Content uploaded by Churl-Hong Chun
Author content
All content in this area was uploaded by Churl-Hong Chun on Jun 13, 2016
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
miR-370 and miR-373 regulate the pathogenesis of
osteoarthritis by modulating one-carbon metabolism via
SHMT-2 and MECP-2, respectively
Jinsoo Song,
1
* Dongkyun Kim,
1
* Churl-Hong Chun
2
and
Eun-Jung Jin
1
1
Department of Biological Sciences, College of Natural Sciences, Wonkwang
University, Iksan, Chunbuk 570-749, Korea
2
Department of Orthopedic Surgery, Wonkwang University School of
Medicine, Iksan, Chunbuk 570-749, Korea
Summary
The aim of this study was to determine the mechanism underlying
the association between one-carbon metabolism and DNA meth-
ylation during chronic degenerative joint disorder, osteoarthritis
(OA). Articular chondrocytes were isolated from human OA
cartilage and normal cartilage biopsied, and the degree of cartilage
degradation was determined by safranin O staining. We found that
the expression levels of SHMT-2 and MECP-2 were increased in OA
chondrocytes, and 30UTR reporter assays showed that SHMT-2 and
MECP-2 are the direct targets of miR-370 and miR-373, respectively,
in human articular chondrocytes. Our experiments showed that
miR-370 and miR-373 levels were significantly lower in OA
chondrocytes compared to normal chondrocytes. Overexpression
of miR-370 or miR-373, or knockdown of SHMT-2 or MECP-2
reduced both MMP-13 expression and apoptotic cell death in
cultured OA chondrocytes. In vivo, we found that introduction of
miR-370 or miR-373 into the cartilage of mice that had undergone
destabilization of the medial meniscus (DMM) surgery significantly
reduced the cartilage destruction in this model, whereas introduc-
tion of SHMT-2 or MECP-2 increased the severity of cartilage
destruction. Together, these results show that miR-370 and miR-
373 contribute to the pathogenesis of OA and act as negative
regulators of SHMT-2 and MECP-2, respectively.
Key words: human articular chondrocyte; miR-370/373;
one-carbon metabolism; osteoarthritis; SHMT-2/MECP-2; total
knee replacement.
Abbreviations
OA osteoarthritis
miR microRNA
SHMT-2 serine hydroxymethyltransferase
MECP-2 methyl-CpG-binding protein 2
MTX methotrexate
ADAMTS-5 A disintegrin and metalloproteinase with thrombospondin
motifs 5
MMP matrix metalloproteinase
5-mC 5-methylcytosine
5-hmC 5-hydroxymethylcytosine
TET Tet methylcytosine dioxygenase
DHFR dihydrofolate reductase
TYMS thymidylate synthase
DNMT DNA methyltransferase
UTR untranslated region
Introduction
Osteoarthritis (OA) was previously considered a degenerative disease but
is now believed to be a metabolically dynamic process; it takes a
worldwide toll in terms of decreased physical ability, increased morbidity,
and a high utilization of healthcare resources (Corti & Rigon, 2003). With
the exception of total knee replacement (TKR) surgery, there is currently
no safe, long-term, and effective treatment for OA pain (Wenham et al.,
2013). Methotrexate (MTX) is probably the most widely used antisynovial
treatment for inflammatory arthritis (Yazici, 2010). MTX treatment of
chondrocytes has been shown to increase the uptake of [
3
H]-thymidine
while decreasing the uptake of [
3
H]-d-uridine, which reflects altered
DNA metabolism (Neidel et al., 1998). This suggests that the mainte-
nance or repair of the thymidylate biosynthetic pathway in OA-afflicted
chondrocytes may be an effective therapeutic target.
The term ‘one-carbon metabolism’ refers to a system of interdepen-
dent metabolic pathways that facilitate the transfer of the one-carbon
units that are needed for DNA methylation, dTMP synthesis, and purine
synthesis (Stover, 2009). It also produces several metabolic intermediates
that are used in the homocysteine and folate metabolic pathways.
Methylation is responsible for controlling gene expression, stabilizing the
chromatin structure, and maintaining genomic stability; these DNA-
methylation-mediated epigenomic changes may regulate the expression
of pathogenic genes involved in various diseases (Nishida & Goel, 2011).
However, the functional roles and regulatory mechanisms that connect
one-carbon metabolism, methylation, and OA have not yet been studied
in detail.
During the past two decades, numerous studies have identified and
emphasized the importance of microRNAs (miRNAs), which are endog-
enous small noncoding RNAs that regulate gene expression by pairing
with a complementary site in the 30UTR to interfere with the translation
or stability of target transcripts (Jackson & Linsley, 2010). Our
Correspondence
Eun-Jung Jin, PhD, Department of Biological Sciences, College of Natural Sciences,
Wonkwang University, Iksan, Chunbuk 570-749, Korea. Tel.: +82-63-850-6197;
fax: +82 63 857 8837; e-mail: jineunjung@wku.ac.kr
and
Churl-Hong Chun, Department of Orthopedic Surgery, Wonkwang University
School of Medicine, Iksan, Chunbuk 570-749, Korea. Tel.: +82-63-859-1363; fax:
+82-63-852-9329; e-mail: cch@wonkwang.ac.kr
*These authors contributed equally to this work.
Accepted for publication 14 May 2015
826 ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited.
Aging Cell (2015) 14, pp826–837 Doi: 10.1111/acel.12363
Aging
Cell
understanding of miRNA biogenesis has increased, and numerous
studies have shown that miRNAs play important roles in the pathogen-
esis of various diseases (Brodeur, 2003; Tonge et al., 2013). In an effort
to define the molecular mechanisms underlying OA, researchers are
currently examining the roles of specific miRNAs in the morphological
transition, apoptosis, and/or regulation of chondrocyte-specific genes
(Miyaki et al., 2010). For example, the major miRNA, miR-140, has been
shown to function in chondrogenesis and cartilage development.
Deletion of miR-140 predisposed mice to develop age-related OA-like
changes and further increased cartilage destruction in an OA animal
model by directly targeting A disintegrin and metalloproteinase with
thrombospondin motifs 5 (ADAMTS-5), aspartyl aminopeptidase
Non-OA OA
MMP-13
(fold of Non-OA)
5-aza
*
*
*
*
**
P
P
atients
1 2 345 6 7 8 910
Normal OA Normal OA Normal OA
TET-1 TET-2 TET-3
(A)
(B)
(C)
Fig. 1 The involvement of methylation during OA pathogenesis. (A) OA cartilages (n=10) were divided into healthy zones (Non-OA) and severely damaged zones (OA)
and stained with safranin O (left panel). Normal chondrocytes were isolated form biopsy of normal cartilage, OA chondrocytes were isolated from OA area from the 10
patient samples, and the expression levels of type I and type II collagen, aggrecan, and COMP were analyzed by real-time PCR (right panel). (B) OA chondrocytes were
treated with the methyltransferase inhibitor, 5-azacytidine (5-aza). Cell images were captured, and MMP-13 expression was analyzed by real-time PCR (left panel). The
expression levels of TET-1, TET-2, and TET-3 were analyzed in chondrocytes isolated from normal biopsy samples (Normal) and OA chondrocytes. (C) Normal and OA
chondrocytes (n=4 samples in each case) were subjected to real-time PCR analysis of SHMT-2, MECP-2, DNMT-3B, MMP-13, and ADAMTS-4 expression. *P<0.05 vs.
control (normal). *, statistically different from normal (P<0.05).
