Over-expression of BMPR-IB reduces the malignancy of glioblastoma cells by upregulation of p21 and p27Kip1.

Abstract

In our previous study, we detected decreased expression of phospho-Smad1/5/8 and its upstream signaling molecule, bone morphogenetic protein receptor IB subunit (BMPR-IB), in certain glioblastoma tissues, unlike normal brain tissues. In order to clarify the functional roles and mechanism of BMPR-IB in the development of glioblastoma, we studied the effects of BMPR-IB overexpression on glioblastoma cell lines in vitro and in vivo.
We selected glioblastoma cell lines U251, U87, SF763, which have different expression of BMPR-IB to be the research subjects. Colony formation analysis and FACS were used to detect the effects of BMPR-IB on the growth and proliferation of glioblastoma cells in vivo. Immunofluresence was used to detect the differentiation changes after BMPR-IB overexpression or knocking-down. Then we used subcutaneous and intracranial tumor models to study the effect of BMPR-IB on the growth and differentiation of glioblastoma cells in vivo. The genetic alterations involved in this process were examined by real-time PCR and western blot analysis.ed.
Forced BMPR-IB expression in malignant human glioma cells, which exhibit lower expression of BMPR-IB, induced the phosphorylation and nuclear localization of smad1/5/8 and arrested the cell cycle in G1. Additionally, BMPR-IB overexpression could suppress anchorage-independent growth and promote differentiation of theses glioblastoma cells. Furthermore, overexpression of BMPR-IB inhibited the growth of subcutaneous and intracranial tumor xenografts and prolonged the survival of mice injected intracranially with BMPR-IB-overexpressing glioblastoma cells. Conversely, inhibition of BMPR-IB caused SF763 malignant glioma cells, a line known to exhibit high BMPR-IB expression that does not form tumors when used for xenografts, to show increased growth and regain tumorigenicity in a nude mouse model system, ultimately shortening the survival of these mice. We also observed significant accumulation of p21 and p27kip1 proteins in response to BMPR-IB overexpression. Our study suggests that overexpression of BMPR-IB may arrest and induce the differentiation of glioblastoma cells due to upregulation of p21 and p27kip1 in vitro and that in vivo and decreased expression of BMPR-IB in human glioblastoma cells contributes to glioma tumorigenicity. BMPR-IB could represent a new potential therapeutic target for malignant human gliomas.

Full text

Page 1

RESEARCHOpen Access
Over-expression of BMPR-IB reduces the
malignancy of glioblastoma cells by upregulation
of p21 and p27Kip1
Shuang Liu1,2, Feng Yin1, Wenhong Fan2, Shuwei Wang1, Xin-ru Guo1, Jian-ning Zhang1*, Zeng-min Tian1*and
Ming Fan2,3*
Abstract
Background: In our previous study, we detected decreased expression of phospho-Smad1/5/8 and its upstream
signaling molecule, bone morphogenetic protein receptor IB subunit (BMPR-IB), in certain glioblastoma tissues,
unlike normal brain tissues. In order to clarify the functional roles and mechanism of BMPR-IB in the development
of glioblastoma, we studied the effects of BMPR-IB overexpression on glioblastoma cell lines in vitro and in vivo.
Methods: We selected glioblastoma cell lines U251, U87, SF763, which have different expression of BMPR-IB to be
the research subjects. Colony formation analysis and FACS were used to detect the effects of BMPR-IB on the
growth and proliferation of glioblastoma cells in vivo. Immunofluresence was used to detect the differentiation
changes after BMPR-IB overexpression or knocking-down. Then we used subcutaneous and intracranial tumor
models to study the effect of BMPR-IB on the growth and differentiation of glioblastoma cells in vivo. The genetic
alterations involved in this process were examined by real-time PCR and western blot analysis.ed.
Results and conclusion: Forced BMPR-IB expression in malignant human glioma cells, which exhibit lower
expression of BMPR-IB, induced the phosphorylation and nuclear localization of smad1/5/8 and arrested the cell
cycle in G1. Additionally, BMPR-IB overexpression could suppress anchorage-independent growth and promote
differentiation of theses glioblastoma cells. Furthermore, overexpression of BMPR-IB inhibited the growth of
subcutaneous and intracranial tumor xenografts and prolonged the survival of mice injected intracranially with
BMPR-IB-overexpressing glioblastoma cells. Conversely, inhibition of BMPR-IB caused SF763 malignant glioma cells, a
line known to exhibit high BMPR-IB expression that does not form tumors when used for xenografts, to show
increased growth and regain tumorigenicity in a nude mouse model system, ultimately shortening the survival of
these mice. We also observed significant accumulation of p21 and p27kip1 proteins in response to BMPR-IB
overexpression. Our study suggests that overexpression of BMPR-IB may arrest and induce the differentiation of
glioblastoma cells due to upregulation of p21 and p27kip1 in vitro and that in vivo and decreased expression of
BMPR-IB in human glioblastoma cells contributes to glioma tumorigenicity. BMPR-IB could represent a new
potential therapeutic target for malignant human gliomas.
Keywords: 1. BMPR-IB, 2. Glioblastoma, 3. Growth inhibition, 4. Differentiation
* Correspondence: jnzhang2005@yahoo.com.cn; tianzengmin@sina.com;
fanmingchina@126.com
1Department of Neurosurgery, Navy General Hospital, 100048 Beijing, China
2Department of Brain Protection & Plasticity Research, Beijing Institute of
Basic Medical Sciences, Taiping Road 27, Beijing 100850, People’s Republic of
China
Full list of author information is available at the end of the article
© 2012 Liu et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.
Liu et al. Journal of Experimental & Clinical Cancer Research 2012, 31:52
http://www.jeccr.com/content/31/1/52

