MOLECULAR MEDICINE REPORTS 15: 2473-2480, 2017
Abstract. Vitamin K
2
has been demonstrated to promote
the osteogenic differentiation of mesenchymal stem cells;
however, the mechanisms underlying this effect remain
unclear. As microRNA (miR)‑133a has been identied as a
negative regulator of osteogenic differentiation, the present
study hypothesized that vitamin K
2
promoted osteogenesis by
inhibiting miR-133a. Using human bone marrow stromal cells
(hBMSCs) overexpressing miR-133a, or a control, the expres-
sion levels of osteogenesis-associated proteins, including
runt-related transcription factor 2, alkaline phosphatase and
osteocalcin, were analyzed. miR‑133a signicantly suppressed
the osteogenic differentiation of hBMSCs. To determine the
effect of vitamin K
2
on miR-133a expression and osteogenesis,
hBMSCs were treated with vitamin K
2
. Vitamin K
2
inhibited
miR-133a expression, which was accompanied by enhanced
osteogenic differentiation. Furthermore, the expression levels
of vitamin K epoxide reductase complex subunit 1, the key
protein in γ-carboxylation, were downregulated by miR-133a
overexpression and upregulated by vitamin K
2
treatment, indi-
cating a positive feedback on γ-carboxylation. The results of
the present study suggested that vitamin K
2
targets miR-133a
to regulate osteogenesis.
Introduction
microRNAs (miRNAs) are endogenous single-stranded
noncoding RNAs that negatively regulate diverse biological
and pathological processes, including apoptosis, myogenesis
and tumorigenesis. miRNAs post-transcriptionally regulate
gene expression via inhibiting mRNA translation or promoting
mRNA degradation following binding to the 3' untranslated
region of target mRNA (1-4). Accumulating evidence suggests
that miRNAs are involved in osteoblast differentiation.
Previous studies have revealed that miRNA (miR)-20a is
upregulated in naringin-induced osteogenesis in bone marrow
stromal cells (BMSCs) (5) and that inhibition of miR-103a
increased runt-related transcription factor 2 (Runx2) protein
expression levels in osteoblasts (6). A recent study has demon-
strated that miR-21 promotes osteogenesis and mineralization
in MC3T3-E1 cells by inhibiting the translation of mothers
against decapentaplegic 7 (7). Additionally, it has been
demonstrated that miR-188 is a key regulator of the age-asso-
ciated switch from osteogenic to adipogenic differentiation of
BMSCs (8). Therefore, miRNAs may be potential therapeutic
targets for the treatment of bone-associated diseases, including
osteoporosis and osteonecrosis.
miR-133a is a highly conserved miRNA. Previous
studies have revealed that mir‑133a is specically expressed
in cardiac and skeletal muscle and acts as a key regulator
during the development of these tissues (3,9). Studies have
demonstrated the expression of miR-133a in mesenchymal
stem cells, where it acts as a negative regulator of osteoblast
differentiation (10-12). Furthermore, Wang et al (13) suggested
that miR-133a expression levels in circulating monocytes were
upregulated in patients with osteoporosis and Wu et al (14)
discovered that mir-133a was significantly downregulated
in non-traumatic osteonecrosis of the femoral head samples
compared with femoral neck fracture samples, indicating that
miR-133a may be a key regulator of bone metabolism.