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al. 827
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
(Dnpep), and insulin-like growth factor binding protein 5 (IGFBP-5).
More recently, miR-140 was shown to directly mediate matrix metallo-
protease (MMP)-13 expression in vitro (Tardif et al., 2009). Recent
studies have also revealed that several other miRNAs are involved in the
pathogenesis of OA. For example, miR-9 contributes to regulating MMP-
13, miR-98 and miR-146, which are important for controlling the
expression of tumor necrosis factor a(Jones et al., 2009). Various epi-
miRNAs, including miR-29, miR-152, and miR-290, have been shown to
play pivotal roles in regulating the epigenetic modifications that occur
through DNA methylation (Ji et al., 2013). miR-29a (Benetti et al.,
2008), miR-152 (Fabbri et al., 2007), and the miR-17~29 cluster
(Dakhlallah et al., 2013) have all been shown to target DNA methyl-
transferase (DNMT), a key regulator of DNA methylation, thereby
contributing to hepatocellular carcinoma, pulmonary fibrosis, and 16HBE
cells. In terms of cartilage, miR-140 targets histone deacetylase (HDAC)
4, which controls chondrocyte hypertrophy during skeletogenesis
(Tuddenham et al., 2006) and controls cartilage homeostasis by regu-
lating ADAMTS-5 and MMP-13 to protect against cartilage damage
(Miyaki & Asahara, 2012). Growing evidence from cartilage-based
experiments suggests that the epi-miRNAs involved in regulating various
pathogenesis-related genes may be useful targets for the diagnosis,
prevention, and treatment of OA. However, we need a detailed
understanding of the regulatory network governing these epi-mRNAs
if we hope to use them in therapeutic applications for controlling OA.
There is currently no disease-modifying therapy available for patients
with OA (Jordan et al., 2003). Growing evidence suggests that unrav-
eling the role of miRNAs in joint physiology and pathology may facilitate
the diagnosis, prevention, and treatment of OA. However, we must
SHMT-2
(Expressed fold of normal)
SHMT2Con
TNF
BCL2L11
TNFRSF10A
SYCP2
ATP6V1G2
CD40LG
FASLG
APAF1
CASP3
NOL3
SPATA2
GADD45A
CD40
CASP7
CFLAR
BAX
CASP1
TNFRSF1A
FAS
CASP9
DFFA
CASP6
CYLD
ABL1
BIRC2
TP3
IGF1
IRGM
ULK1
SNCA
BCL2
TP53
IFNG
BCL2L1
ATG5
PIK3C3
FAS
ATG12
BECN1
BAX
CASP3
NFKB1
APP
AKT1
MAP1LC3A
MAPK8
GAA
ATG16L1
HTT
ATG7
CTSB
SQSTM1
RPS6KB1
ATG3
CTSS
INS
ESR1
TNF
SHMT2Con
Patients
123 4567 8910
*
*
*
*
***
*
*
MMP-2
MMP-13
GAPDH
GAPDH
SHMT-2
*
*
Apoptosis (%)
Si-SHMT-2
– + : SHMT2
SHMT2
5-aza
SHMT2
– A B C D : siRNA
Normal
OA
si-SHMT-2
(A)
(C)
(B)
Fig. 2 SHMT-2 is involved in OA pathogenesis in vitro. (A) OA chondrocytes (n=10 patients) were subjected to real-time PCR analysis of SHMT-2 expression and
represented as fold of normal chondrocytes. (B) SiRNA-mediated knockdown of SHMT-2 was confirmed by immunoblotting (left upper panel), and the protein levels of
MMP-2 and MMP-13 (left lower panel) were analyzed by immunoblotting and apoptosis (left right) was analyzed by MuseTM Cell Cycle assay. (C) Gene profile of apoptosis-
related genes (left panel) and autophagy-related genes (right panel) was examined and expressed as heat map (upper panel) and bar graph (lower panel). ‘Red’ color
represents significant decrease in expression levels. Bars show the mean SD of three individual experiments. (D) Mice were subjected to destabilization of the medial
meniscus (DMM) surgery as a model for OA and infected with lentiviruses expressing SHMT-2 (SHMT2) or its specific siRNA (siSHMT-2). DMM and sham-operated (Sham,
control) cartilages were stained with safranin O. Cartilage destruction was scored according to the guidelines of the OARSI histopathology initiative, and the average
thickness of cartilage was plotted. Inserted image showed the successful lentiviral infection. Mouse cartilage was scrapped into slide and captured image under fluorescent
microscopy. *P<0.05 vs. control (normal). *, statistically different from normal (P<0.05).
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al.
828
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
understand the network of expressed mRNAs and their relationships
with cognate genes to further develop miRNAs for the therapeutic
control of OA. In this prospective study, we examined the potential
functional role of miR-370 and miR-373 in OA pathogenesis. We identify
serine hydroxymethyltransferase (SHMT)-2 and methyl-CpG-binding
protein (MECP)-2 as targets of miR-370 and miR-373, respectively, and
hypothesize that these interactions contribute to the pathogenesis of
OA. Finally, we propose that miR-370 and miR-373 could be potent
therapeutic targets for OA.
Result
Cartilage samples were obtained from patients with OA (n=10;
average age, 72.6) undergoing total knee replacement surgery; they
were divided into non-OA (a relatively heathy area of OA cartilage, non-
OA chondrocytes) and OA regions (a severely damaged area of OA
cartilage, OA chondrocytes), and articular chondrocytes were isolated
and cultured. Normal chondrocytes were isolated from biopsy samples of
normal cartilage (up to 0.1 g) (n=10; average age, 45.6). Consistent
with a previous report (Song et al., 2013), we observed loss of the
cartilage matrix as analyzed by safranin O staining with drastic severe
degradation in damaged area of OA cartilage (OA) compared to
relatively healthy area of OA cartilage (non-OA) and increased MMP-13
levels in OA chondrocytes compared to non-OA chondrocytes (Fig. 1A).
Furthermore, type I collagen, type II collagen, aggrecan, and COMP
levels were analyzed to confirm the intrinsic characteristics of normal or
OA chondrocytes (Fig. 1A). Epigenetic mechanisms (e.g., DNA methyl-
ation, chromatin modification, and noncoding RNAs) have been
suggested to regulate the OA-specific genes responsible for cartilage
degradation (Hashimoto et al., 2009). Here, we found that treatment
with the DNA methyltransferase inhibitor, 5-azacytidine, reduced the
level of MMP-13 (Fig. 1B). DNA methylation at the 5 position of cytosine
(5-mC), which can be converted to 5-hydroxymethylcytosine (5-hmC) by
the ten-eleven translocation (TET) family proteins, has emerged as a key
epigenetic marker that plays essential roles in various biological and
pathological processes (Ito et al., 2011). We also found that the
expression level of TET-1 was significantly lower in OA chondrocytes
compared to normal chondrocytes which were isolated from biopsy
sample of normal cartilages (n=5; average age, 45.3) (Fig. 1B). These
data suggest that dysregulation of methylation may be responsible for
OA-inducing events.
