of cell
cycle progression. Cyclin-dependent kinase inhibitors are tumor
suppressor proteins that down-regulate the cell cycle progres-
sion by binding with active cyclin-dependent kinase-cyclin
complexes and thereby inhibiting their kinase activities
(Yim et al., 2005). A common characteristic of tumor suppres-
sor genes is their ability to inhibit cell proliferation when over-
expressed in sensitive cell lines. The inhibition of cell growth
observed in icaritin-treated cells may be due to induction of G1
cell cycle arrest as a result of multiple gene activities. We found
that two important cyclin-dependent kinase inhibitors, p27Kip1
and p16Ink4a, were up-regulated in PC-3 cells treated with 5–
30 μmol/l icaritin. The p27Kip1 is a member of cyclin-dependent
kinase inhibitors, which could bind and inhibit a broader range
of cyclin-dependent kinases. The expression of p27Kip1 has
been shown to be up-regulated during TGF-β mediated
antiproliferative response, serum starvation or density arrested
cells (Polyak et al., 1994; Slingerland et al., 1994). Recent
studies have also indicated the prognostic significance of
cyclin-dependent kinase inhibitors in prostate cancer. Specifi-
cally, p27Kip1 expression has been shown to be an independent
predictor of prostate-specific antigen failure following radical
On the other hand, we used AO-EB dyes to investigate cell
morphology. After 48 h treatment, a number of cells exhibited a
flattened polygonal morphology, whereas few cells showed
morphological features of apoptosis (membrane broken, and
breaking up of the nuclei) or revealed any orange stain that is

34
X. Huang et al. / European Journal of Pharmacology 564 (2007) 26–36
Previous studies have demonstrated that, pl6Ink4a could
negatively regulate cell proliferation by suppressing hyperpho-
sphorylation and functional inactivation of pRb. In contrast,
cells that lack functional pRb appear not to be affected by high
levels of pl6Ink4a, demonstrating that pRb mediates the growth
suppression by pl6Ink4a. Since pl6Ink4a has been shown to
specifically inhibit cyclin D1/CDK4 complexes, these findings
indicate that pRb is the critical target of these complexes in
order to promote progression through the G1 phase of the cell
cycle. The evidence presented here connects two tumor
suppressors, p16Ink4a and pRb, in one pathway controlling
cell growth. In this pathway cyclin D/CDK complexes would
mediate the inactivation of pRb, either directly or indirectly
(Rene' et al., 1995). Several lines of evidence suggest that
inactivation of this growth-regulatory pathway might be an
absolute requirement for the progression of certain tumors.
Thus, results came from our experiment showed that icaritin
Fig. 7. Inducible effects of icaritin. Icaritin on cell death followed by pRb,
p27Kip1 and p16Ink4a protein up-expressions and phosphorylated pRb, Cyclin D1
and CDK4 protein down-expressions in PC-3 cells. Induction of the cell cycle
regulatory pathways contributed to the inactivation of E2F, which plays an
important role in entering from G1 to S. Meanwhile, icaritin also caused drop of
mitochondrial transmembrane potential, which often exists in early strategy of
apoptosis. Both mechanisms could due to the cell growth inhibition of PC-3 cells.
(5–30 μmol/l) could both up-regulate the expression of p27Kip1
,
p16Ink4a together with pRb and down-regulate the expression of
phosphorylated pRb, cyclin Dl and CDK4. In other words,
icaritin could inhibit PC-3 cell growth by regulating every stage
through the cyclin D1-CDK4-p16Ink4a pathway (Ortega et al.,
2002; Semczuk and Jakowicki, 2004).
prostatectomy, and its low expression is correlated with poor
disease-free survival in prostate cancer patient (Yang et al.,
1998; Freedland et al., 2003). Therefore, the up-regulation of
p27Kip1 may be an important molecular mechanism through
which icaritin inhibits cancer cell growth.
Furthermore, we also observed that 50 μmol/l icaritin
reduces G2-M arrest at the expense of G1 phase cell population.
This suggests that, at this dose, the mechanism of action of
icaritin differs from the action of lower doses. Treatment with
this concentration of icaritin unexpectedly increased expression
of Cyclin D1 but caused similar changes in pRB, phosphory-
lated pRB, and CDK4 expression as 30 μmol/l icaritin.
Most of the presently available cytotoxic anticancer drugs
mediate their effect via apoptosis induction in cancer cells, and
apoptosis is suggested as one of the major mechanisms for the
targeted therapy of various cancers including prostate cancer.