Page 2

Background
Malignant gliomas are the most common primary tumors
in the brain; they are destructive, invasive, and the most
highly vascularized lethal tumors observed in humans.
Gliomas are classified into grades I – IVaccording to their
histological degree of malignancy by the WHO criterion.
Despite recent progress in combination therapies, the me-
dian survival of patients with glioblastoma (WHO grade
IV) is less than 14–15 months [1]. Advances in the treat-
ment of malignant gliomas will require improved under-
standing of the biology and molecular mechanisms of
glioma development and progression. Many studies show
that the malignant transformation of glioma is a conse-
quence of the stepwise accumulation of genetic alterations
that lead to aberrant regulation of proliferation and differ-
entiation signals and disruption of the apoptotic pathway
[1]. Recent research on the molecular basis of gliomas and
the implications for targeted therapeutics has focused on
the PTEN, EGFR and VEGF signaling pathways [2-4].
These pathways, while important, represent only a subset
of the multiple molecular mediators of glioma tumorigen-
esis and malignancy [1].
A previous study by our group showed that the ex-
pression of bone morphogenetic protein receptor IB
subunit (BMPR-IB) is decreased in most malignant
human glioma tissues, including anaplastic astrocytomas
and glioblastomas. Furthermore, the low expression of
BMPR-IB was found to contribute to a lower ratio of
phospho-Smad1/5/8 to Smad1/5/8 expression, which
correlates significantly with poor patient survival [5].
Thus, it would not be unreasonable to speculate that
BMP signals may participate in the development and
progression of gliomas.
BMPs are the subclass of the transforming growth fac-
tor-β (TGF-β) superfamily, including more than 20 mem-
bers. BMP ligands and receptor subunits are present
throughout neural development and mediate a diverse
array of developmental processes, including cellular sur-
vival, proliferation, morphogenesis, lineage commitment,
differentiation and apoptosis of neural stem cells in the
CNS [6-8]. Additionally, during regional and cellular
maturation, BMPs can mediate long-range signaling by
acting as gradient morphogens, or they can mediate
short-range signaling by modulating cell-cell communi-
cation [6,7,9]. BMP signals transduce intracellular signals
through type I (BMP-RIA and BMP-RIB) and type II
(BMP-RII) serine/threonine kinase receptors. Binding of
BMPs to BMPR-II results in phosphorylation of BMPR-I
and downstream Smad proteins. BMPs activate Smad1/
5/8, which can associate with Smad4 in a heterodimeric
complex upon phosphorylation that is translocated to
the nucleus, where it activates transcription [10-13].
Although the BMP pathways have emerged as import-
ant contributors to many human neoplastic conditions
[14,15], the role of BMPs/BMPRs in human glioma has
not been completely defined. In the present study, we
continued to investigate how BMPR-IB regulates the
growth of glioblastomas.
Methods
Cell lines and cell culture
The human malignant glioma cell lines SF126, SF763,
and M17 were obtained from the American Type Cul-
ture Collection. The glioblastoma cell line U-251 and
normal human astrocytes, which were described previ-
ously (5), were also used. These cell lines were cultured
in D/F12 medium supplemented with 10% fetal bovine
serum (FBS), (Hyclone USA).
Animals
The athymic BALB/c nude mice (female), which weight
from 25 to 28 g, were purchased from the Animal Cen-
ter of the Chinese Academy of Medical Science. The
mice were bred in laminar flow cabinets under specific
pathogen-free conditions and handled according to the
policies and standards of Laboratory Animal Care in
China.
Stable transfection of glioma cells
To generate a recombinant AAV serotype 2 –BMPR-IB
(rAAV2-BMPR-IB) viral vector, full-length cDNA for
human BMPR-IB was obtained by EcoRI and BamH1 di-
gestion and subcloned into the pSNAV plasmid (Invitro-
gen) and was then recombined into rAAV2. U87 and
U251MG cells were infected with AAV-BMPR-IB or con-
trol virus to generate BMPR-IB-overexpressing glioblast-
oma cells. To generate a BMPR-IB small interfering
RNA (siRNA) expression vector for gene knockdown
studies, four BMPR-IB siRNAs were designed and
synthesized (Invitrogen, USA). A siRNA with the se-
quence 5′-GGACCCAGUUGUACCUAAUdTdT-3′ was
determined to be the most effective siRNA for inhibiting
BMPR-IB expression. The BMPR-IB siRNA was further
incorporated into the pSilencer plasmid (Ambion, USA).
SF763 cells were transfected with the BMPR-IB siRNA
expression vector (si-BMPR-IB) or the control vector (si-
control). The cell lines, which stably expressed BMPR-IB
siRNA, were isolated by neomycin (G418) selection.
Quantitative real-time RT-PCR
Total RNA, which derived from glioma cells, was pre-
pared using TRIzol (Gibco), and further purified using
the RNeasy Mini Kit (Qiagen). Real-time PCR was per-
formed according to the manufacturer’s instructions
using an ABI Prism 7900 sequence detection system
(Applied Biosystems, USA). Primers and probes for p21,
p27, p53, CDK2, CDK4, Skp2, BMPR-IB (human) and
GAPDH were obtained from Applied Biosystems, USA.
Liu et al. Journal of Experimental & Clinical Cancer Research 2012, 31:52
http://www.jeccr.com/content/31/1/52
Page 2 of 11