Folate status can affect the degree of global DNA methylation and
has been associated with gene silencing, indicating that epigenetic gene
silencing during development might be modified by one-carbon metab-
olism (Gonzalez, 2013). Furthermore, a recent study suggested that one-
carbon metabolism may be involved in OA pathogenesis, and proposed
that regulation of this pathway could be a therapeutic target for OA.
Here, we found that exposure of OA chondrocytes to folate increased
the levels of SHMT-2, MECP-2, DNMT-3B (all of which are involved in the
methylation process), MMP-13, and ADAM-TS (Fig. 1C).
The exact regulatory mechanisms that modulate one-carbon metab-
olism have not been well studied, although SHMT-1, SHMT-2a,
thymidylate synthase (TYMS), and dihydrofolate reductase (DHFR) are
known to be involved (Anderson & Stover, 2009). Methylene-THF, which
is generated by SHMT, acts as a one-carbon donor for the TYMS-
catalyzed conversion of dUMP to thymidylate, generating dihydrofolate.
Thereafter, DHFR catalyzes the NADPH-dependent reduction of dihy-
drofolate to regenerate THF for subsequent cycles of de novo thymidyl-
ate synthesis (Anderson et al., 2012). Therefore, we examined the
expression level of SHMT-2 in OA and normal chondrocytes. The
expression level of SHMT-2 was increased in OA chondrocytes compared
to normal chondrocytes (Fig. 2A). To further investigate the role of
SHMT-2, we introduced four different small interfering RNAs (siRNAs)
against SHMT-2 into the chondrocyte cell line. The knockdown of SHMT-
2 by all four siRNAs, which was confirmed by immunoblotting (Fig. 2B,
left upper panel), was found to reduce protein levels of MMP-2 and
n = 10 mice/group
DMM/SHMT-2
DMM/siSHMT-2
DMM
Sham
(D)
Fig. 2 Continued.
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al. 829
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
MMP-13 in OA chondrocytes compared to normal chondrocytes
(Fig. 2B, left lower panel). Chondrocyte apoptosis was significantly
increased with knockdown of SHMT-2 (Fig. 2B, middle panel). On the
other hand, overexpression of SHMT-2 significantly induced MMP-13
protein level and this increased MMP-13 protein was suppressed by
cotreatment of 5-azacytidine (Fig. 2B, right panel) providing the insight
that SHMT-2 mediates MMP-13 methylation during OA pathogenesis.
Moreover, RNA levels of apoptosis-related genes and autophagy-
related genes were significantly downregulated by knockdown of SHMT-
2 in OA chondrocytes (Fig. 2C). In in vivo DMM OA model, cartilage
destruction as visualized by safranin O and trichrome staining and OARSI
scoring was significantly increased by overexpression of SHMT-2,
whereas knockdown of SHMT-2 successfully blocked cartilage degrada-
tion induced by DMM surgery (Fig. 2D).
In an effort to identify the regulatory molecules that act on SHMT-2,
we focused on miRNAs, which affect many biological and pathological
responses by targeting key functional genes. We analyzed the expression
patterns of miRNAs using normal chondrocytes (n=4) isolated from
biopsy sample of normal cartilage and OA chondrocytes (n=4) isolated
from OA cartilage. We searched miRNAs that involved in Epigenetic
Regulation Database as well as degenerative diseases such as OA using
several miR databases (www.mirbase.org, www.mirdb.org, www.tar-
getscan.org, www.microrna.org, http://210.46.85.180:8080/EpimiR)
and selected miRNAs to analyze. Among we analyzed, miR-370 and
miR-373 seem highly OA specific (Fig. 3A). Furthermore, decreased
levels of miR-370 and miR-373 were observed in OA chondrocytes from
majority of patients, as confirmed by analysis of MMP-13 expression
(Fig. 3B). To see whether these two OA-specific miRNAs are involved in
the regulation of SHMT-2, we cloned the entire 30UTR of SHMT-2 into a
luciferase reporter vector, electroporated the vector into cells along with
the precursor of miR-370 or a cognate nontargeting negative control,
and assayed cell lysates for luciferase expression. We found that
chondrocytes transfected with the SHMT-2 30UTR-driven vector plus
pre-miR-370 exhibited significantly less luciferase activity but mutated
SHMT-2 30UTR-driven vector plus pre-miR-370 did not alter luciferase
activity (Fig. 4A, left panel) compared to cells that received the reporter
plus the nontargeting negative control, suggesting that miR-370 may be
involved in the SHMT-2-regulated pathogenesis of OA. However,
chondrocytes transfected with the SHMT-2 30UTR-driven vector plus
pre-miR-373 showed no difference in luciferase activity compared to
cells that received the reporter plus the nontargeting negative control
(data not shown). In addition, induction of miR-370 decreased SHMT-2
RNA level, whereas knockdown of miR-370 increased SHMT-2 RNA level
in OA chondrocytes (Fig. 4A, right panel).
miR-32-5p
miR-200c-3p
miR-146a-5p
miR-194-5p
miR-126-3p
miR-181b-5p
miR-181a-5p
miR-19a-3p
miR-19b-3p
miR-29b-3p
miR-101-3p
miR-210
miR-18a-5p
miR-423-5p
miR—7-5p
miR-155-5p
miR-99a-5p
miR-125a-5p
miR-16-5p
miR-195-5p
miR-27b-3p
miR-26a-5p
let-7b-5p
T^T\
T^T\
let-7f-5p
miR-26b-5p
miR-30d-5p
miR-17-5p
miR-140-3p
miR-185-5p
miR-15a-5p
miR-28-5p
miR-130a-3p
miR-93-5p
miR-186-5p
miR-425-5p
miR-25-5p
miR-196b-5p
miR-128
miR-30c-5p
miR-30e-5p
miR-30b-5p
miR-424-5p
miR-374a-5p
miR-92a-3p
let-7g-5p
let-7c
miR-15b-5p
miR-103a-3p
miR-376c-3p
miR-30a-5p
miR-151a-5p
miR-191-5p
miR-302a
miR-20a-5p
miR-106b-5p
T^T\
miR-143-3p
miR-29a-3p
miR-29c-3p
miR-125b-5p
miR-24-3p
let-7a-5p
miR-23a-3p
miR-23b-3p
miR-27a-3p
miR-100-5p
miR-222-5p
miR-21-5p
miR-22-3p
miR-96-5p
miR-142-3p
miR-122-5p
miR-142-5p
miR-150-5p
miR-223-3p
miR-141-3p
miR-124-3p
miR-9-5p
miR-302c-3p
miR-302a-3p
miR-302b-3p
miR-144-3p
miR-370
miR-373
miR-373/RNU6B
(fold of normal)
Average
miR-370/RNU6B
(fold of normal)
Average
*
u vh
patients : 123 4 XYZ[
QQ
QQ
Q
Q
QQ
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
(A)
(B)
Fig. 3 miR-370 and miR-373 are involved in OA pathogenesis. (A) Expression levels of miRNAs were analyzed using normal chondrocytes (n=4) isolated from biopsy
sample of normal cartilage and OA chondrocytes (n=4) isolated from severely damaged OA cartilage and presented as heat map. ‘Red’ color represents significant decrease
in expression levels (upper panel). Examples of correlation graph on miRNA profile between normal vs. OA chondrocytes are shown (lower panel). (B) Expression levels of
miR-370 and miR-373 were analyzed in OA chondrocytes (n=10) isolated from severely damaged OA cartilage, compared to average value of normal chondrocytes (n=4),
and represented as fold of normal chondrocytes. *P<0.05 vs. control (normal). *, statistically different from normal (P<0.05).