In case of advanced prostate cancer, cancer cells become
resistant to apoptosis and do not respond to cytotoxic
chemotherapeutic agents (Yim et al., 2005). Therefore, the
agents that induce apoptotic death of hormone-refractory
prostate cancer cells could be useful in controlling this
malignancy. Consistent with this approach, our data showing
an induction of apoptotic death of advanced prostate cancer
cells by icaritin could be of a significance in identifying another
anticancer mechanism (together with a greater significance of
cell cycle arrest) of icaritin for its possible application in
prostate cancer control. Because induction of cyclin-dependent
kinase inhibitors has been reported in anticancer agent-caused
apoptosis in human prostate cancer cells (Don et al., 2001),
the icaritin-induced cyclin-dependent kinase inhibitors could
be, in part, responsible for the observed apoptotic death of
PC-3 cells.
The increased expression of G1 cyclin, Cyclin D1, in cancer
cells provided them an uncontrolled growth advantage because
most of these cells either lack cyclin-dependent kinase
inhibitors, harbor nonfunctional cyclin-dependent kinase inhi-
bitors, or cyclin-dependent kinase inhibitors expression is not at
a sufficient level to control cyclin-dependent kinase activity
(Dulic et al., 1992). Consistent with these reports, cell cycle
analysis data showed that icaritin (5–30 μmol/l) caused a dose-
dependent G1 arrest in cell cycle progression of prostate cancer
cells, and the growth-inhibitory effect of icaritin could also be
directly related to its ability to down-regulate CDK4, which is
one of the critical molecules required for early progression of
the G1 cell cycle.
Binding of pl6Ink4a to CDK4 prevents association of CDK4
with the D-type cyclins and results in an inhibition of the
catalytic activity of the cyclin D/CDK4 enzymes. Thus,
overexpression of pl6Ink4a in PC-3 cells might due to an arrest
in the G1 phase of the cell cycle.
In vitro, complexes of cyclin Dl and CDK4 can phosphor-
ylate the product of the retinoblastoma tumor suppressor gene,
pRb. The phosphorylation of pRb, occurring in mid/late G1,
reverses its growth-inhibitory effect and enables cells to proceed
from G1 to S phase. The timing of cyclin D-dependent kinase
activity and the onset of pRb hyperphosphorylation appear to
coincide in the cell cycle. Taken together, these findings suggest
that cyclin D/CDK complexes play a critical role in pRb
hyperphosphorylation in vivo. Consistent with these, our data
showed that icaritin (5–30 μmol/l) increased the expression of
pRb and decreased phosphorylated pRb in a dose- and time-
dependent manner (Fig. 7).
JC-1 is a non-toxic fluorescence probe to monitor membrane
potential. This dye has no effect on living cells, including their
respiration. JC-1 monomers accumulate selectively in mito-
chondria and subsequently aggregate as a consequence of the
membrane potential. This reaction is pH independent within the
physiological range. The electrochemical gradient is responsi-
ble for this J aggregation. The change in binding of JC-1 is

X. Huang et al. / European Journal of Pharmacology 564 (2007) 26–36
35
manifested by a loss of red fluorescence, which represents the
Acknowledgements
aggregate binding of this dye in mitochondria. On the other
hand, the mitochondrial transmembrane potential decreases
early during apoptosis (Huang et al., 2004; Bedner et al., 1999).
Thus, we demonstrated that early stage of icaritin treatment (5–
50 μmol/l for 24 h) induced dose-dependent depolarization of
the mitochondrial transmembrane might be associated with cell
apoptosis or necrosis, while 30 μmol/l icariin treatment has no
activity of inducing apoptosis.
This work was funded by the National Natural Science
Foundation of China and we sincerely thank the Foundation for
its support. We also sincerely thank Mr. Zhiqiang Wang for his
work of purification of icaritin.
References
However, after 48 h treatment, these apoptotic or necrotic
signals diminished, for the results of flow cytometry analysis
showed that, the percentage of apoptosis cells of icaritin
(5–50 μmol/l) treatment group is similar to that of DMSO
treatment group. Flow cytometry techniques that are widely
used in studies of cell death, and particularly in the iden-
tification of apoptotic cells, generally rely on the measurement
of a single characteristic biochemical or molecular attribute
(Bedner et al., 1999). However, these methods fail to recog-
nize cell death lacking that attribute, as in some examples of
atypical apoptosis. Since apoptosis was originally defined by
morphologic criteria, and mitochondrial metabolism is closely
linked to mechanisms regulating apoptosis, we let the
cytometry defined apoptosis be confirmed by morphologic
examination.
Thus, we used AO-EB staining, which allows the identifi-
cation of viable, apoptotic and necrotic cells based on color and
appearance (Gao et al., 2005). Viable cells were green with
intact nuclei. Non-viable cells had bright orange chromatin.
Apoptosis was demonstrated by the appearance of cell
shrinkage with condensation and breaking up of the nuclei.