In humans, miR-133a appears to have a direct regulatory
effect on the expression of vitamin K epoxide reductase
complex subunit 1 (VKORC1), which is crucial in the vitamin
K (VK)-epoxide cycle (15). The VK-epoxide cycle is closely
associated with the γ-carboxylation of VK-dependent proteins
during bone metabolism, including osteocalcin (OCN), matrix
Gla protein and protein S. OCN is synthesized by osteoblasts
during the mineralization of bone; γ-carboxylation of OCN
confers greater afnity for calcium ions, whereas the failure
of carboxylation results in the accumulation of uncarboxylated
OCN, which has decreased afnity for calcium ions, reducing
Vitamin K
2
promotes mesenchymal stem cell
differentiation by inhibiting miR‑133a expression
YUELEI ZHANG
1,2*
, SHIYANG WENG
1*
, JUNHUI YIN
1
, HAO DING
1
, CHANGQING ZHANG
1
and YOUSHUI GAO
1
1
Department of Orthopedic Surgery, Shanghai Jiao Tong University Afliated Sixth People's Hospital,
Shanghai 200233;
2
Department of Orthopedic Surgery, The Second Afliated Hospital and Yuying
Children's Hospital of Wenzhou Medical University, Wenzhou, Zhejiang 325027, P.R. China
Received January 26, 2016; Accepted January 13, 2017
DOI: 10.3892/mmr.2017.6308
Correspondence to: Dr Changqing Zhang or Dr Youshui
Gao, Department of Orthopedic Surgery, Shanghai Jiao Tong
University Affiliated Sixth People's Hospital, 600 Yishan Road,
Shanghai 200233, P.R. China
E-mail: zhangcq@sjtu.edu.cn
E-mail: acerzhang5233@sjtu.edu.cn
*
Contributed equally
Key words: vitamin K
2
, microRNA-133a, osteogenesis, bone
marrow stromal cells
ZHANG et al: VITAMIN K
2
INHIBITS mir-133a EXPRESSION
2474
the bone quality (16,17). Due to the effect of vitamin K
2
(VK
2
)
on the γ-carboxylation of OCN, and other bone metabolism
effects, including the stimulation of osteogenesis and inhibi-
tion of adipogenesis (18,19), VK
2
has been used as a treatment
for osteoporosis.
As miR-133a is a regulator of osteogenesis, the present
study hypothesized that miR-133a additionally mediates
the effect of VK
2
on bone metabolism, functioning in the
following two ways: Direct suppression of mir-133a by VK
2
,
which further increases Runx2 and promotes osteogenesis
via a positive feedback mechanism, and suppression of the
expression of mir-133a by VK
2
, resulting in the upregulation
of VKORC1 and γ-carboxylation of OCN.
Materials and methods
Cell culture and treatment. Human BMSCs (hBMSCs) were
harvested from the femoral heads of patients undergoing
total hip arthroplasty, as reported in a previous study (18).
Ethical approval was obtained by the Office of Research
Ethics at the Shanghai Sixth People's Hospital of Shanghai
Jiao Tong University (Shanghai, China). hBMSCs were
cultured in complete medium consisting of alpha minimal
essential medium (MEM; Gibco; Thermo Fisher Scientic,
Inc., Waltham, MA, USA) supplemented with 10% fetal
bovine serum (FBS, Invitrogen; Thermo Fisher Scientific,
Inc.), 100 units/ml penicillin and 10 µg/ml streptomycin. The
culture medium was replaced with fresh medium every 3 days.
Cells were passaged at ~90% conuence, and cells at passages
3-6 were used in the experiments. To induce osteogenic
differentiation, hBMSCs were cultured in osteogenic induc-
tion medium, which consisted of complete medium, 10 mM
β-sodium glycerophosphate, 50 µg/ml L-ascorbic acid and
10 nM dexamethasone. The following graded concentrations
of VK
2
: 10
-5
M
(VK
-5
), 10
-6
M
(VK
-6
) and 10
-7
M
(VK
-7
), were
used to treat hBMSCs (18).
Alizarin red S staining. Following culture in osteogenic
medium for 21 days, hBMSCs were rinsed with PBS, xed
with 4% paraformaldehyde for 20 min and subsequently
stained with 40 mM alizarin red working solution for 10 min.
Following this, cells were rinsed twice with PBS, and images
were captured using an inverted phase contrast microscope.
Transfection assay. hBMSCs were seeded at 10
5
cells/well in
a 6-well plate. After 24 h, the culture medium was replaced
with fresh MEM containing 10% FBS, and 5 µg/ml polybrene
(Sigma-Aldrich; Merck Millipore, Darmstadt, Germany),
the cells were transduced with 5x10
6
TU/ml lentiviral vector
(Shanghai GenePharma Co., Ltd., Shanghai, China) carrying
miR-133a (miR-133a-control; sequence, TTT GGT CCC CTT
CAA CCA GCTG) or a miRNA negative control (miRNA
negative; sequence, TT C TCC GAA CGT GTC ACGT) using
the four plasmid system. Briefly, this system involves the
co-transfection of a recombinant virus plasmid encoding
for a transfer vector with three kinds of auxiliary packaging
vector into 293T cells. The supernatant containing the virus
particles become concentrated following 72 h, and then the
high-titer lentivirus concentrate is obtained within 293T cells.