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al.
830
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
To investigate whether modulation of miR-370 induced the same
effects as modulation of SHMT-2, we altered the expression level of miR-
370 using its specific inhibitor or precursors (Fig. 4B). Introduction of
miR-370 as confirmed by real-time PCR into OA chondrocytes reduced
the protein levels of MMP-2 and MMP-13 that are typically increased in
OA chondrocytes. On the other hand, inhibition of miR-370 in OA
chondrocytes increased the expression levels of these proteins. In
addition, annexin V staining revealed that apoptotic cell death was
increased in OA chondrocytes (9.94%) compared to normal chondro-
cytes (3.62%). However, the induction of miR-370 in OA chondrocytes
significantly reduced this apoptotic cell death to 4.56%, whereas
inhibition of miR-370 increased cell death among OA chondrocytes.
Overexpression of miR-370 suppressed the RNA levels of apoptosis-
related and autophagy-related genes (Fig. 4C). Co-induction of miR-370
significantly inhibited SHMT-2-induced MMP-13 expression and SHMT-
2-upregulated apoptosis-related gene cell death in normal chondrocytes
(Fig. 4D). In addition, folate treatment reduced miR-370 levels and
apoptosis of OA chondrocytes (Fig. 4E), suggesting that miR-370 may be
involved in one-carbon metabolism. In in vivo study using DMM mice,
the most severe cartilage destruction (as visualized by safranin O
staining) was observed among mice infected with anti-miR-370- and
SHMT-2-encoding lentiviruses, whereas infection of DMM mice with
miR-370-expressing lentiviruses significantly ameliorated DMM-induced
cartilage destruction (Fig. 4F). The MMP-13 RNA level as examined using
RNA extracted from paraffin sections showed same expression pattern
with the degree of cartilage degradation.
Studies have shown that 5-mC can inhibit gene transcription by
attracting methyl-CpG-binding domain (MBD)-containing proteins (e.g.,
methyl-CpG-binding protein; MECP-2); these proteins ‘read’ methylation
marks, bind to methylated DNA at methylation sites (Srinivasan et al.,
2013), and affect chromatin condensation by recruiting corepressor
proteins (e.g., SIN3A and histone-modifying enzymes) (Kadonaga,
1998). We found that the expression level of MECP-2 was significantly
increased in OA chondrocytes compared to normal chondrocytes
(Fig. 5A). To further examine the potential role of MECP-2 in OA
pathogenesis, we modulated MECP-2 level exogenously using its specific
siRNA or overexpression vector and confirmed the efficiency of MECP-2-
specific siRNA or MECP overexpression vector using real-time PCR
(A)
(C)
(B)
Fig. 4 miR-370 regulates OA pathogenesis by targeting SHMT-2. (A) Luciferase reporter activity was driven by the 30UTR of SHMT-2 or mutated 30UTR of SHMT-2 (30UTR
mt) with or without forced expression of miR-370 (left panel). OA chondrocytes were treated with miR-370 precursor (miR-370) and miR-370 inhibitor (anti-miR-370), and
SHMT-2 level was analyzed by real-time PCR (right panel). (B) The efficiencies of the miR-370 precursor (miR-370) and miR-370 inhibitor (anti-miR-370) in normal
chondrocytes were analyzed by real-time PCR (left panel). Normal and OA chondrocytes were treated with a miR-370 precursor (miR-370) or an inhibitor of miR-370 (anti-
miR-370), and the protein expression levels of MMP-2 and MMP-13 were analyzed by immunoblotting. GAPDH was used as a loading control (middle panel). Apoptosis was
analyzed by MuseTM Cell Cycle assay (right panel). (C) Gene profile of apoptosis-related genes (left panel) and autophagy-related genes (right panel) was examined and
expressed as bar graph. Bars show the mean SD of three individual experiments. (D) OA chondrocytes were infected with SHMT-2-expressing lentiviruses or SHMT-2-
specific siRNAs in the presence of anti-miR-370. The infection efficiency was assessed by fluorescent microscopy as both SHMT-2-expressing lentiviruses and SHMT-2-specific
siRNAs were tagged with GFP, and MMP-13 protein expression was analyzed by immunoblotting (left panel). Gene profile of apoptosis-related genes was examined and
expressed as bar graph. Bars show the mean SD of three individual experiments (right panel). (E) OA chondrocytes were treated with folate, and miR-370 expression was
analyzed by real-time PCR (left panel) and apoptosis was analyzed by MuseTM Cell Cycle analysis (right panel). (F) DMM and sham-operated (Sham) mice were infected with
miR-370 precursor (miR-370)- or the miR-370 inhibitor (anti-miR-370)-encoding lentiviruses in the presence or absence of SHMT2 (SHMT -2)-or SHMT-2-specific siRNA (siSHMT-
2)-encoding lentiviruses, and cartilage samples were stained with safranin O. Cartilage destruction was scored and average cartilage thickness was plotted. RNA was isolated
from cartilage sections, and the expression level of MMP-13 was analyzed by real-time PCR. *P<0.05 vs. control (normal). *, statistically different from normal (P<0.05).
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al. 831
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
(Fig. 5B). Overexpression of MECP-2 increased apoptosis as assessed by
annexin V staining in OA cell chondrocytes, and knockdown of MECP-2
reduced apoptosis and induced the protein level of type II collagen in OA
chondrocytes. In addition, knockdown of MECP-2 significantly inhibited
apoptosis-related genes in OA chondrocytes (Fig. 5C).
To further test whether MECP-2 is a target of miR-370 or miR-373,
we cloned the entire 30UTR of MECP-2 into a luciferase reporter vector,
electroporated the vector into chondrocytes along with the precursor of
miR-370 or miR-373 or a cognate nontargeting negative control, and
assayed cell lysates for luciferase expression. We found that cells
transfected with the MECP-2 30UTR-driven vector plus pre-miR-373
exhibited significantly less luciferase activity, but mutated MECP-2
30UTR-driven vector plus pre-miR-373 did not alter luciferase activity
compared to cells that received the reporter plus the nontargeting
negative control (Fig. 6A, left panel). In addition, induction of miR-373
decreased RNA level of MECP-2, whereas knockdown of miR-373
increased RNA level of MECP-2 in OA chondrocytes (Fig. 6A, right
panel). Next, to investigate whether modulation of miR-373 induced the
same effects as modulation of MECP-2, we altered the expression level
of miR-373 using its specific inhibitor or precursors (Fig. 6B). Annexin V
staining revealed that the induction of miR-373 in OA chondrocytes
significantly reduced this apoptotic cell death to 2.26%, whereas
inhibition of miR-370 increased cell death up to 10% in normal
chondrocytes. Overexpression of MECP-2 significantly increased MMP-
13 mRNA expression and apoptosis, whereas siRNA-mediated knock-
down of MECP-2 significantly decreased MMP-13 mRNA expression and
apoptosis (Fig. 6C). Co-induction of miR-373 significantly inhibited
MECP-2-upregulated apoptosis-related gene cell death in normal chon-
drocytes (Fig. 6D). In addition, folate treatment decreased the level of
miR-373 in OA chondrocytes suggesting that miR-373 may be involved
in one-carbon metabolism (Fig. 6E). In in vivo study using DMM mice,
overexpression of MECP2 and knockdown of miR-373 induced severe
cartilage degradation, whereas knockdown of MECP2 and overexpres-
sion of miR-373 successfully blocked cartilage degradation induced by
DMM surgery. The most severe cartilage destruction (as visualized by
safranin O staining and OARSI scoring) was observed among mice
infected with the combination of anti-miR-373- and MECP2-encoding
lentiviruses (Fig. 6F). The MMP-13 RNA level as examined using RNA
extracted from paraffin sections showed same expression pattern with
the degree of cartilage degradation.