Thus, apoptotic cells were easily distinguished from necrotic
cells because the latter appeared orange with a normal nuclear
structure. Herein, little morphological changes could be
detected in icaritin treated cells by using AO-EB staining.
Cells undergo death by two major mechanisms: necrosis, in
which primary damage to the metabolic or membrane integrity
of the cell occurs; and apoptosis, which is an internal suicide
programme contained in all cells. Taken together, we concluded
that, icaritin principally caused cell death by cell cycle arrest in
the entire process, and induced depolarization of the mitochon-
drial transmembrane, which might be associated with cell
apoptosis or necrosis, in early treatment stage. However, a
detailed molecular mechanism is far from clear and requires
further exploration.
Bedner, E., Li, X., Gorczyca, W., Melamed, M.R., Darzynkiewicz, Z., 1999.
Analysis of apoptosis by laser scanning cytometry. Cytometry 35, 181–195.
Colgate, E.C., Miranda, C.L., Stevens, J.F., Bray, T.M., Ho, E., 2006.
Xanthohumol, a prenylflavonoid derived from hops induces apoptosis and
inhibits NF-kappaB activation in prostate epithelial cells. Cancer Lett. 246,
201–209.
Don, M.J., Chang, Y.H., Chen, K.K., Ho, L.K., Chau, Y.P., 2001. Induction of
CDK inhibitors (p21(WAF1) and p27(Kip1)) and Bak in the beta-lapachone-
induced apoptosis of human prostate cancer cells. Mol. Pharmacol. 59,
784–794.
Dulic, V., Lees, E., Reed, S.I., 1992. Association of human cyclin E with a
periodic G1-S phase protein kinase. Science 257, 1958–1961.
Freedland, S.J., deGregorio, F., Sacoolidge, J.C., Elshimali, Y.I., Csathy, G.S.,
Dorey, F., Reiter, R.E., Aronson, W.J., 2003. Preoperative p27 status is an
independent predictor of prostate specific antigen failure following radical
prostatectomy. J. Urol. 169, 1325–1330.
Gao, L., Shan, B.E., Chen, J., Liu, J.H., Song, D.X., Zhu, B.C., 2005. Effects of
spider Macrothele raven venom on cell proliferation and cytotoxicity in
HeLa cells. Acta Pharmacol. Sin. 26, 369–376.
Greenlee, R.T., Hill-Harmon, M.B., Murray, T., Thun, M., 2001. Cancer
statistics, 2001. CA Cancer J. Clin. 51, 15–36.
Huang, Y.T., Chueh, S.C., Teng, C.M., Guh, J.H., 2004. Investigation of
ouabain-induced anticancer effect in human androgen-independent prostate
cancer PC-3 cells. Biochem. Pharmacol. 67, 727–733.
Liang, Z., 2005. Tumour suppressor retinoblastoma protein Rb: a transcriptional
regulator. Eur. J. Cancer 41, 2415–2427.
Linja, M.J., Savinainen, K.J., Tammela, T.L., Isola, J.J., Visakorpi, T., 2003.
Expression of ERα and ERβ in prostate cancer. Prostate 55, 180–186.
Liu, J., Lou, Y.J., 2004. Determination of icariin and metabolites in rat serum by
capillary zone electrophoresis: rat pharmacokinetic studies after adminis-
tration of icariin. J. Pharm. Biomed. Anal. 36, 365–370.
Marcos, M., Mariano, B., 2005. Mammalian cyclin-dependent kinases. Trends
Biochem. Sci. 30, 630–641.
Noble, M., Barrett, P., Endicott, J., Johnson, L., McDonnell, J., Robertson, G.,
Zawaira, A., 2005. Exploiting structural principles to design cyclin-
dependent kinase inhibitors. Biochim. Biophys. Acta 1754, 58–64.
Ortega, S., Malumbres, M., Barbacid, M., 2002. Cyclin D-dependent kinases,
INK4 inhibitors and cancer. Biochim. Biophys. Acta 1602, 73–87.
Pitrak, D.L., Tsai, H.C., Mullane, K.M., Sutton, S.H., Stevens, P., 1996.
Accelerated neutrophil apoptosis in the acquired immunodeficiency
syndrome. J. Clin. Invest. 98, 2714–2719.
Polyak, K., Kato, J.Y., Solomon, M.J., Sherr, C.J., Massague, J., Roberts, J.M.,
Koff, A., 1994. p27Kip1, a cyclin-Cdk inhibitor, links transforming
growth factor-beta and contact inhibition to cell cycle arrest. Genes Dev. 8,
9–22.