The medium was replaced with complete culture medium 24 h
after transduction. The transduction efciency was assessed
96 h following transduction and cells were selected with
2 µg/ml puromycin until colonies of stably transduced cells
were formed.
Reverse transcription‑quantitative polymerase chain reaction
(RT‑qPCR) analysis. Following culture in osteogenic medium,
total RNA was isolated from cells using TRIzol
®
reagent; 1 µg
RNA was reverse-transcribed with the EasyScript One-step
gDNA Removal and cDNA Synthesis supermix (Beijing
Transgen Biotech Co., Ltd., Beijing, China) in a total volume of
20 µl, according to the manufacturer's protocol. Following this,
qPCR was performed using SYBR
®
Green reagents (Beijing
Transgen Biotech Co., Ltd.) for 40 cycles under the following
conditions: Denaturation at 95˚C for 30 sec, annealing at 95˚C
for 5 sec and extension at 60˚C for 30 sec. A 65‑95˚C solu-
bility curve was constructed with ABI Prism 7900 (Applied
Biosystems; Thermo Fisher Scientic, Inc.). All values were
normalized to those of GAPDH, and the fold change method
(2
-
ΔΔ
Cq
) was used for analysis (19). The primers used for the
amplication of target mRNAs are listed in Table I.
For the detection of mir-133a, miRNAs were isolated using
the EasyPure miRNA kit (Beijing Transgen Biotech Co., Ltd.)
according to the manufacturer's protocol. qPCR analysis of
mir-133a was performed using the SYBR Green RT-qPCR
kit (BioTNT Co., Ltd., Shanghai, China). The stem-loop RT
primer for miR-133a was 5'-CTCA ACT G GT GTC GTG GAG
T C G GC A ATT C AG TT G AG C AG C T GG T-3 ', a n d q P C R
pr i mers were: Forward, 5'-TT T G GT CCC CT T CA AC-3' a nd
reverse, 5'-TCA ACT GGT GTC GTGG-3'. U6 served as a
control with primers as described previously (20). The ther-
mocycling conditions were as follows: Denaturation at 95˚C
for 5 min, annealing at 95˚C for 5 sec and extension at 60˚C for
30 sec, for a total of 40 cycles. A 65‑95˚C solubility curve was
constructed, and the value of mir-133a expression was normal-
ized to that of U6 using the 2
-
ΔΔ
Cq
method.
Western blotting. Total protein extracted with Cell Lysis
buffer (#9803; Cell Signaling Technology, Inc., Danvers,
MA, USA) from osteogenic-induced hBMSCs was quanti-
ed using a Bicinchoninic Acid (BCA) kit (Thermo Fisher
Scientic, Inc.). Following this, total protein was denatured
by heating at 95˚C for 5 min, mixed with loading buffer,
and 30 µg proteins were separated by SDS-PAGE on 10%
acrylamide gels. Proteins were subsequently transferred onto
polyvinylidene diuoride membranes. The membranes were
blocked with 5% dried skimmed milk, and incubated with
rabbit anti-human GAPDH (1:2,000; #2118; Cell Signaling
Technology, Inc., Danvers, MA, USA) and rabbit anti-human
Runx2 (1:1,000; ab48811; Abcam, Cambridge, MA, USA)
antibodies at 4˚C overnight and subsequently incubated with
a horseradish peroxidase-conjugated anti-rabbit secondary
antibody (1:2,000; #7074; Cell Signaling Technology, Inc.) at
37˚C for 1 h. Following chemiluminescence with a commer-
cial assay (Pierce™ ECL Western Blotting substrate, #32106,
Thermo Fisher Scientic, Inc.), protein bands were visualized
by Odyssey scanner (LI-COR Biosciences, Lincoln, NE,
USA). The bands were quantied using Quantity One software
(version 4.52; Bio-Rad Laboratories, Inc., Hercules, CA, USA)
and normalized to GAPDH.