Discussion
One-carbon metabolism is linked to DNA synthesis, DNA methylation,
amino acid metabolism, and cell proliferation, and dysfunctions of
genes involved in one-carbon metabolism have been associated with
increased risks for various diseases, including cancer and anemia (Jones
et al., 1998; Copped
e, 2014). DNA methylation, which is a crucial
(D) (E)
(F)
Fig. 4 Continued.
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al.
832
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
epigenetic alteration, is subject to particular control by one-carbon
metabolism. SHMT is an important protein in one-carbon metabolism;
it catalyzes the reversible conversion of glycine to serine, which
provides metabolites (e.g., formyl groups) that are important for purine
biosynthesis, and the methyl groups needed for pyrimidine synthesis,
homocysteine remethylation, and other reactions that are important for
cellular homeostasis (Locasale, 2013). The folate metabolites generated
via the action of SHMT play important role in maintaining normal
methylation patterns, DNA stability, and genetic variation (Blom &
Smulders, 2011). SHMT-1 and SHMT-2 have been associated with a
wide variety of human phenotypes, including neural tube defects
(Duthie, 2011), childhood acute leukemia (Relton et al., 2004), rectal
carcinoma (Vijayakrishnan & Houlston, 2010), and prostate cancer
(Koml
osi et al., 2010). Shmt1+/mice accumulate uracil within nuclear
DNA (Collin et al., 2009) and show increased susceptibility to neural
tube defects (Collin et al., 2009) and intestinal cancer (MacFarlane
et al., 2008). Another important set of proteins involved in one-carbon
metabolism are the DNA methyltransferases (DNMTs). DNMT dysfunc-
tion has been associated with impaired memory consolidation,
decreased synaptic plasticity, memory deficits, and learning disabilities
(Beaudin et al., 2011). DNMT-3B gene mutations have been shown to
cause the immunodeficiency–centromeric instability–facial anomalies
syndrome, which is characterized by hypomethylation of pericentro-
meric repeats (Macfarlane et al., 2011). DNMT is recruited by MECP-2,
which acts as a transcriptional regulator by modulating the expression
of methylation-sensitive genes. Notably, MECP-2 deficiency has been
associated with Rett syndrome, a neurological disorder in humans. In
the present study, we show that SHMT-2 and MECP-2 are upregulated
in OA chondrocytes, and overexpression of both SHMT-2 and MECP-2
induced severe cartilage degradation in the DMM mouse model. This
supports the idea that misregulation of one-carbon metabolism may be
involved in the pathogenesis of OA.
Recently, Stone and colleagues performed a Monte Carlo-based in
silico analysis and found that several miRNAs could play important roles
in regulating one-carbon metabolism (Monsey et al., 2011). They
suggested miR-22 and miR-125 as possible master regulators, and
miR-344-5p/484 and miR-488 as possible master coregulators that may
influence the genes involved in one-carbon metabolism. However, we do
not yet fully understand the detailed functions of miRNAs in modulating
one-carbon metabolism. Here, we present the first evidence suggesting
that miR-370 and miR-373 may potently regulate one-carbon metabo-
lism by directly targeting SHMT-2 and MECP-2, respectively.
Studies comparing the expression levels of miRNAs between OA
tissue specimens and normal cartilage tissue/non-OA tissue specimens
(A) (B)
(C)
Fig. 5 MECP-2 is involved in OA pathogenesis in vitro. (A) RNA level of MECP-2 was analyzed using OA chondrocytes (n=10), compared to average value of normal
chondrocytes (n=4), and represented as fold of normal. (B) Chondrocytes were infected with MECP-2- or MECP-2 siRNA-encoding lentiviruses. Efficiency was confirmed by
real-time PCR (left panel), viability was analyzed by annexin V staining (middle panel), apoptosis was analyzed with the MuseTM Cell Cycle assay (right panel), and the protein
level of type II collagen was analyzed by immunoblotting (right upper panel). GAPDH was used as a loading control. (C) Gene profile of apoptosis-related genes (left panel)
and autophagy-related genes (right panel) was examined with knockdown of MECP-2 in OA chondrocytes. *P<0.05 vs. control (normal). *, statistically different from
normal (P<0.05).
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al. 833
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
have identified several miRNAs as being involved in the pathogenesis of
OA. For example, Iliopoulos et al. showed that 16 miRNAs (nine
upregulated and seven downregulated) were differentially expressed in
OA cartilage compared with normal controls. Seventeen miRNAs whose
expression varied by fourfold or more in normal vs. late-stage OA
cartilage were identified. Of these differentially expressed miRNAs, miR-
27b (downregulated in OA) directly targets MMP-13 expression (Stone
et al., 2011); miR-22 (upregulated in OA) directly regulates PPARA and
BMP-7 expression in cartilage; miR-9 inhibits MMP13 secretion in
isolated human chondrocytes; and miR-146a is highly expressed in early
OA cartilage and has been shown to control knee joint homeostasis and
OA-associated algesia by balancing inflammatory responses in the
cartilage and synovium. Here, we report the first study showing that the
levels of miR-370 and miR-373 are significantly lower in OA chondro-
cytes. The previously identified targets for miR-370 and miR-373 were
largely associated with cancer pathogenesis (e.g., oncogenes and tumor
suppressors), suggesting that the utilized target genes may contribute to
the tissue- and cell-type-specific performances of these miRNAs (Akhtar
et al., 2010). For example, miR-370 targets FOXO1 in gastric mucosal
cancer and has been associated with disease progression (Voorhoeve
et al., 2007). miR-373 regulates MMP activity by modulating MAPK
signaling or targeting mTOR and SIRT1 in human fibrosarcoma cells (Fan
et al., 2013), and by targeting ZIP4 in pancreatic cancer (Liu & Wilson,
2012). Here, we found that miR-370 and miR-373 were significantly
downregulated in OA chondrocytes compared to normal chondrocytes.