Page 3

Additional file 1: Table S1 shows the forward and reverse
primer sequences of theses genes. All samples were
tested in triplicate. The relative number of target tran-
scripts was normalized to the number of human
GAPDH transcripts in the same sample. The relative
quantitation of target gene expression was performed
using the standard curve or comparative cycle threshold
(Ct) method.
Western blot analysis
Whole-cell lysates were isolated from glioma cells and
the transplanted glioma tissues (5). Standard western
blotting was performed with monoclonal antibodies
against human BMPR-IB, p21, p27KIP1, Skp2, Cdk2,
Cdk4, p53, GFAP, Nestin and β-actin proteins(Santa
Cruz Biotechnology,USA) and the corresponding sec-
ondary antibodies (anti-rabbit IgG, anti-mouse IgG, and
anti-goat IgG; Abcam, USA). Human β-actin was used
as a loading control. These proteins were detected using
the Amersham enhanced chemiluminescence system
according to the manufacturer’s instructions.
Immunofluorescent staining
At 48 h following AAV-BMPR-IB infection, the U251 and
U87 cells were fixed in 4% paraformaldehyde-PBS. After
incubation with 0.1% Triton-PBS for 30 min and blocking
with 1% bovine serum albumin-PBS for 2 h in room
temperature, the cells were then incubated with the pri-
mary antibodies overnight in 4°C at the concentration
recommended by the supplier (a rabbit anti-phospho-
Smad1/5/8 antibody (Cell signal), a goat anti-BMPR-IB
antibody (Santa Cruz Biotechnology) and a mouse anti-
GFAP antibody (Sigma)). After washing with 0.1% Triton-
PBS three times, cells were incubated with RBITC-conjugated
rabbit anti-goat IgG and FITC-conjugated goat anti-
rabbit IgG (Santa Cruz Biotechnology) for 2 h in room
temperature. The cell nuclei were stained with DAPI.
The stained cells were visualized and mounted with a
confocal laser scanning microscope (Olympus).
Analysis of cell cycle distribution
Glioma cells were harveseted and washed with 1×PBS
three times, then fixed by 70% Ethanol for 1 h. Ethanol-
fixed cells were treated with RNase (10 mg/ml in PBS)
and stained with propidium iodide (100 μg/mL in PBS)
for 30 min at 37°C. Stained cells were tranfered to FACS
tubes and detected using flow cytometry (Becton Dickin-
son Immunocytometry Systems, San Jose, CA).
Colony formation in soft agar
U87 and U251 cells were infected with either rAAV2-
BMPR-IB or the control vector rAAV2, SF763 cells were
stably transfected with the BMPR-IB siRNA oligonucleo-
tide or control siRNA. After 48 h of infection, cells were
trypsinized, and 2×104cells were mixed with a 0.5% agar
solution in DMEM/F12 containing 10% FBS and 200 μg/mL
neomycin, then layered on top of 0.70% agar in 35 mm
culture plates. The plates were incubated at 37°C in a
humidified incubator for 10–14 days. Colonies were then
stained with 0.005% Crystal violet for 1 h, and counted
using a dissecting microscopically in 8 randomly chosen
microscope fields. Only colonies containing >50 cells
were scored.
Subcutaneous tumor growth
To study the kinetics of glioma cells growth in vivo, gli-
oma cells (3 × 106or 1 × 107cells in 50 μl of PBS) were
injected s.c. into the right armpit of nude mice. The
diameter of the resulting tumors was measured once
every 5 days. The ability of tumor formation was deter-
mined by measurement of the diameters of subcutane-
ous tumors.
Intracranial human glioma xenograft model
Glioma cells (1 × 106/4μl) were grown in metrigel for
2 h, then implanted into the right striatum of nude mice
by stereotactic injection (0.2 μl/min). The injection coor-