MOLECULAR MEDICINE REPORTS 15: 2473-2480, 2017
2475
Alkaline phosphatase (ALP) activity and staining. hBMSCs
were lysed with 0.2% Triton-X 100 and centrifuged at
14,000 x g for 10 min at 4˚C, , and the supernatant was detected
with an ALP assay kit (Nanjing Jiancheng Bioengineering
Institute, Nanjing, China) according to the manufacturer's
protocol. Absorbance values were measured at a wavelength
of 520 nm using the iMark Microplate Absorbance reader
(Bio-Rad Laboratories, Inc.) and normalized against the
protein concentration determined with the use of a BCA kit.
ALP staining was performed using a 5-Bromo-
4-chloro-3-indolyl Phosphate/Nitro Blue Tetrazolium ALP
Color Development kit (Beyotime Institute of Biotechnology,
Shanghai, China) according to the manufacturer's protocol.
Briefly, cells were rinsed with PBS, fixed with 4% para-
formaldehyde for 20 min and stained with the ALP Color
Development kit for 30 min. Images were captured using an
inverted phase contrast microscope.
OCN detection by ELISA. Following incubation of cells for
1 week, the osteogenic medium was removed and the cultures
were incubated with MEM for 24 h. Concentrations of OCN
in the medium were determined using an ELISA kit (#MK128,
Takara Bio, Inc., Otsu, Japan). The values were normalized
against the total protein concentration determined using a
BCA kit.
Statistical analysis. SPSS software (version, 20.0; IBM SPSS,
Armonk, NY, USA) was used to analyze the values in each
group. All experiments were performed in triplicate, and data
are expressed as the mean ± standard deviation. One-way
analysis of variance followed by Tukey's post hoc test was
performed to determine statistical signicance. P<0.05 was
considered to indicate a statistically signicant difference.
Results
miR‑133a expression levels are reduced during the osteogenic
differentiation of hBMSCs. To detect miR-133a expression
during osteogenic differentiation, RT-qPCR was performed
on samples collected at days 0, 1, 3, 7, 14 of differentiation.
Osteogenic differentiation was conrmed by Alizarin red S
staining (Fig. 1). Compared with undifferentiated cells at day
0, miR-133a expression levels rapidly declined during osteo-
genic induction to 0.05‑fold after 14 days (P<0.05; Fig. 2A).
By contrast, the expression levels of Runx2 (Fig. 2B), ALP
(Fig. 2C) and OCN (Fig. 2D) mRNA steadily increased in
a time-dependent manner, exhibiting a negative correlation
with miR-133a expression levels. In addition, the expression
of VKORC1 mRNA (Fig. 2E) was inversely correlated with
miR-133a expression, as previously reported (14), increasing
13-fold in osteogenic-induced hBMSCs at day 14. These
results suggested that the osteogenic differentiation of BMSCs
was associated with a reduction in miR-133a expression levels,
indicating that miR-133a may be associated with the osteoblast
differentiation of mesenchymal cells.
VK
2
inhibits miRNA‑133a expression and promotes the
osteogenic differentiation of hBMSCs. To evaluate the asso-
ciation between miR-133a and VK
2
, osteogenic induction of
hBMSCs was performed for 1 week in the presence or absence
of gradient concentrations of VK
2
, and miR-133a expression
levels were detected by RT-qPCR. miR-133a expression levels
were markedly downregulated by VK
2
treatment compared
with the control group (P<0.05), to the greatest extent by 10
-5
M
VK
2
(Fig. 3A). In addition, mRNA expression levels of Runx2
(Fig. 3B) and ALP (Fig. 3C) were signicantly upregulated
by VK
2
treatment (P<0.05), whereas those of OCN were not
(P>0.05) (Fig. 3D). ALP and OCN are two late-stage osteo-
genic differentiation markers of mesenchymal cells, and were
therefore further analyzed. Greater ALP activity was detected
in cells treated with 10
-5
M and 10
-6
M VK
2
compared with the
control (P<0.05; Fig. 3E). OCN concentrations in supernatant
Table I. The reverse transcription-quantitative polymerase chain reaction primer sequences.