This dysregulates cell survival by targeting the newly discovered targets,
SHMT-2 and MECP-2, respectively; apoptosis was induced by the
inhibition of miR-370 or miR-373, and also by the overexpression of
SHMT-2 or MECP-2. These results suggest that the decreased functions
of miR-370 and miR-373 in OA chondrocytes lead to the disinhibition of
(A) (B)
(D)(C) (E)
Fig. 6 miR-373 regulates OA pathogenesis by targeting MECP-2. (A) Luciferase reporter assays were performed on OA chondrocytes harboring a vector construct driven by
the 30UTR of MECP-2 or mutated 30UTR of MECP-2 (30UTR mt) with or without forced expression of miR-373 (left panel). RNA level of MECP was examined with treatment of
the miR-373 precursor (miR-373) or miR-373 inhibitor (anti-miR-373) (right panel). (B) Chondrocytes were treated with miR-373 precursor (miR-373) or miR-373 inhibitor
(anti-miR-373), the efficiency was confirmed by real-time PCR (left panel), viability was analyzed by annexin V staining (middle panel), and apoptosis was analyzed with the
MuseTM Cell Cycle assay (right panel). (C) Normal chondrocytes were treated with MECP-2-expressing vector or the MECP-2-specific siRNA (MECP-2 siRNA) in the presence
of the miR-373 precursor (miR-373). RNA level of MMP-13 was analyzed by real-time PCR, and apoptosis was analyzed by MuseTM Cell Cycle assay. (D) Gene profile of
apoptosis-related genes was examined and expressed as bar graph. Bars show the mean SD of three individual experiments. (E) OA chondrocytes were treated with
folate, and miR-373 expression was analyzed by real-time PCR. (F) DMM and sham-operated mice were infected with lentiviruses encoding the miR-370 precursor (miR-373)/
miR-370 inhibitor (anti-miR-370) or MECP-2/MECP-2-specific siRNA (siMECP2), or combination of miR-370 precursor (miR-373)/miR-370 inhibitor (anti-miR-370) and MECP-
2/MECP-2-specific siRNA (siMECP2), and cartilage samples were stained with safranin O. Cartilage destruction was scored according to the guidelines of the OARSI
histopathology initiative, and the average thickness of cartilage was plotted. RNA was isolated from cartilage sections, and the expression level of MMP-13 was analyzed by
real-time PCR. *P<0.05 vs. control (normal). *, statistically different from normal (P<0.05).
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al.
834
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
SHMT-2 and MECP-2, respectively, thereby promoting apoptotic cell
death.
In sum, we herein show for the first time that miR-370 and miR-373
directly target and negatively regulate SHMT-2 and MECP-2, respec-
tively, in human articular chondrocytes, where they contribute to the
pathogenesis of OA. These results provide novel insights into OA and
may facilitate the development of therapeutic approaches based on the
modulation of miR-370 and/or miR-373.
Experimental procedures
Primary cell cultures
Human articular cartilage specimens were obtained from patients
undergoing total knee arthroplasty designated as OA chondrocytes, and
normal chondrocytes were obtained from biopsy sample of normal
cartilage designated as normal chondrocytes in this study. Tissue collection
was approved by the Human Subjects Committee of Wonkwang
University (Chunbuk, Korea). Chondrocytes were extracted and seeded
at 1.5 910
4
cells cm
2
in DMEM (Gibco Invitrogen, Grand Island, NY,
USA) supplemented with 10% fetal bovine serum (FBS), 100 units mL
1
penicillin, and 100 lgmL
1
streptomycin (Gibco Invitrogen).
Cell viability assay
MuseTM Annexin V (Millipore, Billerica, MA, USA) assay kit and MuseTM
apoptosis kit (Millipore) were used to detect apoptotic cell death.
Quantification of miRNA and real-time quantitative RT–PCR
of mRNA
The expression levels of various miRNAs and mRNAs were quantified
using the TaqMan microRNA and gene expression assays, respectively
(Applied Biosystems: Foster City, CA, USA), according to the manu-
facturer’s protocols. miRNA expression was normalized with respect to
that of the RNU43 small nuclear RNA (endogenous control). For
assessment of mRNA, transcripts were quantified by real-time quan-
titative polymerase chain reaction (RT–PCR) and normalized with
respect to the expression of GAPDH.
miRNA inhibitor-mediated knockdown and pre-miR-370/373-
mediated upregulation of miR-370 and miR-373
The precursors or inhibitors of miR-370 and miR-373 (Ambion, Austin,
TX, USA) were electroporated into cells using a square-wave generator
(BTX-830; Gentronics, San Diego, CA, USA) with 20 ms, 200 square
pulses. Scrambled oligos or miRNAs were used as negative controls.
Reporter vectors and DNA constructs
The full length of 30untranslated regions (UTRs) of MECP-2 and SHMT-2
were PCR-amplified and cloned downstream of the CMV-driven firefly
luciferase cassette in the pMIR-Report vector (Ambion). For miRNA target
validation, cells were electroporated as described above with 25–50 ng
of each firefly luciferase reporter construct, 150–175 ng of empty
(F)
Fig. 6 Continued.
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al. 835
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
pcDNA3 vector (Clontech, Mountain View, CA, USA), 200 ng pcDNA3
harboring the Renilla luciferase gene (transfection control), and 30 pmol
of pre-miR-370, pre-miR-373, or pre-miR-neg (Ambion). At 24 h post-
transfection, the activities of the firefly and Renilla luciferase were
assayed using commercially available kits (Promega, San Luis Obispo, CA,
USA). The ratios of firefly luciferase activity vs. Renilla luciferase activity
are given as normalized relative light units (RLUs).
Production of lentiviral particles
Human 293FT cells were transfected with lentiviral vectors encoding miR-
370, miR-373, SHMT-2, or MECP-2 or the negative control lentivirus
(AppliedBiological Materials, Canada)using the third-generationpackaging
mix (Applied Biological Materials, Canada) and Lentifectin (Applied Biolog-
ical Materials, Canada).The cells were then culturedovernight in Opti-MEM
I medium (Gibco Invitrogen). The supernatant was collected and lentiviral
particles were concentrated using a Lenti-X Concentrator (Clontech).
Western blotting
Cell lysates were separated by 10% polyacrylamide gel electrophoresis,
transferred to a nitrocellulose membrane (Schleicher and Schuell, Dassel,
Germany), and probed with antibodies against type II collagen (Santa
Cruz Biotech, Santa Cruz, CA, USA), caspase-7 (Cell Signaling, Danvers,
MA, USA), MMP-2 (Abcam, Cambridge, MA, USA), MMP-13 (Biovision,
San Francisco, CA, USA), and GAPDH (Santa Cruz Biotech). The blots
were developed with a peroxidase-conjugated secondary antibody and
visualized using an electrochemiluminescence (ECL) system (Pierce
Biotechnology Inc., Rockford, MN, USA).
Arthritic cartilage, experimental OA, and histology of OA
cartilage
Human OA cartilage was sourced from individuals undergoing arthroplasty
for OA of the knee joint. The Wonkwang University Hospital Institutional
Review Board approved the use of these materials, and all individuals
provided written informed consent prior to the operative procedure. The
human OA cartilage samples were frozen, sectioned at a thickness of
10 lm, fixed in paraformaldehyde, and stained with Alcian blue.
Experimental OA was induced in 8-week-old male mice by
destabilization of the medial meniscus (DMM) surgery using C57BL/6
mice. Sham-operated animals injected with empty lentiviruses (mock
transduction) were used as controls. DMM surgery was performed in
male mice, and lentiviruses were injected intra-articularly with
1910
9
plaque-forming units (PFU) of lentiviral vectors encoding
miR-370, miR-373, SHMT-2, or MECP-2 every week for 8 weeks.
The mice were sacrificed 2 weeks after stopping injection and
subjected to histological and biochemical analyses. Cartilage destruc-
tion in mice was examined using safranin O staining. Briefly, knee
joints were fixed in 4% paraformaldehyde, decalcified in 0.5 MEDTA
(pH 7.4) for 14 days at 4 °C, and embedded in paraffin. The paraffin
blocks were sectioned at a thickness of 6 lm. The sections were
deparaffinized in xylene, hydrated with graded ethanol, and stained
with safranin O.