dinates were: anteroposterior=0; mediolateral=3.0 mm;
and dorsoventral=4.0 mm. Animals showing general or
local symptoms were killed; the remaining animals were
killed 90 days after glioma cell injection by perfusion of
4% formaldehyde. The brain of each mouse was har-
vested, fixed in 4% formaldehyde, and embedded in par-
affin. Tumor formation and phenotype were determined
by histological analysis of hematoxylin and eosin (H&E)-
stained sections. Two independent experiments were
performed, with five mice per group in each experiment.
Histology and immunohistochemistry of xenograft
tumors
Fixed Brain tissue specimens were embedded in paraffin,
sectioned, and stained with H&E according to standard
protocols. Tissue sections were immunostained using
mouse anti- GFAP and goat anti-CD34 monoclonal
antibodies (Santa Cruz Biotechnology,USA) to detect
the growth, differentiation and angiogenesis of the
xenografts.
Statistics
All of the values were calculated as mean±SE. Student’s
t-test was used to analysis the significance of the results
in vitro, whereas the significance of the results in vivo
was determined by the Mann–Whitney U test. Kaplan–
Meier survival analysis was used to analysis the overall
survival times of the glioblastoma nude mouse.
Liu et al. Journal of Experimental & Clinical Cancer Research 2012, 31:52
http://www.jeccr.com/content/31/1/52
Page 3 of 11

Page 4

Results
Expression of members of the BMPs/Smad1/5/8 signaling
pathway in different malignant glioma cell lines
We examined the mRNA and protein expressions of
BMP2, BMPR-II, BMPR-IA, BMPR-IB and Smad1/5/8
in normal astrocytes and malignant glioma cell lines
using real-time RT-PCR and western blot analysis, re-
spectively. We found that the mRNA expression of
BMPR-IB mRNA in all glioblastoma cell lines decreased
compared to normal astrocytes, while the expression of
the other genes remained similar between normal astro-
cytes and malignant glioma cell lines (Figure 1A). Fur-
thermore, the protein expression of BMPR-IB and
phospho-Smad1/5/8 in all malignant glioma cell lines
was lower than the levels in normal astrocytes; intracel-
lular protein expression of BMPR-IB was moderately
lower in SF763 cells and drastically lower in other ma-
lignant glioma cell lines compared to normal astrocytes
(Figure 1B). We overexpressed BMPR-IB in U87 and
U251 cells following rAAV infection. Forty-eight hours
after infection, a significant increase of BMPR-IB and
phospho-smad1/5/8 protein expression was confirmed
in the rAAV-BMPR-IB-infected U87 and U251 cell lines
by western blot analysis (Figure 1C). Furthermore, im-
munofluorescent staining with an anti-phospho-smad1/
5/8-specific antibody showed nuclear translocation of
phospho-smad1/5/8 after 48 h of AAV-BMPR-IB infec-
tion (Figure 1D).
Figure 1 Determination of BMPR-IB expression in normal human astrocytes and glioma cell lines. (A) Real-time-RT-PCR was used to
determine the mRNA expressions of BMPR-IB and other factors involved in BMP/BMPR signaling pathway. (B) Western blot analyses were
employed to show the protein expression of BMPR-IB, P-Smad1/5/8 and Smad1/5/8 in glioblastoma cell lines(up). Statistical analysis of results
from WB analysis(down). (C) Alterations in the expression of BMPR-IB and P-Smad1/5/8 after 48 h of BMPR-IB overexpression, determined by WB
analysis. (D) Immunofluorescence analysis of the activation of Smad1/5/8 after 48 h of BMPR-IB infection.
Liu et al. Journal of Experimental & Clinical Cancer Research 2012, 31:52
http://www.jeccr.com/content/31/1/52
Page 4 of 11