Rene', H.M., Rafael, E.H., Felix, L., Robert, A.W., 1995. Growth suppression by
p16ink4a requires functional retinoblastoma protein. Cell Biol. 92, 6289–6293.
Ricardo, V.L., Lori, A.E., Long, J., Elzbieta, K., Xiang, Q., John, C.C., 1999.
p27kip1: a multifunctional cyclin-dependent kinase inhibitor with prognostic
significance in human cancers. Am. J. Pathol. 154, 313–323.
Santti, R., Makela, S., Strauss, L., Korkman, J., Kostian, M.L., 1998.
Phytoestrogens: potential endocrine disruptors in males. Toxicol. Ind.
Health 14, 223–237.
In conclusion, the results of our studies provide experimental
evidence that icaritin inhibited cell growth of PC-3 prostate
cancer cells, arrested cell cycle with alterations in phosphory-
lated pRb, pRb, cyclin D1, CDK4, p27Kip1, and p16Ink4a protein
expression but not associated with estrogen receptors, and
decreased mitochondrial transmembrane potential in early stage
treatments. The nature of icaritin in selectively mediating the
above mentioned responses in PC-3 cells, in conjunction with
its nontoxic nature, could make it a potentially effective
chemopreventive or therapeutic agent against prostate cancer.
However, further in vivo studies are needed to establish the role
of icaritin as a chemopreventive or therapeutic agent against
prostate cancer.
Semczuk, A., Jakowicki, J.A., 2004. Alterations of pRb1-cyclin D1-cdk4/6-p16
(INK4A) pathway in endometrial carcinogenesis. Cancer Lett. 203, 1–12.
Sharpless, N.E., DePinho, R.A., 1999. The INK4A/ARF locus and its two gene
products. Curr. Opin. Genet. Dev. 9, 22–30.

36
X. Huang et al. / European Journal of Pharmacology 564 (2007) 26–36
Sim, H.G., Cheng, C.W., 2005. Changing demography of prostate cancer in
Asia. Eur. J. Cancer 41, 834–845.
Yim, D., Singh, R.P., Agarwal, C., Lee, S., Chi, H., Agarwal, R., 2005. A novel
anticancer agent, decursin, induces G1 arrest and apoptosis in human
prostate carcinoma cells. Cancer Res. 65, 1035–1044.
Zhao, Y., Cui, Z., Zhang, L., 1997. Effects of icariin on the differentiation of HL-
60 cells. Zhonghua Zhong Liu Za Zhi 19, 53–55.
Zuliani, T., Duval, R., Jayat, C., Schnébert, S., André, P., Dumas, M., Ratinaud,
M., 2003. Sensitive and reliable JC-1 and TOTO-3 double staining to assess
mitochondrial transmembrane potential and plasma membrane integrity:
interest for cell death investigations. Cytometry 54A, 100–108.
Slingerland, J.M., Hengst, L., Pan, C.H., Alexander, D., Stampfer, M.R., Reed,
S.I., 1994. A novel inhibitor of cyclin-Cdk activity detected in transforming
growth factor beta-arrested epithelial cells. Mol. Cell. Biol. 14, 3683–3694.
Stevens, J.F., Page, J.E., 2004. Xanthohumol and related prenylflavonoids from
hops and beer: to your good health! Phytochemistry 65, 1317–1330.
Szilagyi, G., Simon, L., Koska, P., Telek, G., Nagy, Z., 2006. Visualization of
mitochondrial membrane potential and reactive oxygen species via double
staining. Neurosci. Lett. 399, 206–209.
Wang, Z.Q., Lou, Y.J., 2004. Proliferation-stimulating effects of icaritin and
desmethylicaritin in MCF-7 cells. Eur. J. Pharmacol. 504, 147–153.
Wang, Z.Q., Nadine, W., Lou, Y.J., Peter, P., 2006. Prenylflavonoids as
nonsteroidal phytoestrogens and related structure–activity relationships.
ChemMedChem 1, 482–488.
Yang, R.M., Naitoh, J., Murphy, M., Wang, H.J., Phillipson, J., deKernion, J.B.,
Loda, M., Reiter, R.E., 1998. Low p27 expression predicts poor disease-free
survival in patients with prostate cancer. J. Urol. 159, 941–945.
Ye, H.Y., Lou, Y.J., 2005. Estrogenic effects of two derivatives of icariin on
human breast cancer MCF-7 cells. Phytomedicine 12, 735–741.
Further reading
Lloyd, R.V., Erickson, L.A., Jin, L., Kulig, E., Qian, X., Cheville, J.C.,
Scheithauer, B.W., 1999. p27kip1: a multifunctional cyclin-dependent kinase
inhibitor with prognostic significance in human cancers. Am. J. Pathol. 154,
313–323.