Extraction of RNA from FFPE (formalin-fixed paraffin-
embedded) tissues
RNA was extracted from paraffin-embedded samples using a MasterPure
kit (Epicentre Biotechnologies, Madison, WI, USA). Briefly, each FFPE
tissue was resuspended in MasterPure Tissue and Cell Lysis solution (final
concentration, 0.15 mg mL
1
proteinase K) and incubated at 65 °C for
30 min. The MasterPure MPCTM protein precipitation reagent was then
added, and nucleic acids were precipitated with isopropanol and
pelleted by centrifugation at 10 000 9gfor 10 min at 4 °C. The
supernatant was carefully removed. The pellet was washed twice with
75% ethanol, allowed to dry, and then treated with 30 lL TE buffer
containing 40 units of RNase inhibitor.
Statistical analysis
A two-tailed Student’s t-test or one-way ANOVA followed by the
Student–Newman–Keuls post hoc test was used to determine the
significance of the differences between results, with P<0.05 taken as
indicating a significant difference.
Funding
This works was supported by National Research Foundation (NRF) of
Korea Grant funded by the Korean Governments by the Korea
government (MSIP) [2013R1A1A2011999], [NRF-2013R1A2A2A0106
7194], and [2011-0030130]. The funders had no role in study design,
data collection and analysis, decision to publish, or preparation of the
manuscript.
Author contributions
E-J Jin designed and performed the experiments, analyzed the data, and
wrote the manuscript. J Song and D Kim performed the experiments and
analyzed the data. C-H Chun provided crucial reagents and analyzed the data.
Conflict of interest
All authors state that they have no conflict of interest.
References
Akhtar N, Rasheed Z, Ramamurthy S, Anbazhagan AN, Voss FR, Haqqi TM (2010)
MicroRNA-27b regulates the expression of matrix metalloproteinase 13 in
human osteoarthritis chondrocytes. Arthritis Rheum. 62, 1361–1371.
Anderson DD, Stover PJ (2009) SHMT1 and SHMT2 are functionally redundant in
nuclear de novo thymidylate biosynthesis. PLoS One 4, e5839.
Anderson DD, Woeller CF, Chiang EP, Shane B, Stover PJ (2012) Serine
hydroxymethyltransferase anchors de novo thymidylate synthesis pathway to
nuclear lamina for DNA synthesis. J. Biol. Chem. 287, 7051–7062.
Beaudin AE, Abarinov EV, Noden DM, Perry CA, Chu S, Stabler SP, Allen RH, Stover
PJ (2011) Shmt1 and de novo thymidylate biosynthesis underlie folate-responsive
neural tube defects in mice. Am. J. Clin. Nutr. 93, 789–798.
Benetti R, Gonzalo S, Jaco I, Mu~
noz P, Gonzalez S, Schoeftner S, Murchison E, Andl
T, Chen T, Klatt P, Li E, Serrano M, Millar S, Hannon G, Blasco MA (2008) A
mammalian microRNA cluster controls DNA methylation and telomere recom-
bination via Rbl2-dependent regulation of DNA methyltransferases. Nat. Struct.
Mol. Biol. 15, 268–279.
Blom HJ, Smulders Y (2011) Overview of homocysteine and folate metabolism.
With special references to cardiovascular disease and neural tube defects.
J. Inherit. Metab. Dis. 34,75–81.
Brodeur GM (2003) Neuroblastoma: biological insights into a clinical enigma. Nat.
Rev. Cancer 3, 203–216.
Collin SM, Metcalfe C, Zuccolo L, Lewis SJ, Chen L, Cox A, Davis M, Lane JA,
Donovan J, Smith GD, Neal DE, Hamdy FC, Gudmundsson J, Sulem P, Rafnar T,
Benediktsdottir KR, Eeles RA, Guy M, Kote-Jarai Z, UK Genetic Prostate Cancer
Study Group, Morrison J, Al Olama AA, Stefansson K, Easton DF, Martin RM
(2009) Association of folate-pathway gene polymorphisms with the risk of
prostate cancer: a population-based nested case-control study, systematic
review, and meta-analysis. Cancer Epidemiol. Biomarkers Prev. 18, 2528–2539.
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al.
836
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
Copped
e F (2014) Epigenetic biomarkers of colorectal cancer: focus on DNA
methylation. Cancer Lett. 342, 238–247.
Corti MC, Rigon C (2003) Epidemiology of osteoarthritis: prevalence, risk factors
and functional impact. Aging Clin. Exp. Res. 15, 359–363.
Dakhlallah D, Batte K, Wang Y, Cantemir-Stone CZ, Yan P, Nuovo G, Mikhail A,
Hitchcock CL, Wright VP, Nana-Sinkam SP, Piper MG, Marsh CB (2013)
Epigenetic regulation of miR-17~92 contributes to the pathogenesis of pulmo-
nary fibrosis. Am. J. Respir. Crit. Care Med. 187, 397–405.
Duthie SJ (2011) Folate and cancer: how DNA damage, repair and methylation
impact on colon carcinogenesis. J. Inherit. Metab. Dis. 34, 101–109.
Fabbri M, Garzon R, Cimmino A, Liu Z, Zanesi N, Callegari E, Liu S, Alder H,
Costinean S, Fernandez-Cymering C, Volinia S, Guler G, Morrison CD, Chan KK,
Marcucci G, Calin GA, Huebner K, Croce CM (2007) MicroRNA-29 family reverts
aberrant methylation in lung cancer by targeting DNA methyltransferases 3A
and 3B. Proc. Natl Acad. Sci. USA 104, 15805–15810.
Fan C, Liu S, Zhao Y, Han Y, Yang L, Tao G, Li Q, Zhang L (2013) Upregulation of
miR-370 contributes to the progression of gastric carcinoma via suppression of
FOXO1. Biomed. Pharmacother. 67, 521–526.
Gonzalez A (2013) Osteoarthritis year 2013 in review: genetics and genomics.
Osteoarthritis Cartilage 21, 1443–1451.
Hashimoto K, Oreffo RO, Gibson MB, Goldring MB, Roach HI (2009) DNA
demethylation at specific CpG sites in the IL1B promoter in response to
inflammatory cytokines in human articular chondrocytes. Arthritis Rheum. 60,
3303–3313.
Iliopoulos D, Malizos KN, Oikonomou P, Tsezou A (2008) Integrative microRNA and
proteomic approaches identify novel osteoarthritis genes and their collaborative
metabolic and inflammatory networks. PLoS One 3, e3740.
Ito S, Shen L, Dai Q, Wu SC, Collins LB, Swenberg JA, He C, Zhang Y (2011) Tet
proteins can convert 5-methylcytosine to 5-formylcytosine and 5-carboxylcyto-
sine. Science 333, 1300–1303.
Jackson A, Linsley PS (2010) The therapeutic potential of microRNA modulation.
Discov. Med. 9, 311–318.
Ji W, Yang L, Yuan J, Yang L, Zhang M, Qi D, Duan X, Xuan A, Zhang W, Lu J,
Zhuang Z, Zeng G (2013) MicroRNA-152 targets DNA methyltransferase 1 in
NiS-transformed cells via a feedback mechanism. Carcinogenesis 34, 446–453.