Page 5

Effects of BMPR-IB overexpression and knock-down on
the cell cycle progression of glioblastoma cells
We overexpressed BMPR-IB with rAAV in U87 and
U251 cells and suppressed BMPR-IB expression in
SF763 cells with siBMPR-IB. Forty-eight hours after in-
fection and transfection, a significant increase in BMPR-
IB protein expression in the rAAV-BMPR-IB-infected
U87 and U251 cell lines and a decrease in BMPR-IB pro-
tein expression in the BMPR-IB siRNA-transfected
SF763 cell line were confirmed by western blot analysis
(Figure 2A). Defects in the regulation of cell cycle pro-
gression are thought to be among the most common
features of glioblastoma multiforme [1]. Therefore, we
used flow cytometry to assess whether BMPR-IB expres-
sion could affect the cell cycle progression of glioblast-
oma cells. As shown in Figure 2B, the percentage of
BMPR-IB-infected U87 and U251 cells in G1/G0 phase
was higher compared to that of control vector rAAV-
infected cells. Conversely, the percentage of si-BMPR-IB
transfected SF763 cells in G0/G1 phase was lower rela-
tive to that of si-control-transfected SF763 cells.
Effects of BMPR-IB overexpression and knock-down on
the growth of glioblastoma cells in vitro
After 5 days of BMPR-IB overexpression or knock-down,
the anchorage-independent growth of BMPR-IB-overex-
pressing glioblastoma cells was drastically inhibited, as
shown by a decrease in the number and volume of col-
onies on soft agar compared with control cells, and the
anchorage-independent growth of SF763 cells treated
with siBMPR-IB was 2 times as high as that of the si-
control-treated cells. BMPR-IB overexpression decreased
the colony numbers of U251 and U87 by 55% and 66%,
and BMPR-IB knock-down caused an approximate 94%
increase in colony numbers compared with controls
(Figure 3A, B).
Effects of BMPR-IB overexpression and knock-down on
the differentiation of glioblastoma cells in vitro
The contrast photomicrographs showed that the glio-
blastoma cell lines U87 and U251 were prone to differ-
entiate after2days of
Conversely, BMPR-IB knock-down inhibited the out-
growth of neurites in SF763 cells (Figure 4A). Immuno-
fluorescence analysis showed that BMPR-IB infection
increased the expression of GFAP protein, which is a
recognized marker of astrocytic differentiation, whereas
BMPR-IB knock-down decreased the expression of
GFAP protein (Figure 4A). Further investigation using
western blot analysis showed that BMPR-IB overexpres-
sion increased the expression of GFAP protein and
inhibited the expression of Nestin, which is a marker of
CNS precursor cells. In addition, BMPR-IB knock-down
decreased the expression of GFAP protein and increased
the expression of Nestin protein (Figure 4B).
rAAV-BMPR-IB infection.
Effects of altered BMPR-IB expression on the mRNA and
protein levels of Skp2, p21, p27Kip1, Cdk2, Cdk4 and p53
in glioma cell lines
To identify the mechanisms related to the growth inhib-
ition and differentiation of glioma cells secondary to
BMPR-IB overexpression, we first examined the mRNA
Figure 2 Effects of altered BMPR-IB expression on the cell cycle of human glioblastoma cells. (A) Western blot analysis of BMPR-IB
expression in parental glioma cells, control vector–AAV and AAV-BMPR-IB-infected cells. (B) Cell cycle distribution analysis histogram. (Values are
expressed as the mean±SD, n=3. *, P<0.05).
Liu et al. Journal of Experimental & Clinical Cancer Research 2012, 31:52
http://www.jeccr.com/content/31/1/52
Page 5 of 11