Gene Forward sequence Reverse sequence
Runx2 GCGGTGCAAACTTTCTCCAG TGCTTGCAGCCTTAAATGACTC
ALP GAGAAGCCGGGACACAGTTC CCTCCTCAACTGGGATGATGC
OCN CTCACACTCCTCGCCCTATTG GCTTGGACACAAAGGCTGCAC
GAPDH CCTCGCCTTTGCCGATCC ATCATCATCCATGGTGAGCTGG
Runx2, Runt-related trnascription factor 2; ALP, alkaline phosphatase; OCN, osteocalcin.
Figure 1. Osteogenic differentiation of hBMSCs. Alizarin red S staining of
hBMSCs cultured in osteogenic differentiation medium at days 0 (control)
and 14. hBMSCs, human bone marrow stromal cells.
ZHANG et al: VITAMIN K
2
INHIBITS mir-133a EXPRESSION
2476
Figure 2. Expression levels of miR-133a and osteoblastic markers during the osteogenic differentiation of human bone marrow stromal cells, as assessed by
reverse transcription-quantitative polymerase chain reaction. Expression levels of (A) miR-133a, (B) Runx2, (C) ALP, (D) OCN and (E) VKORC1 at days 0, 1,
3, 7 and 14 of osteogenic differentiation. Data are expressed as the mean ± standard deviation (n=3).
*
P<0.05 vs. day 0. miR, microRNA; Runx2, runt‑related
transcription factor 2; ALP, alkaline phosphatase; OCN, osteocalcin; VKORC1, vitamin K epoxide reductase complex subunit 1.
Figure 3. Expression levels of miR-133a and osteoblastic markers following osteogenic induction of human bone marrow stromal cells in the presence or
absence of VK
2
. Expression levels of (A) miR-133a, (B) Runx2, (C) ALP and (D) OCN were determined by reverse transcription-quantitative polymerase chain
reaction. (E) ALP activity in cell lysates as determined by an ALP activity assay. (F) OCN concentrations in supernatant as determined by ELISA. (G) Runx2
protein expression levels were determined by western blotting and (H) quantied by densitometry; GAPDH served as a control. Data are expressed as the
mean ± standard deviation (n=3).
*
P<0.05 vs. control. miR, microRNA; VK
2
, vitamin K
2
; Runx2, runt-related transcription factor 2; ALP, alkaline phosphatase;
OCN, osteocalcin.
MOLECULAR MEDICINE REPORTS 15: 2473-2480, 2017
2477
were increased by VK
2
treatment, particularly by 10
-5
M VK
2
(P<0.05; Fig. 3F), despite the lack of effect on OCN mRNA
expression levels. Detection of Runx2 protein expression levels
by western blotting (Fig. 3G) further veried the osteogenic
stimulation effect of VK
2
treatment, revealing greater Runx2
protein expression in VK
2
‑treated cells (P<0.05), except at a
concentration of 10
-7
M VK
2
, al t hou g h not signicant for 1 0
-6
M
VK
2
(P>0.05) (Fig. 3H). Although the western blotting results
Figure 4. Expression levels of miR-133a and osteoblastic markers following osteogenic induction of miR-133a-overexpressing human bone marrow stromal cells
in the presence or absence of VK
2
. Expression levels of (A) miR-133a, (B) ALP, (C) OCN and (D) Runx2 were determined by reverse transcription-quantitative
polymerase chain reaction. (E) ALP activity in cell lysates as determined by an ALP activity assay. (F) OCN concentrations in supernatant as determined by
ELISA. (G) ALP expression, as determined by staining. (H) Runx2 protein expression levels were determined by western blotting and (I) quantied by densi-
tometry. Data are expressed as the mean ± standard deviation (n=3).
*
P<0.05 vs. miR‑133a‑control. miR, microRNA; VK
2
, vitamin K
2
; Runx2, runt-related
transcription factor 2; ALP, alkaline phosphatase; OCN, osteocalcin.