Jones SW, Watkins G, Le Good N, Roberts S, Murphy CL, Brockbank SM, Needham
MR, Read SJ, Newham P (2009) The identification of differentially expressed
microRNA in osteoarthritic tissue that modulate the production of TNF-alpha and
MMP13. Osteoarthritis Cartilage 17, 464–472.
Jones PL, Veenstra GJ, Wade PA, Vermaak D, Kass SU, Landsberger N, Strouboulis
J, Wolffe AP (1998) Methylated DNA and MeCP2 recruit histone deacetylase to
repress transcription. Nat. Genet. 19, 187–191.
Jordan KM, Arden NK, Doherty M, Bannwarth B, Bijlsma JW, Dieppe P, Gunther K,
Hauselmann H, Herrero-Beaumont G, Kaklamanis P, Lohmander S, Leeb B,
Lequesne M, Mazieres B, Martin-Mola E, Pavelka K, Pendleton A, Punzi L, Serni
U, Swoboda B, Verbruggen G, Zimmerman-Gorska I, Dougados M, Standing
Committee for International Clinical Studies Including Therapeutic Trials ESCISIT
(2003) EULAR Recommendations 2003: an evidence based approach to the
management of knee osteoarthritis: Report of a Task Force of the Standing
Committee for International Clinical Studies Including Therapeutic Trials (ESCI-
SIT). Ann. Rheum. Dis. 62, 1145–1155.
Kadonaga JT (1998) Eukaryotic transcription: an interlaced network of transcrip-
tion factors and chromatin-modifying machines. Cell 92, 307–313.
Koml
osi V, Hitre E, Pap E, Adleff V, R
eti A, Sz
ekely E, B
ır
o A, Rudnai P, Schoket B,
M€
uller J, T
oth B, Ott
oS,K
asler M, Kralov
anszky J, Budai B (2010) SHMT1 1420
and MTHFR 677 variants are associated with rectal but not colon cancer. BMC
Cancer 10, 525.
Liu P, Wilson MJ (2012) miR-520c and miR-373 upregulate MMP9 expression by
targeting mTOR and SIRT1, and activate the Ras/Raf/MEK/Erk signaling pathway
and NF-jB factor in human fibrosarcoma cells. J. Cell. Physiol. 227, 867–876.
Locasale JW (2013) Serine, glycine and one-carbon units: cancer metabolism in full
circle. Nat. Rev. Cancer 13, 572–583.
MacFarlane AJ, Liu X, Perry CA, Flodby P, Allen RH, Stabler SP, Stover PJ (2008)
Cytoplasmic serine hydroxymethyltransferase regulates the metabolic partition-
ing of methylenetetrahydrofolate but is not essential in mice. J. Biol. Chem. 283,
25846–25853.
Macfarlane AJ, Perry CA, McEntee MF, Lin DM, Stover PJ (2011) Shmt1 heterozy-
gosity impairs folate-dependent thymidylate synthesis capacity and modifies risk of
Apc(min)-mediated intestinal cancer risk. Cancer Res. 71, 2098–2107.
Miyaki S, Asahara H (2012) Macro view of microRNA function in osteoarthritis.
Nat. Rev. Rheumatol. 8, 543–552.
Miyaki S, Sato T, Inoue A, Otsuki S, Ito Y, Yokoyama S, Kato Y, Takemoto F,
Nakasa T, Yamashita S, Takada S, Lotz MK, Ueno-Kudo H, Asahara H (2010)
MicroRNA-140 plays dual roles in both cartilage development and homeostasis.
Genes Dev. 24, 1173–1185.
Monsey MS, Ota KT, Akingbade IF, Hong ES, Schafe GE (2011) Epigenetic
alterations are critical for fear memory consolidation and synaptic plasticity in the
lateral amygdala. PLoS One 6, e19958.
Neidel J, Sova L, Schroers B, Sintermann F, Manzke O, Bohlen H (1998) Effects of
methotrexate on normal articular cartilage in vitro and in vivo. Ann. Rheum. Dis.
57, 414–421.
Nishida N, Goel A (2011) Genetic and epigenetic signatures in human hepatocel-
lular carcinoma: a systematic review. Curr. Genomics 12, 130–137.
Relton CL, Wilding CS, Laffling AJ, Jonas PA, Burgess T, Binks K, Tawn EJ, Burn J
(2004) Low erythrocyte folate status and polymorphic variation in folate-related
genes are associated with risk of neural tube defect pregnancy. Mol. Genet.
Metab. 81, 273–281.
Song J, Lee M, Kim D, Han J, Chun CH, Jin EJ (2013) MicroRNA-181b regulates
articular chondrocytes differentiation and cartilage integrity. Biochem. Biophys.
Res. Commun. 431, 210–214.
Srinivasan S, Selvan ST, Archunan G, Gulyas B, Padmanabhan P (2013) MicroRNAs
–the next generation therapeutic targets in human diseases. Theranostics 3,
930–942.
Stone N, Pangilinan F, Molloy AM, Shane B, Scott JM, Ueland PM, Mills JL, Kirke
PN, Sethupathy P, Brody LC (2011) Bioinformatic and genetic association analysis
of microRNA target sites in one-carbon metabolism genes. PLoS One 6, e21851.
Stover PJ (2009) One-carbon metabolism-genome interactions in folate-associated
pathologies. J. Nutr. 139, 2402–2405.
Tardif G, Hum D, Pelletier JP, Duval N, Martel-Pelletier J (2009) Regulation of the
IGFBP-5 and MMP-13 genes by the microRNAs miR-140 and miR-27a in human
osteoarthritic chondrocytes. BMC Musculoskelet. Disord. 10, 148.
Tonge DP, Tugwood JD, Kelsall J, Gant TW (2013) The role of microRNAs in the
pathogenesis of MMPi-induced skin fibrodysplasia. BMC Genom. 14, 338.
Tuddenham L, Wheeler G, Ntounia-Fousara S, Waters J, Hajihosseini MK, Clark I,
Dalmay T (2006) The cartilage specific microRNA-140 targets histone deacet-
ylase 4 in mouse cells. FEBS Lett. 580, 4214–4217.
Vijayakrishnan J, Houlston RS (2010) Review candidate gene association studies
and risk of childhood acute lymphoblastic leukemia: a systematic review and
meta-analysis. Haematologica 95, 1405–1414.
Voorhoeve PM, le Sage C, Schrier M, Gillis AJ, Stoop H, Nagel R, Liu YP, van Duijse
J, Drost J, Griekspoor A, Zlotorynski E, Yabuta N, De Vita G, Nojima H, Looijenga
LH, Agami R (2007) A genetic screen implicates miRNA-372 and miRNA-373 as
oncogenes in testicular germ cell tumors. Adv. Exp. Med. Biol. 604,17–46.
Wenham CY, Grainger AJ, Hensor EM, Caperon AR, Ash ZR, Conaghan PG (2013)
Methotrexate for pain relief in knee osteoarthritis: an open-label study.
Rheumatology (Oxford) 52, 888–892.
Yazici Y (2010) Long-term safety of methotrexate in the treatment of rheumatoid
arthritis. Clin. Exp. Rheumatol. 28, S65–S67.
ª2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
The roles of miR-370 and miR-373 in OA pathogenesis, J. Song et al. 837
